From 96fe420bebc159599fb8da1080e9ff207bdb650a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jingle=20Wu=20=E5=90=B3=E9=87=91=E5=9C=8B?= Date: Tue, 21 Apr 2026 07:00:10 +0000 Subject: [PATCH 001/411] Input: elan_i2c - add ic type 0x19 The 0x19 is valid 3000 serial ic type too. Signed-off-by: Jingle Wu Link: https://patch.msgid.link/KL1PR01MB511699853D1B66D137C06806DC2C2@KL1PR01MB5116.apcprd01.prod.exchangelabs.com Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index fee1796da3d0..7475803c6ce4 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -162,6 +162,9 @@ static int elan_get_fwinfo(u16 ic_type, u8 iap_version, u16 *validpage_count, case 0x15: *validpage_count = 1024; break; + case 0x19: + *validpage_count = 2032; + break; default: /* unknown ic type clear value */ *validpage_count = 0; From 8f9d6cd6d3916add4c47a9dd1622e4fc057f877b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jingle=20Wu=20=E5=90=B3=E9=87=91=E5=9C=8B?= Date: Tue, 21 Apr 2026 07:02:33 +0000 Subject: [PATCH 002/411] Input: elan_i2c - increase device reset wait timeout after update FW Extend wait_for_completion_timeout from 300ms to 700ms to ensure sufficient time for device reset after firmware update. Signed-off-by: Jingle Wu Link: https://patch.msgid.link/KL1PR01MB5116031986614B3214EF2F30DC2C2@KL1PR01MB5116.apcprd01.prod.exchangelabs.com Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_i2c.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/mouse/elan_i2c_i2c.c b/drivers/input/mouse/elan_i2c_i2c.c index a9057d124a88..88d4070d4b44 100644 --- a/drivers/input/mouse/elan_i2c_i2c.c +++ b/drivers/input/mouse/elan_i2c_i2c.c @@ -690,7 +690,7 @@ static int elan_i2c_finish_fw_update(struct i2c_client *client, if (error) { dev_err(dev, "device reset failed: %d\n", error); } else if (!wait_for_completion_timeout(completion, - msecs_to_jiffies(300))) { + msecs_to_jiffies(700))) { dev_err(dev, "timeout waiting for device reset\n"); error = -ETIMEDOUT; } From 2905281cbda52ec9df540113b35b835feb5fafd3 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 20 Apr 2026 18:00:27 +0200 Subject: [PATCH 003/411] Input: usbtouchscreen - clamp NEXIO data_len/x_len to URB buffer size MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit nexio_read_data() pulls data_len and x_len from a packed __be16 header in the device's interrupt packet and then walks packet->data[0..x_len) and packet->data[x_len..data_len) comparing each byte against a threshold. Both fields are 16-bit on the wire (max 65535). The existing adjustments shave at most 0x100 / 0x80 off, so the loop bound can still reach roughly 0xfeff. The URB transfer buffer for NEXIO is rept_size (1024) bytes from usb_alloc_coherent(), with the first 7 occupied by the packed header — so packet->data[] has 1017 valid bytes. read_data() callbacks are not given urb->actual_length, and nothing else bounds the walk. A device that lies about its length can get a ~64 KiB out-of-bounds read past the coherent DMA allocation. The first index whose byte exceeds NEXIO_THRESHOLD lands in begin_x / begin_y and from there into the reported touch coordinates, so adjacent kernel memory contents leak to userspace as ABS_X / ABS_Y events. Far enough out, the read can also hit an unmapped page and fault. Fix this all by clamping data_len to the buffer's data[] capacity and x_len to data_len. Cc: Dmitry Torokhov Fixes: 5197424cdccc ("Input: usbtouchscreen - add NEXIO (or iNexio) support") Cc: stable Assisted-by: gkh_clanker_t1000 Signed-off-by: Greg Kroah-Hartman Link: https://patch.msgid.link/2026042026-chlorine-epidermis-fd6d@gregkh Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/usbtouchscreen.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/input/touchscreen/usbtouchscreen.c b/drivers/input/touchscreen/usbtouchscreen.c index daa28135f887..0bbacb517c28 100644 --- a/drivers/input/touchscreen/usbtouchscreen.c +++ b/drivers/input/touchscreen/usbtouchscreen.c @@ -1067,6 +1067,11 @@ static int nexio_read_data(struct usbtouch_usb *usbtouch, unsigned char *pkt) if (x_len > 0xff) x_len -= 0x80; + if (data_len > usbtouch->data_size - sizeof(*packet)) + data_len = usbtouch->data_size - sizeof(*packet); + if (x_len > data_len) + x_len = data_len; + /* send ACK */ ret = usb_submit_urb(priv->ack, GFP_ATOMIC); if (ret) From 6cdc46b38cf146ce81d4831b6472dbf7731849a2 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sun, 26 Apr 2026 21:09:33 -0700 Subject: [PATCH 004/411] Input: xpad - fix out-of-bounds access for Share button xpadone_process_packet() receives len directly from urb->actual_length and uses it to index the share-button byte at data[len - 18] or data[len - 26]. Since both len and data[0] are under the device's control, a broken controller can send a GIP_CMD_INPUT packet with actual_length < 18 (e.g. 5 bytes) and reach this code path, causing accesses beyond the actual array. Fix this by calculating the offset and checking bounds against the packet length. Reported-by: Greg Kroah-Hartman Fixes: 4ef46367073b ("Input: xpad - fix Share button on Xbox One controllers") Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 0549fdc5a985..19ce90da89e9 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -1077,10 +1077,10 @@ static void xpadone_process_packet(struct usb_xpad *xpad, u16 cmd, unsigned char input_report_key(dev, BTN_START, data[4] & BIT(2)); input_report_key(dev, BTN_SELECT, data[4] & BIT(3)); if (xpad->mapping & MAP_SHARE_BUTTON) { - if (xpad->mapping & MAP_SHARE_OFFSET) - input_report_key(dev, KEY_RECORD, data[len - 26] & BIT(0)); - else - input_report_key(dev, KEY_RECORD, data[len - 18] & BIT(0)); + u32 offset = (xpad->mapping & MAP_SHARE_OFFSET) ? 26 : 18; + + if (len >= offset) + input_report_key(dev, KEY_RECORD, data[len - offset] & BIT(0)); } /* buttons A,B,X,Y */ From 852534744c2d35626a604f128ff0b8ec12805591 Mon Sep 17 00:00:00 2001 From: Christofer Jonason Date: Wed, 4 Mar 2026 10:07:27 +0100 Subject: [PATCH 005/411] iio: adc: xilinx-xadc: Fix sequencer mode in postdisable for dual mux MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit xadc_postdisable() unconditionally sets the sequencer to continuous mode. For dual external multiplexer configurations this is incorrect: simultaneous sampling mode is required so that ADC-A samples through the mux on VAUX[0-7] while ADC-B simultaneously samples through the mux on VAUX[8-15]. In continuous mode only ADC-A is active, so VAUX[8-15] channels return incorrect data. Since postdisable is also called from xadc_probe() to set the initial idle state, the wrong sequencer mode is active from the moment the driver loads. The preenable path already uses xadc_get_seq_mode() which returns SIMULTANEOUS for dual mux. Fix postdisable to do the same. Fixes: bdc8cda1d010 ("iio:adc: Add Xilinx XADC driver") Cc: stable@vger.kernel.org Signed-off-by: Christofer Jonason Reviewed-by: Andy Shevchenko Reviewed-by: Nuno Sá Reviewed-by: Salih Erim Signed-off-by: Jonathan Cameron --- drivers/iio/adc/xilinx-xadc-core.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/iio/adc/xilinx-xadc-core.c b/drivers/iio/adc/xilinx-xadc-core.c index e257c1b94a5f..3980dfacbcd7 100644 --- a/drivers/iio/adc/xilinx-xadc-core.c +++ b/drivers/iio/adc/xilinx-xadc-core.c @@ -817,6 +817,7 @@ static int xadc_postdisable(struct iio_dev *indio_dev) { struct xadc *xadc = iio_priv(indio_dev); unsigned long scan_mask; + int seq_mode; int ret; int i; @@ -824,6 +825,12 @@ static int xadc_postdisable(struct iio_dev *indio_dev) for (i = 0; i < indio_dev->num_channels; i++) scan_mask |= BIT(indio_dev->channels[i].scan_index); + /* + * Use the correct sequencer mode for the idle state: simultaneous + * mode for dual external mux configurations, continuous otherwise. + */ + seq_mode = xadc_get_seq_mode(xadc, scan_mask); + /* Enable all channels and calibration */ ret = xadc_write_adc_reg(xadc, XADC_REG_SEQ(0), scan_mask & 0xffff); if (ret) @@ -834,11 +841,11 @@ static int xadc_postdisable(struct iio_dev *indio_dev) return ret; ret = xadc_update_adc_reg(xadc, XADC_REG_CONF1, XADC_CONF1_SEQ_MASK, - XADC_CONF1_SEQ_CONTINUOUS); + seq_mode); if (ret) return ret; - return xadc_power_adc_b(xadc, XADC_CONF1_SEQ_CONTINUOUS); + return xadc_power_adc_b(xadc, seq_mode); } static int xadc_preenable(struct iio_dev *indio_dev) From 673478bc29cf72010faaf293c1c8c667393335a0 Mon Sep 17 00:00:00 2001 From: Pengpeng Hou Date: Thu, 2 Apr 2026 13:40:15 +0800 Subject: [PATCH 006/411] iio: chemical: mhz19b: reject oversized serial replies mhz19b_receive_buf() appends each serdev chunk into the fixed MHZ19B_CMD_SIZE receive buffer and advances buf_idx by len without checking that the chunk fits in the remaining space. A large callback can therefore overflow st->buf before the command path validates the reply. Reset the reply state before each command and reject oversized serial replies before copying them into the fixed buffer. When an oversized reply is detected, wake the waiter and report -EMSGSIZE instead of overwriting st->buf. Fixes: 4572a70b3681 ("iio: chemical: Add support for Winsen MHZ19B CO2 sensor") Cc: stable@vger.kernel.org Signed-off-by: Pengpeng Hou Acked-by: Gyeyoung Baek Signed-off-by: Jonathan Cameron --- drivers/iio/chemical/mhz19b.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/iio/chemical/mhz19b.c b/drivers/iio/chemical/mhz19b.c index 3c64154918b1..9d4cf432919e 100644 --- a/drivers/iio/chemical/mhz19b.c +++ b/drivers/iio/chemical/mhz19b.c @@ -52,6 +52,8 @@ struct mhz19b_state { struct completion buf_ready; u8 buf_idx; + bool buf_overflow; + /* * Serdev receive buffer. * When data is received from the MH-Z19B, @@ -106,6 +108,10 @@ static int mhz19b_serdev_cmd(struct iio_dev *indio_dev, int cmd, u16 arg) cmd_buf[8] = mhz19b_get_checksum(cmd_buf); /* Write buf to uart ctrl synchronously */ + st->buf_idx = 0; + st->buf_overflow = false; + reinit_completion(&st->buf_ready); + ret = serdev_device_write(serdev, cmd_buf, MHZ19B_CMD_SIZE, 0); if (ret < 0) return ret; @@ -121,6 +127,9 @@ static int mhz19b_serdev_cmd(struct iio_dev *indio_dev, int cmd, u16 arg) if (!ret) return -ETIMEDOUT; + if (st->buf_overflow) + return -EMSGSIZE; + if (st->buf[8] != mhz19b_get_checksum(st->buf)) { dev_err(dev, "checksum err"); return -EINVAL; @@ -240,6 +249,14 @@ static size_t mhz19b_receive_buf(struct serdev_device *serdev, { struct iio_dev *indio_dev = dev_get_drvdata(&serdev->dev); struct mhz19b_state *st = iio_priv(indio_dev); + size_t remaining = MHZ19B_CMD_SIZE - st->buf_idx; + + if (len > remaining) { + st->buf_idx = 0; + st->buf_overflow = true; + complete(&st->buf_ready); + return len; + } memcpy(st->buf + st->buf_idx, data, len); st->buf_idx += len; From b66f922f6a4fa92840f662fbcfeb4f8a0f774bcc Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Fri, 27 Mar 2026 20:27:54 +0800 Subject: [PATCH 007/411] iio: light: veml6070: Fix resource leak in probe error path The driver calls i2c_new_dummy_device() to create a dummy device, then calls i2c_smbus_write_byte(). If i2c_smbus_write_byte() fails and returns, the cleanup via devm_add_action_or_reset() was never registered, so the dummy device leaks. Switch to devm_i2c_new_dummy_device() which registers cleanup atomically with device creation, eliminating the error-path window. Fixes: 7501bff87c3e ("iio: light: veml6070: add action for i2c_unregister_device") Reviewed-by: Andy Shevchenko Signed-off-by: Felix Gu Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/light/veml6070.c | 14 ++------------ 1 file changed, 2 insertions(+), 12 deletions(-) diff --git a/drivers/iio/light/veml6070.c b/drivers/iio/light/veml6070.c index 74d7246e5225..4bbd86d0cb46 100644 --- a/drivers/iio/light/veml6070.c +++ b/drivers/iio/light/veml6070.c @@ -245,13 +245,6 @@ static const struct iio_info veml6070_info = { .write_raw = veml6070_write_raw, }; -static void veml6070_i2c_unreg(void *p) -{ - struct veml6070_data *data = p; - - i2c_unregister_device(data->client2); -} - static int veml6070_probe(struct i2c_client *client) { struct veml6070_data *data; @@ -281,7 +274,8 @@ static int veml6070_probe(struct i2c_client *client) if (ret < 0) return ret; - data->client2 = i2c_new_dummy_device(client->adapter, VEML6070_ADDR_DATA_LSB); + data->client2 = devm_i2c_new_dummy_device(&client->dev, client->adapter, + VEML6070_ADDR_DATA_LSB); if (IS_ERR(data->client2)) return dev_err_probe(&client->dev, PTR_ERR(data->client2), "i2c device for second chip address failed\n"); @@ -292,10 +286,6 @@ static int veml6070_probe(struct i2c_client *client) if (ret < 0) return ret; - ret = devm_add_action_or_reset(&client->dev, veml6070_i2c_unreg, data); - if (ret < 0) - return ret; - return devm_iio_device_register(&client->dev, indio_dev); } From 1a772719318c11e146f6fbe621fffd230a6f456a Mon Sep 17 00:00:00 2001 From: Radu Sabau Date: Wed, 8 Apr 2026 13:32:13 +0300 Subject: [PATCH 008/411] iio: adc: ad4695: Fix call ordering in offload buffer postenable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ad4695_enter_advanced_sequencer_mode() was called after spi_offload_trigger_enable(). That is wrong because ad4695_enter_advanced_sequencer_mode() issues regular SPI transfers to put the ADC into advanced sequencer mode, and not all SPI offload capable controllers support regular SPI transfers while offloading is enabled. Fix this by calling ad4695_enter_advanced_sequencer_mode() before spi_offload_trigger_enable(), so the ADC is fully configured before the first CNV pulse can occur. This is consistent with the same constraint that already applies to the BUSY_GP_EN write above it. Update the error unwind labels accordingly: add err_exit_conversion_mode so that a failure of spi_offload_trigger_enable() correctly exits conversion mode before clearing BUSY_GP_EN. Fixes: f09f140e3ea8 ("iio: adc: ad4695: Add support for SPI offload") Reviewed-by: Nuno Sá Reviewed-by: David Lechner Signed-off-by: Radu Sabau Cc: Stable@vger.kernel.org Signed-off-by: Jonathan Cameron --- drivers/iio/adc/ad4695.c | 23 ++++++++--------------- 1 file changed, 8 insertions(+), 15 deletions(-) diff --git a/drivers/iio/adc/ad4695.c b/drivers/iio/adc/ad4695.c index cda419638d9a..53642de7330d 100644 --- a/drivers/iio/adc/ad4695.c +++ b/drivers/iio/adc/ad4695.c @@ -876,14 +876,14 @@ static int ad4695_offload_buffer_postenable(struct iio_dev *indio_dev) if (ret) goto err_unoptimize_message; - ret = spi_offload_trigger_enable(st->offload, st->offload_trigger, - &config); + ret = ad4695_enter_advanced_sequencer_mode(st, num_slots); if (ret) goto err_disable_busy_output; - ret = ad4695_enter_advanced_sequencer_mode(st, num_slots); + ret = spi_offload_trigger_enable(st->offload, st->offload_trigger, + &config); if (ret) - goto err_offload_trigger_disable; + goto err_exit_conversion_mode; mutex_lock(&st->cnv_pwm_lock); pwm_get_state(st->cnv_pwm, &state); @@ -895,23 +895,16 @@ static int ad4695_offload_buffer_postenable(struct iio_dev *indio_dev) ret = pwm_apply_might_sleep(st->cnv_pwm, &state); mutex_unlock(&st->cnv_pwm_lock); if (ret) - goto err_offload_exit_conversion_mode; + goto err_offload_trigger_disable; return 0; -err_offload_exit_conversion_mode: - /* - * We have to unwind in a different order to avoid triggering offload. - * ad4695_exit_conversion_mode() triggers a conversion, so it has to be - * done after spi_offload_trigger_disable(). - */ - spi_offload_trigger_disable(st->offload, st->offload_trigger); - ad4695_exit_conversion_mode(st); - goto err_disable_busy_output; - err_offload_trigger_disable: spi_offload_trigger_disable(st->offload, st->offload_trigger); +err_exit_conversion_mode: + ad4695_exit_conversion_mode(st); + err_disable_busy_output: regmap_clear_bits(st->regmap, AD4695_REG_GP_MODE, AD4695_REG_GP_MODE_BUSY_GP_EN); From 761e8b489e6cf166c574034b70637f8a7eadd0ee Mon Sep 17 00:00:00 2001 From: Antoniu Miclaus Date: Tue, 31 Mar 2026 13:13:00 +0300 Subject: [PATCH 009/411] iio: gyro: adis16260: fix division by zero in write_raw MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a validation check for the sampling frequency value before using it as a divisor. A user writing zero to the sampling_frequency sysfs attribute triggers a division by zero in the kernel. Fixes: 089a41985c6c ("staging: iio: adis16260 digital gyro driver") Signed-off-by: Antoniu Miclaus Reviewed-by: Nuno Sá Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/gyro/adis16260.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/iio/gyro/adis16260.c b/drivers/iio/gyro/adis16260.c index 586e6cfa14a9..91b9c5f18ec4 100644 --- a/drivers/iio/gyro/adis16260.c +++ b/drivers/iio/gyro/adis16260.c @@ -287,6 +287,9 @@ static int adis16260_write_raw(struct iio_dev *indio_dev, addr = adis16260_addresses[chan->scan_index][1]; return adis_write_reg_16(adis, addr, val); case IIO_CHAN_INFO_SAMP_FREQ: + if (val <= 0) + return -EINVAL; + if (spi_get_device_id(adis->spi)->driver_data) t = 256 / val; else From bb21ee31f5753a7972148798fd7dfb841dd33bdb Mon Sep 17 00:00:00 2001 From: Svyatoslav Ryhel Date: Thu, 16 Apr 2026 14:14:42 +0300 Subject: [PATCH 010/411] iio: Fix iio_multiply_value use in iio_read_channel_processed_scale The function iio_multiply_value returns IIO_VAL_INT (1) on success or a negative error number on failure, while iio_read_channel_processed_scale should return an error code or 0. This creates a situation where the expected result is treated as an error. Fix this by checking the iio_multiply_value result separately, instead of passing it as a return value. Fixes: 05f958d003c9 ("iio: Improve iio_read_channel_processed_scale() precision") Signed-off-by: Svyatoslav Ryhel Reviewed-by: Hans de Goede Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/inkern.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/iio/inkern.c b/drivers/iio/inkern.c index 0df0ab3de270..9ce20cb05a9b 100644 --- a/drivers/iio/inkern.c +++ b/drivers/iio/inkern.c @@ -738,7 +738,11 @@ int iio_read_channel_processed_scale(struct iio_channel *chan, int *val, if (ret < 0) return ret; - return iio_multiply_value(val, scale, ret, pval, pval2); + ret = iio_multiply_value(val, scale, ret, pval, pval2); + if (ret < 0) + return ret; + + return 0; } else { ret = iio_channel_read(chan, val, NULL, IIO_CHAN_INFO_RAW); if (ret < 0) From 7e5c0f97c66ad538b87c04a640573371fb434b4f Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Thu, 16 Apr 2026 11:01:22 +0200 Subject: [PATCH 011/411] iio: adc: nxp-sar-adc: Avoid division by zero When Common Clock Framework is disabled, clk_get_rate() returns 0. This is used as part of the divisor to perform nanosecond delays with help of ndelay(). When the above condition occurs the compiler, due to unspecified behaviour, is free to do what it wants to. Here it saturates the value, which is logical from mathematics point of view. However, the ndelay() implementation has set a reasonable upper threshold and refuses to provide anything for such a long delay. That's why code may not be linked under these circumstances. To solve the issue, provide a wrapper that calls ndelay() when the value is known not to be zero. Fixes: 4434072a893e ("iio: adc: Add the NXP SAR ADC support for the s32g2/3 platforms") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202603311958.ly6uROit-lkp@intel.com/ Signed-off-by: Andy Shevchenko Acked-by: Daniel Lezcano Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/nxp-sar-adc.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/iio/adc/nxp-sar-adc.c b/drivers/iio/adc/nxp-sar-adc.c index 9d9f2c76bed4..705dd7da1bd2 100644 --- a/drivers/iio/adc/nxp-sar-adc.c +++ b/drivers/iio/adc/nxp-sar-adc.c @@ -198,6 +198,15 @@ static void nxp_sar_adc_irq_cfg(struct nxp_sar_adc *info, bool enable) writel(0, NXP_SAR_ADC_IMR(info->regs)); } +static void nxp_sar_adc_wait_for(struct nxp_sar_adc *info, unsigned int cycles) +{ + u64 rate; + + rate = clk_get_rate(info->clk); + if (rate) + ndelay(div64_u64(NSEC_PER_SEC, rate * cycles)); +} + static bool nxp_sar_adc_set_enabled(struct nxp_sar_adc *info, bool enable) { u32 mcr; @@ -221,7 +230,7 @@ static bool nxp_sar_adc_set_enabled(struct nxp_sar_adc *info, bool enable) * configuration of NCMR and the setting of NSTART. */ if (enable) - ndelay(div64_u64(NSEC_PER_SEC, clk_get_rate(info->clk) * 3)); + nxp_sar_adc_wait_for(info, 3); return pwdn; } @@ -469,7 +478,7 @@ static void nxp_sar_adc_stop_conversion(struct nxp_sar_adc *info) * only when the capture finishes. The delay will be very * short, usec-ish, which is acceptable in the atomic context. */ - ndelay(div64_u64(NSEC_PER_SEC, clk_get_rate(info->clk)) * 80); + nxp_sar_adc_wait_for(info, 80); } static int nxp_sar_adc_start_conversion(struct nxp_sar_adc *info, bool raw) From 0d42e2c0bd6ceb89e44c6e065f9bdf9b1df3ef0c Mon Sep 17 00:00:00 2001 From: David Carlier Date: Tue, 14 Apr 2026 13:30:06 +0100 Subject: [PATCH 012/411] iio: adc: npcm: fix unbalanced clk_disable_unprepare() The driver acquired the ADC clock with devm_clk_get() and read its rate, but never called clk_prepare_enable(). The probe error path and npcm_adc_remove() both called clk_disable_unprepare() unconditionally, causing the clk framework's enable/prepare counts to underflow on probe failure or module unbind. The issue went unnoticed because NPCM BMC firmware leaves the ADC clock enabled at boot, so the driver happened to work in practice. Switch to devm_clk_get_enabled() so the clock is properly enabled during probe and automatically released by the device-managed cleanup, and drop the now-redundant clk_disable_unprepare() from both the probe error path and remove(). While at it, drop the duplicate error message on devm_request_irq() failure since the IRQ core already logs it. Fixes: 9bf85fbc9d8f ("iio: adc: add NPCM ADC driver") Signed-off-by: David Carlier Reviewed-by: Andy Shevchenko Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/npcm_adc.c | 25 ++++++++----------------- 1 file changed, 8 insertions(+), 17 deletions(-) diff --git a/drivers/iio/adc/npcm_adc.c b/drivers/iio/adc/npcm_adc.c index ddabb9600d46..61c8b825bda1 100644 --- a/drivers/iio/adc/npcm_adc.c +++ b/drivers/iio/adc/npcm_adc.c @@ -231,7 +231,7 @@ static int npcm_adc_probe(struct platform_device *pdev) if (IS_ERR(info->reset)) return PTR_ERR(info->reset); - info->adc_clk = devm_clk_get(&pdev->dev, NULL); + info->adc_clk = devm_clk_get_enabled(&pdev->dev, NULL); if (IS_ERR(info->adc_clk)) { dev_warn(&pdev->dev, "ADC clock failed: can't read clk\n"); return PTR_ERR(info->adc_clk); @@ -244,17 +244,13 @@ static int npcm_adc_probe(struct platform_device *pdev) info->adc_sample_hz = clk_get_rate(info->adc_clk) / ((div + 1) * 2); irq = platform_get_irq(pdev, 0); - if (irq < 0) { - ret = irq; - goto err_disable_clk; - } + if (irq < 0) + return irq; ret = devm_request_irq(&pdev->dev, irq, npcm_adc_isr, 0, "NPCM_ADC", indio_dev); - if (ret < 0) { - dev_err(dev, "failed requesting interrupt\n"); - goto err_disable_clk; - } + if (ret < 0) + return ret; reg_con = ioread32(info->regs + NPCM_ADCCON); info->vref = devm_regulator_get_optional(&pdev->dev, "vref"); @@ -262,7 +258,7 @@ static int npcm_adc_probe(struct platform_device *pdev) ret = regulator_enable(info->vref); if (ret) { dev_err(&pdev->dev, "Can't enable ADC reference voltage\n"); - goto err_disable_clk; + return ret; } iowrite32(reg_con & ~NPCM_ADCCON_REFSEL, @@ -272,10 +268,8 @@ static int npcm_adc_probe(struct platform_device *pdev) * Any error which is not ENODEV indicates the regulator * has been specified and so is a failure case. */ - if (PTR_ERR(info->vref) != -ENODEV) { - ret = PTR_ERR(info->vref); - goto err_disable_clk; - } + if (PTR_ERR(info->vref) != -ENODEV) + return PTR_ERR(info->vref); /* Use internal reference */ iowrite32(reg_con | NPCM_ADCCON_REFSEL, @@ -314,8 +308,6 @@ static int npcm_adc_probe(struct platform_device *pdev) iowrite32(reg_con & ~NPCM_ADCCON_ADC_EN, info->regs + NPCM_ADCCON); if (!IS_ERR(info->vref)) regulator_disable(info->vref); -err_disable_clk: - clk_disable_unprepare(info->adc_clk); return ret; } @@ -332,7 +324,6 @@ static void npcm_adc_remove(struct platform_device *pdev) iowrite32(regtemp & ~NPCM_ADCCON_ADC_EN, info->regs + NPCM_ADCCON); if (!IS_ERR(info->vref)) regulator_disable(info->vref); - clk_disable_unprepare(info->adc_clk); } static struct platform_driver npcm_adc_driver = { From 5aba4f94b225617a55fed442a70329b2ee19c0a5 Mon Sep 17 00:00:00 2001 From: Antoniu Miclaus Date: Wed, 1 Apr 2026 14:08:29 +0300 Subject: [PATCH 013/411] iio: chemical: scd30: fix division by zero in write_raw Add a zero check for val2 before using it as a divisor when setting the sampling frequency. A user writing a zero fractional part to the sampling_frequency sysfs attribute triggers a division by zero in the kernel. Fixes: 64b3d8b1b0f5 ("iio: chemical: scd30: add core driver") Signed-off-by: Antoniu Miclaus Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/chemical/scd30_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/chemical/scd30_core.c b/drivers/iio/chemical/scd30_core.c index a665fcb78806..11d6bc1b63e6 100644 --- a/drivers/iio/chemical/scd30_core.c +++ b/drivers/iio/chemical/scd30_core.c @@ -256,7 +256,7 @@ static int scd30_write_raw(struct iio_dev *indio_dev, struct iio_chan_spec const guard(mutex)(&state->lock); switch (mask) { case IIO_CHAN_INFO_SAMP_FREQ: - if (val) + if (val || !val2) return -EINVAL; val = 1000000000 / val2; From 76b0d0baa9ae9c60e726bbe1b6ff0bec2c993634 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Sat, 25 Apr 2026 22:07:06 -0700 Subject: [PATCH 014/411] Input: elan_i2c - validate firmware size before use Ensure that the firmware file is large enough to contain the expected number of pages and the signature (which resides at the end of the firmware blob) before accessing them to prevent potential out-of-bounds reads. Cc: stable@vger.kernel.org Link: https://patch.msgid.link/ae2dOgiFvXRm4BHo@google.com Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/elan_i2c_core.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/input/mouse/elan_i2c_core.c b/drivers/input/mouse/elan_i2c_core.c index 7475803c6ce4..5cba02a156ce 100644 --- a/drivers/input/mouse/elan_i2c_core.c +++ b/drivers/input/mouse/elan_i2c_core.c @@ -648,6 +648,11 @@ static ssize_t elan_sysfs_update_fw(struct device *dev, return error; } + if (fw->size < data->fw_signature_address + sizeof(signature)) { + dev_err(dev, "firmware file too small\n"); + return -EBADF; + } + /* Firmware file must match signature data */ fw_signature = &fw->data[data->fw_signature_address]; if (memcmp(fw_signature, signature, sizeof(signature)) != 0) { From c897cf120696b94f56ed0f3197ba9a77071a59ec Mon Sep 17 00:00:00 2001 From: Dmitriy Zharov Date: Thu, 30 Apr 2026 22:35:22 +0400 Subject: [PATCH 015/411] Input: xpad - add support for ASUS ROG RAIKIRI II Add the VID/PIDs for the ASUS ROG RAIKIRI II controller to xpad_device and the VID to xpad_table. The controller has a physical PC/XBOX toggle which switches between XBOX360 and XBOXONE protocols. Signed-off-by: Dmitriy Zharov Link: https://patch.msgid.link/20260430183522.122151-1-contact@zharov.dev Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 19ce90da89e9..97daec39bfee 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -186,6 +186,10 @@ static const struct xpad_device { { 0x07ff, 0xffff, "Mad Catz GamePad", 0, XTYPE_XBOX360 }, { 0x0b05, 0x1a38, "ASUS ROG RAIKIRI", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x0b05, 0x1abb, "ASUS ROG RAIKIRI PRO", 0, XTYPE_XBOXONE }, + { 0x0b05, 0x1c91, "ASUS ROG RAIKIRI II", 0, XTYPE_XBOX360 }, + { 0x0b05, 0x1c92, "ASUS ROG RAIKIRI II WIRELESS", 0, XTYPE_XBOX360 }, + { 0x0b05, 0x1c96, "ASUS ROG RAIKIRI II XBOX", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, + { 0x0b05, 0x1d04, "ASUS ROG RAIKIRI II XBOX WIRELESS", MAP_SHARE_BUTTON, XTYPE_XBOXONE }, { 0x0c12, 0x0005, "Intec wireless", 0, XTYPE_XBOX }, { 0x0c12, 0x8801, "Nyko Xbox Controller", 0, XTYPE_XBOX }, { 0x0c12, 0x8802, "Zeroplus Xbox Controller", 0, XTYPE_XBOX }, @@ -507,6 +511,7 @@ static const struct usb_device_id xpad_table[] = { { USB_DEVICE(0x0738, 0x4540) }, /* Mad Catz Beat Pad */ XPAD_XBOXONE_VENDOR(0x0738), /* Mad Catz FightStick TE 2 */ XPAD_XBOX360_VENDOR(0x07ff), /* Mad Catz Gamepad */ + XPAD_XBOX360_VENDOR(0x0b05), /* ASUS controllers */ XPAD_XBOXONE_VENDOR(0x0b05), /* ASUS controllers */ XPAD_XBOX360_VENDOR(0x0c12), /* Zeroplus X-Box 360 controllers */ XPAD_XBOX360_VENDOR(0x0db0), /* Micro Star International X-Box 360 controllers */ From 1f6ac0f8441c48c4cc250141e1da8486c13512ba Mon Sep 17 00:00:00 2001 From: Qbeliw Tanaka Date: Thu, 30 Apr 2026 21:44:12 -0700 Subject: [PATCH 016/411] Input: xpad - add "Nova 2 Lite" from GameSir Add support for the gamepad "Nova 2 Lite" from GameSir, compatible with the Xbox 360 gamepad. Signed-off-by: Qbeliw Tanaka Link: https://patch.msgid.link/20260429.162040.930225048583399359.q.tanaka@gmx.com Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/joystick/xpad.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c index 97daec39bfee..feb8f368f834 100644 --- a/drivers/input/joystick/xpad.c +++ b/drivers/input/joystick/xpad.c @@ -395,6 +395,7 @@ static const struct xpad_device { { 0x3285, 0x0662, "Nacon Revolution5 Pro", 0, XTYPE_XBOX360 }, { 0x3285, 0x0663, "Nacon Evol-X", 0, XTYPE_XBOXONE }, { 0x3537, 0x1004, "GameSir T4 Kaleid", 0, XTYPE_XBOX360 }, + { 0x3537, 0x100f, "GameSir Nova 2 Lite", 0, XTYPE_XBOX360 }, { 0x3537, 0x1010, "GameSir G7 SE", 0, XTYPE_XBOXONE }, { 0x3651, 0x1000, "CRKD SG", 0, XTYPE_XBOX360 }, { 0x366c, 0x0005, "ByoWave Proteus Controller", MAP_SHARE_BUTTON, XTYPE_XBOXONE, FLAG_DELAY_INIT }, From d9eeb0ea0d2de658663bfaa9c26eccdd8fd64440 Mon Sep 17 00:00:00 2001 From: Guangshuo Li Date: Mon, 13 Apr 2026 21:46:04 +0800 Subject: [PATCH 017/411] counter: Fix refcount leak in counter_alloc() error path After device_initialize(), the lifetime of the embedded struct device is expected to be managed through the device core reference counting. In counter_alloc(), if dev_set_name() fails after device_initialize(), the error path removes the chrdev, frees the ID, and frees the backing allocation directly instead of releasing the device reference with put_device(). This bypasses the normal device lifetime rules and may leave the reference count of the embedded struct device unbalanced, resulting in a refcount leak. The issue was identified by a static analysis tool I developed and confirmed by manual review. Fix this by using put_device() in the dev_set_name() failure path and let counter_device_release() handle the final cleanup. Fixes: 4da08477ea1f ("counter: Set counter device name") Cc: stable@vger.kernel.org Signed-off-by: Guangshuo Li Link: https://lore.kernel.org/r/20260413134604.2861772-1-lgs201920130244@gmail.com Signed-off-by: William Breathitt Gray --- drivers/counter/counter-core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/counter/counter-core.c b/drivers/counter/counter-core.c index 50bd30ba3d03..0b1dac61b7b5 100644 --- a/drivers/counter/counter-core.c +++ b/drivers/counter/counter-core.c @@ -124,7 +124,8 @@ struct counter_device *counter_alloc(size_t sizeof_priv) err_dev_set_name: - counter_chrdev_remove(counter); + put_device(dev); + return NULL; err_chrdev_add: ida_free(&counter_ida, dev->id); From fb7415f2ab0e3c818254cbf5fb0afda71bef4333 Mon Sep 17 00:00:00 2001 From: Bruce Johnston Date: Tue, 28 Apr 2026 14:39:31 -0400 Subject: [PATCH 018/411] dm vdo: use GFP_NOIO for blkdev_issue_zeroout on format path GFP_NOWAIT is inappropriate when blkdev_issue_zeroout may sleep and bio_alloc can fail under pressure; use GFP_NOIO for clear_partition and vdo_clear_layout zeroout calls. Signed-off-by: Bruce Johnston Signed-off-by: Matthew Sakai Signed-off-by: Mikulas Patocka Fixes: fc1d43826702 ("dm vdo: save the formatted metadata to disk") --- drivers/md/dm-vdo/vdo.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/md/dm-vdo/vdo.c b/drivers/md/dm-vdo/vdo.c index 7bec2418c121..d0d4e0262be2 100644 --- a/drivers/md/dm-vdo/vdo.c +++ b/drivers/md/dm-vdo/vdo.c @@ -965,7 +965,7 @@ static int __must_check clear_partition(struct vdo *vdo, enum partition_id id) return blkdev_issue_zeroout(vdo_get_backing_device(vdo), partition->offset * VDO_SECTORS_PER_BLOCK, partition->count * VDO_SECTORS_PER_BLOCK, - GFP_NOWAIT, 0); + GFP_NOIO, 0); } int vdo_clear_layout(struct vdo *vdo) @@ -976,7 +976,7 @@ int vdo_clear_layout(struct vdo *vdo) result = blkdev_issue_zeroout(vdo_get_backing_device(vdo), VDO_SECTORS_PER_BLOCK, VDO_SECTORS_PER_BLOCK, - GFP_NOWAIT, 0); + GFP_NOIO, 0); if (result != VDO_SUCCESS) return result; From f4268b466190dae95a7585f69b4f1f8ad097632c Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Wed, 29 Apr 2026 13:40:41 +0000 Subject: [PATCH 019/411] nfc: llcp: Fix use-after-free in llcp_sock_release() llcp_sock_release() unconditionally unlinks the socket from the local sockets list. However, if the socket is still in connecting state, it is on the connecting list. Fix this by checking the socket state and unlinking from the correct list. Fixes: b4011239a08e ("NFC: llcp: Fix non blocking sockets connections") Signed-off-by: Lee Jones Link: https://patch.msgid.link/20260429134115.3558604-1-lee@kernel.org Signed-off-by: David Heidelberg --- net/nfc/llcp_sock.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/nfc/llcp_sock.c b/net/nfc/llcp_sock.c index f1be1e84f665..feab29fc62f4 100644 --- a/net/nfc/llcp_sock.c +++ b/net/nfc/llcp_sock.c @@ -633,6 +633,8 @@ static int llcp_sock_release(struct socket *sock) if (sock->type == SOCK_RAW) nfc_llcp_sock_unlink(&local->raw_sockets, sk); + else if (sk->sk_state == LLCP_CONNECTING) + nfc_llcp_sock_unlink(&local->connecting_sockets, sk); else nfc_llcp_sock_unlink(&local->sockets, sk); From b493ea2765cc17cb8aa7e7544a4b6dcb05b6ed77 Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Wed, 29 Apr 2026 13:40:42 +0000 Subject: [PATCH 020/411] nfc: llcp: Fix use-after-free race in nfc_llcp_recv_cc() A race condition exists in the NFC LLCP connection state machine where the connection acceptance packet (CC) can be processed concurrently with socket release. This can lead to a use-after-free of the socket object. When nfc_llcp_recv_cc() moves the socket from the connecting_sockets list to the sockets list, it does so without holding the socket lock. If llcp_sock_release() is executing concurrently, it might have already unlinked the socket and dropped its references, which can result in nfc_llcp_recv_cc() linking a freed socket into the live list. Fix this by holding lock_sock() during the state transition and list movement in nfc_llcp_recv_cc(). After acquiring the lock, check if the socket is still hashed to ensure it hasn't already been unlinked and marked for destruction by the release path. This aligns the locking pattern with recv_hdlc() and recv_disc(). Fixes: a69f32af86e3 ("NFC: Socket linked list") Signed-off-by: Lee Jones Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260429134115.3558604-2-lee@kernel.org Signed-off-by: David Heidelberg --- net/nfc/llcp_core.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/net/nfc/llcp_core.c b/net/nfc/llcp_core.c index db5bc6a878dd..dc65c719f35f 100644 --- a/net/nfc/llcp_core.c +++ b/net/nfc/llcp_core.c @@ -1218,6 +1218,15 @@ static void nfc_llcp_recv_cc(struct nfc_llcp_local *local, sk = &llcp_sock->sk; + lock_sock(sk); + + /* Check if socket was destroyed whilst waiting for the lock */ + if (!sk_hashed(sk)) { + release_sock(sk); + nfc_llcp_sock_put(llcp_sock); + return; + } + /* Unlink from connecting and link to the client array */ nfc_llcp_sock_unlink(&local->connecting_sockets, sk); nfc_llcp_sock_link(&local->sockets, sk); @@ -1229,6 +1238,8 @@ static void nfc_llcp_recv_cc(struct nfc_llcp_local *local, sk->sk_state = LLCP_CONNECTED; sk->sk_state_change(sk); + release_sock(sk); + nfc_llcp_sock_put(llcp_sock); } From 7e2a4f7ca0952820731ef7bdadfc9a9e9d3571b4 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Mon, 4 May 2026 22:27:36 +0800 Subject: [PATCH 021/411] xfrm: route MIGRATE notifications to caller's netns xfrm_send_migrate() in net/xfrm/xfrm_user.c and pfkey_send_migrate() in net/key/af_key.c both hardcode &init_net for the multicast that announces a successful XFRM_MSG_MIGRATE / SADB_X_MIGRATE. XFRM_MSG_MIGRATE arrives on a per-netns NETLINK_XFRM socket, and the rest of the xfrm/af_key netlink path was made netns-aware in 2008. The other 14 multicast paths in xfrm_user.c route their event using xs_net(x), xp_net(xp) or sock_net(skb->sk); only the migrate path was missed. Two consequences of the init_net hardcoding: 1. The notification (selector, old/new endpoint addresses, and the km_address) is delivered to listeners on init_net's XFRMNLGRP_MIGRATE / pfkey BROADCAST_ALL groups rather than on the issuing netns. An IKE daemon running in init_net therefore receives migration notifications originating from any other netns on the host. 2. An IKE daemon running inside a non-init netns and subscribed to its own XFRMNLGRP_MIGRATE / pfkey groups never receives the notification of its own migration. IKEv2 MOBIKE / address-update handling inside a netns is silently broken. Thread struct net through km_migrate() and the xfrm_mgr.migrate function pointer, drop the &init_net override in xfrm_send_migrate() and pfkey_send_migrate(), and pass the caller's net (already in scope in xfrm_migrate() via sock_net(skb->sk)) all the way down. struct xfrm_mgr is in-tree only and not exported as a stable API, so the function-pointer signature change is internal. pfkey_broadcast() is already netns-aware via net_generic(net, pfkey_net_id) since the pernet conversion. The five other pfkey_broadcast() callers in af_key.c already pass xs_net(x), sock_net(sk) or a per-netns net, so this only removes the &init_net outlier. Fixes: 5c79de6e79cd ("[XFRM]: User interface for handling XFRM_MSG_MIGRATE") Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Maoyi Xie Signed-off-by: Steffen Klassert --- include/net/xfrm.h | 3 ++- net/key/af_key.c | 6 +++--- net/xfrm/xfrm_policy.c | 2 +- net/xfrm/xfrm_state.c | 4 ++-- net/xfrm/xfrm_user.c | 5 ++--- 5 files changed, 10 insertions(+), 10 deletions(-) diff --git a/include/net/xfrm.h b/include/net/xfrm.h index 10d3edde6b2f..874409127e29 100644 --- a/include/net/xfrm.h +++ b/include/net/xfrm.h @@ -715,6 +715,7 @@ struct xfrm_mgr { const struct xfrm_migrate *m, int num_bundles, const struct xfrm_kmaddress *k, + struct net *net, const struct xfrm_encap_tmpl *encap); bool (*is_alive)(const struct km_event *c); }; @@ -1891,7 +1892,7 @@ int xfrm_sk_policy_insert(struct sock *sk, int dir, struct xfrm_policy *pol); #ifdef CONFIG_XFRM_MIGRATE int km_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_bundles, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap); struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *net, u32 if_id); diff --git a/net/key/af_key.c b/net/key/af_key.c index a166a88d8788..9cffeef18cd9 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -3564,7 +3564,7 @@ static int set_ipsecrequest(struct sk_buff *skb, #ifdef CONFIG_NET_KEY_MIGRATE static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_bundles, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { int i; @@ -3669,7 +3669,7 @@ static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, } /* broadcast migrate message to sockets */ - pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, &init_net); + pfkey_broadcast(skb, GFP_ATOMIC, BROADCAST_ALL, NULL, net); return 0; @@ -3680,7 +3680,7 @@ static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, #else static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_bundles, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { return -ENOPROTOOPT; diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index c944327ce66c..59968dcbafe1 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4703,7 +4703,7 @@ int xfrm_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, } /* Stage 5 - announce */ - km_migrate(sel, dir, type, m, num_migrate, k, encap); + km_migrate(sel, dir, type, m, num_migrate, k, net, encap); xfrm_pol_put(pol); diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 686014d39429..395d82411a87 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -2837,7 +2837,7 @@ EXPORT_SYMBOL(km_policy_expired); #ifdef CONFIG_XFRM_MIGRATE int km_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_migrate, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { int err = -EINVAL; @@ -2848,7 +2848,7 @@ int km_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, list_for_each_entry_rcu(km, &xfrm_km_list, list) { if (km->migrate) { ret = km->migrate(sel, dir, type, m, num_migrate, k, - encap); + net, encap); if (!ret) err = ret; } diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 38a90e5ee3d9..71a4b7278eba 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -3271,10 +3271,9 @@ static int build_migrate(struct sk_buff *skb, const struct xfrm_migrate *m, static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_migrate, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { - struct net *net = &init_net; struct sk_buff *skb; int err; @@ -3292,7 +3291,7 @@ static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, #else static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, const struct xfrm_migrate *m, int num_migrate, - const struct xfrm_kmaddress *k, + const struct xfrm_kmaddress *k, struct net *net, const struct xfrm_encap_tmpl *encap) { return -ENOPROTOOPT; From f040e590c035bfd9553fe79ee9585caf1b14d67b Mon Sep 17 00:00:00 2001 From: Ashutosh Desai Date: Tue, 5 May 2026 17:07:12 +0000 Subject: [PATCH 022/411] nfc: hci: fix out-of-bounds read in HCP header parsing Both nfc_hci_recv_from_llc() and nci_hci_data_received_cb() read packet->header from skb->data at function entry without first checking that the buffer holds at least one byte. A malicious NFC peer can send a 0-byte HCP frame that passes through the SHDLC layer and reaches these functions, causing an out-of-bounds heap read of packet->header. The same 0-byte frame, if queued as a non-final fragment, also causes the reassembly loop to underflow msg_len to UINT_MAX, triggering skb_over_panic() when the reassembled skb is written. Fix this by adding a pskb_may_pull() check at the entry of each function before packet->header is first accessed. The existing pskb_may_pull() checks before the reassembled hcp_skb is cast to struct hcp_packet remain in place to guard the 2-byte HCP message header. Fixes: 8b8d2e08bf0d ("NFC: HCI support") Fixes: 11f54f228643 ("NFC: nci: Add HCI over NCI protocol support") Cc: stable@vger.kernel.org Reviewed-by: Simon Horman Signed-off-by: Ashutosh Desai Link: https://patch.msgid.link/20260505170712.96560-1-ashutoshdesai993@gmail.com Signed-off-by: David Heidelberg --- net/nfc/hci/core.c | 10 ++++++++++ net/nfc/nci/hci.c | 10 ++++++++++ 2 files changed, 20 insertions(+) diff --git a/net/nfc/hci/core.c b/net/nfc/hci/core.c index 0d33c81a15fe..ba6f0310ffd7 100644 --- a/net/nfc/hci/core.c +++ b/net/nfc/hci/core.c @@ -861,6 +861,11 @@ static void nfc_hci_recv_from_llc(struct nfc_hci_dev *hdev, struct sk_buff *skb) struct sk_buff *frag_skb; int msg_len; + if (!pskb_may_pull(skb, NFC_HCI_HCP_PACKET_HEADER_LEN)) { + kfree_skb(skb); + return; + } + packet = (struct hcp_packet *)skb->data; if ((packet->header & ~NFC_HCI_FRAGMENT) == 0) { skb_queue_tail(&hdev->rx_hcp_frags, skb); @@ -904,6 +909,11 @@ static void nfc_hci_recv_from_llc(struct nfc_hci_dev *hdev, struct sk_buff *skb) * unblock waiting cmd context. Otherwise, enqueue to dispatch * in separate context where handler can also execute command. */ + if (!pskb_may_pull(hcp_skb, NFC_HCI_HCP_HEADER_LEN)) { + kfree_skb(hcp_skb); + return; + } + packet = (struct hcp_packet *)hcp_skb->data; type = HCP_MSG_GET_TYPE(packet->message.header); if (type == NFC_HCI_HCP_RESPONSE) { diff --git a/net/nfc/nci/hci.c b/net/nfc/nci/hci.c index 40ae8e5a7ec7..c03e8a0bd3bd 100644 --- a/net/nfc/nci/hci.c +++ b/net/nfc/nci/hci.c @@ -439,6 +439,11 @@ void nci_hci_data_received_cb(void *context, return; } + if (!pskb_may_pull(skb, NCI_HCI_HCP_PACKET_HEADER_LEN)) { + kfree_skb(skb); + return; + } + packet = (struct nci_hcp_packet *)skb->data; if ((packet->header & ~NCI_HCI_FRAGMENT) == 0) { skb_queue_tail(&ndev->hci_dev->rx_hcp_frags, skb); @@ -482,6 +487,11 @@ void nci_hci_data_received_cb(void *context, * unblock waiting cmd context. Otherwise, enqueue to dispatch * in separate context where handler can also execute command. */ + if (!pskb_may_pull(hcp_skb, NCI_HCI_HCP_HEADER_LEN)) { + kfree_skb(hcp_skb); + return; + } + packet = (struct nci_hcp_packet *)hcp_skb->data; type = NCI_HCP_MSG_GET_TYPE(packet->message.header); if (type == NCI_HCI_HCP_RESPONSE) { From 74d695fd6f9d70df849c555f358ddfd26e2d85bf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Uwe=20Kleine-K=C3=B6nig=20=28The=20Capable=20Hub=29?= Date: Thu, 7 May 2026 18:00:51 +0200 Subject: [PATCH 023/411] Input: fm801-gp - simplify initialisation of pci_device_id array MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Instead of assigning the pci_device_id members using a list (which is hard to read as you need to look at the order of the members in that struct in parallel) use the PCI_VDEVICE() convenience macro to compact the initialisation while improving readability. Also drop trailing zeros that the compiler will care about then. The change doesn't introduce binary changes to the compiled driver, verified on both ARCH=x86 and ARCH=arm64. Signed-off-by: Uwe Kleine-König (The Capable Hub) Link: https://patch.msgid.link/20260507160051.3315630-2-u.kleine-koenig@baylibre.com Signed-off-by: Dmitry Torokhov --- drivers/input/gameport/fm801-gp.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/input/gameport/fm801-gp.c b/drivers/input/gameport/fm801-gp.c index 423cccdea34f..1e8c6c044844 100644 --- a/drivers/input/gameport/fm801-gp.c +++ b/drivers/input/gameport/fm801-gp.c @@ -125,8 +125,8 @@ static void fm801_gp_remove(struct pci_dev *pci) } static const struct pci_device_id fm801_gp_id_table[] = { - { PCI_VENDOR_ID_FORTEMEDIA, PCI_DEVICE_ID_FM801_GP, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, - { 0 } + { PCI_VDEVICE(FORTEMEDIA, PCI_DEVICE_ID_FM801_GP) }, + { } }; MODULE_DEVICE_TABLE(pci, fm801_gp_id_table); From baa0210fb6a9dc3882509a9411b6d284d88fe30e Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Mon, 4 May 2026 11:54:45 -0700 Subject: [PATCH 024/411] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem When a configuration file provides an object size that is larger than the driver's known mxt_obj_size(object), the driver intends to discard the extra bytes. The loop iterates using for (i = 0; i < size; i++). Inside the loop, the condition to skip processing extra bytes is: if (i > mxt_obj_size(object)) continue; Since i is a 0-based index, the valid indices for the object are 0 through mxt_obj_size(object) - 1. When i == mxt_obj_size(object), the condition evaluates to false, and the code processes the byte instead of discarding it. This causes the code to calculate byte_offset = reg + i - cfg->start_ofs and writes the byte there, overwriting exactly one byte of the adjacent instance or object. Update the boundary check to skip extra bytes correctly by using >=. Fixes: 50a77c658b80 ("Input: atmel_mxt_ts - download device config using firmware loader") Cc: stable@vger.kernel.org Assisted-by: Gemini:gemini-3.1-pro Reviewed-by: Ricardo Ribalda Link: https://patch.msgid.link/20260504185448.4055973-1-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/atmel_mxt_ts.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c index 87c6a10381f2..fad1b3f4138b 100644 --- a/drivers/input/touchscreen/atmel_mxt_ts.c +++ b/drivers/input/touchscreen/atmel_mxt_ts.c @@ -1473,7 +1473,7 @@ static int mxt_prepare_cfg_mem(struct mxt_data *data, struct mxt_cfg *cfg) } cfg->raw_pos += offset; - if (i > mxt_obj_size(object)) + if (i >= mxt_obj_size(object)) continue; byte_offset = reg + i - cfg->start_ofs; From a5fd88a5d63f812422e69682f3cb663d9d7f3e9c Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Mon, 4 May 2026 11:54:46 -0700 Subject: [PATCH 025/411] Input: atmel_mxt_ts - check mem_size before calculating config memory size In mxt_update_cfg(), the driver calculates the memory size needed to store the configuration as data->mem_size - cfg.start_ofs. If data->mem_size is less than or equal to cfg.start_ofs, this calculation will underflow or result in a zero-size buffer, neither of which is valid for a configuration update. Add a check to return -EINVAL if data->mem_size is too small. While at it, change the types of start_ofs and mem_size in struct mxt_cfg to u16 to match the device address space. Assisted-by: Gemini:gemini-3.1-pro Link: https://patch.msgid.link/20260504185448.4055973-2-dmitry.torokhov@gmail.com Signed-off-by: Dmitry Torokhov --- drivers/input/touchscreen/atmel_mxt_ts.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c index fad1b3f4138b..f21bf2844112 100644 --- a/drivers/input/touchscreen/atmel_mxt_ts.c +++ b/drivers/input/touchscreen/atmel_mxt_ts.c @@ -275,8 +275,8 @@ struct mxt_cfg { off_t raw_pos; u8 *mem; - size_t mem_size; - int start_ofs; + u16 mem_size; + u16 start_ofs; struct mxt_info info; }; @@ -1627,6 +1627,13 @@ static int mxt_update_cfg(struct mxt_data *data, const struct firmware *fw) cfg.start_ofs = MXT_OBJECT_START + data->info->object_num * sizeof(struct mxt_object) + MXT_INFO_CHECKSUM_SIZE; + + if (data->mem_size <= cfg.start_ofs) { + dev_err(dev, "Memory size too small: %u < %u\n", + data->mem_size, cfg.start_ofs); + return -EINVAL; + } + cfg.mem_size = data->mem_size - cfg.start_ofs; u8 *mem_buf __free(kfree) = cfg.mem = kzalloc(cfg.mem_size, GFP_KERNEL); From 7dbac7680eb629b3b4dc7e98c34f943b8814c0c8 Mon Sep 17 00:00:00 2001 From: Herbert Xu Date: Wed, 6 May 2026 21:23:28 +0800 Subject: [PATCH 026/411] xfrm: ipcomp: Free destination pages on acomp errors Move the out_free_req label up by a couple of lines so that the allocated dst SG list gets freed on error as well as success. Fixes: eb2953d26971 ("xfrm: ipcomp: Use crypto_acomp interface") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Reported-by: Yilin Zhu Signed-off-by: Herbert Xu Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_ipcomp.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/net/xfrm/xfrm_ipcomp.c b/net/xfrm/xfrm_ipcomp.c index 5f38dff16177..671d48f8c937 100644 --- a/net/xfrm/xfrm_ipcomp.c +++ b/net/xfrm/xfrm_ipcomp.c @@ -51,11 +51,15 @@ static int ipcomp_post_acomp(struct sk_buff *skb, int err, int hlen) struct scatterlist *dsg; int len, dlen; - if (unlikely(err)) - goto out_free_req; + if (unlikely(!req)) + return err; extra = acomp_request_extra(req); dsg = extra->sg; + + if (unlikely(err)) + goto out_free_req; + dlen = req->dlen; pskb_trim_unique(skb, 0); @@ -84,10 +88,10 @@ static int ipcomp_post_acomp(struct sk_buff *skb, int err, int hlen) skb_shinfo(skb)->nr_frags++; } while ((dlen -= len)); - for (; dsg; dsg = sg_next(dsg)) +out_free_req: + for (; dsg && sg_page(dsg); dsg = sg_next(dsg)) __free_page(sg_page(dsg)); -out_free_req: acomp_request_free(req); return err; } From 01deda0152066c6c955f0619114ea6afa070aaec Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 10 May 2026 19:16:56 -0400 Subject: [PATCH 027/411] thunderbolt: property: Reject u32 wrap in tb_property_entry_valid() entry->value is u32 and entry->length is u16; the sum is performed in u32 and wraps. A malicious XDomain peer can pick value = 0xffffff00, length = 0x100 so the sum 0x100000000 wraps to 0 and passes the > block_len check. tb_property_parse() then passes entry->value to parse_dwdata() as a dword offset into the property block, reading attacker-directed memory far past the allocation. For TEXT-typed entries with the "deviceid" or "vendorid" keys this lands in xd->device_name / xd->vendor_name and is readable back via the per-XDomain device_name / vendor_name sysfs attributes; the leak is NUL-bounded (kstrdup() stops at the first zero byte) and untargeted (the attacker picks a delta, not an absolute address). DATA-typed entries are parsed into property->value.data but not generically surfaced to userspace. Use check_add_overflow() so a wrapped sum is rejected. Fixes: cdae7c07e3e3 ("thunderbolt: Add support for XDomain properties") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-6 Assisted-by: Codex:gpt-5-4 Signed-off-by: Michael Bommarito Signed-off-by: Mika Westerberg --- drivers/thunderbolt/property.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/thunderbolt/property.c b/drivers/thunderbolt/property.c index 50cbfc92fe65..29cd60c11ac4 100644 --- a/drivers/thunderbolt/property.c +++ b/drivers/thunderbolt/property.c @@ -8,6 +8,7 @@ */ #include +#include #include #include #include @@ -52,13 +53,16 @@ static inline void format_dwdata(void *dst, const void *src, size_t dwords) static bool tb_property_entry_valid(const struct tb_property_entry *entry, size_t block_len) { + u32 end; + switch (entry->type) { case TB_PROPERTY_TYPE_DIRECTORY: case TB_PROPERTY_TYPE_DATA: case TB_PROPERTY_TYPE_TEXT: if (entry->length > block_len) return false; - if (entry->value + entry->length > block_len) + if (check_add_overflow(entry->value, entry->length, &end) || + end > block_len) return false; break; From de21b59c29e31c5108ddc04210631bbfab81b997 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 10 May 2026 19:16:57 -0400 Subject: [PATCH 028/411] thunderbolt: property: Reject dir_len < 4 to prevent size_t underflow On the non-root path, __tb_property_parse_dir() takes dir_len from entry->length (u16 widened to size_t). Two distinct OOB conditions follow when entry->length < 4: 1. The non-root path begins with kmemdup(&block[dir_offset], sizeof(*dir->uuid), ...) which always reads 4 dwords from dir_offset. tb_property_entry_valid() only enforces dir_offset + entry->length <= block_len, so a crafted entry with dir_offset close to the end of the property block and entry->length in 0..3 passes that gate but lets the UUID copy run off the block (e.g. dir_offset = 497, dir_len = 3 in a 500-dword block reads block[497..501]). 2. After the kmemdup, content_len = dir_len - 4 underflows size_t to ~SIZE_MAX, nentries becomes SIZE_MAX / 4, and the entry walk runs OOB on each iteration until an entry fails validation or the kernel oopses on an unmapped page. Reject dir_len < 4 on the non-root path *before* the UUID kmemdup, which closes both holes. Also move INIT_LIST_HEAD(&dir->properties) up to immediately after the dir allocation so the new error-return path (and the existing uuid-alloc failure path) calling tb_property_free_dir() sees a walkable list rather than the zero-initialized NULL next/prev that list_for_each_entry_safe() would oops on. Fixes: cdae7c07e3e3 ("thunderbolt: Add support for XDomain properties") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-6 Assisted-by: Codex:gpt-5-4 Signed-off-by: Michael Bommarito Signed-off-by: Mika Westerberg --- drivers/thunderbolt/property.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/thunderbolt/property.c b/drivers/thunderbolt/property.c index 29cd60c11ac4..74c92f9801ff 100644 --- a/drivers/thunderbolt/property.c +++ b/drivers/thunderbolt/property.c @@ -174,10 +174,16 @@ static struct tb_property_dir *__tb_property_parse_dir(const u32 *block, if (!dir) return NULL; + INIT_LIST_HEAD(&dir->properties); + if (is_root) { content_offset = dir_offset + 2; content_len = dir_len; } else { + if (dir_len < 4) { + tb_property_free_dir(dir); + return NULL; + } dir->uuid = kmemdup(&block[dir_offset], sizeof(*dir->uuid), GFP_KERNEL); if (!dir->uuid) { @@ -191,8 +197,6 @@ static struct tb_property_dir *__tb_property_parse_dir(const u32 *block, entries = (const struct tb_property_entry *)&block[content_offset]; nentries = content_len / (sizeof(*entries) / 4); - INIT_LIST_HEAD(&dir->properties); - for (i = 0; i < nentries; i++) { struct tb_property *property; From 928abe19fbf0127003abcb1ea69cabc1c897d0ab Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 10 May 2026 19:16:58 -0400 Subject: [PATCH 029/411] thunderbolt: property: Cap recursion depth in __tb_property_parse_dir() A DIRECTORY entry's value field is used as the dir_offset for a recursive call into __tb_property_parse_dir() with no depth counter. A crafted peer that chains DIRECTORY entries into a back-reference loop drives the parser until the kernel stack is exhausted and the guard page fires. Any untrusted XDomain peer (cable, dock, in-line inspector, adjacent host) that reaches the PROPERTIES_REQUEST control-plane exchange can trigger this without authentication. Thread a depth counter through tb_property_parse() and __tb_property_parse_dir(), and reject blocks that exceed TB_PROPERTY_MAX_DEPTH = 8. That is comfortably larger than any observed legitimate XDomain layout. Operators who do not need XDomain host-to-host discovery can disable the path entirely with thunderbolt.xdomain=0 on the kernel command line. Fixes: cdae7c07e3e3 ("thunderbolt: Add support for XDomain properties") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-6 Assisted-by: Codex:gpt-5-4 Signed-off-by: Michael Bommarito Signed-off-by: Mika Westerberg --- drivers/thunderbolt/property.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/drivers/thunderbolt/property.c b/drivers/thunderbolt/property.c index 74c92f9801ff..da2c59a17db5 100644 --- a/drivers/thunderbolt/property.c +++ b/drivers/thunderbolt/property.c @@ -35,10 +35,11 @@ struct tb_property_dir_entry { }; #define TB_PROPERTY_ROOTDIR_MAGIC 0x55584401 +#define TB_PROPERTY_MAX_DEPTH 8 static struct tb_property_dir *__tb_property_parse_dir(const u32 *block, size_t block_len, unsigned int dir_offset, size_t dir_len, - bool is_root); + bool is_root, unsigned int depth); static inline void parse_dwdata(void *dst, const void *src, size_t dwords) { @@ -97,7 +98,8 @@ tb_property_alloc(const char *key, enum tb_property_type type) } static struct tb_property *tb_property_parse(const u32 *block, size_t block_len, - const struct tb_property_entry *entry) + const struct tb_property_entry *entry, + unsigned int depth) { char key[TB_PROPERTY_KEY_SIZE + 1]; struct tb_property *property; @@ -118,7 +120,7 @@ static struct tb_property *tb_property_parse(const u32 *block, size_t block_len, switch (property->type) { case TB_PROPERTY_TYPE_DIRECTORY: dir = __tb_property_parse_dir(block, block_len, entry->value, - entry->length, false); + entry->length, false, depth + 1); if (!dir) { kfree(property); return NULL; @@ -163,13 +165,17 @@ static struct tb_property *tb_property_parse(const u32 *block, size_t block_len, } static struct tb_property_dir *__tb_property_parse_dir(const u32 *block, - size_t block_len, unsigned int dir_offset, size_t dir_len, bool is_root) + size_t block_len, unsigned int dir_offset, size_t dir_len, bool is_root, + unsigned int depth) { const struct tb_property_entry *entries; size_t i, content_len, nentries; unsigned int content_offset; struct tb_property_dir *dir; + if (depth > TB_PROPERTY_MAX_DEPTH) + return NULL; + dir = kzalloc_obj(*dir); if (!dir) return NULL; @@ -200,7 +206,7 @@ static struct tb_property_dir *__tb_property_parse_dir(const u32 *block, for (i = 0; i < nentries; i++) { struct tb_property *property; - property = tb_property_parse(block, block_len, &entries[i]); + property = tb_property_parse(block, block_len, &entries[i], depth); if (!property) { tb_property_free_dir(dir); return NULL; @@ -239,7 +245,7 @@ struct tb_property_dir *tb_property_parse_dir(const u32 *block, return NULL; return __tb_property_parse_dir(block, block_len, 0, rootdir->length, - true); + true, 0); } /** From d15cd40cb1858f75846eaafa9a6bca841b790a92 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Fri, 10 Apr 2026 18:19:31 +0100 Subject: [PATCH 030/411] serial: zs: Fix swapped RI/DSR modem line transition counting Fix a thinko in the status interrupt handler that has caused counters for the RI and DSR modem line transitions to be used for the other line each. Fixes: 8b4a40809e53 ("zs: move to the serial subsystem") Cc: stable Signed-off-by: Maciej W. Rozycki Link: https://patch.msgid.link/alpine.DEB.2.21.2604101747110.29980@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/zs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/tty/serial/zs.c b/drivers/tty/serial/zs.c index 72a3c0d90f40..31e12645f66b 100644 --- a/drivers/tty/serial/zs.c +++ b/drivers/tty/serial/zs.c @@ -680,9 +680,9 @@ static void zs_status_handle(struct zs_port *zport, struct zs_port *zport_a) uart_handle_dcd_change(uport, zport->mctrl & TIOCM_CAR); if (delta & TIOCM_RNG) - uport->icount.dsr++; - if (delta & TIOCM_DSR) uport->icount.rng++; + if (delta & TIOCM_DSR) + uport->icount.dsr++; if (delta) wake_up_interruptible(&uport->state->port.delta_msr_wait); From 6fe472c1bbbe238e91141f7cabc1226e96a60d43 Mon Sep 17 00:00:00 2001 From: Zhaoyang Yu <2426767509@qq.com> Date: Thu, 9 Apr 2026 13:41:58 +0800 Subject: [PATCH 031/411] tty: serial: pch_uart: add check for dma_alloc_coherent() Add a check for dma_alloc_coherent() failure to prevent a potential NULL pointer dereference in dma_handle_rx(). Properly release DMA channels and the PCI device reference using a goto ladder if the allocation fails. Fixes: 3c6a483275f4 ("Serial: EG20T: add PCH_UART driver") Cc: stable Signed-off-by: Zhaoyang Yu <2426767509@qq.com> Reviewed-by: Andy Shevchenko Link: https://patch.msgid.link/tencent_E328416B7CFD436F6029F2DF02AD7ED89C08@qq.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/pch_uart.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/drivers/tty/serial/pch_uart.c b/drivers/tty/serial/pch_uart.c index 6729d8e83c3c..ba1fcd663fe2 100644 --- a/drivers/tty/serial/pch_uart.c +++ b/drivers/tty/serial/pch_uart.c @@ -689,8 +689,7 @@ static void pch_request_dma(struct uart_port *port) if (!chan) { dev_err(priv->port.dev, "%s:dma_request_channel FAILS(Tx)\n", __func__); - pci_dev_put(dma_dev); - return; + goto err_pci_get; } priv->chan_tx = chan; @@ -704,18 +703,26 @@ static void pch_request_dma(struct uart_port *port) if (!chan) { dev_err(priv->port.dev, "%s:dma_request_channel FAILS(Rx)\n", __func__); - dma_release_channel(priv->chan_tx); - priv->chan_tx = NULL; - pci_dev_put(dma_dev); - return; + goto err_req_tx; } /* Get Consistent memory for DMA */ priv->rx_buf_virt = dma_alloc_coherent(port->dev, port->fifosize, &priv->rx_buf_dma, GFP_KERNEL); + if (!priv->rx_buf_virt) + goto err_req_rx; priv->chan_rx = chan; pci_dev_put(dma_dev); + return; + +err_req_rx: + dma_release_channel(chan); +err_req_tx: + dma_release_channel(priv->chan_tx); + priv->chan_tx = NULL; +err_pci_get: + pci_dev_put(dma_dev); } static void pch_dma_rx_complete(void *arg) From 92b1ea22454b08a39baef3a7290fb3ec50366616 Mon Sep 17 00:00:00 2001 From: Hongling Zeng Date: Tue, 21 Apr 2026 14:57:37 +0800 Subject: [PATCH 032/411] serial: sh-sci: fix memory region release in error path The sci_request_port() function uses request_mem_region() to reserve I/O memory, but in the error path when sci_remap_port() fails, it incorrectly calls release_resource() instead of release_mem_region(). This mismatch can cause resource accounting issues. Fix it by using the correct release function, consistent with sci_release_port(). Fixes: e2651647080930a1 ("serial: sh-sci: Handle port memory region reservations.") Cc: stable Reported-by: kernel test robot Reported-by: Dan Carpenter Closes: https://lore.kernel.org/r/202604032356.SzEjYkBC-lkp@intel.com/ Signed-off-by: Hongling Zeng Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/20260421065737.724187-1-zenghongling@kylinos.cn Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/sh-sci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/sh-sci.c b/drivers/tty/serial/sh-sci.c index 6c819b6b2425..54db019a5bfc 100644 --- a/drivers/tty/serial/sh-sci.c +++ b/drivers/tty/serial/sh-sci.c @@ -3025,7 +3025,7 @@ int sci_request_port(struct uart_port *port) ret = sci_remap_port(port); if (unlikely(ret != 0)) { - release_resource(res); + release_mem_region(port->mapbase, sport->reg_size); return ret; } From ca2584d841b69391ffc4144840563d2e1a0018df Mon Sep 17 00:00:00 2001 From: Prasanna S Date: Tue, 28 Apr 2026 09:56:13 +0530 Subject: [PATCH 033/411] serial: qcom-geni: fix UART_RX_PAR_EN bit position UART_RX_PAR_EN is incorrectly defined as bit 3, which triggers false framing errors (S_GP_IRQ_1_EN) and causes received data to be dropped when parity is enabled and the parity bit is 0. Define UART_RX_PAR_EN as bit 4 of the SE_UART_RX_TRANS_CFG register, as specified in the reference manual. Fixes: c4f528795d1a ("tty: serial: msm_geni_serial: Add serial driver support for GENI based QUP") Cc: stable Signed-off-by: Prasanna S Reviewed-by: Konrad Dybcio Link: https://patch.msgid.link/20260428-serial-bit-correct-v1-1-9131ad5b97d8@oss.qualcomm.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/qcom_geni_serial.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c index b365dd5da3cb..5139a9d21b2b 100644 --- a/drivers/tty/serial/qcom_geni_serial.c +++ b/drivers/tty/serial/qcom_geni_serial.c @@ -50,7 +50,7 @@ #define TX_STOP_BIT_LEN_2 2 /* SE_UART_RX_TRANS_CFG */ -#define UART_RX_PAR_EN BIT(3) +#define UART_RX_PAR_EN BIT(4) /* SE_UART_RX_WORD_LEN */ #define RX_WORD_LEN_MASK GENMASK(9, 0) From 1e5b50c78d10119be08bf8f7a11d8ea333dd113a Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 6 May 2026 14:43:23 +0200 Subject: [PATCH 034/411] tty: add missing tty_driver include to tty_port.h Include the definition of struct tty_driver in tty_port.h to keep the header self-contained and avoid build breakage in case anyone includes it before tty_driver.h. Fixes: eb3b0d92c9c3 ("tty: tty_port: add workqueue to flip TTY buffer") Cc: Xin Zhao Signed-off-by: Johan Hovold Link: https://patch.msgid.link/20260506124323.186703-1-johan@kernel.org Signed-off-by: Greg Kroah-Hartman --- include/linux/tty_port.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/tty_port.h b/include/linux/tty_port.h index d2a7882c0b58..23cad403bb8f 100644 --- a/include/linux/tty_port.h +++ b/include/linux/tty_port.h @@ -6,10 +6,10 @@ #include #include #include +#include #include struct attribute_group; -struct tty_driver; struct tty_port; struct tty_struct; From 9a9254c4a2a3ca2b3da16d173f3b0dd01f397ff6 Mon Sep 17 00:00:00 2001 From: Shitalkumar Gandhi Date: Mon, 20 Apr 2026 19:29:03 +0530 Subject: [PATCH 035/411] serial: fsl_lpuart: fix rx buffer and DMA map leaks in start_rx_dma lpuart_start_rx_dma() allocates sport->rx_ring.buf with kzalloc() and then maps a scatterlist via dma_map_sg(). On three subsequent error paths the function returns directly without releasing those resources: - when dma_map_sg() returns 0 (-EINVAL): ring->buf is leaked. - when dmaengine_slave_config() fails: ring->buf and the DMA mapping are leaked. - when dmaengine_prep_dma_cyclic() returns NULL: ring->buf and the DMA mapping are leaked. The sole cleanup path, lpuart_dma_rx_free(), is only reached when lpuart_dma_rx_use is set, and the caller lpuart_rx_dma_startup() clears that flag on failure of lpuart_start_rx_dma(). So these resources are permanently leaked on every failure in this function. Repeated port open/close or termios changes under error conditions will slowly consume memory and leave stale streaming DMA mappings behind. Fix it by introducing two error labels that unmap the scatterlist and free the ring buffer as appropriate. While here, replace the misleading -EFAULT (bad userspace pointer) returned when dmaengine_prep_dma_cyclic() fails with the more accurate -ENOMEM, matching how other dmaengine users in the tree treat this failure. No functional change on the success path. Fixes: 5887ad43ee02 ("tty: serial: fsl_lpuart: Use cyclic DMA for Rx") Cc: stable Signed-off-by: Shitalkumar Gandhi Reviewed-by: Frank Li Link: https://patch.msgid.link/20260420135903.2062024-1-shitalkumar.gandhi@cambiumnetworks.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/fsl_lpuart.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/drivers/tty/serial/fsl_lpuart.c b/drivers/tty/serial/fsl_lpuart.c index 1bd7ec9c81ea..b7919c05f0fb 100644 --- a/drivers/tty/serial/fsl_lpuart.c +++ b/drivers/tty/serial/fsl_lpuart.c @@ -1379,7 +1379,8 @@ static inline int lpuart_start_rx_dma(struct lpuart_port *sport) if (!nent) { dev_err(sport->port.dev, "DMA Rx mapping error\n"); - return -EINVAL; + ret = -EINVAL; + goto err_free_buf; } dma_rx_sconfig.src_addr = lpuart_dma_datareg_addr(sport); @@ -1391,7 +1392,7 @@ static inline int lpuart_start_rx_dma(struct lpuart_port *sport) if (ret < 0) { dev_err(sport->port.dev, "DMA Rx slave config failed, err = %d\n", ret); - return ret; + goto err_unmap_sg; } sport->dma_rx_desc = dmaengine_prep_dma_cyclic(chan, @@ -1402,7 +1403,8 @@ static inline int lpuart_start_rx_dma(struct lpuart_port *sport) DMA_PREP_INTERRUPT); if (!sport->dma_rx_desc) { dev_err(sport->port.dev, "Cannot prepare cyclic DMA\n"); - return -EFAULT; + ret = -ENOMEM; + goto err_unmap_sg; } sport->dma_rx_desc->callback = lpuart_dma_rx_complete; @@ -1426,6 +1428,13 @@ static inline int lpuart_start_rx_dma(struct lpuart_port *sport) } return 0; + +err_unmap_sg: + dma_unmap_sg(chan->device->dev, &sport->rx_sgl, 1, DMA_FROM_DEVICE); +err_free_buf: + kfree(ring->buf); + ring->buf = NULL; + return ret; } static void lpuart_dma_rx_free(struct uart_port *port) From 452d6fa37ae9b021f4f6d397dbae077f7296f6f4 Mon Sep 17 00:00:00 2001 From: Viken Dadhaniya Date: Wed, 6 May 2026 10:15:21 +0530 Subject: [PATCH 036/411] serial: qcom_geni: fix kfifo underflow when flush precedes DMA completion IRQ When uart_flush_buffer() runs before the DMA completion IRQ is delivered, the following race can occur (all steps serialized by uart_port_lock): 1. DMA starts: tx_remaining = N, kfifo contains N bytes 2. DMA completes in hardware; IRQ is pending but not yet delivered 3. uart_flush_buffer() acquires the port lock and calls kfifo_reset(), making kfifo_len() = 0 while tx_remaining remains N 4. uart_flush_buffer() releases the port lock 5. DMA IRQ fires; handle_tx_dma() acquires the port lock and calls uart_xmit_advance(uport, tx_remaining) on an empty kfifo uart_xmit_advance() increments kfifo->out by tx_remaining. Since kfifo_reset() already set both in and out to 0, out wraps past in, causing kfifo_len() to return UART_XMIT_SIZE - tx_remaining. The next start_tx_dma() call then submits a DMA transfer of stale buffer data. Fix this by snapshotting kfifo_len() at the start of handle_tx_dma() and skipping uart_xmit_advance() when fifo_len < tx_remaining, which indicates the kfifo was reset by a preceding flush. Fixes: 2aaa43c70778 ("tty: serial: qcom-geni-serial: add support for serial engine DMA") Cc: stable Signed-off-by: Viken Dadhaniya Reviewed-by: Bartosz Golaszewski Link: https://patch.msgid.link/20260506-serial-dma-stale-tx-buf-v1-1-e3ccb360d719@oss.qualcomm.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/qcom_geni_serial.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c index 5139a9d21b2b..17da115b1e78 100644 --- a/drivers/tty/serial/qcom_geni_serial.c +++ b/drivers/tty/serial/qcom_geni_serial.c @@ -1031,8 +1031,20 @@ static void qcom_geni_serial_handle_tx_dma(struct uart_port *uport) { struct qcom_geni_serial_port *port = to_dev_port(uport); struct tty_port *tport = &uport->state->port; + unsigned int fifo_len = kfifo_len(&tport->xmit_fifo); + + /* + * Only advance the kfifo if it still contains the bytes that were + * transferred. uart_flush_buffer() may have run before this IRQ + * fired: it calls kfifo_reset() under the port lock, making + * fifo_len = 0 while tx_remaining remains non-zero. Calling + * uart_xmit_advance() in that case would underflow kfifo->out past + * kfifo->in, making kfifo_len() wrap to UART_XMIT_SIZE - tx_remaining + * and triggering a spurious large DMA transfer of stale data. + */ + if (fifo_len >= port->tx_remaining) + uart_xmit_advance(uport, port->tx_remaining); - uart_xmit_advance(uport, port->tx_remaining); geni_se_tx_dma_unprep(&port->se, port->tx_dma_addr, port->tx_remaining); port->tx_dma_addr = 0; port->tx_remaining = 0; From 6f89d96fff65aec1ff12bc566fca0eb1bb59e16e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 12 May 2026 18:26:54 +0200 Subject: [PATCH 037/411] Input: atlas - check ACPI_COMPANION() against NULL Every platform driver can be forced to match a device that doesn't match its list of device IDs because of device_match_driver_override(), so platform drivers that rely on the existence of a device's ACPI companion object need to verify its presence. Accordingly, add a requisite ACPI_COMPANION() check against NULL to the atlas_btns driver. Fixes: b8303880b641 ("Input: atlas - convert ACPI driver to a platform one") Signed-off-by: Rafael J. Wysocki Link: https://patch.msgid.link/8696590.T7Z3S40VBb@rafael.j.wysocki Signed-off-by: Dmitry Torokhov --- drivers/input/misc/atlas_btns.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/input/misc/atlas_btns.c b/drivers/input/misc/atlas_btns.c index 47b31725e850..835ad45a9d65 100644 --- a/drivers/input/misc/atlas_btns.c +++ b/drivers/input/misc/atlas_btns.c @@ -60,11 +60,15 @@ static acpi_status acpi_atlas_button_handler(u32 function, static int atlas_acpi_button_probe(struct platform_device *pdev) { - struct acpi_device *device = ACPI_COMPANION(&pdev->dev); + struct acpi_device *device; acpi_status status; int i; int err; + device = ACPI_COMPANION(&pdev->dev); + if (!device) + return -ENODEV; + input_dev = input_allocate_device(); if (!input_dev) { pr_err("unable to allocate input device\n"); From 6d3790bc689de9f18fae01c21f02e7d6d425534c Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Mon, 27 Apr 2026 18:25:03 -0700 Subject: [PATCH 038/411] KVM: selftests: Include sys/mman.h *and* linux/mman.h, via kvm_syscalls.h MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Include both linux/mman.h (the kernel provided version) and sys/mman.h (the libc provided version) throughout KVM selftests, by way of kvm_syscalls.h (which should have been including sys/mman.h anyways). Pulling in the kernel's version fixes compilation errors with the guest_memfd test on older versions of libc due to a recent commit adding MADV_COLLAPSE testing. In file included from include/kvm_util.h:8, from guest_memfd_test.c:21: guest_memfd_test.c: In function ‘test_collapse’: guest_memfd_test.c:219:47: error: ‘MADV_COLLAPSE’ undeclared (first use in this function); did you mean ‘MADV_COLD’? 219 | TEST_ASSERT_EQ(madvise(mem, pmd_size, MADV_COLLAPSE), -1); | ^~~~~~~~~~~~~ include/test_util.h:62:16: note: in definition of macro ‘TEST_ASSERT_EQ’ 62 | typeof(a) __a = (a); \ | ^ guest_memfd_test.c:219:47: note: each undeclared identifier is reported only once for each function it appears in 219 | TEST_ASSERT_EQ(madvise(mem, pmd_size, MADV_COLLAPSE), -1); | ^~~~~~~~~~~~~ include/test_util.h:62:16: note: in definition of macro ‘TEST_ASSERT_EQ’ 62 | typeof(a) __a = (a); \ | ^ Route the includes through kvm_syscalls.h to try and avoid a future game of whack-a-mole, i.e. so that future expansion of test coverage doesn't run into the same problem. To discourage use of sys/mman.h, opportunistically include the kernel's version of mman.h in test_util.h as it only needs MAP_SHARED, i.e. only needs the full set of kernel defs, not the libc syscall wrappers. Fixes: 9830209b4ae8 ("KVM: selftests: Test MADV_COLLAPSE on guest_memfd") Reported-by: Rick Edgecombe Closes: https://lore.kernel.org/all/20260427204313.50741-1-rick.p.edgecombe@intel.com Link: https://patch.msgid.link/20260428012503.1213654-1-seanjc@google.com Signed-off-by: Sean Christopherson --- .../testing/selftests/kvm/access_tracking_perf_test.c | 2 +- tools/testing/selftests/kvm/guest_memfd_test.c | 2 +- tools/testing/selftests/kvm/include/kvm_syscalls.h | 10 ++++++++++ tools/testing/selftests/kvm/include/test_util.h | 2 +- tools/testing/selftests/kvm/lib/kvm_util.c | 2 +- tools/testing/selftests/kvm/memslot_perf_test.c | 2 +- .../testing/selftests/kvm/s390/shared_zeropage_test.c | 3 +-- tools/testing/selftests/kvm/s390/tprot.c | 2 +- tools/testing/selftests/kvm/set_memory_region_test.c | 2 +- 9 files changed, 18 insertions(+), 9 deletions(-) diff --git a/tools/testing/selftests/kvm/access_tracking_perf_test.c b/tools/testing/selftests/kvm/access_tracking_perf_test.c index e5bbdb5bbdc3..4415c94b2866 100644 --- a/tools/testing/selftests/kvm/access_tracking_perf_test.c +++ b/tools/testing/selftests/kvm/access_tracking_perf_test.c @@ -41,10 +41,10 @@ #include #include #include -#include #include #include +#include "kvm_syscalls.h" #include "kvm_util.h" #include "test_util.h" #include "memstress.h" diff --git a/tools/testing/selftests/kvm/guest_memfd_test.c b/tools/testing/selftests/kvm/guest_memfd_test.c index 253e748c1d4a..832ef4dfb99f 100644 --- a/tools/testing/selftests/kvm/guest_memfd_test.c +++ b/tools/testing/selftests/kvm/guest_memfd_test.c @@ -14,10 +14,10 @@ #include #include #include -#include #include #include +#include "kvm_syscalls.h" #include "kvm_util.h" #include "numaif.h" #include "test_util.h" diff --git a/tools/testing/selftests/kvm/include/kvm_syscalls.h b/tools/testing/selftests/kvm/include/kvm_syscalls.h index 843c9904c46f..067a4c9cf452 100644 --- a/tools/testing/selftests/kvm/include/kvm_syscalls.h +++ b/tools/testing/selftests/kvm/include/kvm_syscalls.h @@ -2,8 +2,18 @@ #ifndef SELFTEST_KVM_SYSCALLS_H #define SELFTEST_KVM_SYSCALLS_H +/* + * Include both the kernel and libc versions of mman.h. The kernel provides + * the most up-to-date flags and definitions, while libc provides the syscall + * wrappers tests expect. + */ +#include + +#include #include +#include + #define MAP_ARGS0(m,...) #define MAP_ARGS1(m,t,a,...) m(t,a) #define MAP_ARGS2(m,t,a,...) m(t,a), MAP_ARGS1(m,__VA_ARGS__) diff --git a/tools/testing/selftests/kvm/include/test_util.h b/tools/testing/selftests/kvm/include/test_util.h index d9b433b834f1..a56271c237ae 100644 --- a/tools/testing/selftests/kvm/include/test_util.h +++ b/tools/testing/selftests/kvm/include/test_util.h @@ -19,9 +19,9 @@ #include #include #include -#include #include "kselftest.h" +#include #include #define msecs_to_usecs(msec) ((msec) * 1000ULL) diff --git a/tools/testing/selftests/kvm/lib/kvm_util.c b/tools/testing/selftests/kvm/lib/kvm_util.c index 2a76eca7029d..e08967ef7b7b 100644 --- a/tools/testing/selftests/kvm/lib/kvm_util.c +++ b/tools/testing/selftests/kvm/lib/kvm_util.c @@ -5,13 +5,13 @@ * Copyright (C) 2018, Google LLC. */ #include "test_util.h" +#include "kvm_syscalls.h" #include "kvm_util.h" #include "processor.h" #include "ucall_common.h" #include #include -#include #include #include #include diff --git a/tools/testing/selftests/kvm/memslot_perf_test.c b/tools/testing/selftests/kvm/memslot_perf_test.c index 3d02db371422..e977e979470f 100644 --- a/tools/testing/selftests/kvm/memslot_perf_test.c +++ b/tools/testing/selftests/kvm/memslot_perf_test.c @@ -15,7 +15,6 @@ #include #include #include -#include #include #include @@ -23,6 +22,7 @@ #include #include +#include #include #include #include diff --git a/tools/testing/selftests/kvm/s390/shared_zeropage_test.c b/tools/testing/selftests/kvm/s390/shared_zeropage_test.c index a9e5a01200b8..478381e6f84e 100644 --- a/tools/testing/selftests/kvm/s390/shared_zeropage_test.c +++ b/tools/testing/selftests/kvm/s390/shared_zeropage_test.c @@ -4,11 +4,10 @@ * * Copyright (C) 2024, Red Hat, Inc. */ -#include - #include #include "test_util.h" +#include "kvm_syscalls.h" #include "kvm_util.h" #include "kselftest.h" #include "ucall_common.h" diff --git a/tools/testing/selftests/kvm/s390/tprot.c b/tools/testing/selftests/kvm/s390/tprot.c index 8054d2b178f0..d86179827a18 100644 --- a/tools/testing/selftests/kvm/s390/tprot.c +++ b/tools/testing/selftests/kvm/s390/tprot.c @@ -4,8 +4,8 @@ * * Copyright IBM Corp. 2021 */ -#include #include "test_util.h" +#include "kvm_syscalls.h" #include "kvm_util.h" #include "kselftest.h" #include "ucall_common.h" diff --git a/tools/testing/selftests/kvm/set_memory_region_test.c b/tools/testing/selftests/kvm/set_memory_region_test.c index 9b919a231c93..e639a9db51ee 100644 --- a/tools/testing/selftests/kvm/set_memory_region_test.c +++ b/tools/testing/selftests/kvm/set_memory_region_test.c @@ -8,11 +8,11 @@ #include #include #include -#include #include #include +#include #include #include From fff82ea9d900b6bbebc58d34b7a63789de1ad10d Mon Sep 17 00:00:00 2001 From: Mikhail Gavrilov Date: Tue, 5 May 2026 04:54:35 +0500 Subject: [PATCH 039/411] x86/virt: Silence RCU lockdep splat in emergency virt callback path MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit x86_virt_invoke_kvm_emergency_callback() reaches rcu_dereference() through machine_crash_shutdown() with IRQs disabled but with RCU not necessarily watching the crashing CPU, which triggers a suspicious RCU usage splat on debug kernels (CONFIG_PROVE_RCU=y) during panic/kdump: WARNING: suspicious RCU usage arch/x86/virt/hw.c:52 suspicious rcu_dereference_check() usage! rcu_scheduler_active = 2, debug_locks = 1 1 lock held by tee/11119: #0: ffff8881fa32c440 (sb_writers#3){.+.+}-{0:0}, at: ksys_write Call Trace: dump_stack_lvl+0x84/0xd0 lockdep_rcu_suspicious.cold+0x37/0x8f x86_virt_invoke_kvm_emergency_callback+0x5f/0x70 x86_svm_emergency_disable_virtualization_cpu+0x2a/0x30 x86_virt_emergency_disable_virtualization_cpu+0x6b/0x90 native_machine_crash_shutdown+0x72/0x170 __crash_kexec+0x137/0x280 panic+0xce/0xd0 sysrq_handle_crash+0x1f/0x20 __handle_sysrq.cold+0x192/0x335 write_sysrq_trigger+0x8c/0xc0 proc_reg_write+0x1c3/0x3c0 vfs_write+0x1d0/0xf80 ksys_write+0x116/0x250 do_syscall_64+0x11c/0x1480 entry_SYSCALL_64_after_hwframe+0x76/0x7e A truly correct fix is non-trivial: the RCU usage genuinely is wrong in panic context (RCU may ignore the crashing CPU during synchronization), and a concurrent KVM module unload could in principle race with the callback read; see commit 2baa33a8ddd6 ("KVM: x86: Leave user-return notifier registered on reboot/shutdown") which notes that nothing prevents module unload during panic/reboot. However, the alternatives are worse: - smp_store_release()/smp_load_acquire() handles ordering but not liveness; the kernel still needs to keep the module text alive while the callback is in flight. - Taking a lock in the panic path is risky — any lock could be held by a CPU that has already been NMI'd to a halt. Use rcu_dereference_raw() to silence the splat and accept the vanishingly small remaining race. Panic context inherently cannot guarantee complete correctness; the goal here is to keep debug builds quiet on the kdump path so the splat doesn't obscure the actual kernel state being captured. Reproducible on a debug kernel (CONFIG_PROVE_LOCKING=y, CONFIG_PROVE_RCU=y) with kvm_amd or kvm_intel loaded by triggering kdump: echo c > /proc/sysrq-trigger Suggested-by: Sean Christopherson Fixes: 428afac5a8ea ("KVM: x86: Move bulk of emergency virtualizaton logic to virt subsystem") Signed-off-by: Mikhail Gavrilov Acked-by: Sean Christopherson Link: https://patch.msgid.link/20260504235435.90957-1-mikhail.v.gavrilov@gmail.com Signed-off-by: Sean Christopherson --- arch/x86/virt/hw.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/arch/x86/virt/hw.c b/arch/x86/virt/hw.c index f647557d38ac..7e9091c640be 100644 --- a/arch/x86/virt/hw.c +++ b/arch/x86/virt/hw.c @@ -49,7 +49,20 @@ static void x86_virt_invoke_kvm_emergency_callback(void) { cpu_emergency_virt_cb *kvm_callback; - kvm_callback = rcu_dereference(kvm_emergency_callback); + /* + * RCU may not be watching the crashing CPU here, so rcu_dereference() + * triggers a suspicious-RCU-usage splat. In principle, a concurrent + * KVM module unload could race with this read; see commit 2baa33a8ddd6 + * ("KVM: x86: Leave user-return notifier registered on reboot/shutdown") + * which notes that nothing prevents module unload during panic/reboot. + * + * However, taking a lock here would be riskier than the current race: + * the system is going down via NMI shootdown, and any lock could be + * held by an already-stopped CPU. Use rcu_dereference_raw() to silence + * the lockdep splat and accept the comically small remaining race; + * panic context inherently cannot guarantee complete correctness. + */ + kvm_callback = rcu_dereference_raw(kvm_emergency_callback); if (kvm_callback) kvm_callback(); } From 8fe2e698fce4a95a3ac2c25fe59832a3c22534c6 Mon Sep 17 00:00:00 2001 From: Lei Chen Date: Thu, 9 Apr 2026 22:22:26 +0800 Subject: [PATCH 040/411] KVM: x86: Rate-limit global clock updates on vCPU load commit 446fcce2a52b ("Revert "x86: kvm: rate-limit global clock updates"") dropped the rate limiting for KVM_REQ_GLOBAL_CLOCK_UPDATE. As a result, kvm_arch_vcpu_load() can queue global clock update requests every time a vCPU is scheduled when the master clock is disabled or when the vCPU is loaded for the first time. Restore the throttling with a per-VM ratelimit state and gate KVM_REQ_GLOBAL_CLOCK_UPDATE through __ratelimit(), so frequent vCPU scheduling does not generate a steady stream of redundant clock update requests. Fixes: 446fcce2a52b ("Revert "x86: kvm: rate-limit global clock updates"") Signed-off-by: Lei Chen Reported-by: Jaroslav Pulchart Closes: https://lore.kernel.org/all/CAK8fFZ5gY8_Mw2A=iZVFNVKQNrXQzVsn-HTd+Me9K6ZfmdgA+Q@mail.gmail.com/ Link: https://patch.msgid.link/20260409142226.2581-1-lei.chen@smartx.com Signed-off-by: Sean Christopherson --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/x86.c | 11 +++++++++-- 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index c470e40a00aa..f14009f25a3b 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1504,6 +1504,7 @@ struct kvm_arch { bool use_master_clock; u64 master_kernel_ns; u64 master_cycle_now; + struct ratelimit_state kvmclock_update_rs; #ifdef CONFIG_KVM_HYPERV struct kvm_hv hyperv; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0a1b63c63d1a..e01d6984ed04 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5227,8 +5227,13 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) * On a host with synchronized TSC, there is no need to update * kvmclock on vcpu->cpu migration */ - if (!vcpu->kvm->arch.use_master_clock || vcpu->cpu == -1) - kvm_make_request(KVM_REQ_GLOBAL_CLOCK_UPDATE, vcpu); + if (!vcpu->kvm->arch.use_master_clock || vcpu->cpu == -1) { + if (__ratelimit(&vcpu->kvm->arch.kvmclock_update_rs)) + kvm_make_request(KVM_REQ_GLOBAL_CLOCK_UPDATE, vcpu); + else + kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); + } + if (vcpu->cpu != cpu) kvm_make_request(KVM_REQ_MIGRATE_TIMER, vcpu); vcpu->cpu = cpu; @@ -13366,6 +13371,8 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) raw_spin_lock_init(&kvm->arch.tsc_write_lock); mutex_init(&kvm->arch.apic_map_lock); seqcount_raw_spinlock_init(&kvm->arch.pvclock_sc, &kvm->arch.tsc_write_lock); + ratelimit_state_init(&kvm->arch.kvmclock_update_rs, HZ, 10); + ratelimit_set_flags(&kvm->arch.kvmclock_update_rs, RATELIMIT_MSG_ON_RELEASE); kvm->arch.kvmclock_offset = -get_kvmclock_base_ns(); raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags); From 34065a5f3cf94886e59e2a8b5db00515f32d6cf2 Mon Sep 17 00:00:00 2001 From: Hisam Mehboob Date: Thu, 9 Apr 2026 20:38:47 +0500 Subject: [PATCH 041/411] KVM: selftests: Guard execinfo.h inclusion for non-glibc builds The backtrace() function and execinfo.h are GNU extensions available in glibc but not in non-glibc C libraries such as musl. Building KVM selftests with musl-gcc fails with: lib/assert.c:9:10: fatal error: execinfo.h: No such file or directory Fix this by guarding the inclusion of execinfo.h and the stack dumping logic under #ifdef __GLIBC__. For non-glibc builds, provide a local stub for test_dump_stack(). Suggested-by: Aqib Faruqui Suggested-by: Sean Christopherson Signed-off-by: Hisam Mehboob Link: https://patch.msgid.link/20260409153846.1502656-2-hisamshar@gmail.com Signed-off-by: Sean Christopherson --- tools/testing/selftests/kvm/lib/assert.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/kvm/lib/assert.c b/tools/testing/selftests/kvm/lib/assert.c index b49690658c60..8be0d09ecf0f 100644 --- a/tools/testing/selftests/kvm/lib/assert.c +++ b/tools/testing/selftests/kvm/lib/assert.c @@ -6,11 +6,14 @@ */ #include "test_util.h" -#include + #include #include "kselftest.h" +#ifdef __GLIBC__ +#include + /* Dumps the current stack trace to stderr. */ static void __attribute__((noinline)) test_dump_stack(void); static void test_dump_stack(void) @@ -57,6 +60,9 @@ static void test_dump_stack(void) system(cmd); #pragma GCC diagnostic pop } +#else +static void test_dump_stack(void) {} +#endif static pid_t _gettid(void) { From 16ca52bc209fa4bf9239cd9e5643e95533476b58 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Nicol=C3=A1s=20Bazaes?= Date: Wed, 13 May 2026 21:35:49 -0400 Subject: [PATCH 042/411] Input: synaptics - add LEN2058 to SMBus passlist for ThinkPad E490 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The Lenovo ThinkPad E490 (PNP ID: LEN2058) has a Synaptics TM3471-020 touchpad that supports SMBus/RMI4 mode but is not listed in smbus_pnp_ids[]. Without this entry, RMI4 over SMBus is not enabled by default, and the touchpad falls back to PS/2 mode. Adding LEN2058 to the passlist enables automatic RMI4 detection without requiring the psmouse.synaptics_intertouch parameter, and matches the behavior of similar ThinkPad models already in the list (E480/LEN2054, E580/LEN2055). Tested on ThinkPad E490 with kernel 7.0.5-zen1 and Arch Linux. RMI4 over SMBus is confirmed working without any kernel parameters. Signed-off-by: Nicolás Bazaes Assisted-by: Claude:claude-sonnet-4-6 Link: https://patch.msgid.link/20260514013552.14234-1-contacto@bazaes.cl Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/mouse/synaptics.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/input/mouse/synaptics.c b/drivers/input/mouse/synaptics.c index 26071128f43a..c70502e24031 100644 --- a/drivers/input/mouse/synaptics.c +++ b/drivers/input/mouse/synaptics.c @@ -190,6 +190,7 @@ static const char * const smbus_pnp_ids[] = { "LEN2044", /* L470 */ "LEN2054", /* E480 */ "LEN2055", /* E580 */ + "LEN2058", /* E490 */ "LEN2068", /* T14 Gen 1 */ "SYN1221", /* TUXEDO InfinityBook Pro 14 v5 */ "SYN3003", /* HP EliteBook 850 G1 */ From 742b04d0550b0ec89dcbc99537ec88653bd1ad90 Mon Sep 17 00:00:00 2001 From: David Ahern Date: Wed, 13 May 2026 10:49:14 -0600 Subject: [PATCH 043/411] xfrm: Check for underflow in xfrm_state_mtu Leo Lin reported OOB write issue in esp component: xfrm_state_mtu() returns u32 but performs its arithmetic in unsigned modulo-2^32 space using an attacker-influenced "header_len + authsize + net_adj" subtracted from a small "mtu" argument. A nobody user can install an IPv4 ESP tunnel SA with a large authentication key (XFRMA_ALG_AUTH_TRUNC, e.g. hmac(sha512), 64-byte key, 64-byte trunc), configure a small interface MTU (68 bytes), and set XFRMA_TFCPAD to a large value. When a single UDP datagram is then sent through the tunnel, xfrm_state_mtu() underflows to a near-2^32 value, and esp_output() consumes it as a signed int via: padto = min(x->tfcpad, xfrm_state_mtu(x, mtu_cached)) esp.tfclen = padto - skb->len (assigned to int) esp.tfclen ends up negative (e.g. -207). It is sign-extended to size_t when passed to memset() inside esp_output_fill_trailer(), producing a ~16 EB write of zeroes at skb_tail_pointer(skb). KASAN logs it as "Write of size 18446744073709551537 at addr ffff888...". Check for underflow and return 1. This causes the sendmsg attempt to fail with ENETUNREACH. Fixes: c5c252389374 ("[XFRM]: Optimize MTU calculation") Reported-by: Leo Lin Assisted-by: Codex:26.506.31004 Signed-off-by: David Ahern Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 395d82411a87..589c3b6e4679 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -3114,10 +3114,14 @@ u32 xfrm_state_mtu(struct xfrm_state *x, int mtu) const struct xfrm_type *type = READ_ONCE(x->type); struct crypto_aead *aead; u32 blksize, net_adj = 0; + u32 overhead, payload_mtu; if (x->km.state != XFRM_STATE_VALID || - !type || type->proto != IPPROTO_ESP) + !type || type->proto != IPPROTO_ESP) { + if (mtu <= x->props.header_len) + return 1; return mtu - x->props.header_len; + } aead = x->data; blksize = ALIGN(crypto_aead_blocksize(aead), 4); @@ -3140,8 +3144,17 @@ u32 xfrm_state_mtu(struct xfrm_state *x, int mtu) break; } - return ((mtu - x->props.header_len - crypto_aead_authsize(aead) - - net_adj) & ~(blksize - 1)) + net_adj - 2; + overhead = x->props.header_len + crypto_aead_authsize(aead) + net_adj; + if (mtu <= overhead) + return 1; + + payload_mtu = mtu - overhead; + payload_mtu &= ~(blksize - 1); + if (payload_mtu <= 2) + return 1; + + return payload_mtu + net_adj - 2; + } EXPORT_SYMBOL_GPL(xfrm_state_mtu); From a9aba21a539c668a66b58eeb08ad3909e5a54c2a Mon Sep 17 00:00:00 2001 From: Antoniu Miclaus Date: Wed, 1 Apr 2026 18:29:24 +0300 Subject: [PATCH 044/411] iio: adc: nxp-sar-adc: fix division by zero in write_raw Add a validation check for the sampling frequency value before using it as a divisor. A user writing zero or a negative value to the sampling_frequency sysfs attribute triggers a division by zero in the kernel. Also prevent unsigned integer underflow when the computed cycle count is smaller than NXP_SAR_ADC_CONV_TIME, which would wrap the u32 inpsamp to a huge value. Fixes: 4434072a893e ("iio: adc: Add the NXP SAR ADC support for the s32g2/3 platforms") Signed-off-by: Antoniu Miclaus Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/nxp-sar-adc.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/iio/adc/nxp-sar-adc.c b/drivers/iio/adc/nxp-sar-adc.c index 705dd7da1bd2..1711cae7d872 100644 --- a/drivers/iio/adc/nxp-sar-adc.c +++ b/drivers/iio/adc/nxp-sar-adc.c @@ -569,6 +569,9 @@ static int nxp_sar_adc_write_raw(struct iio_dev *indio_dev, struct iio_chan_spec switch (mask) { case IIO_CHAN_INFO_SAMP_FREQ: + if (val <= 0) + return -EINVAL; + /* * Configures the sample period duration in terms of the SAR * controller clock. The minimum acceptable value is 8. @@ -577,7 +580,11 @@ static int nxp_sar_adc_write_raw(struct iio_dev *indio_dev, struct iio_chan_spec * sampling timing which gives us the number of cycles expected. * The value is 8-bit wide, consequently the max value is 0xFF. */ - inpsamp = clk_get_rate(info->clk) / val - NXP_SAR_ADC_CONV_TIME; + inpsamp = clk_get_rate(info->clk) / val; + if (inpsamp < NXP_SAR_ADC_CONV_TIME) + return -EINVAL; + + inpsamp -= NXP_SAR_ADC_CONV_TIME; nxp_sar_adc_conversion_timing_set(info, inpsamp); return 0; From a093999355084bdbfe6e97f1dd232e58a1525f0b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Beno=C3=AEt=20Monin?= Date: Wed, 1 Apr 2026 17:24:58 +0200 Subject: [PATCH 045/411] iio: buffer: Fix DMA fence leak in iio_buffer_enqueue_dmabuf() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit iio_buffer_enqueue_dmabuf() allocates a struct iio_dma_fence (104 bytes, kmalloc-128) via kmalloc_obj()+dma_fence_init(), which sets the initial kref to 1. It then calls dma_resv_add_fence() which takes a second reference (kref=2), and stores a raw pointer in block->fence. On the success path the function returns without calling dma_fence_put() to release the initial reference, so every buffer enqueue permanently leaks one kmalloc-128 allocation. The iio_buffer_cleanup() work item only releases the temporary reference taken during completion signalling by iio_buffer_signal_dmabuf_done(); the initial reference from dma_fence_init() is never released. With four iio_rwdev instances at 240kHz and 512 samples per buffer, this produces ~1875 kmalloc-128 allocations per second matching the observed slab growth exactly. A test with ftrace confirmed that the dma_fence_destroy event was never triggered. Fix by calling dma_fence_put() after dma_resv_add_fence(), transferring ownership of the fence to the DMA reservation object. The DMA fence then gets properly discarded after being signalled. Fixes: 3e26d9f08fbe0 ("iio: core: Add new DMABUF interface infrastructure") Originally-by: James Nuss Signed-off-by: Benoît Monin Reviewed-by: Paul Cercueil Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/industrialio-buffer.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/iio/industrialio-buffer.c b/drivers/iio/industrialio-buffer.c index 46f36a6ed271..5c3df993bea2 100644 --- a/drivers/iio/industrialio-buffer.c +++ b/drivers/iio/industrialio-buffer.c @@ -1909,6 +1909,7 @@ static int iio_buffer_enqueue_dmabuf(struct iio_dev_buffer_pair *ib, dma_resv_add_fence(dmabuf->resv, &fence->base, dma_to_ram ? DMA_RESV_USAGE_WRITE : DMA_RESV_USAGE_READ); + dma_fence_put(&fence->base); dma_resv_unlock(dmabuf->resv); cookie = dma_fence_begin_signalling(); From 49f79cd28f1e3333cbe0d616ce59ead0b24bf34e Mon Sep 17 00:00:00 2001 From: Advait Dhamorikar Date: Tue, 7 Apr 2026 12:50:59 +0530 Subject: [PATCH 046/411] iio: magnetometer: st_magn: fix default DRDY pin selection for LIS2MDL The device tree binding for st,lis2mdl does not support st,drdy-int-pin property. However, when no platform data is provided and the property is absent, the driver falls back to default_magn_pdata which hardcodes drdy_int_pin = 2. This causes `st_sensors_set_drdy_int_pin` to fail with -EINVAL because the LIS2MDL sensor settings have no INT2 DRDY mask defined. Fix this by checking the sensor's INT2 DRDY mask availability at probe time and selecting the appropriate default pin. Sensors that do not support INT2 DRDY will default to INT1, while all others retain the existing default of INT2. Fixes: 38934daf7b5c ("iio: magnetometer: st_magn: Provide default platform data") Signed-off-by: Advait Dhamorikar Reviewed-by: Andy Shevchenko Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/magnetometer/st_magn_core.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/drivers/iio/magnetometer/st_magn_core.c b/drivers/iio/magnetometer/st_magn_core.c index ef348d316c00..7644bd04654b 100644 --- a/drivers/iio/magnetometer/st_magn_core.c +++ b/drivers/iio/magnetometer/st_magn_core.c @@ -506,6 +506,11 @@ static const struct st_sensors_platform_data default_magn_pdata = { .drdy_int_pin = 2, }; +/* LIS2MDL only supports DRDY on INT1 */ +static const struct st_sensors_platform_data alt_magn_pdata = { + .drdy_int_pin = 1, +}; + static int st_magn_read_raw(struct iio_dev *indio_dev, struct iio_chan_spec const *ch, int *val, int *val2, long mask) @@ -628,8 +633,12 @@ int st_magn_common_probe(struct iio_dev *indio_dev) mdata->current_fullscale = &mdata->sensor_settings->fs.fs_avl[0]; mdata->odr = mdata->sensor_settings->odr.odr_avl[0].hz; - if (!pdata) - pdata = (struct st_sensors_platform_data *)&default_magn_pdata; + if (!pdata) { + if (mdata->sensor_settings->drdy_irq.int2.mask) + pdata = (struct st_sensors_platform_data *)&default_magn_pdata; + else + pdata = (struct st_sensors_platform_data *)&alt_magn_pdata; + } err = st_sensors_init_sensor(indio_dev, pdata); if (err < 0) From 1f4f0bcc5255dec5c4c3a1551bf49d8c33b69b20 Mon Sep 17 00:00:00 2001 From: Aldo Conte Date: Tue, 7 Apr 2026 17:17:01 +0200 Subject: [PATCH 047/411] iio: light: cm3323: fix reg_conf not being initialized correctly The code stores the return value of i2c_smbus_write_word_data() in data->reg_conf; however, this value represents the result of the write operation and not the value actually written to the configuration register. This meant that the contents of data->reg_conf did not truly reflect the contents of the hardware register. Instead, save the value of the register before the write and use this value in the I2C write. The bug was found by code inspection: i2c_smbus_write_word_data() returns 0 on success, not the value written to the register. Tested using i2c-stub on a Raspberry Pi 3B running a custom 6.19.10 kernel. Before loading the driver, the configuration register 0x00 CM3323_CMD_CONF was populated with 0x0030 using `i2cset -y 11 0x10 0x00 0x0030 w`, encoding an integration time of 320ms in bits[6:4]. Due to incorrect initialization of data->reg_conf in cm3323_init(), the print of integration_time returns 0.040000 instead of the expected 0.320000. This happens because the read of the integration_time depends on cm3323_get_it_bits() that is based on the value of data->reg_conf, which is erroneously set to 0. With this fix applied, data->reg_conf correctly saves 0x0030 after init and the successive integration_time reports 0.320000 as expected. Fixes: 8b0544263761 ("iio: light: Add support for Capella CM3323 color sensor") Cc: stable@vger.kernel.org Signed-off-by: Aldo Conte Signed-off-by: Jonathan Cameron --- drivers/iio/light/cm3323.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/iio/light/cm3323.c b/drivers/iio/light/cm3323.c index 79ad6e2209ca..0fe61b8a7029 100644 --- a/drivers/iio/light/cm3323.c +++ b/drivers/iio/light/cm3323.c @@ -89,15 +89,14 @@ static int cm3323_init(struct iio_dev *indio_dev) /* enable sensor and set auto force mode */ ret &= ~(CM3323_CONF_SD_BIT | CM3323_CONF_AF_BIT); + data->reg_conf = ret; - ret = i2c_smbus_write_word_data(data->client, CM3323_CMD_CONF, ret); + ret = i2c_smbus_write_word_data(data->client, CM3323_CMD_CONF, data->reg_conf); if (ret < 0) { dev_err(&data->client->dev, "Error writing reg_conf\n"); return ret; } - data->reg_conf = ret; - return 0; } From 8ce176501f836634f9c0419c0820140f968e9dc5 Mon Sep 17 00:00:00 2001 From: Shuvam Pandey Date: Mon, 6 Apr 2026 15:38:24 +0545 Subject: [PATCH 048/411] iio: adc: nxp-sar-adc: zero-initialize dma_slave_config nxp_sar_adc_start_cyclic_dma() only fills the RX-side members of dma_slave_config before passing it to dmaengine_slave_config(). Zero-initialize the structure so unused members do not contain stack garbage. Some DMA engines consult optional dma_slave_config fields, so leaving them uninitialized can cause DMA setup failures. Fixes: 4434072a893e ("iio: adc: Add the NXP SAR ADC support for the s32g2/3 platforms") Signed-off-by: Shuvam Pandey Reviewed-by: David Lechner Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/nxp-sar-adc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/nxp-sar-adc.c b/drivers/iio/adc/nxp-sar-adc.c index 1711cae7d872..8f4ed3db94f0 100644 --- a/drivers/iio/adc/nxp-sar-adc.c +++ b/drivers/iio/adc/nxp-sar-adc.c @@ -676,7 +676,7 @@ static void nxp_sar_adc_dma_cb(void *data) static int nxp_sar_adc_start_cyclic_dma(struct iio_dev *indio_dev) { struct nxp_sar_adc *info = iio_priv(indio_dev); - struct dma_slave_config config; + struct dma_slave_config config = { }; struct dma_async_tx_descriptor *desc; int ret; From 387c86b582e0782ab332e7bfcd4e6e3f93922961 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 9 Apr 2026 15:40:47 +0200 Subject: [PATCH 049/411] iio: pressure: bmp280: fix stack leak in bmp580 trigger handler MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit bmp580_trigger_handler() declares its scan buffer on the stack without an initializer and then memcpy()s 3 bytes of 24-bit sensor data into each 4-byte __le32 field. The high byte of comp_temp and comp_press is left uninitialized, and the channel storagebits is 32, so two bytes of stack are pushed to userspace per scan. This is a regression from when the buffer lived in the private data, the move to a stack-local struct dropped the implicit zeroing. bme280_trigger_handler() was fixed up to handle this bug, but this driver was not fixed because there was no padding hole, but rather a short-fill issue. Fix this all by just zero-initializing the structure on the stack. Cc: Jonathan Cameron Cc: David Lechner Cc: "Nuno Sá" Cc: Andy Shevchenko Fixes: 872c8014e05e ("iio: pressure: bmp280: drop sensor_data array") Cc: stable Assisted-by: gregkh_clanker_t1000 Signed-off-by: Greg Kroah-Hartman Reviewed-by: David Lechner Signed-off-by: Jonathan Cameron --- drivers/iio/pressure/bmp280-core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/pressure/bmp280-core.c b/drivers/iio/pressure/bmp280-core.c index d983ce9c0b99..9b489766e457 100644 --- a/drivers/iio/pressure/bmp280-core.c +++ b/drivers/iio/pressure/bmp280-core.c @@ -2616,7 +2616,7 @@ static irqreturn_t bmp580_trigger_handler(int irq, void *p) __le32 comp_temp; __le32 comp_press; aligned_s64 timestamp; - } buffer; + } buffer = { }; int ret; guard(mutex)(&data->lock); From c9d8e9adaa63150ef7e833480b799d0bab83a276 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 9 Apr 2026 15:40:48 +0200 Subject: [PATCH 050/411] iio: imu: st_lsm6dsx: fix stack leak in tagged FIFO buffer MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The tagged FIFO path declares iio_buff on the stack with __aligned(8) but no initializer, but there is a hole in the structure, which will then leak to userspace as ST_LSM6DSX_SAMPLE_SIZE bytes (6) will be copied, but the space between that and the timestamp are not initialized. Commit c14edb4d0bdc ("iio:imu:st_lsm6dsx Fix alignment and data leak issues") moved the untagged FIFO path to a kzalloc'd buffer in hw->scan, but for the tagged path it only added the alignment qualifier and not the initializer :( Fix this by just zero-initializing the structure on the stack. Cc: Lorenzo Bianconi Cc: Jonathan Cameron Cc: David Lechner Cc: "Nuno Sá" Cc: Andy Shevchenko Fixes: c14edb4d0bdc ("iio:imu:st_lsm6dsx Fix alignment and data leak issues") Cc: stable Assisted-by: gregkh_clanker_t1000 Signed-off-by: Greg Kroah-Hartman Reviewed-by: David Lechner Signed-off-by: Jonathan Cameron --- drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c index 5b28a3ffcc3d..48291203d1cd 100644 --- a/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c +++ b/drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_buffer.c @@ -609,7 +609,7 @@ int st_lsm6dsx_read_tagged_fifo(struct st_lsm6dsx_hw *hw) * must be passed a buffer that is aligned to 8 bytes so * as to allow insertion of a naturally aligned timestamp. */ - u8 iio_buff[ST_LSM6DSX_IIO_BUFF_SIZE] __aligned(8); + u8 iio_buff[ST_LSM6DSX_IIO_BUFF_SIZE] __aligned(8) = { }; u8 tag; bool reset_ts = false; int i, err, read_len; From 474f8928d50b09f7dcf507049f08732640b88b49 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 9 Apr 2026 15:40:49 +0200 Subject: [PATCH 051/411] iio: imu: adis16550: fix stack leak in trigger handler MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit adis16550_trigger_handler() declares the scan data array on the stack without initializing it. The memcpy() at the bottom fills only the first 28 bytes (TEMP + 6 channels of GYRO/ACCEL data), and iio_push_to_buffers_with_timestamp() writes the s64 timestamp at the 8-byte-aligned offset 32. Bytes 28-31 remain uninitialized stack data which leaks to userspace on ever trigger. Fix this all by just zero-initializing the structure on the stack. Cc: Lars-Peter Clausen Cc: Michael Hennerich Cc: Jonathan Cameron Cc: David Lechner Cc: "Nuno Sá" Cc: Andy Shevchenko Fixes: e4570f4bb231 ("iio: imu: adis16550: align buffers for timestamp") Cc: stable Assisted-by: gregkh_clanker_t1000 Signed-off-by: Greg Kroah-Hartman Reviewed-by: David Lechner Signed-off-by: Jonathan Cameron --- drivers/iio/imu/adis16550.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/imu/adis16550.c b/drivers/iio/imu/adis16550.c index 1f2af506f4bd..75679612052f 100644 --- a/drivers/iio/imu/adis16550.c +++ b/drivers/iio/imu/adis16550.c @@ -836,7 +836,7 @@ static irqreturn_t adis16550_trigger_handler(int irq, void *p) u16 dummy; bool valid; struct iio_poll_func *pf = p; - __be32 data[ADIS16550_MAX_SCAN_DATA] __aligned(8); + __be32 data[ADIS16550_MAX_SCAN_DATA] __aligned(8) = { }; struct iio_dev *indio_dev = pf->indio_dev; struct adis16550 *st = iio_priv(indio_dev); struct adis *adis = iio_device_get_drvdata(indio_dev); From 5ace794c3ded38038a1f97f9ea26b9a8c835c111 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Fri, 10 Apr 2026 13:12:13 +0300 Subject: [PATCH 052/411] iio: adc: qcom-spmi-adc5-gen3: Fix off by one in adc5_gen3_get_fw_channel_data() The > in "if (chan > ADC5_MAX_CHANNEL)" should be >= to prevent an out of bound read of the adc->data->adc_chans[] array. Fixes: baff45179e90 ("iio: adc: Add support for QCOM PMIC5 Gen3 ADC") Signed-off-by: Dan Carpenter Reviewed-by: Konrad Dybcio Signed-off-by: Jonathan Cameron --- drivers/iio/adc/qcom-spmi-adc5-gen3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/adc/qcom-spmi-adc5-gen3.c b/drivers/iio/adc/qcom-spmi-adc5-gen3.c index f8168a14b907..48c793b18d11 100644 --- a/drivers/iio/adc/qcom-spmi-adc5-gen3.c +++ b/drivers/iio/adc/qcom-spmi-adc5-gen3.c @@ -482,7 +482,7 @@ static int adc5_gen3_get_fw_channel_data(struct adc5_chip *adc, sid = FIELD_GET(ADC5_GEN3_VIRTUAL_SID_MASK, chan); chan = FIELD_GET(ADC5_GEN3_CHANNEL_MASK, chan); - if (chan > ADC5_MAX_CHANNEL) + if (chan >= ADC5_MAX_CHANNEL) return dev_err_probe(dev, -EINVAL, "%s invalid channel number %d\n", name, chan); From f9bbd943c34a9ad60e593a4b99ce2394e4e2381b Mon Sep 17 00:00:00 2001 From: Salah Triki Date: Mon, 27 Apr 2026 21:12:38 +0100 Subject: [PATCH 053/411] iio: adc: mt6359: fix unchecked return value in mt6358_read_imp In mt6358_read_imp(), the variable val_v is passed to regmap_read() but the return value is not checked. If the read fails, val_v remains uninitialized and its random stack content is subsequently reported as a measurement result. Initialize val_v to zero to ensure a predictable value is reported in case of bus failure and to prevent potential stack data leakage. This also satisfies static analyzers that might otherwise flag the variable as used uninitialized. Fixes: 3587914bf61d ("iio: adc: Add support for MediaTek MT6357/8/9 Auxiliary ADC") Signed-off-by: Salah Triki Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/mt6359-auxadc.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/iio/adc/mt6359-auxadc.c b/drivers/iio/adc/mt6359-auxadc.c index 6b9ed9b1fde2..1d9724ef0983 100644 --- a/drivers/iio/adc/mt6359-auxadc.c +++ b/drivers/iio/adc/mt6359-auxadc.c @@ -497,6 +497,7 @@ static int mt6358_read_imp(struct mt6359_auxadc *adc_dev, return ret; /* Read the params before stopping */ + val_v = 0; regmap_read(regmap, reg_adc0 + (cinfo->imp_adc_num << 1), &val_v); mt6358_stop_imp_conv(adc_dev); From d0a228d903425e653f18a4341e60c0538afb6d41 Mon Sep 17 00:00:00 2001 From: Salah Triki Date: Mon, 27 Apr 2026 22:33:19 +0100 Subject: [PATCH 054/411] iio: dac: max5821: fix return value check in powerdown sync The function max5821_sync_powerdown_mode() returned the result of i2c_master_send() directly. If a partial transfer occurred, it would be incorrectly treated as a success by the caller. While the caller currently handles the positive return value of 2 as success, this patch refactors the function to return 0 on full success and -EIO on short writes. This ensures robust error handling for incomplete transfers and improves code maintainability by using sizeof(outbuf). Fixes: 472988972737 ("iio: add support of the max5821") Signed-off-by: Salah Triki Reviewed-by: Andy Shevchenko Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/dac/max5821.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/iio/dac/max5821.c b/drivers/iio/dac/max5821.c index e7e29359f8fe..dd4e35460195 100644 --- a/drivers/iio/dac/max5821.c +++ b/drivers/iio/dac/max5821.c @@ -90,6 +90,7 @@ static int max5821_sync_powerdown_mode(struct max5821_data *data, const struct iio_chan_spec *chan) { u8 outbuf[2]; + int ret; outbuf[0] = MAX5821_EXTENDED_COMMAND_MODE; @@ -103,7 +104,13 @@ static int max5821_sync_powerdown_mode(struct max5821_data *data, else outbuf[1] |= MAX5821_EXTENDED_POWER_UP; - return i2c_master_send(data->client, outbuf, 2); + ret = i2c_master_send(data->client, outbuf, sizeof(outbuf)); + if (ret < 0) + return ret; + if (ret != sizeof(outbuf)) + return -EIO; + + return 0; } static ssize_t max5821_write_dac_powerdown(struct iio_dev *indio_dev, From ba121d7582361fe74405f32724976aeff5c35177 Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Mon, 27 Apr 2026 19:26:31 +0800 Subject: [PATCH 055/411] iio: adc: meson-saradc: fix calibration buffer leak on error meson_sar_adc_temp_sensor_init() allocates a buffer with nvmem_cell_read(), but the old code leaked it if syscon_regmap_lookup_by_phandle() failed. Fix this by adding missing kfree(buf). Fixes: d6f2eac64403 ("iio: adc: meson: no devm for nvmem_cell_get") Signed-off-by: Felix Gu Signed-off-by: Jonathan Cameron --- drivers/iio/adc/meson_saradc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/iio/adc/meson_saradc.c b/drivers/iio/adc/meson_saradc.c index 23991a3612bd..000e39ca5c62 100644 --- a/drivers/iio/adc/meson_saradc.c +++ b/drivers/iio/adc/meson_saradc.c @@ -817,9 +817,11 @@ static int meson_sar_adc_temp_sensor_init(struct iio_dev *indio_dev) } priv->tsc_regmap = syscon_regmap_lookup_by_phandle(dev->of_node, "amlogic,hhi-sysctrl"); - if (IS_ERR(priv->tsc_regmap)) + if (IS_ERR(priv->tsc_regmap)) { + kfree(buf); return dev_err_probe(dev, PTR_ERR(priv->tsc_regmap), "failed to get amlogic,hhi-sysctrl regmap\n"); + } trimming_bits = priv->param->temperature_trimming_bits; trimming_mask = BIT(trimming_bits) - 1; From eedf7602fbd929e97e0c480da501dc7a34beb2a8 Mon Sep 17 00:00:00 2001 From: Sanjay Chitroda Date: Sun, 26 Apr 2026 14:47:04 +0530 Subject: [PATCH 056/411] iio: ssp_sensors: cancel delayed work_refresh on remove The work_refresh may still be pending or running when the device is removed, cancel the delayed work_refresh in remove path. Fixes: 50dd64d57eee ("iio: common: ssp_sensors: Add sensorhub driver") Signed-off-by: Sanjay Chitroda Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/common/ssp_sensors/ssp_dev.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/iio/common/ssp_sensors/ssp_dev.c b/drivers/iio/common/ssp_sensors/ssp_dev.c index da09c9f3ceb6..e2538a84c812 100644 --- a/drivers/iio/common/ssp_sensors/ssp_dev.c +++ b/drivers/iio/common/ssp_sensors/ssp_dev.c @@ -590,6 +590,7 @@ static void ssp_remove(struct spi_device *spi) ssp_clean_pending_list(data); free_irq(data->spi->irq, data); + cancel_delayed_work_sync(&data->work_refresh); timer_delete_sync(&data->wdt_timer); cancel_work_sync(&data->work_wdt); From ecae2ae606d493cf11457946436335bd0e726663 Mon Sep 17 00:00:00 2001 From: Rodrigo Alencar Date: Fri, 1 May 2026 10:14:54 +0100 Subject: [PATCH 057/411] iio: dac: ad5686: fix ref bit initialization for single-channel parts The reference bit position was ignored when writing the register at the probe() function (!!val was used). When such bit is 1, internal voltage reference is disabled so that an external one can be used. For multi-channel devices, bit 0 of the Internal Reference Setup command behaves the same way, so AD5686_REF_BIT_MSK is created. The issue exists since support for single-channel devices were first introduced. Fixes: be1b24d24541 ("iio:dac:ad5686: Add AD5691R/AD5692R/AD5693/AD5693R support") Reviewed-by: Andy Shevchenko Signed-off-by: Rodrigo Alencar Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/dac/ad5686.c | 6 +++--- drivers/iio/dac/ad5686.h | 1 + 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/iio/dac/ad5686.c b/drivers/iio/dac/ad5686.c index 4b18498aa074..b85d5c5a864b 100644 --- a/drivers/iio/dac/ad5686.c +++ b/drivers/iio/dac/ad5686.c @@ -509,7 +509,7 @@ int ad5686_probe(struct device *dev, break; case AD5686_REGMAP: cmd = AD5686_CMD_INTERNAL_REFER_SETUP; - ref_bit_msk = 0; + ref_bit_msk = AD5686_REF_BIT_MSK; break; case AD5693_REGMAP: cmd = AD5686_CMD_CONTROL_REG; @@ -520,9 +520,9 @@ int ad5686_probe(struct device *dev, return -EINVAL; } - val = (has_external_vref | ref_bit_msk); + val = has_external_vref ? ref_bit_msk : 0; - ret = st->write(st, cmd, 0, !!val); + ret = st->write(st, cmd, 0, val); if (ret) return ret; diff --git a/drivers/iio/dac/ad5686.h b/drivers/iio/dac/ad5686.h index e7d36bae3e59..36e16c5c4581 100644 --- a/drivers/iio/dac/ad5686.h +++ b/drivers/iio/dac/ad5686.h @@ -46,6 +46,7 @@ #define AD5310_REF_BIT_MSK BIT(8) #define AD5683_REF_BIT_MSK BIT(12) +#define AD5686_REF_BIT_MSK BIT(0) #define AD5693_REF_BIT_MSK BIT(12) /** From d01220ee5e43c65a206df827b39bf5cf5f7b9dce Mon Sep 17 00:00:00 2001 From: Rodrigo Alencar Date: Fri, 1 May 2026 10:14:55 +0100 Subject: [PATCH 058/411] iio: dac: ad5686: fix input raw value check Fix range check for input raw value, which is off by one, i.e., for a 10-bit DAC the max valid value is 1023, but 1 << 10 equals 1024, which passes the previous check, allowing an out-of-range write. The issue exists since the ad5686 driver was first introduced. Fixes: c2f37c8dcadc ("iio: dac: New driver for AD5686R, AD5685R, AD5684R Digital to analog converters") Reviewed-by: Andy Shevchenko Signed-off-by: Rodrigo Alencar Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/dac/ad5686.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/dac/ad5686.c b/drivers/iio/dac/ad5686.c index b85d5c5a864b..27878a6318ff 100644 --- a/drivers/iio/dac/ad5686.c +++ b/drivers/iio/dac/ad5686.c @@ -154,7 +154,7 @@ static int ad5686_write_raw(struct iio_dev *indio_dev, switch (mask) { case IIO_CHAN_INFO_RAW: - if (val > (1 << chan->scan_type.realbits) || val < 0) + if (val >= (1 << chan->scan_type.realbits) || val < 0) return -EINVAL; mutex_lock(&st->lock); From 6f5ed4f2c7c83f33344e0ba179f72a12e5dad4a4 Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Thu, 30 Apr 2026 21:29:06 +0800 Subject: [PATCH 059/411] iio: buffer: hw-consumer: fix use-after-free in error path MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In the err_put_buffers cleanup path of iio_hw_consumer_alloc(), the code was using list_for_each_entry() to iterate through buffers while calling iio_buffer_put() which can free the current buffer if refcount drops to 0. The list_for_each_entry() loop macro then evaluates buf->head.next to continue iteration, accessing the freed buffer. Fix this by using list_for_each_entry_safe(). Fixes: 48b66f8f936f ("iio: Add hardware consumer buffer support") Reported-by: sashiko Closes: https://sashiko.dev/#/patchset/20260427-iio_buf-v1-1-2bbdac844647%40gmail.com Signed-off-by: Felix Gu Reviewed-by: Andy Shevchenko Reviewed-by: Nuno Sá Reviewed-by: Maxwell Doose Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/buffer/industrialio-hw-consumer.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/iio/buffer/industrialio-hw-consumer.c b/drivers/iio/buffer/industrialio-hw-consumer.c index 24d7df603760..700528c9a0a4 100644 --- a/drivers/iio/buffer/industrialio-hw-consumer.c +++ b/drivers/iio/buffer/industrialio-hw-consumer.c @@ -85,7 +85,7 @@ static struct hw_consumer_buffer *iio_hw_consumer_get_buffer( */ struct iio_hw_consumer *iio_hw_consumer_alloc(struct device *dev) { - struct hw_consumer_buffer *buf; + struct hw_consumer_buffer *buf, *tmp; struct iio_hw_consumer *hwc; struct iio_channel *chan; int ret; @@ -116,7 +116,7 @@ struct iio_hw_consumer *iio_hw_consumer_alloc(struct device *dev) return hwc; err_put_buffers: - list_for_each_entry(buf, &hwc->buffers, head) + list_for_each_entry_safe(buf, tmp, &hwc->buffers, head) iio_buffer_put(&buf->buffer); iio_channel_release_all(hwc->channels); err_free_hwc: From ebd250c2581ec46c64c73fdfa918c9a7f757505e Mon Sep 17 00:00:00 2001 From: Kim Seer Paller Date: Tue, 5 May 2026 12:34:32 +0800 Subject: [PATCH 060/411] iio: dac: ad3530r: Fix AD3531/AD3531R powerdown mode strings The AD3531/AD3531R has different output operating modes from the AD3530/AD3530R. According to the AD3531/AD3531R datasheet, the powerdown modes are: 01: 500 Ohm output impedance 10: 3.85 kOhm output impedance 11: 16 kOhm output impedance The driver currently uses the AD3530R modes (1k, 7.7k, 32k) for all variants, which is incorrect for AD3531/AD3531R. Add AD3531R-specific powerdown mode strings and assign them to the AD3531/AD3531R chip variants. Fixes: 93583174a3df ("iio: dac: ad3530r: Add driver for AD3530R and AD3531R") Signed-off-by: Kim Seer Paller Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/dac/ad3530r.c | 54 +++++++++++++++++++++++++++++---------- 1 file changed, 40 insertions(+), 14 deletions(-) diff --git a/drivers/iio/dac/ad3530r.c b/drivers/iio/dac/ad3530r.c index b97b46090d80..d9db3226ecd6 100644 --- a/drivers/iio/dac/ad3530r.c +++ b/drivers/iio/dac/ad3530r.c @@ -105,6 +105,12 @@ static const char * const ad3530r_powerdown_modes[] = { "32kohm_to_gnd", }; +static const char * const ad3531r_powerdown_modes[] = { + "500ohm_to_gnd", + "3.85kohm_to_gnd", + "16kohm_to_gnd", +}; + static int ad3530r_get_powerdown_mode(struct iio_dev *indio_dev, const struct iio_chan_spec *chan) { @@ -133,6 +139,13 @@ static const struct iio_enum ad3530r_powerdown_mode_enum = { .set = ad3530r_set_powerdown_mode, }; +static const struct iio_enum ad3531r_powerdown_mode_enum = { + .items = ad3531r_powerdown_modes, + .num_items = ARRAY_SIZE(ad3531r_powerdown_modes), + .get = ad3530r_get_powerdown_mode, + .set = ad3530r_set_powerdown_mode, +}; + static ssize_t ad3530r_get_dac_powerdown(struct iio_dev *indio_dev, uintptr_t private, const struct iio_chan_spec *chan, @@ -276,7 +289,20 @@ static const struct iio_chan_spec_ext_info ad3530r_ext_info[] = { { } }; -#define AD3530R_CHAN(_chan) \ +static const struct iio_chan_spec_ext_info ad3531r_ext_info[] = { + { + .name = "powerdown", + .shared = IIO_SEPARATE, + .read = ad3530r_get_dac_powerdown, + .write = ad3530r_set_dac_powerdown, + }, + IIO_ENUM("powerdown_mode", IIO_SEPARATE, &ad3531r_powerdown_mode_enum), + IIO_ENUM_AVAILABLE("powerdown_mode", IIO_SHARED_BY_TYPE, + &ad3531r_powerdown_mode_enum), + { } +}; + +#define AD3530R_CHAN(_chan, _ext_info) \ { \ .type = IIO_VOLTAGE, \ .indexed = 1, \ @@ -284,25 +310,25 @@ static const struct iio_chan_spec_ext_info ad3530r_ext_info[] = { .output = 1, \ .info_mask_separate = BIT(IIO_CHAN_INFO_RAW) | \ BIT(IIO_CHAN_INFO_SCALE), \ - .ext_info = ad3530r_ext_info, \ + .ext_info = _ext_info, \ } static const struct iio_chan_spec ad3530r_channels[] = { - AD3530R_CHAN(0), - AD3530R_CHAN(1), - AD3530R_CHAN(2), - AD3530R_CHAN(3), - AD3530R_CHAN(4), - AD3530R_CHAN(5), - AD3530R_CHAN(6), - AD3530R_CHAN(7), + AD3530R_CHAN(0, ad3530r_ext_info), + AD3530R_CHAN(1, ad3530r_ext_info), + AD3530R_CHAN(2, ad3530r_ext_info), + AD3530R_CHAN(3, ad3530r_ext_info), + AD3530R_CHAN(4, ad3530r_ext_info), + AD3530R_CHAN(5, ad3530r_ext_info), + AD3530R_CHAN(6, ad3530r_ext_info), + AD3530R_CHAN(7, ad3530r_ext_info), }; static const struct iio_chan_spec ad3531r_channels[] = { - AD3530R_CHAN(0), - AD3530R_CHAN(1), - AD3530R_CHAN(2), - AD3530R_CHAN(3), + AD3530R_CHAN(0, ad3531r_ext_info), + AD3530R_CHAN(1, ad3531r_ext_info), + AD3530R_CHAN(2, ad3531r_ext_info), + AD3530R_CHAN(3, ad3531r_ext_info), }; static const struct ad3530r_chip_info ad3530_chip = { From 4701e471c16866e7aa8f5e6a3a6b0d31e097e2c9 Mon Sep 17 00:00:00 2001 From: Salah Triki Date: Tue, 5 May 2026 08:10:24 +0100 Subject: [PATCH 061/411] iio: temperature: tsys01: fix broken PROM checksum validation The current implementation of tsys01_crc_valid() incorrectly sums the first word (n_prom[0]) repeatedly instead of iterating over the 8 words retrieved from the PROM. This leads to a checksum mismatch and probe failure on hardware. According to the TSYS01 datasheet, the PROM consists of 8 words. A valid check must iterate through all 8 words to verify the integrity of the calibration data. The current driver only checks the first word 8 times. Note: This fix was identified during a code audit and is based on datasheet specifications. It has not been tested on real hardware. Fixes: 43e53407f680 ("Add tsys01 meas-spec driver support") Signed-off-by: Salah Triki Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/temperature/tsys01.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/temperature/tsys01.c b/drivers/iio/temperature/tsys01.c index 334bba6fdae6..104dd45598b0 100644 --- a/drivers/iio/temperature/tsys01.c +++ b/drivers/iio/temperature/tsys01.c @@ -119,7 +119,7 @@ static bool tsys01_crc_valid(u16 *n_prom) u8 sum = 0; for (cnt = 0; cnt < TSYS01_PROM_WORDS_NB; cnt++) - sum += ((n_prom[0] >> 8) + (n_prom[0] & 0xFF)); + sum += ((n_prom[cnt] >> 8) + (n_prom[cnt] & 0xFF)); return (sum == 0); } From 5237c3175cae5ab05f18878cec3301a04403859e Mon Sep 17 00:00:00 2001 From: Rodrigo Alencar Date: Tue, 5 May 2026 13:35:04 +0100 Subject: [PATCH 062/411] iio: dac: ad5686: acquire lock when doing powerdown control Protect access of pwr_down_mode and pwr_down_mask fields with existing mutex lock. Each channel exposes their own attributes for controlling powerdown modes and powerdown state. This fixes potential race conditions as those the write functions perform non-atomic read-modify-write operations to those pwr_down_* fields. This issue exists since the ad5686 driver was first introduced. Fixes: c2f37c8dcadc ("iio: dac: New driver for AD5686R, AD5685R, AD5684R Digital to analog converters") Signed-off-by: Rodrigo Alencar Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/dac/ad5686.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/iio/dac/ad5686.c b/drivers/iio/dac/ad5686.c index 27878a6318ff..2e443fcfeb39 100644 --- a/drivers/iio/dac/ad5686.c +++ b/drivers/iio/dac/ad5686.c @@ -30,6 +30,8 @@ static int ad5686_get_powerdown_mode(struct iio_dev *indio_dev, { struct ad5686_state *st = iio_priv(indio_dev); + guard(mutex)(&st->lock); + return ((st->pwr_down_mode >> (chan->channel * 2)) & 0x3) - 1; } @@ -39,6 +41,8 @@ static int ad5686_set_powerdown_mode(struct iio_dev *indio_dev, { struct ad5686_state *st = iio_priv(indio_dev); + guard(mutex)(&st->lock); + st->pwr_down_mode &= ~(0x3 << (chan->channel * 2)); st->pwr_down_mode |= ((mode + 1) << (chan->channel * 2)); @@ -57,6 +61,8 @@ static ssize_t ad5686_read_dac_powerdown(struct iio_dev *indio_dev, { struct ad5686_state *st = iio_priv(indio_dev); + guard(mutex)(&st->lock); + return sysfs_emit(buf, "%d\n", !!(st->pwr_down_mask & (0x3 << (chan->channel * 2)))); } @@ -77,6 +83,8 @@ static ssize_t ad5686_write_dac_powerdown(struct iio_dev *indio_dev, if (ret) return ret; + guard(mutex)(&st->lock); + if (readin) st->pwr_down_mask |= (0x3 << (chan->channel * 2)); else From 8aeaf25a85263a7a43357e16ad78ab969f6f8aeb Mon Sep 17 00:00:00 2001 From: Rodrigo Alencar Date: Tue, 5 May 2026 13:35:05 +0100 Subject: [PATCH 063/411] iio: dac: ad5686: fix powerdown control on dual-channel devices Fix powerdown control by using a proper bit shift for the powerdown mask values. During initialization, powerdown bits are initialized so that unused bits are set to 1 and the correct bit shift is used. Dual-channel devices use one-hot encoding in the address and that reflects on the position of the powerdown bits, which are not channel-index based for that case. Quad-channel devices also use one-hot encoding for the channel address but the result of log2(address) coincides with the channel index value. Mask as 0x3U is used rather than 0x3, because shift can reach value of 30 (last channel of a 16-channel device), which would mess with the sign bit. The issue was introduced when first adding support for dual-channel devices, which overlooked powerdown control differences. Fixes: 7dc8faeab3e3 ("iio: dac: ad5686: add support for AD5338R") Signed-off-by: Rodrigo Alencar Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/dac/ad5686.c | 40 ++++++++++++++++++++++++++++++---------- 1 file changed, 30 insertions(+), 10 deletions(-) diff --git a/drivers/iio/dac/ad5686.c b/drivers/iio/dac/ad5686.c index 2e443fcfeb39..a7213bc6b156 100644 --- a/drivers/iio/dac/ad5686.c +++ b/drivers/iio/dac/ad5686.c @@ -25,26 +25,37 @@ static const char * const ad5686_powerdown_modes[] = { "three_state" }; +static inline unsigned int ad5686_pd_mask_shift(const struct iio_chan_spec *chan) +{ + if (chan->channel == chan->address) + return chan->channel * 2; + + /* one-hot encoding is used in dual/quad channel devices */ + return __ffs(chan->address) * 2; +} + static int ad5686_get_powerdown_mode(struct iio_dev *indio_dev, const struct iio_chan_spec *chan) { + unsigned int shift = ad5686_pd_mask_shift(chan); struct ad5686_state *st = iio_priv(indio_dev); guard(mutex)(&st->lock); - return ((st->pwr_down_mode >> (chan->channel * 2)) & 0x3) - 1; + return ((st->pwr_down_mode >> shift) & 0x3U) - 1; } static int ad5686_set_powerdown_mode(struct iio_dev *indio_dev, const struct iio_chan_spec *chan, unsigned int mode) { + unsigned int shift = ad5686_pd_mask_shift(chan); struct ad5686_state *st = iio_priv(indio_dev); guard(mutex)(&st->lock); - st->pwr_down_mode &= ~(0x3 << (chan->channel * 2)); - st->pwr_down_mode |= ((mode + 1) << (chan->channel * 2)); + st->pwr_down_mode &= ~(0x3U << shift); + st->pwr_down_mode |= (mode + 1) << shift; return 0; } @@ -59,12 +70,12 @@ static const struct iio_enum ad5686_powerdown_mode_enum = { static ssize_t ad5686_read_dac_powerdown(struct iio_dev *indio_dev, uintptr_t private, const struct iio_chan_spec *chan, char *buf) { + unsigned int shift = ad5686_pd_mask_shift(chan); struct ad5686_state *st = iio_priv(indio_dev); guard(mutex)(&st->lock); - return sysfs_emit(buf, "%d\n", !!(st->pwr_down_mask & - (0x3 << (chan->channel * 2)))); + return sysfs_emit(buf, "%d\n", !!(st->pwr_down_mask & (0x3U << shift))); } static ssize_t ad5686_write_dac_powerdown(struct iio_dev *indio_dev, @@ -86,9 +97,9 @@ static ssize_t ad5686_write_dac_powerdown(struct iio_dev *indio_dev, guard(mutex)(&st->lock); if (readin) - st->pwr_down_mask |= (0x3 << (chan->channel * 2)); + st->pwr_down_mask |= 0x3U << ad5686_pd_mask_shift(chan); else - st->pwr_down_mask &= ~(0x3 << (chan->channel * 2)); + st->pwr_down_mask &= ~(0x3U << ad5686_pd_mask_shift(chan)); switch (st->chip_info->regmap_type) { case AD5310_REGMAP: @@ -468,7 +479,7 @@ int ad5686_probe(struct device *dev, { struct ad5686_state *st; struct iio_dev *indio_dev; - unsigned int val, ref_bit_msk; + unsigned int val, ref_bit_msk, shift; bool has_external_vref; u8 cmd; int ret, i; @@ -492,9 +503,18 @@ int ad5686_probe(struct device *dev, has_external_vref = ret != -ENODEV; st->vref_mv = has_external_vref ? ret / 1000 : st->chip_info->int_vref_mv; + /* Initialize masks to all ones provided the max shift (last channel) */ + shift = ad5686_pd_mask_shift(&st->chip_info->channels[st->chip_info->num_channels - 1]); + st->pwr_down_mask = GENMASK(shift + 1, 0); + st->pwr_down_mode = GENMASK(shift + 1, 0); + /* Set all the power down mode for all channels to 1K pulldown */ - for (i = 0; i < st->chip_info->num_channels; i++) - st->pwr_down_mode |= (0x01 << (i * 2)); + for (i = 0; i < st->chip_info->num_channels; i++) { + shift = ad5686_pd_mask_shift(&st->chip_info->channels[i]); + st->pwr_down_mask &= ~(0x3U << shift); /* powered up state */ + st->pwr_down_mode &= ~(0x3U << shift); + st->pwr_down_mode |= 0x01U << shift; + } indio_dev->name = name; indio_dev->info = &ad5686_info; From 6bdc3023d62ed5c7d591f0eb27a5adb37fb892ae Mon Sep 17 00:00:00 2001 From: David Carlier Date: Tue, 5 May 2026 14:37:48 +0100 Subject: [PATCH 064/411] iio: gyro: itg3200: fix i2c read into the wrong stack location itg3200_read_all_channels() takes `__be16 *buf' as a parameter and fills the i2c_msg destination as `(char *)&buf'. Since `buf' is the parameter (a pointer), `&buf' is the address of the local pointer slot on the stack of itg3200_read_all_channels(), not the address of the caller's scan buffer. The (char *) cast hides the type mismatch. i2c_transfer() therefore writes ITG3200_SCAN_ELEMENTS * sizeof(s16) = 8 bytes into the parameter's stack slot, which is discarded when the function returns. The caller's scan buffer in itg3200_trigger_handler() is never written to, so iio_push_to_buffers_with_timestamp() pushes uninitialised stack contents to userspace via /dev/iio:deviceX every scan -- both a functional bug (no actual gyroscope or temperature data is delivered through the triggered buffer) and an information leak. The non-buffered read_raw() path is unaffected: it goes through itg3200_read_reg_s16() which uses `&out' on a local s16 value, where that is correct. Drop the spurious `&' so the i2c read writes into the caller's buffer. Fixes: 9dbf091da080 ("iio: gyro: Add itg3200") Cc: stable@vger.kernel.org Signed-off-by: David Carlier Reviewed-by: Andy Shevchenko Signed-off-by: Jonathan Cameron --- drivers/iio/gyro/itg3200_buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iio/gyro/itg3200_buffer.c b/drivers/iio/gyro/itg3200_buffer.c index cf97adfa9727..87efa2c74ca4 100644 --- a/drivers/iio/gyro/itg3200_buffer.c +++ b/drivers/iio/gyro/itg3200_buffer.c @@ -34,7 +34,7 @@ static int itg3200_read_all_channels(struct i2c_client *i2c, __be16 *buf) .addr = i2c->addr, .flags = i2c->flags | I2C_M_RD, .len = ITG3200_SCAN_ELEMENTS * sizeof(s16), - .buf = (char *)&buf, + .buf = (char *)buf, }, }; From 422b5bbf333f75fb486855ad0eedc23cf21f3277 Mon Sep 17 00:00:00 2001 From: Salah Triki Date: Thu, 7 May 2026 20:07:51 +0100 Subject: [PATCH 065/411] iio: adc: viperboard: Fix error handling in vprbrd_iio_read_raw MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The driver proceeds to the reception phase even if the preceding transmission fails. This uses a goto error label for an early bail out and ensures the mutex is properly unlocked in case of failure. Fixes: ffd8a6e7a778 ("iio: adc: Add viperboard adc driver") Signed-off-by: Salah Triki Reviewed-by: Joshua Crofts Reviewed-by: Maxwell Doose Reviewed-by: Nuno Sá Cc: Signed-off-by: Jonathan Cameron --- drivers/iio/adc/viperboard_adc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/iio/adc/viperboard_adc.c b/drivers/iio/adc/viperboard_adc.c index 9bb0b83c8f67..6efe1c618ef7 100644 --- a/drivers/iio/adc/viperboard_adc.c +++ b/drivers/iio/adc/viperboard_adc.c @@ -70,8 +70,10 @@ static int vprbrd_iio_read_raw(struct iio_dev *iio_dev, VPRBRD_USB_TYPE_OUT, 0x0000, 0x0000, admsg, sizeof(struct vprbrd_adc_msg), VPRBRD_USB_TIMEOUT_MS); if (ret != sizeof(struct vprbrd_adc_msg)) { - dev_err(&iio_dev->dev, "usb send error on adc read\n"); + mutex_unlock(&vb->lock); error = -EREMOTEIO; + dev_err(&iio_dev->dev, "usb send error on adc read\n"); + goto error; } ret = usb_control_msg(vb->usb_dev, From 79d8be262377f7112cfa3088dfc4142d5a2533f3 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 15 May 2026 11:45:31 -0400 Subject: [PATCH 066/411] xfrm: ah: use skb_to_full_sk in async output callbacks When AH output is offloaded to an asynchronous crypto provider (hardware accelerators such as AMD CCP, or a forced-async software shim used for testing), the digest completion fires ah_output_done() / ah6_output_done() on a workqueue. The egress skb at that point may have been originated by a TCP listener sending a SYN-ACK, which sets skb->sk to a request_sock via skb_set_owner_edemux(); it may also have been originated by an inet_timewait_sock retransmit. Neither is a full struct sock, and passing the raw skb->sk to xfrm_output_resume() then forwards a non-full socket through the rest of the xfrm output chain. xfrm_output_resume() and its downstream consumers expect a full sk where they dereference at all. The natural egress path through ah_output_done() does not crash today because the consumers that read past sock_common are either gated by sk_fullsock() or short-circuit on flags that are clear on a fresh request_sock; an exhaustive walk of the 50 most plausible consumers under sch_fq, dev_queue_xmit, netfilter, tc-egress and cgroup-egress BPF found no current unguarded deref. The bug is still a real type confusion that future consumer changes could turn into a memory-corruption primitive. This is the same bug class fixed for ESP in commit 1620c88887b1 ("xfrm: Fix the usage of skb->sk"). Apply the analogous fix to AH: convert skb->sk to a full socket pointer (or NULL) via skb_to_full_sk() before handing it to xfrm_output_resume(). The same async AH callbacks were touched recently for an independent ESN-related ICV layout bug in commit ec54093e6a8f ("xfrm: ah: account for ESN high bits in async callbacks"); the sk type-confusion addressed here is orthogonal. This patch is part of an ongoing audit of the AH callback paths; an ah_output ihl-validation hardening series is also currently under review on netdev. Reproduced under UML + KASAN + lockdep with a forced-async hmac(sha1) shim that registers at priority 9999 and wraps the sync in-tree hmac-sha1-lib. With the shim loaded, ah_output_done runs on every SYN-ACK egress through a transport-mode AH SA and skb->sk arrives as a request_sock (TCP_NEW_SYN_RECV); after this patch, xfrm_output_resume() receives the listener (the result of sk_to_full_sk()) and consumer derefs land on full-sock fields as intended. Fixes: 9ab1265d5231 ("xfrm: Use actual socket sk instead of skb socket for xfrm_output_resume") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Signed-off-by: Steffen Klassert --- net/ipv4/ah4.c | 2 +- net/ipv6/ah6.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c index 4366cbac3f06..6fd642d2278d 100644 --- a/net/ipv4/ah4.c +++ b/net/ipv4/ah4.c @@ -143,7 +143,7 @@ static void ah_output_done(void *data, int err) } kfree(AH_SKB_CB(skb)->tmp); - xfrm_output_resume(skb->sk, skb, err); + xfrm_output_resume(skb_to_full_sk(skb), skb, err); } static int ah_output(struct xfrm_state *x, struct sk_buff *skb) diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index de1e68199a01..76f7a2de9108 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -337,7 +337,7 @@ static void ah6_output_done(void *data, int err) ah6_restore_hdrs(top_iph, iph_ext, extlen); kfree(AH_SKB_CB(skb)->tmp); - xfrm_output_resume(skb->sk, skb, err); + xfrm_output_resume(skb_to_full_sk(skb), skb, err); } static int ah6_output(struct xfrm_state *x, struct sk_buff *skb) From b02900c85a6423cf9b3dcc6b47bf060c85075e69 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Tue, 12 May 2026 13:14:59 +0300 Subject: [PATCH 067/411] usb: typec: tipd: Fix error code in tps6598x_probe() Set the error code on these two error paths. The existing code returns success. Fixes: 77ed2f4538da ("usb: typec: tipd: Use read_power_status function in probe") Fixes: 04041fd7d6ec ("usb: typec: tipd: Read data status in probe and cache its value") Cc: stable Signed-off-by: Dan Carpenter Reviewed-by: Heikki Krogerus Link: https://patch.msgid.link/agL9o7wUK1dOVBTy@stanley.mountain Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tipd/core.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/usb/typec/tipd/core.c b/drivers/usb/typec/tipd/core.c index 43faec794b95..d0b769333bd9 100644 --- a/drivers/usb/typec/tipd/core.c +++ b/drivers/usb/typec/tipd/core.c @@ -1835,6 +1835,7 @@ static int tps6598x_probe(struct i2c_client *client) goto err_role_put; if (status & TPS_STATUS_PLUG_PRESENT) { + ret = -EINVAL; if (!tps6598x_read_power_status(tps)) goto err_unregister_port; if (!tps->data->read_data_status(tps)) From 6c5dbc104dadd79fc2923497c20bae759a18758c Mon Sep 17 00:00:00 2001 From: Jeremy Erazo Date: Tue, 12 May 2026 16:05:30 +0000 Subject: [PATCH 068/411] usb: gadget: composite: fix integer underflow in WebUSB GET_URL handling The WebUSB GET_URL handler in composite_setup() narrows landing_page_length to fit the host-supplied wLength using landing_page_length = w_length - WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH + landing_page_offset; If wLength is smaller than WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH the unsigned subtraction wraps, and the subsequent memcpy(url_descriptor->URL, cdev->landing_page + landing_page_offset, landing_page_length - landing_page_offset); ends up copying close to UINT_MAX bytes from cdev->landing_page into cdev->req->buf. KASAN reports a slab-out-of-bounds in composite_setup on the kmalloc-2k gadget_info allocation, and FORTIFY_SOURCE traps the memcpy as a 4294967293-byte field-spanning write into url_descriptor->URL (size 252). A USB host can reach this from a single SETUP packet against any gadget that has webusb/use=1 and a landingPage configured. Handle the small-wLength case before the math: when the host requested fewer bytes than the URL descriptor header, only the header is meaningful and no URL bytes need to be copied. Setting landing_page_length to landing_page_offset makes the existing memcpy a no-op and leaves the descriptor returned to the host unchanged for all larger wLength values. Fixes: 93c473948c58 ("usb: gadget: add WebUSB landing page support") Cc: stable Signed-off-by: Jeremy Erazo Link: https://patch.msgid.link/20260512160530.352318-1-mendozayt13@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/composite.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c index a902184bdf82..dc3664374596 100644 --- a/drivers/usb/gadget/composite.c +++ b/drivers/usb/gadget/composite.c @@ -2172,7 +2172,10 @@ composite_setup(struct usb_gadget *gadget, const struct usb_ctrlrequest *ctrl) sizeof(url_descriptor->URL) - WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH + landing_page_offset); - if (w_length < WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH + landing_page_length) + if (w_length < WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH) + landing_page_length = landing_page_offset; + else if (w_length < + WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH + landing_page_length) landing_page_length = w_length - WEBUSB_URL_DESCRIPTOR_HEADER_LENGTH + landing_page_offset; From d7486952bf74e546ee3748fb14b2d07881fa6273 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 14 May 2026 19:10:06 +0200 Subject: [PATCH 069/411] usb: typec: ucsi: ccg: reject firmware images without a ':' record header do_flash() locates the first .cyacd record with p = strnchr(fw->data, fw->size, ':'); while (p < eof) { s = strnchr(p + 1, eof - p - 1, ':'); ... } If the firmware image contains no ':' byte, strnchr() returns NULL. NULL compares less than the valid kernel pointer eof, so the loop body runs and strnchr() is called with p + 1 == (void *)1 and a length of roughly (unsigned long)eof, causing a wonderful crash. The not_signed_fw fallthrough earlier in do_flash() and the chip-state branches in ccg_fw_update_needed() allow an unsigned blob to reach this loop, so a root user who can place a crafted file under /lib/firmware and write the do_flash sysfs attribute can trigger the oops. Bail out with -EINVAL when the initial strnchr() returns NULL. Assisted-by: gkh_clanker_t1000 Cc: stable Cc: Heikki Krogerus Reviewed-by: Heikki Krogerus Signed-off-by: Greg Kroah-Hartman Link: https://patch.msgid.link/2026051405-posture-shrill-7884@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/ucsi_ccg.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/usb/typec/ucsi/ucsi_ccg.c b/drivers/usb/typec/ucsi/ucsi_ccg.c index 199799b319c2..4463c1ae96bd 100644 --- a/drivers/usb/typec/ucsi/ucsi_ccg.c +++ b/drivers/usb/typec/ucsi/ucsi_ccg.c @@ -1243,6 +1243,11 @@ static int do_flash(struct ucsi_ccg *uc, enum enum_flash_mode mode) *****************************************************************/ p = strnchr(fw->data, fw->size, ':'); + if (!p) { + dev_err(dev, "Bad FW format: no ':' record header found\n"); + err = -EINVAL; + goto release_mem; + } while (p < eof) { s = strnchr(p + 1, eof - p - 1, ':'); From d1e280334b7f0a1df441e08bd1f6a1bcc36b3bbb Mon Sep 17 00:00:00 2001 From: Michal Pecio Date: Mon, 18 May 2026 07:31:21 +0200 Subject: [PATCH 070/411] usb: core: Fix SuperSpeed root hub wMaxPacketSize There is no good reason to have wBytesPerInterval < wMaxPacketSize - either one is too low or the other too high, and we may want to warn about such descriptors. Start with cleaning up our own root hubs. USB 3.2 section 10.15.1 sets wMaxPacketSize and wBytesPerInterval of SuperSpeed hub status endpoints at 2 bytes, so reduce wMaxPacketSize from its former value of 4, which was derived from USB 2.0 spec and the kernel's USB_MAXCHILDREN limit. They don't apply because USB 3.2 10.15.2.1 specifies SuperSpeed hubs to have up to 15 ports. Suggested-by: Mathias Nyman Signed-off-by: Michal Pecio Link: https://patch.msgid.link/20260518073121.7bc1da0f.michal.pecio@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/hcd.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c index 89221f1ce769..b181b43a35dc 100644 --- a/drivers/usb/core/hcd.c +++ b/drivers/usb/core/hcd.c @@ -328,9 +328,7 @@ static const u8 ss_rh_config_descriptor[] = { USB_DT_ENDPOINT, /* __u8 ep_bDescriptorType; Endpoint */ 0x81, /* __u8 ep_bEndpointAddress; IN Endpoint 1 */ 0x03, /* __u8 ep_bmAttributes; Interrupt */ - /* __le16 ep_wMaxPacketSize; 1 + (MAX_ROOT_PORTS / 8) - * see hub.c:hub_configure() for details. */ - (USB_MAXCHILDREN + 1 + 7) / 8, 0x00, + 0x02, 0x00, /* __le16 ep_wMaxPacketSize; 2 bytes per USB3 10.15.1 */ 0x0c, /* __u8 ep_bInterval; (256ms -- usb 2.0 spec) */ /* one SuperSpeed endpoint companion descriptor */ From 727d045d064b7c9a24db3bce9c0485a382cb768b Mon Sep 17 00:00:00 2001 From: Michal Pecio Date: Mon, 18 May 2026 07:32:07 +0200 Subject: [PATCH 071/411] usb: core: Fix up Interrupt IN endpoints with bogus wBytesPerInterval Tao Xue found that some common devices violate USB 3.x section 9.6.7 by reporting wBytesPerInterval lower than the size of packets they actually send. I confirmed that AX88179 may set it to 0 and RTL8153 CDC configuration sets it to 8 but sends both 8 and 16 byte packets: S Ii:11:007:3 -115:128 16 < C Ii:11:007:3 0:128 8 = a1000000 01000000 S Ii:11:007:3 -115:128 16 < C Ii:11:007:3 0:128 16 = a12a0000 01000800 00000000 00000000 Most xHCI host controllers neglect interrupt bandwidth reservations and let such devices exceed theirs, some fail the URB with EOVERFLOW. Assume that wBytesPerInterval lower than wMaxPacketSize is bogus and increase it to the worst case maximum on interrupt IN endpoints. This solves xHCI problems and appears to have no other effect. Interrupt transfers are not limited to one interval and drivers submit URBs of class defined size without looking at wBytesPerInterval. Any multi- interval transfer is considered terminated by a packet shorter than wMaxPacketSize regardless of wBytesPerInterval - see USB3 8.10.3. Stay in spec on OUT endpoints and isochronous. No buggy devices are known and we don't want to risk sending more data than the device is prepared to handle or confusing isoc drivers regarding altsetting capacities guaranteed by the device itself. And don't complain when wMaxPacketSize <= wBytesPerInterval < wMaxPacketSize * (bMaxBurst+1) because enabling this seems to be the exact goal of the spec. Reported-and-tested-by: Tao Xue Closes: https://lore.kernel.org/linux-usb/20260402021400.28853-1-xuetao09@huawei.com/ Cc: stable@vger.kernel.org Signed-off-by: Michal Pecio Link: https://patch.msgid.link/20260518073207.5b7d26e7.michal.pecio@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/config.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c index 417140b012bb..d9171bf7bc88 100644 --- a/drivers/usb/core/config.c +++ b/drivers/usb/core/config.c @@ -191,7 +191,14 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno, (desc->bMaxBurst + 1); else max_tx = 999999; - if (le16_to_cpu(desc->wBytesPerInterval) > max_tx) { + /* + * wBytesPerInterval > max_tx is bogus, but USB3 spec doesn't forbid the opposite. + * Experience shows that wBytesPerInterval < wMaxPacketSize on common interrupt IN + * endpoints is usually bogus too, and recent HCs enforce interrupt BW limits. + */ + if (le16_to_cpu(desc->wBytesPerInterval) > max_tx || + (le16_to_cpu(desc->wBytesPerInterval) < usb_endpoint_maxp(&ep->desc) && + usb_endpoint_is_int_in(&ep->desc))) { dev_notice(ddev, "%s endpoint with wBytesPerInterval of %d in " "config %d interface %d altsetting %d ep %d: " "setting to %d\n", From af8c5aa7a9c6f503d81f103d7ab4f8d759521de3 Mon Sep 17 00:00:00 2001 From: Michal Pecio Date: Mon, 18 May 2026 07:32:58 +0200 Subject: [PATCH 072/411] usb: core: Clean up SuperSpeed/eUSB2 descriptor validation logging Core usually prints endpoint addresses with 0x%X format. Change this code to use it too, instead of just %d. Particularly for IN, 0x83 seems more readable than 131. While at that, fix checkpatch warnings about multi-line quoted strings, as well as missing or doubled whitespace in those strings. Signed-off-by: Michal Pecio Link: https://patch.msgid.link/20260518073258.6532bdd5.michal.pecio@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/config.c | 37 +++++++++++-------------------------- 1 file changed, 11 insertions(+), 26 deletions(-) diff --git a/drivers/usb/core/config.c b/drivers/usb/core/config.c index d9171bf7bc88..45e20c6d76c0 100644 --- a/drivers/usb/core/config.c +++ b/drivers/usb/core/config.c @@ -56,8 +56,7 @@ static void usb_parse_ssp_isoc_endpoint_companion(struct device *ddev, desc = (struct usb_ssp_isoc_ep_comp_descriptor *) buffer; if (size < USB_DT_SSP_ISOC_EP_COMP_SIZE || desc->bDescriptorType != USB_DT_SSP_ISOC_ENDPOINT_COMP) { - dev_notice(ddev, "Invalid SuperSpeedPlus isoc endpoint companion" - "for config %d interface %d altsetting %d ep %d.\n", + dev_notice(ddev, "Invalid SuperSpeedPlus isoc endpoint companion for config %d interface %d altsetting %d ep 0x%X.\n", cfgno, inum, asnum, ep->desc.bEndpointAddress); return; } @@ -91,7 +90,7 @@ static void usb_parse_eusb2_isoc_endpoint_companion(struct device *ddev, size -= h->bLength; } - dev_notice(ddev, "No eUSB2 isoc ep %d companion for config %d interface %d altsetting %d\n", + dev_notice(ddev, "No eUSB2 isoc ep 0x%X companion for config %d interface %d altsetting %d\n", ep->desc.bEndpointAddress, cfgno, inum, asnum); } @@ -115,9 +114,7 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno, } if (desc->bDescriptorType != USB_DT_SS_ENDPOINT_COMP) { - dev_notice(ddev, "No SuperSpeed endpoint companion for config %d " - " interface %d altsetting %d ep %d: " - "using minimum values\n", + dev_notice(ddev, "No SuperSpeed endpoint companion for config %d interface %d altsetting %d ep 0x%X: using minimum values\n", cfgno, inum, asnum, ep->desc.bEndpointAddress); /* Fill in some default values. @@ -141,42 +138,32 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno, /* Check the various values */ if (usb_endpoint_xfer_control(&ep->desc) && desc->bMaxBurst != 0) { - dev_notice(ddev, "Control endpoint with bMaxBurst = %d in " - "config %d interface %d altsetting %d ep %d: " - "setting to zero\n", desc->bMaxBurst, - cfgno, inum, asnum, ep->desc.bEndpointAddress); + dev_notice(ddev, "Control endpoint with bMaxBurst = %d in config %d interface %d altsetting %d ep 0x%X: setting to zero\n", + desc->bMaxBurst, cfgno, inum, asnum, ep->desc.bEndpointAddress); ep->ss_ep_comp.bMaxBurst = 0; } else if (desc->bMaxBurst > 15) { - dev_notice(ddev, "Endpoint with bMaxBurst = %d in " - "config %d interface %d altsetting %d ep %d: " - "setting to 15\n", desc->bMaxBurst, - cfgno, inum, asnum, ep->desc.bEndpointAddress); + dev_notice(ddev, "Endpoint with bMaxBurst = %d in config %d interface %d altsetting %d ep 0x%X: setting to 15\n", + desc->bMaxBurst, cfgno, inum, asnum, ep->desc.bEndpointAddress); ep->ss_ep_comp.bMaxBurst = 15; } if ((usb_endpoint_xfer_control(&ep->desc) || usb_endpoint_xfer_int(&ep->desc)) && desc->bmAttributes != 0) { - dev_notice(ddev, "%s endpoint with bmAttributes = %d in " - "config %d interface %d altsetting %d ep %d: " - "setting to zero\n", + dev_notice(ddev, "%s endpoint with bmAttributes = %d in config %d interface %d altsetting %d ep 0x%X: setting to zero\n", usb_endpoint_xfer_control(&ep->desc) ? "Control" : "Bulk", desc->bmAttributes, cfgno, inum, asnum, ep->desc.bEndpointAddress); ep->ss_ep_comp.bmAttributes = 0; } else if (usb_endpoint_xfer_bulk(&ep->desc) && desc->bmAttributes > 16) { - dev_notice(ddev, "Bulk endpoint with more than 65536 streams in " - "config %d interface %d altsetting %d ep %d: " - "setting to max\n", + dev_notice(ddev, "Bulk endpoint with more than 65536 streams in config %d interface %d altsetting %d ep 0x%X: setting to max\n", cfgno, inum, asnum, ep->desc.bEndpointAddress); ep->ss_ep_comp.bmAttributes = 16; } else if (usb_endpoint_xfer_isoc(&ep->desc) && !USB_SS_SSP_ISOC_COMP(desc->bmAttributes) && USB_SS_MULT(desc->bmAttributes) > 3) { - dev_notice(ddev, "Isoc endpoint has Mult of %d in " - "config %d interface %d altsetting %d ep %d: " - "setting to 3\n", + dev_notice(ddev, "Isoc endpoint has Mult of %d in config %d interface %d altsetting %d ep 0x%X: setting to 3\n", USB_SS_MULT(desc->bmAttributes), cfgno, inum, asnum, ep->desc.bEndpointAddress); ep->ss_ep_comp.bmAttributes = 2; @@ -199,9 +186,7 @@ static void usb_parse_ss_endpoint_companion(struct device *ddev, int cfgno, if (le16_to_cpu(desc->wBytesPerInterval) > max_tx || (le16_to_cpu(desc->wBytesPerInterval) < usb_endpoint_maxp(&ep->desc) && usb_endpoint_is_int_in(&ep->desc))) { - dev_notice(ddev, "%s endpoint with wBytesPerInterval of %d in " - "config %d interface %d altsetting %d ep %d: " - "setting to %d\n", + dev_notice(ddev, "%s endpoint with wBytesPerInterval of %d in config %d interface %d altsetting %d ep 0x%X: setting to %d\n", usb_endpoint_xfer_isoc(&ep->desc) ? "Isoc" : "Int", le16_to_cpu(desc->wBytesPerInterval), cfgno, inum, asnum, ep->desc.bEndpointAddress, From f23bf992d65a42007c517b060ca35cebdea3525a Mon Sep 17 00:00:00 2001 From: Carl Lee Date: Sat, 16 May 2026 19:55:18 +0800 Subject: [PATCH 073/411] nfc: nxp-nci: i2c: use rising-edge IRQ on ACPI systems Some ACPI-based platforms report incorrect IRQ trigger types (e.g. IRQF_TRIGGER_HIGH), which can lead to interrupt storms. Use the historically working rising-edge trigger on ACPI systems to avoid this regression. Device Tree-based systems continue to use the firmware-provided trigger type. Fixes: 57be33f85e36 ("nfc: nxp-nci: remove interrupt trigger type") Signed-off-by: Carl Lee Tested-by: Bartosz Golaszewski Reviewed-by: Bartosz Golaszewski Reviewed-by: Mark Pearson Tested-by: Mark Pearson Tested-by: Luca Stefani Link: https://patch.msgid.link/20260516-nfc-nxp-nci-i2c-restore-irq-trigger-fallback-v3-1-37ba4b6e9086@amd.com Signed-off-by: David Heidelberg --- drivers/nfc/nxp-nci/i2c.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/drivers/nfc/nxp-nci/i2c.c b/drivers/nfc/nxp-nci/i2c.c index b3d34433bd14..a6c08175d9dd 100644 --- a/drivers/nfc/nxp-nci/i2c.c +++ b/drivers/nfc/nxp-nci/i2c.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -267,6 +268,7 @@ static int nxp_nci_i2c_probe(struct i2c_client *client) { struct device *dev = &client->dev; struct nxp_nci_i2c_phy *phy; + unsigned long irqflags; int r; if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) { @@ -303,9 +305,26 @@ static int nxp_nci_i2c_probe(struct i2c_client *client) if (r < 0) return r; + /* + * ACPI platforms may report incorrect IRQ trigger types + * (e.g. level-high), which can lead to interrupt storms. + * + * Use the historically stable rising-edge trigger for ACPI devices. + * + * On non-ACPI systems (e.g. Device Tree), prefer the firmware- + * provided trigger type, falling back to rising-edge if not set. + */ + if (ACPI_COMPANION(dev)) { + irqflags = IRQF_TRIGGER_RISING; + } else { + irqflags = irq_get_trigger_type(client->irq); + if (!irqflags) + irqflags = IRQF_TRIGGER_RISING; + } + r = request_threaded_irq(client->irq, NULL, nxp_nci_i2c_irq_thread_fn, - IRQF_ONESHOT, + irqflags | IRQF_ONESHOT, NXP_NCI_I2C_DRIVER_NAME, phy); if (r < 0) nfc_err(&client->dev, "Unable to register IRQ handler\n"); From 9785df3fd67083ac10f6bde83a316286044a66f1 Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 7 Apr 2026 14:45:22 +0800 Subject: [PATCH 074/411] iommu/vt-d: Simplify calculate_psi_aligned_address() This is doing far too much math for the simple task of finding a power of 2 that fully spans the given range. Use fls directly on the xor which computes the common binary prefix. Signed-off-by: Jason Gunthorpe Reviewed-by: Kevin Tian Link: https://lore.kernel.org/r/0-v2-895748900b39+5303-iommupt_inv_vtd_jgg@nvidia.com Signed-off-by: Lu Baolu Signed-off-by: Joerg Roedel --- drivers/iommu/intel/cache.c | 49 ++++++++++++++----------------------- 1 file changed, 18 insertions(+), 31 deletions(-) diff --git a/drivers/iommu/intel/cache.c b/drivers/iommu/intel/cache.c index be8410f0e841..fdc88817709f 100644 --- a/drivers/iommu/intel/cache.c +++ b/drivers/iommu/intel/cache.c @@ -254,37 +254,29 @@ void cache_tag_unassign_domain(struct dmar_domain *domain, } static unsigned long calculate_psi_aligned_address(unsigned long start, - unsigned long end, - unsigned long *_mask) + unsigned long last, + unsigned long *size_order) { - unsigned long pages = aligned_nrpages(start, end - start + 1); - unsigned long aligned_pages = __roundup_pow_of_two(pages); - unsigned long bitmask = aligned_pages - 1; - unsigned long mask = ilog2(aligned_pages); - unsigned long pfn = IOVA_PFN(start); - - /* - * PSI masks the low order bits of the base address. If the - * address isn't aligned to the mask, then compute a mask value - * needed to ensure the target range is flushed. - */ - if (unlikely(bitmask & pfn)) { - unsigned long end_pfn = pfn + pages - 1, shared_bits; + unsigned int sz_lg2; + /* Compute a sz_lg2 that spans start and last */ + start &= GENMASK(BITS_PER_LONG - 1, VTD_PAGE_SHIFT); + sz_lg2 = fls_long(start ^ last); + if (sz_lg2 <= 12) { + *size_order = 0; + return start; + } + if (unlikely(sz_lg2 >= BITS_PER_LONG)) { /* - * Since end_pfn <= pfn + bitmask, the only way bits - * higher than bitmask can differ in pfn and end_pfn is - * by carrying. This means after masking out bitmask, - * high bits starting with the first set bit in - * shared_bits are all equal in both pfn and end_pfn. + * MAX_AGAW_PFN_WIDTH triggers full invalidation in all + * downstream users. */ - shared_bits = ~(pfn ^ end_pfn) & ~bitmask; - mask = shared_bits ? __ffs(shared_bits) : MAX_AGAW_PFN_WIDTH; + *size_order = MAX_AGAW_PFN_WIDTH; + return 0; } - *_mask = mask; - - return ALIGN_DOWN(start, VTD_PAGE_SIZE << mask); + *size_order = sz_lg2 - VTD_PAGE_SHIFT; + return start & GENMASK(BITS_PER_LONG - 1, sz_lg2); } static void qi_batch_flush_descs(struct intel_iommu *iommu, struct qi_batch *batch) @@ -441,12 +433,7 @@ void cache_tag_flush_range(struct dmar_domain *domain, unsigned long start, struct cache_tag *tag; unsigned long flags; - if (start == 0 && end == ULONG_MAX) { - addr = 0; - mask = MAX_AGAW_PFN_WIDTH; - } else { - addr = calculate_psi_aligned_address(start, end, &mask); - } + addr = calculate_psi_aligned_address(start, end, &mask); spin_lock_irqsave(&domain->cache_lock, flags); list_for_each_entry(tag, &domain->cache_tags, node) { From 4af7ad0e6d7aa4403dbb1dac7b9659b0421efcaa Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:48 +0200 Subject: [PATCH 075/411] usb: typec: wcove: don't write past struct pd_message in wcove_read_rx_buffer() wcove_read_rx_buffer() copies the PD RX FIFO into the caller's struct pd_message with for (i = 0; i < USBC_RXINFO_RXBYTES(info); i++) regmap_read(wcove->regmap, USBC_RX_DATA + i, msg + i); which has two problems: USBC_RXINFO_RXBYTES() is a 5-bit field (max 31) while struct pd_message is 30 bytes (__le16 header + __le32 payload[PD_MAX_PAYLOAD], packed). The byte count latched in RXINFO is the number of bytes the port partner put on the wire, so a malicious partner that transmits a 31-byte frame can drive the loop one byte past the destination if the WCOVE BMC receiver does not enforce the PD object-count limit in hardware. The existing FIXME flagged this as unverified. Independently, regmap_read() takes an unsigned int * and stores a full unsigned int at the destination. Passing the byte pointer msg + i means each iteration writes four bytes; the high three are zero (val_bits is 8) and are normally overwritten by the next iteration, but the final iteration's high bytes are not. With RXBYTES == 30 the i == 29 iteration already writes three zero bytes past msg, which sits on the IRQ thread's stack in wcove_typec_irq(). Clamp the loop to sizeof(struct pd_message) and read each register into a local before storing only its low byte, so the copy can never exceed the destination regardless of what RXINFO reports. Assisted-by: gkh_clanker_t1000 Cc: stable Reviewed-by: Heikki Krogerus Link: https://patch.msgid.link/2026051347-clustered-deflected-9543@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/wcove.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/usb/typec/tcpm/wcove.c b/drivers/usb/typec/tcpm/wcove.c index 759c982bb16a..0e5a3e277c3e 100644 --- a/drivers/usb/typec/tcpm/wcove.c +++ b/drivers/usb/typec/tcpm/wcove.c @@ -444,9 +444,11 @@ static int wcove_start_toggling(struct tcpc_dev *tcpc, return regmap_write(wcove->regmap, USBC_CONTROL1, usbc_ctrl); } -static int wcove_read_rx_buffer(struct wcove_typec *wcove, void *msg) +static int wcove_read_rx_buffer(struct wcove_typec *wcove, + struct pd_message *msg) { - unsigned int info; + unsigned int info, val, len; + u8 *buf = (u8 *)msg; int ret; int i; @@ -454,12 +456,13 @@ static int wcove_read_rx_buffer(struct wcove_typec *wcove, void *msg) if (ret) return ret; - /* FIXME: Check that USBC_RXINFO_RXBYTES(info) matches the header */ + len = min(USBC_RXINFO_RXBYTES(info), sizeof(*msg)); - for (i = 0; i < USBC_RXINFO_RXBYTES(info); i++) { - ret = regmap_read(wcove->regmap, USBC_RX_DATA + i, msg + i); + for (i = 0; i < len; i++) { + ret = regmap_read(wcove->regmap, USBC_RX_DATA + i, &val); if (ret) return ret; + buf[i] = val; } return regmap_write(wcove->regmap, USBC_RXSTATUS, From 8a18f896e667df491331371b55d4ad644dc51d60 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:49 +0200 Subject: [PATCH 076/411] usb: typec: altmodes/displayport: validate count before reading Status Update VDO A broken/malicious device can send the incorrect count for a status update VDO, which will cause the kernel to read uninitialized stack data and send it off elsewhere. Fix this up by correctly verifying the count for the update object. Assisted-by: gkh_clanker_t1000 Cc: stable Reviewed-by: Heikki Krogerus Link: https://patch.msgid.link/2026051350-reacquire-sculpture-4244@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/altmodes/displayport.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/usb/typec/altmodes/displayport.c b/drivers/usb/typec/altmodes/displayport.c index 35d9c3086990..263a89c5f324 100644 --- a/drivers/usb/typec/altmodes/displayport.c +++ b/drivers/usb/typec/altmodes/displayport.c @@ -405,6 +405,8 @@ static int dp_altmode_vdm(struct typec_altmode *alt, dp->state = DP_STATE_EXIT_PRIME; break; case DP_CMD_STATUS_UPDATE: + if (count < 2) + break; dp->data.status = *vdo; ret = dp_altmode_status_update(dp); break; From aa2f716327be1818e1cb156da8a2844804aaec2f Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:50 +0200 Subject: [PATCH 077/411] usb: typec: tcpm/tcpci_maxim: validate header NDO against RX_BYTE_CNT MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A broken/malicious port can transmit a CRC-valid frame whose header advertises up to seven data objects but whose body carries fewer than that. Check for this, and rightfully reject the message, instead of reading from uninitialized stack memory. Assisted-by: gkh_clanker_t1000 Cc: Heikki Krogerus Cc: "André Draszik" Cc: Badhri Jagan Sridharan Cc: Amit Sunil Dhamne Cc: stable Link: https://patch.msgid.link/2026051350-sitter-canopener-9045@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/tcpci_maxim_core.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/usb/typec/tcpm/tcpci_maxim_core.c b/drivers/usb/typec/tcpm/tcpci_maxim_core.c index c0ee7e6959ed..7324139d51c8 100644 --- a/drivers/usb/typec/tcpm/tcpci_maxim_core.c +++ b/drivers/usb/typec/tcpm/tcpci_maxim_core.c @@ -181,6 +181,15 @@ static void process_rx(struct max_tcpci_chip *chip, u16 status) rx_buf_ptr = rx_buf + TCPC_RECEIVE_BUFFER_RX_BYTE_BUF_OFFSET; msg.header = cpu_to_le16(*(u16 *)rx_buf_ptr); rx_buf_ptr = rx_buf_ptr + sizeof(msg.header); + + if (count < TCPC_RECEIVE_BUFFER_RX_BYTE_BUF_OFFSET + sizeof(msg.header) + + pd_header_cnt_le(msg.header) * sizeof(msg.payload[0])) { + max_tcpci_write16(chip, TCPC_ALERT, TCPC_ALERT_RX_STATUS); + dev_err(chip->dev, "Invalid TCPC_RX_BYTE_CNT %d for header cnt %d\n", + count, pd_header_cnt_le(msg.header)); + return; + } + for (payload_index = 0; payload_index < pd_header_cnt_le(msg.header); payload_index++, rx_buf_ptr += sizeof(msg.payload[0])) msg.payload[payload_index] = cpu_to_le32(*(u32 *)rx_buf_ptr); From 8fbc349e8383125dd2d8de1c1e926279d398ab17 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:51 +0200 Subject: [PATCH 078/411] usb: typec: tcpm: validate VDO count in Discover Identity ACK handlers Properly validate the count passed from a device when calling svdm_consume_identity() or svdm_consume_identity_sop_prime() as the device-controlled value could index off of the static arrays, which could leak data. Assisted-by: gkh_clanker_t1000 Cc: Heikki Krogerus Cc: stable Reviewed-by: Badhri Jagan Sridharan Link: https://patch.msgid.link/2026051350-plated-salute-0efe@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/tcpm.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c index 55fee96d3342..44dab6c32c33 100644 --- a/drivers/usb/typec/tcpm/tcpm.c +++ b/drivers/usb/typec/tcpm/tcpm.c @@ -1855,6 +1855,9 @@ static void svdm_consume_identity(struct tcpm_port *port, const u32 *p, int cnt) u32 vdo = p[VDO_INDEX_IDH]; u32 product = p[VDO_INDEX_PRODUCT]; + if (cnt <= VDO_INDEX_PRODUCT) + return; + memset(&port->mode_data, 0, sizeof(port->mode_data)); port->partner_ident.id_header = vdo; @@ -1875,6 +1878,9 @@ static void svdm_consume_identity_sop_prime(struct tcpm_port *port, const u32 *p u32 product = p[VDO_INDEX_PRODUCT]; int svdm_version; + if (cnt <= VDO_INDEX_CABLE_1) + return; + /* * Attempt to consume identity only if cable currently is not set */ @@ -1898,7 +1904,7 @@ static void svdm_consume_identity_sop_prime(struct tcpm_port *port, const u32 *p switch (port->negotiated_rev_prime) { case PD_REV30: port->cable_desc.pd_revision = 0x0300; - if (port->cable_desc.active) + if (port->cable_desc.active && cnt > VDO_INDEX_CABLE_2) port->cable_ident.vdo[1] = p[VDO_INDEX_CABLE_2]; break; case PD_REV20: From 3389c149c68c3fea61910ad5d34f7bf3bff44e32 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:53 +0200 Subject: [PATCH 079/411] usb: typec: tcpm: bound altmode_desc[] per iteration in svdm_consume_modes() svdm_consume_modes() checks pmdata->altmodes against the array size once before the loop over the count, but forgot to check the bound at every point in the loop. In the well-behaved SVDM discovery flow this is harmless because each of at most SVID_DISCOVERY_MAX SVIDs contributes at most MODE_DISCOVERY_MAX modes, exactly filling altmode_desc[ALTMODE_DISCOVERY_MAX]. But the CMDT_RSP_ACK handler in tcpm_pd_svdm() does not correlate an incoming ACK with any request the port actually sent. Once port->partner is set, an unsolicited Discover Modes ACK is consumed unconditionally. A broken or malicious port partner can therefore drive altmodes to ALTMODE_DISCOVERY_MAX - 1 via the normal flow, and then send one extra Discover Modes ACK with seven VDOs. Because the pre-loop check passes, the loop could then writes up to five entries past altmode_desc[]. For mode_data_prime the next field in struct tcpm_port is the partner_altmode[] pointer array, which then receives partner-chosen SVID/VDO bytes. Move the bound check inside the loop so the array can never be indexed past ALTMODE_DISCOVERY_MAX regardless of how many VDOs the partner supplies or how the function was reached. Assisted-by: gkh_clanker_t1000 Cc: Badhri Jagan Sridharan Cc: Heikki Krogerus Cc: stable Link: https://patch.msgid.link/2026051351-reshuffle-skillful-90af@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/tcpm.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c index 44dab6c32c33..ed5f745a8231 100644 --- a/drivers/usb/typec/tcpm/tcpm.c +++ b/drivers/usb/typec/tcpm/tcpm.c @@ -1992,23 +1992,19 @@ static void svdm_consume_modes(struct tcpm_port *port, const u32 *p, int cnt, switch (rx_sop_type) { case TCPC_TX_SOP_PRIME: pmdata = &port->mode_data_prime; - if (pmdata->altmodes >= ARRAY_SIZE(port->plug_prime_altmode)) { - /* Already logged in svdm_consume_svids() */ - return; - } break; case TCPC_TX_SOP: pmdata = &port->mode_data; - if (pmdata->altmodes >= ARRAY_SIZE(port->partner_altmode)) { - /* Already logged in svdm_consume_svids() */ - return; - } break; default: return; } for (i = 1; i < cnt; i++) { + if (pmdata->altmodes >= ALTMODE_DISCOVERY_MAX) { + /* Already logged in svdm_consume_svids() */ + return; + } paltmode = &pmdata->altmode_desc[pmdata->altmodes]; memset(paltmode, 0, sizeof(*paltmode)); From 167dd8d12226587ee554f520aed0256b7769cd5d Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:54 +0200 Subject: [PATCH 080/411] usb: typec: ucsi: displayport: NAK DP_CMD_CONFIGURE without a payload VDO ucsi_displayport_vdm() handles a DP_CMD_CONFIGURE by copying the first payload VDO from data[], but unlike the equivalent handler in altmodes/displayport.c it does not check that count covers a VDO beyond the header. A header-only Configure VDM (count == 1) would read one u32 past the caller's array. In the normal UCSI path the caller controls count, so this is hardening for non-standard delivery paths. NAK and bail when no configuration VDO is present, matching the generic DP altmode driver's existing guard. Assisted-by: gkh_clanker_t1000 Cc: Pooja Katiyar Cc: Johan Hovold Cc: stable Reviewed-by: Heikki Krogerus Link: https://patch.msgid.link/2026051351-vividly-flattered-eb3d@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/displayport.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/typec/ucsi/displayport.c b/drivers/usb/typec/ucsi/displayport.c index 8aae80b457d7..67a0991a7b76 100644 --- a/drivers/usb/typec/ucsi/displayport.c +++ b/drivers/usb/typec/ucsi/displayport.c @@ -240,6 +240,10 @@ static int ucsi_displayport_vdm(struct typec_altmode *alt, dp->header |= VDO_CMDT(CMDT_RSP_ACK); break; case DP_CMD_CONFIGURE: + if (count < 2) { + dp->header |= VDO_CMDT(CMDT_RSP_NAK); + break; + } dp->data.conf = *data; if (ucsi_displayport_configure(dp)) { dp->header |= VDO_CMDT(CMDT_RSP_NAK); From 288a81a8507052bcfbf884d39a463c44c42c5fd9 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 13 May 2026 17:52:55 +0200 Subject: [PATCH 081/411] usb: typec: ucsi: validate connector number in ucsi_connector_change() The connector number in a UCSI CCI notification is a 7-bit field supplied by the PPM. ucsi_connector_change() uses it to index the ucsi->connector[] array without checking it against the number of connectors the PPM reported at init time, so a buggy or malicious PPM (EC firmware, or an I2C-attached UCSI controller on the ccg / stm32g0 / glink transports) can drive schedule_work() on memory past the end of the array. Reject connector numbers that are zero or exceed cap.num_connectors before dereferencing the array. Assisted-by: gkh_clanker_t1000 Cc: Heikki Krogerus Cc: Benson Leung Cc: Jameson Thies Cc: Nathan Rebello Cc: Johan Hovold Cc: Pooja Katiyar Cc: Hsin-Te Yuan Cc: Abel Vesa Cc: stable Reviewed-by: Abel Vesa Reviewed-by: Heikki Krogerus Reviewed-by: Benson Leung Link: https://patch.msgid.link/2026051351-truck-steadfast-df48@gregkh Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/ucsi.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c index 5b7ad9e99cb9..539dc706798d 100644 --- a/drivers/usb/typec/ucsi/ucsi.c +++ b/drivers/usb/typec/ucsi/ucsi.c @@ -1380,13 +1380,22 @@ static void ucsi_handle_connector_change(struct work_struct *work) */ void ucsi_connector_change(struct ucsi *ucsi, u8 num) { - struct ucsi_connector *con = &ucsi->connector[num - 1]; + struct ucsi_connector *con; if (!(ucsi->ntfy & UCSI_ENABLE_NTFY_CONNECTOR_CHANGE)) { dev_dbg(ucsi->dev, "Early connector change event\n"); return; } + if (!num || num > ucsi->cap.num_connectors) { + dev_warn_ratelimited(ucsi->dev, + "Bogus connector change on %u (max %u)\n", + num, ucsi->cap.num_connectors); + return; + } + + con = &ucsi->connector[num - 1]; + if (!test_and_set_bit(EVENT_PENDING, &ucsi->flags)) schedule_work(&con->work); } From 4ce058df2ee02cc2a0f0fd5cd64ce6f1482a0b65 Mon Sep 17 00:00:00 2001 From: Zhang Cen Date: Tue, 19 May 2026 19:11:50 +0800 Subject: [PATCH 082/411] USB: serial: belkin_sa: validate interrupt status length The Belkin interrupt callback treats interrupt data as a four-byte status report and reads LSR/MSR fields at offsets 2 and 3. The interrupt-in buffer length is derived from endpoint wMaxPacketSize, and short interrupt transfers may complete successfully with a smaller actual_length. Check the completed interrupt packet length before parsing status fields so short interrupt endpoints and short successful packets are ignored instead of causing out-of-bounds or stale status-byte reads. KASAN report as below: BUG: KASAN: slab-out-of-bounds in belkin_sa_read_int_callback() Read of size 1 Call trace: belkin_sa_read_int_callback() (drivers/usb/serial/belkin_sa.c:202) __usb_hcd_giveback_urb() (drivers/usb/core/hcd.c:1630) dummy_timer() (?:?) Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Assisted-by: Codex:gpt-5.5 Signed-off-by: Zhang Cen Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold --- drivers/usb/serial/belkin_sa.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/serial/belkin_sa.c b/drivers/usb/serial/belkin_sa.c index 38ac910b1082..7bbd9523d4e9 100644 --- a/drivers/usb/serial/belkin_sa.c +++ b/drivers/usb/serial/belkin_sa.c @@ -194,6 +194,9 @@ static void belkin_sa_read_int_callback(struct urb *urb) usb_serial_debug_data(&port->dev, __func__, urb->actual_length, data); + if (urb->actual_length < BELKIN_SA_MSR_INDEX + 1) + goto exit; + /* Handle known interrupt data */ /* ignore data[0] and data[1] */ From cb3560e8eab1dfa1cac1ed52631adf8ec6ff2cd5 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 20 May 2026 16:26:22 +0200 Subject: [PATCH 083/411] USB: serial: digi_acceleport: fix memory corruption with small endpoints Add the missing bulk-out buffer size sanity checks to avoid out-of-bounds memory accesses or slab corruption should a malicious device report smaller buffers than expected. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/digi_acceleport.c | 23 +++++++++++++++++++++-- 1 file changed, 21 insertions(+), 2 deletions(-) diff --git a/drivers/usb/serial/digi_acceleport.c b/drivers/usb/serial/digi_acceleport.c index d515df045c4c..c481208255eb 100644 --- a/drivers/usb/serial/digi_acceleport.c +++ b/drivers/usb/serial/digi_acceleport.c @@ -1229,15 +1229,34 @@ static int digi_port_init(struct usb_serial_port *port, unsigned port_num) static int digi_startup(struct usb_serial *serial) { struct digi_serial *serial_priv; + int oob_port_num; int ret; + int i; + + /* + * The port bulk-out buffers must be large enough for header and + * buffered data. + */ + for (i = 0; i < serial->type->num_ports; i++) { + if (serial->port[i]->bulk_out_size < DIGI_OUT_BUF_SIZE + 2) + return -EINVAL; + } + + /* + * The OOB port bulk-out buffer must be large enough for the two + * commands in digi_set_modem_signals(). + */ + oob_port_num = serial->type->num_ports; + if (serial->port[oob_port_num]->bulk_out_size < 8) + return -EINVAL; serial_priv = kzalloc_obj(*serial_priv); if (!serial_priv) return -ENOMEM; spin_lock_init(&serial_priv->ds_serial_lock); - serial_priv->ds_oob_port_num = serial->type->num_ports; - serial_priv->ds_oob_port = serial->port[serial_priv->ds_oob_port_num]; + serial_priv->ds_oob_port_num = oob_port_num; + serial_priv->ds_oob_port = serial->port[oob_port_num]; ret = digi_port_init(serial_priv->ds_oob_port, serial_priv->ds_oob_port_num); From ab8336a7e414f018430aa1af3a46944032f7ff96 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 20 May 2026 16:26:48 +0200 Subject: [PATCH 084/411] USB: serial: keyspan: fix missing indat transfer sanity check Add the missing sanity check on the size of usa49wg indat transfers to avoid parsing stale or uninitialised slab data. Fixes: 0ca1268e109a ("USB Serial Keyspan: add support for USA-49WG & USA-28XG") Cc: stable@vger.kernel.org # 2.6.23 Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/keyspan.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/serial/keyspan.c b/drivers/usb/serial/keyspan.c index 46448843541a..28b80607cebd 100644 --- a/drivers/usb/serial/keyspan.c +++ b/drivers/usb/serial/keyspan.c @@ -1187,6 +1187,10 @@ static void usa49wg_indat_callback(struct urb *urb) len = 0; while (i < urb->actual_length) { + if (urb->actual_length - i < 3) { + dev_warn_ratelimited(&urb->dev->dev, "malformed indat packet\n"); + break; + } /* Check port number from message */ if (data[i] >= serial->num_ports) { From 915b36d701950503c4ea0f6e314b10868e59fce3 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 20 May 2026 16:27:00 +0200 Subject: [PATCH 085/411] USB: serial: mct_u232: fix memory corruption with small endpoint The driver overrides the maximum transfer size for a specific device which only accepts 16 byte packets for its 32 byte bulk-out endpoint. Make sure to never increase the maximum transfer size to prevent slab corruption should a malicious device report a smaller endpoint max packet size than expected. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/mct_u232.c | 21 +++++++++++---------- 1 file changed, 11 insertions(+), 10 deletions(-) diff --git a/drivers/usb/serial/mct_u232.c b/drivers/usb/serial/mct_u232.c index 18844b92bd08..ca1530da6e77 100644 --- a/drivers/usb/serial/mct_u232.c +++ b/drivers/usb/serial/mct_u232.c @@ -378,6 +378,7 @@ static int mct_u232_port_probe(struct usb_serial_port *port) { struct usb_serial *serial = port->serial; struct mct_u232_private *priv; + u16 pid; /* check first to simplify error handling */ if (!serial->port[1] || !serial->port[1]->interrupt_in_urb) { @@ -385,6 +386,16 @@ static int mct_u232_port_probe(struct usb_serial_port *port) return -ENODEV; } + /* + * Compensate for a hardware bug: although the Sitecom U232-P25 + * device reports a maximum output packet size of 32 bytes, + * it seems to be able to accept only 16 bytes (and that's what + * SniffUSB says too...) + */ + pid = le16_to_cpu(serial->dev->descriptor.idProduct); + if (pid == MCT_U232_SITECOM_PID) + port->bulk_out_size = min(16, port->bulk_out_size); + priv = kzalloc_obj(*priv); if (!priv) return -ENOMEM; @@ -410,7 +421,6 @@ static void mct_u232_port_remove(struct usb_serial_port *port) static int mct_u232_open(struct tty_struct *tty, struct usb_serial_port *port) { - struct usb_serial *serial = port->serial; struct mct_u232_private *priv = usb_get_serial_port_data(port); int retval = 0; unsigned int control_state; @@ -418,15 +428,6 @@ static int mct_u232_open(struct tty_struct *tty, struct usb_serial_port *port) unsigned char last_lcr; unsigned char last_msr; - /* Compensate for a hardware bug: although the Sitecom U232-P25 - * device reports a maximum output packet size of 32 bytes, - * it seems to be able to accept only 16 bytes (and that's what - * SniffUSB says too...) - */ - if (le16_to_cpu(serial->dev->descriptor.idProduct) - == MCT_U232_SITECOM_PID) - port->bulk_out_size = 16; - /* Do a defined restart: the normal serial device seems to * always turn on DTR and RTS here, so do the same. I'm not * sure if this is really necessary. But it should not harm From 245aba83e3c288e176ed037a1f6b618b09e92ed8 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Wed, 20 May 2026 16:27:10 +0200 Subject: [PATCH 086/411] USB: serial: mct_u232: fix missing interrupt-in transfer sanity check Add the missing sanity check on the size of interrupt-in transfers to avoid parsing stale or uninitialised slab data (and leaking it to user space). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/mct_u232.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/usb/serial/mct_u232.c b/drivers/usb/serial/mct_u232.c index ca1530da6e77..163161881d2d 100644 --- a/drivers/usb/serial/mct_u232.c +++ b/drivers/usb/serial/mct_u232.c @@ -544,6 +544,11 @@ static void mct_u232_read_int_callback(struct urb *urb) goto exit; } + if (urb->actual_length < 2) { + dev_warn_ratelimited(&port->dev, "short interrupt-in packet\n"); + goto exit; + } + /* * The interrupt-in pipe signals exceptional conditions (modem line * signal changes and errors). data[0] holds MSR, data[1] holds LSR. From f42d01aadcedd7bbf4f9a466cabe25c1781dedad Mon Sep 17 00:00:00 2001 From: Hongtao Lee Date: Wed, 20 May 2026 11:01:26 +0800 Subject: [PATCH 087/411] tools/bootconfig: Fix buf leaks in apply_xbc If data calloc failed, free the buf before return. Link: https://lore.kernel.org/all/20260520030126.147782-1-lihongtao@kylinos.cn/ Fixes: 950313ebf79c ("tools: bootconfig: Add bootconfig command") Signed-off-by: Hongtao Lee Signed-off-by: Masami Hiramatsu (Google) --- tools/bootconfig/main.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/bootconfig/main.c b/tools/bootconfig/main.c index 643f707b8f1d..ddabde20585f 100644 --- a/tools/bootconfig/main.c +++ b/tools/bootconfig/main.c @@ -390,8 +390,10 @@ static int apply_xbc(const char *path, const char *xbc_path) /* Backup the bootconfig data */ data = calloc(size + BOOTCONFIG_ALIGN + BOOTCONFIG_FOOTER_SIZE, 1); - if (!data) + if (!data) { + free(buf); return -ENOMEM; + } memcpy(data, buf, size); /* Check the data format */ From 7d2b37d3e42d19071b62f4ddbee6e16e905efbf1 Mon Sep 17 00:00:00 2001 From: Jan Volckaert Date: Sun, 17 May 2026 17:32:37 +0200 Subject: [PATCH 088/411] USB: serial: option: add MeiG SRM813Q Add support for the Qualcomm Technology Snapdragon X35-based MeiG SRM813Q module. The module can be put in different modes via AT commands to enable/disable GPS functionality: MODEM - PPP mode(2dee:4d63): AT+SER=1,1 If#= 0: RMNET If#= 1: DIAG/ADB If#= 2: MODEM If#= 3: AT P: Vendor=2dee ProdID=4d63 Rev=05.15 S: Manufacturer=MEIG S: Product=LTE-A Module S: SerialNumber=1bd51f0e C: #Ifs= 4 Cfg#= 1 Atr=80 MxPwr=500mA I: If#= 0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=50 Driver=qmi_wwan E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=82(I) Atr=03(Int.) MxPS= 8 Ivl=32ms I: If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=30 Driver=option E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms I: If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=85(I) Atr=03(Int.) MxPS= 10 Ivl=32ms I: If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=86(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=87(I) Atr=03(Int.) MxPS= 10 Ivl=32ms NMEA mode(2dee:4d64): AT+SER=51,1 If#= 0: RMNET If#= 1: DIAG/ADB If#= 2: NMEA If#= 3: AT P: Vendor=2dee ProdID=4d64 Rev=05.15 S: Manufacturer=MEIG S: Product=LTE-A Module S: SerialNumber=1bd51f0e C: #Ifs= 4 Cfg#= 1 Atr=80 MxPwr=500mA I: If#= 0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=50 Driver=qmi_wwan E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=82(I) Atr=03(Int.) MxPS= 8 Ivl=32ms I: If#= 1 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=30 Driver=option E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms I: If#= 2 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=60 Driver=option E: Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=85(I) Atr=03(Int.) MxPS= 10 Ivl=32ms I: If#= 3 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=86(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=87(I) Atr=03(Int.) MxPS= 10 Ivl=32ms Signed-off-by: Jan Volckaert Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold --- drivers/usb/serial/option.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 42e4cecd28ac..92c2feda3e3d 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -2450,6 +2450,12 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d38, 0xff, 0xff, 0x30) }, /* MeiG Smart SRM825WN (Diag) */ { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d38, 0xff, 0xff, 0x40) }, /* MeiG Smart SRM825WN (AT) */ { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d38, 0xff, 0xff, 0x60) }, /* MeiG Smart SRM825WN (NMEA) */ + { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d63, 0xff, 0xff, 0x30) }, /* MeiG SRM813Q (Diag) */ + { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d63, 0xff, 0xff, 0x40) }, /* MeiG SRM813Q (AT) */ + { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d64, 0xff, 0xff, 0x30) }, /* MeiG SRM813Q (Diag) */ + { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d64, 0xff, 0xff, 0x40) }, /* MeiG SRM813Q (AT) */ + { USB_DEVICE_AND_INTERFACE_INFO(0x2dee, 0x4d64, 0xff, 0xff, 0x60) }, /* MeiG SRM813Q (NMEA) */ + { USB_DEVICE_INTERFACE_CLASS(0x2df3, 0x9d03, 0xff) }, /* LongSung M5710 */ { USB_DEVICE_INTERFACE_CLASS(0x305a, 0x1404, 0xff) }, /* GosunCn GM500 RNDIS */ { USB_DEVICE_INTERFACE_CLASS(0x305a, 0x1405, 0xff) }, /* GosunCn GM500 MBIM */ From 689f2facc689c8add11d7ff69fbbad17d65ee596 Mon Sep 17 00:00:00 2001 From: Wanquan Zhong Date: Wed, 20 May 2026 19:32:45 +0800 Subject: [PATCH 089/411] USB: serial: option: add missing RSVD(5) flag for Rolling RW135R-GL The RW135R-GL entry added in commit 01e8d0f74222 ("USB: serial: option: add support for Rolling Wireless RW135R-GL") was missing the .driver_info = RSVD(5) flag used by other Rolling Wireless MBIM laptop modules (e.g. RW135-GL and RW350-GL). Without this flag, the option driver incorrectly binds to the reserved ADB interface (If#5) in multi-interface USB modes, causing AT/MBIM communication failures after mode switching. This matches the handling of other Rolling Wireless MBIM devices. - VID:PID 33f8:1003, RW135R-GL for laptop debug M.2 cards (with MBIM interface for Linux/Chrome OS) 0x1003: mbim, diag, AT, pipe Here are the outputs of usb-devices: T: Bus=03 Lev=01 Prnt=01 Port=04 Cnt=02 Dev#= 8 Spd=480 MxCh= 0 D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1 P: Vendor=33f8 ProdID=1003 Rev= 5.15 S: Manufacturer=Rolling Wireless S.a.r.l. S: Product=Rolling RW135R-GL Module S: SerialNumber=12345678 C:* #Ifs= 5 Cfg#= 1 Atr=a0 MxPwr=500mA A: FirstIf#= 0 IfCount= 2 Cls=02(comm.) Sub=0e Prot=00 I:* If#= 0 Alt= 0 #EPs= 1 Cls=02(comm.) Sub=0e Prot=00 Driver=cdc_mbim E: Ad=82(I) Atr=03(Int.) MxPS= 64 Ivl=32ms I: If#= 1 Alt= 0 #EPs= 0 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim I:* If#= 1 Alt= 1 #EPs= 2 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=00 Prot=00 Driver=option E: Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 3 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=30 Driver=option E: Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=86(I) Atr=03(Int.) MxPS= 10 Ivl=32ms E: Ad=85(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms - VID:PID 33f8:1003, RW135R-GL for laptop debug M.2 cards (with MBIM interface for Linux/Chrome OS) 0x1003: mbim, diag, AT, ADB, pipe Here are the outputs of usb-devices: T: Bus=03 Lev=01 Prnt=01 Port=04 Cnt=02 Dev#= 7 Spd=480 MxCh= 0 D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1 P: Vendor=33f8 ProdID=1003 Rev= 5.15 S: Manufacturer=Rolling Wireless S.a.r.l. S: Product=Rolling RW135R-GL Module S: SerialNumber=12345678 C:* #Ifs= 6 Cfg#= 1 Atr=a0 MxPwr=500mA A: FirstIf#= 0 IfCount= 2 Cls=02(comm.) Sub=0e Prot=00 I:* If#= 0 Alt= 0 #EPs= 1 Cls=02(comm.) Sub=0e Prot=00 Driver=cdc_mbim E: Ad=82(I) Atr=03(Int.) MxPS= 64 Ivl=32ms I: If#= 1 Alt= 0 #EPs= 0 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim I:* If#= 1 Alt= 1 #EPs= 2 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=00 Prot=00 Driver=option E: Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 3 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=30 Driver=option E: Ad=03(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 4 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=ff Prot=40 Driver=option E: Ad=86(I) Atr=03(Int.) MxPS= 10 Ivl=32ms E: Ad=85(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=04(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 5 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=42 Prot=01 Driver=(none) E: Ad=05(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=87(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms - VID:PID 33f8:1003, RW135R-GL for laptop debug M.2 cards (with MBIM interface for Linux/Chrome OS) 0x1003: mbim, pipe Here are the outputs of usb-devices: T: Bus=03 Lev=01 Prnt=01 Port=04 Cnt=02 Dev#= 9 Spd=480 MxCh= 0 D: Ver= 2.00 Cls=00(>ifc ) Sub=00 Prot=00 MxPS=64 #Cfgs= 1 P: Vendor=33f8 ProdID=1003 Rev= 5.15 S: Manufacturer=Rolling Wireless S.a.r.l. S: Product=Rolling RW135R-GL Module S: SerialNumber=12345678 C:* #Ifs= 3 Cfg#= 1 Atr=a0 MxPwr=500mA A: FirstIf#= 0 IfCount= 2 Cls=02(comm.) Sub=0e Prot=00 I:* If#= 0 Alt= 0 #EPs= 1 Cls=02(comm.) Sub=0e Prot=00 Driver=cdc_mbim E: Ad=82(I) Atr=03(Int.) MxPS= 64 Ivl=32ms I: If#= 1 Alt= 0 #EPs= 0 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim I:* If#= 1 Alt= 1 #EPs= 2 Cls=0a(data ) Sub=00 Prot=02 Driver=cdc_mbim E: Ad=81(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=01(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms I:* If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=00 Prot=00 Driver=option E: Ad=83(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms E: Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms Fixes: 01e8d0f74222 ("USB: serial: option: add support for Rolling Wireless RW135R-GL") Signed-off-by: Wanquan Zhong Cc: stable@vger.kernel.org Signed-off-by: Johan Hovold --- drivers/usb/serial/option.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c index 92c2feda3e3d..48ae0188f2e9 100644 --- a/drivers/usb/serial/option.c +++ b/drivers/usb/serial/option.c @@ -2476,7 +2476,8 @@ static const struct usb_device_id option_ids[] = { { USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x0302, 0xff) }, /* Rolling RW101R-GL (laptop MBIM) */ { USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x0802, 0xff), /* Rolling RW350-GL (laptop MBIM) */ .driver_info = RSVD(5) }, - { USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x1003, 0xff) }, /* Rolling RW135R-GL (laptop MBIM) */ + { USB_DEVICE_INTERFACE_CLASS(0x33f8, 0x1003, 0xff), /* Rolling RW135R-GL (laptop MBIM) */ + .driver_info = RSVD(5) }, { USB_DEVICE_AND_INTERFACE_INFO(0x3731, 0x0100, 0xff, 0xff, 0x30) }, /* NetPrisma LCUK54-WWD for Global */ { USB_DEVICE_AND_INTERFACE_INFO(0x3731, 0x0100, 0xff, 0x00, 0x40) }, { USB_DEVICE_AND_INTERFACE_INFO(0x3731, 0x0100, 0xff, 0xff, 0x40) }, From 8a3bee801d420be8a7a0bae4a26547b353b8fe22 Mon Sep 17 00:00:00 2001 From: Ian Abbott Date: Wed, 22 Apr 2026 15:46:37 +0100 Subject: [PATCH 090/411] comedi: comedi_test: Fix limiting of convert_arg in waveform_ai_cmdtest() The function checks and possibly modifies the description of an asynchronous command to be run on the analog input subdevice of a comedi device attached to the "comedi_test" driver, returning 0 if no modifications were required, or a positive value that indicates which step of the checking process it failed on. Step 4 fixes up various argument values for various trigger sources. There are two bugs in the fixing up of the `convert_arg` value to keep the `scan_begin_arg` value within the range of `unsigned int` when `scan_begin_src` and `convert_src` both have the value `TRIG_TIMER`, which indicates that the corresponding `_arg` values hold a time period in nanoseconds. The code also uses `scan_end_arg` which hold the number of "conversions" within each "scan". The goal is to end up with the scan period being less than or equal to the convert period multiplied by the number of conversions per scan. It intends to do that by clamping the `convert_arg` value to a maximum value of `UINT_MAX / scan_end_arg` rounded down to a multiple of 1000 (`NSEC_PER_USEC`). (The rounding from nanoseconds to microseconds is because the driver is modelling a device that uses a 1 MHz clock for timing. This is partly because that is a more typical timing base for real hardware devices driven by comedi, and partly because the driver used to use `struct timeval` internally.) The first bug is that the code checks if `scan_begin_arg == TRIG_TIMER` when it should be checking if `scan_begin_src == TRIG_TIMER`. The bugged check will always fail because if `scan_begin_src == TRIG_TIMER`, then `scan_begin_arg` will be at least 1000 (`NSEC_PER_USEC`), otherwise `scan_begin_src == TRIG_FOLLOW` and `scan_begin_arg` will be 0. (N.B `TRIG_TIMER` is defined as `0x10`.) The second bug is that is rounding the maximum value down to a multiple of 1000000000 (`NSEC_PER_SEC`) instead of 1000 (`NSEC_PER_USEC`), however this bug is not reached due to the first bug. This patch fixes both bugs. Fixes: 783ddaebd397 ("staging: comedi: comedi_test: support scan_begin_src == TRIG_FOLLOW") Fixes: 5afdcad2f818 ("staging: comedi: comedi_test: limit maximum convert_arg") Cc: stable Signed-off-by: Ian Abbott Link: https://patch.msgid.link/20260422144637.27692-1-abbotti@mev.co.uk Signed-off-by: Greg Kroah-Hartman --- drivers/comedi/drivers/comedi_test.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/comedi/drivers/comedi_test.c b/drivers/comedi/drivers/comedi_test.c index 01aafce20ef8..4050f66193e5 100644 --- a/drivers/comedi/drivers/comedi_test.c +++ b/drivers/comedi/drivers/comedi_test.c @@ -324,10 +324,10 @@ static int waveform_ai_cmdtest(struct comedi_device *dev, arg = min(arg, rounddown(UINT_MAX, (unsigned int)NSEC_PER_USEC)); arg = NSEC_PER_USEC * DIV_ROUND_CLOSEST(arg, NSEC_PER_USEC); - if (cmd->scan_begin_arg == TRIG_TIMER) { + if (cmd->scan_begin_src == TRIG_TIMER) { /* limit convert_arg to keep scan_begin_arg in range */ limit = UINT_MAX / cmd->scan_end_arg; - limit = rounddown(limit, (unsigned int)NSEC_PER_SEC); + limit = rounddown(limit, (unsigned int)NSEC_PER_USEC); arg = min(arg, limit); } err |= comedi_check_trigger_arg_is(&cmd->convert_arg, arg); From 542f5248cb481073203e0dadab5bcbd28aeae308 Mon Sep 17 00:00:00 2001 From: Ian Abbott Date: Wed, 22 Apr 2026 17:21:19 +0100 Subject: [PATCH 091/411] comedi: comedi_test: fix check for valid scan_begin_src in waveform_ai_cmdtest() Commit 783ddaebd397 ("staging: comedi: comedi_test: support scan_begin_src == TRIG_FOLLOW") neglected to add a test that `scan_begin_src` has only one bit set. The allowed values are `TRIG_FOLLOW` and `TRIG_TIMER`, but the code incorrectly also allows `TRIG_FOLLOW | TRIG_TIMER`. Add a call to `comedi_check_trigger_is_unique()` to check that only one trigger source bit is set. Fixes: 783ddaebd397 ("staging: comedi: comedi_test: support scan_begin_src == TRIG_FOLLOW") Cc: stable Signed-off-by: Ian Abbott Link: https://patch.msgid.link/20260422162138.36003-1-abbotti@mev.co.uk Signed-off-by: Greg Kroah-Hartman --- drivers/comedi/drivers/comedi_test.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/comedi/drivers/comedi_test.c b/drivers/comedi/drivers/comedi_test.c index 4050f66193e5..1f430ffc7bd9 100644 --- a/drivers/comedi/drivers/comedi_test.c +++ b/drivers/comedi/drivers/comedi_test.c @@ -274,6 +274,7 @@ static int waveform_ai_cmdtest(struct comedi_device *dev, /* Step 2a : make sure trigger sources are unique */ err |= comedi_check_trigger_is_unique(cmd->convert_src); + err |= comedi_check_trigger_is_unique(cmd->scan_begin_src); err |= comedi_check_trigger_is_unique(cmd->stop_src); /* Step 2b : and mutually compatible */ From 2e78b21864dd6e21b76160753ea632b5e758fdbd Mon Sep 17 00:00:00 2001 From: Myeonghun Pak Date: Fri, 24 Apr 2026 22:21:31 +0900 Subject: [PATCH 092/411] HID: u2fzero: free allocated URB on probe errors u2fzero_fill_in_urb() allocates dev->urb with usb_alloc_urb(), but u2fzero_probe() ignored its return value and only freed the URB from u2fzero_remove(). If LED or hwrng registration fails after the URB allocation, probe returns an error and the driver core does not call .remove(), leaking the URB. A failed URB setup was also allowed to continue probing with an unusable device. Check the URB setup result and add the missing probe-error unwind so the URB is freed before returning from later errors. Signed-off-by: Myeonghun Pak Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-u2fzero.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/drivers/hid/hid-u2fzero.c b/drivers/hid/hid-u2fzero.c index 744a91e6e78c..82404b6e2d25 100644 --- a/drivers/hid/hid-u2fzero.c +++ b/drivers/hid/hid-u2fzero.c @@ -341,29 +341,33 @@ static int u2fzero_probe(struct hid_device *hdev, if (ret) return ret; - u2fzero_fill_in_urb(dev); + ret = u2fzero_fill_in_urb(dev); + if (ret) + goto err_hid_hw_stop; dev->present = true; minor = ((struct hidraw *) hdev->hidraw)->minor; ret = u2fzero_init_led(dev, minor); - if (ret) { - hid_hw_stop(hdev); - return ret; - } + if (ret) + goto err_free_urb; hid_info(hdev, "%s LED initialised\n", hw_configs[dev->hw_revision].name); ret = u2fzero_init_hwrng(dev, minor); - if (ret) { - hid_hw_stop(hdev); - return ret; - } + if (ret) + goto err_free_urb; hid_info(hdev, "%s RNG initialised\n", hw_configs[dev->hw_revision].name); return 0; + +err_free_urb: + usb_free_urb(dev->urb); +err_hid_hw_stop: + hid_hw_stop(hdev); + return ret; } static void u2fzero_remove(struct hid_device *hdev) From dd2147375a8fe7c5bc3f1f1b1d3a9567c26faefa Mon Sep 17 00:00:00 2001 From: Liu Kai Date: Thu, 7 May 2026 16:32:04 +0800 Subject: [PATCH 093/411] HID: remove duplicate hid_warn_ratelimited definition The hid_warn_ratelimited macro is defined twice in include/linux/hid.h: - first one added by commit 4051ead99888 ("HID: rate-limit hid_warn to prevent log flooding") - second one added by commit 1d64624243af ("HID: core: Add printk_ratelimited variants to hid_warn() etc")). The second definition is correctly grouped with other ratelimited macros. Remove the duplicate definition. Fixes: 1d64624243af ("HID: core: Add printk_ratelimited variants to hid_warn() etc") Signed-off-by: Liu Kai [bentiss: edited commit message] Signed-off-by: Benjamin Tissoires --- include/linux/hid.h | 2 -- 1 file changed, 2 deletions(-) diff --git a/include/linux/hid.h b/include/linux/hid.h index bfb9859f391e..47dc0bc89fa4 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -1316,8 +1316,6 @@ void hid_quirks_exit(__u16 bus); dev_notice(&(hid)->dev, fmt, ##__VA_ARGS__) #define hid_warn(hid, fmt, ...) \ dev_warn(&(hid)->dev, fmt, ##__VA_ARGS__) -#define hid_warn_ratelimited(hid, fmt, ...) \ - dev_warn_ratelimited(&(hid)->dev, fmt, ##__VA_ARGS__) #define hid_info(hid, fmt, ...) \ dev_info(&(hid)->dev, fmt, ##__VA_ARGS__) #define hid_dbg(hid, fmt, ...) \ From c1a0ecbf32c4b397353204e2ec94c5bb9f3300ed Mon Sep 17 00:00:00 2001 From: Radhey Shyam Pandey Date: Tue, 19 May 2026 17:25:29 +0530 Subject: [PATCH 094/411] usb: dwc3: xilinx: fix error handling in zynqmp init error paths Fix error handling and resource cleanup i.e remove invalid phy_exit() after failed phy_init(), route failures through proper cleanup paths and return 0 explicitly on success. Fixes: 84770f028fab ("usb: dwc3: Add driver for Xilinx platforms") Cc: stable@vger.kernel.org Acked-by: Thinh Nguyen Signed-off-by: Radhey Shyam Pandey Link: https://patch.msgid.link/20260519115529.2980421-1-radhey.shyam.pandey@amd.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc3/dwc3-xilinx.c | 27 +++++++++++++++------------ 1 file changed, 15 insertions(+), 12 deletions(-) diff --git a/drivers/usb/dwc3/dwc3-xilinx.c b/drivers/usb/dwc3/dwc3-xilinx.c index f41b0da5e89d..9b9525592a85 100644 --- a/drivers/usb/dwc3/dwc3-xilinx.c +++ b/drivers/usb/dwc3/dwc3-xilinx.c @@ -184,15 +184,13 @@ static int dwc3_xlnx_init_zynqmp(struct dwc3_xlnx *priv_data) } ret = phy_init(priv_data->usb3_phy); - if (ret < 0) { - phy_exit(priv_data->usb3_phy); + if (ret < 0) goto err; - } ret = reset_control_deassert(apbrst); if (ret < 0) { dev_err(dev, "Failed to release APB reset\n"); - goto err; + goto err_phy_exit; } if (priv_data->usb3_phy) { @@ -208,26 +206,24 @@ static int dwc3_xlnx_init_zynqmp(struct dwc3_xlnx *priv_data) ret = reset_control_deassert(crst); if (ret < 0) { dev_err(dev, "Failed to release core reset\n"); - goto err; + goto err_phy_exit; } ret = reset_control_deassert(hibrst); if (ret < 0) { dev_err(dev, "Failed to release hibernation reset\n"); - goto err; + goto err_phy_exit; } ret = phy_power_on(priv_data->usb3_phy); - if (ret < 0) { - phy_exit(priv_data->usb3_phy); - goto err; - } + if (ret < 0) + goto err_phy_exit; /* ulpi reset via gpio-modepin or gpio-framework driver */ reset_gpio = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH); if (IS_ERR(reset_gpio)) { - return dev_err_probe(dev, PTR_ERR(reset_gpio), - "Failed to request reset GPIO\n"); + ret = PTR_ERR(reset_gpio); + goto err_phy_power_off; } if (reset_gpio) { @@ -237,6 +233,13 @@ static int dwc3_xlnx_init_zynqmp(struct dwc3_xlnx *priv_data) } dwc3_xlnx_set_coherency(priv_data, XLNX_USB_TRAFFIC_ROUTE_CONFIG); + + return 0; + +err_phy_power_off: + phy_power_off(priv_data->usb3_phy); +err_phy_exit: + phy_exit(priv_data->usb3_phy); err: return ret; } From ca927fc45e4906bdab42426da044fba4d3584f34 Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Tue, 28 Apr 2026 21:18:21 +0800 Subject: [PATCH 095/411] usb: typec: fusb302: Fix resource leak when devm_drm_dp_hpd_bridge_add() fails If devm_drm_dp_hpd_bridge_add() fails during fusb302_probe(), the original code returned directly without cleaning up the resources. Move bridge registration before the IRQ is requested and route bridge registration failures through the existing TCPM unregister and fwnode cleanup path. Fixes: 5d79c525405d ("usb: typec: fusb302: add DRM DP HPD bridge support") Signed-off-by: Felix Gu Reviewed-by: Heikki Krogerus Reviewed-by: Sebastian Reichel Link: https://patch.msgid.link/20260428-fusb-v2-1-aa3b5942cabb@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/fusb302.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/usb/typec/tcpm/fusb302.c b/drivers/usb/typec/tcpm/fusb302.c index 889c4c29c1b8..9ab1277b7ed1 100644 --- a/drivers/usb/typec/tcpm/fusb302.c +++ b/drivers/usb/typec/tcpm/fusb302.c @@ -1751,19 +1751,22 @@ static int fusb302_probe(struct i2c_client *client) bridge_dev = devm_drm_dp_hpd_bridge_alloc(chip->dev, to_of_node(chip->tcpc_dev.fwnode)); if (IS_ERR(bridge_dev)) { - ret = PTR_ERR(bridge_dev); - dev_err_probe(chip->dev, ret, "failed to alloc bridge\n"); - goto destroy_workqueue; + ret = dev_err_probe(chip->dev, PTR_ERR(bridge_dev), + "failed to alloc bridge\n"); + goto fwnode_put; } chip->tcpm_port = tcpm_register_port(&client->dev, &chip->tcpc_dev); if (IS_ERR(chip->tcpm_port)) { - fwnode_handle_put(chip->tcpc_dev.fwnode); ret = dev_err_probe(dev, PTR_ERR(chip->tcpm_port), "cannot register tcpm port\n"); - goto destroy_workqueue; + goto fwnode_put; } + ret = devm_drm_dp_hpd_bridge_add(chip->dev, bridge_dev); + if (ret) + goto tcpm_unregister_port; + ret = request_threaded_irq(chip->gpio_int_n_irq, NULL, fusb302_irq_intn, IRQF_ONESHOT | IRQF_TRIGGER_LOW, "fsc_interrupt_int_n", chip); @@ -1774,14 +1777,11 @@ static int fusb302_probe(struct i2c_client *client) enable_irq_wake(chip->gpio_int_n_irq); i2c_set_clientdata(client, chip); - ret = devm_drm_dp_hpd_bridge_add(chip->dev, bridge_dev); - if (ret) - return ret; - - return ret; + return 0; tcpm_unregister_port: tcpm_unregister_port(chip->tcpm_port); +fwnode_put: fwnode_handle_put(chip->tcpc_dev.fwnode); destroy_workqueue: fusb302_debugfs_exit(chip); From 07466fc91c55532edcfb5c6a7ccd2ea52728d6bd Mon Sep 17 00:00:00 2001 From: hlleng Date: Tue, 12 May 2026 09:57:37 +0800 Subject: [PATCH 096/411] HID: quirks: Add ALWAYS_POLL quirk for SIGMACHIP USB mouse The SIGMACHIP USB mouse with VID/PID 1c4f:0034 can disconnect and re-enumerate repeatedly after it has been enumerated if its interrupt endpoint is not continuously polled. This was observed with the device reporting itself as "SIGMACHIP Usb Mouse". Keeping the input event device open avoids the disconnects. Add HID_QUIRK_ALWAYS_POLL for this device so the HID core keeps polling it even when there is no userspace input consumer. Cc: stable@vger.kernel.org Signed-off-by: hlleng Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-ids.h | 1 + drivers/hid/hid-quirks.c | 1 + 2 files changed, 2 insertions(+) diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h index 4657d96fb083..426ff78c1c03 100644 --- a/drivers/hid/hid-ids.h +++ b/drivers/hid/hid-ids.h @@ -1284,6 +1284,7 @@ #define USB_VENDOR_ID_SIGMA_MICRO 0x1c4f #define USB_DEVICE_ID_SIGMA_MICRO_KEYBOARD 0x0002 +#define USB_DEVICE_ID_SIGMA_MICRO_USB_MOUSE 0x0034 #define USB_DEVICE_ID_SIGMA_MICRO_KEYBOARD2 0x0059 #define USB_VENDOR_ID_SIGMATEL 0x066F diff --git a/drivers/hid/hid-quirks.c b/drivers/hid/hid-quirks.c index 512049963978..57d8efdd9b89 100644 --- a/drivers/hid/hid-quirks.c +++ b/drivers/hid/hid-quirks.c @@ -187,6 +187,7 @@ static const struct hid_device_id hid_quirks[] = { { HID_USB_DEVICE(USB_VENDOR_ID_SEMICO, USB_DEVICE_ID_SEMICO_USB_KEYKOARD), HID_QUIRK_NO_INIT_REPORTS }, { HID_USB_DEVICE(USB_VENDOR_ID_SENNHEISER, USB_DEVICE_ID_SENNHEISER_BTD500USB), HID_QUIRK_NOGET }, { HID_USB_DEVICE(USB_VENDOR_ID_SIGMA_MICRO, USB_DEVICE_ID_SIGMA_MICRO_KEYBOARD), HID_QUIRK_NO_INIT_REPORTS }, + { HID_USB_DEVICE(USB_VENDOR_ID_SIGMA_MICRO, USB_DEVICE_ID_SIGMA_MICRO_USB_MOUSE), HID_QUIRK_ALWAYS_POLL }, { HID_USB_DEVICE(USB_VENDOR_ID_SIGMATEL, USB_DEVICE_ID_SIGMATEL_STMP3780), HID_QUIRK_NOGET }, { HID_USB_DEVICE(USB_VENDOR_ID_SIS_TOUCH, USB_DEVICE_ID_SIS1030_TOUCH), HID_QUIRK_NOGET }, { HID_USB_DEVICE(USB_VENDOR_ID_SIS_TOUCH, USB_DEVICE_ID_SIS817_TOUCH), HID_QUIRK_NOGET }, From e6970cda63fd4b4546aeed9d0e2f53a7c95cd09c Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Wed, 13 May 2026 16:53:09 +0800 Subject: [PATCH 097/411] usb: cdns3: plat: fix leaked usb2_phy initialization on usb3_phy acquisition failure Move usb2_phy initialization after usb3_phy acquisition. Fixes: f738957277ba ("usb: cdns3: Split core.c into cdns3-plat and core.c file") Cc: stable Reported-by: sashiko-bot Closes: https://lore.kernel.org/linux-devicetree/agKaEePSFknhDBg2@nchen-desktop/T/#m21e1d9c1574eb127ce03c0c2a1a49002ce435b52 Signed-off-by: Peter Chen Link: https://patch.msgid.link/20260513085310.2217547-2-peter.chen@cixtech.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/cdns3/cdns3-plat.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/usb/cdns3/cdns3-plat.c b/drivers/usb/cdns3/cdns3-plat.c index 735df88774e4..d2e8d1e9007b 100644 --- a/drivers/usb/cdns3/cdns3-plat.c +++ b/drivers/usb/cdns3/cdns3-plat.c @@ -126,15 +126,15 @@ static int cdns3_plat_probe(struct platform_device *pdev) return dev_err_probe(dev, PTR_ERR(cdns->usb2_phy), "Failed to get cdn3,usb2-phy\n"); - ret = phy_init(cdns->usb2_phy); - if (ret) - return ret; - cdns->usb3_phy = devm_phy_optional_get(dev, "cdns3,usb3-phy"); if (IS_ERR(cdns->usb3_phy)) return dev_err_probe(dev, PTR_ERR(cdns->usb3_phy), "Failed to get cdn3,usb3-phy\n"); + ret = phy_init(cdns->usb2_phy); + if (ret) + return ret; + ret = phy_init(cdns->usb3_phy); if (ret) goto err_phy3_init; From ae6f3b82324e4f39ad8443c9020787e6fc889637 Mon Sep 17 00:00:00 2001 From: Peter Chen Date: Wed, 13 May 2026 16:53:10 +0800 Subject: [PATCH 098/411] usb: cdns3: plat: fix unbalanced pm_runtime_forbid() call permanently leaks the runtime PM usage counter across bind/unbind cycles Call pm_runtime_allow(dev) conditionally at cdns3_plat_remove. Fixes: f738957277ba ("usb: cdns3: Split core.c into cdns3-plat and core.c file") Cc: stable Reported-by: sashiko-bot Closes: https://lore.kernel.org/linux-devicetree/agKaEePSFknhDBg2@nchen-desktop/T/#m21e1d9c1574eb127ce03c0c2a1a49002ce435b52 Signed-off-by: Peter Chen Link: https://patch.msgid.link/20260513085310.2217547-3-peter.chen@cixtech.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/cdns3/cdns3-plat.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/usb/cdns3/cdns3-plat.c b/drivers/usb/cdns3/cdns3-plat.c index d2e8d1e9007b..94e9706a1806 100644 --- a/drivers/usb/cdns3/cdns3-plat.c +++ b/drivers/usb/cdns3/cdns3-plat.c @@ -186,6 +186,9 @@ static void cdns3_plat_remove(struct platform_device *pdev) struct device *dev = cdns->dev; pm_runtime_get_sync(dev); + if (!(cdns->pdata && (cdns->pdata->quirks & CDNS3_DEFAULT_PM_RUNTIME_ALLOW))) + pm_runtime_allow(dev); + pm_runtime_disable(dev); pm_runtime_put_noidle(dev); cdns_remove(cdns); From c8778ff817a7047d6848fefba99dcb27b1bf01fe Mon Sep 17 00:00:00 2001 From: Yongchao Wu Date: Thu, 14 May 2026 00:00:12 +0800 Subject: [PATCH 099/411] usb: cdns3: gadget: fix request skipping after clearing halt According to the cdns3 datasheet, the EPRST (Endpoint Reset) command causes the DMA engine to reposition its internal pointer to the next Transfer Descriptor (TD) if it was already processing one. This issue is consistently observed during the ADB identification process on macOS hosts, where the host issues a Clear_Halt. Although commit 4bf2dd65135a ("usb: cdns3: gadget: toggle cycle bit before reset endpoint") attempted to avoid DMA advance by toggling the cycle bit, trace logs show that on certain hosts like macOS, the DMA pointer (EP_TRADDR) still shifts after EPRST: cdns3_ctrl_req: Clear Endpoint Feature(Halt ep1out) cdns3_doorbell_epx: ep1out, ep_trbaddr f9c04030 <-- Should be f9c04000 cdns3_gadget_giveback: ep1out: req: ... length: 16384/16384 As shown above, the DMA pointer jumped to the next TD, causing the controller to skip the initial TRBs of the request. This leads to data misalignment and ADB protocol hangs on macOS. Fix this by manually restoring the EP_TRADDR register to the starting physical address of the current request after the EPRST operation is complete. Fixes: 7733f6c32e36 ("usb: cdns3: Add Cadence USB3 DRD Driver") Cc: stable Cc: Peter Chen Signed-off-by: Yongchao Wu Acked-by: Peter Chen Link: https://patch.msgid.link/20260513160012.2547894-1-yongchao.wu@autochips.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/cdns3/cdns3-gadget.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/usb/cdns3/cdns3-gadget.c b/drivers/usb/cdns3/cdns3-gadget.c index 8382231af357..1db8db1b7cc3 100644 --- a/drivers/usb/cdns3/cdns3-gadget.c +++ b/drivers/usb/cdns3/cdns3-gadget.c @@ -2817,9 +2817,19 @@ int __cdns3_gadget_ep_clear_halt(struct cdns3_endpoint *priv_ep) priv_ep->flags &= ~(EP_STALLED | EP_STALL_PENDING); if (request) { - if (trb) + if (trb) { *trb = trb_tmp; + /* + * Per datasheet, EPRST causes DMA to reposition to the next TD. + * Manually reset EP_TRADDR to the current TRB to prevent + * the hardware from skipping the interrupted request. + */ + writel(EP_TRADDR_TRADDR(priv_ep->trb_pool_dma + + priv_req->start_trb * TRB_SIZE), + &priv_dev->regs->ep_traddr); + } + cdns3_rearm_transfer(priv_ep, 1); } From c06e6cd488194e37ed4dc29d1488d1ffb760de60 Mon Sep 17 00:00:00 2001 From: Sebastian Reichel Date: Wed, 29 Apr 2026 18:33:32 +0200 Subject: [PATCH 100/411] usb: typec: tcpm: improve handling of DISCOVER_MODES failures UGREEN USB-C Multifunction Adapter Model CM512 (AKA "Revodok 107") exposes two SVIDs: 0xff01 (DP Alt Mode) and 0x1d5c. The DISCOVER_MODES step succeeds for 0xff01 and gets a NAK for 0x1d5c. Currently this results in DP Alt Mode not being registered either, since the modes are only registered once all of them have been discovered. The NAK results in the processing being stopped and thus no Alt modes being registered. Improve the situation by handling the NAK gracefully and continue processing the other modes. Before this change, the TCPM log ends like this: (more log entries before this) [ 5.028287] AMS DISCOVER_SVIDS finished [ 5.028291] cc:=4 [ 5.040040] SVID 1: 0xff01 [ 5.040054] SVID 2: 0x1d5c [ 5.040082] AMS DISCOVER_MODES start [ 5.040096] PD TX, header: 0x1b6f [ 5.050946] PD TX complete, status: 0 [ 5.059609] PD RX, header: 0x264f [1] [ 5.059626] Rx VDM cmd 0xff018043 type 1 cmd 3 len 2 [ 5.059640] AMS DISCOVER_MODES finished [ 5.059644] cc:=4 [ 5.069994] Alternate mode 0: SVID 0xff01, VDO 1: 0x000c0045 [ 5.070029] AMS DISCOVER_MODES start [ 5.070043] PD TX, header: 0x1d6f [ 5.081139] PD TX complete, status: 0 [ 5.087498] PD RX, header: 0x184f [1] [ 5.087515] Rx VDM cmd 0x1d5c8083 type 2 cmd 3 len 1 [ 5.087529] AMS DISCOVER_MODES finished [ 5.087534] cc:=4 (no further log entries after this point) After this patch the TCPM log looks exactly the same, but then continues like this: [ 5.100222] Skip SVID 0x1d5c (failed to discover mode) [ 5.101699] AMS DFP_TO_UFP_ENTER_MODE start (log goes on as the system initializes DP AltMode) Cc: stable Fixes: 41d9d75344d9 ("usb: typec: tcpm: add discover svids and discover modes support for sop'") Reviewed-by: Heikki Krogerus Signed-off-by: Sebastian Reichel Reviewed-by: RD Babiera Reviewed-by: Badhri Jagan Sridharan Link: https://patch.msgid.link/20260429-tcpm-discover-modes-nak-fix-v4-1-75945d0ed30f@collabora.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/tcpm/tcpm.c | 97 ++++++++++++++++++++++------------- 1 file changed, 61 insertions(+), 36 deletions(-) diff --git a/drivers/usb/typec/tcpm/tcpm.c b/drivers/usb/typec/tcpm/tcpm.c index ed5f745a8231..7ef746a90a17 100644 --- a/drivers/usb/typec/tcpm/tcpm.c +++ b/drivers/usb/typec/tcpm/tcpm.c @@ -2149,6 +2149,55 @@ static bool tcpm_cable_vdm_supported(struct tcpm_port *port) tcpm_can_communicate_sop_prime(port); } +static int tcpm_handle_discover_mode(struct tcpm_port *port, u32 *response, + enum tcpm_transmit_type rx_sop_type, + enum tcpm_transmit_type *response_tx_sop_type) +{ + struct typec_port *typec = port->typec_port; + struct pd_mode_data *modep; + + if (rx_sop_type == TCPC_TX_SOP) { + modep = &port->mode_data; + modep->svid_index++; + + if (modep->svid_index < modep->nsvids) { + u16 svid = modep->svids[modep->svid_index]; + *response_tx_sop_type = TCPC_TX_SOP; + response[0] = VDO(svid, 1, + typec_get_negotiated_svdm_version(typec), + CMD_DISCOVER_MODES); + return 1; + } + + if (tcpm_cable_vdm_supported(port)) { + *response_tx_sop_type = TCPC_TX_SOP_PRIME; + response[0] = VDO(USB_SID_PD, 1, + typec_get_cable_svdm_version(typec), + CMD_DISCOVER_SVID); + return 1; + } + + tcpm_register_partner_altmodes(port); + } else if (rx_sop_type == TCPC_TX_SOP_PRIME) { + modep = &port->mode_data_prime; + modep->svid_index++; + + if (modep->svid_index < modep->nsvids) { + u16 svid = modep->svids[modep->svid_index]; + *response_tx_sop_type = TCPC_TX_SOP_PRIME; + response[0] = VDO(svid, 1, + typec_get_cable_svdm_version(typec), + CMD_DISCOVER_MODES); + return 1; + } + + tcpm_register_plug_altmodes(port); + tcpm_register_partner_altmodes(port); + } + + return 0; +} + static int tcpm_pd_svdm(struct tcpm_port *port, struct typec_altmode *adev, const u32 *p, int cnt, u32 *response, enum adev_actions *adev_action, @@ -2406,41 +2455,11 @@ static int tcpm_pd_svdm(struct tcpm_port *port, struct typec_altmode *adev, } break; case CMD_DISCOVER_MODES: - if (rx_sop_type == TCPC_TX_SOP) { - /* 6.4.4.3.3 */ - svdm_consume_modes(port, p, cnt, rx_sop_type); - modep->svid_index++; - if (modep->svid_index < modep->nsvids) { - u16 svid = modep->svids[modep->svid_index]; - *response_tx_sop_type = TCPC_TX_SOP; - response[0] = VDO(svid, 1, svdm_version, - CMD_DISCOVER_MODES); - rlen = 1; - } else if (tcpm_cable_vdm_supported(port)) { - *response_tx_sop_type = TCPC_TX_SOP_PRIME; - response[0] = VDO(USB_SID_PD, 1, - typec_get_cable_svdm_version(typec), - CMD_DISCOVER_SVID); - rlen = 1; - } else { - tcpm_register_partner_altmodes(port); - } - } else if (rx_sop_type == TCPC_TX_SOP_PRIME) { - /* 6.4.4.3.3 */ - svdm_consume_modes(port, p, cnt, rx_sop_type); - modep_prime->svid_index++; - if (modep_prime->svid_index < modep_prime->nsvids) { - u16 svid = modep_prime->svids[modep_prime->svid_index]; - *response_tx_sop_type = TCPC_TX_SOP_PRIME; - response[0] = VDO(svid, 1, - typec_get_cable_svdm_version(typec), - CMD_DISCOVER_MODES); - rlen = 1; - } else { - tcpm_register_plug_altmodes(port); - tcpm_register_partner_altmodes(port); - } - } + /* 6.4.4.3.3 */ + svdm_consume_modes(port, p, cnt, rx_sop_type); + rlen = tcpm_handle_discover_mode(port, response, + rx_sop_type, + response_tx_sop_type); break; case CMD_ENTER_MODE: *response_tx_sop_type = rx_sop_type; @@ -2483,9 +2502,15 @@ static int tcpm_pd_svdm(struct tcpm_port *port, struct typec_altmode *adev, switch (cmd) { case CMD_DISCOVER_IDENT: case CMD_DISCOVER_SVID: - case CMD_DISCOVER_MODES: case VDO_CMD_VENDOR(0) ... VDO_CMD_VENDOR(15): break; + case CMD_DISCOVER_MODES: + tcpm_log(port, "Skip SVID 0x%04x (failed to discover mode)", + PD_VDO_SVID_SVID0(p[0])); + rlen = tcpm_handle_discover_mode(port, response, + rx_sop_type, + response_tx_sop_type); + break; case CMD_ENTER_MODE: /* Back to USB Operation */ *adev_action = ADEV_NOTIFY_USB_AND_QUEUE_VDM; From b80e7d34c7ea6a564525119d6138fbb577a23dba Mon Sep 17 00:00:00 2001 From: Myrrh Periwinkle Date: Tue, 19 May 2026 18:41:39 +0700 Subject: [PATCH 101/411] usb: typec: ucsi: Check if power role change actually happened before handling The CrOS EC may send a connector status change event with the power direction changed flag set even if the power direction hasn't actually changed after initiating a SET_PDR command internally [1]. In practice this happens on every system suspend due to other changes performed by the EC [2][3][4], causing suspend to fail. Fix this by checking if the power role change actually happened before handling it. [1]: https://source.chromium.org/chromiumos/chromiumos/codesearch/+/main:src/platform/ec/zephyr/subsys/pd_controller/pdc_power_mgmt.c;l=1689;drc=2d5a1cffce4e5ac8a39442cb3b764d2d5e1cf794 [2]: https://source.chromium.org/chromiumos/chromiumos/codesearch/+/main:src/platform/ec/zephyr/subsys/pd_controller/pdc_power_mgmt.c;l=3923;drc=2d5a1cffce4e5ac8a39442cb3b764d2d5e1cf794 [3]: https://source.chromium.org/chromiumos/chromiumos/codesearch/+/main:src/platform/ec/zephyr/subsys/pd_controller/pdc_power_mgmt.c;l=5094;drc=2d5a1cffce4e5ac8a39442cb3b764d2d5e1cf794 [4]: https://source.chromium.org/chromiumos/chromiumos/codesearch/+/main:src/platform/ec/zephyr/subsys/pd_controller/pdc_power_mgmt.c;l=2229;drc=2d5a1cffce4e5ac8a39442cb3b764d2d5e1cf794 Cc: stable Fixes: 7616f006db07 ("usb: typec: ucsi: Update power_supply on power role change") Signed-off-by: Myrrh Periwinkle Reported-and-tested-by: Sergey Senozhatsky Reviewed-by: Heikki Krogerus Link: https://patch.msgid.link/20260519-ucsi-fix-2-v1-1-6f1239535187@qtmlabs.xyz Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/ucsi.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c index 539dc706798d..53c101bd48e2 100644 --- a/drivers/usb/typec/ucsi/ucsi.c +++ b/drivers/usb/typec/ucsi/ucsi.c @@ -1277,7 +1277,7 @@ static void ucsi_handle_connector_change(struct work_struct *work) work); struct ucsi *ucsi = con->ucsi; u8 curr_scale, volt_scale; - enum typec_role role; + enum typec_role role, prev_role; u16 change; int ret; u32 val; @@ -1288,6 +1288,8 @@ static void ucsi_handle_connector_change(struct work_struct *work) dev_err_once(ucsi->dev, "%s entered without EVENT_PENDING\n", __func__); + prev_role = UCSI_CONSTAT(con, PWR_DIR); + ret = ucsi_get_connector_status(con, true); if (ret) { dev_err(ucsi->dev, "%s: GET_CONNECTOR_STATUS failed (%d)\n", @@ -1304,7 +1306,7 @@ static void ucsi_handle_connector_change(struct work_struct *work) change = UCSI_CONSTAT(con, CHANGE); role = UCSI_CONSTAT(con, PWR_DIR); - if (change & UCSI_CONSTAT_POWER_DIR_CHANGE) { + if ((change & UCSI_CONSTAT_POWER_DIR_CHANGE) && role != prev_role) { typec_set_pwr_role(con->port, role); ucsi_port_psy_changed(con); From d98d413ca65d0790a8f3695d0a5845538958ab84 Mon Sep 17 00:00:00 2001 From: Myrrh Periwinkle Date: Tue, 19 May 2026 18:41:40 +0700 Subject: [PATCH 102/411] usb: typec: ucsi: Don't update power_supply on power role change if not connected We only need to update the power_supply on power role change if the port is connected, because otherwise the online status should be the same for both cases. Cc: stable Fixes: 7616f006db07 ("usb: typec: ucsi: Update power_supply on power role change") Signed-off-by: Myrrh Periwinkle Reported-and-tested-by: Sergey Senozhatsky Link: https://patch.msgid.link/20260519-ucsi-fix-2-v1-2-6f1239535187@qtmlabs.xyz Signed-off-by: Greg Kroah-Hartman --- drivers/usb/typec/ucsi/ucsi.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/usb/typec/ucsi/ucsi.c b/drivers/usb/typec/ucsi/ucsi.c index 53c101bd48e2..61cb24ed820f 100644 --- a/drivers/usb/typec/ucsi/ucsi.c +++ b/drivers/usb/typec/ucsi/ucsi.c @@ -1308,7 +1308,12 @@ static void ucsi_handle_connector_change(struct work_struct *work) if ((change & UCSI_CONSTAT_POWER_DIR_CHANGE) && role != prev_role) { typec_set_pwr_role(con->port, role); - ucsi_port_psy_changed(con); + + /* Some power_supply properties vary depending on the power direction when + * connected + */ + if (UCSI_CONSTAT(con, CONNECTED)) + ucsi_port_psy_changed(con); /* Complete pending power role swap */ if (!completion_done(&con->complete)) From 373452ac0649846431ca0f88574a2fa6382d2045 Mon Sep 17 00:00:00 2001 From: Vincent Donnefort Date: Wed, 20 May 2026 23:08:30 +0100 Subject: [PATCH 103/411] KVM: arm64: Fix CONFIG_PKVM_DISABLE_STAGE2_ON_PANIC A typo in the config guard in __hyp_do_panic broke the stage-2 disabling and made backtraces for pKVM quite unreliable. Fix that typo. Fixes: 9019e82c7e46 ("KVM: arm64: Add PKVM_DISABLE_STAGE2_ON_PANIC") Signed-off-by: Vincent Donnefort Link: https://patch.msgid.link/20260520220830.273289-1-vdonnefort@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp/nvhe/host.S | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/nvhe/host.S b/arch/arm64/kvm/hyp/nvhe/host.S index f337770ec459..9393fe3ea6a1 100644 --- a/arch/arm64/kvm/hyp/nvhe/host.S +++ b/arch/arm64/kvm/hyp/nvhe/host.S @@ -120,7 +120,7 @@ SYM_FUNC_START(__hyp_do_panic) mov x29, x0 -#ifdef PKVM_DISABLE_STAGE2_ON_PANIC +#ifdef CONFIG_PKVM_DISABLE_STAGE2_ON_PANIC /* Ensure host stage-2 is disabled */ mrs x0, hcr_el2 bic x0, x0, #HCR_VM From c7ee0b73c8c4dfb7eafa49aaef5247890862a948 Mon Sep 17 00:00:00 2001 From: Kean Date: Thu, 14 May 2026 20:58:38 +0800 Subject: [PATCH 104/411] HID: lenovo: Fix buffer over-read and unaligned access in X12 Tab raw_event handler In lenovo_raw_event(), the X12 Tab keyboard handler reads a 4-byte little-endian value from the raw HID report buffer but: 1. The size guard is size >= 3, while the access reads 4 bytes. A malformed 3-byte report with ID 0x03 would over-read the buffer by one byte. 2. Casting u8 *data directly to __le32 * can trigger unaligned access faults on architectures like ARM, MIPS, and SPARC, because HID input buffers carry no alignment guarantee. (e.g. uhid payloads start at offset 6 in struct uhid_event, giving only 2-byte alignment.) Fix both by tightening the size check to >= 4 and replacing the open-coded cast + le32_to_cpu() with get_unaligned_le32(), which handles the LE-to-CPU conversion safely regardless of alignment. Link: https://sashiko.dev/#/message/20260512044911.99B6DC2BCB0%40smtp.kernel.org Assisted-by: CLAUDE:claude-4-sonnet Signed-off-by: Kean Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-lenovo.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/hid/hid-lenovo.c b/drivers/hid/hid-lenovo.c index a6b73e03c16b..c11957ae8b77 100644 --- a/drivers/hid/hid-lenovo.c +++ b/drivers/hid/hid-lenovo.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include "hid-ids.h" @@ -793,8 +794,8 @@ static int lenovo_raw_event(struct hid_device *hdev, */ if (unlikely((hdev->product == USB_DEVICE_ID_LENOVO_X12_TAB || hdev->product == USB_DEVICE_ID_LENOVO_X12_TAB2) - && size >= 3 && report->id == 0x03)) - return lenovo_raw_event_TP_X12_tab(hdev, le32_to_cpu(*(__le32 *)data)); + && size >= 4 && report->id == 0x03)) + return lenovo_raw_event_TP_X12_tab(hdev, get_unaligned_le32(data)); return 0; } From 2ee7e632405b022319f42c01635eb6fbbd86414a Mon Sep 17 00:00:00 2001 From: Louis Clinckx Date: Fri, 15 May 2026 14:57:39 +0000 Subject: [PATCH 105/411] HID: lenovo-go: reject non-USB transports in probe These drivers only match HID_USB_DEVICE() entries and assume the underlying bus is USB. Make that explicit at probe by rejecting any non-USB hdev, following the pattern used by other HID drivers. Signed-off-by: Louis Clinckx Reviewed-by: Derek J. Clark Tested-by: Derek J. Clark Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-lenovo-go-s.c | 3 +++ drivers/hid/hid-lenovo-go.c | 3 +++ 2 files changed, 6 insertions(+) diff --git a/drivers/hid/hid-lenovo-go-s.c b/drivers/hid/hid-lenovo-go-s.c index ff1782a75191..0444d84498bd 100644 --- a/drivers/hid/hid-lenovo-go-s.c +++ b/drivers/hid/hid-lenovo-go-s.c @@ -1461,6 +1461,9 @@ static int hid_gos_probe(struct hid_device *hdev, { int ret, ep; + if (!hid_is_usb(hdev)) + return -EINVAL; + ret = hid_parse(hdev); if (ret) { hid_err(hdev, "Parse failed\n"); diff --git a/drivers/hid/hid-lenovo-go.c b/drivers/hid/hid-lenovo-go.c index d4d26c783356..3fa1fe83f7e5 100644 --- a/drivers/hid/hid-lenovo-go.c +++ b/drivers/hid/hid-lenovo-go.c @@ -2419,6 +2419,9 @@ static int hid_go_probe(struct hid_device *hdev, const struct hid_device_id *id) { int ret, ep; + if (!hid_is_usb(hdev)) + return -EINVAL; + hdev->quirks |= HID_QUIRK_INPUT_PER_APP | HID_QUIRK_MULTI_INPUT; ret = hid_parse(hdev); From da7f96a68c39de9eb1c351a261e7fbf716375c91 Mon Sep 17 00:00:00 2001 From: Louis Clinckx Date: Fri, 15 May 2026 14:57:40 +0000 Subject: [PATCH 106/411] HID: lenovo-go: drop dead NULL check on to_usb_interface() to_usb_interface() is a container_of_const() macro: it performs pointer arithmetic and never returns NULL. The if (!intf) and if (intf) tests in get_endpoint_address() can never fire. Remove them in both drivers. No functional change. Suggested-by: Derek J. Clark Signed-off-by: Louis Clinckx Reviewed-by: Derek J. Clark Tested-by: Derek J. Clark Signed-off-by: Benjamin Tissoires --- drivers/hid/hid-lenovo-go-s.c | 8 +++----- drivers/hid/hid-lenovo-go.c | 3 --- 2 files changed, 3 insertions(+), 8 deletions(-) diff --git a/drivers/hid/hid-lenovo-go-s.c b/drivers/hid/hid-lenovo-go-s.c index 0444d84498bd..a72f7f748cb5 100644 --- a/drivers/hid/hid-lenovo-go-s.c +++ b/drivers/hid/hid-lenovo-go-s.c @@ -382,11 +382,9 @@ static int get_endpoint_address(struct hid_device *hdev) struct usb_interface *intf = to_usb_interface(hdev->dev.parent); struct usb_host_endpoint *ep; - if (intf) { - ep = intf->cur_altsetting->endpoint; - if (ep) - return ep->desc.bEndpointAddress; - } + ep = intf->cur_altsetting->endpoint; + if (ep) + return ep->desc.bEndpointAddress; return -ENODEV; } diff --git a/drivers/hid/hid-lenovo-go.c b/drivers/hid/hid-lenovo-go.c index 3fa1fe83f7e5..e0c9d5ec9451 100644 --- a/drivers/hid/hid-lenovo-go.c +++ b/drivers/hid/hid-lenovo-go.c @@ -641,9 +641,6 @@ static int get_endpoint_address(struct hid_device *hdev) struct usb_interface *intf = to_usb_interface(hdev->dev.parent); struct usb_host_endpoint *ep; - if (!intf) - return -ENODEV; - ep = intf->cur_altsetting->endpoint; if (!ep) return -ENODEV; From fb6988b83b4cafe8db63999c1ddff1b7c66d2ff5 Mon Sep 17 00:00:00 2001 From: Florian Schmaus Date: Thu, 7 May 2026 10:48:54 +0200 Subject: [PATCH 107/411] kunit: fix use-after-free in debugfs when using kunit.filter When the kernel is booted with a kunit filter (e.g., kunit.filter="speed!=slow"), the kunit executor dynamically allocates copies of the filtered test suites using kmalloc/kmemdup. During the initial boot execution, kunit_debugfs_create_suite() creates debugfs files (such as /sys/kernel/debug/kunit//run) and permanently stores a pointer to the dynamically allocated suite in the inode's i_private field. Previously, the executor freed this dynamically allocated suite_set immediately after executing the boot-time tests. Because the debugfs nodes were not destroyed, any subsequent interaction with the debugfs `run` file from userspace triggered a use-after-free (UAF). On systems with architectural capabilities, like CHERI RISC-V, this resulted in an immediate fatal hardware exception due to the invalidation of the capability tags on the reclaimed memory. On other architectures, it resulted in silent memory corruption. Fix this UAF by properly coupling the lifetime of the filtered suite memory allocation to the lifetime of the kunit subsystem and its associated VFS nodes. Ownership of the boot-time suite_set is now transferred to a global tracker ('kunit_boot_suites'), and the memory is cleanly released in kunit_exit() during module teardown. Link: https://lore.kernel.org/r/20260507084854.233984-1-florian.schmaus@codasip.com Fixes: e2219db280e3 ("kunit: add debugfs /sys/kernel/debug/kunit//results display") Signed-off-by: Florian Schmaus Reviewed-by: Martin Kaiser Reviewed-by: David Gow Signed-off-by: Shuah Khan --- include/kunit/test.h | 1 + lib/kunit/executor.c | 19 ++++++++++++++++--- lib/kunit/test.c | 1 + 3 files changed, 18 insertions(+), 3 deletions(-) diff --git a/include/kunit/test.h b/include/kunit/test.h index 9cd1594ab697..ce0573e196ce 100644 --- a/include/kunit/test.h +++ b/include/kunit/test.h @@ -613,6 +613,7 @@ unsigned long kunit_vm_mmap(struct kunit *test, struct file *file, unsigned long offset); void kunit_cleanup(struct kunit *test); +void kunit_free_boot_suites(void); void __printf(2, 3) kunit_log_append(struct string_stream *log, const char *fmt, ...); diff --git a/lib/kunit/executor.c b/lib/kunit/executor.c index 1fef217de11d..b0f8a41d61d3 100644 --- a/lib/kunit/executor.c +++ b/lib/kunit/executor.c @@ -15,6 +15,16 @@ extern struct kunit_suite * const __kunit_suites_end[]; extern struct kunit_suite * const __kunit_init_suites_start[]; extern struct kunit_suite * const __kunit_init_suites_end[]; +static struct kunit_suite_set kunit_boot_suites; + +void kunit_free_boot_suites(void) +{ + if (kunit_boot_suites.start) { + kunit_free_suite_set(kunit_boot_suites); + kunit_boot_suites = (struct kunit_suite_set){ NULL, NULL }; + } +} + static char *action_param; module_param_named(action, action_param, charp, 0400); @@ -411,9 +421,12 @@ int kunit_run_all_tests(void) pr_err("kunit executor: unknown action '%s'\n", action_param); free_out: - if (filter_glob_param || filter_param) - kunit_free_suite_set(suite_set); - else if (init_num_suites > 0) + if (filter_glob_param || filter_param) { + if (err) + kunit_free_suite_set(suite_set); + else + kunit_boot_suites = suite_set; + } else if (init_num_suites > 0) /* Don't use kunit_free_suite_set because suites aren't individually allocated */ kfree(suite_set.start); diff --git a/lib/kunit/test.c b/lib/kunit/test.c index 41e1c89799b6..99773e000e1b 100644 --- a/lib/kunit/test.c +++ b/lib/kunit/test.c @@ -1075,6 +1075,7 @@ static void __exit kunit_exit(void) kunit_bus_shutdown(); kunit_debugfs_cleanup(); + kunit_free_boot_suites(); } module_exit(kunit_exit); From f706e6a4ce75585af979aec3dcbdce68bc76306b Mon Sep 17 00:00:00 2001 From: Dhabaleshwar Das Date: Thu, 21 May 2026 00:00:00 +0530 Subject: [PATCH 108/411] accel/rocket: fix UAF via dangling GEM handle in create_bo rocket_ioctl_create_bo() inserts a GEM handle into the file's IDR via drm_gem_handle_create() early on, then performs several operations that can fail (sgt allocation, drm_mm insert, iommu_map). If any fail after the handle is live, the error path calls drm_gem_shmem_object_free() which kfree's the object without removing the handle from the IDR. This leaves a dangling handle pointing to freed slab memory. Any subsequent ioctl using that handle (PREP_BO, FINI_BO, SUBMIT) calls drm_gem_object_lookup() and dereferences freed memory (UAF). Fix by moving drm_gem_handle_create() to after all fallible operations succeed, matching the pattern used by panfrost, lima, and etnaviv. Also fix drm_mm_insert_node_generic() whose return value was silently overwritten by iommu_map_sgtable() on the next line. Add the missing error check. [tomeu: Move handle creation to the very end] Fixes: 658ebeac3351 ("accel/rocket: Add IOCTL for BO creation") Reported-by: Dhabaleshwar Das Signed-off-by: Dhabaleshwar Das Reviewed-by: Tomeu Vizoso Link: https://patch.msgid.link/20260521165720.2113571-1-tomeu@tomeuvizoso.net Signed-off-by: Tomeu Vizoso --- drivers/accel/rocket/rocket_gem.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/drivers/accel/rocket/rocket_gem.c b/drivers/accel/rocket/rocket_gem.c index c8084719208a..a5fffa51ff35 100644 --- a/drivers/accel/rocket/rocket_gem.c +++ b/drivers/accel/rocket/rocket_gem.c @@ -79,11 +79,6 @@ int rocket_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file * rkt_obj->size = args->size; rkt_obj->offset = 0; - ret = drm_gem_handle_create(file, gem_obj, &args->handle); - drm_gem_object_put(gem_obj); - if (ret) - goto err; - sgt = drm_gem_shmem_get_pages_sgt(shmem_obj); if (IS_ERR(sgt)) { ret = PTR_ERR(sgt); @@ -95,6 +90,8 @@ int rocket_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file * rkt_obj->size, PAGE_SIZE, 0, 0); mutex_unlock(&rocket_priv->mm_lock); + if (ret) + goto err; ret = iommu_map_sgtable(rocket_priv->domain->domain, rkt_obj->mm.start, @@ -112,8 +109,18 @@ int rocket_ioctl_create_bo(struct drm_device *dev, void *data, struct drm_file * args->offset = drm_vma_node_offset_addr(&gem_obj->vma_node); args->dma_address = rkt_obj->mm.start; + ret = drm_gem_handle_create(file, gem_obj, &args->handle); + if (ret) + goto err_unmap; + + drm_gem_object_put(gem_obj); + return 0; +err_unmap: + iommu_unmap(rocket_priv->domain->domain, + rkt_obj->mm.start, rkt_obj->size); + err_remove_node: mutex_lock(&rocket_priv->mm_lock); drm_mm_remove_node(&rkt_obj->mm); From a8878e19d2f5205ad1f170fc230c2cc25a3b9390 Mon Sep 17 00:00:00 2001 From: Lizhi Hou Date: Wed, 20 May 2026 15:35:31 -0700 Subject: [PATCH 109/411] accel/amdxdna: Block running when IOMMU is off The AIE2 device firmware requires IOMMU on. Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5319 Reviewed-by: Mario Limonciello Signed-off-by: Lizhi Hou Link: https://patch.msgid.link/20260520223531.1403302-1-lizhi.hou@amd.com --- drivers/accel/amdxdna/aie2_pci.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/accel/amdxdna/aie2_pci.c b/drivers/accel/amdxdna/aie2_pci.c index f1ac4e00bd9f..4500b9ccb02e 100644 --- a/drivers/accel/amdxdna/aie2_pci.c +++ b/drivers/accel/amdxdna/aie2_pci.c @@ -511,6 +511,11 @@ static int aie2_init(struct amdxdna_dev *xdna) return -EINVAL; } + if (!xdna->group) { + XDNA_ERR(xdna, "Running without IOMMU not supported"); + return -EINVAL; + } + ndev = drmm_kzalloc(&xdna->ddev, sizeof(*ndev), GFP_KERNEL); if (!ndev) return -ENOMEM; From ea25e3c7915b24e0ef93ee85190f3fada037dfb1 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 21 Apr 2026 12:11:25 -0400 Subject: [PATCH 110/411] sunrpc: prevent out-of-bounds read in __cache_seq_start() Commit 7b546bd89975 ("sunrpc/cache: improve RCU safety in cache_list walking.") changed the tail of __cache_seq_start() to unconditionally store *pos = ((long long)hash << 32) + 1 before returning, dropping a prior "hash >= cd->hash_size" guard. When the while loop exits because every remaining bucket was empty, hash equals cd->hash_size, so the stored *pos is one position past the table's last valid bucket. seq_read_iter() caches that index in m->index. A subsequent pread(2) at the same file offset skips traverse() and hands the stored index back to __cache_seq_start(), which decodes hash = cd->hash_size and dereferences cd->hash_table[cd->hash_size] -- one hlist_head past the end of the kzalloc'd table. KASAN reports an eight-byte slab-out-of-bounds read at the tail of the 2048-byte hash_table allocation for the NFSD export cache (EXPORT_HASHMAX * sizeof(struct hlist_head) == 256 * 8). Reject an input hash that is out of range before touching the hash table. cache_seq_next() already bounds-checks its own loop; the start routine needs to be symmetric. Reported-by: syzbot+60cfa08822470bbebe44@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=60cfa08822470bbebe44 Fixes: 7b546bd89975 ("sunrpc/cache: improve RCU safety in cache_list walking.") Reviewed-by: Benjamin Coddington Reviewed-by: NeilBrown Signed-off-by: Chuck Lever --- net/sunrpc/cache.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index b5474ce534fb..27dd6b58b8ff 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -1348,6 +1348,9 @@ static void *__cache_seq_start(struct seq_file *m, loff_t *pos) hash = n >> 32; entry = n & ((1LL<<32) - 1); + if (hash >= cd->hash_size) + return NULL; + hlist_for_each_entry_rcu(ch, &cd->hash_table[hash], cache_list) if (!entry--) return ch; From fc151100098d2899b7aed99aa1bcfe27bf00d58d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 21 Apr 2026 15:20:21 -0400 Subject: [PATCH 111/411] NFSD: Report whether fh_key was actually updated The nfsd_ctl_fh_key_set tracepoint was introduced to capture operator activity on the filehandle signing key. Earlier revisions logged the key bytes verbatim; the version that landed hashes the 16 key bytes through crc32_le and stores the result. CRC32 is a linear projection of its input rather than a one-way function, and truncating any hash of fixed-size secret material leaves the key recoverable under offline brute force when the threat model includes an attacker with access to the trace ring. The operational question the fingerprint was meant to answer is whether a NFSD_CMD_THREADS_SET call that carries an NFSD_A_SERVER_FH_KEY attribute actually replaced the active key or re-installed the value already in place. Answer that question directly: compare the incoming key bytes against the current nn->fh_key inside nfsd_nl_fh_key_set() and surface a single bit to the tracepoint. The event now prints "updated" when the stored key changed and "unmodified" otherwise. A first set that fails kmalloc reports "unmodified" because no key was installed. Reported-by: jaeyeong Fixes: 62346217fd72 ("NFSD: Add a key for signing filehandles") Cc: Benjamin Coddington Reviewed-by: Benjamin Coddington Reviewed-by: Jeff Layton Signed-off-by: Chuck Lever --- fs/nfsd/nfsctl.c | 18 ++++++++++++++---- fs/nfsd/trace.h | 16 +++++++--------- 2 files changed, 21 insertions(+), 13 deletions(-) diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 39e7012a60d8..04e3954d54bd 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -1594,16 +1594,27 @@ int nfsd_nl_rpc_status_get_dumpit(struct sk_buff *skb, static int nfsd_nl_fh_key_set(const struct nlattr *attr, struct nfsd_net *nn) { siphash_key_t *fh_key = nn->fh_key; + u64 k0, k1; + bool changed; + + k0 = get_unaligned_le64(nla_data(attr)); + k1 = get_unaligned_le64(nla_data(attr) + 8); if (!fh_key) { fh_key = kmalloc(sizeof(siphash_key_t), GFP_KERNEL); - if (!fh_key) + if (!fh_key) { + trace_nfsd_ctl_fh_key_set(false, -ENOMEM); return -ENOMEM; + } nn->fh_key = fh_key; + changed = true; + } else { + changed = fh_key->key[0] != k0 || fh_key->key[1] != k1; } - fh_key->key[0] = get_unaligned_le64(nla_data(attr)); - fh_key->key[1] = get_unaligned_le64(nla_data(attr) + 8); + fh_key->key[0] = k0; + fh_key->key[1] = k1; + trace_nfsd_ctl_fh_key_set(changed, 0); return 0; } @@ -1682,7 +1693,6 @@ int nfsd_nl_threads_set_doit(struct sk_buff *skb, struct genl_info *info) attr = info->attrs[NFSD_A_SERVER_FH_KEY]; if (attr) { ret = nfsd_nl_fh_key_set(attr, nn); - trace_nfsd_ctl_fh_key_set((const char *)nn->fh_key, ret); if (ret) goto out_unlock; } diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index 5ad38f50836d..b631a472222b 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -2243,23 +2243,21 @@ TRACE_EVENT(nfsd_end_grace, TRACE_EVENT(nfsd_ctl_fh_key_set, TP_PROTO( - const char *key, + bool changed, int result ), - TP_ARGS(key, result), + TP_ARGS(changed, result), TP_STRUCT__entry( - __field(u32, key_hash) + __field(bool, changed) __field(int, result) ), TP_fast_assign( - if (key) - __entry->key_hash = ~crc32_le(0xFFFFFFFF, key, 16); - else - __entry->key_hash = 0; + __entry->changed = changed; __entry->result = result; ), - TP_printk("key=0x%08x result=%d", - __entry->key_hash, __entry->result + TP_printk("key %s, result=%d", + __entry->changed ? "updated" : "unmodified", + __entry->result ) ); From 0b474240327cebeff08ad429e8ed3cfc6c8ee816 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Tue, 28 Apr 2026 15:47:44 -0400 Subject: [PATCH 112/411] lockd: fix TEST handling when not all permissions are available. The F_GETLK fcntl can work with either read access or write access or both. It can query F_RDLCK and F_WRLCK locks in either case. However lockd currently treats F_GETLK similar to F_SETLK in that read access is required to query an F_RDLCK lock and write access is required to query a F_WRLCK lock. This is wrong and can cause problems - e.g. when qemu accesses a read-only (e.g. iso) filesystem image over NFS (though why it queries if it can get a write lock - I don't know. But it does, and this works with local filesystems). So we need TEST requests to be handled differently. To do this: - change nlm_do_fopen() to accept O_RDWR as a mode and in that case succeed if either a O_RDONLY or O_WRONLY file can be opened. - change nlm_lookup_file() to accept a mode argument from caller, instead of deducing base on lock time, and pass that on to nlm_do_fopen() - change nlm4svc_retrieve_args() and nlmsvc_retrieve_args() to detect TEST requests and pass O_RDWR as a mode to nlm_lookup_file, passing the same mode as before for other requests. Also set lock->fl.c.flc_file to whichever file is available for TEST requests. - change nlmsvc_testlock() to also not calculate the mode, but to use whatever was stored in lock->fl.c.flc_file. This behaviour of lockd - requesting O_WRONLY access to TEST for exclusive locks - has been present at least since git history began. However it was hidden until recently because knfsd ignored the access requested by lockd and required only READ access for all locking requests (unless the underlying filesystem provided an f_op->open function which checked access permissions). The commit mentioned in Fixes: below changed nfsd_permission() to NOT override the access request for LOCK requests and this exposed the bug that we are now fixing. Note that there is another issue that this patch does not address. The flock(.., LOCK_EX) call is permitted on a read-only file descriptor. Linux NFS maps this to NLM locking as whole-file byte-range locks. nfsd will see this as though it were fcntl( F_SETLK (F_WRLCK)) and will now require write access, which it might not be able to get. It is not clear if this is a problem in practice, or what the best solution might be. So no attempt is made to address it. Reported-by: Tj Link: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1128861 Fixes: 4cc9b9f2bf4d ("nfsd: refine and rename NFSD_MAY_LOCK") Reviewed-by: Jeff Layton Signed-off-by: NeilBrown Signed-off-by: Chuck Lever --- fs/lockd/lockd.h | 2 +- fs/lockd/svc4proc.c | 9 +++++++-- fs/lockd/svclock.c | 4 +--- fs/lockd/svcproc.c | 15 ++++++++++++--- fs/lockd/svcsubs.c | 31 +++++++++++++++++++++---------- 5 files changed, 42 insertions(+), 19 deletions(-) diff --git a/fs/lockd/lockd.h b/fs/lockd/lockd.h index a7c85ab6d4b5..1db6cb352542 100644 --- a/fs/lockd/lockd.h +++ b/fs/lockd/lockd.h @@ -332,7 +332,7 @@ int nlmsvc_dispatch(struct svc_rqst *rqstp); * File handling for the server personality */ __be32 nlm_lookup_file(struct svc_rqst *, struct nlm_file **, - struct nlm_lock *); + struct nlm_lock *, int); void nlm_release_file(struct nlm_file *); void nlmsvc_put_lockowner(struct nlm_lockowner *); void nlmsvc_release_lockowner(struct nlm_lock *); diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 5de41e249534..41cab858de57 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -146,8 +146,11 @@ nlm4svc_lookup_file(struct svc_rqst *rqstp, struct nlm_host *host, struct nlm_lock *lock, struct nlm_file **filp, struct nlm4_lock *xdr_lock, unsigned char type) { + bool is_test = (rqstp->rq_proc == NLMPROC4_TEST || + rqstp->rq_proc == NLMPROC4_TEST_MSG); struct file_lock *fl = &lock->fl; struct nlm_file *file = NULL; + int mode; __be32 error; if (xdr_lock->fh.len > NFS_MAXFHSIZE) @@ -170,7 +173,8 @@ nlm4svc_lookup_file(struct svc_rqst *rqstp, struct nlm_host *host, fl->c.flc_type = type; lockd_set_file_lock_range4(fl, lock->lock_start, lock->lock_len); - error = nlm_lookup_file(rqstp, &file, lock); + mode = is_test ? O_RDWR : lock_to_openmode(fl); + error = nlm_lookup_file(rqstp, &file, lock, mode); switch (error) { case nlm_granted: break; @@ -184,7 +188,8 @@ nlm4svc_lookup_file(struct svc_rqst *rqstp, struct nlm_host *host, *filp = file; fl->c.flc_flags = FL_POSIX; - fl->c.flc_file = file->f_file[lock_to_openmode(fl)]; + fl->c.flc_file = is_test ? nlmsvc_file_file(file) + : file->f_file[mode]; fl->c.flc_pid = current->tgid; fl->fl_lmops = &nlmsvc_lock_operations; nlmsvc_locks_init_private(fl, host, (pid_t)lock->svid); diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index ee23f5802af1..44bc20837062 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -613,7 +613,6 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file, struct nlm_lock *conflock) { int error; - int mode; __be32 ret; dprintk("lockd: nlmsvc_testlock(%s/%ld, ty=%d, %Ld-%Ld)\n", @@ -631,14 +630,13 @@ nlmsvc_testlock(struct svc_rqst *rqstp, struct nlm_file *file, goto out; } - mode = lock_to_openmode(&lock->fl); locks_init_lock(&conflock->fl); /* vfs_test_lock only uses start, end, and owner, but tests flc_file */ conflock->fl.c.flc_file = lock->fl.c.flc_file; conflock->fl.fl_start = lock->fl.fl_start; conflock->fl.fl_end = lock->fl.fl_end; conflock->fl.c.flc_owner = lock->fl.c.flc_owner; - error = vfs_test_lock(file->f_file[mode], &conflock->fl); + error = vfs_test_lock(lock->fl.c.flc_file, &conflock->fl); if (error) { ret = nlm_lck_denied_nolocks; goto out; diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 749abf8886ba..c0a3487719e2 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -68,6 +68,8 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, struct nlm_host *host = NULL; struct nlm_file *file = NULL; struct nlm_lock *lock = &argp->lock; + bool is_test = (rqstp->rq_proc == NLMPROC_TEST || + rqstp->rq_proc == NLMPROC_TEST_MSG); int mode; __be32 error = 0; @@ -83,15 +85,22 @@ nlmsvc_retrieve_args(struct svc_rqst *rqstp, struct nlm_args *argp, /* Obtain file pointer. Not used by FREE_ALL call. */ if (filp != NULL) { - error = cast_status(nlm_lookup_file(rqstp, &file, lock)); + mode = lock_to_openmode(&lock->fl); + + if (is_test) + mode = O_RDWR; + + error = cast_status(nlm_lookup_file(rqstp, &file, lock, mode)); if (error != 0) goto no_locks; *filp = file; /* Set up the missing parts of the file_lock structure */ - mode = lock_to_openmode(&lock->fl); lock->fl.c.flc_flags = FL_POSIX; - lock->fl.c.flc_file = file->f_file[mode]; + if (is_test) + lock->fl.c.flc_file = nlmsvc_file_file(file); + else + lock->fl.c.flc_file = file->f_file[mode]; lock->fl.c.flc_pid = current->tgid; lock->fl.fl_lmops = &nlmsvc_lock_operations; nlmsvc_locks_init_private(&lock->fl, host, (pid_t)lock->svid); diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 71eaec5ed8d7..976cc66d0c19 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -83,23 +83,36 @@ int lock_to_openmode(struct file_lock *lock) * * We have to make sure we have the right credential to open * the file. + * + * @mode is O_RDONLY, O_WRONLY, or O_RDWR. O_RDWR means success + * is achieved with EITHER O_RDONLY or O_WRONLY; it does not + * require both. */ static __be32 nlm_do_fopen(struct svc_rqst *rqstp, struct nlm_file *file, int mode) { - struct file **fp = &file->f_file[mode]; - __be32 nlmerr = nlm_granted; + __be32 nlmerr = nlm__int__failed; + __be32 deferred = 0; int error; + int m; - if (*fp) - return nlmerr; + for (m = O_RDONLY; m <= O_WRONLY; m++) { + struct file **fp = &file->f_file[m]; + + if (mode != O_RDWR && mode != m) + continue; + if (*fp) + return nlm_granted; + + error = nlmsvc_ops->fopen(rqstp, &file->f_handle, fp, m); + if (!error) + return nlm_granted; - error = nlmsvc_ops->fopen(rqstp, &file->f_handle, fp, mode); - if (error) { dprintk("lockd: open failed (errno %d)\n", error); switch (error) { case -EWOULDBLOCK: nlmerr = nlm__int__drop_reply; + deferred = nlmerr; break; case -ESTALE: nlmerr = nlm__int__stale_fh; @@ -110,7 +123,7 @@ static __be32 nlm_do_fopen(struct svc_rqst *rqstp, } } - return nlmerr; + return deferred ? deferred : nlmerr; } /* @@ -119,17 +132,15 @@ static __be32 nlm_do_fopen(struct svc_rqst *rqstp, */ __be32 nlm_lookup_file(struct svc_rqst *rqstp, struct nlm_file **result, - struct nlm_lock *lock) + struct nlm_lock *lock, int mode) { struct nlm_file *file; unsigned int hash; __be32 nfserr; - int mode; nlm_debug_print_fh("nlm_lookup_file", &lock->fh); hash = file_hash(&lock->fh); - mode = lock_to_openmode(&lock->fl); /* Lock file table */ mutex_lock(&nlm_file_mutex); From b60621c5121c9435eda99af7dc2100f5c0f88695 Mon Sep 17 00:00:00 2001 From: Emily Ehlert Date: Mon, 18 May 2026 13:59:56 +0000 Subject: [PATCH 113/411] KVM: x86: Fix ERAPS RAP clear on INVPCID single-context invalidation Use kvm_register_mark_dirty() instead of kvm_register_is_dirty() to actually mark VCPU_EXREG_ERAPS as dirty when emulating INVPCID_TYPE_SINGLE_CTXT. kvm_register_is_dirty() is a read-only predicate whose return value is discarded, making the call a no-op. Without this fix, a single-context INVPCID will not trigger a RAP clear on the next VMRUN, breaking the ERAPS security guarantee. Fixes: db5e82496492 ("KVM: SVM: Virtualize and advertise support for ERAPS") Signed-off-by: Emily Ehlert Link: https://patch.msgid.link/20260518135956.82569-1-ehemily@amazon.de Signed-off-by: Sean Christopherson --- arch/x86/kvm/x86.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e01d6984ed04..108318e1b3f0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -14330,7 +14330,7 @@ int kvm_handle_invpcid(struct kvm_vcpu *vcpu, unsigned long type, gva_t gva) * the RAP (Return Address Predicator). */ if (guest_cpu_cap_has(vcpu, X86_FEATURE_ERAPS)) - kvm_register_is_dirty(vcpu, VCPU_EXREG_ERAPS); + kvm_register_mark_dirty(vcpu, VCPU_EXREG_ERAPS); kvm_invalidate_pcid(vcpu, operand.pcid); return kvm_skip_emulated_instruction(vcpu); From a9e18aa3263f356edae305e29830e5fe63d8597a Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 15 May 2026 10:15:36 -0700 Subject: [PATCH 114/411] KVM: SVM: Flush the current TLB when transitioning from xAVIC => x2AVIC Flush the current TLB when xAVIC *or* x2AVIC is activated, as KVM is (apparently) responsible for purging TLB entries when transitioning from xAVIC to x2AVIC. The APM says a whole lot of nothing about TLB flushing with respect to (x2)AVIC, but empirical data strongly suggests hardware also does a whole lot of nothing. Failure to flush the TLB when enabling x2AVIC can lead to guest accesses to the APIC base address getting incorrectly redirected to the virtual APIC page. The flaw most visibly manifests as failures in KVM-Unit-Test's verify_disabled_apic_mmio() testcase when x2APIC is enabled (though for reasons unknown, the test only reliably fails with EFI builds). Fixes: 0ccf3e7cb95a ("KVM: SVM: Flush the "current" TLB when activating AVIC") Fixes: 4d1d7942e36a ("KVM: SVM: Introduce logic to (de)activate x2AVIC mode") Cc: stable@vger.kernel.org Cc: Naveen N Rao (AMD) Link: https://patch.msgid.link/20260515171536.1841645-1-seanjc@google.com Signed-off-by: Sean Christopherson --- arch/x86/kvm/svm/avic.c | 35 +++++++++++++++++++++++++++++------ 1 file changed, 29 insertions(+), 6 deletions(-) diff --git a/arch/x86/kvm/svm/avic.c b/arch/x86/kvm/svm/avic.c index adf211860949..e8bd60156941 100644 --- a/arch/x86/kvm/svm/avic.c +++ b/arch/x86/kvm/svm/avic.c @@ -206,6 +206,35 @@ static void avic_activate_vmcb(struct vcpu_svm *svm) svm_clr_intercept(svm, INTERCEPT_CR8_WRITE); + /* + * Flush the TLB when enabling (x2)AVIC and when transitioning between + * xAVIC and x2AVIC, as the CPU may have inserted a TLB entry for the + * "wrong" mapping. + * + * KVM uses a per-VM "scratch" page to back the APIC memslot, because + * KVM also uses per-VM page tables *and* maintains the page table (NPT + * or shadow page) mappings for said memslot even if one or more vCPUs + * have their local APIC hardware-disabled or are in x2APIC mode, i.e. + * even if one or more vCPUs' APIC MMIO BAR is effectively disabled. + * + * If xAVIC is fully enabled, hardware ignores the physical address in + * KVM's page tables, i.e. in the leaf SPTE for the APIC memslot, and + * instead redirects the access to the AVIC backing page, i.e. to the + * vCPU's virtual APIC page. If xAVIC is not enabled (APIC is either + * hardware-disabled or in x2APIC mode), then guest accesses will use + * the page table mapping verbatim, i.e. will access the per-VM scratch + * page, as normal memory. + * + * In both cases, the CPU is allowed to cache TLB entries for the APIC + * base GPA. So, KVM needs to flush the TLB when enabling xAVIC, as + * accesses need to be redirected to the virtual APIC page, but the TLB + * may contain entries pointing at the scratch page. KVM also needs to + * flush the TLB when enabling x2AVIC, as accesses need to go to the + * scratch page, but the TLB may contain entries tagged as xAVIC, i.e. + * entries pointing to the vCPU's virtual APIC page. + */ + kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, &svm->vcpu); + /* * Note: KVM supports hybrid-AVIC mode, where KVM emulates x2APIC MSR * accesses, while interrupt injection to a running vCPU can be @@ -219,12 +248,6 @@ static void avic_activate_vmcb(struct vcpu_svm *svm) /* Disabling MSR intercept for x2APIC registers */ avic_set_x2apic_msr_interception(svm, false); } else { - /* - * Flush the TLB, the guest may have inserted a non-APIC - * mapping into the TLB while AVIC was disabled. - */ - kvm_make_request(KVM_REQ_TLB_FLUSH_CURRENT, &svm->vcpu); - /* Enabling MSR intercept for x2APIC registers */ avic_set_x2apic_msr_interception(svm, true); } From 83ec6eeb74a592e6568cb0723bac99fb8b3810b4 Mon Sep 17 00:00:00 2001 From: Ian Ray Date: Wed, 6 May 2026 09:33:35 +0300 Subject: [PATCH 115/411] MAINTAINERS: .mailmap: update after GEHC spin-off Update my email address from @ge.com to @gehealthcare.com after GE HealthCare was spun-off from GE. Link: https://lore.kernel.org/20260506063335.3-1-ian.ray@gehealthcare.com Signed-off-by: Ian Ray Reviewed-by: Luca Ceresoli Cc: Neil Armstrong Signed-off-by: Andrew Morton --- .mailmap | 1 + MAINTAINERS | 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff --git a/.mailmap b/.mailmap index bd6a72f29d9c..de41cfdae5d0 100644 --- a/.mailmap +++ b/.mailmap @@ -339,6 +339,7 @@ Henrik Rydberg Herbert Xu Huacai Chen Huacai Chen +Ian Ray Ignat Korchagin Igor Korotin Ike Panhc diff --git a/MAINTAINERS b/MAINTAINERS index 10e8253181d3..9eb15dacb939 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -16504,7 +16504,7 @@ F: drivers/usb/mtu3/ MEGACHIPS STDPXXXX-GE-B850V3-FW LVDS/DP++ BRIDGES M: Peter Senna Tschudin -M: Ian Ray +M: Ian Ray M: Martyn Welch S: Maintained F: Documentation/devicetree/bindings/display/bridge/megachips-stdpxxxx-ge-b850v3-fw.txt From 83f9efcce93f8574be2279090ee2aec58b86cda7 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Tue, 12 May 2026 17:06:43 +0100 Subject: [PATCH 116/411] Revert "mm/hugetlbfs: update hugetlbfs to use mmap_prepare" This reverts commit ea52cb24cd3f ("mm/hugetlbfs: update hugetlbfs to use mmap_prepare") with conflict resolution to account for changes in commit ea52cb24cd3f ("mm/hugetlbfs: update hugetlbfs to use mmap_prepare"). The patch incorrectly handled hugetlb VMA lock allocation at the mmap_prepare stage, where a failed allocation occurring after mmap_prepare is called might result in the lock leaking. There is no risk of a merge causing a similar issues, as VMA_DONTEXPAND_BIT is set for hugetlb mappings. As a first step in addressing this issue, simply revert the change so we can rework how we do this having corrected the underlying issues. We maintain the VMA flags changes as best we can, accounting for the fact that we were working with a VMA descriptor previously and propagating like-for-like changes for this. Note that we invoke vma_set_flags() and do not call vma_start_write() as vm_flags_set() does. This is OK as it's being done in an .mmap hook where the VMA is not yet linked into the tree so nobody else can be accessing it. Link: https://lore.kernel.org/20260512160643.266960-1-ljs@kernel.org Fixes: ea52cb24cd3f ("mm/hugetlbfs: update hugetlbfs to use mmap_prepare") Signed-off-by: Lorenzo Stoakes Reported-by: Mingyu Wang <25181214217@stu.xidian.edu.cn> Closes: https://lore.kernel.org/linux-mm/20260425070700.562229-1-25181214217@stu.xidian.edu.cn/ Acked-by: Muchun Song Acked-by: Oscar Salvador Cc: David Hildenbrand Cc: Liam R. Howlett Cc: Pedro Falcato Cc: Signed-off-by: Andrew Morton --- fs/hugetlbfs/inode.c | 46 ++++++--------------- include/linux/hugetlb.h | 8 +--- include/linux/hugetlb_inline.h | 14 +------ mm/hugetlb.c | 75 ++++++++++++++-------------------- 4 files changed, 47 insertions(+), 96 deletions(-) diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 8b05bec08e04..78d61bf2bd9b 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -96,15 +96,8 @@ static const struct fs_parameter_spec hugetlb_fs_parameters[] = { #define PGOFF_LOFFT_MAX \ (((1UL << (PAGE_SHIFT + 1)) - 1) << (BITS_PER_LONG - (PAGE_SHIFT + 1))) -static int hugetlb_file_mmap_prepare_success(const struct vm_area_struct *vma) +static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) { - /* Unfortunate we have to reassign vma->vm_private_data. */ - return hugetlb_vma_lock_alloc((struct vm_area_struct *)vma); -} - -static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) -{ - struct file *file = desc->file; struct inode *inode = file_inode(file); loff_t len, vma_len; int ret; @@ -119,8 +112,8 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) * way when do_mmap unwinds (may be important on powerpc * and ia64). */ - vma_desc_set_flags(desc, VMA_HUGETLB_BIT, VMA_DONTEXPAND_BIT); - desc->vm_ops = &hugetlb_vm_ops; + vma_set_flags(vma, VMA_HUGETLB_BIT, VMA_DONTEXPAND_BIT); + vma->vm_ops = &hugetlb_vm_ops; /* * page based offset in vm_pgoff could be sufficiently large to @@ -129,16 +122,16 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) * sizeof(unsigned long). So, only check in those instances. */ if (sizeof(unsigned long) == sizeof(loff_t)) { - if (desc->pgoff & PGOFF_LOFFT_MAX) + if (vma->vm_pgoff & PGOFF_LOFFT_MAX) return -EINVAL; } /* must be huge page aligned */ - if (desc->pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) + if (vma->vm_pgoff & (~huge_page_mask(h) >> PAGE_SHIFT)) return -EINVAL; - vma_len = (loff_t)vma_desc_size(desc); - len = vma_len + ((loff_t)desc->pgoff << PAGE_SHIFT); + vma_len = (loff_t)(vma->vm_end - vma->vm_start); + len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT); /* check for overflow */ if (len < vma_len) return -EINVAL; @@ -148,7 +141,7 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) ret = -ENOMEM; - vma_flags = desc->vma_flags; + vma_flags = vma->flags; /* * for SHM_HUGETLB, the pages are reserved in the shmget() call so skip * reserving here. Note: only for SHM hugetlbfs file, the inode @@ -158,30 +151,17 @@ static int hugetlbfs_file_mmap_prepare(struct vm_area_desc *desc) vma_flags_set(&vma_flags, VMA_NORESERVE_BIT); if (hugetlb_reserve_pages(inode, - desc->pgoff >> huge_page_order(h), - len >> huge_page_shift(h), desc, - vma_flags) < 0) + vma->vm_pgoff >> huge_page_order(h), + len >> huge_page_shift(h), vma, + vma_flags) < 0) goto out; ret = 0; - if (vma_desc_test(desc, VMA_WRITE_BIT) && inode->i_size < len) + if (vma_test(vma, VMA_WRITE_BIT) && inode->i_size < len) i_size_write(inode, len); out: inode_unlock(inode); - if (!ret) { - /* Allocate the VMA lock after we set it up. */ - desc->action.success_hook = hugetlb_file_mmap_prepare_success; - /* - * We cannot permit the rmap finding this VMA in the time - * between the VMA being inserted into the VMA tree and the - * completion/success hook being invoked. - * - * This is because we establish a per-VMA hugetlb lock which can - * be raced by rmap. - */ - desc->action.hide_from_rmap_until_complete = true; - } return ret; } @@ -1227,7 +1207,7 @@ static void init_once(void *foo) static const struct file_operations hugetlbfs_file_operations = { .read_iter = hugetlbfs_read_iter, - .mmap_prepare = hugetlbfs_file_mmap_prepare, + .mmap = hugetlbfs_file_mmap, .fsync = noop_fsync, .get_unmapped_area = hugetlb_get_unmapped_area, .llseek = default_llseek, diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 93418625d3c5..5957bc25efa8 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -148,7 +148,7 @@ int hugetlb_mfill_atomic_pte(pte_t *dst_pte, struct folio **foliop); #endif /* CONFIG_USERFAULTFD */ long hugetlb_reserve_pages(struct inode *inode, long from, long to, - struct vm_area_desc *desc, vma_flags_t vma_flags); + struct vm_area_struct *vma, vma_flags_t vma_flags); long hugetlb_unreserve_pages(struct inode *inode, long start, long end, long freed); bool folio_isolate_hugetlb(struct folio *folio, struct list_head *list); @@ -276,7 +276,6 @@ long hugetlb_change_protection(struct vm_area_struct *vma, void hugetlb_unshare_all_pmds(struct vm_area_struct *vma); void fixup_hugetlb_reservations(struct vm_area_struct *vma); void hugetlb_split(struct vm_area_struct *vma, unsigned long addr); -int hugetlb_vma_lock_alloc(struct vm_area_struct *vma); unsigned int arch_hugetlb_cma_order(void); @@ -469,11 +468,6 @@ static inline void fixup_hugetlb_reservations(struct vm_area_struct *vma) static inline void hugetlb_split(struct vm_area_struct *vma, unsigned long addr) {} -static inline int hugetlb_vma_lock_alloc(struct vm_area_struct *vma) -{ - return 0; -} - #endif /* !CONFIG_HUGETLB_PAGE */ #ifndef pgd_write diff --git a/include/linux/hugetlb_inline.h b/include/linux/hugetlb_inline.h index 565b473fd135..5c29cd3223a1 100644 --- a/include/linux/hugetlb_inline.h +++ b/include/linux/hugetlb_inline.h @@ -6,23 +6,13 @@ #ifdef CONFIG_HUGETLB_PAGE -static inline bool is_vm_hugetlb_flags(vm_flags_t vm_flags) -{ - return !!(vm_flags & VM_HUGETLB); -} - static inline bool is_vma_hugetlb_flags(const vma_flags_t *flags) { - return vma_flags_test_any(flags, VMA_HUGETLB_BIT); + return vma_flags_test(flags, VMA_HUGETLB_BIT); } #else -static inline bool is_vm_hugetlb_flags(vm_flags_t vm_flags) -{ - return false; -} - static inline bool is_vma_hugetlb_flags(const vma_flags_t *flags) { return false; @@ -32,7 +22,7 @@ static inline bool is_vma_hugetlb_flags(const vma_flags_t *flags) static inline bool is_vm_hugetlb_page(const struct vm_area_struct *vma) { - return is_vm_hugetlb_flags(vma->vm_flags); + return is_vma_hugetlb_flags(&vma->flags); } #endif diff --git a/mm/hugetlb.c b/mm/hugetlb.c index f24bf49be047..4b80b167cc9c 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -116,6 +116,7 @@ struct mutex *hugetlb_fault_mutex_table __ro_after_init; /* Forward declaration */ static int hugetlb_acct_memory(struct hstate *h, long delta); static void hugetlb_vma_lock_free(struct vm_area_struct *vma); +static void hugetlb_vma_lock_alloc(struct vm_area_struct *vma); static void __hugetlb_vma_unlock_write_free(struct vm_area_struct *vma); static void hugetlb_unshare_pmds(struct vm_area_struct *vma, unsigned long start, unsigned long end, bool take_locks); @@ -413,21 +414,17 @@ static void hugetlb_vma_lock_free(struct vm_area_struct *vma) } } -/* - * vma specific semaphore used for pmd sharing and fault/truncation - * synchronization - */ -int hugetlb_vma_lock_alloc(struct vm_area_struct *vma) +static void hugetlb_vma_lock_alloc(struct vm_area_struct *vma) { struct hugetlb_vma_lock *vma_lock; /* Only establish in (flags) sharable vmas */ if (!vma || !(vma->vm_flags & VM_MAYSHARE)) - return 0; + return; /* Should never get here with non-NULL vm_private_data */ if (vma->vm_private_data) - return -EINVAL; + return; vma_lock = kmalloc_obj(*vma_lock); if (!vma_lock) { @@ -442,15 +439,13 @@ int hugetlb_vma_lock_alloc(struct vm_area_struct *vma) * allocation failure. */ pr_warn_once("HugeTLB: unable to allocate vma specific lock\n"); - return -EINVAL; + return; } kref_init(&vma_lock->refs); init_rwsem(&vma_lock->rw_sema); vma_lock->vma = vma; vma->vm_private_data = vma_lock; - - return 0; } /* Helper that removes a struct file_region from the resv_map cache and returns @@ -1147,30 +1142,22 @@ static struct resv_map *vma_resv_map(struct vm_area_struct *vma) } } +static void set_vma_resv_map(struct vm_area_struct *vma, struct resv_map *map) +{ + VM_WARN_ON_ONCE_VMA(!is_vm_hugetlb_page(vma), vma); + VM_WARN_ON_ONCE_VMA(vma_test(vma, VMA_MAYSHARE_BIT), vma); + + set_vma_private_data(vma, (unsigned long)map); +} + static void set_vma_resv_flags(struct vm_area_struct *vma, unsigned long flags) { VM_WARN_ON_ONCE_VMA(!is_vm_hugetlb_page(vma), vma); - VM_WARN_ON_ONCE_VMA(vma->vm_flags & VM_MAYSHARE, vma); + VM_WARN_ON_ONCE_VMA(vma_test(vma, VMA_MAYSHARE_BIT), vma); set_vma_private_data(vma, get_vma_private_data(vma) | flags); } -static void set_vma_desc_resv_map(struct vm_area_desc *desc, struct resv_map *map) -{ - VM_WARN_ON_ONCE(!is_vma_hugetlb_flags(&desc->vma_flags)); - VM_WARN_ON_ONCE(vma_desc_test(desc, VMA_MAYSHARE_BIT)); - - desc->private_data = map; -} - -static void set_vma_desc_resv_flags(struct vm_area_desc *desc, unsigned long flags) -{ - VM_WARN_ON_ONCE(!is_vma_hugetlb_flags(&desc->vma_flags)); - VM_WARN_ON_ONCE(vma_desc_test(desc, VMA_MAYSHARE_BIT)); - - desc->private_data = (void *)((unsigned long)desc->private_data | flags); -} - static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag) { VM_BUG_ON_VMA(!is_vm_hugetlb_page(vma), vma); @@ -1178,13 +1165,6 @@ static int is_vma_resv_set(struct vm_area_struct *vma, unsigned long flag) return (get_vma_private_data(vma) & flag) != 0; } -static bool is_vma_desc_resv_set(struct vm_area_desc *desc, unsigned long flag) -{ - VM_WARN_ON_ONCE(!is_vma_hugetlb_flags(&desc->vma_flags)); - - return ((unsigned long)desc->private_data) & flag; -} - bool __vma_private_lock(struct vm_area_struct *vma) { return !(vma->vm_flags & VM_MAYSHARE) && @@ -6553,7 +6533,7 @@ long hugetlb_change_protection(struct vm_area_struct *vma, long hugetlb_reserve_pages(struct inode *inode, long from, long to, - struct vm_area_desc *desc, + struct vm_area_struct *vma, vma_flags_t vma_flags) { long chg = -1, add = -1, spool_resv, gbl_resv; @@ -6570,6 +6550,12 @@ long hugetlb_reserve_pages(struct inode *inode, return -EINVAL; } + /* + * vma specific semaphore used for pmd sharing and fault/truncation + * synchronization + */ + hugetlb_vma_lock_alloc(vma); + /* * Only apply hugepage reservation if asked. At fault time, an * attempt will be made for VM_NORESERVE to allocate a page @@ -6582,9 +6568,9 @@ long hugetlb_reserve_pages(struct inode *inode, * Shared mappings base their reservation on the number of pages that * are already allocated on behalf of the file. Private mappings need * to reserve the full area even if read-only as mprotect() may be - * called to make the mapping read-write. Assume !desc is a shm mapping + * called to make the mapping read-write. Assume !vma is a shm mapping */ - if (!desc || vma_desc_test(desc, VMA_MAYSHARE_BIT)) { + if (!vma || vma_test(vma, VMA_MAYSHARE_BIT)) { /* * resv_map can not be NULL as hugetlb_reserve_pages is only * called for inodes for which resv_maps were created (see @@ -6603,8 +6589,8 @@ long hugetlb_reserve_pages(struct inode *inode, chg = to - from; - set_vma_desc_resv_map(desc, resv_map); - set_vma_desc_resv_flags(desc, HPAGE_RESV_OWNER); + set_vma_resv_map(vma, resv_map); + set_vma_resv_flags(vma, HPAGE_RESV_OWNER); } if (chg < 0) { @@ -6618,7 +6604,7 @@ long hugetlb_reserve_pages(struct inode *inode, if (err < 0) goto out_err; - if (desc && !vma_desc_test(desc, VMA_MAYSHARE_BIT) && h_cg) { + if (vma && !vma_test(vma, VMA_MAYSHARE_BIT) && h_cg) { /* For private mappings, the hugetlb_cgroup uncharge info hangs * of the resv_map. */ @@ -6655,7 +6641,7 @@ long hugetlb_reserve_pages(struct inode *inode, * consumed reservations are stored in the map. Hence, nothing * else has to be done for private mappings here */ - if (!desc || vma_desc_test(desc, VMA_MAYSHARE_BIT)) { + if (!vma || vma_test(vma, VMA_MAYSHARE_BIT)) { add = region_add(resv_map, from, to, regions_needed, h, h_cg); if (unlikely(add < 0)) { @@ -6719,15 +6705,16 @@ long hugetlb_reserve_pages(struct inode *inode, hugetlb_cgroup_uncharge_cgroup_rsvd(hstate_index(h), chg * pages_per_huge_page(h), h_cg); out_err: - if (!desc || vma_desc_test(desc, VMA_MAYSHARE_BIT)) + hugetlb_vma_lock_free(vma); + if (!vma || vma_test(vma, VMA_MAYSHARE_BIT)) /* Only call region_abort if the region_chg succeeded but the * region_add failed or didn't run. */ if (chg >= 0 && add < 0) region_abort(resv_map, from, to, regions_needed); - if (desc && is_vma_desc_resv_set(desc, HPAGE_RESV_OWNER)) { + if (vma && is_vma_resv_set(vma, HPAGE_RESV_OWNER)) { kref_put(&resv_map->refs, resv_map_release); - set_vma_desc_resv_map(desc, NULL); + set_vma_resv_map(vma, NULL); } return err; } From fa0b9b2b7ae3539908d69c2b9ac0d144d9bc5139 Mon Sep 17 00:00:00 2001 From: Linpu Yu Date: Sun, 10 May 2026 13:43:30 +0800 Subject: [PATCH 117/411] ipc: limit next_id allocation to the valid ID range The checkpoint/restore sysctl path can request the next SysV IPC id through ids->next_id. ipc_idr_alloc() currently forwards that request to idr_alloc() with an open-ended upper bound. If the valid tail of the SysV IPC id space is full, the allocation can spill beyond ipc_mni. The returned SysV IPC id still uses the normal index encoding, so later lookup and removal can target the wrong slot. This leaves the real IDR entry behind and breaks the IDR state for the object. The bug is in ipc_idr_alloc() in the checkpoint/restore path. 1. ids->next_id is passed to: idr_alloc(&ids->ipcs_idr, new, ipcid_to_idx(next_id), 0, ...) 2. The zero upper bound makes the allocation effectively open-ended. Once the valid SysV IPC tail is occupied, idr_alloc() can spill past ipc_mni and allocate an entry beyond the valid IPC id range. 3. The new object id is still encoded with the narrower SysV IPC index width: new->id = (new->seq << ipcmni_seq_shift()) + idx 4. Later removal goes through ipc_rmid(), which uses: ipcid_to_idx(ipcp->id) That truncates the real IDR index. An object actually stored at a high index can then be removed as if it lived at a low in-range index. 5. For shared memory, shm_destroy() frees the current object anyway, but the real high IDR slot is left behind as a dangling pointer. 6. A subsequent walk of /proc/sysvipc/shm reaches the stale IDR entry and dereferences freed memory. Prevent this by bounding the requested allocation to ipc_mni so the checkpoint/restore path fails once the valid range is exhausted. Link: https://lore.kernel.org/cover.1778336914.git.linpu5433@gmail.com Link: https://lore.kernel.org/2eebe949bfa7d1f6e13b5be6a92c64c850ce9d45.1778336914.git.linpu5433@gmail.com Fixes: 03f595668017 ("ipc: add sysctl to specify desired next object id") Signed-off-by: Linpu Yu Signed-off-by: Ren Wei Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Cc: Kees Cook Cc: Stanislav Kinsbursky Cc: Davidlohr Bueso Cc: Signed-off-by: Andrew Morton --- ipc/util.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/ipc/util.c b/ipc/util.c index 9eb89820594e..1737d776bc08 100644 --- a/ipc/util.c +++ b/ipc/util.c @@ -253,7 +253,7 @@ static inline int ipc_idr_alloc(struct ipc_ids *ids, struct kern_ipc_perm *new) } else { new->seq = ipcid_to_seqx(next_id); idx = idr_alloc(&ids->ipcs_idr, new, ipcid_to_idx(next_id), - 0, GFP_NOWAIT); + ipc_mni, GFP_NOWAIT); } if (idx >= 0) new->id = (new->seq << ipcmni_seq_shift()) + idx; From 3b041514cb6eae45869b020f743c14d983363222 Mon Sep 17 00:00:00 2001 From: "Pratyush Yadav (Google)" Date: Tue, 5 May 2026 15:39:20 +0200 Subject: [PATCH 118/411] memfd: deny writeable mappings when implying SEAL_WRITE When SEAL_EXEC is added, SEAL_WRITE is implied to make W^X. But the implied seal is set after the check that makes sure the memfd can not have any writable mappings. This means one can use SEAL_EXEC to apply SEAL_WRITE while having writeable mappings. This breaks the contract that SEAL_WRITE provides and can be used by an attacker to pass a memfd that appears to be write sealed but can still be modified arbitrarily. Fix this by adding the implied seals before the call for mapping_deny_writable() is done. Link: https://lore.kernel.org/20260505133922.797635-1-pratyush@kernel.org Fixes: c4f75bc8bd6b ("mm/memfd: add write seals when apply SEAL_EXEC to executable memfd") Signed-off-by: Pratyush Yadav (Google) Reviewed-by: Pasha Tatashin Acked-by: Jeff Xu Cc: Baolin Wang Cc: Brendan Jackman Cc: Greg Thelen Cc: Hugh Dickins Cc: Kees Cook Cc: "David Hildenbrand (Arm)" Cc: Signed-off-by: Andrew Morton --- mm/memfd.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/mm/memfd.c b/mm/memfd.c index fb425f4e315f..abe13b291ddc 100644 --- a/mm/memfd.c +++ b/mm/memfd.c @@ -283,6 +283,12 @@ int memfd_add_seals(struct file *file, unsigned int seals) goto unlock; } + /* + * SEAL_EXEC implies SEAL_WRITE, making W^X from the start. + */ + if (seals & F_SEAL_EXEC && inode->i_mode & 0111) + seals |= F_SEAL_SHRINK|F_SEAL_GROW|F_SEAL_WRITE|F_SEAL_FUTURE_WRITE; + if ((seals & F_SEAL_WRITE) && !(*file_seals & F_SEAL_WRITE)) { error = mapping_deny_writable(file->f_mapping); if (error) @@ -295,12 +301,6 @@ int memfd_add_seals(struct file *file, unsigned int seals) } } - /* - * SEAL_EXEC implies SEAL_WRITE, making W^X from the start. - */ - if (seals & F_SEAL_EXEC && inode->i_mode & 0111) - seals |= F_SEAL_SHRINK|F_SEAL_GROW|F_SEAL_WRITE|F_SEAL_FUTURE_WRITE; - *file_seals |= seals; error = 0; From bf62f69574b19720ae5fbbbcdf24a0c4e3e05e43 Mon Sep 17 00:00:00 2001 From: Richard Chang Date: Tue, 12 May 2026 07:49:18 +0000 Subject: [PATCH 119/411] zram: fix use-after-free in zram_writeback_endio A crash was observed in zram_writeback_endio due to a NULL pointer dereference in wake_up. The root cause is a race condition between the bio completion handler (zram_writeback_endio) and the writeback task. In zram_writeback_endio, wake_up() is called on &wb_ctl->done_wait after releasing wb_ctl->done_lock. This creates a race window where the writeback task can see num_inflight become 0, return, and free wb_ctl before zram_writeback_endio calls wake_up(). CPU 0 (zram_writeback_endio) CPU 1 (writeback_store) ============================ ============================ zram_writeback_slots zram_submit_wb_request zram_submit_wb_request wait_event(wb_ctl->done_wait) spin_lock(&wb_ctl->done_lock); list_add(&req->entry, &wb_ctl->done_reqs); spin_unlock(&wb_ctl->done_lock); wake_up(&wb_ctl->done_wait); zram_complete_done_reqs spin_lock(&wb_ctl->done_lock); list_add(&req->entry, &wb_ctl->done_reqs); spin_unlock(&wb_ctl->done_lock); while (num_inflight) > 0) spin_lock(&wb_ctl->done_lock); list_del(&req->entry); spin_unlock(&wb_ctl->done_lock); // num_inflight becomes 0 atomic_dec(num_inflight); // Leave zram_writeback_slots // Free wb_ctl release_wb_ctl(wb_ctl); // UAF crash! wake_up(&wb_ctl->done_wait); This patch fixes this race by using RCU. By protecting wb_ctl with rcu_read_lock() in zram_writeback_endio and using kfree_rcu() to free it, we ensure that wb_ctl remains valid during the execution of zram_writeback_endio. Link: https://lore.kernel.org/20260512074918.2606208-1-richardycc@google.com Fixes: f405066a1f0d ("zram: introduce writeback bio batching") Signed-off-by: Richard Chang Suggested-by: Sergey Senozhatsky Suggested-by: Minchan Kim Acked-by: Sergey Senozhatsky Acked-by: Minchan Kim Cc: Brian Geffon Cc: Jens Axboe Cc: Martin Liu Cc: wang wei Cc: Signed-off-by: Andrew Morton --- drivers/block/zram/zram_drv.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index aebc710f0d6a..07111455eecf 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -33,6 +33,7 @@ #include #include #include +#include #include "zram_drv.h" @@ -504,6 +505,7 @@ struct zram_wb_ctl { wait_queue_head_t done_wait; spinlock_t done_lock; atomic_t num_inflight; + struct rcu_head rcu; }; struct zram_wb_req { @@ -847,7 +849,7 @@ static void release_wb_ctl(struct zram_wb_ctl *wb_ctl) release_wb_req(req); } - kfree(wb_ctl); + kfree_rcu(wb_ctl, rcu); } static struct zram_wb_ctl *init_wb_ctl(struct zram *zram) @@ -964,11 +966,13 @@ static void zram_writeback_endio(struct bio *bio) struct zram_wb_ctl *wb_ctl = bio->bi_private; unsigned long flags; + rcu_read_lock(); spin_lock_irqsave(&wb_ctl->done_lock, flags); list_add(&req->entry, &wb_ctl->done_reqs); spin_unlock_irqrestore(&wb_ctl->done_lock, flags); wake_up(&wb_ctl->done_wait); + rcu_read_unlock(); } static void zram_submit_wb_request(struct zram *zram, From 3f8968e9cbf95d5d87d32218906cab0b9b9eddbe Mon Sep 17 00:00:00 2001 From: Dev Jain Date: Mon, 18 May 2026 12:06:56 +0530 Subject: [PATCH 120/411] mm/rmap: initialize nr_pages to 1 at loop start in try_to_unmap_one Initialize nr_pages to 1 at the start of each loop iteration, like folio_referenced_one() does. Without this, nr_pages computed by a previous folio_unmap_pte_batch() call can be reused on a later iteration that does not run folio_unmap_pte_batch() again. mmap a 64K large folio with MAP_ANONYMOUS | MAP_DROPPABLE, then call madvise(MADV_FREE), then make the last page device-exclusive via HMM_DMIRROR_EXCLUSIVE. Trigger node reclaim through sysfs. Now, in try_to_unmap_one(), we will first clear the first 15 out of 16 entries mapping the lazyfree folio. This will set nr_pages to 15. In the next pvmw walk, this nr_pages gets reused on a device-exclusive pte, thus potentially corrupting folio refcount/mapcount. At the moment, I have a userspace program which can make the kernel spit out a trace, but the blow up is in folio_referenced_one(), because there are existing bugs in the interaction between device-private and rmap (which too I am investigating). I did a one liner kernel change to avoid going into folio_referenced_one(), and the kernel blows up at folio_remove_rmap_ptes in try_to_unmap_one which is what I wanted. Note that the bug is there not since file folio batching but lazyfree folio batching, since device-exclusive only works for anonymous folios. Userspace visible effect is simply kernel crashing somewhere due to refcount/mapcount corruption. Link: https://lore.kernel.org/20260518063656.3721056-1-dev.jain@arm.com Fixes: 354dffd29575 ("mm: support batched unmap for lazyfree large folios during reclamation") Signed-off-by: Dev Jain Acked-by: Barry Song Acked-by: David Hildenbrand (Arm) Reviewed-by: Lorenzo Stoakes Cc: Anshuman Khandual Cc: Barry Song Cc: Dev Jain Cc: Harry Yoo Cc: Jann Horn Cc: Liam R. Howlett Cc: Rik van Riel Cc: Ryan Roberts Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- mm/rmap.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/rmap.c b/mm/rmap.c index 78b7fb5f367c..99e1b3dc390b 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -2030,6 +2030,8 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma, mmu_notifier_invalidate_range_start(&range); while (page_vma_mapped_walk(&pvmw)) { + nr_pages = 1; + /* * If the folio is in an mlock()d vma, we must not swap it out. */ From 441f92f7d386b85bad16de49db95a307cba048a2 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Mon, 18 May 2026 08:25:58 -0700 Subject: [PATCH 121/411] mm/damon/sysfs-schemes: delete tried region in regions_rmdirs() DAMON sysfs maintains the DAMOS tried region directory objects via a linked list. When the user requests refresh of the directories, DAMON sysfs removes all the region directories first, and then generate updated regions directory on the empty space. The removal function (damon_sysfs_scheme_regions_rm_dirs()) only puts the kobj objects. Deletion of the container region object from the linked list is done inside the kobj release callback function. If somehow the callback invocation is delayed, the list will contain regions list that gonna be freed. If the updated region directories creation is started in this situation, the list can be corrupted and use-after-free can happen. Because the kobj objects are managed by only DAMON sysfs, the issue cannot happen in normal situation. But, such delays can be made on kernels that built with CONFIG_DEBUG_KOBJECT_RELEASE. On the kernel, the issue can indeed be reproduced like below. # damo start --damos_action stat # cd /sys/kernel/mm/damon/admin/kdamonds/0/ # for i in {1..10}; do echo update_schemes_tried_regions > state; done # dmesg | grep underflow [ 89.296152] refcount_t: underflow; use-after-free. Fix the issue by removing the region object from the list when decrementing the reference count. Also update damos_sysfs_populate_region_dir() to add the region object to the list only after the kobject_init_and_add() is success, so that fail of kobject_init_and_add() is not leaving the deallocated object on the list. The issue was discovered [1] by Sashiko. Link: https://lore.kernel.org/20260518152559.93038-1-sj@kernel.org Link: https://lore.kernel.org/20260513011920.119183-1-sj@kernel.org [1] Fixes: 9277d0367ba1 ("mm/damon/sysfs-schemes: implement scheme region directory") Signed-off-by: SeongJae Park Cc: # 6.2.x Signed-off-by: Andrew Morton --- mm/damon/sysfs-schemes.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/mm/damon/sysfs-schemes.c b/mm/damon/sysfs-schemes.c index 04746cbb3327..a8014780edae 100644 --- a/mm/damon/sysfs-schemes.c +++ b/mm/damon/sysfs-schemes.c @@ -88,7 +88,6 @@ static void damon_sysfs_scheme_region_release(struct kobject *kobj) struct damon_sysfs_scheme_region *region = container_of(kobj, struct damon_sysfs_scheme_region, kobj); - list_del(®ion->list); kfree(region); } @@ -164,7 +163,7 @@ static void damon_sysfs_scheme_regions_rm_dirs( struct damon_sysfs_scheme_region *r, *next; list_for_each_entry_safe(r, next, ®ions->regions_list, list) { - /* release function deletes it from the list */ + list_del(&r->list); kobject_put(&r->kobj); regions->nr_regions--; } @@ -2928,14 +2927,15 @@ void damos_sysfs_populate_region_dir(struct damon_sysfs_schemes *sysfs_schemes, if (!region) return; region->sz_filter_passed = sz_filter_passed; - list_add_tail(®ion->list, &sysfs_regions->regions_list); - sysfs_regions->nr_regions++; if (kobject_init_and_add(®ion->kobj, &damon_sysfs_scheme_region_ktype, &sysfs_regions->kobj, "%d", sysfs_regions->nr_regions++)) { kobject_put(®ion->kobj); + return; } + list_add_tail(®ion->list, &sysfs_regions->regions_list); + sysfs_regions->nr_regions++; } int damon_sysfs_schemes_clear_regions( From e16f17a9c5af50221184d1ef4be4056bf3c4209e Mon Sep 17 00:00:00 2001 From: Alexandre Ghiti Date: Mon, 18 May 2026 10:28:19 +0200 Subject: [PATCH 122/411] mm: memcontrol: propagate NMI slab stats to memcg vmstats flush_nmi_stats() drains per-node NMI slab atomics into the per-node lruvec_stats, but does not propagate them to the memcg-level vmstats. For non NMI case, account_slab_nmi_safe() calls mod_memcg_lruvec_state() which updates both per-node lruvec_stats and memcg-level vmstats, so flush_nmi_stats() needs to flush to per-node lruvec_stats as well as memcg-level vmstats. So fix this by flushing to the memcg-level vmstats for NMI too. Link: https://lore.kernel.org/20260518082830.599102-1-alex@ghiti.fr Fixes: 940b01fc8dc1 ("memcg: nmi safe memcg stats for specific archs") Signed-off-by: Alexandre Ghiti Acked-by: Shakeel Butt Acked-by: Johannes Weiner Reviewed-by: Harry Yoo (Oracle) Cc: Michal Hocko Cc: Muchun Song Cc: Roman Gushchin Cc: Signed-off-by: Andrew Morton --- mm/memcontrol.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index c03d4787d466..507be1cca392 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -4352,6 +4352,9 @@ static void flush_nmi_stats(struct mem_cgroup *memcg, struct mem_cgroup *parent, lstats->state[index] += slab; if (plstats) plstats->state_pending[index] += slab; + memcg->vmstats->state[index] += slab; + if (parent) + parent->vmstats->state_pending[index] += slab; } if (atomic_read(&pn->slab_unreclaimable)) { int slab = atomic_xchg(&pn->slab_unreclaimable, 0); @@ -4360,6 +4363,9 @@ static void flush_nmi_stats(struct mem_cgroup *memcg, struct mem_cgroup *parent, lstats->state[index] += slab; if (plstats) plstats->state_pending[index] += slab; + memcg->vmstats->state[index] += slab; + if (parent) + parent->vmstats->state_pending[index] += slab; } } } From 09e7827e785729f391c8d46dc71becce70d296ab Mon Sep 17 00:00:00 2001 From: Deepanshu Kartikey Date: Mon, 16 Mar 2026 20:49:56 +0530 Subject: [PATCH 123/411] kernel/fork: validate exit_signal in kernel_clone() When a child process exits, it sends exit_signal to its parent via do_notify_parent(). The clone() syscall constructs exit_signal as: (lower_32_bits(clone_flags) & CSIGNAL) CSIGNAL is 0xff, so values in the range 65-255 are possible. However, valid_signal() only accepts signals up to _NSIG (64 on x86_64). A non-zero non-valid exit_signal acts the same as exit_signal == 0: the parent process is not signaled when the child terminates. The syzkaller reproducer triggers this by calling clone() with flags=0x80, resulting in exit_signal = (0x80 & CSIGNAL) = 128, which exceeds _NSIG and is not a valid signal. The v1 of this patch added the check only in the clone() syscall handler, which is incomplete. kernel_clone() has other callers such as sys_ia32_clone() which would remain unprotected. Move the check to kernel_clone() to cover all callers. Since the valid_signal() check is now in kernel_clone() and covers all callers including clone3(), the same check in copy_clone_args_from_user() becomes redundant and is removed. The higher 32bits check for clone3() is kept as it is clone3() specific. Note that this is a user-visible change: previously, passing an invalid exit_signal to clone() was silently accepted. The man page for clone() does not document any defined behavior for invalid exit_signal values, so rejecting them with -EINVAL is the correct behavior. It is unlikely that any sane application relies on passing an invalid exit_signal. [oleg@redhat.com: the comment above kernel_clone() should be updated] Link: https://lore.kernel.org/abwvgU17W8wuW2-J@redhat.com Link: https://lore.kernel.org/20260316151956.563558-1-kartikey406@gmail.com Fixes: 3f2c788a1314 ("fork: prevent accidental access to clone3 features") Signed-off-by: Deepanshu Kartikey Signed-off-by: Oleg Nesterov Reported-by: syzbot+bbe6b99feefc3a0842de@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=bbe6b99feefc3a0842de Tested-by: syzbot+bbe6b99feefc3a0842de@syzkaller.appspotmail.com Link: https://lore.kernel.org/all/20260307064202.353405-1-kartikey406@gmail.com/T/ [v1] Link: https://lore.kernel.org/all/20260316104536.558108-1-kartikey406@gmail.com/T/ [v2] Acked-by: Oleg Nesterov Acked-by: Michal Hocko Cc: Ben Segall Cc: Christian Brauner Cc: David Hildenbrand Cc: Dietmar Eggemann Cc: Ingo Molnar Cc: Juri Lelli Cc: Kees Cook Cc: Liam Howlett Cc: Lorenzo Stoakes (Oracle) Cc: Mel Gorman Cc: Mike Rapoport Cc: Peter Zijlstra Cc: Steven Rostedt Cc: Suren Baghdasaryan Cc: Valentin Schneider Cc: Vincent Guittot Cc: Vlastimil Babka Cc: Tetsuo Handa Signed-off-by: Andrew Morton --- kernel/fork.c | 11 +++++------ 1 file changed, 5 insertions(+), 6 deletions(-) diff --git a/kernel/fork.c b/kernel/fork.c index 5f3fdfdb14c7..8ac38beae360 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -2664,8 +2664,6 @@ struct task_struct *create_io_thread(int (*fn)(void *), void *arg, int node) * * It copies the process, and if successful kick-starts * it and waits for it to finish using the VM if required. - * - * args->exit_signal is expected to be checked for sanity by the caller. */ pid_t kernel_clone(struct kernel_clone_args *args) { @@ -2700,6 +2698,9 @@ pid_t kernel_clone(struct kernel_clone_args *args) (args->pidfd == args->parent_tid)) return -EINVAL; + if (!valid_signal(args->exit_signal)) + return -EINVAL; + /* * Determine whether and which event to report to ptracer. When * called from kernel_thread or CLONE_UNTRACED is explicitly @@ -2898,11 +2899,9 @@ static noinline int copy_clone_args_from_user(struct kernel_clone_args *kargs, return -EINVAL; /* - * Verify that higher 32bits of exit_signal are unset and that - * it is a valid signal + * Verify that higher 32bits of exit_signal are unset */ - if (unlikely((args.exit_signal & ~((u64)CSIGNAL)) || - !valid_signal(args.exit_signal))) + if (unlikely(args.exit_signal & ~((u64)CSIGNAL))) return -EINVAL; if ((args.flags & CLONE_INTO_CGROUP) && From 2c6f81d58741349298f51ff697d988cb42881453 Mon Sep 17 00:00:00 2001 From: Sunny Patel Date: Fri, 1 May 2026 17:21:16 +0530 Subject: [PATCH 124/411] mm/migrate_device: fix pgtable leak in migrate_vma_insert_huge_pmd_page When migrate_vma_insert_huge_pmd_page() jumps to unlock_abort due to a PMD check failure, the pgtable allocated earlier via pte_alloc_one() is never freed, causing a memory leak. Added free_abort label to release the pgtable in error path. Link: https://lore.kernel.org/20260501115122.23288-1-nueralspacetech@gmail.com Fixes: a30b48bf1b24 ("mm/migrate_device: implement THP migration of zone device pages") Signed-off-by: Sunny Patel Acked-by: David Hildenbrand (Arm) Reviewed-by: Huang Ying Cc: Alistair Popple Cc: Balbir Singh Cc: Byungchul Park Cc: Gregory Price Cc: Joshua Hahn Cc: Matthew Brost Cc: Rakie Kim Cc: Zi Yan Cc: Signed-off-by: Andrew Morton --- mm/migrate_device.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/migrate_device.c b/mm/migrate_device.c index ab49d4dcdb60..19cd14b34114 100644 --- a/mm/migrate_device.c +++ b/mm/migrate_device.c @@ -840,7 +840,7 @@ static int migrate_vma_insert_huge_pmd_page(struct migrate_vma *migrate, } else { if (folio_is_zone_device(folio) && !folio_is_device_coherent(folio)) { - goto abort; + goto free_abort; } entry = folio_mk_pmd(folio, vma->vm_page_prot); if (vma->vm_flags & VM_WRITE) @@ -893,6 +893,8 @@ static int migrate_vma_insert_huge_pmd_page(struct migrate_vma *migrate, unlock_abort: spin_unlock(ptl); +free_abort: + pte_free(vma->vm_mm, pgtable); abort: for (i = 0; i < HPAGE_PMD_NR; i++) src[i] &= ~MIGRATE_PFN_MIGRATE; From f0af98ff6b3077278974a460becbd05bbc710e60 Mon Sep 17 00:00:00 2001 From: Eugen Hristev Date: Sat, 25 Apr 2026 16:06:48 +0300 Subject: [PATCH 125/411] MAINTAINERS, mailmap: change email for Eugen Hristev Replace old bouncing emails with ehristev@kernel.org Link: https://lore.kernel.org/20260425-eh-mailmap-v1-1-58788d401eef@kernel.org Signed-off-by: Eugen Hristev Signed-off-by: Andrew Morton --- .mailmap | 5 +++-- MAINTAINERS | 12 ++++++------ 2 files changed, 9 insertions(+), 8 deletions(-) diff --git a/.mailmap b/.mailmap index de41cfdae5d0..0b51bce3c05a 100644 --- a/.mailmap +++ b/.mailmap @@ -263,8 +263,9 @@ Enric Balletbo i Serra Enric Balletbo i Serra Erik Kaneda Ethan Carter Edwards Ethan Edwards -Eugen Hristev -Eugen Hristev +Eugen Hristev +Eugen Hristev +Eugen Hristev Evgeniy Polyakov Ezequiel Garcia Faith Ekstrand diff --git a/MAINTAINERS b/MAINTAINERS index 9eb15dacb939..8cf9ba51d981 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -10832,7 +10832,7 @@ F: include/linux/generic-radix-tree.h F: lib/generic-radix-tree.c GENERIC RESISTIVE TOUCHSCREEN ADC DRIVER -M: Eugen Hristev +M: Eugen Hristev L: linux-input@vger.kernel.org S: Maintained F: drivers/input/touchscreen/resistive-adc-touch.c @@ -17343,7 +17343,7 @@ F: Documentation/devicetree/bindings/sound/mikroe,mikroe-proto.txt F: sound/soc/atmel MICROCHIP CSI2DC DRIVER -M: Eugen Hristev +M: Eugen Hristev L: linux-media@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/media/microchip,csi2dc.yaml @@ -17370,7 +17370,7 @@ F: drivers/i2c/busses/i2c-at91-*.c F: drivers/i2c/busses/i2c-at91.h MICROCHIP ISC DRIVER -M: Eugen Hristev +M: Eugen Hristev L: linux-media@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/media/atmel,isc.yaml @@ -17382,7 +17382,7 @@ F: drivers/staging/media/deprecated/atmel/atmel-sama*-isc* F: include/linux/atmel-isc-media.h MICROCHIP ISI DRIVER -M: Eugen Hristev +M: Eugen Hristev L: linux-media@vger.kernel.org S: Supported F: drivers/media/platform/atmel/atmel-isi.c @@ -17572,7 +17572,7 @@ F: Documentation/devicetree/bindings/display/bridge/microchip,sam9x75-lvds.yaml F: drivers/gpu/drm/bridge/microchip-lvds.c MICROCHIP SAMA5D2-COMPATIBLE ADC DRIVER -M: Eugen Hristev +M: Eugen Hristev L: linux-iio@vger.kernel.org S: Supported F: Documentation/devicetree/bindings/iio/adc/atmel,sama5d2-adc.yaml @@ -24123,7 +24123,7 @@ F: drivers/mmc/host/sdhci* SECURE DIGITAL HOST CONTROLLER INTERFACE (SDHCI) MICROCHIP DRIVER M: Aubin Constans -R: Eugen Hristev +R: Eugen Hristev L: linux-mmc@vger.kernel.org S: Supported F: drivers/mmc/host/sdhci-of-at91.c From 04aa71da5f35aacdc9ae9cb5150947daa624f641 Mon Sep 17 00:00:00 2001 From: "Uladzislau Rezki (Sony)" Date: Fri, 15 May 2026 17:30:09 +0200 Subject: [PATCH 126/411] mm/vmalloc: do not trigger BUG() on BH disabled context __get_vm_area_node() currently triggers a BUG() if in_interrupt() returns true. However, in_interrupt() also reports true when BH are disabled. The bridge code can call rhashtable_lookup_insert_fast() with bottom halves disabled: __vlan_add() -> br_fdb_add_local() spin_lock_bh(&br->hash_lock); <-- Disable BH -> fdb_add_local() -> fdb_create() -> rhashtable_lookup_insert_fast() -> kvmalloc() -> vmalloc() -> __get_vm_area_node() -> BUG_ON(in_interrupt()) spin_unlock_bh(&br->hash_lock) this triggers the BUG() despite the caller not being in NMI or hard IRQ context. Replace the in_interrupt() check with in_nmi() || in_hardirq(). Link: https://lore.kernel.org/20260515153009.2296191-1-urezki@gmail.com Fixes: c6307674ed82 ("mm: kvmalloc: add non-blocking support for vmalloc") Signed-off-by: Uladzislau Rezki (Sony) Cc: Ido Schimmel Reported-by: syzbot+8b12fc6e0fb139765b58@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/69ff8c7c.050a0220.1036b8.000b.GAE@google.com/ Reviewed-by: Baoquan He Cc: Signed-off-by: Andrew Morton --- mm/vmalloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/vmalloc.c b/mm/vmalloc.c index c31a8615a832..bb6ae08d18f5 100644 --- a/mm/vmalloc.c +++ b/mm/vmalloc.c @@ -3203,7 +3203,7 @@ struct vm_struct *__get_vm_area_node(unsigned long size, struct vm_struct *area; unsigned long requested_size = size; - BUG_ON(in_interrupt()); + BUG_ON(in_nmi() || in_hardirq()); size = ALIGN(size, 1ul << shift); if (unlikely(!size)) return NULL; From 54cf41c969da6637cce790b7400da1451609db9b Mon Sep 17 00:00:00 2001 From: Byungchul Park Date: Fri, 15 May 2026 12:47:01 +0900 Subject: [PATCH 127/411] Revert "mm: introduce a new page type for page pool in page type" This reverts commit db359fccf212 ("mm: introduce a new page type for page pool in page type") and a part of 735a309b4bfb9e ("net: add net_iov_init() and use it to initialize ->page_type"). Netpp page_type'ed pages might be used in mapping so as to use @_mapcount. However, since @page_type and @_mapcount are union'ed in struct page, these two can't be used at the same time. Revert the commit introducing page_type for Netpp for now. The patch will be retried once @page_type and @_mapcount get allowed to be used at the same time. The revert also includes removal of @page_type initialization part introduced by commit 735a309b4bfb9e ("net: add net_iov_init() and use it to initialize ->page_type"), which will be restored on the retry. Link: https://lore.kernel.org/20260515034701.17027-1-byungchul@sk.com Fixes: db359fccf212 ("mm: introduce a new page type for page pool in page type") Signed-off-by: Byungchul Park Reported-by: Dragos Tatulea Closes: https://lore.kernel.org/all/982b9bc1-0a0a-4fc5-8e3a-3672db2b29a1@nvidia.com Acked-by: Jakub Kicinski Acked-by: David Hildenbrand (Arm) Acked-by: Harry Yoo (Oracle) Reviewed-by: Lorenzo Stoakes Cc: Alexei Starovoitov Cc: Baolin Wang Cc: Brendan Jackman Cc: David S. Miller Cc: Eric Dumazet Cc: Ilias Apalodimas Cc: Jesper Dangaard Brouer Cc: Johannes Weiner Cc: John Fastabend Cc: Leon Romanovsky Cc: Liam R. Howlett Cc: Mark Bloch Cc: Matthew Wilcox (Oracle) Cc: Michal Hocko Cc: Mike Rapoport Cc: Paolo Abeni Cc: Pavel Begunkov Cc: Saeed Mahameed Cc: Simon Horman Cc: Stanislav Fomichev Cc: Suren Baghdasaryan Cc: Tariq Toukan Cc: Toke Hoiland-Jorgensen Cc: Vlastimil Babka Cc: Zi Yan Signed-off-by: Andrew Morton --- .../net/ethernet/mellanox/mlx5/core/en/xdp.c | 2 +- include/linux/mm.h | 27 ++++++++++++++++--- include/linux/page-flags.h | 6 ----- include/net/netmem.h | 19 ++----------- mm/page_alloc.c | 13 +++------ net/core/netmem_priv.h | 23 +++++++++------- net/core/page_pool.c | 24 ++--------------- 7 files changed, 46 insertions(+), 68 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c index 190b8b66b3ce..d3bab198c99c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/xdp.c @@ -708,7 +708,7 @@ static void mlx5e_free_xdpsq_desc(struct mlx5e_xdpsq *sq, xdpi = mlx5e_xdpi_fifo_pop(xdpi_fifo); page = xdpi.page.page; - /* No need to check PageNetpp() as we + /* No need to check page_pool_page_is_pp() as we * know this is a page_pool page. */ page_pool_recycle_direct(pp_page_to_nmdesc(page)->pp, diff --git a/include/linux/mm.h b/include/linux/mm.h index af23453e9dbd..06bbe9eba636 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -5174,9 +5174,10 @@ int arch_lock_shadow_stack_status(struct task_struct *t, unsigned long status); * DMA mapping IDs for page_pool * * When DMA-mapping a page, page_pool allocates an ID (from an xarray) and - * stashes it in the upper bits of page->pp_magic. Non-PP pages can have - * arbitrary kernel pointers stored in the same field as pp_magic (since - * it overlaps with page->lru.next), so we must ensure that we cannot + * stashes it in the upper bits of page->pp_magic. We always want to be able to + * unambiguously identify page pool pages (using page_pool_page_is_pp()). Non-PP + * pages can have arbitrary kernel pointers stored in the same field as pp_magic + * (since it overlaps with page->lru.next), so we must ensure that we cannot * mistake a valid kernel pointer with any of the values we write into this * field. * @@ -5211,6 +5212,26 @@ int arch_lock_shadow_stack_status(struct task_struct *t, unsigned long status); #define PP_DMA_INDEX_MASK GENMASK(PP_DMA_INDEX_BITS + PP_DMA_INDEX_SHIFT - 1, \ PP_DMA_INDEX_SHIFT) +/* Mask used for checking in page_pool_page_is_pp() below. page->pp_magic is + * OR'ed with PP_SIGNATURE after the allocation in order to preserve bit 0 for + * the head page of compound page and bit 1 for pfmemalloc page, as well as the + * bits used for the DMA index. page_is_pfmemalloc() is checked in + * __page_pool_put_page() to avoid recycling the pfmemalloc page. + */ +#define PP_MAGIC_MASK ~(PP_DMA_INDEX_MASK | 0x3UL) + +#ifdef CONFIG_PAGE_POOL +static inline bool page_pool_page_is_pp(const struct page *page) +{ + return (page->pp_magic & PP_MAGIC_MASK) == PP_SIGNATURE; +} +#else +static inline bool page_pool_page_is_pp(const struct page *page) +{ + return false; +} +#endif + #define PAGE_SNAPSHOT_FAITHFUL (1 << 0) #define PAGE_SNAPSHOT_PG_BUDDY (1 << 1) #define PAGE_SNAPSHOT_PG_IDLE (1 << 2) diff --git a/include/linux/page-flags.h b/include/linux/page-flags.h index 0e03d816e8b9..7223f6f4e2b4 100644 --- a/include/linux/page-flags.h +++ b/include/linux/page-flags.h @@ -923,7 +923,6 @@ enum pagetype { PGTY_zsmalloc = 0xf6, PGTY_unaccepted = 0xf7, PGTY_large_kmalloc = 0xf8, - PGTY_netpp = 0xf9, PGTY_mapcount_underflow = 0xff }; @@ -1056,11 +1055,6 @@ PAGE_TYPE_OPS(Zsmalloc, zsmalloc, zsmalloc) PAGE_TYPE_OPS(Unaccepted, unaccepted, unaccepted) PAGE_TYPE_OPS(LargeKmalloc, large_kmalloc, large_kmalloc) -/* - * Marks page_pool allocated pages. - */ -PAGE_TYPE_OPS(Netpp, netpp, netpp) - /** * PageHuge - Determine if the page belongs to hugetlbfs * @page: The page to test. diff --git a/include/net/netmem.h b/include/net/netmem.h index 78fe51e5756b..bccacd21b6c3 100644 --- a/include/net/netmem.h +++ b/include/net/netmem.h @@ -94,20 +94,10 @@ enum net_iov_type { */ struct net_iov { struct netmem_desc desc; - unsigned int page_type; enum net_iov_type type; struct net_iov_area *owner; }; -/* Make sure 'the offset of page_type in struct page == the offset of - * type in struct net_iov'. - */ -#define NET_IOV_ASSERT_OFFSET(pg, iov) \ - static_assert(offsetof(struct page, pg) == \ - offsetof(struct net_iov, iov)) -NET_IOV_ASSERT_OFFSET(page_type, page_type); -#undef NET_IOV_ASSERT_OFFSET - struct net_iov_area { /* Array of net_iovs for this area. */ struct net_iov *niovs; @@ -127,11 +117,7 @@ static inline unsigned int net_iov_idx(const struct net_iov *niov) return niov - net_iov_owner(niov)->niovs; } -/* Initialize a niov: stamp the owning area, the memory provider type, - * and the page_type "no type" sentinel expected by the page-type API - * (see PAGE_TYPE_OPS in ) so that - * page_pool_set_pp_info() can later call __SetPageNetpp() on a niov - * cast to struct page. +/* Initialize a niov: stamp the owning area, the memory provider type. */ static inline void net_iov_init(struct net_iov *niov, struct net_iov_area *owner, @@ -139,7 +125,6 @@ static inline void net_iov_init(struct net_iov *niov, { niov->owner = owner; niov->type = type; - niov->page_type = UINT_MAX; } /* netmem */ @@ -245,7 +230,7 @@ static inline unsigned long netmem_pfn_trace(netmem_ref netmem) */ #define pp_page_to_nmdesc(p) \ ({ \ - DEBUG_NET_WARN_ON_ONCE(!PageNetpp(p)); \ + DEBUG_NET_WARN_ON_ONCE(!page_pool_page_is_pp(p)); \ __pp_page_to_nmdesc(p); \ }) diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 23c7298d3be2..d49c254174da 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1035,6 +1035,7 @@ static inline bool page_expected_state(struct page *page, #ifdef CONFIG_MEMCG page->memcg_data | #endif + page_pool_page_is_pp(page) | (page->flags.f & check_flags))) return false; @@ -1061,6 +1062,8 @@ static const char *page_bad_reason(struct page *page, unsigned long flags) if (unlikely(page->memcg_data)) bad_reason = "page still charged to cgroup"; #endif + if (unlikely(page_pool_page_is_pp(page))) + bad_reason = "page_pool leak"; return bad_reason; } @@ -1377,17 +1380,9 @@ __always_inline bool __free_pages_prepare(struct page *page, mod_mthp_stat(order, MTHP_STAT_NR_ANON, -1); folio->mapping = NULL; } - if (unlikely(page_has_type(page))) { - /* networking expects to clear its page type before releasing */ - if (is_check_pages_enabled()) { - if (unlikely(PageNetpp(page))) { - bad_page(page, "page_pool leak"); - return false; - } - } + if (unlikely(page_has_type(page))) /* Reset the page_type (which overlays _mapcount) */ page->page_type = UINT_MAX; - } if (is_check_pages_enabled()) { if (free_page_is_bad(page)) diff --git a/net/core/netmem_priv.h b/net/core/netmem_priv.h index 3e6fde8f1726..23175cb2bd86 100644 --- a/net/core/netmem_priv.h +++ b/net/core/netmem_priv.h @@ -8,18 +8,21 @@ static inline unsigned long netmem_get_pp_magic(netmem_ref netmem) return netmem_to_nmdesc(netmem)->pp_magic & ~PP_DMA_INDEX_MASK; } +static inline void netmem_or_pp_magic(netmem_ref netmem, unsigned long pp_magic) +{ + netmem_to_nmdesc(netmem)->pp_magic |= pp_magic; +} + +static inline void netmem_clear_pp_magic(netmem_ref netmem) +{ + WARN_ON_ONCE(netmem_to_nmdesc(netmem)->pp_magic & PP_DMA_INDEX_MASK); + + netmem_to_nmdesc(netmem)->pp_magic = 0; +} + static inline bool netmem_is_pp(netmem_ref netmem) { - struct page *page; - - /* XXX: Now that the offset of page_type is shared between - * struct page and net_iov, just cast the netmem to struct page - * unconditionally by clearing NET_IOV if any, no matter whether - * it comes from struct net_iov or struct page. This should be - * adjusted once the offset is no longer shared. - */ - page = (struct page *)((__force unsigned long)netmem & ~NET_IOV); - return PageNetpp(page); + return (netmem_get_pp_magic(netmem) & PP_MAGIC_MASK) == PP_SIGNATURE; } static inline void netmem_set_pp(netmem_ref netmem, struct page_pool *pool) diff --git a/net/core/page_pool.c b/net/core/page_pool.c index 6e576dec80db..8171d1173221 100644 --- a/net/core/page_pool.c +++ b/net/core/page_pool.c @@ -707,18 +707,8 @@ s32 page_pool_inflight(const struct page_pool *pool, bool strict) void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) { - struct page *page; - netmem_set_pp(netmem, pool); - - /* XXX: Now that the offset of page_type is shared between - * struct page and net_iov, just cast the netmem to struct page - * unconditionally by clearing NET_IOV if any, no matter whether - * it comes from struct net_iov or struct page. This should be - * adjusted once the offset is no longer shared. - */ - page = (struct page *)((__force unsigned long)netmem & ~NET_IOV); - __SetPageNetpp(page); + netmem_or_pp_magic(netmem, PP_SIGNATURE); /* Ensuring all pages have been split into one fragment initially: * page_pool_set_pp_info() is only called once for every page when it @@ -733,17 +723,7 @@ void page_pool_set_pp_info(struct page_pool *pool, netmem_ref netmem) void page_pool_clear_pp_info(netmem_ref netmem) { - struct page *page; - - /* XXX: Now that the offset of page_type is shared between - * struct page and net_iov, just cast the netmem to struct page - * unconditionally by clearing NET_IOV if any, no matter whether - * it comes from struct net_iov or struct page. This should be - * adjusted once the offset is no longer shared. - */ - page = (struct page *)((__force unsigned long)netmem & ~NET_IOV); - __ClearPageNetpp(page); - + netmem_clear_pp_magic(netmem); netmem_set_pp(netmem, NULL); } From 2982e599fff6faa21c8df147d96fc7af6c1a2f24 Mon Sep 17 00:00:00 2001 From: e521588 Date: Wed, 20 May 2026 09:27:17 +0200 Subject: [PATCH 128/411] esp: fix page frag reference leak on skb_to_sgvec failure In esp_output_tail(), when esp->inplace is false, the old skb page frags are replaced with a new page from the xfrm page_frag cache. The source scatterlist (sg) is built from the old frags before the replacement, and esp_ssg_unref() is responsible for releasing the old page references after the crypto operation completes. However, if the second skb_to_sgvec() call (which builds the destination scatterlist from the new page) fails, the code jumps to error_free which only calls kfree(tmp). The old page frag references captured in the source scatterlist are never released: 1. sg[] is built from old frags via skb_to_sgvec() (no extra get_page) 2. nr_frags is set to 1 and frag[0] is replaced with the new page 3. Second skb_to_sgvec() fails -> goto error_free 4. kfree(tmp) frees the sg[] memory but old frags are not unref'd 5. kfree_skb() only releases frag[0] (the new page), not the old ones Fix this by adding a bool parameter to esp_ssg_unref() that, when true, unconditionally unrefs the source scatterlist frags without checking req->src and req->dst, since those fields are not yet initialized by aead_request_set_crypt() at the point of the error. Existing callers pass false to preserve the original behavior. The same issue exists in both esp4 and esp6 as the code is identical. Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible") Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible") Signed-off-by: Alessandro Schino <7991aleschino@gmail.com> Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 12 +++++++----- net/ipv6/esp6.c | 12 +++++++----- 2 files changed, 14 insertions(+), 10 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 6a5febbdbee4..8314d7bddcb7 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -96,7 +96,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -113,7 +113,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (req->src != req->dst) + if (already_unref || req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -220,7 +220,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); kfree(tmp); if (xo && (xo->flags & XFRM_DEV_RESUME)) { @@ -569,8 +569,10 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) + if (unlikely(err < 0)) { + esp_ssg_unref(x, tmp, skb, true); goto error_free; + } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -602,7 +604,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * } if (sg != dsg) - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9c06c5a1419d..9d0c4957ac62 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -113,7 +113,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -130,7 +130,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (req->src != req->dst) + if (already_unref || req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -254,7 +254,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); kfree(tmp); esp_output_encap_csum(skb); @@ -600,8 +600,10 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) + if (unlikely(err < 0)) { + esp_ssg_unref(x, tmp, skb, true); goto error_free; + } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -634,7 +636,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info } if (sg != dsg) - esp_ssg_unref(x, tmp, skb); + esp_ssg_unref(x, tmp, skb, false); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); From dfa0d7b0ff1eb6b2c416b8fdb9b4f2cefba57a40 Mon Sep 17 00:00:00 2001 From: Jingguo Tan Date: Mon, 18 May 2026 17:06:48 +0800 Subject: [PATCH 129/411] xfrm: esp: restore combined single-frag length gate The ESP out-of-place fast path appends the trailer in esp_output_head() before esp_output_tail() allocates the destination page frag. The head-side gate currently checks skb->data_len and tailen separately, but the tail code allocates a single destination frag from the combined post-trailer skb->data_len. Reject the page-frag fast path when the combined aligned length exceeds a page. Otherwise skb_page_frag_refill() may fall back to a single page while the destination sg still spans the combined skb->data_len. Restore this combined-length page gate for both IPv4 and IPv6. Fixes: 5bd8baab087d ("esp: limit skb_page_frag_refill use to a single page") Cc: stable@vger.kernel.org Signed-off-by: Lin Ma Signed-off-by: Chenyuan Mi Signed-off-by: Jingguo Tan Reviewed-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 4 ++-- net/ipv6/esp6.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 8314d7bddcb7..5d3a8656687e 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -419,8 +419,8 @@ int esp_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * return err; } - if (ALIGN(tailen, L1_CACHE_BYTES) > PAGE_SIZE || - ALIGN(skb->data_len, L1_CACHE_BYTES) > PAGE_SIZE) + if (ALIGN(skb->data_len + tailen, L1_CACHE_BYTES) > + PAGE_SIZE) goto cow; if (!skb_cloned(skb)) { diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 9d0c4957ac62..b963b8e72604 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -448,8 +448,8 @@ int esp6_output_head(struct xfrm_state *x, struct sk_buff *skb, struct esp_info return err; } - if (ALIGN(tailen, L1_CACHE_BYTES) > PAGE_SIZE || - ALIGN(skb->data_len, L1_CACHE_BYTES) > PAGE_SIZE) + if (ALIGN(skb->data_len + tailen, L1_CACHE_BYTES) > + PAGE_SIZE) goto cow; if (!skb_cloned(skb)) { From e194ce048f5a6c549b3a23a8c568c6470f40f772 Mon Sep 17 00:00:00 2001 From: Wentao Liang Date: Thu, 9 Apr 2026 10:11:04 +0000 Subject: [PATCH 130/411] usb: musb: omap2430: Fix use-after-free in omap2430_probe() In omap2430_probe(), of_node_put(np) is called prematurely before the last access to np, leading to a use-after-free if the node's reference count drops to zero. Move the of_node_put() calls after the last use of np in both the success and error paths. Fixes: ffbe2feac59b ("usb: musb: omap2430: Fix probe regression for missing resources") Cc: stable Signed-off-by: Wentao Liang Link: https://patch.msgid.link/20260409101104.480623-1-vulab@iscas.ac.cn Signed-off-by: Greg Kroah-Hartman --- drivers/usb/musb/omap2430.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/usb/musb/omap2430.c b/drivers/usb/musb/omap2430.c index 48bb9bfb2204..333ab79f0ca9 100644 --- a/drivers/usb/musb/omap2430.c +++ b/drivers/usb/musb/omap2430.c @@ -337,7 +337,6 @@ static int omap2430_probe(struct platform_device *pdev) } else { device_set_of_node_from_dev(&musb->dev, &pdev->dev); } - of_node_put(np); glue->dev = &pdev->dev; glue->musb = musb; @@ -455,6 +454,7 @@ static int omap2430_probe(struct platform_device *pdev) dev_err(&pdev->dev, "failed to register musb device\n"); goto err_disable_rpm; } + of_node_put(np); return 0; @@ -464,6 +464,7 @@ static int omap2430_probe(struct platform_device *pdev) if (!IS_ERR(glue->control_otghs)) put_device(glue->control_otghs); err_put_musb: + of_node_put(np); platform_device_put(musb); return ret; From 4f88d65def6f3c90121601b4f62a4c967f3063a6 Mon Sep 17 00:00:00 2001 From: Guangshuo Li Date: Mon, 13 Apr 2026 22:21:19 +0800 Subject: [PATCH 131/411] usb: gadget: f_hid: fix device reference leak in hidg_alloc() hidg_alloc() initializes hidg->dev with device_initialize() before calling dev_set_name(). If dev_set_name() fails, the function currently jumps to err_unlock and returns without calling put_device(). This leaves the device reference unbalanced and prevents hidg_release() from being called. Calling put_device() here is also safe, since hidg_release() only frees resources owned by hidg. The issue was identified by a static analysis tool I developed and confirmed by manual review. Route the dev_set_name() failure path through err_put_device so the device reference is dropped properly. Fixes: 89ff3dfac604 ("usb: gadget: f_hid: fix f_hidg lifetime vs cdev") Cc: stable Reviewed-by: Johan Hovold Signed-off-by: Guangshuo Li Reviewed-by: Johan Hovold johan@kernel.org Link: https://patch.msgid.link/20260413142119.2977716-1-lgs201920130244@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_hid.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/usb/gadget/function/f_hid.c b/drivers/usb/gadget/function/f_hid.c index c5a12a6760ea..3c6b43d06a6d 100644 --- a/drivers/usb/gadget/function/f_hid.c +++ b/drivers/usb/gadget/function/f_hid.c @@ -1622,7 +1622,7 @@ static struct usb_function *hidg_alloc(struct usb_function_instance *fi) hidg->dev.devt = MKDEV(major, opts->minor); ret = dev_set_name(&hidg->dev, "hidg%d", opts->minor); if (ret) - goto err_unlock; + goto err_put_device; hidg->bInterfaceSubClass = opts->subclass; hidg->bInterfaceProtocol = opts->protocol; @@ -1659,7 +1659,6 @@ static struct usb_function *hidg_alloc(struct usb_function_instance *fi) err_put_device: put_device(&hidg->dev); -err_unlock: mutex_unlock(&opts->lock); return ERR_PTR(ret); } From c8547c74988e0b5f4cbb1b895e2a57aae084f070 Mon Sep 17 00:00:00 2001 From: Guangshuo Li Date: Mon, 27 Apr 2026 23:36:51 +0800 Subject: [PATCH 132/411] usb: gadget: net2280: Fix double free in probe error path usb_initialize_gadget() installs gadget_release() as the release callback for the embedded gadget device. The struct net2280 instance is therefore released through gadget_release() when the gadget device's last reference is dropped. The probe error path calls net2280_remove(), which tears down the partially initialized device and drops the gadget reference with usb_put_gadget(). Calling kfree(dev) afterwards can free the same object again. Drop the explicit kfree() and let the gadget device release callback handle the final free. This issue was found by a static analysis tool I am developing. Fixes: f770fbec4165 ("USB: UDC: net2280: Fix memory leaks") Cc: stable Signed-off-by: Guangshuo Li Reviewed-by: Alan Stern Link: https://patch.msgid.link/20260427153651.337846-1-lgs201920130244@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc/net2280.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/usb/gadget/udc/net2280.c b/drivers/usb/gadget/udc/net2280.c index d02765bd49ce..7c5f30cfd24d 100644 --- a/drivers/usb/gadget/udc/net2280.c +++ b/drivers/usb/gadget/udc/net2280.c @@ -3790,10 +3790,8 @@ static int net2280_probe(struct pci_dev *pdev, const struct pci_device_id *id) return 0; done: - if (dev) { + if (dev) net2280_remove(pdev); - kfree(dev); - } return retval; } From 68aa70648b625fa684bc0b71bbfd905f4943ca20 Mon Sep 17 00:00:00 2001 From: Kai Aizen Date: Thu, 30 Apr 2026 20:56:43 +0300 Subject: [PATCH 133/411] usb: gadget: uvc: hold opts->lock across XU walks in uvc_function_bind uvc_function_bind() walks &opts->extension_units twice without holding opts->lock: - directly, for the iExtension string-descriptor fixup loop; - indirectly, four times via uvc_copy_descriptors() (once per speed), where the helper iterates uvc->desc.extension_units (which aliases &opts->extension_units) to size and emit XU descriptors. The configfs side (uvcg_extension_make / uvcg_extension_drop, in drivers/usb/gadget/function/uvc_configfs.c) takes opts->lock around its list_add_tail / list_del operations. A privileged userspace process that holds the configfs subtree open and writes the gadget UDC name to bind the function while concurrently rmdir()'ing an extensions subdir can race uvcg_extension_drop() against the bind-time list walks and dereference a freed struct uvcg_extension. Hold opts->lock from the start of the XU string-descriptor fixup through the last uvc_copy_descriptors() call, releasing on the descriptor-error path via a new error_unlock label that drops the lock before falling through to the existing error label. This matches the locking discipline of the configfs callbacks and removes the only remaining unsynchronised reader of the XU list during bind. Reachability: only privileged processes that can mount configfs and write to gadget UDC files can trigger the race, so this is a correctness fix rather than a security boundary. Fixes: 0525210c9840 ("usb: gadget: uvc: Allow definition of XUs in configfs") Cc: stable Signed-off-by: Kai Aizen Link: https://patch.msgid.link/20260430175643.67120-1-kai.aizen.dev@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_uvc.c | 28 +++++++++++++++++++++------- 1 file changed, 21 insertions(+), 7 deletions(-) diff --git a/drivers/usb/gadget/function/f_uvc.c b/drivers/usb/gadget/function/f_uvc.c index 8d404d88391c..73dc7e42875f 100644 --- a/drivers/usb/gadget/function/f_uvc.c +++ b/drivers/usb/gadget/function/f_uvc.c @@ -768,6 +768,16 @@ uvc_function_bind(struct usb_configuration *c, struct usb_function *f) uvc_hs_streaming_ep.bEndpointAddress = uvc->video.ep->address; uvc_ss_streaming_ep.bEndpointAddress = uvc->video.ep->address; + /* + * Hold opts->lock across both the XU string-descriptor fixup below and + * the descriptor-copy block further down. Without this, configfs + * uvcg_extension_drop() (which takes opts->lock) can race with the + * list_for_each_entry() walks here and inside uvc_copy_descriptors(), + * leading to a UAF on a freed struct uvcg_extension. See + * drivers/usb/gadget/function/uvc_configfs.c::uvcg_extension_drop(). + */ + mutex_lock(&opts->lock); + /* * XUs can have an arbitrary string descriptor describing them. If they * have one pick up the ID. @@ -785,7 +795,7 @@ uvc_function_bind(struct usb_configuration *c, struct usb_function *f) ARRAY_SIZE(uvc_en_us_strings)); if (IS_ERR(us)) { ret = PTR_ERR(us); - goto error; + goto error_unlock; } uvc_iad.iFunction = opts->iad_index ? cdev->usb_strings[opts->iad_index].id : @@ -799,14 +809,14 @@ uvc_function_bind(struct usb_configuration *c, struct usb_function *f) /* Allocate interface IDs. */ if ((ret = usb_interface_id(c, f)) < 0) - goto error; + goto error_unlock; uvc_iad.bFirstInterface = ret; uvc_control_intf.bInterfaceNumber = ret; uvc->control_intf = ret; opts->control_interface = ret; if ((ret = usb_interface_id(c, f)) < 0) - goto error; + goto error_unlock; uvc_streaming_intf_alt0.bInterfaceNumber = ret; uvc_streaming_intf_alt1.bInterfaceNumber = ret; uvc->streaming_intf = ret; @@ -817,30 +827,32 @@ uvc_function_bind(struct usb_configuration *c, struct usb_function *f) if (IS_ERR(f->fs_descriptors)) { ret = PTR_ERR(f->fs_descriptors); f->fs_descriptors = NULL; - goto error; + goto error_unlock; } f->hs_descriptors = uvc_copy_descriptors(uvc, USB_SPEED_HIGH); if (IS_ERR(f->hs_descriptors)) { ret = PTR_ERR(f->hs_descriptors); f->hs_descriptors = NULL; - goto error; + goto error_unlock; } f->ss_descriptors = uvc_copy_descriptors(uvc, USB_SPEED_SUPER); if (IS_ERR(f->ss_descriptors)) { ret = PTR_ERR(f->ss_descriptors); f->ss_descriptors = NULL; - goto error; + goto error_unlock; } f->ssp_descriptors = uvc_copy_descriptors(uvc, USB_SPEED_SUPER_PLUS); if (IS_ERR(f->ssp_descriptors)) { ret = PTR_ERR(f->ssp_descriptors); f->ssp_descriptors = NULL; - goto error; + goto error_unlock; } + mutex_unlock(&opts->lock); + /* Preallocate control endpoint request. */ uvc->control_req = usb_ep_alloc_request(cdev->gadget->ep0, GFP_KERNEL); uvc->control_buf = kmalloc(UVC_MAX_REQUEST_SIZE, GFP_KERNEL); @@ -872,6 +884,8 @@ uvc_function_bind(struct usb_configuration *c, struct usb_function *f) return 0; +error_unlock: + mutex_unlock(&opts->lock); v4l2_error: v4l2_device_unregister(&uvc->v4l2_dev); error: From 5a4c828b8b29b47534814ade26d9aee09d5101fc Mon Sep 17 00:00:00 2001 From: Wei-Cheng Chen Date: Tue, 5 May 2026 19:26:30 +0800 Subject: [PATCH 134/411] xhci: tegra: Fix ghost USB device on dual-role port unplug When a USB device is unplugged from the dual-role port, the device-mode path in tegra_xhci_id_work() explicitly clears both SS and HS port power via direct hub_control ClearPortFeature(POWER) calls. This preempts the xHCI controller's normal disconnect processing -- PORT_CSC is never generated, the USB core never sees the disconnect, and the device remains in its internal tree as a ghost visible in lsusb. Add an otg_set_port_power flag to control whether the dual-role switch path performs explicit port power management. SoCs that need it (Tegra124 / Tegra210 / Tegra186) set the flag; later SoCs (Tegra194 and beyond) rely on the PHY mode change to handle disconnect naturally and skip all port power calls. Within the port power path, otg_reset_sspi additionally gates the SSPI reset sequence on host-mode entry for SoCs that require it. Flags set per SoC: Tegra124, Tegra186 -> otg_set_port_power Tegra210 -> otg_set_port_power, otg_reset_sspi Tegra194 and later -> (none) Fixes: f836e7843036 ("usb: xhci-tegra: Add OTG support") Cc: stable Signed-off-by: Wei-Cheng Chen Link: https://patch.msgid.link/20260505112630.217704-1-weichengc@nvidia.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/host/xhci-tegra.c | 71 ++++++++++++++++++++--------------- 1 file changed, 40 insertions(+), 31 deletions(-) diff --git a/drivers/usb/host/xhci-tegra.c b/drivers/usb/host/xhci-tegra.c index d2214d309e96..d5637b376367 100644 --- a/drivers/usb/host/xhci-tegra.c +++ b/drivers/usb/host/xhci-tegra.c @@ -247,6 +247,7 @@ struct tegra_xusb_soc { bool has_ipfs; bool lpm_support; bool otg_reset_sspi; + bool otg_set_port_power; bool has_bar2; }; @@ -1352,12 +1353,13 @@ static void tegra_xhci_id_work(struct work_struct *work) struct tegra_xusb_mbox_msg msg; struct phy *phy = tegra_xusb_get_phy(tegra, "usb2", tegra->otg_usb2_port); + bool host_mode = tegra->host_mode; u32 status; int ret; - dev_dbg(tegra->dev, "host mode %s\n", str_on_off(tegra->host_mode)); + dev_dbg(tegra->dev, "host mode %s\n", str_on_off(host_mode)); - if (tegra->host_mode) + if (host_mode) phy_set_mode_ext(phy, PHY_MODE_USB_OTG, USB_ROLE_HOST); else phy_set_mode_ext(phy, PHY_MODE_USB_OTG, USB_ROLE_NONE); @@ -1366,41 +1368,43 @@ static void tegra_xhci_id_work(struct work_struct *work) tegra->otg_usb2_port); pm_runtime_get_sync(tegra->dev); - if (tegra->host_mode) { - /* switch to host mode */ - if (tegra->otg_usb3_port >= 0) { - if (tegra->soc->otg_reset_sspi) { - /* set PP=0 */ - tegra_xhci_hc_driver.hub_control( - xhci->shared_hcd, GetPortStatus, - 0, tegra->otg_usb3_port+1, - (char *) &status, sizeof(status)); - if (status & USB_SS_PORT_STAT_POWER) - tegra_xhci_set_port_power(tegra, false, - false); + if (tegra->soc->otg_set_port_power) { + if (host_mode) { + /* switch to host mode */ + if (tegra->otg_usb3_port >= 0) { + if (tegra->soc->otg_reset_sspi) { + /* set PP=0 */ + tegra_xhci_hc_driver.hub_control( + xhci->shared_hcd, GetPortStatus, + 0, tegra->otg_usb3_port+1, + (char *) &status, sizeof(status)); + if (status & USB_SS_PORT_STAT_POWER) + tegra_xhci_set_port_power(tegra, false, + false); - /* reset OTG port SSPI */ - msg.cmd = MBOX_CMD_RESET_SSPI; - msg.data = tegra->otg_usb3_port+1; + /* reset OTG port SSPI */ + msg.cmd = MBOX_CMD_RESET_SSPI; + msg.data = tegra->otg_usb3_port+1; - ret = tegra_xusb_mbox_send(tegra, &msg); - if (ret < 0) { - dev_info(tegra->dev, - "failed to RESET_SSPI %d\n", - ret); + ret = tegra_xusb_mbox_send(tegra, &msg); + if (ret < 0) { + dev_info(tegra->dev, + "failed to RESET_SSPI %d\n", + ret); + } } + + tegra_xhci_set_port_power(tegra, false, true); } - tegra_xhci_set_port_power(tegra, false, true); + tegra_xhci_set_port_power(tegra, true, true); + + } else { + if (tegra->otg_usb3_port >= 0) + tegra_xhci_set_port_power(tegra, false, false); + + tegra_xhci_set_port_power(tegra, true, false); } - - tegra_xhci_set_port_power(tegra, true, true); - - } else { - if (tegra->otg_usb3_port >= 0) - tegra_xhci_set_port_power(tegra, false, false); - - tegra_xhci_set_port_power(tegra, true, false); } pm_runtime_put_autosuspend(tegra->dev); } @@ -2553,6 +2557,7 @@ static const struct tegra_xusb_soc tegra124_soc = { .scale_ss_clock = true, .has_ipfs = true, .otg_reset_sspi = false, + .otg_set_port_power = true, .ops = &tegra124_ops, .mbox = { .cmd = 0xe4, @@ -2593,6 +2598,7 @@ static const struct tegra_xusb_soc tegra210_soc = { .scale_ss_clock = false, .has_ipfs = true, .otg_reset_sspi = true, + .otg_set_port_power = true, .ops = &tegra124_ops, .mbox = { .cmd = 0xe4, @@ -2640,6 +2646,7 @@ static const struct tegra_xusb_soc tegra186_soc = { .scale_ss_clock = false, .has_ipfs = false, .otg_reset_sspi = false, + .otg_set_port_power = true, .ops = &tegra124_ops, .mbox = { .cmd = 0xe4, @@ -2673,6 +2680,7 @@ static const struct tegra_xusb_soc tegra194_soc = { .scale_ss_clock = false, .has_ipfs = false, .otg_reset_sspi = false, + .otg_set_port_power = false, .ops = &tegra124_ops, .mbox = { .cmd = 0x68, @@ -2708,6 +2716,7 @@ static const struct tegra_xusb_soc tegra234_soc = { .scale_ss_clock = false, .has_ipfs = false, .otg_reset_sspi = false, + .otg_set_port_power = false, .ops = &tegra234_ops, .mbox = { .cmd = XUSB_BAR2_ARU_MBOX_CMD, From 52f2ad3f7e5eb3b5908e1d685d4342519dc9cfcd Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 5 May 2026 15:56:03 -0300 Subject: [PATCH 135/411] usb: usbtmc: check URB actual_length for interrupt-IN notifications USBTMC devices can use an optional interrupt endpoint for notification messages. These typically contain two-byte headers indicating the payload format, but the driver does not check if these headers are present before accessing the data buffers. In cases where the URB actual_length is not enough to fit these headers, the driver will either cause an out-of-bounds read, or consume stale leftover data from a previous notification. Fix by checking if actual_data contains enough bytes for the headers, otherwise resubmit URB to the interrupt endpoint. Fixes: dbf3e7f654c0 ("Implement an ioctl to support the USMTMC-USB488 READ_STATUS_BYTE operation.") Reported-by: syzbot+abbfd103085885cf16a2@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=abbfd103085885cf16a2 Cc: stable Suggested-by: Michal Pecio Signed-off-by: Heitor Alves de Siqueira Link: https://patch.msgid.link/20260505-usbtmc-iin-size-v3-1-a36113f62db7@igalia.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/usbtmc.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index bd9347804dec..e15efd0c5ca7 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -2306,6 +2306,14 @@ static void usbtmc_interrupt(struct urb *urb) switch (status) { case 0: /* SUCCESS */ + /* ensure at least two bytes of headers were transferred */ + if (urb->actual_length < 2) { + dev_warn(dev, + "actual length %d not sufficient for interrupt headers\n", + urb->actual_length); + goto exit; + } + /* check for valid STB notification */ if (data->iin_buffer[0] > 0x81) { data->bNotify1 = data->iin_buffer[0]; From 121d2f682ba912b1427cddca7cf84840f41cc620 Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 5 May 2026 15:56:04 -0300 Subject: [PATCH 136/411] usb: usbtmc: reject interrupt endpoints with small wMaxPacketSize The USB488 subclass specification requires interrupt wMaxPacketSize to be 0x02, unless the device sends vendor-specific notifications. Endpoints that advertise less than 2 bytes for wMaxPacketSize are unlikely to work with the current driver, as URBs will not have enough space for interrupt headers. Considering that any notification URBs will be ignored by the driver, reject these endpoints early during probe. Fixes: 041370cce889 ("USB: usbtmc: refactor endpoint retrieval") Cc: stable Suggested-by: Michal Pecio Signed-off-by: Heitor Alves de Siqueira Link: https://patch.msgid.link/20260505-usbtmc-iin-size-v3-2-a36113f62db7@igalia.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/usbtmc.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/usb/class/usbtmc.c b/drivers/usb/class/usbtmc.c index e15efd0c5ca7..af9ae55dae14 100644 --- a/drivers/usb/class/usbtmc.c +++ b/drivers/usb/class/usbtmc.c @@ -2440,6 +2440,12 @@ static int usbtmc_probe(struct usb_interface *intf, data->iin_ep = int_in->bEndpointAddress; data->iin_wMaxPacketSize = usb_endpoint_maxp(int_in); data->iin_interval = int_in->bInterval; + /* wMaxPacketSize should be 0x02 or more as per USB488 Table 22 */ + if (iface_desc->desc.bInterfaceProtocol == 1 && + data->iin_wMaxPacketSize < 2) { + retcode = -EINVAL; + goto err_put; + } dev_dbg(&intf->dev, "Found Int in endpoint at %u\n", data->iin_ep); } From 9ddb9c0deca48d2c2a22ebf4d2f35c925a520328 Mon Sep 17 00:00:00 2001 From: "Stephen J. Fuhry" Date: Wed, 13 May 2026 13:14:19 -0400 Subject: [PATCH 137/411] USB: quirks: add NO_LPM for Lenovo ThinkPad USB-C Dock Gen2 hub controllers The Lenovo ThinkPad USB-C Dock Gen2 (17ef:a391, 17ef:a392) hub controllers exhibit link instability when USB Link Power Management is enabled, similar to the dock's Ethernet adapter (17ef:a387) which already carries USB_QUIRK_NO_LPM. When the dock reconnects after a transient disconnect, the hub controllers enter LPM states between re-enumeration retries, causing repeated disconnect/reconnect cycles lasting up to two minutes. Disabling LPM for these devices restores stable enumeration. Signed-off-by: Stephen J. Fuhry Cc: stable Link: https://patch.msgid.link/20260513171419.44849-1-fuhrysteve@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/core/quirks.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/core/quirks.c b/drivers/usb/core/quirks.c index 0ffdaefba508..87810eff974e 100644 --- a/drivers/usb/core/quirks.c +++ b/drivers/usb/core/quirks.c @@ -513,6 +513,10 @@ static const struct usb_device_id usb_quirk_list[] = { /* Lenovo ThinkPad USB-C Dock Gen2 Ethernet (RTL8153 GigE) */ { USB_DEVICE(0x17ef, 0xa387), .driver_info = USB_QUIRK_NO_LPM }, + /* Lenovo ThinkPad USB-C Dock Gen2 USB 3.1 and USB 2.0 hub controllers */ + { USB_DEVICE(0x17ef, 0xa391), .driver_info = USB_QUIRK_NO_LPM }, + { USB_DEVICE(0x17ef, 0xa392), .driver_info = USB_QUIRK_NO_LPM }, + /* BUILDWIN Photo Frame */ { USB_DEVICE(0x1908, 0x1315), .driver_info = USB_QUIRK_HONOR_BNUMINTERFACES }, From b53ebb811e00be50a779ce4e7aee604178b4a825 Mon Sep 17 00:00:00 2001 From: Sam Burkels Date: Fri, 1 May 2026 15:23:46 +0200 Subject: [PATCH 138/411] usb: storage: Add quirks for PNY Elite Portable SSD The PNY Elite Portable SSD (USB ID 154b:f009) is a sibling of the already-quirked PNY Pro Elite SSDs (154b:f00b and 154b:f00d). Like its siblings, it uses a Phison-based USB-SATA bridge that exhibits firmware bugs when bound to the uas driver. Without quirks, the device fails to complete READ CAPACITY commands when accessed over UAS on a SuperSpeed (USB 3) port. The device enumerates and reports as a SCSI direct-access device, but reports zero logical blocks and never finishes spin-up: usb 2-3: new SuperSpeed USB device number 8 using xhci_hcd usb 2-3: New USB device found, idVendor=154b, idProduct=f009 usb 2-3: Product: PNY ELITE PSSD usb 2-3: Manufacturer: PNY scsi host0: uas scsi 0:0:0:0: Direct-Access PNY PNY ELITE PSSD 0 sd 0:0:0:0: [sda] Spinning up disk... [...10+ seconds of polling, no progress...] sd 0:0:0:0: [sda] Read Capacity(16) failed: hostbyte=DID_ERROR sd 0:0:0:0: [sda] Read Capacity(10) failed: hostbyte=DID_ERROR sd 0:0:0:0: [sda] 0 512-byte logical blocks: (0 B/0 B) Tested each individual quirk to find the minimum that fixes this: - US_FL_NO_ATA_1X alone: device hangs on spin-up - US_FL_NO_REPORT_OPCODES alone: works on USB 2.0, hangs on USB 3.0 - US_FL_NO_ATA_1X | US_FL_NO_REPORT_OPCODES: works on both With both quirks the device enumerates correctly while still using the uas driver, and delivers full UAS throughput (~281 MB/s sequential read on a USB 3.0 Gen 1 port). The existing PNY Pro Elite entries (f00b, f00d) only set NO_ATA_1X, but this device additionally chokes on REPORT OPCODES under SuperSpeed. Signed-off-by: Sam Burkels Acked-by: Oliver Neukum Cc: stable Link: https://patch.msgid.link/20260501132346.86572-1-sam@1a38.nl Signed-off-by: Greg Kroah-Hartman --- drivers/usb/storage/unusual_uas.h | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/usb/storage/unusual_uas.h b/drivers/usb/storage/unusual_uas.h index 939a98c2d3f7..d6f86d5db3bf 100644 --- a/drivers/usb/storage/unusual_uas.h +++ b/drivers/usb/storage/unusual_uas.h @@ -132,6 +132,13 @@ UNUSUAL_DEV(0x152d, 0x0583, 0x0000, 0x9999, USB_SC_DEVICE, USB_PR_DEVICE, NULL, US_FL_NO_REPORT_OPCODES), +/* Reported-by: Sam Burkels */ +UNUSUAL_DEV(0x154b, 0xf009, 0x0000, 0x9999, + "PNY", + "PNY ELITE PSSD", + USB_SC_DEVICE, USB_PR_DEVICE, NULL, + US_FL_NO_ATA_1X | US_FL_NO_REPORT_OPCODES), + /* Reported-by: Thinh Nguyen */ UNUSUAL_DEV(0x154b, 0xf00b, 0x0000, 0x9999, "PNY", From 0b5bde9e1c4df3bbf93a01a0e00c05085f3449fd Mon Sep 17 00:00:00 2001 From: "Rob Herring (Arm)" Date: Fri, 8 May 2026 13:25:55 -0500 Subject: [PATCH 139/411] dt-bindings: usb: ti,omap4-musb: Drop duplicate 'usb-phy' property constraints The deprecated 'usb-phy' property is documented already in usb.yaml and doesn't need a type definition here. It just needs constraints on how many entries there are. As this is a host controller, reference usb-hcd.yaml which then references usb.yaml. Fixes: 70fcdc82cf4a ("dt-bindings: usb: ti,omap4-musb: convert to DT schema") Signed-off-by: Rob Herring (Arm) Link: https://patch.msgid.link/20260508182556.1759173-1-robh@kernel.org Signed-off-by: Greg Kroah-Hartman --- Documentation/devicetree/bindings/usb/ti,omap4-musb.yaml | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/usb/ti,omap4-musb.yaml b/Documentation/devicetree/bindings/usb/ti,omap4-musb.yaml index a3d15f217658..e1887e490e02 100644 --- a/Documentation/devicetree/bindings/usb/ti,omap4-musb.yaml +++ b/Documentation/devicetree/bindings/usb/ti,omap4-musb.yaml @@ -81,9 +81,7 @@ properties: const: usb2-phy usb-phy: - $ref: /schemas/types.yaml#/definitions/phandle-array - description: Phandle for the PHY device. - deprecated: true + maxItems: 1 ctrl-module: $ref: /schemas/types.yaml#/definitions/phandle @@ -96,6 +94,9 @@ required: - interrupts - interrupt-names +allOf: + - $ref: usb-hcd.yaml# + unevaluatedProperties: false examples: From d96209626a29ea64666be98c30b30ac82e5f1be6 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 17 Apr 2026 12:35:52 -0400 Subject: [PATCH 140/411] usbip: vudc: Fix use after free bug in vudc_remove due to race condition This patch follows up Zheng Wang's 2023 report of a use-after-free in vudc_remove(). The original thread stalled on Shuah Khan's request for runtime testing of the unplug/unbind path. This patch supplies that testing and keeps Zheng's original fix shape. In vudc_probe(), v_init_timer() binds udc->tr_timer.timer to v_timer(). usbip_sockfd_store() starts the timer via v_start_timer()/v_kick_timer(). vudc_remove() can then free the containing struct vudc while the timer is still pending or executing. KASAN confirms the race on an unpatched x86_64 QEMU guest with CONFIG_KASAN=y, CONFIG_USBIP_VUDC=y, CONFIG_USB_ZERO=y, and a tight loop that repeatedly writes a socket fd to usbip_sockfd, closes the socket pair, and unbinds/rebinds usbip-vudc.0: BUG: KASAN: slab-use-after-free in __run_timer_base.part.0+0x8ba/0x8e0 Write of size 8 at addr ffff888001b80740 by task trigger_and_unb/239 Allocated by task 239: vudc_probe+0x4d/0xaa0 Freed by task 239: kfree+0x18f/0x520 device_release_driver_internal+0x388/0x540 unbind_store+0xd9/0x100 This lands in the timer core rather than v_timer() itself because the embedded timer_list is being walked after its containing struct vudc has already been freed. The underlying lifetime bug is the same one Zheng reported. With v_stop_timer() called from vudc_remove() and the timer deleted synchronously, the same harness completed 5000 bind/unbind iterations with no KASAN report. Fixes: b6a0ca111867 ("usbip: vudc: Add UDC specific ops") Cc: stable Reported-by: Zheng Wang Closes: https://lore.kernel.org/linux-usb/20230317100954.2626573-1-zyytlz.wz@163.com/ Signed-off-by: Michael Bommarito Acked-by: Shuah Khan Link: https://patch.msgid.link/20260417163552.807548-1-michael.bommarito@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/usbip/vudc_dev.c | 1 + drivers/usb/usbip/vudc_transfer.c | 3 ++- 2 files changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/usb/usbip/vudc_dev.c b/drivers/usb/usbip/vudc_dev.c index 90383107b660..c5f079c5a1ea 100644 --- a/drivers/usb/usbip/vudc_dev.c +++ b/drivers/usb/usbip/vudc_dev.c @@ -632,6 +632,7 @@ void vudc_remove(struct platform_device *pdev) { struct vudc *udc = platform_get_drvdata(pdev); + v_stop_timer(udc); usb_del_gadget_udc(&udc->gadget); cleanup_vudc_hw(udc); kfree(udc); diff --git a/drivers/usb/usbip/vudc_transfer.c b/drivers/usb/usbip/vudc_transfer.c index a4f02ea3e3ef..d4ce85c4c6a2 100644 --- a/drivers/usb/usbip/vudc_transfer.c +++ b/drivers/usb/usbip/vudc_transfer.c @@ -490,7 +490,8 @@ void v_stop_timer(struct vudc *udc) { struct transfer_timer *t = &udc->tr_timer; - /* timer itself will take care of stopping */ + /* Delete the timer synchronously before teardown frees udc. */ dev_dbg(&udc->pdev->dev, "timer stop"); + timer_delete_sync(&t->timer); t->state = VUDC_TR_STOPPED; } From f1ecb0e563595d4ba9a3b8e39ed52a3dc2d8e328 Mon Sep 17 00:00:00 2001 From: Hang Cao Date: Wed, 15 Apr 2026 14:42:38 +0800 Subject: [PATCH 141/411] dt-bindings: usb: Fix EIC7700 USB reset's issue The EIC7700 USB requires a USB PHY reset operation; otherwise, the USB will not work. The reason why the USB driver that was applied can work properly is that the USB PHY has already been reset in ESWIN's U-Boot. However, the proper functioning of the USB driver should not be dependent on the bootloader. Therefore, it is necessary to incorporate the USB PHY reset signal into the DT bindings. This patch does not introduce any backward incompatibility since the dts is not upstream yet. As array of reset operations are used in the driver, no modifications to the USB controller driver are needed. Fixes: c640a4239db5 ("dt-bindings: usb: Add ESWIN EIC7700 USB controller") Cc: stable Signed-off-by: Hang Cao Acked-by: Conor Dooley Link: https://patch.msgid.link/20260415064238.1784-1-caohang@eswincomputing.com Signed-off-by: Greg Kroah-Hartman --- .../devicetree/bindings/usb/eswin,eic7700-usb.yaml | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/Documentation/devicetree/bindings/usb/eswin,eic7700-usb.yaml b/Documentation/devicetree/bindings/usb/eswin,eic7700-usb.yaml index 41c3b1b98991..658260619423 100644 --- a/Documentation/devicetree/bindings/usb/eswin,eic7700-usb.yaml +++ b/Documentation/devicetree/bindings/usb/eswin,eic7700-usb.yaml @@ -41,12 +41,13 @@ properties: - const: usb_en resets: - maxItems: 2 + maxItems: 3 reset-names: items: - const: vaux - const: usb_rst + - const: usb_phy eswin,hsp-sp-csr: description: @@ -85,8 +86,8 @@ examples: interrupt-parent = <&plic>; interrupts = <85>; interrupt-names = "peripheral"; - resets = <&reset 84>, <&hspcrg 2>; - reset-names = "vaux", "usb_rst"; + resets = <&reset 84>, <&hspcrg 2>, <&hspcrg 4>; + reset-names = "vaux", "usb_rst", "usb_phy"; dr_mode = "peripheral"; maximum-speed = "high-speed"; phy_type = "utmi"; From 7d9633528dd40e33964d2dc74a5abbf5c4d116ce Mon Sep 17 00:00:00 2001 From: Seungjin Bae Date: Mon, 18 May 2026 19:43:14 -0400 Subject: [PATCH 142/411] usb: gadget: dummy_hcd: Reject hub port requests for non-existent ports The `dummy_hub_control()` function handles USB hub class requests to the virtual root hub. The `GetPortStatus` case returns -EPIPE for requests with `wIndex != 1`, since the virtual root hub has only a single port. However, the `ClearPortFeature` and `SetPortFeature` cases lack the same check. Fix this by extending the `wIndex != 1` rejection to both cases, matching the existing behavior of `GetPortStatus`. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable Suggested-by: Alan Stern Signed-off-by: Seungjin Bae Reviewed-by: Alan Stern Link: https://patch.msgid.link/20260518234314.1889396-1-eeodqql09@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/udc/dummy_hcd.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/usb/gadget/udc/dummy_hcd.c b/drivers/usb/gadget/udc/dummy_hcd.c index f094491b1041..f47903461ed5 100644 --- a/drivers/usb/gadget/udc/dummy_hcd.c +++ b/drivers/usb/gadget/udc/dummy_hcd.c @@ -2134,6 +2134,8 @@ static int dummy_hub_control( case ClearHubFeature: break; case ClearPortFeature: + if (wIndex != 1) + goto error; switch (wValue) { case USB_PORT_FEAT_SUSPEND: if (hcd->speed == HCD_USB3) { @@ -2248,6 +2250,8 @@ static int dummy_hub_control( retval = -EPIPE; break; case SetPortFeature: + if (wIndex != 1) + goto error; switch (wValue) { case USB_PORT_FEAT_LINK_STATE: if (hcd->speed != HCD_USB3) { From 4e036c10e7f4df5d951c69cc3697bc8e209c6d02 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 19 Apr 2026 12:03:59 -0400 Subject: [PATCH 143/411] usb: gadget: f_fs: copy only received bytes on short ep0 read ffs_ep0_read() allocates its control-OUT data buffer with kmalloc() (not kzalloc) at the Length value from the Setup packet, then copies that full len to userspace regardless of how many bytes were actually received: data = kmalloc(len, GFP_KERNEL); ... ret = __ffs_ep0_queue_wait(ffs, data, len); if ((ret > 0) && (copy_to_user(buf, data, len))) ret = -EFAULT; __ffs_ep0_queue_wait() returns req->actual, which on a short control OUT transfer is strictly less than len. The copy_to_user() call still copies len bytes, so on a short OUT the last (len - ret) bytes of the kmalloc() buffer -- uninitialised slab residue -- are delivered to the FunctionFS daemon. Short ep0 OUT completions are specified USB control-transfer behavior and are produced by in-tree UDCs: * dwc2 continues on req->actual < req->length for ep0 DATA OUT (short-not-ok is the only ep0-OUT stall path). * aspeed_udc ends ep0 OUT on rx_len < ep->ep.maxpacket. * renesas_usbf logs "ep0 short packet" and completes the request. * dwc3 stalls on short IN but not on short OUT. A short ep0 OUT is therefore not evidence of a broken UDC; it is a normal condition f_fs has to cope with. The sibling gadgetfs implementation in drivers/usb/gadget/legacy/inode.c already does this correctly via min(len, dev->req->actual) before copy_to_user(). This patch brings f_fs.c to the same safe pattern rather than trimming at a defensive layer. The bug is reached from the FunctionFS device node, which in real deployments is owned by the privileged gadget daemon (adbd, UMS, composite gadget services, etc.); it is not reachable from unprivileged userspace. Linux host stacks normally reject short-wLength control OUTs before they reach the gadget, so reproducing this required a build that bypasses that host-side check. With the bypass in place, a 1-byte payload on a 64-byte Setup produces 63 bytes of non-canary slab residue in the daemon's read buffer. Fix by copying only ret (actually received) bytes to userspace. Fixes: ddf8abd25994 ("USB: f_fs: the FunctionFS driver") Cc: stable Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260419160359.1577270-1-michael.bommarito@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_fs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index 002c3441bea3..815639506520 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -619,7 +619,7 @@ static ssize_t ffs_ep0_read(struct file *file, char __user *buf, /* unlocks spinlock */ ret = __ffs_ep0_queue_wait(ffs, data, len); - if ((ret > 0) && (copy_to_user(buf, data, len))) + if ((ret > 0) && (copy_to_user(buf, data, ret))) ret = -EFAULT; goto done_mutex; From 2796646f6d892c1eb6818c7ca41fdfa12568e8d1 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 19 Apr 2026 12:12:27 -0400 Subject: [PATCH 144/411] usb: gadget: f_fs: serialize DMABUF cancel against request completion ffs_epfile_dmabuf_io_complete() calls usb_ep_free_request() on the completed request but leaves priv->req, the back-pointer that ffs_dmabuf_transfer() set on submission, pointing at the freed memory. A later FUNCTIONFS_DMABUF_DETACH ioctl or ffs_epfile_release() on the close path still sees priv->req non-NULL under ffs->eps_lock: if (priv->ep && priv->req) usb_ep_dequeue(priv->ep, priv->req); so usb_ep_dequeue() is called on a freed usb_request. On dummy_hcd the dequeue path only walks a live queue and pointer-compares, so the freed pointer reads without faulting and KASAN requires an explicit check at the FunctionFS call site to surface the use-after-free. On SG-capable in-tree UDCs the dequeue path dereferences the supplied request immediately: * chipidea's ep_dequeue() does container_of(req, struct ci_hw_req, req) and reads hwreq->req.status before acquiring its own lock. * cdnsp's cdnsp_gadget_ep_dequeue() reads request->status first. The narrower option of clearing priv->req via cmpxchg() in the completion does not close the race: the completion runs without eps_lock, so a cancel path holding eps_lock can still observe priv->req non-NULL, race a concurrent completion that clears and frees, and pass the freed pointer to usb_ep_dequeue(). A slightly longer fix that moves the free into the cleanup work is needed. Same class of lifetime race as the recent usbip-vudc timer fix [1]. Take eps_lock in the sole place that mutates priv->req from the callback direction by moving usb_ep_free_request() out of the completion into ffs_dmabuf_cleanup(), the existing work handler scheduled by ffs_dmabuf_signal_done() on ffs->io_completion_wq. Clear priv->req there under eps_lock before freeing, and only clear if priv->req still names our request (a subsequent ffs_dmabuf_transfer() on the same attachment may have queued a new one). This keeps the existing dummy_hcd sync-dequeue invariant: the completion callback is still invoked by the UDC without eps_lock held (dummy_hcd drops its own lock before calling the callback), and the callback now takes no f_fs lock at all. Serialization against the cancel path happens in cleanup, which runs from the workqueue with no f_fs lock held on entry. The priv ref count protects the containing ffs_dmabuf_priv: ffs_dmabuf_transfer() takes a ref via ffs_dmabuf_get(), cleanup drops it via ffs_dmabuf_put(), so priv stays live for the cleanup even after the cancel path's list_del + ffs_dmabuf_put. The ffs_dmabuf_transfer() error path no longer frees usb_req inline: fence->req and fence->ep are set before usb_ep_queue(), so ffs_dmabuf_cleanup() (scheduled by the error-path ffs_dmabuf_signal_done()) owns the free regardless of whether the queue succeeded. Reproduced under KASAN on both detach and close paths against dummy_hcd with an observability hook (kasan_check_byte(priv->req) immediately before usb_ep_dequeue) at the two FunctionFS cancel sites to surface the stale-pointer access; the hook is not part of this patch. The KASAN allocator / free stacks in the captured splats identify the same request: alloc in dummy_alloc_request, free in dummy_timer, fault reached from ffs_epfile_release (close) and from the FUNCTIONFS_DMABUF_DETACH ioctl (detach). With the patch applied, both paths are silent under the same hook. The bug is reached from the FunctionFS device node, which in real deployments is owned by the privileged gadget daemon (adbd, UMS, composite gadget services, etc.); it is not reachable from unprivileged userspace or from a USB host on the cable. FunctionFS mounts default to GLOBAL_ROOT_UID, but the filesystem supports uid=, gid=, and fmode= delegation to a non-root gadget daemon, so on real deployments the attacker may be a less-privileged service rather than root. Fixes: 7b07a2a7ca02 ("usb: gadget: functionfs: Add DMABUF import interface") Link: https://lore.kernel.org/all/20260417163552.807548-1-michael.bommarito@gmail.com/ [1] Cc: stable Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260419161227.1587668-1-michael.bommarito@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/gadget/function/f_fs.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-) diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c index 815639506520..75912ce6ab55 100644 --- a/drivers/usb/gadget/function/f_fs.c +++ b/drivers/usb/gadget/function/f_fs.c @@ -150,6 +150,8 @@ struct ffs_dma_fence { struct dma_fence base; struct ffs_dmabuf_priv *priv; struct work_struct work; + struct usb_ep *ep; + struct usb_request *req; }; struct ffs_epfile { @@ -1385,6 +1387,21 @@ static void ffs_dmabuf_cleanup(struct work_struct *work) struct ffs_dmabuf_priv *priv = dma_fence->priv; struct dma_buf_attachment *attach = priv->attach; struct dma_fence *fence = &dma_fence->base; + struct usb_request *req = dma_fence->req; + struct usb_ep *ep = dma_fence->ep; + + /* + * eps_lock pairs with the cancel paths so they cannot pass a freed + * req to usb_ep_dequeue(). Only clear if priv->req still names ours; + * a re-queue on the same attachment may have taken that slot. + */ + spin_lock_irq(&priv->ffs->eps_lock); + if (priv->req == req) + priv->req = NULL; + spin_unlock_irq(&priv->ffs->eps_lock); + + if (ep && req) + usb_ep_free_request(ep, req); ffs_dmabuf_put(attach); dma_fence_put(fence); @@ -1414,8 +1431,8 @@ static void ffs_epfile_dmabuf_io_complete(struct usb_ep *ep, struct usb_request *req) { pr_vdebug("FFS: DMABUF transfer complete, status=%d\n", req->status); + /* req is freed by ffs_dmabuf_cleanup() under eps_lock. */ ffs_dmabuf_signal_done(req->context, req->status); - usb_ep_free_request(ep, req); } static const char *ffs_dmabuf_get_driver_name(struct dma_fence *fence) @@ -1699,6 +1716,10 @@ static int ffs_dmabuf_transfer(struct file *file, usb_req->context = fence; usb_req->complete = ffs_epfile_dmabuf_io_complete; + /* ffs_dmabuf_cleanup() frees usb_req via these two fields. */ + fence->req = usb_req; + fence->ep = ep->ep; + cookie = dma_fence_begin_signalling(); ret = usb_ep_queue(ep->ep, usb_req, GFP_ATOMIC); dma_fence_end_signalling(cookie); @@ -1708,7 +1729,6 @@ static int ffs_dmabuf_transfer(struct file *file, } else { pr_warn("FFS: Failed to queue DMABUF: %d\n", ret); ffs_dmabuf_signal_done(fence, ret); - usb_ep_free_request(ep->ep, usb_req); } spin_unlock_irq(&epfile->ffs->eps_lock); From 8f6aa392653e52a45858cff5c063df550028836b Mon Sep 17 00:00:00 2001 From: Xu Yang Date: Mon, 27 Apr 2026 15:57:55 +0800 Subject: [PATCH 145/411] usb: chipidea: core: convert ci_role_switch to local variable When a system contains multiple USB controllers, the global ci_role_switch variable may be overwritten by subsequent driver initialization code. This can cause issues in the following cases: - The 2nd ci_hdrc_probe() sees ci_role_switch.fwnode as non-NULL even though the "usb-role-switch" property is not present for the controller. - When the ci_hdrc device is unbound and bound again, ci_role_switch fwnode will not be reassigned, and the old value will be used instead. Convert ci_role_switch to a local variable to fix these issues. Fixes: 05559f10ed79 ("usb: chipidea: add role switch class support") Cc: stable Acked-by: Peter Chen Reviewed-by: Frank Li Signed-off-by: Xu Yang Link: https://patch.msgid.link/20260427075755.3611217-1-xu.yang_2@nxp.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/chipidea/core.c | 16 ++++++---------- 1 file changed, 6 insertions(+), 10 deletions(-) diff --git a/drivers/usb/chipidea/core.c b/drivers/usb/chipidea/core.c index 7cfabb04a4fb..2ab3db3c1015 100644 --- a/drivers/usb/chipidea/core.c +++ b/drivers/usb/chipidea/core.c @@ -655,12 +655,6 @@ static enum ci_role ci_get_role(struct ci_hdrc *ci) return role; } -static struct usb_role_switch_desc ci_role_switch = { - .set = ci_usb_role_switch_set, - .get = ci_usb_role_switch_get, - .allow_userspace_control = true, -}; - static int ci_get_platdata(struct device *dev, struct ci_hdrc_platform_data *platdata) { @@ -787,9 +781,6 @@ static int ci_get_platdata(struct device *dev, cable->connected = false; } - if (device_property_read_bool(dev, "usb-role-switch")) - ci_role_switch.fwnode = dev->fwnode; - platdata->pctl = devm_pinctrl_get(dev); if (!IS_ERR(platdata->pctl)) { struct pinctrl_state *p; @@ -1033,6 +1024,7 @@ ATTRIBUTE_GROUPS(ci); static int ci_hdrc_probe(struct platform_device *pdev) { + struct usb_role_switch_desc ci_role_switch = {}; struct device *dev = &pdev->dev; struct ci_hdrc *ci; struct resource *res; @@ -1179,7 +1171,11 @@ static int ci_hdrc_probe(struct platform_device *pdev) } } - if (ci_role_switch.fwnode) { + if (device_property_read_bool(dev, "usb-role-switch")) { + ci_role_switch.set = ci_usb_role_switch_set; + ci_role_switch.get = ci_usb_role_switch_get; + ci_role_switch.allow_userspace_control = true; + ci_role_switch.fwnode = dev_fwnode(dev); ci_role_switch.driver_data = ci; ci->role_switch = usb_role_switch_register(dev, &ci_role_switch); From 9ea06a3fbf9f16e0d98c52cb3b99642be15ec281 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Wed, 20 May 2026 08:59:28 +0300 Subject: [PATCH 146/411] usb: dwc2: Fix use after free in debug code We're not allowed to dereference "urb" after calling usb_hcd_giveback_urb() so save the urb->status ahead of time. Fixes: 7359d482eb4d ("staging: HCD files for the DWC2 driver") Cc: stable Signed-off-by: Dan Carpenter Link: https://patch.msgid.link/ag1NwBpqT4IEQcdJ@stanley.mountain Signed-off-by: Greg Kroah-Hartman --- drivers/usb/dwc2/hcd.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/usb/dwc2/hcd.c b/drivers/usb/dwc2/hcd.c index 1a763ad4f721..2414291aa908 100644 --- a/drivers/usb/dwc2/hcd.c +++ b/drivers/usb/dwc2/hcd.c @@ -4804,6 +4804,7 @@ static int _dwc2_hcd_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, struct dwc2_hsotg *hsotg = dwc2_hcd_to_hsotg(hcd); int rc; unsigned long flags; + int urb_status; dev_dbg(hsotg->dev, "DWC OTG HCD URB Dequeue\n"); dwc2_dump_urb_info(hcd, urb, "urb_dequeue"); @@ -4828,11 +4829,12 @@ static int _dwc2_hcd_urb_dequeue(struct usb_hcd *hcd, struct urb *urb, /* Higher layer software sets URB status */ spin_unlock(&hsotg->lock); + urb_status = urb->status; usb_hcd_giveback_urb(hcd, urb, status); spin_lock(&hsotg->lock); dev_dbg(hsotg->dev, "Called usb_hcd_giveback_urb()\n"); - dev_dbg(hsotg->dev, " urb->status = %d\n", urb->status); + dev_dbg(hsotg->dev, " urb->status = %d\n", urb_status); out: spin_unlock_irqrestore(&hsotg->lock, flags); From 5eb070769ea5e18405535609d1d3f6886f3755bd Mon Sep 17 00:00:00 2001 From: Wentao Guan Date: Fri, 22 May 2026 17:13:58 +0800 Subject: [PATCH 147/411] USB: cdc-acm: Fix bit overlap and move quirk definitions to header The VENDOR_CLASS_DATA_IFACE and ALWAYS_POLL_CTRL quirk flags added in commit f58752ebcb35 ("USB: cdc-acm: Add quirks for Yoga Book 9 14IAH10 INGENIC touchscreen") were placed inside the acm_ctrl_msg() function rather than in the header with the other quirk flags. Then, their values (BIT(9) and BIT(10)) collided with NO_UNION_12 which is already BIT(9). Move the definitions to drivers/usb/class/cdc-acm.h where they belong and shift them to BIT(10) and BIT(11) to avoid the overlap. Fixes: f58752ebcb35 ("USB: cdc-acm: Add quirks for Yoga Book 9 14IAH10 INGENIC touchscreen") Cc: stable Signed-off-by: Wentao Guan Link: https://patch.msgid.link/20260522091357.1301196-1-guanwentao@uniontech.com Signed-off-by: Greg Kroah-Hartman --- drivers/usb/class/cdc-acm.c | 2 -- drivers/usb/class/cdc-acm.h | 2 ++ 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c index 54059e4fc6ed..ddf0b5963859 100644 --- a/drivers/usb/class/cdc-acm.c +++ b/drivers/usb/class/cdc-acm.c @@ -114,8 +114,6 @@ static int acm_ctrl_msg(struct acm *acm, int request, int value, int retval; retval = usb_autopm_get_interface(acm->control); -#define VENDOR_CLASS_DATA_IFACE BIT(9) /* data interface uses vendor-specific class */ -#define ALWAYS_POLL_CTRL BIT(10) /* keep ctrl URB active even without an open TTY */ if (retval) return retval; diff --git a/drivers/usb/class/cdc-acm.h b/drivers/usb/class/cdc-acm.h index 25fd5329a878..01f448a783c0 100644 --- a/drivers/usb/class/cdc-acm.h +++ b/drivers/usb/class/cdc-acm.h @@ -115,3 +115,5 @@ struct acm { #define DISABLE_ECHO BIT(7) #define MISSING_CAP_BRK BIT(8) #define NO_UNION_12 BIT(9) +#define VENDOR_CLASS_DATA_IFACE BIT(10) /* data interface uses vendor-specific class */ +#define ALWAYS_POLL_CTRL BIT(11) /* keep ctrl URB active even without an open TTY */ From ea66be25f0e934f49d24cd0c5845d13cdba3520b Mon Sep 17 00:00:00 2001 From: Myeonghun Pak Date: Tue, 12 May 2026 15:56:57 +0900 Subject: [PATCH 148/411] serial: altera_jtaguart: handle uart_add_one_port() failures altera_jtaguart_probe() maps the register window before registering the UART port, but it ignores failures from uart_add_one_port(). If port registration fails, probe still returns success and the mapping remains live until a later remove path that is not part of probe failure cleanup. Return the uart_add_one_port() error and unmap the register window on that failure path. This issue was identified during our ongoing static-analysis research while reviewing kernel code. Fixes: 5bcd601049c6 ("serial: Add driver for the Altera JTAG UART") Cc: stable Co-developed-by: Ijae Kim Signed-off-by: Ijae Kim Signed-off-by: Myeonghun Pak Link: https://patch.msgid.link/20260512065837.79528-1-mhun512@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/altera_jtaguart.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/altera_jtaguart.c b/drivers/tty/serial/altera_jtaguart.c index d47a62d1c9f7..20f079fe11d8 100644 --- a/drivers/tty/serial/altera_jtaguart.c +++ b/drivers/tty/serial/altera_jtaguart.c @@ -379,6 +379,7 @@ static int altera_jtaguart_probe(struct platform_device *pdev) struct resource *res_mem; int i = pdev->id; int irq; + int ret; /* -1 emphasizes that the platform must have one port, no .N suffix */ if (i == -1) @@ -418,7 +419,11 @@ static int altera_jtaguart_probe(struct platform_device *pdev) port->flags = UPF_BOOT_AUTOCONF; port->dev = &pdev->dev; - uart_add_one_port(&altera_jtaguart_driver, port); + ret = uart_add_one_port(&altera_jtaguart_driver, port); + if (ret) { + iounmap(port->membase); + return ret; + } return 0; } From a3bb136bff5e6a5e48cdd813246c9c4686feaaa9 Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Fri, 15 May 2026 12:41:21 +0000 Subject: [PATCH 149/411] tty: serial: samsung: Remove redundant port lock acquisition in rx helpers Sashiko identified a deadlock when the console flow is engaged [1]. When console flow control is enabled (UPF_CONS_FLOW), s3c24xx_serial_stop_tx() calls s3c24xx_serial_rx_enable() and s3c24xx_serial_start_tx() calls s3c24xx_serial_rx_disable(). The serial core framework invokes the .stop_tx() and .start_tx() callbacks with the port->lock spinlock already held. Furthermore, all internal driver paths that invoke stop_tx (such as the DMA TX completion handler s3c24xx_serial_tx_dma_complete() or the PIO TX IRQ handler s3c24xx_serial_tx_irq()) also acquire port->lock prior to calling it. (Note that s3c24xx_serial_start_tx() is only invoked by the serial core). However, s3c24xx_serial_rx_enable() and s3c24xx_serial_rx_disable() unconditionally attempt to acquire port->lock again using uart_port_lock_irqsave(). Since spinlocks are not recursive, this causes a deadlock on the same CPU when console flow control is engaged. Remove the redundant lock acquisition from both rx helper functions. Cc: stable Fixes: b497549a035c ("[ARM] S3C24XX: Split serial driver into core and per-cpu drivers") Reported-by: John Ogness Closes: https://sashiko.dev/#/patchset/20260506121606.5805-1-john.ogness%40linutronix.de [1] Signed-off-by: Tudor Ambarus Link: https://patch.msgid.link/20260515-samsung-tty-flow-control-deadlock-v1-1-93255edbc9bc@linaro.org Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/samsung_tty.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/tty/serial/samsung_tty.c b/drivers/tty/serial/samsung_tty.c index e27806bf2cf3..17cd5bb100b1 100644 --- a/drivers/tty/serial/samsung_tty.c +++ b/drivers/tty/serial/samsung_tty.c @@ -245,12 +245,9 @@ static bool s3c24xx_serial_txempty_nofifo(const struct uart_port *port) static void s3c24xx_serial_rx_enable(struct uart_port *port) { struct s3c24xx_uart_port *ourport = to_ourport(port); - unsigned long flags; int count = 10000; u32 ucon, ufcon; - uart_port_lock_irqsave(port, &flags); - while (--count && !s3c24xx_serial_txempty_nofifo(port)) udelay(100); @@ -263,23 +260,18 @@ static void s3c24xx_serial_rx_enable(struct uart_port *port) wr_regl(port, S3C2410_UCON, ucon); ourport->rx_enabled = 1; - uart_port_unlock_irqrestore(port, flags); } static void s3c24xx_serial_rx_disable(struct uart_port *port) { struct s3c24xx_uart_port *ourport = to_ourport(port); - unsigned long flags; u32 ucon; - uart_port_lock_irqsave(port, &flags); - ucon = rd_regl(port, S3C2410_UCON); ucon &= ~S3C2410_UCON_RXIRQMODE; wr_regl(port, S3C2410_UCON, ucon); ourport->rx_enabled = 0; - uart_port_unlock_irqrestore(port, flags); } static void s3c24xx_serial_stop_tx(struct uart_port *port) From c3cce2e67bb22a223f5b8ef05db0fcde70994068 Mon Sep 17 00:00:00 2001 From: Jacques Nilo Date: Wed, 13 May 2026 15:30:23 +0200 Subject: [PATCH 150/411] serial: core: introduce guard(uart_port_lock_check_sysrq_irqsave) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit uart_handle_break() and uart_prepare_sysrq_char() (in include/linux/serial_core.h) capture a SysRq character into port->sysrq_ch while the port lock is held and rely on the unlock helper -- uart_unlock_and_check_sysrq_irqrestore() -- to dispatch the captured character to handle_sysrq() on scope exit. The existing guard(uart_port_lock_irqsave) cannot be used by IRQ handlers that process RX, because its destructor calls plain uart_port_unlock_irqrestore() and silently drops port->sysrq_ch. Add a dedicated guard(uart_port_lock_check_sysrq_irqsave) variant whose destructor is the sysrq-aware unlock helper. The lock side is identical to uart_port_lock_irqsave -- only the unlock-time behaviour differs. Callers that may capture SysRq characters must use guard(uart_port_lock_check_sysrq_irqsave); the existing guard(uart_port_lock_irqsave) keeps its current plain-unlock semantics for the many callers that do not process RX. The new macro is placed after the CONFIG_MAGIC_SYSRQ_SERIAL block so both definitions of uart_unlock_and_check_sysrq_irqrestore() (sysrq enabled and disabled) are visible at expansion time. When CONFIG_MAGIC_SYSRQ_SERIAL=n the destructor degenerates to plain uart_port_unlock_irqrestore(), so there is no overhead. No functional change on its own; users are converted in the following patches. Cc: stable@vger.kernel.org Signed-off-by: Jacques Nilo Reviewed-by: Ilpo Järvinen Link: https://patch.msgid.link/3849af4bc55d5d2a424fa850844e94d641b2f8a6.1778675349.git.jnilo@free.fr Signed-off-by: Greg Kroah-Hartman --- include/linux/serial_core.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index 666430b47899..110ad4e2aef9 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -1274,6 +1274,18 @@ static inline void uart_unlock_and_check_sysrq_irqrestore(struct uart_port *port } #endif /* CONFIG_MAGIC_SYSRQ_SERIAL */ +/* + * Variant of guard(uart_port_lock_irqsave) for IRQ handlers that may capture + * a SysRq character via uart_prepare_sysrq_char(). The destructor uses the + * sysrq-aware unlock helper so that a captured port->sysrq_ch is dispatched + * to handle_sysrq() on scope exit. The plain guard variant silently drops + * sysrq_ch and must not be used by callers that process RX. + */ +DEFINE_LOCK_GUARD_1(uart_port_lock_check_sysrq_irqsave, struct uart_port, + uart_port_lock_irqsave(_T->lock, &_T->flags), + uart_unlock_and_check_sysrq_irqrestore(_T->lock, _T->flags), + unsigned long flags); + /* * We do the SysRQ and SAK checking like this... */ From 71f42b2149a1307a97165b409493665579462ea0 Mon Sep 17 00:00:00 2001 From: Jacques Nilo Date: Wed, 13 May 2026 15:30:24 +0200 Subject: [PATCH 151/411] serial: 8250: dispatch SysRq character in serial8250_handle_irq() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit serial8250_handle_irq() captures a SysRq character into port->sysrq_ch inside serial8250_handle_irq_locked() via uart_prepare_sysrq_char() (reached from serial8250_read_char()). Dispatch of that captured character to handle_sysrq() is expected to happen at port-unlock time, through uart_unlock_and_check_sysrq[_irqrestore](). After commit 8324a54f604d ("serial: 8250: Add serial8250_handle_irq_locked()") the function was reduced to a wrapper that takes the port lock via guard(uart_port_lock_irqsave) whose destructor is plain uart_port_unlock_irqrestore(). The sysrq-aware unlock helper is no longer called, so port->sysrq_ch is captured but never dispatched: BREAK + SysRq key is consumed silently. This was the very condition Johan Hovold's 853a9ae29e978 ("serial: 8250: fix handle_irq locking", 2021) introduced uart_unlock_and_check_sysrq_irqrestore() to address. Switch to the new guard(uart_port_lock_check_sysrq_irqsave), whose destructor is the sysrq-aware unlock helper, restoring the pre-split behaviour. Update the Context: comment on serial8250_handle_irq_locked() so future HW-specific 8250 wrappers know to use the same guard or the explicit sysrq-aware unlock. Verified on RTL8196E with CONFIG_MAGIC_SYSRQ_SERIAL=y: BREAK + 'h' on the console UART produces the SysRq help dump in dmesg and the brk counter in /proc/tty/driver/serial increments correctly. Fixes: 8324a54f604d ("serial: 8250: Add serial8250_handle_irq_locked()") Cc: stable@vger.kernel.org Reviewed-by: Ilpo Järvinen Signed-off-by: Jacques Nilo Link: https://patch.msgid.link/52692ae6c3501f7940347cef364ad7fcacaab7e5.1778675349.git.jnilo@free.fr Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_port.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index af78cc02f38e..c66ba714caa5 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -1784,7 +1784,10 @@ static bool handle_rx_dma(struct uart_8250_port *up, unsigned int iir) } /* - * Context: port's lock must be held by the caller. + * Context: port's lock must be held by the caller. The caller must + * release it via guard(uart_port_lock_check_sysrq_irqsave) or + * uart_unlock_and_check_sysrq_irqrestore(), which captures SysRq + * character on unlock. */ void serial8250_handle_irq_locked(struct uart_port *port, unsigned int iir) { @@ -1837,7 +1840,7 @@ int serial8250_handle_irq(struct uart_port *port, unsigned int iir) if (iir & UART_IIR_NO_INT) return 0; - guard(uart_port_lock_irqsave)(port); + guard(uart_port_lock_check_sysrq_irqsave)(port); serial8250_handle_irq_locked(port, iir); return 1; From 2e211723953f7740e54b53f3d3a0d5e351a5e223 Mon Sep 17 00:00:00 2001 From: Jacques Nilo Date: Wed, 13 May 2026 15:30:25 +0200 Subject: [PATCH 152/411] serial: 8250_dw: dispatch SysRq character in dw8250_handle_irq() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit dw8250_handle_irq() calls serial8250_handle_irq_locked() with the port lock held via guard(uart_port_lock_irqsave). The guard destructor is plain uart_port_unlock_irqrestore(), so a SysRq character captured into port->sysrq_ch by uart_prepare_sysrq_char() is dropped without ever being dispatched to handle_sysrq(). This is the same regression pattern as in serial8250_handle_irq(), introduced when 883c5a2bc934 ("serial: 8250_dw: Rework dw8250_handle_irq() locking and IIR handling") moved the function to the guard()-based locking scheme without using the sysrq-aware unlock helper. Switch to guard(uart_port_lock_check_sysrq_irqsave) so that captured sysrq_ch is dispatched on scope exit, matching the fix in serial8250_handle_irq(). Fixes: 883c5a2bc934 ("serial: 8250_dw: Rework dw8250_handle_irq() locking and IIR handling") Cc: stable@vger.kernel.org Reviewed-by: Ilpo Järvinen Signed-off-by: Jacques Nilo Link: https://patch.msgid.link/ed56fcaf4af24e4ed011a7bce206e0182acb761c.1778675349.git.jnilo@free.fr Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_dw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c index 94beadb4024d..2af0c4d0ad82 100644 --- a/drivers/tty/serial/8250/8250_dw.c +++ b/drivers/tty/serial/8250/8250_dw.c @@ -427,7 +427,7 @@ static int dw8250_handle_irq(struct uart_port *p) unsigned int quirks = d->pdata->quirks; unsigned int status; - guard(uart_port_lock_irqsave)(p); + guard(uart_port_lock_check_sysrq_irqsave)(p); switch (FIELD_GET(DW_UART_IIR_IID, iir)) { case UART_IIR_NO_INT: From ca904f4b42355287bc5ce8b7550ebe909cda4c2c Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:31 +0100 Subject: [PATCH 153/411] serial: dz: Fix bootconsole message clobbering at chip reset In the DZ interface as implemented by the DC7085 gate array the serial transmitters are double buffered, meaning that at the time a transmitter is ready to accept the next character there is one in the transmit shift register still being sent to the line. Issuing a master clear at this time causes this character to be lost, so wait an extra amount of time sufficient for the transmit shift register to drain at 9600bps, which is the baud rate setting used by the firmware console. Mind the specified 1.4us TRDY recovery time in the course and continue using iob() as the completion barrier, since the platforms involved use a write buffer that can delay and combine writes, and reorder them with respect to reads regardless of the MMIO locations accessed and we still lack a platform-independent handler for that. When called from dz_serial_console_init() this is too early for fsleep() to work and even before lpj has been calculated and therefore the delay is actually not sufficient for the transmitter to drain and is merely a placeholder now. This will be addressed in a follow-up change. Fixes: e6ee512f5a77 ("dz.c: Resource management") Signed-off-by: Maciej W. Rozycki Cc: stable@vger.kernel.org # v2.6.25+ Link: https://patch.msgid.link/alpine.DEB.2.21.2605062259080.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/dz.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/drivers/tty/serial/dz.c b/drivers/tty/serial/dz.c index e53c54353c3e..d14a40e8cc0a 100644 --- a/drivers/tty/serial/dz.c +++ b/drivers/tty/serial/dz.c @@ -542,10 +542,31 @@ static int dz_encode_baud_rate(unsigned int baud) static void dz_reset(struct dz_port *dport) { struct dz_mux *mux = dport->mux; + unsigned short tcr; + int loops = 10000; if (mux->initialised) return; + tcr = dz_in(dport, DZ_TCR); + + /* Do not disturb any ongoing transmissions. */ + if (dz_in(dport, DZ_CSR) & DZ_MSE) { + unsigned short csr, mask; + + mask = tcr; + while ((mask & DZ_LNENB) && loops--) { + csr = dz_in(dport, DZ_CSR); + if (!(csr & DZ_TRDY)) + continue; + mask &= ~(1 << ((csr & DZ_TLINE) >> 8)); + dz_out(dport, DZ_TCR, mask); + iob(); + udelay(2); /* 1.4us TRDY recovery. */ + } + udelay(1200); /* Transmitter drain. */ + } + dz_out(dport, DZ_CSR, DZ_CLR); while (dz_in(dport, DZ_CSR) & DZ_CLR); iob(); From 7f127b2208e5e2b817243cad41fe4211a6d5a7a3 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:35 +0100 Subject: [PATCH 154/411] serial: dz: Fix bootconsole handover lockup Calling dz_reset() in the course of setting up the serial device causes line parameters to be reset and the transmitter disabled. We've been lucky in that no message is usually produced to the kernel log between this call and the later call to uart_set_options() in the course of console setup done by dz_serial_console_init(), or the system would hang as the console output handler in the firmware tried to access a port the transmitter of which has been disabled and line parameters messed up. This will change with the next change to the driver, so fix dz_reset() such that line parameters are set for 9600n8 console operation as with the system firmware and the transmitter re-enabled after reset. This also means dz_pm() serves no purpose anymore, so drop it. Fixes: e6ee512f5a77 ("dz.c: Resource management") Signed-off-by: Maciej W. Rozycki Cc: stable@vger.kernel.org # v2.6.25+ Link: https://patch.msgid.link/alpine.DEB.2.21.2605062302010.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/dz.c | 36 ++++++++++++------------------------ 1 file changed, 12 insertions(+), 24 deletions(-) diff --git a/drivers/tty/serial/dz.c b/drivers/tty/serial/dz.c index d14a40e8cc0a..03f4fc9248b8 100644 --- a/drivers/tty/serial/dz.c +++ b/drivers/tty/serial/dz.c @@ -571,6 +571,18 @@ static void dz_reset(struct dz_port *dport) while (dz_in(dport, DZ_CSR) & DZ_CLR); iob(); + /* + * Set parameters across all lines such as not to interfere + * with the initial PROM-based console. Otherwise any output + * produced before the console handover would cause the system + * firmware to produce rubbish. + */ + for (int line = 0; line < DZ_NB_PORT; line++) + dz_out(dport, DZ_LPR, DZ_B9600 | DZ_CS8 | line); + + /* Re-enable transmission for the initial PROM-based console. */ + dz_out(dport, DZ_TCR, tcr); + /* Enable scanning. */ dz_out(dport, DZ_CSR, DZ_MSE); @@ -654,26 +666,6 @@ static void dz_set_termios(struct uart_port *uport, struct ktermios *termios, uart_port_unlock_irqrestore(&dport->port, flags); } -/* - * Hack alert! - * Required solely so that the initial PROM-based console - * works undisturbed in parallel with this one. - */ -static void dz_pm(struct uart_port *uport, unsigned int state, - unsigned int oldstate) -{ - struct dz_port *dport = to_dport(uport); - unsigned long flags; - - uart_port_lock_irqsave(&dport->port, &flags); - if (state < 3) - dz_start_tx(&dport->port); - else - dz_stop_tx(&dport->port); - uart_port_unlock_irqrestore(&dport->port, flags); -} - - static const char *dz_type(struct uart_port *uport) { return "DZ"; @@ -769,7 +761,6 @@ static const struct uart_ops dz_ops = { .startup = dz_startup, .shutdown = dz_shutdown, .set_termios = dz_set_termios, - .pm = dz_pm, .type = dz_type, .release_port = dz_release_port, .request_port = dz_request_port, @@ -894,10 +885,7 @@ static int __init dz_console_setup(struct console *co, char *options) if (ret) return ret; - spin_lock_init(&dport->port.lock); /* For dz_pm(). */ - dz_reset(dport); - dz_pm(uport, 0, -1); if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); From 6c05cf72e13314ce9b770b5951695dc5a2152920 Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:39 +0100 Subject: [PATCH 155/411] serial: zs: Fix bootconsole handover lockup Calling zs_reset() in the course of setting up the serial device causes line parameters to be reset and the transmitter disabled. We've been lucky in that no message is usually produced to the kernel log between this call and the later call to uart_set_options() in the course of console setup done by zs_serial_console_init(), or the system would hang as the console output handler in the firmware tried to access a port the transmitter of which has been disabled and line parameters messed up. This will change with the next change to the driver, so fix zs_reset() such that line parameters are set for 9600n8 console operation as with the system firmware and the transmitter re-enabled after reset. This also means zs_pm() serves no purpose anymore, so drop it. Fixes: 8b4a40809e53 ("zs: move to the serial subsystem") Signed-off-by: Maciej W. Rozycki Cc: stable@vger.kernel.org # v2.6.23+ Link: https://patch.msgid.link/alpine.DEB.2.21.2605062308040.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/zs.c | 29 ++++++++--------------------- 1 file changed, 8 insertions(+), 21 deletions(-) diff --git a/drivers/tty/serial/zs.c b/drivers/tty/serial/zs.c index 31e12645f66b..bb10cd8aa7dc 100644 --- a/drivers/tty/serial/zs.c +++ b/drivers/tty/serial/zs.c @@ -105,18 +105,24 @@ struct zs_parms { static struct zs_scc zs_sccs[ZS_NUM_SCCS]; +/* + * Set parameters in WR5, WR12, WR13 such as not to interfere + * with the initial PROM-based console. Otherwise any output + * produced before the console handover would cause the system + * firmware to hang (TxENAB) or produce rubbish (Tx8, B9600). + */ static u8 zs_init_regs[ZS_NUM_REGS] __initdata = { 0, /* write 0 */ PAR_SPEC, /* write 1 */ 0, /* write 2 */ 0, /* write 3 */ X16CLK | SB1, /* write 4 */ - 0, /* write 5 */ + Tx8 | TxENAB, /* write 5 */ 0, 0, 0, /* write 6, 7, 8 */ MIE | DLC | NV, /* write 9 */ NRZ, /* write 10 */ TCBR | RCBR, /* write 11 */ - 0, 0, /* BRG time constant, write 12 + 13 */ + 0x16, 0x00, /* BRG time constant, write 12 + 13 */ BRSRC | BRENABL, /* write 14 */ 0, /* write 15 */ }; @@ -956,23 +962,6 @@ static void zs_set_termios(struct uart_port *uport, struct ktermios *termios, spin_unlock_irqrestore(&scc->zlock, flags); } -/* - * Hack alert! - * Required solely so that the initial PROM-based console - * works undisturbed in parallel with this one. - */ -static void zs_pm(struct uart_port *uport, unsigned int state, - unsigned int oldstate) -{ - struct zs_port *zport = to_zport(uport); - - if (state < 3) - zport->regs[5] |= TxENAB; - else - zport->regs[5] &= ~TxENAB; - write_zsreg(zport, R5, zport->regs[5]); -} - static const char *zs_type(struct uart_port *uport) { @@ -1055,7 +1044,6 @@ static const struct uart_ops zs_ops = { .startup = zs_startup, .shutdown = zs_shutdown, .set_termios = zs_set_termios, - .pm = zs_pm, .type = zs_type, .release_port = zs_release_port, .request_port = zs_request_port, @@ -1210,7 +1198,6 @@ static int __init zs_console_setup(struct console *co, char *options) return ret; zs_reset(zport); - zs_pm(uport, 0, -1); if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); From 8572955630f30948837088aa98bcbe0532d1ceac Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:43 +0100 Subject: [PATCH 156/411] serial: zs: Switch to using channel reset Switch the driver to using the channel reset rather than hardware reset, simplifying handling by removing an interference between channels that causes the other channel to become uninitialised afterwards. There is little difference between the two kinds of reset in terms of register settings that result, and we initialise the whole register set right away anyway. However this prevents a hang from happening should the console output handler in the firmware try to access the other port whose transmitter has been disabled and line parameters messed up. For example this will happen if the keyboard port (port A) is chosen for the system console, unusually but not insanely for a headless system, as the port is wired to a standard DA-15 connector and an adapter can be easily made. Or with the next change in place this would happen for the regular console port (port B), since the keyboard port (port A) will be initialised first. Just remove the unnecessary complication then, a channel reset is good enough. We still need the initialisation marker, now per channel rather than per SCC, as for the console port zs_reset() will be called twice: once early on via zs_serial_console_init() for the console setup only, and then again via zs_config_port() as the port is associated with a TTY device. Fixes: 8b4a40809e53 ("zs: move to the serial subsystem") Signed-off-by: Maciej W. Rozycki Cc: stable@vger.kernel.org # v2.6.23+ Link: https://patch.msgid.link/alpine.DEB.2.21.2605062323430.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/zs.c | 7 ++++--- drivers/tty/serial/zs.h | 2 +- 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/tty/serial/zs.c b/drivers/tty/serial/zs.c index bb10cd8aa7dc..71cab10a33c3 100644 --- a/drivers/tty/serial/zs.c +++ b/drivers/tty/serial/zs.c @@ -832,21 +832,22 @@ static void zs_shutdown(struct uart_port *uport) static void zs_reset(struct zs_port *zport) { + struct zs_port *zport_a = &zport->scc->zport[ZS_CHAN_A]; struct zs_scc *scc = zport->scc; int irq; unsigned long flags; spin_lock_irqsave(&scc->zlock, flags); irq = !irqs_disabled_flags(flags); - if (!scc->initialised) { + if (!zport->initialised) { /* Reset the pointer first, just in case... */ read_zsreg(zport, R0); /* And let the current transmission finish. */ zs_line_drain(zport, irq); - write_zsreg(zport, R9, FHWRES); + write_zsreg(zport, R9, zport == zport_a ? CHRA : CHRB); udelay(10); write_zsreg(zport, R9, 0); - scc->initialised = 1; + zport->initialised = 1; } load_zsregs(zport, zport->regs, irq); spin_unlock_irqrestore(&scc->zlock, flags); diff --git a/drivers/tty/serial/zs.h b/drivers/tty/serial/zs.h index 26ef8eafa1c1..8e51f847bc03 100644 --- a/drivers/tty/serial/zs.h +++ b/drivers/tty/serial/zs.h @@ -22,6 +22,7 @@ struct zs_port { struct zs_scc *scc; /* Containing SCC. */ struct uart_port port; /* Underlying UART. */ + int initialised; /* For the console port. */ int clk_mode; /* May be 1, 16, 32, or 64. */ @@ -41,7 +42,6 @@ struct zs_scc { struct zs_port zport[2]; spinlock_t zlock; atomic_t irq_guard; - int initialised; }; #endif /* __KERNEL__ */ From 5d7a49d60b8fda66da60e240fd7315232fa1754f Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:48 +0100 Subject: [PATCH 157/411] serial: dz: Convert to use a platform device Prevent a crash from happening as the first serial port is initialised: Console: switching to colour frame buffer device 160x64 tgafb: SFB+ detected, rev=0x02 fb0: Digital ZLX-E1 frame buffer device at 0x1e000000 DECstation DZ serial driver version 1.04 CPU 0 Unable to handle kernel paging request at virtual address 000000bc, epc == 8048b3a4, ra == 80470a78 Oops[#1]: CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.19.0-dirty #35 NONE $ 0 : 00000000 1000ac00 00000004 804707ac $ 4 : 00000000 80e20850 80e20858 81000030 $ 8 : 00000000 8072c81c 00000008 fefefeff $12 : 6c616972 00000006 80c5917f 69726420 $16 : 80e20800 00000000 808f8968 80e20800 $20 : 00000000 807f5a90 808b0094 808d3bc8 $24 : 00000018 80479030 $28 : 80c2e000 80c2fd70 00000069 80470a78 Hi : 00000004 Lo : 00000000 epc : 8048b3a4 __dev_fwnode+0x0/0xc ra : 80470a78 serial_base_ctrl_add+0xa0/0x168 Status: 1000ac04 IEp Cause : 30000008 (ExcCode 02) BadVA : 000000bc PrId : 00000220 (R3000) Modules linked in: Process swapper/0 (pid: 1, threadinfo=(ptrval), task=(ptrval), tls=00000000) Stack : 00400044 00400040 8046f4cc 00000000 808a6148 808a0000 808f8968 8086983c 808e0000 8046fc84 1000ac01 00000028 80e20700 802ba3f8 80e20700 80d34a94 80c1b900 80e20700 80e20700 80e20700 80e20700 80444650 00000000 00000000 00000000 807f5a90 808b0094 80447080 00400040 808e0000 80d34a94 808a6148 80d34a94 00000004 80e20700 00000000 8076974c 80469810 80c2fe3c 1000ac01 ... Call Trace: [<8048b3a4>] __dev_fwnode+0x0/0xc [<80470a78>] serial_base_ctrl_add+0xa0/0x168 [<8046fc84>] serial_core_register_port+0x1c8/0x974 [<808c6af0>] dz_init+0x74/0xc8 [<800470e0>] do_one_initcall+0x44/0x2d4 [<808b111c>] kernel_init_freeable+0x258/0x308 [<8072e434>] kernel_init+0x20/0x114 [<80049cd0>] ret_from_kernel_thread+0x14/0x1c Code: 27bd0018 03e00008 2402ffea <8c8200bc> 03e00008 00000000 27bdffc0 afbe0038 afb30024 ---[ end trace 0000000000000000 ]--- -- where a pointer is dereferenced that has been derived from a null pointer to the port's parent device. Since no device is available with legacy probing and it's not anymore a preferable way to discover devices anyway, switch the driver to using a platform device and use it as the port's parent device. Update resource handling accordingly and only request the actual span of addresses used within the slot, which will have had its resource already requested by generic platform device code. Use platform_driver_probe() not just because the DZ device is fixed with solder on board and not straightforward to remove, but foremost because the associated TTY's major device number is the same as used by the zs driver and the first driver to claim it will prevent the other one from using it. Either one DZ device or some SCC devices will be present in a given system but never both at a time, and therefore we want the major device number to be claimed by the first driver to actually successfully bind to its device and platform_driver_probe() is a way to fulfil that. An unfortunate consequence of the switch to a platform device is we now hand the console over from the bootconsole much later in the bootstrap. The firmware console handler appears good enough though to work so late and in particular with interrupts enabled. Conversely only starting the console port so late lets the reset code fully utilise our delay handlers, so switch from udelay() to fsleep() for transmitter draining so as to avoid busy-waiting for an excessive amount of time. Fixes: 84a9582fd203 ("serial: core: Start managing serial controllers to enable runtime PM") Signed-off-by: Maciej W. Rozycki Cc: stable@vger.kernel.org # needs to use .remove_new for <= 6.10 Link: https://patch.msgid.link/alpine.DEB.2.21.2605062326540.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- arch/mips/dec/platform.c | 55 ++++++++++++++++++- drivers/tty/serial/dz.c | 116 +++++++++++++++++++-------------------- 2 files changed, 110 insertions(+), 61 deletions(-) diff --git a/arch/mips/dec/platform.c b/arch/mips/dec/platform.c index c4fcb8c58e01..fdecc91ee22a 100644 --- a/arch/mips/dec/platform.c +++ b/arch/mips/dec/platform.c @@ -10,6 +10,13 @@ #include #include +#include + +#include +#include +#include +#include + static struct resource dec_rtc_resources[] = { { .name = "rtc", @@ -30,11 +37,57 @@ static struct platform_device dec_rtc_device = { .num_resources = ARRAY_SIZE(dec_rtc_resources), }; +static struct resource dec_dz_resources[] = { + { .name = "dz", .flags = IORESOURCE_MEM, }, + { .name = "dz", .flags = IORESOURCE_IRQ, }, +}; + +static struct platform_device dec_dz_device = { + .name = "dz", + .id = PLATFORM_DEVID_NONE, + .resource = dec_dz_resources, + .num_resources = ARRAY_SIZE(dec_dz_resources), +}; + +static struct platform_device *dec_dz_devices[] __initdata = { + &dec_dz_device, +}; + static int __init dec_add_devices(void) { + int ret1, ret2; + int num_dz; + int irq, i; + dec_rtc_resources[0].start = RTC_PORT(0); dec_rtc_resources[0].end = RTC_PORT(0) + dec_kn_slot_size - 1; - return platform_device_register(&dec_rtc_device); + + i = 0; + irq = dec_interrupt[DEC_IRQ_DZ11]; + if (IS_ENABLED(CONFIG_32BIT) && irq >= 0) { + resource_size_t base; + + switch (mips_machtype) { + case MACH_DS23100: + case MACH_DS5100: + base = dec_kn_slot_base + KN01_DZ11; + break; + default: + base = dec_kn_slot_base + KN02_DZ11; + break; + } + dec_dz_device.resource[0].start = base; + dec_dz_device.resource[0].end = base + dec_kn_slot_size - 1; + dec_dz_device.resource[1].start = irq; + dec_dz_device.resource[1].end = irq; + i++; + } + num_dz = i; + + ret1 = platform_device_register(&dec_rtc_device); + ret2 = IS_ENABLED(CONFIG_32BIT) ? + platform_add_devices(dec_dz_devices, num_dz) : 0; + return ret1 ? ret1 : ret2; } device_initcall(dec_add_devices); diff --git a/drivers/tty/serial/dz.c b/drivers/tty/serial/dz.c index 03f4fc9248b8..39d93e9c2d15 100644 --- a/drivers/tty/serial/dz.c +++ b/drivers/tty/serial/dz.c @@ -40,6 +40,7 @@ #include #include #include +#include #include #include #include @@ -48,14 +49,6 @@ #include #include -#include - -#include -#include -#include -#include -#include -#include #include "dz.h" @@ -65,7 +58,9 @@ MODULE_LICENSE("GPL"); static char dz_name[] __initdata = "DECstation DZ serial driver version "; -static char dz_version[] __initdata = "1.04"; +static char dz_version[] __initdata = "1.05"; + +#define DZ_IO_SIZE 0x20 /* IOMEM space size. */ struct dz_port { struct dz_mux *mux; @@ -81,6 +76,7 @@ struct dz_mux { }; static struct dz_mux dz_mux; +static struct uart_driver dz_reg; static inline struct dz_port *to_dport(struct uart_port *uport) { @@ -564,7 +560,7 @@ static void dz_reset(struct dz_port *dport) iob(); udelay(2); /* 1.4us TRDY recovery. */ } - udelay(1200); /* Transmitter drain. */ + fsleep(1200); /* Transmitter drain. */ } dz_out(dport, DZ_CSR, DZ_CLR); @@ -681,14 +677,13 @@ static void dz_release_port(struct uart_port *uport) map_guard = atomic_add_return(-1, &mux->map_guard); if (!map_guard) - release_mem_region(uport->mapbase, dec_kn_slot_size); + release_mem_region(uport->mapbase, DZ_IO_SIZE); } static int dz_map_port(struct uart_port *uport) { if (!uport->membase) - uport->membase = ioremap(uport->mapbase, - dec_kn_slot_size); + uport->membase = ioremap(uport->mapbase, DZ_IO_SIZE); if (!uport->membase) { printk(KERN_ERR "dz: Cannot map MMIO\n"); return -ENOMEM; @@ -704,8 +699,7 @@ static int dz_request_port(struct uart_port *uport) map_guard = atomic_add_return(1, &mux->map_guard); if (map_guard == 1) { - if (!request_mem_region(uport->mapbase, dec_kn_slot_size, - "dz")) { + if (!request_mem_region(uport->mapbase, DZ_IO_SIZE, "dz")) { atomic_add(-1, &mux->map_guard); printk(KERN_ERR "dz: Unable to reserve MMIO resource\n"); @@ -716,7 +710,7 @@ static int dz_request_port(struct uart_port *uport) if (ret) { map_guard = atomic_add_return(-1, &mux->map_guard); if (!map_guard) - release_mem_region(uport->mapbase, dec_kn_slot_size); + release_mem_region(uport->mapbase, DZ_IO_SIZE); return ret; } return 0; @@ -768,20 +762,15 @@ static const struct uart_ops dz_ops = { .verify_port = dz_verify_port, }; -static void __init dz_init_ports(void) +static int __init dz_probe(struct platform_device *pdev) { - static int first = 1; - unsigned long base; + struct resource *mem_resource, *irq_resource; int line; - if (!first) - return; - first = 0; - - if (mips_machtype == MACH_DS23100 || mips_machtype == MACH_DS5100) - base = dec_kn_slot_base + KN01_DZ11; - else - base = dec_kn_slot_base + KN02_DZ11; + mem_resource = platform_get_resource(pdev, IORESOURCE_MEM, 0); + irq_resource = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + if (!mem_resource || !irq_resource) + return -ENODEV; for (line = 0; line < DZ_NB_PORT; line++) { struct dz_port *dport = &dz_mux.dport[line]; @@ -789,14 +778,33 @@ static void __init dz_init_ports(void) dport->mux = &dz_mux; - uport->irq = dec_interrupt[DEC_IRQ_DZ11]; + uport->dev = &pdev->dev; + uport->irq = irq_resource->start; uport->fifosize = 1; uport->iotype = UPIO_MEM; uport->flags = UPF_BOOT_AUTOCONF; uport->ops = &dz_ops; uport->line = line; - uport->mapbase = base; + uport->mapbase = mem_resource->start; uport->has_sysrq = IS_ENABLED(CONFIG_SERIAL_DZ_CONSOLE); + + if (uart_add_one_port(&dz_reg, uport)) + uport->dev = NULL; + } + + return 0; +} + +static void __exit dz_remove(struct platform_device *pdev) +{ + int line; + + for (line = DZ_NB_PORT - 1; line >= 0; line--) { + struct dz_port *dport = &dz_mux.dport[line]; + struct uart_port *uport = &dport->port; + + if (uport->dev) + uart_remove_one_port(&dz_reg, uport); } } @@ -879,21 +887,14 @@ static int __init dz_console_setup(struct console *co, char *options) int bits = 8; int parity = 'n'; int flow = 'n'; - int ret; - - ret = dz_map_port(uport); - if (ret) - return ret; - - dz_reset(dport); + if (!dport->mux) + return -ENODEV; if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); - - return uart_set_options(&dport->port, co, baud, parity, bits, flow); + return uart_set_options(uport, co, baud, parity, bits, flow); } -static struct uart_driver dz_reg; static struct console dz_console = { .name = "ttyS", .write = dz_console_print, @@ -904,18 +905,6 @@ static struct console dz_console = { .data = &dz_reg, }; -static int __init dz_serial_console_init(void) -{ - if (!IOASIC) { - dz_init_ports(); - register_console(&dz_console); - return 0; - } else - return -ENXIO; -} - -console_initcall(dz_serial_console_init); - #define SERIAL_DZ_CONSOLE &dz_console #else #define SERIAL_DZ_CONSOLE NULL @@ -931,25 +920,32 @@ static struct uart_driver dz_reg = { .cons = SERIAL_DZ_CONSOLE, }; +static struct platform_driver dz_driver = { + .remove = __exit_p(dz_remove), + .driver = { .name = "dz" }, +}; + static int __init dz_init(void) { - int ret, i; - - if (IOASIC) - return -ENXIO; + int ret; printk("%s%s\n", dz_name, dz_version); - dz_init_ports(); - ret = uart_register_driver(&dz_reg); if (ret) return ret; + ret = platform_driver_probe(&dz_driver, dz_probe); + if (ret) + uart_unregister_driver(&dz_reg); - for (i = 0; i < DZ_NB_PORT; i++) - uart_add_one_port(&dz_reg, &dz_mux.dport[i].port); + return ret; +} - return 0; +static void __exit dz_exit(void) +{ + platform_driver_unregister(&dz_driver); + uart_unregister_driver(&dz_reg); } module_init(dz_init); +module_exit(dz_exit); From 7cac59d08a73cb866ec51a483a6f3fe0f531947c Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:52 +0100 Subject: [PATCH 158/411] serial: zs: Convert to use a platform device Prevent a crash from happening as the first serial port is initialised: Console: switching to mono frame buffer device 160x64 fb0: PMAG-AA frame buffer device at tc0 DECstation Z85C30 serial driver version 0.10 CPU 0 Unable to handle kernel paging request at virtual address 0000002c, epc == 803ab00c, ra == 803aafe0 Oops[#1]: CPU: 0 PID: 1 Comm: swapper Not tainted 6.4.0-rc3-00031-g84a9582fd203-dirty #57 $ 0 : 00000000 10012c00 803aaeb0 00000000 $ 4 : 80e12f60 80e12f50 80e12f58 81000030 $ 8 : 00000000 805ff37c 00000000 33433538 $12 : 65732030 00000006 80c2915d 6c616972 $16 : 80e12f00 807b7630 00000000 00000000 $20 : 00000004 00000348 000001a0 807623b8 $24 : 00000018 00000000 $28 : 80c24000 80c25d60 8078b148 803aafe0 Hi : 00000000 Lo : 00000000 epc : 803ab00c serial_base_ctrl_add+0x78/0xf4 ra : 803aafe0 serial_base_ctrl_add+0x4c/0xf4 Status: 10012c03 KERNEL EXL IE Cause : 00000008 (ExcCode 02) BadVA : 0000002c PrId : 00000440 (R4400SC) Modules linked in: Process swapper (pid: 1, threadinfo=(ptrval), task=(ptrval), tls=00000000) Stack : 80760000 00000cc0 00400044 00400040 803aa02c 80d61ab8 00000000 807b7630 80760000 807623b8 807b7628 803aa644 80386998 00000000 80e17780 80220f68 80e17780 80d61ab8 80c17d80 80e17780 80e17780 8063c798 80e17780 80383fa0 00000010 80e17780 00000000 80386998 807a0000 00000000 00400040 8038f848 807623b8 80d61ab8 00000004 80e17780 00000000 803a68e4 80c25e2c 803bb884 ... Call Trace: [<803ab00c>] serial_base_ctrl_add+0x78/0xf4 [<803aa644>] serial_core_register_port+0x174/0x69c [<8077e9ac>] zs_init+0xc8/0xfc [<800404d4>] do_one_initcall+0x40/0x2ac [<8076cecc>] kernel_init_freeable+0x1e4/0x270 [<80605bec>] kernel_init+0x20/0x108 [<800431e8>] ret_from_kernel_thread+0x14/0x1c Code: 2442aeb0 ae120024 ae0200d0 <8c67002c> 50e00001 8c670000 3c06806e 3c05806e afb30010 ---[ end trace 0000000000000000 ]--- (report at the offending commit) -- where a pointer is dereferenced that has been derived from a null pointer to the port's parent device. Since no device is available with legacy probing and it's not anymore a preferable way to discover devices anyway, switch the driver to using a platform device and use it as the port's parent device. Update resource handling accordingly and only request the actual span of addresses used within the slot, which will have had its resource already requested by generic platform device code. Use platform_driver_probe() not just because SCC devices are fixed with solder on board and not straightforward to remove, but foremost because the associated TTY's major device number is the same as used by the dz driver and the first driver to claim it will prevent the other one from using it. Either one DZ device or some SCC devices will be present in a given system but never both at a time, and therefore we want the major device number to be claimed by the first driver to actually successfully bind to its device and platform_driver_probe() is a way to fulfil that. An unfortunate consequence of the switch to a platform device is we now hand the console over from the bootconsole much later in the bootstrap. The firmware console handler appears good enough though to work so late and in particular with interrupts enabled. Since there is one way only remaining to reach zs_reset() now, remove the port initialisation marker as no longer needed and go through the channel reset unconditionally. Fixes: 84a9582fd203 ("serial: core: Start managing serial controllers to enable runtime PM") Signed-off-by: Maciej W. Rozycki Cc: stable@vger.kernel.org # needs to use .remove_new for <= 6.10 Link: https://patch.msgid.link/alpine.DEB.2.21.2605062328480.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- arch/mips/dec/platform.c | 60 ++++++++++++- drivers/tty/serial/zs.c | 184 +++++++++++++++------------------------ drivers/tty/serial/zs.h | 1 - 3 files changed, 125 insertions(+), 120 deletions(-) diff --git a/arch/mips/dec/platform.c b/arch/mips/dec/platform.c index fdecc91ee22a..723ce16cbfc0 100644 --- a/arch/mips/dec/platform.c +++ b/arch/mips/dec/platform.c @@ -13,6 +13,7 @@ #include #include +#include #include #include #include @@ -53,10 +54,37 @@ static struct platform_device *dec_dz_devices[] __initdata = { &dec_dz_device, }; +static struct resource dec_zs_resources[][2] = { + { + { .name = "scc0", .flags = IORESOURCE_MEM, }, + { .name = "scc0", .flags = IORESOURCE_IRQ, }, + }, + { + { .name = "scc1", .flags = IORESOURCE_MEM, }, + { .name = "scc1", .flags = IORESOURCE_IRQ, }, + }, +}; + +static struct platform_device dec_zs_device[] = { + { + .name = "zs", + .id = 0, + .resource = dec_zs_resources[0], + .num_resources = ARRAY_SIZE(dec_zs_resources[0]), + }, + { + .name = "zs", + .id = 1, + .resource = dec_zs_resources[1], + .num_resources = ARRAY_SIZE(dec_zs_resources[1]), + }, +}; + static int __init dec_add_devices(void) { - int ret1, ret2; - int num_dz; + struct platform_device *dec_zs_devices[ARRAY_SIZE(dec_zs_device)]; + int ret1, ret2, ret3; + int num_dz, num_zs; int irq, i; dec_rtc_resources[0].start = RTC_PORT(0); @@ -84,10 +112,36 @@ static int __init dec_add_devices(void) } num_dz = i; + i = 0; + irq = dec_interrupt[DEC_IRQ_SCC0]; + if (irq >= 0) { + resource_size_t base = dec_kn_slot_base + IOASIC_SCC0; + + dec_zs_device[i].resource[0].start = base; + dec_zs_device[i].resource[0].end = base + dec_kn_slot_size - 1; + dec_zs_device[i].resource[1].start = irq; + dec_zs_device[i].resource[1].end = irq; + dec_zs_devices[i] = &dec_zs_device[i]; + i++; + } + irq = dec_interrupt[DEC_IRQ_SCC1]; + if (irq >= 0) { + resource_size_t base = dec_kn_slot_base + IOASIC_SCC1; + + dec_zs_device[i].resource[0].start = base; + dec_zs_device[i].resource[0].end = base + dec_kn_slot_size - 1; + dec_zs_device[i].resource[1].start = irq; + dec_zs_device[i].resource[1].end = irq; + dec_zs_devices[i] = &dec_zs_device[i]; + i++; + } + num_zs = i; + ret1 = platform_device_register(&dec_rtc_device); ret2 = IS_ENABLED(CONFIG_32BIT) ? platform_add_devices(dec_dz_devices, num_dz) : 0; - return ret1 ? ret1 : ret2; + ret3 = platform_add_devices(dec_zs_devices, num_zs); + return ret1 ? ret1 : ret2 ? ret2 : ret3; } device_initcall(dec_add_devices); diff --git a/drivers/tty/serial/zs.c b/drivers/tty/serial/zs.c index 71cab10a33c3..8f92b4129a38 100644 --- a/drivers/tty/serial/zs.c +++ b/drivers/tty/serial/zs.c @@ -56,6 +56,7 @@ #include #include #include +#include #include #include #include @@ -66,10 +67,6 @@ #include -#include -#include -#include - #include "zs.h" @@ -79,7 +76,7 @@ MODULE_LICENSE("GPL"); static char zs_name[] __initdata = "DECstation Z85C30 serial driver version "; -static char zs_version[] __initdata = "0.10"; +static char zs_version[] __initdata = "0.11"; /* * It would be nice to dynamically allocate everything that @@ -98,12 +95,8 @@ static char zs_version[] __initdata = "0.10"; #define to_zport(uport) container_of(uport, struct zs_port, port) -struct zs_parms { - resource_size_t scc[ZS_NUM_SCCS]; - int irq[ZS_NUM_SCCS]; -}; - static struct zs_scc zs_sccs[ZS_NUM_SCCS]; +static struct uart_driver zs_reg; /* * Set parameters in WR5, WR12, WR13 such as not to interfere @@ -839,16 +832,15 @@ static void zs_reset(struct zs_port *zport) spin_lock_irqsave(&scc->zlock, flags); irq = !irqs_disabled_flags(flags); - if (!zport->initialised) { - /* Reset the pointer first, just in case... */ - read_zsreg(zport, R0); - /* And let the current transmission finish. */ - zs_line_drain(zport, irq); - write_zsreg(zport, R9, zport == zport_a ? CHRA : CHRB); - udelay(10); - write_zsreg(zport, R9, 0); - zport->initialised = 1; - } + + /* Reset the pointer first, just in case... */ + read_zsreg(zport, R0); + /* And let the current transmission finish. */ + zs_line_drain(zport, irq); + write_zsreg(zport, R9, zport == zport_a ? CHRA : CHRB); + udelay(10); + write_zsreg(zport, R9, 0); + load_zsregs(zport, zport->regs, irq); spin_unlock_irqrestore(&scc->zlock, flags); } @@ -1055,63 +1047,62 @@ static const struct uart_ops zs_ops = { /* * Initialize Z85C30 port structures. */ -static int __init zs_probe_sccs(void) +static int __init zs_probe(struct platform_device *pdev) { - static int probed; - struct zs_parms zs_parms; - int chip, side, irq; - int n_chips = 0; + struct resource *mem_resource, *irq_resource; + int chip, side; int i; - if (probed) - return 0; + mem_resource = platform_get_resource(pdev, IORESOURCE_MEM, 0); + irq_resource = platform_get_resource(pdev, IORESOURCE_IRQ, 0); + if (!mem_resource || !irq_resource) + return -ENODEV; - irq = dec_interrupt[DEC_IRQ_SCC0]; - if (irq >= 0) { - zs_parms.scc[n_chips] = IOASIC_SCC0; - zs_parms.irq[n_chips] = dec_interrupt[DEC_IRQ_SCC0]; - n_chips++; - } - irq = dec_interrupt[DEC_IRQ_SCC1]; - if (irq >= 0) { - zs_parms.scc[n_chips] = IOASIC_SCC1; - zs_parms.irq[n_chips] = dec_interrupt[DEC_IRQ_SCC1]; - n_chips++; - } - if (!n_chips) - return -ENXIO; + chip = pdev->id; + spin_lock_init(&zs_sccs[chip].zlock); + for (side = 0; side < ZS_NUM_CHAN; side++) { + struct zs_port *zport = &zs_sccs[chip].zport[side]; + struct uart_port *uport = &zport->port; - probed = 1; + zport->scc = &zs_sccs[chip]; + zport->clk_mode = 16; - for (chip = 0; chip < n_chips; chip++) { - spin_lock_init(&zs_sccs[chip].zlock); - for (side = 0; side < ZS_NUM_CHAN; side++) { - struct zs_port *zport = &zs_sccs[chip].zport[side]; - struct uart_port *uport = &zport->port; + uport->dev = &pdev->dev; + uport->has_sysrq = IS_ENABLED(CONFIG_SERIAL_ZS_CONSOLE); + uport->irq = irq_resource->start; + uport->uartclk = ZS_CLOCK; + uport->fifosize = 1; + uport->iotype = UPIO_MEM; + uport->flags = UPF_BOOT_AUTOCONF; + uport->ops = &zs_ops; + uport->line = chip * ZS_NUM_CHAN + side; + uport->mapbase = mem_resource->start + + (side ^ ZS_CHAN_B) * ZS_CHAN_IO_SIZE; - zport->scc = &zs_sccs[chip]; - zport->clk_mode = 16; + for (i = 0; i < ZS_NUM_REGS; i++) + zport->regs[i] = zs_init_regs[i]; - uport->has_sysrq = IS_ENABLED(CONFIG_SERIAL_ZS_CONSOLE); - uport->irq = zs_parms.irq[chip]; - uport->uartclk = ZS_CLOCK; - uport->fifosize = 1; - uport->iotype = UPIO_MEM; - uport->flags = UPF_BOOT_AUTOCONF; - uport->ops = &zs_ops; - uport->line = chip * ZS_NUM_CHAN + side; - uport->mapbase = dec_kn_slot_base + - zs_parms.scc[chip] + - (side ^ ZS_CHAN_B) * ZS_CHAN_IO_SIZE; - - for (i = 0; i < ZS_NUM_REGS; i++) - zport->regs[i] = zs_init_regs[i]; - } + if (uart_add_one_port(&zs_reg, uport)) + uport->dev = NULL; } return 0; } +static void __exit zs_remove(struct platform_device *pdev) +{ + int chip, side; + + chip = pdev->id; + for (side = ZS_NUM_CHAN - 1; side >= 0; side--) { + struct zs_port *zport = &zs_sccs[chip].zport[side]; + struct uart_port *uport = &zport->port; + + if (uport->dev) + uart_remove_one_port(&zs_reg, uport); + } +} + #ifdef CONFIG_SERIAL_ZS_CONSOLE static void zs_console_putchar(struct uart_port *uport, unsigned char ch) @@ -1192,20 +1183,14 @@ static int __init zs_console_setup(struct console *co, char *options) int bits = 8; int parity = 'n'; int flow = 'n'; - int ret; - - ret = zs_map_port(uport); - if (ret) - return ret; - - zs_reset(zport); + if (!zport->scc) + return -ENODEV; if (options) uart_parse_options(options, &baud, &parity, &bits, &flow); return uart_set_options(uport, co, baud, parity, bits, flow); } -static struct uart_driver zs_reg; static struct console zs_console = { .name = "ttyS", .write = zs_console_write, @@ -1216,23 +1201,6 @@ static struct console zs_console = { .data = &zs_reg, }; -/* - * Register console. - */ -static int __init zs_serial_console_init(void) -{ - int ret; - - ret = zs_probe_sccs(); - if (ret) - return ret; - register_console(&zs_console); - - return 0; -} - -console_initcall(zs_serial_console_init); - #define SERIAL_ZS_CONSOLE &zs_console #else #define SERIAL_ZS_CONSOLE NULL @@ -1248,47 +1216,31 @@ static struct uart_driver zs_reg = { .cons = SERIAL_ZS_CONSOLE, }; +static struct platform_driver zs_driver = { + .remove = __exit_p(zs_remove), + .driver = { .name = "zs" }, +}; + /* zs_init inits the driver. */ static int __init zs_init(void) { - int i, ret; + int ret; pr_info("%s%s\n", zs_name, zs_version); - /* Find out how many Z85C30 SCCs we have. */ - ret = zs_probe_sccs(); - if (ret) - return ret; - ret = uart_register_driver(&zs_reg); if (ret) return ret; + ret = platform_driver_probe(&zs_driver, zs_probe); + if (ret) + uart_unregister_driver(&zs_reg); - for (i = 0; i < ZS_NUM_SCCS * ZS_NUM_CHAN; i++) { - struct zs_scc *scc = &zs_sccs[i / ZS_NUM_CHAN]; - struct zs_port *zport = &scc->zport[i % ZS_NUM_CHAN]; - struct uart_port *uport = &zport->port; - - if (zport->scc) - uart_add_one_port(&zs_reg, uport); - } - - return 0; + return ret; } static void __exit zs_exit(void) { - int i; - - for (i = ZS_NUM_SCCS * ZS_NUM_CHAN - 1; i >= 0; i--) { - struct zs_scc *scc = &zs_sccs[i / ZS_NUM_CHAN]; - struct zs_port *zport = &scc->zport[i % ZS_NUM_CHAN]; - struct uart_port *uport = &zport->port; - - if (zport->scc) - uart_remove_one_port(&zs_reg, uport); - } - + platform_driver_unregister(&zs_driver); uart_unregister_driver(&zs_reg); } diff --git a/drivers/tty/serial/zs.h b/drivers/tty/serial/zs.h index 8e51f847bc03..e0d3c189b33f 100644 --- a/drivers/tty/serial/zs.h +++ b/drivers/tty/serial/zs.h @@ -22,7 +22,6 @@ struct zs_port { struct zs_scc *scc; /* Containing SCC. */ struct uart_port port; /* Underlying UART. */ - int initialised; /* For the console port. */ int clk_mode; /* May be 1, 16, 32, or 64. */ From e4240d8845445d58b4b96f7066adfe175a61bd0c Mon Sep 17 00:00:00 2001 From: "Maciej W. Rozycki" Date: Wed, 6 May 2026 23:42:56 +0100 Subject: [PATCH 159/411] serial: dz: Enable modular build Enable modular build since the driver now has a proper module_exit() handler. There's nothing specific to DZ hardware to prevent driver unloading and reloading from working. Signed-off-by: Maciej W. Rozycki Link: https://patch.msgid.link/alpine.DEB.2.21.2605062331420.46195@angie.orcam.me.uk Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/Kconfig b/drivers/tty/serial/Kconfig index 9aa61c93d7bc..ec284aceb909 100644 --- a/drivers/tty/serial/Kconfig +++ b/drivers/tty/serial/Kconfig @@ -334,7 +334,7 @@ config SERIAL_MAX310X Say Y here if you want to support this ICs. config SERIAL_DZ - bool "DECstation DZ serial driver" + tristate "DECstation DZ serial driver" depends on MACH_DECSTATION && 32BIT select SERIAL_CORE default y From 4c19719eb8b8df08c5bec7c499f73ddaea6f09fc Mon Sep 17 00:00:00 2001 From: Alice Ryhl Date: Tue, 14 Apr 2026 12:02:34 +0000 Subject: [PATCH 160/411] rust_binder: avoid calling pending_oneway_finished() on TF_UPDATE_TXN When an outdated transaction is removed from `oneway_todo` due to `TF_UPDATE_TXN`, its `Allocation` is dropped. The current implementation of `Allocation::drop` calls `pending_oneway_finished()`, assuming the transaction was executed. This leads to premature execution of the next queued one-way transaction. Fix this by taking the `oneway_node` from the `Allocation` of the outdated transaction before it is dropped. This prevents `Allocation::drop` from signaling completion. We do not call `take_oneway_node()` from `Transaction::cancel` because it's actually correct to call `pending_oneway_finished()` on cancel if the transaction did not come from `oneway_todo`. This ensures that if `BINDER_THREAD_EXIT` is invoked and cancels a oneway transaction, then the next transaction is taken from `oneway_todo`. This bug does not lead to any issues in the kernel, but may lead to Binder delivering transactions to userspace earlier than userspace expected to receive them. Cc: stable Fixes: eafedbc7c050 ("rust_binder: add Rust Binder driver") Assisted-by: Antigravity:gemini Signed-off-by: Alice Ryhl Acked-by: Carlos Llamas Link: https://patch.msgid.link/20260414-tf-update-txn-fix-v1-1-d2b83303acc9@google.com Signed-off-by: Greg Kroah-Hartman --- drivers/android/binder/allocation.rs | 8 ++++++++ drivers/android/binder/transaction.rs | 11 ++++++++++- 2 files changed, 18 insertions(+), 1 deletion(-) diff --git a/drivers/android/binder/allocation.rs b/drivers/android/binder/allocation.rs index 0cab959e4b7e..b7b05e72970a 100644 --- a/drivers/android/binder/allocation.rs +++ b/drivers/android/binder/allocation.rs @@ -157,6 +157,14 @@ pub(crate) fn set_info_target_node(&mut self, target_node: NodeRef) { self.get_or_init_info().target_node = Some(target_node); } + pub(crate) fn take_oneway_node(&mut self) -> Option> { + if let Some(info) = self.allocation_info.as_mut() { + info.oneway_node.take() + } else { + None + } + } + /// Reserve enough space to push at least `num_fds` fds. pub(crate) fn info_add_fd_reserve(&mut self, num_fds: usize) -> Result { self.get_or_init_info() diff --git a/drivers/android/binder/transaction.rs b/drivers/android/binder/transaction.rs index 47d5e4d88b07..1d9b66920a21 100644 --- a/drivers/android/binder/transaction.rs +++ b/drivers/android/binder/transaction.rs @@ -270,7 +270,8 @@ fn drop_outstanding_txn(&self) { /// Not used for replies. pub(crate) fn submit(self: DLArc, info: &mut TransactionInfo) -> BinderResult { // Defined before `process_inner` so that the destructor runs after releasing the lock. - let mut _t_outdated; + let _t_outdated; + let _oneway_node; let oneway = self.flags & TF_ONE_WAY != 0; let process = self.to.clone(); @@ -287,6 +288,14 @@ pub(crate) fn submit(self: DLArc, info: &mut TransactionInfo) -> BinderRes if let Some(t_outdated) = target_node.take_outdated_transaction(&self, &mut process_inner) { + let mut alloc_guard = t_outdated.allocation.lock(); + if let Some(alloc) = (*alloc_guard).as_mut() { + // Take the oneway node to prevent `Allocation::drop` from calling + // `pending_oneway_finished()`, which would be incorrect as this + // transaction is not being submitted. + _oneway_node = alloc.take_oneway_node(); + } + drop(alloc_guard); // Save the transaction to be dropped after locks are released. _t_outdated = t_outdated; } From f6d8fea9e3953151a4adb4f603503dc3dc9c69da Mon Sep 17 00:00:00 2001 From: Matthew Maurer Date: Fri, 3 Apr 2026 18:18:58 +0000 Subject: [PATCH 161/411] rust_binder: Avoid holding lock when dropping delivered_death In 6c37bebd8c926, we switched to looping over the list and dropping each individual node, ostensibly without the lock held in the loop body. If the kernel were using Rust Edition 2024, the comment would be accurate, and the lock would not be held across the drop. However, the kernel is currently using 2021, so tail expression lifetime extension results in the lock being held across the drop. Explicitly binding the expression result to a variable makes the lockguard no longer part of a tail expression, causing the lock to be dropped before entering the loop body. This was detected via `CONFIG_PROVE_LOCKING` identifying an invalid wait context at the drop site. Reported-by: David Stevens Signed-off-by: Matthew Maurer Cc: stable Fixes: 6c37bebd8c92 ("rust_binder: avoid mem::take on delivered_deaths") Reviewed-by: Alice Ryhl Acked-by: Carlos Llamas Link: https://patch.msgid.link/20260403-lockhold-v1-1-c332b56cd8ae@google.com Signed-off-by: Greg Kroah-Hartman --- drivers/android/binder/process.rs | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/android/binder/process.rs b/drivers/android/binder/process.rs index 820cbd541435..96b8440ceac6 100644 --- a/drivers/android/binder/process.rs +++ b/drivers/android/binder/process.rs @@ -1402,7 +1402,12 @@ fn deferred_release(self: Arc) { // Clear delivered_deaths list. // // Scope ensures that MutexGuard is dropped while executing the body. - while let Some(delivered_death) = { self.inner.lock().delivered_deaths.pop_front() } { + while let Some(delivered_death) = { + // Explicitly bind to avoid tail expression lifetime extension of the lockguard + // Can be removed when the kernel moves to edition 2024 + let maybe_death = self.inner.lock().delivered_deaths.pop_front(); + maybe_death + } { drop(delivered_death); } From f74c8696f14149d5e43cc28b015326a759c48f00 Mon Sep 17 00:00:00 2001 From: Guangshuo Li Date: Tue, 5 May 2026 23:02:56 +0800 Subject: [PATCH 162/411] uio: uio_pci_generic_sva: fix double free of devm_kzalloc() memory uio_pci_sva allocates struct uio_pci_sva_dev with devm_kzalloc() in probe(), but then calls kfree(udev) both on the probe() error path (label out_free) and again in remove(). Because devm_kzalloc() allocations are devres-managed and are freed automatically when the device is detached (including after a failing probe() and during driver unbind), the explicit kfree() can lead to a double free. If probe() fails after devm_kzalloc(), the error path frees udev and devres cleanup will free it again when the core unwinds the partially bound device. On normal driver removal, remove() frees udev and devres will free it again when the device is detached. This issue was identified by a static analysis tool I developed and confirmed by manual review. Fix by removing the manual kfree() calls and dropping the now-unused label. Fixes: 3397c3cd859a2 ("uio: Add SVA support for PCI devices via uio_pci_generic_sva.c") Cc: stable Signed-off-by: Guangshuo Li Link: https://patch.msgid.link/20260505150256.614071-1-lgs201920130244@gmail.com Signed-off-by: Greg Kroah-Hartman --- drivers/uio/uio_pci_generic_sva.c | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/drivers/uio/uio_pci_generic_sva.c b/drivers/uio/uio_pci_generic_sva.c index 4a46acd994a8..d05ef77f7e32 100644 --- a/drivers/uio/uio_pci_generic_sva.c +++ b/drivers/uio/uio_pci_generic_sva.c @@ -129,15 +129,13 @@ static int probe(struct pci_dev *pdev, const struct pci_device_id *id) ret = devm_uio_register_device(&pdev->dev, &udev->info); if (ret) { dev_err(&pdev->dev, "Failed to register uio device\n"); - goto out_free; + goto out_disable; } pci_set_drvdata(pdev, udev); return 0; -out_free: - kfree(udev); out_disable: pci_disable_device(pdev); @@ -146,11 +144,8 @@ static int probe(struct pci_dev *pdev, const struct pci_device_id *id) static void remove(struct pci_dev *pdev) { - struct uio_pci_sva_dev *udev = pci_get_drvdata(pdev); - pci_release_regions(pdev); pci_disable_device(pdev); - kfree(udev); } static ssize_t pasid_show(struct device *dev, From ef15ccbb3e8640a723c42ad90eaf81d66ae02017 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Tue, 5 May 2026 20:45:12 +0200 Subject: [PATCH 163/411] parport: Fix race between port and client registration The parport subsystem registers port devices before they are fully initialised, resulting in a race condition where client drivers such as lp can attach to ports that are not completely initialised or even being torn down. When the port and client drivers are built as modules and loaded around the same time during boot, this occasionally results in a crash. I was able to make this happen reliably in a VM with a PC-style parallel port by patching parport_pc to fail probing: > --- a/drivers/parport/parport_pc.c > +++ b/drivers/parport/parport_pc.c > @@ -2069,7 +2069,7 @@ static struct parport *__parport_pc_probe_port(unsigned long int base, > if (!p) > goto out3; > > - base_res = request_region(base, 3, p->name); > + base_res = NULL; > if (!base_res) > goto out4; > and then running: while true; do modprobe lp & modprobe parport_pc wait rmmod lp parport_pc done for a few seconds. In the long term I think port registration should be changed to put the call to device_add() inside parport_announce_port(), but since the latter currently cannot fail this will require changing all port drivers. For now, add a flag to indicate whether a port has been "announced" and only try to attach client drivers to ports when the flag is set. Fixes: 6fa45a226897 ("parport: add device-model to parport subsystem") Closes: https://bugs.debian.org/1130365 Closes: https://lore.kernel.org/all/6ba903ad-9897-42bb-8c2d-337385cc3746@molgen.mpg.de/ Cc: stable Signed-off-by: Ben Hutchings Acked-by: Sudip Mukherjee Link: https://patch.msgid.link/afo6uBv68GDevbMD@decadent.org.uk Signed-off-by: Greg Kroah-Hartman --- drivers/parport/share.c | 11 +++++++++-- include/linux/parport.h | 1 + 2 files changed, 10 insertions(+), 2 deletions(-) diff --git a/drivers/parport/share.c b/drivers/parport/share.c index ba5292828703..eb0977ca1605 100644 --- a/drivers/parport/share.c +++ b/drivers/parport/share.c @@ -214,10 +214,14 @@ static void get_lowlevel_driver(void) static int port_check(struct device *dev, void *dev_drv) { struct parport_driver *drv = dev_drv; + struct parport *port; /* only send ports, do not send other devices connected to bus */ - if (is_parport(dev)) - drv->match_port(to_parport_dev(dev)); + if (is_parport(dev)) { + port = to_parport_dev(dev); + if (test_bit(PARPORT_ANNOUNCED, &port->devflags)) + drv->match_port(port); + } return 0; } @@ -532,6 +536,7 @@ void parport_announce_port(struct parport *port) if (slave) attach_driver_chain(slave); } + set_bit(PARPORT_ANNOUNCED, &port->devflags); mutex_unlock(®istration_lock); } EXPORT_SYMBOL(parport_announce_port); @@ -561,6 +566,8 @@ void parport_remove_port(struct parport *port) mutex_lock(®istration_lock); + clear_bit(PARPORT_ANNOUNCED, &port->devflags); + /* Spread the word. */ detach_driver_chain(port); diff --git a/include/linux/parport.h b/include/linux/parport.h index 464c2ad28039..f64cb0676e3b 100644 --- a/include/linux/parport.h +++ b/include/linux/parport.h @@ -240,6 +240,7 @@ struct parport { unsigned long devflags; #define PARPORT_DEVPROC_REGISTERED 0 +#define PARPORT_ANNOUNCED 1 struct pardevice *proc_device; /* Currently register proc device */ struct list_head full_list; From 2eae90a457baa0048a96ed38ad93090ee38c8b2f Mon Sep 17 00:00:00 2001 From: Hongling Zeng Date: Mon, 18 May 2026 10:29:39 +0800 Subject: [PATCH 164/411] gpib: cb7210: Fix region leak when request_irq fails When request_irq() fails, the region allocated by request_region() is not released. Fix this by adding an error handling path with proper goto labels to release the region. Fixes: e9dc69956d4d ("staging: gpib: Add Computer Boards GPIB driver") Closes: https://lore.kernel.org/oe-kbuild-all/202605160620.ReBOadPX-lkp@intel.com/ Signed-off-by: Hongling Zeng Cc: stable Link: https://patch.msgid.link/20260518022939.16881-1-zenghongling@kylinos.cn Signed-off-by: Greg Kroah-Hartman --- drivers/gpib/cb7210/cb7210.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/gpib/cb7210/cb7210.c b/drivers/gpib/cb7210/cb7210.c index 6dd8637c5964..673b5bfe2e7d 100644 --- a/drivers/gpib/cb7210/cb7210.c +++ b/drivers/gpib/cb7210/cb7210.c @@ -1049,7 +1049,8 @@ static int cb_isa_attach(struct gpib_board *board, const struct gpib_board_confi if (!request_region(config->ibbase, cb7210_iosize, DRV_NAME)) { dev_err(board->gpib_dev, "ioports starting at 0x%x are already in use\n", config->ibbase); - return -EBUSY; + retval = -EBUSY; + goto err_release_region; } nec_priv->iobase = config->ibbase; cb_priv->fifo_iobase = nec7210_iobase(cb_priv); @@ -1062,11 +1063,16 @@ static int cb_isa_attach(struct gpib_board *board, const struct gpib_board_confi // install interrupt handler if (request_irq(config->ibirq, cb7210_interrupt, isr_flags, DRV_NAME, board)) { dev_err(board->gpib_dev, "failed to obtain IRQ %d\n", config->ibirq); - return -EBUSY; + retval = -EBUSY; + goto err_release_region; } cb_priv->irq = config->ibirq; return cb7210_init(cb_priv, board); + +err_release_region: + release_region(nec7210_iobase(cb_priv), cb7210_iosize); + return retval; } static void cb_isa_detach(struct gpib_board *board) From 36770417153644bc88281c7284730ef1d14d8d3c Mon Sep 17 00:00:00 2001 From: Xiaolei Wang Date: Mon, 18 May 2026 15:34:05 +0800 Subject: [PATCH 165/411] misc: rp1: Send IACK on IRQ activate to fix kdump/kexec MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After a kexec/kdump reboot, the macb Ethernet controller fails to receive any packets, causing DHCP to hang indefinitely and the network interface to be unusable despite link being up. The root cause is that RP1's level-triggered MSI-X interrupt sources (such as macb on hwirq 6) may have their internal state machines stuck in the "waiting for IACK" state. This happens because the previous kernel crashed before sending the acknowledgment for a pending level interrupt. In this stuck state, RP1 will not generate new MSI-X writes even though the interrupt source remains asserted. Since no new MSI-X is sent, the GIC never sees a new edge, the chained IRQ handler is never invoked, and the interrupt is permanently lost. Fix this by sending MSIX_CFG_IACK in rp1_irq_activate(). This unconditionally resets the MSI-X state machine back to idle when a child device requests its interrupt. If the interrupt source is still asserted, RP1 will immediately issue a new MSI-X with the freshly configured msg_addr/msg_data, and normal interrupt delivery resumes. Writing IACK when the state machine is already idle (i.e., on a normal cold boot) is harmless — it has no effect. Fixes: 49d63971f963 ("misc: rp1: RaspberryPi RP1 misc driver") Cc: stable Signed-off-by: Xiaolei Wang Link: https://patch.msgid.link/20260518073405.2115003-1-xiaolei.wang@windriver.com Signed-off-by: Greg Kroah-Hartman --- drivers/misc/rp1/rp1_pci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/misc/rp1/rp1_pci.c b/drivers/misc/rp1/rp1_pci.c index d210da84c30a..81685e3f3296 100644 --- a/drivers/misc/rp1/rp1_pci.c +++ b/drivers/misc/rp1/rp1_pci.c @@ -143,6 +143,7 @@ static int rp1_irq_activate(struct irq_domain *d, struct irq_data *irqd, struct rp1_dev *rp1 = d->host_data; msix_cfg_set(rp1, (unsigned int)irqd->hwirq, MSIX_CFG_ENABLE); + msix_cfg_set(rp1, (unsigned int)irqd->hwirq, MSIX_CFG_IACK); return 0; } From bed6e04be8e6b9133d8b16d5a42d0e0ce674fa9a Mon Sep 17 00:00:00 2001 From: Hamza Mahfooz Date: Mon, 11 May 2026 10:43:14 -0400 Subject: [PATCH 166/411] netfilter: conntrack: tcp: do not force CLOSE on invalid-seq RST without direction check An unintended behavior in the TCP conntrack state machine allows a connection to be forced into the CLOSE state using an RST packet with an invalid sequence number. Specifically, after a SYN packet is observed, an RST with an invalid SEQ can transition the conntrack entry to TCP_CONNTRACK_CLOSE, regardless of whether the RST corresponds to the expected reply direction. The relevant code path assumes the RST is a response to an outgoing SYN, but does not validate packet direction or ensure that a matching SYN was actually sent in the opposite direction. As a result, a crafted packet sequence consisting of a SYN followed by an invalid-sequence RST can prematurely terminate an active NAT entry. This makes connection teardown easier than intended. So, tighten the state transition logic to ensure that RST-triggered CLOSE transitions only occur when the RST is a valid response to a previously observed SYN in the correct direction. Cc: stable@vger.kernel.org Fixes: 9fb9cbb1082d ("[NETFILTER]: Add nf_conntrack subsystem.") Signed-off-by: Hamza Mahfooz Signed-off-by: Florian Westphal --- net/netfilter/nf_conntrack_proto_tcp.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index b67426c2189b..e99ab1e88e9f 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1221,7 +1221,8 @@ int nf_conntrack_tcp_packet(struct nf_conn *ct, new_state = old_state; } if (((test_bit(IPS_SEEN_REPLY_BIT, &ct->status) - && ct->proto.tcp.last_index == TCP_SYN_SET) + && ct->proto.tcp.last_index == TCP_SYN_SET + && ct->proto.tcp.last_dir != dir) || (!test_bit(IPS_ASSURED_BIT, &ct->status) && ct->proto.tcp.last_index == TCP_ACK_SET)) && ntohl(th->ack_seq) == ct->proto.tcp.last_end) { From 92170e6afe927ab2792a3f71902845789c8e31b1 Mon Sep 17 00:00:00 2001 From: Chris Mason Date: Tue, 19 May 2026 12:36:14 -0700 Subject: [PATCH 167/411] netfilter: synproxy: refresh tcphdr after skb_ensure_writable synproxy_tstamp_adjust() rewrites the TCP timestamp option in place and then patches the TCP checksum via inet_proto_csum_replace4() on the caller-supplied tcphdr pointer. Both ipv4_synproxy_hook() and ipv6_synproxy_hook() obtain that pointer with skb_header_pointer() before calling in, so it may either alias skb->head directly or point at the caller's on-stack _tcph buffer. Between obtaining the pointer and using it, the function calls skb_ensure_writable(skb, optend), which on a cloned or non-linear skb invokes pskb_expand_head() and frees the old skb->head. After that point the cached th is stale: caller (ipv[46]_synproxy_hook) th = skb_header_pointer(skb, ..., &_tcph) synproxy_tstamp_adjust(skb, protoff, th, ...) skb_ensure_writable(skb, optend) pskb_expand_head() /* kfree(old skb->head) */ ... inet_proto_csum_replace4(&th->check, ...) /* writes into freed head, or into the caller's stack copy leaving the on-wire checksum stale */ The option bytes are written through skb->data and are fine; only the checksum update goes through th and so lands in the wrong place. The result is either a write into freed slab memory or a packet leaving with a checksum that does not match its payload. Fix by re-deriving th from skb->data + protoff immediately after skb_ensure_writable() succeeds, so the subsequent checksum update targets the linear, writable header. Fixes: 48b1de4c110a ("netfilter: add SYNPROXY core/target") Assisted-by: kres (claude-opus-4-7) Signed-off-by: Chris Mason Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- net/netfilter/nf_synproxy_core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c index 57f57e2fc80a..036c8586f49b 100644 --- a/net/netfilter/nf_synproxy_core.c +++ b/net/netfilter/nf_synproxy_core.c @@ -200,6 +200,8 @@ synproxy_tstamp_adjust(struct sk_buff *skb, unsigned int protoff, if (skb_ensure_writable(skb, optend)) return 0; + th = (struct tcphdr *)(skb->data + protoff); + while (optoff < optend) { unsigned char *op = skb->data + optoff; From 47980b6dbf83961eec1c1363ea986e9c06ff8054 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Thu, 14 May 2026 14:21:57 +0200 Subject: [PATCH 168/411] netfilter: nf_conntrack_gre: fix gre keymap list corruption Quoting reporter: A race between GRE keymap insertion and destruction can corrupt the kernel list or use a freed object. `nf_ct_gre_keymap_add()` publishes a new keymap pointer before the embedded `list_head` is linked, while `nf_ct_gre_keymap_destroy()` can concurrently delete and free that same object. An unprivileged user can reach this through the PPTP conntrack helper by racing PPTP control messages or helper teardown, leading to KASAN-detectable list corruption/UAF in kernel context. ## Root Cause Analysis `exp_gre()` installs GRE expectations for a PPTP control flow and then adds two GRE keymap entries [..] The add path publishes `ct_pptp_info->keymap[dir]` before linking the embedded list node [..] Concurrent teardown deletes that partially initialized object. Make add/destroy symmetric: install both, destroy both while under lock. Furthermore, we should refuse to publish a new mapping in case ct is going away, else we may leak the allocation. The "retrans" detection is strange: existing mapping is checked for key equality with the new mapping, then for "is on the list" via list walk. But I can't see how an existing keymap entry can be NOT on list. Change this to only check if we're asked to map same tuple again -- if so, skip re-install, else signal failure. Last, add a bug trap for the keymap list; it has to be empty when namespace is going away. Reported-by: Leo Lin Signed-off-by: Florian Westphal --- .../linux/netfilter/nf_conntrack_proto_gre.h | 7 +- net/netfilter/nf_conntrack_core.c | 8 ++ net/netfilter/nf_conntrack_pptp.c | 8 +- net/netfilter/nf_conntrack_proto_gre.c | 106 +++++++++++++----- 4 files changed, 95 insertions(+), 34 deletions(-) diff --git a/include/linux/netfilter/nf_conntrack_proto_gre.h b/include/linux/netfilter/nf_conntrack_proto_gre.h index 9ee7014400e8..ad5563f0f864 100644 --- a/include/linux/netfilter/nf_conntrack_proto_gre.h +++ b/include/linux/netfilter/nf_conntrack_proto_gre.h @@ -18,9 +18,10 @@ struct nf_ct_gre_keymap { struct rcu_head rcu; }; -/* add new tuple->key_reply pair to keymap */ -int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, - struct nf_conntrack_tuple *t); +/* add tuple->key_reply pairs to keymap */ +bool nf_ct_gre_keymap_add(struct nf_conn *ct, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl); /* delete keymap entries */ void nf_ct_gre_keymap_destroy(struct nf_conn *ct); diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c index 8ba5b22a1eef..b521b5ebd664 100644 --- a/net/netfilter/nf_conntrack_core.c +++ b/net/netfilter/nf_conntrack_core.c @@ -568,6 +568,13 @@ static void destroy_gre_conntrack(struct nf_conn *ct) #endif } +static void warn_on_keymap_list_leak(const struct net *net) +{ +#ifdef CONFIG_NF_CT_PROTO_GRE + WARN_ON_ONCE(!list_empty(&net->ct.nf_ct_proto.gre.keymap_list)); +#endif +} + void nf_ct_destroy(struct nf_conntrack *nfct) { struct nf_conn *ct = (struct nf_conn *)nfct; @@ -2510,6 +2517,7 @@ void nf_conntrack_cleanup_net_list(struct list_head *net_exit_list) } list_for_each_entry(net, net_exit_list, exit_list) { + warn_on_keymap_list_leak(net); nf_conntrack_ecache_pernet_fini(net); nf_conntrack_expect_pernet_fini(net); free_percpu(net->ct.stat); diff --git a/net/netfilter/nf_conntrack_pptp.c b/net/netfilter/nf_conntrack_pptp.c index 4c679638df06..dc23e4181618 100644 --- a/net/netfilter/nf_conntrack_pptp.c +++ b/net/netfilter/nf_conntrack_pptp.c @@ -225,13 +225,9 @@ static int exp_gre(struct nf_conn *ct, __be16 callid, __be16 peer_callid) if (nf_ct_expect_related(exp_reply, 0) != 0) goto out_unexpect_orig; - /* Add GRE keymap entries */ - if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_ORIGINAL, &exp_orig->tuple) != 0) + if (!nf_ct_gre_keymap_add(ct, &exp_orig->tuple, + &exp_reply->tuple)) goto out_unexpect_both; - if (nf_ct_gre_keymap_add(ct, IP_CT_DIR_REPLY, &exp_reply->tuple) != 0) { - nf_ct_gre_keymap_destroy(ct); - goto out_unexpect_both; - } ret = 0; out_put_both: diff --git a/net/netfilter/nf_conntrack_proto_gre.c b/net/netfilter/nf_conntrack_proto_gre.c index 94c19bc4edc5..35e22082d65a 100644 --- a/net/netfilter/nf_conntrack_proto_gre.c +++ b/net/netfilter/nf_conntrack_proto_gre.c @@ -87,41 +87,97 @@ static __be16 gre_keymap_lookup(struct net *net, struct nf_conntrack_tuple *t) return key; } -/* add a single keymap entry, associate with specified master ct */ -int nf_ct_gre_keymap_add(struct nf_conn *ct, enum ip_conntrack_dir dir, - struct nf_conntrack_tuple *t) +enum nf_ct_gre_km_act { + NF_CT_GRE_KM_NEW, + NF_CT_GRE_KM_BAD, + NF_CT_GRE_KM_DUP +}; + +static enum nf_ct_gre_km_act +nf_ct_gre_km_acceptable(const struct nf_ct_pptp_master *ct_pptp_info, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl) +{ + struct nf_ct_gre_keymap *km_orig, *km_repl; + + lockdep_assert_held(&keymap_lock); + + km_orig = ct_pptp_info->keymap[IP_CT_DIR_ORIGINAL]; + km_repl = ct_pptp_info->keymap[IP_CT_DIR_REPLY]; + + if (km_orig && km_repl) { + if (!gre_key_cmpfn(km_orig, orig)) + return NF_CT_GRE_KM_BAD; + + if (!gre_key_cmpfn(km_repl, repl)) + return NF_CT_GRE_KM_BAD; + + return NF_CT_GRE_KM_DUP; + } + + DEBUG_NET_WARN_ON_ONCE(km_orig); + DEBUG_NET_WARN_ON_ONCE(km_repl); + return NF_CT_GRE_KM_NEW; +} + +/* add keymap entries, associate with specified master ct */ +bool nf_ct_gre_keymap_add(struct nf_conn *ct, + const struct nf_conntrack_tuple *orig, + const struct nf_conntrack_tuple *repl) { struct net *net = nf_ct_net(ct); struct nf_gre_net *net_gre = gre_pernet(net); struct nf_ct_pptp_master *ct_pptp_info = nfct_help_data(ct); - struct nf_ct_gre_keymap **kmp, *km; + struct nf_ct_gre_keymap *km_orig, *km_repl; + bool ret = false; - kmp = &ct_pptp_info->keymap[dir]; - if (*kmp) { - /* check whether it's a retransmission */ - list_for_each_entry_rcu(km, &net_gre->keymap_list, list) { - if (gre_key_cmpfn(km, t) && km == *kmp) - return 0; - } - pr_debug("trying to override keymap_%s for ct %p\n", - dir == IP_CT_DIR_REPLY ? "reply" : "orig", ct); - return -EEXIST; - } + km_orig = kmalloc_obj(*km_orig, GFP_ATOMIC); + if (!km_orig) + return false; + km_repl = kmalloc_obj(*km_repl, GFP_ATOMIC); + if (!km_repl) + goto km_free; - km = kmalloc_obj(*km, GFP_ATOMIC); - if (!km) - return -ENOMEM; - memcpy(&km->tuple, t, sizeof(*t)); - *kmp = km; - - pr_debug("adding new entry %p: ", km); - nf_ct_dump_tuple(&km->tuple); + memcpy(&km_orig->tuple, orig, sizeof(*orig)); + memcpy(&km_repl->tuple, repl, sizeof(*repl)); spin_lock_bh(&keymap_lock); - list_add_tail(&km->list, &net_gre->keymap_list); + if (nf_ct_is_dying(ct)) + goto unlock_free; + + switch (nf_ct_gre_km_acceptable(ct_pptp_info, orig, repl)) { + case NF_CT_GRE_KM_NEW: + break; + case NF_CT_GRE_KM_DUP: + ret = true; + goto unlock_free; + case NF_CT_GRE_KM_BAD: + pr_debug("trying to override keymap for ct %p\n", ct); + goto unlock_free; + } + + if (ct_pptp_info->keymap[IP_CT_DIR_ORIGINAL] || + ct_pptp_info->keymap[IP_CT_DIR_REPLY]) + goto unlock_free; + + pr_debug("adding new entries %p,%p: ", km_orig, km_repl); + nf_ct_dump_tuple(&km_orig->tuple); + nf_ct_dump_tuple(&km_repl->tuple); + + list_add_tail_rcu(&km_orig->list, &net_gre->keymap_list); + list_add_tail_rcu(&km_repl->list, &net_gre->keymap_list); + ct_pptp_info->keymap[IP_CT_DIR_ORIGINAL] = km_orig; + ct_pptp_info->keymap[IP_CT_DIR_REPLY] = km_repl; spin_unlock_bh(&keymap_lock); - return 0; + return true; + +unlock_free: + spin_unlock_bh(&keymap_lock); +km_free: + kfree(km_orig); + kfree(km_repl); + return ret; } EXPORT_SYMBOL_GPL(nf_ct_gre_keymap_add); From c376f07e16c02239ed44cabb97145d03f65b4d15 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 19 May 2026 20:10:08 +0200 Subject: [PATCH 169/411] netfilter: xt_cpu: prefer raw_smp_processor_id With PREEMPT_RCU we get splat: BUG: using smp_processor_id() in preemptible [..] caller is cpu_mt+0x53/0xd0 net/netfilter/xt_cpu.c:37 CPU: 1 .. Comm: syz.3.1377 #0 PREEMPT(full) Call Trace: dump_stack_lvl+0xe8/0x150 lib/dump_stack.c:120 check_preemption_disabled+0xd3/0xe0 lib/smp_processor_id.c:47 cpu_mt+0x53/0xd0 net/netfilter/xt_cpu.c:37 [..] Just use raw version instead. This is similar to 14d14a5d2957 ("netfilter: nft_meta: use raw_smp_processor_id()"). Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables") Reported-by: syzbot+690d3e3ffa7335ac10eb@syzkaller.appspotmail.com Signed-off-by: Florian Westphal --- net/netfilter/xt_cpu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/xt_cpu.c b/net/netfilter/xt_cpu.c index 3bdc302a0f91..9cb259902a58 100644 --- a/net/netfilter/xt_cpu.c +++ b/net/netfilter/xt_cpu.c @@ -34,7 +34,7 @@ static bool cpu_mt(const struct sk_buff *skb, struct xt_action_param *par) { const struct xt_cpu_info *info = par->matchinfo; - return (info->cpu == smp_processor_id()) ^ info->invert; + return (info->cpu == raw_smp_processor_id()) ^ info->invert; } static struct xt_match cpu_mt_reg __read_mostly = { From 968cc2c96390f06e56ed6a43f935bfebdefed28f Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Sat, 16 May 2026 23:23:21 +0800 Subject: [PATCH 170/411] netfilter: disable payload mangling in userns Several parts of network stack rely on iph->ihl validation done by network stack before PRE_ROUTING. Disable this feature for user namespaces for now. tcp option handling is likely safe even for LOCAL_IN, so this this leaves tcp option mangling via nft_exthdr.c as-is. I don't think these are the only means to alter packets, but these appear to be relatively prominent. This could be relaxed later. Example: - allow userns for ingress hook. - allow userns if base is transport header. Also, we should revalidate or restrict generally: - Don't allow linklayer writes to spill into network header - restrict ipv4 and ipv6 to 'known safe' writes, e.g. saddr/daddr/check/tos Reported-by: Qi Tang Reported-by: Tong Liu Tested-by: Qi Tang Link: https://lore.kernel.org/netfilter-devel/20260515100411.3141-1-fw@strlen.de/ Signed-off-by: Florian Westphal --- net/netfilter/nfnetlink_queue.c | 6 ++++-- net/netfilter/nft_payload.c | 3 +++ 2 files changed, 7 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nfnetlink_queue.c b/net/netfilter/nfnetlink_queue.c index 984a0eb9e149..60ab88d45096 100644 --- a/net/netfilter/nfnetlink_queue.c +++ b/net/netfilter/nfnetlink_queue.c @@ -1141,6 +1141,9 @@ nfqnl_mangle(void *data, unsigned int data_len, struct nf_queue_entry *e, int di { struct sk_buff *nskb; + if (e->state.net->user_ns != &init_user_ns) + return -EPERM; + if (diff < 0) { unsigned int min_len = skb_transport_offset(e->skb); @@ -1537,8 +1540,7 @@ static int nfqnl_recv_verdict(struct sk_buff *skb, const struct nfnl_info *info, if (nfqnl_mangle(nla_data(nfqa[NFQA_PAYLOAD]), payload_len, entry, diff) < 0) verdict = NF_DROP; - - if (ct && diff) + else if (ct && diff) nfnl_ct->seq_adjust(entry->skb, ct, ctinfo, diff); } diff --git a/net/netfilter/nft_payload.c b/net/netfilter/nft_payload.c index 01e13e5255a9..484a5490832e 100644 --- a/net/netfilter/nft_payload.c +++ b/net/netfilter/nft_payload.c @@ -917,6 +917,9 @@ static int nft_payload_set_init(const struct nft_ctx *ctx, struct nft_payload_set *priv = nft_expr_priv(expr); int err; + if (ctx->net->user_ns != &init_user_ns) + return -EPERM; + priv->base = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_BASE])); priv->len = ntohl(nla_get_be32(tb[NFTA_PAYLOAD_LEN])); From f438d1786d657d57790c5d138d6db3fc9fdac392 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 19 May 2026 22:52:07 +0200 Subject: [PATCH 171/411] netfilter: ebtables: fix OOB read in compat_mtw_from_user Luxiao Xu says: The function compat_mtw_from_user() converts ebtables extensions from 32-bit user structures to kernel native structures. However, it lacks proper validation of the user-supplied match_size/target_size. When certain extensions are processed, the kernel-side translation logic may perform memory accesses based on the extension's expected size. If the user provides a size smaller than what the extension requires, it results in an out-of-bounds read as reported by KASAN. This fix introduces a check to ensure match_size is at least as large as the extension's required compatsize. This covers matches, watchers, and targets, while maintaining compatibility with standard targets. AFAIU this is relevant for matches that need to go though match->compat_from_user() call. Those that use plain memcpy with the user-provided size are ok because the caller checks that size vs the start of the next rule entry offset (which itself is checked vs. total size copied from userspace). The ->compat_from_user() callbacks assume they can read compatsize bytes, so they need this extra check. Based on an earlier patch from Luxiao Xu. Fixes: 81e675c227ec ("netfilter: ebtables: add CONFIG_COMPAT support") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Luxiao Xu Signed-off-by: Ren Wei Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- net/bridge/netfilter/ebtables.c | 30 ++++++++++++++++++++++++++++++ 1 file changed, 30 insertions(+) diff --git a/net/bridge/netfilter/ebtables.c b/net/bridge/netfilter/ebtables.c index b9f4daac09af..8a6a069329d2 100644 --- a/net/bridge/netfilter/ebtables.c +++ b/net/bridge/netfilter/ebtables.c @@ -1956,6 +1956,25 @@ enum compat_mwt { EBT_COMPAT_TARGET, }; +static bool match_size_ok(const struct xt_match *match, unsigned int match_size) +{ + u16 csize; + + if (match->matchsize == -1) /* cannot validate ebt_among */ + return true; + + csize = match->compatsize ? : match->matchsize; + + return match_size >= csize; +} + +static bool tgt_size_ok(const struct xt_target *tgt, unsigned int tgt_size) +{ + u16 csize = tgt->compatsize ? : tgt->targetsize; + + return tgt_size >= csize; +} + static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, enum compat_mwt compat_mwt, struct ebt_entries_buf_state *state, @@ -1981,6 +2000,11 @@ static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, if (IS_ERR(match)) return PTR_ERR(match); + if (!match_size_ok(match, match_size)) { + module_put(match->me); + return -EINVAL; + } + off = ebt_compat_match_offset(match, match_size); if (dst) { if (match->compat_from_user) @@ -2000,6 +2024,12 @@ static int compat_mtw_from_user(const struct compat_ebt_entry_mwt *mwt, mwt->u.revision); if (IS_ERR(wt)) return PTR_ERR(wt); + + if (!tgt_size_ok(wt, match_size)) { + module_put(wt->me); + return -EINVAL; + } + off = xt_compat_target_offset(wt); if (dst) { From 1d001b0a6182b0d2f41a8d687f7522b6f1e94280 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 20 May 2026 10:34:09 +0800 Subject: [PATCH 172/411] netfilter: nft_fib_ipv6: walk fib6_siblings under RCU nft_fib6_info_nh_uses_dev() runs from nft_fib6_eval() in softirq under rcu_read_lock(). fib6_siblings is modified by writers that hold tb6_lock but do not wait for RCU readers, so the sibling walk should use list_for_each_entry_rcu(): it adds READ_ONCE() on the ->next pointer and lets CONFIG_PROVE_RCU_LIST validate the locking. No functional change for non-debug builds. Fixes: 1c32b24c234b ("netfilter: nft_fib_ipv6: switch to fib6_lookup") Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal --- net/ipv6/netfilter/nft_fib_ipv6.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index 8b2dba88ee96..5e192a446ec8 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -170,7 +170,7 @@ static bool nft_fib6_info_nh_uses_dev(struct fib6_info *rt, if (nft_fib6_info_nh_dev_match(nh_dev, dev)) return true; - list_for_each_entry(iter, &rt->fib6_siblings, fib6_siblings) { + list_for_each_entry_rcu(iter, &rt->fib6_siblings, fib6_siblings) { nh_dev = fib6_info_nh_dev(iter); if (nft_fib6_info_nh_dev_match(nh_dev, dev)) From f81b0c2d281faa93e4c2b7247047922aaf3e4ba6 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 20 May 2026 10:34:10 +0800 Subject: [PATCH 173/411] netfilter: nft_fib_ipv6: handle routes via external nexthop fib6_info has a union: union { struct list_head fib6_siblings; struct list_head nh_list; }; Old-style multipath (ip -6 route add ... nexthop ... nexthop ...) uses fib6_siblings. External nexthop (ip -6 route add ... nhid N) uses nh_list, linked into &nh->f6i_list. nft_fib6_info_nh_uses_dev() blindly walks &rt->fib6_siblings, causing an OOB read past the struct nexthop slab when rt->nh is set: ================================================================== BUG: KASAN: slab-out-of-bounds in nft_fib6_eval+0x1362/0x16c0 Read of size 8 at addr ffff888103a099d0 by task ping/386 CPU: 2 UID: 0 PID: 386 Comm: ping Not tainted 7.1.0-rc3+ #251 PREEMPT Call Trace: dump_stack_lvl+0x76/0xa0 print_report+0xd1/0x5f0 kasan_report+0xe7/0x130 __asan_report_load8_noabort+0x14/0x30 nft_fib6_eval+0x1362/0x16c0 nft_do_chain+0x279/0x18c0 nft_do_chain_ipv6+0x1a8/0x230 nf_hook_slow+0xad/0x200 ipv6_rcv+0x152/0x380 __netif_receive_skb_one_core+0x118/0x1c0 ================================================================== Branch by route shape: when rt->nh is set, walk via nexthop_for_each_fib6_nh() (also covers nh groups, which the original code missed); otherwise walk fib6_siblings, guarded by READ_ONCE() of rt->fib6_nsiblings as required by commit 31d7d67ba127 ("ipv6: annotate data-races around rt->fib6_nsiblings"). Fixes: 1c32b24c234b ("netfilter: nft_fib_ipv6: switch to fib6_lookup") Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal --- net/ipv6/netfilter/nft_fib_ipv6.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index 5e192a446ec8..c0a0075e2590 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -160,16 +160,32 @@ static bool nft_fib6_info_nh_dev_match(const struct net_device *nh_dev, l3mdev_master_ifindex_rcu(nh_dev) == dev->ifindex; } +static int nft_fib6_nh_match_dev_cb(struct fib6_nh *nh, void *arg) +{ + const struct net_device *dev = arg; + + return nft_fib6_info_nh_dev_match(nh->fib_nh_dev, dev); +} + static bool nft_fib6_info_nh_uses_dev(struct fib6_info *rt, const struct net_device *dev) { const struct net_device *nh_dev; struct fib6_info *iter; + /* External nexthop: fib6_siblings slot aliases nh_list, walk via nh. */ + if (rt->nh) + return nexthop_for_each_fib6_nh(rt->nh, + nft_fib6_nh_match_dev_cb, + (void *)dev); + nh_dev = fib6_info_nh_dev(rt); if (nft_fib6_info_nh_dev_match(nh_dev, dev)) return true; + if (!READ_ONCE(rt->fib6_nsiblings)) + return false; + list_for_each_entry_rcu(iter, &rt->fib6_siblings, fib6_siblings) { nh_dev = fib6_info_nh_dev(iter); From a40aaaef2f8f5a17a779eeac7032f2f7d5322406 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 20 May 2026 10:34:11 +0800 Subject: [PATCH 174/411] selftests: netfilter: add nft_fib_nexthop test Functional coverage of nft_fib6_eval()'s nexthop enumeration over three route shapes: 1) single external nexthop (nhid) 2) external nexthop group (nhid -> group) 3) old-style multipath (nexthop ... nexthop ...) Each scenario places one nexthop on the input device (veth0). For (2) and (3) the matching nexthop is the second member, so the walk has to traverse beyond the primary nh. Two nft counters on prerouting verify the data path: one increments only when fib reports veth0 as the oif, the other counts "missing" results and must stay at zero. ./nft_fib_nexthop.sh PASS: single external nexthop (nhid -> veth0) PASS: nexthop group (dummy0 + veth0) PASS: old-style multipath (sibling on veth0) Suggested-by: Florian Westphal Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal --- .../testing/selftests/net/netfilter/Makefile | 1 + .../net/netfilter/nft_fib_nexthop.sh | 152 ++++++++++++++++++ 2 files changed, 153 insertions(+) create mode 100755 tools/testing/selftests/net/netfilter/nft_fib_nexthop.sh diff --git a/tools/testing/selftests/net/netfilter/Makefile b/tools/testing/selftests/net/netfilter/Makefile index ee2d1a5254f8..d953ee218c0f 100644 --- a/tools/testing/selftests/net/netfilter/Makefile +++ b/tools/testing/selftests/net/netfilter/Makefile @@ -26,6 +26,7 @@ TEST_PROGS := \ nft_concat_range.sh \ nft_conntrack_helper.sh \ nft_fib.sh \ + nft_fib_nexthop.sh \ nft_flowtable.sh \ nft_interface_stress.sh \ nft_meta.sh \ diff --git a/tools/testing/selftests/net/netfilter/nft_fib_nexthop.sh b/tools/testing/selftests/net/netfilter/nft_fib_nexthop.sh new file mode 100755 index 000000000000..c4f203057382 --- /dev/null +++ b/tools/testing/selftests/net/netfilter/nft_fib_nexthop.sh @@ -0,0 +1,152 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# shellcheck disable=SC2154 +# +# Exercise nft_fib6_eval()'s sibling/nh enumeration on three route shapes: +# 1) route via a single external nexthop (nhid) +# 2) route via an external nexthop group (nhid -> group, two members) +# 3) route via old-style multipath (nexthop ... nexthop ...) +# +# In each scenario the route's nexthop set contains veth0 (the iif of the +# test packet). nft_fib6_info_nh_uses_dev() must walk the set and report +# veth0 as a valid oif. For (2) and (3) the matching nexthop is the second +# member, so the walk has to traverse beyond the primary nh. +# +# After sending $PKTS ICMPv6 echo requests from ns1, check two counters on +# nsrouter: +# nf_ok -- `fib daddr . iif oif eq "veth0"` must equal $PKTS +# nf_bad -- `fib daddr . iif oif missing` must stay at 0 +# Both rules also match on iif veth0 and ip6 daddr dead:dead::/64 so that +# kernel-generated ND/MLD/RA traffic cannot pollute the counters. +# +# Topology similar to nft_fib.sh, without ns2; two dummy interfaces on +# nsrouter host extra nh devices: +# +# dead:1::99 dead:1::1 +# ns1 <----veth----> nsrouter --- dummy0 dead:2::1 +# \-- dummy1 dead:9::1 + +source lib.sh + +ret=0 +PKTS=3 + +checktool "nft --version" "run test without nft" +checktool "ip -V" "run test without iproute2" + +setup_ns nsrouter ns1 +trap cleanup_all_ns EXIT + +if ! ip link add veth0 netns "$nsrouter" type veth peer name eth0 netns "$ns1" \ + > /dev/null 2>&1; then + echo "SKIP: No virtual ethernet pair device support in kernel" + exit $ksft_skip +fi + +ip -net "$ns1" link set lo up +ip -net "$ns1" link set eth0 up +ip -net "$ns1" -6 addr add dead:1::99/64 dev eth0 nodad +ip -net "$ns1" -6 route add default via dead:1::1 + +ip -net "$nsrouter" link set lo up +ip -net "$nsrouter" link set veth0 up +ip -net "$nsrouter" -6 addr add dead:1::1/64 dev veth0 nodad + +if ! ip -net "$nsrouter" link add dummy0 type dummy 2>/dev/null; then + echo "SKIP: dummy netdev not available" + exit $ksft_skip +fi +ip -net "$nsrouter" link set dummy0 up +ip -net "$nsrouter" -6 addr add dead:2::1/64 dev dummy0 nodad + +ip -net "$nsrouter" link add dummy1 type dummy +ip -net "$nsrouter" link set dummy1 up +ip -net "$nsrouter" -6 addr add dead:9::1/64 dev dummy1 nodad + +ip netns exec "$nsrouter" sysctl -q net.ipv6.conf.all.forwarding=1 + +load_fib_rule() { + # filter on iif + daddr so the counters only see our test packets + ip netns exec "$nsrouter" nft -f /dev/stdin <&2 + ip netns exec "$nsrouter" nft list counter ip6 t "$counter" 1>&2 +} + +run_scenario() { + local what="$1"; shift + # counter output format is "packets PACKET_NUM bytes BYTES_NUM"; + # we only care about the packet count + local expect_ok="packets $PKTS bytes" + local expect_bad="packets 0 bytes" + local lret=0 + + # reset route + nexthop state between scenarios + ip -net "$nsrouter" -6 route del dead:dead::/64 > /dev/null 2>&1 || true + ip -net "$nsrouter" nexthop flush > /dev/null 2>&1 || true + + # run the scenario function passed by the caller + "$@" || echo "WARN ($what): scenario setup returned non-zero" + + load_fib_rule || { echo "FAIL ($what): nft load"; ret=1; return; } + + # ping a daddr inside dead:dead::/64 so fib has to walk the nh set + ip netns exec "$ns1" ping -6 -c "$PKTS" -i 0.1 -W 1 dead:dead::1 \ + > /dev/null 2>&1 || true + + # verify the packets went through the expected fib path + if ! ip netns exec "$nsrouter" nft list counter ip6 t nf_ok | grep -q "$expect_ok"; then + bad_counter nf_ok "$expect_ok" "$what" + lret=1 + fi + if ! ip netns exec "$nsrouter" nft list counter ip6 t nf_bad | grep -q "$expect_bad"; then + bad_counter nf_bad "$expect_bad" "$what" + lret=1 + fi + + if [ $lret -eq 0 ]; then + echo "PASS: $what" + else + ret=1 + fi +} + +scenario_single_nh() { + ip -net "$nsrouter" nexthop add id 1 via dead:1::99 dev veth0 + ip -net "$nsrouter" -6 route add dead:dead::/64 nhid 1 +} +run_scenario "single external nexthop (nhid -> veth0)" scenario_single_nh + +scenario_nh_group() { + ip -net "$nsrouter" nexthop add id 1 via dead:2::2 dev dummy0 + ip -net "$nsrouter" nexthop add id 2 via dead:1::99 dev veth0 + ip -net "$nsrouter" nexthop add id 100 group 1/2 + ip -net "$nsrouter" -6 route add dead:dead::/64 nhid 100 +} +run_scenario "nexthop group (dummy0 + veth0)" scenario_nh_group + +scenario_old_multipath() { + ip -net "$nsrouter" -6 route add dead:dead::/64 \ + nexthop via dead:2::2 dev dummy0 \ + nexthop via dead:1::99 dev veth0 +} +run_scenario "old-style multipath (sibling on veth0)" scenario_old_multipath + +exit $ret From 18014147d3ee7831dce53fe65d7fc8d428b02552 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Mon, 11 May 2026 16:37:56 +0200 Subject: [PATCH 175/411] netfilter: nf_tables: fix dst corruption in same register operation For lshift and rshift, the shift operations are performed in a loop over 32-bit words. The loop calculates the shifted value and write it to dst, and then immediately reads from src to calculate the carry for the next iteration. Because src and dst could point to the same memory location, the carry is incorrectly calculated using the newly modified dst value instead of the original src value. Adding a temporary local variable to cache the original value before writing to dst and using it for the carry calculation solves the problem. In addition, partial overlap is rejected from control plane for all kind of operations including byteorder. This was tested with the following bytecode: table test_table ip flags 0 use 1 handle 1 ip test_table test_chain use 3 type filter hook input prio 0 policy accept packets 0 bytes 0 flags 1 ip test_table test_chain 2 [ immediate reg 1 0x44332211 0x88776655 ] [ bitwise reg 1 = ( reg 1 << 0x08000000 ) ] [ cmp eq reg 1 0x66443322 0x00887766 ] [ counter pkts 0 bytes 0 ] ip test_table test_chain 4 3 [ immediate reg 1 0x44332211 0x88776655 ] [ bitwise reg 1 = ( reg 1 << 0x08000000 ) ] [ cmp eq reg 1 0x55443322 0x00887766 ] [ counter pkts 21794 bytes 1917798 ] Fixes: 567d746b55bc ("netfilter: bitwise: add support for shifts.") Acked-by: Jeremy Sowden Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal --- include/net/netfilter/nf_tables.h | 7 +++++++ net/netfilter/nft_bitwise.c | 18 ++++++++++++++---- net/netfilter/nft_byteorder.c | 13 ++++++++++--- 3 files changed, 31 insertions(+), 7 deletions(-) diff --git a/include/net/netfilter/nf_tables.h b/include/net/netfilter/nf_tables.h index cff7b773e972..9d844354c4d9 100644 --- a/include/net/netfilter/nf_tables.h +++ b/include/net/netfilter/nf_tables.h @@ -180,6 +180,13 @@ static inline u64 nft_reg_load64(const u32 *sreg) return get_unaligned((u64 *)sreg); } +static inline bool nft_reg_overlap(u8 src, u8 dst, u32 len) +{ + unsigned int n = DIV_ROUND_UP(len, sizeof(u32)); + + return src != dst && src < dst + n && dst < src + n; +} + static inline void nft_data_copy(u32 *dst, const struct nft_data *src, unsigned int len) { diff --git a/net/netfilter/nft_bitwise.c b/net/netfilter/nft_bitwise.c index 94dccdcfa06b..785b8e9731d1 100644 --- a/net/netfilter/nft_bitwise.c +++ b/net/netfilter/nft_bitwise.c @@ -43,8 +43,10 @@ static void nft_bitwise_eval_lshift(u32 *dst, const u32 *src, u32 carry = 0; for (i = DIV_ROUND_UP(priv->len, sizeof(u32)); i > 0; i--) { - dst[i - 1] = (src[i - 1] << shift) | carry; - carry = src[i - 1] >> (BITS_PER_TYPE(u32) - shift); + u32 tmp_src = src[i - 1]; + + dst[i - 1] = (tmp_src << shift) | carry; + carry = tmp_src >> (BITS_PER_TYPE(u32) - shift); } } @@ -56,8 +58,10 @@ static void nft_bitwise_eval_rshift(u32 *dst, const u32 *src, u32 carry = 0; for (i = 0; i < DIV_ROUND_UP(priv->len, sizeof(u32)); i++) { - dst[i] = carry | (src[i] >> shift); - carry = src[i] << (BITS_PER_TYPE(u32) - shift); + u32 tmp_src = src[i]; + + dst[i] = carry | (tmp_src >> shift); + carry = tmp_src << (BITS_PER_TYPE(u32) - shift); } } @@ -235,6 +239,9 @@ static int nft_bitwise_init_bool(const struct nft_ctx *ctx, &priv->sreg2, priv->len); if (err < 0) return err; + + if (nft_reg_overlap(priv->sreg2, priv->dreg, priv->len)) + return -EINVAL; } return 0; @@ -265,6 +272,9 @@ static int nft_bitwise_init(const struct nft_ctx *ctx, if (err < 0) return err; + if (nft_reg_overlap(priv->sreg, priv->dreg, priv->len)) + return -EINVAL; + if (tb[NFTA_BITWISE_OP]) { priv->op = ntohl(nla_get_be32(tb[NFTA_BITWISE_OP])); switch (priv->op) { diff --git a/net/netfilter/nft_byteorder.c b/net/netfilter/nft_byteorder.c index e00dddfa2fc0..2316c77f4228 100644 --- a/net/netfilter/nft_byteorder.c +++ b/net/netfilter/nft_byteorder.c @@ -144,9 +144,16 @@ static int nft_byteorder_init(const struct nft_ctx *ctx, if (err < 0) return err; - return nft_parse_register_store(ctx, tb[NFTA_BYTEORDER_DREG], - &priv->dreg, NULL, NFT_DATA_VALUE, - priv->len); + err = nft_parse_register_store(ctx, tb[NFTA_BYTEORDER_DREG], + &priv->dreg, NULL, NFT_DATA_VALUE, + priv->len); + if (err < 0) + return err; + + if (nft_reg_overlap(priv->sreg, priv->dreg, priv->len)) + return -EINVAL; + + return 0; } static int nft_byteorder_dump(struct sk_buff *skb, From e1a9d791fd66ab2431b9e6f6f835823809869047 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Fri, 22 May 2026 12:16:21 +0200 Subject: [PATCH 176/411] USB: serial: cypress_m8: fix memory corruption with small endpoint Make sure that the interrupt-out endpoint max packet size is at least eight bytes to avoid user-controlled slab corruption or NULL-pointer dereference should a malicious device report a smaller size. Fixes: 3416eaa1f8f8 ("USB: cypress_m8: Packet format is separate from characteristic size") Cc: stable@vger.kernel.org # 2.6.26 Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/cypress_m8.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c index afff1a0f4298..0b8a4e9d7bc5 100644 --- a/drivers/usb/serial/cypress_m8.c +++ b/drivers/usb/serial/cypress_m8.c @@ -445,6 +445,14 @@ static int cypress_generic_port_probe(struct usb_serial_port *port) return -ENODEV; } + /* + * The buffer must be large enough for the one or two-byte header (and + * following data), but assume anything smaller than eight bytes is + * broken. + */ + if (port->interrupt_out_size < 8) + return -EINVAL; + priv = kzalloc_obj(struct cypress_private); if (!priv) return -ENOMEM; From f4feb1e20058e407cb00f45aff47f5b7e19a6bbf Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Wed, 20 May 2026 09:00:21 -0700 Subject: [PATCH 177/411] tun: free page on short-frame rejection in tun_xdp_one() tun_xdp_one() returns -EINVAL on a frame shorter than ETH_HLEN without freeing the page that vhost_net_build_xdp() allocated for it. tun_sendmsg() discards that -EINVAL and still returns total_len, so vhost_tx_batch() takes the success path and never frees the page; each short frame in a batch leaks one page-frag chunk. A local process that can open /dev/net/tun and /dev/vhost-net can hit this path: it attaches a tun/tap device as the vhost-net backend and feeds TX descriptors whose length minus the virtio-net header is below ETH_HLEN. Each kick leaks the page-frag chunks for that batch, and a tight submission loop exhausts host memory and triggers an OOM panic. Free the page before returning -EINVAL, matching the XDP-program error path in the same function. Fixes: 049584807f1d ("tun: add missing verification for short frame") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Reviewed-by: Dongli Zhang Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260520160020.375349-2-bestswngs@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/tun.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index b183189f1853..f594360d66d6 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -2394,8 +2394,10 @@ static int tun_xdp_one(struct tun_struct *tun, bool skb_xdp = false; struct page *page; - if (unlikely(datasize < ETH_HLEN)) + if (unlikely(datasize < ETH_HLEN)) { + put_page(virt_to_head_page(xdp->data)); return -EINVAL; + } xdp_prog = rcu_dereference(tun->xdp_prog); if (xdp_prog) { From dab48a7e74e6a394f3aa0461a2b1fb0c7b38fcb8 Mon Sep 17 00:00:00 2001 From: Thomas Fourier Date: Fri, 22 May 2026 10:54:04 +0200 Subject: [PATCH 178/411] Input: ims-pcu - fix usb_free_coherent() size in ims_pcu_buffers_free() The input buffer size is pcu->max_in_size, but pcu->max_out_size is passed to usb_free_coherent(). Change size to match the allocation size. Fixes: 628329d52474 ("Input: add IMS Passenger Control Unit driver") Cc: stable@vger.kernel.org Signed-off-by: Thomas Fourier Link: https://patch.msgid.link/20260522085412.45430-2-fourier.thomas@gmail.com Signed-off-by: Dmitry Torokhov --- drivers/input/misc/ims-pcu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/input/misc/ims-pcu.c b/drivers/input/misc/ims-pcu.c index 4c022a36dbe8..7a1cb9333f53 100644 --- a/drivers/input/misc/ims-pcu.c +++ b/drivers/input/misc/ims-pcu.c @@ -1624,7 +1624,7 @@ static void ims_pcu_buffers_free(struct ims_pcu *pcu) usb_kill_urb(pcu->urb_in); usb_free_urb(pcu->urb_in); - usb_free_coherent(pcu->udev, pcu->max_out_size, + usb_free_coherent(pcu->udev, pcu->max_in_size, pcu->urb_in_buf, pcu->read_dma); kfree(pcu->urb_out_buf); From 3bcf7aec6a9d16438f2cec29f5d7c8d5b8edf9b2 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Thu, 21 May 2026 09:32:31 -0700 Subject: [PATCH 179/411] tap: free page on error paths in tap_get_user_xdp() tap_get_user_xdp() rejects a frame shorter than ETH_HLEN with -EINVAL, and returns -ENOMEM when build_skb() fails. Both paths jump to the err label without freeing the page that vhost_net_build_xdp() allocated for the frame. tap_sendmsg() discards the per-buffer return value and always returns 0, so vhost_tx_batch() takes the success path and never frees the page; each rejected frame in a batch leaks one page-frag chunk. Free the page on both error paths, before the skb is built. This is the tap counterpart of the same leak in tun_xdp_one(). Fixes: 0efac27791ee ("tap: accept an array of XDP buffs through sendmsg()") Fixes: ed7f2afdd0e0 ("tap: add missing verification for short frame") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Reviewed-by: Dongli Zhang Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260521163230.1478627-2-bestswngs@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/tap.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/tap.c b/drivers/net/tap.c index a590e07ce0a9..fae115915c8e 100644 --- a/drivers/net/tap.c +++ b/drivers/net/tap.c @@ -1052,6 +1052,7 @@ static int tap_get_user_xdp(struct tap_queue *q, struct xdp_buff *xdp) int err, depth; if (unlikely(xdp->data_end - xdp->data < ETH_HLEN)) { + put_page(virt_to_head_page(xdp->data)); err = -EINVAL; goto err; } @@ -1061,6 +1062,7 @@ static int tap_get_user_xdp(struct tap_queue *q, struct xdp_buff *xdp) skb = build_skb(xdp->data_hard_start, buflen); if (!skb) { + put_page(virt_to_head_page(xdp->data)); err = -ENOMEM; goto err; } From aa8963fdce667a42fb7f0bdd2909fadcab02f9a8 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Thu, 21 May 2026 09:33:13 -0700 Subject: [PATCH 180/411] tun: free page on build_skb failure in tun_xdp_one() When build_skb() fails in tun_xdp_one(), the function sets ret to -ENOMEM and jumps to the out label, which returns without freeing the page that vhost_net_build_xdp() allocated for the frame. As with the short-frame rejection path, tun_sendmsg() discards the per-buffer error and still returns total_len, so vhost_tx_batch() takes the success path and never frees the page. Each build_skb() failure in a batch leaks one page-frag chunk. Free the page before taking the error path, matching the put_page() the other error exits of tun_xdp_one() already perform. Fixes: 043d222f93ab ("tuntap: accept an array of XDP buffs through sendmsg()") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Reviewed-by: Dongli Zhang Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260521163312.1479805-2-bestswngs@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/tun.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/tun.c b/drivers/net/tun.c index f594360d66d6..9e7744eb57a3 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -2439,6 +2439,7 @@ static int tun_xdp_one(struct tun_struct *tun, build: skb = build_skb(xdp->data_hard_start, buflen); if (!skb) { + put_page(virt_to_head_page(xdp->data)); ret = -ENOMEM; goto out; } From aae9d8a5528b8ee9ff8dc5d3558b8a9f852a724a Mon Sep 17 00:00:00 2001 From: Ziyu Zhang Date: Wed, 20 May 2026 00:56:36 +0800 Subject: [PATCH 181/411] vsock: keep poll shutdown state consistent vsock_poll() reads vsk->peer_shutdown before taking the socket lock to set EPOLLHUP and EPOLLRDHUP, then reads it again after taking the lock to report EOF readability. A shutdown packet can update peer_shutdown while poll is waiting for the lock, so one poll invocation can report EOF readability without the corresponding HUP/RDHUP bits. For connectible sockets, take one peer_shutdown snapshot after lock_sock() and use it for all peer-shutdown-derived poll bits. For datagram sockets, which do not take lock_sock() in poll(), take one lockless READ_ONCE() snapshot and pair it with WRITE_ONCE() on the writer side. This keeps the peer-shutdown-derived bits internally consistent for each poll pass. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Signed-off-by: Ziyu Zhang Link: https://patch.msgid.link/20260519165636.62542-1-ziyuzhang201@gmail.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/af_vsock.c | 49 ++++++++++++++++--------- net/vmw_vsock/hyperv_transport.c | 9 +++-- net/vmw_vsock/virtio_transport_common.c | 14 ++++--- net/vmw_vsock/vmci_transport.c | 8 ++-- 4 files changed, 52 insertions(+), 28 deletions(-) diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c index 44037b066a5f..2ce1063d4a67 100644 --- a/net/vmw_vsock/af_vsock.c +++ b/net/vmw_vsock/af_vsock.c @@ -642,7 +642,7 @@ int vsock_assign_transport(struct vsock_sock *vsk, struct vsock_sock *psk) */ sock_reset_flag(sk, SOCK_DONE); sk->sk_state = TCP_CLOSE; - vsk->peer_shutdown = 0; + WRITE_ONCE(vsk->peer_shutdown, 0); } if (sk->sk_type == SOCK_SEQPACKET) { @@ -933,7 +933,7 @@ static struct sock *__vsock_create(struct net *net, vsk->rejected = false; vsk->sent_request = false; vsk->ignore_connecting_rst = false; - vsk->peer_shutdown = 0; + WRITE_ONCE(vsk->peer_shutdown, 0); INIT_DELAYED_WORK(&vsk->connect_work, vsock_connect_timeout); INIT_DELAYED_WORK(&vsk->pending_work, vsock_pending_work); @@ -1241,6 +1241,25 @@ static int vsock_shutdown(struct socket *sock, int mode) return err; } +static __poll_t vsock_poll_shutdown(struct sock *sk, u32 peer_shutdown) +{ + __poll_t mask = 0; + + /* INET sockets treat local write shutdown and peer write shutdown as a + * case of EPOLLHUP set. + */ + if (sk->sk_shutdown == SHUTDOWN_MASK || + ((sk->sk_shutdown & SEND_SHUTDOWN) && + (peer_shutdown & SEND_SHUTDOWN))) + mask |= EPOLLHUP; + + if (sk->sk_shutdown & RCV_SHUTDOWN || + peer_shutdown & SEND_SHUTDOWN) + mask |= EPOLLRDHUP; + + return mask; +} + static __poll_t vsock_poll(struct file *file, struct socket *sock, poll_table *wait) { @@ -1258,24 +1277,17 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, /* Signify that there has been an error on this socket. */ mask |= EPOLLERR; - /* INET sockets treat local write shutdown and peer write shutdown as a - * case of EPOLLHUP set. - */ - if ((sk->sk_shutdown == SHUTDOWN_MASK) || - ((sk->sk_shutdown & SEND_SHUTDOWN) && - (vsk->peer_shutdown & SEND_SHUTDOWN))) { - mask |= EPOLLHUP; - } - - if (sk->sk_shutdown & RCV_SHUTDOWN || - vsk->peer_shutdown & SEND_SHUTDOWN) { - mask |= EPOLLRDHUP; - } - if (sk_is_readable(sk)) mask |= EPOLLIN | EPOLLRDNORM; if (sock->type == SOCK_DGRAM) { + u32 peer_shutdown = READ_ONCE(vsk->peer_shutdown); + + /* DGRAM sockets do not take lock_sock() in poll(), so use one + * lockless snapshot for all shutdown-derived mask bits. + */ + mask |= vsock_poll_shutdown(sk, peer_shutdown); + /* For datagram sockets we can read if there is something in * the queue and write as long as the socket isn't shutdown for * sending. @@ -1290,6 +1302,7 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, } else if (sock_type_connectible(sk->sk_type)) { const struct vsock_transport *transport; + u32 peer_shutdown; lock_sock(sk); @@ -1322,8 +1335,10 @@ static __poll_t vsock_poll(struct file *file, struct socket *sock, * terminated should also be considered read, and we check the * shutdown flag for that. */ + peer_shutdown = READ_ONCE(vsk->peer_shutdown); + mask |= vsock_poll_shutdown(sk, peer_shutdown); if (sk->sk_shutdown & RCV_SHUTDOWN || - vsk->peer_shutdown & SEND_SHUTDOWN) { + peer_shutdown & SEND_SHUTDOWN) { mask |= EPOLLIN | EPOLLRDNORM; } diff --git a/net/vmw_vsock/hyperv_transport.c b/net/vmw_vsock/hyperv_transport.c index 7a8963595bf9..b3394946b2ed 100644 --- a/net/vmw_vsock/hyperv_transport.c +++ b/net/vmw_vsock/hyperv_transport.c @@ -264,7 +264,7 @@ static void hvs_do_close_lock_held(struct vsock_sock *vsk, struct sock *sk = sk_vsock(vsk); sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); if (vsock_stream_has_data(vsk) <= 0) sk->sk_state = TCP_CLOSING; sk->sk_state_change(sk); @@ -593,7 +593,9 @@ static int hvs_update_recv_data(struct hvsock *hvs) return -EIO; if (payload_len == 0) - hvs->vsk->peer_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(hvs->vsk->peer_shutdown, + READ_ONCE(hvs->vsk->peer_shutdown) | + SEND_SHUTDOWN); hvs->recv_data_len = payload_len; hvs->recv_data_off = 0; @@ -736,7 +738,8 @@ static s64 hvs_stream_has_data(struct vsock_sock *vsk) return ret; return hvs->recv_data_len; case 0: - vsk->peer_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(vsk->peer_shutdown, + READ_ONCE(vsk->peer_shutdown) | SEND_SHUTDOWN); ret = 0; break; default: /* -1 */ diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index df3b418e0392..d4d26fba9e37 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -1228,7 +1228,7 @@ static void virtio_transport_do_close(struct vsock_sock *vsk, struct sock *sk = sk_vsock(vsk); sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); if (vsock_stream_has_data(vsk) <= 0) sk->sk_state = TCP_CLOSING; sk->sk_state_change(sk); @@ -1431,12 +1431,15 @@ virtio_transport_recv_connected(struct sock *sk, case VIRTIO_VSOCK_OP_CREDIT_UPDATE: sk->sk_write_space(sk); break; - case VIRTIO_VSOCK_OP_SHUTDOWN: + case VIRTIO_VSOCK_OP_SHUTDOWN: { + u32 peer_shutdown = READ_ONCE(vsk->peer_shutdown); + if (le32_to_cpu(hdr->flags) & VIRTIO_VSOCK_SHUTDOWN_RCV) - vsk->peer_shutdown |= RCV_SHUTDOWN; + peer_shutdown |= RCV_SHUTDOWN; if (le32_to_cpu(hdr->flags) & VIRTIO_VSOCK_SHUTDOWN_SEND) - vsk->peer_shutdown |= SEND_SHUTDOWN; - if (vsk->peer_shutdown == SHUTDOWN_MASK) { + peer_shutdown |= SEND_SHUTDOWN; + WRITE_ONCE(vsk->peer_shutdown, peer_shutdown); + if (peer_shutdown == SHUTDOWN_MASK) { if (vsock_stream_has_data(vsk) <= 0 && !sock_flag(sk, SOCK_DONE)) { (void)virtio_transport_reset(vsk, NULL); virtio_transport_do_close(vsk, true); @@ -1451,6 +1454,7 @@ virtio_transport_recv_connected(struct sock *sk, if (le32_to_cpu(virtio_vsock_hdr(skb)->flags)) sk->sk_state_change(sk); break; + } case VIRTIO_VSOCK_OP_RST: virtio_transport_do_close(vsk, true); break; diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index d2579380f51e..5c1ecd5bfdbc 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -819,7 +819,7 @@ static void vmci_transport_handle_detach(struct sock *sk) /* On a detach the peer will not be sending or receiving * anymore. */ - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); /* We should not be sending anymore since the peer won't be * there to receive, but we can still receive if there is data @@ -1542,7 +1542,9 @@ static int vmci_transport_recv_connected(struct sock *sk, if (pkt->u.mode) { vsk = vsock_sk(sk); - vsk->peer_shutdown |= pkt->u.mode; + WRITE_ONCE(vsk->peer_shutdown, + READ_ONCE(vsk->peer_shutdown) | + pkt->u.mode); sk->sk_state_change(sk); } break; @@ -1559,7 +1561,7 @@ static int vmci_transport_recv_connected(struct sock *sk, * a clean shutdown. */ sock_set_flag(sk, SOCK_DONE); - vsk->peer_shutdown = SHUTDOWN_MASK; + WRITE_ONCE(vsk->peer_shutdown, SHUTDOWN_MASK); if (vsock_stream_has_data(vsk) <= 0) sk->sk_state = TCP_CLOSING; From 70f8592ee90585272018a725054b6eb2ab7e99ca Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Wed, 20 May 2026 19:22:35 +0200 Subject: [PATCH 182/411] net: netlink: fix sending unassigned nsid after assigned one If the current skb is not shared, it is re-used directly for all the sockets subscribed to the notification. If we have remote all-nsid socket receiving a message first, then the 'nsid_is_set' will be set to 'true'. If the nsid is NOT_ASSIGNED for the next socket in the list, the 'nsid_is_set' will remain 'true' and the negative value is be delivered to the user space. All subsequent nsid values will be delivered as well, since there is no code path that sets the flag back to 'false'. Fix that by always dropping the flag to 'false' first. Fixes: 7212462fa6fd ("netlink: don't send unknown nsid") Signed-off-by: Ilya Maximets Acked-by: Nicolas Dichtel Link: https://patch.msgid.link/20260520172317.175168-2-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 2aeb0680807d..0742e97f256e 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1482,6 +1482,7 @@ static void do_one_broadcast(struct sock *sk, p->skb2 = NULL; goto out; } + NETLINK_CB(p->skb2).nsid_is_set = false; NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) NETLINK_CB(p->skb2).nsid_is_set = true; From 88b126b39f9757e9debc322d4679239e9af089c7 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Wed, 20 May 2026 19:22:36 +0200 Subject: [PATCH 183/411] net: netlink: don't set nsid on local notifications In most cases, notifications on sockets with NETLINK_LISTEN_ALL_NSID do not contain NSID in their ancillary data in case the event is local to the listener. However, when a self-referential NSID is allocated for a namespace, every local notification starts sending this ID to the user space. This is problematic, because the listener cannot tell if those notifications are local or not anymore without making extra requests to figure out if the provided NSID is local or not. The listener can also not figure out the local NSID beforehand as it can be allocated at any point in time by other processes, changing the structure of the future notifications for everyone. The value is practically not useful, since it's the namespace's own ID that the application has to obtain from other sources in order to figure out if it's the same or not. So, for the application it's just an extra busy work with no benefits. Moreover, applications that do not know about this quirk may be mishandling notifications with NSID set as notifications from remote namespaces. This is the case for ovs-vswitchd and the iproute2's 'ip monitor' that stops printing 'current' and starts printing the nsid number mid-session. Lack of clear documentation for this behavior is also not helping. A search though open-source projects doesn't reveal any projects that use NETNSA_NSID_NOT_ASSIGNED and rely on metadata to contain self-referential NSIDs (expected, since the value is not useful). Quite the opposite, as already mentioned, there are few applications that rely on NSID to not be present in local events. Since the value is not useful and actively harmful in some cases, let's not report it for local events, making the notifications more consistent. Also adding some blank lines for readability. Fixes: 59324cf35aba ("netlink: allow to listen "all" netns") Reported-by: Matteo Perin Signed-off-by: Ilya Maximets Acked-by: Nicolas Dichtel Link: https://patch.msgid.link/20260520172317.175168-3-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- net/netlink/af_netlink.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/netlink/af_netlink.c b/net/netlink/af_netlink.c index 0742e97f256e..7269e23b578d 100644 --- a/net/netlink/af_netlink.c +++ b/net/netlink/af_netlink.c @@ -1482,10 +1482,14 @@ static void do_one_broadcast(struct sock *sk, p->skb2 = NULL; goto out; } + NETLINK_CB(p->skb2).nsid_is_set = false; - NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); - if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) - NETLINK_CB(p->skb2).nsid_is_set = true; + if (!net_eq(sock_net(sk), p->net)) { + NETLINK_CB(p->skb2).nsid = peernet2id(sock_net(sk), p->net); + if (NETLINK_CB(p->skb2).nsid != NETNSA_NSID_NOT_ASSIGNED) + NETLINK_CB(p->skb2).nsid_is_set = true; + } + val = netlink_broadcast_deliver(sk, p->skb2); if (val < 0) { netlink_overrun(sk); From 2e43b64248909c617281921d6d9ba3bfc0159473 Mon Sep 17 00:00:00 2001 From: Ilya Maximets Date: Wed, 20 May 2026 19:22:38 +0200 Subject: [PATCH 184/411] selftests: net: add a test case for nsid in all nsid notifications The test subscribes to link events from all namespaces and makes sure that local events do not carry NSID in their ancillary data (even if there is a self-referential NSID allocated for the local namespace), and remote events do. Signed-off-by: Ilya Maximets Acked-by: Nicolas Dichtel Link: https://patch.msgid.link/20260520172317.175168-5-i.maximets@ovn.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/link_netns.py | 61 ++++++++++++++++++++++- 1 file changed, 59 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/link_netns.py b/tools/testing/selftests/net/link_netns.py index aab043c59d69..6d1f863b6262 100755 --- a/tools/testing/selftests/net/link_netns.py +++ b/tools/testing/selftests/net/link_netns.py @@ -3,13 +3,14 @@ import time -from lib.py import ksft_run, ksft_exit, ksft_true +from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_true from lib.py import ip from lib.py import NetNS, NetNSEnter from lib.py import RtnlFamily LINK_NETNSID = 100 +LINK_NETNSID2 = 200 def test_event() -> None: @@ -32,6 +33,57 @@ def test_event() -> None: "Received unexpected link notification") +def test_event_all_nsid() -> None: + """NETLINK_LISTEN_ALL_NSID notifications: local events must not + carry nsid even with a self-referential mapping. Remote events + must carry the correct nsid.""" + + with NetNS() as ns1, NetNS() as ns2: + net1, net2 = str(ns1), str(ns2) + + with NetNSEnter(net1): + rtnl = RtnlFamily() + rtnl.ntf_listen_all_nsid() + rtnl.ntf_subscribe("rtnlgrp-link") + + # Case 1: no nsid assigned, local event, no nsid expected. + ip("link add dummy-lo type dummy", ns=net1) + + # Case 2: self-referential nsid, local event, still no nsid. + ip(f"netns set {net1} {LINK_NETNSID}", ns=net1) + ip("link add dummy-sr type dummy", ns=net1) + + # Case 3: remote event, nsid present. + ip(f"netns set {net2} {LINK_NETNSID2}", ns=net1) + ip("link add dummy-re type dummy", ns=net2) + + # Collect the three newlink events, ignoring unrelated noise. + events = {} + for msg in rtnl.poll_ntf(duration=1): + if msg['name'] == 'getlink': + ifname = msg['msg'].get('ifname') + if ifname in ('dummy-lo', 'dummy-sr', 'dummy-re'): + events[ifname] = msg + if len(events) == 3: + break + + ksft_true('dummy-lo' in events, "missing local event") + ksft_true(events['dummy-lo'].get('nsid') is None, + "local event without nsid should not carry nsid") + + ksft_true('dummy-sr' in events, "missing self-ref event") + ksft_true(events['dummy-sr'].get('nsid') is None, + "local event with self-ref nsid should not carry nsid") + + ksft_true('dummy-re' in events, "missing remote event") + ksft_eq(events['dummy-re'].get('nsid'), LINK_NETNSID2, + "remote event should carry nsid") + + ip("link del dummy-lo", ns=net1) + ip("link del dummy-sr", ns=net1) + ip("link del dummy-re", ns=net2) + + def validate_link_netns(netns, ifname, link_netnsid) -> bool: link_info = ip(f"-d link show dev {ifname}", ns=netns, json=True) if not link_info: @@ -133,7 +185,12 @@ def test_peer_net() -> None: def main() -> None: - ksft_run([test_event, test_link_net, test_peer_net]) + ksft_run([ + test_event, + test_event_all_nsid, + test_link_net, + test_peer_net, + ]) ksft_exit() From 9e4389b0038781f19f97895186ed941ff8ac1678 Mon Sep 17 00:00:00 2001 From: Alexandra Winter Date: Thu, 21 May 2026 16:56:39 +0200 Subject: [PATCH 185/411] net/smc: Do not re-initialize smc hashtables INIT_HLIST_HEAD(&smc_v*_hashinfo.ht) are called after smc_nl_init(), proto_register() and sock_register(). This can lead to smc_v*_hashinfo.ht being reset even though hash entries already exist and are being used, possibly resulting in a corrupted list. Remove unnecessary and dangerous re-initialisation of smc_v*_hashinfo.ht in smc_init(); it is implicitly initialised to zero anyhow. Add HLIST_HEAD_INIT to the definitions for clarity. Fixes: f16a7dd5cf27 ("smc: netlink interface for SMC sockets") Suggested-by: Halil Pasic Signed-off-by: Alexandra Winter Acked-by: Halil Pasic Reviewed-by: Mahanta Jambigi Link: https://patch.msgid.link/20260521145639.10317-1-wintera@linux.ibm.com Signed-off-by: Jakub Kicinski --- net/smc/af_smc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/smc/af_smc.c b/net/smc/af_smc.c index dffbd529762d..b5db69073e20 100644 --- a/net/smc/af_smc.c +++ b/net/smc/af_smc.c @@ -188,10 +188,12 @@ static bool smc_hs_congested(const struct sock *sk) struct smc_hashinfo smc_v4_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v4_hashinfo.lock), + .ht = HLIST_HEAD_INIT, }; struct smc_hashinfo smc_v6_hashinfo = { .lock = __RW_LOCK_UNLOCKED(smc_v6_hashinfo.lock), + .ht = HLIST_HEAD_INIT, }; int smc_hash_sk(struct sock *sk) @@ -3517,8 +3519,6 @@ static int __init smc_init(void) pr_err("%s: sock_register fails with %d\n", __func__, rc); goto out_proto6; } - INIT_HLIST_HEAD(&smc_v4_hashinfo.ht); - INIT_HLIST_HEAD(&smc_v6_hashinfo.ht); rc = smc_ib_register_client(); if (rc) { From 3589d20a666caf30ad100c960a2de7de390fce88 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Thu, 21 May 2026 07:11:45 -0700 Subject: [PATCH 186/411] net/iucv: fix locking in .getsockopt Mirror iucv_sock_setsockopt() and wrap the whole switch in lock_sock()/release_sock(). The pre-existing SO_MSGLIMIT-only lock becomes redundant and is removed. Any AF_IUCV HIPER user can potentially crash the kernel by racing recvmsg() with getsockopt(SO_MSGSIZE): the SO_MSGSIZE arm dereferences iucv->hs_dev->mtu after iucv_sock_close() (called from the racing recvmsg()) has set hs_dev to NULL, producing a NULL pointer dereference oops. Suggested-by: Stanislav Fomichev Fixes: 51363b8751a6 ("af_iucv: allow retrieval of maximum message size") Signed-off-by: Breno Leitao Reviewed-by: Alexandra Winter Tested-by: Alexandra Winter Link: https://patch.msgid.link/20260521-af_iucv_fix2-v1-1-f16b1c510aa9@debian.org Signed-off-by: Jakub Kicinski --- net/iucv/af_iucv.c | 20 ++++++++++++++------ 1 file changed, 14 insertions(+), 6 deletions(-) diff --git a/net/iucv/af_iucv.c b/net/iucv/af_iucv.c index 72dfccd4e3d5..c2dc3338670e 100644 --- a/net/iucv/af_iucv.c +++ b/net/iucv/af_iucv.c @@ -1540,7 +1540,7 @@ static int iucv_sock_getsockopt(struct socket *sock, int level, int optname, struct sock *sk = sock->sk; struct iucv_sock *iucv = iucv_sk(sk); unsigned int val; - int len; + int len, rc; if (level != SOL_IUCV) return -ENOPROTOOPT; @@ -1553,26 +1553,34 @@ static int iucv_sock_getsockopt(struct socket *sock, int level, int optname, len = min_t(unsigned int, len, sizeof(int)); + rc = 0; + + lock_sock(sk); switch (optname) { case SO_IPRMDATA_MSG: val = (iucv->flags & IUCV_IPRMDATA) ? 1 : 0; break; case SO_MSGLIMIT: - lock_sock(sk); val = (iucv->path != NULL) ? iucv->path->msglim /* connected */ : iucv->msglimit; /* default */ - release_sock(sk); break; case SO_MSGSIZE: - if (sk->sk_state == IUCV_OPEN) - return -EBADFD; + if (sk->sk_state == IUCV_OPEN) { + rc = -EBADFD; + break; + } val = (iucv->hs_dev) ? iucv->hs_dev->mtu - sizeof(struct af_iucv_trans_hdr) - ETH_HLEN : 0x7fffffff; break; default: - return -ENOPROTOOPT; + rc = -ENOPROTOOPT; + break; } + release_sock(sk); + + if (rc) + return rc; if (put_user(len, optlen)) return -EFAULT; From 7205b58702273baf21d6ba7992e6ba15852325f7 Mon Sep 17 00:00:00 2001 From: David Jeffery Date: Fri, 15 May 2026 14:09:41 -0400 Subject: [PATCH 187/411] scsi: core: Run queues for all non-SDEV_DEL devices from scsi_run_host_queues While a SCSI host is in a recovery state, scsi_mq_requeue_cmd() will not set the requeue list for a requeued command to be kicked in the future. The expectation is a call to scsi_run_host_queues() will kick all SCSI devices once the recovery state is cleared. However, scsi_run_host_queues() uses shost_for_each_device() which uses scsi_device_get() and so will ignore devices in a partially removed state like SDEV_CANCEL. But these devices may also have requeued requests, leaving their requests stuck from not being kicked and causing the removal process of the device to hang. scsi_run_host_queues() needs to run against more devices than the macro shost_for_each_device() allows. Instead of using the too limiting scsi_device_get() state checks, only ignore devices in SDEV_DEL state or when unable to acquire a reference. Attempt to run the queues for all other devices when scsi_run_host_queues() is called. Fixes: 8b566edbdbfb ("scsi: core: Only kick the requeue list if necessary") Signed-off-by: David Jeffery Reviewed-by: Bart Van Assche Link: https://patch.msgid.link/20260515180941.9698-1-djeffery@redhat.com Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_lib.c | 27 +++++++++++++++++++++++++-- 1 file changed, 25 insertions(+), 2 deletions(-) diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index 6e8c7a42603e..85eef401925a 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -575,10 +575,33 @@ void scsi_requeue_run_queue(struct work_struct *work) void scsi_run_host_queues(struct Scsi_Host *shost) { - struct scsi_device *sdev; + struct scsi_device *sdev, *prev = NULL; + unsigned long flags; - shost_for_each_device(sdev, shost) + spin_lock_irqsave(shost->host_lock, flags); + __shost_for_each_device(sdev, shost) { + /* + * Only skip devices so deep into removal they will never need + * another kick to their queues. Thus scsi_device_get() cannot + * be used as it would skip devices in SDEV_CANCEL state which + * may need a queue kick. + */ + if (sdev->sdev_state == SDEV_DEL || + !get_device(&sdev->sdev_gendev)) + continue; + spin_unlock_irqrestore(shost->host_lock, flags); + + if (prev) + put_device(&prev->sdev_gendev); scsi_run_queue(sdev->request_queue); + + prev = sdev; + + spin_lock_irqsave(shost->host_lock, flags); + } + spin_unlock_irqrestore(shost->host_lock, flags); + if (prev) + put_device(&prev->sdev_gendev); } static void scsi_uninit_cmd(struct scsi_cmnd *cmd) From adda8a44e1e43aceba058839f56fa1c599f6f99b Mon Sep 17 00:00:00 2001 From: Alexander Perlis Date: Tue, 12 May 2026 18:12:54 -0500 Subject: [PATCH 188/411] scsi: devinfo: Add BLIST_NO_RSOC for Promise VTrak E310f The extremely slow boots reported July 2014 in bug 79901: https://bugzilla.kernel.org/show_bug.cgi?id=79901 for Promise VTrak E610f 3U 16-bay FC RAID enclosure occur also with the Promise VTrak E310f 2U 12-bay FC RAID enclosure. The 2014 patch: https://bugzilla.kernel.org/attachment.cgi?id=144101&action=diff added support for the BLIST_NO_RSOC flag and specified that flag for the Promise VTrak E610f. This current patch simply adds the E310f to that same list. One curiosity is the additional BLIST_SPARSELUN flag. This was also in the 2014 patch for the E610f, and was already in place for *all* Promise devices since 2007 due to commit e0b2e597d5dd ("[SCSI] stex: fix id mapping issue") which added the line: {"Promise", "", NULL, BLIST_SPARSELUN} The 2007 commit message talks of issues with SuperTrak EX (stex) but the added line did not limit itself to that particular device family. The current patch for E310F, like the 2014 patch for E610f, adds BLIST_NO_RSOC while preserving BLIST_SPARSELUN from 2007. Signed-off-by: Alexander Perlis Suggested-by: Nikkos Svoboda Link: https://patch.msgid.link/20260512231254.27530-1-aperlis@math.lsu.edu Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_devinfo.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c index 68a992494b12..c6defe1c3152 100644 --- a/drivers/scsi/scsi_devinfo.c +++ b/drivers/scsi/scsi_devinfo.c @@ -218,6 +218,7 @@ static struct { {"PIONEER", "CD-ROM DRM-602X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, {"PIONEER", "CD-ROM DRM-604X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, {"PIONEER", "CD-ROM DRM-624X", NULL, BLIST_FORCELUN | BLIST_SINGLELUN}, + {"Promise", "VTrak E310f", NULL, BLIST_SPARSELUN | BLIST_NO_RSOC}, {"Promise", "VTrak E610f", NULL, BLIST_SPARSELUN | BLIST_NO_RSOC}, {"Promise", "", NULL, BLIST_SPARSELUN}, {"QEMU", "QEMU CD-ROM", NULL, BLIST_SKIP_VPD_PAGES}, From a4719ae23fb5b1b6229120c7ea4b6143a501a62e Mon Sep 17 00:00:00 2001 From: "Milan P. Gandhi" Date: Thu, 14 May 2026 13:27:54 +0530 Subject: [PATCH 189/411] scsi: megaraid_sas: Fix NULL pointer dereference on firmware duplicate completion Add NULL check for scmd_local in the MPI2_FUNCTION_SCSI_IO_REQUEST case to handle firmware duplicate/stale completions. When firmware sends a duplicate completion for a command that was already processed and returned to the pool, the driver accesses NULL scmd pointer causing a crash. Timeline of the bug: 1. Command completes normally, megasas_return_cmd_fusion() called 2. This sets cmd->scmd = NULL and clears io_request with memset(..., 0, ...) 3. Firmware sends duplicate/stale completion for same SMID (firmware bug) 4. Driver processes reply descriptor again 5. Cleared io_request has Function = 0 (MPI2_FUNCTION_SCSI_IO_REQUEST) 6. Switch statement matches SCSI_IO_REQUEST case by accident 7. Accesses megasas_priv(NULL scmd)->status -> crash at offset 0x228 The offset 0x228 = sizeof(struct scsi_cmnd) 0x220 + offsetof(status) 0x8. This issue was observed on PERC H330 Mini running firmware 25.5.9.0001 after 3+ days of heavy I/O load. Crash signature: BUG: unable to handle kernel NULL pointer dereference at 0x228 RIP: complete_cmd_fusion+0x428 Function: megasas_priv(cmd_fusion->scmd)->status Add defensive check to skip processing when scmd_local is NULL. This handles duplicate completions from firmware and prevents accessing freed command structures. The check protects all scmd_local uses in both the SCSI_IO path and the fallthrough LDIO path. Signed-off-by: Milan P. Gandhi Link: https://patch.msgid.link/agWAgtk6rtHqNWb5@machine1 Signed-off-by: Martin K. Petersen --- drivers/scsi/megaraid/megaraid_sas_fusion.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c index 2699e4e09b5b..056cbe50e19e 100644 --- a/drivers/scsi/megaraid/megaraid_sas_fusion.c +++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c @@ -3612,6 +3612,15 @@ complete_cmd_fusion(struct megasas_instance *instance, u32 MSIxIndex, complete(&cmd_fusion->done); break; case MPI2_FUNCTION_SCSI_IO_REQUEST: /*Fast Path IO.*/ + /* + * Firmware can send stale/duplicate completions for + * commands already returned to the pool. scmd_local + * would be NULL for such cases. Skip processing to + * avoid NULL pointer access. + */ + if (!scmd_local) + break; + /* Update load balancing info */ if (fusion->load_balance_info && (megasas_priv(cmd_fusion->scmd)->status & From e4bb73bf3ac11b4a93634660345b9d764a4a80df Mon Sep 17 00:00:00 2001 From: "Ewan D. Milne" Date: Tue, 19 May 2026 16:53:56 -0400 Subject: [PATCH 190/411] scsi: scsi_debug: Add missing newline in scsi_debug_device_reset() A "\n" at the end of the sdev_printk() string appears to have been inadvertently removed. Add it back for correct log message formatting. Fixes: a743b120227a ("scsi: scsi_debug: Stop printing extra function name in debug logs") Assisted-by: Claude:claude-opus-4-6 Signed-off-by: Ewan D. Milne Reviewed-by: Bart Van Assche Reviewed-by: John Garry Link: https://patch.msgid.link/20260519205356.1040855-1-emilne@redhat.com Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c index 1515495fd9ea..040c5e1e713a 100644 --- a/drivers/scsi/scsi_debug.c +++ b/drivers/scsi/scsi_debug.c @@ -6953,7 +6953,7 @@ static int scsi_debug_device_reset(struct scsi_cmnd *SCpnt) ++num_dev_resets; if (SDEBUG_OPT_ALL_NOISE & sdebug_opts) - sdev_printk(KERN_INFO, sdp, "doing device reset"); + sdev_printk(KERN_INFO, sdp, "doing device reset\n"); scsi_debug_stop_all_queued(sdp); if (devip) { From a9a39233ec1fc9f97ea1340a4d09bb7ec2be5153 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Wed, 20 May 2026 09:30:15 -0400 Subject: [PATCH 191/411] scsi: scsi_transport_fc: Widen FPIN pname walker counter to u32 An adjacent Fibre Channel fabric actor that can deliver an FPIN ELS frame to an lpfc or qla2xxx Linux initiator can trigger a non-return in the generic FC transport. This is not a local userspace or IP network path; the attacker must be able to inject fabric traffic, for example as a compromised switch or fabric controller, or as a same-zone N_Port on a fabric that permits source spoofing. The Link-Integrity and Peer-Congestion FPIN walkers used a u8 loop counter against the 32-bit on-wire pname_count field, and did not bound pname_count by the descriptor body already validated by the TLV walker. A pname_count of 256 therefore wraps the counter and keeps the loop condition true indefinitely. Factor the shared pname_list[] walk into one helper, widen the counter to u32, and clamp pname_count against the entries that fit in the descriptor body before iterating. Fixes: 3dcfe0de5a97 ("scsi: fc: Parse FPIN packets and update statistics") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Reviewed-by: Christoph Hellwig Reviewed-by: John Garry Link: https://patch.msgid.link/20260520133015.1018937-1-michael.bommarito@gmail.com Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_transport_fc.c | 77 +++++++++++++++++--------------- 1 file changed, 41 insertions(+), 36 deletions(-) diff --git a/drivers/scsi/scsi_transport_fc.c b/drivers/scsi/scsi_transport_fc.c index dce95e361daf..173ed6373f04 100644 --- a/drivers/scsi/scsi_transport_fc.c +++ b/drivers/scsi/scsi_transport_fc.c @@ -737,6 +737,37 @@ fc_cn_stats_update(u16 event_type, struct fc_fpin_stats *stats) } } +static void +fc_fpin_pname_stats_update(struct Scsi_Host *shost, + struct fc_rport *attach_rport, u16 event_type, + u32 desc_len, u32 fixed_len, u32 pname_count, + __be64 *pname_list, + void (*stats_update)(u16 event_type, + struct fc_fpin_stats *stats)) +{ + u32 i; + struct fc_rport *rport; + u64 wwpn; + + if (desc_len < fixed_len) + pname_count = 0; + else + pname_count = min(pname_count, (desc_len - fixed_len) / + sizeof(pname_list[0])); + + for (i = 0; i < pname_count; i++) { + wwpn = be64_to_cpu(pname_list[i]); + rport = fc_find_rport_by_wwpn(shost, wwpn); + if (rport && + (rport->roles & FC_PORT_ROLE_FCP_TARGET || + rport->roles & FC_PORT_ROLE_NVME_TARGET)) { + if (rport == attach_rport) + continue; + stats_update(event_type, &rport->fpin_stats); + } + } +} + /* * fc_fpin_li_stats_update - routine to update Link Integrity * event statistics. @@ -747,13 +778,11 @@ fc_cn_stats_update(u16 event_type, struct fc_fpin_stats *stats) static void fc_fpin_li_stats_update(struct Scsi_Host *shost, struct fc_tlv_desc *tlv) { - u8 i; struct fc_rport *rport = NULL; struct fc_rport *attach_rport = NULL; struct fc_host_attrs *fc_host = shost_to_fc_host(shost); struct fc_fn_li_desc *li_desc = (struct fc_fn_li_desc *)tlv; u16 event_type = be16_to_cpu(li_desc->event_type); - u64 wwpn; rport = fc_find_rport_by_wwpn(shost, be64_to_cpu(li_desc->attached_wwpn)); @@ -764,22 +793,11 @@ fc_fpin_li_stats_update(struct Scsi_Host *shost, struct fc_tlv_desc *tlv) fc_li_stats_update(event_type, &attach_rport->fpin_stats); } - if (be32_to_cpu(li_desc->pname_count) > 0) { - for (i = 0; - i < be32_to_cpu(li_desc->pname_count); - i++) { - wwpn = be64_to_cpu(li_desc->pname_list[i]); - rport = fc_find_rport_by_wwpn(shost, wwpn); - if (rport && - (rport->roles & FC_PORT_ROLE_FCP_TARGET || - rport->roles & FC_PORT_ROLE_NVME_TARGET)) { - if (rport == attach_rport) - continue; - fc_li_stats_update(event_type, - &rport->fpin_stats); - } - } - } + fc_fpin_pname_stats_update(shost, attach_rport, event_type, + be32_to_cpu(li_desc->desc_len), + FC_TLV_DESC_LENGTH_FROM_SZ(*li_desc), + be32_to_cpu(li_desc->pname_count), + li_desc->pname_list, fc_li_stats_update); if (fc_host->port_name == be64_to_cpu(li_desc->attached_wwpn)) fc_li_stats_update(event_type, &fc_host->fpin_stats); @@ -827,13 +845,11 @@ static void fc_fpin_peer_congn_stats_update(struct Scsi_Host *shost, struct fc_tlv_desc *tlv) { - u8 i; struct fc_rport *rport = NULL; struct fc_rport *attach_rport = NULL; struct fc_fn_peer_congn_desc *pc_desc = (struct fc_fn_peer_congn_desc *)tlv; u16 event_type = be16_to_cpu(pc_desc->event_type); - u64 wwpn; rport = fc_find_rport_by_wwpn(shost, be64_to_cpu(pc_desc->attached_wwpn)); @@ -844,22 +860,11 @@ fc_fpin_peer_congn_stats_update(struct Scsi_Host *shost, fc_cn_stats_update(event_type, &attach_rport->fpin_stats); } - if (be32_to_cpu(pc_desc->pname_count) > 0) { - for (i = 0; - i < be32_to_cpu(pc_desc->pname_count); - i++) { - wwpn = be64_to_cpu(pc_desc->pname_list[i]); - rport = fc_find_rport_by_wwpn(shost, wwpn); - if (rport && - (rport->roles & FC_PORT_ROLE_FCP_TARGET || - rport->roles & FC_PORT_ROLE_NVME_TARGET)) { - if (rport == attach_rport) - continue; - fc_cn_stats_update(event_type, - &rport->fpin_stats); - } - } - } + fc_fpin_pname_stats_update(shost, attach_rport, event_type, + be32_to_cpu(pc_desc->desc_len), + FC_TLV_DESC_LENGTH_FROM_SZ(*pc_desc), + be32_to_cpu(pc_desc->pname_count), + pc_desc->pname_list, fc_cn_stats_update); } /* From 9eed1bd59937e6828b00d2f2dfef631d964f3636 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Mon, 18 May 2026 10:43:07 -0400 Subject: [PATCH 192/411] scsi: fcoe: Reject FIP descriptors with zero fip_dlen in CVL walker drivers/scsi/fcoe/fcoe_ctlr.c::fcoe_ctlr_recv_clr_vlink() advanced the descriptor cursor by an attacker-supplied fip_dlen without ever requiring dlen >= sizeof(struct fip_desc) in the default branch. The named descriptor cases (FIP_DT_MAC, FIP_DT_NAME, FIP_DT_VN_ID) checked their per-type minimum lengths, but a FIP_DT_NON_CRITICAL descriptor (fip_dtype >= 128, which the standard requires receivers to silently ignore) skipped that check entirely. An unauthenticated L2 peer on the FCoE control VLAN could hang fcoe_ctlr_recv_work on an fcoe, qedf, or bnx2fc initiator indefinitely by emitting one FIP CVL frame whose single descriptor had fip_dtype == FIP_DT_NON_CRITICAL and fip_dlen == 0: the cursor advanced zero bytes per iteration and the loop condition rlen >= sizeof(*desc) stayed true forever, blocking every subsequent FIP frame on that controller. Tighten the outer dlen guard to also reject dlen < sizeof(struct fip_desc), so a malformed descriptor whose length cannot even cover the descriptor header is rejected before the switch. This is the same lower-bound the named cases already apply and is the minimum scope that closes the loop. Fixes: 97c8389d54b9 ("[SCSI] fcoe, libfcoe: Add support for FIP. FCoE discovery and keep-alive.") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260518144307.2820961-1-michael.bommarito@gmail.com Signed-off-by: Martin K. Petersen --- drivers/scsi/fcoe/fcoe_ctlr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/scsi/fcoe/fcoe_ctlr.c b/drivers/scsi/fcoe/fcoe_ctlr.c index 02cd4410efca..496ddd45f74d 100644 --- a/drivers/scsi/fcoe/fcoe_ctlr.c +++ b/drivers/scsi/fcoe/fcoe_ctlr.c @@ -1385,7 +1385,7 @@ static void fcoe_ctlr_recv_clr_vlink(struct fcoe_ctlr *fip, while (rlen >= sizeof(*desc)) { dlen = desc->fip_dlen * FIP_BPW; - if (dlen > rlen) + if (dlen < sizeof(*desc) || dlen > rlen) goto err; /* Drop CVL if there are duplicate critical descriptors */ if ((desc->fip_dtype < 32) && From 4157501b9a8ff1bbe32ff5a7d8aece7ab18eff40 Mon Sep 17 00:00:00 2001 From: Stefano Garzarella Date: Thu, 21 May 2026 14:47:32 +0200 Subject: [PATCH 193/411] vsock/virtio: fix skb overhead overflow on 32-bit builds On 32-bit architectures, both skb_queue_len() and SKB_TRUESIZE(0) evaluate to 32-bit values. The multiplication can overflow before being assigned to the u64 skb_overhead variable, making the skb overhead check ineffective. Cast skb_queue_len() to u64 so the multiplication is always performed in 64-bit arithmetic. This issue was reported by Sashiko while reviewing another patch. Fixes: 059b7dbd20a6 ("vsock/virtio: fix potential unbounded skb queue") Closes: https://sashiko.dev/#/patchset/20260518090656.134588-1-sgarzare%40redhat.com Cc: stable@vger.kernel.org Signed-off-by: Stefano Garzarella Acked-by: Michael S. Tsirkin Link: https://patch.msgid.link/20260521124732.125771-1-sgarzare@redhat.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/virtio_transport_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index d4d26fba9e37..b143290a311d 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -417,7 +417,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, static bool virtio_transport_inc_rx_pkt(struct virtio_vsock_sock *vvs, u32 len) { - u64 skb_overhead = (skb_queue_len(&vvs->rx_queue) + 1) * SKB_TRUESIZE(0); + u64 skb_overhead = ((u64)skb_queue_len(&vvs->rx_queue) + 1) * SKB_TRUESIZE(0); /* Allow at most buf_alloc * 2 total budget (payload + overhead), * similar to how SO_RCVBUF is doubled to reserve space for sk_buff From 87a1e0fe7776da7ab411be332b4be58ac8840d10 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Thu, 21 May 2026 12:21:47 +0000 Subject: [PATCH 194/411] ipv4: free net->ipv4.sysctl_local_reserved_ports after unregister_net_sysctl_table() ipv4_sysctl_exit_net() is currently freeing net->ipv4.sysctl_local_reserved_ports too soon. Only after unregister_net_sysctl_table() we can be sure no threads can possibly use the sysctls, including /proc/sys/net/ipv4/ip_local_reserved_ports. Fixes: 122ff243f5f1 ("ipv4: make ip_local_reserved_ports per netns") Reported-by: Ji'an Zhou Signed-off-by: Eric Dumazet Cc: Cong Wang Reviewed-by: Jason Xing Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260521122147.3584624-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/sysctl_net_ipv4.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c index d8bdb1bdbff1..c0e85cc171ae 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -1705,10 +1705,10 @@ static __net_exit void ipv4_sysctl_exit_net(struct net *net) { const struct ctl_table *table; - kfree(net->ipv4.sysctl_local_reserved_ports); table = net->ipv4.ipv4_hdr->ctl_table_arg; unregister_net_sysctl_table(net->ipv4.ipv4_hdr); kfree(table); + kfree(net->ipv4.sysctl_local_reserved_ports); } static __net_initdata struct pernet_operations ipv4_sysctl_ops = { From 778c2ab142c625a8a8afa570e0f9b7873f445d99 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sat, 18 Apr 2026 11:49:27 -0400 Subject: [PATCH 195/411] scsi: target: iscsi: Fix CRC overread and double-free in iscsit_handle_text_cmd() Two latent bugs in the Text-phase handler, both present since the original LIO integration in commit e48354ce078c ("iscsi-target: Add iSCSI fabric support for target v4.1"): 1) DataDigest CRC buffer overread (4 bytes past text_in). text_in is kzalloc()'d at ALIGN(payload_length, 4). rx_size is then incremented by ISCSI_CRC_LEN to make room for the received DataDigest in the iovec, but the same (now-bumped) rx_size is passed as the buffer length to iscsit_crc_buf(): if (conn->conn_ops->DataDigest) { ... rx_size += ISCSI_CRC_LEN; } ... if (conn->conn_ops->DataDigest) { data_crc = iscsit_crc_buf(text_in, rx_size, 0, NULL); iscsit_crc_buf() walks rx_size bytes of text_in with crc32c(), so when DataDigest is negotiated it reads 4 bytes past the end of the text_in allocation. KASAN reproduces this directly on the unpatched mainline tree as slab-out-of-bounds in crc32c() called from the Text PDU path. The OOB bytes feed crc32c() and are then compared against the initiator-supplied checksum, so the value does not flow back to the attacker, but the kernel does read past the buffer on every Text PDU with DataDigest=CRC32C. Fix by passing the actual padded payload length (ALIGN(payload_length, 4)) that was used for the kzalloc(). 2) Stale cmd->text_in_ptr re-free (double-free) on ERL>0 bad DataDigest drop. On DataDigest mismatch with ErrorRecoveryLevel > 0 the handler silently drops the PDU and lets the initiator plug the CmdSN gap: kfree(text_in); return 0; cmd->text_in_ptr still points at the freed buffer. The next Text Request on the same ITT re-enters iscsit_setup_text_cmd(), which unconditionally does kfree(cmd->text_in_ptr); cmd->text_in_ptr = NULL; freeing the same pointer a second time. Session teardown via iscsit_release_cmd() has the same shape and hits the same double-free if the connection is dropped before a second Text Request arrives. On an unmodified mainline tree the bug-1 CRC overread fires first on the initial valid Text Request and perturbs the subsequent state, so #4 was isolated by building a kernel with only the bug-1 hunk of this patch applied plus temporary printk() observability around the three relevant kfree() sites. The observability prints are not part of this patch. On that build, a three-PDU Text Request sequence after login produces two back-to-back splats: BUG: KASAN: double-free in iscsit_setup_text_cmd+0x?? BUG: KASAN: double-free in iscsit_release_cmd+0x?? showing the same pointer freed in the ERL>0 drop path and again in iscsit_setup_text_cmd() (next Text Request on the same ITT) and once more in iscsit_release_cmd() (session teardown). On distro kernels with CONFIG_SLAB_FREELIST_HARDENED=y (default) the double-free becomes a remote kernel BUG(); on non-hardened kernels it corrupts the slab freelist. Fix by clearing cmd->text_in_ptr after the kfree() in the ERL>0 drop path. With both hunks applied #4 is directly observable on the stock tree without observability printks; fixing bug-1 alone would mask #4 less, not more, so the hunks are submitted together. Both fixes are one-liners. The Text PDU state machine is unchanged and the wire protocol is unaffected. Fixes: e48354ce078c ("iscsi-target: Add iSCSI fabric support for target v4.1") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Tested-by: John Garry Reviewed-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/target/iscsi/iscsi_target.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index cb832fd523af..62ada3a52210 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -2295,7 +2295,9 @@ iscsit_handle_text_cmd(struct iscsit_conn *conn, struct iscsit_cmd *cmd, goto reject; if (conn->conn_ops->DataDigest) { - data_crc = iscsit_crc_buf(text_in, rx_size, 0, NULL); + data_crc = iscsit_crc_buf(text_in, + ALIGN(payload_length, 4), + 0, NULL); if (checksum != data_crc) { pr_err("Text data CRC32C DataDigest" " 0x%08x does not match computed" @@ -2314,6 +2316,7 @@ iscsit_handle_text_cmd(struct iscsit_conn *conn, struct iscsit_cmd *cmd, " Command CmdSN: 0x%08x due to" " DataCRC error.\n", hdr->cmdsn); kfree(text_in); + cmd->text_in_ptr = NULL; return 0; } } else { From bf33e01f88388c43e285492a63e539df6ffed64c Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Mon, 11 May 2026 14:49:14 -0400 Subject: [PATCH 196/411] scsi: target: iscsi: Bound iscsi_encode_text_output() appends to rsp_buf iscsi_encode_text_output() concatenates "key=value\0" records into login->rsp_buf, an 8192-byte kzalloc(MAX_KEY_VALUE_PAIRS) buffer allocated in iscsit_alloc_login_setup_buffer(). The three sprintf() call sites in this function (lines 1398, 1411, 1424 in v7.1-rc2) never check the remaining buffer capacity: *length += sprintf(output_buf, "%s=%s", er->key, er->value); *length += 1; output_buf = textbuf + *length; The 8192-byte ceiling at iscsi_target_check_login_request() bounds the *input* Login PDU payload, but a single PDU can carry up to 2048 minimal four-byte "a=b\0" pairs, each unknown key expanding to a 16-byte "a=NotUnderstood\0" output record via iscsi_add_notunderstood_response(). 2048 * 16 = 32 KiB of output into an 8 KiB buffer, producing a ~24 KiB heap overrun in the kmalloc-8k slab. The fix introduces a static iscsi_encode_text_record() helper that uses snprintf() with a per-call bounds check against the remaining buffer, and threads a u32 textbuf_size parameter through iscsi_encode_text_output(). Both call sites in iscsi_target_handle_csg_zero() (PHASE_SECURITY) and iscsi_target_handle_csg_one() (PHASE_OPERATIONAL) pass MAX_KEY_VALUE_PAIRS. On overflow the encoder logs the condition, calls iscsi_release_extra_responses() to drop queued records, and returns -1; both caller sites now emit ISCSI_STATUS_CLS_INITIATOR_ERR / ISCSI_LOGIN_STATUS_INIT_ERR via iscsit_tx_login_rsp() before returning, so the initiator sees an explicit failed-login response rather than a silent connection drop. (Prior to this patch only the PHASE_OPERATIONAL caller did that; the PHASE_SECURITY caller is converted to the same shape.) Fixes: e48354ce078c ("iscsi-target: Add iSCSI fabric support for target v4.1") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Michael Bommarito Tested-by: John Garry Reviewed-by: John Garry Signed-off-by: Martin K. Petersen --- drivers/target/iscsi/iscsi_target_nego.c | 7 ++- .../target/iscsi/iscsi_target_parameters.c | 62 ++++++++++++++----- .../target/iscsi/iscsi_target_parameters.h | 2 +- 3 files changed, 55 insertions(+), 16 deletions(-) diff --git a/drivers/target/iscsi/iscsi_target_nego.c b/drivers/target/iscsi/iscsi_target_nego.c index 832588f21f91..b03ed154ca34 100644 --- a/drivers/target/iscsi/iscsi_target_nego.c +++ b/drivers/target/iscsi/iscsi_target_nego.c @@ -899,10 +899,14 @@ static int iscsi_target_handle_csg_zero( SENDER_TARGET, login->rsp_buf, &login->rsp_length, + MAX_KEY_VALUE_PAIRS, conn->param_list, conn->tpg->tpg_attrib.login_keys_workaround); - if (ret < 0) + if (ret < 0) { + iscsit_tx_login_rsp(conn, ISCSI_STATUS_CLS_INITIATOR_ERR, + ISCSI_LOGIN_STATUS_INIT_ERR); return -1; + } if (!iscsi_check_negotiated_keys(conn->param_list)) { bool auth_required = iscsi_conn_auth_required(conn); @@ -986,6 +990,7 @@ static int iscsi_target_handle_csg_one(struct iscsit_conn *conn, struct iscsi_lo SENDER_TARGET, login->rsp_buf, &login->rsp_length, + MAX_KEY_VALUE_PAIRS, conn->param_list, conn->tpg->tpg_attrib.login_keys_workaround); if (ret < 0) { diff --git a/drivers/target/iscsi/iscsi_target_parameters.c b/drivers/target/iscsi/iscsi_target_parameters.c index 4ed578c7b98d..2b318b13268e 100644 --- a/drivers/target/iscsi/iscsi_target_parameters.c +++ b/drivers/target/iscsi/iscsi_target_parameters.c @@ -1371,19 +1371,42 @@ int iscsi_decode_text_input( return -1; } +/* + * Append "key=value" plus a trailing NUL into @textbuf at *@length. + * Returns 0 on success and advances *@length, or -EMSGSIZE if the + * record (including the NUL) would not fit in the remaining buffer. + */ +static int iscsi_encode_text_record(char *textbuf, u32 *length, + u32 textbuf_size, + const char *key, const char *value) +{ + int n; + u32 avail; + + if (*length >= textbuf_size) + return -EMSGSIZE; + + avail = textbuf_size - *length; + n = snprintf(textbuf + *length, avail, "%s=%s", key, value); + if (n < 0 || (u32)n + 1 > avail) + return -EMSGSIZE; + + *length += n + 1; + return 0; +} + int iscsi_encode_text_output( u8 phase, u8 sender, char *textbuf, u32 *length, + u32 textbuf_size, struct iscsi_param_list *param_list, bool keys_workaround) { - char *output_buf = NULL; struct iscsi_extra_response *er; struct iscsi_param *param; - - output_buf = textbuf + *length; + int ret; if (iscsi_enforce_integrity_rules(phase, param_list) < 0) return -1; @@ -1395,10 +1418,12 @@ int iscsi_encode_text_output( !IS_PSTATE_RESPONSE_SENT(param) && !IS_PSTATE_REPLY_OPTIONAL(param) && (param->phase & phase)) { - *length += sprintf(output_buf, "%s=%s", - param->name, param->value); - *length += 1; - output_buf = textbuf + *length; + ret = iscsi_encode_text_record(textbuf, length, + textbuf_size, + param->name, + param->value); + if (ret < 0) + goto err_overflow; SET_PSTATE_RESPONSE_SENT(param); pr_debug("Sending key: %s=%s\n", param->name, param->value); @@ -1408,10 +1433,12 @@ int iscsi_encode_text_output( !IS_PSTATE_ACCEPTOR(param) && !IS_PSTATE_PROPOSER(param) && (param->phase & phase)) { - *length += sprintf(output_buf, "%s=%s", - param->name, param->value); - *length += 1; - output_buf = textbuf + *length; + ret = iscsi_encode_text_record(textbuf, length, + textbuf_size, + param->name, + param->value); + if (ret < 0) + goto err_overflow; SET_PSTATE_PROPOSER(param); iscsi_check_proposer_for_optional_reply(param, keys_workaround); @@ -1421,14 +1448,21 @@ int iscsi_encode_text_output( } list_for_each_entry(er, ¶m_list->extra_response_list, er_list) { - *length += sprintf(output_buf, "%s=%s", er->key, er->value); - *length += 1; - output_buf = textbuf + *length; + ret = iscsi_encode_text_record(textbuf, length, textbuf_size, + er->key, er->value); + if (ret < 0) + goto err_overflow; pr_debug("Sending key: %s=%s\n", er->key, er->value); } iscsi_release_extra_responses(param_list); return 0; + +err_overflow: + pr_err("iSCSI login response buffer (%u bytes) exhausted, dropping login.\n", + textbuf_size); + iscsi_release_extra_responses(param_list); + return -1; } int iscsi_check_negotiated_keys(struct iscsi_param_list *param_list) diff --git a/drivers/target/iscsi/iscsi_target_parameters.h b/drivers/target/iscsi/iscsi_target_parameters.h index c672a971fcb7..38d2238dfe08 100644 --- a/drivers/target/iscsi/iscsi_target_parameters.h +++ b/drivers/target/iscsi/iscsi_target_parameters.h @@ -43,7 +43,7 @@ extern struct iscsi_param *iscsi_find_param_from_key(char *, struct iscsi_param_ extern int iscsi_extract_key_value(char *, char **, char **); extern int iscsi_update_param_value(struct iscsi_param *, char *); extern int iscsi_decode_text_input(u8, u8, char *, u32, struct iscsit_conn *); -extern int iscsi_encode_text_output(u8, u8, char *, u32 *, +extern int iscsi_encode_text_output(u8, u8, char *, u32 *, u32, struct iscsi_param_list *, bool); extern int iscsi_check_negotiated_keys(struct iscsi_param_list *); extern void iscsi_set_connection_parameters(struct iscsi_conn_ops *, From 85db7391310b1304d2dc8ae3b0b12105a9567147 Mon Sep 17 00:00:00 2001 From: Alexandru Hossu Date: Thu, 21 May 2026 17:11:21 +0200 Subject: [PATCH 197/411] scsi: target: iscsi: Validate CHAP_R length before base64 decode chap_server_compute_hash() allocates client_digest as kzalloc(chap->digest_size) and then, for BASE64-encoded responses, passes chap_r directly to chap_base64_decode() without checking whether the input length could produce more than digest_size bytes of output. chap_base64_decode() writes to the destination unconditionally as long as there is input to consume. With MAX_RESPONSE_LENGTH set to 128 and the "0b" prefix stripped by extract_param(), up to 127 base64 characters can reach the decoder. 127 characters decode to 95 bytes. For SHA-256 (digest_size=32) this overflows client_digest by 63 bytes; for MD5 (digest_size=16) the overflow is 79 bytes. The length check at line 344 fires after the write has already happened. The HEX branch in the same switch statement already validates the length up front. Apply the same approach to the BASE64 branch: strip trailing base64 padding characters, then reject any input whose data length exceeds DIV_ROUND_UP(digest_size * 4, 3) before calling the decoder. Stripping trailing '=' before the comparison handles both padded and unpadded encodings. chap_base64_decode() already returns early on '=', so the full original string is still passed to the decoder unchanged. The mutual CHAP path decodes CHAP_C into initiatorchg_binhex, which is kzalloc(CHAP_CHALLENGE_STR_LEN). extract_param() caps initiatorchg at CHAP_CHALLENGE_STR_LEN characters, so at most CHAP_CHALLENGE_STR_LEN-1 base64 characters reach the decoder. The maximum decoded size, DIV_ROUND_UP((CHAP_CHALLENGE_STR_LEN-1) * 3, 4), is less than CHAP_CHALLENGE_STR_LEN, so no overflow is possible there. A comment is added at the call site to document this. Fixes: 1e5733883421 ("scsi: target: iscsi: Support base64 in CHAP") Cc: stable@vger.kernel.org Signed-off-by: Alexandru Hossu Reviewed-by: David Disseldorp Link: https://patch.msgid.link/20260521151121.808477-1-hossu.alexandru@gmail.com Signed-off-by: Martin K. Petersen --- drivers/target/iscsi/iscsi_target_auth.c | 19 ++++++++++++++++++- 1 file changed, 18 insertions(+), 1 deletion(-) diff --git a/drivers/target/iscsi/iscsi_target_auth.c b/drivers/target/iscsi/iscsi_target_auth.c index c46c69a28e97..a3ad2d244dbe 100644 --- a/drivers/target/iscsi/iscsi_target_auth.c +++ b/drivers/target/iscsi/iscsi_target_auth.c @@ -340,13 +340,22 @@ static int chap_server_compute_hash( goto out; } break; - case BASE64: + case BASE64: { + size_t r_len = strlen(chap_r); + + while (r_len > 0 && chap_r[r_len - 1] == '=') + r_len--; + if (r_len > DIV_ROUND_UP(chap->digest_size * 4, 3)) { + pr_err("Malformed CHAP_R: base64 payload too long\n"); + goto out; + } if (chap_base64_decode(client_digest, chap_r, strlen(chap_r)) != chap->digest_size) { pr_err("Malformed CHAP_R: invalid BASE64\n"); goto out; } break; + } default: pr_err("Could not find CHAP_R\n"); goto out; @@ -473,6 +482,14 @@ static int chap_server_compute_hash( } break; case BASE64: + /* + * No overflow check needed: initiatorchg_binhex is + * CHAP_CHALLENGE_STR_LEN bytes and extract_param() caps + * initiatorchg at CHAP_CHALLENGE_STR_LEN characters, so + * the decoded output is at most DIV_ROUND_UP( + * (CHAP_CHALLENGE_STR_LEN - 1) * 3, 4) bytes, which is + * less than CHAP_CHALLENGE_STR_LEN. + */ initiatorchg_len = chap_base64_decode(initiatorchg_binhex, initiatorchg, strlen(initiatorchg)); From 4085f0dbb1ce2251c9a5938d693de6593f0ab2bd Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Fri, 22 May 2026 16:19:50 +0200 Subject: [PATCH 198/411] USB: serial: mxuport: fix memory corruption with small endpoint Make sure that the bulk-out endpoint max packet size is at least eight bytes to avoid user-controlled slab corruption should a malicious device report a smaller size. Fixes: ee467a1f2066 ("USB: serial: add Moxa UPORT 12XX/14XX/16XX driver") Cc: stable@vger.kernel.org # 3.14 Cc: Andrew Lunn Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/mxuport.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/usb/serial/mxuport.c b/drivers/usb/serial/mxuport.c index ad5fdf55a02e..c9b9928c473a 100644 --- a/drivers/usb/serial/mxuport.c +++ b/drivers/usb/serial/mxuport.c @@ -962,6 +962,14 @@ static int mxuport_calc_num_ports(struct usb_serial *serial, */ BUILD_BUG_ON(ARRAY_SIZE(epds->bulk_out) < 16); + /* + * The bulk-out buffers must be large enough for the four-byte header + * (and following data), but assume anything smaller than eight bytes + * is broken. + */ + if (usb_endpoint_maxp(epds->bulk_out[0]) < 8) + return -EINVAL; + for (i = 1; i < num_ports; ++i) epds->bulk_out[i] = epds->bulk_out[0]; From 60df93d30f9bdd27db17c4d80ed80ef718d7226b Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Fri, 22 May 2026 16:20:58 +0200 Subject: [PATCH 199/411] USB: serial: omninet: fix memory corruption with small endpoint Make sure that the bulk-out buffers are at least as large as the hardcoded transfer size to avoid user-controlled slab corruption should a malicious device report a smaller endpoint max packet size than expected. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/omninet.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/usb/serial/omninet.c b/drivers/usb/serial/omninet.c index aa1e9745f967..b59982ed8b25 100644 --- a/drivers/usb/serial/omninet.c +++ b/drivers/usb/serial/omninet.c @@ -30,6 +30,10 @@ /* This one seems to be a re-branded ZyXEL device */ #define BT_IGNITIONPRO_ID 0x2000 +#define OMNINET_HEADERLEN 4 +#define OMNINET_BULKOUTSIZE 64 +#define OMNINET_PAYLOADSIZE (OMNINET_BULKOUTSIZE - OMNINET_HEADERLEN) + /* function prototypes */ static void omninet_process_read_urb(struct urb *urb); static int omninet_prepare_write_buffer(struct usb_serial_port *port, @@ -54,6 +58,7 @@ static struct usb_serial_driver zyxel_omninet_device = { .description = "ZyXEL - omni.net usb", .id_table = id_table, .num_bulk_out = 2, + .bulk_out_size = OMNINET_BULKOUTSIZE, .calc_num_ports = omninet_calc_num_ports, .port_probe = omninet_port_probe, .port_remove = omninet_port_remove, @@ -130,10 +135,6 @@ static void omninet_port_remove(struct usb_serial_port *port) kfree(od); } -#define OMNINET_HEADERLEN 4 -#define OMNINET_BULKOUTSIZE 64 -#define OMNINET_PAYLOADSIZE (OMNINET_BULKOUTSIZE - OMNINET_HEADERLEN) - static void omninet_process_read_urb(struct urb *urb) { struct usb_serial_port *port = urb->context; From 438061ed1ad85e6743e2dce826671772d81089ec Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Fri, 22 May 2026 16:22:18 +0200 Subject: [PATCH 200/411] USB: serial: safe_serial: fix memory corruption with small endpoint Make sure that the bulk-out buffer size is at least eight bytes to avoid user-controlled slab corruption in "safe" mode should a malicious device report a smaller size. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Reviewed-by: Greg Kroah-Hartman Signed-off-by: Johan Hovold --- drivers/usb/serial/safe_serial.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/usb/serial/safe_serial.c b/drivers/usb/serial/safe_serial.c index 238b54993446..d267a31dcccf 100644 --- a/drivers/usb/serial/safe_serial.c +++ b/drivers/usb/serial/safe_serial.c @@ -259,6 +259,7 @@ static int safe_prepare_write_buffer(struct usb_serial_port *port, static int safe_startup(struct usb_serial *serial) { struct usb_interface_descriptor *desc; + int bulk_out_size; if (serial->dev->descriptor.bDeviceClass != CDC_DEVICE_CLASS) return -ENODEV; @@ -279,6 +280,16 @@ static int safe_startup(struct usb_serial *serial) default: return -EINVAL; } + + /* + * The bulk-out buffer needs to be large enough for the two-byte + * trailer in safe mode, but assume anything smaller than eight bytes + * is broken. + */ + bulk_out_size = serial->port[0]->bulk_out_size; + if (bulk_out_size > 0 && bulk_out_size < 8) + return -EINVAL; + return 0; } From 9f9bfc80c67f35a275820da7e83a35dface08281 Mon Sep 17 00:00:00 2001 From: Zhang Cen Date: Fri, 22 May 2026 22:54:42 +0800 Subject: [PATCH 201/411] USB: serial: cypress_m8: validate interrupt packet headers cypress_read_int_callback() parses the interrupt-in buffer according to the selected Cypress packet format. Format 1 has a two-byte status/count header and format 2 has a one-byte combined status/count header. The usb-serial core sizes the interrupt-in buffer from the endpoint descriptor's wMaxPacketSize, and successful interrupt transfers can complete short when URB_SHORT_NOT_OK is not set. Check that the completed packet contains the selected header before reading it. Malformed short reports are ignored and the interrupt URB is resubmitted through the existing retry path, preventing out-of-bounds header-byte reads. KASAN report as below: KASAN slab-out-of-bounds in cypress_read_int_callback+0x240/0x7f0 Read of size 1 Call trace: cypress_read_int_callback() (drivers/usb/serial/cypress_m8.c:1009) __usb_hcd_giveback_urb() dummy_timer() Fixes: 3416eaa1f8f8 ("USB: cypress_m8: Packet format is separate from characteristic size") Assisted-by: Codex:gpt-5.5 Signed-off-by: Zhang Cen Fixes: 3416eaa1f8f8 ("USB: cypress_m8: Packet format is separate from characteristic size") Cc: stable@vger.kernel.org # 2.6.26 [ johan: use constants in header length sanity checks ] Signed-off-by: Johan Hovold --- drivers/usb/serial/cypress_m8.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/drivers/usb/serial/cypress_m8.c b/drivers/usb/serial/cypress_m8.c index 0b8a4e9d7bc5..bcf302e88ca4 100644 --- a/drivers/usb/serial/cypress_m8.c +++ b/drivers/usb/serial/cypress_m8.c @@ -1025,8 +1025,8 @@ static void cypress_read_int_callback(struct urb *urb) char tty_flag = TTY_NORMAL; int bytes = 0; int result; - int i = 0; int status = urb->status; + int i; switch (status) { case 0: /* success */ @@ -1064,22 +1064,32 @@ static void cypress_read_int_callback(struct urb *urb) spin_lock_irqsave(&priv->lock, flags); result = urb->actual_length; + i = 0; switch (priv->pkt_fmt) { default: case packet_format_1: /* This is for the CY7C64013... */ + if (result < 2) + break; priv->current_status = data[0] & 0xF8; bytes = data[1] + 2; i = 2; break; case packet_format_2: /* This is for the CY7C63743... */ + if (result < 1) + break; priv->current_status = data[0] & 0xF8; bytes = (data[0] & 0x07) + 1; i = 1; break; } spin_unlock_irqrestore(&priv->lock, flags); + if (i == 0) { + dev_dbg(dev, "%s - short packet received: %d bytes\n", + __func__, result); + goto continue_read; + } if (result < bytes) { dev_dbg(dev, "%s - wrong packet size - received %d bytes but packet said %d bytes\n", From a88c02915d9c6160cfc7ab1b26ed64b2993e2b94 Mon Sep 17 00:00:00 2001 From: Lim HyeonJun Date: Sun, 24 May 2026 20:08:53 +0900 Subject: [PATCH 202/411] io_uring/tctx: set ->io_uring before publishing the tctx node io_register_iowq_max_workers() walks ctx->tctx_list under ctx->tctx_lock and dereferences each node's task->io_uring without a NULL check: list_for_each_entry(node, &ctx->tctx_list, ctx_node) { tctx = node->task->io_uring; if (WARN_ON_ONCE(!tctx->io_wq)) continue; ... } __io_uring_add_tctx_node() installs the node into ctx->tctx_list (via io_tctx_install_node(), which does the list_add() under tctx_lock) and only assigns current->io_uring = tctx afterwards. A task doing its first io_uring operation on a shared ring therefore has a window in which its node is already visible on ctx->tctx_list while node->task->io_uring is still NULL. A concurrent IORING_REGISTER_IOWQ_MAX_WORKERS on the same ring reads that NULL and dereferences tctx->io_wq: KASAN: null-ptr-deref in range [0x0000000000000018-0x000000000000001f] RIP: io_register_iowq_max_workers io_uring/register.c:423 Publish current->io_uring = tctx before installing the node, so any node visible on ctx->tctx_list always has a valid task->io_uring. Fixes: 7880174e1e5e ("io_uring/tctx: clean up __io_uring_add_tctx_node() error handling") Signed-off-by: Lim HyeonJun Link: https://patch.msgid.link/20260524110853.115634-1-shja0831@gmail.com Signed-off-by: Jens Axboe --- io_uring/tctx.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/io_uring/tctx.c b/io_uring/tctx.c index 6af62ca9baba..42b219b34aa8 100644 --- a/io_uring/tctx.c +++ b/io_uring/tctx.c @@ -139,12 +139,14 @@ static int io_tctx_install_node(struct io_ring_ctx *ctx, int __io_uring_add_tctx_node(struct io_ring_ctx *ctx) { struct io_uring_task *tctx = current->io_uring; + bool new_tctx = false; int ret; if (unlikely(!tctx)) { tctx = io_uring_alloc_task_context(current, ctx); if (IS_ERR(tctx)) return PTR_ERR(tctx); + new_tctx = true; if (data_race(ctx->int_flags) & IO_RING_F_IOWQ_LIMITS_SET) { unsigned int limits[2]; @@ -168,13 +170,15 @@ int __io_uring_add_tctx_node(struct io_ring_ctx *ctx) if (tctx->io_wq) io_wq_set_exit_on_idle(tctx->io_wq, false); - ret = io_tctx_install_node(ctx, tctx); - if (!ret) { + if (new_tctx) current->io_uring = tctx; + + ret = io_tctx_install_node(ctx, tctx); + if (!ret) return 0; - } - if (!current->io_uring) { err_free: + if (new_tctx) { + current->io_uring = NULL; if (tctx->io_wq) { io_wq_exit_start(tctx->io_wq); io_wq_put_and_exit(tctx->io_wq); From 98fb1c1bb11e29eb609b7200a25e136e05aa4498 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A1ssio=20Gabriel?= Date: Thu, 21 May 2026 08:01:23 -0300 Subject: [PATCH 203/411] ALSA: firewire-motu: Protect register DSP event queue positions MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The register DSP event queue is updated under parser->lock, but snd_motu_register_dsp_message_parser_count_event() reads pull_pos and push_pos without the lock. snd_motu_register_dsp_message_parser_copy_event() also reads both queue positions before taking the lock. Protect these accesses with parser->lock as well. This keeps the hwdep poll/read path consistent with the producer side and with the cached meter/parameter accessors. Fixes: 634ec0b2906e ("ALSA: firewire-motu: notify event for parameter change in register DSP model") Cc: stable@vger.kernel.org Signed-off-by: Cássio Gabriel Reviewed-by: Takashi Sakamoto Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20260521-alsa-firewire-motu-event-locking-v1-1-708e1c2b5e56@gmail.com --- .../firewire/motu/motu-register-dsp-message-parser.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/sound/firewire/motu/motu-register-dsp-message-parser.c b/sound/firewire/motu/motu-register-dsp-message-parser.c index a8053e3ef065..4ec23e6880d9 100644 --- a/sound/firewire/motu/motu-register-dsp-message-parser.c +++ b/sound/firewire/motu/motu-register-dsp-message-parser.c @@ -386,6 +386,8 @@ unsigned int snd_motu_register_dsp_message_parser_count_event(struct snd_motu *m { struct msg_parser *parser = motu->message_parser; + guard(spinlock_irqsave)(&parser->lock); + if (parser->pull_pos > parser->push_pos) return EVENT_QUEUE_SIZE - parser->pull_pos + parser->push_pos; else @@ -395,13 +397,14 @@ unsigned int snd_motu_register_dsp_message_parser_count_event(struct snd_motu *m bool snd_motu_register_dsp_message_parser_copy_event(struct snd_motu *motu, u32 *event) { struct msg_parser *parser = motu->message_parser; - unsigned int pos = parser->pull_pos; - - if (pos == parser->push_pos) - return false; + unsigned int pos; guard(spinlock_irqsave)(&parser->lock); + if (parser->pull_pos == parser->push_pos) + return false; + + pos = parser->pull_pos; *event = parser->event_queue[pos]; ++pos; From 4e273bca232ec71bbf072f10f7d395a28e85e4e1 Mon Sep 17 00:00:00 2001 From: Zhang Heng Date: Fri, 22 May 2026 14:07:42 +0800 Subject: [PATCH 204/411] ALSA: hda/realtek: Fix incorrect comment for ALC299_FIXUP_PREDATOR_SPK The comment for the pin configuration 0x21 in the fixup ALC299_FIXUP_PREDATOR_SPK states "use as headset mic, without its own jack detect", but the fixup name and the actual usage indicate that the pin is meant to be used as internal speaker. Correct the comment to avoid confusion. Signed-off-by: Zhang Heng Link: https://patch.msgid.link/20260522060742.1384390-1-zhangheng@kylinos.cn Signed-off-by: Takashi Iwai --- sound/hda/codecs/realtek/alc269.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/hda/codecs/realtek/alc269.c b/sound/hda/codecs/realtek/alc269.c index f180d6a72021..53585e20c494 100644 --- a/sound/hda/codecs/realtek/alc269.c +++ b/sound/hda/codecs/realtek/alc269.c @@ -5458,7 +5458,7 @@ static const struct hda_fixup alc269_fixups[] = { [ALC299_FIXUP_PREDATOR_SPK] = { .type = HDA_FIXUP_PINS, .v.pins = (const struct hda_pintbl[]) { - { 0x21, 0x90170150 }, /* use as headset mic, without its own jack detect */ + { 0x21, 0x90170150 }, /* use as internal speaker */ { } } }, From f7b1f71566ff9d5344a516fa745118cee924c8d0 Mon Sep 17 00:00:00 2001 From: Kris Kater Date: Fri, 22 May 2026 08:09:02 +0200 Subject: [PATCH 205/411] ALSA: hda/realtek: Add HDA_CODEC_QUIRK for Lenovo Yoga Slim 7 14AGP11 The BIOS on the Lenovo Yoga Slim 7 14AGP11 (AMD Ryzen AI / Kraken Point chassis; board LNVNB161216, product 83QS) programs the PCI subsystem ID of the HDA function as 17aa:0000. As a result no entry in alc269_fixup_tbl[] matches via SND_PCI_QUIRK, the fixup falls back to the generic auto-routing path, and the bass speaker pin is left mis-routed. Laptop speakers sound noticeably thin. The codec's own internal subsystem ID register reports 0x17aa394c correctly, so an HDA_CODEC_QUIRK entry (which matches on the codec SSID rather than on the PCI SSID) binds the chassis to the existing ALC287_FIXUP_YOGA9_14IAP7_BASS_SPK_PIN fixup. This mirrors the same workaround already in place for the closely-related Yoga 7 2-in-1 14AKP10 and 16AKP10 entries earlier in the table. With this change the kernel log goes from ALC287: picked fixup for PCI SSID 17aa:0000 to ALC287: picked fixup alc287-yoga9-bass-spk-pin and speaker routing matches what the firmware intended. Verified by the reporter against the equivalent modprobe override (model=,alc287-yoga9-bass-spk-pin). Link: https://bugzilla.kernel.org/show_bug.cgi?id=221438 Signed-off-by: Kris Kater Link: https://patch.msgid.link/20260522060902.9423-1-kris@kater.nu Signed-off-by: Takashi Iwai --- sound/hda/codecs/realtek/alc269.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/hda/codecs/realtek/alc269.c b/sound/hda/codecs/realtek/alc269.c index 53585e20c494..9cc4463af05e 100644 --- a/sound/hda/codecs/realtek/alc269.c +++ b/sound/hda/codecs/realtek/alc269.c @@ -7778,6 +7778,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x17aa, 0x3920, "Yoga S990-16 pro Quad VECO Quad", ALC287_FIXUP_TXNW2781_I2C), SND_PCI_QUIRK(0x17aa, 0x3929, "Thinkbook 13x Gen 5", ALC287_FIXUP_MG_RTKC_CSAMP_CS35L41_I2C_THINKPAD), SND_PCI_QUIRK(0x17aa, 0x392b, "Thinkbook 13x Gen 5", ALC287_FIXUP_MG_RTKC_CSAMP_CS35L41_I2C_THINKPAD), + HDA_CODEC_QUIRK(0x17aa, 0x394c, "Lenovo Yoga Slim 7 14AGP11", ALC287_FIXUP_YOGA9_14IAP7_BASS_SPK_PIN), SND_PCI_QUIRK(0x17aa, 0x3977, "IdeaPad S210", ALC283_FIXUP_INT_MIC), SND_PCI_QUIRK(0x17aa, 0x3978, "Lenovo B50-70", ALC269_FIXUP_DMIC_THINKPAD_ACPI), SND_PCI_QUIRK(0x17aa, 0x3bf8, "Quanta FL1", ALC269_FIXUP_PCM_44K), From a0d9e8df2ebca290c2efff70abc05426e5a476b0 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A1ssio=20Gabriel?= Date: Fri, 22 May 2026 09:49:30 -0300 Subject: [PATCH 206/411] ALSA: hda: cs35l56: Fix system name string leaks MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit cs35l56_hda_read_acpi() gets an allocated ACPI _SUB string from acpi_get_subsystem_id(). On success, that string is used to create the firmware system name. Several error paths after the _SUB lookup can return without releasing the allocated string. This includes speaker ID lookup errors other than -ENOENT, and errors after a firmware system name has been allocated. Use scoped cleanup for the temporary _SUB string and make cs35l56->system_name device-managed. This releases the temporary _SUB string on every error path and lets devres release the firmware system name on probe failure and device removal. Fixes: 6f03b446cbae ("ALSA: hda: cs35l56: Add support for speaker id") Fixes: 40b1c2f9b299 ("ALSA: hda/cs35l56: Workaround bad dev-index on Lenovo Yoga Book 9i GenX") Signed-off-by: Cássio Gabriel Reviewed-by: Richard Fitzgerald Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/20260522-alsa-cs35l56-system-name-leak-v4-1-a6154dd09cd9@gmail.com --- sound/hda/codecs/side-codecs/cs35l56_hda.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/sound/hda/codecs/side-codecs/cs35l56_hda.c b/sound/hda/codecs/side-codecs/cs35l56_hda.c index cdbc576569ef..a0ea08eb96a9 100644 --- a/sound/hda/codecs/side-codecs/cs35l56_hda.c +++ b/sound/hda/codecs/side-codecs/cs35l56_hda.c @@ -1025,7 +1025,7 @@ static int cs35l56_hda_read_acpi(struct cs35l56_hda *cs35l56, int hid, int id) u32 values[HDA_MAX_COMPONENTS]; char hid_string[8]; struct acpi_device *adev; - const char *property, *sub; + const char *property; int i, ret; /* @@ -1047,7 +1047,8 @@ static int cs35l56_hda_read_acpi(struct cs35l56_hda *cs35l56, int hid, int id) /* Initialize things that could be overwritten by a fixup */ cs35l56->index = -1; - sub = acpi_get_subsystem_id(ACPI_HANDLE(cs35l56->base.dev)); + const char *sub __free(kfree) = acpi_get_subsystem_id(ACPI_HANDLE(cs35l56->base.dev)); + ret = cs35l56_hda_apply_platform_fixups(cs35l56, sub, &id); if (ret) return ret; @@ -1095,15 +1096,16 @@ static int cs35l56_hda_read_acpi(struct cs35l56_hda *cs35l56, int hid, int id) ret = cirrus_scodec_get_speaker_id(cs35l56->base.dev, cs35l56->index, cs35l56->num_amps, -1); if (ret == -ENOENT) { - cs35l56->system_name = sub; + cs35l56->system_name = devm_kstrdup(cs35l56->base.dev, sub, GFP_KERNEL); } else if (ret >= 0) { - cs35l56->system_name = kasprintf(GFP_KERNEL, "%s-spkid%d", sub, ret); - kfree(sub); - if (!cs35l56->system_name) - return -ENOMEM; + cs35l56->system_name = devm_kasprintf(cs35l56->base.dev, GFP_KERNEL, + "%s-spkid%d", sub, ret); } else { return ret; } + + if (!cs35l56->system_name) + return -ENOMEM; } cs35l56->base.reset_gpio = devm_gpiod_get_index_optional(cs35l56->base.dev, @@ -1254,7 +1256,6 @@ void cs35l56_hda_remove(struct device *dev) cs_dsp_remove(&cs35l56->cs_dsp); - kfree(cs35l56->system_name); pm_runtime_put_noidle(cs35l56->base.dev); gpiod_set_value_cansleep(cs35l56->base.reset_gpio, 0); From 4cc54bdd54b337e77115be5b55577d1c58608eae Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A1ssio=20Gabriel?= Date: Fri, 22 May 2026 22:09:40 -0300 Subject: [PATCH 207/411] ALSA: pcm: oss: Fix setup list UAF on proc write error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit snd_pcm_oss_proc_write() links a newly allocated setup entry into the OSS setup list before duplicating the task name. If the task-name allocation fails, the error path frees the already linked entry and leaves setup_list pointing at freed memory. A later OSS device open can then walk the stale list entry in snd_pcm_oss_look_for_setup() and dereference freed memory. Allocate the task name and initialize the setup entry before publishing the entry on setup_list. Also fetch the initial proc read iterator only after taking setup_mutex, so all setup_list traversal follows the same list lifetime rules. Reported-by: syzbot+8e498074a794999eb41c@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/6a1062b7.170a0220.35b2b7.0003.GAE@google.com Closes: https://syzkaller.appspot.com/bug?extid=8e498074a794999eb41c Fixes: 060d77b9c04a ("[ALSA] Fix / clean up PCM-OSS setup hooks") Signed-off-by: Cássio Gabriel Link: https://patch.msgid.link/20260522-alsa-pcm-oss-setup-uaf-v1-1-40bdcc4d17e8@gmail.com Signed-off-by: Takashi Iwai --- sound/core/oss/pcm_oss.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c index 33fd34f0d615..746eaf93e1a5 100644 --- a/sound/core/oss/pcm_oss.c +++ b/sound/core/oss/pcm_oss.c @@ -2974,8 +2974,10 @@ static void snd_pcm_oss_proc_read(struct snd_info_entry *entry, struct snd_info_buffer *buffer) { struct snd_pcm_str *pstr = entry->private_data; - struct snd_pcm_oss_setup *setup = pstr->oss.setup_list; + struct snd_pcm_oss_setup *setup; + guard(mutex)(&pstr->oss.setup_mutex); + setup = pstr->oss.setup_list; while (setup) { snd_iprintf(buffer, "%s %u %u%s%s%s%s%s%s\n", setup->task_name, @@ -3060,6 +3062,13 @@ static void snd_pcm_oss_proc_write(struct snd_info_entry *entry, buffer->error = -ENOMEM; return; } + template.task_name = kstrdup(task_name, GFP_KERNEL); + if (!template.task_name) { + kfree(setup); + buffer->error = -ENOMEM; + return; + } + *setup = template; if (pstr->oss.setup_list == NULL) pstr->oss.setup_list = setup; else { @@ -3067,12 +3076,7 @@ static void snd_pcm_oss_proc_write(struct snd_info_entry *entry, setup1->next; setup1 = setup1->next); setup1->next = setup; } - template.task_name = kstrdup(task_name, GFP_KERNEL); - if (! template.task_name) { - kfree(setup); - buffer->error = -ENOMEM; - return; - } + continue; } *setup = template; } From db37cf47b67e38ade40de5cd74a4d4d772ff1416 Mon Sep 17 00:00:00 2001 From: "Geoffrey D. Bennett" Date: Sun, 24 May 2026 06:34:14 +0930 Subject: [PATCH 208/411] ALSA: scarlett2: Fix 2i2 Gen 4 direct monitor gain on firmware 2417 Firmware 2417 for the Scarlett 4th Gen 2i2 moved the direct monitor gain parameter by 4 bytes, from offset 0x2a0 to 0x2a4, breaking the "Direct Monitor X Mix Y" controls. Special-case the offset in the get/set config helpers when the running firmware is 2417 or later. Fixes: 4e809a299677 ("ALSA: scarlett2: Add support for Solo, 2i2, and 4i4 Gen 4") Cc: Signed-off-by: Geoffrey D. Bennett Link: https://patch.msgid.link/ahIWTueUlWA5xiV+@m.b4.vu Signed-off-by: Takashi Iwai --- sound/usb/mixer_scarlett2.c | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) diff --git a/sound/usb/mixer_scarlett2.c b/sound/usb/mixer_scarlett2.c index 8e80a7165faf..a4fac4652201 100644 --- a/sound/usb/mixer_scarlett2.c +++ b/sound/usb/mixer_scarlett2.c @@ -2504,6 +2504,27 @@ static int scarlett2_has_config_item( return !!private->config_set->items[config_item_num].offset; } +/* Return the configuration item's offset, applying any per-firmware + * overrides. + * + * Firmware 2417 for the 2i2 Gen 4 moved DIRECT_MONITOR_GAIN by 4 + * bytes. Apply that shift here so that the rest of the driver can + * keep using the single config set. This override can be removed + * once the multi-config-set framework lands. + */ +static int scarlett2_config_item_offset( + struct scarlett2_data *private, int config_item_num) +{ + int offset = private->config_set->items[config_item_num].offset; + + if (config_item_num == SCARLETT2_CONFIG_DIRECT_MONITOR_GAIN && + private->info == &s2i2_gen4_info && + private->firmware_version >= 2417) + offset = 0x2a4; + + return offset; +} + /* Send a USB message to get configuration parameters; result placed in *buf */ static int scarlett2_usb_get_config( struct usb_mixer_interface *mixer, @@ -2513,6 +2534,7 @@ static int scarlett2_usb_get_config( const struct scarlett2_config *config_item = &private->config_set->items[config_item_num]; int size, err, i; + int item_offset; u8 *buf_8; u8 value; @@ -2522,13 +2544,15 @@ static int scarlett2_usb_get_config( if (!config_item->offset) return -EFAULT; + item_offset = scarlett2_config_item_offset(private, config_item_num); + /* Writes to the parameter buffer are always 1 byte */ size = config_item->size ? config_item->size : 8; /* For byte-sized parameters, retrieve directly into buf */ if (size >= 8) { size = size / 8 * count; - err = scarlett2_usb_get(mixer, config_item->offset, buf, size); + err = scarlett2_usb_get(mixer, item_offset, buf, size); if (err < 0) return err; if (config_item->size == 16) { @@ -2546,7 +2570,7 @@ static int scarlett2_usb_get_config( } /* For bit-sized parameters, retrieve into value */ - err = scarlett2_usb_get(mixer, config_item->offset, &value, 1); + err = scarlett2_usb_get(mixer, item_offset, &value, 1); if (err < 0) return err; @@ -2696,7 +2720,8 @@ static int scarlett2_usb_set_config( */ if (config_item->size >= 8) { size = config_item->size / 8; - offset = config_item->offset + index * size; + offset = scarlett2_config_item_offset(private, config_item_num) + + index * size; /* If updating a bit, retrieve the old value, set/clear the * bit as needed, and update value @@ -2705,7 +2730,7 @@ static int scarlett2_usb_set_config( u8 tmp; size = 1; - offset = config_item->offset; + offset = scarlett2_config_item_offset(private, config_item_num); err = scarlett2_usb_get(mixer, offset, &tmp, 1); if (err < 0) From 7740c6cedb2599d8e4534574a0a72042ba63c1c7 Mon Sep 17 00:00:00 2001 From: Edson Juliano Drosdeck Date: Sun, 24 May 2026 15:53:24 -0300 Subject: [PATCH 209/411] ALSA: hda/realtek: Limit mic boost on Positivo DN140 The internal mic boost on the Positivo DN140 is too high. Fix this by applying the ALC269_FIXUP_LIMIT_INT_MIC_BOOST fixup to the machine to limit the gain. Signed-off-by: Edson Juliano Drosdeck Link: https://patch.msgid.link/20260524185324.28959-1-edson.drosdeck@gmail.com Signed-off-by: Takashi Iwai --- sound/hda/codecs/realtek/alc269.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/hda/codecs/realtek/alc269.c b/sound/hda/codecs/realtek/alc269.c index 9cc4463af05e..b4c9a5584a04 100644 --- a/sound/hda/codecs/realtek/alc269.c +++ b/sound/hda/codecs/realtek/alc269.c @@ -7846,6 +7846,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1d72, 0x1947, "RedmiBook Air", ALC255_FIXUP_XIAOMI_HEADSET_MIC), SND_PCI_QUIRK(0x1e39, 0xca14, "MEDION NM14LNL", ALC233_FIXUP_MEDION_MTL_SPK), SND_PCI_QUIRK(0x1e50, 0x7007, "Positivo DN50E", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), + SND_PCI_QUIRK(0x1e50, 0x7038, "Positivo DN140", ALC269_FIXUP_LIMIT_INT_MIC_BOOST), SND_PCI_QUIRK(0x1ee7, 0x2078, "HONOR BRB-X M1010", ALC2XX_FIXUP_HEADSET_MIC), SND_PCI_QUIRK(0x1ee7, 0x2081, "HONOR MRB-XXX M1020", ALC256_FIXUP_HONOR_MRB_XXX_M1020_AUDIO), SND_PCI_QUIRK(0x1f4c, 0xe001, "Minisforum V3 (SE)", ALC245_FIXUP_BASS_HP_DAC), From fe80251152fed5b185f795ef2cd9f7fe9c3162e0 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 22 May 2026 16:49:44 +0200 Subject: [PATCH 210/411] ACPI: button: Fix ACPI GPE handler leak during removal Commit a7e23ec17fee ("ACPI: button: Install notifier for system events as well") changed the ACPI notify handler type for ACPI buttons to ACPI_ALL_NOTIFY, but it forgot to update acpi_button_remove() to reflect that change. This leads to leaking the notify handler past driver removal, which may cause a kernel crash to occur if ACPI notify on the given device is triggered after removing the driver, and causes a subsequent probe of the given device with the same driver to fail. Address this by updating the acpi_remove_notify_handler() call in acpi_button_remove() as appropriate. Fixes: a7e23ec17fee ("ACPI: button: Install notifier for system events as well") Signed-off-by: Rafael J. Wysocki Reviewed-by: Mario Limonciello (AMD) Cc: 6.15+ # 6.15+ Link: https://patch.msgid.link/7954431.EvYhyI6sBW@rafael.j.wysocki --- drivers/acpi/button.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c index b47301ee4c8a..7c2e1a422ba0 100644 --- a/drivers/acpi/button.c +++ b/drivers/acpi/button.c @@ -689,7 +689,7 @@ static void acpi_button_remove(struct platform_device *pdev) acpi_button_event); break; default: - acpi_remove_notify_handler(adev->handle, ACPI_DEVICE_NOTIFY, + acpi_remove_notify_handler(adev->handle, ACPI_ALL_NOTIFY, button->type == ACPI_BUTTON_TYPE_LID ? acpi_lid_notify : acpi_button_notify); From a004b8f0d3bc5d82d3f2c91ff93f4b4b7ccb8f76 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 22 May 2026 16:52:10 +0200 Subject: [PATCH 211/411] ACPI: button: Enable wakeup GPEs for ACPI buttons at probe time Prior to commit 57c31e6d620f ("ACPI: scan: Use acpi_setup_gpe_for_wake() for buttons"), ACPI button wakeup GPEs having handler methods remained enabled after acpi_wakeup_gpe_init(), but currently they are not enabled because acpi_setup_gpe_for_wake() disables them. That causes function keys to stop working on some systems [1] and there may be other related issues elsewhere. To address that, make the ACPI button driver enable wakeup GPEs for ACPI buttons so long as they have handler methods. While this does not restore the old behavior exactly (the ACPI button driver needs to be bound to the button devices for the GPEs to be enabled), it should be sufficient to restore the missing functionality. For this purpose, introduce acpi_enable_gpe_cond() that enables a GPE if its dispatch type matches the supplied one and modify acpi_button_probe() to use that function for enabling the GPEs in question. Fixes: 57c31e6d620f ("ACPI: scan: Use acpi_setup_gpe_for_wake() for buttons") Reported-by: Nick Closes: https://lore.kernel.org/linux-acpi/E2OXET.4X5GTP37VTNC3@kousu.ca/ [1] Signed-off-by: Rafael J. Wysocki Tested-by: Nick Cc: 7.0+ # 7.0+ Link: https://patch.msgid.link/9629117.CDJkKcVGEf@rafael.j.wysocki --- drivers/acpi/acpica/evxfgpe.c | 50 ++++++++++++++++++++++++++++------- drivers/acpi/button.c | 22 +++++++++++++++ include/acpi/acpixf.h | 5 ++++ 3 files changed, 68 insertions(+), 9 deletions(-) diff --git a/drivers/acpi/acpica/evxfgpe.c b/drivers/acpi/acpica/evxfgpe.c index 60dacec1b121..4074b5908db3 100644 --- a/drivers/acpi/acpica/evxfgpe.c +++ b/drivers/acpi/acpica/evxfgpe.c @@ -78,18 +78,22 @@ ACPI_EXPORT_SYMBOL(acpi_update_all_gpes) /******************************************************************************* * - * FUNCTION: acpi_enable_gpe + * FUNCTION: acpi_enable_gpe_cond * * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1 * gpe_number - GPE level within the GPE block + * dispatch_type - GPE dispatch type to match * * RETURN: Status * - * DESCRIPTION: Add a reference to a GPE. On the first reference, the GPE is - * hardware-enabled. + * DESCRIPTION: Add a reference to a GPE so long as its dispatch type matches + * the supplied one, or it is different from ACPI_GPE_DISPATCH_NONE + * if the supplied one is ACPI_GPE_DISPATCH_MASK. On the first + * reference, the GPE is hardware-enabled. * ******************************************************************************/ -acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number) +acpi_status acpi_enable_gpe_cond(acpi_handle gpe_device, u32 gpe_number, + u8 dispatch_type) { acpi_status status = AE_BAD_PARAMETER; struct acpi_gpe_event_info *gpe_event_info; @@ -100,14 +104,18 @@ acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number) flags = acpi_os_acquire_lock(acpi_gbl_gpe_lock); /* - * Ensure that we have a valid GPE number and that there is some way - * of handling the GPE (handler or a GPE method). In other words, we - * won't allow a valid GPE to be enabled if there is no way to handle it. + * Ensure that we have a valid GPE number and that the dispatch type of + * the GPE matches the supplied one (or it is not ACPI_GPE_DISPATCH_NONE + * if the supplied one is ACPI_GPE_DISPATCH_MASK). */ gpe_event_info = acpi_ev_get_gpe_event_info(gpe_device, gpe_number); if (gpe_event_info) { - if (ACPI_GPE_DISPATCH_TYPE(gpe_event_info->flags) != - ACPI_GPE_DISPATCH_NONE) { + if (dispatch_type == ACPI_GPE_DISPATCH_MASK) + dispatch_type = ACPI_GPE_DISPATCH_TYPE(gpe_event_info->flags); + else if (dispatch_type != ACPI_GPE_DISPATCH_TYPE(gpe_event_info->flags)) + dispatch_type = ACPI_GPE_DISPATCH_NONE; + + if (dispatch_type != ACPI_GPE_DISPATCH_NONE) { status = acpi_ev_add_gpe_reference(gpe_event_info, TRUE); if (ACPI_SUCCESS(status) && ACPI_GPE_IS_POLLING_NEEDED(gpe_event_info)) { @@ -128,6 +136,30 @@ acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number) acpi_os_release_lock(acpi_gbl_gpe_lock, flags); return_ACPI_STATUS(status); } +ACPI_EXPORT_SYMBOL(acpi_enable_gpe_cond) + +/******************************************************************************* + * + * FUNCTION: acpi_enable_gpe + * + * PARAMETERS: gpe_device - Parent GPE Device. NULL for GPE0/GPE1 + * gpe_number - GPE level within the GPE block + * + * RETURN: Status + * + * DESCRIPTION: Add a reference to a GPE. On the first reference, the GPE is + * hardware-enabled. + * + ******************************************************************************/ +acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number) +{ + /* + * Ensure that there is some way of handling the GPE (handler or a GPE + * method). In other words, we won't allow a valid GPE to be enabled if + * there is no way to handle it. + */ + return acpi_enable_gpe_cond(gpe_device, gpe_number, ACPI_GPE_DISPATCH_MASK); +} ACPI_EXPORT_SYMBOL(acpi_enable_gpe) /******************************************************************************* diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c index 7c2e1a422ba0..e8dd306e17ed 100644 --- a/drivers/acpi/button.c +++ b/drivers/acpi/button.c @@ -179,6 +179,7 @@ struct acpi_button { ktime_t last_time; bool suspended; bool lid_state_initialized; + bool gpe_enabled; }; static struct acpi_device *lid_device; @@ -646,6 +647,21 @@ static int acpi_button_probe(struct platform_device *pdev) status = acpi_install_notify_handler(device->handle, ACPI_ALL_NOTIFY, handler, button); + if (ACPI_SUCCESS(status) && device->wakeup.flags.valid) { + acpi_status st; + + /* + * If the wakeup GPE has a handler method, enable it in + * case it is also used for signaling runtime events. + */ + st = acpi_enable_gpe_cond(device->wakeup.gpe_device, + device->wakeup.gpe_number, + ACPI_GPE_DISPATCH_METHOD); + button->gpe_enabled = ACPI_SUCCESS(st); + if (button->gpe_enabled) + dev_dbg(button->dev, "Enabled ACPI GPE%02llx\n", + device->wakeup.gpe_number); + } break; } if (ACPI_FAILURE(status)) { @@ -689,6 +705,12 @@ static void acpi_button_remove(struct platform_device *pdev) acpi_button_event); break; default: + if (button->gpe_enabled) { + dev_dbg(button->dev, "Disabling ACPI GPE%02llx\n", + adev->wakeup.gpe_number); + acpi_disable_gpe(adev->wakeup.gpe_device, + adev->wakeup.gpe_number); + } acpi_remove_notify_handler(adev->handle, ACPI_ALL_NOTIFY, button->type == ACPI_BUTTON_TYPE_LID ? acpi_lid_notify : diff --git a/include/acpi/acpixf.h b/include/acpi/acpixf.h index 49d1749f30bb..a4b562700151 100644 --- a/include/acpi/acpixf.h +++ b/include/acpi/acpixf.h @@ -725,6 +725,11 @@ ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status */ ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status acpi_update_all_gpes(void)) +ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status + acpi_enable_gpe_cond(acpi_handle gpe_device, + u32 gpe_number, + u8 dispatch_type)) + ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status acpi_enable_gpe(acpi_handle gpe_device, u32 gpe_number)) From 3109f9f38800841e46769e95e1ba11f1f8c7b230 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Fri, 22 May 2026 16:53:48 +0200 Subject: [PATCH 212/411] ACPI: button: Add missing device class clearing on probe failures Commit e18947038bf4 ("ACPI: driver: Do not set acpi_device_class() unnecessarily") modified acpi_button_remove() to clear the device class field in struct acpi_device on driver removal, but it should also have updated the rollback path in acpi_button_probe(), which it didn't do, so do it now. Fixes: e18947038bf4 ("ACPI: driver: Do not set acpi_device_class() unnecessarily") Signed-off-by: Rafael J. Wysocki Reviewed-by: Mario Limonciello (AMD) Link: https://patch.msgid.link/6167713.MhkbZ0Pkbq@rafael.j.wysocki --- drivers/acpi/button.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/acpi/button.c b/drivers/acpi/button.c index e8dd306e17ed..d80276368b81 100644 --- a/drivers/acpi/button.c +++ b/drivers/acpi/button.c @@ -687,6 +687,7 @@ static int acpi_button_probe(struct platform_device *pdev) acpi_button_remove_fs(button); err_free_button: kfree(button); + memset(acpi_device_class(device), 0, sizeof(acpi_device_class)); return error; } From 13d33b9ef67066c77c84273fac5a1d3fde3533d1 Mon Sep 17 00:00:00 2001 From: Berkant Koc Date: Tue, 19 May 2026 22:08:17 +0200 Subject: [PATCH 213/411] drm/hyperv: validate resolution_count and fix WIN8 fallback A SYNTHVID_RESOLUTION_RESPONSE with resolution_count > 64 walks past the supported_resolution[SYNTHVID_MAX_RESOLUTION_COUNT] array in the parse loop. Bound resolution_count against the array size, folded into the existing zero-check. When the WIN10 resolution probe fails, the caller in hyperv_connect_vsp() left hv->screen_*_max / preferred_* unpopulated, which sets mode_config.max_width / max_height to 0 and makes drm_internal_framebuffer_create() reject every userspace framebuffer with -EINVAL. The pre-WIN10 branch had the same gap for preferred_width / preferred_height. Use a single post-probe fallback guarded by screen_width_max == 0 so both paths converge on the WIN8 defaults. Signed-off-by: Berkant Koc Assisted-by: Claude:claude-opus-4-7 berkoc-pipeline Fixes: 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic video device") Cc: stable@vger.kernel.org # 5.14+ Reviewed-by: Michael Kelley Tested-by: Michael Kelley Signed-off-by: Hamza Mahfooz Link: https://patch.msgid.link/6945b22419c7d404b4954a113de2ac9c900dba93.1779542874.git.me@berkoc.com --- drivers/gpu/drm/hyperv/hyperv_drm_proto.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_proto.c b/drivers/gpu/drm/hyperv/hyperv_drm_proto.c index 051ecc526832..c3d0ff229e3d 100644 --- a/drivers/gpu/drm/hyperv/hyperv_drm_proto.c +++ b/drivers/gpu/drm/hyperv/hyperv_drm_proto.c @@ -391,8 +391,11 @@ static int hyperv_get_supported_resolution(struct hv_device *hdev) return -ETIMEDOUT; } - if (msg->resolution_resp.resolution_count == 0) { - drm_err(dev, "No supported resolutions\n"); + if (msg->resolution_resp.resolution_count == 0 || + msg->resolution_resp.resolution_count > + SYNTHVID_MAX_RESOLUTION_COUNT) { + drm_err(dev, "Invalid resolution count: %d\n", + msg->resolution_resp.resolution_count); return -ENODEV; } @@ -508,9 +511,13 @@ int hyperv_connect_vsp(struct hv_device *hdev) ret = hyperv_get_supported_resolution(hdev); if (ret) drm_err(dev, "Failed to get supported resolution from host, use default\n"); - } else { + } + + if (!hv->screen_width_max) { hv->screen_width_max = SYNTHVID_WIDTH_WIN8; hv->screen_height_max = SYNTHVID_HEIGHT_WIN8; + hv->preferred_width = SYNTHVID_WIDTH_WIN8; + hv->preferred_height = SYNTHVID_HEIGHT_WIN8; } hv->mmio_megabytes = hdev->channel->offermsg.offer.mmio_megabytes; From 7f87763f47a3c22fb50265a00619ef10f2394b18 Mon Sep 17 00:00:00 2001 From: Berkant Koc Date: Sat, 23 May 2026 15:27:47 +0200 Subject: [PATCH 214/411] drm/hyperv: validate VMBus packet size in receive callback hyperv_receive_sub() reads msg->vid_hdr.type and dispatches into one of four message-type branches without knowing how many bytes the host wrote into hv->recv_buf. The completion path then runs memcpy(hv->init_buf, msg, VMBUS_MAX_PACKET_SIZE), so the consumer that wakes on wait_for_completion_timeout() can read up to 16 KiB of residue from a prior message as if it were the response payload. Pass bytes_recvd into hyperv_receive_sub() and reject any packet that does not cover the pipe + synthvid header. A single switch on msg->vid_hdr.type then computes the type-specific payload size: the three completion-driving types (SYNTHVID_VERSION_RESPONSE, SYNTHVID_RESOLUTION_RESPONSE, SYNTHVID_VRAM_LOCATION_ACK) fall through to a shared exit that requires that size before memcpy/complete, while SYNTHVID_FEATURE_CHANGE validates its own payload and returns before reading is_dirt_needed. Unknown types are dropped. SYNTHVID_RESOLUTION_RESPONSE is variable length: the host fills resolution_count entries, not the full SYNTHVID_MAX_RESOLUTION_COUNT array. Validate the fixed prefix first so resolution_count can be read, bound it against the array, then require only the count-sized array, so the shorter responses the host actually sends are accepted. Only run the sub-handler when vmbus_recvpacket() returned success. The memcpy length is bytes_recvd, which is bounded by VMBUS_MAX_PACKET_SIZE only on a successful receive; on -ENOBUFS vmbus_recvpacket() instead reports the required length, which can exceed hv->recv_buf, so copying bytes_recvd would read and write past the 16 KiB buffers. Gating on the success return keeps the copy bounded. The nonzero-return path is itself a malformed-message case and is now logged rather than silently skipped; channel recovery is not attempted. Rejected packets are reported via drm_err_ratelimited() rather than silently dropped, matching the CoCo-hardened pattern in hv_kvp_onchannelcallback(). Fixes: 76c56a5affeb ("drm/hyperv: Add DRM driver for hyperv synthetic video device") Cc: stable@vger.kernel.org # 5.14+ Signed-off-by: Berkant Koc Assisted-by: Claude:claude-opus-4-7 berkoc-pipeline Reviewed-by: Michael Kelley Tested-by: Michael Kelley Signed-off-by: Hamza Mahfooz Link: https://patch.msgid.link/8200dbc199c7a9b75ac7e8af6c748d2189b5ebd5.1779542874.git.me@berkoc.com --- drivers/gpu/drm/hyperv/hyperv_drm_proto.c | 100 +++++++++++++++++++--- 1 file changed, 87 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_proto.c b/drivers/gpu/drm/hyperv/hyperv_drm_proto.c index c3d0ff229e3d..4e6f703a1b33 100644 --- a/drivers/gpu/drm/hyperv/hyperv_drm_proto.c +++ b/drivers/gpu/drm/hyperv/hyperv_drm_proto.c @@ -420,30 +420,92 @@ static int hyperv_get_supported_resolution(struct hv_device *hdev) return 0; } -static void hyperv_receive_sub(struct hv_device *hdev) +static void hyperv_receive_sub(struct hv_device *hdev, u32 bytes_recvd) { struct hyperv_drm_device *hv = hv_get_drvdata(hdev); struct synthvid_msg *msg; + size_t hdr_size; + size_t need; if (!hv) return; - msg = (struct synthvid_msg *)hv->recv_buf; - - /* Complete the wait event */ - if (msg->vid_hdr.type == SYNTHVID_VERSION_RESPONSE || - msg->vid_hdr.type == SYNTHVID_RESOLUTION_RESPONSE || - msg->vid_hdr.type == SYNTHVID_VRAM_LOCATION_ACK) { - memcpy(hv->init_buf, msg, VMBUS_MAX_PACKET_SIZE); - complete(&hv->wait); + hdr_size = sizeof(struct pipe_msg_hdr) + + sizeof(struct synthvid_msg_hdr); + if (bytes_recvd < hdr_size) { + drm_err_ratelimited(&hv->dev, + "synthvid packet too small for header: %u\n", + bytes_recvd); return; } - if (msg->vid_hdr.type == SYNTHVID_FEATURE_CHANGE) { + msg = (struct synthvid_msg *)hv->recv_buf; + need = hdr_size; + + switch (msg->vid_hdr.type) { + case SYNTHVID_VERSION_RESPONSE: + need += sizeof(struct synthvid_version_resp); + break; + case SYNTHVID_RESOLUTION_RESPONSE: + /* + * The resolution response is variable length: the host + * fills resolution_count entries, not the full + * SYNTHVID_MAX_RESOLUTION_COUNT array. Require the fixed + * prefix first so resolution_count can be read, then + * demand exactly the count-sized array. + */ + need += offsetof(struct synthvid_supported_resolution_resp, + supported_resolution); + if (bytes_recvd < need) + break; + if (msg->resolution_resp.resolution_count > + SYNTHVID_MAX_RESOLUTION_COUNT) { + drm_err_ratelimited(&hv->dev, + "synthvid resolution count too large: %u\n", + msg->resolution_resp.resolution_count); + return; + } + need += msg->resolution_resp.resolution_count * + sizeof(struct hvd_screen_info); + break; + case SYNTHVID_VRAM_LOCATION_ACK: + need += sizeof(struct synthvid_vram_location_ack); + break; + case SYNTHVID_FEATURE_CHANGE: + /* + * Not a completion-driving message: validate its own payload + * and consume it here rather than falling through to the + * memcpy/complete shared by the wait-event responses. + */ + if (bytes_recvd < need + + sizeof(struct synthvid_feature_change)) { + drm_err_ratelimited(&hv->dev, + "synthvid feature change packet too small: %u\n", + bytes_recvd); + return; + } hv->dirt_needed = msg->feature_chg.is_dirt_needed; if (hv->dirt_needed) hyperv_hide_hw_ptr(hv->hdev); + return; + default: + return; } + + /* + * Shared completion path for the wait-event responses + * (VERSION_RESPONSE, RESOLUTION_RESPONSE, VRAM_LOCATION_ACK): + * require the type-specific payload before handing the buffer to + * the waiter. + */ + if (bytes_recvd < need) { + drm_err_ratelimited(&hv->dev, + "synthvid packet too small for type %u: %u < %zu\n", + msg->vid_hdr.type, bytes_recvd, need); + return; + } + memcpy(hv->init_buf, msg, bytes_recvd); + complete(&hv->wait); } static void hyperv_receive(void *ctx) @@ -464,9 +526,21 @@ static void hyperv_receive(void *ctx) ret = vmbus_recvpacket(hdev->channel, recv_buf, VMBUS_MAX_PACKET_SIZE, &bytes_recvd, &req_id); - if (bytes_recvd > 0 && - recv_buf->pipe_hdr.type == PIPE_MSG_DATA) - hyperv_receive_sub(hdev); + if (ret) { + /* + * A nonzero return (e.g. -ENOBUFS for an oversized + * packet) is itself a malformed message: bytes_recvd + * then reports the required length rather than a copied + * payload, so it must not be forwarded to the + * sub-handler. Channel recovery is not attempted. + */ + drm_err_ratelimited(&hv->dev, + "vmbus_recvpacket failed: %d (need %u)\n", + ret, bytes_recvd); + } else if (bytes_recvd > 0 && + recv_buf->pipe_hdr.type == PIPE_MSG_DATA) { + hyperv_receive_sub(hdev, bytes_recvd); + } } while (bytes_recvd > 0 && ret == 0); } From afb2a3a9d8369d18122a0d7cd294eba9a98259c6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A1ssio=20Gabriel?= Date: Tue, 19 May 2026 13:51:47 -0300 Subject: [PATCH 215/411] ASoC: Intel: bytcht_es8316: Fix MCLK leak on init errors MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit byt_cht_es8316_init() enables MCLK before configuring the codec sysclk and creating the headset jack. If either of those later steps fails, the function returns without disabling MCLK, leaving the clock enabled after card registration fails. Track whether this driver enabled MCLK and disable it on the init error paths. Add the matching DAI link exit callback so the same clock enable is also balanced when ASoC cleans up a successfully initialized link. Fixes: a03bdaa565cb ("ASoC: Intel: add machine driver for BYT/CHT + ES8316") Signed-off-by: Cássio Gabriel Link: https://patch.msgid.link/20260519-asoc-bytcht-es8316-mclk-leak-v1-1-b4a11cdc2afd@gmail.com Signed-off-by: Mark Brown --- sound/soc/intel/boards/bytcht_es8316.c | 29 ++++++++++++++++++++++++-- 1 file changed, 27 insertions(+), 2 deletions(-) diff --git a/sound/soc/intel/boards/bytcht_es8316.c b/sound/soc/intel/boards/bytcht_es8316.c index 192e2a394ff3..ea387dc74273 100644 --- a/sound/soc/intel/boards/bytcht_es8316.c +++ b/sound/soc/intel/boards/bytcht_es8316.c @@ -40,6 +40,7 @@ struct byt_cht_es8316_private { struct gpio_desc *speaker_en_gpio; struct device *codec_dev; bool speaker_en; + bool mclk_enabled; }; enum { @@ -170,6 +171,15 @@ static struct snd_soc_jack_pin byt_cht_es8316_jack_pins[] = { }, }; +static void byt_cht_es8316_disable_mclk(struct byt_cht_es8316_private *priv) +{ + if (!priv->mclk_enabled) + return; + + clk_disable_unprepare(priv->mclk); + priv->mclk_enabled = false; +} + static int byt_cht_es8316_init(struct snd_soc_pcm_runtime *runtime) { struct snd_soc_component *codec = snd_soc_rtd_to_codec(runtime, 0)->component; @@ -227,12 +237,14 @@ static int byt_cht_es8316_init(struct snd_soc_pcm_runtime *runtime) ret = clk_prepare_enable(priv->mclk); if (ret) dev_err(card->dev, "unable to enable MCLK\n"); + else + priv->mclk_enabled = true; ret = snd_soc_dai_set_sysclk(snd_soc_rtd_to_codec(runtime, 0), 0, 19200000, SND_SOC_CLOCK_IN); if (ret < 0) { dev_err(card->dev, "can't set codec clock %d\n", ret); - return ret; + goto err_disable_mclk; } ret = snd_soc_card_jack_new_pins(card, "Headset", @@ -241,13 +253,25 @@ static int byt_cht_es8316_init(struct snd_soc_pcm_runtime *runtime) ARRAY_SIZE(byt_cht_es8316_jack_pins)); if (ret) { dev_err(card->dev, "jack creation failed %d\n", ret); - return ret; + goto err_disable_mclk; } snd_jack_set_key(priv->jack.jack, SND_JACK_BTN_0, KEY_PLAYPAUSE); snd_soc_component_set_jack(codec, &priv->jack, NULL); return 0; + +err_disable_mclk: + byt_cht_es8316_disable_mclk(priv); + return ret; +} + +static void byt_cht_es8316_exit(struct snd_soc_pcm_runtime *runtime) +{ + struct snd_soc_card *card = runtime->card; + struct byt_cht_es8316_private *priv = snd_soc_card_get_drvdata(card); + + byt_cht_es8316_disable_mclk(priv); } static int byt_cht_es8316_codec_fixup(struct snd_soc_pcm_runtime *rtd, @@ -353,6 +377,7 @@ static struct snd_soc_dai_link byt_cht_es8316_dais[] = { | SND_SOC_DAIFMT_CBC_CFC, .be_hw_params_fixup = byt_cht_es8316_codec_fixup, .init = byt_cht_es8316_init, + .exit = byt_cht_es8316_exit, SND_SOC_DAILINK_REG(ssp2_port, ssp2_codec, platform), }, }; From cee3e63e7106c3c81b2053371fdf14240bfba2fc Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Mon, 18 May 2026 09:23:43 +0000 Subject: [PATCH 216/411] ASoC: qcom: q6asm-dai: do not set stream state in event and trigger callbacks The q6asm-dai stream state is used by prepare() to decide whether an existing stream setup needs to be closed before opening/configuring a new one. Updating the state from trigger or asynchronous DSP callbacks can make that state stale or incorrect relative to the actual setup lifetime. In particular, setting Q6ASM_STREAM_STOPPED on STOP or EOS completion can make prepare() believe there is no active setup to close, which can result in opening/configuring the same stream more than once. Keep stream state updates tied to prepare(), where the stream is actually closed and reopened, and stop changing it from trigger and EOS callbacks. Fixes: bfbb12dfa144 ("ASoC: qcom: q6asm-dai: perform correct state check before closing") Cc: Stable@vger.kernel.org Closes: https://lore.kernel.org/all/afS7rTHdc9TyIeLx@rdacayan/ Signed-off-by: Srinivas Kandagatla Link: https://patch.msgid.link/20260518092347.3446946-2-srinivas.kandagatla@oss.qualcomm.com Signed-off-by: Mark Brown --- sound/soc/qcom/qdsp6/q6asm-dai.c | 5 ----- 1 file changed, 5 deletions(-) diff --git a/sound/soc/qcom/qdsp6/q6asm-dai.c b/sound/soc/qcom/qdsp6/q6asm-dai.c index 4f8f7db6c3d3..56f0d8913904 100644 --- a/sound/soc/qcom/qdsp6/q6asm-dai.c +++ b/sound/soc/qcom/qdsp6/q6asm-dai.c @@ -186,7 +186,6 @@ static void event_handler(uint32_t opcode, uint32_t token, case ASM_CLIENT_EVENT_CMD_RUN_DONE: break; case ASM_CLIENT_EVENT_CMD_EOS_DONE: - prtd->state = Q6ASM_STREAM_STOPPED; break; case ASM_CLIENT_EVENT_DATA_WRITE_DONE: { snd_pcm_period_elapsed(substream); @@ -341,7 +340,6 @@ static int q6asm_dai_trigger(struct snd_soc_component *component, 0, 0, 0); break; case SNDRV_PCM_TRIGGER_STOP: - prtd->state = Q6ASM_STREAM_STOPPED; ret = q6asm_cmd_nowait(prtd->audio_client, prtd->stream_id, CMD_EOS); break; @@ -555,8 +553,6 @@ static void compress_event_handler(uint32_t opcode, uint32_t token, snd_compr_drain_notify(prtd->cstream); prtd->notify_on_drain = false; - } else { - prtd->state = Q6ASM_STREAM_STOPPED; } break; @@ -1014,7 +1010,6 @@ static int q6asm_dai_compr_trigger(struct snd_soc_component *component, 0, 0, 0); break; case SNDRV_PCM_TRIGGER_STOP: - prtd->state = Q6ASM_STREAM_STOPPED; ret = q6asm_cmd_nowait(prtd->audio_client, prtd->stream_id, CMD_EOS); break; From 048c540ee76ded666bda74f9dae1ca3254e0633c Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Mon, 18 May 2026 09:23:44 +0000 Subject: [PATCH 217/411] ASoC: qcom: q6asm-dai: close stream only when running q6asm_dai_close() and q6asm_dai_compr_free() currently issue CMD_CLOSE whenever prtd->state is non-zero. After prepare() closes an existing stream, the state is updated to Q6ASM_STREAM_STOPPED. Since this state is also non-zero, the close and free paths can send CMD_CLOSE again for a stream that has already been closed. Restrict CMD_CLOSE to the Q6ASM_STREAM_RUNNING state so the command is sent only when the ASM stream is still active. Fixes: 2a9e92d371db ("ASoC: qdsp6: q6asm: Add q6asm dai driver") Cc: Stable@vger.kernel.org Signed-off-by: Srinivas Kandagatla Link: https://patch.msgid.link/20260518092347.3446946-3-srinivas.kandagatla@oss.qualcomm.com Signed-off-by: Mark Brown --- sound/soc/qcom/qdsp6/q6asm-dai.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/sound/soc/qcom/qdsp6/q6asm-dai.c b/sound/soc/qcom/qdsp6/q6asm-dai.c index 56f0d8913904..ef86b5b9a951 100644 --- a/sound/soc/qcom/qdsp6/q6asm-dai.c +++ b/sound/soc/qcom/qdsp6/q6asm-dai.c @@ -455,12 +455,12 @@ static int q6asm_dai_close(struct snd_soc_component *component, struct q6asm_dai_rtd *prtd = runtime->private_data; if (prtd->audio_client) { - if (prtd->state) + if (prtd->state == Q6ASM_STREAM_RUNNING) { q6asm_cmd(prtd->audio_client, prtd->stream_id, CMD_CLOSE); - - q6asm_unmap_memory_regions(substream->stream, + q6asm_unmap_memory_regions(substream->stream, prtd->audio_client); + } q6asm_audio_client_free(prtd->audio_client); prtd->audio_client = NULL; } @@ -670,7 +670,7 @@ static int q6asm_dai_compr_free(struct snd_soc_component *component, struct snd_soc_pcm_runtime *rtd = stream->private_data; if (prtd->audio_client) { - if (prtd->state) { + if (prtd->state == Q6ASM_STREAM_RUNNING) { q6asm_cmd(prtd->audio_client, prtd->stream_id, CMD_CLOSE); if (prtd->next_track_stream_id) { @@ -678,11 +678,11 @@ static int q6asm_dai_compr_free(struct snd_soc_component *component, prtd->next_track_stream_id, CMD_CLOSE); } - } - snd_dma_free_pages(&prtd->dma_buffer); - q6asm_unmap_memory_regions(stream->direction, + q6asm_unmap_memory_regions(stream->direction, prtd->audio_client); + } + snd_dma_free_pages(&prtd->dma_buffer); q6asm_audio_client_free(prtd->audio_client); prtd->audio_client = NULL; } From 4b4db09f283df65d780bc7cee66cb4a7e9bf4770 Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Mon, 18 May 2026 09:23:45 +0000 Subject: [PATCH 218/411] ASoC: qcom: q6asm-dai: fix error handling in prepare and set_params Fix error handling in q6asm_dai_compr_set_params() and q6asm_dai_prepare() for both CMD_CLOSE and q6asm_unmap_memory_regions(). In both the functions, we are doing q6asm_audio_client_free in failure cases, which means if prepare or set_params fail, we can never recover. Now open and close are done in respective dai_open/close functions. Fixes: 2a9e92d371db ("ASoC: qdsp6: q6asm: Add q6asm dai driver") Cc: Stable@vger.kernel.org Signed-off-by: Srinivas Kandagatla Link: https://patch.msgid.link/20260518092347.3446946-4-srinivas.kandagatla@oss.qualcomm.com Signed-off-by: Mark Brown --- sound/soc/qcom/qdsp6/q6asm-dai.c | 24 ++++++++++++++++-------- 1 file changed, 16 insertions(+), 8 deletions(-) diff --git a/sound/soc/qcom/qdsp6/q6asm-dai.c b/sound/soc/qcom/qdsp6/q6asm-dai.c index ef86b5b9a951..fd691004a657 100644 --- a/sound/soc/qcom/qdsp6/q6asm-dai.c +++ b/sound/soc/qcom/qdsp6/q6asm-dai.c @@ -226,9 +226,19 @@ static int q6asm_dai_prepare(struct snd_soc_component *component, /* rate and channels are sent to audio driver */ if (prtd->state == Q6ASM_STREAM_RUNNING) { /* clear the previous setup if any */ - q6asm_cmd(prtd->audio_client, prtd->stream_id, CMD_CLOSE); - q6asm_unmap_memory_regions(substream->stream, - prtd->audio_client); + ret = q6asm_cmd(prtd->audio_client, prtd->stream_id, CMD_CLOSE); + if (ret < 0) { + dev_err(dev, "Failed to close q6asm stream %d\n", prtd->stream_id); + return ret; + } + + ret = q6asm_unmap_memory_regions(substream->stream, prtd->audio_client); + if (ret < 0) { + dev_err(dev, "Failed to unmap memory regions for q6asm stream %d\n", + prtd->stream_id); + return ret; + } + q6routing_stream_close(soc_prtd->dai_link->id, substream->stream); prtd->state = Q6ASM_STREAM_STOPPED; @@ -296,8 +306,6 @@ static int q6asm_dai_prepare(struct snd_soc_component *component, q6asm_cmd(prtd->audio_client, prtd->stream_id, CMD_CLOSE); open_err: q6asm_unmap_memory_regions(substream->stream, prtd->audio_client); - q6asm_audio_client_free(prtd->audio_client); - prtd->audio_client = NULL; return ret; } @@ -912,7 +920,7 @@ static int q6asm_dai_compr_set_params(struct snd_soc_component *component, prtd->session_id, dir); if (ret) { dev_err(dev, "Stream reg failed ret:%d\n", ret); - goto q6_err; + goto routing_err; } ret = __q6asm_dai_compr_set_codec_params(component, stream, @@ -938,11 +946,11 @@ static int q6asm_dai_compr_set_params(struct snd_soc_component *component, return 0; q6_err: + q6routing_stream_close(rtd->dai_link->id, dir); +routing_err: q6asm_cmd(prtd->audio_client, prtd->stream_id, CMD_CLOSE); open_err: - q6asm_audio_client_free(prtd->audio_client); - prtd->audio_client = NULL; return ret; } From c92d880cde739a3fb6346f42fc5feb2f093c063c Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Mon, 18 May 2026 09:23:46 +0000 Subject: [PATCH 219/411] ASoC: qcom: q6asm-dai: remove unnecessary braces The ASM_CLIENT_EVENT_DATA_WRITE_DONE case does not declare any local variables or require a separate scope, so drop the unnecessary braces. Signed-off-by: Srinivas Kandagatla Link: https://patch.msgid.link/20260518092347.3446946-5-srinivas.kandagatla@oss.qualcomm.com Signed-off-by: Mark Brown --- sound/soc/qcom/qdsp6/q6asm-dai.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/sound/soc/qcom/qdsp6/q6asm-dai.c b/sound/soc/qcom/qdsp6/q6asm-dai.c index fd691004a657..5774c2611197 100644 --- a/sound/soc/qcom/qdsp6/q6asm-dai.c +++ b/sound/soc/qcom/qdsp6/q6asm-dai.c @@ -187,10 +187,9 @@ static void event_handler(uint32_t opcode, uint32_t token, break; case ASM_CLIENT_EVENT_CMD_EOS_DONE: break; - case ASM_CLIENT_EVENT_DATA_WRITE_DONE: { + case ASM_CLIENT_EVENT_DATA_WRITE_DONE: snd_pcm_period_elapsed(substream); break; - } case ASM_CLIENT_EVENT_DATA_READ_DONE: snd_pcm_period_elapsed(substream); if (prtd->state == Q6ASM_STREAM_RUNNING) From 909595c288af2304d0902488698e91c8aeee36a3 Mon Sep 17 00:00:00 2001 From: Srinivas Kandagatla Date: Mon, 18 May 2026 09:23:47 +0000 Subject: [PATCH 220/411] ASoC: qcom: q6asm-dai: use pointer type with kzalloc_obj() Use kzalloc_obj(*prtd) instead of explicitly naming the structure type. Signed-off-by: Srinivas Kandagatla Link: https://patch.msgid.link/20260518092347.3446946-6-srinivas.kandagatla@oss.qualcomm.com Signed-off-by: Mark Brown --- sound/soc/qcom/qdsp6/q6asm-dai.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/qcom/qdsp6/q6asm-dai.c b/sound/soc/qcom/qdsp6/q6asm-dai.c index 5774c2611197..4f09fdd40905 100644 --- a/sound/soc/qcom/qdsp6/q6asm-dai.c +++ b/sound/soc/qcom/qdsp6/q6asm-dai.c @@ -383,7 +383,7 @@ static int q6asm_dai_open(struct snd_soc_component *component, return -EINVAL; } - prtd = kzalloc_obj(struct q6asm_dai_rtd); + prtd = kzalloc_obj(*prtd); if (prtd == NULL) return -ENOMEM; From 974820a59efde7c1a7e1260bcfe9bb81f833cc9f Mon Sep 17 00:00:00 2001 From: Mikulas Patocka Date: Mon, 25 May 2026 14:48:58 +0200 Subject: [PATCH 221/411] hpfs: fix a crash if hpfs_map_dnode_bitmap fails If hpfs_map_dnode_bitmap fails, the code would call hpfs_brelse4 on uninitialized quad buffer head, causing a crash. Signed-off-by: Mikulas Patocka Reported-by: Farhad Alemi Cc: stable@vger.kernel.org --- fs/hpfs/alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/hpfs/alloc.c b/fs/hpfs/alloc.c index 66617b1557c6..f5150372618e 100644 --- a/fs/hpfs/alloc.c +++ b/fs/hpfs/alloc.c @@ -372,8 +372,8 @@ int hpfs_check_free_dnodes(struct super_block *s, int n) return 0; } } + hpfs_brelse4(&qbh); } - hpfs_brelse4(&qbh); i = 0; if (hpfs_sb(s)->sb_c_bitmap != -1) { bmp = hpfs_map_bitmap(s, b, &qbh, "chkdn1"); From 86f1d0f063e423a5c1982db1e5e7a8eac511e603 Mon Sep 17 00:00:00 2001 From: Prathamesh Deshpande Date: Wed, 6 May 2026 01:00:31 +0100 Subject: [PATCH 222/411] net/mlx5: HWS: Reject unsupported remove-header action mlx5_cmd_hws_packet_reformat_alloc() handles MLX5_REFORMAT_TYPE_REMOVE_HDR by looking up a matching HWS remove-header action. If mlx5_fs_get_action_remove_header_vlan() returns NULL, the code only logs an error and continues. The function then returns success with a NULL HWS action stored in the packet-reformat object. Return an error when no matching remove-header action is available. Fixes: aecd9d1020e3 ("net/mlx5: fs, add HWS packet reformat API function") Signed-off-by: Prathamesh Deshpande Reviewed-by: Simon Horman Reviewed-by: Yevgeny Kliteynik Acked-by: Tariq Toukan Link: https://patch.msgid.link/20260506000054.51797-1-prathameshdeshpande7@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c index aca77853abb8..5a172c572a68 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/hws/fs_hws.c @@ -1320,8 +1320,10 @@ mlx5_cmd_hws_packet_reformat_alloc(struct mlx5_flow_root_namespace *ns, break; case MLX5_REFORMAT_TYPE_REMOVE_HDR: hws_action = mlx5_fs_get_action_remove_header_vlan(fs_ctx, params); - if (!hws_action) + if (!hws_action) { mlx5_core_err(dev, "Only vlan remove header supported\n"); + return -EOPNOTSUPP; + } break; default: mlx5_core_err(ns->dev, "Packet-reformat not supported(%d)\n", From 7c2eee9c136734825ff524dd8b2146438a4f8250 Mon Sep 17 00:00:00 2001 From: "Mike Rapoport (Microsoft)" Date: Wed, 13 May 2026 13:51:22 +0300 Subject: [PATCH 223/411] memblock: don't touch memblock arrays when memblock_free() is called late When memblock_free() is called after memblock_discard() on architectures that don't select ARCH_KEEP_MEMBLOCK, it tries to update memblock.reserved that was already discarded and it causes use-after-free, for example [ 8.514775] BUG: KASAN: use-after-free in memblock_isolate_range+0x4ac/0x650 [ 8.514775] Read of size 8 at addr ffff88a07fe6a000 by task swapper/0/1 [ 8.514775] Call Trace: [ 8.514775] [ 8.514775] kasan_report+0xb2/0x1b0 [ 8.514775] memblock_isolate_range+0x4ac/0x650 [ 8.514775] memblock_phys_free+0xc4/0x190 [ 8.514775] housekeeping_late_init+0x257/0x280 [ 8.514775] do_one_initcall+0xaa/0x470 [ 8.514775] do_initcalls+0x1b4/0x1f0 [ 8.514775] kernel_init_freeable+0x4b5/0x550 [ 8.514775] kernel_init+0x1c/0x150 [ 8.514775] ret_from_fork+0x5dc/0x8e0 [ 8.514775] ret_from_fork_asm+0x1a/0x30 [ 8.514775] Make sure memblock_free() updates memblock.reserved only when called early enough or when ARCH_KEEP_MEMBLOCK is enabled. Reported-by: Waiman Long Reported-by: Breno Leitao Closes: https://lore.kernel.org/all/20260505051821.1107133-1-longman@redhat.com Tested-by: Waiman Long Tested-by: Breno Leitao Fixes: 87ce9e83ab8b ("memblock, treewide: make memblock_free() handle late freeing") Link: https://patch.msgid.link/20260513105122.502506-1-rppt@kernel.org Signed-off-by: Mike Rapoport (Microsoft) --- mm/memblock.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/mm/memblock.c b/mm/memblock.c index a6a1c91e276d..ccd43f3abb82 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -989,13 +989,15 @@ void __init_memblock memblock_free(void *ptr, size_t size) int __init_memblock memblock_phys_free(phys_addr_t base, phys_addr_t size) { phys_addr_t end = base + size - 1; - int ret; + int ret = 0; memblock_dbg("%s: [%pa-%pa] %pS\n", __func__, &base, &end, (void *)_RET_IP_); kmemleak_free_part_phys(base, size); - ret = memblock_remove_range(&memblock.reserved, base, size); + + if (!slab_is_available() || IS_ENABLED(CONFIG_ARCH_KEEP_MEMBLOCK)) + ret = memblock_remove_range(&memblock.reserved, base, size); if (slab_is_available()) __free_reserved_area(base, base + size, -1); From f7b52afe3592eae66e160586b45a3f2242972c63 Mon Sep 17 00:00:00 2001 From: Zhengchuan Liang Date: Fri, 22 May 2026 17:42:26 +0800 Subject: [PATCH 224/411] ipv6: exthdrs: refresh nh after handling HAO option ip6_parse_tlv() caches skb_network_header(skb) in nh while walking IPv6 TLVs. ipv6_dest_hao() may call pskb_expand_head() for a cloned skb, which can move the skb head and invalidate the cached network header pointer. Refresh nh after ipv6_dest_hao() returns so any trailing padding or TLVs are parsed from the current skb head. This matches the existing pattern used in ip6_parse_tlv() after helpers that can modify skb header storage. Fixes: a831f5bbc89a ("[IPV6] MIP6: Add inbound interface of home address option.") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Xin Liu Co-developed-by: Luxing Yin Signed-off-by: Luxing Yin Signed-off-by: Zhengchuan Liang Signed-off-by: Ren Wei Reviewed-by: Justin Iurman Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/7aba1debc2196189172499e5769802b026f8caf8.1779247873.git.zcliangcn@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index cf90f933ca1a..6d92c02d0e3d 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -201,6 +201,8 @@ static bool ip6_parse_tlv(bool hopbyhop, case IPV6_TLV_HAO: if (!ipv6_dest_hao(skb, off)) return false; + + nh = skb_network_header(skb); break; #endif default: From d47548a36639095939f4747d4c43f2271366f565 Mon Sep 17 00:00:00 2001 From: Justin Iurman Date: Fri, 22 May 2026 13:20:13 +0200 Subject: [PATCH 225/411] ipv6: exthdrs: refresh nh pointer after ipv6_hop_jumbo() ipv6_hop_jumbo() calls pskb_trim_rcsum(), which can change skb pointers. Let's recompute nh pointer to make sure any change won't mess things up. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Signed-off-by: Justin Iurman Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260522112013.12342-1-justin.iurman@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index 6d92c02d0e3d..aca2a2abd2df 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -184,6 +184,8 @@ static bool ip6_parse_tlv(bool hopbyhop, case IPV6_TLV_JUMBO: if (!ipv6_hop_jumbo(skb, off)) return false; + + nh = skb_network_header(skb); break; case IPV6_TLV_CALIPSO: if (!ipv6_hop_calipso(skb, off)) From e68842b3356471ba56c882209f324613dac47f64 Mon Sep 17 00:00:00 2001 From: Junrui Luo Date: Wed, 20 May 2026 11:47:55 +0800 Subject: [PATCH 226/411] macsec: fix replay protection at XPN lower-PN wrap In macsec_post_decrypt(), when pn is U32_MAX, pn + 1 overflows u32 to 0 and the first branch never fires. If next_pn_halves.lower is also in the upper half, pn_same_half(pn, lower) is true and the XPN else-if does not fire either, leaving next_pn_halves unchanged. An attacker that captures the legitimate frame carrying pn == 0xFFFFFFFF on an XPN association can then replay it indefinitely, since lowest_pn never rises above the captured pn and macsec_decrypt() reconstructs the same IV. Extend the XPN else-if to also fire when pn + 1 wraps to 0, so receipt of pn == U32_MAX advances next_pn_halves to (upper + 1, 0). Fixes: a21ecf0e0338 ("macsec: Support XPN frame handling - IEEE 802.1AEbw") Reported-by: Yuhao Jiang Cc: stable@vger.kernel.org Signed-off-by: Junrui Luo Link: https://patch.msgid.link/SYBPR01MB78813FD49E58F253B989F197AF012@SYBPR01MB7881.ausprd01.prod.outlook.com Signed-off-by: Jakub Kicinski --- drivers/net/macsec.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index f904f4d16b45..fb009120a924 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -808,7 +808,8 @@ static bool macsec_post_decrypt(struct sk_buff *skb, struct macsec_secy *secy, u if (pn + 1 > rx_sa->next_pn_halves.lower) { rx_sa->next_pn_halves.lower = pn + 1; } else if (secy->xpn && - !pn_same_half(pn, rx_sa->next_pn_halves.lower)) { + (pn + 1 == 0 || + !pn_same_half(pn, rx_sa->next_pn_halves.lower))) { rx_sa->next_pn_halves.upper++; rx_sa->next_pn_halves.lower = pn + 1; } From 2156a29aecfffa2eb7c558255690084efbe9f3b0 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Wed, 20 May 2026 11:41:57 -0400 Subject: [PATCH 227/411] octeontx2-af: validate body pcifunc in rvu_mbox_handler_rep_event_notify rvu_mbox_handler_rep_event_notify() in drivers/net/ethernet/marvell/ octeontx2/af/rvu_rep.c queues a sender-controlled REP_EVENT_NOTIFY request body verbatim, and rvu_rep_up_notify() then forwards event->pcifunc (the nested body field, distinct from the AF-normalised header pcifunc) into rvu_get_pfvf(), rvu_get_pf() and the AF->PF mailbox device index without any bounds check. A VF attached to a PF that has been put into switchdev representor mode reaches this path: the VF mailbox handler otx2_pfvf_mbox_handler() forwards every message id including MBOX_MSG_REP_EVENT_NOTIFY to AF without an allowlist, and the AF dispatcher rewrites only msg->pcifunc, leaving struct rep_event::pcifunc attacker-controlled. The sibling rvu_mbox_handler_esw_cfg() refuses requests whose header pcifunc is not rvu->rep_pcifunc; this handler has no equivalent gate. An out-of-range body pcifunc selects an &rvu->pf[]/&rvu->hwvf[] element past the allocated array and, for RVU_EVENT_MAC_ADDR_CHANGE, turns into a six-byte attacker-chosen OOB ether_addr_copy() target inside the queued worker; KASAN reports a slab-out-of-bounds write in rvu_rep_wq_handler. Reject malformed requests at the handler entry by gating on is_pf_func_valid(), which is already the canonical PF/VF range check in this driver; expose it via rvu.h so callers in rvu_rep.c can use it instead of open-coding the same range arithmetic. Fixes: b8fea84a0468 ("octeontx2-pf: Add support to sync link state between representor and VFs") Cc: stable@vger.kernel.org Signed-off-by: Michael Bommarito Link: https://patch.msgid.link/20260520154157.1439319-1-michael.bommarito@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/rvu.c | 2 +- drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 1 + drivers/net/ethernet/marvell/octeontx2/af/rvu_rep.c | 8 ++++++++ 3 files changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c index e40b79076358..3cf131508ecf 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c @@ -436,7 +436,7 @@ struct rvu_pfvf *rvu_get_pfvf(struct rvu *rvu, int pcifunc) return &rvu->pf[rvu_get_pf(rvu->pdev, pcifunc)]; } -static bool is_pf_func_valid(struct rvu *rvu, u16 pcifunc) +bool is_pf_func_valid(struct rvu *rvu, u16 pcifunc) { int pf, vf, nvfs; u64 cfg; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index a466181cf908..de3fbd3d15d6 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -917,6 +917,7 @@ u16 rvu_get_rsrc_mapcount(struct rvu_pfvf *pfvf, int blkaddr); struct rvu_pfvf *rvu_get_pfvf(struct rvu *rvu, int pcifunc); void rvu_get_pf_numvfs(struct rvu *rvu, int pf, int *numvfs, int *hwvf); bool is_block_implemented(struct rvu_hwinfo *hw, int blkaddr); +bool is_pf_func_valid(struct rvu *rvu, u16 pcifunc); bool is_pffunc_map_valid(struct rvu *rvu, u16 pcifunc, int blktype); int rvu_get_lf(struct rvu *rvu, struct rvu_block *block, u16 pcifunc, u16 slot); int rvu_lf_reset(struct rvu *rvu, struct rvu_block *block, int lf); diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_rep.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_rep.c index 901f6fd40fd4..a2781e0f504e 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_rep.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_rep.c @@ -97,6 +97,14 @@ int rvu_mbox_handler_rep_event_notify(struct rvu *rvu, struct rep_event *req, { struct rep_evtq_ent *qentry; + /* The mailbox dispatcher normalises only the header pcifunc; the + * nested struct rep_event::pcifunc body field is sender-controlled + * and is later used by rvu_rep_up_notify() to index rvu->pf[] / + * rvu->hwvf[]. Reject out-of-range body selectors before queueing. + */ + if (!is_pf_func_valid(rvu, req->pcifunc)) + return -EINVAL; + qentry = kmalloc_obj(*qentry, GFP_ATOMIC); if (!qentry) return -ENOMEM; From f229426072fc865654a60978bb7fda790a051ff3 Mon Sep 17 00:00:00 2001 From: Luka Gejak Date: Sat, 23 May 2026 15:03:30 +0200 Subject: [PATCH 228/411] net: hsr: fix potential OOB access in supervision frame handling Ensure the entire TLV header is linearized before access by adding sizeof(struct hsr_sup_tlv) to the pskb_may_pull() calls. Without this, a truncated frame could cause an out-of-bounds access. Fixes: eafaa88b3eb7 ("net: hsr: Add support for redbox supervision frames") Signed-off-by: Luka Gejak Reviewed-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20260523130330.61880-1-luka.gejak@linux.dev Signed-off-by: Jakub Kicinski --- net/hsr/hsr_forward.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/hsr/hsr_forward.c b/net/hsr/hsr_forward.c index 0aca859c88cb..f669a226d728 100644 --- a/net/hsr/hsr_forward.c +++ b/net/hsr/hsr_forward.c @@ -84,7 +84,7 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb) /* Get next tlv */ total_length += hsr_sup_tag->tlv.HSR_TLV_length; - if (!pskb_may_pull(skb, total_length)) + if (!pskb_may_pull(skb, total_length + sizeof(struct hsr_sup_tlv))) return false; skb_pull(skb, total_length); hsr_sup_tlv = (struct hsr_sup_tlv *)skb->data; @@ -100,7 +100,7 @@ static bool is_supervision_frame(struct hsr_priv *hsr, struct sk_buff *skb) /* make sure another tlv follows */ total_length += sizeof(struct hsr_sup_tlv) + hsr_sup_tlv->HSR_TLV_length; - if (!pskb_may_pull(skb, total_length)) + if (!pskb_may_pull(skb, total_length + sizeof(struct hsr_sup_tlv))) return false; /* get next tlv */ From 80501dff814eeccebf44a59340c3fe3a205eb120 Mon Sep 17 00:00:00 2001 From: "Borislav Petkov (AMD)" Date: Wed, 20 May 2026 13:25:07 -0700 Subject: [PATCH 229/411] Documentation/arch/x86: Hide clearcpuid= This option was never meant to be used in production because it solely clears the X86_FEATURE kernel-internal representation of what CPUID bits it has detected and doesn't do any *proper* feature disablement like clearing CR4.CET in the user shadow stack case, for example. So remove its documentation so that it doesn't get used in production and people get silly ideas. It is meant strictly for debugging; and if a chicken bit for properly disabling a feature is warranted, then that would need proper enablement. No functional changes. Signed-off-by: Borislav Petkov (AMD) Signed-off-by: Ingo Molnar Cc: Mathias Krause Cc: Linus Torvalds Link: https://patch.msgid.link/20260520202508.160112-1-bp@kernel.org --- .../admin-guide/kernel-parameters.txt | 18 ------------------ Documentation/arch/x86/cpuinfo.rst | 4 ++++ 2 files changed, 4 insertions(+), 18 deletions(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 4d0f545fb3ec..97007f4f69d4 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -789,24 +789,6 @@ Kernel parameters cio_ignore= [S390] See Documentation/arch/s390/common_io.rst for details. - clearcpuid=X[,X...] [X86] - Disable CPUID feature X for the kernel. See - arch/x86/include/asm/cpufeatures.h for the valid bit - numbers X. Note the Linux-specific bits are not necessarily - stable over kernel options, but the vendor-specific - ones should be. - X can also be a string as appearing in the flags: line - in /proc/cpuinfo which does not have the above - instability issue. However, not all features have names - in /proc/cpuinfo. - Note that using this option will taint your kernel. - Also note that user programs calling CPUID directly - or using the feature without checking anything - will still see it. This just prevents it from - being used by the kernel or shown in /proc/cpuinfo. - Also note the kernel might malfunction if you disable - some critical bits. - clk_ignore_unused [CLK] Prevents the clock framework from automatically gating diff --git a/Documentation/arch/x86/cpuinfo.rst b/Documentation/arch/x86/cpuinfo.rst index 9f2e47c4b1c8..17fce95367e6 100644 --- a/Documentation/arch/x86/cpuinfo.rst +++ b/Documentation/arch/x86/cpuinfo.rst @@ -187,6 +187,10 @@ to disable features using the feature number as defined in Protection can be disabled using clearcpuid=514. The number 514 is calculated from #define X86_FEATURE_UMIP (16*32 + 2). +DO NOT USE this cmdline option in production - it is meant to be used only as +a quick'n'dirty debugging aid to rule out a feature-enabling code is the +culprit. If you use it, it'll taint the kernel. + In addition, there exists a variety of custom command-line parameters that disable specific features. The list of parameters includes, but is not limited to, nofsgsbase, nosgx, noxsave, etc. 5-level paging can also be disabled using From 20587302f8d700f26ee2c8a60ffb0a69ae0edf16 Mon Sep 17 00:00:00 2001 From: Zhang Heng Date: Tue, 26 May 2026 09:36:11 +0800 Subject: [PATCH 230/411] ALSA: hda/realtek: Fix speaker output on ASUS ROG Strix G615LP Add quirk for ALC294 codec on ASUS ROG Strix G615LP (SSID 1043:1214) using ALC287_FIXUP_TXNW2781_I2C_ASUS to fix speaker output. Link: https://bugzilla.kernel.org/show_bug.cgi?id=221173 Cc: Signed-off-by: Zhang Heng Link: https://patch.msgid.link/20260526013611.1954949-1-zhangheng@kylinos.cn Signed-off-by: Takashi Iwai --- sound/hda/codecs/realtek/alc269.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/hda/codecs/realtek/alc269.c b/sound/hda/codecs/realtek/alc269.c index b4c9a5584a04..212adce0f8e6 100644 --- a/sound/hda/codecs/realtek/alc269.c +++ b/sound/hda/codecs/realtek/alc269.c @@ -7324,6 +7324,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x1043, 0x11c0, "ASUS X556UR", ALC255_FIXUP_ASUS_MIC_NO_PRESENCE), HDA_CODEC_QUIRK(0x1043, 0x1204, "ASUS Strix G16 G615JMR", ALC287_FIXUP_TXNW2781_I2C_ASUS), SND_PCI_QUIRK(0x1043, 0x1204, "ASUS Strix G615JHR_JMR_JPR", ALC287_FIXUP_TAS2781_I2C), + HDA_CODEC_QUIRK(0x1043, 0x1214, "ASUS ROG Strix G615LP", ALC287_FIXUP_TXNW2781_I2C_ASUS), SND_PCI_QUIRK(0x1043, 0x1214, "ASUS Strix G615LH_LM_LP", ALC287_FIXUP_TAS2781_I2C), SND_PCI_QUIRK(0x1043, 0x125e, "ASUS Q524UQK", ALC255_FIXUP_ASUS_MIC_NO_PRESENCE), SND_PCI_QUIRK(0x1043, 0x1271, "ASUS X430UN", ALC256_FIXUP_ASUS_MIC_NO_PRESENCE), From 44e151be23deb788d9f6124de93823faf6e04e99 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 25 May 2026 10:14:42 +0300 Subject: [PATCH 231/411] accel/ivpu: prevent uninitialized data bug in debugfs The simple_write_to_buffer() will only initialize data starting from the *pos offset so if it's non-zero then the first part of the buffer uninitialized. Really, if *pos is non-zero then this code won't work so just check for that at the start of the function. Fixes: 320323d2e545 ("accel/ivpu: Add debugfs interface for setting HWS priority bands") Signed-off-by: Dan Carpenter Reviewed-by: Karol Wachowski Signed-off-by: Karol Wachowski Link: https://patch.msgid.link/ahP24m6Mii9EDL7Q@stanley.mountain --- drivers/accel/ivpu/ivpu_debugfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ivpu/ivpu_debugfs.c b/drivers/accel/ivpu/ivpu_debugfs.c index 189dbe94cf14..dc20bc73c6ed 100644 --- a/drivers/accel/ivpu/ivpu_debugfs.c +++ b/drivers/accel/ivpu/ivpu_debugfs.c @@ -450,7 +450,7 @@ priority_bands_fops_write(struct file *file, const char __user *user_buf, size_t u32 band; int ret; - if (size >= sizeof(buf)) + if (*pos != 0 || size >= sizeof(buf)) return -EINVAL; ret = simple_write_to_buffer(buf, sizeof(buf) - 1, pos, user_buf, size); From d64b0372760e09de6d18a0616d7bc652c8c6891d Mon Sep 17 00:00:00 2001 From: George Guo Date: Sat, 9 May 2026 10:44:15 +0800 Subject: [PATCH 232/411] kho: fix KHO_TREE_MAX_DEPTH for non-4KB page sizes KHO_TREE_MAX_DEPTH is calculated as: DIV_ROUND_UP(KHO_ORDER_0_LOG2 - KHO_BITMAP_SIZE_LOG2, KHO_TABLE_SIZE_LOG2) + 1 For systems with 16KB pages (e.g. arm64 with CONFIG_ARM64_16K_PAGES=y or LoongArch), this gives a depth of 4. Since levels are 0 based, with depth = 4 the effective top level is 3 and the top-level shift at bit 39. PAGE_SHIFT = 14 KHO_BITMAP_SIZE_LOG2 = PAGE_SHIFT + 3 = 17 KHO_TABLE_SIZE_LOG2 = log(2; (1 << PAGE_SHIFT) / 8) = 11 shift = ((3 - 1) * KHO_TABLE_SIZE_LOG2) + KHO_BITMAP_SIZE_LOG2 = 39 The order-0 bit sits at bit 50 (KHO_ORDER_0_LOG2 = 64 - PAGE_SHIFT = 50). When inserting or reading a key, the index extracted at the top level is: (1 << 50) >> 39 = 2048 2048 is exactly the table size (PAGE_SIZE / sizeof(phys_addr_t) = 2048 for 16KB pages), so it wraps to 0, aliasing the order bit to index 0 and losing it silently. On the second kernel, kho_radix_decode_key() sees a key without the order bit, calls fls64() on the wrong bit, computes a wrong order and thus a garbage physical address. phys_to_page() of that address faults in kho_preserved_memory_reserve(), causing a kernel panic early in boot. Fix by adding +1 to the DIV_ROUND_UP numerator so the formula accounts for the order bit itself, giving depth 5 for 16KB pages. The top-level shift becomes 50, and (1 << 50) >> 50 = 1, which is nonzero and unambiguous. For 4KB and 64KB page sizes the depth is unchanged. Link: https://patch.msgid.link/20260509024415.33190-1-dongtai.guo@linux.dev Fixes: 3f2ad90060f6 ("kho: adopt radix tree for preserved memory tracking") Tested-by: Kexin Liu Signed-off-by: George Guo Reviewed-by: Pasha Tatashin [rppt: added actual math to the changelog] Signed-off-by: Mike Rapoport (Microsoft) --- include/linux/kho/abi/kexec_handover.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/kho/abi/kexec_handover.h b/include/linux/kho/abi/kexec_handover.h index 7e847a2339b0..db9bda6dd310 100644 --- a/include/linux/kho/abi/kexec_handover.h +++ b/include/linux/kho/abi/kexec_handover.h @@ -274,7 +274,7 @@ enum kho_radix_consts { * and 1 bitmap level. */ KHO_TREE_MAX_DEPTH = - DIV_ROUND_UP(KHO_ORDER_0_LOG2 - KHO_BITMAP_SIZE_LOG2, + DIV_ROUND_UP(KHO_ORDER_0_LOG2 - KHO_BITMAP_SIZE_LOG2 + 1, KHO_TABLE_SIZE_LOG2) + 1, }; From 8fd2f26fa2a33cfe8ac043f976137ecf5b567f03 Mon Sep 17 00:00:00 2001 From: "Pratyush Yadav (Google)" Date: Tue, 19 May 2026 15:33:30 +0200 Subject: [PATCH 233/411] kho: fix order calculation for kho_unpreserve_pages() Commit 91e74fa8b1bc ("kho: make sure preservations do not span multiple NUMA nodes") made sure preservations from kho_preserve_pages() do not span multiple NUMA nodes. If they do, the order is reduced and tried again. The same logic was not implemented for kho_unpreserve_pages(). This can result in unpreserve calculating a different order than preserve, and thus not actually unpreserving the pages. Fix this by moving the order calculation logic to __kho_preserve_pages_order() and use it from both preserve and unpreserve paths. Move __kho_unpreserve() down to avoid having a forward declaration. Its users are further down in the file anyway. Also, it results in grouping for all the page-level preservation and unpreservation functions. This unfortunately makes the diff hard to read, but the main change in __kho_unpreserve() is to call __kho_preserve_pages_order() instead of open-coding the order calculation. Fixes: 91e74fa8b1bc ("kho: make sure preservations do not span multiple NUMA nodes") Cc: stable@vger.kernel.org Signed-off-by: Pratyush Yadav (Google) Reviewed-by: Samiullah Khawaja Reviewed-by: Pasha Tatashin Link: https://patch.msgid.link/20260519133332.2498092-1-pratyush@kernel.org Signed-off-by: Mike Rapoport (Microsoft) --- kernel/liveupdate/kexec_handover.c | 56 +++++++++++++++++------------- 1 file changed, 32 insertions(+), 24 deletions(-) diff --git a/kernel/liveupdate/kexec_handover.c b/kernel/liveupdate/kexec_handover.c index 2592f7ca16e2..1b592d86dc48 100644 --- a/kernel/liveupdate/kexec_handover.c +++ b/kernel/liveupdate/kexec_handover.c @@ -357,20 +357,6 @@ int kho_radix_walk_tree(struct kho_radix_tree *tree, } EXPORT_SYMBOL_GPL(kho_radix_walk_tree); -static void __kho_unpreserve(struct kho_radix_tree *tree, - unsigned long pfn, unsigned long end_pfn) -{ - unsigned int order; - - while (pfn < end_pfn) { - order = min(count_trailing_zeros(pfn), ilog2(end_pfn - pfn)); - - kho_radix_del_page(tree, pfn, order); - - pfn += 1 << order; - } -} - /* For physically contiguous 0-order pages. */ static void kho_init_pages(struct page *page, unsigned long nr_pages) { @@ -860,6 +846,37 @@ void kho_unpreserve_folio(struct folio *folio) } EXPORT_SYMBOL_GPL(kho_unpreserve_folio); +static unsigned int __kho_preserve_pages_order(unsigned long start_pfn, + unsigned long end_pfn) +{ + unsigned int order = min(count_trailing_zeros(start_pfn), + ilog2(end_pfn - start_pfn)); + + /* + * Make sure all the pages in a single preservation are in the same NUMA + * node. The restore machinery can not cope with a preservation spanning + * multiple NUMA nodes. + */ + while (pfn_to_nid(start_pfn) != pfn_to_nid(start_pfn + (1UL << order) - 1)) + order--; + + return order; +} + +static void __kho_unpreserve(struct kho_radix_tree *tree, + unsigned long pfn, unsigned long end_pfn) +{ + unsigned int order; + + while (pfn < end_pfn) { + order = __kho_preserve_pages_order(pfn, end_pfn); + + kho_radix_del_page(tree, pfn, order); + + pfn += 1 << order; + } +} + /** * kho_preserve_pages - preserve contiguous pages across kexec * @page: first page in the list. @@ -885,16 +902,7 @@ int kho_preserve_pages(struct page *page, unsigned long nr_pages) } while (pfn < end_pfn) { - unsigned int order = - min(count_trailing_zeros(pfn), ilog2(end_pfn - pfn)); - - /* - * Make sure all the pages in a single preservation are in the - * same NUMA node. The restore machinery can not cope with a - * preservation spanning multiple NUMA nodes. - */ - while (pfn_to_nid(pfn) != pfn_to_nid(pfn + (1UL << order) - 1)) - order--; + unsigned int order = __kho_preserve_pages_order(pfn, end_pfn); err = kho_radix_add_page(tree, pfn, order); if (err) { From dac917ed5aead741004db8d0d5151dd577802df8 Mon Sep 17 00:00:00 2001 From: Alexander Stein Date: Tue, 26 May 2026 08:35:01 +0200 Subject: [PATCH 234/411] gpio: mxc: fix irq_high handling If port->irq_high is -1 (fsl,imx21-gpio compatible) and gpio_idx is >= 16 enable_irq_wake() is called with -1 which is wrong. Fixes: 5f6d1998adeb ("gpio: mxc: release the parent IRQ in runtime suspend") Signed-off-by: Alexander Stein Reviewed-by: Frank Li Link: https://patch.msgid.link/20260526063504.25916-1-alexander.stein@ew.tq-group.com Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-mxc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpio/gpio-mxc.c b/drivers/gpio/gpio-mxc.c index 647b6f4861b7..12f11a6c9665 100644 --- a/drivers/gpio/gpio-mxc.c +++ b/drivers/gpio/gpio-mxc.c @@ -469,7 +469,7 @@ static int mxc_gpio_probe(struct platform_device *pdev) * the handler is needed only once, but doing it for every port * is more robust and easier. */ - port->irq_high = -1; + port->irq_high = 0; port->mx_irq_handler = mx2_gpio_irq_handler; } else port->mx_irq_handler = mx3_gpio_irq_handler; From 5c4063c87a619e4df954c179d24628636f5db15f Mon Sep 17 00:00:00 2001 From: Janusz Krzysztofik Date: Fri, 8 May 2026 14:23:51 +0200 Subject: [PATCH 235/411] drm/i915: Fix potential UAF in TTM object purge MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit TLDR: The bo->ttm object might be changed by calling ttm_bo_validate(), move casting it to an i915_tt object later to actually get the right pointer. A user reported hitting the following bug under heavy use on DG2: [26620.095550] Oops: general protection fault, probably for non-canonical address 0xa56b6b6b6b6b6b8b: 0000 1 SMP NOPTI [26620.095556] CPU: 2 UID: 0 PID: 631 Comm: Xorg Not tainted 6.18.8 #1 PREEMPT(lazy) [26620.095558] Hardware name: ASRock B850M Steel Legend WiFi/B850M Steel Legend WiFi, BIOS 3.50 09/18/2025 [26620.095559] RIP: 0010:i915_ttm_purge+0x84/0x100 [i915] [26620.095604] Code: 00 00 00 48 8d 54 24 10 48 89 e6 48 89 fb e8 83 aa ae ff 85 c0 75 6f 48 83 bb a8 01 00 00 00 74 2c 48 8b 45 78 48 85 c0 74 23 <48> 8b 78 20 48 c7 c2 ff ff ff ff 31 f6 e8 7a 73 e3 e0 48 8b 7d 78 [26620.095605] RSP: 0018:ffffc90005fd7430 EFLAGS: 00010282 [26620.095607] RAX: a56b6b6b6b6b6b6b RBX: ffff8881f46c3dc0 RCX: 0000000000000000 [26620.095608] RDX: 0000000000000000 RSI: 0000000000000246 RDI: 00000000ffffffff [26620.095609] RBP: ffff888289610f00 R08: 0000000000000001 R09: ffff88823b022000 [26620.095609] R10: ffff888103029b28 R11: ffff8881fc7f3800 R12: ffff88810b6150d0 [26620.095609] R13: ffff888289610f00 R14: 0000000000000000 R15: ffff8881f46c3dc0 [26620.095610] FS: 00007f1004d86900(0000) GS:ffff88901c858000(0000) knlGS:0000000000000000 [26620.095611] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [26620.095611] CR2: 00007f0fdf489000 CR3: 000000035b0c1000 CR4: 0000000000750ef0 [26620.095612] PKRU: 55555554 [26620.095612] Call Trace: [26620.095615] [26620.095615] i915_ttm_move+0x2b9/0x420 [i915] [26620.095642] ? ttm_tt_init+0x65/0x80 [ttm] [26620.095644] ? i915_ttm_tt_create+0xc6/0x150 [i915] [26620.095667] ttm_bo_handle_move_mem+0xb6/0x160 [ttm] [26620.095669] ttm_bo_evict+0x100/0x150 [ttm] [26620.095671] ? preempt_count_add+0x64/0xa0 [26620.095673] ? _raw_spin_lock+0xe/0x30 [26620.095675] ? _raw_spin_unlock+0xd/0x30 [26620.095675] ? i915_gem_object_evictable+0xb7/0xd0 [i915] [26620.095704] ttm_bo_evict_cb+0x6e/0xd0 [ttm] [26620.095705] ttm_lru_walk_for_evict+0xa6/0x200 [ttm] [26620.095708] ttm_bo_alloc_resource+0x185/0x4f0 [ttm] [26620.095709] ? init_object+0x62/0xd0 [26620.095712] ttm_bo_validate+0x7a/0x180 [ttm] [26620.095713] ? _raw_spin_unlock_irqrestore+0x16/0x30 [26620.095714] __i915_ttm_get_pages+0xb0/0x170 [i915] [26620.095737] i915_ttm_get_pages+0x9f/0x150 [i915] [26620.095759] ? i915_gem_do_execbuffer+0xedc/0x2b40 [i915] [26620.095786] ? alloc_debug_processing+0xd0/0x100 [26620.095787] ? _raw_spin_unlock_irqrestore+0x16/0x30 [26620.095788] ? i915_vma_instance+0xa0/0x4e0 [i915] [26620.095822] __i915_gem_object_get_pages+0x2f/0x40 [i915] [26620.095848] i915_vma_pin_ww+0x706/0x980 [i915] [26620.095875] ? i915_gem_do_execbuffer+0xedc/0x2b40 [i915] [26620.095904] eb_validate_vmas+0x170/0xa00 [i915] [26620.095930] i915_gem_do_execbuffer+0x1201/0x2b40 [i915] [26620.095953] ? alloc_debug_processing+0xd0/0x100 [26620.095954] ? _raw_spin_unlock_irqrestore+0x16/0x30 [26620.095955] ? i915_gem_execbuffer2_ioctl+0xc9/0x240 [i915] [26620.095977] ? __wake_up_sync_key+0x32/0x50 [26620.095979] ? i915_gem_execbuffer2_ioctl+0xc9/0x240 [i915] [26620.096001] ? __slab_alloc.isra.0+0x67/0xc0 [26620.096003] i915_gem_execbuffer2_ioctl+0x11a/0x240 [i915] Results from decode_stacktrace.sh pointed to dereference of a file pointer field of a i915 TTM page vector container associated with an object being purged on eviction. That path is taken when the object is marked as no longer needed. Code analysis revealed a possibility of the i915 TTM page vector container being replaced with a new instance inside a function that purges content of the object, should it be still busy. That function is called, indirectly via a more general function that changes the object's placement and caching policy, before the problematic dereference, but still after a pointer to the container is captured, rendering the pointer no longer valid. Fix the issue by capturing the pointer to the container only after its potential replacement. v2: Move the container_of() inside the if block (Sebastian), - a simplified version of the commit description that explains briefly why the change is necessary (Christian). Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/work_items/14882 Fixes: 7ae034590ceae ("drm/i915/ttm: add tt shmem backend") Signed-off-by: Janusz Krzysztofik Cc: stable@vger.kernel.org # v5.17+ Cc: Matthew Auld Cc: Thomas Hellström Cc: Sebastian Brzezinka Cc: Christian König Reviewed-by: Andi Shyti Reviewed-by: Christian König Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/20260508122612.469227-2-janusz.krzysztofik@linux.intel.com (cherry picked from commit 4462966a93eb185849b7f174f0d0de53476d00a4) Signed-off-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/gem/i915_gem_ttm.c | 28 ++++++++++++++----------- 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index de70517b4ef2..df3fcc2b1248 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -419,8 +419,6 @@ void i915_ttm_free_cached_io_rsgt(struct drm_i915_gem_object *obj) int i915_ttm_purge(struct drm_i915_gem_object *obj) { struct ttm_buffer_object *bo = i915_gem_to_ttm(obj); - struct i915_ttm_tt *i915_tt = - container_of(bo->ttm, typeof(*i915_tt), ttm); struct ttm_operation_ctx ctx = { .interruptible = true, .no_wait_gpu = false, @@ -435,16 +433,22 @@ int i915_ttm_purge(struct drm_i915_gem_object *obj) if (ret) return ret; - if (bo->ttm && i915_tt->filp) { - /* - * The below fput(which eventually calls shmem_truncate) might - * be delayed by worker, so when directly called to purge the - * pages(like by the shrinker) we should try to be more - * aggressive and release the pages immediately. - */ - shmem_truncate_range(file_inode(i915_tt->filp), - 0, (loff_t)-1); - fput(fetch_and_zero(&i915_tt->filp)); + if (bo->ttm) { + struct i915_ttm_tt *i915_tt = + container_of(bo->ttm, typeof(*i915_tt), ttm); + + if (i915_tt->filp) { + /* + * The below fput(which eventually calls shmem_truncate) + * might be delayed by worker, so when directly called + * to purge the pages(like by the shrinker) we should + * try to be more aggressive and release the pages + * immediately. + */ + shmem_truncate_range(file_inode(i915_tt->filp), + 0, (loff_t)-1); + fput(fetch_and_zero(&i915_tt->filp)); + } } obj->write_domain = 0; From 202e77cf2e839e1adc804433322dc5c9ee511c9f Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Micha=C5=82=20Grzelak?= Date: Thu, 16 Apr 2026 18:37:44 +0200 Subject: [PATCH 236/411] drm/i915/aux: use polling when irqs are unavailable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit PTL with physically disconnected display was observed to have 40s longer execution time when testing xe_fault_injection@xe_guc_mmio_send_recv. The issue has not been seen when reverting commit 40a9f77a28fa ("Revert "drm/i915/dp: change aux_ctl reg read to polling read""). Apparently the configuration suffers from not having AUX enabled when using interrupts. One probable cause can be xe enabling interrupts too late: interrupts need memory allocations which currently can't be done before the display FB takeover is done. As for now, use polling for AUX in case interrupts are unavailable. Fixes: 40a9f77a28fa ("Revert "drm/i915/dp: change aux_ctl reg read to polling read"") Suggested-by: Ville Syrjälä Signed-off-by: Michał Grzelak Signed-off-by: Ville Syrjälä Link: https://patch.msgid.link/20260416163744.288107-1-michal.grzelak@intel.com (cherry picked from commit 05e0550b65cd1604bd515fbc65f522bce4c10a87) Signed-off-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/display/intel_dp_aux.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux.c b/drivers/gpu/drm/i915/display/intel_dp_aux.c index b20ec3e589fa..9c9b6410366d 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux.c @@ -12,6 +12,7 @@ #include "intel_dp.h" #include "intel_dp_aux.h" #include "intel_dp_aux_regs.h" +#include "intel_parent.h" #include "intel_pps.h" #include "intel_quirks.h" #include "intel_tc.h" @@ -60,18 +61,29 @@ intel_dp_aux_wait_done(struct intel_dp *intel_dp) struct intel_display *display = to_intel_display(intel_dp); i915_reg_t ch_ctl = intel_dp->aux_ch_ctl_reg(intel_dp); const unsigned int timeout_ms = 10; + bool done = true; u32 status; - bool done; + int ret; + if (intel_parent_irq_enabled(display)) { #define C (((status = intel_de_read_notrace(display, ch_ctl)) & DP_AUX_CH_CTL_SEND_BUSY) == 0) - done = wait_event_timeout(display->gmbus.wait_queue, C, - msecs_to_jiffies_timeout(timeout_ms)); + done = wait_event_timeout(display->gmbus.wait_queue, C, + msecs_to_jiffies_timeout(timeout_ms)); + +#undef C + } else { + ret = intel_de_wait_ms(display, ch_ctl, + DP_AUX_CH_CTL_SEND_BUSY, 0, + timeout_ms, &status); + + if (ret == -ETIMEDOUT) + done = false; + } if (!done) drm_err(display->drm, "%s: did not complete or timeout within %ums (status 0x%08x)\n", intel_dp->aux.name, timeout_ms, status); -#undef C return status; } From d196136a988051173f68f91de0b5a1bd32122dd7 Mon Sep 17 00:00:00 2001 From: Pranay Samala Date: Tue, 19 May 2026 13:23:08 +0530 Subject: [PATCH 237/411] drm/i915/color: Fix HDR pre-CSC LUT programming loop The integer lut programming loop never executes completely due to incorrect condition (i++ > 130). Fix to properly program 129th+ entries for values > 1.0. Cc: #v6.19 Fixes: 82caa1c8813f ("drm/i915/color: Program Pre-CSC registers") Signed-off-by: Pranay Samala Signed-off-by: Chaitanya Kumar Borah Reviewed-by: Uma Shankar Signed-off-by: Suraj Kandpal Link: https://patch.msgid.link/20260519075308.383877-1-pranay.samala@intel.com (cherry picked from commit f33862ec3e8849ad7c0a3dd46719083b13ade248) Signed-off-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/display/intel_color.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/display/intel_color.c b/drivers/gpu/drm/i915/display/intel_color.c index e7950655434b..6d1cffc6d2be 100644 --- a/drivers/gpu/drm/i915/display/intel_color.c +++ b/drivers/gpu/drm/i915/display/intel_color.c @@ -3976,7 +3976,7 @@ xelpd_program_plane_pre_csc_lut(struct intel_dsb *dsb, intel_de_write_dsb(display, dsb, PLANE_PRE_CSC_GAMC_DATA_ENH(pipe, plane, 0), (1 << 24)); - } while (i++ > 130); + } while (i++ < 130); } else { for (i = 0; i < lut_size; i++) { u32 v = (i * ((1 << 24) - 1)) / (lut_size - 1); From 8bb9093df555f9e89fdbe1405118b11384c03e04 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jouni=20H=C3=B6gander?= Date: Wed, 20 May 2026 13:49:43 +0300 Subject: [PATCH 238/411] drm/i915/psr: Block DC states on vblank enable when Panel Replay supported MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Currently we are blocking DC states only when Panel Replay is enabled on vblank enable. It may happen that Panel Replay is getting enabled when vblank is already enabled. Fix this by blocking DC states always if Panel Replay is supported. While at it take care of possible dual eDP case by looping all encoders supporting PSR. Fixes: 0c427ac78a1d ("drm/i915/psr: Add interface to notify PSR of vblank enable/disable") Cc: # v6.16+ Signed-off-by: Jouni Högander Reviewed-by: Michał Grzelak Link: https://patch.msgid.link/20260520104944.239797-1-jouni.hogander@intel.com (cherry picked from commit eb5911f990554f7ce947dd53df00c114362e4465) Signed-off-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/display/intel_psr.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 29904a037575..bd5a8c6ac6ef 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -4151,32 +4151,33 @@ void intel_psr_notify_vblank_enable_disable(struct intel_display *display, bool enable) { struct intel_encoder *encoder; + bool block_dc_states = false; for_each_intel_encoder_with_psr(display->drm, encoder) { struct intel_dp *intel_dp = enc_to_intel_dp(encoder); mutex_lock(&intel_dp->psr.lock); - if (intel_dp->psr.panel_replay_enabled) { - mutex_unlock(&intel_dp->psr.lock); - break; - } + if (CAN_PANEL_REPLAY(intel_dp)) + block_dc_states = true; - if (intel_dp->psr.enabled && intel_dp->psr.pkg_c_latency_used) + if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled && + intel_dp->psr.pkg_c_latency_used) intel_psr_apply_underrun_on_idle_wa_locked(intel_dp); mutex_unlock(&intel_dp->psr.lock); - return; } /* * NOTE: intel_display_power_set_target_dc_state is used - * only by PSR * code for DC3CO handling. DC3CO target + * only by PSR code for DC3CO handling. DC3CO target * state is currently disabled in * PSR code. If DC3CO * is taken into use we need take that into account here * as well. */ - intel_display_power_set_target_dc_state(display, enable ? DC_STATE_DISABLE : - DC_STATE_EN_UPTO_DC6); + if (block_dc_states) + intel_display_power_set_target_dc_state(display, enable ? + DC_STATE_DISABLE : + DC_STATE_EN_UPTO_DC6); } static void From 3549a9649dc7c5fc586ab12f675279283cdcb2a7 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jouni=20H=C3=B6gander?= Date: Wed, 20 May 2026 13:49:44 +0300 Subject: [PATCH 239/411] drm/i915/psr: Use DC_OFF wake reference to block DC6 on vblank enable MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We are observing following warnings: *ERROR* power well DC_off state mismatch (refcount 0/enabled 1) gen9_dc_off_power_well_enabled is considering target state DC_STATE_DISABLE as DC_OFF power well being enabled. Fix this by using wakeref for the purpose. To achieve this we need to modify notification code as well. Currently it is possible that PSR gets notified vblank enable/disable twice on same status. This is currently not a problem as it is just triggering call to intel_display_power_set_target_dc_state with same target state as a parameter. When using wakeref this becomes a problem due to reference counting. Fix this storing vbank status on last notification and use that to ensure there are no more than one notification with same vblank status. v2: ensure there is no subsequent notifications with same status Fixes: aa451abcffb5 ("drm/i915/display: Prevent DC6 while vblank is enabled for Panel Replay") Cc: # v6.13+ Signed-off-by: Jouni Högander Reviewed-by: Michał Grzelak Link: https://patch.msgid.link/20260520104944.239797-2-jouni.hogander@intel.com (cherry picked from commit 35485ac56d878192a3829a58cb26503125ec7104) Signed-off-by: Tvrtko Ursulin --- .../gpu/drm/i915/display/intel_display_core.h | 1 + .../gpu/drm/i915/display/intel_display_irq.c | 8 +++++-- .../drm/i915/display/intel_display_types.h | 2 ++ drivers/gpu/drm/i915/display/intel_psr.c | 24 +++++++------------ 4 files changed, 18 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_display_core.h b/drivers/gpu/drm/i915/display/intel_display_core.h index d9baca2d5aaf..78afcd42f44c 100644 --- a/drivers/gpu/drm/i915/display/intel_display_core.h +++ b/drivers/gpu/drm/i915/display/intel_display_core.h @@ -497,6 +497,7 @@ struct intel_display { u8 vblank_enabled; int vblank_enable_count; + bool vblank_status_last_notified; struct work_struct vblank_notify_work; diff --git a/drivers/gpu/drm/i915/display/intel_display_irq.c b/drivers/gpu/drm/i915/display/intel_display_irq.c index 70c1bba7c0a8..aedf3928a089 100644 --- a/drivers/gpu/drm/i915/display/intel_display_irq.c +++ b/drivers/gpu/drm/i915/display/intel_display_irq.c @@ -1773,8 +1773,12 @@ static void intel_display_vblank_notify_work(struct work_struct *work) struct intel_display *display = container_of(work, typeof(*display), irq.vblank_notify_work); int vblank_enable_count = READ_ONCE(display->irq.vblank_enable_count); + bool vblank_status = !!vblank_enable_count; - intel_psr_notify_vblank_enable_disable(display, vblank_enable_count); + if (display->irq.vblank_status_last_notified != vblank_status) { + intel_psr_notify_vblank_enable_disable(display, vblank_status); + display->irq.vblank_status_last_notified = vblank_status; + } } int bdw_enable_vblank(struct drm_crtc *_crtc) @@ -1787,10 +1791,10 @@ int bdw_enable_vblank(struct drm_crtc *_crtc) if (gen11_dsi_configure_te(crtc, true)) return 0; + spin_lock_irqsave(&display->irq.lock, irqflags); if (crtc->vblank_psr_notify && display->irq.vblank_enable_count++ == 0) schedule_work(&display->irq.vblank_notify_work); - spin_lock_irqsave(&display->irq.lock, irqflags); bdw_enable_pipe_irq(display, pipe, GEN8_PIPE_VBLANK); spin_unlock_irqrestore(&display->irq.lock, irqflags); diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 9c7c357afb09..2e6a85708555 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -1790,6 +1790,8 @@ struct intel_psr { u8 active_non_psr_pipes; const char *no_psr_reason; + + struct ref_tracker *vblank_wakeref; }; struct intel_dp { diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index bd5a8c6ac6ef..598fe769a402 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -4151,14 +4151,20 @@ void intel_psr_notify_vblank_enable_disable(struct intel_display *display, bool enable) { struct intel_encoder *encoder; - bool block_dc_states = false; for_each_intel_encoder_with_psr(display->drm, encoder) { struct intel_dp *intel_dp = enc_to_intel_dp(encoder); mutex_lock(&intel_dp->psr.lock); - if (CAN_PANEL_REPLAY(intel_dp)) - block_dc_states = true; + if (CAN_PANEL_REPLAY(intel_dp)) { + if (enable) + intel_dp->psr.vblank_wakeref = + intel_display_power_get(display, + POWER_DOMAIN_DC_OFF); + else + intel_display_power_put(display, POWER_DOMAIN_DC_OFF, + intel_dp->psr.vblank_wakeref); + } if (intel_dp->psr.enabled && !intel_dp->psr.panel_replay_enabled && intel_dp->psr.pkg_c_latency_used) @@ -4166,18 +4172,6 @@ void intel_psr_notify_vblank_enable_disable(struct intel_display *display, mutex_unlock(&intel_dp->psr.lock); } - - /* - * NOTE: intel_display_power_set_target_dc_state is used - * only by PSR code for DC3CO handling. DC3CO target - * state is currently disabled in * PSR code. If DC3CO - * is taken into use we need take that into account here - * as well. - */ - if (block_dc_states) - intel_display_power_set_target_dc_state(display, enable ? - DC_STATE_DISABLE : - DC_STATE_EN_UPTO_DC6); } static void From 7f83d174073234839aea176f265e517e0d50a1d2 Mon Sep 17 00:00:00 2001 From: Shaomin Chen Date: Thu, 21 May 2026 02:07:23 +0800 Subject: [PATCH 240/411] xfrm: iptfs: reset runtime state when cloning SAs iptfs_clone_state() clones the IPTFS mode data with kmemdup(). This copies runtime objects which must not be shared with the original SA, including the embedded sk_buff_head, hrtimers, spinlock, and in-flight reassembly/reorder state. If xfrm_state_migrate() fails after clone_state() but before the later init_state() call has reinitialized those fields, the cloned state can be destroyed by xfrm_state_gc_task() with list and timer state copied from the original SA. With queued packets this lets the clone splice and free skbs owned by the original IPTFS queue, leading to use-after-free and double-free reports in iptfs_destroy_state() and skb release paths. Reinitialize the clone's runtime state before publishing it through x->mode_data. Because clone_state() now publishes a destroyable mode_data object before init_state(), take the mode callback module reference there. Avoid taking it again from __iptfs_init_state() for the same object. Fixes: 0e4fbf013fa5 ("xfrm: iptfs: add user packet (tunnel ingress) handling") Cc: stable@vger.kernel.org Signed-off-by: Shaomin Chen Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_iptfs.c | 28 +++++++++++++++++++++++----- 1 file changed, 23 insertions(+), 5 deletions(-) diff --git a/net/xfrm/xfrm_iptfs.c b/net/xfrm/xfrm_iptfs.c index 97bc979e55ba..6c6bbc040517 100644 --- a/net/xfrm/xfrm_iptfs.c +++ b/net/xfrm/xfrm_iptfs.c @@ -2650,7 +2650,8 @@ static void __iptfs_init_state(struct xfrm_state *x, x->props.enc_hdr_len = sizeof(struct ip_iptfs_hdr); /* Always keep a module reference when x->mode_data is set */ - __module_get(x->mode_cbs->owner); + if (x->mode_data != xtfs) + __module_get(x->mode_cbs->owner); x->mode_data = xtfs; xtfs->x = x; @@ -2658,22 +2659,39 @@ static void __iptfs_init_state(struct xfrm_state *x, static int iptfs_clone_state(struct xfrm_state *x, struct xfrm_state *orig) { + struct skb_wseq *w_saved = NULL; struct xfrm_iptfs_data *xtfs; xtfs = kmemdup(orig->mode_data, sizeof(*xtfs), GFP_KERNEL); if (!xtfs) return -ENOMEM; - xtfs->ra_newskb = NULL; if (xtfs->cfg.reorder_win_size) { - xtfs->w_saved = kzalloc_objs(*xtfs->w_saved, - xtfs->cfg.reorder_win_size); - if (!xtfs->w_saved) { + w_saved = kzalloc_objs(*w_saved, xtfs->cfg.reorder_win_size); + if (!w_saved) { kfree_sensitive(xtfs); return -ENOMEM; } } + xtfs->w_saved = w_saved; + __skb_queue_head_init(&xtfs->queue); + xtfs->queue_size = 0; + hrtimer_setup(&xtfs->iptfs_timer, iptfs_delay_timer, CLOCK_MONOTONIC, + IPTFS_HRTIMER_MODE); + + spin_lock_init(&xtfs->drop_lock); + hrtimer_setup(&xtfs->drop_timer, iptfs_drop_timer, CLOCK_MONOTONIC, + IPTFS_HRTIMER_MODE); + + xtfs->w_seq_set = false; + xtfs->w_wantseq = 0; + xtfs->w_savedlen = 0; + xtfs->ra_newskb = NULL; + xtfs->ra_wantseq = 0; + xtfs->ra_runtlen = 0; + + __module_get(x->mode_cbs->owner); x->mode_data = xtfs; xtfs->x = x; From 3e52417318473782012b236d0325bf7d2266a597 Mon Sep 17 00:00:00 2001 From: Usama Arif Date: Thu, 21 May 2026 03:29:26 -0700 Subject: [PATCH 241/411] xfrm: move policy_bydst RCU sync from per-netns .exit to .pre_exit The struct pernet_operations docstring in include/net/net_namespace.h explicitly warns against blocking RCU primitives in .exit handlers: Exit methods using blocking RCU primitives, such as synchronize_rcu(), should be implemented via exit_batch. [...] Please, avoid synchronize_rcu() at all, where it's possible. Note that a combination of pre_exit() and exit() can be used, since a synchronize_rcu() is guaranteed between the calls. xfrm_policy_fini() violates this: it calls synchronize_rcu() before freeing the policy_bydst hash tables (so no RCU reader is mid- traversal at free time), but runs from xfrm_net_ops.exit -- once per namespace -- so a cleanup_net() of N namespaces pays N full RCU grace periods serially. Use the documented pre_exit/exit split. Move the policy flush (and the workqueue drains it depends on) into a new .pre_exit handler; xfrm_policy_fini() then runs in .exit and frees the hash tables after the synchronize_rcu_expedited() that cleanup_net() guarantees between the two phases. Providing O(1) RCU grace periods per batch instead of O(N). Observed on Linux 6.18 with a workload doing unshare(CLONE_NEWNET) at ~13/sec sustained: cleanup_net() and the netns_wq rescuer kthread both stuck in xfrm_policy_fini()'s synchronize_rcu(), >300k struct net accumulated in the cleanup queue, Percpu in /proc/meminfo climbed to 130+ GB on 256-CPU hosts, and memcg OOMs followed. setup_net and __put_net counts were balanced, ruling out a refcount leak. Fixes: 069daad4f2ae ("xfrm: Wait for RCU readers during policy netns exit") Signed-off-by: Usama Arif Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 15 ++++++++------- 1 file changed, 8 insertions(+), 7 deletions(-) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 59968dcbafe1..dd09d2063da2 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4276,21 +4276,21 @@ static int __net_init xfrm_policy_init(struct net *net) return -ENOMEM; } -static void xfrm_policy_fini(struct net *net) +static void __net_exit xfrm_net_pre_exit(struct net *net) { - struct xfrm_pol_inexact_bin *b, *t; - unsigned int sz; - int dir; - disable_work_sync(&net->xfrm.policy_hthresh.work); - flush_work(&net->xfrm.policy_hash_work); #ifdef CONFIG_XFRM_SUB_POLICY xfrm_policy_flush(net, XFRM_POLICY_TYPE_SUB, false); #endif xfrm_policy_flush(net, XFRM_POLICY_TYPE_MAIN, false); +} - synchronize_rcu(); +static void xfrm_policy_fini(struct net *net) +{ + struct xfrm_pol_inexact_bin *b, *t; + unsigned int sz; + int dir; WARN_ON(!list_empty(&net->xfrm.policy_all)); @@ -4368,6 +4368,7 @@ static void __net_exit xfrm_net_exit(struct net *net) static struct pernet_operations __net_initdata xfrm_net_ops = { .init = xfrm_net_init, + .pre_exit = xfrm_net_pre_exit, .exit = xfrm_net_exit, }; From c16f74dc1d75d0e2e7670076d5375deda110ebeb Mon Sep 17 00:00:00 2001 From: Zhengchuan Liang Date: Fri, 22 May 2026 17:31:55 +0800 Subject: [PATCH 242/411] xfrm: input: hold netns during deferred transport reinjection Transport-mode reinjection stores a struct net pointer in skb->cb and uses it later from xfrm_trans_reinject(). That pointer must stay valid until the deferred callback runs. Take a netns reference when queueing deferred reinjection work and drop it after the callback completes. Use maybe_get_net() so the queueing path does not revive a namespace that is already being torn down. This keeps the existing workqueue design and fixes the netns lifetime handling in one place for all users of xfrm_trans_queue_net(). Fixes: 7b3801927e52 ("xfrm: introduce xfrm_trans_queue_net") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Xin Liu Co-developed-by: Luxing Yin Signed-off-by: Luxing Yin Signed-off-by: Zhengchuan Liang Signed-off-by: Ren Wei Assisted-by: Codex:gpt-5.4 Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_input.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index f65291eba1f6..e4c2cd24936d 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -797,9 +797,12 @@ static void xfrm_trans_reinject(struct work_struct *work) spin_unlock_bh(&trans->queue_lock); local_bh_disable(); - while ((skb = __skb_dequeue(&queue))) - XFRM_TRANS_SKB_CB(skb)->finish(XFRM_TRANS_SKB_CB(skb)->net, - NULL, skb); + while ((skb = __skb_dequeue(&queue))) { + struct net *net = XFRM_TRANS_SKB_CB(skb)->net; + + XFRM_TRANS_SKB_CB(skb)->finish(net, NULL, skb); + put_net(net); + } local_bh_enable(); } @@ -808,6 +811,7 @@ int xfrm_trans_queue_net(struct net *net, struct sk_buff *skb, struct sk_buff *)) { struct xfrm_trans_tasklet *trans; + struct net *hold_net; trans = this_cpu_ptr(&xfrm_trans_tasklet); @@ -816,8 +820,12 @@ int xfrm_trans_queue_net(struct net *net, struct sk_buff *skb, BUILD_BUG_ON(sizeof(struct xfrm_trans_cb) > sizeof(skb->cb)); + hold_net = maybe_get_net(net); + if (!hold_net) + return -ENODEV; + XFRM_TRANS_SKB_CB(skb)->finish = finish; - XFRM_TRANS_SKB_CB(skb)->net = net; + XFRM_TRANS_SKB_CB(skb)->net = hold_net; spin_lock_bh(&trans->queue_lock); __skb_queue_tail(&trans->queue, skb); spin_unlock_bh(&trans->queue_lock); From 25fe708bbc59289d3d1ea4b126fbc1b460a072a5 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Thu, 21 May 2026 01:12:01 -0700 Subject: [PATCH 243/411] net: team: fix NULL pointer dereference in team_xmit during mode change __team_change_mode() clears team->ops with memset() before restoring safe dummy handlers via team_adjust_ops(). A concurrent team_xmit() running under RCU on another CPU can read team->ops.transmit during this window and call a NULL function pointer, crashing the kernel. The race requires a mode change (CAP_NET_ADMIN) concurrent with transmit on the team device. BUG: kernel NULL pointer dereference, address: 0000000000000000 Oops: 0010 [#1] SMP KASAN NOPTI RIP: 0010:0x0 Call Trace: team_xmit (drivers/net/team/team_core.c:1853) dev_hard_start_xmit (net/core/dev.c:3904) __dev_queue_xmit (net/core/dev.c:4871) packet_sendmsg (net/packet/af_packet.c:3109) __sys_sendto (net/socket.c:2265) The original code assumed that no ports means no traffic, so mode changes could freely memset()/memcpy() the ops. AF_PACKET with forced carrier breaks that assumption. Prevent the race instead of making it safe: replace memset()/memcpy() with per-field updates that never touch transmit or receive. Those two handlers are managed solely by team_adjust_ops(), which already installs dummies when tx_en_port_count == 0 (always true during mode change since no ports are present). WRITE_ONCE/READ_ONCE prevent store/load tearing on the handler pointers. synchronize_net() before exit_op() drains in-flight readers that may still reference old mode state from before port removal switched the handlers to dummies. Fixes: 3d249d4ca7d0 ("net: introduce ethernet teaming device") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260521081159.1491563-3-bestswngs@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/team/team_core.c | 45 +++++++++++++++++++++++++----------- 1 file changed, 32 insertions(+), 13 deletions(-) diff --git a/drivers/net/team/team_core.c b/drivers/net/team/team_core.c index 0c87f9972457..f51388d50307 100644 --- a/drivers/net/team/team_core.c +++ b/drivers/net/team/team_core.c @@ -534,21 +534,23 @@ static void team_adjust_ops(struct team *team) if (!team->tx_en_port_count || !team_is_mode_set(team) || !team->mode->ops->transmit) - team->ops.transmit = team_dummy_transmit; + WRITE_ONCE(team->ops.transmit, team_dummy_transmit); else - team->ops.transmit = team->mode->ops->transmit; + WRITE_ONCE(team->ops.transmit, team->mode->ops->transmit); if (!team->rx_en_port_count || !team_is_mode_set(team) || !team->mode->ops->receive) - team->ops.receive = team_dummy_receive; + WRITE_ONCE(team->ops.receive, team_dummy_receive); else - team->ops.receive = team->mode->ops->receive; + WRITE_ONCE(team->ops.receive, team->mode->ops->receive); } /* - * We can benefit from the fact that it's ensured no port is present - * at the time of mode change. Therefore no packets are in fly so there's no - * need to set mode operations in any special way. + * team_change_mode() ensures no ports are present during mode change, + * but lockless readers can still reach team_xmit(). Avoid touching + * transmit/receive -- they are already set to dummies by + * team_adjust_ops() since no ports are enabled. synchronize_net() + * drains in-flight readers before destroying old mode state. */ static int __team_change_mode(struct team *team, const struct team_mode *new_mode) @@ -557,9 +559,21 @@ static int __team_change_mode(struct team *team, if (team_is_mode_set(team)) { void (*exit_op)(struct team *team) = team->ops.exit; - /* Clear ops area so no callback is called any longer */ - memset(&team->ops, 0, sizeof(struct team_mode_ops)); - team_adjust_ops(team); + /* Clear cold-path ops used only under RTNL. transmit and + * receive are already dummies (no ports) so leave them + * alone -- overwriting them is the source of the race. + */ + team->ops.init = NULL; + team->ops.exit = NULL; + team->ops.port_enter = NULL; + team->ops.port_leave = NULL; + team->ops.port_change_dev_addr = NULL; + team->ops.port_tx_disabled = NULL; + + /* Wait for in-flight readers before tearing down mode + * state they may reference. + */ + synchronize_net(); if (exit_op) exit_op(team); @@ -582,7 +596,12 @@ static int __team_change_mode(struct team *team, } team->mode = new_mode; - memcpy(&team->ops, new_mode->ops, sizeof(struct team_mode_ops)); + team->ops.init = new_mode->ops->init; + team->ops.exit = new_mode->ops->exit; + team->ops.port_enter = new_mode->ops->port_enter; + team->ops.port_leave = new_mode->ops->port_leave; + team->ops.port_change_dev_addr = new_mode->ops->port_change_dev_addr; + team->ops.port_tx_disabled = new_mode->ops->port_tx_disabled; team_adjust_ops(team); return 0; @@ -743,7 +762,7 @@ static rx_handler_result_t team_handle_frame(struct sk_buff **pskb) /* allow exact match delivery for disabled ports */ res = RX_HANDLER_EXACT; } else { - res = team->ops.receive(team, port, skb); + res = READ_ONCE(team->ops.receive)(team, port, skb); } if (res == RX_HANDLER_ANOTHER) { struct team_pcpu_stats *pcpu_stats; @@ -1845,7 +1864,7 @@ static netdev_tx_t team_xmit(struct sk_buff *skb, struct net_device *dev) tx_success = team_queue_override_transmit(team, skb); if (!tx_success) - tx_success = team->ops.transmit(team, skb); + tx_success = READ_ONCE(team->ops.transmit)(team, skb); if (tx_success) { struct team_pcpu_stats *pcpu_stats; From 11b326fb0a374f4654f9be22d0f0f7abd9f7d3fe Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Thu, 21 May 2026 21:05:54 +0800 Subject: [PATCH 244/411] ip6: vti: Use ip6_tnl.net in vti6_changelink(). ip netns add ns1 ip netns add ns2 ip -n ns1 link add vti6_test type vti6 remote ::1 local ::2 key 7 ip -n ns1 link set vti6_test netns ns2 ip -n ns2 link set vti6_test type vti6 remote ::3 local ::4 key 9 ip netns del ns2 ip netns del ns1 [ 132.495484] ------------[ cut here ]------------ [ 132.497609] kernel BUG at net/core/dev.c:12376! Commit 61220ab34948 ("vti6: Enable namespace changing") dropped NETIF_F_NETNS_LOCAL from vti6 devices. A vti6 tunnel can then move through IFLA_NET_NS_FD. After the move dev_net(dev) points at the new netns while t->net stays at the creation netns. vti6_changelink() and vti6_update() still use dev_net(dev) and dev_net(t->dev). They unlink from one per netns hash and relink into another. The creation netns is left with a stale entry. cleanup_net() of that netns later walks freed memory. Reachable from an unprivileged user namespace (unshare --user --map-root-user --net). Cross tenant scope on container hosts. Fixes: 61220ab34948 ("vti6: Enable namespace changing") Reported-by: Maoyi Xie Reviewed-by: Eric Dumazet Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260521130555.3421684-2-maoyixie.tju@gmail.com Signed-off-by: Paolo Abeni --- net/ipv6/ip6_vti.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index ad5290be4dd6..dcb257411d6e 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -722,10 +722,11 @@ vti6_tnl_change(struct ip6_tnl *t, const struct __ip6_tnl_parm *p, static int vti6_update(struct ip6_tnl *t, struct __ip6_tnl_parm *p, bool keep_mtu) { - struct net *net = dev_net(t->dev); - struct vti6_net *ip6n = net_generic(net, vti6_net_id); + struct net *net = t->net; + struct vti6_net *ip6n; int err; + ip6n = net_generic(net, vti6_net_id); vti6_tnl_unlink(ip6n, t); synchronize_net(); err = vti6_tnl_change(t, p, keep_mtu); @@ -1031,11 +1032,12 @@ static int vti6_changelink(struct net_device *dev, struct nlattr *tb[], struct nlattr *data[], struct netlink_ext_ack *extack) { - struct ip6_tnl *t; + struct ip6_tnl *t = netdev_priv(dev); + struct net *net = t->net; struct __ip6_tnl_parm p; - struct net *net = dev_net(dev); - struct vti6_net *ip6n = net_generic(net, vti6_net_id); + struct vti6_net *ip6n; + ip6n = net_generic(net, vti6_net_id); if (dev == ip6n->fb_tnl_dev) return -EINVAL; From 8b484efd5cb4eeef9021a661e198edc5349dacf6 Mon Sep 17 00:00:00 2001 From: Maoyi Xie Date: Thu, 21 May 2026 21:05:55 +0800 Subject: [PATCH 245/411] ip6: vti: Use ip6_tnl.net in vti6_siocdevprivate(). After patch 1/2 in this series, vti6_update() unlinks and relinks the tunnel through t->net. vti6_siocdevprivate() still uses dev_net(dev) for the collision lookup. For a tunnel moved through IFLA_NET_NS_FD, dev_net(dev) is the new netns, not t->net. SIOCCHGTUNNEL on a migrated tunnel then runs: net = dev_net(dev) /* migrated netns */ t = vti6_locate(net, &p1, false) /* misses target in t->net */ ... t = netdev_priv(dev) vti6_update(t, &p1, false) /* mutates t->net's hash */ A caller in the migrated netns picks params that match a tunnel in the creation netns. The lookup in dev_net(dev) finds nothing. vti6_update() prepends the migrated tunnel at the head of the creation netns hash bucket for those params. Later lookups in the creation netns resolve to the migrated device. xfrm receive delivers the matched packets through a device the caller controls. Reachable from an unprivileged user namespace (unshare --user --map-root-user --net). Cross tenant scope on container hosts. Switch the SIOCCHGTUNNEL path on a non fallback device to use t->net for the lookup. The lookup now matches the netns vti6_update() operates on. Also add ns_capable(self->net->user_ns, CAP_NET_ADMIN) before the lookup. The check at the top of the case is against dev_net(dev)->user_ns, which after migration is the attacker's netns. A caller there can pick params absent from self->net, the lookup returns NULL, t becomes self, and vti6_update() inserts the device into the creation netns hash. The new check requires CAP_NET_ADMIN in the creation netns user_ns too. SIOCADDTUNNEL and SIOCCHGTUNNEL on the fallback device keep dev_net(dev), which equals init_net there. Fixes: 61220ab34948 ("vti6: Enable namespace changing") Suggested-by: Jakub Kicinski Suggested-by: Xiao Liang Cc: stable@vger.kernel.org # v5.15+ Signed-off-by: Maoyi Xie Link: https://patch.msgid.link/20260521130555.3421684-3-maoyixie.tju@gmail.com Signed-off-by: Paolo Abeni --- net/ipv6/ip6_vti.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c index dcb257411d6e..df793c8bfffb 100644 --- a/net/ipv6/ip6_vti.c +++ b/net/ipv6/ip6_vti.c @@ -835,17 +835,24 @@ vti6_siocdevprivate(struct net_device *dev, struct ifreq *ifr, void __user *data if (p.proto != IPPROTO_IPV6 && p.proto != 0) break; vti6_parm_from_user(&p1, &p); - t = vti6_locate(net, &p1, cmd == SIOCADDTUNNEL); if (dev != ip6n->fb_tnl_dev && cmd == SIOCCHGTUNNEL) { + struct ip6_tnl *self = netdev_priv(dev); + + err = -EPERM; + if (!ns_capable(self->net->user_ns, CAP_NET_ADMIN)) + break; + t = vti6_locate(self->net, &p1, false); if (t) { if (t->dev != dev) { err = -EEXIST; break; } } else - t = netdev_priv(dev); + t = self; err = vti6_update(t, &p1, false); + } else { + t = vti6_locate(net, &p1, cmd == SIOCADDTUNNEL); } if (t) { err = 0; From df488cac6140aa04ae52af9b4507d8f99a3762be Mon Sep 17 00:00:00 2001 From: K Prateek Nayak Date: Sat, 23 May 2026 05:55:03 +0000 Subject: [PATCH 246/411] cpufreq/amd-pstate-ut: Disable dynamic_epp after the mode switch Dan reported a possible NULL pointer dereference in amd-pstate-ut.c from static analysis and sure enough, running amd-pstate-ut in active mode with amd_dynamic_epp=enable results in a crash as a reult of the policy reference being set to NULL early, before disabling dynamic EPP. Kalpana also reported seeing amd-pstate-ut error out with -EBUSY for "amd_pstate_ut_epp" test when starting from the passive mode and amd_dynamic_epp=enable in the command line. The reason for the failure is that the command line enables dynamic_epp by default after the mode switch and the modifications to EPP values are blocked when running in dynamic EPP mode. Solution to both problems is to toggle off dynamic_epp *after* the mode switch when the driver grabs the policy reference again since the unit test is in full control of the policy after that point. The final restoration step will reset the dynamic_epp state via mode switch based on the initial conditions of the system. Reported-by: Kalpana Shetty Reported-by: Dan Carpenter Closes: https://lore.kernel.org/linux-pm/ahEq0CvdBX0T7_cO@stanley.mountain/ Fixes: f9f16835d4dc ("cpufreq/amd-pstate-ut: Drop policy reference before driver switch") Signed-off-by: K Prateek Nayak Link: https://patch.msgid.link/20260523055503.7651-1-kprateek.nayak@amd.com Signed-off-by: Rafael J. Wysocki --- drivers/cpufreq/amd-pstate-ut.c | 16 ++++++++++------ 1 file changed, 10 insertions(+), 6 deletions(-) diff --git a/drivers/cpufreq/amd-pstate-ut.c b/drivers/cpufreq/amd-pstate-ut.c index 13a23dac477d..735b29f76438 100644 --- a/drivers/cpufreq/amd-pstate-ut.c +++ b/drivers/cpufreq/amd-pstate-ut.c @@ -302,12 +302,6 @@ static int amd_pstate_ut_epp(u32 index) cpufreq_cpu_put(policy); policy = NULL; - /* disable dynamic EPP before running test */ - if (cpudata->dynamic_epp) { - pr_debug("Dynamic EPP is enabled, disabling it\n"); - amd_pstate_clear_dynamic_epp(policy); - } - buf = (char *)__get_free_page(GFP_KERNEL); if (!buf) return -ENOMEM; @@ -327,6 +321,16 @@ static int amd_pstate_ut_epp(u32 index) orig_policy = cpudata->policy; cpudata->policy = CPUFREQ_POLICY_POWERSAVE; + /* + * Disable dynamic EPP before running test. If "orig_dynamic_epp" is + * true, the driver will do a redundant switch at the end and there + * is no need for enabling it again at the end of the test. + */ + if (cpudata->dynamic_epp) { + pr_debug("Dynamic EPP is enabled, disabling it\n"); + amd_pstate_clear_dynamic_epp(policy); + } + for (epp = 0; epp <= U8_MAX; epp++) { u8 val; From 2e357f002c61fd76fd8f12468744a06a5ec48eaa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bj=C3=B6rn=20T=C3=B6pel?= Date: Fri, 22 May 2026 14:06:40 +0200 Subject: [PATCH 247/411] net: Avoid checksumming unreadable skb tail on trim MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit pskb_trim_rcsum_slow() keeps CHECKSUM_COMPLETE valid by subtracting the checksum of the bytes removed from the skb tail. That assumes the removed bytes can be read. io_uring zcrx skbs may contain unreadable net_iov frags. With fbnic header/data split, small TCP/IPv4 packets can carry Ethernet padding in such a frag. ip_rcv_core() trims the skb to iph->tot_len before TCP sees it, and the CHECKSUM_COMPLETE adjustment then calls skb_checksum() on the padding. This is exposed by IPv4 because small TCP/IPv4 frames can be shorter than the Ethernet minimum payload. TCP/IPv6 frames are large enough in the normal zcrx path, so they do not hit the same padding trim. Keep the existing checksum adjustment for readable skbs. If the remaining packet is fully linear, drop CHECKSUM_COMPLETE and let the stack validate the packet after trimming. If unreadable payload would remain, fail the trim; the checksum cannot be adjusted without reading the trimmed tail. Also clear skb->unreadable when trimming removes all frags. Fixes: 65249feb6b3d ("net: add support for skbs with unreadable frags") Signed-off-by: Björn Töpel Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20260522120643.242974-1-bjorn@kernel.org Signed-off-by: Paolo Abeni --- net/core/skbuff.c | 31 +++++++++++++++++++++++++++---- 1 file changed, 27 insertions(+), 4 deletions(-) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 44ac121cfccb..d247acd447e4 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -2787,6 +2787,8 @@ int ___pskb_trim(struct sk_buff *skb, unsigned int len) skb->data_len = 0; skb_set_tail_pointer(skb, len); } + if (!skb_shinfo(skb)->nr_frags && !skb_has_frag_list(skb)) + skb->unreadable = 0; if (!skb->sk || skb->destructor == sock_edemux) skb_condense(skb); @@ -2794,16 +2796,37 @@ int ___pskb_trim(struct sk_buff *skb, unsigned int len) } EXPORT_SYMBOL(___pskb_trim); +static int pskb_trim_rcsum_complete(struct sk_buff *skb, unsigned int len) +{ + int delta = skb->len - len; + + if (skb_frags_readable(skb)) { + skb->csum = csum_block_sub(skb->csum, + skb_checksum(skb, len, delta, 0), + len); + return 0; + } + + if (len > skb_headlen(skb)) + return -EFAULT; + + /* The trimmed bytes are unreadable, but the remaining packet can be + * checksummed by software after trimming. + */ + skb->ip_summed = CHECKSUM_NONE; + return 0; +} + /* Note : use pskb_trim_rcsum() instead of calling this directly */ int pskb_trim_rcsum_slow(struct sk_buff *skb, unsigned int len) { if (skb->ip_summed == CHECKSUM_COMPLETE) { - int delta = skb->len - len; + int err; - skb->csum = csum_block_sub(skb->csum, - skb_checksum(skb, len, delta, 0), - len); + err = pskb_trim_rcsum_complete(skb, len); + if (err) + return err; } else if (skb->ip_summed == CHECKSUM_PARTIAL) { int hdlen = (len > skb_headlen(skb)) ? skb_headlen(skb) : len; int offset = skb_checksum_start_offset(skb) + skb->csum_offset; From c75b6f6eaacd0b74b832414cc3b9289c3686e941 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:42 -0700 Subject: [PATCH 248/411] ethtool: rss: avoid modifying the RSS context response Gemini says that we're modifying the RSS_CREATE response skb. I think it's right, the comment says that unicast() should unshare the skb but I'm not entirely sure what I meant there. netlink_trim() does a copy but only if skb is not well sized (it's at least 2x larger than necessary for the payload). Fixes: a166ab7816c5 ("ethtool: rss: support creating contexts via Netlink") Link: https://patch.msgid.link/20260522230647.1705600-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 353110b862ab..8ffec9785efa 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -981,11 +981,17 @@ ethnl_rss_create_validate(struct net_device *dev, struct genl_info *info) } static void -ethnl_rss_create_send_ntf(struct sk_buff *rsp, struct net_device *dev) +ethnl_rss_create_send_ntf(const struct sk_buff *rsp, struct net_device *dev) { - struct nlmsghdr *nlh = (void *)rsp->data; struct genlmsghdr *genl_hdr; + struct nlmsghdr *nlh; + struct sk_buff *ntf; + ntf = skb_copy_expand(rsp, 0, 0, GFP_KERNEL); + if (!ntf) + return; + + nlh = nlmsg_hdr(ntf); /* Convert the reply into a notification */ nlh->nlmsg_pid = 0; nlh->nlmsg_seq = ethnl_bcast_seq_next(); @@ -993,7 +999,7 @@ ethnl_rss_create_send_ntf(struct sk_buff *rsp, struct net_device *dev) genl_hdr = nlmsg_data(nlh); genl_hdr->cmd = ETHTOOL_MSG_RSS_CREATE_NTF; - ethnl_multicast(rsp, dev); + ethnl_multicast(ntf, dev); } int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) @@ -1104,12 +1110,8 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) genlmsg_end(rsp, hdr); - /* Use the same skb for the response and the notification, - * genlmsg_reply() will copy the skb if it has elevated user count. - */ - skb_get(rsp); - ret = genlmsg_reply(rsp, info); ethnl_rss_create_send_ntf(rsp, dev); + ret = genlmsg_reply(rsp, info); rsp = NULL; exit_unlock: From 3e6c6e9782ff8a8d8ded774b07ad4590cd61d04c Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:43 -0700 Subject: [PATCH 249/411] ethtool: rss: add missing errno on RSS context delete Remember to set ret before jumping out if someone tries to delete a context on a device which doesn't support contexts. Fixes: fbe09277fa63 ("ethtool: rss: support removing contexts via Netlink") Link: https://patch.msgid.link/20260522230647.1705600-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 8ffec9785efa..a16ee1e8e640 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -1170,8 +1170,10 @@ int ethnl_rss_delete_doit(struct sk_buff *skb, struct genl_info *info) dev = req.dev; ops = dev->ethtool_ops; - if (!ops->create_rxfh_context) + if (!ops->create_rxfh_context) { + ret = -EOPNOTSUPP; goto exit_free_dev; + } rtnl_lock(); netdev_lock_ops(dev); From 8d60141a32875248ef71d49c9920fa5e2aa40b29 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:44 -0700 Subject: [PATCH 250/411] ethtool: rss: fix falsely ignoring indir table updates rss_set_prep_indir() compares the new indirection table against the current one to determine whether any update is needed. The memcmp call passes data->indir_size as the length argument, but indir_size is the number of u32 entries, not the byte count. Fixes: c0ae03588bbb ("ethtool: rss: initial RSS_SET (indirection table handling)") Link: https://patch.msgid.link/20260522230647.1705600-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index a16ee1e8e640..458a4a7907e4 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -686,7 +686,7 @@ rss_set_prep_indir(struct net_device *dev, struct genl_info *info, ethtool_rxfh_indir_default(i, num_rx_rings); } - *mod |= memcmp(rxfh->indir, data->indir_table, data->indir_size); + *mod |= memcmp(rxfh->indir, data->indir_table, alloc_size); return user_size; From 266297692f97008ca48bc311775c087c59bd7fe3 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:45 -0700 Subject: [PATCH 251/411] ethtool: rss: fix indir_table and hkey leak on get_rxfh failure rss_prepare_get() allocates the indirection table and hash key buffer via rss_get_data_alloc(), then calls ops->get_rxfh() to populate them. If get_rxfh() fails, the function returns an error without freeing the allocation. Fixes: 4f038a6a02d2 ("net: ethtool: Don't call .cleanup_data when prepare_data fails") Link: https://patch.msgid.link/20260522230647.1705600-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 458a4a7907e4..9fb675d29232 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -170,8 +170,10 @@ rss_prepare_get(const struct rss_req_info *request, struct net_device *dev, rxfh.key = data->hkey; ret = ops->get_rxfh(dev, &rxfh); - if (ret) + if (ret) { + rss_get_data_free(data); goto out_unlock; + } data->hfunc = rxfh.hfunc; data->input_xfrm = rxfh.input_xfrm; From 78ccf1a70c6378e1f5073a8c2209b5129067b925 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:46 -0700 Subject: [PATCH 252/411] ethtool: rss: fix hkey leak when indir_size is 0 rss_get_data_alloc() allocates a single buffer that backs both the indirection table and the hash key, but only assigned data->indir_table when indir_size was nonzero. The expectation was that no driver implements RSS without supporting indirection table but apparently enic does just that (it's the only such in-tree driver). enic has get_rxfh_key_size but no get_rxfh_indir_size. data->indir_table stays as NULL, hkey gets set but rss_get_data_free() kfree(data->indir_table) is a nop and the allocation leaks. Always store the allocation base in data->indir_table so the free path is unambiguous. No caller treats indir_table as a sentinel; everything keys off indir_size. Fixes: 7112a04664bf ("ethtool: add netlink based get rss support") Link: https://patch.msgid.link/20260522230647.1705600-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index 9fb675d29232..f5cf214f8f85 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -134,8 +134,7 @@ rss_get_data_alloc(struct net_device *dev, struct rss_reply_data *data) if (!rss_config) return -ENOMEM; - if (data->indir_size) - data->indir_table = (u32 *)rss_config; + data->indir_table = (u32 *)rss_config; if (data->hkey_size) data->hkey = rss_config + indir_bytes; From 32a9ecde62731c9f7412507709192c84dafc38d1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:06:47 -0700 Subject: [PATCH 253/411] ethtool: rss: avoid device context leak on reply-build failure We wait with filling the reply for new RSS context creation until after the driver ->create_rxfh_context call. The driver needs to fill some of the defaults in the context. The failure of rss_fill_reply() is somewhat theoretical, but doesn't take much effort to handle it properly. Call ->remove_rxfh_context(). If the driver's remove callback fails (some implementations like sfc can return real command errors from firmware RPCs) - skip the xa_erase and kfree, leaving the context in the xarray. This matches how ethnl_rss_delete_doit() behaves. Fixes: a166ab7816c5 ("ethtool: rss: support creating contexts via Netlink") Link: https://patch.msgid.link/20260522230647.1705600-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/rss.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/net/ethtool/rss.c b/net/ethtool/rss.c index f5cf214f8f85..53792f53f922 100644 --- a/net/ethtool/rss.c +++ b/net/ethtool/rss.c @@ -1106,7 +1106,7 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) ntf_fail |= rss_fill_reply(rsp, &req.base, &data.base); if (WARN_ON(!hdr || ntf_fail)) { ret = -EMSGSIZE; - goto exit_unlock; + goto err_remove_ctx; } genlmsg_end(rsp, hdr); @@ -1134,6 +1134,10 @@ int ethnl_rss_create_doit(struct sk_buff *skb, struct genl_info *info) nlmsg_free(rsp); return ret; +err_remove_ctx: + if (ops->remove_rxfh_context(dev, ctx, req.rss_context, NULL)) + /* leave the context on failure, like ethnl_rss_delete_doit() */ + goto exit_unlock; err_ctx_id_free: xa_erase(&dev->ethtool->rss_ctx, req.rss_context); err_unlock_free_ctx: From 84371fb58423f997939aacdcbc02d128d76a54e5 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:04 -0700 Subject: [PATCH 254/411] ethtool: module: call ethnl_ops_complete() on module flash errors When validate() fails we are skipping over ethnl_ops_complete() even tho we already called ethnl_ops_begin(). Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/net/ethtool/module.c b/net/ethtool/module.c index cad2eb25b5a4..741f6fb25d45 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -427,10 +427,11 @@ int ethnl_act_module_fw_flash(struct sk_buff *skb, struct genl_info *info) ret = ethnl_module_fw_flash_validate(dev, info->extack); if (ret < 0) - goto out_unlock; + goto out_complete; ret = module_flash_fw(dev, tb, skb, info); +out_complete: ethnl_ops_complete(dev); out_unlock: From fb7f511d62692661846c47f199e0afe25c2982db Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:05 -0700 Subject: [PATCH 255/411] ethtool: module: avoid leaking a netdev ref on module flash errors module_flash_fw_schedule() is missing undo for setting the "in_progress" flag and taking the netdev reference. Delay taking these, the device can't disappear while we are holding rtnl_lock. Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/ethtool/module.c b/net/ethtool/module.c index 741f6fb25d45..392c03935e5e 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -319,8 +319,6 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name, if (err < 0) goto err_release_firmware; - dev->ethtool->module_fw_flash_in_progress = true; - netdev_hold(dev, &module_fw->dev_tracker, GFP_KERNEL); fw_update->dev = dev; fw_update->ntf_params.portid = info->snd_portid; fw_update->ntf_params.seq = info->snd_seq; @@ -335,6 +333,9 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name, if (err < 0) goto err_release_firmware; + dev->ethtool->module_fw_flash_in_progress = true; + netdev_hold(dev, &module_fw->dev_tracker, GFP_KERNEL); + schedule_work(&module_fw->work); return 0; From 7a84b965ffc12030af63cd10a8f3a1123ff39b7a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:06 -0700 Subject: [PATCH 256/411] ethtool: module: avoid racy updates to dev->ethtool bitfield When reviewing other changes Gemini points out that we currently update module_fw_flash_in_progress without holding any locks. Since module_fw_flash_in_progress is part of a bitfield this is not great, updates to other fields may be lost. We could use a bool and sprinkle some READ_ONCE/WRITE_ONCE here but seems like the issue is rather than the work is an unusual writer. The other writers already hold the right locks. So just very briefly take these locks when the work completes. Note that nothing ever cancels the FW update work, so there's no concern with deadlocks vs cancel. Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/net/ethtool/module.c b/net/ethtool/module.c index 392c03935e5e..cdb85e19a23b 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -221,14 +221,22 @@ static void module_flash_fw_work_list_del(struct list_head *list) static void module_flash_fw_work(struct work_struct *work) { struct ethtool_module_fw_flash *module_fw; + struct net_device *dev; module_fw = container_of(work, struct ethtool_module_fw_flash, work); + dev = module_fw->fw_update.dev; ethtool_cmis_fw_update(&module_fw->fw_update); module_flash_fw_work_list_del(&module_fw->list); - module_fw->fw_update.dev->ethtool->module_fw_flash_in_progress = false; - netdev_put(module_fw->fw_update.dev, &module_fw->dev_tracker); + + rtnl_lock(); + netdev_lock_ops(dev); + dev->ethtool->module_fw_flash_in_progress = false; + netdev_unlock_ops(dev); + rtnl_unlock(); + + netdev_put(dev, &module_fw->dev_tracker); release_firmware(module_fw->fw_update.fw); kfree(module_fw); } From 504eaefa44c8dec50f7499edcb36d24f3aefab2a Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:07 -0700 Subject: [PATCH 257/411] ethtool: module: check fw_flash_in_progress under rtnl_lock ethnl_set_module_validate() inspects module_fw_flash_in_progress but validate is meant for _input_ validation, not state validation. rtnl_lock is not held, yet. Move the check into ethnl_set_module(). Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/net/ethtool/module.c b/net/ethtool/module.c index cdb85e19a23b..5b49004ddf60 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -120,12 +120,6 @@ ethnl_set_module_validate(struct ethnl_req_info *req_info, if (!tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY]) return 0; - if (req_info->dev->ethtool->module_fw_flash_in_progress) { - NL_SET_ERR_MSG(info->extack, - "Module firmware flashing is in progress"); - return -EBUSY; - } - if (!ops->get_module_power_mode || !ops->set_module_power_mode) { NL_SET_ERR_MSG_ATTR(info->extack, tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY], @@ -148,6 +142,12 @@ ethnl_set_module(struct ethnl_req_info *req_info, struct genl_info *info) ops = dev->ethtool_ops; + if (dev->ethtool->module_fw_flash_in_progress) { + NL_SET_ERR_MSG(info->extack, + "Module firmware flashing is in progress"); + return -EBUSY; + } + power_new.policy = nla_get_u8(tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY]); ret = ops->get_module_power_mode(dev, &power, info->extack); if (ret < 0) From 760d04ebad5c4304f22c0d2251c9623b87a117c8 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:08 -0700 Subject: [PATCH 258/411] ethtool: module: fix cleanup if socket used for flashing multiple devices When a single Netlink socket issues MODULE_FW_FLASH_ACT against multiple devices, ethnl_sock_priv_set() overwrites sk_priv->dev on each call, retaining only the last one. The socket priv is used on socket close, to walk the global work list and mark the uncompleted flashing work as "orphaned". Otherwise if another socket reuses the PID it will unexpectedly receive the flashing notifications. Don't record the device, record net pointer instead. The purpose of the dev is to scope the work to a netns, anyway. If we store netns the overrides are safe/a nop since all flashed devices must be in the same netns as the socket. Fixes: 32b4c8b53ee7 ("ethtool: Add ability to flash transceiver modules' firmware") Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/module.c | 9 ++++----- net/ethtool/netlink.c | 4 ++-- net/ethtool/netlink.h | 4 ++-- 3 files changed, 8 insertions(+), 9 deletions(-) diff --git a/net/ethtool/module.c b/net/ethtool/module.c index 5b49004ddf60..ea4fb2a76650 100644 --- a/net/ethtool/module.c +++ b/net/ethtool/module.c @@ -291,11 +291,9 @@ void ethnl_module_fw_flash_sock_destroy(struct ethnl_sock_priv *sk_priv) spin_lock(&module_fw_flash_work_list_lock); list_for_each_entry(work, &module_fw_flash_work_list, list) { - if (work->fw_update.dev == sk_priv->dev && - work->fw_update.ntf_params.portid == sk_priv->portid) { + if (work->fw_update.ntf_params.portid == sk_priv->portid && + dev_net(work->fw_update.dev) == sk_priv->net) work->fw_update.ntf_params.closed_sock = true; - break; - } } spin_unlock(&module_fw_flash_work_list_lock); } @@ -332,7 +330,8 @@ module_flash_fw_schedule(struct net_device *dev, const char *file_name, fw_update->ntf_params.seq = info->snd_seq; fw_update->ntf_params.closed_sock = false; - err = ethnl_sock_priv_set(skb, dev, fw_update->ntf_params.portid, + err = ethnl_sock_priv_set(skb, dev_net(dev), + fw_update->ntf_params.portid, ETHTOOL_SOCK_TYPE_MODULE_FW_FLASH); if (err < 0) goto err_release_firmware; diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c index 5046023a30b1..7d45f9a884e5 100644 --- a/net/ethtool/netlink.c +++ b/net/ethtool/netlink.c @@ -53,7 +53,7 @@ const struct nla_policy ethnl_header_policy_phy_stats[] = { [ETHTOOL_A_HEADER_PHY_INDEX] = NLA_POLICY_MIN(NLA_U32, 1), }; -int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid, +int ethnl_sock_priv_set(struct sk_buff *skb, struct net *net, u32 portid, enum ethnl_sock_type type) { struct ethnl_sock_priv *sk_priv; @@ -62,7 +62,7 @@ int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid, if (IS_ERR(sk_priv)) return PTR_ERR(sk_priv); - sk_priv->dev = dev; + sk_priv->net = net; sk_priv->portid = portid; sk_priv->type = type; diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h index aaf6f2468768..fd2198e45d2b 100644 --- a/net/ethtool/netlink.h +++ b/net/ethtool/netlink.h @@ -318,12 +318,12 @@ enum ethnl_sock_type { }; struct ethnl_sock_priv { - struct net_device *dev; + struct net *net; u32 portid; enum ethnl_sock_type type; }; -int ethnl_sock_priv_set(struct sk_buff *skb, struct net_device *dev, u32 portid, +int ethnl_sock_priv_set(struct sk_buff *skb, struct net *net, u32 portid, enum ethnl_sock_type type); /** From 6c3f999a9d1338c6c89a9ff4549eafe72bc2e7b1 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:09 -0700 Subject: [PATCH 259/411] ethtool: cmis: require exact CDB reply length Malicious SFP module could respond with rpl_len longer than what cmis_cdb_process_reply() expected, leading to OOB writes. Malicious HW is a bit theoretical but some modules may just be buggy and/or the reads may occasionally get corrupted, so let's protect the kernel. The existing check protects from short replies. We need to protect from long ones, too. All callers that pass a non-zero rpl_exp_len cast the reply payload to a fixed-layout struct and read fields at fixed offsets, with no version negotiation or short-reply handling: - cmis_cdb_validate_password() - cmis_cdb_module_features_get() - cmis_fw_update_fw_mng_features_get() so let's assume that responses longer than expected do not have to be handled gracefully here. Add a warning message to make the debug easier in case my understanding is wrong... Note that page_data->length (argument of kmalloc) comes from last arg to ethtool_cmis_page_init() which is rpl_exp_len. Note2 that AIs also like to point out overflows in args->req.payload itself (which is a fixed-size 120 B buffer, on the stack), but callers should be reading structs defined by the standard, so protecting from requests for more data than max seem like defensive programming. Fixes: a39c84d79625 ("ethtool: cmis_cdb: Add a layer for supporting CDB commands") Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis_cdb.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/ethtool/cmis_cdb.c b/net/ethtool/cmis_cdb.c index 3670ca42dd40..f3a53a984460 100644 --- a/net/ethtool/cmis_cdb.c +++ b/net/ethtool/cmis_cdb.c @@ -513,8 +513,13 @@ static int cmis_cdb_process_reply(struct net_device *dev, } rpl = (struct ethtool_cmis_cdb_rpl *)page_data->data; - if ((args->rpl_exp_len > rpl->hdr.rpl_len + rpl_hdr_len) || - !rpl->hdr.rpl_chk_code) { + if (rpl->hdr.rpl_len != args->rpl_exp_len) { + netdev_warn(dev, "CDB reply length mismatch, expected %u got %u\n", + args->rpl_exp_len, rpl->hdr.rpl_len); + err = -EIO; + goto out; + } + if (!rpl->hdr.rpl_chk_code) { err = -EIO; goto out; } From 3e8c3d464c36bb342fe377b026577c7ec27fdbb4 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:10 -0700 Subject: [PATCH 260/411] ethtool: cmis: fix u16-to-u8 truncation of msleep_pre_rpl ethtool_cmis_cdb_compose_args() accepts msleep_pre_rpl as u16 but stores it into the u8 field ethtool_cmis_cdb_cmd_args::msleep_pre_rpl, silently truncating values >= 256. Seven of the nine call sites pass 1000 ms (it's the third argument from the end). Fixes: a39c84d79625 ("ethtool: cmis_cdb: Add a layer for supporting CDB commands") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-8-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ethtool/cmis.h b/net/ethtool/cmis.h index 4a9a946cabf0..778783a0f23c 100644 --- a/net/ethtool/cmis.h +++ b/net/ethtool/cmis.h @@ -63,9 +63,9 @@ struct ethtool_cmis_cdb_request { * struct ethtool_cmis_cdb_cmd_args - CDB commands execution arguments * @req: CDB command fields as described in the CMIS standard. * @max_duration: Maximum duration time for command completion in msec. + * @msleep_pre_rpl: Waiting time before checking reply in msec. * @read_write_len_ext: Allowable additional number of byte octets to the LPL * in a READ or a WRITE commands. - * @msleep_pre_rpl: Waiting time before checking reply in msec. * @rpl_exp_len: Expected reply length in bytes. * @flags: Validation flags for CDB commands. * @err_msg: Error message to be sent to user space. @@ -73,8 +73,8 @@ struct ethtool_cmis_cdb_request { struct ethtool_cmis_cdb_cmd_args { struct ethtool_cmis_cdb_request req; u16 max_duration; + u16 msleep_pre_rpl; u8 read_write_len_ext; - u8 msleep_pre_rpl; u8 rpl_exp_len; u8 flags; char *err_msg; From 12c2496a71f82f63617971ca9b730dffa05cf58b Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:11 -0700 Subject: [PATCH 261/411] ethtool: cmis: validate start_cmd_payload_size from module The CMIS firmware update code reads start_cmd_payload_size from the module's FW Management Features CDB reply and uses it directly as the byte count for memcpy. The destination buffer is 112 bytes (ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - 8). So a malicious module (or corrupted response) can cause a OOB write later on in cmis_fw_update_start_download(). Let's error out. If modules that expect longer LPL writes actually exist we should revisit. struct cmis_cdb_start_fw_download_pl's definition has to move, no change there. Fixes: c4f78134d45c ("ethtool: cmis_fw_update: add a layer for supporting firmware update using CDB") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-9-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis_fw_update.c | 36 ++++++++++++++++++++++-------------- 1 file changed, 22 insertions(+), 14 deletions(-) diff --git a/net/ethtool/cmis_fw_update.c b/net/ethtool/cmis_fw_update.c index df5f344209c4..16190c97e1f7 100644 --- a/net/ethtool/cmis_fw_update.c +++ b/net/ethtool/cmis_fw_update.c @@ -44,6 +44,20 @@ enum cmis_cdb_fw_write_mechanism { CMIS_CDB_FW_WRITE_MECHANISM_BOTH = 0x11, }; +/* See section 9.7.2 "CMD 0101h: Start Firmware Download" in CMIS standard + * revision 5.2. + * struct cmis_cdb_start_fw_download_pl is a structured layout of the + * flat array, ethtool_cmis_cdb_request::payload. + */ +struct cmis_cdb_start_fw_download_pl { + __struct_group(cmis_cdb_start_fw_download_pl_h, head, /* no attrs */, + __be32 image_size; + __be32 resv1; + ); + u8 vendor_data[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - + sizeof(struct cmis_cdb_start_fw_download_pl_h)]; +}; + static int cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb, struct net_device *dev, @@ -86,6 +100,14 @@ cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb, */ cdb->read_write_len_ext = rpl->read_write_len_ext; fw_mng->start_cmd_payload_size = rpl->start_cmd_payload_size; + if (fw_mng->start_cmd_payload_size > + sizeof_field(struct cmis_cdb_start_fw_download_pl, vendor_data)) { + ethnl_module_fw_flash_ntf_err(dev, ntf_params, + "Start cmd payload size exceeds max LPL payload", + NULL); + return -EINVAL; + } + fw_mng->write_mechanism = rpl->write_mechanism == CMIS_CDB_FW_WRITE_MECHANISM_LPL ? CMIS_CDB_FW_WRITE_MECHANISM_LPL : @@ -97,20 +119,6 @@ cmis_fw_update_fw_mng_features_get(struct ethtool_cmis_cdb *cdb, return 0; } -/* See section 9.7.2 "CMD 0101h: Start Firmware Download" in CMIS standard - * revision 5.2. - * struct cmis_cdb_start_fw_download_pl is a structured layout of the - * flat array, ethtool_cmis_cdb_request::payload. - */ -struct cmis_cdb_start_fw_download_pl { - __struct_group(cmis_cdb_start_fw_download_pl_h, head, /* no attrs */, - __be32 image_size; - __be32 resv1; - ); - u8 vendor_data[ETHTOOL_CMIS_CDB_LPL_MAX_PL_LENGTH - - sizeof(struct cmis_cdb_start_fw_download_pl_h)]; -}; - static int cmis_fw_update_start_download(struct ethtool_cmis_cdb *cdb, struct ethtool_cmis_fw_update_params *fw_update, From d5551f4c1800dc714cec86647bdd651ae0de923e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Fri, 22 May 2026 16:13:12 -0700 Subject: [PATCH 262/411] ethtool: cmis: validate fw->size against start_cmd_payload_size cmis_fw_update_start_download() copies start_cmd_payload_size bytes from the firmware blob into the CDB LPL vendor_data[] payload without validating that the FW has enough data. Since the start_cmd_payload_size can only be ~120B an image too short is most likely corrupted, so reject it. Fixes: c4f78134d45c ("ethtool: cmis_fw_update: add a layer for supporting firmware update using CDB") Reviewed-by: Maxime Chevallier Reviewed-by: Danielle Ratson Link: https://patch.msgid.link/20260522231312.1710836-10-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/cmis_fw_update.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/ethtool/cmis_fw_update.c b/net/ethtool/cmis_fw_update.c index 16190c97e1f7..291d04d2776a 100644 --- a/net/ethtool/cmis_fw_update.c +++ b/net/ethtool/cmis_fw_update.c @@ -130,6 +130,14 @@ cmis_fw_update_start_download(struct ethtool_cmis_cdb *cdb, u8 lpl_len; int err; + if (fw_update->fw->size < vendor_data_size) { + ethnl_module_fw_flash_ntf_err(fw_update->dev, + &fw_update->ntf_params, + "Firmware image too small for module's start payload", + NULL); + return -EINVAL; + } + pl.image_size = cpu_to_be32(fw_update->fw->size); memcpy(pl.vendor_data, fw_update->fw->data, vendor_data_size); From 60474437af5fe12abd20607766c809e0ad4e7245 Mon Sep 17 00:00:00 2001 From: Varadarajan Narayanan Date: Thu, 14 May 2026 12:15:31 +0530 Subject: [PATCH 263/411] spi: dt-bindings: spi-qpic-snand: Add ipq5210 compatible Since the QPIC-SPI-NAND flash controller present in ipq5210 is the same as the one found in ipq9574, document the ipq5210 compatible and with ipq9574 as the fallback. Signed-off-by: Varadarajan Narayanan Link: https://patch.msgid.link/20260514-ipq5210-nand-v1-1-cbdd7492e826@oss.qualcomm.com Signed-off-by: Mark Brown --- Documentation/devicetree/bindings/spi/qcom,spi-qpic-snand.yaml | 1 + 1 file changed, 1 insertion(+) diff --git a/Documentation/devicetree/bindings/spi/qcom,spi-qpic-snand.yaml b/Documentation/devicetree/bindings/spi/qcom,spi-qpic-snand.yaml index 7d0571feb46d..829da22537d4 100644 --- a/Documentation/devicetree/bindings/spi/qcom,spi-qpic-snand.yaml +++ b/Documentation/devicetree/bindings/spi/qcom,spi-qpic-snand.yaml @@ -25,6 +25,7 @@ properties: - items: - enum: - qcom,ipq5018-snand + - qcom,ipq5210-snand - qcom,ipq5332-snand - qcom,ipq5424-snand - const: qcom,ipq9574-snand From cda64169bade79427f264e43d0f422eaed9dc116 Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Wed, 13 May 2026 22:06:01 +0200 Subject: [PATCH 264/411] x86/microcode: Do not access MSR_IA32_PLATFORM_ID when running as a guest Patch in Fixes: causes the usual: unchecked MSR access error: RDMSR from 0x17 at ... (intel_get_platform_id) Call Trace: early_init_intel early_cpu_init setup_arch _printk start_kernel x86_64_start_reservations x86_64_start_kernel common_startup_64 because the kernel is booted in a guest. In order to avoid it, this MSR access needs to be prevented when running virtualized. That is usually done by checking X86_FEATURE_HYPERVISOR but for this particular case it is too early yet. The platform ID needs to be read as early as when microcode is loaded on the BSP: load_ucode_bsp ... -> get_microcode_blob ... -> intel_find_matching_signature and by that time, CPUID leafs haven't been parsed yet. The microcode loader already has logic to check early whether the kernel is running virtualized so make that globally available to arch/x86/. The query whether running virtualized is getting more and more prominent in recent times so might as well make it an arch-global var which the rest of the code can use. Fixes: d8630b67ca1ed ("x86/cpu: Add platform ID to CPU info structure") Reported-by: Vishal Verma Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Binbin Wu Reviewed-by: Xiaoyao Li Tested-by: Binbin Wu Link: https://lore.kernel.org/all/20260430020953.1405535-1-binbin.wu@linux.intel.com --- arch/x86/include/asm/processor.h | 1 + arch/x86/kernel/cpu/microcode/amd.c | 4 ++-- arch/x86/kernel/cpu/microcode/core.c | 22 ++++++++++------------ arch/x86/kernel/cpu/microcode/intel.c | 3 +++ arch/x86/kernel/cpu/microcode/internal.h | 1 - 5 files changed, 16 insertions(+), 15 deletions(-) diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 10b5355b323e..67dd932305db 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -733,6 +733,7 @@ bool xen_set_default_idle(void); #endif void __noreturn stop_this_cpu(void *dummy); +extern bool x86_hypervisor_present; void microcode_check(struct cpuinfo_x86 *prev_info); void store_cpu_caps(struct cpuinfo_x86 *info); diff --git a/arch/x86/kernel/cpu/microcode/amd.c b/arch/x86/kernel/cpu/microcode/amd.c index e533881284a1..5c0afae75e9f 100644 --- a/arch/x86/kernel/cpu/microcode/amd.c +++ b/arch/x86/kernel/cpu/microcode/amd.c @@ -322,7 +322,7 @@ static u32 get_patch_level(void) { u32 rev, dummy __always_unused; - if (IS_ENABLED(CONFIG_MICROCODE_DBG) && hypervisor_present) { + if (IS_ENABLED(CONFIG_MICROCODE_DBG) && x86_hypervisor_present) { int cpu = smp_processor_id(); if (!microcode_rev[cpu]) { @@ -714,7 +714,7 @@ static bool __apply_microcode_amd(struct microcode_amd *mc, u32 *cur_rev, invlpg(p_addr_end); } - if (IS_ENABLED(CONFIG_MICROCODE_DBG) && hypervisor_present) + if (IS_ENABLED(CONFIG_MICROCODE_DBG) && x86_hypervisor_present) microcode_rev[smp_processor_id()] = mc->hdr.patch_id; /* verify patch application was successful */ diff --git a/arch/x86/kernel/cpu/microcode/core.c b/arch/x86/kernel/cpu/microcode/core.c index 651202e6fefb..45ca406a8112 100644 --- a/arch/x86/kernel/cpu/microcode/core.c +++ b/arch/x86/kernel/cpu/microcode/core.c @@ -57,7 +57,7 @@ bool force_minrev = IS_ENABLED(CONFIG_MICROCODE_LATE_FORCE_MINREV); u32 base_rev; u32 microcode_rev[NR_CPUS] = {}; -bool hypervisor_present; +bool __ro_after_init x86_hypervisor_present; /* * Synchronization. @@ -118,14 +118,9 @@ bool __init microcode_loader_disabled(void) /* * Disable when: * - * 1) The CPU does not support CPUID. - */ - if (!cpuid_feature()) { - dis_ucode_ldr = true; - return dis_ucode_ldr; - } - - /* + * 1) The CPU does not support CPUID, detected below in + * load_ucode_bsp(). + * * 2) Bit 31 in CPUID[1]:ECX is clear * The bit is reserved for hypervisor use. This is still not * completely accurate as XEN PV guests don't see that CPUID bit @@ -135,9 +130,7 @@ bool __init microcode_loader_disabled(void) * 3) Certain AMD patch levels are not allowed to be * overwritten. */ - hypervisor_present = native_cpuid_ecx(1) & BIT(31); - - if ((hypervisor_present && !IS_ENABLED(CONFIG_MICROCODE_DBG)) || + if ((x86_hypervisor_present && !IS_ENABLED(CONFIG_MICROCODE_DBG)) || amd_check_current_patch_level()) dis_ucode_ldr = true; @@ -179,6 +172,11 @@ void __init load_ucode_bsp(void) early_parse_cmdline(); + if (!cpuid_feature()) + dis_ucode_ldr = true; + else + x86_hypervisor_present = native_cpuid_ecx(1) & BIT(31); + if (microcode_loader_disabled()) return; diff --git a/arch/x86/kernel/cpu/microcode/intel.c b/arch/x86/kernel/cpu/microcode/intel.c index 37ac4afe0972..a4c0a0cf928b 100644 --- a/arch/x86/kernel/cpu/microcode/intel.c +++ b/arch/x86/kernel/cpu/microcode/intel.c @@ -138,6 +138,9 @@ u32 intel_get_platform_id(void) { unsigned int val[2]; + if (x86_hypervisor_present) + return 0; + /* * This can be called early. Use CPUID directly instead of * relying on cpuinfo_x86 which may not be fully initialized. diff --git a/arch/x86/kernel/cpu/microcode/internal.h b/arch/x86/kernel/cpu/microcode/internal.h index 3b93c0676b4f..a10b547eda1e 100644 --- a/arch/x86/kernel/cpu/microcode/internal.h +++ b/arch/x86/kernel/cpu/microcode/internal.h @@ -48,7 +48,6 @@ extern struct early_load_data early_data; extern struct ucode_cpu_info ucode_cpu_info[]; extern u32 microcode_rev[NR_CPUS]; extern u32 base_rev; -extern bool hypervisor_present; struct cpio_data find_microcode_in_initrd(const char *path); From d90f236f8b9e354848bd226f581db27755ab901d Mon Sep 17 00:00:00 2001 From: Li Ming Date: Wed, 20 May 2026 20:14:57 +0800 Subject: [PATCH 265/411] cxl/test: Update mock dev array before calling platform_device_add() CXL test environment hits the following error sometimes. cxl_mem mem9: endpoint7 failed probe All mock memdevs are platform firmware devices added by cxl_test module, and cxl_test module also provides a platform device driver for them to create a memdev device to CXL subsystem. cxl_test module uses cxl_rcd/mem_single/mem arrays to store different types of mock memdevs. CXL drivers calls registered mock functions for a mock memdev by checking if a given memdev is in these arrays. When cxl_test module adds these mock memdevs, it always calls platform_device_add() before adding them to a suitable mock memdev array. However, there is a small window where CXL drivers calls mock function for a added memdev before it added to a mock memdev array. In above case, cxl endpoint driver considers a added memdev was not a mock memdev, then calling devm_cxl_endpoint_decoders_setup() for it rather than mock_endpoint_decoders_setup(). An appropriate solution is that adding a new mock device to a mock device array before calling platform_device_add() for it. It can guarantee the new mock device is visible to CXL subsystem. This patch introduces a new helped called cxl_mock_platform_device_add() to handle the issue, and uses the function for all mock devices addition. Fixes: 3a2b97b3210b ("cxl/test: Improve init-order fidelity relative to real-world systems") Signed-off-by: Li Ming Tested-by: Alison Schofield Reviewed-by: Alison Schofield Link: https://patch.msgid.link/20260520121457.234404-1-ming.li@zohomail.com Signed-off-by: Dave Jiang --- tools/testing/cxl/test/cxl.c | 105 ++++++++++++++--------------------- 1 file changed, 43 insertions(+), 62 deletions(-) diff --git a/tools/testing/cxl/test/cxl.c b/tools/testing/cxl/test/cxl.c index 418669927fb0..296516eecfd6 100644 --- a/tools/testing/cxl/test/cxl.c +++ b/tools/testing/cxl/test/cxl.c @@ -1523,6 +1523,23 @@ static void mock_companion(struct acpi_device *adev, struct device *dev) #define SZ_64G (SZ_32G * 2) #endif +static int cxl_mock_platform_device_add(struct platform_device *pdev, + struct platform_device **ppdev) +{ + int rc; + + if (ppdev) + *ppdev = pdev; + rc = platform_device_add(pdev); + if (rc) { + platform_device_put(pdev); + if (ppdev) + *ppdev = NULL; + } + + return rc; +} + static __init int cxl_rch_topo_init(void) { int rc, i; @@ -1537,13 +1554,10 @@ static __init int cxl_rch_topo_init(void) goto err_bridge; mock_companion(adev, &pdev->dev); - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_rch[i]); + if (rc) goto err_bridge; - } - cxl_rch[i] = pdev; mock_pci_bus[idx].bridge = &pdev->dev; rc = sysfs_create_link(&pdev->dev.kobj, &pdev->dev.kobj, "firmware_node"); @@ -1595,13 +1609,10 @@ static __init int cxl_single_topo_init(void) goto err_bridge; mock_companion(adev, &pdev->dev); - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_hb_single[i]); + if (rc) goto err_bridge; - } - cxl_hb_single[i] = pdev; mock_pci_bus[i + NR_CXL_HOST_BRIDGES].bridge = &pdev->dev; rc = sysfs_create_link(&pdev->dev.kobj, &pdev->dev.kobj, "physical_node"); @@ -1620,12 +1631,9 @@ static __init int cxl_single_topo_init(void) goto err_port; pdev->dev.parent = &bridge->dev; - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_root_single[i]); + if (rc) goto err_port; - } - cxl_root_single[i] = pdev; } for (i = 0; i < ARRAY_SIZE(cxl_swu_single); i++) { @@ -1638,12 +1646,9 @@ static __init int cxl_single_topo_init(void) goto err_uport; pdev->dev.parent = &root_port->dev; - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_swu_single[i]); + if (rc) goto err_uport; - } - cxl_swu_single[i] = pdev; } for (i = 0; i < ARRAY_SIZE(cxl_swd_single); i++) { @@ -1657,12 +1662,9 @@ static __init int cxl_single_topo_init(void) goto err_dport; pdev->dev.parent = &uport->dev; - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_swd_single[i]); + if (rc) goto err_dport; - } - cxl_swd_single[i] = pdev; } return 0; @@ -1735,12 +1737,9 @@ static int cxl_mem_init(void) pdev->dev.parent = &dport->dev; set_dev_node(&pdev->dev, i % 2); - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_mem[i]); + if (rc) goto err_mem; - } - cxl_mem[i] = pdev; } for (i = 0; i < ARRAY_SIZE(cxl_mem_single); i++) { @@ -1753,12 +1752,9 @@ static int cxl_mem_init(void) pdev->dev.parent = &dport->dev; set_dev_node(&pdev->dev, i % 2); - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_mem_single[i]); + if (rc) goto err_single; - } - cxl_mem_single[i] = pdev; } for (i = 0; i < ARRAY_SIZE(cxl_rcd); i++) { @@ -1772,12 +1768,9 @@ static int cxl_mem_init(void) pdev->dev.parent = &rch->dev; set_dev_node(&pdev->dev, i % 2); - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_rcd[i]); + if (rc) goto err_rcd; - } - cxl_rcd[i] = pdev; } return 0; @@ -1869,13 +1862,10 @@ static __init int cxl_test_init(void) goto err_bridge; mock_companion(adev, &pdev->dev); - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_host_bridge[i]); + if (rc) goto err_bridge; - } - cxl_host_bridge[i] = pdev; mock_pci_bus[i].bridge = &pdev->dev; rc = sysfs_create_link(&pdev->dev.kobj, &pdev->dev.kobj, "physical_node"); @@ -1893,12 +1883,9 @@ static __init int cxl_test_init(void) goto err_port; pdev->dev.parent = &bridge->dev; - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_root_port[i]); + if (rc) goto err_port; - } - cxl_root_port[i] = pdev; } BUILD_BUG_ON(ARRAY_SIZE(cxl_switch_uport) != ARRAY_SIZE(cxl_root_port)); @@ -1911,12 +1898,9 @@ static __init int cxl_test_init(void) goto err_uport; pdev->dev.parent = &root_port->dev; - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_switch_uport[i]); + if (rc) goto err_uport; - } - cxl_switch_uport[i] = pdev; } for (i = 0; i < ARRAY_SIZE(cxl_switch_dport); i++) { @@ -1929,12 +1913,9 @@ static __init int cxl_test_init(void) goto err_dport; pdev->dev.parent = &uport->dev; - rc = platform_device_add(pdev); - if (rc) { - platform_device_put(pdev); + rc = cxl_mock_platform_device_add(pdev, &cxl_switch_dport[i]); + if (rc) goto err_dport; - } - cxl_switch_dport[i] = pdev; } rc = cxl_single_topo_init(); @@ -1953,9 +1934,9 @@ static __init int cxl_test_init(void) acpi0017_mock.dev.bus = &platform_bus_type; cxl_acpi->dev.groups = cxl_acpi_groups; - rc = platform_device_add(cxl_acpi); + rc = cxl_mock_platform_device_add(cxl_acpi, NULL); if (rc) - goto err_root; + goto err_rch; rc = cxl_mem_init(); if (rc) From b051bb6bf0a231117036aa607cadf55be8e63910 Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Tue, 26 May 2026 08:35:31 -0700 Subject: [PATCH 266/411] blk-mq: reinsert cached request to the list A previous commit removed an optimization out of caution for a scenario that turns out not to be real: all the "queue_exit" goto's are safe to reinsert the request into the cached_rq's plug list as they are either from a non-blocking path, or a successful merge that already holds the queue reference. This optimization is most needed for small sequential workloads that successfully merge into larger requests. Fixes: dc278e9bf2b9 ("blk-mq: pop cached request if it is usable") Suggested-by: Ming Lei Suggested-by: Christoph Hellwig Signed-off-by: Keith Busch Reviewed-by: Chaitanya Kulkarni Link: https://patch.msgid.link/20260526153531.2365935-1-kbusch@meta.com Signed-off-by: Jens Axboe --- block/blk-mq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 28c2d931e75e..a24175441380 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -3246,7 +3246,7 @@ void blk_mq_submit_bio(struct bio *bio) if (!rq) blk_queue_exit(q); else - blk_mq_free_request(rq); + rq_list_add_head(&plug->cached_rqs, rq); } #ifdef CONFIG_BLK_MQ_STACKING From 030675aa54cf757769b3db65642433d626b3ed7c Mon Sep 17 00:00:00 2001 From: Chaitanya Sabnis Date: Tue, 26 May 2026 15:52:40 +0530 Subject: [PATCH 267/411] i2c: davinci: fix division by zero on missing clock-frequency When the 'clock-frequency' property is missing from the device tree, the driver falls back to DAVINCI_I2C_DEFAULT_BUS_FREQ. However, this macro was defined in kHz (100), whereas the device tree property is expected in Hz. The probe function divided the fallback value by 1000, causing integer truncation that resulted in dev->bus_freq = 0. This triggered a deterministic division-by-zero kernel panic when calculating clock dividers later in the probe sequence. Fix this by redefining DAVINCI_I2C_DEFAULT_BUS_FREQ in Hz (100000) to match the expected device tree property unit, allowing the existing division logic to work correctly for both cases. Fixes: b04ce6385979 ("i2c: davinci: kill platform data") Reported-by: Sashiko Closes: https://lore.kernel.org/all/20260514044726.57297C2BCB7@smtp.kernel.org/ Signed-off-by: Chaitanya Sabnis Cc: # v6.14+ Reviewed-by: Bartosz Golaszewski Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/20260526102240.4949-1-chaitanya.msabnis@gmail.com --- drivers/i2c/busses/i2c-davinci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-davinci.c b/drivers/i2c/busses/i2c-davinci.c index a773ba082321..66c23535656b 100644 --- a/drivers/i2c/busses/i2c-davinci.c +++ b/drivers/i2c/busses/i2c-davinci.c @@ -117,7 +117,7 @@ /* timeout for pm runtime autosuspend */ #define DAVINCI_I2C_PM_TIMEOUT 1000 /* ms */ -#define DAVINCI_I2C_DEFAULT_BUS_FREQ 100 +#define DAVINCI_I2C_DEFAULT_BUS_FREQ 100000 struct davinci_i2c_dev { struct device *dev; From 05f95729ca844704d15e49ce14868af4b403b32b Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Fri, 22 May 2026 22:34:23 -0400 Subject: [PATCH 268/411] l2tp: use refcount_inc_not_zero in l2tp_session_get_by_ifname A reader in l2tp_session_get_by_ifname() can return a pointer to a session whose refcount has reached zero. The getter takes its reference with plain refcount_inc(), but every other session getter in the same file (l2tp_v2_session_get, l2tp_v3_session_get, and the corresponding _get_next variants) uses refcount_inc_not_zero() because the IDR/RCU lookup can race with refcount_dec_and_test() -> l2tp_session_free() -> kfree_rcu(). The ifname getter is the only outlier; the inconsistency was raised on-list after 979c017803c4 ("l2tp: use list_del_rcu in l2tp_session_unhash"). A reader inside rcu_read_lock_bh() that matches session->ifname can be preempted between the strcmp() and the refcount_inc(). If the last reference drops on another CPU in that window, the reader's refcount_inc() runs on a counter that has reached zero. refcount_t catches the addition-on-zero, prints "refcount_t: addition on 0; use-after-free", saturates the counter, and returns the saturated pointer to the caller. Session memory is held live by the in-flight RCU read section, but the kfree_rcu() callback queued from l2tp_session_free() will free it once the grace period closes; a caller that dereferences the returned session past that point hits a slab-use-after-free. On PREEMPT_RT local_bh_disable() is a per-CPU sleeping lock and the preemption window is real; on stock PREEMPT kernels local_bh_disable() is a preempt_count increment that closes the cross-CPU race in practice (see below). Use refcount_inc_not_zero() and continue the list walk on failure, matching the other session getters in the file. The ifname getter is the only session getter in net/l2tp/ that still uses the bare refcount_inc() pattern; this change restores file-internal consistency. The success path is unchanged. Fixes: abe7a1a7d0b6 ("l2tp: improve tunnel/session refcount helpers") Cc: stable@vger.kernel.org Signed-off-by: Michael Bommarito Reviewed-by: James Chapman Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260523023423.2568972-1-michael.bommarito@gmail.com Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_core.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/net/l2tp/l2tp_core.c b/net/l2tp/l2tp_core.c index 1455f67e01dd..9419c8555d22 100644 --- a/net/l2tp/l2tp_core.c +++ b/net/l2tp/l2tp_core.c @@ -441,12 +441,13 @@ struct l2tp_session *l2tp_session_get_by_ifname(const struct net *net, idr_for_each_entry_ul(&pn->l2tp_tunnel_idr, tunnel, tmp, tunnel_id) { if (tunnel) { list_for_each_entry_rcu(session, &tunnel->session_list, list) { - if (!strcmp(session->ifname, ifname)) { - refcount_inc(&session->ref_count); - rcu_read_unlock_bh(); + if (strcmp(session->ifname, ifname)) + continue; + if (!refcount_inc_not_zero(&session->ref_count)) + continue; + rcu_read_unlock_bh(); - return session; - } + return session; } } } From d895767c337814cf4b97d5ad5375e5ed7e12018d Mon Sep 17 00:00:00 2001 From: "Lucien.Jheng" Date: Sun, 24 May 2026 14:39:15 +0800 Subject: [PATCH 269/411] net: phy: air_en8811h: add AN8811HB MCU assert/deassert support AN8811HB needs a MCU soft-reset cycle before firmware loading begins. Assert the MCU (hold it in reset) and immediately deassert (release) via a dedicated PBUS register pair (0x5cf9f8 / 0x5cf9fc), accessed through a registered mdio_device at PHY-addr+8. Add __air_pbus_reg_write() as a low-level helper taking a struct mdio_device *, create and register the PBUS mdio_device in an8811hb_probe() and store it in priv->pbusdev, then implement an8811hb_mcu_assert() / _deassert() on top of it. Add an8811hb_remove() to unregister the PBUS device on teardown. Wire both calls into an8811hb_load_firmware() and en8811h_restart_mcu() so every firmware load or MCU restart on AN8811HB correctly sequences the reset control registers. Fixes: 5afda1d734ed ("net: phy: air_en8811h: add Airoha AN8811HB support") Signed-off-by: Lucien Jheng Reviewed-by: Andrew Lunn Link: https://patch.msgid.link/20260524063915.47961-1-lucienzx159@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/air_en8811h.c | 153 +++++++++++++++++++++++++++++++++- 1 file changed, 149 insertions(+), 4 deletions(-) diff --git a/drivers/net/phy/air_en8811h.c b/drivers/net/phy/air_en8811h.c index 29ae73e65caa..a86129ce693c 100644 --- a/drivers/net/phy/air_en8811h.c +++ b/drivers/net/phy/air_en8811h.c @@ -17,6 +17,7 @@ #include #include #include +#include #include #include #include @@ -170,9 +171,23 @@ #define AN8811HB_CLK_DRV_CKO_LDPWD BIT(13) #define AN8811HB_CLK_DRV_CKO_LPPWD BIT(14) +#define AN8811HB_MCU_SW_RST 0x5cf9f8 +#define AN8811HB_MCU_SW_RST_HOLD BIT(16) +#define AN8811HB_MCU_SW_RST_RUN (BIT(16) | BIT(0)) +#define AN8811HB_MCU_SW_START 0x5cf9fc +#define AN8811HB_MCU_SW_START_EN BIT(16) + +/* MII register constants for PBUS access (PHY addr + 8) */ +#define AIR_PBUS_ADDR_HIGH 0x1c +#define AIR_PBUS_DATA_HIGH 0x10 +#define AIR_PBUS_REG_ADDR_HIGH_MASK GENMASK(15, 6) +#define AIR_PBUS_REG_ADDR_LOW_MASK GENMASK(5, 2) + /* Led definitions */ #define EN8811H_LED_COUNT 3 +#define EN8811H_PBUS_ADDR_OFFS 8 + /* Default LED setup: * GPIO5 <-> LED0 On: Link detected, blink Rx/Tx * GPIO4 <-> LED1 On: Link detected at 2500 or 1000 Mbps @@ -201,6 +216,7 @@ struct en8811h_priv { struct clk_hw hw; struct phy_device *phydev; unsigned int cko_is_enabled; + struct mdio_device *pbusdev; }; enum { @@ -254,6 +270,31 @@ static int air_phy_write_page(struct phy_device *phydev, int page) return __phy_write(phydev, AIR_EXT_PAGE_ACCESS, page); } +static int __air_pbus_reg_write(struct mdio_device *mdiodev, + u32 pbus_reg, u32 pbus_data) +{ + int ret; + + ret = __mdiobus_write(mdiodev->bus, mdiodev->addr, AIR_EXT_PAGE_ACCESS, + upper_16_bits(pbus_reg)); + if (ret < 0) + return ret; + + ret = __mdiobus_write(mdiodev->bus, mdiodev->addr, AIR_PBUS_ADDR_HIGH, + FIELD_GET(AIR_PBUS_REG_ADDR_HIGH_MASK, pbus_reg)); + if (ret < 0) + return ret; + + ret = __mdiobus_write(mdiodev->bus, mdiodev->addr, + FIELD_GET(AIR_PBUS_REG_ADDR_LOW_MASK, pbus_reg), + lower_16_bits(pbus_data)); + if (ret < 0) + return ret; + + return __mdiobus_write(mdiodev->bus, mdiodev->addr, AIR_PBUS_DATA_HIGH, + upper_16_bits(pbus_data)); +} + static int __air_buckpbus_reg_write(struct phy_device *phydev, u32 pbus_address, u32 pbus_data) { @@ -570,10 +611,67 @@ static int an8811hb_load_file(struct phy_device *phydev, const char *name, return ret; } +static int an8811hb_mcu_assert(struct phy_device *phydev) +{ + struct en8811h_priv *priv = phydev->priv; + int ret; + + phy_lock_mdio_bus(phydev); + + ret = __air_pbus_reg_write(priv->pbusdev, AN8811HB_MCU_SW_RST, + AN8811HB_MCU_SW_RST_HOLD); + if (ret < 0) + goto unlock; + + ret = __air_pbus_reg_write(priv->pbusdev, AN8811HB_MCU_SW_START, 0); + if (ret < 0) + goto unlock; + + msleep(50); + phydev_dbg(phydev, "MCU asserted\n"); + +unlock: + phy_unlock_mdio_bus(phydev); + return ret; +} + +static int an8811hb_mcu_deassert(struct phy_device *phydev) +{ + struct en8811h_priv *priv = phydev->priv; + int ret; + + phy_lock_mdio_bus(phydev); + + ret = __air_pbus_reg_write(priv->pbusdev, AN8811HB_MCU_SW_START, + AN8811HB_MCU_SW_START_EN); + if (ret < 0) + goto unlock; + + ret = __air_pbus_reg_write(priv->pbusdev, AN8811HB_MCU_SW_RST, + AN8811HB_MCU_SW_RST_RUN); + if (ret < 0) + goto unlock; + + msleep(50); + phydev_dbg(phydev, "MCU deasserted\n"); + +unlock: + phy_unlock_mdio_bus(phydev); + return ret; +} + static int an8811hb_load_firmware(struct phy_device *phydev) { int ret; + ret = an8811hb_mcu_assert(phydev); + if (ret < 0) + return ret; + + ret = an8811hb_mcu_deassert(phydev); + if (ret < 0) + return ret; + ret = air_buckpbus_reg_write(phydev, EN8811H_FW_CTRL_1, EN8811H_FW_CTRL_1_START); if (ret < 0) @@ -662,6 +760,16 @@ static int en8811h_restart_mcu(struct phy_device *phydev) { int ret; + if (phy_id_compare_model(phydev->phy_id, AN8811HB_PHY_ID)) { + ret = an8811hb_mcu_assert(phydev); + if (ret < 0) + return ret; + + ret = an8811hb_mcu_deassert(phydev); + if (ret < 0) + return ret; + } + ret = air_buckpbus_reg_write(phydev, EN8811H_FW_CTRL_1, EN8811H_FW_CTRL_1_START); if (ret < 0) @@ -1166,6 +1274,7 @@ static int en8811h_leds_setup(struct phy_device *phydev) static int an8811hb_probe(struct phy_device *phydev) { + struct mdio_device *mdiodev; struct en8811h_priv *priv; int ret; @@ -1175,10 +1284,28 @@ static int an8811hb_probe(struct phy_device *phydev) return -ENOMEM; phydev->priv = priv; + /* + * The AN8811HB PHY address is restricted to 8-15 (decimal), + * depending on the board hardware strapping. + * This means the PBUS address is only in the range 16-21 (decimal), + * so we do not need to handle the case + * where the PBUS address exceeds 31 (decimal). + */ + mdiodev = mdio_device_create(phydev->mdio.bus, + phydev->mdio.addr + EN8811H_PBUS_ADDR_OFFS); + if (IS_ERR(mdiodev)) + return PTR_ERR(mdiodev); + + ret = mdio_device_register(mdiodev); + if (ret) + goto err_dev_free; + + priv->pbusdev = mdiodev; + ret = an8811hb_load_firmware(phydev); if (ret < 0) { phydev_err(phydev, "Load firmware failed: %d\n", ret); - return ret; + goto err_dev_create; } en8811h_print_fw_version(phydev); @@ -1191,22 +1318,29 @@ static int an8811hb_probe(struct phy_device *phydev) ret = en8811h_leds_setup(phydev); if (ret < 0) - return ret; + goto err_dev_create; priv->phydev = phydev; /* Co-Clock Output */ ret = an8811hb_clk_provider_setup(&phydev->mdio.dev, &priv->hw); if (ret) - return ret; + goto err_dev_create; /* Configure led gpio pins as output */ ret = air_buckpbus_reg_modify(phydev, AN8811HB_GPIO_OUTPUT, AN8811HB_GPIO_OUTPUT_345, AN8811HB_GPIO_OUTPUT_345); if (ret < 0) - return ret; + goto err_dev_create; return 0; + +err_dev_create: + mdio_device_remove(mdiodev); + +err_dev_free: + mdio_device_free(mdiodev); + return ret; } static int en8811h_probe(struct phy_device *phydev) @@ -1561,6 +1695,16 @@ static int en8811h_suspend(struct phy_device *phydev) return genphy_suspend(phydev); } +static void an8811hb_remove(struct phy_device *phydev) +{ + struct en8811h_priv *priv = phydev->priv; + + if (priv->pbusdev) { + mdio_device_remove(priv->pbusdev); + mdio_device_free(priv->pbusdev); + } +} + static struct phy_driver en8811h_driver[] = { { PHY_ID_MATCH_MODEL(EN8811H_PHY_ID), @@ -1587,6 +1731,7 @@ static struct phy_driver en8811h_driver[] = { PHY_ID_MATCH_MODEL(AN8811HB_PHY_ID), .name = "Airoha AN8811HB", .probe = an8811hb_probe, + .remove = an8811hb_remove, .get_features = en8811h_get_features, .config_init = an8811hb_config_init, .get_rate_matching = en8811h_get_rate_matching, From b4bc94353050b1fa7b702bd4c6600710dd926cff Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 25 May 2026 20:13:35 +0000 Subject: [PATCH 270/411] tunnels: load network headers after skb_cow() in iptunnel_pmtud_build_icmp[v6]() Sashiko found that iptunnel_pmtud_build_icmp() and iptunnel_pmtud_build_icmpv6() were caching ip_hdr() and ipv6_hdr() before an skb_cow() call which can reallocate skb->head. Fix this possible UAF by initializing the local variables after the skb_cow() call. Remove skb_reset_network_header() calls which were not needed. Fixes: 4cb47a8644cc ("tunnels: PMTU discovery support for directly bridged IP packets") Signed-off-by: Eric Dumazet Reviewed-by: Stefano Brivio Link: https://patch.msgid.link/20260525201335.2361845-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_tunnel_core.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index 2667f53482bd..c77a4c3fbe75 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -212,7 +212,7 @@ EXPORT_SYMBOL_GPL(iptunnel_handle_offloads); */ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) { - const struct iphdr *iph = ip_hdr(skb); + const struct iphdr *iph; struct icmphdr *icmph; struct iphdr *niph; struct ethhdr eh; @@ -226,7 +226,6 @@ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) skb_copy_bits(skb, skb_mac_offset(skb), &eh, ETH_HLEN); pskb_pull(skb, ETH_HLEN); - skb_reset_network_header(skb); err = pskb_trim(skb, 576 - sizeof(*niph) - sizeof(*icmph)); if (err) @@ -236,7 +235,7 @@ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) err = skb_cow(skb, sizeof(*niph) + sizeof(*icmph) + ETH_HLEN); if (err) return err; - + iph = ip_hdr(skb); icmph = skb_push(skb, sizeof(*icmph)); *icmph = (struct icmphdr) { .type = ICMP_DEST_UNREACH, @@ -308,7 +307,7 @@ static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) */ static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) { - const struct ipv6hdr *ip6h = ipv6_hdr(skb); + const struct ipv6hdr *ip6h; struct icmp6hdr *icmp6h; struct ipv6hdr *nip6h; struct ethhdr eh; @@ -323,7 +322,6 @@ static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) skb_copy_bits(skb, skb_mac_offset(skb), &eh, ETH_HLEN); pskb_pull(skb, ETH_HLEN); - skb_reset_network_header(skb); err = pskb_trim(skb, IPV6_MIN_MTU - sizeof(*nip6h) - sizeof(*icmp6h)); if (err) @@ -334,6 +332,7 @@ static int iptunnel_pmtud_build_icmpv6(struct sk_buff *skb, int mtu) if (err) return err; + ip6h = ipv6_hdr(skb); icmp6h = skb_push(skb, sizeof(*icmp6h)); *icmp6h = (struct icmp6hdr) { .icmp6_type = ICMPV6_PKT_TOOBIG, From 7d9ef0cb271555d8cf39fefe6c981e1493b25ecf Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Mon, 25 May 2026 20:36:42 +0000 Subject: [PATCH 271/411] vxlan: do not reuse cached ip_hdr() value after skb_tunnel_check_pmtu() skb_tunnel_check_pmtu() can change skb->head. Reusing old_iph afer skb_tunnel_check_pmtu() can cause an UAF. Use instead ip_hdr(skb) as done in drivers/net/bareudp.c and drivers/net/geneve.c. Found by Sashiko. Fixes: 4cb47a8644cc ("tunnels: PMTU discovery support for directly bridged IP packets") Signed-off-by: Eric Dumazet Reviewed-by: Stefano Brivio Link: https://patch.msgid.link/20260525203642.2389723-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/net/vxlan/vxlan_core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/vxlan/vxlan_core.c b/drivers/net/vxlan/vxlan_core.c index e88798497503..b5b1253ac08b 100644 --- a/drivers/net/vxlan/vxlan_core.c +++ b/drivers/net/vxlan/vxlan_core.c @@ -2531,7 +2531,7 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, goto out_unlock; } - tos = ip_tunnel_ecn_encap(tos, old_iph, skb); + tos = ip_tunnel_ecn_encap(tos, ip_hdr(skb), skb); ttl = ttl ? : ip4_dst_hoplimit(&rt->dst); err = vxlan_build_skb(skb, ndst, sizeof(struct iphdr), vni, md, flags, udp_sum); @@ -2605,7 +2605,7 @@ void vxlan_xmit_one(struct sk_buff *skb, struct net_device *dev, goto out_unlock; } - tos = ip_tunnel_ecn_encap(tos, old_iph, skb); + tos = ip_tunnel_ecn_encap(tos, ip_hdr(skb), skb); ttl = ttl ? : ip6_dst_hoplimit(ndst); skb_scrub_packet(skb, xnet); err = vxlan_build_skb(skb, ndst, sizeof(struct ipv6hdr), From 509323077ef79a26ba0c60bb556e45c12c398b2d Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Fri, 22 May 2026 11:55:12 +0000 Subject: [PATCH 272/411] tunnels: do not assume transport header in iptunnel_pmtud_check_icmp() In some cases, iptunnel_pmtud_check_icmp() can be called while skb transport header is not set. This triggers an out-of-bound access, because (typeof(skb->transport_header))~0U is 65535. Access the icmp header based on IPv4 network header, after making sure icmp->type is present in skb linear part. Note that iptunnel_pmtud_check_icmpv6()) is fine. Fixes: 4cb47a8644cc ("tunnels: PMTU discovery support for directly bridged IP packets") Reported-by: Damiano Melotti Signed-off-by: Eric Dumazet Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260522115512.1519110-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_tunnel_core.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c index c77a4c3fbe75..d3c677e9bff2 100644 --- a/net/ipv4/ip_tunnel_core.c +++ b/net/ipv4/ip_tunnel_core.c @@ -280,7 +280,6 @@ static int iptunnel_pmtud_build_icmp(struct sk_buff *skb, int mtu) */ static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) { - const struct icmphdr *icmph = icmp_hdr(skb); const struct iphdr *iph = ip_hdr(skb); if (mtu < 576 || iph->frag_off != htons(IP_DF)) @@ -291,9 +290,17 @@ static int iptunnel_pmtud_check_icmp(struct sk_buff *skb, int mtu) ipv4_is_lbcast(iph->saddr) || ipv4_is_multicast(iph->saddr)) return 0; - if (iph->protocol == IPPROTO_ICMP && icmp_is_err(icmph->type)) - return 0; + if (iph->protocol == IPPROTO_ICMP) { + const struct icmphdr *icmph; + if (!pskb_network_may_pull(skb, iph->ihl * 4 + + offsetofend(struct icmphdr, type))) + return 0; + iph = ip_hdr(skb); + icmph = (void *)iph + iph->ihl * 4; + if (icmp_is_err(icmph->type)) + return 0; + } return iptunnel_pmtud_build_icmp(skb, mtu); } From 0e60dafe97eca61721f3db456f97d97a80c6c8ae Mon Sep 17 00:00:00 2001 From: Ali Ganiyev Date: Mon, 25 May 2026 10:23:47 +0900 Subject: [PATCH 273/411] ksmbd: OOB read regression in smb_check_perm_dacl() ACE-walk loops Commit d07b26f39246 ("ksmbd: require minimum ACE size in smb_check_perm_dacl()") introduced a transposed bounds check: if (offsetof(struct smb_ace, sid) + aces_size < CIFS_SID_BASE_SIZE) Since offsetof(..sid) is 8 and CIFS_SID_BASE_SIZE is 8, this evaluates to `aces_size < 0`. Because `aces_size` is always non-negative, this check becomes dead code and never breaks the loop. Worse, that commit removed the old 4-byte guard, meaning the loop now reads `ace->size` (offset 2) even when `aces_size` is 0-3 bytes. This re-opens a 2-byte heap out-of-bounds (OOB) read past the pntsd allocation during subsequent SMB2_CREATE operations. Fix this by properly transposing the comparison to require at least 16 bytes (8-byte offset + 8-byte SID base), matching the correct form used in smb_inherit_dacl(). Fixes: d07b26f39246 ("ksmbd: require minimum ACE size in smb_check_perm_dacl()") Cc: stable@vger.kernel.org Signed-off-by: Ali Ganiyev Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smbacl.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/smb/server/smbacl.c b/fs/smb/server/smbacl.c index c2d9be52a311..664b1b4a3233 100644 --- a/fs/smb/server/smbacl.c +++ b/fs/smb/server/smbacl.c @@ -1446,8 +1446,8 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path, ace = (struct smb_ace *)((char *)pdacl + sizeof(struct smb_acl)); aces_size = acl_size - sizeof(struct smb_acl); for (i = 0; i < le16_to_cpu(pdacl->num_aces); i++) { - if (offsetof(struct smb_ace, sid) + - aces_size < CIFS_SID_BASE_SIZE) + if (aces_size < offsetof(struct smb_ace, sid) + + CIFS_SID_BASE_SIZE) break; ace_size = le16_to_cpu(ace->size); if (ace_size > aces_size || @@ -1467,8 +1467,8 @@ int smb_check_perm_dacl(struct ksmbd_conn *conn, const struct path *path, ace = (struct smb_ace *)((char *)pdacl + sizeof(struct smb_acl)); aces_size = acl_size - sizeof(struct smb_acl); for (i = 0; i < le16_to_cpu(pdacl->num_aces); i++) { - if (offsetof(struct smb_ace, sid) + - aces_size < CIFS_SID_BASE_SIZE) + if (aces_size < offsetof(struct smb_ace, sid) + + CIFS_SID_BASE_SIZE) break; ace_size = le16_to_cpu(ace->size); if (ace_size > aces_size || From 2f15dcd0d4b502c704a52f5c7de128b163677978 Mon Sep 17 00:00:00 2001 From: Aleksandr Golovnya Date: Tue, 26 May 2026 01:50:18 +0700 Subject: [PATCH 274/411] ksmbd: release ksmbd_inode ref via ksmbd_inode_put on lookup paths ksmbd_query_inode_status() and ksmbd_lookup_fd_inode() both take a reference on a ksmbd_inode via __ksmbd_inode_lookup() (which performs atomic_inc_not_zero()) and later release it using a bare atomic_dec(&ci->m_count). Unlike ksmbd_inode_put(), a bare atomic_dec() does not check whether the reference count has reached zero, so if the caller happens to drop the last reference, the ksmbd_inode is leaked: it stays in the global inode hash table with m_count == 0, future __ksmbd_inode_lookup() calls reject it via atomic_inc_not_zero(), and ksmbd_inode_free() is never invoked. The race is: T1: __ksmbd_inode_lookup() -> atomic_inc_not_zero(): m_count = 2 T2: ksmbd_inode_put() -> atomic_dec_and_test(): m_count = 1 (not freed) T1: atomic_dec(&ci->m_count) -> m_count = 0 return (LEAK) In ksmbd_lookup_fd_inode() the matched-fp path (which now also uses ksmbd_inode_put()) cannot currently reach m_count == 0 because the matched ksmbd_file holds its own reference on ci, but converting it to the proper API keeps the three call sites consistent and avoids future regressions if the locking changes. Because ksmbd_inode_put() may free the ksmbd_inode if this drops the last reference, the call must happen after up_read(&ci->m_lock) on the two affected paths in ksmbd_lookup_fd_inode(). On the no-match path this is a pure reordering; on the matched path ksmbd_fp_get() is moved above the unlock so that the returned ksmbd_file is pinned before the inode reference is released. Signed-off-by: Aleksandr Golovnya Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/vfs_cache.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index 5a232d94f567..4d2d33df6231 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -217,7 +217,7 @@ int ksmbd_query_inode_status(struct dentry *dentry) ret = KSMBD_INODE_STATUS_OK; up_read(&ci->m_lock); - atomic_dec(&ci->m_count); + ksmbd_inode_put(ci); return ret; } @@ -719,14 +719,14 @@ struct ksmbd_file *ksmbd_lookup_fd_inode(struct dentry *dentry) down_read(&ci->m_lock); list_for_each_entry(lfp, &ci->m_fp_list, node) { if (inode == file_inode(lfp->filp)) { - atomic_dec(&ci->m_count); lfp = ksmbd_fp_get(lfp); up_read(&ci->m_lock); + ksmbd_inode_put(ci); return lfp; } } - atomic_dec(&ci->m_count); up_read(&ci->m_lock); + ksmbd_inode_put(ci); return NULL; } From cc57232cae23c0df91b4a59d0f519141ce9b5b02 Mon Sep 17 00:00:00 2001 From: Sean Shen Date: Tue, 26 May 2026 22:07:16 +0900 Subject: [PATCH 275/411] ksmbd: fix FSCTL permission bypass by adding a permission check for FSCTL_SET_SPARSE FSCTL_SET_SPARSE in fsctl_set_sparse() modifies the file's sparse attribute and saves it through xattr without any permission checks. This exposes two issues: 1) A client on a read-only share can change the sparse attribute on files it opened, even though the share is read-only. Other FSCTL write operations already check test_tree_conn_flag(work->tcon, KSMBD_TREE_CONN_FLAG_WRITABLE), but FSCTL_SET_SPARSE does not. 2) Even on writable shares, clients without FILE_WRITE_DATA or FILE_WRITE_ATTRIBUTES access should not modify the sparse attribute. Similar handle-level checks exist in other functions but are missing here. Add both share-level writable check and per-handle access check. Use goto out on error to avoid leaking file references. Fixes: e2f34481b24d ("cifsd: add server-side procedures for SMB3") Cc: Namjae Jeon Cc: Sergey Senozhatsky Cc: Steve French Signed-off-by: Sean Shen Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 5128a693aca6..620bcfbbfd92 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -8202,9 +8202,20 @@ static inline int fsctl_set_sparse(struct ksmbd_work *work, u64 id, int ret = 0; __le32 old_fattr; + if (!test_tree_conn_flag(work->tcon, KSMBD_TREE_CONN_FLAG_WRITABLE)) { + ksmbd_debug(SMB, "User does not have write permission\n"); + return -EACCES; + } + fp = ksmbd_lookup_fd_fast(work, id); if (!fp) return -ENOENT; + + if (!(fp->daccess & (FILE_WRITE_DATA_LE | FILE_WRITE_ATTRIBUTES_LE))) { + ret = -EACCES; + goto out; + } + idmap = file_mnt_idmap(fp->filp); old_fattr = fp->f_ci->m_fattr; From dd433671fef381fdaf7b530c631e6b782d66e224 Mon Sep 17 00:00:00 2001 From: Qi Tang Date: Sat, 23 May 2026 22:32:45 +0800 Subject: [PATCH 276/411] ipv6: validate extension header length before copying to cmsg ip6_datagram_recv_specific_ctl() builds IPV6_{HOPOPTS,DSTOPTS,RTHDR} cmsgs (and their IPV6_2292* legacy counterparts) by trusting the on-wire hdrlen byte (ptr[1]) when computing the put_cmsg() length. The length was validated only at parse time (ipv6_parse_hopopts(), etc.). An nftables payload-write expression can rewrite hdrlen after parsing and before the skb reaches recvmsg; the write itself is in-bounds but put_cmsg() then reads up to ((hdrlen+1) << 3) = 2040 bytes from an 8-byte header. nftables is reachable from an unprivileged user namespace, so this is an unprivileged slab-out-of-bounds read: BUG: KASAN: slab-out-of-bounds in put_cmsg+0x3ac/0x540 put_cmsg+0x3ac/0x540 udpv6_recvmsg+0xca0/0x1250 sock_recvmsg+0xdf/0x190 ____sys_recvmsg+0x1b1/0x620 Add ipv6_get_exthdr_len() which validates that at least two bytes are accessible before reading the hdrlen field, then checks the computed length against skb_tail_pointer(skb), returning 0 on failure. Extension headers are kept in the linear skb area by pskb_may_pull() during input, so skb_tail_pointer() is the correct bound. Use ipv6_get_exthdr_len() at all non-AH call sites: the five standalone cmsg blocks (HbH, 2292HbH, 2292DSTOPTS x2, 2292RTHDR) and the three standard cases in the extension-header walk loop (DSTOPTS, ROUTING, default). AH retains an inline bounds check because its length formula differs ((ptr[1]+2)<<2). The walk loop also gets a pre-read bounds check at the top to validate ptr before any case accesses ptr[0] or ptr[1]. When the walk loop detects a corrupted header, return from the function instead of continuing to process later socket options. Cc: stable@vger.kernel.org Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Qi Tang Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260523143245.2281415-1-tpluszz77@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv6/datagram.c | 54 ++++++++++++++++++++++++++++++++++++++------- 1 file changed, 46 insertions(+), 8 deletions(-) diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index ca3605acb433..38d7b4845281 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -617,6 +617,18 @@ void ip6_datagram_recv_common_ctl(struct sock *sk, struct msghdr *msg, } } +static u16 ipv6_get_exthdr_len(const struct sk_buff *skb, const u8 *ptr) +{ + u16 len; + + if (ptr + 2 > skb_tail_pointer(skb)) + return 0; + + len = (ptr[1] + 1) << 3; + + return (len <= skb_tail_pointer(skb) - ptr) ? len : 0; +} + void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, struct sk_buff *skb) { @@ -643,7 +655,10 @@ void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, /* HbH is allowed only once */ if (np->rxopt.bits.hopopts && (opt->flags & IP6SKB_HOPBYHOP)) { u8 *ptr = nh + sizeof(struct ipv6hdr); - put_cmsg(msg, SOL_IPV6, IPV6_HOPOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_HOPOPTS, len, ptr); } if (opt->lastopt && @@ -664,26 +679,37 @@ void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, unsigned int len; u8 *ptr = nh + off; + if (ptr + 2 > skb_tail_pointer(skb)) + return; + switch (nexthdr) { case IPPROTO_DSTOPTS: nexthdr = ptr[0]; - len = (ptr[1] + 1) << 3; + len = ipv6_get_exthdr_len(skb, ptr); + if (!len) + return; if (np->rxopt.bits.dstopts) put_cmsg(msg, SOL_IPV6, IPV6_DSTOPTS, len, ptr); break; case IPPROTO_ROUTING: nexthdr = ptr[0]; - len = (ptr[1] + 1) << 3; + len = ipv6_get_exthdr_len(skb, ptr); + if (!len) + return; if (np->rxopt.bits.srcrt) put_cmsg(msg, SOL_IPV6, IPV6_RTHDR, len, ptr); break; case IPPROTO_AH: nexthdr = ptr[0]; len = (ptr[1] + 2) << 2; + if (ptr + len > skb_tail_pointer(skb)) + return; break; default: nexthdr = ptr[0]; - len = (ptr[1] + 1) << 3; + len = ipv6_get_exthdr_len(skb, ptr); + if (!len) + return; break; } @@ -705,19 +731,31 @@ void ip6_datagram_recv_specific_ctl(struct sock *sk, struct msghdr *msg, } if (np->rxopt.bits.ohopopts && (opt->flags & IP6SKB_HOPBYHOP)) { u8 *ptr = nh + sizeof(struct ipv6hdr); - put_cmsg(msg, SOL_IPV6, IPV6_2292HOPOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292HOPOPTS, len, ptr); } if (np->rxopt.bits.odstopts && opt->dst0) { u8 *ptr = nh + opt->dst0; - put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, len, ptr); } if (np->rxopt.bits.osrcrt && opt->srcrt) { struct ipv6_rt_hdr *rthdr = (struct ipv6_rt_hdr *)(nh + opt->srcrt); - put_cmsg(msg, SOL_IPV6, IPV6_2292RTHDR, (rthdr->hdrlen+1) << 3, rthdr); + u16 len = ipv6_get_exthdr_len(skb, (u8 *)rthdr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292RTHDR, len, rthdr); } if (np->rxopt.bits.odstopts && opt->dst1) { u8 *ptr = nh + opt->dst1; - put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, (ptr[1]+1)<<3, ptr); + u16 len = ipv6_get_exthdr_len(skb, ptr); + + if (len) + put_cmsg(msg, SOL_IPV6, IPV6_2292DSTOPTS, len, ptr); } if (np->rxopt.bits.rxorigdstaddr) { struct sockaddr_in6 sin6; From 0a10faad5ca58332ad70f7663ba82611f4daf736 Mon Sep 17 00:00:00 2001 From: Fabian Lippold Date: Tue, 26 May 2026 17:41:01 +0200 Subject: [PATCH 277/411] ALSA: hda/realtek: add quirk for HP Dragonfly Folio G3 2-in-1 Add PCI quirk for HP Dragonfly Folio G3 (PCI ID 103c:8a06) to select the CS35L41 SPI4 & GPIO LED fixup variant. Signed-off-by: Fabian Lippold Link: https://patch.msgid.link/20260526154418.1850568-3-fabianlippold1184@gmail.com Signed-off-by: Takashi Iwai --- sound/hda/codecs/realtek/alc269.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/hda/codecs/realtek/alc269.c b/sound/hda/codecs/realtek/alc269.c index 212adce0f8e6..dcbc669842e0 100644 --- a/sound/hda/codecs/realtek/alc269.c +++ b/sound/hda/codecs/realtek/alc269.c @@ -7070,6 +7070,7 @@ static const struct hda_quirk alc269_fixup_tbl[] = { SND_PCI_QUIRK(0x103c, 0x89d3, "HP EliteBook 645 G9 (MB 89D2)", ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF), SND_PCI_QUIRK(0x103c, 0x89da, "HP Spectre x360 14t-ea100", ALC245_FIXUP_HP_SPECTRE_X360_EU0XXX), SND_PCI_QUIRK(0x103c, 0x89e7, "HP Elite x2 G9", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED), + SND_PCI_QUIRK(0x103c, 0x8a06, "HP Dragonfly Folio G3 2-in-1", ALC245_FIXUP_CS35L41_SPI_4_HP_GPIO_LED), SND_PCI_QUIRK(0x103c, 0x8a0f, "HP Pavilion 14-ec1xxx", ALC287_FIXUP_HP_GPIO_LED), SND_PCI_QUIRK(0x103c, 0x8a1f, "HP Laptop 14s-dr5xxx", ALC236_FIXUP_HP_MUTE_LED_COEFBIT2), SND_PCI_QUIRK(0x103c, 0x8a20, "HP Laptop 15s-fq5xxx", ALC236_FIXUP_HP_MUTE_LED_COEFBIT2), From 4db42e5fb9327c27b41f26bb9427ba0b97ecec30 Mon Sep 17 00:00:00 2001 From: Jakub Pisarczyk Date: Tue, 26 May 2026 22:18:30 +0200 Subject: [PATCH 278/411] ALSA: hda/cs420x: Add CS4208 fixup for iMac16,1 The 21.5" Retina 4K iMac (Late 2015, DMI product name "iMac16,1") ships with a Cirrus Logic CS4208 codec wired to an external speaker amplifier enabled through codec GPIO0 -- the same arrangement as the late-2013 MacBookPro 11,x. Without a matching entry in cs4208_mac_fixup_tbl[] the fixup picker logs: snd_hda_codec_cs420x hdaudioC1D0: CS4208: picked fixup for codec SSID 106b:0000 i.e. an empty fixup name, GPIO0 stays low, the external amp is never powered up, and the internal speakers are silent on a stock kernel. The codec SSID reported by hardware is 0x106b:0x7f00. Reusing CS4208_MBP11 (GPIO0 + SPDIF switch fixup) makes the internal speakers and S/PDIF output work out of the box, removing the need for users to set `options snd_hda_intel model=mbp11` via /etc/modprobe.d/. Tested on iMac16,1 (kernel 6.17.0): four internal drivers (Left tweeter, Left woofer, Right tweeter, Right woofer, exposed as the 4 channels of the analog-surround-40 ALSA profile) produce audio after the fixup is applied. Signed-off-by: Jakub Pisarczyk Link: https://patch.msgid.link/20260526201830.34097-1-pisarz77@gmail.com Signed-off-by: Takashi Iwai --- sound/hda/codecs/cirrus/cs420x.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/hda/codecs/cirrus/cs420x.c b/sound/hda/codecs/cirrus/cs420x.c index 42559edbba05..85c2ecf46d38 100644 --- a/sound/hda/codecs/cirrus/cs420x.c +++ b/sound/hda/codecs/cirrus/cs420x.c @@ -582,6 +582,7 @@ static const struct hda_quirk cs4208_mac_fixup_tbl[] = { SND_PCI_QUIRK(0x106b, 0x7200, "MacBookAir 6,2", CS4208_MBA6), SND_PCI_QUIRK(0x106b, 0x7800, "MacPro 6,1", CS4208_MACMINI), SND_PCI_QUIRK(0x106b, 0x7b00, "MacBookPro 12,1", CS4208_MBP11), + SND_PCI_QUIRK(0x106b, 0x7f00, "iMac 16,1", CS4208_MBP11), {} /* terminator */ }; From 14912d497188283f5a0aa5daaa161e52f79c7f34 Mon Sep 17 00:00:00 2001 From: Lianqin Hu Date: Wed, 27 May 2026 03:33:08 +0000 Subject: [PATCH 279/411] ALSA: usb-audio: Add iface reset and delay quirk for TAE1160 USB Audio Setting up the interface when suspended/resumeing fail on this card. Adding a reset and delay quirk will eliminate this problem. usb 1-1: new full-speed USB device number 2 using xhci-hcd usb 1-1: New USB device found, idVendor=25aa, idProduct=600b usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-1: Product: TAE1159 usb 1-1: Manufacturer: Generic usb 1-1: SerialNumber: 20210726905926 Signed-off-by: Lianqin Hu Link: https://patch.msgid.link/TYUPR06MB621736D7C85D43200E54E740D2082@TYUPR06MB6217.apcprd06.prod.outlook.com Signed-off-by: Takashi Iwai --- sound/usb/quirks.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 31cbe383ae65..3d1b3523b020 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -2449,6 +2449,8 @@ static const struct usb_audio_quirk_flags_table quirk_flags_table[] = { QUIRK_FLAG_DSD_RAW), DEVICE_FLG(0x2522, 0x0007, /* LH Labs Geek Out HD Audio 1V5 */ QUIRK_FLAG_SET_IFACE_FIRST), + DEVICE_FLG(0x25aa, 0x600b, /* TAE1159 */ + QUIRK_FLAG_FORCE_IFACE_RESET | QUIRK_FLAG_IFACE_DELAY), DEVICE_FLG(0x262a, 0x9302, /* ddHiFi TC44C */ QUIRK_FLAG_DSD_RAW), DEVICE_FLG(0x2708, 0x0002, /* Audient iD14 */ From 1750ad1388e03fb27068cd1f22c9c8b4590fe936 Mon Sep 17 00:00:00 2001 From: Qiang Ma Date: Tue, 26 May 2026 15:46:40 +0800 Subject: [PATCH 280/411] KVM: arm64: PMU: Preserve AArch32 counter low bits AArch32 writes to PMU event counters cannot update the top 32 bits, even when PMUv3p5 makes the counters 64-bit. KVM therefore needs to preserve the existing high half and only update the low half written by the guest, unless the caller explicitly forces a full reset through PMCR.P. The current code masks @val down to the old high half before taking lower_32_bits(val), which means the low half is always zero. As a result, AArch32 writes to event counters discard the guest-provided low 32 bits instead of storing them. Build the new value from the old high 32 bits and the low 32 bits of the value supplied by the guest. Fixes: 26d2d0594d70 ("KVM: arm64: PMU: Do not let AArch32 change the counters' top 32 bits") Signed-off-by: Qiang Ma Signed-off-by: Marc Zyngier Link: https://patch.msgid.link/20260526074640.791991-1-maqianga@uniontech.com Cc: stable@vger.kernel.org --- arch/arm64/kvm/pmu-emul.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/pmu-emul.c b/arch/arm64/kvm/pmu-emul.c index e1860acae641..c816db5d6761 100644 --- a/arch/arm64/kvm/pmu-emul.c +++ b/arch/arm64/kvm/pmu-emul.c @@ -174,8 +174,8 @@ static void kvm_pmu_set_pmc_value(struct kvm_pmc *pmc, u64 val, bool force) * action is to use PMCR.P, which will reset them to * 0 (the only use of the 'force' parameter). */ - val = __vcpu_sys_reg(vcpu, reg) & GENMASK(63, 32); - val |= lower_32_bits(val); + val = (__vcpu_sys_reg(vcpu, reg) & GENMASK(63, 32)) | + lower_32_bits(val); } __vcpu_assign_sys_reg(vcpu, reg, val); From f63ad68e18d774a5d15cd7e405ead63f6b322679 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?C=C3=A1ssio=20Gabriel?= Date: Wed, 27 May 2026 09:24:00 -0300 Subject: [PATCH 281/411] ASoC: codecs: simple-mux: Fix enum control bounds check MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit simple_mux_control_put() rejects values greater than e->items, but enum control values are zero based. For the two-entry mux used by this driver, valid values are 0 and 1, so value 2 must be rejected as well. Accepting e->items can store an invalid mux state, pass it to the GPIO setter, and pass it on to the DAPM mux update path where it is used as an index into the enum text array. Use the same >= e->items check used by the ASoC enum helpers. Fixes: 342fbb7578d1 ("ASoC: add simple-mux") Signed-off-by: Cássio Gabriel Link: https://patch.msgid.link/20260527-asoc-simple-mux-enum-bounds-v1-1-3f805b9fc671@gmail.com Signed-off-by: Mark Brown --- sound/soc/codecs/simple-mux.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/codecs/simple-mux.c b/sound/soc/codecs/simple-mux.c index 069555f35f73..c2f906a3f074 100644 --- a/sound/soc/codecs/simple-mux.c +++ b/sound/soc/codecs/simple-mux.c @@ -51,7 +51,7 @@ static int simple_mux_control_put(struct snd_kcontrol *kcontrol, struct snd_soc_component *c = snd_soc_dapm_to_component(dapm); struct simple_mux *priv = snd_soc_component_get_drvdata(c); - if (ucontrol->value.enumerated.item[0] > e->items) + if (ucontrol->value.enumerated.item[0] >= e->items) return -EINVAL; if (priv->mux == ucontrol->value.enumerated.item[0]) From 0f7abb6eaa3c3965f925e231c18409dac4f5a0c1 Mon Sep 17 00:00:00 2001 From: Vincent Donnefort Date: Thu, 21 May 2026 13:46:11 +0100 Subject: [PATCH 282/411] KVM: arm64: Fix meta-page unsharing in pKVM hyp tracing As the hyp_trace_buffer_unshare_hyp() function name suggests we should unshare all the previously shared pages, otherwise we leak hyp-shared pages which won't be reusable for hyp memory. Fix the typo by calling __unshare_page() on the meta-page, ensuring all previously shared pages are correctly unshared. Fixes: 3aed038aac8d ("KVM: arm64: Add trace remote for the nVHE/pKVM hyp") Signed-off-by: Vincent Donnefort Link: https://patch.msgid.link/20260521124613.911067-2-vdonnefort@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp_trace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp_trace.c b/arch/arm64/kvm/hyp_trace.c index 8b7f2bf2fba8..06805b426101 100644 --- a/arch/arm64/kvm/hyp_trace.c +++ b/arch/arm64/kvm/hyp_trace.c @@ -189,7 +189,7 @@ static void hyp_trace_buffer_unshare_hyp(struct hyp_trace_buffer *trace_buffer, if (cpu > last_cpu) break; - __share_page(rb_desc->meta_va); + __unshare_page(rb_desc->meta_va); for (p = 0; p < rb_desc->nr_page_va; p++) __unshare_page(rb_desc->page_va[p]); } From a23780ea9db3f3cadbb52ff6151384bff89d95d2 Mon Sep 17 00:00:00 2001 From: Vincent Donnefort Date: Thu, 21 May 2026 13:46:12 +0100 Subject: [PATCH 283/411] KVM: arm64: Fix rollback in hyp_trace_buffer_share_hyp() When sharing the trace buffer with the hypervisor, if sharing a page fails, the rollback path in hyp_trace_buffer_share_hyp() misses unsharing the metadata page (meta_va) which was successfully shared before entering the page sharing loop. Additionally, if a failure occurs, the cleanup calls hyp_trace_buffer_unshare_hyp() with an incorrect CPU index. Since that CPU's pages were already rolled back locally in the loop, this leads to duplicate unsharing attempts. Fix both issues affecting the rollback. Fixes: 3aed038aac8d ("KVM: arm64: Add trace remote for the nVHE/pKVM hyp") Reported-by: Sashiko Signed-off-by: Vincent Donnefort Link: https://patch.msgid.link/20260521124613.911067-3-vdonnefort@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp_trace.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/hyp_trace.c b/arch/arm64/kvm/hyp_trace.c index 06805b426101..8595f9bdb3dc 100644 --- a/arch/arm64/kvm/hyp_trace.c +++ b/arch/arm64/kvm/hyp_trace.c @@ -212,14 +212,15 @@ static int hyp_trace_buffer_share_hyp(struct hyp_trace_buffer *trace_buffer) } if (ret) { - for (p--; p >= 0; p--) + while (--p >= 0) __unshare_page(rb_desc->page_va[p]); + __unshare_page(rb_desc->meta_va); break; } } if (ret) - hyp_trace_buffer_unshare_hyp(trace_buffer, cpu--); + hyp_trace_buffer_unshare_hyp(trace_buffer, --cpu); return ret; } From adae9996c04fea3b1791099b6d79e1df76d50849 Mon Sep 17 00:00:00 2001 From: Vincent Donnefort Date: Thu, 21 May 2026 13:46:13 +0100 Subject: [PATCH 284/411] KVM: arm64: Fix memory leak in hyp_trace_unload() During trace remote loading, hyp_trace_load() allocates the descriptor pages but fails to store the allocated size in trace_buffer->desc_size. As a result, when unloading the trace buffer, hyp_trace_unload() calls free_pages_exact() with a size of 0 which fails to free the memory. Fix this by updating the descriptor size in trace_buffer->desc_size. Fixes: 3aed038aac8d ("KVM: arm64: Add trace remote for the nVHE/pKVM hyp") Reported-by: Sashiko Signed-off-by: Vincent Donnefort Link: https://patch.msgid.link/20260521124613.911067-4-vdonnefort@google.com Signed-off-by: Marc Zyngier --- arch/arm64/kvm/hyp_trace.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/kvm/hyp_trace.c b/arch/arm64/kvm/hyp_trace.c index 8595f9bdb3dc..c4b3ee552131 100644 --- a/arch/arm64/kvm/hyp_trace.c +++ b/arch/arm64/kvm/hyp_trace.c @@ -249,6 +249,7 @@ static struct trace_buffer_desc *hyp_trace_load(unsigned long size, void *priv) goto err_free_desc; trace_buffer->desc = desc; + trace_buffer->desc_size = desc_size; ret = hyp_trace_buffer_alloc_bpages_backing(trace_buffer, size); if (ret) @@ -298,6 +299,7 @@ static void hyp_trace_unload(struct trace_buffer_desc *desc, void *priv) hyp_trace_buffer_free_bpages_backing(trace_buffer); free_pages_exact(trace_buffer->desc, trace_buffer->desc_size); trace_buffer->desc = NULL; + trace_buffer->desc_size = 0; } static int hyp_trace_enable_tracing(bool enable, void *priv) From f657a6a3ba4c20bc01f5be3752d53498ee1bfe35 Mon Sep 17 00:00:00 2001 From: Balasubramani Vivekanandan Date: Fri, 22 May 2026 22:05:32 +0530 Subject: [PATCH 285/411] drm/xe: Restore IDLEDLY regiter on engine reset Wa_16023105232 programs the register IDLEDLY. The register is reset whenever the engine is reset. Therefore it should be added to the GuC save-restore register list for it to be restored after reset. Fixes: 7c53ff050ba8 ("drm/xe: Apply Wa_16023105232") Reviewed-by: Matt Roper Link: https://patch.msgid.link/20260522163531.1365540-2-balasubramani.vivekanandan@intel.com Signed-off-by: Balasubramani Vivekanandan (cherry picked from commit df1cfe24743a93b71eab27687e148ab8ae9b69e3) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_ads.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_guc_ads.c b/drivers/gpu/drm/xe/xe_guc_ads.c index 2b835d48b565..5760251cb685 100644 --- a/drivers/gpu/drm/xe/xe_guc_ads.c +++ b/drivers/gpu/drm/xe/xe_guc_ads.c @@ -767,6 +767,11 @@ static unsigned int guc_mmio_regset_write(struct xe_guc_ads *ads, } } + if (XE_GT_WA(hwe->gt, 16023105232)) + guc_mmio_regset_write_one(ads, regset_map, + RING_IDLEDLY(hwe->mmio_base), + count++); + return count; } From 2e7f55eb408c3f72ee1957a0d0ad11d8648a6379 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Sun, 17 May 2026 09:17:42 -0400 Subject: [PATCH 286/411] drm/amdgpu: fix lock leak on ENOMEM in AMDGPU_GEM_OP_GET_MAPPING_INFO The AMDGPU_GEM_OP_GET_MAPPING_INFO branch of amdgpu_gem_op_ioctl() holds three cleanup-tracked resources before calling kvcalloc(): the drm_gem_object reference from drm_gem_object_lookup(), the drm_exec lock on the looked-up GEM via drm_exec_lock_obj(), and the drm_exec lock on the per-process VM root page directory via amdgpu_vm_lock_pd(). All three are released by the out_exec label that every other error path in this function jumps to. The kvcalloc() failure path returns -ENOMEM directly, skipping out_exec and leaking all three. The leaked per-process VM root PD dma_resv lock is the load-bearing leak: any subsequent operation on the same VM (further GEM ops, command-submission, eviction, TTM shrinker callbacks) blocks on the held lock. DRM_IOCTL_AMDGPU_GEM_OP is DRM_AUTH | DRM_RENDER_ALLOW, so this is an unprivileged-local denial of service against the caller's GPU context, reachable by any process with /dev/dri/renderD* access. Route the failure through out_exec so drm_exec_fini() and drm_gem_object_put() run. Reproduced on stock 7.0.0-10, Ryzen 7 5700U / Radeon Vega (Lucienne): the failing ioctl returns -ENOMEM and a second GET_MAPPING_INFO on the same fd then blocks in drm_exec_lock_obj() on the leaked dma_resv. SIGKILL on the caller does not reap the task; the fd-release path during process exit goes through amdgpu_gem_object_close() -> drm_exec_prepare_obj() on the same lock, leaving the task in D state until the box is rebooted. The patched kernel was not rebuilt and re-tested on this hardware; the fix is mechanical. Tested on a single Lucienne / Vega box only. Ziyi Guo posted an independent INT_MAX-bound check for args->num_entries in the same branch [1]; the two patches are complementary and can land in either order. Fixes: 4d82724f7f2b ("drm/amdgpu: Add mapping info option for GEM_OP ioctl") Link: https://lore.kernel.org/all/20260208000255.4073363-1-n7l8m4@u.northwestern.edu/ # [1] Signed-off-by: Michael Bommarito Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Alex Deucher (cherry picked from commit b69d3256d79de15f54c322986ff4da68f1d65b0a) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 123d4a09114d..06dd2e8a5b47 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -1094,8 +1094,10 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data, * be retried. */ vm_entries = kvcalloc(args->num_entries, sizeof(*vm_entries), GFP_KERNEL); - if (!vm_entries) - return -ENOMEM; + if (!vm_entries) { + r = -ENOMEM; + goto out_exec; + } amdgpu_vm_bo_va_for_each_valid_mapping(bo_va, mapping) { if (num_mappings < args->num_entries) { From a1ba4594232c87c3b8defd6f89a2e40f8b08395d Mon Sep 17 00:00:00 2001 From: Ziyi Guo Date: Sun, 8 Feb 2026 00:02:55 +0000 Subject: [PATCH 287/411] drm/amdgpu: check num_entries in GEM_OP GET_MAPPING_INFO kvcalloc(args->num_entries, sizeof(*vm_entries), GFP_KERNEL) at amdgpu_gem.c:1050 uses the user-supplied num_entries directly without any upper bounds check. Since num_entries is a __u32 and sizeof(drm_amdgpu_gem_vm_entry) is 32 bytes, a large num_entries produces an allocation exceeding INT_MAX, triggering WARNING in __kvmalloc_node_noprof(), causing a kernel WARNING, TAINT_WARN, and panic on CONFIG_PANIC_ON_WARN=y systems. Add a size bounds check before we invoke the kvzalloc() to reject oversized num_entries early with -EINVAL. Fixes: 4d82724f7f2b ("drm/amdgpu: Add mapping info option for GEM_OP ioctl") Signed-off-by: Ziyi Guo Signed-off-by: Alex Deucher (cherry picked from commit 1fe7bf5457f6efd7be60b17e23163ba54341d73d) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c index 06dd2e8a5b47..fe6d988e7f24 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c @@ -1093,6 +1093,11 @@ int amdgpu_gem_op_ioctl(struct drm_device *dev, void *data, * If that number is larger than the size of the array, the ioctl must * be retried. */ + if (args->num_entries > INT_MAX / sizeof(*vm_entries)) { + r = -EINVAL; + goto out_exec; + } + vm_entries = kvcalloc(args->num_entries, sizeof(*vm_entries), GFP_KERNEL); if (!vm_entries) { r = -ENOMEM; From 4a03d23ce6ad474cb15862563bc9132e16e3e31e Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Tue, 19 May 2026 15:02:00 +0530 Subject: [PATCH 288/411] drm/amdgpu/userq: Fix doorbell object cleanup of queue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Unpin and unref the door bell obj if queue creation fails before initialization is complete. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 8c7506f7ba945f21e5abe7f8eac0a3acca6b5330) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index f070ea37d918..2301a44a03b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -787,7 +787,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) r = uq_funcs->mqd_create(queue, &args->in); if (r) { drm_file_err(uq_mgr->file, "Failed to create Queue\n"); - goto clean_mapping; + goto clean_doorbell_bo; } /* Update VM owner at userq submit-time for page-fault attribution. */ @@ -808,7 +808,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) if (r) { drm_file_err(uq_mgr->file, "Failed to map Queue\n"); mutex_unlock(&uq_mgr->userq_mutex); - goto clean_doorbell; + goto erase_doorbell; } } @@ -831,10 +831,15 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) args->out.queue_id = qid; return 0; -clean_doorbell: +erase_doorbell: xa_erase_irq(&adev->userq_doorbell_xa, index); clean_mqd: uq_funcs->mqd_destroy(queue); +clean_doorbell_bo: + amdgpu_bo_reserve(queue->db_obj.obj, true); + amdgpu_bo_unpin(queue->db_obj.obj); + amdgpu_bo_unreserve(queue->db_obj.obj); + amdgpu_bo_unref(&queue->db_obj.obj); clean_mapping: amdgpu_bo_reserve(fpriv->vm.root.bo, true); amdgpu_userq_buffer_vas_list_cleanup(adev, queue); From ba4c0ff47ee098c8e17d25f9dc050e6276bf9979 Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Tue, 19 May 2026 15:12:42 +0530 Subject: [PATCH 289/411] drm/amdgpu/userq: Fix the mutex_init cleanup for fence_drv_lock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit mutex fence_drv_lock is destroyed in amdgpu_userq_fence_driver_free also in one of the jump condition mutex_destroy is also called leading to double mutex_destroy. So rearranging the code so amdgpu_userq_fence_driver_free takes care of the clean up along with mutex_destroy. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 384dbef269d101e5b671fc7b942c56734cd1d186) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 2301a44a03b1..d6390cc5a798 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -748,12 +748,12 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) INIT_DELAYED_WORK(&queue->hang_detect_work, amdgpu_userq_hang_detect_work); - mutex_init(&queue->fence_drv_lock); - xa_init_flags(&queue->fence_drv_xa, XA_FLAGS_ALLOC); r = amdgpu_userq_fence_driver_alloc(adev, &queue->fence_drv); if (r) goto free_queue; + xa_init_flags(&queue->fence_drv_xa, XA_FLAGS_ALLOC); + mutex_init(&queue->fence_drv_lock); /* Make sure the queue can actually run with those virtual addresses. */ r = amdgpu_bo_reserve(fpriv->vm.root.bo, false); if (r) @@ -844,7 +844,6 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) amdgpu_bo_reserve(fpriv->vm.root.bo, true); amdgpu_userq_buffer_vas_list_cleanup(adev, queue); amdgpu_bo_unreserve(fpriv->vm.root.bo); - mutex_destroy(&queue->fence_drv_lock); free_fence_drv: amdgpu_userq_fence_driver_free(queue); free_queue: From ec78a85d95e9c37b6ca16d6ed1639fa64d5dd6dc Mon Sep 17 00:00:00 2001 From: Ivan Lipski Date: Thu, 14 May 2026 11:53:50 -0400 Subject: [PATCH 290/411] drm/amd/display: Write REFCLK to 48MHz on DCN21 [Why&How] dccg21_init() calls dccg2_init() which hardcodes 100MHz refclk values for MICROSECOND_TIME_BASE_DIV and MILLISECOND_TIME_BASE_DIV. DCN21 uses 48MHz refclk, so the wrong values corrupt DCCG timing and cause eDP link training failure on cold boot. Write the correct 48MHz values directly instead of calling dccg2_init(). v2: Fixed typo Fixes: e6e2b956fc81 ("drm/amd/display: Add missing DCCG register entries for DCN20-DCN316") Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5272 Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5311 Reported-by: Max Chernoff Tested-by: Max Chernoff Signed-off-by: Ivan Lipski Acked-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 08236c3ef284cd2d110e5e3d51fc9615e551f9dc) Cc: stable@vger.kernel.org --- .../drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c b/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c index c4d4eea140f3..1f23dfccf07a 100644 --- a/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dccg/dcn21/dcn21_dccg.c @@ -105,15 +105,26 @@ static void dccg21_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppcl * dccg2_init() unconditionally overwrites MICROSECOND_TIME_BASE_DIV to * 0x00120264, destroying the marker before it can be read. * - * Guard the call: if the S0i3 marker is present, skip dccg2_init() so the + * Guard the call: if the S0i3 marker is present, skip init so the * WA can function correctly. bios_golden_init() will handle init in that case. + * + * DCN21 uses 48MHz refclk, not 100MHz, so we must explicitly set the correct + * values (48MHz is taken from rn_clk_mgr_construct()). */ static void dccg21_init(struct dccg *dccg) { + struct dcn_dccg *dccg_dcn = TO_DCN_DCCG(dccg); + if (dccg2_is_s0i3_golden_init_wa_done(dccg)) return; - dccg2_init(dccg); + /* 48MHz refclk from rn_clk_mgr_construct() */ + REG_WRITE(MICROSECOND_TIME_BASE_DIV, 0x00120230); + REG_WRITE(MILLISECOND_TIME_BASE_DIV, 0x0010bb80); + REG_WRITE(DISPCLK_FREQ_CHANGE_CNTL, 0x0e01003c); + + if (REG(REFCLK_CNTL)) + REG_WRITE(REFCLK_CNTL, 0); } static const struct dccg_funcs dccg21_funcs = { From e984d61d92e702096058f0f828f4b2b8563b88ce Mon Sep 17 00:00:00 2001 From: Eric Huang Date: Thu, 7 May 2026 15:51:49 -0400 Subject: [PATCH 291/411] drm/amdkfd: fix NULL pointer bug in svm_range_set_attr The process_info could be NULL if user doesn't call kfd_ioctl_acquire_vm before calling kfd_ioctl_svm. Signed-off-by: Eric Huang Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 83a26c812e0529eb040d31a76f73e33e637243d4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 35ec67d9739b..3841943da5ec 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -3732,6 +3732,9 @@ svm_range_set_attr(struct kfd_process *p, struct mm_struct *mm, svms = &p->svms; + if (!process_info) + return -EINVAL; + mutex_lock(&process_info->lock); svm_range_list_lock_and_flush_work(svms, mm); From d8d9c820405eb1fcbde959de8898ad7d716a2d7b Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Mon, 18 May 2026 17:42:15 +0530 Subject: [PATCH 292/411] drm/amdgpu: simplify return value in amdgpu_userq_get_doorbell_index MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit amdgpu_userq_get_doorbell_index returns a uint64 type index as well as a int type failure values. Simplifying this and using a int type return value and getting the index in input pointer of type uint64 type. Also since it's used at once place making it static would be better. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit e947ec9d0529d5f93dbdb33cd197347f6a7b2922) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 21 +++++++++++---------- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 4 ---- 2 files changed, 11 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index d6390cc5a798..34c0d9ee94f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -536,12 +536,13 @@ void amdgpu_userq_destroy_object(struct amdgpu_userq_mgr *uq_mgr, amdgpu_bo_unref(&userq_obj->obj); } -uint64_t +static int amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_db_info *db_info, - struct drm_file *filp) + struct drm_file *filp, + u64 *index) { - uint64_t index; + u64 doorbell_index; struct drm_gem_object *gobj; struct amdgpu_userq_obj *db_obj = db_info->db_obj; int r, db_size; @@ -588,12 +589,13 @@ amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr, goto unpin_bo; } - index = amdgpu_doorbell_index_on_bar(uq_mgr->adev, db_obj->obj, - db_info->doorbell_offset, db_size); + doorbell_index = amdgpu_doorbell_index_on_bar(uq_mgr->adev, db_obj->obj, + db_info->doorbell_offset, db_size); drm_dbg_driver(adev_to_drm(uq_mgr->adev), - "[Usermode queues] doorbell index=%lld\n", index); + "[Usermode queues] doorbell index=%lld\n", doorbell_index); amdgpu_bo_unreserve(db_obj->obj); - return index; + *index = doorbell_index; + return 0; unpin_bo: amdgpu_bo_unpin(db_obj->obj); @@ -776,10 +778,9 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) db_info.doorbell_handle = queue->doorbell_handle; db_info.db_obj = &queue->db_obj; db_info.doorbell_offset = args->in.doorbell_offset; - index = amdgpu_userq_get_doorbell_index(uq_mgr, &db_info, filp); - if (index == (uint64_t)-EINVAL) { + r = amdgpu_userq_get_doorbell_index(uq_mgr, &db_info, filp, &index); + if (r) { drm_file_err(uq_mgr->file, "Failed to get doorbell for queue\n"); - r = -EINVAL; goto clean_mapping; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index 49b33e2d6932..033b8a0de6b1 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -163,10 +163,6 @@ void amdgpu_userq_evict(struct amdgpu_userq_mgr *uq_mgr); void amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *userq_mgr, struct amdgpu_eviction_fence_mgr *evf_mgr); -uint64_t amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr, - struct amdgpu_db_info *db_info, - struct drm_file *filp); - u32 amdgpu_userq_get_supported_ip_mask(struct amdgpu_device *adev); bool amdgpu_userq_enabled(struct drm_device *dev); From cedee93d43f893ce67e39b57c67240965c7c5a69 Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Mon, 18 May 2026 18:33:00 +0530 Subject: [PATCH 293/411] drm/amdgpu/userq: add amdgpu_bo_unpin when amdgpu_ttm_alloc_gart fails MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Unpin the wptr_obj->obj when amdgpu_ttm_alloc_gart fails. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit d8145c437ccdc2d91c579787290f82788172bea0) --- drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c index 5b4121ddc78c..026940fad524 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c @@ -81,7 +81,7 @@ mes_userq_create_wptr_mapping(struct amdgpu_device *adev, ret = amdgpu_ttm_alloc_gart(&wptr_obj->obj->tbo); if (ret) { DRM_ERROR("Failed to bind bo to GART. ret %d\n", ret); - goto fail_map; + goto fail_alloc_gart; } queue->wptr_obj.gpu_addr = amdgpu_bo_gpu_offset(wptr_obj->obj); @@ -89,6 +89,8 @@ mes_userq_create_wptr_mapping(struct amdgpu_device *adev, drm_exec_fini(&exec); return 0; +fail_alloc_gart: + amdgpu_bo_unpin(wptr_obj->obj); fail_map: amdgpu_bo_unref(&wptr_obj->obj); fail_lock: From a00caed2302c604c19a5cab781e34d7ba4fa7558 Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Mon, 18 May 2026 18:55:25 +0530 Subject: [PATCH 294/411] drm/amdgpu/userq: reserve root bo without interruption MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the code to make it an uninterruptible reservation for root bo. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit d409ab4e387d94b2e593d558b54b7bfd315e0e75) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 34c0d9ee94f2..5da107dffcce 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -620,11 +620,7 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que /* Cancel any pending hang detection work and cleanup */ cancel_delayed_work_sync(&queue->hang_detect_work); - r = amdgpu_bo_reserve(vm->root.bo, false); - if (r) { - drm_file_err(uq_mgr->file, "Failed to reserve root bo during userqueue destroy\n"); - return r; - } + amdgpu_bo_reserve(vm->root.bo, true); amdgpu_userq_buffer_vas_list_cleanup(adev, queue); amdgpu_bo_unreserve(vm->root.bo); From cf4aafdccefccc7f8236fed028d06725246e289e Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Mon, 18 May 2026 19:58:08 +0530 Subject: [PATCH 295/411] drm/amdgpu/userq: make sure queue is valid in the hang_detect_work MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Thread 1: Running amdgpu_userq_destroy which eventually remove the queue from door bell and set userq_mgr = NULL. Thread2: An interrupt might have scheduled the hang_detect_work which still need userq_mgr to be valid but could get an NULL ptrs. To fix that make sure we cancel the hang_detect_work again before setting userq_mgr to NULL. Along with that we also need all the queue va to remain valid till we could be running anything on the queue and hence moving the userq_va post hang_detect handler is cancelled. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 1a66ceb98b137d18d303b9889f0e7d8c4db73943) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 5da107dffcce..8d7dad1d30eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -427,8 +427,6 @@ static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue) xa_erase_irq(&adev->userq_doorbell_xa, queue->doorbell_index); amdgpu_userq_fence_driver_free(queue); queue->fence_drv = NULL; - queue->userq_mgr = NULL; - list_del(&queue->userq_va_list); up_read(&adev->reset_domain->sem); } @@ -619,11 +617,6 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que /* Cancel any pending hang detection work and cleanup */ cancel_delayed_work_sync(&queue->hang_detect_work); - - amdgpu_bo_reserve(vm->root.bo, true); - amdgpu_userq_buffer_vas_list_cleanup(adev, queue); - amdgpu_bo_unreserve(vm->root.bo); - mutex_lock(&uq_mgr->userq_mutex); amdgpu_userq_wait_for_last_fence(queue); @@ -635,6 +628,13 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que amdgpu_userq_cleanup(queue); mutex_unlock(&uq_mgr->userq_mutex); + cancel_delayed_work_sync(&queue->hang_detect_work); + amdgpu_bo_reserve(vm->root.bo, true); + amdgpu_userq_buffer_vas_list_cleanup(adev, queue); + amdgpu_bo_unreserve(vm->root.bo); + list_del(&queue->userq_va_list); + queue->userq_mgr = NULL; + amdgpu_bo_reserve(queue->db_obj.obj, true); amdgpu_bo_unpin(queue->db_obj.obj); amdgpu_bo_unreserve(queue->db_obj.obj); From 0fb4a8e64a9db74eeda8da7d0b78985392ae483b Mon Sep 17 00:00:00 2001 From: "Stanley.Yang" Date: Mon, 11 May 2026 16:49:19 +0800 Subject: [PATCH 296/411] drm/amdgpu: fix potential overflow in fs_info.debugfs_name Use snprintf() with sizeof(fs_info.debugfs_name) so a long RAS block name plus the "_err_inject" suffix cannot overflow the 32-byte buffer. Signed-off-by: Stanley.Yang Reviewed-by: Tao Zhou Signed-off-by: Alex Deucher (cherry picked from commit 1a58070fda26857a8f6acc0ab05428e60d5c6844) --- drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 6c644cfe6695..fc9f3adf9912 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -2280,7 +2280,8 @@ void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev) list_for_each_entry(obj, &con->head, node) { if (amdgpu_ras_is_supported(adev, obj->head.block) && (obj->attr_inuse == 1)) { - sprintf(fs_info.debugfs_name, "%s_err_inject", + snprintf(fs_info.debugfs_name, sizeof(fs_info.debugfs_name), + "%s_err_inject", get_ras_block_str(&obj->head)); fs_info.head = obj->head; amdgpu_ras_debugfs_create(adev, &fs_info, dir); From 6842b6a4b72da9b2906ffc5ca9d846ace2c54c14 Mon Sep 17 00:00:00 2001 From: David Francis Date: Thu, 14 May 2026 10:31:20 -0400 Subject: [PATCH 297/411] drm/amdkfd: Check for pdd drm file first in CRIU restore path CRIU restore ioctls are meant to be called by CRIU with no existing drm file. There's an error path for if the drm file unexpectedly exists. It was positioned so it was missing a fput(drm_file). Do that check earlier, as soon as we have the pdd. Signed-off-by: David Francis Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 2bab781dac78916c5cc8de76345a4102449267d7) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c index 03b266b26738..8785f7810157 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c @@ -2300,6 +2300,11 @@ static int criu_restore_devices(struct kfd_process *p, ret = -EINVAL; goto exit; } + + if (pdd->drm_file) { + ret = -EINVAL; + goto exit; + } pdd->user_gpu_id = device_buckets[i].user_gpu_id; drm_file = fget(device_buckets[i].drm_fd); @@ -2310,11 +2315,6 @@ static int criu_restore_devices(struct kfd_process *p, goto exit; } - if (pdd->drm_file) { - ret = -EINVAL; - goto exit; - } - /* create the vm using render nodes for kfd pdd */ if (kfd_process_device_init_vm(pdd, drm_file)) { pr_err("could not init vm for given pdd\n"); From dd4f3ee535b3b0ac027f75dbf9dc5fc88733c765 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Timur=20Krist=C3=B3f?= Date: Tue, 19 May 2026 10:41:54 +0200 Subject: [PATCH 298/411] drm/amd/pm/si: Disregard vblank time when no displays are connected MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When no displays are connected, there is no vblank happening so the power management code shouldn't worry about it. This fixes a regression that caused the memory clock to be stuck at maximum when there were no displays connected to a SI GPU. Fixes: 9003a0746864 ("drm/amd/pm: Treat zero vblank time as too short in si_dpm (v3)") Fixes: 9d73b107a61b ("drm/amd/pm: Use pm_display_cfg in legacy DPM (v2)") Reviewed-by: Alex Deucher Tested-by: Jeremy Klarenbeek Signed-off-by: Timur Kristóf Signed-off-by: Alex Deucher (cherry picked from commit 6d87e0199f7b83735b56e422d59f170a201897a8) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c index 36942467d4ad..c3aff5d0c53d 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -3076,6 +3076,10 @@ static bool si_dpm_vblank_too_short(void *handle) /* we never hit the non-gddr5 limit so disable it */ u32 switch_limit = adev->gmc.vram_type == AMDGPU_VRAM_TYPE_GDDR5 ? 450 : 0; + /* Disregard vblank time when there are no displays connected */ + if (!adev->pm.pm_display_cfg.num_display) + return false; + /* Consider zero vblank time too short and disable MCLK switching. * Note that the vblank time is set to maximum when no displays are attached, * so we'll still enable MCLK switching in that case. From ca8e7a119a2e4045324cffb8f9f58bedcc3dc928 Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Wed, 20 May 2026 16:13:09 +0530 Subject: [PATCH 299/411] drm/amdgpu/userq: remove amdgpu_userq_create/destroy_object wrapper MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Remove the amdgpu_userq_create/destroy_object wrappers and use directly the kernel bo allocation function which does all the things which are done in wrapper. Signed-off-by: Sunil Khatri Suggested-by: Christian König Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit deb02080ca5d3f015cf71e56067a39ef2f141998) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 67 ---------------------- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 7 --- drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 29 +++++++--- 3 files changed, 21 insertions(+), 82 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 8d7dad1d30eb..a93f5c238e55 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -465,74 +465,7 @@ amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *uq_mgr, dma_fence_put(ev_fence); } -int amdgpu_userq_create_object(struct amdgpu_userq_mgr *uq_mgr, - struct amdgpu_userq_obj *userq_obj, - int size) -{ - struct amdgpu_device *adev = uq_mgr->adev; - struct amdgpu_bo_param bp; - int r; - memset(&bp, 0, sizeof(bp)); - bp.byte_align = PAGE_SIZE; - bp.domain = AMDGPU_GEM_DOMAIN_GTT; - bp.flags = AMDGPU_GEM_CREATE_VRAM_CONTIGUOUS | - AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED; - bp.type = ttm_bo_type_kernel; - bp.size = size; - bp.resv = NULL; - bp.bo_ptr_size = sizeof(struct amdgpu_bo); - - r = amdgpu_bo_create(adev, &bp, &userq_obj->obj); - if (r) { - drm_file_err(uq_mgr->file, "Failed to allocate BO for userqueue (%d)", r); - return r; - } - - r = amdgpu_bo_reserve(userq_obj->obj, true); - if (r) { - drm_file_err(uq_mgr->file, "Failed to reserve BO to map (%d)", r); - goto free_obj; - } - - r = amdgpu_bo_pin(userq_obj->obj, AMDGPU_GEM_DOMAIN_GTT); - if (r) - goto unresv; - - r = amdgpu_ttm_alloc_gart(&(userq_obj->obj)->tbo); - if (r) { - drm_file_err(uq_mgr->file, "Failed to alloc GART for userqueue object (%d)", r); - goto unpin_bo; - } - - r = amdgpu_bo_kmap(userq_obj->obj, &userq_obj->cpu_ptr); - if (r) { - drm_file_err(uq_mgr->file, "Failed to map BO for userqueue (%d)", r); - goto unpin_bo; - } - - userq_obj->gpu_addr = amdgpu_bo_gpu_offset(userq_obj->obj); - amdgpu_bo_unreserve(userq_obj->obj); - memset(userq_obj->cpu_ptr, 0, size); - return 0; - -unpin_bo: - amdgpu_bo_unpin(userq_obj->obj); -unresv: - amdgpu_bo_unreserve(userq_obj->obj); -free_obj: - amdgpu_bo_unref(&userq_obj->obj); - - return r; -} - -void amdgpu_userq_destroy_object(struct amdgpu_userq_mgr *uq_mgr, - struct amdgpu_userq_obj *userq_obj) -{ - amdgpu_bo_kunmap(userq_obj->obj); - amdgpu_bo_unpin(userq_obj->obj); - amdgpu_bo_unref(&userq_obj->obj); -} static int amdgpu_userq_get_doorbell_index(struct amdgpu_userq_mgr *uq_mgr, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index 033b8a0de6b1..76ef5cfab52e 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -151,13 +151,6 @@ void amdgpu_userq_mgr_cancel_reset_work(struct amdgpu_device *adev); void amdgpu_userq_mgr_cancel_resume(struct amdgpu_userq_mgr *userq_mgr); void amdgpu_userq_mgr_fini(struct amdgpu_userq_mgr *userq_mgr); -int amdgpu_userq_create_object(struct amdgpu_userq_mgr *uq_mgr, - struct amdgpu_userq_obj *userq_obj, - int size); - -void amdgpu_userq_destroy_object(struct amdgpu_userq_mgr *uq_mgr, - struct amdgpu_userq_obj *userq_obj); - void amdgpu_userq_evict(struct amdgpu_userq_mgr *uq_mgr); void amdgpu_userq_ensure_ev_fence(struct amdgpu_userq_mgr *userq_mgr, diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c index 026940fad524..71251370c8b3 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c @@ -192,12 +192,16 @@ static int mes_userq_create_ctx_space(struct amdgpu_userq_mgr *uq_mgr, * for the same. */ size = AMDGPU_USERQ_PROC_CTX_SZ + AMDGPU_USERQ_GANG_CTX_SZ; - r = amdgpu_userq_create_object(uq_mgr, ctx, size); + r = amdgpu_bo_create_kernel(uq_mgr->adev, size, 0, + AMDGPU_GEM_DOMAIN_GTT, + &ctx->obj, &ctx->gpu_addr, + &ctx->cpu_ptr); if (r) { DRM_ERROR("Failed to allocate ctx space bo for userqueue, err:%d\n", r); return r; } + memset(ctx->cpu_ptr, 0, size); return 0; } @@ -270,13 +274,19 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, return -ENOMEM; } - r = amdgpu_userq_create_object(uq_mgr, &queue->mqd, - AMDGPU_MQD_SIZE_ALIGN(mqd_hw_default->mqd_size)); + r = amdgpu_bo_create_kernel(adev, + AMDGPU_MQD_SIZE_ALIGN(mqd_hw_default->mqd_size), + 0, AMDGPU_GEM_DOMAIN_GTT, + &queue->mqd.obj, &queue->mqd.gpu_addr, + &queue->mqd.cpu_ptr); if (r) { DRM_ERROR("Failed to create MQD object for userqueue\n"); goto free_props; } + memset(queue->mqd.cpu_ptr, 0, + AMDGPU_MQD_SIZE_ALIGN(mqd_hw_default->mqd_size)); + /* Initialize the MQD BO with user given values */ userq_props->wptr_gpu_addr = mqd_user->wptr_va; userq_props->rptr_gpu_addr = mqd_user->rptr_va; @@ -432,10 +442,12 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, return 0; free_ctx: - amdgpu_userq_destroy_object(uq_mgr, &queue->fw_obj); + amdgpu_bo_free_kernel(&queue->fw_obj.obj, &queue->fw_obj.gpu_addr, + &queue->fw_obj.cpu_ptr); free_mqd: - amdgpu_userq_destroy_object(uq_mgr, &queue->mqd); + amdgpu_bo_free_kernel(&queue->mqd.obj, &queue->mqd.gpu_addr, + &queue->mqd.cpu_ptr); free_props: kfree(userq_props); @@ -445,11 +457,12 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, static void mes_userq_mqd_destroy(struct amdgpu_usermode_queue *queue) { - struct amdgpu_userq_mgr *uq_mgr = queue->userq_mgr; - amdgpu_userq_destroy_object(uq_mgr, &queue->fw_obj); + amdgpu_bo_free_kernel(&queue->fw_obj.obj, &queue->fw_obj.gpu_addr, + &queue->fw_obj.cpu_ptr); kfree(queue->userq_prop); - amdgpu_userq_destroy_object(uq_mgr, &queue->mqd); + amdgpu_bo_free_kernel(&queue->mqd.obj, &queue->mqd.gpu_addr, + &queue->mqd.cpu_ptr); } static int mes_userq_preempt(struct amdgpu_usermode_queue *queue) From 93f5534b35a05ef8a0109c1eefa800062fee810a Mon Sep 17 00:00:00 2001 From: Eric Huang Date: Tue, 12 May 2026 10:19:52 -0400 Subject: [PATCH 300/411] drm/amdkfd: fix a vulnerability of integer overflow in kfd debugger get_queue_ids() computes array_size = num_queues * sizeof(uint32_t), which could overflow on 32-bit size_t build. using array_size() instead, it saturates to SIZE_MAX on overflow. Signed-off-by: Eric Huang Acked-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 2d57a0475f085c08b49312dfd8edcb461845f285) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index e0a31e11f0ff..0d7296c739ed 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -3308,12 +3308,14 @@ static void copy_context_work_handler(struct work_struct *work) static uint32_t *get_queue_ids(uint32_t num_queues, uint32_t *usr_queue_id_array) { - size_t array_size = num_queues * sizeof(uint32_t); - if (!usr_queue_id_array) return NULL; - return memdup_user(usr_queue_id_array, array_size); + if (num_queues > KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) + return ERR_PTR(-EINVAL); + + return memdup_user(usr_queue_id_array, + array_size(num_queues, sizeof(uint32_t))); } int resume_queues(struct kfd_process *p, From 0b8a1600ab50f331aeeba47c777a1b34cba606bf Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Wed, 20 May 2026 16:25:50 +0530 Subject: [PATCH 301/411] drm/amdgpu/userq: move mqd_destroy to later stage to keep core obj valid MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit mqd_destroy cleans up queue core objects like mqd and fw_object which are needed for any pending fence to signal properly. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 4ad65d610096498c8e265615aba42b3c47441bb5) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index a93f5c238e55..28a1849e7dcd 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -417,12 +417,10 @@ static void amdgpu_userq_cleanup(struct amdgpu_usermode_queue *queue) { struct amdgpu_userq_mgr *uq_mgr = queue->userq_mgr; struct amdgpu_device *adev = uq_mgr->adev; - const struct amdgpu_userq_funcs *uq_funcs = adev->userq_funcs[queue->queue_type]; /* Wait for mode-1 reset to complete */ down_read(&adev->reset_domain->sem); - uq_funcs->mqd_destroy(queue); /* Use interrupt-safe locking since IRQ handlers may access these XArrays */ xa_erase_irq(&adev->userq_doorbell_xa, queue->doorbell_index); amdgpu_userq_fence_driver_free(queue); @@ -541,15 +539,15 @@ static int amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_queue *queue) { struct amdgpu_device *adev = uq_mgr->adev; - struct amdgpu_fpriv *fpriv = uq_mgr_to_fpriv(uq_mgr); - struct amdgpu_vm *vm = &fpriv->vm; - + const struct amdgpu_userq_funcs *uq_funcs = adev->userq_funcs[queue->queue_type]; + struct amdgpu_vm *vm = queue->vm; int r = 0; cancel_delayed_work_sync(&uq_mgr->resume_work); /* Cancel any pending hang detection work and cleanup */ cancel_delayed_work_sync(&queue->hang_detect_work); + mutex_lock(&uq_mgr->userq_mutex); amdgpu_userq_wait_for_last_fence(queue); @@ -566,6 +564,7 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que amdgpu_userq_buffer_vas_list_cleanup(adev, queue); amdgpu_bo_unreserve(vm->root.bo); list_del(&queue->userq_va_list); + uq_funcs->mqd_destroy(queue); queue->userq_mgr = NULL; amdgpu_bo_reserve(queue->db_obj.obj, true); From 181307acf8ea597ad63fd574b44d0f98a329a61b Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Wed, 20 May 2026 16:39:49 +0530 Subject: [PATCH 302/411] drm/amdgpu/userq: use array instead of list for userq_vas MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use arrays instead of list for userq_vas since we have fixed no of bos. Also, we dont have to worry to free that memory later since this array would be free along with queue only. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit ef7dc711a664b0c548ecfdf13a00436b7446b8e7) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 88 ++++++---------------- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 20 +++-- drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 14 ++-- 3 files changed, 45 insertions(+), 77 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 28a1849e7dcd..cf192500800f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -215,33 +215,15 @@ void amdgpu_userq_process_fence_irq(struct amdgpu_device *adev, u32 doorbell) xa_unlock_irqrestore(xa, flags); } -static int amdgpu_userq_buffer_va_list_add(struct amdgpu_usermode_queue *queue, - struct amdgpu_bo_va_mapping *va_map, u64 addr) -{ - struct amdgpu_userq_va_cursor *va_cursor; - struct userq_va_list; - - va_cursor = kzalloc_obj(*va_cursor); - if (!va_cursor) - return -ENOMEM; - - INIT_LIST_HEAD(&va_cursor->list); - va_cursor->gpu_addr = addr; - va_map->bo_va->userq_va_mapped = true; - list_add(&va_cursor->list, &queue->userq_va_list); - - return 0; -} - int amdgpu_userq_input_va_validate(struct amdgpu_device *adev, struct amdgpu_usermode_queue *queue, - u64 addr, u64 expected_size) + u64 addr, u64 expected_size, + u64 *va_out) { struct amdgpu_bo_va_mapping *va_map; struct amdgpu_vm *vm = queue->vm; u64 user_addr; u64 size; - int r = 0; /* Caller must hold vm->root.bo reservation */ dma_resv_assert_held(queue->vm->root.bo->tbo.base.resv); @@ -250,20 +232,18 @@ int amdgpu_userq_input_va_validate(struct amdgpu_device *adev, size = expected_size >> AMDGPU_GPU_PAGE_SHIFT; va_map = amdgpu_vm_bo_lookup_mapping(vm, user_addr); - if (!va_map) { - r = -EINVAL; - goto out_err; - } + if (!va_map) + return -EINVAL; + /* Only validate the userq whether resident in the VM mapping range */ if (user_addr >= va_map->start && va_map->last - user_addr + 1 >= size) { - amdgpu_userq_buffer_va_list_add(queue, va_map, user_addr); + va_map->bo_va->userq_va_mapped = true; + *va_out = user_addr; return 0; } - r = -EINVAL; -out_err: - return r; + return -EINVAL; } static bool amdgpu_userq_buffer_va_mapped(struct amdgpu_vm *vm, u64 addr) @@ -284,14 +264,16 @@ static bool amdgpu_userq_buffer_va_mapped(struct amdgpu_vm *vm, u64 addr) static bool amdgpu_userq_buffer_vas_mapped(struct amdgpu_usermode_queue *queue) { - struct amdgpu_userq_va_cursor *va_cursor, *tmp; - int r = 0; + int i, r = 0; - list_for_each_entry_safe(va_cursor, tmp, &queue->userq_va_list, list) { - r += amdgpu_userq_buffer_va_mapped(queue->vm, va_cursor->gpu_addr); + for (i = 0; i < ARRAY_SIZE(queue->userq_vas.va_array); i++) { + if (!queue->userq_vas.va_array[i]) + continue; + r += amdgpu_userq_buffer_va_mapped(queue->vm, + queue->userq_vas.va_array[i]); dev_dbg(queue->userq_mgr->adev->dev, "validate the userq mapping:%p va:%llx r:%d\n", - queue, va_cursor->gpu_addr, r); + queue, queue->userq_vas.va_array[i], r); } if (r != 0) @@ -300,24 +282,7 @@ static bool amdgpu_userq_buffer_vas_mapped(struct amdgpu_usermode_queue *queue) return false; } -static void amdgpu_userq_buffer_vas_list_cleanup(struct amdgpu_device *adev, - struct amdgpu_usermode_queue *queue) -{ - struct amdgpu_userq_va_cursor *va_cursor, *tmp; - struct amdgpu_bo_va_mapping *mapping; - /* Caller must hold vm->root.bo reservation */ - dma_resv_assert_held(queue->vm->root.bo->tbo.base.resv); - - list_for_each_entry_safe(va_cursor, tmp, &queue->userq_va_list, list) { - mapping = amdgpu_vm_bo_lookup_mapping(queue->vm, va_cursor->gpu_addr); - if (mapping) - dev_dbg(adev->dev, "delete the userq:%p va:%llx\n", - queue, va_cursor->gpu_addr); - list_del(&va_cursor->list); - kfree(va_cursor); - } -} static int amdgpu_userq_preempt_helper(struct amdgpu_usermode_queue *queue) { @@ -540,7 +505,6 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que { struct amdgpu_device *adev = uq_mgr->adev; const struct amdgpu_userq_funcs *uq_funcs = adev->userq_funcs[queue->queue_type]; - struct amdgpu_vm *vm = queue->vm; int r = 0; cancel_delayed_work_sync(&uq_mgr->resume_work); @@ -560,10 +524,6 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que mutex_unlock(&uq_mgr->userq_mutex); cancel_delayed_work_sync(&queue->hang_detect_work); - amdgpu_bo_reserve(vm->root.bo, true); - amdgpu_userq_buffer_vas_list_cleanup(adev, queue); - amdgpu_bo_unreserve(vm->root.bo); - list_del(&queue->userq_va_list); uq_funcs->mqd_destroy(queue); queue->userq_mgr = NULL; @@ -669,7 +629,6 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) } kref_init(&queue->refcount); - INIT_LIST_HEAD(&queue->userq_va_list); queue->doorbell_handle = args->in.doorbell_handle; queue->queue_type = args->in.ip_type; queue->vm = &fpriv->vm; @@ -690,14 +649,17 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) goto free_fence_drv; if (amdgpu_userq_input_va_validate(adev, queue, args->in.queue_va, - args->in.queue_size) || + args->in.queue_size, + &queue->userq_vas.va.queue_rb) || amdgpu_userq_input_va_validate(adev, queue, args->in.rptr_va, - AMDGPU_GPU_PAGE_SIZE) || + AMDGPU_GPU_PAGE_SIZE, + &queue->userq_vas.va.rptr) || amdgpu_userq_input_va_validate(adev, queue, args->in.wptr_va, - AMDGPU_GPU_PAGE_SIZE)) { + AMDGPU_GPU_PAGE_SIZE, + &queue->userq_vas.va.wptr)) { r = -EINVAL; amdgpu_bo_unreserve(fpriv->vm.root.bo); - goto clean_mapping; + goto free_fence_drv; } amdgpu_bo_unreserve(fpriv->vm.root.bo); @@ -709,7 +671,7 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) r = amdgpu_userq_get_doorbell_index(uq_mgr, &db_info, filp, &index); if (r) { drm_file_err(uq_mgr->file, "Failed to get doorbell for queue\n"); - goto clean_mapping; + goto free_fence_drv; } queue->doorbell_index = index; @@ -769,10 +731,6 @@ amdgpu_userq_create(struct drm_file *filp, union drm_amdgpu_userq *args) amdgpu_bo_unpin(queue->db_obj.obj); amdgpu_bo_unreserve(queue->db_obj.obj); amdgpu_bo_unref(&queue->db_obj.obj); -clean_mapping: - amdgpu_bo_reserve(fpriv->vm.root.bo, true); - amdgpu_userq_buffer_vas_list_cleanup(adev, queue); - amdgpu_bo_unreserve(fpriv->vm.root.bo); free_fence_drv: amdgpu_userq_fence_driver_free(queue); free_queue: diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index 76ef5cfab52e..28cfc6682333 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -48,11 +48,6 @@ struct amdgpu_userq_obj { struct amdgpu_bo *obj; }; -struct amdgpu_userq_va_cursor { - u64 gpu_addr; - struct list_head list; -}; - struct amdgpu_usermode_queue { int queue_type; enum amdgpu_userq_state state; @@ -93,7 +88,17 @@ struct amdgpu_usermode_queue { struct delayed_work hang_detect_work; struct kref refcount; - struct list_head userq_va_list; + union { + struct { + u64 queue_rb; + u64 wptr; + u64 rptr; + u64 eop; + u64 shadow; + u64 csa; + } va; + u64 va_array[6]; + } userq_vas; }; struct amdgpu_userq_funcs { @@ -174,7 +179,8 @@ void amdgpu_userq_process_fence_irq(struct amdgpu_device *adev, u32 doorbell); int amdgpu_userq_input_va_validate(struct amdgpu_device *adev, struct amdgpu_usermode_queue *queue, - u64 addr, u64 expected_size); + u64 addr, u64 expected_size, u64 *va_out); + void amdgpu_userq_gem_va_unmap_validate(struct amdgpu_device *adev, struct amdgpu_bo_va_mapping *mapping, uint64_t saddr); diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c index 71251370c8b3..98aa00eeb2f4 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c @@ -318,8 +318,9 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, kfree(compute_mqd); goto free_mqd; } - r = amdgpu_userq_input_va_validate(adev, queue, compute_mqd->eop_va, - 2048); + r = amdgpu_userq_input_va_validate(adev, queue, + compute_mqd->eop_va, 2048, + &queue->userq_vas.va.eop); amdgpu_bo_unreserve(queue->vm->root.bo); if (r) { kfree(compute_mqd); @@ -368,7 +369,8 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, goto free_mqd; } r = amdgpu_userq_input_va_validate(adev, queue, mqd_gfx_v11->shadow_va, - shadow_info.shadow_size); + shadow_info.shadow_size, + &queue->userq_vas.va.shadow); if (r) { amdgpu_bo_unreserve(queue->vm->root.bo); kfree(mqd_gfx_v11); @@ -376,7 +378,8 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, } r = amdgpu_userq_input_va_validate(adev, queue, mqd_gfx_v11->csa_va, - shadow_info.csa_size); + shadow_info.csa_size, + &queue->userq_vas.va.csa); amdgpu_bo_unreserve(queue->vm->root.bo); if (r) { kfree(mqd_gfx_v11); @@ -406,7 +409,8 @@ static int mes_userq_mqd_create(struct amdgpu_usermode_queue *queue, goto free_mqd; } r = amdgpu_userq_input_va_validate(adev, queue, mqd_sdma_v11->csa_va, - 32); + 32, + &queue->userq_vas.va.csa); amdgpu_bo_unreserve(queue->vm->root.bo); if (r) { kfree(mqd_sdma_v11); From 962d684b5dc0741dcd93485d41b450de402d5592 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Wed, 18 Feb 2026 12:53:27 +0100 Subject: [PATCH 303/411] drm/amdgpu: fix amdgpu_hmm_range_get_pages MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The notifier sequence must only be read once or otherwise we could work with invalid pages. While at it also fix the coding style, e.g. drop the pre-initialized return value and use the common define for 2G range. Signed-off-by: Christian König Reviewed-by: Vitaly Prosyak Tested-by: Vitaly Prosyak Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit c08972f555945cda57b0adb72272a37910153390) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c index f72990ac046e..0a9582da3a33 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c @@ -51,8 +51,6 @@ #include "amdgpu_amdkfd.h" #include "amdgpu_hmm.h" -#define MAX_WALK_BYTE (2UL << 30) - /** * amdgpu_hmm_invalidate_gfx - callback to notify about mm change * @@ -170,11 +168,13 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier, void *owner, struct amdgpu_hmm_range *range) { - unsigned long end; + const u64 max_bytes = SZ_2G; + + struct hmm_range *hmm_range = &range->hmm_range; unsigned long timeout; unsigned long *pfns; - int r = 0; - struct hmm_range *hmm_range = &range->hmm_range; + unsigned long end; + int r; pfns = kvmalloc_array(npages, sizeof(*pfns), GFP_KERNEL); if (unlikely(!pfns)) { @@ -191,8 +191,9 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier, end = start + npages * PAGE_SIZE; hmm_range->dev_private_owner = owner; + hmm_range->notifier_seq = mmu_interval_read_begin(notifier); do { - hmm_range->end = min(hmm_range->start + MAX_WALK_BYTE, end); + hmm_range->end = min(hmm_range->start + max_bytes, end); pr_debug("hmm range: start = 0x%lx, end = 0x%lx", hmm_range->start, hmm_range->end); @@ -200,7 +201,6 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier, timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT); retry: - hmm_range->notifier_seq = mmu_interval_read_begin(notifier); r = hmm_range_fault(hmm_range); if (unlikely(r)) { if (r == -EBUSY && !time_after(jiffies, timeout)) @@ -210,7 +210,7 @@ int amdgpu_hmm_range_get_pages(struct mmu_interval_notifier *notifier, if (hmm_range->end == end) break; - hmm_range->hmm_pfns += MAX_WALK_BYTE >> PAGE_SHIFT; + hmm_range->hmm_pfns += max_bytes >> PAGE_SHIFT; hmm_range->start = hmm_range->end; } while (hmm_range->end < end); From 1c824497d8acd3187d585d6187cedc1897dcc871 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Wed, 18 Feb 2026 12:31:29 +0100 Subject: [PATCH 304/411] drm/amdgpu: fix calling VM invalidation in amdgpu_hmm_invalidate_gfx MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Otherwise we don't invalidate page tables on next CS. Signed-off-by: Christian König Reviewed-by: Vitaly Prosyak Tested-by: Vitaly Prosyak Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit b6444d1bcbc34f6f2a31a3aab3059be082f3683e) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c | 1 + drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 7 +++++-- 2 files changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c index 0a9582da3a33..5bfa5a84b09c 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c @@ -76,6 +76,7 @@ static bool amdgpu_hmm_invalidate_gfx(struct mmu_interval_notifier *mni, mmu_interval_set_seq(mni, cur_seq); + amdgpu_vm_bo_invalidate(bo, false); r = dma_resv_wait_timeout(bo->tbo.base.resv, DMA_RESV_USAGE_BOOKKEEP, false, MAX_SCHEDULE_TIMEOUT); mutex_unlock(&adev->notifier_lock); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index fccd758b6699..c9f88ecce1a7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1631,6 +1631,7 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev, { struct amdgpu_bo_va *bo_va; struct dma_resv *resv; + struct amdgpu_bo *bo; bool clear, unlock; int r; @@ -1650,11 +1651,13 @@ int amdgpu_vm_handle_moved(struct amdgpu_device *adev, while (!list_empty(&vm->invalidated)) { bo_va = list_first_entry(&vm->invalidated, struct amdgpu_bo_va, base.vm_status); - resv = bo_va->base.bo->tbo.base.resv; + bo = bo_va->base.bo; + resv = bo->tbo.base.resv; spin_unlock(&vm->status_lock); /* Try to reserve the BO to avoid clearing its ptes */ - if (!adev->debug_vm && dma_resv_trylock(resv)) { + if (!adev->debug_vm && !amdgpu_ttm_tt_get_usermm(bo->tbo.ttm) && + dma_resv_trylock(resv)) { clear = false; unlock = true; /* The caller is already holding the reservation lock */ From bfa9d28960ed677d556bdf097073bc3129686229 Mon Sep 17 00:00:00 2001 From: Pavitra Jha Date: Thu, 21 May 2026 04:04:14 -0400 Subject: [PATCH 305/411] Bluetooth: hci_conn: Fix memory leak in hci_le_big_terminate() hci_le_big_terminate() allocates iso_list_data via kzalloc_obj but returns 0 without freeing it when neither pa_sync_term nor big_sync_term flags are set after evaluating the PA and BIG sync connection state. This early-return path was introduced when hci_le_big_terminate() was refactored to take struct hci_conn instead of raw u8 parameters, adding PA/BIG flag evaluation logic. The existing kfree() on hci_cmd_sync_queue failure does not cover this path. Fixes: a7bcffc673de ("Bluetooth: Add PA_LINK to distinguish BIG sync and PA sync connections") Cc: stable@vger.kernel.org Signed-off-by: Pavitra Jha Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 17b46ad6a349..54eabaa46960 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -870,8 +870,10 @@ static int hci_le_big_terminate(struct hci_dev *hdev, struct hci_conn *conn) d->big_sync_term = true; } - if (!d->pa_sync_term && !d->big_sync_term) + if (!d->pa_sync_term && !d->big_sync_term) { + kfree(d); return 0; + } ret = hci_cmd_sync_queue(hdev, big_terminate_sync, d, terminate_big_destroy); From 9dbd84990394c51f5cee1e8871bb5ff8af5ed939 Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 20 May 2026 22:30:36 -0400 Subject: [PATCH 306/411] Bluetooth: L2CAP: fix chan ref leak in l2cap_chan_timeout() on !conn __set_chan_timer() takes a l2cap_chan reference via l2cap_chan_hold() before scheduling the delayed work. The normal path in l2cap_chan_timeout() drops this reference with l2cap_chan_put() at the end, but the early return when chan->conn is NULL skips the put, leaking the reference. Add the missing l2cap_chan_put() before the early return. Fixes: adf0398cee86 ("Bluetooth: l2cap: fix null-ptr-deref in l2cap_chan_timeout") Cc: stable@vger.kernel.org Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index fdccd62ccca8..5668c92b3f58 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -411,8 +411,10 @@ static void l2cap_chan_timeout(struct work_struct *work) BT_DBG("chan %p state %s", chan, state_to_string(chan->state)); - if (!conn) + if (!conn) { + l2cap_chan_put(chan); return; + } mutex_lock(&conn->lock); /* __set_chan_timer() calls l2cap_chan_hold(chan) while scheduling From 8c8e620467a7b51562dbcefbd1f09f288d7d710d Mon Sep 17 00:00:00 2001 From: Siwei Zhang Date: Wed, 20 May 2026 22:12:20 -0400 Subject: [PATCH 307/411] Bluetooth: L2CAP: use chan timer to close channels in cleanup_listen() l2cap_chan_close() removes the channel from conn->chan_l, which must be done under conn->lock. cleanup_listen() runs under the parent sk_lock, so acquiring conn->lock would invert the established conn->lock -> chan->lock -> sk_lock order. Instead of calling l2cap_chan_close() directly, schedule l2cap_chan_timeout with delay 0 to close the channel asynchronously. The timeout handler already acquires conn->lock and chan->lock in the correct order. The timer is only armed when chan->conn is still set: if it is already NULL, l2cap_conn_del() has already processed this channel (l2cap_chan_del + l2cap_sock_teardown_cb + l2cap_sock_close_cb), so there is nothing left to do. If l2cap_conn_del() races in after the timer is armed, __clear_chan_timer() inside l2cap_chan_del() cancels it; if the timer has already fired, the handler returns harmlessly because chan->conn was cleared. Fixes: 3df91ea20e74 ("Bluetooth: Revert to mutexes from RCU list") Cc: # 0b58004: Bluetooth: fix UAF in l2cap_sock_cleanup_listen() vs l2cap_conn_del() Signed-off-by: Siwei Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_sock.c | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index b34e7da8d906..c138aa4ae266 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1499,6 +1499,10 @@ static void l2cap_sock_cleanup_listen(struct sock *parent) * pin it (hold_unless_zero() additionally skips a chan already past * its last reference). We then drop the sk lock before taking * chan->lock, so sk and chan locks are never held together. + * + * Since we cannot call l2cap_chan_close() without conn->lock, + * schedule l2cap_chan_timeout to close the channel; it already + * acquires conn->lock -> chan->lock in the correct order. */ while ((sk = bt_accept_dequeue(parent, NULL))) { struct l2cap_chan *chan; @@ -1516,14 +1520,12 @@ static void l2cap_sock_cleanup_listen(struct sock *parent) state_to_string(chan->state)); l2cap_chan_lock(chan); - __clear_chan_timer(chan); - l2cap_chan_close(chan, ECONNRESET); - /* l2cap_conn_del() may already have killed this socket - * (it sets SOCK_DEAD); skip the duplicate to avoid a - * double sock_put()/l2cap_chan_put(). + /* Since we cannot call l2cap_chan_close() without + * conn->lock, schedule its timer to trigger the close + * and cleanup of this channel. */ - if (!sock_flag(sk, SOCK_DEAD)) - l2cap_sock_kill(sk); + if (chan->conn) + __set_chan_timer(chan, 0); l2cap_chan_unlock(chan); l2cap_chan_put(chan); From 2a3ac9ee11dbb9845f3947cef4a79dba658cf6f6 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Wed, 20 May 2026 18:56:43 -0400 Subject: [PATCH 308/411] Bluetooth: HIDP: fix missing length checks in hidp_input_report() hidp_input_report() reads keyboard and mouse payload data from an skb without first verifying that skb->len contains enough data. hidp_recv_intr_frame() pulls the 1-byte HIDP header before dispatching to hidp_input_report(). If a paired device sends a truncated packet, the handler reads beyond the valid skb data, resulting in an out-of-bounds read of skb data. The OOB bytes may be interpreted as phantom key presses or spurious mouse movement. Replace the open-coded length tracking and pointer arithmetic with skb_pull_data() calls. skb_pull_data() returns NULL if the requested bytes are not present, eliminating the need for a manual size variable and the separate skb->len guard. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hidp/core.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/net/bluetooth/hidp/core.c b/net/bluetooth/hidp/core.c index 976f91eeb745..70344bd3248a 100644 --- a/net/bluetooth/hidp/core.c +++ b/net/bluetooth/hidp/core.c @@ -179,12 +179,21 @@ static void hidp_input_report(struct hidp_session *session, struct sk_buff *skb) { struct input_dev *dev = session->input; unsigned char *keys = session->keys; - unsigned char *udata = skb->data + 1; - signed char *sdata = skb->data + 1; - int i, size = skb->len - 1; + unsigned char *udata; + signed char *sdata; + u8 *hdr; + int i; - switch (skb->data[0]) { + hdr = skb_pull_data(skb, 1); + if (!hdr) + return; + + switch (*hdr) { case 0x01: /* Keyboard report */ + udata = skb_pull_data(skb, 8); + if (!udata) + break; + for (i = 0; i < 8; i++) input_report_key(dev, hidp_keycode[i + 224], (udata[0] >> i) & 1); @@ -213,6 +222,10 @@ static void hidp_input_report(struct hidp_session *session, struct sk_buff *skb) break; case 0x02: /* Mouse report */ + sdata = skb_pull_data(skb, 3); + if (!sdata) + break; + input_report_key(dev, BTN_LEFT, sdata[0] & 0x01); input_report_key(dev, BTN_RIGHT, sdata[0] & 0x02); input_report_key(dev, BTN_MIDDLE, sdata[0] & 0x04); @@ -222,7 +235,7 @@ static void hidp_input_report(struct hidp_session *session, struct sk_buff *skb) input_report_rel(dev, REL_X, sdata[1]); input_report_rel(dev, REL_Y, sdata[2]); - if (size > 3) + if (skb->len > 0) input_report_rel(dev, REL_WHEEL, sdata[3]); break; } From 82855073c1081732656734b74d7d1d5e4cfd0da7 Mon Sep 17 00:00:00 2001 From: Shuai Zhang Date: Thu, 21 May 2026 13:25:47 +0800 Subject: [PATCH 309/411] Bluetooth: btusb: Allow firmware re-download when version matches The Bluetooth host decides whether to download firmware by reading the controller firmware download completion flag and firmware version information. If a USB error occurs during the firmware download process (for example due to a USB disconnect), the download is aborted immediately. An incomplete firmware transfer does not cause the controller to set the download completion flag, but the firmware version information may be updated at an early stage of the download process. In this case, after USB reconnection, the host attempts to re-download the firmware because the download completion flag is not set. However, since the controller reports the same firmware version as the target firmware, the download is skipped. This ultimately results in the firmware not being properly updated on the controller. This change removes the restriction that skips firmware download when the versions are equal. It covers scenarios where the USB connection can be disconnected at any time and ensures that firmware download can be retriggered after USB reconnection, allowing the Bluetooth firmware to be correctly and completely updated. Fixes: 3267c884cefa ("Bluetooth: btusb: Add support for QCA ROME chipset family") Cc: stable@vger.kernel.org Signed-off-by: Shuai Zhang Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index 7f5fce93d984..830fefb342c6 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -3540,7 +3540,13 @@ static int btusb_setup_qca_load_rampatch(struct hci_dev *hdev, "firmware rome 0x%x build 0x%x", rver_rom, rver_patch, ver_rom, ver_patch); - if (rver_rom != ver_rom || rver_patch <= ver_patch) { + /* Allow rampatch when the patch version equals the firmware version. + * A firmware download may be aborted by a transient USB error (e.g. + * disconnect) after the controller updates version info but before + * completion. + * Allowing equal versions enables re-flashing during recovery. + */ + if (rver_rom != ver_rom || rver_patch < ver_patch) { bt_dev_err(hdev, "rampatch file version did not match with firmware"); err = -EINVAL; goto done; From 3c40d381ce04f9575a5d8b542898183c3b4b38dc Mon Sep 17 00:00:00 2001 From: Zhao Dongdong Date: Tue, 26 May 2026 11:21:39 +0800 Subject: [PATCH 310/411] Bluetooth: 6lowpan: check skb_clone() return value in send_mcast_pkt() The skb_clone() function can return NULL if memory allocation fails. send_mcast_pkt() calls skb_clone() without checking the return value, which can lead to a NULL pointer dereference in send_pkt() when it dereferences skb->data. Add a NULL check after skb_clone() and skip the peer if the clone fails. Fixes: 18722c247023 ("Bluetooth: Enable 6LoWPAN support for BT LE devices") Signed-off-by: Zhao Dongdong Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/6lowpan.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/bluetooth/6lowpan.c b/net/bluetooth/6lowpan.c index 2f03b780b40d..960a19b3e26d 100644 --- a/net/bluetooth/6lowpan.c +++ b/net/bluetooth/6lowpan.c @@ -486,6 +486,8 @@ static int send_mcast_pkt(struct sk_buff *skb, struct net_device *netdev) int ret; local_skb = skb_clone(skb, GFP_ATOMIC); + if (!local_skb) + continue; BT_DBG("xmit %s to %pMR type %u IP %pI6c chan %p", netdev->name, From bfea6091e0fffb270c20e74384b660910277eb6c Mon Sep 17 00:00:00 2001 From: Doruk Tan Ozturk Date: Mon, 25 May 2026 18:24:38 +0200 Subject: [PATCH 311/411] Bluetooth: hci_sync: fix UAF in hci_le_create_cis_sync hci_le_create_cis_sync() dereferences conn->conn_timeout after releasing both rcu_read_lock() and hci_dev_lock(hdev). The conn pointer was obtained from an RCU-protected iteration over hdev->conn_hash.list and is not valid once these locks are dropped. A concurrent disconnect can free the hci_conn between the unlock and the dereference, causing a use-after-free read. The cancellation mechanism in hci_conn_del() cannot prevent this because hci_le_create_cis_pending() queues hci_create_cis_sync with data=NULL: hci_cmd_sync_queue(hdev, hci_create_cis_sync, NULL, NULL); While hci_conn_del() dequeues with data=conn: hci_cmd_sync_dequeue(hdev, NULL, conn, NULL); Since NULL != conn, the lookup in _hci_cmd_sync_lookup_entry() never matches, and the pending work item is not cancelled. Fix this by saving conn->conn_timeout into a local variable while the locks are still held, so the stale conn pointer is never dereferenced after unlock. This is the same class of bug as the one fixed by commit 035c25007c9e ("Bluetooth: hci_sync: Fix UAF on le_read_features_complete") which addressed the identical pattern in a different function. This vulnerability was identified using 0sec.ai, an open-source automated security auditing platform (https://github.com/0sec-labs). Fixes: c09b80be6ffc ("Bluetooth: hci_conn: Fix not waiting for HCI_EVT_LE_CIS_ESTABLISHED") Cc: stable@vger.kernel.org Reported-by: Doruk Tan Ozturk Signed-off-by: Doruk Tan Ozturk Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index aff8562a8690..1faf8df6d159 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -6699,6 +6699,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) DEFINE_FLEX(struct hci_cp_le_create_cis, cmd, cis, num_cis, 0x1f); size_t aux_num_cis = 0; struct hci_conn *conn; + u16 timeout = 0; u8 cig = BT_ISO_QOS_CIG_UNSET; /* The spec allows only one pending LE Create CIS command at a time. If @@ -6769,6 +6770,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) set_bit(HCI_CONN_CREATE_CIS, &conn->flags); cis->acl_handle = cpu_to_le16(conn->parent->handle); cis->cis_handle = cpu_to_le16(conn->handle); + timeout = conn->conn_timeout; aux_num_cis++; if (aux_num_cis >= cmd->num_cis) @@ -6788,7 +6790,7 @@ int hci_le_create_cis_sync(struct hci_dev *hdev) return __hci_cmd_sync_status_sk(hdev, HCI_OP_LE_CREATE_CIS, struct_size(cmd, cis, cmd->num_cis), cmd, HCI_EVT_LE_CIS_ESTABLISHED, - conn->conn_timeout, NULL); + timeout, NULL); } int hci_le_remove_cig_sync(struct hci_dev *hdev, u8 handle) From fa21e86caba2347e89eb65af926205a36a097c53 Mon Sep 17 00:00:00 2001 From: Shuai Zhang Date: Mon, 25 May 2026 14:51:56 +0800 Subject: [PATCH 312/411] Bluetooth: hci_qca: Use 100 ms SSR delay for rampatch and NVM loading When bt_en is pulled high by hardware, the host does not re-download the firmware after SSR. The controller loads the rampatch and NVM internally. On HMT chip, the rampatch is ~264 KB and the NVM is ~9.4 KB. The loading process takes approximately 70 ms. The previous 50 ms delay is too short, causing the controller to not respond to the reset command sent by the host, which leads to BT initialization failure: Bluetooth: hci0: QCA memdump Done, received 458752, total 458752 Bluetooth: hci0: mem_dump_status: 2 Bluetooth: hci0: Opcode 0x0c03 failed: -110 Increase the delay to 100 ms, which was confirmed as a safe value by the controller, to ensure the controller has finished loading the firmware before the host sends commands. Steps to reproduce: 1. Trigger SSR and wait for SSR to complete: hcitool cmd 0x3f 0c 26 2. Run "bluetoothctl power on" and observe that BT fails to start. Fixes: fce1a9244a0f ("Bluetooth: hci_qca: Fix SSR (SubSystem Restart) fail when BT_EN is pulled up by hw") Cc: stable@vger.kernel.org Reviewed-by: Dmitry Baryshkov Signed-off-by: Shuai Zhang Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/hci_qca.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/bluetooth/hci_qca.c b/drivers/bluetooth/hci_qca.c index ed280399bf47..34500137df2c 100644 --- a/drivers/bluetooth/hci_qca.c +++ b/drivers/bluetooth/hci_qca.c @@ -1680,8 +1680,8 @@ static void qca_hw_error(struct hci_dev *hdev, u8 code) mod_timer(&qca->tx_idle_timer, jiffies + msecs_to_jiffies(qca->tx_idle_delay)); - /* Controller reset completion time is 50ms */ - msleep(50); + /* Wait for the controller to load the rampatch and NVM. */ + msleep(100); clear_bit(QCA_SSR_TRIGGERED, &qca->flags); clear_bit(QCA_IBS_DISABLED, &qca->flags); From 426a35d7530722ffa2d89d759c39f5157e0e500d Mon Sep 17 00:00:00 2001 From: Jeremy Erazo Date: Wed, 20 May 2026 18:23:31 +0000 Subject: [PATCH 313/411] smb: client: detect short folioq copy in cifs_copy_folioq_to_iter() cifs_copy_folioq_to_iter() copies a requested number of bytes from a folio queue into the destination iterator. Since the encrypted SMB2 READ path was changed to pass the server-declared payload length (data_len) instead of the larger folioq buffer length, the caller can ask for fewer bytes than the folio queue holds. In that case the helper continues walking the remaining folios after data_size has reached zero and calls copy_folio_to_iter() with len = 0, which is unnecessary work. The helper also returns 0 (success) when the folio queue is exhausted before data_size bytes have been copied. The caller has no way to distinguish that from a full copy and the reported transfer count ends up larger than the amount of data placed in the iterator. Add an early exit when data_size reaches zero, and return an error when the folio queue is exhausted before all requested bytes have been copied. Signed-off-by: Jeremy Erazo Reviewed-by: David Howells Signed-off-by: Steve French --- fs/smb/client/smb2ops.c | 18 +++++++++++++++--- 1 file changed, 15 insertions(+), 3 deletions(-) diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 61b60114e4b8..d4875f9532b4 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -4706,9 +4706,15 @@ cifs_copy_folioq_to_iter(struct folio_queue *folioq, size_t data_size, { for (; folioq; folioq = folioq->next) { for (int s = 0; s < folioq_count(folioq); s++) { - struct folio *folio = folioq_folio(folioq, s); - size_t fsize = folio_size(folio); - size_t n, len = umin(fsize - skip, data_size); + struct folio *folio; + size_t fsize, n, len; + + if (data_size == 0) + return 0; + + folio = folioq_folio(folioq, s); + fsize = folio_size(folio); + len = umin(fsize - skip, data_size); n = copy_folio_to_iter(folio, skip, len, iter); if (n != len) { @@ -4721,6 +4727,12 @@ cifs_copy_folioq_to_iter(struct folio_queue *folioq, size_t data_size, } } + if (data_size != 0) { + cifs_dbg(VFS, "%s: short copy, %zu bytes missing\n", + __func__, data_size); + return smb_EIO2(smb_eio_trace_rx_copy_to_iter, 0, data_size); + } + return 0; } From 9d2491197a00acf8c423512078458c2855102b66 Mon Sep 17 00:00:00 2001 From: Steve French Date: Fri, 22 May 2026 18:28:49 -0500 Subject: [PATCH 314/411] smb: client: fix uninitialized variable in smb2_writev_callback compiling with W=2 pointed out that "written may be used uninitialized" Fixes: 20d72b00ca81 ("netfs: Fix the request's work item to not require a ref") Cc: stable@vger.kernel.org Reviewed-by: David Howells Signed-off-by: Steve French --- fs/smb/client/smb2pdu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/smb/client/smb2pdu.c b/fs/smb/client/smb2pdu.c index 3bd300347f16..fbeb2156ddb6 100644 --- a/fs/smb/client/smb2pdu.c +++ b/fs/smb/client/smb2pdu.c @@ -4955,7 +4955,7 @@ smb2_writev_callback(struct TCP_Server_Info *server, struct mid_q_entry *mid) unsigned int rreq_debug_id = wdata->rreq->debug_id; unsigned int subreq_debug_index = wdata->subreq.debug_index; ssize_t result = 0; - size_t written; + size_t written = 0; WARN_ONCE(wdata->server != server, "wdata server %p != mid server %p", From a17dc12bfed8868e6a86f3b45c16065a70641acb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Alexis=20Lothor=C3=A9=20=28eBPF=20Foundation=29?= Date: Wed, 27 May 2026 21:12:31 +0200 Subject: [PATCH 315/411] x86/ftrace: Relocate %rip-relative percpu refs in dynamic trampolines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit With CONFIG_CALL_DEPTH_TRACKING enabled on an x86 retbleed-affected platform (eg: Skylake), with retbleed=stuff, registering a dynamic ftrace trampoline crashes on the first call into the traced function: BUG: unable to handle page fault for address: ffff88817ae18880 #PF: supervisor write access in kernel mode #PF: error_code(0x0002) - not-present page PGD 4b53067 P4D 4b53067 PUD 0 Oops: Oops: 0002 [#1] SMP PTI CPU: 3 UID: 0 PID: 187 Comm: usleep Not tainted 7.0.10 #243 PREEMPT(full) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Arch Linux 1.17.0-2-2 04/01/2014 Code: 24 78 00 00 00 00 48 89 ea 48 89 54 24 20 48 8b b4 24 b8 00 00 00 48 8b bc 24 b0 00 00 00 48 89 bc 24 80 00 00 00 48 83 ef 05 <65> 48 c1 3d 1f a8 b6 02 05 48 8b 15 f6 00 00 00 4c 89 3c 24 4c 89 Call Trace: ? find_held_lock ? exc_page_fault ? lock_release ? __x64_sys_clock_nanosleep ? lockdep_hardirqs_on_prepare ? trace_hardirqs_on __x64_sys_clock_nanosleep do_syscall_64 ? exc_page_fault ? call_depth_return_thunk entry_SYSCALL_64_after_hwframe ... Kernel panic - not syncing: Fatal exception This small reproducer allows to easily trigger the crash: # echo 'p __x64_sys_clock_nanosleep' > /sys/kernel/tracing/kprobe_events # echo 1 > /sys/kernel/tracing/events/kprobes/p___x64_sys_clock_nanosleep_0/enable # usleep 1 Monitoring the crash under GDB points to the exact instruction in charge of incrementing the call depth: sarq $5, %gs:__x86_call_depth(%rip) This instruction matches the one inserted by the ftrace_regs_caller from ftrace_64.S. This emitted code was likely working fine until the introduction of 59bec00ace28 ("x86/percpu: Introduce %rip-relative addressing to PER_CPU_VAR()"): it has made the call depth accounting addressing relative to $rip, instead of being based on an absolute address. As this code exact location depends on where the trampoline lives in memory, the corresponding displacement needs to be adjusted at runtime to actually correctly find the per-cpu __x86_call_depth value, otherwise the targeted address is wrong, leading to the page fault seen above. Fix the %rip-relative displacement of the copied CALL_DEPTH_ACCOUNT instruction (from ftrace_regs_caller) by calling text_poke_apply_relocation(), as it is done for example by the x86 BPF JIT compiler through x86_call_depth_emit_accounting(). This corrects both CALL_DEPTH_ACCOUNT slots, in ftrace_caller and ftrace_regs_caller. [ bp: Massage. ] Fixes: 59bec00ace28 ("x86/percpu: Introduce %rip-relative addressing to PER_CPU_VAR()") Signed-off-by: Alexis Lothoré (eBPF Foundation) Signed-off-by: Borislav Petkov (AMD) Acked-by: Peter Zijlstra (Intel) Acked-by: Steven Rostedt Cc: Link: https://patch.msgid.link/20260527-fix_call_depth_in_trampoline-v1-1-1c1abc8ae310@bootlin.com --- arch/x86/kernel/ftrace.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 0543b57f54ee..17d6edfcb7e0 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -375,6 +375,13 @@ create_trampoline(struct ftrace_ops *ops, unsigned int *tramp_size) goto fail; } + /* + * Generated trampoline may contain rIP-relative addressing which + * displacement needs to be fixed. + */ + text_poke_apply_relocation(trampoline, trampoline, size, + (void *)start_offset, size); + /* * The address of the ftrace_ops that is used for this trampoline * is stored at the end of the trampoline. This will be used to From 8ba68464e4787b6a7ec938826e16124df20fd23d Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Tue, 26 May 2026 21:33:19 +0200 Subject: [PATCH 316/411] bonding: refuse to enslave CAN devices syzbot reported a kernel paging request crash in can_rx_unregister() inside net/can/af_can.c. The crash occurs because a virtual CAN device (vxcan) is being enslaved to a bonding master. During the enslavement process, the bonding driver mutates and modifies the network device states to fit an Ethernet-like aggregation model. However, CAN devices operate on a completely different Layer 2 architecture, relying on the CAN mid-layer private data structure (can_ml_priv) instead of standard Ethernet structures. Since bonding does not initialize or maintain these CAN structures, subsequent operations on the half-enslaved interface (such as closing associated sockets via isotp_release) lead to a null-pointer dereference when accessing the CAN receiver lists. Bonding CAN interfaces is architecturally invalid as CAN lacks MAC addresses, ARP capabilities, and standard Ethernet link-layer mechanisms. While generic loopback devices are blocked globally in net/core/dev.c, virtual CAN devices bypass this check because they do not carry the IFF_LOOPBACK flag, despite acting as local software-loopbacks. Fix this by explicitly blocking network devices of type ARPHRD_CAN from being enslaved at the very beginning of bond_enslave(). This prevents illegal state mutations, eliminates the resulting KASAN crashes, and avoids potential memory leaks from incomplete socket cleanups. As the CAN support has been added a long time after bonding the Fixes-tag points to the introduction of ARPHRD_CAN that would have needed a specific handling in bonding_main.c. Fixes: cd05acfe65ed ("[CAN]: Allocate protocol numbers for PF_CAN") Reported-by: syzbot+8ed98cbd0161632bce95@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=8ed98cbd0161632bce95 Signed-off-by: Oliver Hartkopp Acked-by: Jay Vosburgh Link: https://patch.msgid.link/20260526-bonding-candev-v1-1-ba1df400918a@hartkopp.net Signed-off-by: Jakub Kicinski --- drivers/net/bonding/bond_main.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index af82a3df2c5d..82e779f7916b 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1890,6 +1890,12 @@ int bond_enslave(struct net_device *bond_dev, struct net_device *slave_dev, struct sockaddr_storage ss; int res = 0, i; + if (slave_dev->type == ARPHRD_CAN) { + BOND_NL_ERR(bond_dev, extack, + "CAN devices cannot be enslaved"); + return -EPERM; + } + if (slave_dev->flags & IFF_MASTER && !netif_is_bond_master(slave_dev)) { BOND_NL_ERR(bond_dev, extack, From 5eec4427b89c2fb2beac54920101e55a2f1c0c21 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 26 May 2026 09:48:16 +0300 Subject: [PATCH 317/411] bridge: Fix sleep in atomic context in netlink path Since the introduction of the netlink configuration path for bridge ports in commit 25c71c75ac87 ("bridge: bridge port parameters over netlink"), br_setport() was always called with the bridge lock held around it. Back then this decision made sense: The bridge lock protects the STP state of the bridge and its ports and at that time the function only processed three STP related netlink attributes (cost, priority and state). Nowadays, br_setport() processes a lot more attributes and most of them do not need the bridge lock: * Bridge flags: Only require RTNL. Read locklessly by the data path. Annotations can be added in net-next. * FDB port flushing: Only requires the FDB lock. * Multicast attributes: Only require the multicast lock. * Group forward mask: Only requires RTNL. Read locklessly by the data path. Annotations can be added in net-next. * Backup port and NHID: Only require RTNL. Read locklessly by the data path. This is a problem as the bridge calls dev_set_promiscuity() when certain bridge port flags change and this function can sleep since the commit cited below, resulting in a splat such as [1]. Fix this by reducing the scope of the bridge lock and only take it when processing the three STP related attributes that require it. This is consistent with the multicast attributes where each attribute acquires the multicast lock instead of having one critical section for all relevant attributes. [1] BUG: sleeping function called from invalid context at net/core/dev_addr_lists.c:1262 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 356, name: bridge preempt_count: 201, expected: 0 RCU nest depth: 0, expected: 0 2 locks held by bridge/356: #0: ffffffff919473a0 (rtnl_mutex){+.+.}-{4:4}, at: rtnetlink_rcv_msg (net/core/rtnetlink.c:80 net/core/rtnetlink.c:7002) #1: ffff888115072d58 (&br->lock){+...}-{3:3}, at: br_setlink (./include/linux/spinlock.h:348 net/bridge/br_netlink.c:1117) Preemption disabled at: 0x0 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 Call Trace: dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) __might_resched.cold (kernel/sched/core.c:9163) netif_rx_mode_run (net/core/dev_addr_lists.c:1262) netif_rx_mode_sync (net/core/dev_addr_lists.c:1428) dev_set_promiscuity (net/core/dev_api.c:289) br_manage_promisc (net/bridge/br_if.c:135 net/bridge/br_if.c:172) br_port_flags_change (net/bridge/br_if.c:242 net/bridge/br_if.c:747) br_setport (net/bridge/br_netlink.c:1000) br_setlink (net/bridge/br_netlink.c:1118) rtnl_bridge_setlink (net/core/rtnetlink.c:5572) rtnetlink_rcv_msg (net/core/rtnetlink.c:7005) netlink_rcv_skb (net/netlink/af_netlink.c:2550) netlink_unicast (net/netlink/af_netlink.c:1318 net/netlink/af_netlink.c:1344) netlink_sendmsg (net/netlink/af_netlink.c:1894) __sock_sendmsg (net/socket.c:787 (discriminator 4) net/socket.c:802 (discriminator 4)) ____sys_sendmsg (net/socket.c:2698) ___sys_sendmsg (net/socket.c:2752) __sys_sendmsg (net/socket.c:2784) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) Fixes: 78cd408356fe ("net: add missing instance lock to dev_set_promiscuity") Reviewed-by: Nikolay Aleksandrov Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20260526064818.272516-2-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_netlink.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-) diff --git a/net/bridge/br_netlink.c b/net/bridge/br_netlink.c index c04a4d0889ae..b9591dd755f9 100644 --- a/net/bridge/br_netlink.c +++ b/net/bridge/br_netlink.c @@ -1000,19 +1000,25 @@ static int br_setport(struct net_bridge_port *p, struct nlattr *tb[], br_port_flags_change(p, changed_mask); if (tb[IFLA_BRPORT_COST]) { + spin_lock_bh(&p->br->lock); err = br_stp_set_path_cost(p, nla_get_u32(tb[IFLA_BRPORT_COST])); + spin_unlock_bh(&p->br->lock); if (err) return err; } if (tb[IFLA_BRPORT_PRIORITY]) { + spin_lock_bh(&p->br->lock); err = br_stp_set_port_priority(p, nla_get_u16(tb[IFLA_BRPORT_PRIORITY])); + spin_unlock_bh(&p->br->lock); if (err) return err; } if (tb[IFLA_BRPORT_STATE]) { + spin_lock_bh(&p->br->lock); err = br_set_port_state(p, nla_get_u8(tb[IFLA_BRPORT_STATE])); + spin_unlock_bh(&p->br->lock); if (err) return err; } @@ -1114,9 +1120,7 @@ int br_setlink(struct net_device *dev, struct nlmsghdr *nlh, u16 flags, if (err) return err; - spin_lock_bh(&p->br->lock); err = br_setport(p, tb, extack); - spin_unlock_bh(&p->br->lock); } else { /* Binary compatibility with old RSTP */ if (nla_len(protinfo) < sizeof(u8)) @@ -1203,17 +1207,10 @@ static int br_port_slave_changelink(struct net_device *brdev, struct nlattr *data[], struct netlink_ext_ack *extack) { - struct net_bridge *br = netdev_priv(brdev); - int ret; - if (!data) return 0; - spin_lock_bh(&br->lock); - ret = br_setport(br_port_get_rtnl(dev), data, extack); - spin_unlock_bh(&br->lock); - - return ret; + return br_setport(br_port_get_rtnl(dev), data, extack); } static int br_port_fill_slave_info(struct sk_buff *skb, From 6d34594cc619d0d4b07d5afcad8b5984f3526dcf Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 26 May 2026 09:48:17 +0300 Subject: [PATCH 318/411] bridge: Fix sleep in atomic context in sysfs path Since the start of the git history, brport_store() always acquired the bridge lock. Back then this decision made sense: The bridge lock protects the STP state of the bridge and its ports and at that time the function was only used by two STP related attributes (cost and priority). Nowadays, brport_store() processes a lot more attributes and most of them do not need the bridge lock: * Bridge flags: Only require RTNL. Read locklessly by the data path. Annotations can be added in net-next. * FDB port flushing: Only requires the FDB lock. * Multicast attributes: Only require the multicast lock. * Group forward mask: Only requires RTNL. Read locklessly by the data path. Annotations can be added in net-next. * Backup port: Only requires RTNL. Read locklessly by the data path. This is a problem as the bridge calls dev_set_promiscuity() when certain bridge port flags change and this function can sleep since the commit cited below, resulting in a splat such as [1]. Fix this by reducing the scope of the bridge lock and only take it when processing the two STP related attributes that require it. Remove the now stale comment from br_switchdev_set_port_flag(). The SWITCHDEV_F_DEFER flag can be removed in net-next. [1] BUG: sleeping function called from invalid context at net/core/dev_addr_lists.c:1262 in_atomic(): 1, irqs_disabled(): 0, non_block: 0, pid: 372, name: bash preempt_count: 201, expected: 0 RCU nest depth: 0, expected: 0 5 locks held by bash/372: #0: ffff88810c51c3f0 (sb_writers#7){.+.+}-{0:0}, at: ksys_write (fs/read_write.c:740) #1: ffff888115ce9480 (&of->mutex){+.+.}-{4:4}, at: kernfs_fop_write_iter (fs/kernfs/file.c:343) #2: ffff88810b9fd330 (kn->active#37){.+.+}-{0:0}, at: kernfs_fop_write_iter (fs/kernfs/file.c:80 fs/kernfs/file.c:344) #3: ffffffffa59473a0 (rtnl_mutex){+.+.}-{4:4}, at: brport_store (net/bridge/br_sysfs_if.c:326) #4: ffff8881099d2d58 (&br->lock){+...}-{3:3}, at: brport_store (./include/linux/spinlock.h:348 net/bridge/br_sysfs_if.c:345) Preemption disabled at: 0x0 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 Call Trace: dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) __might_resched.cold (kernel/sched/core.c:9163) netif_rx_mode_run (net/core/dev_addr_lists.c:1262) netif_rx_mode_sync (net/core/dev_addr_lists.c:1428) dev_set_promiscuity (net/core/dev_api.c:289) br_manage_promisc (net/bridge/br_if.c:135 net/bridge/br_if.c:172) br_port_flags_change (net/bridge/br_if.c:242 net/bridge/br_if.c:747) store_learning (net/bridge/br_sysfs_if.c:79 net/bridge/br_sysfs_if.c:235) brport_store (net/bridge/br_sysfs_if.c:346) kernfs_fop_write_iter (fs/kernfs/file.c:352) new_sync_write (fs/read_write.c:595) vfs_write (fs/read_write.c:688) ksys_write (fs/read_write.c:740) do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94) entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121) Fixes: 78cd408356fe ("net: add missing instance lock to dev_set_promiscuity") Reviewed-by: Nikolay Aleksandrov Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20260526064818.272516-3-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/bridge/br_switchdev.c | 1 - net/bridge/br_sysfs_if.c | 30 ++++++++++++++++++++++-------- 2 files changed, 22 insertions(+), 9 deletions(-) diff --git a/net/bridge/br_switchdev.c b/net/bridge/br_switchdev.c index 18b558a931ad..ee3ad9dfbab9 100644 --- a/net/bridge/br_switchdev.c +++ b/net/bridge/br_switchdev.c @@ -99,7 +99,6 @@ int br_switchdev_set_port_flag(struct net_bridge_port *p, attr.u.brport_flags.val = flags; attr.u.brport_flags.mask = mask; - /* We run from atomic context here */ err = call_switchdev_notifiers(SWITCHDEV_PORT_ATTR_SET, p->dev, &info.info, extack); err = notifier_to_errno(err); diff --git a/net/bridge/br_sysfs_if.c b/net/bridge/br_sysfs_if.c index 1f57c36a7fc0..d6df81fa0d13 100644 --- a/net/bridge/br_sysfs_if.c +++ b/net/bridge/br_sysfs_if.c @@ -86,16 +86,34 @@ static ssize_t show_path_cost(struct net_bridge_port *p, char *buf) return sysfs_emit(buf, "%d\n", p->path_cost); } -static BRPORT_ATTR(path_cost, 0644, - show_path_cost, br_stp_set_path_cost); +static int store_path_cost(struct net_bridge_port *p, unsigned long v) +{ + int ret; + + spin_lock_bh(&p->br->lock); + ret = br_stp_set_path_cost(p, v); + spin_unlock_bh(&p->br->lock); + return ret; +} + +static BRPORT_ATTR(path_cost, 0644, show_path_cost, store_path_cost); static ssize_t show_priority(struct net_bridge_port *p, char *buf) { return sysfs_emit(buf, "%d\n", p->priority); } -static BRPORT_ATTR(priority, 0644, - show_priority, br_stp_set_port_priority); +static int store_priority(struct net_bridge_port *p, unsigned long v) +{ + int ret; + + spin_lock_bh(&p->br->lock); + ret = br_stp_set_port_priority(p, v); + spin_unlock_bh(&p->br->lock); + return ret; +} + +static BRPORT_ATTR(priority, 0644, show_priority, store_priority); static ssize_t show_designated_root(struct net_bridge_port *p, char *buf) { @@ -334,17 +352,13 @@ static ssize_t brport_store(struct kobject *kobj, ret = -ENOMEM; goto out_unlock; } - spin_lock_bh(&p->br->lock); ret = brport_attr->store_raw(p, buf_copy); - spin_unlock_bh(&p->br->lock); kfree(buf_copy); } else if (brport_attr->store) { val = simple_strtoul(buf, &endp, 0); if (endp == buf) goto out_unlock; - spin_lock_bh(&p->br->lock); ret = brport_attr->store(p, val); - spin_unlock_bh(&p->br->lock); } if (!ret) { From 147f3b1f23cbd74f1022cc5689570a06f6bc47c8 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Tue, 26 May 2026 09:48:18 +0300 Subject: [PATCH 319/411] selftests: rtnetlink: Add bridge promiscuity tests Add two test cases that always pass, but trigger sleeping in atomic context BUGs without "bridge: Fix sleep in atomic context in netlink path" and "bridge: Fix sleep in atomic context in sysfs path". Reviewed-by: Nikolay Aleksandrov Signed-off-by: Ido Schimmel Link: https://patch.msgid.link/20260526064818.272516-4-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/rtnetlink.sh | 63 ++++++++++++++++++++++++ 1 file changed, 63 insertions(+) diff --git a/tools/testing/selftests/net/rtnetlink.sh b/tools/testing/selftests/net/rtnetlink.sh index c499953d4885..ace3a99023ed 100755 --- a/tools/testing/selftests/net/rtnetlink.sh +++ b/tools/testing/selftests/net/rtnetlink.sh @@ -24,6 +24,8 @@ ALL_TESTS=" kci_test_macsec kci_test_macsec_vlan kci_test_team_bridge_macvlan + kci_test_bridge_promisc_netlink + kci_test_bridge_promisc_sysfs kci_test_ipsec kci_test_ipsec_offload kci_test_fdb_get @@ -61,6 +63,14 @@ check_fail() fi } +sysfs_write() +{ + local val="$1" + local path="$2" + + echo "$val" > "$path" +} + run_cmd_common() { local cmd="$*" @@ -680,6 +690,59 @@ kci_test_team_bridge_macvlan() end_test "PASS: team_bridge_macvlan" } +# Test that changing bridge port flags via the netlink path does not sleep with +# the bridge spin lock held. +kci_test_bridge_promisc_netlink() +{ + local dummy="test_dummy1" + local bridge="test_br1" + local team="test_team1" + local ret=0 + + run_cmd ip link add $team up type team + run_cmd ip link add $bridge up type bridge vlan_filtering 1 + run_cmd ip link add $dummy up type dummy + run_cmd ip link set $dummy master $bridge + run_cmd ip link set $team master $bridge + + # This causes the bridge driver to sync all the static FDB entries to + # the team device (which supports unicast filtering) and remove it from + # promiscuous mode. The call to dev_set_promiscuity() can sleep due to + # Rx mode inlining, which is a problem if the bridge spin lock is held. + run_cmd bridge link set dev $dummy flood off learning off + + run_cmd ip link del $dummy + run_cmd ip link del $bridge + run_cmd ip link del $team + + end_test "PASS: bridge_promisc_netlink" +} + +# Same as kci_test_bridge_promisc_netlink(), but the flags are changed via the +# sysfs path. +kci_test_bridge_promisc_sysfs() +{ + local dummy="test_dummy1" + local bridge="test_br1" + local team="test_team1" + local ret=0 + + run_cmd ip link add $team up type team + run_cmd ip link add $bridge up type bridge vlan_filtering 1 + run_cmd ip link add $dummy up type dummy + run_cmd ip link set $dummy master $bridge + run_cmd ip link set $team master $bridge + + run_cmd sysfs_write 0 /sys/class/net/$dummy/brport/unicast_flood + run_cmd sysfs_write 0 /sys/class/net/$dummy/brport/learning + + run_cmd ip link del $dummy + run_cmd ip link del $bridge + run_cmd ip link del $team + + end_test "PASS: bridge_promisc_sysfs" +} + #------------------------------------------------------------------- # Example commands # ip x s add proto esp src 14.0.0.52 dst 14.0.0.70 \ From 7281b096b072f6c6e30420e3467d738f2e4c4b57 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:24 -0700 Subject: [PATCH 320/411] ethtool: coalesce: cap profile updates at NET_DIM_PARAMS_NUM_PROFILES ethnl_update_profile() walks the ETHTOOL_A_PROFILE_IRQ_MODERATION nest list with an index 'i' and writes new_profile[i++] without bounding i. The destination is kmemdup()'d at NET_DIM_PARAMS_NUM_PROFILES entries (5), but the Netlink nest count is entirely user-controlled. Netlink policies do not have support for constraining the number of nested entries (or number of multi-attr entries). Fixes: f750dfe825b9 ("ethtool: provide customized dim profile management") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-2-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/coalesce.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/ethtool/coalesce.c b/net/ethtool/coalesce.c index 1e2c5c7048a8..e73fc3e5a02b 100644 --- a/net/ethtool/coalesce.c +++ b/net/ethtool/coalesce.c @@ -472,6 +472,12 @@ static int ethnl_update_profile(struct net_device *dev, nla_for_each_nested_type(nest, ETHTOOL_A_PROFILE_IRQ_MODERATION, nests, rem) { + if (i >= NET_DIM_PARAMS_NUM_PROFILES) { + NL_SET_BAD_ATTR(extack, nest); + ret = -E2BIG; + goto err_out; + } + ret = nla_parse_nested(tb, len_irq_moder - 1, nest, coalesce_irq_moderation_policy, extack); From a888bbd43940cada72f7686337741ce86d1cf869 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:25 -0700 Subject: [PATCH 321/411] ethtool: tsconfig: fix reply error handling A couple of trivial bugs in error handling in tsconfig_send_reply(). If we failed to allocate rskb we need to set the error. If we did allocate it but failed to send it - we need to remember to free it. Fixes: 6e9e2eed4f39 ("net: ethtool: Add support for tsconfig command to get/set hwtstamp config") Reviewed-by: Vadim Fedorenko Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20260526153533.2779187-3-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsconfig.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/net/ethtool/tsconfig.c b/net/ethtool/tsconfig.c index e4f518e49d4c..e9db4ee2299d 100644 --- a/net/ethtool/tsconfig.c +++ b/net/ethtool/tsconfig.c @@ -224,16 +224,21 @@ static int tsconfig_send_reply(struct net_device *dev, struct genl_info *info) reply_len = ret + ethnl_reply_header_size(); rskb = ethnl_reply_init(reply_len, dev, ETHTOOL_MSG_TSCONFIG_SET_REPLY, ETHTOOL_A_TSCONFIG_HEADER, info, &reply_payload); - if (!rskb) + if (!rskb) { + ret = -ENOMEM; goto err_cleanup; + } ret = tsconfig_fill_reply(rskb, &req_info->base, &reply_data->base); if (ret < 0) - goto err_cleanup; + goto err_free_msg; genlmsg_end(rskb, reply_payload); ret = genlmsg_reply(rskb, info); + rskb = NULL; +err_free_msg: + nlmsg_free(rskb); err_cleanup: kfree(reply_data); kfree(req_info); From 596c51ed9e125b12c4d85b4530dfd4c7847634b7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:26 -0700 Subject: [PATCH 322/411] ethtool: linkstate: fix unbalanced ethnl_ops_complete() on PHY lookup error linkstate_prepare_data() calls ethnl_req_get_phydev() before ethnl_ops_begin(), but routes its error path through "goto out" which calls ethnl_ops_complete(). Fixes: fe55b1d401c6 ("ethtool: linkstate: migrate linkstate functions to support multi-PHY setups") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-4-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/linkstate.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/ethtool/linkstate.c b/net/ethtool/linkstate.c index 8a5985fd7712..24569e92942c 100644 --- a/net/ethtool/linkstate.c +++ b/net/ethtool/linkstate.c @@ -106,10 +106,8 @@ static int linkstate_prepare_data(const struct ethnl_req_info *req_base, phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_LINKSTATE_HEADER, info->extack); - if (IS_ERR(phydev)) { - ret = PTR_ERR(phydev); - goto out; - } + if (IS_ERR(phydev)) + return PTR_ERR(phydev); ret = ethnl_ops_begin(dev); if (ret < 0) From ab5bf428fb6bd361163c7247b92750d1d24ca2ed Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:27 -0700 Subject: [PATCH 323/411] ethtool: pse-pd: fix missing ethnl_ops_complete() pse_prepare_data() is missing ethnl_ops_complete() if ethnl_req_get_phydev() returned an error. Move getting phydev up so that we don't have to worry about this (similar order to linkstate_prepare_data()). Note that phydev may still be NULL (this is checked in pse_get_pse_attributes()), the goal isn't really to avoid the _begin() / _complete() calls, only to simplify the error handling. While at it propagate the original error. Why this code overrides the error with -ENODEV but !phydev generates -EOPNOTSUPP is unclear to me... Fixes: 31748765bed3 ("net: ethtool: pse-pd: Target the command to the requested PHY") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-5-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/pse-pd.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/net/ethtool/pse-pd.c b/net/ethtool/pse-pd.c index 2eb9bdc2dcb9..757c9e0cc856 100644 --- a/net/ethtool/pse-pd.c +++ b/net/ethtool/pse-pd.c @@ -62,14 +62,14 @@ static int pse_prepare_data(const struct ethnl_req_info *req_base, struct phy_device *phydev; int ret; - ret = ethnl_ops_begin(dev); - if (ret < 0) - return ret; - phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_PSE_HEADER, info->extack); if (IS_ERR(phydev)) - return -ENODEV; + return PTR_ERR(phydev); + + ret = ethnl_ops_begin(dev); + if (ret < 0) + return ret; ret = pse_get_pse_attributes(phydev, info->extack, data); From 6386bd772de64e6760306eb91c7e86163af6c22f Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:28 -0700 Subject: [PATCH 324/411] ethtool: tsconfig: fix missing ethnl_ops_complete() tsconfig_prepare_data() calls ethnl_ops_begin(), we need to call ethnl_ops_complete() before returning the error. Fixes: 6e9e2eed4f39 ("net: ethtool: Add support for tsconfig command to get/set hwtstamp config") Reviewed-by: Vadim Fedorenko Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20260526153533.2779187-6-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsconfig.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/ethtool/tsconfig.c b/net/ethtool/tsconfig.c index e9db4ee2299d..fc4f93cfa459 100644 --- a/net/ethtool/tsconfig.c +++ b/net/ethtool/tsconfig.c @@ -69,8 +69,10 @@ static int tsconfig_prepare_data(const struct ethnl_req_info *req_base, if (ret) goto out; - if (ts_info.phc_index == -1) - return -ENODEV; + if (ts_info.phc_index == -1) { + ret = -ENODEV; + goto out; + } data->hwprov_desc.index = ts_info.phc_index; data->hwprov_desc.qualifier = ts_info.phc_qualifier; From 1de405699c62c3a9544bcdcfb9eff8a01cfc7582 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:29 -0700 Subject: [PATCH 325/411] ethtool: tsinfo: fix uninitialized stats on the by-PHC path tsinfo_prepare_data() has two code paths: a "by-PHC" path for user-specified hardware timestamping providers, and the old path. Commit 89e281ebff72 ("ethtool: init tsinfo stats if requested") added ethtool_stats_init() to mark stat slots as ETHTOOL_STAT_NOT_SET before the driver callback populates them, but placed the call inside the old-path block. When commit b9e3f7dc9ed9 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") added the by-PHC early return, it landed above the stats initialization. On that path the stats array retains the zero-fill from ethnl_init_reply_data()'s zalloc. This leads to the reply including a stats nest with four zero-valued attributes that should have been absent. Reject GET requests for stats with HWTSTAMP_PROVIDER or dump. Fixes: b9e3f7dc9ed9 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-7-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsinfo.c | 13 +++++++++++++ 1 file changed, 13 insertions(+) diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c index a865f0fdd26b..f54fe6b662b2 100644 --- a/net/ethtool/tsinfo.c +++ b/net/ethtool/tsinfo.c @@ -83,6 +83,11 @@ tsinfo_parse_request(struct ethnl_req_info *req_base, if (!tb[ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER]) return 0; + if (req_base->flags & ETHTOOL_FLAG_STATS) { + NL_SET_ERR_MSG(extack, "can't query statistics for a provider"); + return -EOPNOTSUPP; + } + return ts_parse_hwtst_provider(tb[ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER], &req->hwprov_desc, extack, &mod); } @@ -523,6 +528,12 @@ int ethnl_tsinfo_start(struct netlink_callback *cb) if (ret < 0) goto free_reply_data; + if (req_info->base.flags & ETHTOOL_FLAG_STATS) { + NL_SET_ERR_MSG(cb->extack, "stats not supported in dump"); + ret = -EOPNOTSUPP; + goto err_dev_put; + } + ctx->req_info = req_info; ctx->reply_data = reply_data; ctx->pos_ifindex = 0; @@ -532,6 +543,8 @@ int ethnl_tsinfo_start(struct netlink_callback *cb) return 0; +err_dev_put: + ethnl_parse_header_dev_put(&req_info->base); free_reply_data: kfree(reply_data); free_req_info: From c3fc9976f686f9a95baf87db9d387f218fd65394 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:30 -0700 Subject: [PATCH 326/411] ethtool: tsinfo: don't pass ERR_PTR to genlmsg_cancel on prepare failure The goto err label leads to: genlmsg_cancel(skb, ehdr); return ret; If ethnl_tsinfo_prepare_dump() failed, it has not started a genlmsg. There's nothing to cancel, and passing an error pointer to genlmsg_cancel() would cause a crash. Fixes: b9e3f7dc9ed9 ("net: ethtool: tsinfo: Enhance tsinfo to support several hwtstamp by net topology") Reviewed-by: Maxime Chevallier Reviewed-by: Kory Maincent Link: https://patch.msgid.link/20260526153533.2779187-8-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/tsinfo.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c index f54fe6b662b2..14bf01e3b55c 100644 --- a/net/ethtool/tsinfo.c +++ b/net/ethtool/tsinfo.c @@ -407,10 +407,8 @@ static int ethnl_tsinfo_dump_one_netdev(struct sk_buff *skb, continue; ehdr = ethnl_tsinfo_prepare_dump(skb, dev, reply_data, cb); - if (IS_ERR(ehdr)) { - ret = PTR_ERR(ehdr); - goto err; - } + if (IS_ERR(ehdr)) + return PTR_ERR(ehdr); reply_data->ts_info.phc_qualifier = ctx->pos_phcqualifier; ret = ops->get_ts_info(dev, &reply_data->ts_info); From a8d8bef6b45bf7cc0b1f6110c5cd8d0160a9bad7 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:31 -0700 Subject: [PATCH 327/411] ethtool: strset: fix header attribute index in ethnl_req_get_phydev() strset_prepare_data() passes ETHTOOL_A_HEADER_FLAGS (3) as the header attribute to ethnl_req_get_phydev(). This is incorrect, in the main attr space 3 is ETHTOOL_A_STRSET_COUNTS_ONLY, not the request header attr. The correct constant is ETHTOOL_A_STRSET_HEADER (1). ethnl_req_get_phydev() only uses this value for the extack, so this is not a "functionally visible"(?) bug. Fixes: e96c93aa4be9 ("net: ethtool: strset: Allow querying phy stats by index") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-9-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/strset.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ethtool/strset.c b/net/ethtool/strset.c index bb1e829ba099..94c4718d31ae 100644 --- a/net/ethtool/strset.c +++ b/net/ethtool/strset.c @@ -311,7 +311,7 @@ static int strset_prepare_data(const struct ethnl_req_info *req_base, return 0; } - phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_HEADER_FLAGS, + phydev = ethnl_req_get_phydev(req_base, tb, ETHTOOL_A_STRSET_HEADER, info->extack); /* phydev can be NULL, check for errors only */ From 2376586f85f972fefe701f095bb37dcfe7405d21 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:32 -0700 Subject: [PATCH 328/411] ethtool: eeprom: add missing ethnl_ops_begin() / _complete() during fallback All ethtool driver op calls should be sandwiched between ethnl_ops_begin() / ethnl_ops_complete(). In Netlink eeprom code, if the paged access failed we fall back to old API, but we first call _complete() and the fallback never does its own ethnl_ops_begin(). Move the fallback into the _begin() / _complete() section. Fixes: 96d971e307cc ("ethtool: Add fallback to get_module_eeprom from netlink command") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-10-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/eeprom.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/net/ethtool/eeprom.c b/net/ethtool/eeprom.c index a557e3996c85..836316df3092 100644 --- a/net/ethtool/eeprom.c +++ b/net/ethtool/eeprom.c @@ -141,12 +141,11 @@ static int eeprom_prepare_data(const struct ethnl_req_info *req_base, return 0; err_ops: + if (ret == -EOPNOTSUPP) + ret = eeprom_fallback(request, reply); ethnl_ops_complete(dev); err_free: kfree(page_data.data); - - if (ret == -EOPNOTSUPP) - return eeprom_fallback(request, reply); return ret; } From 67cfdd9210b99f260b3e0afeb9525e0acc7be31e Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 26 May 2026 08:35:33 -0700 Subject: [PATCH 329/411] ethtool: eeprom: add more safeties to EEPROM Netlink fallback The Netlink fallback path for reading module EEPROM (fallback_set_params()) validates that offset < eeprom_len, but does not check that offset + length stays within eeprom_len. The ioctl equivalent (ethtool_get_any_eeprom() in ioctl.c) has always enforced both bounds: if (eeprom.offset + eeprom.len > total_len) return -EINVAL; This could lead to surprises in both drivers and device FW. Add the missing offset + length validation to fallback_set_params(), mirroring the ioctl. Similarly - ethtool core in general, and ethtool_get_any_eeprom() in particular tries to zero-init all buffers passed to the drivers to avoid any extra work of zeroing things out. eeprom_fallback() uses a plain kmalloc(), change it to zalloc. Fixes: 96d971e307cc ("ethtool: Add fallback to get_module_eeprom from netlink command") Reviewed-by: Maxime Chevallier Link: https://patch.msgid.link/20260526153533.2779187-11-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/ethtool/eeprom.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/ethtool/eeprom.c b/net/ethtool/eeprom.c index 836316df3092..0b8cfeddb014 100644 --- a/net/ethtool/eeprom.c +++ b/net/ethtool/eeprom.c @@ -44,6 +44,9 @@ static int fallback_set_params(struct eeprom_req_info *request, if (offset >= modinfo->eeprom_len) return -EINVAL; + if (length > modinfo->eeprom_len - offset) + return -EINVAL; + eeprom->cmd = ETHTOOL_GMODULEEEPROM; eeprom->len = length; eeprom->offset = offset; @@ -69,7 +72,7 @@ static int eeprom_fallback(struct eeprom_req_info *request, if (err < 0) return err; - data = kmalloc(eeprom.len, GFP_KERNEL); + data = kzalloc(eeprom.len, GFP_KERNEL); if (!data) return -ENOMEM; err = ethtool_get_module_eeprom_call(dev, &eeprom, data); From 9d5e7a46a9f6d8f503b41bfefef70659845f1679 Mon Sep 17 00:00:00 2001 From: Rahul Chandelkar Date: Mon, 25 May 2026 21:10:31 +0530 Subject: [PATCH 330/411] ipv6: rpl: fix hdrlen overflow in ipv6_rpl_srh_decompress() ipv6_rpl_srh_decompress() computes: outhdr->hdrlen = (((n + 1) * sizeof(struct in6_addr)) >> 3); hdrlen is __u8. For n >= 127 the result exceeds 255 and silently truncates. With n=127 (cmpri=15, cmpre=15, pad=0, hdrlen=16): (128 * 16) >> 3 = 256, truncated to 0 as __u8 The caller in ipv6_rpl_srh_rcv() then places the compressed header at buf + ((ohdr->hdrlen + 1) << 3). With hdrlen=0 this is buf + 8, but the decompressed region occupies buf[0..2055] (8-byte header plus 128 full addresses). The compressed header overlaps the decompressed data, and ipv6_rpl_srh_compress() writes into this overlap, corrupting the routing header of the forwarded packet. The existing guard at exthdrs.c:546 checks (n + 1) > 255, which prevents n+1 from overflowing unsigned char (the segments_left field), but does not prevent the computed hdrlen from overflowing __u8. n=127 passes because 128 <= 255, yet hdrlen=256 does not fit. Tighten the bound to (n + 1) > 127. This caps n at 126, giving hdrlen = (127 * 16) >> 3 = 254, which fits in __u8. The compressed header then lands at buf + ((254 + 1) << 3) = buf + 2040, exactly past the decompressed region (buf[0..2039]). No overlap. 127 segments is well beyond any realistic RPL deployment. Fixes: 8610c7c6e3bd ("net: ipv6: add support for rpl sr exthdr") Signed-off-by: Rahul Chandelkar Link: https://patch.msgid.link/20260525154031.2290876-1-rc@rexion.ai Signed-off-by: Jakub Kicinski --- net/ipv6/exthdrs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index aca2a2abd2df..43f46ef9c53b 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -548,7 +548,7 @@ static int ipv6_rpl_srh_rcv(struct sk_buff *skb) * unsigned char which is segments_left field. Should not be * higher than that. */ - if (r || (n + 1) > 255) { + if (r || (n + 1) > 127) { kfree_skb(skb); return -1; } From 4c9ad387aa2d6785299722e54224d34764edaeb3 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 13 May 2026 16:53:54 +0200 Subject: [PATCH 331/411] iommu, debugobjects: avoid gcc-16.1 section mismatch warnings gcc-16 has gained some more advanced inter-procedual optimization techniques that enable it to inline the dummy_tlb_add_page() and dummy_tlb_flush() function pointers into a specialized version of __arm_v7s_unmap: WARNING: modpost: vmlinux: section mismatch in reference: __arm_v7s_unmap+0x2cc (section: .text) -> dummy_tlb_add_page (section: .init.text) ERROR: modpost: Section mismatches detected. >From what I can tell, the transformation is correct, as this is only called when __arm_v7s_unmap() is called from arm_v7s_do_selftests(), which is also __init. Since __arm_v7s_unmap() however is not __init, gcc cannot inline the inner function calls directly. In debug_objects_selftest(), the same thing happens. Both the caller and the leaf function are __init, but the IPA pulls it into a non-init one: WARNING: modpost: vmlinux: section mismatch in reference: lookup_object_or_alloc+0x7c (section: .text.lookup_object_or_alloc) -> is_static_object (section: .init.text) Marking the affected functions as not "__init" would reliably avoid this issue but is not a good solution because it removes an otherwise correct annotation. I tried marking the functions as 'noinline', but that ended up not covering all the affected configurations. With some more experimenting, I found that marking these functions as __attribute__((noipa)) is both logical and reliable. In order to keep the syntax readable, add a custom macro for this in include/linux/compiler_attributes.h next to other related macros and use it to annotate both files. Link: https://lore.kernel.org/all/abRB6g-48ZX6Yl2r@willie-the-truck/ Cc: Will Deacon Cc: Thomas Gleixner Cc: Andrew Morton Cc: Miguel Ojeda Cc: linux-kbuild@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Arnd Bergmann Acked-by: Will Deacon Acked-by: Thomas Gleixner Acked-by: Miguel Ojeda Signed-off-by: Joerg Roedel --- drivers/iommu/io-pgtable-arm-v7s.c | 18 ++++++++++++------ include/linux/compiler_attributes.h | 11 +++++++++++ lib/debugobjects.c | 2 +- 3 files changed, 24 insertions(+), 7 deletions(-) diff --git a/drivers/iommu/io-pgtable-arm-v7s.c b/drivers/iommu/io-pgtable-arm-v7s.c index 40e33257d3c2..1dbef8c55007 100644 --- a/drivers/iommu/io-pgtable-arm-v7s.c +++ b/drivers/iommu/io-pgtable-arm-v7s.c @@ -777,21 +777,27 @@ struct io_pgtable_init_fns io_pgtable_arm_v7s_init_fns = { static struct io_pgtable_cfg *cfg_cookie __initdata; -static void __init dummy_tlb_flush_all(void *cookie) +/* + * __noipa prevents gcc from turning indirect iommu_flush_ops calls + * into direct calls from a specialized __arm_v7s_unmap() that triggers + * a build time section mismatch assertion. + */ +static __noipa void __init dummy_tlb_flush_all(void *cookie) { WARN_ON(cookie != cfg_cookie); } -static void __init dummy_tlb_flush(unsigned long iova, size_t size, - size_t granule, void *cookie) +static __noipa void __init dummy_tlb_flush(unsigned long iova, size_t size, + size_t granule, void *cookie) { WARN_ON(cookie != cfg_cookie); WARN_ON(!(size & cfg_cookie->pgsize_bitmap)); } -static void __init dummy_tlb_add_page(struct iommu_iotlb_gather *gather, - unsigned long iova, size_t granule, - void *cookie) +static __noipa void __init dummy_tlb_add_page(struct iommu_iotlb_gather *gather, + unsigned long iova, + size_t granule, + void *cookie) { dummy_tlb_flush(iova, granule, granule, cookie); } diff --git a/include/linux/compiler_attributes.h b/include/linux/compiler_attributes.h index c16d4199bf92..836a50f5917a 100644 --- a/include/linux/compiler_attributes.h +++ b/include/linux/compiler_attributes.h @@ -396,6 +396,17 @@ # define __disable_sanitizer_instrumentation #endif +/* + * Optional: not supported by clang + * + * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Attributes.html#index-noipa + */ +#if __has_attribute(noipa) +# define __noipa __attribute__((noipa)) +#else +# define __noipa +#endif + /* * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Function-Attributes.html#index-weak-function-attribute * gcc: https://gcc.gnu.org/onlinedocs/gcc/Common-Variable-Attributes.html#index-weak-variable-attribute diff --git a/lib/debugobjects.c b/lib/debugobjects.c index 12e2e42e6a31..c93b7ca3e1ab 100644 --- a/lib/debugobjects.c +++ b/lib/debugobjects.c @@ -1212,7 +1212,7 @@ struct self_test { static __initconst const struct debug_obj_descr descr_type_test; -static bool __init is_static_object(void *addr) +static __noipa bool __init is_static_object(void *addr) { struct self_test *obj = addr; From 8aeb879baf12fe64889f019da9a4f8347c604e91 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 26 May 2026 11:06:31 +0200 Subject: [PATCH 332/411] x86/kvm/vmx: Fix x86_64 CFI build It was missed that idt_do_interrupt_irqoff() gets compiled on x84_64; this is a problem for CFI builds because it includes an unadorned indirect call. It is however completely dead code. Rework things to not emit this function at all. Fixes: 0701c9e17bd9 ("x86/kvm/vmx: Move IRQ/NMI dispatch from KVM into x86 core") Reported-by: Nathan Chancellor Reported-by: Calvin Owens Signed-off-by: Peter Zijlstra (Intel) Tested-by: Nathan Chancellor Link: https://patch.msgid.link/20260526090631.GA4149641@noisy.programming.kicks-ass.net --- arch/x86/entry/common.c | 2 +- arch/x86/entry/entry.S | 2 ++ arch/x86/kernel/idt.c | 12 ++---------- 3 files changed, 5 insertions(+), 11 deletions(-) diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c index 06c7c6ebd6f9..14cd43d4da6c 100644 --- a/arch/x86/entry/common.c +++ b/arch/x86/entry/common.c @@ -55,7 +55,7 @@ noinstr void x86_entry_from_kvm(unsigned int event_type, unsigned int vector) * The FRED NMI context is significantly different and will not work * right (specifically FRED fixed the NMI recursion issue). */ - idt_entry_from_kvm(vector); + idt_do_nmi_irqoff(); } EXPORT_SYMBOL_FOR_KVM(x86_entry_from_kvm); #endif diff --git a/arch/x86/entry/entry.S b/arch/x86/entry/entry.S index a56e043b266d..2bc217bb5475 100644 --- a/arch/x86/entry/entry.S +++ b/arch/x86/entry/entry.S @@ -109,11 +109,13 @@ EXPORT_SYMBOL(__ref_stack_chk_guard); RET .endm +#ifndef CONFIG_X86_64 .pushsection .text, "ax" SYM_FUNC_START(idt_do_interrupt_irqoff) IDT_DO_EVENT_IRQOFF CALL_NOSPEC _ASM_ARG1 SYM_FUNC_END(idt_do_interrupt_irqoff) .popsection +#endif .pushsection .noinstr.text, "ax" SYM_FUNC_START(idt_do_nmi_irqoff) diff --git a/arch/x86/kernel/idt.c b/arch/x86/kernel/idt.c index 7bcf1decc034..90a22e24a9eb 100644 --- a/arch/x86/kernel/idt.c +++ b/arch/x86/kernel/idt.c @@ -268,18 +268,10 @@ void __init idt_setup_early_pf(void) } #endif -#if IS_ENABLED(CONFIG_KVM_INTEL) -noinstr void idt_entry_from_kvm(unsigned int vector) +#if IS_ENABLED(CONFIG_KVM_INTEL) && !defined(CONFIG_X86_64) +void idt_entry_from_kvm(unsigned int vector) { - if (vector == NMI_VECTOR) - return idt_do_nmi_irqoff(); - - /* - * Only the NMI path requires noinstr. - */ - instrumentation_begin(); idt_do_interrupt_irqoff(gate_offset(idt_table + vector)); - instrumentation_end(); } #endif From 98b34f3e8c3492cfc89ff943c9d92b4d52863d1d Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:48 -0400 Subject: [PATCH 333/411] net: Introduce skb tc depth field to track packet loops Add a 2-bit per-skb tc depth field to track packet loops across the stack. The previous per-CPU loop counters like MIRRED_NEST_LIMIT assume a single call stack and lose state in two cases: 1) When a packet is queued and reprocessed later (e.g., egress->ingress via backlog), the per-cpu state is gone by the time it is dequeued. 2) With XPS/RPS a packet may arrive on one CPU and be processed on another. A per-skb field solves both by travelling with the packet itself. The field fits in existing padding, using 2 bits that were previously a hole: pahole before(-) and after (+) diff looks like: __u8 slow_gro:1; /* 132: 3 1 */ __u8 csum_not_inet:1; /* 132: 4 1 */ __u8 unreadable:1; /* 132: 5 1 */ + __u8 tc_depth:2; /* 132: 6 1 */ - /* XXX 2 bits hole, try to pack */ /* XXX 1 byte hole, try to pack */ __u16 tc_index; /* 134 2 */ There used to be a ttl field which was removed as part of tc_verd in commit aec745e2c520 ("net-tc: remove unused tc_verd fields"). It was already unused by that time, due to remove earlier in commit c19ae86a510c ("tc: remove unused redirect ttl"). The first user of this field is netem, which increments tc_depth on duplicated packets before re-enqueueing them at the root qdisc. On re-entry, netem skips duplication for any skb with tc_depth already set, bounding recursion to a single level regardless of tree topology. The other user is mirred which increments it on each pass and limits to depth to MIRRED_DEFER_LIMIT (3). The new field was called ttl in earlier versions of this patch but renamed to tc_depth to avoid confusion with IP ttl. Note (looking at you Sashiko! Dont ignore me and continue bringing this up): 1. Since both mirred and netem utilize the same 2-bit tc_depth field it is possible when netem and mirred are used together that netem qdisc to skip the duplication step. This is a known trade-off, as a 2-bit field cannot independently track both features' recursion depths and it is not considered sane to have a setup that addresses both features on at the same time. 2. skb_scrub_packet does not clear tc_depth. This means a packet's loop history is preserved even across namespaces. While this might be restrictive for some topologies, it is also design intent to provide robustness against loops across namespaces. Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-2-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- include/linux/skbuff.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index 2bcf78a4de7b..3f06254ab1b7 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -821,6 +821,7 @@ enum skb_tstamp_type { * @_sk_redir: socket redirection information for skmsg * @_nfct: Associated connection, if any (with nfctinfo bits) * @skb_iif: ifindex of device we arrived on + * @tc_depth: counter for packet duplication * @tc_index: Traffic control index * @hash: the packet hash * @queue_mapping: Queue mapping for multiqueue devices @@ -1030,6 +1031,7 @@ struct sk_buff { __u8 csum_not_inet:1; #endif __u8 unreadable:1; + __u8 tc_depth:2; #if defined(CONFIG_NET_SCHED) || defined(CONFIG_NET_XGRESS) __u16 tc_index; /* traffic control index */ #endif From eda0b7f203bb166c98d1418b204135bd566ac83b Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:49 -0400 Subject: [PATCH 334/411] net/sched: Revert "net/sched: Restrict conditions for adding duplicating netems to qdisc tree" This reverts commit ec8e0e3d7adef940cdf9475e2352c0680189d14e. The original patch rejects any tree containing two netems when either has duplication set, even when they sit on unrelated classes of the same classful parent. That broke configurations that have worked since netem was introduced. The re-entrancy problem the original commit was trying to solve is handled by later patch using tc_depth flag. Doing this revert will (re)expose the original bug with multiple netem duplication. When this patch is backported make sure and get the full series. Fixes: ec8e0e3d7ade ("net/sched: Restrict conditions for adding duplicating netems to qdisc tree") Reported-by: Ji-Soo Chung Reported-by: Gerlinde Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220774 Reported-by: zyc zyc Closes: https://lore.kernel.org/all/19adda5a1e2.12410b78222774.9191120410578703463@zohomail.cn/ Reported-by: Manas Ghandat Closes: https://lore.kernel.org/netdev/f69b2c8f-8325-4c2e-a011-6dbc089f30e4@gmail.com/ Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-3-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/sch_netem.c | 40 ---------------------------------------- 1 file changed, 40 deletions(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index bc18e1976b6e..d97acd2f3923 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -1007,41 +1007,6 @@ static int parse_attr(struct nlattr *tb[], int maxtype, struct nlattr *nla, return 0; } -static const struct Qdisc_class_ops netem_class_ops; - -static int check_netem_in_tree(struct Qdisc *sch, bool duplicates, - struct netlink_ext_ack *extack) -{ - struct Qdisc *root, *q; - unsigned int i; - - root = qdisc_root_sleeping(sch); - - if (sch != root && root->ops->cl_ops == &netem_class_ops) { - if (duplicates || - ((struct netem_sched_data *)qdisc_priv(root))->duplicate) - goto err; - } - - if (!qdisc_dev(root)) - return 0; - - hash_for_each(qdisc_dev(root)->qdisc_hash, i, q, hash) { - if (sch != q && q->ops->cl_ops == &netem_class_ops) { - if (duplicates || - ((struct netem_sched_data *)qdisc_priv(q))->duplicate) - goto err; - } - } - - return 0; - -err: - NL_SET_ERR_MSG(extack, - "netem: cannot mix duplicating netems with other netems in tree"); - return -EINVAL; -} - /* Parse netlink message to set options */ static int netem_change(struct Qdisc *sch, struct nlattr *opt, struct netlink_ext_ack *extack) @@ -1118,11 +1083,6 @@ static int netem_change(struct Qdisc *sch, struct nlattr *opt, q->gap = qopt->gap; q->counter = 0; q->loss = qopt->loss; - - ret = check_netem_in_tree(sch, qopt->duplicate, extack); - if (ret) - goto unlock; - q->duplicate = qopt->duplicate; /* for compatibility with earlier versions. From b213a4c6074fc4ee4f1cdef9a73b34732606b637 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:50 -0400 Subject: [PATCH 335/411] Revert "selftests/tc-testing: Add tests for restrictions on netem duplication" This reverts commit ecdec65ec78d67d3ebd17edc88b88312054abe0d. The tests added were related to check_netem_in_tree() which was just reverted in the previous patch. Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-4-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- .../tc-testing/tc-tests/infra/qdiscs.json | 5 +- .../tc-testing/tc-tests/qdiscs/netem.json | 81 ------------------- 2 files changed, 3 insertions(+), 83 deletions(-) diff --git a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json index 848696c373fc..82c38a13dfbf 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json +++ b/tools/testing/selftests/tc-testing/tc-tests/infra/qdiscs.json @@ -702,6 +702,7 @@ "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 netem duplicate 100%", "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 1 u32 match ip dst 10.10.10.1/32 flowid 1:1", "$TC class add dev $DUMMY parent 1:0 classid 1:2 hfsc ls m2 10Mbit", + "$TC qdisc add dev $DUMMY parent 1:2 handle 3:0 netem duplicate 100%", "$TC filter add dev $DUMMY parent 1:0 protocol ip prio 2 u32 match ip dst 10.10.10.2/32 flowid 1:2", "ping -c 1 10.10.10.1 -I$DUMMY > /dev/null || true", "$TC filter del dev $DUMMY parent 1:0 protocol ip prio 1", @@ -714,8 +715,8 @@ { "kind": "hfsc", "handle": "1:", - "bytes": 294, - "packets": 3 + "bytes": 392, + "packets": 4 } ], "matchCount": "1", diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json index 718d2df2aafa..3c4444961488 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json @@ -336,86 +336,5 @@ "teardown": [ "$TC qdisc del dev $DUMMY handle 1: root" ] - }, - { - "id": "d34d", - "name": "NETEM test qdisc duplication restriction in qdisc tree in netem_change root", - "category": ["qdisc", "netem"], - "plugins": { - "requires": "nsPlugin" - }, - "setup": [ - "$TC qdisc add dev $DUMMY root handle 1: netem limit 1", - "$TC qdisc add dev $DUMMY parent 1: handle 2: netem limit 1" - ], - "cmdUnderTest": "$TC qdisc change dev $DUMMY handle 1: netem duplicate 50%", - "expExitCode": "2", - "verifyCmd": "$TC -s qdisc show dev $DUMMY", - "matchPattern": "qdisc netem", - "matchCount": "2", - "teardown": [ - "$TC qdisc del dev $DUMMY handle 1:0 root" - ] - }, - { - "id": "b33f", - "name": "NETEM test qdisc duplication restriction in qdisc tree in netem_change non-root", - "category": ["qdisc", "netem"], - "plugins": { - "requires": "nsPlugin" - }, - "setup": [ - "$TC qdisc add dev $DUMMY root handle 1: netem limit 1", - "$TC qdisc add dev $DUMMY parent 1: handle 2: netem limit 1" - ], - "cmdUnderTest": "$TC qdisc change dev $DUMMY handle 2: netem duplicate 50%", - "expExitCode": "2", - "verifyCmd": "$TC -s qdisc show dev $DUMMY", - "matchPattern": "qdisc netem", - "matchCount": "2", - "teardown": [ - "$TC qdisc del dev $DUMMY handle 1:0 root" - ] - }, - { - "id": "cafe", - "name": "NETEM test qdisc duplication restriction in qdisc tree", - "category": ["qdisc", "netem"], - "plugins": { - "requires": "nsPlugin" - }, - "setup": [ - "$TC qdisc add dev $DUMMY root handle 1: netem limit 1 duplicate 100%" - ], - "cmdUnderTest": "$TC qdisc add dev $DUMMY parent 1: handle 2: netem duplicate 100%", - "expExitCode": "2", - "verifyCmd": "$TC -s qdisc show dev $DUMMY", - "matchPattern": "qdisc netem", - "matchCount": "1", - "teardown": [ - "$TC qdisc del dev $DUMMY handle 1:0 root" - ] - }, - { - "id": "1337", - "name": "NETEM test qdisc duplication restriction in qdisc tree across branches", - "category": ["qdisc", "netem"], - "plugins": { - "requires": "nsPlugin" - }, - "setup": [ - "$TC qdisc add dev $DUMMY parent root handle 1:0 hfsc", - "$TC class add dev $DUMMY parent 1:0 classid 1:1 hfsc rt m2 10Mbit", - "$TC qdisc add dev $DUMMY parent 1:1 handle 2:0 netem", - "$TC class add dev $DUMMY parent 1:0 classid 1:2 hfsc rt m2 10Mbit" - ], - "cmdUnderTest": "$TC qdisc add dev $DUMMY parent 1:2 handle 3:0 netem duplicate 100%", - "expExitCode": "2", - "verifyCmd": "$TC -s qdisc show dev $DUMMY", - "matchPattern": "qdisc netem", - "matchCount": "1", - "teardown": [ - "$TC qdisc del dev $DUMMY handle 1:0 root" - ] } ] From 9552b11e3edabc97cfcd9f29103d5afbce7ae183 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:51 -0400 Subject: [PATCH 336/411] net/sched: fix packet loop on netem when duplicate is on When netem duplicates a packet it re-enqueues the copy at the root qdisc. If another netem sits in the tree the copy can be duplicated again, recursing until the stack or memory is exhausted. The original duplication guard temporarily zeroed q->duplicate around the re-enqueue, but that does not cover all cases because it is per-qdisc state shared across all concurrent enqueue paths and is not safe without additional locking. Use the skb tc_depth field introduced in an earlier patch: - increment it on the duplicate before re-enqueue - skip duplication for any skb whose tc_depth is already non-zero. This marks the packet itself rather than mutating qdisc state, therefore it is safe regardless of tree topology or concurrency. Fixes: 0afb51e72855 ("[PKT_SCHED]: netem: reinsert for duplication") Reported-by: William Liu Reported-by: Savino Dicanosa Closes: https://lore.kernel.org/netdev/8DuRWwfqjoRDLDmBMlIfbrsZg9Gx50DHJc1ilxsEBNe2D6NMoigR_eIRIG0LOjMc3r10nUUZtArXx4oZBIdUfZQrwjcQhdinnMis_0G7VEk=@willsroot.io/ Co-developed-by: Victor Nogueira Signed-off-by: Victor Nogueira Reviewed-by: William Liu Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-5-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/sch_netem.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/net/sched/sch_netem.c b/net/sched/sch_netem.c index d97acd2f3923..17a79fe2f091 100644 --- a/net/sched/sch_netem.c +++ b/net/sched/sch_netem.c @@ -461,7 +461,8 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, skb->prev = NULL; /* Random duplication */ - if (q->duplicate && q->duplicate >= get_crandom(&q->dup_cor, &q->prng)) + if (q->duplicate && skb->tc_depth == 0 && + q->duplicate >= get_crandom(&q->dup_cor, &q->prng)) ++count; /* Drop packet? */ @@ -540,11 +541,9 @@ static int netem_enqueue(struct sk_buff *skb, struct Qdisc *sch, */ if (skb2) { struct Qdisc *rootq = qdisc_root_bh(sch); - u32 dupsave = q->duplicate; /* prevent duplicating a dup... */ - q->duplicate = 0; + skb2->tc_depth++; /* prevent duplicating a dup... */ rootq->enqueue(skb2, rootq, to_free); - q->duplicate = dupsave; skb2 = NULL; } From db875221ab08d213a83bf30196ae8b64d55a3403 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Mon, 25 May 2026 08:25:52 -0400 Subject: [PATCH 337/411] net/sched: Fix ethx:ingress -> ethy:egress -> ethx:ingress mirred loop When mirred redirects to ingress (from either ingress or egress) the loop state from sched_mirred_dev array dev is lost because of 1) the packet deferral into the backlog and 2) the fact the sched_mirred_dev array is cleared. In such cases, if there was a loop we won't discover it. Here's a simple test to reproduce: ip a add dev port0 10.10.10.11/24 tc qdisc add dev port0 clsact tc filter add dev port0 egress protocol ip \ prio 10 matchall action mirred ingress redirect dev port1 tc qdisc add dev port1 clsact tc filter add dev port1 ingress protocol ip \ prio 10 matchall action mirred egress redirect dev port0 ping -c 1 -W0.01 10.10.10.10 Fixes: fe946a751d9b ("net/sched: act_mirred: add loop detection") Tested-by: Victor Nogueira Reviewed-by: Stephen Hemminger Signed-off-by: Jamal Hadi Salim Link: https://patch.msgid.link/20260525122556.973584-6-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_mirred.c | 47 +++++++++++++++++++++++++++--------------- 1 file changed, 30 insertions(+), 17 deletions(-) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index 2c5a7a321a94..dd5e7ea7ef26 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -26,6 +26,10 @@ #include #include +#define MIRRED_DEFER_LIMIT 3 +_Static_assert(MIRRED_DEFER_LIMIT <= 3, + "MIRRED_DEFER_LIMIT exceeds tc_depth bitfield width"); + static LIST_HEAD(mirred_list); static DEFINE_SPINLOCK(mirred_list_lock); @@ -234,12 +238,15 @@ tcf_mirred_forward(bool at_ingress, bool want_ingress, struct sk_buff *skb) { int err; - if (!want_ingress) + if (!want_ingress) { err = tcf_dev_queue_xmit(skb, dev_queue_xmit); - else if (!at_ingress) - err = netif_rx(skb); - else - err = netif_receive_skb(skb); + } else { + skb->tc_depth++; + if (!at_ingress) + err = netif_rx(skb); + else + err = netif_receive_skb(skb); + } return err; } @@ -426,6 +433,7 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, struct netdev_xmit *xmit; bool m_mac_header_xmit; struct net_device *dev; + bool want_ingress; int i, m_eaction; u32 blockid; @@ -434,7 +442,8 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, #else xmit = this_cpu_ptr(&softnet_data.xmit); #endif - if (unlikely(xmit->sched_mirred_nest >= MIRRED_NEST_LIMIT)) { + if (unlikely(xmit->sched_mirred_nest >= MIRRED_NEST_LIMIT || + skb->tc_depth >= MIRRED_DEFER_LIMIT)) { net_warn_ratelimited("Packet exceeded mirred recursion limit on dev %s\n", netdev_name(skb->dev)); return TC_ACT_SHOT; @@ -453,23 +462,27 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, tcf_action_inc_overlimit_qstats(&m->common); return retval; } - for (i = 0; i < xmit->sched_mirred_nest; i++) { - if (xmit->sched_mirred_dev[i] != dev) - continue; - pr_notice_once("tc mirred: loop on device %s\n", - netdev_name(dev)); - tcf_action_inc_overlimit_qstats(&m->common); - return retval; + + m_eaction = READ_ONCE(m->tcfm_eaction); + want_ingress = tcf_mirred_act_wants_ingress(m_eaction); + if (!want_ingress) { + for (i = 0; i < xmit->sched_mirred_nest; i++) { + if (xmit->sched_mirred_dev[i] != dev) + continue; + pr_notice_once("tc mirred: loop on device %s\n", + netdev_name(dev)); + tcf_action_inc_overlimit_qstats(&m->common); + return retval; + } + xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; } - xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; - m_mac_header_xmit = READ_ONCE(m->tcfm_mac_header_xmit); - m_eaction = READ_ONCE(m->tcfm_eaction); retval = tcf_mirred_to_dev(skb, m, dev, m_mac_header_xmit, m_eaction, retval); - xmit->sched_mirred_nest--; + if (!want_ingress) + xmit->sched_mirred_nest--; return retval; } From a005fa5d7502eefec7ee6e1c01adadc06de2f9ad Mon Sep 17 00:00:00 2001 From: "Kito Xu (veritas501)" Date: Mon, 25 May 2026 08:25:53 -0400 Subject: [PATCH 338/411] net/sched: act_mirred: Fix blockcast recursion bypass leading to stack overflow tcf_mirred_act() checks sched_mirred_nest against MIRRED_NEST_LIMIT (4) to prevent deep recursion. However, when the action uses blockcast (tcfm_blockid != 0), the function returns at the tcf_blockcast() call BEFORE reaching the counter increment. As a result, the recursion counter never advances and the limit check is entirely bypassed. When two devices share a TC egress block with a mirred blockcast rule, a packet egressing on device A is mirrored to device B via blockcast; device B's egress TC re-enters tcf_mirred_act() via blockcast and mirrors back to A, creating an unbounded recursion loop: tcf_mirred_act -> tcf_blockcast -> tcf_mirred_to_dev -> dev_queue_xmit -> sch_handle_egress -> tcf_classify -> tcf_mirred_act -> (repeat) This recursion continues until the kernel stack overflows. The bug is reachable from an unprivileged user via unshare(CLONE_NEWUSER | CLONE_NEWNET): user namespaces grant CAP_NET_ADMIN in the new network namespace, which is sufficient to create dummy devices, attach clsact qdiscs with shared blocks, and install mirred blockcast filters. BUG: TASK stack guard page was hit at ffffc90000b7fff8 Oops: stack guard page: 0000 [#1] SMP KASAN NOPTI CPU: 2 UID: 1000 PID: 169 Comm: poc Not tainted 7.0.0-rc7-next-20260410 RIP: 0010:xas_find+0x17/0x480 Call Trace: xa_find+0x17b/0x1d0 tcf_mirred_act+0x640/0x1060 tcf_action_exec+0x400/0x530 basic_classify+0x128/0x1d0 tcf_classify+0xd83/0x1150 tc_run+0x328/0x620 __dev_queue_xmit+0x797/0x3100 tcf_mirred_to_dev+0x7b1/0xf70 tcf_mirred_act+0x68a/0x1060 [repeating ~30+ times until stack overflow] Kernel panic - not syncing: Fatal exception in interrupt Fix this by incrementing sched_mirred_nest before calling tcf_blockcast() and decrementing it on return, mirroring the non-blockcast path. This ensures subsequent recursive entries see the updated counter and are correctly limited by MIRRED_NEST_LIMIT. Fixes: fe946a751d9b ("net/sched: act_mirred: add loop detection") Signed-off-by: Kito Xu (veritas501) Link: https://patch.msgid.link/20260525122556.973584-7-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_mirred.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index dd5e7ea7ef26..dbe4a4ff3e08 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -396,14 +396,12 @@ static int tcf_blockcast_mirror(struct sk_buff *skb, struct tcf_mirred *m, static int tcf_blockcast(struct sk_buff *skb, struct tcf_mirred *m, const u32 blockid, struct tcf_result *res, - int retval) + int m_eaction, int retval) { const u32 exception_ifindex = skb->dev->ifindex; struct tcf_block *block; bool is_redirect; - int m_eaction; - m_eaction = READ_ONCE(m->tcfm_eaction); is_redirect = tcf_mirred_is_act_redirect(m_eaction); /* we are already under rcu protection, so can call block lookup @@ -453,8 +451,16 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, tcf_action_update_bstats(&m->common, skb); blockid = READ_ONCE(m->tcfm_blockid); - if (blockid) - return tcf_blockcast(skb, m, blockid, res, retval); + m_eaction = READ_ONCE(m->tcfm_eaction); + want_ingress = tcf_mirred_act_wants_ingress(m_eaction); + if (blockid) { + if (!want_ingress) + xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = NULL; + retval = tcf_blockcast(skb, m, blockid, res, m_eaction, retval); + if (!want_ingress) + xmit->sched_mirred_nest--; + return retval; + } dev = rcu_dereference_bh(m->tcfm_dev); if (unlikely(!dev)) { @@ -463,8 +469,6 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, return retval; } - m_eaction = READ_ONCE(m->tcfm_eaction); - want_ingress = tcf_mirred_act_wants_ingress(m_eaction); if (!want_ingress) { for (i = 0; i < xmit->sched_mirred_nest; i++) { if (xmit->sched_mirred_dev[i] != dev) From e80ad525fc7e8c933ad78478c5dda286cfd55c60 Mon Sep 17 00:00:00 2001 From: Victor Nogueira Date: Mon, 25 May 2026 08:25:54 -0400 Subject: [PATCH 339/411] net/sched: act_mirred: Fix return code in early mirred redirect error paths Since retval is set as TC_ACT_STOLEN in the mirred redirect case, returning retval in cases where redirect failed will make the callers not register the skb as being dropped. Fix this by returning TC_ACT_SHOT instead in such scenarios. Fixes: 16085e48cb48 ("net/sched: act_mirred: Create function tcf_mirred_to_dev and improve readability") Reported-by: Sashiko Closes: https://sashiko.dev/#/patchset/20260413082027.2244884-1-hxzene%40gmail.com Signed-off-by: Victor Nogueira Link: https://patch.msgid.link/20260525122556.973584-8-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- net/sched/act_mirred.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/net/sched/act_mirred.c b/net/sched/act_mirred.c index dbe4a4ff3e08..553342c55cf7 100644 --- a/net/sched/act_mirred.c +++ b/net/sched/act_mirred.c @@ -372,7 +372,8 @@ static int tcf_blockcast_redir(struct sk_buff *skb, struct tcf_mirred *m, dev_is_mac_header_xmit(dev_prev), m_eaction, retval); - return retval; + /* If the packet wasn't redirected, we have to register as a drop */ + return TC_ACT_SHOT; } static int tcf_blockcast_mirror(struct sk_buff *skb, struct tcf_mirred *m, @@ -410,7 +411,7 @@ static int tcf_blockcast(struct sk_buff *skb, struct tcf_mirred *m, block = tcf_block_lookup(dev_net(skb->dev), blockid); if (!block || xa_empty(&block->ports)) { tcf_action_inc_overlimit_qstats(&m->common); - return retval; + return is_redirect ? TC_ACT_SHOT : retval; } if (is_redirect) @@ -428,8 +429,8 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, { struct tcf_mirred *m = to_mirred(a); int retval = READ_ONCE(m->tcf_action); + bool m_mac_header_xmit, is_redirect; struct netdev_xmit *xmit; - bool m_mac_header_xmit; struct net_device *dev; bool want_ingress; int i, m_eaction; @@ -462,11 +463,13 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, return retval; } + is_redirect = tcf_mirred_is_act_redirect(m_eaction); + dev = rcu_dereference_bh(m->tcfm_dev); if (unlikely(!dev)) { pr_notice_once("tc mirred: target device is gone\n"); tcf_action_inc_overlimit_qstats(&m->common); - return retval; + goto err_out; } if (!want_ingress) { @@ -476,7 +479,7 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, pr_notice_once("tc mirred: loop on device %s\n", netdev_name(dev)); tcf_action_inc_overlimit_qstats(&m->common); - return retval; + goto err_out; } xmit->sched_mirred_dev[xmit->sched_mirred_nest++] = dev; } @@ -489,6 +492,11 @@ TC_INDIRECT_SCOPE int tcf_mirred_act(struct sk_buff *skb, xmit->sched_mirred_nest--; return retval; + +err_out: + if (is_redirect) + retval = TC_ACT_SHOT; + return retval; } static void tcf_stats_update(struct tc_action *a, u64 bytes, u64 packets, From d38dc56a0225664e494221b5b251931b35d125ef Mon Sep 17 00:00:00 2001 From: Victor Nogueira Date: Mon, 25 May 2026 08:25:55 -0400 Subject: [PATCH 340/411] selftests/tc-testing: Add mirred test cases exercising loops Add mirred loop test cases to validate that those will be caught and other test cases that were previously misinterpreted as loops by mirred. This commit adds 12 test cases: - Redirect multiport: dummy egress -> dev1 ingress -> dummy egress (Loop) - Redirect singleport: dev1 ingress -> dev1 egress -> dev1 ingress (Loop) - Redirect multiport: dev1 ingress -> dummy ingress -> dev1 egress (No Loop) - Redirect multiport: dev1 ingress -> dummy ingress -> dev1 ingress (Loop) - Redirect multiport: dev1 ingress -> dummy egress -> dev1 ingress (Loop) - Redirect multiport: dummy egress -> dev1 ingress -> dummy egress, different prios (Loop) - Redirect multiport: dev1 ingress -> dummy ingress -> dummy egress -> dev1 egress (No Loop) - Redirect multiport: dev1 ingress -> dummy egress -> dev1 egress (No Loop) - Redirect multiport: dev1 ingress -> dummy egress -> dummy ingress (No Loop) - Redirect singleport: dev1 ingress -> dev1 ingress (Loop) - Redirect singleport: dummy egress -> dummy ingress (No Loop) - Redirect multiport: dev1 ingress -> dummy ingress -> dummy egress (No Loop) Acked-by: Jamal Hadi Salim Acked-by: Stephen Hemminger Signed-off-by: Victor Nogueira Link: https://patch.msgid.link/20260525122556.973584-9-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- .../tc-testing/tc-tests/actions/mirred.json | 616 +++++++++++++++++- 1 file changed, 615 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/tc-testing/tc-tests/actions/mirred.json b/tools/testing/selftests/tc-testing/tc-tests/actions/mirred.json index b056eb966871..d0cad6571691 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/actions/mirred.json +++ b/tools/testing/selftests/tc-testing/tc-tests/actions/mirred.json @@ -1144,6 +1144,620 @@ "teardown": [ "$TC qdisc del dev $DUMMY clsact" ] + }, + { + "id": "531c", + "name": "Redirect multiport: dummy egress -> dev1 ingress -> dummy egress (Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin" + ] + }, + "setup": [ + "$IP link set dev $DUMMY up || true", + "$IP addr add 10.10.10.10/24 dev $DUMMY || true", + "$TC qdisc add dev $DUMMY clsact", + "$TC filter add dev $DUMMY egress protocol ip prio 10 matchall action mirred ingress redirect dev $DEV1 index 1", + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred egress redirect dev $DUMMY index 2" + ], + "cmdUnderTest": "ping -c1 -W0.01 -I $DUMMY 10.10.10.1", + "expExitCode": "1", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 3 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DUMMY clsact", + "$TC qdisc del dev $DEV1 clsact" + ] + }, + { + "id": "b1d7", + "name": "Redirect singleport: dev1 ingress -> dev1 egress -> dev1 ingress (Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred egress redirect dev $DEV1 index 1" + ], + "cmdUnderTest": "$TC filter add dev $DEV1 egress protocol ip prio 11 matchall action mirred ingress redirect dev $DEV1 index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "egress", + "index": 1, + "stats": { + "packets": 3 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact" + ] + }, + { + "id": "c66d", + "name": "Redirect multiport: dev1 ingress -> dummy ingress -> dev1 egress (No Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred ingress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY ingress protocol ip prio 11 matchall action mirred egress redirect dev $DEV1 index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "aa99", + "name": "Redirect multiport: dev1 ingress -> dummy ingress -> dev1 ingress (Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred ingress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY ingress protocol ip prio 11 matchall action mirred ingress redirect dev $DEV1 index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 2, + "overlimits": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "37d7", + "name": "Redirect multiport: dev1 ingress -> dummy egress -> dev1 ingress (Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred egress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY egress protocol ip prio 11 matchall action mirred ingress redirect dev $DEV1 index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "egress", + "index": 1, + "stats": { + "packets": 3 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "6d02", + "name": "Redirect multiport: dummy egress -> dev1 ingress -> dummy egress, different prios (Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin" + ] + }, + "setup": [ + "$IP link set dev $DUMMY up || true", + "$IP addr add 10.10.10.10/24 dev $DUMMY || true", + "$TC qdisc add dev $DUMMY clsact", + "$TC filter add dev $DUMMY egress protocol ip prio 10 matchall action mirred ingress redirect dev $DEV1 index 1", + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 11 matchall action mirred egress redirect dev $DUMMY index 2" + ], + "cmdUnderTest": "ping -c1 -W0.01 -I $DUMMY 10.10.10.1", + "expExitCode": "1", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 3 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DUMMY clsact", + "$TC qdisc del dev $DEV1 clsact" + ] + }, + { + "id": "8115", + "name": "Redirect multiport: dev1 ingress -> dummy ingress -> dummy egress -> dev1 egress (No Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred ingress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact", + "$TC filter add dev $DUMMY ingress protocol ip prio 11 matchall action mirred egress redirect dev $DUMMY index 2" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY egress protocol ip prio 12 matchall action mirred egress redirect dev $DEV1 index 3", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "9eb3", + "name": "Redirect multiport: dev1 ingress -> dummy egress -> dev1 egress (No Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred egress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY egress protocol ip prio 11 matchall action mirred egress redirect dev $DEV1 index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "egress", + "index": 1, + "stats": { + "packets": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "d837", + "name": "Redirect multiport: dev1 ingress -> dummy egress -> dummy ingress (No Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred egress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY egress protocol ip prio 11 matchall action mirred ingress redirect dev $DUMMY index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "egress", + "index": 1, + "stats": { + "packets": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "2071", + "name": "Redirect singleport: dev1 ingress -> dev1 ingress (Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact" + ], + "cmdUnderTest": "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred ingress redirect dev $DEV1 index 1", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 1, + "overlimits": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact" + ] + }, + { + "id": "0101", + "name": "Redirect singleport: dummy egress -> dummy ingress (No Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin" + ] + }, + "setup": [ + "$IP addr add 10.10.10.10/24 dev $DUMMY || true", + "$TC qdisc add dev $DUMMY clsact", + "$TC filter add dev $DUMMY egress protocol ip prio 11 matchall action mirred ingress redirect dev $DUMMY index 1" + ], + "cmdUnderTest": "ping -c1 -W0.01 -I $DUMMY 10.10.10.1", + "expExitCode": "1", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DUMMY clsact" + ] + }, + { + "id": "cf97", + "name": "Redirect multiport: dev1 ingress -> dummy ingress -> dummy egress (No Loop)", + "category": [ + "filter", + "mirred" + ], + "plugins": { + "requires": [ + "nsPlugin", + "scapyPlugin" + ] + }, + "setup": [ + "$TC qdisc add dev $DEV1 clsact", + "$TC filter add dev $DEV1 ingress protocol ip prio 10 matchall action mirred ingress redirect dev $DUMMY index 1", + "$TC qdisc add dev $DUMMY clsact" + ], + "cmdUnderTest": "$TC filter add dev $DUMMY ingress protocol ip prio 11 matchall action mirred egress redirect dev $DUMMY index 2", + "scapy": [ + { + "iface": "$DEV0", + "count": 1, + "packet": "Ether()/IP(dst='10.10.10.1', src='10.10.10.10')/ICMP()" + } + ], + "expExitCode": "0", + "verifyCmd": "$TC -j -s actions get action mirred index 1", + "matchJSON": [ + { + "total acts": 0 + }, + { + "actions": [ + { + "order": 1, + "kind": "mirred", + "mirred_action": "redirect", + "direction": "ingress", + "index": 1, + "stats": { + "packets": 1 + }, + "not_in_hw": true + } + ] + } + ], + "teardown": [ + "$TC qdisc del dev $DEV1 clsact", + "$TC qdisc del dev $DUMMY clsact" + ] } - ] From 0f6e00aa5f652f5653e0039b9c9a8835f4b4174b Mon Sep 17 00:00:00 2001 From: Victor Nogueira Date: Mon, 25 May 2026 08:25:56 -0400 Subject: [PATCH 341/411] selftests/tc-testing: Add netem test case exercising loops Add a netem nested duplicate test case to validate that it won't cause an infinite loop Acked-by: Jamal Hadi Salim Acked-by: Stephen Hemminger Signed-off-by: Victor Nogueira Link: https://patch.msgid.link/20260525122556.973584-10-jhs@mojatatu.com Signed-off-by: Paolo Abeni --- .../tc-testing/tc-tests/qdiscs/netem.json | 33 ++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json index 3c4444961488..472b672a600d 100644 --- a/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json +++ b/tools/testing/selftests/tc-testing/tc-tests/qdiscs/netem.json @@ -336,5 +336,36 @@ "teardown": [ "$TC qdisc del dev $DUMMY handle 1: root" ] - } + }, + { + "id": "8c17", + "name": "Test netem's recursive duplicate", + "category": [ + "qdisc", + "netem" + ], + "plugins": { + "requires": "nsPlugin" + }, + "setup": [ + "$IP link set dev $DUMMY up || true", + "$IP addr add 10.10.11.10/24 dev $DUMMY || true", + "$TC qdisc add dev $DUMMY root handle 1: netem limit 1000 duplicate 100%", + "$TC qdisc add dev $DUMMY parent 1: handle 2: netem limit 1000 duplicate 100%" + ], + "cmdUnderTest": "ping -c 1 10.10.11.11 -W 0.01", + "expExitCode": "1", + "verifyCmd": "$TC -s -j qdisc ls dev $DUMMY root", + "matchJSON": [ + { + "kind": "netem", + "handle": "1:", + "bytes": 294, + "packets": 3 + } + ], + "teardown": [ + "$TC qdisc del dev $DUMMY handle 1: root" + ] + } ] From 463a1271aa26eac992851b9d98cc75bc3cd4a1ed Mon Sep 17 00:00:00 2001 From: Jijie Shao Date: Mon, 25 May 2026 22:45:24 +0800 Subject: [PATCH 342/411] net: hibmcge: disable Relaxed Ordering to fix RX packet corruption When SMMU is disabled, the hibmcge driver may receive corrupted packets. The hardware writes packet data and descriptors to the same page, but with Relaxed Ordering enabled, PCI write transactions may not be strictly ordered. This can cause the driver to observe a valid descriptor before the corresponding packet data is fully written. Fix this by clearing PCI_EXP_DEVCTL_RELAX_EN in the PCI bridge control register to ensure strict write ordering between packet data and descriptors. Fixes: f72e25594061 ("net: hibmcge: Implement rx_poll function to receive packets") Signed-off-by: Jijie Shao Link: https://patch.msgid.link/20260525144525.94884-2-shaojijie@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c index 068da2fd1fea..f721e9893804 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_main.c @@ -420,6 +420,9 @@ static int hbg_pci_init(struct pci_dev *pdev) return -ENOMEM; pci_set_master(pdev); + pcie_capability_clear_word(pdev, PCI_EXP_DEVCTL, + PCI_EXP_DEVCTL_RELAX_EN); + pci_save_state(pdev); return 0; } From b545b6ea1802b32436fa97f1d2918718212cc831 Mon Sep 17 00:00:00 2001 From: Jijie Shao Date: Mon, 25 May 2026 22:45:25 +0800 Subject: [PATCH 343/411] net: hibmcge: move dma_rmb() after dma_sync_single_for_cpu() in RX path The dma_rmb() barrier was placed before dma_sync_single_for_cpu(), which is incorrect. DMA sync must complete first to make the buffer accessible to the CPU, then the rmb barrier ensures subsequent descriptor reads observe the latest data written by the hardware. Reorder the operations so dma_sync_single_for_cpu() is called before dma_rmb() to guarantee the driver reads consistent data from the DMA buffer. Fixes: f72e25594061 ("net: hibmcge: Implement rx_poll function to receive packets") Signed-off-by: Jijie Shao Link: https://patch.msgid.link/20260525144525.94884-3-shaojijie@huawei.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c index a4ea92c31c2f..0ae314994676 100644 --- a/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c +++ b/drivers/net/ethernet/hisilicon/hibmcge/hbg_txrx.c @@ -452,12 +452,12 @@ static bool hbg_sync_data_from_hw(struct hbg_priv *priv, { struct hbg_rx_desc *rx_desc; - /* make sure HW write desc complete */ - dma_rmb(); - dma_sync_single_for_cpu(&priv->pdev->dev, buffer->page_dma, buffer->page_size, DMA_FROM_DEVICE); + /* make sure HW write desc complete */ + dma_rmb(); + rx_desc = (struct hbg_rx_desc *)buffer->page_addr; return FIELD_GET(HBG_RX_DESC_W2_PKT_LEN_M, rx_desc->word2) != 0; } From 98d0912e9f841e5529a5b89a972805f34cb1c69d Mon Sep 17 00:00:00 2001 From: Minh Nguyen Date: Tue, 26 May 2026 11:12:39 +0700 Subject: [PATCH 344/411] net: skbuff: fix missing zerocopy reference in pskb_carve helpers pskb_carve_inside_header() and pskb_carve_inside_nonlinear() both copy the old skb_shared_info header into a new buffer via memcpy(), which includes the destructor_arg pointer (uarg) for MSG_ZEROCOPY skbs. Neither function calls net_zcopy_get() for the new shinfo, creating an unaccounted holder: every skb_shared_info with destructor_arg set will call skb_zcopy_clear() once when freed, but the corresponding net_zcopy_get() was never called for the new copy. Repeated calls drive uarg->refcnt to zero prematurely, freeing ubuf_info_msgzc while TX skbs still hold live destructor_arg pointers. KASAN reports use-after-free on a freed ubuf_info_msgzc: BUG: KASAN: slab-use-after-free in skb_release_data+0x77b/0x810 Read of size 8 at addr ffff88801574d3e8 by task poc/220 Call Trace: skb_release_data+0x77b/0x810 kfree_skb_list_reason+0x13e/0x610 skb_release_data+0x4cd/0x810 sk_skb_reason_drop+0xf3/0x340 skb_queue_purge_reason+0x282/0x440 rds_tcp_inc_free+0x1e/0x30 rds_recvmsg+0x354/0x1780 __sys_recvmsg+0xdf/0x180 Allocated by task 219: msg_zerocopy_realloc+0x157/0x7b0 tcp_sendmsg_locked+0x2892/0x3ba0 Freed by task 219: ip_recv_error+0x74a/0xb10 tcp_recvmsg+0x475/0x530 The skb consuming the late access still referenced the same uarg via shinfo->destructor_arg copied by pskb_carve_inside_nonlinear() without a refcount bump. This has been verified to be reliably exploitable: a working proof-of-concept achieves full root privilege escalation from an unprivileged local user on a default kernel configuration. The fix follows the pattern of pskb_expand_head() which has the same memcpy/cloned structure. For pskb_carve_inside_header(), net_zcopy_get() is placed after skb_orphan_frags() succeeds, so the orphan error path needs no cleanup. For pskb_carve_inside_nonlinear(), net_zcopy_get() is placed after all failure points and just before skb_release_data(), so no error path needs cleanup at all -- matching pskb_expand_head() more closely and avoiding the need for a balancing net_zcopy_put(). Fixes: 6fa01ccd8830 ("skbuff: Add pskb_extract() helper function") Cc: stable@vger.kernel.org Assisted-by: Claude:claude-sonnet-4-6 Signed-off-by: Minh Nguyen Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260526041240.329462-1-minhnguyen.080505@gmail.com Signed-off-by: Paolo Abeni --- net/core/skbuff.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index d247acd447e4..0d3cc115f2e7 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6833,6 +6833,8 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off, skb_kfree_head(data); return -ENOMEM; } + if (skb_zcopy(skb)) + net_zcopy_get(skb_zcopy(skb)); for (i = 0; i < skb_shinfo(skb)->nr_frags; i++) skb_frag_ref(skb, i); if (skb_has_frag_list(skb)) @@ -6976,6 +6978,8 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off, skb_kfree_head(data); return -ENOMEM; } + if (skb_zcopy(skb)) + net_zcopy_get(skb_zcopy(skb)); skb_release_data(skb, SKB_CONSUMED); skb->head = data; From cc993e0927ec8bd98ea33377ada03295fcda0f24 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:15 -0400 Subject: [PATCH 345/411] net/handshake: Use spin_lock_bh for hn_lock nvmet_tcp_state_change(), a socket callback that runs in BH context, can reach handshake_req_cancel() via nvmet_tcp_schedule_release_queue() and tls_handshake_cancel(). handshake_req_cancel() acquires hn->hn_lock with plain spin_lock(). If a process-context thread on the same CPU holds hn->hn_lock when a softirq invokes the cancel path, the lock attempt deadlocks. This is the only caller that invokes tls_handshake_cancel() from BH context; every other consumer calls it from process context. Deferring the cancel to process context in the NVMe target is not straightforward: nvmet_tcp_schedule_release_queue() must call tls_handshake_cancel() atomically with its state transition to DISCONNECTING. If the cancel were deferred, the handshake completion callback could fire in the window before the cancel runs, observe the unexpected state, and return without dropping its kref on the queue. Reworking that interlock is considerably more invasive than hardening the handshake lock. Convert all hn->hn_lock acquisitions from spin_lock/spin_unlock to spin_lock_bh/spin_unlock_bh so the lock is never taken with softirqs enabled. Fixes: 675b453e0241 ("nvmet-tcp: enable TLS handshake upcall") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-1-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/netlink.c | 4 ++-- net/handshake/request.c | 14 +++++++------- net/handshake/tlshd.c | 2 ++ 3 files changed, 11 insertions(+), 9 deletions(-) diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index b989456fc4c5..97114ec8027a 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -202,10 +202,10 @@ static void __net_exit handshake_net_exit(struct net *net) * accepted and are in progress will be destroyed when * the socket is closed. */ - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags); list_splice_init(&requests, &hn->hn_requests); - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); while (!list_empty(&requests)) { req = list_first_entry(&requests, struct handshake_req, hr_list); diff --git a/net/handshake/request.c b/net/handshake/request.c index 2829adbeb149..5d4a17f902d2 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -167,12 +167,12 @@ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) { bool ret = false; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); if (!list_empty(&req->hr_list)) { __remove_pending_locked(hn, req); ret = true; } - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); return ret; } @@ -182,7 +182,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) struct handshake_req *req, *pos; req = NULL; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); list_for_each_entry(pos, &hn->hn_requests, hr_list) { if (pos->hr_proto->hp_handler_class != class) continue; @@ -190,7 +190,7 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) req = pos; break; } - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); return req; } @@ -249,7 +249,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, if (READ_ONCE(hn->hn_pending) >= hn->hn_pending_max) goto out_err; - spin_lock(&hn->hn_lock); + spin_lock_bh(&hn->hn_lock); ret = -EOPNOTSUPP; if (test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags)) goto out_unlock; @@ -258,7 +258,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_unlock; if (!__add_pending_locked(hn, req)) goto out_unlock; - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); ret = handshake_genl_notify(net, req->hr_proto, flags); if (ret) { @@ -274,7 +274,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, return 0; out_unlock: - spin_unlock(&hn->hn_lock); + spin_unlock_bh(&hn->hn_lock); out_err: /* Restore original destructor so socket teardown still runs on failure */ req->hr_sk->sk_destruct = req->hr_odestruct; diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index 8f9532a15f43..af294c6cc717 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -425,6 +425,8 @@ EXPORT_SYMBOL(tls_server_hello_psk); * Request cancellation races with request completion. To determine * who won, callers examine the return value from this function. * + * Context: May be called from process or softirq context. + * * Return values: * %true - Uncompleted handshake request was canceled * %false - Handshake request already completed or not found From 9015985b5eb1a90eb86caf5bce1dfcf1aa38f8ad Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:16 -0400 Subject: [PATCH 346/411] nvme-tcp: store negative errno in queue->tls_err nvme_tcp_tls_done() assigns queue->tls_err in three branches. The ENOKEY lookup failure and the EOPNOTSUPP initializer both store negative errnos. The third branch, reached when the handshake layer reports a non-zero status, stores -status. The handshake layer delivers status to the consumer callback as a negative errno; the other in-tree consumers -- xs_tls_handshake_done() and the nvmet target callback -- treat their status argument that way. The extra negation in nvme_tcp_tls_done() flips the sign, leaving tls_err as a positive value (for instance, +EIO), which nvme_tcp_start_tls() then returns to its caller. Drop the extra negation so queue->tls_err uniformly carries a negative errno on failure. Fixes: be8e82caa685 ("nvme-tcp: enable TLS handshake upcall") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Reviewed-by: Alistair Francis Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-2-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- drivers/nvme/host/tcp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/nvme/host/tcp.c b/drivers/nvme/host/tcp.c index 15d36d6a728e..68a1d7640494 100644 --- a/drivers/nvme/host/tcp.c +++ b/drivers/nvme/host/tcp.c @@ -1702,7 +1702,7 @@ static void nvme_tcp_tls_done(void *data, int status, key_serial_t pskid) qid, pskid, status); if (status) { - queue->tls_err = -status; + queue->tls_err = status; goto out_complete; } From 6b22d433aa13f68e3cd9534ca9a5f4277bfa01c2 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:17 -0400 Subject: [PATCH 347/411] net/handshake: Pass negative errno through handshake_complete() handshake_complete() declares status as unsigned int and tls_handshake_done() negates that value (-status) before handing it to the TLS consumer. Consumers match on negative errno constants -- xs_tls_handshake_done() has switch (status) { case 0: case -EACCES: case -ETIMEDOUT: lower_transport->xprt_err = status; break; default: lower_transport->xprt_err = -EACCES; } so the API as designed expects callers to pass positive errno values that the tlshd shim then negates. Three internal callers in handshake_nl_accept_doit(), the net-exit drain, and a kunit test follow kernel convention and pass negative errnos -- -EIO, -ETIMEDOUT, -ETIMEDOUT. The implicit conversion to unsigned int turns -ETIMEDOUT into 0xFFFFFF92; the subsequent -status in tls_handshake_done() wraps back to 110, the consumer's switch falls through, and the xprt reports -EACCES on what should be -ETIMEDOUT or -EIO. Fix the API rather than the call sites. The natural kernel convention is negative errno in, negative errno out. Change handshake_complete() and hp_done to take int status, drop the negation in tls_handshake_done(), and negate once in handshake_nl_done_doit() where status arrives from the wire as an unsigned netlink attribute. The three internal callers were already correct under that convention and need no change. At the same wire boundary, declare MAX_ERRNO as the netlink policy upper bound for HANDSHAKE_A_DONE_STATUS. Attribute validation rejects out-of-range values before handshake_nl_done_doit() runs, and negating a bounded u32 there stays within int range -- closing the UBSAN-visible signed- integer overflow that an unconstrained u32 would invoke. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-3-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- Documentation/netlink/specs/handshake.yaml | 8 ++++++++ net/handshake/genl.c | 3 ++- net/handshake/genl.h | 1 + net/handshake/handshake-test.c | 2 +- net/handshake/handshake.h | 4 ++-- net/handshake/netlink.c | 2 +- net/handshake/request.c | 2 +- net/handshake/tlshd.c | 4 ++-- 8 files changed, 18 insertions(+), 8 deletions(-) diff --git a/Documentation/netlink/specs/handshake.yaml b/Documentation/netlink/specs/handshake.yaml index 95c3fade7a8d..1024297b3851 100644 --- a/Documentation/netlink/specs/handshake.yaml +++ b/Documentation/netlink/specs/handshake.yaml @@ -12,6 +12,12 @@ protocol: genetlink doc: Netlink protocol to request a transport layer security handshake. definitions: + - + type: const + name: max-errno + value: 4095 + header: linux/err.h + scope: kernel - type: enum name: handler-class @@ -80,6 +86,8 @@ attribute-sets: - name: status type: u32 + checks: + max: max-errno - name: sockfd type: s32 diff --git a/net/handshake/genl.c b/net/handshake/genl.c index 870612609491..4b20cd9cdd0e 100644 --- a/net/handshake/genl.c +++ b/net/handshake/genl.c @@ -10,6 +10,7 @@ #include "genl.h" #include +#include /* HANDSHAKE_CMD_ACCEPT - do */ static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HANDLER_CLASS + 1] = { @@ -18,7 +19,7 @@ static const struct nla_policy handshake_accept_nl_policy[HANDSHAKE_A_ACCEPT_HAN /* HANDSHAKE_CMD_DONE - do */ static const struct nla_policy handshake_done_nl_policy[HANDSHAKE_A_DONE_REMOTE_AUTH + 1] = { - [HANDSHAKE_A_DONE_STATUS] = { .type = NLA_U32, }, + [HANDSHAKE_A_DONE_STATUS] = NLA_POLICY_MAX(NLA_U32, MAX_ERRNO), [HANDSHAKE_A_DONE_SOCKFD] = { .type = NLA_S32, }, [HANDSHAKE_A_DONE_REMOTE_AUTH] = { .type = NLA_U32, }, }; diff --git a/net/handshake/genl.h b/net/handshake/genl.h index 8d3e18672daf..46b65f131669 100644 --- a/net/handshake/genl.h +++ b/net/handshake/genl.h @@ -11,6 +11,7 @@ #include #include +#include int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info); int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info); diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index 55442b2f518a..df3948e807a0 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -25,7 +25,7 @@ static int test_accept_func(struct handshake_req *req, struct genl_info *info, return 0; } -static void test_done_func(struct handshake_req *req, unsigned int status, +static void test_done_func(struct handshake_req *req, int status, struct genl_info *info) { } diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h index a48163765a7a..2289b0e274f4 100644 --- a/net/handshake/handshake.h +++ b/net/handshake/handshake.h @@ -57,7 +57,7 @@ struct handshake_proto { int (*hp_accept)(struct handshake_req *req, struct genl_info *info, int fd); void (*hp_done)(struct handshake_req *req, - unsigned int status, + int status, struct genl_info *info); void (*hp_destroy)(struct handshake_req *req); }; @@ -86,7 +86,7 @@ struct handshake_req *handshake_req_hash_lookup(struct sock *sk); struct handshake_req *handshake_req_next(struct handshake_net *hn, int class); int handshake_req_submit(struct socket *sock, struct handshake_req *req, gfp_t flags); -void handshake_complete(struct handshake_req *req, unsigned int status, +void handshake_complete(struct handshake_req *req, int status, struct genl_info *info); bool handshake_req_cancel(struct sock *sk); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 97114ec8027a..039344979de9 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -160,7 +160,7 @@ int handshake_nl_done_doit(struct sk_buff *skb, struct genl_info *info) status = -EIO; if (info->attrs[HANDSHAKE_A_DONE_STATUS]) - status = nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]); + status = -(int)nla_get_u32(info->attrs[HANDSHAKE_A_DONE_STATUS]); handshake_complete(req, status, info); sockfd_put(sock); diff --git a/net/handshake/request.c b/net/handshake/request.c index 5d4a17f902d2..97f9f8239949 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -284,7 +284,7 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, } EXPORT_SYMBOL(handshake_req_submit); -void handshake_complete(struct handshake_req *req, unsigned int status, +void handshake_complete(struct handshake_req *req, int status, struct genl_info *info) { struct sock *sk = req->hr_sk; diff --git a/net/handshake/tlshd.c b/net/handshake/tlshd.c index af294c6cc717..7567150c2a4f 100644 --- a/net/handshake/tlshd.c +++ b/net/handshake/tlshd.c @@ -93,7 +93,7 @@ static void tls_handshake_remote_peerids(struct tls_handshake_req *treq, * */ static void tls_handshake_done(struct handshake_req *req, - unsigned int status, struct genl_info *info) + int status, struct genl_info *info) { struct tls_handshake_req *treq = handshake_req_private(req); @@ -104,7 +104,7 @@ static void tls_handshake_done(struct handshake_req *req, if (!status) set_bit(HANDSHAKE_F_REQ_SESSION, &req->hr_flags); - treq->th_consumer_done(treq->th_consumer_data, -status, + treq->th_consumer_done(treq->th_consumer_data, status, treq->th_peerid[0]); } From 09dba37eee70d0596e26645015f1aa95a9848e9d Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:18 -0400 Subject: [PATCH 348/411] net/handshake: Take a long-lived file reference at submit handshake_nl_accept_doit() needs the file pointer backing req->hr_sk->sk_socket to survive the window between handshake_req_next() and the subsequent FD_PREPARE() and get_file(). The submit-side sock_hold() does not provide that. sk_refcnt keeps struct sock alive, but struct socket is owned by sock->file: when the consumer fputs the last file reference, sock_release() tears the socket down regardless of any sock_hold. Add an hr_file pointer to struct handshake_req and acquire an explicit reference on sock->file during handshake_req_submit(). handshake_complete() and handshake_req_cancel() release the reference on the completion-bit-winning path. The submit error path must also release the file reference, but after rhashtable insertion a concurrent handshake_req_cancel() can discover the request and race the error path. Gate the error-path cleanup -- sk_destruct restoration, fput, and request destruction -- with test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED), the same serialization handshake_complete() and handshake_req_cancel() already use. When cancel has already claimed ownership, the submit error path returns without touching the request; socket teardown handles final destruction. The accept-side dereferences are not yet retargeted; that change comes in the next patch. Signed-off-by: Chuck Lever Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-4-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/handshake.h | 2 ++ net/handshake/netlink.c | 6 ------ net/handshake/request.c | 42 ++++++++++++++++++++++++++++++++------- 3 files changed, 37 insertions(+), 13 deletions(-) diff --git a/net/handshake/handshake.h b/net/handshake/handshake.h index 2289b0e274f4..da61cadd1ad3 100644 --- a/net/handshake/handshake.h +++ b/net/handshake/handshake.h @@ -24,6 +24,7 @@ enum hn_flags_bits { HANDSHAKE_F_NET_DRAINING, }; +struct file; struct handshake_proto; /* One handshake request */ @@ -32,6 +33,7 @@ struct handshake_req { struct rhash_head hr_rhash; unsigned long hr_flags; const struct handshake_proto *hr_proto; + struct file *hr_file; struct sock *hr_sk; void (*hr_odestruct)(struct sock *sk); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 039344979de9..1a5821eb7184 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -210,12 +210,6 @@ static void __net_exit handshake_net_exit(struct net *net) while (!list_empty(&requests)) { req = list_first_entry(&requests, struct handshake_req, hr_list); list_del(&req->hr_list); - - /* - * Requests on this list have not yet been - * accepted, so they do not have an fd to put. - */ - handshake_complete(req, -ETIMEDOUT, NULL); } } diff --git a/net/handshake/request.c b/net/handshake/request.c index 97f9f8239949..da064511ab86 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -215,9 +216,16 @@ EXPORT_SYMBOL_IF_KUNIT(handshake_req_next); * A zero return value from handshake_req_submit() means that * exactly one subsequent completion callback is guaranteed. * - * A negative return value from handshake_req_submit() means that - * no completion callback will be done and that @req has been - * destroyed. + * A negative return value from handshake_req_submit() guarantees that + * no completion callback will occur and that @req is no longer owned by + * the caller. If cancellation wins the completion race after the request + * has been published, final destruction is deferred until socket teardown. + * + * The caller must hold a reference on @sock->file for the duration + * of this call. Once the request is published to the accept side, a + * concurrent completion or cancellation may release the request's pin on + * @sock->file; the caller's reference is what keeps @sock->sk valid until + * handshake_req_submit() returns. */ int handshake_req_submit(struct socket *sock, struct handshake_req *req, gfp_t flags) @@ -236,6 +244,14 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, kfree(req); return -EINVAL; } + + /* + * Pin sock->file for the lifetime of the request so the + * accept side does not race a consumer that releases the + * socket while a handshake is pending. + */ + req->hr_file = get_file(sock->file); + req->hr_odestruct = req->hr_sk->sk_destruct; req->hr_sk->sk_destruct = handshake_sk_destruct; @@ -267,7 +283,11 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_err; } - /* Prevent socket release while a handshake request is pending */ + /* + * Pin struct sock so sk_destruct does not run until the + * handshake completion path releases it; struct socket is + * held separately via hr_file above. + */ sock_hold(req->hr_sk); trace_handshake_submit(net, req, req->hr_sk); @@ -276,10 +296,13 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, out_unlock: spin_unlock_bh(&hn->hn_lock); out_err: - /* Restore original destructor so socket teardown still runs on failure */ - req->hr_sk->sk_destruct = req->hr_odestruct; trace_handshake_submit_err(net, req, req->hr_sk, ret); - handshake_req_destroy(req); + if (!test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) { + /* Restore original destructor so socket teardown still runs. */ + req->hr_sk->sk_destruct = req->hr_odestruct; + fput(req->hr_file); + handshake_req_destroy(req); + } return ret; } EXPORT_SYMBOL(handshake_req_submit); @@ -291,11 +314,15 @@ void handshake_complete(struct handshake_req *req, int status, struct net *net = sock_net(sk); if (!test_and_set_bit(HANDSHAKE_F_REQ_COMPLETED, &req->hr_flags)) { + struct file *file = req->hr_file; + trace_handshake_complete(net, req, sk, status); req->hr_proto->hp_done(req, status, info); /* Handshake request is no longer pending */ sock_put(sk); + + fput(file); } } EXPORT_SYMBOL_IF_KUNIT(handshake_complete); @@ -344,6 +371,7 @@ bool handshake_req_cancel(struct sock *sk) /* Handshake request is no longer pending */ sock_put(sk); + fput(req->hr_file); return true; } EXPORT_SYMBOL(handshake_req_cancel); From f4251190e58b209999c1ba9e6d2976136a1be055 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:19 -0400 Subject: [PATCH 349/411] net/handshake: hand off the pinned file reference to accept_doit handshake_req_next() removes the request from the per-net pending list and drops hn_lock before handshake_nl_accept_doit() reads req->hr_sk->sk_socket and dereferences sock->file (once in FD_PREPARE() and again in get_file()). In that window a consumer running tls_handshake_cancel() followed by sockfd_put() (svc_sock_free) or __fput_sync() (xs_reset_transport) releases sock->file. sock_release() then runs sock_orphan(), zeroing sk_socket, and frees the struct socket. The accept-side code either reads NULL through sk_socket or chases freed memory. The submit-side sock_hold() does not prevent this. sk_refcnt protects struct sock, but struct socket and sock->file are independently refcounted via the file descriptor the consumer owns. Pinning sk leaves sock and sock->file unprotected. Retarget the accept-side dereferences at req->hr_file, which was pinned at submit time, instead of req->hr_sk->sk_socket->file. Pinning on its own is not sufficient: a consumer that cancels between handshake_req_next() returning and accept_doit reaching FD_PREPARE() takes the !remove_pending() branch in handshake_req_cancel() and drops hr_file before the accept side takes its own reference. Hand off an additional file reference inside handshake_req_next(), under hn_lock, so the accept side operates on a reference that no concurrent handshake_req_cancel() can revoke. FD_PREPARE() consumes that handed-off reference, either by transferring it to the new fd in fd_publish() or by dropping it in the cleanup destructor on error; the explicit get_file() that previously balanced FD_PREPARE() is therefore redundant and goes away. Update handshake_req_cancel_test2 and _test3 to simulate the FD_PREPARE() consumption with an fput() so the kunit file-count assertions stay balanced. Reported-by: Chris Mason Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-5-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/handshake-test.c | 8 ++++++++ net/handshake/netlink.c | 7 ++----- net/handshake/request.c | 18 ++++++++++++++++++ 3 files changed, 28 insertions(+), 5 deletions(-) diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index df3948e807a0..9cc7a95f4120 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -375,6 +375,10 @@ static void handshake_req_cancel_test2(struct kunit *test) /* Pretend to accept this request */ next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD); KUNIT_ASSERT_PTR_EQ(test, req, next); + /* Simulate FD_PREPARE() consuming the file reference handed + * off by handshake_req_next(); see handshake_nl_accept_doit(). + */ + fput(filp); /* Act */ result = handshake_req_cancel(sock->sk); @@ -417,6 +421,10 @@ static void handshake_req_cancel_test3(struct kunit *test) /* Pretend to accept this request */ next = handshake_req_next(hn, HANDSHAKE_HANDLER_CLASS_TLSHD); KUNIT_ASSERT_PTR_EQ(test, req, next); + /* Simulate FD_PREPARE() consuming the file reference handed + * off by handshake_req_next(); see handshake_nl_accept_doit(). + */ + fput(filp); /* Pretend to complete this request */ handshake_complete(next, -ETIMEDOUT, NULL); diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 1a5821eb7184..21d6cbd52fcd 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -92,7 +92,6 @@ int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info) struct net *net = sock_net(skb->sk); struct handshake_net *hn = handshake_pernet(net); struct handshake_req *req = NULL; - struct socket *sock; int class, err; err = -EOPNOTSUPP; @@ -107,15 +106,13 @@ int handshake_nl_accept_doit(struct sk_buff *skb, struct genl_info *info) err = -EAGAIN; req = handshake_req_next(hn, class); if (req) { - sock = req->hr_sk->sk_socket; - - FD_PREPARE(fdf, O_CLOEXEC, sock->file); + FD_PREPARE(fdf, O_CLOEXEC, req->hr_file); if (fdf.err) { + fput(req->hr_file); /* drop ref from handshake_req_next() */ err = fdf.err; goto out_complete; } - get_file(sock->file); /* FD_PREPARE() consumes a reference. */ err = req->hr_proto->hp_accept(req, info, fd_prepare_fd(fdf)); if (err) goto out_complete; /* Automatic cleanup handles fput */ diff --git a/net/handshake/request.c b/net/handshake/request.c index da064511ab86..e2d7ee7ce6e0 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -178,6 +178,17 @@ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) return ret; } +/** + * handshake_req_next - Return the next queued handshake request + * @hn: per-net handshake state + * @class: handler class to match + * + * On a non-NULL return, the caller owns an extra reference + * on @req->hr_file. FD_PREPARE() consumes it on success; on + * the FD_PREPARE() failure path the caller must fput() it. + * + * Return: pointer to a removed handshake_req, or NULL. + */ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) { struct handshake_req *req, *pos; @@ -188,6 +199,13 @@ struct handshake_req *handshake_req_next(struct handshake_net *hn, int class) if (pos->hr_proto->hp_handler_class != class) continue; __remove_pending_locked(hn, pos); + /* Hand off a file reference to the accept side under + * hn_lock. A concurrent handshake_req_cancel() can drop + * hr_file before accept reaches FD_PREPARE(); this extra + * reference keeps the file alive until FD_PREPARE() takes + * ownership. + */ + get_file(pos->hr_file); req = pos; break; } From 5da98f55b13173c08f003011b76531b25c821c07 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:20 -0400 Subject: [PATCH 350/411] net/handshake: Close the submit-side sock_hold race handshake_req_submit() publishes the request via handshake_req_hash_add() and __add_pending_locked(), drops hn_lock, and calls handshake_genl_notify() (which can sleep) before taking sock_hold() on req->hr_sk. A fast tlshd ACCEPT followed by DONE can drive handshake_complete()'s sock_put() into the window between the spin_unlock and the late sock_hold(); on a system where the consumer's fd held the only sk reference, the late sock_hold() then operates on an sk whose refcount has reached zero. The preceding two patches install an explicit file reference on struct handshake_req. That file pins sock->file, which pins the embedded struct socket, which defers inet_release()'s sock_put(). As long as hr_file is held, sk cannot reach refcount zero from the consumer side, and the submit-side sock_hold() with its matching sock_put() calls in handshake_complete() and handshake_req_cancel() is now redundant. Drop all three. The file reference already keeps each request's socket alive, and the lifetime story is contained in a single get_file()/fput() pair. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-6-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/request.c | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/net/handshake/request.c b/net/handshake/request.c index e2d7ee7ce6e0..bd3d9467ab91 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -301,13 +301,6 @@ int handshake_req_submit(struct socket *sock, struct handshake_req *req, goto out_err; } - /* - * Pin struct sock so sk_destruct does not run until the - * handshake completion path releases it; struct socket is - * held separately via hr_file above. - */ - sock_hold(req->hr_sk); - trace_handshake_submit(net, req, req->hr_sk); return 0; @@ -337,9 +330,6 @@ void handshake_complete(struct handshake_req *req, int status, trace_handshake_complete(net, req, sk, status); req->hr_proto->hp_done(req, status, info); - /* Handshake request is no longer pending */ - sock_put(sk); - fput(file); } } @@ -387,8 +377,6 @@ bool handshake_req_cancel(struct sock *sk) out_true: trace_handshake_cancel(net, req, sk); - /* Handshake request is no longer pending */ - sock_put(sk); fput(req->hr_file); return true; } From 204a5efde5ed52932840ee1d15d3b581cfda48e2 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:21 -0400 Subject: [PATCH 351/411] net/handshake: Verify file-reference balance in submit paths The new file-reference contract on struct handshake_req is silently breakable: a missing get_file() at submit or a missing fput() on an error path leaves the file leaked but does not crash the test, so the existing absence-of-crash checks pass either way. Snapshot file_count(filp) before each handshake_req_submit() in the submit-success, EAGAIN, EBUSY, and cancel tests, and assert the expected balance after submit and again after cancel. The already-completed cancel test also asserts the post-complete balance, which pins down that handshake_complete() drops the reference and that the subsequent cancel does not double-fput. The destroy test gets the same treatment before __fput_sync(), which double-checks that cancel's fput() ran and the only remaining reference is the one sock_alloc_file() established. Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-7-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/handshake-test.c | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/net/handshake/handshake-test.c b/net/handshake/handshake-test.c index 9cc7a95f4120..3dd507470d5f 100644 --- a/net/handshake/handshake-test.c +++ b/net/handshake/handshake-test.c @@ -208,6 +208,7 @@ static void handshake_req_submit_test3(struct kunit *test) static void handshake_req_submit_test4(struct kunit *test) { struct handshake_req *req, *result; + unsigned long fcount_before; struct socket *sock; struct file *filp; int err; @@ -224,8 +225,10 @@ static void handshake_req_submit_test4(struct kunit *test) KUNIT_ASSERT_NOT_NULL(test, sock->sk); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); /* Act */ result = handshake_req_hash_lookup(sock->sk); @@ -235,11 +238,13 @@ static void handshake_req_submit_test4(struct kunit *test) KUNIT_EXPECT_PTR_EQ(test, req, result); handshake_req_cancel(sock->sk); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_submit_test5(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req; struct handshake_net *hn; struct socket *sock; @@ -265,12 +270,14 @@ static void handshake_req_submit_test5(struct kunit *test) saved = hn->hn_pending; hn->hn_pending = hn->hn_pending_max + 1; + fcount_before = file_count(filp); /* Act */ err = handshake_req_submit(sock, req, GFP_KERNEL); /* Assert */ KUNIT_EXPECT_EQ(test, err, -EAGAIN); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); hn->hn_pending = saved; @@ -279,6 +286,7 @@ static void handshake_req_submit_test5(struct kunit *test) static void handshake_req_submit_test6(struct kunit *test) { struct handshake_req *req1, *req2; + unsigned long fcount_before; struct socket *sock; struct file *filp; int err; @@ -296,21 +304,26 @@ static void handshake_req_submit_test6(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); KUNIT_ASSERT_NOT_NULL(test, sock->sk); sock->file = filp; + fcount_before = file_count(filp); /* Act */ err = handshake_req_submit(sock, req1, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); err = handshake_req_submit(sock, req2, GFP_KERNEL); /* Assert */ KUNIT_EXPECT_EQ(test, err, -EBUSY); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); handshake_req_cancel(sock->sk); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_cancel_test1(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req; struct socket *sock; struct file *filp; @@ -329,8 +342,10 @@ static void handshake_req_cancel_test1(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); /* NB: handshake_req hasn't been accepted */ @@ -339,12 +354,14 @@ static void handshake_req_cancel_test1(struct kunit *test) /* Assert */ KUNIT_EXPECT_TRUE(test, result); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_cancel_test2(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req, *next; struct handshake_net *hn; struct socket *sock; @@ -365,8 +382,10 @@ static void handshake_req_cancel_test2(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); net = sock_net(sock->sk); hn = handshake_pernet(net); @@ -385,12 +404,14 @@ static void handshake_req_cancel_test2(struct kunit *test) /* Assert */ KUNIT_EXPECT_TRUE(test, result); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } static void handshake_req_cancel_test3(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req, *next; struct handshake_net *hn; struct socket *sock; @@ -411,8 +432,10 @@ static void handshake_req_cancel_test3(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before + 1); net = sock_net(sock->sk); hn = handshake_pernet(net); @@ -428,12 +451,14 @@ static void handshake_req_cancel_test3(struct kunit *test) /* Pretend to complete this request */ handshake_complete(next, -ETIMEDOUT, NULL); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); /* Act */ result = handshake_req_cancel(sock->sk); /* Assert */ KUNIT_EXPECT_FALSE(test, result); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); fput(filp); } @@ -454,6 +479,7 @@ static struct handshake_proto handshake_req_alloc_proto_destroy = { static void handshake_req_destroy_test1(struct kunit *test) { + unsigned long fcount_before; struct handshake_req *req; struct socket *sock; struct file *filp; @@ -473,10 +499,12 @@ static void handshake_req_destroy_test1(struct kunit *test) KUNIT_ASSERT_NOT_ERR_OR_NULL(test, filp); sock->file = filp; + fcount_before = file_count(filp); err = handshake_req_submit(sock, req, GFP_KERNEL); KUNIT_ASSERT_EQ(test, err, 0); handshake_req_cancel(sock->sk); + KUNIT_EXPECT_EQ(test, file_count(filp), fcount_before); /* Act */ /* Ensure the close/release/put process has run to From ea5fe6a73ca57e5150b8a38b341aef2636eb72f0 Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Mon, 25 May 2026 12:51:22 -0400 Subject: [PATCH 352/411] net/handshake: Drain pending requests at net namespace exit The arguments to list_splice_init() in handshake_net_exit() are reversed. The call moves the local empty "requests" list onto hn->hn_requests, leaving the local list empty, so the subsequent drain loop runs zero iterations. Pending handshake requests that had not yet been accepted are not torn down when the net namespace is destroyed; each one keeps a reference on a socket file and on the handshake_req allocation. Pass the source and destination in the documented order (list_splice_init(list, head) moves list onto head) so the pending list is transferred to the local scratch list and drained through handshake_complete(). Fixing the splice direction exposes a list-corruption race. After the splice each req->hr_list still has non-empty link pointers, threading the stack-local scratch list rather than hn_requests. A concurrent handshake_req_cancel() -- for example, from sunrpc's TLS timeout on a kernel socket whose netns reference was not taken -- finds the request through the rhashtable, calls remove_pending(), and sees !list_empty(&req->hr_list). __remove_pending_locked() then list_del_init()s an entry off the scratch list while the drain iterates, corrupting it. The same call arriving after the drain loop has run list_del() on an entry hits LIST_POISON instead. Have remove_pending() check HANDSHAKE_F_NET_DRAINING under hn_lock and report not-found when drain is in progress. The drain has already taken ownership; handshake_complete()'s existing test_and_set on HANDSHAKE_F_REQ_COMPLETED still arbitrates between drain and cancel for who calls the consumer's hp_done. Use list_del_init() rather than list_del() in the drain so req->hr_list does not carry LIST_POISON after drain releases the entry. The DRAINING guard in remove_pending() makes cancel return false, but cancel still falls through to test_and_set_bit on HANDSHAKE_F_REQ_COMPLETED and drops the request's hr_file reference. Without another pin, if that is the last reference, sk_destruct frees the request while it is still linked on the drain loop's local list. Pin each request's hr_file under hn_lock before releasing the list, and drop that drain pin after the loop finishes with the request. Fixes: 3b3009ea8abb ("net/handshake: Create a NETLINK service for handling handshake requests") Signed-off-by: Chuck Lever Reviewed-by: Hannes Reinecke Link: https://patch.msgid.link/20260525-handshake-file-pin-v3-8-66c616906ead@oracle.com Signed-off-by: Paolo Abeni --- net/handshake/netlink.c | 10 ++++++++-- net/handshake/request.c | 5 ++++- 2 files changed, 12 insertions(+), 3 deletions(-) diff --git a/net/handshake/netlink.c b/net/handshake/netlink.c index 21d6cbd52fcd..3fd4fef9bab1 100644 --- a/net/handshake/netlink.c +++ b/net/handshake/netlink.c @@ -201,13 +201,19 @@ static void __net_exit handshake_net_exit(struct net *net) */ spin_lock_bh(&hn->hn_lock); set_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags); - list_splice_init(&requests, &hn->hn_requests); + list_splice_init(&hn->hn_requests, &requests); + list_for_each_entry(req, &requests, hr_list) + get_file(req->hr_file); spin_unlock_bh(&hn->hn_lock); while (!list_empty(&requests)) { + struct file *file; + req = list_first_entry(&requests, struct handshake_req, hr_list); - list_del(&req->hr_list); + file = req->hr_file; + list_del_init(&req->hr_list); handshake_complete(req, -ETIMEDOUT, NULL); + fput(file); } } diff --git a/net/handshake/request.c b/net/handshake/request.c index bd3d9467ab91..cd30d54d0501 100644 --- a/net/handshake/request.c +++ b/net/handshake/request.c @@ -163,13 +163,16 @@ static void __remove_pending_locked(struct handshake_net *hn, * otherwise %false. * * If @req was on a pending list, it has not yet been accepted. + * Returns %false when the net namespace is draining; the drain + * loop has taken ownership of the pending list. */ static bool remove_pending(struct handshake_net *hn, struct handshake_req *req) { bool ret = false; spin_lock_bh(&hn->hn_lock); - if (!list_empty(&req->hr_list)) { + if (!test_bit(HANDSHAKE_F_NET_DRAINING, &hn->hn_flags) && + !list_empty(&req->hr_list)) { __remove_pending_locked(hn, req); ret = true; } From 2e4eed207cb4ba513e4a1fdcecbb3732e98f4914 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 28 May 2026 09:53:17 +0200 Subject: [PATCH 353/411] MAINTAINERS: Add Vasant Hegde to reviewers of AMD IOMMU Vasant has a long history of providing valuable feedback and testing results for the AMD IOMMU code. Still, too often he gets not Cc'ed on code changes, so make his reviewer status official. Acked-by: Vasant Hegde Signed-off-by: Joerg Roedel --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index c2c6d79275c6..34293e53129f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1152,6 +1152,7 @@ F: drivers/platform/x86/amd/hfi/ AMD IOMMU (AMD-VI) M: Joerg Roedel R: Suravee Suthikulpanit +R: Vasant Hegde L: iommu@lists.linux.dev S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/iommu/linux.git From 00c257948900fae69dae2a055b378edf09aacf6e Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 28 May 2026 09:53:18 +0200 Subject: [PATCH 354/411] MAINTAINERS: Add my employer to my entries AMD pays for my IOMMU maintainer work, so mention that in the MAINTAINERS file as well. Signed-off-by: Joerg Roedel --- MAINTAINERS | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/MAINTAINERS b/MAINTAINERS index 34293e53129f..52e538030cf9 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1150,7 +1150,7 @@ F: Documentation/arch/x86/amd-hfi.rst F: drivers/platform/x86/amd/hfi/ AMD IOMMU (AMD-VI) -M: Joerg Roedel +M: Joerg Roedel (AMD) R: Suravee Suthikulpanit R: Vasant Hegde L: iommu@lists.linux.dev @@ -13482,7 +13482,7 @@ F: include/linux/iommu-dma.h F: include/linux/iova.h IOMMU SUBSYSTEM -M: Joerg Roedel +M: Joerg Roedel (AMD) M: Will Deacon R: Robin Murphy L: iommu@lists.linux.dev From 20040b2a3cb992f84d3db4c086b909eb9b906b31 Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Tue, 26 May 2026 09:45:23 +0200 Subject: [PATCH 355/411] dpll: export __dpll_device_change_ntf() for use under dpll_lock Export __dpll_device_change_ntf() so that drivers can send device change notifications from within device callbacks, which are already called under dpll_lock. Using dpll_device_change_ntf() in that context would deadlock. Add lockdep_assert_held() to catch misuse without the lock held. Signed-off-by: Ivan Vecera Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20260526074525.1451008-2-ivecera@redhat.com Signed-off-by: Paolo Abeni --- drivers/dpll/dpll_netlink.c | 13 +++++++++++-- include/linux/dpll.h | 1 + 2 files changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/dpll/dpll_netlink.c b/drivers/dpll/dpll_netlink.c index 0ff1658c2dc1..75e3ae0c16d0 100644 --- a/drivers/dpll/dpll_netlink.c +++ b/drivers/dpll/dpll_netlink.c @@ -829,12 +829,21 @@ int dpll_device_delete_ntf(struct dpll_device *dpll) return dpll_device_event_send(DPLL_CMD_DEVICE_DELETE_NTF, dpll); } -static int -__dpll_device_change_ntf(struct dpll_device *dpll) +/** + * __dpll_device_change_ntf - notify that the dpll device has been changed + * @dpll: registered dpll pointer + * + * Context: caller must hold dpll_lock. Suitable for use inside device + * callbacks which are already invoked under dpll_lock. + * Return: 0 if succeeds, error code otherwise. + */ +int __dpll_device_change_ntf(struct dpll_device *dpll) { + lockdep_assert_held(&dpll_lock); dpll_device_notify(dpll, DPLL_DEVICE_CHANGED); return dpll_device_event_send(DPLL_CMD_DEVICE_CHANGE_NTF, dpll); } +EXPORT_SYMBOL_GPL(__dpll_device_change_ntf); /** * dpll_device_change_ntf - notify that the dpll device has been changed diff --git a/include/linux/dpll.h b/include/linux/dpll.h index f8037f1ab20b..2dbe8567eafc 100644 --- a/include/linux/dpll.h +++ b/include/linux/dpll.h @@ -284,6 +284,7 @@ void dpll_pin_on_pin_unregister(struct dpll_pin *parent, struct dpll_pin *pin, int dpll_pin_ref_sync_pair_add(struct dpll_pin *pin, struct dpll_pin *ref_sync_pin); +int __dpll_device_change_ntf(struct dpll_device *dpll); int dpll_device_change_ntf(struct dpll_device *dpll); int __dpll_pin_change_ntf(struct dpll_pin *pin); From d733f519f6443540f8359461a34e3b0042099bbe Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Tue, 26 May 2026 09:45:24 +0200 Subject: [PATCH 356/411] dpll: zl3073x: use __dpll_device_change_ntf() and remove change_work The change_work was introduced to send device change notifications from DPLL device callbacks without deadlocking on dpll_lock, since the callbacks are already invoked under that lock. Now that __dpll_device_change_ntf() is exported for callers that already hold dpll_lock, use it directly and remove the change_work infrastructure entirely. This eliminates a race condition where change_work could be re-scheduled after cancel_work_sync() during device teardown, potentially causing the handler to dereference a freed or NULL dpll_dev pointer. Fixes: 9363b4837659 ("dpll: zl3073x: Allow to configure phase offset averaging factor") Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20260526074525.1451008-3-ivecera@redhat.com Signed-off-by: Paolo Abeni --- drivers/dpll/zl3073x/dpll.c | 26 +++++++++----------------- drivers/dpll/zl3073x/dpll.h | 2 -- 2 files changed, 9 insertions(+), 19 deletions(-) diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index 64b4e9e3e8fe..0770bd895de9 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -1079,15 +1079,6 @@ zl3073x_dpll_phase_offset_avg_factor_get(const struct dpll_device *dpll, return 0; } -static void -zl3073x_dpll_change_work(struct work_struct *work) -{ - struct zl3073x_dpll *zldpll; - - zldpll = container_of(work, struct zl3073x_dpll, change_work); - dpll_device_change_ntf(zldpll->dpll_dev); -} - static int zl3073x_dpll_phase_offset_avg_factor_set(const struct dpll_device *dpll, void *dpll_priv, u32 factor, @@ -1113,8 +1104,10 @@ zl3073x_dpll_phase_offset_avg_factor_set(const struct dpll_device *dpll, * we have to send a notification for other DPLL devices. */ list_for_each_entry(item, &zldpll->dev->dplls, list) { - if (item != zldpll) - schedule_work(&item->change_work); + struct dpll_device *dpll_dev = READ_ONCE(item->dpll_dev); + + if (item != zldpll && dpll_dev) + __dpll_device_change_ntf(dpll_dev); } return 0; @@ -1627,13 +1620,13 @@ zl3073x_dpll_device_register(struct zl3073x_dpll *zldpll) static void zl3073x_dpll_device_unregister(struct zl3073x_dpll *zldpll) { - WARN(!zldpll->dpll_dev, "DPLL device is not registered\n"); + struct dpll_device *dpll_dev = READ_ONCE(zldpll->dpll_dev); - cancel_work_sync(&zldpll->change_work); + WARN(!dpll_dev, "DPLL device is not registered\n"); - dpll_device_unregister(zldpll->dpll_dev, &zldpll->ops, zldpll); - dpll_device_put(zldpll->dpll_dev, &zldpll->tracker); - zldpll->dpll_dev = NULL; + WRITE_ONCE(zldpll->dpll_dev, NULL); + dpll_device_unregister(dpll_dev, &zldpll->ops, zldpll); + dpll_device_put(dpll_dev, &zldpll->tracker); } /** @@ -1926,7 +1919,6 @@ zl3073x_dpll_alloc(struct zl3073x_dev *zldev, u8 ch) zldpll->dev = zldev; zldpll->id = ch; INIT_LIST_HEAD(&zldpll->pins); - INIT_WORK(&zldpll->change_work, zl3073x_dpll_change_work); return zldpll; } diff --git a/drivers/dpll/zl3073x/dpll.h b/drivers/dpll/zl3073x/dpll.h index 434c32a7db12..c8bc8437a709 100644 --- a/drivers/dpll/zl3073x/dpll.h +++ b/drivers/dpll/zl3073x/dpll.h @@ -21,7 +21,6 @@ * @tracker: tracking object for the acquired reference * @lock_status: last saved DPLL lock status * @pins: list of pins - * @change_work: device change notification work */ struct zl3073x_dpll { struct list_head list; @@ -35,7 +34,6 @@ struct zl3073x_dpll { dpll_tracker tracker; enum dpll_lock_status lock_status; struct list_head pins; - struct work_struct change_work; }; struct zl3073x_dpll *zl3073x_dpll_alloc(struct zl3073x_dev *zldev, u8 ch); From c1224569cef038b040db0459510cd7948ecd467b Mon Sep 17 00:00:00 2001 From: Ivan Vecera Date: Tue, 26 May 2026 09:45:25 +0200 Subject: [PATCH 357/411] dpll: zl3073x: make frequency monitor a per-device attribute The frequency monitoring feature uses shared hardware registers that measure input reference frequencies independently of individual DPLL channels. However, the freq_monitor flag was incorrectly placed in the per-DPLL structure, causing each channel to track its own enable/disable state independently. Since the DPLL core calls measured_freq_get() only for the first pin registration, the measured_freq_check() in the periodic worker was gated by the per-DPLL freq_monitor flag of whichever channel happens to be checked. If the first DPLL channel had frequency monitoring disabled while another had it enabled, measurements were never reported. Move freq_monitor from struct zl3073x_dpll to struct zl3073x_dev so all DPLL channels share a single flag, matching the hardware behavior. Update freq_monitor_set() to notify other DPLL devices about the change (like phase_offset_avg_factor_set() already does) and remove the mode-dependent guard in zl3073x_dpll_changes_check() since all input pin monitoring (pin state, phase offset, FFO, and measured frequency) works correctly in all DPLL modes. Fixes: bfc923b642874 ("dpll: zl3073x: implement frequency monitoring") Signed-off-by: Ivan Vecera Link: https://patch.msgid.link/20260526074525.1451008-4-ivecera@redhat.com Signed-off-by: Paolo Abeni --- drivers/dpll/zl3073x/core.c | 19 ++++++++----------- drivers/dpll/zl3073x/core.h | 4 +++- drivers/dpll/zl3073x/dpll.c | 29 ++++++++++++++--------------- drivers/dpll/zl3073x/dpll.h | 2 -- 4 files changed, 25 insertions(+), 29 deletions(-) diff --git a/drivers/dpll/zl3073x/core.c b/drivers/dpll/zl3073x/core.c index 5f1e70f3e40a..0a133b0f2d97 100644 --- a/drivers/dpll/zl3073x/core.c +++ b/drivers/dpll/zl3073x/core.c @@ -762,18 +762,15 @@ zl3073x_dev_periodic_work(struct kthread_work *work) dev_warn(zldev->dev, "Failed to update phase offsets: %pe\n", ERR_PTR(rc)); - /* Update measured input reference frequencies if any DPLL has - * frequency monitoring enabled. + /* Update measured input reference frequencies if frequency + * monitoring is enabled. */ - list_for_each_entry(zldpll, &zldev->dplls, list) { - if (zldpll->freq_monitor) { - rc = zl3073x_ref_freq_meas_update(zldev); - if (rc) - dev_warn(zldev->dev, - "Failed to update measured frequencies: %pe\n", - ERR_PTR(rc)); - break; - } + if (zldev->freq_monitor) { + rc = zl3073x_ref_freq_meas_update(zldev); + if (rc) + dev_warn(zldev->dev, + "Failed to update measured frequencies: %pe\n", + ERR_PTR(rc)); } /* Update references' fractional frequency offsets */ diff --git a/drivers/dpll/zl3073x/core.h b/drivers/dpll/zl3073x/core.h index 99440620407d..addba378b0df 100644 --- a/drivers/dpll/zl3073x/core.h +++ b/drivers/dpll/zl3073x/core.h @@ -57,6 +57,7 @@ struct zl3073x_chip_info { * @work: periodic work * @clock_id: clock id of the device * @phase_avg_factor: phase offset measurement averaging factor + * @freq_monitor: is frequency monitor enabled */ struct zl3073x_dev { struct device *dev; @@ -77,9 +78,10 @@ struct zl3073x_dev { struct kthread_worker *kworker; struct kthread_delayed_work work; - /* Devlink parameters */ + /* Per-chip parameters */ u64 clock_id; u8 phase_avg_factor; + bool freq_monitor; }; extern const struct regmap_config zl3073x_regmap_config; diff --git a/drivers/dpll/zl3073x/dpll.c b/drivers/dpll/zl3073x/dpll.c index 0770bd895de9..0bfcbae2109f 100644 --- a/drivers/dpll/zl3073x/dpll.c +++ b/drivers/dpll/zl3073x/dpll.c @@ -1212,7 +1212,7 @@ zl3073x_dpll_freq_monitor_get(const struct dpll_device *dpll, { struct zl3073x_dpll *zldpll = dpll_priv; - if (zldpll->freq_monitor) + if (zldpll->dev->freq_monitor) *state = DPLL_FEATURE_STATE_ENABLE; else *state = DPLL_FEATURE_STATE_DISABLE; @@ -1226,9 +1226,19 @@ zl3073x_dpll_freq_monitor_set(const struct dpll_device *dpll, enum dpll_feature_state state, struct netlink_ext_ack *extack) { - struct zl3073x_dpll *zldpll = dpll_priv; + struct zl3073x_dpll *item, *zldpll = dpll_priv; - zldpll->freq_monitor = (state == DPLL_FEATURE_STATE_ENABLE); + zldpll->dev->freq_monitor = (state == DPLL_FEATURE_STATE_ENABLE); + + /* The frequency monitoring is common for all DPLL channels so after + * change we have to send a notification for other DPLL devices. + */ + list_for_each_entry(item, &zldpll->dev->dplls, list) { + struct dpll_device *dpll_dev = READ_ONCE(item->dpll_dev); + + if (item != zldpll && dpll_dev) + __dpll_device_change_ntf(dpll_dev); + } return 0; } @@ -1745,7 +1755,7 @@ zl3073x_dpll_pin_measured_freq_check(struct zl3073x_dpll_pin *pin) u8 ref_id; u32 freq; - if (!zldpll->freq_monitor) + if (!zldpll->dev->freq_monitor) return false; ref_id = zl3073x_input_pin_ref_get(pin->id); @@ -1778,10 +1788,8 @@ zl3073x_dpll_changes_check(struct zl3073x_dpll *zldpll) struct zl3073x_dev *zldev = zldpll->dev; enum dpll_lock_status lock_status; struct device *dev = zldev->dev; - const struct zl3073x_chan *chan; struct zl3073x_dpll_pin *pin; int rc; - u8 mode; zldpll->check_count++; @@ -1800,15 +1808,6 @@ zl3073x_dpll_changes_check(struct zl3073x_dpll *zldpll) dpll_device_change_ntf(zldpll->dpll_dev); } - /* Input pin monitoring does make sense only in automatic - * or forced reference modes. - */ - chan = zl3073x_chan_state_get(zldev, zldpll->id); - mode = zl3073x_chan_mode_get(chan); - if (mode != ZL_DPLL_MODE_REFSEL_MODE_AUTO && - mode != ZL_DPLL_MODE_REFSEL_MODE_REFLOCK) - return; - /* Update phase offset latch registers for this DPLL if the phase * offset monitor feature is enabled. */ diff --git a/drivers/dpll/zl3073x/dpll.h b/drivers/dpll/zl3073x/dpll.h index c8bc8437a709..21adcc18e45e 100644 --- a/drivers/dpll/zl3073x/dpll.h +++ b/drivers/dpll/zl3073x/dpll.h @@ -15,7 +15,6 @@ * @id: DPLL index * @check_count: periodic check counter * @phase_monitor: is phase offset monitor enabled - * @freq_monitor: is frequency monitor enabled * @ops: DPLL device operations for this instance * @dpll_dev: pointer to registered DPLL device * @tracker: tracking object for the acquired reference @@ -28,7 +27,6 @@ struct zl3073x_dpll { u8 id; u8 check_count; bool phase_monitor; - bool freq_monitor; struct dpll_device_ops ops; struct dpll_device *dpll_dev; dpll_tracker tracker; From 79378db6a86c7014cce40b65252e6c18f5b8bcc2 Mon Sep 17 00:00:00 2001 From: Santhosh Kumar K Date: Wed, 27 May 2026 23:07:36 +0530 Subject: [PATCH 358/411] spi: spi-mem: avoid mutating op template in spi_mem_supports_op() spi_mem_supports_op() accepts a const struct spi_mem_op pointer but casts away const internally to call spi_mem_adjust_op_freq(). This mutates the caller's op template, which causes stale max_freq values when callers reuse persistent templates - subsequent calls won't re-apply the device frequency cap since spi_mem_adjust_op_freq() skips non-zero values. Fix by operating on a stack-local copy instead. Fixes: a4f8e70d75dd ("spi: spi-mem: add spi_mem_adjust_op_freq() in spi_mem_supports_op()") Cc: Tianyu Xu Cc: stable@vger.kernel.org Signed-off-by: Santhosh Kumar K Reviewed-by: Miquel Raynal Link: https://patch.msgid.link/20260527173736.2243004-1-s-k6@ti.com Signed-off-by: Mark Brown --- drivers/spi/spi-mem.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) diff --git a/drivers/spi/spi-mem.c b/drivers/spi/spi-mem.c index a09371a075d2..93266848c6df 100644 --- a/drivers/spi/spi-mem.c +++ b/drivers/spi/spi-mem.c @@ -279,13 +279,20 @@ static bool spi_mem_internal_supports_op(struct spi_mem *mem, */ bool spi_mem_supports_op(struct spi_mem *mem, const struct spi_mem_op *op) { - /* Make sure the operation frequency is correct before going futher */ - spi_mem_adjust_op_freq(mem, (struct spi_mem_op *)op); + struct spi_mem_op eval_op = *op; - if (spi_mem_check_op(op)) + /* + * Work on a local copy; this is a pure capability check and must + * not modify the caller's op. Stored templates with max_freq == 0 + * must remain unset so their frequency is always re-capped to the + * current device maximum at execution time. + */ + spi_mem_adjust_op_freq(mem, &eval_op); + + if (spi_mem_check_op(&eval_op)) return false; - return spi_mem_internal_supports_op(mem, op); + return spi_mem_internal_supports_op(mem, &eval_op); } EXPORT_SYMBOL_GPL(spi_mem_supports_op); From 00e1950716c6ed67d74777b2db286b0fa23b4be9 Mon Sep 17 00:00:00 2001 From: Zhenghang Xiao Date: Tue, 26 May 2026 18:51:52 +0800 Subject: [PATCH 359/411] Bluetooth: l2cap: clear chan->ident on ECRED reconfiguration success l2cap_ecred_reconf_rsp() returns early on success without clearing chan->ident. Every other L2CAP response handler (l2cap_ecred_conn_rsp, l2cap_le_connect_rsp, l2cap_config_rsp) clears chan->ident after a successful transaction to prevent the channel from matching subsequent responses with the recycled ident value. A remote attacker that completed a reconfiguration as the peer can replay a failure response with the stale ident, causing the kernel to match and destroy the already-established channel via l2cap_chan_del(chan, ECONNRESET). Clear chan->ident for all matching channels on success, and harden the failure path by using l2cap_chan_hold_unless_zero() consistent with other L2CAP handlers (l2cap_le_command_rej, __l2cap_get_chan_by_ident). Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Zhenghang Xiao Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 5668c92b3f58..ff13c43e3588 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5460,14 +5460,20 @@ static inline int l2cap_ecred_reconf_rsp(struct l2cap_conn *conn, BT_DBG("result 0x%4.4x", result); - if (!result) + if (!result) { + list_for_each_entry(chan, &conn->chan_l, list) { + if (chan->ident == cmd->ident) + chan->ident = 0; + } return 0; + } list_for_each_entry_safe(chan, tmp, &conn->chan_l, list) { if (chan->ident != cmd->ident) continue; - l2cap_chan_hold(chan); + if (!l2cap_chan_hold_unless_zero(chan)) + continue; l2cap_chan_lock(chan); l2cap_chan_del(chan, ECONNRESET); From 41c2713b204e6cb6a94587bc6bf6935107df5479 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 11 May 2026 12:09:42 -0400 Subject: [PATCH 360/411] Bluetooth: L2CAP: Fix possible crash on l2cap_ecred_conn_rsp If dcid is received for an already-assigned destination CID the spec requires that both channels to be discarded, but calling l2cap_chan_del may invalidate the tmp cursor created by list_for_each_entry_safe and in fact it is the wrong procedure as the chan->dcid may be assigned previously it really needs to be disconnected. Calling l2cap_chan_clone directly may still lead to l2cap_chan_del so instead schedule l2cap_chan_timeout with delay 0 to close the channel asynchronously. Fixes: 15f02b910562 ("Bluetooth: L2CAP: Add initial code for Enhanced Credit Based Mode") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 27 ++++++++++++++++++++++----- 1 file changed, 22 insertions(+), 5 deletions(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index ff13c43e3588..45b175399e8d 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5262,6 +5262,7 @@ static inline int l2cap_ecred_conn_rsp(struct l2cap_conn *conn, cmd_len -= sizeof(*rsp); list_for_each_entry_safe(chan, tmp, &conn->chan_l, list) { + struct l2cap_chan *orig; u16 dcid; if (chan->ident != cmd->ident || @@ -5283,8 +5284,10 @@ static inline int l2cap_ecred_conn_rsp(struct l2cap_conn *conn, BT_DBG("dcid[%d] 0x%4.4x", i, dcid); + orig = __l2cap_get_chan_by_dcid(conn, dcid); + /* Check if dcid is already in use */ - if (dcid && __l2cap_get_chan_by_dcid(conn, dcid)) { + if (dcid && orig) { /* If a device receives a * L2CAP_CREDIT_BASED_CONNECTION_RSP packet with an * already-assigned Destination CID, then both the @@ -5293,10 +5296,24 @@ static inline int l2cap_ecred_conn_rsp(struct l2cap_conn *conn, */ l2cap_chan_del(chan, ECONNREFUSED); l2cap_chan_unlock(chan); - chan = __l2cap_get_chan_by_dcid(conn, dcid); - l2cap_chan_lock(chan); - l2cap_chan_del(chan, ECONNRESET); - l2cap_chan_unlock(chan); + + /* Check that the dcid channel mode is + * L2CAP_MODE_EXT_FLOWCTL since this procedure is only + * valid for that mode and shouldn't disconnect a dcid + * in other modes. + */ + if (orig->mode == L2CAP_MODE_EXT_FLOWCTL) { + l2cap_chan_lock(orig); + /* Disconnect the original channel as it may be + * considered connected since dcid has already + * been assigned; don't call l2cap_chan_close + * directly since that could lead to + * l2cap_chan_del and then removing the channel + * from the list while we're iterating over it. + */ + __set_chan_timer(orig, 0); + l2cap_chan_unlock(orig); + } continue; } From 47f23a259517abbdb8032c057a1e8a6bf3734878 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Wed, 27 May 2026 04:59:17 +0000 Subject: [PATCH 361/411] Bluetooth: ISO: fix UAF in iso_recv_frame iso_recv_frame reads conn->sk under iso_conn_lock but releases the lock before using sk, with no reference held. A concurrent iso_sock_kill() can free sk in that window, causing use-after-free on sk->sk_state and sock_queue_rcv_skb(). Fix by replacing the bare pointer read with iso_sock_hold(conn), which calls sock_hold() while the spinlock is held, atomically elevating the refcount before the lock drops. Add a drop_put label so sock_put() is called on all exit paths where the hold succeeded. Fixes: ccf74f2390d60a2f9a75ef496d2564abb478f46a ("Bluetooth: Add BTPROTO_ISO socket type") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index d7af617cda45..f03b7fa5dccc 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -564,7 +564,7 @@ static void iso_recv_frame(struct iso_conn *conn, struct sk_buff *skb) struct sock *sk; iso_conn_lock(conn); - sk = conn->sk; + sk = iso_sock_hold(conn); iso_conn_unlock(conn); if (!sk) @@ -573,11 +573,15 @@ static void iso_recv_frame(struct iso_conn *conn, struct sk_buff *skb) BT_DBG("sk %p len %d", sk, skb->len); if (sk->sk_state != BT_CONNECTED) - goto drop; + goto drop_put; - if (!sock_queue_rcv_skb(sk, skb)) + if (!sock_queue_rcv_skb(sk, skb)) { + sock_put(sk); return; + } +drop_put: + sock_put(sk); drop: kfree_skb(skb); } From 4b5f8e608749b7e8fa386c6e4301cf9272595859 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Wed, 27 May 2026 04:59:18 +0000 Subject: [PATCH 362/411] Bluetooth: ISO: serialize iso_sock_clear_timer with socket lock iso_sock_close() calls iso_sock_clear_timer() before acquiring lock_sock(sk). iso_sock_clear_timer() reads iso_pi(sk)->conn twice without the socket lock held: if (!iso_pi(sk)->conn) return; cancel_delayed_work(&iso_pi(sk)->conn->timeout_work); Concurrently, iso_conn_del() executes under lock_sock(sk) and calls iso_chan_del(), which sets iso_pi(sk)->conn to NULL and may result in the final reference to the connection being dropped: CPU0 CPU1 ---- ---- iso_sock_clear_timer() if (conn != NULL) ... lock_sock(sk) iso_chan_del() iso_pi(sk)->conn = NULL cancel_delayed_work(conn) /* NULL deref or UAF */ iso_pi(sk)->conn is not stable across the unlock window, causing a NULL pointer dereference or use-after-free. Serialize iso_sock_clear_timer() with the socket lock by moving it inside lock_sock()/release_sock(), matching the pattern used in iso_conn_del() and all other call sites. Fixes: ccf74f2390d60a2f9a75ef496d2564abb478f46a ("Bluetooth: Add BTPROTO_ISO socket type") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index f03b7fa5dccc..876649556d3c 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -864,8 +864,8 @@ static void __iso_sock_close(struct sock *sk) /* Must be called on unlocked socket. */ static void iso_sock_close(struct sock *sk) { - iso_sock_clear_timer(sk); lock_sock(sk); + iso_sock_clear_timer(sk); __iso_sock_close(sk); release_sock(sk); iso_sock_kill(sk); From 40b87657200cfae93e48904fd9c9c8fc3e192cae Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 26 May 2026 10:50:57 -0300 Subject: [PATCH 363/411] Bluetooth: hci_core: Rework hci_dev_do_reset() to use hci_sync functions The current HCI reset function in hci_core.c duplicates most of the work done by hci_dev_close_sync(), and doesn't handle LE, advertising or discovery. Instead of porting these to hci_dev_do_reset(), directly call the close/open functions from hci_sync to reset the hdev. MGMT now notifies when a user performs a reset. Suggested-by: Luiz Augusto von Dentz Signed-off-by: Heitor Alves de Siqueira Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_core.c | 43 +++------------------------------------- 1 file changed, 3 insertions(+), 40 deletions(-) diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index c46c1236ebfa..28d7929dc593 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -539,46 +539,9 @@ static int hci_dev_do_reset(struct hci_dev *hdev) hci_req_sync_lock(hdev); - /* Drop queues */ - skb_queue_purge(&hdev->rx_q); - skb_queue_purge(&hdev->cmd_q); - - /* Cancel these to avoid queueing non-chained pending work */ - hci_dev_set_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); - /* Wait for - * - * if (!hci_dev_test_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE)) - * queue_delayed_work(&hdev->{cmd,ncmd}_timer) - * - * inside RCU section to see the flag or complete scheduling. - */ - synchronize_rcu(); - /* Explicitly cancel works in case scheduled after setting the flag. */ - cancel_delayed_work(&hdev->cmd_timer); - cancel_delayed_work(&hdev->ncmd_timer); - - /* Avoid potential lockdep warnings from the *_flush() calls by - * ensuring the workqueue is empty up front. - */ - drain_workqueue(hdev->workqueue); - - hci_dev_lock(hdev); - hci_inquiry_cache_flush(hdev); - hci_conn_hash_flush(hdev); - hci_dev_unlock(hdev); - - if (hdev->flush) - hdev->flush(hdev); - - hci_dev_clear_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); - - atomic_set(&hdev->cmd_cnt, 1); - hdev->acl_cnt = 0; - hdev->sco_cnt = 0; - hdev->le_cnt = 0; - hdev->iso_cnt = 0; - - ret = hci_reset_sync(hdev); + ret = hci_dev_close_sync(hdev); + if (!ret) + ret = hci_dev_open_sync(hdev); hci_req_sync_unlock(hdev); return ret; From 525daaea459fc215f432de1b8debbd9144bf97b0 Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 26 May 2026 10:50:58 -0300 Subject: [PATCH 364/411] Bluetooth: hci_sync: Set HCI_CMD_DRAIN_WORKQUEUE during device close Since hci_dev_close_sync() can now be called during the reset path, we should also set HCI_CMD_DRAIN_WORKQUEUE. This avoids queuing timeouts while the hdev workqueue is being drained. Fixes: 877afadad2dc ("Bluetooth: When HCI work queue is drained, only queue chained work") Signed-off-by: Heitor Alves de Siqueira Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 1faf8df6d159..0f016d269c62 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5301,6 +5301,12 @@ int hci_dev_close_sync(struct hci_dev *hdev) bt_dev_dbg(hdev, ""); + /* Set HCI_DRAIN_WORKQUEUE flag to prevent queuing work during + * reset/close. See hci_cmd_work() and handle_cmd_cnt_and_timer(). + */ + hci_dev_set_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); + synchronize_rcu(); + if (hci_dev_test_flag(hdev, HCI_UNREGISTER)) { disable_delayed_work(&hdev->power_off); disable_delayed_work(&hdev->ncmd_timer); @@ -5324,6 +5330,7 @@ int hci_dev_close_sync(struct hci_dev *hdev) if (!test_and_clear_bit(HCI_UP, &hdev->flags)) { cancel_delayed_work_sync(&hdev->cmd_timer); + hci_dev_clear_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); return err; } @@ -5423,6 +5430,7 @@ int hci_dev_close_sync(struct hci_dev *hdev) /* Clear flags */ hdev->flags &= BIT(HCI_RAW); hci_dev_clear_volatile_flags(hdev); + hci_dev_clear_flag(hdev, HCI_CMD_DRAIN_WORKQUEUE); memset(hdev->eir, 0, sizeof(hdev->eir)); memset(hdev->dev_class, 0, sizeof(hdev->dev_class)); From cdf88b35e06f1b385f7f6228060ae541d44fbb72 Mon Sep 17 00:00:00 2001 From: Heitor Alves de Siqueira Date: Tue, 26 May 2026 10:50:59 -0300 Subject: [PATCH 365/411] Bluetooth: hci_sync: Reset device counters in hci_dev_close_sync() Before resetting or closing the device, protocol counters should also be zeroed. Fixes: d0b137062b2d ("Bluetooth: hci_sync: Rework init stages") Signed-off-by: Heitor Alves de Siqueira Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 0f016d269c62..aeccd8084cba 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -5393,6 +5393,10 @@ int hci_dev_close_sync(struct hci_dev *hdev) /* Reset device */ skb_queue_purge(&hdev->cmd_q); atomic_set(&hdev->cmd_cnt, 1); + hdev->acl_cnt = 0; + hdev->sco_cnt = 0; + hdev->le_cnt = 0; + hdev->iso_cnt = 0; if (hci_test_quirk(hdev, HCI_QUIRK_RESET_ON_CLOSE) && !auto_off && !hci_dev_test_flag(hdev, HCI_UNCONFIGURED)) { set_bit(HCI_INIT, &hdev->flags); From bbec30f7e19d9a1c604da7164b8057ccee590e72 Mon Sep 17 00:00:00 2001 From: Bartosz Golaszewski Date: Fri, 22 May 2026 09:49:35 +0200 Subject: [PATCH 366/411] gpio: shared: undo the vote of the proxy on GPIO free When the user of a shared GPIO managed by gpio-shared-proxy calls gpiod_put() to release it, we never undo the potential "vote" for driving the shared line "high". In the free() callback, check if this proxy voted for "high" and - if so - decrease the number of votes and potentially revert the value to low if this is the last user. Cc: stable@vger.kernel.org Fixes: e992d54c6f97 ("gpio: shared-proxy: implement the shared GPIO proxy driver") Closes: https://sashiko.dev/#/patchset/20260513-gpio-shared-dynamic-voting-v1-1-8e1c49961b7d%40oss.qualcomm.com Reviewed-by: Linus Walleij Link: https://patch.msgid.link/20260522-gpio-shared-free-vote-v3-1-8a4fddc6bedb@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-shared-proxy.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/gpio/gpio-shared-proxy.c b/drivers/gpio/gpio-shared-proxy.c index 29d7d2e4dfc0..6941e4be6cf1 100644 --- a/drivers/gpio/gpio-shared-proxy.c +++ b/drivers/gpio/gpio-shared-proxy.c @@ -103,9 +103,18 @@ static void gpio_shared_proxy_free(struct gpio_chip *gc, unsigned int offset) { struct gpio_shared_proxy_data *proxy = gpiochip_get_data(gc); struct gpio_shared_desc *shared_desc = proxy->shared_desc; + int ret; guard(gpio_shared_desc_lock)(shared_desc); + if (proxy->voted_high) { + ret = gpio_shared_proxy_set_unlocked(proxy, + shared_desc->can_sleep ? gpiod_set_value_cansleep : gpiod_set_value, 0); + if (ret) + dev_err(proxy->dev, + "Failed to unset the shared GPIO value on release: %d\n", ret); + } + proxy->shared_desc->usecnt--; dev_dbg(proxy->dev, "Shared GPIO freed, number of users: %u\n", From a5c627d90809b793fc053849b3a00609db305776 Mon Sep 17 00:00:00 2001 From: Bartosz Golaszewski Date: Fri, 22 May 2026 09:35:27 +0200 Subject: [PATCH 367/411] gpio: adnp: fix flow control regression caused by scoped_guard() scoped_guard() is implemented as a for loop. Using it to protect code using the continue statement changes the flow as we now only break out of the hidden loop inside scoped_guard(), not the original for loop. Use a regular code block instead. Fixes: c7fe19ed3973 ("gpio: adnp: use lock guards for the I2C lock") Reported-by: David Lechner Closes: https://lore.kernel.org/all/cde2abb2-4cc8-4fc9-b34a-0c5d2b95779f@baylibre.com/ Reviewed-by: Linus Walleij Link: https://patch.msgid.link/20260522073527.9812-1-bartosz.golaszewski@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-adnp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpio/gpio-adnp.c b/drivers/gpio/gpio-adnp.c index e5ac2d211013..fe5bcaa90496 100644 --- a/drivers/gpio/gpio-adnp.c +++ b/drivers/gpio/gpio-adnp.c @@ -237,7 +237,9 @@ static irqreturn_t adnp_irq(int irq, void *data) unsigned long pending; int err; - scoped_guard(mutex, &adnp->i2c_lock) { + { + guard(mutex)(&adnp->i2c_lock); + err = adnp_read(adnp, GPIO_PLR(adnp) + i, &level); if (err < 0) continue; From a1b836607304f71051f9f9dcccf8b5097b86a1fb Mon Sep 17 00:00:00 2001 From: Bartosz Golaszewski Date: Fri, 22 May 2026 11:12:36 +0200 Subject: [PATCH 368/411] gpio: shared: fix deadlock on shared proxy's parent removal Commit 710abda58055 ("gpio: shared: call gpio_chip::of_xlate() if set") used the mutex embedded in struct gpio_shared_entry to protect the offset field which now can be modified after assignment. The critical section however is too wide and introduced a potential deadlock on the removal of the shared GPIO proxy's parent. Make the critical section shorter - only protect the offset when it's being read. While at it: mention the fact that the entry lock is now also used to protect against concurrent access to the offset field in the structure's documentation. Cc: stable@vger.kernel.org Fixes: 710abda58055 ("gpio: shared: call gpio_chip::of_xlate() if set") Reviewed-by: Linus Walleij Link: https://patch.msgid.link/20260522-gpio-shared-deadlock-v1-1-76bca088f8c0@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpiolib-shared.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/drivers/gpio/gpiolib-shared.c b/drivers/gpio/gpiolib-shared.c index e02d6b93a4ab..087b64c06c9f 100644 --- a/drivers/gpio/gpiolib-shared.c +++ b/drivers/gpio/gpiolib-shared.c @@ -53,7 +53,7 @@ struct gpio_shared_entry { unsigned int offset; /* Index in the property value array. */ size_t index; - /* Synchronizes the modification of shared_desc. */ + /* Synchronizes the modification of shared_desc and offset. */ struct mutex lock; struct gpio_shared_desc *shared_desc; struct kref ref; @@ -598,12 +598,11 @@ void gpio_device_teardown_shared(struct gpio_device *gdev) struct gpio_shared_ref *ref; list_for_each_entry(entry, &gpio_shared_list, list) { - guard(mutex)(&entry->lock); - if (!device_match_fwnode(&gdev->dev, entry->fwnode)) continue; - gpiod_free_commit(&gdev->descs[entry->offset]); + scoped_guard(mutex, &entry->lock) + gpiod_free_commit(&gdev->descs[entry->offset]); list_for_each_entry(ref, &entry->refs, list) { guard(mutex)(&ref->lock); From 9d7697fabbc72428f981c01ddbe0a6be0ce8b6fa Mon Sep 17 00:00:00 2001 From: Bartosz Golaszewski Date: Fri, 22 May 2026 11:12:37 +0200 Subject: [PATCH 369/411] gpio: shared: fix lockdep false positive by removing unneeded lock By the time gpio_device_teardown_shared() is called, the parent device is gone from the global list of GPIO devices and all outstanding SRCU read-side critical sections have completed. That means that no concurrent gpio_find_and_request() can call gpio_shared_add_proxy_lookup() for this device at this time. There's also no risk of the parent device being re-bound to the driver before the unbinding completes (including the child devices). Lockdep produces a false-positive report about a possible circular dependency as it doesn't know the ordering guarantee. Not taking the ref->lock in gpio_device_teardown_shared() silences it and is safe to do. Cc: stable@vger.kernel.org Fixes: ea513dd3c066 ("gpio: shared: make locking more fine-grained") Reviewed-by: Linus Walleij Link: https://patch.msgid.link/20260522-gpio-shared-deadlock-v1-2-76bca088f8c0@oss.qualcomm.com Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpiolib-shared.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/gpio/gpiolib-shared.c b/drivers/gpio/gpiolib-shared.c index 087b64c06c9f..de72776fb154 100644 --- a/drivers/gpio/gpiolib-shared.c +++ b/drivers/gpio/gpiolib-shared.c @@ -605,8 +605,6 @@ void gpio_device_teardown_shared(struct gpio_device *gdev) gpiod_free_commit(&gdev->descs[entry->offset]); list_for_each_entry(ref, &entry->refs, list) { - guard(mutex)(&ref->lock); - if (ref->lookup) { gpiod_remove_lookup_table(ref->lookup); kfree(ref->lookup->table[0].key); From 8a122b5e72cc0043705f0d524bcd15f0c0b3ec15 Mon Sep 17 00:00:00 2001 From: Dan Carpenter Date: Mon, 25 May 2026 10:15:16 +0300 Subject: [PATCH 370/411] gpio: virtuser: Fix uninitialized data bug in gpio_virtuser_direction_do_write() If *ppos is non-zero (user-space write split over multiple calls to write()) then simple_write_to_buffer() won't initialize the start of the buffer. Really, non-zero values for *ppos aren't going to work at all. Check for that and return -EINVAL at the start of the function. Fixes: 91581c4b3f29 ("gpio: virtuser: new virtual testing driver for the GPIO API") Signed-off-by: Dan Carpenter Link: https://patch.msgid.link/ahP3BJWWy-m_qI0X@stanley.mountain Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-virtuser.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpio/gpio-virtuser.c b/drivers/gpio/gpio-virtuser.c index 128520d340d4..846f8688fec5 100644 --- a/drivers/gpio/gpio-virtuser.c +++ b/drivers/gpio/gpio-virtuser.c @@ -397,7 +397,7 @@ static ssize_t gpio_virtuser_direction_do_write(struct file *file, char buf[32], *trimmed; int ret, dir, val = 0; - if (count >= sizeof(buf)) + if (*ppos != 0 || count >= sizeof(buf)) return -EINVAL; ret = simple_write_to_buffer(buf, sizeof(buf) - 1, ppos, user_buf, count); @@ -622,7 +622,7 @@ static ssize_t gpio_virtuser_consumer_write(struct file *file, char buf[GPIO_VIRTUSER_NAME_BUF_LEN + 2]; int ret; - if (count >= sizeof(buf)) + if (*ppos != 0 || count >= sizeof(buf)) return -EINVAL; ret = simple_write_to_buffer(buf, GPIO_VIRTUSER_NAME_BUF_LEN, ppos, From 3e46c18d5d87f063a93ae0fe7662fbf6660459d5 Mon Sep 17 00:00:00 2001 From: Marco Scardovi Date: Tue, 26 May 2026 19:02:45 +0200 Subject: [PATCH 371/411] gpio: rockchip: convert bank->clk to devm_clk_get_enabled() The bank->clk was previously obtained via of_clk_get() and manually prepared/enabled. However, it was missing a corresponding clk_put() in both the error paths and the remove function, leading to a reference leak. Convert the allocation to devm_clk_get_enabled(), which also properly propagates failures from clk_prepare_enable() that were previously ignored. The GPIO bank device uses the same OF node as the previous of_clk_get() call, so devm_clk_get_enabled(dev, NULL) correctly resolves the same clock provider entry. Fix the reference leak and simplify the code by removing the manual clk_disable_unprepare() calls in the probe error paths and in the remove function. Fixes: 936ee2675eee ("gpio/rockchip: add driver for rockchip gpio") Assisted-by: Antigravity:gemini-3.5-flash Signed-off-by: Marco Scardovi Link: https://patch.msgid.link/20260526171050.12785-2-scardracs@disroot.org Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-rockchip.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/drivers/gpio/gpio-rockchip.c b/drivers/gpio/gpio-rockchip.c index 44d7ebd12724..33580093a4e7 100644 --- a/drivers/gpio/gpio-rockchip.c +++ b/drivers/gpio/gpio-rockchip.c @@ -656,11 +656,10 @@ static int rockchip_get_bank_data(struct rockchip_pin_bank *bank) if (!bank->irq) return -EINVAL; - bank->clk = of_clk_get(bank->of_node, 0); + bank->clk = devm_clk_get_enabled(bank->dev, NULL); if (IS_ERR(bank->clk)) return PTR_ERR(bank->clk); - clk_prepare_enable(bank->clk); id = readl(bank->reg_base + gpio_regs_v2.version_id); switch (id) { @@ -672,7 +671,6 @@ static int rockchip_get_bank_data(struct rockchip_pin_bank *bank) bank->db_clk = of_clk_get(bank->of_node, 1); if (IS_ERR(bank->db_clk)) { dev_err(bank->dev, "cannot find debounce clk\n"); - clk_disable_unprepare(bank->clk); return -EINVAL; } break; @@ -751,7 +749,6 @@ static int rockchip_gpio_probe(struct platform_device *pdev) ret = rockchip_gpiolib_register(bank); if (ret) { - clk_disable_unprepare(bank->clk); mutex_unlock(&bank->deferred_lock); return ret; } @@ -792,7 +789,6 @@ static void rockchip_gpio_remove(struct platform_device *pdev) { struct rockchip_pin_bank *bank = platform_get_drvdata(pdev); - clk_disable_unprepare(bank->clk); gpiochip_remove(&bank->gpio_chip); } From 9500077678230e36d22bf16d2b9539c13e59a801 Mon Sep 17 00:00:00 2001 From: Marco Scardovi Date: Tue, 26 May 2026 19:02:46 +0200 Subject: [PATCH 372/411] gpio: rockchip: teardown bugs and resource leaks Address several teardown issues and resource leaks in the driver's remove path and error handling: 1. Debounce clock reference leak: The debounce clock (bank->db_clk) is obtained using of_clk_get() which increments the clock's reference count, but clk_put() is never called. Register a devm action to cleanly release it on unbind. Note that of_clk_get(..., 1) remains necessary over devm_clk_get() because the DT binding does not define clock-names, precluding name-based lookup. 2. Unregistered chained IRQ handler: The chained IRQ handler is not disconnected in remove(). If a stray interrupt fires after the driver is removed, the kernel attempts to execute a stale handler, leading to a panic. Fix this by clearing the handler in remove(). 3. IRQ domain leak: The linear IRQ domain and its generic chips are allocated manually during probe but never removed. Remove the IRQ domain during driver teardown to free the associated generic chips and mappings. Fixes: 936ee2675eee ("gpio/rockchip: add driver for rockchip gpio") Assisted-by: Antigravity:gemini-3.5-flash Signed-off-by: Marco Scardovi Link: https://patch.msgid.link/20260526171050.12785-3-scardracs@disroot.org [Bartosz: don't emit an error message on devres allocation failure] Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-rockchip.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/gpio/gpio-rockchip.c b/drivers/gpio/gpio-rockchip.c index 33580093a4e7..bc97d5d5d329 100644 --- a/drivers/gpio/gpio-rockchip.c +++ b/drivers/gpio/gpio-rockchip.c @@ -638,10 +638,17 @@ static int rockchip_gpiolib_register(struct rockchip_pin_bank *bank) return ret; } +static void rockchip_clk_put(void *data) +{ + struct clk *clk = data; + + clk_put(clk); +} + static int rockchip_get_bank_data(struct rockchip_pin_bank *bank) { struct resource res; - int id = 0; + int id = 0, ret; if (of_address_to_resource(bank->of_node, 0, &res)) { dev_err(bank->dev, "cannot find IO resource for bank\n"); @@ -673,6 +680,11 @@ static int rockchip_get_bank_data(struct rockchip_pin_bank *bank) dev_err(bank->dev, "cannot find debounce clk\n"); return -EINVAL; } + + ret = devm_add_action_or_reset(bank->dev, rockchip_clk_put, + bank->db_clk); + if (ret) + return ret; break; case GPIO_TYPE_V1: bank->gpio_regs = &gpio_regs_v1; @@ -789,6 +801,9 @@ static void rockchip_gpio_remove(struct platform_device *pdev) { struct rockchip_pin_bank *bank = platform_get_drvdata(pdev); + irq_set_chained_handler_and_data(bank->irq, NULL, NULL); + if (bank->domain) + irq_domain_remove(bank->domain); gpiochip_remove(&bank->gpio_chip); } From 175db11786bde9061db526bf1ac5107d915f5163 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Sat, 16 May 2026 04:34:14 +0900 Subject: [PATCH 373/411] Disable -Wattribute-alias for clang-23 and newer Clang recently added support for -Wattribute-alias [1], which results in the same warnings that necessitated commit bee20031772a ("disable -Wattribute-alias warning for SYSCALL_DEFINEx()") for GCC. kernel/time/itimer.c:325:1: error: alias and aliasee have different types 'long (unsigned int)' and 'long (typeof (__builtin_choose_expr((__builtin_types_compatible_p(typeof ((unsigned int)0), typeof (0LL)) || __builtin_types_compatible_p(typeof ((unsigned int)0), typeof (0ULL))), 0LL, 0L)))' (aka 'long (long)') [-Werror,-Wattribute-alias] 325 | SYSCALL_DEFINE1(alarm, unsigned int, seconds) | ^ include/linux/syscalls.h:225:36: note: expanded from macro 'SYSCALL_DEFINE1' 225 | #define SYSCALL_DEFINE1(name, ...) SYSCALL_DEFINEx(1, _##name, __VA_ARGS__) | ^ include/linux/syscalls.h:236:2: note: expanded from macro 'SYSCALL_DEFINEx' 236 | __SYSCALL_DEFINEx(x, sname, __VA_ARGS__) | ^ include/linux/syscalls.h:251:18: note: expanded from macro '__SYSCALL_DEFINEx' 251 | __attribute__((alias(__stringify(__se_sys##name)))); \ | ^ kernel/time/itimer.c:325:1: note: aliasee is declared here include/linux/syscalls.h:225:36: note: expanded from macro 'SYSCALL_DEFINE1' 225 | #define SYSCALL_DEFINE1(name, ...) SYSCALL_DEFINEx(1, _##name, __VA_ARGS__) | ^ include/linux/syscalls.h:236:2: note: expanded from macro 'SYSCALL_DEFINEx' 236 | __SYSCALL_DEFINEx(x, sname, __VA_ARGS__) | ^ include/linux/syscalls.h:255:18: note: expanded from macro '__SYSCALL_DEFINEx' 255 | asmlinkage long __se_sys##name(__MAP(x,__SC_LONG,__VA_ARGS__)) \ | ^ :16:1: note: expanded from here 16 | __se_sys_alarm | ^ Disable the warnings in the same way for clang-23 and newer. Disable the warning about unknown warning options to avoid breaking the build for versions of clang-23 that do not have -Wattribute-alias, such as ones deployed by vendors like Android or CI systems or when bisecting LLVM between llvmorg-23-init and release/23.x. Cc: stable@vger.kernel.org Closes: https://github.com/ClangBuiltLinux/linux/issues/2163 Link: https://github.com/llvm/llvm-project/commit/40da6920a0d71d49dfa2392b09153600b0759f5e [1] Link: https://patch.msgid.link/20260515-syscall-disable-attribute-alias-for-clang-v1-1-9a9d95d41df6@kernel.org Signed-off-by: Nathan Chancellor --- arch/riscv/include/asm/syscall_wrapper.h | 4 ++++ include/linux/compat.h | 4 ++++ include/linux/compiler-clang.h | 6 ++++++ include/linux/compiler_types.h | 4 ++++ include/linux/syscalls.h | 4 ++++ 5 files changed, 22 insertions(+) diff --git a/arch/riscv/include/asm/syscall_wrapper.h b/arch/riscv/include/asm/syscall_wrapper.h index ac80216549ff..226289c3b5c8 100644 --- a/arch/riscv/include/asm/syscall_wrapper.h +++ b/arch/riscv/include/asm/syscall_wrapper.h @@ -32,6 +32,10 @@ asmlinkage long __riscv_sys_ni_syscall(const struct pt_regs *); __diag_push(); \ __diag_ignore(GCC, 8, "-Wattribute-alias", \ "Type aliasing is used to sanitize syscall arguments"); \ + __diag_ignore(clang, 23, "-Wunknown-warning-option", \ + "Avoid breaking versions without -Wattribute-alias"); \ + __diag_ignore(clang, 23, "-Wattribute-alias", \ + "Type aliasing is used to sanitize syscall arguments"); \ static long __se_##prefix##name(ulong, ulong, ulong, ulong, ulong, ulong, \ ulong) \ __attribute__((alias(__stringify(___se_##prefix##name)))); \ diff --git a/include/linux/compat.h b/include/linux/compat.h index 56cebaff0c91..8da0a15c95f4 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -72,6 +72,10 @@ __diag_push(); \ __diag_ignore(GCC, 8, "-Wattribute-alias", \ "Type aliasing is used to sanitize syscall arguments");\ + __diag_ignore(clang, 23, "-Wunknown-warning-option", \ + "Avoid breaking versions without -Wattribute-alias"); \ + __diag_ignore(clang, 23, "-Wattribute-alias", \ + "Type aliasing is used to sanitize syscall arguments"); \ asmlinkage long compat_sys##name(__MAP(x,__SC_DECL,__VA_ARGS__)) \ __attribute__((alias(__stringify(__se_compat_sys##name)))); \ ALLOW_ERROR_INJECTION(compat_sys##name, ERRNO); \ diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h index e1123dd28486..527e4e136020 100644 --- a/include/linux/compiler-clang.h +++ b/include/linux/compiler-clang.h @@ -131,6 +131,12 @@ #define __diag_str(s) __diag_str1(s) #define __diag(s) _Pragma(__diag_str(clang diagnostic s)) +#if CONFIG_CLANG_VERSION >= 230000 +#define __diag_clang_23(s) __diag(s) +#else +#define __diag_clang_23(s) +#endif + #define __diag_clang_13(s) __diag(s) #define __diag_ignore_all(option, comment) \ diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h index e8fd77593b68..369966598a2c 100644 --- a/include/linux/compiler_types.h +++ b/include/linux/compiler_types.h @@ -711,6 +711,10 @@ struct ftrace_likely_data { #define __diag_GCC(version, severity, string) #endif +#ifndef __diag_clang +#define __diag_clang(version, severity, string) +#endif + #define __diag_push() __diag(push) #define __diag_pop() __diag(pop) diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index f5639d5ac331..4fb7291f54b6 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -247,6 +247,10 @@ static inline int is_syscall_trace_event(struct trace_event_call *tp_event) __diag_push(); \ __diag_ignore(GCC, 8, "-Wattribute-alias", \ "Type aliasing is used to sanitize syscall arguments");\ + __diag_ignore(clang, 23, "-Wunknown-warning-option", \ + "Avoid breaking versions without -Wattribute-alias");\ + __diag_ignore(clang, 23, "-Wattribute-alias", \ + "Type aliasing is used to sanitize syscall arguments");\ asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__)) \ __attribute__((alias(__stringify(__se_sys##name)))); \ ALLOW_ERROR_INJECTION(sys##name, ERRNO); \ From 006c66d1d52f1905e6ccfb615cf27235e4e6e745 Mon Sep 17 00:00:00 2001 From: bui duc phuc Date: Thu, 28 May 2026 12:32:04 +0700 Subject: [PATCH 374/411] regmap: reject volatile update_bits() in cache-only mode Prevent _regmap_update_bits() from accessing hardware when the register map is in cache-only mode. Unlike regmap_raw_read() and _regmap_read(), the volatile _regmap_update_bits() fast path bypasses the cache_only check. This can result in unexpected hardware accesses while the device is suspended. Return -EBUSY to ensure behavior is consistent with other cache-only access paths. Signed-off-by: bui duc phuc Link: https://patch.msgid.link/20260528053204.46783-1-phucduc.bui@gmail.com Signed-off-by: Mark Brown --- drivers/base/regmap/regmap.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/base/regmap/regmap.c b/drivers/base/regmap/regmap.c index b2b26f07f4e3..e6e022b02637 100644 --- a/drivers/base/regmap/regmap.c +++ b/drivers/base/regmap/regmap.c @@ -3257,6 +3257,9 @@ static int _regmap_update_bits(struct regmap *map, unsigned int reg, *change = false; if (regmap_volatile(map, reg) && map->reg_update_bits) { + if (map->cache_only) + return -EBUSY; + reg = regmap_reg_addr(map, reg); ret = map->reg_update_bits(map->bus_context, reg, mask, val); if (ret == 0 && change) From ead6680f354f83966c796fc7f9463a3171789616 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Sat, 23 May 2026 19:14:46 +0100 Subject: [PATCH 375/411] dma-buf: fix UAF in dma_buf_fd() tracepoint MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Once FD_ADD() returns, the fd is live in the file descriptor table and a thread sharing that table can close() it before DMA_BUF_TRACE() runs. The close drops the last reference, __fput() frees the dma_buf, and the tracepoint then dereferences dmabuf to take dmabuf->name_lock -- slab-use-after-free. Split FD_ADD() back into get_unused_fd_flags() + fd_install() and emit the tracepoint between them. While the fdtable slot is reserved with a NULL file pointer, a racing close() returns -EBADF without entering __fput(), so the dma_buf stays alive across the trace. Same approach as commit 2d76319c4cbb ("dma-buf: fix UAF in dma_buf_put() tracepoint"). This undoes the FD_ADD() conversion done in commit 34dfce523c90 ("dma: convert dma_buf_fd() to FD_ADD()"); FD_ADD() has no place to hook the tracepoint safely. Reported-by: syzbot+7f4987d0afb97dd090cb@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=7f4987d0afb97dd090cb Fixes: 281a22631423 ("dma-buf: add some tracepoints to debug.") Cc: stable@vger.kernel.org # 7.0.x Signed-off-by: David Carlier Reviewed-by: Christian König Signed-off-by: Sumit Semwal Link: https://patch.msgid.link/20260523181446.69525-1-devnexen@gmail.com --- drivers/dma-buf/dma-buf.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/dma-buf/dma-buf.c b/drivers/dma-buf/dma-buf.c index 71f37544a5c6..d504c636dc29 100644 --- a/drivers/dma-buf/dma-buf.c +++ b/drivers/dma-buf/dma-buf.c @@ -792,9 +792,13 @@ int dma_buf_fd(struct dma_buf *dmabuf, int flags) if (!dmabuf || !dmabuf->file) return -EINVAL; - fd = FD_ADD(flags, dmabuf->file); + fd = get_unused_fd_flags(flags); + if (fd < 0) + return fd; + DMA_BUF_TRACE(trace_dma_buf_fd, dmabuf, fd); + fd_install(fd, dmabuf->file); return fd; } EXPORT_SYMBOL_NS_GPL(dma_buf_fd, "DMA_BUF"); From c0a8899e02ddebd51e2589835182c239c2e224ae Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Wed, 27 May 2026 17:05:26 +0100 Subject: [PATCH 376/411] HID: wacom: Fix OOB write in wacom_hid_set_device_mode() wacom_hid_set_device_mode() currently assumes that the HID_DG_INPUTMODE usage is always located in the first field (field[0]) of the feature report. However, a device can specify HID_DG_INPUTMODE in a different field. If HID_DG_INPUTMODE is in a field other than the first one and the first field has a report_count smaller than the usage_index of HID_DG_INPUTMODE, this leads to an out-of-bounds write to r->field[0]->value. Fix this by storing the field index of HID_DG_INPUTMODE in 'struct hid_data' during feature mapping. In wacom_hid_set_device_mode(), use this stored field index to access the correct field and add bounds checks to ensure both the field index and the value index are within valid ranges before writing. Cc: stable@vger.kernel.org Fixes: 5ae6e89f7409 ("HID: wacom: implement the finger part of the HID generic handling") Tested-by: Ping Cheng Reviewed-by: Ping Cheng Signed-off-by: Lee Jones Signed-off-by: Benjamin Tissoires --- drivers/hid/wacom_sys.c | 13 ++++++++++--- drivers/hid/wacom_wac.h | 1 + 2 files changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/hid/wacom_sys.c b/drivers/hid/wacom_sys.c index a32320b351e3..2220168bf116 100644 --- a/drivers/hid/wacom_sys.c +++ b/drivers/hid/wacom_sys.c @@ -356,6 +356,7 @@ static void wacom_feature_mapping(struct hid_device *hdev, hid_data->inputmode = field->report->id; hid_data->inputmode_index = usage->usage_index; + hid_data->inputmode_field_index = field->index; break; case HID_UP_DIGITIZER: @@ -571,9 +572,14 @@ static int wacom_hid_set_device_mode(struct hid_device *hdev) re = &(hdev->report_enum[HID_FEATURE_REPORT]); r = re->report_id_hash[hid_data->inputmode]; - if (r) { - r->field[0]->value[hid_data->inputmode_index] = 2; - hid_hw_request(hdev, r, HID_REQ_SET_REPORT); + if (r && hid_data->inputmode_field_index >= 0 && + hid_data->inputmode_field_index < r->maxfield) { + struct hid_field *field = r->field[hid_data->inputmode_field_index]; + + if (field && hid_data->inputmode_index < field->report_count) { + field->value[hid_data->inputmode_index] = 2; + hid_hw_request(hdev, r, HID_REQ_SET_REPORT); + } } return 0; } @@ -2846,6 +2852,7 @@ static int wacom_probe(struct hid_device *hdev, return -ENODEV; wacom_wac->hid_data.inputmode = -1; + wacom_wac->hid_data.inputmode_field_index = -1; wacom_wac->mode_report = -1; if (hid_is_usb(hdev)) { diff --git a/drivers/hid/wacom_wac.h b/drivers/hid/wacom_wac.h index d4f7d8ca1e7e..126bec6e5c0c 100644 --- a/drivers/hid/wacom_wac.h +++ b/drivers/hid/wacom_wac.h @@ -295,6 +295,7 @@ struct wacom_shared { struct hid_data { __s16 inputmode; /* InputMode HID feature, -1 if non-existent */ __s16 inputmode_index; /* InputMode HID feature index in the report */ + __s16 inputmode_field_index; /* InputMode HID feature field index in the report */ bool sense_state; bool inrange_state; bool eraser; From 43a1e3744548e6fd85873e6fb43e293eb4010694 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Thu, 28 May 2026 11:45:41 -0700 Subject: [PATCH 377/411] security/keys: fix missed RCU read section on lookup Nicholas Carlini reports that the keyring code calls assoc_array_find() in find_key_to_update() without holding the RCU read lock, while the assoc_array_gc() code really is designed around removing the node from the tree and then freeing it after an RCU grace-period. The regular key handling doesn't see this because holding the keyring semaphore hides any lifetime issues, but the persistent key handling uses a different model. Instead of extending the keyring locking, just do the simple RCU locking that the assoc_array was designed for. Reported-by: Nicholas Carlini Cc: David Howells Cc: Jarkko Sakkinen Cc: Paul Moore Cc: James Morris James Morris Cc: Serge E. Hallyn Signed-off-by: Linus Torvalds --- security/keys/keyring.c | 1 + 1 file changed, 1 insertion(+) diff --git a/security/keys/keyring.c b/security/keys/keyring.c index b39038f7dd31..5a9887d6b7be 100644 --- a/security/keys/keyring.c +++ b/security/keys/keyring.c @@ -1109,6 +1109,7 @@ key_ref_t find_key_to_update(key_ref_t keyring_ref, kenter("{%d},{%s,%s}", keyring->serial, index_key->type->name, index_key->description); + guard(rcu)(); object = assoc_array_find(&keyring->keys, &keyring_assoc_array_ops, index_key); From 17bfe0a8c014ee1d542ad352cd6a0a505361664a Mon Sep 17 00:00:00 2001 From: Dipayaan Roy Date: Mon, 25 May 2026 01:08:24 -0700 Subject: [PATCH 378/411] net: mana: Add NULL guards in teardown path to prevent panic on attach failure When queue allocation fails partway through, the error cleanup frees and NULLs apc->tx_qp and apc->rxqs. Multiple teardown paths such as mana_remove(), mana_change_mtu() recovery, and internal error handling in mana_alloc_queues() can subsequently call into functions that dereference these pointers without NULL checks: - mana_chn_setxdp() dereferences apc->rxqs[0], causing a NULL pointer dereference panic (CR2: 0000000000000000 at mana_chn_setxdp+0x26). - mana_destroy_vport() iterates apc->rxqs without a NULL check. - mana_fence_rqs() iterates apc->rxqs without a NULL check. - mana_dealloc_queues() iterates apc->tx_qp without a NULL check. Add NULL guards for apc->rxqs in mana_fence_rqs(), mana_destroy_vport(), and before the mana_chn_setxdp() call. Add a NULL guard for apc->tx_qp in mana_dealloc_queues() to skip TX queue draining when TX queues were never allocated or already freed. Fixes: ca9c54d2d6a5 ("net: mana: Add a driver for Microsoft Azure Network Adapter (MANA)") Reviewed-by: Haiyang Zhang Signed-off-by: Dipayaan Roy Link: https://patch.msgid.link/20260525081129.1230035-2-dipayanroy@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 72 +++++++++++-------- 1 file changed, 42 insertions(+), 30 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 9afc786b297a..9e7e4bf526bf 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -1727,6 +1727,9 @@ static void mana_fence_rqs(struct mana_port_context *apc) struct mana_rxq *rxq; int err; + if (!apc->rxqs) + return; + for (rxq_idx = 0; rxq_idx < apc->num_queues; rxq_idx++) { rxq = apc->rxqs[rxq_idx]; err = mana_fence_rq(apc, rxq); @@ -2858,13 +2861,16 @@ static void mana_destroy_vport(struct mana_port_context *apc) struct mana_rxq *rxq; u32 rxq_idx; - for (rxq_idx = 0; rxq_idx < apc->num_queues; rxq_idx++) { - rxq = apc->rxqs[rxq_idx]; - if (!rxq) - continue; + if (apc->rxqs) { - mana_destroy_rxq(apc, rxq, true); - apc->rxqs[rxq_idx] = NULL; + for (rxq_idx = 0; rxq_idx < apc->num_queues; rxq_idx++) { + rxq = apc->rxqs[rxq_idx]; + if (!rxq) + continue; + + mana_destroy_rxq(apc, rxq, true); + apc->rxqs[rxq_idx] = NULL; + } } mana_destroy_txq(apc); @@ -3269,7 +3275,8 @@ static int mana_dealloc_queues(struct net_device *ndev) if (apc->port_is_up) return -EINVAL; - mana_chn_setxdp(apc, NULL); + if (apc->rxqs) + mana_chn_setxdp(apc, NULL); if (gd->gdma_context->is_pf && !apc->ac->bm_hostmode) mana_pf_deregister_filter(apc); @@ -3287,33 +3294,38 @@ static int mana_dealloc_queues(struct net_device *ndev) * number of queues. */ - for (i = 0; i < apc->num_queues; i++) { - txq = &apc->tx_qp[i].txq; - tsleep = 1000; - while (atomic_read(&txq->pending_sends) > 0 && - time_before(jiffies, timeout)) { - usleep_range(tsleep, tsleep + 1000); - tsleep <<= 1; - } - if (atomic_read(&txq->pending_sends)) { - err = pcie_flr(to_pci_dev(gd->gdma_context->dev)); - if (err) { - netdev_err(ndev, "flr failed %d with %d pkts pending in txq %u\n", - err, atomic_read(&txq->pending_sends), - txq->gdma_txq_id); + if (apc->tx_qp) { + for (i = 0; i < apc->num_queues; i++) { + txq = &apc->tx_qp[i].txq; + tsleep = 1000; + while (atomic_read(&txq->pending_sends) > 0 && + time_before(jiffies, timeout)) { + usleep_range(tsleep, tsleep + 1000); + tsleep <<= 1; } - break; + if (atomic_read(&txq->pending_sends)) { + err = + pcie_flr(to_pci_dev(gd->gdma_context->dev)); + if (err) { + netdev_err(ndev, "flr failed %d with %d pkts pending in txq %u\n", + err, + atomic_read(&txq->pending_sends), + txq->gdma_txq_id); + } + break; + } + } + + for (i = 0; i < apc->num_queues; i++) { + txq = &apc->tx_qp[i].txq; + while ((skb = skb_dequeue(&txq->pending_skbs))) { + mana_unmap_skb(skb, apc); + dev_kfree_skb_any(skb); + } + atomic_set(&txq->pending_sends, 0); } } - for (i = 0; i < apc->num_queues; i++) { - txq = &apc->tx_qp[i].txq; - while ((skb = skb_dequeue(&txq->pending_skbs))) { - mana_unmap_skb(skb, apc); - dev_kfree_skb_any(skb); - } - atomic_set(&txq->pending_sends, 0); - } /* We're 100% sure the queues can no longer be woken up, because * we're sure now mana_poll_tx_cq() can't be running. */ From 5b05aa36ee24297d7296ca58dfd8c448d0e4cda3 Mon Sep 17 00:00:00 2001 From: Dipayaan Roy Date: Mon, 25 May 2026 01:08:25 -0700 Subject: [PATCH 379/411] net: mana: Skip redundant detach on already-detached port When mana_per_port_queue_reset_work_handler() runs after a previous detach succeeded but attach failed, the port is left in a detached state with apc->tx_qp and apc->rxqs already freed. Calling mana_detach() again unconditionally leads to NULL pointer dereferences during queue teardown. Add an early exit in mana_detach() when the port is already in detached state (!netif_device_present) for non-close callers, making it safe to call idempotently. This allows the queue reset handler and other recovery paths to simply retry mana_attach() without redundant teardown. Fixes: 3b194343c250 ("net: mana: Implement ndo_tx_timeout and serialize queue resets per port.") Reviewed-by: Haiyang Zhang Signed-off-by: Dipayaan Roy Link: https://patch.msgid.link/20260525081129.1230035-3-dipayanroy@linux.microsoft.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 9e7e4bf526bf..c9b1df1ed109 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -3350,6 +3350,12 @@ int mana_detach(struct net_device *ndev, bool from_close) ASSERT_RTNL(); + /* If already detached (indicates detach succeeded but attach failed + * previously). Now skip mana detach and just retry mana_attach. + */ + if (!from_close && !netif_device_present(ndev)) + return 0; + apc->port_st_save = apc->port_is_up; apc->port_is_up = false; From f14fe6395a8b3d961a61e138ad7b36ba3626dd4e Mon Sep 17 00:00:00 2001 From: Zhenghang Xiao Date: Wed, 27 May 2026 11:24:11 +0800 Subject: [PATCH 380/411] sctp: fix race between sctp_wait_for_connect and peeloff MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit sctp_wait_for_connect() drops and re-acquires the socket lock while waiting for the association to reach ESTABLISHED state. During this window, another thread can peeloff the association to a new socket via getsockopt(SCTP_SOCKOPT_PEELOFF), changing asoc->base.sk. After re-acquiring the old socket lock, sctp_wait_for_connect() returns success without noticing the migration — the caller then accesses the association under the wrong lock in sctp_datamsg_from_user(). Add the same sk != asoc->base.sk check that sctp_wait_for_sndbuf() already has, returning an error if the association was migrated while we slept. Fixes: 668c9beb9020 ("sctp: implement assign_number for sctp_stream_interleave") Signed-off-by: Zhenghang Xiao Acked-by: Xin Long Link: https://patch.msgid.link/20260527032411.60959-1-kipreyyy@gmail.com Signed-off-by: Jakub Kicinski --- net/sctp/socket.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index 1d2568bb6bc2..66e12fb0c646 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -9403,6 +9403,8 @@ static int sctp_wait_for_connect(struct sctp_association *asoc, long *timeo_p) release_sock(sk); current_timeo = schedule_timeout(current_timeo); lock_sock(sk); + if (sk != asoc->base.sk) + goto do_error; *timeo_p = current_timeo; } From 422b5233b607476ac7176bfa2a101b9a103d7653 Mon Sep 17 00:00:00 2001 From: Frank Wunderlich Date: Tue, 26 May 2026 17:32:38 +0200 Subject: [PATCH 381/411] net: pcs: pcs-mtk-lynxi: fix bpi-r3 serdes configuration Commit 8871389da151 introduces common pcs dts properties which writes rx=normal,tx=normal polarity to register SGMSYS_QPHY_WRAP_CTRL of switch. This is initialized with tx-bit set and so change inverts polarity compared to before. It looks like mt7531 has tx polarity inverted in hardware and set tx-bit by default to restore the normal polarity. The MT7531 datasheet quite clearly states: Register 000050EC QPHY_WRAP_CTRL -- QPHY wrapper control Reset value: 0x00000501 BIT 1 RX_BIT_POLARITY -- RX bit polarity control 1'b0: normal 1'b1: inverted BIT 0 TX_BIT_POLARITY -- TX bit polarity control (TX default inversed in MT7531) 1'b0: normal 1'b1: inverted Till this patch the register write was only called when mediatek,pnswap property was set which cannot be done for switch because the fw-node param was always NULL from switch driver in the mtk_pcs_lynxi_create call. Do not configure switch side like it's done before. Fixes: 8871389da151 ("net: pcs: pcs-mtk-lynxi: deprecate "mediatek,pnswap"") Signed-off-by: Frank Wunderlich Reviewed-by: Vladimir Oltean Link: https://patch.msgid.link/20260526153239.30194-1-linux@fw-web.de Signed-off-by: Jakub Kicinski --- drivers/net/pcs/pcs-mtk-lynxi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/net/pcs/pcs-mtk-lynxi.c b/drivers/net/pcs/pcs-mtk-lynxi.c index c12f8087af9b..a753bd88cbc2 100644 --- a/drivers/net/pcs/pcs-mtk-lynxi.c +++ b/drivers/net/pcs/pcs-mtk-lynxi.c @@ -129,6 +129,9 @@ static int mtk_pcs_config_polarity(struct mtk_pcs_lynxi *mpcs, unsigned int val = 0; int ret; + if (!fwnode) + return 0; + if (fwnode_property_read_bool(fwnode, "mediatek,pnswap")) default_pol = PHY_POL_INVERT; From 5ab62dd3687bcc2cc542b99385aabac5c996db6f Mon Sep 17 00:00:00 2001 From: Rajat Gupta Date: Wed, 20 May 2026 22:11:21 -0700 Subject: [PATCH 382/411] drm: prevent integer overflows in dumb buffer creation helpers Fix integer overflow issues in the dumb buffer creation path: 1. drm_mode_create_dumb() does not bound width, height, or bpp before passing them to driver callbacks. Downstream helpers (e.g. drm_gem_dma_dumb_create_internal) perform pitch/size alignment in u32 arithmetic that can overflow for extreme values. Add hard limits: width and height < 8192, bpp <= 32. No legitimate software rendering use case exceeds these. 2. drm_mode_align_dumb() uses roundup(pitch, hw_pitch_align) without checking for overflow. If pitch is near U32_MAX, roundup() wraps to a small value, making subsequent check_mul_overflow() pass with a much smaller pitch than intended. Add an overflow check after roundup. 3. drm_mode_align_dumb() uses ALIGN(size, hw_size_align) which only works correctly for power-of-two alignment values. Replace with roundup() which works for any alignment. Suggested-by: Thomas Zimmermann Signed-off-by: Rajat Gupta Signed-off-by: Thomas Zimmermann --- drivers/gpu/drm/drm_dumb_buffers.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/drm_dumb_buffers.c b/drivers/gpu/drm/drm_dumb_buffers.c index e2b62e5fb891..cc99681a9ed0 100644 --- a/drivers/gpu/drm/drm_dumb_buffers.c +++ b/drivers/gpu/drm/drm_dumb_buffers.c @@ -70,8 +70,11 @@ static int drm_mode_align_dumb(struct drm_mode_create_dumb *args, if (!pitch) return -EINVAL; - if (hw_pitch_align) + if (hw_pitch_align) { pitch = roundup(pitch, hw_pitch_align); + if (pitch < hw_pitch_align) + return -EINVAL; + } if (!hw_size_align) hw_size_align = PAGE_SIZE; @@ -80,7 +83,7 @@ static int drm_mode_align_dumb(struct drm_mode_create_dumb *args, if (check_mul_overflow(args->height, pitch, &size)) return -EINVAL; - size = ALIGN(size, hw_size_align); + size = roundup(size, hw_size_align); if (!size) return -EINVAL; @@ -199,6 +202,13 @@ int drm_mode_create_dumb(struct drm_device *dev, if (!args->width || !args->height || !args->bpp) return -EINVAL; + /* Reject unreasonable inputs early. Dumb buffers are for software + * rendering; nothing legitimate needs more than 8192x8192 at 32bpp. + * This prevents overflows in downstream alignment helpers. + */ + if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32) + return -EINVAL; + /* overflow checks for 32bit size calculations */ if (args->bpp > U32_MAX - 8) return -EINVAL; From 6851161feb01cea41358c9ec304bd2f981fc8505 Mon Sep 17 00:00:00 2001 From: Steffen Klassert Date: Fri, 29 May 2026 10:23:25 +0200 Subject: [PATCH 383/411] Revert "esp: fix page frag reference leak on skb_to_sgvec failure" This reverts commit 2982e599fff6faa21c8df147d96fc7af6c1a2f24. The patch does not fully fix the issue and the Author does not match the 'Signed-off-by:' tag, so revert it for now. Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 12 +++++------- net/ipv6/esp6.c | 12 +++++------- 2 files changed, 10 insertions(+), 14 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 5d3a8656687e..513c8215c947 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -96,7 +96,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -113,7 +113,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (already_unref || req->src != req->dst) + if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -220,7 +220,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); kfree(tmp); if (xo && (xo->flags & XFRM_DEV_RESUME)) { @@ -569,10 +569,8 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) { - esp_ssg_unref(x, tmp, skb, true); + if (unlikely(err < 0)) goto error_free; - } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -604,7 +602,7 @@ int esp_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info * } if (sg != dsg) - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index b963b8e72604..57481e423e59 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -113,7 +113,7 @@ static inline struct scatterlist *esp_req_sg(struct crypto_aead *aead, __alignof__(struct scatterlist)); } -static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, bool already_unref) +static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb) { struct crypto_aead *aead = x->data; int extralen = 0; @@ -130,7 +130,7 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb, /* Unref skb_frag_pages in the src scatterlist if necessary. * Skip the first sg which comes from skb->data. */ - if (already_unref || req->src != req->dst) + if (req->src != req->dst) for (sg = sg_next(req->src); sg; sg = sg_next(sg)) skb_page_unref(page_to_netmem(sg_page(sg)), skb->pp_recycle); @@ -254,7 +254,7 @@ static void esp_output_done(void *data, int err) } tmp = ESP_SKB_CB(skb)->tmp; - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); kfree(tmp); esp_output_encap_csum(skb); @@ -600,10 +600,8 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info err = skb_to_sgvec(skb, dsg, (unsigned char *)esph - skb->data, assoclen + ivlen + esp->clen + alen); - if (unlikely(err < 0)) { - esp_ssg_unref(x, tmp, skb, true); + if (unlikely(err < 0)) goto error_free; - } } if ((x->props.flags & XFRM_STATE_ESN)) @@ -636,7 +634,7 @@ int esp6_output_tail(struct xfrm_state *x, struct sk_buff *skb, struct esp_info } if (sg != dsg) - esp_ssg_unref(x, tmp, skb, false); + esp_ssg_unref(x, tmp, skb); if (!err && x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) err = esp_output_tail_tcp(x, skb); From 83726330748981372bde86ed5411d7b306612991 Mon Sep 17 00:00:00 2001 From: Mark Brown Date: Fri, 29 May 2026 00:01:44 +0100 Subject: [PATCH 384/411] KVM: arm64: Correctly cap ZCR_EL2 provided by a guest hypervisor ZCR_EL2 can be updated by a VHE guest hypervisor either using ZCR_EL2 (which traps) or ZCR_EL1 (which does not trap). KVM handles both in different way: - on ZCR_EL2 trap, ZCR_EL2.LEN is immediately capped at the VM's own VL limit. This has the potential to break existing SW that relies on the full LEN field to be stateful. - on ZCR_EL1 access, we do absolutely nothing. On restoring the SVE context for an L2 guest, we directly restore the guest hypervisor's view of ZCR_EL2 into the physical ZCR_EL2. If the guest's view of the register was updated using the ZCR_EL2 accessor, the value has already been sanitised (with the caveat mentioned above). But if the guest used ZCR_EL1, the raw value is written into the HW, and the L2 guest can now access VLs that it shouldn't. Fix all the above by moving the VL capping to the restore points, ensuring that: - the HW is always programmed with a capped value, irrespective of the accessor being used, - the ZCR_EL2.LEN field is always completely stateful, irrespective of the accessor being used. Additionally, move ZCR_EL2 to be a sanitised register, ensuring that only the LEN field is actually stateful. This requires some creative construction of the RES0 mask, as the sysreg generation script does not yet generate RAZ/WI fields. Fixes: b3d29a823099 ("KVM: arm64: nv: Handle ZCR_EL2 traps") Signed-off-by: Mark Brown Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260529-kvm-arm64-fix-zcr-len-nv-v2-1-86cad51992bd@kernel.org [maz: rewrote commit message, tidy up access_zcr_el2()] Signed-off-by: Marc Zyngier --- arch/arm64/include/asm/kvm_host.h | 2 +- arch/arm64/kvm/hyp/include/hyp/switch.h | 16 ++++++++++------ arch/arm64/kvm/nested.c | 5 +++++ arch/arm64/kvm/sys_regs.c | 11 +++-------- 4 files changed, 19 insertions(+), 15 deletions(-) diff --git a/arch/arm64/include/asm/kvm_host.h b/arch/arm64/include/asm/kvm_host.h index 65eead8362e0..a49042bfa801 100644 --- a/arch/arm64/include/asm/kvm_host.h +++ b/arch/arm64/include/asm/kvm_host.h @@ -511,7 +511,6 @@ enum vcpu_sysreg { ACTLR_EL2, /* Auxiliary Control Register (EL2) */ CPTR_EL2, /* Architectural Feature Trap Register (EL2) */ HACR_EL2, /* Hypervisor Auxiliary Control Register */ - ZCR_EL2, /* SVE Control Register (EL2) */ TTBR0_EL2, /* Translation Table Base Register 0 (EL2) */ TTBR1_EL2, /* Translation Table Base Register 1 (EL2) */ TCR_EL2, /* Translation Control Register (EL2) */ @@ -543,6 +542,7 @@ enum vcpu_sysreg { SCTLR2_EL2, /* System Control Register 2 (EL2) */ MDCR_EL2, /* Monitor Debug Configuration Register (EL2) */ CNTHCTL_EL2, /* Counter-timer Hypervisor Control register */ + ZCR_EL2, /* SVE Control Register (EL2) */ /* Any VNCR-capable reg goes after this point */ MARKER(__VNCR_START__), diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index bf0eb5e43427..320cd45d49c5 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -462,11 +462,13 @@ static inline bool kvm_hyp_handle_mops(struct kvm_vcpu *vcpu, u64 *exit_code) static inline void __hyp_sve_restore_guest(struct kvm_vcpu *vcpu) { + u64 zcr_el2 = vcpu_sve_max_vq(vcpu) - 1; + /* * The vCPU's saved SVE state layout always matches the max VL of the * vCPU. Start off with the max VL so we can load the SVE state. */ - sve_cond_update_zcr_vq(vcpu_sve_max_vq(vcpu) - 1, SYS_ZCR_EL2); + sve_cond_update_zcr_vq(zcr_el2, SYS_ZCR_EL2); __sve_restore_state(vcpu_sve_pffr(vcpu), &vcpu->arch.ctxt.fp_regs.fpsr, true); @@ -476,8 +478,10 @@ static inline void __hyp_sve_restore_guest(struct kvm_vcpu *vcpu) * nested guest, as the guest hypervisor could select a smaller VL. Slap * that into hardware before wrapping up. */ - if (is_nested_ctxt(vcpu)) - sve_cond_update_zcr_vq(__vcpu_sys_reg(vcpu, ZCR_EL2), SYS_ZCR_EL2); + if (is_nested_ctxt(vcpu)) { + zcr_el2 = min(zcr_el2, __vcpu_sys_reg(vcpu, ZCR_EL2)); + sve_cond_update_zcr_vq(zcr_el2, SYS_ZCR_EL2); + } write_sysreg_el1(__vcpu_sys_reg(vcpu, vcpu_sve_zcr_elx(vcpu)), SYS_ZCR); } @@ -501,11 +505,11 @@ static inline void fpsimd_lazy_switch_to_guest(struct kvm_vcpu *vcpu) return; if (vcpu_has_sve(vcpu)) { + zcr_el2 = vcpu_sve_max_vq(vcpu) - 1; + /* A guest hypervisor may restrict the effective max VL. */ if (is_nested_ctxt(vcpu)) - zcr_el2 = __vcpu_sys_reg(vcpu, ZCR_EL2); - else - zcr_el2 = vcpu_sve_max_vq(vcpu) - 1; + zcr_el2 = min(zcr_el2, __vcpu_sys_reg(vcpu, ZCR_EL2)); write_sysreg_el2(zcr_el2, SYS_ZCR); diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index 883b6c1008fb..38f672e94087 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -1834,6 +1834,11 @@ int kvm_init_nv_sysregs(struct kvm_vcpu *vcpu) resx.res1 = VNCR_EL2_RES1; set_sysreg_masks(kvm, VNCR_EL2, resx); + /* ZCR_EL2 - bits 8:4 are RAZ/WI so treat them as RES0 */ + resx.res0 = ZCR_ELx_RES0 | GENMASK_ULL(8, 4); + resx.res1 = ZCR_ELx_RES1; + set_sysreg_masks(kvm, ZCR_EL2, resx); + out: for (enum vcpu_sysreg sr = __SANITISED_REG_START__; sr < NR_SYS_REGS; sr++) __vcpu_rmw_sys_reg(vcpu, sr, |=, 0); diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c index 148fc3400ea8..fa5c93c7a135 100644 --- a/arch/arm64/kvm/sys_regs.c +++ b/arch/arm64/kvm/sys_regs.c @@ -2862,21 +2862,16 @@ static bool access_zcr_el2(struct kvm_vcpu *vcpu, struct sys_reg_params *p, const struct sys_reg_desc *r) { - unsigned int vq; - if (guest_hyp_sve_traps_enabled(vcpu)) { kvm_inject_nested_sve_trap(vcpu); return false; } - if (!p->is_write) { + if (!p->is_write) p->regval = __vcpu_sys_reg(vcpu, ZCR_EL2); - return true; - } + else + __vcpu_assign_sys_reg(vcpu, ZCR_EL2, p->regval); - vq = SYS_FIELD_GET(ZCR_ELx, LEN, p->regval) + 1; - vq = min(vq, vcpu_sve_max_vq(vcpu)); - __vcpu_assign_sys_reg(vcpu, ZCR_EL2, vq - 1); return true; } From a0d8f7ac03e387634c5a7efe3dc162f7d605e2cd Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Thu, 7 May 2026 00:56:49 +0300 Subject: [PATCH 385/411] Revert "media: renesas: vsp1: Initialize format on all pads" This reverts commit 133ac42af0a1b389e8b7b3dc7c1cc8c30ff162b6. The change to format initialization, along with the change to format propagation in the BRx in commit 937f3e6b51f1 ("media: renesas: vsp1: brx: Fix format propagation"), broke configuration of the DRM pipeline. Revert it to fix the regression. The original commit was meant to fix a v4l2-compliance failure, with no known userspace applications being affected beside test tools. Reverting is the simplest option, a more comprehensive fix can be developed (and tested more thoroughly) later. Fixes: 133ac42af0a1 ("media: renesas: vsp1: Initialize format on all pads") Tested-by: Lad Prabhakar # On RZ/T2H Reviewed-by: Lad Prabhakar Link: https://patch.msgid.link/20260506215650.1897177-2-laurent.pinchart+renesas@ideasonboard.com Signed-off-by: Laurent Pinchart Signed-off-by: Hans Verkuil --- drivers/media/platform/renesas/vsp1/vsp1_entity.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/platform/renesas/vsp1/vsp1_entity.c b/drivers/media/platform/renesas/vsp1/vsp1_entity.c index 1dad9589768c..839b75b62ceb 100644 --- a/drivers/media/platform/renesas/vsp1/vsp1_entity.c +++ b/drivers/media/platform/renesas/vsp1/vsp1_entity.c @@ -380,7 +380,7 @@ static int vsp1_entity_init_state(struct v4l2_subdev *subdev, unsigned int pad; /* Initialize all pad formats with default values. */ - for (pad = 0; pad < subdev->entity.num_pads; ++pad) { + for (pad = 0; pad < subdev->entity.num_pads - 1; ++pad) { struct v4l2_subdev_format format = { .pad = pad, .which = sd_state ? V4L2_SUBDEV_FORMAT_TRY From f78073e84c800ae146ce62447e7a685a5ceeb92d Mon Sep 17 00:00:00 2001 From: Laurent Pinchart Date: Thu, 7 May 2026 00:56:50 +0300 Subject: [PATCH 386/411] Revert "media: renesas: vsp1: brx: Fix format propagation" This reverts commit 937f3e6b51f1cea079be9ba642665f2bf8bcc31f. The change to format propagation in the BRx broke configuration of the DRM pipeline. Revert it to fix the regression. The original commit was meant to fix a v4l2-compliance failure, with no known userspace applications being affected beside test tools. Reverting is the simplest option, a more comprehensive fix can be developed (and tested more thoroughly) later. Reported-by: Lad Prabhakar Closes: https://lore.kernel.org/linux-media/CA+V-a8t481xuwava0nb7uY9CUPqFWZ_8EP0xrK3BgumP7HDcLg@mail.gmail.com Fixes: 937f3e6b51f1 ("media: renesas: vsp1: brx: Fix format propagation") Tested-by: Lad Prabhakar # On RZ/T2H Reviewed-by: Lad Prabhakar Link: https://patch.msgid.link/20260506215650.1897177-3-laurent.pinchart+renesas@ideasonboard.com Signed-off-by: Laurent Pinchart Signed-off-by: Hans Verkuil --- drivers/media/platform/renesas/vsp1/vsp1_brx.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/media/platform/renesas/vsp1/vsp1_brx.c b/drivers/media/platform/renesas/vsp1/vsp1_brx.c index b1a2c68e9944..9d93cb8b8e82 100644 --- a/drivers/media/platform/renesas/vsp1/vsp1_brx.c +++ b/drivers/media/platform/renesas/vsp1/vsp1_brx.c @@ -156,20 +156,14 @@ static int brx_set_format(struct v4l2_subdev *subdev, compose->height = format->height; } - /* - * Propagate the format code to all pads, and the whole format to the - * source pad. - */ + /* Propagate the format code to all pads. */ if (fmt->pad == BRX_PAD_SINK(0)) { unsigned int i; - for (i = 0; i < brx->entity.source_pad; ++i) { + for (i = 0; i <= brx->entity.source_pad; ++i) { format = v4l2_subdev_state_get_format(state, i); format->code = fmt->format.code; } - - format = v4l2_subdev_state_get_format(state, i); - *format = fmt->format; } done: From db3f2195d29344a3cf1e9dd9ab7f21ced7308cf7 Mon Sep 17 00:00:00 2001 From: Michael Roth Date: Fri, 1 May 2026 13:22:26 -0700 Subject: [PATCH 387/411] KVM: SEV: Require in-GHCB scratch area if GHCB v2+ is in use As per the GHCB spec, when using GHCB v2+ require the software scratch area to reside in the GHCB's shared buffer. Note, things like Page State Change (PSC) requests _rely_ on this behavior, as the guest can't provide a length when making the request, i.e. the size of the guest payload is bounded by the size of the shared buffer. Failure to force usage of the GHCB, and a slew of other flaws, lets a malicious SNP guest corrupt host kernel heap memory, and leak host heap layout information. setup_vmgexit_scratch() allocates a buffer via kvzalloc(exit_info_2), where exit_info_2 is guest-controlled. With exit_info_2=24, this yields a 24-byte allocation in kmalloc-cg-32 (32-byte slab objects). The buffer holds an 8-byte psc_hdr followed by 8-byte psc_entry structs, so only entries[0] and entries[1] are in-bounds. snp_begin_psc() validates end_entry against VMGEXIT_PSC_MAX_COUNT (253) but NOT against the actual buffer size: idx_end = hdr->end_entry; if (idx_end >= VMGEXIT_PSC_MAX_COUNT) { // checks 253, not buffer snp_complete_psc(svm, ...); return 1; } for (idx = idx_start; idx <= idx_end; idx++) { entry_start = entries[idx]; // OOB when idx >= 2 The guest sets end_entry=10+, causing the host to iterate entries[2+] which are OOB into adjacent slab objects. For each OOB entry: - The host reads 8 bytes (OOB READ / info leak oracle) - If the data passes PSC validation, __snp_complete_one_psc() writes cur_page = 1 or 512 into the entry (OOB WRITE, sev.c:3806) - If validation fails, the error response reveals whether adjacent memory is zero vs non-zero (information disclosure to guest) The guest controls allocation size (exit_info_2), entry range (cur_entry/end_entry), and can fire unlimited VMGEXITs to repeatedly hit different slab positions. By exploiting the variety of bugs, a malicious SEV-SNP guest can: - OOB read adjacent kmalloc-cg-32 objects (heap layout disclosure) - OOB write cur_page bits into adjacent objects (heap corruption) - Trigger use-after-free conditions across VMGEXITs E.g. with KASAN enabled, a single insmod of the PoC guest module produces 73 KASAN reports: BUG: KASAN: slab-out-of-bounds in snp_begin_psc+0x126/0x890 Read of size 8 at addr ffff888219ffb5e0 by task qemu-system-x86/2199 BUG: KASAN: slab-out-of-bounds in snp_begin_psc+0x468/0x890 Write of size 8 at addr ffff888351566648 by task qemu-system-x86/2199 The buggy address belongs to the object at ffff888XXXXXXXXX which belongs to the cache kmalloc-cg-32 of size 32 The buggy address is located N bytes to the right of allocated 32-byte region [ffff888XXXXXXXXX, ffff888XXXXXXXXX) Breakdown: 62 slab-out-of-bounds (reads + writes past allocation) 7 slab-use-after-free 4 use-after-free All credit to Stan for the wonderful description and reproducer! Reported-by: Stan Shaw Cc: Michael Roth Cc: Tom Lendacky Cc: Peter Gonda Cc: Jacky Li Fixes: 4af663c2f64a ("KVM: SEV: Allow per-guest configuration of GHCB protocol version") Cc: stable@vger.kernel.org Signed-off-by: Michael Roth [sean: write changelog] Reviewed-by: Tom Lendacky Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-2-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index c2126b3c3072..23170b64f4a3 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3703,6 +3703,10 @@ static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) scratch_va = (void *)svm->sev_es.ghcb; scratch_va += (scratch_gpa_beg - control->ghcb_gpa); } else { + /* GHCB v2 requires the scratch area to be within the GHCB. */ + if (to_kvm_sev_info(svm->vcpu.kvm)->ghcb_version >= 2) + goto e_scratch; + /* * The guest memory must be read into a kernel buffer, so * limit the size From 1aa8a6dc7dac8b83234b53518311bf78231f4fa5 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:27 -0700 Subject: [PATCH 388/411] KVM: SEV: Ignore MMIO requests of length '0' Explicitly ignore MMIO requests of length '0', so that setting up the software scratch area (and other code) doesn't have to worry about underflowing the length, and to allow for special casing '0' in the future. Fixes: 8f423a80d299 ("KVM: SVM: Support MMIO for an SEV-ES guest") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-3-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 23170b64f4a3..fb2174b6d1ba 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -4497,13 +4497,17 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) case SVM_VMGEXIT_MMIO_READ: case SVM_VMGEXIT_MMIO_WRITE: { bool is_write = control->exit_code == SVM_VMGEXIT_MMIO_WRITE; + u64 len = control->exit_info_2; - ret = setup_vmgexit_scratch(svm, !is_write, control->exit_info_2); + if (!len) + return 1; + + ret = setup_vmgexit_scratch(svm, !is_write, len); if (ret) break; - ret = kvm_sev_es_mmio(vcpu, is_write, control->exit_info_1, - control->exit_info_2, svm->sev_es.ghcb_sa); + ret = kvm_sev_es_mmio(vcpu, is_write, control->exit_info_1, len, + svm->sev_es.ghcb_sa); break; } case SVM_VMGEXIT_NMI_COMPLETE: From dcf1b2d4b0564a27e4ca7c654871aab4f9620046 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:28 -0700 Subject: [PATCH 389/411] KVM: SEV: Reject MMIO requests larger than 8 bytes with GHCB v2+ When using GHCB v2+, reject MMIO requests that are larger than 8 bytes. Per the GHCB spec: SW_EXITINFO2 must be less than or equal to 0x7fffffff for version 1 and less than or equal to 0x8 for all other versions. Fixes: 4af663c2f64a ("KVM: SEV: Allow per-guest configuration of GHCB protocol version") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-4-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index fb2174b6d1ba..e6579ca9f364 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -4502,6 +4502,11 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) if (!len) return 1; + if (to_kvm_sev_info(vcpu->kvm)->ghcb_version >= 2 && len > 8) { + svm_vmgexit_bad_input(svm, GHCB_ERR_INVALID_INPUT); + return 1; + } + ret = setup_vmgexit_scratch(svm, !is_write, len); if (ret) break; From 3988bd2723de407ae90fa7a6f6029b4e60238c58 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:29 -0700 Subject: [PATCH 390/411] KVM: SEV: Ignore Port I/O requests of length '0' Explicitly ignore Port I/O requests of length '0' (or count '0'), so that setting up the software scratch area (and other code) doesn't have to worry about underflowing the length, and to allow for WARNing on trying to configure the scratch area with len==0. Fixes: 291bd20d5d88 ("KVM: SVM: Add initial support for a VMGEXIT VMEXIT") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-5-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index e6579ca9f364..52703c954856 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -4585,6 +4585,11 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) control->exit_info_1, control->exit_info_2); ret = -EINVAL; break; + case SVM_EXIT_IOIO: + if (!((control->exit_info_1 & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT)) + return 1; + + fallthrough; default: ret = svm_invoke_exit_handler(vcpu, control->exit_code); } @@ -4605,6 +4610,9 @@ int sev_es_string_io(struct vcpu_svm *svm, int size, unsigned int port, int in) if (unlikely(check_mul_overflow(count, size, &bytes))) return -EINVAL; + if (!bytes) + return 1; + r = setup_vmgexit_scratch(svm, in, bytes); if (r) return r; From 2be54670bdc017004c4a4b8bddb6ff02ebe7dbe2 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:30 -0700 Subject: [PATCH 391/411] KVM: SEV: Use the size of the PSC header as the minimum size for PSC requests When handling a Page State Change (PSC) #VMGEXIT use the size of the PSC header as the minimum size for the scratch area. Per the GHCB spec, PSC requests do NOT provide the length, i.e. using control->exit_info_2 for the length is completely made up behavior. The existing code "works", e.g. even though Linux-as-a-guest always passes '0', because KVM doesn't do anything with the length when the request is in the GHCB's shared buffer. Use the header as the min length. Once the header is retrieved, KVM can use the specified indices to compute the full size of the request. Fixes: 9b54e248d264 ("KVM: SEV: Add support to handle Page State Change VMGEXIT") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Reviewed-by: Michael Roth Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-6-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 52703c954856..cbb3040e0778 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -4559,7 +4559,7 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) vcpu->run->system_event.data[0] = control->ghcb_gpa; break; case SVM_VMGEXIT_PSC: - ret = setup_vmgexit_scratch(svm, true, control->exit_info_2); + ret = setup_vmgexit_scratch(svm, true, sizeof(struct psc_hdr)); if (ret) break; From 5867d7e202e09f037cefe77f7af4413c7c0fa088 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:31 -0700 Subject: [PATCH 392/411] KVM: SEV: Compute the correct max length of the in-GHCB scratch area When setting the length of the GHCB scratch area, and the area is in the GHCB shared buffer, set the effective length of the scratch area to the max possible size given the start of the guest-provided pointer, and the end of the shared buffer. The code was "fine" when first introduced, as KVM doesn't consult the length of the buffer when emulating MMIO, because the passed in @len always specifies the *max* size required. But for PSC requests, the incoming @len is just the minimum length (to process the header), and KVM needs to know the full size of the scratch area to avoid buffer overflows (spoiler alert). Opportunistically rename @len => @min_len to better reflect its role. Fixes: 9b54e248d264 ("KVM: SEV: Add support to handle Page State Change VMGEXIT") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Reviewed-by: Michael Roth Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-7-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index cbb3040e0778..6072fecfe994 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3662,7 +3662,7 @@ int pre_sev_run(struct vcpu_svm *svm, int cpu) } #define GHCB_SCRATCH_AREA_LIMIT (16ULL * PAGE_SIZE) -static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) +static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 min_len) { struct vmcb_control_area *control = &svm->vmcb->control; u64 ghcb_scratch_beg, ghcb_scratch_end; @@ -3675,10 +3675,10 @@ static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) goto e_scratch; } - scratch_gpa_end = scratch_gpa_beg + len; + scratch_gpa_end = scratch_gpa_beg + min_len; if (scratch_gpa_end < scratch_gpa_beg) { pr_err("vmgexit: scratch length (%#llx) not valid for scratch address (%#llx)\n", - len, scratch_gpa_beg); + min_len, scratch_gpa_beg); goto e_scratch; } @@ -3702,6 +3702,8 @@ static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) scratch_va = (void *)svm->sev_es.ghcb; scratch_va += (scratch_gpa_beg - control->ghcb_gpa); + + svm->sev_es.ghcb_sa_len = ghcb_scratch_end - scratch_gpa_beg; } else { /* GHCB v2 requires the scratch area to be within the GHCB. */ if (to_kvm_sev_info(svm->vcpu.kvm)->ghcb_version >= 2) @@ -3711,16 +3713,16 @@ static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) * The guest memory must be read into a kernel buffer, so * limit the size */ - if (len > GHCB_SCRATCH_AREA_LIMIT) { + if (min_len > GHCB_SCRATCH_AREA_LIMIT) { pr_err("vmgexit: scratch area exceeds KVM limits (%#llx requested, %#llx limit)\n", - len, GHCB_SCRATCH_AREA_LIMIT); + min_len, GHCB_SCRATCH_AREA_LIMIT); goto e_scratch; } - scratch_va = kvzalloc(len, GFP_KERNEL_ACCOUNT); + scratch_va = kvzalloc(min_len, GFP_KERNEL_ACCOUNT); if (!scratch_va) return -ENOMEM; - if (kvm_read_guest(svm->vcpu.kvm, scratch_gpa_beg, scratch_va, len)) { + if (kvm_read_guest(svm->vcpu.kvm, scratch_gpa_beg, scratch_va, min_len)) { /* Unable to copy scratch area from guest */ pr_err("vmgexit: kvm_read_guest for scratch area failed\n"); @@ -3736,11 +3738,10 @@ static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 len) */ svm->sev_es.ghcb_sa_sync = sync; svm->sev_es.ghcb_sa_free = true; + svm->sev_es.ghcb_sa_len = min_len; } svm->sev_es.ghcb_sa = scratch_va; - svm->sev_es.ghcb_sa_len = len; - return 0; e_scratch: From f185e05dce6f170f83c4ba602e969b1c3c7a22e6 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:32 -0700 Subject: [PATCH 393/411] KVM: SEV: WARN if KVM attempts to setup scratch area with min_len==0 Now that all paths in KVM properly validate the length needed for the scratch area, and are guaranteed to pass in a non-zero length, WARN if KVM attempts to configured the scratch area with min_len==0 to guard against future bugs. Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Reviewed-by: Michael Roth Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-8-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 6072fecfe994..a3e85348ace9 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3669,6 +3669,9 @@ static int setup_vmgexit_scratch(struct vcpu_svm *svm, bool sync, u64 min_len) u64 scratch_gpa_beg, scratch_gpa_end; void *scratch_va; + if (WARN_ON_ONCE(!min_len)) + goto e_scratch; + scratch_gpa_beg = svm->sev_es.sw_scratch; if (!scratch_gpa_beg) { pr_err("vmgexit: scratch gpa not provided\n"); From ebe4b2dc9cfbfb2d8f665667c4d08f4c6c9bec05 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:33 -0700 Subject: [PATCH 394/411] KVM: SEV: Don't explicitly pass PSC buffer to snp_begin_psc() Stop explicitly passing the PSC buffer to snp_begin_psc(): it *must* be the scratch area. This will allow fixing a variety of bugs without further complicating the code. No functional change intended. Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Reviewed-by: Michael Roth Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-9-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index a3e85348ace9..8577451b82b2 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3841,7 +3841,7 @@ struct psc_buffer { struct psc_entry entries[]; } __packed; -static int snp_begin_psc(struct vcpu_svm *svm, struct psc_buffer *psc); +static int snp_begin_psc(struct vcpu_svm *svm); static void snp_complete_psc(struct vcpu_svm *svm, u64 psc_ret) { @@ -3883,7 +3883,6 @@ static void __snp_complete_one_psc(struct vcpu_svm *svm) static int snp_complete_one_psc(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - struct psc_buffer *psc = svm->sev_es.ghcb_sa; if (vcpu->run->hypercall.ret) { snp_complete_psc(svm, VMGEXIT_PSC_ERROR_GENERIC); @@ -3893,11 +3892,13 @@ static int snp_complete_one_psc(struct kvm_vcpu *vcpu) __snp_complete_one_psc(svm); /* Handle the next range (if any). */ - return snp_begin_psc(svm, psc); + return snp_begin_psc(svm); } -static int snp_begin_psc(struct vcpu_svm *svm, struct psc_buffer *psc) +static int snp_begin_psc(struct vcpu_svm *svm) { + struct vcpu_sev_es_state *sev_es = &svm->sev_es; + struct psc_buffer *psc = sev_es->ghcb_sa; struct psc_entry *entries = psc->entries; struct kvm_vcpu *vcpu = &svm->vcpu; struct psc_hdr *hdr = &psc->hdr; @@ -4567,7 +4568,7 @@ int sev_handle_vmgexit(struct kvm_vcpu *vcpu) if (ret) break; - ret = snp_begin_psc(svm, svm->sev_es.ghcb_sa); + ret = snp_begin_psc(svm); break; case SVM_VMGEXIT_AP_CREATION: ret = sev_snp_ap_creation(svm); From 121d88de56bc5c0ba0ce2f6381af67f948a7e7c1 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:34 -0700 Subject: [PATCH 395/411] KVM: SEV: Check PSC request indices against the actual size of the buffer When processing Page State Change (PSC) requests, validate the PSC buffer against the effective size of the scratch area, which could be less than the maximum size if the guest provided a pointer that isn't exactly at the start of the GHCB shared buffer. Fixes: 9b54e248d264 ("KVM: SEV: Add support to handle Page State Change VMGEXIT") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Reviewed-by: Michael Roth Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-10-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 8577451b82b2..6e8cbae2135a 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3903,7 +3903,7 @@ static int snp_begin_psc(struct vcpu_svm *svm) struct kvm_vcpu *vcpu = &svm->vcpu; struct psc_hdr *hdr = &psc->hdr; struct psc_entry entry_start; - u16 idx, idx_start, idx_end; + u16 idx, idx_start, idx_end, max_nr_entries; int npages; bool huge; u64 gfn; @@ -3913,6 +3913,19 @@ static int snp_begin_psc(struct vcpu_svm *svm) return 1; } + /* + * GHCB v2 requires the scratch area to reside within the GHCB itself, + * and PSC requests are only supported for GHCB v2+. Thus it should be + * impossible to exceed the max PSC entry count (which is derived from + * the size of the shared GHCB buffer). + */ + max_nr_entries = (sev_es->ghcb_sa_len - sizeof(struct psc_hdr)) / + sizeof(struct psc_entry); + if (WARN_ON_ONCE(max_nr_entries > VMGEXIT_PSC_MAX_COUNT)) { + snp_complete_psc(svm, VMGEXIT_PSC_ERROR_GENERIC); + return 1; + } + next_range: /* There should be no other PSCs in-flight at this point. */ if (WARN_ON_ONCE(svm->sev_es.psc_inflight)) { @@ -3928,7 +3941,7 @@ static int snp_begin_psc(struct vcpu_svm *svm) idx_start = hdr->cur_entry; idx_end = hdr->end_entry; - if (idx_end >= VMGEXIT_PSC_MAX_COUNT) { + if (idx_end >= max_nr_entries) { snp_complete_psc(svm, VMGEXIT_PSC_ERROR_INVALID_HDR); return 1; } From c8cc238093ca6c99267032f6cfe78f59389f3157 Mon Sep 17 00:00:00 2001 From: Sean Christopherson Date: Fri, 1 May 2026 13:22:35 -0700 Subject: [PATCH 396/411] KVM: SEV: Use READ_ONCE() when reading entries/indices from PSC buffer Use READ_ONCE() when reading entries/indices from the guest-accessible Page State Change buffer to defend against TOCTOU bugs. Don't bother with READ_ONCE()/WRITE_ONCE() for cases where KVM is writing (and not consuming the result!), as the guest isn't supposed to touch the buffer while it's being processed. I.e. using READ_ONCE() is all about protecting against misbehaving guests. Fixes: 9b54e248d264 ("KVM: SEV: Add support to handle Page State Change VMGEXIT") Cc: stable@vger.kernel.org Reviewed-by: Tom Lendacky Signed-off-by: Sean Christopherson Message-ID: <20260501202250.2115252-11-seanjc@google.com> Signed-off-by: Paolo Bonzini --- arch/x86/kvm/svm/sev.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/x86/kvm/svm/sev.c b/arch/x86/kvm/svm/sev.c index 6e8cbae2135a..62b5befe0eed 100644 --- a/arch/x86/kvm/svm/sev.c +++ b/arch/x86/kvm/svm/sev.c @@ -3872,9 +3872,9 @@ static void __snp_complete_one_psc(struct vcpu_svm *svm) */ for (idx = svm->sev_es.psc_idx; svm->sev_es.psc_inflight; svm->sev_es.psc_inflight--, idx++) { - struct psc_entry *entry = &entries[idx]; + struct psc_entry entry = READ_ONCE(entries[idx]); - entry->cur_page = entry->pagesize ? 512 : 1; + entries[idx].cur_page = entry.pagesize ? 512 : 1; } hdr->cur_entry = idx; @@ -3938,8 +3938,8 @@ static int snp_begin_psc(struct vcpu_svm *svm) * validation, so take care to only use validated copies of values used * for things like array indexing. */ - idx_start = hdr->cur_entry; - idx_end = hdr->end_entry; + idx_start = READ_ONCE(hdr->cur_entry); + idx_end = READ_ONCE(hdr->end_entry); if (idx_end >= max_nr_entries) { snp_complete_psc(svm, VMGEXIT_PSC_ERROR_INVALID_HDR); @@ -3948,7 +3948,7 @@ static int snp_begin_psc(struct vcpu_svm *svm) /* Find the start of the next range which needs processing. */ for (idx = idx_start; idx <= idx_end; idx++, hdr->cur_entry++) { - entry_start = entries[idx]; + entry_start = READ_ONCE(entries[idx]); gfn = entry_start.gfn; huge = entry_start.pagesize; @@ -3992,7 +3992,7 @@ static int snp_begin_psc(struct vcpu_svm *svm) * KVM_HC_MAP_GPA_RANGE exit. */ while (++idx <= idx_end) { - struct psc_entry entry = entries[idx]; + struct psc_entry entry = READ_ONCE(entries[idx]); if (entry.operation != entry_start.operation || entry.gfn != entry_start.gfn + npages || From 1e584c304cfb94a759417130b1fc6d30b30c4cce Mon Sep 17 00:00:00 2001 From: Jingguo Tan Date: Wed, 27 May 2026 10:33:01 +0800 Subject: [PATCH 397/411] vsock/virtio: bind uarg before filling zerocopy skb virtio_transport_send_pkt_info() allocates or reuses the zerocopy uarg before entering the send loop, but virtio_transport_alloc_skb() still fills the skb before it inherits that uarg. When fixed-buffer vectored zerocopy hits MAX_SKB_FRAGS, io_sg_from_iter() may partially attach managed frags and return -EMSGSIZE. The rollback path call kfree_skb() to free an skb that carries SKBFL_MANAGED_FRAG_REFS but no uarg, so skb_release_data() falls through to ordinary frag unref. Pass the uarg into virtio_transport_alloc_skb() and bind it immediately before virtio_transport_fill_skb(). This keeps control or no-payload skbs untouched while ensuring success and rollback share one lifetime rule. Fixes: 581512a6dc93 ("vsock/virtio: MSG_ZEROCOPY flag support") Signed-off-by: Lin Ma Signed-off-by: Rongzhen Cui Signed-off-by: Jingguo Tan Acked-by: Arseniy Krasnov Acked-by: Michael S. Tsirkin Reviewed-by: Stefano Garzarella Link: https://patch.msgid.link/20260527023301.1075581-1-malin89@huawei.com Signed-off-by: Jakub Kicinski --- net/vmw_vsock/virtio_transport_common.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/net/vmw_vsock/virtio_transport_common.c b/net/vmw_vsock/virtio_transport_common.c index b143290a311d..b10666937c49 100644 --- a/net/vmw_vsock/virtio_transport_common.c +++ b/net/vmw_vsock/virtio_transport_common.c @@ -205,6 +205,7 @@ static u16 virtio_transport_get_type(struct sock *sk) static struct sk_buff *virtio_transport_alloc_skb(struct virtio_vsock_pkt_info *info, size_t payload_len, bool zcopy, + struct ubuf_info *uarg, u32 src_cid, u32 src_port, u32 dst_cid, @@ -245,6 +246,12 @@ static struct sk_buff *virtio_transport_alloc_skb(struct virtio_vsock_pkt_info * if (info->msg && payload_len > 0) { int err; + /* Bind the zerocopy lifetime before filling frags so error + * rollback frees managed fixed-buffer pages through + * the uarg-aware path. + */ + skb_zcopy_set(skb, uarg, NULL); + err = virtio_transport_fill_skb(skb, info, payload_len, zcopy); if (err) goto out; @@ -364,6 +371,7 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, skb_len = min(max_skb_len, rest_len); skb = virtio_transport_alloc_skb(info, skb_len, can_zcopy, + uarg, src_cid, src_port, dst_cid, dst_port); if (!skb) { @@ -371,8 +379,6 @@ static int virtio_transport_send_pkt_info(struct vsock_sock *vsk, break; } - skb_zcopy_set(skb, uarg, NULL); - virtio_transport_inc_tx_pkt(vvs, skb); ret = t_ops->send_pkt(skb, info->net); @@ -1183,7 +1189,7 @@ static int virtio_transport_reset_no_sock(const struct virtio_transport *t, if (!t) return -ENOTCONN; - reply = virtio_transport_alloc_skb(&info, 0, false, + reply = virtio_transport_alloc_skb(&info, 0, false, NULL, le64_to_cpu(hdr->dst_cid), le32_to_cpu(hdr->dst_port), le64_to_cpu(hdr->src_cid), From f72eed9b84fb771019a955908132410a9ba9ea3f Mon Sep 17 00:00:00 2001 From: Yuqi Xu Date: Wed, 27 May 2026 11:48:15 +0800 Subject: [PATCH 398/411] bpf: sockmap: fix tail fragment offset in bpf_msg_push_data When bpf_msg_push_data() inserts data in the middle of a scatterlist entry, it splits the original entry into a left fragment and a right fragment. The right fragment offset is page-local, but the code advances it with `start`, which is the message-global insertion point. For inserts into a non-first SG entry, this over-advances the offset and leaves the split layout inconsistent. Advance the right fragment offset by the fragment-local delta, `start - offset`, which matches the length removed from the front of the original entry. Fixes: 6fff607e2f14 ("bpf: sk_msg program helper bpf_msg_push_data") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Zhengchuan Liang Reported-by: Xin Liu Signed-off-by: Yuqi Xu Signed-off-by: Ren Wei Link: https://patch.msgid.link/8b129d10566aa3eb43f61a8f9757bcf51707d324.1779636774.git.xuyq21@lenovo.com Signed-off-by: Jakub Kicinski --- net/core/filter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/core/filter.c b/net/core/filter.c index 9590877b0714..80439767e0ee 100644 --- a/net/core/filter.c +++ b/net/core/filter.c @@ -2869,7 +2869,7 @@ BPF_CALL_4(bpf_msg_push_data, struct sk_msg *, msg, u32, start, psge->length = start - offset; rsge.length -= psge->length; - rsge.offset += start; + rsge.offset += start - offset; sk_msg_iter_var_next(i); sg_unmark_end(psge); From 9f72412bcf60144f252b0d6205106abf14344abc Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 27 May 2026 13:31:30 +0800 Subject: [PATCH 399/411] ipv6: fix possible infinite loop in rt6_fill_node() Sashiko reported this issue [1]. Apply the same fix as commit f8d8ce1b515a ("ipv6: fix possible infinite loop in fib6_info_uses_dev()"). Writers holding tb6_lock can list_del_rcu(&rt->fib6_siblings) without waiting for RCU readers; rt->fib6_siblings.next then still points into the old ring and this softirq-side walker never reaches &rt->fib6_siblings, causing a CPU stall. fib6_del_route() always WRITE_ONCE()s rt->fib6_nsiblings to 0 before list_del_rcu(), so an inside-loop check is a reliable detach signal. [1] https://sashiko.dev/#/patchset/20260526020227.4857-1-jiayuan.chen%40linux.dev Fixes: d9ccb18f83ea ("ipv6: Fix soft lockups in fib6_select_path under high next hop churn") Signed-off-by: Jiayuan Chen Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260527053133.180695-1-jiayuan.chen@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index b106e5fef9cb..dad416fdc585 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -5902,6 +5902,8 @@ static int rt6_fill_node(struct net *net, struct sk_buff *skb, goto nla_put_failure; } + if (!READ_ONCE(rt->fib6_nsiblings)) + break; } rcu_read_unlock(); From 9c7da87c2dc860bb17ca1ece942495d28b1ce3b9 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 27 May 2026 13:31:31 +0800 Subject: [PATCH 400/411] ipv6: fix possible infinite loop in fib6_select_path() Found while auditing the same pattern Sashiko reported in rt6_fill_node() [1]. Apply the same fix as commit f8d8ce1b515a ("ipv6: fix possible infinite loop in fib6_info_uses_dev()"). Writers holding tb6_lock can list_del_rcu(&first->fib6_siblings) without waiting for RCU readers; first->fib6_siblings.next then still points into the old ring and this softirq-side walker never reaches &first->fib6_siblings as its terminator. fib6_purge_rt() always WRITE_ONCE()s first->fib6_nsiblings to 0 before list_del_rcu(), so an inside-loop check is a reliable detach signal. [1] https://sashiko.dev/#/patchset/20260526020227.4857-1-jiayuan.chen%40linux.dev Fixes: d9ccb18f83ea ("ipv6: Fix soft lockups in fib6_select_path under high next hop churn") Signed-off-by: Jiayuan Chen Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260527053133.180695-2-jiayuan.chen@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv6/route.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index dad416fdc585..636f0120d7e3 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -481,6 +481,9 @@ void fib6_select_path(const struct net *net, struct fib6_result *res, const struct fib6_nh *nh = sibling->fib6_nh; int nh_upper_bound; + if (!READ_ONCE(first->fib6_nsiblings)) + break; + nh_upper_bound = atomic_read(&nh->fib_nh_upper_bound); if (hash > nh_upper_bound) continue; From ff6e798c2eac3ebd0501ad7e796f583fab928de8 Mon Sep 17 00:00:00 2001 From: Pavel Begunkov Date: Thu, 28 May 2026 19:43:53 +0100 Subject: [PATCH 401/411] net: skbuff: fix pskb_carve leaking zcopy pages When SKBFL_MANAGED_FRAG_REFS is set, frag pages are not refcounted but their lifetime is controlled by the attached ubuf_info. To make a copy of the skb_shared_info, we either should clear the flag and reference the frags, or keep the flag and have frags unreferenced. pskb_carve_inside_header() and pskb_carve_inside_nonlinear() don't follow the rule and thus can leak page references. Let's clear SKBFL_MANAGED_FRAG_REFS from the original skb to fix it. It's the simplest way to address it, but there are more performant ways to do that if it ever becomes a problem. Link: https://lore.kernel.org/all/20260523085809.26331-1-nvminh232@clc.fitus.edu.vn/ Fixes: 753f1ca4e1e50 ("net: introduce managed frags infrastructure") Reported-by: Minh Nguyen Reported-by: Willem de Bruijn Signed-off-by: Pavel Begunkov Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/1e2086aa69217d7f9c8da3d38f5be7160f1b4cd1.1779993185.git.asml.silence@gmail.com Signed-off-by: Jakub Kicinski --- net/core/skbuff.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/net/core/skbuff.c b/net/core/skbuff.c index 0d3cc115f2e7..c02f0a507ba8 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -6823,6 +6823,11 @@ static int pskb_carve_inside_header(struct sk_buff *skb, const u32 off, skb_copy_from_linear_data_offset(skb, off, data, new_hlen); skb->len -= off; + /* Remove SKBFL_MANAGED_FRAG_REFS instead of trying to honour it + * while refcounting frags below. + */ + skb_zcopy_downgrade_managed(skb); + memcpy((struct skb_shared_info *)(data + size), skb_shinfo(skb), offsetof(struct skb_shared_info, @@ -6936,6 +6941,11 @@ static int pskb_carve_inside_nonlinear(struct sk_buff *skb, const u32 off, return -ENOMEM; size = SKB_WITH_OVERHEAD(size); + /* Remove SKBFL_MANAGED_FRAG_REFS instead of trying to honour it + * while refcounting frags below. + */ + skb_zcopy_downgrade_managed(skb); + memcpy((struct skb_shared_info *)(data + size), skb_shinfo(skb), offsetof(struct skb_shared_info, frags[0])); if (skb_orphan_frags(skb, gfp_mask)) { From 072aa0f5c3d8f11f3159037418ec45edce7440b8 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Fri, 29 May 2026 13:23:57 +0200 Subject: [PATCH 402/411] Revert "ipv6: preserve insertion order for same-scope addresses" Chris Adams reported that preserving insertion order for same-scope addresses is causing SSH connections to be dropped after stopping a VM while running NetworkManager. NetworkManager caches the IPv6 address configuration, when a RA arrives, it determines the list of addresses to configure and checks if the addresses are already in the right order in the kernel. If they aren't, NetworkManager removes and re-adds them to achieve the desired order. As the order changes, NetworkManager is confused and reconfigures the addresses on every update. In addition, this would also affect to cloud tooling that relies on IPv6 addresses order to identify primary and secondaries addresses. This reverts commit cb3de96eea66f5e4a580086c6a1be46e765f97f4. Fixes: cb3de96eea66 ("ipv6: preserve insertion order for same-scope addresses") Reported-by: Chris Adams Closes: https://lore.kernel.org/netdev/20260521135310.GC977@cmadams.net/ Signed-off-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20260529112357.5079-1-fmancera@suse.de Signed-off-by: Jakub Kicinski --- net/ipv6/addrconf.c | 2 +- tools/testing/selftests/net/ioam6.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 5476b6536eb7..bb84a78b80f6 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -1013,7 +1013,7 @@ ipv6_link_dev_addr(struct inet6_dev *idev, struct inet6_ifaddr *ifp) list_for_each(p, &idev->addr_list) { struct inet6_ifaddr *ifa = list_entry(p, struct inet6_ifaddr, if_list); - if (ifp_scope > ipv6_addr_src_scope(&ifa->addr)) + if (ifp_scope >= ipv6_addr_src_scope(&ifa->addr)) break; } diff --git a/tools/testing/selftests/net/ioam6.sh b/tools/testing/selftests/net/ioam6.sh index b2b99889942f..845c26dd01a9 100755 --- a/tools/testing/selftests/net/ioam6.sh +++ b/tools/testing/selftests/net/ioam6.sh @@ -273,8 +273,8 @@ setup() ip -netns $ioam_node_beta link set ioam-veth-betaR name veth1 &>/dev/null ip -netns $ioam_node_gamma link set ioam-veth-gamma name veth0 &>/dev/null - ip -netns $ioam_node_alpha addr add 2001:db8:1::2/64 dev veth0 &>/dev/null ip -netns $ioam_node_alpha addr add 2001:db8:1::50/64 dev veth0 &>/dev/null + ip -netns $ioam_node_alpha addr add 2001:db8:1::2/64 dev veth0 &>/dev/null ip -netns $ioam_node_alpha link set veth0 up &>/dev/null ip -netns $ioam_node_alpha link set lo up &>/dev/null ip -netns $ioam_node_alpha route add 2001:db8:2::/64 \ From f75e3eb08fe31d30a9af6ed80cdd22e6772837e2 Mon Sep 17 00:00:00 2001 From: "Jason A. Donenfeld" Date: Fri, 29 May 2026 19:31:34 +0200 Subject: [PATCH 403/411] wireguard: send: append trailer after expanding head With how this is currently written, we add the trailer, zero it out, and then add the header space on. If that header space requires a reallocation + copy, the zeros in the trailer aren't copied, because the skb len hasn't actually been yet expanded to cover that. Instead add the padding at the end of the process rather than at the beginning. Fixes: e7096c131e51 ("net: WireGuard secure network tunnel") Cc: stable@vger.kernel.org Signed-off-by: Jason A. Donenfeld Link: https://patch.msgid.link/20260529173134.3080773-2-Jason@zx2c4.com Signed-off-by: Jakub Kicinski --- drivers/net/wireguard/send.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/drivers/net/wireguard/send.c b/drivers/net/wireguard/send.c index 26e09c30d596..67d01478eb76 100644 --- a/drivers/net/wireguard/send.c +++ b/drivers/net/wireguard/send.c @@ -177,16 +177,6 @@ static bool encrypt_packet(struct sk_buff *skb, struct noise_keypair *keypair) trailer_len = padding_len + noise_encrypted_len(0); plaintext_len = skb->len + padding_len; - /* Expand data section to have room for padding and auth tag. */ - num_frags = skb_cow_data(skb, trailer_len, &trailer); - if (unlikely(num_frags < 0 || num_frags > ARRAY_SIZE(sg))) - return false; - - /* Set the padding to zeros, and make sure it and the auth tag are part - * of the skb. - */ - memset(skb_tail_pointer(trailer), 0, padding_len); - /* Expand head section to have room for our header and the network * stack's headers. */ @@ -198,6 +188,16 @@ static bool encrypt_packet(struct sk_buff *skb, struct noise_keypair *keypair) skb_checksum_help(skb))) return false; + /* Expand data section to have room for padding and auth tag. */ + num_frags = skb_cow_data(skb, trailer_len, &trailer); + if (unlikely(num_frags < 0 || num_frags > ARRAY_SIZE(sg))) + return false; + + /* Set the padding to zeros, and make sure it and the auth tag are part + * of the skb. + */ + memset(skb_tail_pointer(trailer), 0, padding_len); + /* Only after checksumming can we safely add on the padding at the end * and the header. */ From 7164d78559b0ff29931a366a840a9e5dd53d4b7c Mon Sep 17 00:00:00 2001 From: Zhenghang Xiao Date: Tue, 26 May 2026 16:53:13 +0800 Subject: [PATCH 404/411] drm/gem: fix race between change_handle and handle_delete drm_gem_change_handle_ioctl leaves the old handle live in the IDR during the window between spin_unlock(table_lock) and the final spin_lock(table_lock). A concurrent drm_gem_handle_delete on the old handle succeeds in this window, decrements handle_count to 0, and frees the GEM object while the new handle's IDR entry still references it. NULL the old handle's IDR entry before dropping table_lock so that any concurrent GEM_CLOSE on the old handle sees NULL and returns -EINVAL. Restore the old entry on the prime-bookkeeping error path. Fixes: 5e28b7b94408 ("drm: Set old handle to NULL before prime swap in change_handle") Signed-off-by: Zhenghang Xiao Cc: stable@vger.kernel.org Signed-off-by: Dave Airlie Link: https://patch.msgid.link/20260526085313.26791-1-kipreyyy@gmail.com --- drivers/gpu/drm/drm_gem.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index e3b8a1f353cb..e12cdf91f4dc 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -1065,6 +1065,7 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data, goto out_unlock; } + idr_replace(&file_priv->object_idr, NULL, args->handle); spin_unlock(&file_priv->table_lock); if (obj->dma_buf) { @@ -1073,6 +1074,7 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data, if (ret < 0) { spin_lock(&file_priv->table_lock); idr_remove(&file_priv->object_idr, handle); + idr_replace(&file_priv->object_idr, obj, args->handle); spin_unlock(&file_priv->table_lock); goto out_unlock; } From 44eeff9bc467bc7d1fec34fc3f6001f385fe462c Mon Sep 17 00:00:00 2001 From: Andrei Vagin Date: Tue, 26 May 2026 20:50:43 +0000 Subject: [PATCH 405/411] Revert "x86/fpu: Refine and simplify the magic number check during signal return" This reverts dc8aa31a7ac2 ("x86/fpu: Refine and simplify the magic number check during signal return"). The aforementioned commit broke applications that construct signal frames in userspace (such as CRIU and gVisor) if the frame's xstate size is smaller than the kernel's fpstate->user_size. Furthermore, this introduces a critical issue for checkpoint/restore tools like CRIU. If a process is checkpointed while inside a signal handler, its stack contains a signal frame formatted according to the source host's xstate capabilities. If that process is later restored on a destination host with larger xstate capabilities (e.g., a newer CPU with more features enabled, resulting in a larger fpstate->user_size), the kernel will look for FP_XSTATE_MAGIC2 at the destination host's larger user_size offset instead of the offset encoded in the frame's fx_sw->xstate_size. This causes the magic2 check to fail, forcing sigreturn to silently fall back to "FX-only" mode. Upon return from the signal handler, the process's extended state is reset to initial values instead of being restored, leading to silent data corruption. The aforementioned commit cited d877550eaf2d ("x86/fpu: Stop relying on userspace for info to fault in xsave buffer") as justification to stop relying on userspace for the magic number check. However, these two changes are fundamentally different. The last one only changed how much memory the kernel ensures is paged-in before running XRSTOR to prevent an infinite loop. It did not change the signal frame format or how the layout is validated. Reverting this change restores the use of fx_sw->xstate_size for locating magic2 and restores the necessary sanity checks, ensuring that the signal frame remains self-describing and portable. [ bp: Massage commit message. ] Fixes: dc8aa31a7ac2 ("x86/fpu: Refine and simplify the magic number check during signal return") Signed-off-by: Andrei Vagin Signed-off-by: Borislav Petkov (AMD) Acked-by: Chang S. Bae Cc: stable@vger.kernel.org Link: https://lore.kernel.org/all/20260429000623.3356606-1-avagin@google.com --- arch/x86/kernel/fpu/signal.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/arch/x86/kernel/fpu/signal.c b/arch/x86/kernel/fpu/signal.c index c3ec2512f2bb..20b638c507ca 100644 --- a/arch/x86/kernel/fpu/signal.c +++ b/arch/x86/kernel/fpu/signal.c @@ -27,14 +27,19 @@ static inline bool check_xstate_in_sigframe(struct fxregs_state __user *fxbuf, struct _fpx_sw_bytes *fx_sw) { + int min_xstate_size = sizeof(struct fxregs_state) + + sizeof(struct xstate_header); void __user *fpstate = fxbuf; unsigned int magic2; if (__copy_from_user(fx_sw, &fxbuf->sw_reserved[0], sizeof(*fx_sw))) return false; - /* Check for the first magic field */ - if (fx_sw->magic1 != FP_XSTATE_MAGIC1) + /* Check for the first magic field and other error scenarios. */ + if (fx_sw->magic1 != FP_XSTATE_MAGIC1 || + fx_sw->xstate_size < min_xstate_size || + fx_sw->xstate_size > x86_task_fpu(current)->fpstate->user_size || + fx_sw->xstate_size > fx_sw->extended_size) goto setfx; /* @@ -43,7 +48,7 @@ static inline bool check_xstate_in_sigframe(struct fxregs_state __user *fxbuf, * fpstate layout with out copying the extended state information * in the memory layout. */ - if (__get_user(magic2, (__u32 __user *)(fpstate + x86_task_fpu(current)->fpstate->user_size))) + if (__get_user(magic2, (__u32 __user *)(fpstate + fx_sw->xstate_size))) return false; if (likely(magic2 == FP_XSTATE_MAGIC2)) From 05d5d79440c2cc0784f91b61f2012753e66be472 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Sat, 30 May 2026 12:25:36 +0200 Subject: [PATCH 406/411] Revert "gpib: cb7210: Fix region leak when request_irq fails" MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This reverts commit 2eae90a457baa0048a96ed38ad93090ee38c8b2f. Turns out not to be correct. Link: https://lore.kernel.org/r/PpNUbGhrvT8I_KayoDvQYI2PYjmMw1QEkuVBDZz2PwBsVVgPkBXJarc2mBM0IhiH3AQG0GtgqEsDRXNj3yUKEDBaZa25u73pAjvcE6vfRsg=@protonmail.com Reported-by: Dominik Karol Piątkowski Cc: Mark Brown Cc: Hongling Zeng Cc: Hongling Zeng Signed-off-by: Greg Kroah-Hartman --- drivers/gpib/cb7210/cb7210.c | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/drivers/gpib/cb7210/cb7210.c b/drivers/gpib/cb7210/cb7210.c index 673b5bfe2e7d..6dd8637c5964 100644 --- a/drivers/gpib/cb7210/cb7210.c +++ b/drivers/gpib/cb7210/cb7210.c @@ -1049,8 +1049,7 @@ static int cb_isa_attach(struct gpib_board *board, const struct gpib_board_confi if (!request_region(config->ibbase, cb7210_iosize, DRV_NAME)) { dev_err(board->gpib_dev, "ioports starting at 0x%x are already in use\n", config->ibbase); - retval = -EBUSY; - goto err_release_region; + return -EBUSY; } nec_priv->iobase = config->ibbase; cb_priv->fifo_iobase = nec7210_iobase(cb_priv); @@ -1063,16 +1062,11 @@ static int cb_isa_attach(struct gpib_board *board, const struct gpib_board_confi // install interrupt handler if (request_irq(config->ibirq, cb7210_interrupt, isr_flags, DRV_NAME, board)) { dev_err(board->gpib_dev, "failed to obtain IRQ %d\n", config->ibirq); - retval = -EBUSY; - goto err_release_region; + return -EBUSY; } cb_priv->irq = config->ibirq; return cb7210_init(cb_priv, board); - -err_release_region: - release_region(nec7210_iobase(cb_priv), cb7210_iosize); - return retval; } static void cb_isa_detach(struct gpib_board *board) From 1d774589f924056b8403e271fdecaf9a803a50fc Mon Sep 17 00:00:00 2001 From: Alexis Bouzigues Date: Fri, 29 May 2026 09:28:14 -0500 Subject: [PATCH 407/411] i2c: virtio: mark device ready before registering the adapter virtio_i2c_probe() synchronously probes child i2c drivers on the bus, but peripherals may use the bus at probe for tasks like reading a chip id. The vhost-user-i2c backend stalls at such probes unless DRIVER_OK is already set before the virtqueue is first kicked. Set DRIVER_OK explicitly before i2c_add_adapter(), as done for the same reason in commit f5866db64f34 ("virtio_console: enable VQs early") and commit 71e4b8bf0482 ("virtio_rpmsg: set DRIVER_OK before using device"). Signed-off-by: Alexis Bouzigues Signed-off-by: Wolfram Sang --- drivers/i2c/busses/i2c-virtio.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/i2c/busses/i2c-virtio.c b/drivers/i2c/busses/i2c-virtio.c index 7b0b0bff8000..5da6fef92bec 100644 --- a/drivers/i2c/busses/i2c-virtio.c +++ b/drivers/i2c/busses/i2c-virtio.c @@ -222,6 +222,8 @@ static int virtio_i2c_probe(struct virtio_device *vdev) */ ACPI_COMPANION_SET(&vi->adap.dev, ACPI_COMPANION(vdev->dev.parent)); + virtio_device_ready(vdev); + ret = i2c_add_adapter(&vi->adap); if (ret) virtio_i2c_del_vqs(vdev); From 171022c7d594c133a45f92357a2a91475edabe20 Mon Sep 17 00:00:00 2001 From: Henri A Date: Wed, 20 May 2026 10:25:44 -0400 Subject: [PATCH 408/411] media: rc: igorplugusb: fix control request setup packet Commit eac69475b01f ("media: rc: igorplugusb: heed coherency rules") changed the control request storage from an embedded struct to an allocated pointer so it can obey DMA coherency rules. However, the driver still passes &ir->request to usb_fill_control_urb(). That points the URB setup packet at the pointer field itself rather than at the allocated struct usb_ctrlrequest. USB core then interprets pointer bytes as the setup packet. This can produce an invalid bRequestType and trigger the control direction warning reported by syzbot: usb 2-1: BOGUS control dir, pipe 80003580 doesn't match bRequestType 0 Pass ir->request itself as the setup packet. Fixes: eac69475b01f ("media: rc: igorplugusb: heed coherency rules") Reported-by: syzbot+11f0e4f957c7c3bf3d51@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=11f0e4f957c7c3bf3d51 Tested-by: syzbot+11f0e4f957c7c3bf3d51@syzkaller.appspotmail.com Cc: stable@vger.kernel.org Assisted-by: Codex:GPT-5.5 Signed-off-by: Henri A Signed-off-by: Sean Young Signed-off-by: Hans Verkuil --- drivers/media/rc/igorplugusb.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/rc/igorplugusb.c b/drivers/media/rc/igorplugusb.c index 3e10f6fe89f8..b5117ee9f5fa 100644 --- a/drivers/media/rc/igorplugusb.c +++ b/drivers/media/rc/igorplugusb.c @@ -184,7 +184,7 @@ static int igorplugusb_probe(struct usb_interface *intf, if (!ir->buf_in) goto fail; usb_fill_control_urb(ir->urb, udev, - usb_rcvctrlpipe(udev, 0), (uint8_t *)&ir->request, + usb_rcvctrlpipe(udev, 0), (uint8_t *)ir->request, ir->buf_in, MAX_PACKET, igorplugusb_callback, ir); usb_make_path(udev, ir->phys, sizeof(ir->phys)); From e43ffb69e0438cddd72aaa30898b4dc446f664f8 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 31 May 2026 15:14:24 -0700 Subject: [PATCH 409/411] Linux 7.1-rc6 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index f056c921ea9c..d8da45199699 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 7 PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = -rc5 +EXTRAVERSION = -rc6 NAME = Baby Opossum Posse # *DOCUMENTATION* From caba9259a22b4da6beeeb48c84292870dbcff978 Mon Sep 17 00:00:00 2001 From: Alexey Charkov Date: Mon, 25 May 2026 13:20:45 +0400 Subject: [PATCH 410/411] ASoC: dt-bindings: nau8822: Add supply regulators NAU8822 has 4 power supply pins: VDDA, VDDB, VDDC and VDDSPK, which need to be online and stable before communication with the device is attempted. List them (as optional) so that device tree users can ensure correct power sequencing. Signed-off-by: Alexey Charkov Link: https://patch.msgid.link/20260525-nau8822-reg-v2-1-7d37ae393e46@flipper.net Signed-off-by: Mark Brown --- .../bindings/sound/nuvoton,nau8822.yaml | 18 ++++++++++++++++++ 1 file changed, 18 insertions(+) diff --git a/Documentation/devicetree/bindings/sound/nuvoton,nau8822.yaml b/Documentation/devicetree/bindings/sound/nuvoton,nau8822.yaml index cb8182bbc491..cf4c13038241 100644 --- a/Documentation/devicetree/bindings/sound/nuvoton,nau8822.yaml +++ b/Documentation/devicetree/bindings/sound/nuvoton,nau8822.yaml @@ -30,6 +30,20 @@ properties: clock-names: const: mclk + vdda-supply: + description: Analog power supply + + vddb-supply: + description: Digital buffer (input/output) supply + + vddc-supply: + description: Digital core supply + + vddspk-supply: + description: + Speaker supply (power supply pin for RSPKOUT, LSPKOUT, AUXOUT2 and + AUXTOUT1 output drivers) + nuvoton,spk-btl: description: If set, configure the two loudspeaker outputs as a Bridge Tied Load output @@ -54,5 +68,9 @@ examples: codec@1a { compatible = "nuvoton,nau8822"; reg = <0x1a>; + vdda-supply = <&vcca_3v3_s0>; + vddb-supply = <&vcca_3v3_s0>; + vddc-supply = <&vcca_3v3_s0>; + vddspk-supply = <&vcca_3v3_s0>; }; }; From f8e7cd48e5b3b38ec5e1542b73b670a7770d2d7c Mon Sep 17 00:00:00 2001 From: Alexey Charkov Date: Mon, 25 May 2026 13:20:46 +0400 Subject: [PATCH 411/411] ASoC: codecs: nau8822: add support for supply regulators NAU8822 has four power supply pins: VDDA, VDDB, VDDC, and VDDSPK, which need to be online and stable before communication with the device is attempted. Request and enable these regulators at init time, if provided. Also wait for 100 us after powering up the supply regulators before attempting to access the device registers, as recommended by the datasheet. This helps avoid -ENXIO errors when the codec is probed before the regulators are ready. Signed-off-by: Alexey Charkov Link: https://patch.msgid.link/20260525-nau8822-reg-v2-2-7d37ae393e46@flipper.net Signed-off-by: Mark Brown --- sound/soc/codecs/nau8822.c | 46 +++++++++++++++++++++++++++++++++++--- sound/soc/codecs/nau8822.h | 3 +++ 2 files changed, 46 insertions(+), 3 deletions(-) diff --git a/sound/soc/codecs/nau8822.c b/sound/soc/codecs/nau8822.c index a11759f85eac..cd4a7de47939 100644 --- a/sound/soc/codecs/nau8822.c +++ b/sound/soc/codecs/nau8822.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include #include @@ -108,6 +109,10 @@ static const struct reg_default nau8822_reg_defaults[] = { { NAU8822_REG_OUTPUT_TIEOFF, 0x0000 }, }; +static const char * const nau8822_supply_names[NAU8822_NUM_SUPPLIES] = { + "vdda", "vddb", "vddc", "vddspk", +}; + static bool nau8822_readable_reg(struct device *dev, unsigned int reg) { switch (reg) { @@ -1056,6 +1061,7 @@ static int nau8822_suspend(struct snd_soc_component *component) struct snd_soc_dapm_context *dapm = snd_soc_component_to_dapm(component); snd_soc_dapm_force_bias_level(dapm, SND_SOC_BIAS_OFF); + regulator_bulk_disable(NAU8822_NUM_SUPPLIES, nau8822->supplies); regcache_mark_dirty(nau8822->regmap); @@ -1066,6 +1072,15 @@ static int nau8822_resume(struct snd_soc_component *component) { struct nau8822 *nau8822 = snd_soc_component_get_drvdata(component); struct snd_soc_dapm_context *dapm = snd_soc_component_to_dapm(component); + int ret = regulator_bulk_enable(NAU8822_NUM_SUPPLIES, nau8822->supplies); + + if (ret) { + dev_err(component->dev, + "Failed to enable regulators: %d\n", ret); + return ret; + } + + fsleep(100); regcache_sync(nau8822->regmap); @@ -1153,7 +1168,7 @@ static int nau8822_i2c_probe(struct i2c_client *i2c) { struct device *dev = &i2c->dev; struct nau8822 *nau8822 = dev_get_platdata(dev); - int ret; + int ret, i; if (!nau8822) { nau8822 = devm_kzalloc(dev, sizeof(*nau8822), GFP_KERNEL); @@ -1167,6 +1182,13 @@ static int nau8822_i2c_probe(struct i2c_client *i2c) return dev_err_probe(&i2c->dev, PTR_ERR(nau8822->mclk), "Error getting mclk\n"); + for (i = 0; i < NAU8822_NUM_SUPPLIES; i++) + nau8822->supplies[i].supply = nau8822_supply_names[i]; + + ret = devm_regulator_bulk_get(dev, NAU8822_NUM_SUPPLIES, nau8822->supplies); + if (ret) + return dev_err_probe(dev, ret, "Failed to get regulators\n"); + nau8822->regmap = devm_regmap_init_i2c(i2c, &nau8822_regmap_config); if (IS_ERR(nau8822->regmap)) { ret = PTR_ERR(nau8822->regmap); @@ -1175,21 +1197,38 @@ static int nau8822_i2c_probe(struct i2c_client *i2c) } nau8822->dev = dev; + ret = regulator_bulk_enable(NAU8822_NUM_SUPPLIES, nau8822->supplies); + if (ret) + return dev_err_probe(dev, ret, "Failed to enable regulators\n"); + + fsleep(100); + /* Reset the codec */ ret = regmap_write(nau8822->regmap, NAU8822_REG_RESET, 0x00); if (ret != 0) { dev_err(&i2c->dev, "Failed to issue reset: %d\n", ret); - return ret; + goto err_reg; } ret = devm_snd_soc_register_component(dev, &soc_component_dev_nau8822, &nau8822_dai, 1); if (ret != 0) { dev_err(&i2c->dev, "Failed to register CODEC: %d\n", ret); - return ret; + goto err_reg; } return 0; + +err_reg: + regulator_bulk_disable(NAU8822_NUM_SUPPLIES, nau8822->supplies); + return ret; +} + +static void nau8822_i2c_remove(struct i2c_client *i2c) +{ + struct nau8822 *nau8822 = i2c_get_clientdata(i2c); + + regulator_bulk_disable(NAU8822_NUM_SUPPLIES, nau8822->supplies); } static const struct i2c_device_id nau8822_i2c_id[] = { @@ -1212,6 +1251,7 @@ static struct i2c_driver nau8822_i2c_driver = { .of_match_table = of_match_ptr(nau8822_of_match), }, .probe = nau8822_i2c_probe, + .remove = nau8822_i2c_remove, .id_table = nau8822_i2c_id, }; module_i2c_driver(nau8822_i2c_driver); diff --git a/sound/soc/codecs/nau8822.h b/sound/soc/codecs/nau8822.h index 13fe0a091e9e..24799c7b5931 100644 --- a/sound/soc/codecs/nau8822.h +++ b/sound/soc/codecs/nau8822.h @@ -211,6 +211,8 @@ struct nau8822_pll { int freq_out; }; +#define NAU8822_NUM_SUPPLIES 4 + /* Codec Private Data */ struct nau8822 { struct device *dev; @@ -219,6 +221,7 @@ struct nau8822 { struct nau8822_pll pll; int sysclk; int div_id; + struct regulator_bulk_data supplies[NAU8822_NUM_SUPPLIES]; }; #endif /* __NAU8822_H__ */