From 4ced4cf5c9d172d91f181df3accdf949d3761aab Mon Sep 17 00:00:00 2001 From: Andrei Vagin Date: Tue, 17 Feb 2026 18:01:05 +0000 Subject: [PATCH 001/352] binfmt_elf_fdpic: fix AUXV size calculation for ELF_HWCAP3 and ELF_HWCAP4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Commit 4e6e8c2b757f ("binfmt_elf: Wire up AT_HWCAP3 at AT_HWCAP4") added support for AT_HWCAP3 and AT_HWCAP4, but it missed updating the AUX vector size calculation in create_elf_fdpic_tables() and AT_VECTOR_SIZE_BASE in include/linux/auxvec.h. Similar to the fix for AT_HWCAP2 in commit c6a09e342f8e ("binfmt_elf_fdpic: fix AUXV size calculation when ELF_HWCAP2 is defined"), this omission leads to a mismatch between the reserved space and the actual number of AUX entries, eventually triggering a kernel BUG_ON(csp != sp). Fix this by incrementing nitems when ELF_HWCAP3 or ELF_HWCAP4 are defined and updating AT_VECTOR_SIZE_BASE. Cc: Mark Brown Cc: Max Filippov Reviewed-by: Michal Koutný Reviewed-by: Mark Brown Reviewed-by: Cyrill Gorcunov Reviewed-by: Alexander Mikhalitsyn Fixes: 4e6e8c2b757f ("binfmt_elf: Wire up AT_HWCAP3 at AT_HWCAP4") Signed-off-by: Andrei Vagin Link: https://patch.msgid.link/20260217180108.1420024-2-avagin@google.com Signed-off-by: Kees Cook --- fs/binfmt_elf_fdpic.c | 6 ++++++ include/linux/auxvec.h | 2 +- 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index 95b1d0852135..95b65aab7daa 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -595,6 +595,12 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, #ifdef ELF_HWCAP2 nitems++; #endif +#ifdef ELF_HWCAP3 + nitems++; +#endif +#ifdef ELF_HWCAP4 + nitems++; +#endif csp = sp; sp -= nitems * 2 * sizeof(unsigned long); diff --git a/include/linux/auxvec.h b/include/linux/auxvec.h index 407f7005e6d6..8bcb9b726262 100644 --- a/include/linux/auxvec.h +++ b/include/linux/auxvec.h @@ -4,6 +4,6 @@ #include -#define AT_VECTOR_SIZE_BASE 22 /* NEW_AUX_ENT entries in auxiliary table */ +#define AT_VECTOR_SIZE_BASE 24 /* NEW_AUX_ENT entries in auxiliary table */ /* number of "#define AT_.*" above, minus {AT_NULL, AT_IGNORE, AT_NOTELF} */ #endif /* _LINUX_AUXVEC_H */ From f69cfd8e8fd13b5d57e638fa1542fcd56f594ef0 Mon Sep 17 00:00:00 2001 From: Wei Liu Date: Tue, 24 Feb 2026 17:08:28 +0000 Subject: [PATCH 002/352] x86/hyperv: print out reserved vectors in hexadecimal Signed-off-by: Wei Liu --- arch/x86/kernel/cpu/mshyperv.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c index 89a2eb8a0722..9befdc557d9e 100644 --- a/arch/x86/kernel/cpu/mshyperv.c +++ b/arch/x86/kernel/cpu/mshyperv.c @@ -496,8 +496,9 @@ static void hv_reserve_irq_vectors(void) test_and_set_bit(HYPERV_DBG_FASTFAIL_VECTOR, system_vectors)) BUG(); - pr_info("Hyper-V: reserve vectors: %d %d %d\n", HYPERV_DBG_ASSERT_VECTOR, - HYPERV_DBG_SERVICE_VECTOR, HYPERV_DBG_FASTFAIL_VECTOR); + pr_info("Hyper-V: reserve vectors: 0x%x 0x%x 0x%x\n", + HYPERV_DBG_ASSERT_VECTOR, HYPERV_DBG_SERVICE_VECTOR, + HYPERV_DBG_FASTFAIL_VECTOR); } static void __init ms_hyperv_init_platform(void) From aa8a3f3c67235422a0c3608a8772f69ca3b7b63f Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 24 Feb 2026 00:05:11 +0100 Subject: [PATCH 003/352] xfrm: add missing extack for XFRMA_SA_PCPU in add_acquire and allocspi We're returning an error caused by invalid user input without setting an extack. Add one. Fixes: 1ddf9916ac09 ("xfrm: Add support for per cpu xfrm state handling.") Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 403b5ecac2c5..3e6477c6082e 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1850,6 +1850,7 @@ static int xfrm_alloc_userspi(struct sk_buff *skb, struct nlmsghdr *nlh, pcpu_num = nla_get_u32(attrs[XFRMA_SA_PCPU]); if (pcpu_num >= num_possible_cpus()) { err = -EINVAL; + NL_SET_ERR_MSG(extack, "pCPU number too big"); goto out_noput; } } @@ -3001,8 +3002,10 @@ static int xfrm_add_acquire(struct sk_buff *skb, struct nlmsghdr *nlh, if (attrs[XFRMA_SA_PCPU]) { x->pcpu_num = nla_get_u32(attrs[XFRMA_SA_PCPU]); err = -EINVAL; - if (x->pcpu_num >= num_possible_cpus()) + if (x->pcpu_num >= num_possible_cpus()) { + NL_SET_ERR_MSG(extack, "pCPU number too big"); goto free_state; + } } err = verify_newpolicy_info(&ua->policy, extack); From b57defcf8f109da5ba9cf59b2a736606faf3d846 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 24 Feb 2026 00:05:12 +0100 Subject: [PATCH 004/352] xfrm: fix the condition on x->pcpu_num in xfrm_sa_len pcpu_num = 0 is a valid value. The marker for "unset pcpu_num" which makes copy_to_user_state_extra not add the XFRMA_SA_PCPU attribute is UINT_MAX. Fixes: 1ddf9916ac09 ("xfrm: Add support for per cpu xfrm state handling.") Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_user.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 3e6477c6082e..4dd8341225bc 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -3676,7 +3676,7 @@ static inline unsigned int xfrm_sa_len(struct xfrm_state *x) } if (x->if_id) l += nla_total_size(sizeof(x->if_id)); - if (x->pcpu_num) + if (x->pcpu_num != UINT_MAX) l += nla_total_size(sizeof(x->pcpu_num)); /* Must count x->lastused as it may become non-zero behind our back. */ From 7d2fc41f91bc69acb6e01b0fa23cd7d0109a6a23 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 24 Feb 2026 00:05:13 +0100 Subject: [PATCH 005/352] xfrm: call xdo_dev_state_delete during state update When we update an SA, we construct a new state and call xdo_dev_state_add, but never insert it. The existing state is updated, then we immediately destroy the new state. Since we haven't added it, we don't go through the standard state delete code, and we're skipping removing it from the device (but xdo_dev_state_free will get called when we destroy the temporary state). This is similar to commit c5d4d7d83165 ("xfrm: Fix deletion of offloaded SAs on failure."). Fixes: d77e38e612a0 ("xfrm: Add an IPsec hardware offloading API") Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 98b362d51836..a00c4fe1ab0c 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -2264,6 +2264,7 @@ int xfrm_state_update(struct xfrm_state *x) err = 0; x->km.state = XFRM_STATE_DEAD; + xfrm_dev_state_delete(x); __xfrm_state_put(x); } From 0c0eef8ccd2413b0a10eb6bbd3442333b1e64dd2 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Tue, 24 Feb 2026 00:05:14 +0100 Subject: [PATCH 006/352] esp: fix skb leak with espintcp and async crypto When the TX queue for espintcp is full, esp_output_tail_tcp will return an error and not free the skb, because with synchronous crypto, the common xfrm output code will drop the packet for us. With async crypto (esp_output_done), we need to drop the skb when esp_output_tail_tcp returns an error. Fixes: e27cca96cd68 ("xfrm: add espintcp (RFC 8229)") Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/ipv4/esp4.c | 9 ++++++--- net/ipv6/esp6.c | 9 ++++++--- 2 files changed, 12 insertions(+), 6 deletions(-) diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c index 2c922afadb8f..6dfc0bcdef65 100644 --- a/net/ipv4/esp4.c +++ b/net/ipv4/esp4.c @@ -235,10 +235,13 @@ static void esp_output_done(void *data, int err) xfrm_dev_resume(skb); } else { if (!err && - x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) - esp_output_tail_tcp(x, skb); - else + x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) { + err = esp_output_tail_tcp(x, skb); + if (err != -EINPROGRESS) + kfree_skb(skb); + } else { xfrm_output_resume(skb_to_full_sk(skb), skb, err); + } } } diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index e75da98f5283..9f75313734f8 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -271,10 +271,13 @@ static void esp_output_done(void *data, int err) xfrm_dev_resume(skb); } else { if (!err && - x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) - esp_output_tail_tcp(x, skb); - else + x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) { + err = esp_output_tail_tcp(x, skb); + if (err != -EINPROGRESS) + kfree_skb(skb); + } else { xfrm_output_resume(skb_to_full_sk(skb), skb, err); + } } } From 2b476739f93d286dc7c2b9d14301eaccccd789d3 Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Fri, 13 Feb 2026 17:02:33 +0000 Subject: [PATCH 007/352] MAINTAINERS: remove Tudor Ambarus as SPI NOR maintainer I have not been actively involved in SPI NOR development recently and would like to step down to focus on my current day-to-day work. The subsystem remains in good hands with Pratyush and Michael. Signed-off-by: Tudor Ambarus Acked-by: Michael Walle Signed-off-by: Miquel Raynal --- MAINTAINERS | 1 - 1 file changed, 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 55af015174a5..e1694b2c3400 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -24909,7 +24909,6 @@ F: drivers/clk/spear/ F: drivers/pinctrl/spear/ SPI NOR SUBSYSTEM -M: Tudor Ambarus M: Pratyush Yadav M: Michael Walle L: linux-mtd@lists.infradead.org From 82d938d5266256d2f4d90ed3733a421493eb3623 Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Fri, 13 Feb 2026 17:02:34 +0000 Subject: [PATCH 008/352] MAINTAINERS: add Takahiro Kuwano as SPI NOR reviewer Takahiro has been an active contributor to the SPI NOR subsystem, providing valuable patches and reviews. Add him as a designated reviewer to help facilitate patch processing and maintenance. Cc: Takahiro Kuwano Signed-off-by: Tudor Ambarus Acked-by: Takahiro Kuwano Acked-by: Michael Walle Signed-off-by: Miquel Raynal --- MAINTAINERS | 1 + 1 file changed, 1 insertion(+) diff --git a/MAINTAINERS b/MAINTAINERS index e1694b2c3400..fc1b0d70fd1f 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -24911,6 +24911,7 @@ F: drivers/pinctrl/spear/ SPI NOR SUBSYSTEM M: Pratyush Yadav M: Michael Walle +R: Takahiro Kuwano L: linux-mtd@lists.infradead.org S: Maintained W: http://www.linux-mtd.infradead.org/ From 8e2f8020270af7777d49c2e7132260983e4fc566 Mon Sep 17 00:00:00 2001 From: Finn Thain Date: Mon, 16 Feb 2026 18:01:30 +1100 Subject: [PATCH 009/352] mtd: Avoid boot crash in RedBoot partition table parser Given CONFIG_FORTIFY_SOURCE=y and a recent compiler, commit 439a1bcac648 ("fortify: Use __builtin_dynamic_object_size() when available") produces the warning below and an oops. Searching for RedBoot partition table in 50000000.flash at offset 0x7e0000 ------------[ cut here ]------------ WARNING: lib/string_helpers.c:1035 at 0xc029e04c, CPU#0: swapper/0/1 memcmp: detected buffer overflow: 15 byte read of buffer size 14 Modules linked in: CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.19.0 #1 NONE As Kees said, "'names' is pointing to the final 'namelen' many bytes of the allocation ... 'namelen' could be basically any length at all. This fortify warning looks legit to me -- this code used to be reading beyond the end of the allocation." Since the size of the dynamic allocation is calculated with strlen() we can use strcmp() instead of memcmp() and remain within bounds. Cc: Kees Cook Cc: stable@vger.kernel.org Cc: linux-hardening@vger.kernel.org Link: https://lore.kernel.org/all/202602151911.AD092DFFCD@keescook/ Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Suggested-by: Kees Cook Signed-off-by: Finn Thain Signed-off-by: Miquel Raynal --- drivers/mtd/parsers/redboot.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/mtd/parsers/redboot.c b/drivers/mtd/parsers/redboot.c index 558905160ddb..bf162c44eafe 100644 --- a/drivers/mtd/parsers/redboot.c +++ b/drivers/mtd/parsers/redboot.c @@ -270,9 +270,9 @@ static int parse_redboot_partitions(struct mtd_info *master, strcpy(names, fl->img->name); #ifdef CONFIG_MTD_REDBOOT_PARTS_READONLY - if (!memcmp(names, "RedBoot", 8) || - !memcmp(names, "RedBoot config", 15) || - !memcmp(names, "FIS directory", 14)) { + if (!strcmp(names, "RedBoot") || + !strcmp(names, "RedBoot config") || + !strcmp(names, "FIS directory")) { parts[i].mask_flags = MTD_WRITEABLE; } #endif From 0410e1a4c545c769c59c6eda897ad5d574d0c865 Mon Sep 17 00:00:00 2001 From: Chen Ni Date: Mon, 9 Feb 2026 15:56:18 +0800 Subject: [PATCH 010/352] mtd: rawnand: cadence: Fix error check for dma_alloc_coherent() in cadence_nand_init() Fix wrong variable used for error checking after dma_alloc_coherent() call. The function checks cdns_ctrl->dma_cdma_desc instead of cdns_ctrl->cdma_desc, which could lead to incorrect error handling. Fixes: ec4ba01e894d ("mtd: rawnand: Add new Cadence NAND driver to MTD subsystem") Cc: stable@vger.kernel.org Signed-off-by: Chen Ni Reviewed-by: Alok Tiwari Signed-off-by: Miquel Raynal --- drivers/mtd/nand/raw/cadence-nand-controller.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/nand/raw/cadence-nand-controller.c b/drivers/mtd/nand/raw/cadence-nand-controller.c index 99135ec23010..d53b35a8b3cb 100644 --- a/drivers/mtd/nand/raw/cadence-nand-controller.c +++ b/drivers/mtd/nand/raw/cadence-nand-controller.c @@ -3133,7 +3133,7 @@ static int cadence_nand_init(struct cdns_nand_ctrl *cdns_ctrl) sizeof(*cdns_ctrl->cdma_desc), &cdns_ctrl->dma_cdma_desc, GFP_KERNEL); - if (!cdns_ctrl->dma_cdma_desc) + if (!cdns_ctrl->cdma_desc) return -ENOMEM; cdns_ctrl->buf_size = SZ_16K; From 5a674ef871fe9d4c7477127340941f2c4d9a2741 Mon Sep 17 00:00:00 2001 From: "Anirudh Rayabharam (Microsoft)" Date: Wed, 25 Feb 2026 12:44:02 +0000 Subject: [PATCH 011/352] mshv: refactor synic init and cleanup Rename mshv_synic_init() to mshv_synic_cpu_init() and mshv_synic_cleanup() to mshv_synic_cpu_exit() to better reflect that these functions handle per-cpu synic setup and teardown. Use mshv_synic_init/cleanup() to perform init/cleanup that is not per-cpu. Move all the synic related setup from mshv_parent_partition_init. Move the reboot notifier to mshv_synic.c because it currently only operates on the synic cpuhp state. Move out synic_pages from the global mshv_root since its use is now completely local to mshv_synic.c. This is in preparation for adding more stuff to mshv_synic_init(). No functional change. Reviewed-by: Michael Kelley Signed-off-by: Anirudh Rayabharam (Microsoft) Signed-off-by: Wei Liu --- drivers/hv/mshv_root.h | 5 ++- drivers/hv/mshv_root_main.c | 64 +++++---------------------------- drivers/hv/mshv_synic.c | 71 +++++++++++++++++++++++++++++++++---- 3 files changed, 75 insertions(+), 65 deletions(-) diff --git a/drivers/hv/mshv_root.h b/drivers/hv/mshv_root.h index 04c2a1910a8a..826798f1a8ec 100644 --- a/drivers/hv/mshv_root.h +++ b/drivers/hv/mshv_root.h @@ -190,7 +190,6 @@ struct hv_synic_pages { }; struct mshv_root { - struct hv_synic_pages __percpu *synic_pages; spinlock_t pt_ht_lock; DECLARE_HASHTABLE(pt_htable, MSHV_PARTITIONS_HASH_BITS); struct hv_partition_property_vmm_capabilities vmm_caps; @@ -249,8 +248,8 @@ int mshv_register_doorbell(u64 partition_id, doorbell_cb_t doorbell_cb, void mshv_unregister_doorbell(u64 partition_id, int doorbell_portid); void mshv_isr(void); -int mshv_synic_init(unsigned int cpu); -int mshv_synic_cleanup(unsigned int cpu); +int mshv_synic_init(struct device *dev); +void mshv_synic_exit(void); static inline bool mshv_partition_encrypted(struct mshv_partition *partition) { diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c index 82ff823ef0ca..54c3e44d24ee 100644 --- a/drivers/hv/mshv_root_main.c +++ b/drivers/hv/mshv_root_main.c @@ -2064,7 +2064,6 @@ mshv_dev_release(struct inode *inode, struct file *filp) return 0; } -static int mshv_cpuhp_online; static int mshv_root_sched_online; static const char *scheduler_type_to_string(enum hv_scheduler_type type) @@ -2249,27 +2248,6 @@ root_scheduler_deinit(void) free_percpu(root_scheduler_output); } -static int mshv_reboot_notify(struct notifier_block *nb, - unsigned long code, void *unused) -{ - cpuhp_remove_state(mshv_cpuhp_online); - return 0; -} - -struct notifier_block mshv_reboot_nb = { - .notifier_call = mshv_reboot_notify, -}; - -static void mshv_root_partition_exit(void) -{ - unregister_reboot_notifier(&mshv_reboot_nb); -} - -static int __init mshv_root_partition_init(struct device *dev) -{ - return register_reboot_notifier(&mshv_reboot_nb); -} - static int __init mshv_init_vmm_caps(struct device *dev) { int ret; @@ -2314,39 +2292,21 @@ static int __init mshv_parent_partition_init(void) MSHV_HV_MAX_VERSION); } - mshv_root.synic_pages = alloc_percpu(struct hv_synic_pages); - if (!mshv_root.synic_pages) { - dev_err(dev, "Failed to allocate percpu synic page\n"); - ret = -ENOMEM; + ret = mshv_synic_init(dev); + if (ret) goto device_deregister; - } - - ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "mshv_synic", - mshv_synic_init, - mshv_synic_cleanup); - if (ret < 0) { - dev_err(dev, "Failed to setup cpu hotplug state: %i\n", ret); - goto free_synic_pages; - } - - mshv_cpuhp_online = ret; ret = mshv_init_vmm_caps(dev); if (ret) - goto remove_cpu_state; + goto synic_cleanup; ret = mshv_retrieve_scheduler_type(dev); if (ret) - goto remove_cpu_state; - - if (hv_root_partition()) - ret = mshv_root_partition_init(dev); - if (ret) - goto remove_cpu_state; + goto synic_cleanup; ret = root_scheduler_init(dev); if (ret) - goto exit_partition; + goto synic_cleanup; ret = mshv_debugfs_init(); if (ret) @@ -2367,13 +2327,8 @@ static int __init mshv_parent_partition_init(void) mshv_debugfs_exit(); deinit_root_scheduler: root_scheduler_deinit(); -exit_partition: - if (hv_root_partition()) - mshv_root_partition_exit(); -remove_cpu_state: - cpuhp_remove_state(mshv_cpuhp_online); -free_synic_pages: - free_percpu(mshv_root.synic_pages); +synic_cleanup: + mshv_synic_exit(); device_deregister: misc_deregister(&mshv_dev); return ret; @@ -2387,10 +2342,7 @@ static void __exit mshv_parent_partition_exit(void) misc_deregister(&mshv_dev); mshv_irqfd_wq_cleanup(); root_scheduler_deinit(); - if (hv_root_partition()) - mshv_root_partition_exit(); - cpuhp_remove_state(mshv_cpuhp_online); - free_percpu(mshv_root.synic_pages); + mshv_synic_exit(); } module_init(mshv_parent_partition_init); diff --git a/drivers/hv/mshv_synic.c b/drivers/hv/mshv_synic.c index 216065e21d28..617e8c02e365 100644 --- a/drivers/hv/mshv_synic.c +++ b/drivers/hv/mshv_synic.c @@ -12,11 +12,16 @@ #include #include #include +#include +#include #include #include "mshv_eventfd.h" #include "mshv.h" +static int synic_cpuhp_online; +static struct hv_synic_pages __percpu *synic_pages; + static u32 synic_event_ring_get_queued_port(u32 sint_index) { struct hv_synic_event_ring_page **event_ring_page; @@ -26,7 +31,7 @@ static u32 synic_event_ring_get_queued_port(u32 sint_index) u32 message; u8 tail; - spages = this_cpu_ptr(mshv_root.synic_pages); + spages = this_cpu_ptr(synic_pages); event_ring_page = &spages->synic_event_ring_page; synic_eventring_tail = (u8 **)this_cpu_ptr(hv_synic_eventring_tail); @@ -393,7 +398,7 @@ mshv_intercept_isr(struct hv_message *msg) void mshv_isr(void) { - struct hv_synic_pages *spages = this_cpu_ptr(mshv_root.synic_pages); + struct hv_synic_pages *spages = this_cpu_ptr(synic_pages); struct hv_message_page **msg_page = &spages->hyp_synic_message_page; struct hv_message *msg; bool handled; @@ -446,7 +451,7 @@ void mshv_isr(void) } } -int mshv_synic_init(unsigned int cpu) +static int mshv_synic_cpu_init(unsigned int cpu) { union hv_synic_simp simp; union hv_synic_siefp siefp; @@ -455,7 +460,7 @@ int mshv_synic_init(unsigned int cpu) union hv_synic_sint sint; #endif union hv_synic_scontrol sctrl; - struct hv_synic_pages *spages = this_cpu_ptr(mshv_root.synic_pages); + struct hv_synic_pages *spages = this_cpu_ptr(synic_pages); struct hv_message_page **msg_page = &spages->hyp_synic_message_page; struct hv_synic_event_flags_page **event_flags_page = &spages->synic_event_flags_page; @@ -542,14 +547,14 @@ int mshv_synic_init(unsigned int cpu) return -EFAULT; } -int mshv_synic_cleanup(unsigned int cpu) +static int mshv_synic_cpu_exit(unsigned int cpu) { union hv_synic_sint sint; union hv_synic_simp simp; union hv_synic_siefp siefp; union hv_synic_sirbp sirbp; union hv_synic_scontrol sctrl; - struct hv_synic_pages *spages = this_cpu_ptr(mshv_root.synic_pages); + struct hv_synic_pages *spages = this_cpu_ptr(synic_pages); struct hv_message_page **msg_page = &spages->hyp_synic_message_page; struct hv_synic_event_flags_page **event_flags_page = &spages->synic_event_flags_page; @@ -663,3 +668,57 @@ mshv_unregister_doorbell(u64 partition_id, int doorbell_portid) mshv_portid_free(doorbell_portid); } + +static int mshv_synic_reboot_notify(struct notifier_block *nb, + unsigned long code, void *unused) +{ + if (!hv_root_partition()) + return 0; + + cpuhp_remove_state(synic_cpuhp_online); + return 0; +} + +static struct notifier_block mshv_synic_reboot_nb = { + .notifier_call = mshv_synic_reboot_notify, +}; + +int __init mshv_synic_init(struct device *dev) +{ + int ret = 0; + + synic_pages = alloc_percpu(struct hv_synic_pages); + if (!synic_pages) { + dev_err(dev, "Failed to allocate percpu synic page\n"); + return -ENOMEM; + } + + ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "mshv_synic", + mshv_synic_cpu_init, + mshv_synic_cpu_exit); + if (ret < 0) { + dev_err(dev, "Failed to setup cpu hotplug state: %i\n", ret); + goto free_synic_pages; + } + + synic_cpuhp_online = ret; + + ret = register_reboot_notifier(&mshv_synic_reboot_nb); + if (ret) + goto remove_cpuhp_state; + + return 0; + +remove_cpuhp_state: + cpuhp_remove_state(synic_cpuhp_online); +free_synic_pages: + free_percpu(synic_pages); + return ret; +} + +void mshv_synic_exit(void) +{ + unregister_reboot_notifier(&mshv_synic_reboot_nb); + cpuhp_remove_state(synic_cpuhp_online); + free_percpu(synic_pages); +} From 622d68772ddf07573cf88e833afe8ba6c70ac748 Mon Sep 17 00:00:00 2001 From: "Anirudh Rayabharam (Microsoft)" Date: Wed, 25 Feb 2026 12:44:03 +0000 Subject: [PATCH 012/352] mshv: add arm64 support for doorbell & intercept SINTs On x86, the HYPERVISOR_CALLBACK_VECTOR is used to receive synthetic interrupts (SINTs) from the hypervisor for doorbells and intercepts. There is no such vector reserved for arm64. On arm64, the hypervisor exposes a synthetic register that can be read to find the INTID that should be used for SINTs. This INTID is in the PPI range. To better unify the code paths, introduce mshv_sint_vector_init() that either reads the synthetic register and obtains the INTID (arm64) or just uses HYPERVISOR_CALLBACK_VECTOR as the interrupt vector (x86). Reviewed-by: Michael Kelley Reviewed-by: Stanislav Kinsburskii Signed-off-by: Anirudh Rayabharam (Microsoft) Signed-off-by: Wei Liu --- drivers/hv/mshv_synic.c | 119 +++++++++++++++++++++++++++++++++--- include/hyperv/hvgdk_mini.h | 2 + 2 files changed, 111 insertions(+), 10 deletions(-) diff --git a/drivers/hv/mshv_synic.c b/drivers/hv/mshv_synic.c index 617e8c02e365..43f1bcbbf2d3 100644 --- a/drivers/hv/mshv_synic.c +++ b/drivers/hv/mshv_synic.c @@ -10,17 +10,21 @@ #include #include #include +#include #include #include #include #include #include +#include #include "mshv_eventfd.h" #include "mshv.h" static int synic_cpuhp_online; static struct hv_synic_pages __percpu *synic_pages; +static int mshv_sint_vector = -1; /* hwirq for the SynIC SINTs */ +static int mshv_sint_irq = -1; /* Linux IRQ for mshv_sint_vector */ static u32 synic_event_ring_get_queued_port(u32 sint_index) { @@ -442,9 +446,7 @@ void mshv_isr(void) if (msg->header.message_flags.msg_pending) hv_set_non_nested_msr(HV_MSR_EOM, 0); -#ifdef HYPERVISOR_CALLBACK_VECTOR - add_interrupt_randomness(HYPERVISOR_CALLBACK_VECTOR); -#endif + add_interrupt_randomness(mshv_sint_vector); } else { pr_warn_once("%s: unknown message type 0x%x\n", __func__, msg->header.message_type); @@ -456,9 +458,7 @@ static int mshv_synic_cpu_init(unsigned int cpu) union hv_synic_simp simp; union hv_synic_siefp siefp; union hv_synic_sirbp sirbp; -#ifdef HYPERVISOR_CALLBACK_VECTOR union hv_synic_sint sint; -#endif union hv_synic_scontrol sctrl; struct hv_synic_pages *spages = this_cpu_ptr(synic_pages); struct hv_message_page **msg_page = &spages->hyp_synic_message_page; @@ -501,10 +501,12 @@ static int mshv_synic_cpu_init(unsigned int cpu) hv_set_non_nested_msr(HV_MSR_SIRBP, sirbp.as_uint64); -#ifdef HYPERVISOR_CALLBACK_VECTOR + if (mshv_sint_irq != -1) + enable_percpu_irq(mshv_sint_irq, 0); + /* Enable intercepts */ sint.as_uint64 = 0; - sint.vector = HYPERVISOR_CALLBACK_VECTOR; + sint.vector = mshv_sint_vector; sint.masked = false; sint.auto_eoi = hv_recommend_using_aeoi(); hv_set_non_nested_msr(HV_MSR_SINT0 + HV_SYNIC_INTERCEPTION_SINT_INDEX, @@ -512,13 +514,12 @@ static int mshv_synic_cpu_init(unsigned int cpu) /* Doorbell SINT */ sint.as_uint64 = 0; - sint.vector = HYPERVISOR_CALLBACK_VECTOR; + sint.vector = mshv_sint_vector; sint.masked = false; sint.as_intercept = 1; sint.auto_eoi = hv_recommend_using_aeoi(); hv_set_non_nested_msr(HV_MSR_SINT0 + HV_SYNIC_DOORBELL_SINT_INDEX, sint.as_uint64); -#endif /* Enable global synic bit */ sctrl.as_uint64 = hv_get_non_nested_msr(HV_MSR_SCONTROL); @@ -573,6 +574,9 @@ static int mshv_synic_cpu_exit(unsigned int cpu) hv_set_non_nested_msr(HV_MSR_SINT0 + HV_SYNIC_DOORBELL_SINT_INDEX, sint.as_uint64); + if (mshv_sint_irq != -1) + disable_percpu_irq(mshv_sint_irq); + /* Disable Synic's event ring page */ sirbp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIRBP); sirbp.sirbp_enabled = false; @@ -683,14 +687,106 @@ static struct notifier_block mshv_synic_reboot_nb = { .notifier_call = mshv_synic_reboot_notify, }; +#ifndef HYPERVISOR_CALLBACK_VECTOR +static DEFINE_PER_CPU(long, mshv_evt); + +static irqreturn_t mshv_percpu_isr(int irq, void *dev_id) +{ + mshv_isr(); + return IRQ_HANDLED; +} + +#ifdef CONFIG_ACPI +static int __init mshv_acpi_setup_sint_irq(void) +{ + return acpi_register_gsi(NULL, mshv_sint_vector, ACPI_EDGE_SENSITIVE, + ACPI_ACTIVE_HIGH); +} + +static void mshv_acpi_cleanup_sint_irq(void) +{ + acpi_unregister_gsi(mshv_sint_vector); +} +#else +static int __init mshv_acpi_setup_sint_irq(void) +{ + return -ENODEV; +} + +static void mshv_acpi_cleanup_sint_irq(void) +{ +} +#endif + +static int __init mshv_sint_vector_setup(void) +{ + int ret; + struct hv_register_assoc reg = { + .name = HV_ARM64_REGISTER_SINT_RESERVED_INTERRUPT_ID, + }; + union hv_input_vtl input_vtl = { 0 }; + + if (acpi_disabled) + return -ENODEV; + + ret = hv_call_get_vp_registers(HV_VP_INDEX_SELF, HV_PARTITION_ID_SELF, + 1, input_vtl, ®); + if (ret || !reg.value.reg64) + return -ENODEV; + + mshv_sint_vector = reg.value.reg64; + ret = mshv_acpi_setup_sint_irq(); + if (ret < 0) { + pr_err("Failed to setup IRQ for MSHV SINT vector %d: %d\n", + mshv_sint_vector, ret); + goto out_fail; + } + + mshv_sint_irq = ret; + + ret = request_percpu_irq(mshv_sint_irq, mshv_percpu_isr, "MSHV", + &mshv_evt); + if (ret) + goto out_unregister; + + return 0; + +out_unregister: + mshv_acpi_cleanup_sint_irq(); +out_fail: + return ret; +} + +static void mshv_sint_vector_cleanup(void) +{ + free_percpu_irq(mshv_sint_irq, &mshv_evt); + mshv_acpi_cleanup_sint_irq(); +} +#else /* !HYPERVISOR_CALLBACK_VECTOR */ +static int __init mshv_sint_vector_setup(void) +{ + mshv_sint_vector = HYPERVISOR_CALLBACK_VECTOR; + return 0; +} + +static void mshv_sint_vector_cleanup(void) +{ +} +#endif /* HYPERVISOR_CALLBACK_VECTOR */ + int __init mshv_synic_init(struct device *dev) { int ret = 0; + ret = mshv_sint_vector_setup(); + if (ret) + return ret; + synic_pages = alloc_percpu(struct hv_synic_pages); if (!synic_pages) { dev_err(dev, "Failed to allocate percpu synic page\n"); - return -ENOMEM; + ret = -ENOMEM; + goto sint_vector_cleanup; } ret = cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "mshv_synic", @@ -713,6 +809,8 @@ int __init mshv_synic_init(struct device *dev) cpuhp_remove_state(synic_cpuhp_online); free_synic_pages: free_percpu(synic_pages); +sint_vector_cleanup: + mshv_sint_vector_cleanup(); return ret; } @@ -721,4 +819,5 @@ void mshv_synic_exit(void) unregister_reboot_notifier(&mshv_synic_reboot_nb); cpuhp_remove_state(synic_cpuhp_online); free_percpu(synic_pages); + mshv_sint_vector_cleanup(); } diff --git a/include/hyperv/hvgdk_mini.h b/include/hyperv/hvgdk_mini.h index 056ef7b6b360..8bb3dd71c5b4 100644 --- a/include/hyperv/hvgdk_mini.h +++ b/include/hyperv/hvgdk_mini.h @@ -1121,6 +1121,8 @@ enum hv_register_name { HV_X64_REGISTER_MSR_MTRR_FIX4KF8000 = 0x0008007A, HV_X64_REGISTER_REG_PAGE = 0x0009001C, +#elif defined(CONFIG_ARM64) + HV_ARM64_REGISTER_SINT_RESERVED_INTERRUPT_ID = 0x00070001, #endif }; From 0d10393d5eac33cbd92f7a41fddca12c41d3cb7e Mon Sep 17 00:00:00 2001 From: Roshan Kumar Date: Sun, 1 Mar 2026 10:56:38 +0000 Subject: [PATCH 013/352] xfrm: iptfs: validate inner IPv4 header length in IPTFS payload Add validation of the inner IPv4 packet tot_len and ihl fields parsed from decrypted IPTFS payloads in __input_process_payload(). A crafted ESP packet containing an inner IPv4 header with tot_len=0 causes an infinite loop: iplen=0 leads to capturelen=min(0, remaining)=0, so the data offset never advances and the while(data < tail) loop never terminates, spinning forever in softirq context. Reject inner IPv4 packets where tot_len < ihl*4 or ihl*4 < sizeof(struct iphdr), which catches both the tot_len=0 case and malformed ihl values. The normal IP stack performs this validation in ip_rcv_core(), but IPTFS extracts and processes inner packets before they reach that layer. Reported-by: Roshan Kumar Fixes: 6c82d2433671 ("xfrm: iptfs: add basic receive packet (tunnel egress) handling") Cc: stable@vger.kernel.org Signed-off-by: Roshan Kumar Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_iptfs.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/xfrm/xfrm_iptfs.c b/net/xfrm/xfrm_iptfs.c index 3b6d7284fc70..0747d1cfa333 100644 --- a/net/xfrm/xfrm_iptfs.c +++ b/net/xfrm/xfrm_iptfs.c @@ -991,6 +991,11 @@ static bool __input_process_payload(struct xfrm_state *x, u32 data, iplen = be16_to_cpu(iph->tot_len); iphlen = iph->ihl << 2; + if (iplen < iphlen || iphlen < sizeof(*iph)) { + XFRM_INC_STATS(net, + LINUX_MIB_XFRMINHDRERROR); + goto done; + } protocol = cpu_to_be16(ETH_P_IP); XFRM_MODE_SKB_CB(skbseq->root_skb)->tos = iph->tos; } else if (iph->version == 0x6) { From 19d2f0b97a131198efc2c4ca3eb7f980bba8c2b4 Mon Sep 17 00:00:00 2001 From: Alison Schofield Date: Thu, 26 Feb 2026 10:44:36 -0800 Subject: [PATCH 014/352] cxl/port: Fix use after free of parent_port in cxl_detach_ep() cxl_detach_ep() is called during bottom-up removal when all CXL memory devices beneath a switch port have been removed. For each port in the hierarchy it locks both the port and its parent, removes the endpoint, and if the port is now empty, marks it dead and unregisters the port by calling delete_switch_port(). There are two places during this work where the parent_port may be used after freeing: First, a concurrent detach may have already processed a port by the time a second worker finds it via bus_find_device(). Without pinning parent_port, it may already be freed when we discover port->dead and attempt to unlock the parent_port. In a production kernel that's a silent memory corruption, with lock debug, it looks like this: []DEBUG_LOCKS_WARN_ON(__owner_task(owner) != get_current()) []WARNING: kernel/locking/mutex.c:949 at __mutex_unlock_slowpath+0x1ee/0x310 []Call Trace: []mutex_unlock+0xd/0x20 []cxl_detach_ep+0x180/0x400 [cxl_core] []devm_action_release+0x10/0x20 []devres_release_all+0xa8/0xe0 []device_unbind_cleanup+0xd/0xa0 []really_probe+0x1a6/0x3e0 Second, delete_switch_port() releases three devm actions registered against parent_port. The last of those is unregister_port() and it calls device_unregister() on the child port, which can cascade. If parent_port is now also empty the device core may unregister and free it too. So by the time delete_switch_port() returns, parent_port may be free, and the subsequent device_unlock(&parent_port->dev) operates on freed memory. The kernel log looks same as above, with a different offset in cxl_detach_ep(). Both of these issues stem from the absence of a lifetime guarantee between a child port and its parent port. Establish a lifetime rule for ports: child ports hold a reference to their parent device until release. Take the reference when the port is allocated and drop it when released. This ensures the parent is valid for the full lifetime of the child and eliminates the use after free window in cxl_detach_ep(). This is easily reproduced with a reload of cxl_acpi in QEMU with CXL devices present. Fixes: 2345df54249c ("cxl/memdev: Fix endpoint port removal") Reviewed-by: Dave Jiang Reviewed-by: Li Ming Signed-off-by: Alison Schofield Reviewed-by: Jonathan Cameron Link: https://patch.msgid.link/20260226184439.1732841-1-alison.schofield@intel.com Signed-off-by: Dave Jiang --- drivers/cxl/core/port.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/cxl/core/port.c b/drivers/cxl/core/port.c index 0c5957d1d329..c5aacd7054f1 100644 --- a/drivers/cxl/core/port.c +++ b/drivers/cxl/core/port.c @@ -552,10 +552,13 @@ static void cxl_port_release(struct device *dev) xa_destroy(&port->dports); xa_destroy(&port->regions); ida_free(&cxl_port_ida, port->id); - if (is_cxl_root(port)) + + if (is_cxl_root(port)) { kfree(to_cxl_root(port)); - else + } else { + put_device(dev->parent); kfree(port); + } } static ssize_t decoders_committed_show(struct device *dev, @@ -707,6 +710,7 @@ static struct cxl_port *cxl_port_alloc(struct device *uport_dev, struct cxl_port *iter; dev->parent = &parent_port->dev; + get_device(dev->parent); port->depth = parent_port->depth + 1; port->parent_dport = parent_dport; From 2b76e0cc7803e5ab561c875edaba7f6bbd87fbb0 Mon Sep 17 00:00:00 2001 From: Matthew Schwartz Date: Mon, 2 Mar 2026 13:07:17 -0800 Subject: [PATCH 015/352] mmc: sdhci-pci-gli: fix GL9750 DMA write corruption The GL9750 SD host controller has intermittent data corruption during DMA write operations. The GM_BURST register's R_OSRC_Lmt field (bits 17:16), which limits outstanding DMA read requests from system memory, is not being cleared during initialization. The Windows driver sets R_OSRC_Lmt to zero, limiting requests to the smallest unit. Clear R_OSRC_Lmt to match the Windows driver behavior. This eliminates write corruption verified with f3write/f3read tests while maintaining DMA performance. Cc: stable@vger.kernel.org Fixes: e51df6ce668a ("mmc: host: sdhci-pci: Add Genesys Logic GL975x support") Closes: https://lore.kernel.org/linux-mmc/33d12807-5c72-41ce-8679-57aa11831fad@linux.dev/ Acked-by: Adrian Hunter Signed-off-by: Matthew Schwartz Reviewed-by: Ben Chuang Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-pci-gli.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/drivers/mmc/host/sdhci-pci-gli.c b/drivers/mmc/host/sdhci-pci-gli.c index b0f91cc9e40e..6e4084407662 100644 --- a/drivers/mmc/host/sdhci-pci-gli.c +++ b/drivers/mmc/host/sdhci-pci-gli.c @@ -68,6 +68,9 @@ #define GLI_9750_MISC_TX1_DLY_VALUE 0x5 #define SDHCI_GLI_9750_MISC_SSC_OFF BIT(26) +#define SDHCI_GLI_9750_GM_BURST_SIZE 0x510 +#define SDHCI_GLI_9750_GM_BURST_SIZE_R_OSRC_LMT GENMASK(17, 16) + #define SDHCI_GLI_9750_TUNING_CONTROL 0x540 #define SDHCI_GLI_9750_TUNING_CONTROL_EN BIT(4) #define GLI_9750_TUNING_CONTROL_EN_ON 0x1 @@ -345,10 +348,16 @@ static void gli_set_9750(struct sdhci_host *host) u32 misc_value; u32 parameter_value; u32 control_value; + u32 burst_value; u16 ctrl2; gl9750_wt_on(host); + /* clear R_OSRC_Lmt to avoid DMA write corruption */ + burst_value = sdhci_readl(host, SDHCI_GLI_9750_GM_BURST_SIZE); + burst_value &= ~SDHCI_GLI_9750_GM_BURST_SIZE_R_OSRC_LMT; + sdhci_writel(host, burst_value, SDHCI_GLI_9750_GM_BURST_SIZE); + driving_value = sdhci_readl(host, SDHCI_GLI_9750_DRIVING); pll_value = sdhci_readl(host, SDHCI_GLI_9750_PLL); sw_ctrl_value = sdhci_readl(host, SDHCI_GLI_9750_SW_CTRL); From 77b310bb7b5ff8c017524df83292e0242ba89791 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Mon, 2 Feb 2026 11:13:30 -0800 Subject: [PATCH 016/352] cxl/region: Fix leakage in __construct_region() Failing the first sysfs_update_group() needs to explicitly kfree the resource as it is too early for cxl_region_iomem_release() to do so. Signed-off-by: Davidlohr Bueso Reviewed-by: Ira Weiny Reviewed-by: Gregory Price Fixes: d6602e25819d (cxl/region: Add support to indicate region has extended linear cache) Link: https://patch.msgid.link/20260202191330.245608-1-dave@stgolabs.net Signed-off-by: Dave Jiang --- drivers/cxl/core/region.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/cxl/core/region.c b/drivers/cxl/core/region.c index 42874948b589..c37ae0b28bbb 100644 --- a/drivers/cxl/core/region.c +++ b/drivers/cxl/core/region.c @@ -3854,8 +3854,10 @@ static int __construct_region(struct cxl_region *cxlr, } rc = sysfs_update_group(&cxlr->dev.kobj, &cxl_region_group); - if (rc) + if (rc) { + kfree(res); return rc; + } rc = insert_resource(cxlrd->res, res); if (rc) { From f303406efd0b6b8740ce5c47e852097bbcf54879 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 3 Mar 2026 18:52:06 +0100 Subject: [PATCH 017/352] KVM: s390: Fix a deadlock In some scenarios, a deadlock can happen, involving _do_shadow_pte(). Convert all usages of pgste_get_lock() to pgste_get_trylock() in _do_shadow_pte() and return -EAGAIN. All callers can already deal with -EAGAIN being returned. Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") Tested-by: Christian Borntraeger Reviewed-by: Janosch Frank Reviewed-by: Christoph Schlameuss Signed-off-by: Claudio Imbrenda --- arch/s390/kvm/gaccess.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 4630b2a067ea..a9da9390867d 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -1434,7 +1434,8 @@ static int _do_shadow_pte(struct gmap *sg, gpa_t raddr, union pte *ptep_h, union if (rc) return rc; - pgste = pgste_get_lock(ptep_h); + if (!pgste_get_trylock(ptep_h, &pgste)) + return -EAGAIN; newpte = _pte(f->pfn, f->writable, !p, 0); newpte.s.d |= ptep->s.d; newpte.s.sd |= ptep->s.sd; @@ -1444,7 +1445,8 @@ static int _do_shadow_pte(struct gmap *sg, gpa_t raddr, union pte *ptep_h, union pgste_set_unlock(ptep_h, pgste); newpte = _pte(f->pfn, 0, !p, 0); - pgste = pgste_get_lock(ptep); + if (!pgste_get_trylock(ptep, &pgste)) + return -EAGAIN; pgste = __dat_ptep_xchg(ptep, pgste, newpte, gpa_to_gfn(raddr), sg->asce, uses_skeys(sg)); pgste_set_unlock(ptep, pgste); From e825c79ef914bd55cf7c2476ddcfb2738eb689c3 Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Tue, 27 Jan 2026 00:35:47 +0800 Subject: [PATCH 018/352] pinctrl: renesas: rzt2h: Fix device node leak in rzt2h_gpio_register() When calling of_parse_phandle_with_fixed_args(), the caller is responsible for calling of_node_put() to release the device node reference. In rzt2h_gpio_register(), the driver fails to call of_node_put() to release the reference in of_args.np, which causes a memory leak. Add the missing of_node_put() call to fix the leak. Fixes: 34d4d093077a ("pinctrl: renesas: Add support for RZ/T2H") Signed-off-by: Felix Gu Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/20260127-rzt2h-v1-1-86472e7421b8@gmail.com Signed-off-by: Geert Uytterhoeven --- drivers/pinctrl/renesas/pinctrl-rzt2h.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pinctrl/renesas/pinctrl-rzt2h.c b/drivers/pinctrl/renesas/pinctrl-rzt2h.c index 9949108a35bb..5e0586d45cb7 100644 --- a/drivers/pinctrl/renesas/pinctrl-rzt2h.c +++ b/drivers/pinctrl/renesas/pinctrl-rzt2h.c @@ -833,6 +833,7 @@ static int rzt2h_gpio_register(struct rzt2h_pinctrl *pctrl) if (ret) return dev_err_probe(dev, ret, "Unable to parse gpio-ranges\n"); + of_node_put(of_args.np); if (of_args.args[0] != 0 || of_args.args[1] != 0 || of_args.args[2] != pctrl->data->n_port_pins) return dev_err_probe(dev, -EINVAL, From ebe7561e9b9203611cea72a764bc321ff308f737 Mon Sep 17 00:00:00 2001 From: Cosmin Tanislav Date: Thu, 5 Feb 2026 12:39:30 +0200 Subject: [PATCH 019/352] pinctrl: renesas: rzt2h: Fix invalid wait context The rzt2h_gpio_get_direction() function is called from gpiod_get_direction(), which ends up being used within the __setup_irq() call stack when requesting an interrupt. __setup_irq() holds a raw_spinlock_t with IRQs disabled, which creates an atomic context. spinlock_t cannot be used within atomic context when PREEMPT_RT is enabled, since it may become a sleeping lock. An "[ BUG: Invalid wait context ]" splat is observed when running with CONFIG_PROVE_LOCKING enabled, describing exactly the aforementioned call stack. __setup_irq() needs to hold a raw_spinlock_t with IRQs disabled to serialize access against a concurrent hard interrupt. Switch to raw_spinlock_t to fix this. Fixes: 829dde3369a9 ("pinctrl: renesas: rzt2h: Add GPIO IRQ chip to handle interrupts") Signed-off-by: Cosmin Tanislav Reviewed-by: Sebastian Andrzej Siewior Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/20260205103930.666051-1-cosmin-gabriel.tanislav.xa@renesas.com Signed-off-by: Geert Uytterhoeven --- drivers/pinctrl/renesas/pinctrl-rzt2h.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/drivers/pinctrl/renesas/pinctrl-rzt2h.c b/drivers/pinctrl/renesas/pinctrl-rzt2h.c index 5e0586d45cb7..5927744c7a96 100644 --- a/drivers/pinctrl/renesas/pinctrl-rzt2h.c +++ b/drivers/pinctrl/renesas/pinctrl-rzt2h.c @@ -85,7 +85,7 @@ struct rzt2h_pinctrl { struct gpio_chip gpio_chip; struct pinctrl_gpio_range gpio_range; DECLARE_BITMAP(used_irqs, RZT2H_INTERRUPTS_NUM); - spinlock_t lock; /* lock read/write registers */ + raw_spinlock_t lock; /* lock read/write registers */ struct mutex mutex; /* serialize adding groups and functions */ bool safety_port_enabled; atomic_t wakeup_path; @@ -145,7 +145,7 @@ static void rzt2h_pinctrl_set_pfc_mode(struct rzt2h_pinctrl *pctrl, u64 reg64; u16 reg16; - guard(spinlock_irqsave)(&pctrl->lock); + guard(raw_spinlock_irqsave)(&pctrl->lock); /* Set pin to 'Non-use (Hi-Z input protection)' */ reg16 = rzt2h_pinctrl_readw(pctrl, port, PM(port)); @@ -474,7 +474,7 @@ static int rzt2h_gpio_request(struct gpio_chip *chip, unsigned int offset) if (ret) return ret; - guard(spinlock_irqsave)(&pctrl->lock); + guard(raw_spinlock_irqsave)(&pctrl->lock); /* Select GPIO mode in PMC Register */ rzt2h_pinctrl_set_gpio_en(pctrl, port, bit, true); @@ -487,7 +487,7 @@ static void rzt2h_gpio_set_direction(struct rzt2h_pinctrl *pctrl, u32 port, { u16 reg; - guard(spinlock_irqsave)(&pctrl->lock); + guard(raw_spinlock_irqsave)(&pctrl->lock); reg = rzt2h_pinctrl_readw(pctrl, port, PM(port)); reg &= ~PM_PIN_MASK(bit); @@ -509,7 +509,7 @@ static int rzt2h_gpio_get_direction(struct gpio_chip *chip, unsigned int offset) if (ret) return ret; - guard(spinlock_irqsave)(&pctrl->lock); + guard(raw_spinlock_irqsave)(&pctrl->lock); if (rzt2h_pinctrl_readb(pctrl, port, PMC(port)) & BIT(bit)) { /* @@ -547,7 +547,7 @@ static int rzt2h_gpio_set(struct gpio_chip *chip, unsigned int offset, u8 bit = RZT2H_PIN_ID_TO_PIN(offset); u8 reg; - guard(spinlock_irqsave)(&pctrl->lock); + guard(raw_spinlock_irqsave)(&pctrl->lock); reg = rzt2h_pinctrl_readb(pctrl, port, P(port)); if (value) @@ -965,7 +965,7 @@ static int rzt2h_pinctrl_probe(struct platform_device *pdev) if (ret) return ret; - spin_lock_init(&pctrl->lock); + raw_spin_lock_init(&pctrl->lock); mutex_init(&pctrl->mutex); platform_set_drvdata(pdev, pctrl); From 93d0fcdddc9e7be9d4f42acbe57bc90dbb0fe75d Mon Sep 17 00:00:00 2001 From: Keith Busch Date: Thu, 5 Mar 2026 12:40:56 -0800 Subject: [PATCH 020/352] cxl/acpi: Fix CXL_ACPI and CXL_PMEM Kconfig tristate mismatch Commit e7e222ad73d9 ("cxl: Move devm_cxl_add_nvdimm_bridge() to cxl_pmem.ko") moves devm_cxl_add_nvdimm_bridge() into the cxl_pmem file, which has independent config compile options for built-in or module. The call from cxl_acpi_probe() is guarded by IS_ENABLED(CONFIG_CXL_PMEM), which evaluates to true for both =y and =m. When CONFIG_CXL_PMEM=m, a built-in cxl_acpi attempts to reference a symbol exported by a module, which fails to link. CXL_PMEM cannot simply be promoted to =y in this configuration because it depends on LIBNVDIMM, which may itself be =m. Add a Kconfig dependency to prevent CXL_ACPI from being built-in when CXL_PMEM is a module. This contrains CXL_ACPI to =m when CXL_PMEM=m, while still allowing CXL_ACPI to be freely configured when CXL_PMEM is either built-in or disabled. [ dj: Fix up commit reference formatting. ] Fixes: e7e222ad73d9 ("cxl: Move devm_cxl_add_nvdimm_bridge() to cxl_pmem.ko") Signed-off-by: Keith Busch Reviewed-by: Jonathan Cameron Reviewed-by: Dan Williams Link: https://patch.msgid.link/20260305204057.1516948-1-kbusch@meta.com Signed-off-by: Dave Jiang --- drivers/cxl/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/cxl/Kconfig b/drivers/cxl/Kconfig index 4589bf11d3fe..80aeb0d556bd 100644 --- a/drivers/cxl/Kconfig +++ b/drivers/cxl/Kconfig @@ -59,6 +59,7 @@ config CXL_ACPI tristate "CXL ACPI: Platform Support" depends on ACPI depends on ACPI_NUMA + depends on CXL_PMEM || !CXL_PMEM default CXL_BUS select ACPI_TABLE_LIB select ACPI_HMAT From 0b352f83cabfefdaafa806d6471f0eca117dc7d5 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Wed, 4 Mar 2026 15:09:35 +0100 Subject: [PATCH 021/352] xfrm: iptfs: fix skb_put() panic on non-linear skb during reassembly In iptfs_reassem_cont(), IP-TFS attempts to append data to the new inner packet 'newskb' that is being reassembled. First a zero-copy approach is tried if it succeeds then newskb becomes non-linear. When a subsequent fragment in the same datagram does not meet the fast-path conditions, a memory copy is performed. It calls skb_put() to append the data and as newskb is non-linear it triggers SKB_LINEAR_ASSERT check. Oops: invalid opcode: 0000 [#1] SMP NOPTI [...] RIP: 0010:skb_put+0x3c/0x40 [...] Call Trace: iptfs_reassem_cont+0x1ab/0x5e0 [xfrm_iptfs] iptfs_input_ordered+0x2af/0x380 [xfrm_iptfs] iptfs_input+0x122/0x3e0 [xfrm_iptfs] xfrm_input+0x91e/0x1a50 xfrm4_esp_rcv+0x3a/0x110 ip_protocol_deliver_rcu+0x1d7/0x1f0 ip_local_deliver_finish+0xbe/0x1e0 __netif_receive_skb_core.constprop.0+0xb56/0x1120 __netif_receive_skb_list_core+0x133/0x2b0 netif_receive_skb_list_internal+0x1ff/0x3f0 napi_complete_done+0x81/0x220 virtnet_poll+0x9d6/0x116e [virtio_net] __napi_poll.constprop.0+0x2b/0x270 net_rx_action+0x162/0x360 handle_softirqs+0xdc/0x510 __irq_exit_rcu+0xe7/0x110 irq_exit_rcu+0xe/0x20 common_interrupt+0x85/0xa0 Fix this by checking if the skb is non-linear. If it is, linearize it by calling skb_linearize(). As the initial allocation of newskb originally reserved enough tailroom for the entire reassembled packet we do not need to check if we have enough tailroom or extend it. Fixes: 5f2b6a909574 ("xfrm: iptfs: add skb-fragment sharing code") Reported-by: Hao Long Closes: https://lore.kernel.org/netdev/DGRCO9SL0T5U.JTINSHJQ9KPK@imlonghao.com/ Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_iptfs.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/xfrm/xfrm_iptfs.c b/net/xfrm/xfrm_iptfs.c index 0747d1cfa333..2c87290fe06c 100644 --- a/net/xfrm/xfrm_iptfs.c +++ b/net/xfrm/xfrm_iptfs.c @@ -901,6 +901,12 @@ static u32 iptfs_reassem_cont(struct xfrm_iptfs_data *xtfs, u64 seq, iptfs_skb_can_add_frags(newskb, fragwalk, data, copylen)) { iptfs_skb_add_frags(newskb, fragwalk, data, copylen); } else { + if (skb_linearize(newskb)) { + XFRM_INC_STATS(xs_net(xtfs->x), + LINUX_MIB_XFRMINBUFFERERROR); + goto abandon; + } + /* copy fragment data into newskb */ if (skb_copy_seq_read(st, data, skb_put(newskb, copylen), copylen)) { From a76e30c2479ce6ffa2aa6c8a8462897afc82bc90 Mon Sep 17 00:00:00 2001 From: Charles Mirabile Date: Sat, 7 Mar 2026 23:43:30 -0500 Subject: [PATCH 022/352] kbuild: Delete .builtin-dtbs.S when running make clean The makefile tries to delete a file named ".builtin-dtb.S" but the file created by scripts/Makefile.vmlinux is actually called ".builtin-dtbs.S". Fixes: 654102df2ac2a ("kbuild: add generic support for built-in boot DTBs") Cc: stable@vger.kernel.org Signed-off-by: Charles Mirabile Reviewed-by: Nicolas Schier Link: https://patch.msgid.link/20260308044338.181403-1-cmirabil@redhat.com [nathan: Small commit message adjustments] Signed-off-by: Nathan Chancellor --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index 2b15f0b4a0cb..2317618fd074 100644 --- a/Makefile +++ b/Makefile @@ -1650,7 +1650,7 @@ CLEAN_FILES += vmlinux.symvers modules-only.symvers \ modules.builtin.ranges vmlinux.o.map vmlinux.unstripped \ compile_commands.json rust/test \ rust-project.json .vmlinux.objs .vmlinux.export.c \ - .builtin-dtbs-list .builtin-dtb.S + .builtin-dtbs-list .builtin-dtbs.S # Directories & files removed with 'make mrproper' MRPROPER_FILES += include/config include/generated \ From 3b4a3a00de8770f3a60c1fa483921ce37415132d Mon Sep 17 00:00:00 2001 From: Weizhao Ouyang Date: Mon, 9 Mar 2026 20:15:05 +0800 Subject: [PATCH 023/352] scripts: kconfig: merge_config.sh: fix unexpected operator warning Fix a warning for: $ ./scripts/kconfig/merge_config.sh .config extra.config Using .config as base Merging extra.config ./scripts/kconfig/merge_config.sh: 384: [: false: unexpected operator The shellcheck report is also attached: if [ "$STRICT" == "true" ] && [ "$STRICT_MODE_VIOLATED" == "true" ]; then ^-- SC3014 (warning): In POSIX sh, == in place of = is undefined. ^-- SC3014 (warning): In POSIX sh, == in place of = is undefined. Fixes: dfc97e1c5da5 ("scripts: kconfig: merge_config.sh: use awk in checks too") Signed-off-by: Weizhao Ouyang Reviewed-by: Mikko Rapeli Link: https://patch.msgid.link/20260309121505.40454-1-o451686892@gmail.com Signed-off-by: Nathan Chancellor --- scripts/kconfig/merge_config.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/scripts/kconfig/merge_config.sh b/scripts/kconfig/merge_config.sh index 735e1de450c6..073c6bec5245 100755 --- a/scripts/kconfig/merge_config.sh +++ b/scripts/kconfig/merge_config.sh @@ -381,7 +381,7 @@ END { STRICT_MODE_VIOLATED=true fi -if [ "$STRICT" == "true" ] && [ "$STRICT_MODE_VIOLATED" == "true" ]; then +if [ "$STRICT" = "true" ] && [ "$STRICT_MODE_VIOLATED" = "true" ]; then echo "Requested and effective config differ" exit 1 fi From 263447532463cf4444a3595e835b99a4e90952fa Mon Sep 17 00:00:00 2001 From: Neil Armstrong Date: Wed, 4 Mar 2026 18:41:08 +0100 Subject: [PATCH 024/352] pinctrl: qcom: spmi-gpio: implement .get_direction() GPIO controller driver should typically implement the .get_direction() callback as GPIOLIB internals may try to use it to determine the state of a pin. Since introduction of shared proxy, it prints a warning splat when using a shared spmi gpio. The implementation is not easy because the controller supports enabling the input and output logic at the same time, so we aligns on the behaviour of the .get() operation and return -EINVAL in other situations. Fixes: eadff3024472 ("pinctrl: Qualcomm SPMI PMIC GPIO pin controller driver") Fixes: d7b5f5cc5eb4 ("pinctrl: qcom: spmi-gpio: Add support for GPIO LV/MV subtype") Signed-off-by: Neil Armstrong Reviewed-by: Konrad Dybcio Reviewed-by: Bartosz Golaszewski Signed-off-by: Linus Walleij --- drivers/pinctrl/qcom/pinctrl-spmi-gpio.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c index 83f940fe30b2..d02d42513ebb 100644 --- a/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c +++ b/drivers/pinctrl/qcom/pinctrl-spmi-gpio.c @@ -723,6 +723,21 @@ static const struct pinconf_ops pmic_gpio_pinconf_ops = { .pin_config_group_dbg_show = pmic_gpio_config_dbg_show, }; +static int pmic_gpio_get_direction(struct gpio_chip *chip, unsigned pin) +{ + struct pmic_gpio_state *state = gpiochip_get_data(chip); + struct pmic_gpio_pad *pad; + + pad = state->ctrl->desc->pins[pin].drv_data; + + if (!pad->is_enabled || pad->analog_pass || + (!pad->input_enabled && !pad->output_enabled)) + return -EINVAL; + + /* Make sure the state is aligned on what pmic_gpio_get() returns */ + return pad->input_enabled ? GPIO_LINE_DIRECTION_IN : GPIO_LINE_DIRECTION_OUT; +} + static int pmic_gpio_direction_input(struct gpio_chip *chip, unsigned pin) { struct pmic_gpio_state *state = gpiochip_get_data(chip); @@ -801,6 +816,7 @@ static void pmic_gpio_dbg_show(struct seq_file *s, struct gpio_chip *chip) } static const struct gpio_chip pmic_gpio_gpio_template = { + .get_direction = pmic_gpio_get_direction, .direction_input = pmic_gpio_direction_input, .direction_output = pmic_gpio_direction_output, .get = pmic_gpio_get, From fb22bb9701d48c4b0e81fe204c2f96a37a520568 Mon Sep 17 00:00:00 2001 From: Dmitry Torokhov Date: Wed, 18 Feb 2026 12:58:09 -0800 Subject: [PATCH 025/352] pinctrl: renesas: rza1: Normalize return value of gpio_get() The GPIO .get() callback is expected to return 0 or 1 (or a negative error code). Ensure that the value returned by rza1_gpio_get() is normalized to the [0, 1] range. Fixes: 86ef402d805d606a ("gpiolib: sanitize the return value of gpio_chip::get()") Signed-off-by: Dmitry Torokhov Reviewed-by: Andy Shevchenko Reviewed-by: Bartosz Golaszewski Reviewed-by: Linus Walleij Reviewed-by: Geert Uytterhoeven Link: https://patch.msgid.link/aZYnyl-Nf4S1U2yj@google.com Signed-off-by: Geert Uytterhoeven --- drivers/pinctrl/renesas/pinctrl-rza1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pinctrl/renesas/pinctrl-rza1.c b/drivers/pinctrl/renesas/pinctrl-rza1.c index bc8b9b9ad05b..d2949e4dbaf7 100644 --- a/drivers/pinctrl/renesas/pinctrl-rza1.c +++ b/drivers/pinctrl/renesas/pinctrl-rza1.c @@ -589,7 +589,7 @@ static inline unsigned int rza1_get_bit(struct rza1_port *port, { void __iomem *mem = RZA1_ADDR(port->base, reg, port->id); - return ioread16(mem) & BIT(bit); + return !!(ioread16(mem) & BIT(bit)); } /** From a3125bc01884431d30d731461634c8295b6f0529 Mon Sep 17 00:00:00 2001 From: Yazhou Tang Date: Wed, 4 Mar 2026 16:32:27 +0800 Subject: [PATCH 026/352] bpf: Reset register ID for BPF_END value tracking When a register undergoes a BPF_END (byte swap) operation, its scalar value is mutated in-place. If this register previously shared a scalar ID with another register (e.g., after an `r1 = r0` assignment), this tie must be broken. Currently, the verifier misses resetting `dst_reg->id` to 0 for BPF_END. Consequently, if a conditional jump checks the swapped register, the verifier incorrectly propagates the learned bounds to the linked register, leading to false confidence in the linked register's value and potentially allowing out-of-bounds memory accesses. Fix this by explicitly resetting `dst_reg->id` to 0 in the BPF_END case to break the scalar tie, similar to how BPF_NEG handles it via `__mark_reg_known`. Fixes: 9d2119984224 ("bpf: Add bitwise tracking for BPF_END") Closes: https://lore.kernel.org/bpf/AMBPR06MB108683CFEB1CB8D9E02FC95ECF17EA@AMBPR06MB10868.eurprd06.prod.outlook.com/ Link: https://lore.kernel.org/bpf/4be25f7442a52244d0dd1abb47bc6750e57984c9.camel@gmail.com/ Reported-by: Guillaume Laporte Co-developed-by: Tianci Cao Signed-off-by: Tianci Cao Co-developed-by: Shenghao Yuan Signed-off-by: Shenghao Yuan Signed-off-by: Yazhou Tang Acked-by: Eduard Zingerman Link: https://lore.kernel.org/r/20260304083228.142016-2-tangyazhou@zju.edu.cn Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 159b25f8269d..df22bfc572e2 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -15910,6 +15910,13 @@ static void scalar_byte_swap(struct bpf_reg_state *dst_reg, struct bpf_insn *ins /* Apply bswap if alu64 or switch between big-endian and little-endian machines */ bool need_bswap = alu64 || (to_le == is_big_endian); + /* + * If the register is mutated, manually reset its scalar ID to break + * any existing ties and avoid incorrect bounds propagation. + */ + if (need_bswap || insn->imm == 16 || insn->imm == 32) + dst_reg->id = 0; + if (need_bswap) { if (insn->imm == 16) dst_reg->var_off = tnum_bswap16(dst_reg->var_off); From ea1989746b77c3f63bce43af247e1de29ed6bf4a Mon Sep 17 00:00:00 2001 From: Yazhou Tang Date: Wed, 4 Mar 2026 16:32:28 +0800 Subject: [PATCH 027/352] selftests/bpf: Add test for BPF_END register ID reset Add a test case to ensure that BPF_END operations correctly break register's scalar ID ties. The test creates a scenario where r1 is a copy of r0, r0 undergoes a byte swap, and then r0 is checked against a constant. - Without the fix in the verifier, the bounds learned from r0 are incorrectly propagated to r1, making the verifier believe r1 is bounded and wrongly allowing subsequent pointer arithmetic. - With the fix, r1 remains an unbounded scalar, and the verifier correctly rejects the arithmetic operation between the frame pointer and the unbounded register. Co-developed-by: Tianci Cao Signed-off-by: Tianci Cao Co-developed-by: Shenghao Yuan Signed-off-by: Shenghao Yuan Signed-off-by: Yazhou Tang Acked-by: Eduard Zingerman Link: https://lore.kernel.org/r/20260304083228.142016-3-tangyazhou@zju.edu.cn Signed-off-by: Alexei Starovoitov --- .../selftests/bpf/progs/verifier_bswap.c | 22 +++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/tools/testing/selftests/bpf/progs/verifier_bswap.c b/tools/testing/selftests/bpf/progs/verifier_bswap.c index 4b779deee767..cffaf36192bc 100644 --- a/tools/testing/selftests/bpf/progs/verifier_bswap.c +++ b/tools/testing/selftests/bpf/progs/verifier_bswap.c @@ -91,6 +91,28 @@ BSWAP_RANGE_TEST(le32_range, "le32", 0x3f00, 0x3f0000) BSWAP_RANGE_TEST(le64_range, "le64", 0x3f00, 0x3f000000000000) #endif +SEC("socket") +__description("BSWAP, reset reg id") +__failure __msg("math between fp pointer and register with unbounded min value is not allowed") +__naked void bswap_reset_reg_id(void) +{ + asm volatile (" \ + call %[bpf_ktime_get_ns]; \ + r1 = r0; \ + r0 = be16 r0; \ + if r0 != 1 goto l0_%=; \ + r2 = r10; \ + r2 += -512; \ + r2 += r1; \ + *(u8 *)(r2 + 0) = 0; \ +l0_%=: \ + r0 = 0; \ + exit; \ +" : + : __imm(bpf_ktime_get_ns) + : __clobber_all); +} + #else SEC("socket") From 2321a9596d2260310267622e0ad8fbfa6f95378f Mon Sep 17 00:00:00 2001 From: Sachin Kumar Date: Mon, 9 Mar 2026 18:25:42 +0000 Subject: [PATCH 028/352] bpf: Fix constant blinding for PROBE_MEM32 stores BPF_ST | BPF_PROBE_MEM32 immediate stores are not handled by bpf_jit_blind_insn(), allowing user-controlled 32-bit immediates to survive unblinded into JIT-compiled native code when bpf_jit_harden >= 1. The root cause is that convert_ctx_accesses() rewrites BPF_ST|BPF_MEM to BPF_ST|BPF_PROBE_MEM32 for arena pointer stores during verification, before bpf_jit_blind_constants() runs during JIT compilation. The blinding switch only matches BPF_ST|BPF_MEM (mode 0x60), not BPF_ST|BPF_PROBE_MEM32 (mode 0xa0). The instruction falls through unblinded. Add BPF_ST|BPF_PROBE_MEM32 cases to bpf_jit_blind_insn() alongside the existing BPF_ST|BPF_MEM cases. The blinding transformation is identical: load the blinded immediate into BPF_REG_AX via mov+xor, then convert the immediate store to a register store (BPF_STX). The rewritten STX instruction must preserve the BPF_PROBE_MEM32 mode so the architecture JIT emits the correct arena addressing (R12-based on x86-64). Cannot use the BPF_STX_MEM() macro here because it hardcodes BPF_MEM mode; construct the instruction directly instead. Fixes: 6082b6c328b5 ("bpf: Recognize addr_space_cast instruction in the verifier.") Reviewed-by: Puranjay Mohan Reviewed-by: Emil Tsalapatis Signed-off-by: Sachin Kumar Acked-by: Daniel Borkmann Link: https://lore.kernel.org/r/Y6IT5VvNRchPBLI5D7JZHBzZrU9rb0ycRJPJzJSXGj7kJlX8RJwZFSM2YZjcDxoQKABkxt1T8Os2gi23PYyFuQe6KkZGWVyfz8K5afdy9ak=@protonmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/core.c | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+) diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 3ece2da55625..9e126be33755 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -1422,6 +1422,27 @@ static int bpf_jit_blind_insn(const struct bpf_insn *from, *to++ = BPF_ALU64_IMM(BPF_XOR, BPF_REG_AX, imm_rnd); *to++ = BPF_STX_MEM(from->code, from->dst_reg, BPF_REG_AX, from->off); break; + + case BPF_ST | BPF_PROBE_MEM32 | BPF_DW: + case BPF_ST | BPF_PROBE_MEM32 | BPF_W: + case BPF_ST | BPF_PROBE_MEM32 | BPF_H: + case BPF_ST | BPF_PROBE_MEM32 | BPF_B: + *to++ = BPF_ALU64_IMM(BPF_MOV, BPF_REG_AX, imm_rnd ^ + from->imm); + *to++ = BPF_ALU64_IMM(BPF_XOR, BPF_REG_AX, imm_rnd); + /* + * Cannot use BPF_STX_MEM() macro here as it + * hardcodes BPF_MEM mode, losing PROBE_MEM32 + * and breaking arena addressing in the JIT. + */ + *to++ = (struct bpf_insn) { + .code = BPF_STX | BPF_PROBE_MEM32 | + BPF_SIZE(from->code), + .dst_reg = from->dst_reg, + .src_reg = BPF_REG_AX, + .off = from->off, + }; + break; } out: return to - to_buff; From e06e6b8001233241eb5b2e2791162f0585f50f4b Mon Sep 17 00:00:00 2001 From: Paul Chaignon Date: Tue, 10 Mar 2026 12:39:51 +0100 Subject: [PATCH 029/352] selftests/bpf: Fix pkg-config call on static builds For commit b0dcdcb9ae75 ("resolve_btfids: Fix linker flags detection"), I suggested setting HOSTPKG_CONFIG to $PKG_CONFIG when compiling resolve_btfids, but I forgot the quotes around that variable. As a result, when running vmtest.sh with static linking, it fails as follows: $ LDLIBS=-static PKG_CONFIG='pkg-config --static' ./vmtest.sh [...] make: unrecognized option '--static' Usage: make [options] [target] ... [...] This worked when I tested it because HOSTPKG_CONFIG didn't have a default value in the resolve_btfids Makefile, but once it does, the quotes aren't preserved and it fails on the next make call. Fixes: b0dcdcb9ae75 ("resolve_btfids: Fix linker flags detection") Signed-off-by: Paul Chaignon Acked-by: Mykyta Yatsenko Acked-by: Ihor Solodrai Link: https://lore.kernel.org/r/abADBwn_ykblpABE@mail.gmail.com Signed-off-by: Alexei Starovoitov --- tools/testing/selftests/bpf/Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile index d5acbeba0383..65485967c968 100644 --- a/tools/testing/selftests/bpf/Makefile +++ b/tools/testing/selftests/bpf/Makefile @@ -409,7 +409,7 @@ $(RESOLVE_BTFIDS): $(HOST_BPFOBJ) | $(HOST_BUILD_DIR)/resolve_btfids \ CC="$(HOSTCC)" LD="$(HOSTLD)" AR="$(HOSTAR)" \ LIBBPF_INCLUDE=$(HOST_INCLUDE_DIR) \ EXTRA_LDFLAGS='$(SAN_LDFLAGS) $(EXTRA_LDFLAGS)' \ - HOSTPKG_CONFIG=$(PKG_CONFIG) \ + HOSTPKG_CONFIG='$(PKG_CONFIG)' \ OUTPUT=$(HOST_BUILD_DIR)/resolve_btfids/ BPFOBJ=$(HOST_BPFOBJ) # Get Clang's default includes on this system, as opposed to those seen by From 5f47be1b44bf2754c45e8c58ca036b474c9ecbc7 Mon Sep 17 00:00:00 2001 From: Daniel Gomez Date: Tue, 10 Mar 2026 12:36:23 +0100 Subject: [PATCH 030/352] scripts: kconfig: merge_config.sh: pass output file as awk variable The refactoring commit 5fa9b82cbcfc5 ("scripts: kconfig: merge_config.sh: refactor from shell/sed/grep to awk") passes $TMP_FILE.new as ARGV[3] to awk, using it as both an output destination and an input file argument. When the base file is empty, nothing is written to ARGV[3] during processing, so awk fails trying to open it for reading: awk: cmd. line:52: fatal: cannot open file `./.tmp.config.grcQin34jb.new' for reading: No such file or directory mv: cannot stat './.tmp.config.grcQin34jb.new': No such file or directory Pass the output path via -v outfile instead and drop the FILENAME == ARGV[3] { nextfile }. Fixes: 5fa9b82cbcfc5 ("scripts: kconfig: merge_config.sh: refactor from shell/sed/grep to awk") Signed-off-by: Daniel Gomez Link: https://patch.msgid.link/20260310-fixes-merge-config-v1-1-beaeeaded6bd@samsung.com Signed-off-by: Nathan Chancellor --- scripts/kconfig/merge_config.sh | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) diff --git a/scripts/kconfig/merge_config.sh b/scripts/kconfig/merge_config.sh index 073c6bec5245..45e6400c68b6 100755 --- a/scripts/kconfig/merge_config.sh +++ b/scripts/kconfig/merge_config.sh @@ -151,6 +151,7 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do if ! "$AWK" -v prefix="$CONFIG_PREFIX" \ -v warnoverride="$WARNOVERRIDE" \ -v strict="$STRICT" \ + -v outfile="$TMP_FILE.new" \ -v builtin="$BUILTIN" \ -v warnredun="$WARNREDUN" ' BEGIN { @@ -212,7 +213,7 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do # Not a config or not in merge file - keep it if (cfg == "" || !(cfg in merge_cfg)) { - print $0 >> ARGV[3] + print $0 >> outfile next } @@ -222,7 +223,7 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do # BUILTIN: do not demote y to m if (builtin == "true" && new_val ~ /=m$/ && prev_val ~ /=y$/) { warn_builtin(cfg, prev_val, new_val) - print $0 >> ARGV[3] + print $0 >> outfile skip_merge[merge_cfg_line[cfg]] = 1 next } @@ -235,7 +236,7 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do # "=n" is the same as "is not set" if (prev_val ~ /=n$/ && new_val ~ / is not set$/) { - print $0 >> ARGV[3] + print $0 >> outfile next } @@ -246,25 +247,20 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do } } - # output file, skip all lines - FILENAME == ARGV[3] { - nextfile - } - END { # Newline in case base file lacks trailing newline - print "" >> ARGV[3] + print "" >> outfile # Append merge file, skipping lines marked for builtin preservation for (i = 1; i <= merge_total; i++) { if (!(i in skip_merge)) { - print merge_lines[i] >> ARGV[3] + print merge_lines[i] >> outfile } } if (strict_violated) { exit 1 } }' \ - "$ORIG_MERGE_FILE" "$TMP_FILE" "$TMP_FILE.new"; then + "$ORIG_MERGE_FILE" "$TMP_FILE"; then # awk exited non-zero, strict mode was violated STRICT_MODE_VIOLATED=true fi From 775af5cbb22c1de2ad0f486959739c35cfc55ac8 Mon Sep 17 00:00:00 2001 From: Daniel Gomez Date: Tue, 10 Mar 2026 12:36:24 +0100 Subject: [PATCH 031/352] scripts: kconfig: merge_config.sh: fix indentation Replace spaces with tabs for consistency with the rest of the script. Fixes: 5fa9b82cbcfc5 ("scripts: kconfig: merge_config.sh: refactor from shell/sed/grep to awk") Signed-off-by: Daniel Gomez Link: https://patch.msgid.link/20260310-fixes-merge-config-v1-2-beaeeaded6bd@samsung.com Signed-off-by: Nathan Chancellor --- scripts/kconfig/merge_config.sh | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/kconfig/merge_config.sh b/scripts/kconfig/merge_config.sh index 45e6400c68b6..f08e0863b712 100755 --- a/scripts/kconfig/merge_config.sh +++ b/scripts/kconfig/merge_config.sh @@ -196,7 +196,7 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do # First pass: read merge file, store all lines and index FILENAME == ARGV[1] { - mergefile = FILENAME + mergefile = FILENAME merge_lines[FNR] = $0 merge_total = FNR cfg = get_cfg($0) @@ -217,7 +217,7 @@ for ORIG_MERGE_FILE in $MERGE_LIST ; do next } - prev_val = $0 + prev_val = $0 new_val = merge_cfg[cfg] # BUILTIN: do not demote y to m From db9a26765010c55712d8cff32ea9d99732407c55 Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Wed, 25 Feb 2026 17:24:43 +0100 Subject: [PATCH 032/352] dt-bindings: mtd: st,spear600-smi: Fix description The description mixes two nodes. There is the controller, and there is the flash. Describe the flash (which itself can be considered an mtd device, unlike the top level controller), and move the st,smi-fast-mode property inside, as this property is flash specific and should not live in the parent controller node. Fixes: 68cd8ef48452 ("dt-bindings: mtd: st,spear600-smi: convert to DT schema") Reviewed-by: Rob Herring (Arm) Signed-off-by: Miquel Raynal --- .../bindings/mtd/st,spear600-smi.yaml | 22 ++++++++++++++----- 1 file changed, 16 insertions(+), 6 deletions(-) diff --git a/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml b/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml index 8fe27aae7527..d065df269657 100644 --- a/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml +++ b/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml @@ -19,9 +19,6 @@ description: Flash sub nodes describe the memory range and optional per-flash properties. -allOf: - - $ref: mtd.yaml# - properties: compatible: const: st,spear600-smi @@ -42,9 +39,22 @@ properties: $ref: /schemas/types.yaml#/definitions/uint32 description: Functional clock rate of the SMI controller in Hz. - st,smi-fast-mode: - type: boolean - description: Indicates that the attached flash supports fast read mode. +patternProperties: + "^flash@.*$": + $ref: /schemas/mtd/mtd.yaml# + + properties: + reg: + maxItems: 1 + + st,smi-fast-mode: + type: boolean + description: Indicates that the attached flash supports fast read mode. + + unevaluatedProperties: false + + required: + - reg required: - compatible From c21cac8cdcdcb7940c0aab85246ffbb649b73937 Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Wed, 25 Feb 2026 17:24:44 +0100 Subject: [PATCH 033/352] dt-bindings: mtd: st,spear600-smi: #address/size-cells is mandatory These properties must be set because they overwrite the default values, especially #size-cells which is 0 for most controllers and is 'const: 1' here. Fixes: 68cd8ef48452 ("dt-bindings: mtd: st,spear600-smi: convert to DT schema") Reviewed-by: Rob Herring (Arm) Signed-off-by: Miquel Raynal --- Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml | 2 ++ 1 file changed, 2 insertions(+) diff --git a/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml b/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml index d065df269657..62d4359908f2 100644 --- a/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml +++ b/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml @@ -60,6 +60,8 @@ required: - compatible - reg - clock-rate + - "#address-cells" + - "#size-cells" unevaluatedProperties: false From 073b2db72426adee591a0f5a967009ea459ef688 Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Wed, 25 Feb 2026 17:24:45 +0100 Subject: [PATCH 034/352] dt-bindings: mtd: st,spear600-smi: Fix example Example is wrong, the reg property of the flash is always matching the node name. Fixes: 68cd8ef48452 ("dt-bindings: mtd: st,spear600-smi: convert to DT schema") Reviewed-by: Rob Herring (Arm) Signed-off-by: Miquel Raynal --- Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml b/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml index 62d4359908f2..e7385d906591 100644 --- a/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml +++ b/Documentation/devicetree/bindings/mtd/st,spear600-smi.yaml @@ -76,7 +76,7 @@ examples: interrupts = <12>; clock-rate = <50000000>; /* 50 MHz */ - flash@f8000000 { + flash@fc000000 { reg = <0xfc000000 0x1000>; st,smi-fast-mode; }; From bab2bc6e850a697a23b9e5f0e21bb8c187615e95 Mon Sep 17 00:00:00 2001 From: Kamal Dasu Date: Thu, 5 Mar 2026 14:49:06 -0500 Subject: [PATCH 035/352] mtd: rawnand: serialize lock/unlock against other NAND operations nand_lock() and nand_unlock() call into chip->ops.lock_area/unlock_area without holding the NAND device lock. On controllers that implement SET_FEATURES via multiple low-level PIO commands, these can race with concurrent UBI/UBIFS background erase/write operations that hold the device lock, resulting in cmd_pending conflicts on the NAND controller. Add nand_get_device()/nand_release_device() around the lock/unlock operations to serialize them against all other NAND controller access. Fixes: 92270086b7e5 ("mtd: rawnand: Add support for manufacturer specific lock/unlock operation") Signed-off-by: Kamal Dasu Reviewed-by: William Zhang Signed-off-by: Miquel Raynal --- drivers/mtd/nand/raw/nand_base.c | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/drivers/mtd/nand/raw/nand_base.c b/drivers/mtd/nand/raw/nand_base.c index 38429363251c..dfd8361bdd36 100644 --- a/drivers/mtd/nand/raw/nand_base.c +++ b/drivers/mtd/nand/raw/nand_base.c @@ -4737,11 +4737,16 @@ static void nand_shutdown(struct mtd_info *mtd) static int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct nand_chip *chip = mtd_to_nand(mtd); + int ret; if (!chip->ops.lock_area) return -ENOTSUPP; - return chip->ops.lock_area(chip, ofs, len); + nand_get_device(chip); + ret = chip->ops.lock_area(chip, ofs, len); + nand_release_device(chip); + + return ret; } /** @@ -4753,11 +4758,16 @@ static int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len) static int nand_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len) { struct nand_chip *chip = mtd_to_nand(mtd); + int ret; if (!chip->ops.unlock_area) return -ENOTSUPP; - return chip->ops.unlock_area(chip, ofs, len); + nand_get_device(chip); + ret = chip->ops.unlock_area(chip, ofs, len); + nand_release_device(chip); + + return ret; } /* Set default functions */ From da9ba4dcc01e7cf52b7676f0ee9607b8358c2171 Mon Sep 17 00:00:00 2001 From: Kamal Dasu Date: Thu, 5 Mar 2026 15:21:57 -0500 Subject: [PATCH 036/352] mtd: rawnand: brcmnand: skip DMA during panic write When oops_panic_write is set, the driver disables interrupts and switches to PIO polling mode but still falls through into the DMA path. DMA cannot be used reliably in panic context, so make the DMA path an else branch to ensure only PIO is used during panic writes. Fixes: c1ac2dc34b51 ("mtd: rawnand: brcmnand: When oops in progress use pio and interrupt polling") Signed-off-by: Kamal Dasu Reviewed-by: William Zhang Reviewed-by: Florian Fainelli Signed-off-by: Miquel Raynal --- drivers/mtd/nand/raw/brcmnand/brcmnand.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/mtd/nand/raw/brcmnand/brcmnand.c b/drivers/mtd/nand/raw/brcmnand/brcmnand.c index 0427d76f45d0..5b9dadd5405e 100644 --- a/drivers/mtd/nand/raw/brcmnand/brcmnand.c +++ b/drivers/mtd/nand/raw/brcmnand/brcmnand.c @@ -2350,14 +2350,12 @@ static int brcmnand_write(struct mtd_info *mtd, struct nand_chip *chip, for (i = 0; i < ctrl->max_oob; i += 4) oob_reg_write(ctrl, i, 0xffffffff); - if (mtd->oops_panic_write) + if (mtd->oops_panic_write) { /* switch to interrupt polling and PIO mode */ disable_ctrl_irqs(ctrl); - - if (use_dma(ctrl) && (has_edu(ctrl) || !oob) && flash_dma_buf_ok(buf)) { + } else if (use_dma(ctrl) && (has_edu(ctrl) || !oob) && flash_dma_buf_ok(buf)) { if (ctrl->dma_trans(host, addr, (u32 *)buf, oob, mtd->writesize, CMD_PROGRAM_PAGE)) - ret = -EIO; goto out; From edd20cb693d9cb5e3d6fcecd858093dab4e2b0aa Mon Sep 17 00:00:00 2001 From: Wei Liu Date: Wed, 11 Mar 2026 16:51:00 +0000 Subject: [PATCH 037/352] Revert "mshv: expose the scrub partition hypercall" This reverts commit 36d6cbb62133fc6eea28f380409e0fb190f3dfbe. Calling this as a passthrough hypercall leaves the VM in an inconsistent state. Revert before it is released. Signed-off-by: Wei Liu --- drivers/hv/mshv_root_main.c | 1 - include/hyperv/hvgdk_mini.h | 1 - 2 files changed, 2 deletions(-) diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c index 54c3e44d24ee..9d1b881764ed 100644 --- a/drivers/hv/mshv_root_main.c +++ b/drivers/hv/mshv_root_main.c @@ -120,7 +120,6 @@ static u16 mshv_passthru_hvcalls[] = { HVCALL_SET_VP_REGISTERS, HVCALL_TRANSLATE_VIRTUAL_ADDRESS, HVCALL_CLEAR_VIRTUAL_INTERRUPT, - HVCALL_SCRUB_PARTITION, HVCALL_REGISTER_INTERCEPT_RESULT, HVCALL_ASSERT_VIRTUAL_INTERRUPT, HVCALL_GET_GPA_PAGES_ACCESS_STATES, diff --git a/include/hyperv/hvgdk_mini.h b/include/hyperv/hvgdk_mini.h index 8bb3dd71c5b4..1823a290a7b7 100644 --- a/include/hyperv/hvgdk_mini.h +++ b/include/hyperv/hvgdk_mini.h @@ -477,7 +477,6 @@ union hv_vp_assist_msr_contents { /* HV_REGISTER_VP_ASSIST_PAGE */ #define HVCALL_NOTIFY_PARTITION_EVENT 0x0087 #define HVCALL_ENTER_SLEEP_STATE 0x0084 #define HVCALL_NOTIFY_PORT_RING_EMPTY 0x008b -#define HVCALL_SCRUB_PARTITION 0x008d #define HVCALL_REGISTER_INTERCEPT_RESULT 0x0091 #define HVCALL_ASSERT_VIRTUAL_INTERRUPT 0x0094 #define HVCALL_CREATE_PORT 0x0095 From 3fde5281b805370a6c3bd2ef462ebff70a0ea2c6 Mon Sep 17 00:00:00 2001 From: Ard Biesheuvel Date: Mon, 2 Mar 2026 17:45:31 +0100 Subject: [PATCH 038/352] x86/hyperv: Use __naked attribute to fix stackless C function hv_crash_c_entry() is a C function that is entered without a stack, and this is only allowed for functions that have the __naked attribute, which informs the compiler that it must not emit the usual prologue and epilogue or emit any other kind of instrumentation that relies on a stack frame. So split up the function, and set the __naked attribute on the initial part that sets up the stack, GDT, IDT and other pieces that are needed for ordinary C execution. Given that function calls are not permitted either, use the existing long return coded in an asm() block to call the second part of the function, which is an ordinary function that is permitted to call other functions as usual. Reviewed-by: Andrew Cooper # asm parts, not hv parts Reviewed-by: Mukesh Rathor Acked-by: Uros Bizjak Cc: Wei Liu Cc: linux-hyperv@vger.kernel.org Fixes: 94212d34618c ("x86/hyperv: Implement hypervisor RAM collection into vmcore") Signed-off-by: Ard Biesheuvel Signed-off-by: Wei Liu --- arch/x86/hyperv/hv_crash.c | 100 +++++++++++++++++++------------------ 1 file changed, 52 insertions(+), 48 deletions(-) diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c index 92da1b4f2e73..fdb277bf73d8 100644 --- a/arch/x86/hyperv/hv_crash.c +++ b/arch/x86/hyperv/hv_crash.c @@ -107,14 +107,12 @@ static void __noreturn hv_panic_timeout_reboot(void) cpu_relax(); } -/* This cannot be inlined as it needs stack */ -static noinline __noclone void hv_crash_restore_tss(void) +static void hv_crash_restore_tss(void) { load_TR_desc(); } -/* This cannot be inlined as it needs stack */ -static noinline void hv_crash_clear_kernpt(void) +static void hv_crash_clear_kernpt(void) { pgd_t *pgd; p4d_t *p4d; @@ -125,50 +123,9 @@ static noinline void hv_crash_clear_kernpt(void) native_p4d_clear(p4d); } -/* - * This is the C entry point from the asm glue code after the disable hypercall. - * We enter here in IA32-e long mode, ie, full 64bit mode running on kernel - * page tables with our below 4G page identity mapped, but using a temporary - * GDT. ds/fs/gs/es are null. ss is not usable. bp is null. stack is not - * available. We restore kernel GDT, and rest of the context, and continue - * to kexec. - */ -static asmlinkage void __noreturn hv_crash_c_entry(void) + +static void __noreturn hv_crash_handle(void) { - struct hv_crash_ctxt *ctxt = &hv_crash_ctxt; - - /* first thing, restore kernel gdt */ - native_load_gdt(&ctxt->gdtr); - - asm volatile("movw %%ax, %%ss" : : "a"(ctxt->ss)); - asm volatile("movq %0, %%rsp" : : "m"(ctxt->rsp)); - - asm volatile("movw %%ax, %%ds" : : "a"(ctxt->ds)); - asm volatile("movw %%ax, %%es" : : "a"(ctxt->es)); - asm volatile("movw %%ax, %%fs" : : "a"(ctxt->fs)); - asm volatile("movw %%ax, %%gs" : : "a"(ctxt->gs)); - - native_wrmsrq(MSR_IA32_CR_PAT, ctxt->pat); - asm volatile("movq %0, %%cr0" : : "r"(ctxt->cr0)); - - asm volatile("movq %0, %%cr8" : : "r"(ctxt->cr8)); - asm volatile("movq %0, %%cr4" : : "r"(ctxt->cr4)); - asm volatile("movq %0, %%cr2" : : "r"(ctxt->cr4)); - - native_load_idt(&ctxt->idtr); - native_wrmsrq(MSR_GS_BASE, ctxt->gsbase); - native_wrmsrq(MSR_EFER, ctxt->efer); - - /* restore the original kernel CS now via far return */ - asm volatile("movzwq %0, %%rax\n\t" - "pushq %%rax\n\t" - "pushq $1f\n\t" - "lretq\n\t" - "1:nop\n\t" : : "m"(ctxt->cs) : "rax"); - - /* We are in asmlinkage without stack frame, hence make C function - * calls which will buy stack frames. - */ hv_crash_restore_tss(); hv_crash_clear_kernpt(); @@ -177,7 +134,54 @@ static asmlinkage void __noreturn hv_crash_c_entry(void) hv_panic_timeout_reboot(); } -/* Tell gcc we are using lretq long jump in the above function intentionally */ + +/* + * __naked functions do not permit function calls, not even to __always_inline + * functions that only contain asm() blocks themselves. So use a macro instead. + */ +#define hv_wrmsr(msr, val) \ + asm volatile("wrmsr" :: "c"(msr), "a"((u32)val), "d"((u32)(val >> 32)) : "memory") + +/* + * This is the C entry point from the asm glue code after the disable hypercall. + * We enter here in IA32-e long mode, ie, full 64bit mode running on kernel + * page tables with our below 4G page identity mapped, but using a temporary + * GDT. ds/fs/gs/es are null. ss is not usable. bp is null. stack is not + * available. We restore kernel GDT, and rest of the context, and continue + * to kexec. + */ +static void __naked hv_crash_c_entry(void) +{ + /* first thing, restore kernel gdt */ + asm volatile("lgdt %0" : : "m" (hv_crash_ctxt.gdtr)); + + asm volatile("movw %0, %%ss\n\t" + "movq %1, %%rsp" + :: "m"(hv_crash_ctxt.ss), "m"(hv_crash_ctxt.rsp)); + + asm volatile("movw %0, %%ds" : : "m"(hv_crash_ctxt.ds)); + asm volatile("movw %0, %%es" : : "m"(hv_crash_ctxt.es)); + asm volatile("movw %0, %%fs" : : "m"(hv_crash_ctxt.fs)); + asm volatile("movw %0, %%gs" : : "m"(hv_crash_ctxt.gs)); + + hv_wrmsr(MSR_IA32_CR_PAT, hv_crash_ctxt.pat); + asm volatile("movq %0, %%cr0" : : "r"(hv_crash_ctxt.cr0)); + + asm volatile("movq %0, %%cr8" : : "r"(hv_crash_ctxt.cr8)); + asm volatile("movq %0, %%cr4" : : "r"(hv_crash_ctxt.cr4)); + asm volatile("movq %0, %%cr2" : : "r"(hv_crash_ctxt.cr2)); + + asm volatile("lidt %0" : : "m" (hv_crash_ctxt.idtr)); + hv_wrmsr(MSR_GS_BASE, hv_crash_ctxt.gsbase); + hv_wrmsr(MSR_EFER, hv_crash_ctxt.efer); + + /* restore the original kernel CS now via far return */ + asm volatile("pushq %q0\n\t" + "pushq %q1\n\t" + "lretq" + :: "r"(hv_crash_ctxt.cs), "r"(hv_crash_handle)); +} +/* Tell objtool we are using lretq long jump in the above function intentionally */ STACK_FRAME_NON_STANDARD(hv_crash_c_entry); static void hv_mark_tss_not_busy(void) From 3484127c19aca9e93ef6631e7a47bc4f56212da9 Mon Sep 17 00:00:00 2001 From: Uros Bizjak Date: Wed, 11 Mar 2026 11:25:58 +0100 Subject: [PATCH 039/352] x86/hyperv: Save segment registers directly to memory in hv_hvcrash_ctxt_save() hv_hvcrash_ctxt_save() in arch/x86/hyperv/hv_crash.c currently saves segment registers via a general-purpose register (%eax). Update the code to save segment registers (cs, ss, ds, es, fs, gs) directly to the crash context memory using movw. This avoids unnecessary use of a general-purpose register, making the code simpler and more efficient. The size of the corresponding object file improves as follows: text data bss dec hex filename 4167 176 200 4543 11bf hv_crash-old.o 4151 176 200 4527 11af hv_crash-new.o No functional change occurs to the saved context contents; this is purely a code-quality improvement. Signed-off-by: Uros Bizjak Cc: K. Y. Srinivasan Cc: Haiyang Zhang Cc: Wei Liu Cc: Dexuan Cui Cc: Long Li Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Dave Hansen Cc: H. Peter Anvin Signed-off-by: Wei Liu --- arch/x86/hyperv/hv_crash.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c index fdb277bf73d8..2c7ea7e70854 100644 --- a/arch/x86/hyperv/hv_crash.c +++ b/arch/x86/hyperv/hv_crash.c @@ -207,12 +207,12 @@ static void hv_hvcrash_ctxt_save(void) asm volatile("movq %%cr2, %0" : "=a"(ctxt->cr2)); asm volatile("movq %%cr8, %0" : "=a"(ctxt->cr8)); - asm volatile("movl %%cs, %%eax" : "=a"(ctxt->cs)); - asm volatile("movl %%ss, %%eax" : "=a"(ctxt->ss)); - asm volatile("movl %%ds, %%eax" : "=a"(ctxt->ds)); - asm volatile("movl %%es, %%eax" : "=a"(ctxt->es)); - asm volatile("movl %%fs, %%eax" : "=a"(ctxt->fs)); - asm volatile("movl %%gs, %%eax" : "=a"(ctxt->gs)); + asm volatile("movw %%cs, %0" : "=m"(ctxt->cs)); + asm volatile("movw %%ss, %0" : "=m"(ctxt->ss)); + asm volatile("movw %%ds, %0" : "=m"(ctxt->ds)); + asm volatile("movw %%es, %0" : "=m"(ctxt->es)); + asm volatile("movw %%fs, %0" : "=m"(ctxt->fs)); + asm volatile("movw %%gs, %0" : "=m"(ctxt->gs)); native_store_gdt(&ctxt->gdtr); store_idt(&ctxt->idtr); From 2536091d585ac0b60ccf80cbe7a0bf4bfb75ec00 Mon Sep 17 00:00:00 2001 From: Uros Bizjak Date: Wed, 11 Mar 2026 11:25:59 +0100 Subject: [PATCH 040/352] x86/hyperv: Use current_stack_pointer to avoid asm() in hv_hvcrash_ctxt_save() Use current_stack_pointer to avoid asm() when saving %rsp to the crash context memory in hv_hvcrash_ctxt_save(). The new code is more readable and results in exactly the same object file. Signed-off-by: Uros Bizjak Cc: K. Y. Srinivasan Cc: Haiyang Zhang Cc: Wei Liu Cc: Dexuan Cui Cc: Long Li Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Dave Hansen Cc: H. Peter Anvin Signed-off-by: Wei Liu --- arch/x86/hyperv/hv_crash.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c index 2c7ea7e70854..d0f95a278fdb 100644 --- a/arch/x86/hyperv/hv_crash.c +++ b/arch/x86/hyperv/hv_crash.c @@ -199,7 +199,7 @@ static void hv_hvcrash_ctxt_save(void) { struct hv_crash_ctxt *ctxt = &hv_crash_ctxt; - asm volatile("movq %%rsp,%0" : "=m"(ctxt->rsp)); + ctxt->rsp = current_stack_pointer; ctxt->cr0 = native_read_cr0(); ctxt->cr4 = native_read_cr4(); From afeb96cb188d44a61033d838fda0acaa11d1ff24 Mon Sep 17 00:00:00 2001 From: Uros Bizjak Date: Wed, 11 Mar 2026 11:26:00 +0100 Subject: [PATCH 041/352] x86/hyperv: Use any general-purpose register when saving %cr2 and %cr8 hv_hvcrash_ctxt_save() in arch/x86/hyperv/hv_crash.c currently saves %cr2 and %cr8 using %eax ("=a"). This unnecessarily forces a specific register. Update the inline assembly to use a general-purpose register ("=r") for both %cr2 and %cr8. This makes the code more flexible for the compiler while producing the same saved context contents. No functional changes. Signed-off-by: Uros Bizjak Cc: K. Y. Srinivasan Cc: Haiyang Zhang Cc: Wei Liu Cc: Dexuan Cui Cc: Long Li Cc: Thomas Gleixner Cc: Ingo Molnar Cc: Borislav Petkov Cc: Dave Hansen Cc: H. Peter Anvin Signed-off-by: Wei Liu --- arch/x86/hyperv/hv_crash.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/x86/hyperv/hv_crash.c b/arch/x86/hyperv/hv_crash.c index d0f95a278fdb..5ffcc23255de 100644 --- a/arch/x86/hyperv/hv_crash.c +++ b/arch/x86/hyperv/hv_crash.c @@ -204,8 +204,8 @@ static void hv_hvcrash_ctxt_save(void) ctxt->cr0 = native_read_cr0(); ctxt->cr4 = native_read_cr4(); - asm volatile("movq %%cr2, %0" : "=a"(ctxt->cr2)); - asm volatile("movq %%cr8, %0" : "=a"(ctxt->cr8)); + asm volatile("movq %%cr2, %0" : "=r"(ctxt->cr2)); + asm volatile("movq %%cr8, %0" : "=r"(ctxt->cr8)); asm volatile("movw %%cs, %0" : "=m"(ctxt->cs)); asm volatile("movw %%ss, %0" : "=m"(ctxt->ss)); From 0fc773b0e4c1d9fe7cbf56d4df08d7bf90b58fb2 Mon Sep 17 00:00:00 2001 From: Mukesh R Date: Tue, 3 Mar 2026 16:02:51 -0800 Subject: [PATCH 042/352] mshv: pass struct mshv_user_mem_region by reference For unstated reasons, function mshv_partition_ioctl_set_memory passes struct mshv_user_mem_region by value instead of by reference. Change it to pass by reference. Signed-off-by: Mukesh R Reviewed-by: Michael Kelley Signed-off-by: Wei Liu --- drivers/hv/mshv_root_main.c | 26 +++++++++++++------------- 1 file changed, 13 insertions(+), 13 deletions(-) diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c index 9d1b881764ed..e281311b6a9d 100644 --- a/drivers/hv/mshv_root_main.c +++ b/drivers/hv/mshv_root_main.c @@ -1288,7 +1288,7 @@ static int mshv_prepare_pinned_region(struct mshv_mem_region *region) */ static long mshv_map_user_memory(struct mshv_partition *partition, - struct mshv_user_mem_region mem) + struct mshv_user_mem_region *mem) { struct mshv_mem_region *region; struct vm_area_struct *vma; @@ -1296,12 +1296,12 @@ mshv_map_user_memory(struct mshv_partition *partition, ulong mmio_pfn; long ret; - if (mem.flags & BIT(MSHV_SET_MEM_BIT_UNMAP) || - !access_ok((const void __user *)mem.userspace_addr, mem.size)) + if (mem->flags & BIT(MSHV_SET_MEM_BIT_UNMAP) || + !access_ok((const void __user *)mem->userspace_addr, mem->size)) return -EINVAL; mmap_read_lock(current->mm); - vma = vma_lookup(current->mm, mem.userspace_addr); + vma = vma_lookup(current->mm, mem->userspace_addr); is_mmio = vma ? !!(vma->vm_flags & (VM_IO | VM_PFNMAP)) : 0; mmio_pfn = is_mmio ? vma->vm_pgoff : 0; mmap_read_unlock(current->mm); @@ -1309,7 +1309,7 @@ mshv_map_user_memory(struct mshv_partition *partition, if (!vma) return -EINVAL; - ret = mshv_partition_create_region(partition, &mem, ®ion, + ret = mshv_partition_create_region(partition, mem, ®ion, is_mmio); if (ret) return ret; @@ -1354,25 +1354,25 @@ mshv_map_user_memory(struct mshv_partition *partition, /* Called for unmapping both the guest ram and the mmio space */ static long mshv_unmap_user_memory(struct mshv_partition *partition, - struct mshv_user_mem_region mem) + struct mshv_user_mem_region *mem) { struct mshv_mem_region *region; - if (!(mem.flags & BIT(MSHV_SET_MEM_BIT_UNMAP))) + if (!(mem->flags & BIT(MSHV_SET_MEM_BIT_UNMAP))) return -EINVAL; spin_lock(&partition->pt_mem_regions_lock); - region = mshv_partition_region_by_gfn(partition, mem.guest_pfn); + region = mshv_partition_region_by_gfn(partition, mem->guest_pfn); if (!region) { spin_unlock(&partition->pt_mem_regions_lock); return -ENOENT; } /* Paranoia check */ - if (region->start_uaddr != mem.userspace_addr || - region->start_gfn != mem.guest_pfn || - region->nr_pages != HVPFN_DOWN(mem.size)) { + if (region->start_uaddr != mem->userspace_addr || + region->start_gfn != mem->guest_pfn || + region->nr_pages != HVPFN_DOWN(mem->size)) { spin_unlock(&partition->pt_mem_regions_lock); return -EINVAL; } @@ -1403,9 +1403,9 @@ mshv_partition_ioctl_set_memory(struct mshv_partition *partition, return -EINVAL; if (mem.flags & BIT(MSHV_SET_MEM_BIT_UNMAP)) - return mshv_unmap_user_memory(partition, mem); + return mshv_unmap_user_memory(partition, &mem); - return mshv_map_user_memory(partition, mem); + return mshv_map_user_memory(partition, &mem); } static long From 9f455aac17db0aa1486c94dd2c231353ebc9d8bc Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:34 +0100 Subject: [PATCH 043/352] xfrm: state: fix sparse warnings on xfrm_state_hold_rcu In all callers, x is not an __rcu pointer. We can drop the annotation to avoid sparse warnings: net/xfrm/xfrm_state.c:58:39: warning: incorrect type in argument 1 (different address spaces) net/xfrm/xfrm_state.c:58:39: expected struct refcount_struct [usertype] *r net/xfrm/xfrm_state.c:58:39: got struct refcount_struct [noderef] __rcu * net/xfrm/xfrm_state.c:1166:42: warning: incorrect type in argument 1 (different address spaces) net/xfrm/xfrm_state.c:1166:42: expected struct xfrm_state [noderef] __rcu *x net/xfrm/xfrm_state.c:1166:42: got struct xfrm_state *[assigned] x (repeated for each caller) Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index a00c4fe1ab0c..ad32085267a5 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -53,7 +53,7 @@ static DECLARE_WORK(xfrm_state_gc_work, xfrm_state_gc_task); static HLIST_HEAD(xfrm_state_gc_list); static HLIST_HEAD(xfrm_state_dev_gc_list); -static inline bool xfrm_state_hold_rcu(struct xfrm_state __rcu *x) +static inline bool xfrm_state_hold_rcu(struct xfrm_state *x) { return refcount_inc_not_zero(&x->refcnt); } From e2f845f672782b2522062cf1c9aad774276250d7 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:35 +0100 Subject: [PATCH 044/352] xfrm: state: fix sparse warnings in xfrm_state_init Use rcu_assign_pointer, and tmp variables for freeing on the error path without accessing net->xfrm.state_by*. Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 32 ++++++++++++++++++++------------ 1 file changed, 20 insertions(+), 12 deletions(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index ad32085267a5..b81303cccc5e 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -3259,6 +3259,7 @@ EXPORT_SYMBOL(xfrm_init_state); int __net_init xfrm_state_init(struct net *net) { + struct hlist_head *ndst, *nsrc, *nspi, *nseq; unsigned int sz; if (net_eq(net, &init_net)) @@ -3269,18 +3270,25 @@ int __net_init xfrm_state_init(struct net *net) sz = sizeof(struct hlist_head) * 8; - net->xfrm.state_bydst = xfrm_hash_alloc(sz); - if (!net->xfrm.state_bydst) + ndst = xfrm_hash_alloc(sz); + if (!ndst) goto out_bydst; - net->xfrm.state_bysrc = xfrm_hash_alloc(sz); - if (!net->xfrm.state_bysrc) + rcu_assign_pointer(net->xfrm.state_bydst, ndst); + + nsrc = xfrm_hash_alloc(sz); + if (!nsrc) goto out_bysrc; - net->xfrm.state_byspi = xfrm_hash_alloc(sz); - if (!net->xfrm.state_byspi) + rcu_assign_pointer(net->xfrm.state_bysrc, nsrc); + + nspi = xfrm_hash_alloc(sz); + if (!nspi) goto out_byspi; - net->xfrm.state_byseq = xfrm_hash_alloc(sz); - if (!net->xfrm.state_byseq) + rcu_assign_pointer(net->xfrm.state_byspi, nspi); + + nseq = xfrm_hash_alloc(sz); + if (!nseq) goto out_byseq; + rcu_assign_pointer(net->xfrm.state_byseq, nseq); net->xfrm.state_cache_input = alloc_percpu(struct hlist_head); if (!net->xfrm.state_cache_input) @@ -3296,13 +3304,13 @@ int __net_init xfrm_state_init(struct net *net) return 0; out_state_cache_input: - xfrm_hash_free(net->xfrm.state_byseq, sz); + xfrm_hash_free(nseq, sz); out_byseq: - xfrm_hash_free(net->xfrm.state_byspi, sz); + xfrm_hash_free(nspi, sz); out_byspi: - xfrm_hash_free(net->xfrm.state_bysrc, sz); + xfrm_hash_free(nsrc, sz); out_bysrc: - xfrm_hash_free(net->xfrm.state_bydst, sz); + xfrm_hash_free(ndst, sz); out_bydst: return -ENOMEM; } From 55b5bc03148b26ce8156bc47b637a7337aa7d257 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:36 +0100 Subject: [PATCH 045/352] xfrm: state: fix sparse warnings around XFRM_STATE_INSERT We're under xfrm_state_lock in all those cases, use xfrm_state_deref_prot(state_by*) to avoid sparse warnings: net/xfrm/xfrm_state.c:2597:25: warning: cast removes address space '__rcu' of expression net/xfrm/xfrm_state.c:2597:25: warning: incorrect type in argument 2 (different address spaces) net/xfrm/xfrm_state.c:2597:25: expected struct hlist_head *h net/xfrm/xfrm_state.c:2597:25: got struct hlist_head [noderef] __rcu * Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 30 +++++++++++++++++++----------- 1 file changed, 19 insertions(+), 11 deletions(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index b81303cccc5e..34cf9f361683 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1563,23 +1563,23 @@ xfrm_state_find(const xfrm_address_t *daddr, const xfrm_address_t *saddr, list_add(&x->km.all, &net->xfrm.state_all); h = xfrm_dst_hash(net, daddr, saddr, tmpl->reqid, encap_family); XFRM_STATE_INSERT(bydst, &x->bydst, - net->xfrm.state_bydst + h, + xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, x->xso.type); h = xfrm_src_hash(net, daddr, saddr, encap_family); XFRM_STATE_INSERT(bysrc, &x->bysrc, - net->xfrm.state_bysrc + h, + xfrm_state_deref_prot(net->xfrm.state_bysrc, net) + h, x->xso.type); INIT_HLIST_NODE(&x->state_cache); if (x->id.spi) { h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, encap_family); XFRM_STATE_INSERT(byspi, &x->byspi, - net->xfrm.state_byspi + h, + xfrm_state_deref_prot(net->xfrm.state_byspi, net) + h, x->xso.type); } if (x->km.seq) { h = xfrm_seq_hash(net, x->km.seq); XFRM_STATE_INSERT(byseq, &x->byseq, - net->xfrm.state_byseq + h, + xfrm_state_deref_prot(net->xfrm.state_byseq, net) + h, x->xso.type); } x->lft.hard_add_expires_seconds = net->xfrm.sysctl_acq_expires; @@ -1730,25 +1730,29 @@ static void __xfrm_state_insert(struct xfrm_state *x) h = xfrm_dst_hash(net, &x->id.daddr, &x->props.saddr, x->props.reqid, x->props.family); - XFRM_STATE_INSERT(bydst, &x->bydst, net->xfrm.state_bydst + h, + XFRM_STATE_INSERT(bydst, &x->bydst, + xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, x->xso.type); h = xfrm_src_hash(net, &x->id.daddr, &x->props.saddr, x->props.family); - XFRM_STATE_INSERT(bysrc, &x->bysrc, net->xfrm.state_bysrc + h, + XFRM_STATE_INSERT(bysrc, &x->bysrc, + xfrm_state_deref_prot(net->xfrm.state_bysrc, net) + h, x->xso.type); if (x->id.spi) { h = xfrm_spi_hash(net, &x->id.daddr, x->id.spi, x->id.proto, x->props.family); - XFRM_STATE_INSERT(byspi, &x->byspi, net->xfrm.state_byspi + h, + XFRM_STATE_INSERT(byspi, &x->byspi, + xfrm_state_deref_prot(net->xfrm.state_byspi, net) + h, x->xso.type); } if (x->km.seq) { h = xfrm_seq_hash(net, x->km.seq); - XFRM_STATE_INSERT(byseq, &x->byseq, net->xfrm.state_byseq + h, + XFRM_STATE_INSERT(byseq, &x->byseq, + xfrm_state_deref_prot(net->xfrm.state_byseq, net) + h, x->xso.type); } @@ -1868,10 +1872,12 @@ static struct xfrm_state *__find_acq_core(struct net *net, ktime_set(net->xfrm.sysctl_acq_expires, 0), HRTIMER_MODE_REL_SOFT); list_add(&x->km.all, &net->xfrm.state_all); - XFRM_STATE_INSERT(bydst, &x->bydst, net->xfrm.state_bydst + h, + XFRM_STATE_INSERT(bydst, &x->bydst, + xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, x->xso.type); h = xfrm_src_hash(net, daddr, saddr, family); - XFRM_STATE_INSERT(bysrc, &x->bysrc, net->xfrm.state_bysrc + h, + XFRM_STATE_INSERT(bysrc, &x->bysrc, + xfrm_state_deref_prot(net->xfrm.state_bysrc, net) + h, x->xso.type); net->xfrm.state_num++; @@ -2603,7 +2609,9 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high, if (!x0) { x->id.spi = newspi; h = xfrm_spi_hash(net, &x->id.daddr, newspi, x->id.proto, x->props.family); - XFRM_STATE_INSERT(byspi, &x->byspi, net->xfrm.state_byspi + h, x->xso.type); + XFRM_STATE_INSERT(byspi, &x->byspi, + xfrm_state_deref_prot(net->xfrm.state_byspi, net) + h, + x->xso.type); spin_unlock_bh(&net->xfrm.xfrm_state_lock); err = 0; goto unlock; From 33cefb76a8edee8af257abfe6f42fb987c77132f Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:37 +0100 Subject: [PATCH 046/352] xfrm: state: add xfrm_state_deref_prot to state_by* walk under lock We're under xfrm_state_lock for all those walks, we can use xfrm_state_deref_prot to silence sparse warnings such as: net/xfrm/xfrm_state.c:933:17: warning: dereference of noderef expression Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 22 +++++++++++----------- 1 file changed, 11 insertions(+), 11 deletions(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 34cf9f361683..27192b11be43 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -870,7 +870,7 @@ xfrm_state_flush_secctx_check(struct net *net, u8 proto, bool task_valid) for (i = 0; i <= net->xfrm.state_hmask; i++) { struct xfrm_state *x; - hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + i, bydst) { if (xfrm_id_proto_match(x->id.proto, proto) && (err = security_xfrm_state_delete(x)) != 0) { xfrm_audit_state_delete(x, 0, task_valid); @@ -891,7 +891,7 @@ xfrm_dev_state_flush_secctx_check(struct net *net, struct net_device *dev, bool struct xfrm_state *x; struct xfrm_dev_offload *xso; - hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + i, bydst) { xso = &x->xso; if (xso->dev == dev && @@ -931,7 +931,7 @@ int xfrm_state_flush(struct net *net, u8 proto, bool task_valid) for (i = 0; i <= net->xfrm.state_hmask; i++) { struct xfrm_state *x; restart: - hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + i, bydst) { if (!xfrm_state_kern(x) && xfrm_id_proto_match(x->id.proto, proto)) { xfrm_state_hold(x); @@ -973,7 +973,7 @@ int xfrm_dev_state_flush(struct net *net, struct net_device *dev, bool task_vali err = -ESRCH; for (i = 0; i <= net->xfrm.state_hmask; i++) { restart: - hlist_for_each_entry(x, net->xfrm.state_bydst+i, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + i, bydst) { xso = &x->xso; if (!xfrm_state_kern(x) && xso->dev == dev) { @@ -1652,7 +1652,7 @@ xfrm_stateonly_find(struct net *net, u32 mark, u32 if_id, spin_lock_bh(&net->xfrm.xfrm_state_lock); h = xfrm_dst_hash(net, daddr, saddr, reqid, family); - hlist_for_each_entry(x, net->xfrm.state_bydst+h, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, bydst) { if (x->props.family == family && x->props.reqid == reqid && (mark & x->mark.m) == x->mark.v && @@ -1779,7 +1779,7 @@ static void __xfrm_state_bump_genids(struct xfrm_state *xnew) u32 cpu_id = xnew->pcpu_num; h = xfrm_dst_hash(net, &xnew->id.daddr, &xnew->props.saddr, reqid, family); - hlist_for_each_entry(x, net->xfrm.state_bydst+h, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, bydst) { if (x->props.family == family && x->props.reqid == reqid && x->if_id == if_id && @@ -1815,7 +1815,7 @@ static struct xfrm_state *__find_acq_core(struct net *net, struct xfrm_state *x; u32 mark = m->v & m->m; - hlist_for_each_entry(x, net->xfrm.state_bydst+h, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, bydst) { if (x->props.reqid != reqid || x->props.mode != mode || x->props.family != family || @@ -2097,7 +2097,7 @@ struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *n if (m->reqid) { h = xfrm_dst_hash(net, &m->old_daddr, &m->old_saddr, m->reqid, m->old_family); - hlist_for_each_entry(x, net->xfrm.state_bydst+h, bydst) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + h, bydst) { if (x->props.mode != m->mode || x->id.proto != m->proto) continue; @@ -2116,7 +2116,7 @@ struct xfrm_state *xfrm_migrate_state_find(struct xfrm_migrate *m, struct net *n } else { h = xfrm_src_hash(net, &m->old_daddr, &m->old_saddr, m->old_family); - hlist_for_each_entry(x, net->xfrm.state_bysrc+h, bysrc) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bysrc, net) + h, bysrc) { if (x->props.mode != m->mode || x->id.proto != m->proto) continue; @@ -2319,7 +2319,7 @@ void xfrm_state_update_stats(struct net *net) spin_lock_bh(&net->xfrm.xfrm_state_lock); for (i = 0; i <= net->xfrm.state_hmask; i++) { - hlist_for_each_entry(x, net->xfrm.state_bydst + i, bydst) + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_bydst, net) + i, bydst) xfrm_dev_state_update_stats(x); } spin_unlock_bh(&net->xfrm.xfrm_state_lock); @@ -2510,7 +2510,7 @@ static struct xfrm_state *__xfrm_find_acq_byseq(struct net *net, u32 mark, u32 s unsigned int h = xfrm_seq_hash(net, seq); struct xfrm_state *x; - hlist_for_each_entry_rcu(x, net->xfrm.state_byseq + h, byseq) { + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_byseq, net) + h, byseq) { if (x->km.seq == seq && (mark & x->mark.m) == x->mark.v && x->pcpu_num == pcpu_num && From f468fdd52b97a63c4fb916fb882b936d8b43b8ae Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:38 +0100 Subject: [PATCH 047/352] xfrm: remove rcu/state_hold from xfrm_state_lookup_spi_proto xfrm_state_lookup_spi_proto is called under xfrm_state_lock by xfrm_alloc_spi, no need to take a reference on the state and pretend to be under RCU. Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index 27192b11be43..f28cbe249c05 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -1703,18 +1703,12 @@ static struct xfrm_state *xfrm_state_lookup_spi_proto(struct net *net, __be32 sp struct xfrm_state *x; unsigned int i; - rcu_read_lock(); for (i = 0; i <= net->xfrm.state_hmask; i++) { - hlist_for_each_entry_rcu(x, &net->xfrm.state_byspi[i], byspi) { - if (x->id.spi == spi && x->id.proto == proto) { - if (!xfrm_state_hold_rcu(x)) - continue; - rcu_read_unlock(); + hlist_for_each_entry(x, xfrm_state_deref_prot(net->xfrm.state_byspi, net) + i, byspi) { + if (x->id.spi == spi && x->id.proto == proto) return x; - } } } - rcu_read_unlock(); return NULL; } @@ -2616,7 +2610,6 @@ int xfrm_alloc_spi(struct xfrm_state *x, u32 low, u32 high, err = 0; goto unlock; } - xfrm_state_put(x0); spin_unlock_bh(&net->xfrm.xfrm_state_lock); next: From 05b8673963c492fe36533e99a4a3c6661ca09ed0 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:39 +0100 Subject: [PATCH 048/352] xfrm: state: silence sparse warnings during netns exit Silence sparse warnings in xfrm_state_fini: net/xfrm/xfrm_state.c:3327:9: warning: incorrect type in argument 1 (different address spaces) net/xfrm/xfrm_state.c:3327:9: expected struct hlist_head const *h net/xfrm/xfrm_state.c:3327:9: got struct hlist_head [noderef] __rcu *state_byseq Add xfrm_state_deref_netexit() to wrap those calls. The netns is going away, we don't have to worry about the state_by* pointers being changed behind our backs. Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_state.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c index f28cbe249c05..1748d374abca 100644 --- a/net/xfrm/xfrm_state.c +++ b/net/xfrm/xfrm_state.c @@ -3316,6 +3316,8 @@ int __net_init xfrm_state_init(struct net *net) return -ENOMEM; } +#define xfrm_state_deref_netexit(table) \ + rcu_dereference_protected((table), true /* netns is going away */) void xfrm_state_fini(struct net *net) { unsigned int sz; @@ -3328,17 +3330,17 @@ void xfrm_state_fini(struct net *net) WARN_ON(!list_empty(&net->xfrm.state_all)); for (i = 0; i <= net->xfrm.state_hmask; i++) { - WARN_ON(!hlist_empty(net->xfrm.state_byseq + i)); - WARN_ON(!hlist_empty(net->xfrm.state_byspi + i)); - WARN_ON(!hlist_empty(net->xfrm.state_bysrc + i)); - WARN_ON(!hlist_empty(net->xfrm.state_bydst + i)); + WARN_ON(!hlist_empty(xfrm_state_deref_netexit(net->xfrm.state_byseq) + i)); + WARN_ON(!hlist_empty(xfrm_state_deref_netexit(net->xfrm.state_byspi) + i)); + WARN_ON(!hlist_empty(xfrm_state_deref_netexit(net->xfrm.state_bysrc) + i)); + WARN_ON(!hlist_empty(xfrm_state_deref_netexit(net->xfrm.state_bydst) + i)); } sz = (net->xfrm.state_hmask + 1) * sizeof(struct hlist_head); - xfrm_hash_free(net->xfrm.state_byseq, sz); - xfrm_hash_free(net->xfrm.state_byspi, sz); - xfrm_hash_free(net->xfrm.state_bysrc, sz); - xfrm_hash_free(net->xfrm.state_bydst, sz); + xfrm_hash_free(xfrm_state_deref_netexit(net->xfrm.state_byseq), sz); + xfrm_hash_free(xfrm_state_deref_netexit(net->xfrm.state_byspi), sz); + xfrm_hash_free(xfrm_state_deref_netexit(net->xfrm.state_bysrc), sz); + xfrm_hash_free(xfrm_state_deref_netexit(net->xfrm.state_bydst), sz); free_percpu(net->xfrm.state_cache_input); } From b1f9c67781efd8a0ebd5019f14fbbac981cff7c1 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:40 +0100 Subject: [PATCH 049/352] xfrm: policy: fix sparse warnings in xfrm_policy_{init,fini} In xfrm_policy_init: add rcu_assign_pointer to fix warning: net/xfrm/xfrm_policy.c:4238:29: warning: incorrect type in assignment (different address spaces) net/xfrm/xfrm_policy.c:4238:29: expected struct hlist_head [noderef] __rcu *table net/xfrm/xfrm_policy.c:4238:29: got struct hlist_head * add rcu_dereference_protected to silence warning: net/xfrm/xfrm_policy.c:4265:36: warning: incorrect type in argument 1 (different address spaces) net/xfrm/xfrm_policy.c:4265:36: expected struct hlist_head *n net/xfrm/xfrm_policy.c:4265:36: got struct hlist_head [noderef] __rcu *table The netns is being created, no concurrent access is possible yet. In xfrm_policy_fini, net is going away, there shouldn't be any concurrent changes to the hashtables, so we can use rcu_dereference_protected to silence warnings: net/xfrm/xfrm_policy.c:4291:17: warning: incorrect type in argument 1 (different address spaces) net/xfrm/xfrm_policy.c:4291:17: expected struct hlist_head const *h net/xfrm/xfrm_policy.c:4291:17: got struct hlist_head [noderef] __rcu *table net/xfrm/xfrm_policy.c:4292:36: warning: incorrect type in argument 1 (different address spaces) net/xfrm/xfrm_policy.c:4292:36: expected struct hlist_head *n net/xfrm/xfrm_policy.c:4292:36: got struct hlist_head [noderef] __rcu *table Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 5428185196a1..49de5a6f4b85 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4242,7 +4242,7 @@ static int __net_init xfrm_policy_init(struct net *net) net->xfrm.policy_count[XFRM_POLICY_MAX + dir] = 0; htab = &net->xfrm.policy_bydst[dir]; - htab->table = xfrm_hash_alloc(sz); + rcu_assign_pointer(htab->table, xfrm_hash_alloc(sz)); if (!htab->table) goto out_bydst; htab->hmask = hmask; @@ -4269,7 +4269,7 @@ static int __net_init xfrm_policy_init(struct net *net) struct xfrm_policy_hash *htab; htab = &net->xfrm.policy_bydst[dir]; - xfrm_hash_free(htab->table, sz); + xfrm_hash_free(rcu_dereference_protected(htab->table, true), sz); } xfrm_hash_free(net->xfrm.policy_byidx, sz); out_byidx: @@ -4295,8 +4295,8 @@ static void xfrm_policy_fini(struct net *net) htab = &net->xfrm.policy_bydst[dir]; sz = (htab->hmask + 1) * sizeof(struct hlist_head); - WARN_ON(!hlist_empty(htab->table)); - xfrm_hash_free(htab->table, sz); + WARN_ON(!hlist_empty(rcu_dereference_protected(htab->table, true))); + xfrm_hash_free(rcu_dereference_protected(htab->table, true), sz); } sz = (net->xfrm.policy_idx_hmask + 1) * sizeof(struct hlist_head); From 2da6901866e7137f4e1a51a5f0bd1fbd0848a4eb Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:41 +0100 Subject: [PATCH 050/352] xfrm: policy: silence sparse warning in xfrm_policy_unregister_afinfo xfrm_policy_afinfo is __rcu, use rcu_access_pointer to silence: net/xfrm/xfrm_policy.c:4152:43: error: incompatible types in comparison expression (different address spaces): net/xfrm/xfrm_policy.c:4152:43: struct xfrm_policy_afinfo const [noderef] __rcu * net/xfrm/xfrm_policy.c:4152:43: struct xfrm_policy_afinfo const * Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 49de5a6f4b85..66ac93e65264 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4156,7 +4156,7 @@ void xfrm_policy_unregister_afinfo(const struct xfrm_policy_afinfo *afinfo) int i; for (i = 0; i < ARRAY_SIZE(xfrm_policy_afinfo); i++) { - if (xfrm_policy_afinfo[i] != afinfo) + if (rcu_access_pointer(xfrm_policy_afinfo[i]) != afinfo) continue; RCU_INIT_POINTER(xfrm_policy_afinfo[i], NULL); break; From 103b4f5b4007cb484f40b1c8095a7e0526e5aff6 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:42 +0100 Subject: [PATCH 051/352] xfrm: add rcu_access_pointer to silence sparse warning for xfrm_input_afinfo xfrm_input_afinfo is __rcu, we should use rcu_access_pointer to avoid a sparse warning: net/xfrm/xfrm_input.c:78:21: error: incompatible types in comparison expression (different address spaces): net/xfrm/xfrm_input.c:78:21: struct xfrm_input_afinfo const [noderef] __rcu * net/xfrm/xfrm_input.c:78:21: struct xfrm_input_afinfo const * Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_input.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c index 4ed346e682c7..dc1312ed5a09 100644 --- a/net/xfrm/xfrm_input.c +++ b/net/xfrm/xfrm_input.c @@ -75,7 +75,10 @@ int xfrm_input_unregister_afinfo(const struct xfrm_input_afinfo *afinfo) spin_lock_bh(&xfrm_input_afinfo_lock); if (likely(xfrm_input_afinfo[afinfo->is_ipip][afinfo->family])) { - if (unlikely(xfrm_input_afinfo[afinfo->is_ipip][afinfo->family] != afinfo)) + const struct xfrm_input_afinfo *cur; + + cur = rcu_access_pointer(xfrm_input_afinfo[afinfo->is_ipip][afinfo->family]); + if (unlikely(cur != afinfo)) err = -EINVAL; else RCU_INIT_POINTER(xfrm_input_afinfo[afinfo->is_ipip][afinfo->family], NULL); From d87f8bc47fbf012a7f115e311d0603d97e47c34c Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 9 Mar 2026 11:32:43 +0100 Subject: [PATCH 052/352] xfrm: avoid RCU warnings around the per-netns netlink socket net->xfrm.nlsk is used in 2 types of contexts: - fully under RCU, with rcu_read_lock + rcu_dereference and a NULL check - in the netlink handlers, with requests coming from a userspace socket In the 2nd case, net->xfrm.nlsk is guaranteed to stay non-NULL and the object is alive, since we can't enter the netns destruction path while the user socket holds a reference on the netns. After adding the __rcu annotation to netns_xfrm.nlsk (which silences sparse warnings in the RCU users and __net_init code), we need to tell sparse that the 2nd case is safe. Add a helper for that. Signed-off-by: Sabrina Dubroca Reviewed-by: Simon Horman Signed-off-by: Steffen Klassert --- include/net/netns/xfrm.h | 2 +- net/xfrm/xfrm_user.c | 25 +++++++++++++++++-------- 2 files changed, 18 insertions(+), 9 deletions(-) diff --git a/include/net/netns/xfrm.h b/include/net/netns/xfrm.h index 23dd647fe024..b73983a17e08 100644 --- a/include/net/netns/xfrm.h +++ b/include/net/netns/xfrm.h @@ -59,7 +59,7 @@ struct netns_xfrm { struct list_head inexact_bins; - struct sock *nlsk; + struct sock __rcu *nlsk; struct sock *nlsk_stash; u32 sysctl_aevent_etime; diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 4dd8341225bc..1656b487f833 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -35,6 +35,15 @@ #endif #include +static struct sock *xfrm_net_nlsk(const struct net *net, const struct sk_buff *skb) +{ + /* get the source of this request, see netlink_unicast_kernel */ + const struct sock *sk = NETLINK_CB(skb).sk; + + /* sk is refcounted, the netns stays alive and nlsk with it */ + return rcu_dereference_protected(net->xfrm.nlsk, sk->sk_net_refcnt); +} + static int verify_one_alg(struct nlattr **attrs, enum xfrm_attr_type_t type, struct netlink_ext_ack *extack) { @@ -1727,7 +1736,7 @@ static int xfrm_get_spdinfo(struct sk_buff *skb, struct nlmsghdr *nlh, err = build_spdinfo(r_skb, net, sportid, seq, *flags); BUG_ON(err < 0); - return nlmsg_unicast(net->xfrm.nlsk, r_skb, sportid); + return nlmsg_unicast(xfrm_net_nlsk(net, skb), r_skb, sportid); } static inline unsigned int xfrm_sadinfo_msgsize(void) @@ -1787,7 +1796,7 @@ static int xfrm_get_sadinfo(struct sk_buff *skb, struct nlmsghdr *nlh, err = build_sadinfo(r_skb, net, sportid, seq, *flags); BUG_ON(err < 0); - return nlmsg_unicast(net->xfrm.nlsk, r_skb, sportid); + return nlmsg_unicast(xfrm_net_nlsk(net, skb), r_skb, sportid); } static int xfrm_get_sa(struct sk_buff *skb, struct nlmsghdr *nlh, @@ -1807,7 +1816,7 @@ static int xfrm_get_sa(struct sk_buff *skb, struct nlmsghdr *nlh, if (IS_ERR(resp_skb)) { err = PTR_ERR(resp_skb); } else { - err = nlmsg_unicast(net->xfrm.nlsk, resp_skb, NETLINK_CB(skb).portid); + err = nlmsg_unicast(xfrm_net_nlsk(net, skb), resp_skb, NETLINK_CB(skb).portid); } xfrm_state_put(x); out_noput: @@ -1898,7 +1907,7 @@ static int xfrm_alloc_userspi(struct sk_buff *skb, struct nlmsghdr *nlh, } } - err = nlmsg_unicast(net->xfrm.nlsk, resp_skb, NETLINK_CB(skb).portid); + err = nlmsg_unicast(xfrm_net_nlsk(net, skb), resp_skb, NETLINK_CB(skb).portid); out: xfrm_state_put(x); @@ -2543,7 +2552,7 @@ static int xfrm_get_default(struct sk_buff *skb, struct nlmsghdr *nlh, r_up->out = net->xfrm.policy_default[XFRM_POLICY_OUT]; nlmsg_end(r_skb, r_nlh); - return nlmsg_unicast(net->xfrm.nlsk, r_skb, portid); + return nlmsg_unicast(xfrm_net_nlsk(net, skb), r_skb, portid); } static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, @@ -2609,7 +2618,7 @@ static int xfrm_get_policy(struct sk_buff *skb, struct nlmsghdr *nlh, if (IS_ERR(resp_skb)) { err = PTR_ERR(resp_skb); } else { - err = nlmsg_unicast(net->xfrm.nlsk, resp_skb, + err = nlmsg_unicast(xfrm_net_nlsk(net, skb), resp_skb, NETLINK_CB(skb).portid); } } else { @@ -2782,7 +2791,7 @@ static int xfrm_get_ae(struct sk_buff *skb, struct nlmsghdr *nlh, err = build_aevent(r_skb, x, &c); BUG_ON(err < 0); - err = nlmsg_unicast(net->xfrm.nlsk, r_skb, NETLINK_CB(skb).portid); + err = nlmsg_unicast(xfrm_net_nlsk(net, skb), r_skb, NETLINK_CB(skb).portid); spin_unlock_bh(&x->lock); xfrm_state_put(x); return err; @@ -3486,7 +3495,7 @@ static int xfrm_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh, goto err; } - err = netlink_dump_start(net->xfrm.nlsk, skb, nlh, &c); + err = netlink_dump_start(xfrm_net_nlsk(net, skb), skb, nlh, &c); goto err; } From 14de1552a4e3fece78bb20314887e70888c9d448 Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Wed, 11 Mar 2026 16:14:55 -0700 Subject: [PATCH 053/352] include/linux/local_lock_internal.h: Make this header file again compatible with sparse There are two versions of the __this_cpu_local_lock() definitions in include/linux/local_lock_internal.h: one version that relies on the Clang overloading functionality and another version that does not. Select the latter version when using sparse. This patch fixes the following errors reported by sparse: include/linux/local_lock_internal.h:331:40: sparse: sparse: multiple definitions for function '__this_cpu_local_lock' include/linux/local_lock_internal.h:325:37: sparse: the previous one is here Closes: https://lore.kernel.org/oe-kbuild-all/202603062334.wgI5htP0-lkp@intel.com/ Fixes: d3febf16dee2 ("locking/local_lock: Support Clang's context analysis") Reported-by: kernel test robot Signed-off-by: Bart Van Assche Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Marco Elver Link: https://patch.msgid.link/20260311231455.1961413-1-bvanassche@acm.org --- include/linux/local_lock_internal.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/linux/local_lock_internal.h b/include/linux/local_lock_internal.h index eff711bf973f..234be7f12c15 100644 --- a/include/linux/local_lock_internal.h +++ b/include/linux/local_lock_internal.h @@ -315,7 +315,7 @@ do { \ #endif /* CONFIG_PREEMPT_RT */ -#if defined(WARN_CONTEXT_ANALYSIS) +#if defined(WARN_CONTEXT_ANALYSIS) && !defined(__CHECKER__) /* * Because the compiler only knows about the base per-CPU variable, use this * helper function to make the compiler think we lock/unlock the @base variable, From 8d5fae6011260de209aaf231120e8146b14bc8e0 Mon Sep 17 00:00:00 2001 From: Breno Leitao Date: Tue, 10 Mar 2026 03:13:16 -0700 Subject: [PATCH 054/352] perf/x86: Move event pointer setup earlier in x86_pmu_enable() A production AMD EPYC system crashed with a NULL pointer dereference in the PMU NMI handler: BUG: kernel NULL pointer dereference, address: 0000000000000198 RIP: x86_perf_event_update+0xc/0xa0 Call Trace: amd_pmu_v2_handle_irq+0x1a6/0x390 perf_event_nmi_handler+0x24/0x40 The faulting instruction is `cmpq $0x0, 0x198(%rdi)` with RDI=0, corresponding to the `if (unlikely(!hwc->event_base))` check in x86_perf_event_update() where hwc = &event->hw and event is NULL. drgn inspection of the vmcore on CPU 106 showed a mismatch between cpuc->active_mask and cpuc->events[]: active_mask: 0x1e (bits 1, 2, 3, 4) events[1]: 0xff1100136cbd4f38 (valid) events[2]: 0x0 (NULL, but active_mask bit 2 set) events[3]: 0xff1100076fd2cf38 (valid) events[4]: 0xff1100079e990a90 (valid) The event that should occupy events[2] was found in event_list[2] with hw.idx=2 and hw.state=0x0, confirming x86_pmu_start() had run (which clears hw.state and sets active_mask) but events[2] was never populated. Another event (event_list[0]) had hw.state=0x7 (STOPPED|UPTODATE|ARCH), showing it was stopped when the PMU rescheduled events, confirming the throttle-then-reschedule sequence occurred. The root cause is commit 7e772a93eb61 ("perf/x86: Fix NULL event access and potential PEBS record loss") which moved the cpuc->events[idx] assignment out of x86_pmu_start() and into step 2 of x86_pmu_enable(), after the PERF_HES_ARCH check. This broke any path that calls pmu->start() without going through x86_pmu_enable() -- specifically the unthrottle path: perf_adjust_freq_unthr_events() -> perf_event_unthrottle_group() -> perf_event_unthrottle() -> event->pmu->start(event, 0) -> x86_pmu_start() // sets active_mask but not events[] The race sequence is: 1. A group of perf events overflows, triggering group throttle via perf_event_throttle_group(). All events are stopped: active_mask bits cleared, events[] preserved (x86_pmu_stop no longer clears events[] after commit 7e772a93eb61). 2. While still throttled (PERF_HES_STOPPED), x86_pmu_enable() runs due to other scheduling activity. Stopped events that need to move counters get PERF_HES_ARCH set and events[old_idx] cleared. In step 2 of x86_pmu_enable(), PERF_HES_ARCH causes these events to be skipped -- events[new_idx] is never set. 3. The timer tick unthrottles the group via pmu->start(). Since commit 7e772a93eb61 removed the events[] assignment from x86_pmu_start(), active_mask[new_idx] is set but events[new_idx] remains NULL. 4. A PMC overflow NMI fires. The handler iterates active counters, finds active_mask[2] set, reads events[2] which is NULL, and crashes dereferencing it. Move the cpuc->events[hwc->idx] assignment in x86_pmu_enable() to before the PERF_HES_ARCH check, so that events[] is populated even for events that are not immediately started. This ensures the unthrottle path via pmu->start() always finds a valid event pointer. Fixes: 7e772a93eb61 ("perf/x86: Fix NULL event access and potential PEBS record loss") Signed-off-by: Breno Leitao Signed-off-by: Peter Zijlstra (Intel) Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260310-perf-v2-1-4a3156fce43c@debian.org --- arch/x86/events/core.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index 03ce1bc7ef2e..54b4c315d927 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -1372,6 +1372,8 @@ static void x86_pmu_enable(struct pmu *pmu) else if (i < n_running) continue; + cpuc->events[hwc->idx] = event; + if (hwc->state & PERF_HES_ARCH) continue; @@ -1379,7 +1381,6 @@ static void x86_pmu_enable(struct pmu *pmu) * if cpuc->enabled = 0, then no wrmsr as * per x86_pmu_enable_event() */ - cpuc->events[hwc->idx] = event; x86_pmu_start(event, PERF_EF_RELOAD); } cpuc->n_added = 0; From f1cac6ac62d28a9a57b17f51ac5795bf250c12d3 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 11 Mar 2026 21:29:14 +0100 Subject: [PATCH 055/352] x86/perf: Make sure to program the counter value for stopped events on migration Both Mi Dapeng and Ian Rogers noted that not everything that sets HES_STOPPED is required to EF_UPDATE. Specifically the 'step 1' loop of rescheduling explicitly does EF_UPDATE to ensure the counter value is read. However, then 'step 2' simply leaves the new counter uninitialized when HES_STOPPED, even though, as noted above, the thing that stopped them might not be aware it needs to EF_RELOAD -- since it didn't EF_UPDATE on stop. One such location that is affected is throttling, throttle does pmu->stop(, 0); and unthrottle does pmu->start(, 0); possibly restarting an uninitialized counter. Fixes: a4eaf7f14675 ("perf: Rework the PMU methods") Reported-by: Dapeng Mi Reported-by: Ian Rogers Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Dapeng Mi Link: https://patch.msgid.link/20260311204035.GX606826@noisy.programming.kicks-ass.net --- arch/x86/events/core.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/x86/events/core.c b/arch/x86/events/core.c index 54b4c315d927..810ab21ffd99 100644 --- a/arch/x86/events/core.c +++ b/arch/x86/events/core.c @@ -1374,8 +1374,10 @@ static void x86_pmu_enable(struct pmu *pmu) cpuc->events[hwc->idx] = event; - if (hwc->state & PERF_HES_ARCH) + if (hwc->state & PERF_HES_ARCH) { + static_call(x86_pmu_set_period)(event); continue; + } /* * if cpuc->enabled = 0, then no wrmsr as From 4b9ce671960627b2505b3f64742544ae9801df97 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 9 Mar 2026 13:55:46 +0100 Subject: [PATCH 056/352] perf: Make sure to use pmu_ctx->pmu for groups Oliver reported that x86_pmu_del() ended up doing an out-of-bound memory access when group_sched_in() fails and needs to roll back. This *should* be handled by the transaction callbacks, but he found that when the group leader is a software event, the transaction handlers of the wrong PMU are used. Despite the move_group case in perf_event_open() and group_sched_in() using pmu_ctx->pmu. Turns out, inherit uses event->pmu to clone the events, effectively undoing the move_group case for all inherited contexts. Fix this by also making inherit use pmu_ctx->pmu, ensuring all inherited counters end up in the same pmu context. Similarly, __perf_event_read() should use equally use pmu_ctx->pmu for the group case. Fixes: bd2756811766 ("perf: Rewrite core context handling") Reported-by: Oliver Rosenberg Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ian Rogers Link: https://patch.msgid.link/20260309133713.GB606826@noisy.programming.kicks-ass.net --- kernel/events/core.c | 19 ++++++++----------- 1 file changed, 8 insertions(+), 11 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index 1f5699b339ec..89b40e439717 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -4813,7 +4813,7 @@ static void __perf_event_read(void *info) struct perf_event *sub, *event = data->event; struct perf_event_context *ctx = event->ctx; struct perf_cpu_context *cpuctx = this_cpu_ptr(&perf_cpu_context); - struct pmu *pmu = event->pmu; + struct pmu *pmu; /* * If this is a task context, we need to check whether it is @@ -4825,7 +4825,7 @@ static void __perf_event_read(void *info) if (ctx->task && cpuctx->task_ctx != ctx) return; - raw_spin_lock(&ctx->lock); + guard(raw_spinlock)(&ctx->lock); ctx_time_update_event(ctx, event); perf_event_update_time(event); @@ -4833,25 +4833,22 @@ static void __perf_event_read(void *info) perf_event_update_sibling_time(event); if (event->state != PERF_EVENT_STATE_ACTIVE) - goto unlock; + return; if (!data->group) { - pmu->read(event); + perf_pmu_read(event); data->ret = 0; - goto unlock; + return; } + pmu = event->pmu_ctx->pmu; pmu->start_txn(pmu, PERF_PMU_TXN_READ); - pmu->read(event); - + perf_pmu_read(event); for_each_sibling_event(sub, event) perf_pmu_read(sub); data->ret = pmu->commit_txn(pmu); - -unlock: - raw_spin_unlock(&ctx->lock); } static inline u64 perf_event_count(struct perf_event *event, bool self) @@ -14744,7 +14741,7 @@ inherit_event(struct perf_event *parent_event, get_ctx(child_ctx); child_event->ctx = child_ctx; - pmu_ctx = find_get_pmu_context(child_event->pmu, child_ctx, child_event); + pmu_ctx = find_get_pmu_context(parent_event->pmu_ctx->pmu, child_ctx, child_event); if (IS_ERR(pmu_ctx)) { free_event(child_event); return ERR_CAST(pmu_ctx); From 1d07bbd7ea36ea0b8dfa8068dbe67eb3a32d9590 Mon Sep 17 00:00:00 2001 From: Dapeng Mi Date: Sat, 28 Feb 2026 13:33:20 +0800 Subject: [PATCH 057/352] perf/x86/intel: Add missing branch counters constraint apply When running the command: 'perf record -e "{instructions,instructions:p}" -j any,counter sleep 1', a "shift-out-of-bounds" warning is reported on CWF. UBSAN: shift-out-of-bounds in /kbuild/src/consumer/arch/x86/events/intel/lbr.c:970:15 shift exponent 64 is too large for 64-bit type 'long long unsigned int' ...... intel_pmu_lbr_counters_reorder.isra.0.cold+0x2a/0xa7 intel_pmu_lbr_save_brstack+0xc0/0x4c0 setup_arch_pebs_sample_data+0x114b/0x2400 The warning occurs because the second "instructions:p" event, which involves branch counters sampling, is incorrectly programmed to fixed counter 0 instead of the general-purpose (GP) counters 0-3 that support branch counters sampling. Currently only GP counters 0-3 support branch counters sampling on CWF, any event involving branch counters sampling should be programed on GP counters 0-3. Since the counter index of fixed counter 0 is 32, it leads to the "src" value in below code is right shifted 64 bits and trigger the "shift-out-of-bounds" warning. cnt = (src >> (order[j] * LBR_INFO_BR_CNTR_BITS)) & LBR_INFO_BR_CNTR_MASK; The root cause is the loss of the branch counters constraint for the new event in the branch counters sampling event group. Since it isn't yet part of the sibling list. This results in the second "instructions:p" event being programmed on fixed counter 0 incorrectly instead of the appropriate GP counters 0-3. To address this, we apply the missing branch counters constraint for the last event in the group. Additionally, we introduce a new function, `intel_set_branch_counter_constr()`, to apply the branch counters constraint and avoid code duplication. Fixes: 33744916196b ("perf/x86/intel: Support branch counters logging") Reported-by: Xudong Hao Signed-off-by: Dapeng Mi Signed-off-by: Peter Zijlstra (Intel) Link: https://patch.msgid.link/20260228053320.140406-2-dapeng1.mi@linux.intel.com Cc: stable@vger.kernel.org --- arch/x86/events/intel/core.c | 31 +++++++++++++++++++++---------- 1 file changed, 21 insertions(+), 10 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index cf3a4fe06ff2..36c68210d4d2 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -4628,6 +4628,19 @@ static inline void intel_pmu_set_acr_caused_constr(struct perf_event *event, event->hw.dyn_constraint &= hybrid(event->pmu, acr_cause_mask64); } +static inline int intel_set_branch_counter_constr(struct perf_event *event, + int *num) +{ + if (branch_sample_call_stack(event)) + return -EINVAL; + if (branch_sample_counters(event)) { + (*num)++; + event->hw.dyn_constraint &= x86_pmu.lbr_counters; + } + + return 0; +} + static int intel_pmu_hw_config(struct perf_event *event) { int ret = x86_pmu_hw_config(event); @@ -4698,21 +4711,19 @@ static int intel_pmu_hw_config(struct perf_event *event) * group, which requires the extra space to store the counters. */ leader = event->group_leader; - if (branch_sample_call_stack(leader)) + if (intel_set_branch_counter_constr(leader, &num)) return -EINVAL; - if (branch_sample_counters(leader)) { - num++; - leader->hw.dyn_constraint &= x86_pmu.lbr_counters; - } leader->hw.flags |= PERF_X86_EVENT_BRANCH_COUNTERS; for_each_sibling_event(sibling, leader) { - if (branch_sample_call_stack(sibling)) + if (intel_set_branch_counter_constr(sibling, &num)) + return -EINVAL; + } + + /* event isn't installed as a sibling yet. */ + if (event != leader) { + if (intel_set_branch_counter_constr(event, &num)) return -EINVAL; - if (branch_sample_counters(sibling)) { - num++; - sibling->hw.dyn_constraint &= x86_pmu.lbr_counters; - } } if (num > fls(x86_pmu.lbr_counters)) From e7fcc54524f04e42641de99028edd9c69dc19f8c Mon Sep 17 00:00:00 2001 From: Dapeng Mi Date: Wed, 11 Mar 2026 15:52:00 +0800 Subject: [PATCH 058/352] perf/x86/intel: Fix OMR snoop information parsing issues When omr_source is 0x2, the omr_snoop (bit[6]) and omr_promoted (bit[7]) fields are combined to represent the snoop information. However, the omr_promoted field was not left-shifted by 1 bit, resulting in incorrect snoop information. Besides, the snoop information parsing is not accurate for some OMR sources, like the snoop information should be SNOOP_NONE for these memory access (omr_source >= 7) instead of SNOOP_HIT. Fix these issues. Closes: https://lore.kernel.org/all/CAP-5=fW4zLWFw1v38zCzB9-cseNSTTCtup=p2SDxZq7dPayVww@mail.gmail.com/ Fixes: d2bdcde9626c ("perf/x86/intel: Add support for PEBS memory auxiliary info field in DMR") Reported-by: Ian Rogers Signed-off-by: Dapeng Mi Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ian Rogers Link: https://patch.msgid.link/20260311075201.2951073-1-dapeng1.mi@linux.intel.com --- arch/x86/events/intel/ds.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c index 5027afc97b65..7f0d515c07c5 100644 --- a/arch/x86/events/intel/ds.c +++ b/arch/x86/events/intel/ds.c @@ -345,12 +345,12 @@ static u64 parse_omr_data_source(u8 dse) if (omr.omr_remote) val |= REM; - val |= omr.omr_hitm ? P(SNOOP, HITM) : P(SNOOP, HIT); - if (omr.omr_source == 0x2) { - u8 snoop = omr.omr_snoop | omr.omr_promoted; + u8 snoop = omr.omr_snoop | (omr.omr_promoted << 1); - if (snoop == 0x0) + if (omr.omr_hitm) + val |= P(SNOOP, HITM); + else if (snoop == 0x0) val |= P(SNOOP, NA); else if (snoop == 0x1) val |= P(SNOOP, MISS); @@ -359,7 +359,10 @@ static u64 parse_omr_data_source(u8 dse) else if (snoop == 0x3) val |= P(SNOOP, NONE); } else if (omr.omr_source > 0x2 && omr.omr_source < 0x7) { + val |= omr.omr_hitm ? P(SNOOP, HITM) : P(SNOOP, HIT); val |= omr.omr_snoop ? P(SNOOPX, FWD) : 0; + } else { + val |= P(SNOOP, NONE); } return val; From b22c526569e6af84008b674e66378e771bfbdd94 Mon Sep 17 00:00:00 2001 From: Adam Ford Date: Mon, 9 Feb 2026 23:37:01 -0600 Subject: [PATCH 059/352] pmdomain: mediatek: Fix power domain count The wrong value of the number of domains is wrong which leads to failures when trying to enumerate nested power domains. PM: genpd_xlate_onecell: invalid domain index 0 PM: genpd_xlate_onecell: invalid domain index 1 PM: genpd_xlate_onecell: invalid domain index 3 PM: genpd_xlate_onecell: invalid domain index 4 PM: genpd_xlate_onecell: invalid domain index 5 PM: genpd_xlate_onecell: invalid domain index 13 PM: genpd_xlate_onecell: invalid domain index 14 Attempts to use these power domains fail, so fix this by using the correct value of calculated power domains. Signed-off-by: Adam Ford Fixes: 88914db077b6 ("pmdomain: mediatek: Add support for Hardware Voter power domains") Reviewed-by: AngeloGioacchino Del Regno Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/pmdomain/mediatek/mtk-pm-domains.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/pmdomain/mediatek/mtk-pm-domains.c b/drivers/pmdomain/mediatek/mtk-pm-domains.c index f64f24d520dd..e2800aa1bc59 100644 --- a/drivers/pmdomain/mediatek/mtk-pm-domains.c +++ b/drivers/pmdomain/mediatek/mtk-pm-domains.c @@ -1203,7 +1203,7 @@ static int scpsys_probe(struct platform_device *pdev) scpsys->soc_data = soc; scpsys->pd_data.domains = scpsys->domains; - scpsys->pd_data.num_domains = soc->num_domains; + scpsys->pd_data.num_domains = num_domains; parent = dev->parent; if (!parent) { From a424a34b8faddf97b5af41689087e7a230f79ba7 Mon Sep 17 00:00:00 2001 From: Raul E Rangel Date: Mon, 9 Feb 2026 13:58:18 -0700 Subject: [PATCH 060/352] serial: 8250: Fix TX deadlock when using DMA `dmaengine_terminate_async` does not guarantee that the `__dma_tx_complete` callback will run. The callback is currently the only place where `dma->tx_running` gets cleared. If the transaction is canceled and the callback never runs, then `dma->tx_running` will never get cleared and we will never schedule new TX DMA transactions again. This change makes it so we clear `dma->tx_running` after we terminate the DMA transaction. This is "safe" because `serial8250_tx_dma_flush` is holding the UART port lock. The first thing the callback does is also grab the UART port lock, so access to `dma->tx_running` is serialized. Fixes: 9e512eaaf8f4 ("serial: 8250: Fix fifo underflow on flush") Cc: stable Signed-off-by: Raul E Rangel Link: https://patch.msgid.link/20260209135815.1.I16366ecb0f62f3c96fe3dd5763fcf6f3c2b4d8cd@changeid Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_dma.c | 15 +++++++++++++++ 1 file changed, 15 insertions(+) diff --git a/drivers/tty/serial/8250/8250_dma.c b/drivers/tty/serial/8250/8250_dma.c index bdd26c9f34bd..3b6452e759d5 100644 --- a/drivers/tty/serial/8250/8250_dma.c +++ b/drivers/tty/serial/8250/8250_dma.c @@ -162,7 +162,22 @@ void serial8250_tx_dma_flush(struct uart_8250_port *p) */ dma->tx_size = 0; + /* + * We can't use `dmaengine_terminate_sync` because `uart_flush_buffer` is + * holding the uart port spinlock. + */ dmaengine_terminate_async(dma->txchan); + + /* + * The callback might or might not run. If it doesn't run, we need to ensure + * that `tx_running` is cleared so that we can schedule new transactions. + * If it does run, then the zombie callback will clear `tx_running` again + * and perform a no-op since `tx_size` was cleared above. + * + * In either case, we ASSUME the DMA transaction will terminate before we + * issue a new `serial8250_tx_dma`. + */ + dma->tx_running = 0; } int serial8250_rx_dma(struct uart_8250_port *p) From 24b98e8664e157aff0814a0f49895ee8223f382f Mon Sep 17 00:00:00 2001 From: Peng Zhang Date: Tue, 24 Feb 2026 13:16:39 +0100 Subject: [PATCH 061/352] serial: 8250: always disable IRQ during THRE test commit 039d4926379b ("serial: 8250: Toggle IER bits on only after irq has been set up") moved IRQ setup before the THRE test, in combination with commit 205d300aea75 ("serial: 8250: change lock order in serial8250_do_startup()") the interrupt handler can run during the test and race with its IIR reads. This can produce wrong THRE test results and cause spurious registration of the serial8250_backup_timeout timer. Unconditionally disable the IRQ for the short duration of the test and re-enable it afterwards to avoid the race. Fixes: 039d4926379b ("serial: 8250: Toggle IER bits on only after irq has been set up") Depends-on: 205d300aea75 ("serial: 8250: change lock order in serial8250_do_startup()") Cc: stable Signed-off-by: Peng Zhang Reviewed-by: Muchun Song Signed-off-by: Alban Bedel Tested-by: Maximilian Lueer Link: https://patch.msgid.link/20260224121639.579404-1-alban.bedel@lht.dlh.de Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_port.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index cc94af2d578a..a743964c9d22 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -2147,8 +2147,7 @@ static void serial8250_THRE_test(struct uart_port *port) if (up->port.flags & UPF_NO_THRE_TEST) return; - if (port->irqflags & IRQF_SHARED) - disable_irq_nosync(port->irq); + disable_irq(port->irq); /* * Test for UARTs that do not reassert THRE when the transmitter is idle and the interrupt @@ -2170,8 +2169,7 @@ static void serial8250_THRE_test(struct uart_port *port) serial_port_out(port, UART_IER, 0); } - if (port->irqflags & IRQF_SHARED) - enable_irq(port->irq); + enable_irq(port->irq); /* * If the interrupt is not reasserted, or we otherwise don't trust the iir, setup a timer to From d54801cd509515f674a5aac1d3ea1401d2a05863 Mon Sep 17 00:00:00 2001 From: Maciej Andrzejewski ICEYE Date: Thu, 5 Mar 2026 13:37:51 +0100 Subject: [PATCH 062/352] serial: uartlite: fix PM runtime usage count underflow on probe ulite_probe() calls pm_runtime_put_autosuspend() at the end of probe without holding a corresponding PM runtime reference for non-console ports. During ulite_assign(), uart_add_one_port() triggers uart_configure_port() which calls ulite_pm() via uart_change_pm(). For non-console ports, the UART core performs a balanced get/put cycle: uart_change_pm(ON) -> ulite_pm() -> pm_runtime_get_sync() +1 uart_change_pm(OFF) -> ulite_pm() -> pm_runtime_put_autosuspend() -1 This leaves no spare reference for the pm_runtime_put_autosuspend() at the end of probe. The PM runtime core prevents the count from actually going below zero, and instead triggers a "Runtime PM usage count underflow!" warning. For console ports the bug is masked: the UART core skips the uart_change_pm(OFF) call, so the UART core's unbalanced get happens to pair with probe's trailing put. Add pm_runtime_get_noresume() before pm_runtime_enable() to take an explicit probe-owned reference that the trailing pm_runtime_put_autosuspend() can release. This ensures a correct usage count regardless of whether the port is a console. Fixes: 5bbe10a6942d ("tty: serial: uartlite: Add runtime pm support") Cc: stable Signed-off-by: Maciej Andrzejewski ICEYE Link: https://patch.msgid.link/20260305123746.4152800-1-maciej.andrzejewski@m-works.net Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/uartlite.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/tty/serial/uartlite.c b/drivers/tty/serial/uartlite.c index 39c1fd1ff9ce..6240c3d4dfd7 100644 --- a/drivers/tty/serial/uartlite.c +++ b/drivers/tty/serial/uartlite.c @@ -878,6 +878,7 @@ static int ulite_probe(struct platform_device *pdev) pm_runtime_use_autosuspend(&pdev->dev); pm_runtime_set_autosuspend_delay(&pdev->dev, UART_AUTOSUSPEND_TIMEOUT); pm_runtime_set_active(&pdev->dev); + pm_runtime_get_noresume(&pdev->dev); pm_runtime_enable(&pdev->dev); ret = ulite_assign(&pdev->dev, id, res->start, irq, pdata); From 455ce986fa356ff43a43c0d363ba95fa152f21d5 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Wed, 4 Feb 2026 15:43:20 +0800 Subject: [PATCH 063/352] serial: core: fix infinite loop in handle_tx() for PORT_UNKNOWN uart_write_room() and uart_write() behave inconsistently when xmit_buf is NULL (which happens for PORT_UNKNOWN ports that were never properly initialized): - uart_write_room() returns kfifo_avail() which can be > 0 - uart_write() checks xmit_buf and returns 0 if NULL This inconsistency causes an infinite loop in drivers that rely on tty_write_room() to determine if they can write: while (tty_write_room(tty) > 0) { written = tty->ops->write(...); // written is always 0, loop never exits } For example, caif_serial's handle_tx() enters an infinite loop when used with PORT_UNKNOWN serial ports, causing system hangs. Fix by making uart_write_room() also check xmit_buf and return 0 if it's NULL, consistent with uart_write(). Reproducer: https://gist.github.com/mrpre/d9a694cc0e19828ee3bc3b37983fde13 Signed-off-by: Jiayuan Chen Cc: stable Link: https://patch.msgid.link/20260204074327.226165-1-jiayuan.chen@linux.dev Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/serial_core.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c index 487756947a96..89cebdd27841 100644 --- a/drivers/tty/serial/serial_core.c +++ b/drivers/tty/serial/serial_core.c @@ -643,7 +643,10 @@ static unsigned int uart_write_room(struct tty_struct *tty) unsigned int ret; port = uart_port_ref_lock(state, &flags); - ret = kfifo_avail(&state->port.xmit_fifo); + if (!state->port.xmit_buf) + ret = 0; + else + ret = kfifo_avail(&state->port.xmit_fifo); uart_port_unlock_deref(port, flags); return ret; } From 9c0072bc33d349c83d223e64be30794e11938a6b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Martin=20Roukala=20=28n=C3=A9=20Peres=29?= Date: Mon, 9 Mar 2026 15:53:10 +0200 Subject: [PATCH 064/352] serial: 8250_pci: add support for the AX99100 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This is found in popular brands such as StarTech.com or Delock, and has been a source of frustration to quite a few people, if I can trust Amazon comments complaining about Linux support via the official out-of-the-tree driver. Signed-off-by: Martin Roukala (né Peres) Cc: stable Link: https://patch.msgid.link/20260309-8250_pci_ax99100-v1-1-3328bdfd8e94@mupuf.org Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_pci.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c index aa1ab4da9ff1..6cfd1b2af5b7 100644 --- a/drivers/tty/serial/8250/8250_pci.c +++ b/drivers/tty/serial/8250/8250_pci.c @@ -137,6 +137,8 @@ struct serial_private { }; #define PCI_DEVICE_ID_HPE_PCI_SERIAL 0x37e +#define PCIE_VENDOR_ID_ASIX 0x125B +#define PCIE_DEVICE_ID_AX99100 0x9100 static const struct pci_device_id pci_use_msi[] = { { PCI_DEVICE_SUB(PCI_VENDOR_ID_NETMOS, PCI_DEVICE_ID_NETMOS_9900, @@ -149,6 +151,8 @@ static const struct pci_device_id pci_use_msi[] = { 0xA000, 0x1000) }, { PCI_DEVICE_SUB(PCI_VENDOR_ID_HP_3PAR, PCI_DEVICE_ID_HPE_PCI_SERIAL, PCI_ANY_ID, PCI_ANY_ID) }, + { PCI_DEVICE_SUB(PCIE_VENDOR_ID_ASIX, PCIE_DEVICE_ID_AX99100, + 0xA000, 0x1000) }, { } }; @@ -920,6 +924,7 @@ static int pci_netmos_init(struct pci_dev *dev) case PCI_DEVICE_ID_NETMOS_9912: case PCI_DEVICE_ID_NETMOS_9922: case PCI_DEVICE_ID_NETMOS_9900: + case PCIE_DEVICE_ID_AX99100: num_serial = pci_netmos_9900_numports(dev); break; @@ -2544,6 +2549,14 @@ static struct pci_serial_quirk pci_serial_quirks[] = { .init = pci_netmos_init, .setup = pci_netmos_9900_setup, }, + { + .vendor = PCIE_VENDOR_ID_ASIX, + .device = PCI_ANY_ID, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + .init = pci_netmos_init, + .setup = pci_netmos_9900_setup, + }, /* * EndRun Technologies */ @@ -6065,6 +6078,10 @@ static const struct pci_device_id serial_pci_tbl[] = { 0xA000, 0x3002, 0, 0, pbn_NETMOS9900_2s_115200 }, + { PCIE_VENDOR_ID_ASIX, PCIE_DEVICE_ID_AX99100, + 0xA000, 0x1000, + 0, 0, pbn_b0_1_115200 }, + /* * Best Connectivity and Rosewill PCI Multi I/O cards */ From 59a33d83bbe6d73d2071d7ae21590b29faed0503 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:43 +0200 Subject: [PATCH 065/352] serial: 8250: Protect LCR write in shutdown MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The 8250_dw driver needs to potentially perform very complex operations during LCR writes because its BUSY handling prevents updates to LCR while UART is BUSY (which is not fully under our control without those complex operations). Thus, LCR writes should occur under port's lock. Move LCR write under port's lock in serial8250_do_shutdown(). Also split the LCR RMW so that the logic is on a separate line for clarity. Reported-by: Bandal, Shankar Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Cc: stable Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-2-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_port.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index a743964c9d22..20cf123a0540 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -2348,6 +2348,7 @@ static int serial8250_startup(struct uart_port *port) void serial8250_do_shutdown(struct uart_port *port) { struct uart_8250_port *up = up_to_u8250p(port); + u32 lcr; serial8250_rpm_get(up); /* @@ -2374,13 +2375,13 @@ void serial8250_do_shutdown(struct uart_port *port) port->mctrl &= ~TIOCM_OUT2; serial8250_set_mctrl(port, port->mctrl); + + /* Disable break condition */ + lcr = serial_port_in(port, UART_LCR); + lcr &= ~UART_LCR_SBC; + serial_port_out(port, UART_LCR, lcr); } - /* - * Disable break condition and FIFOs - */ - serial_port_out(port, UART_LCR, - serial_port_in(port, UART_LCR) & ~UART_LCR_SBC); serial8250_clear_fifos(up); rsa_disable(up); From 8002d6d6d0d8a36a7d6ca523b17a51cb0fa7c3c3 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:44 +0200 Subject: [PATCH 066/352] serial: 8250_dw: Avoid unnecessary LCR writes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When DW UART is configured with BUSY flag, LCR writes may not always succeed which can make any LCR write complex and very expensive. Performing write directly can trigger IRQ and the driver has to perform complex and distruptive sequence while retrying the write. Therefore, it's better to avoid doing LCR write that would not change the value of the LCR register. Add LCR write avoidance code into the 8250_dw driver's .serial_out() functions. Reported-by: Bandal, Shankar Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Cc: stable Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-3-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_dw.c | 31 +++++++++++++++++++++++++++++++ 1 file changed, 31 insertions(+) diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c index db73b2ae17fa..f47ede4d1968 100644 --- a/drivers/tty/serial/8250/8250_dw.c +++ b/drivers/tty/serial/8250/8250_dw.c @@ -181,6 +181,22 @@ static void dw8250_check_lcr(struct uart_port *p, unsigned int offset, u32 value */ } +/* + * With BUSY, LCR writes can be very expensive (IRQ + complex retry logic). + * If the write does not change the value of the LCR register, skip it entirely. + */ +static bool dw8250_can_skip_reg_write(struct uart_port *p, unsigned int offset, u32 value) +{ + struct dw8250_data *d = to_dw8250_data(p->private_data); + u32 lcr; + + if (offset != UART_LCR || d->uart_16550_compatible) + return false; + + lcr = serial_port_in(p, offset); + return lcr == value; +} + /* Returns once the transmitter is empty or we run out of retries */ static void dw8250_tx_wait_empty(struct uart_port *p) { @@ -207,12 +223,18 @@ static void dw8250_tx_wait_empty(struct uart_port *p) static void dw8250_serial_out(struct uart_port *p, unsigned int offset, u32 value) { + if (dw8250_can_skip_reg_write(p, offset, value)) + return; + writeb(value, p->membase + (offset << p->regshift)); dw8250_check_lcr(p, offset, value); } static void dw8250_serial_out38x(struct uart_port *p, unsigned int offset, u32 value) { + if (dw8250_can_skip_reg_write(p, offset, value)) + return; + /* Allow the TX to drain before we reconfigure */ if (offset == UART_LCR) dw8250_tx_wait_empty(p); @@ -237,6 +259,9 @@ static u32 dw8250_serial_inq(struct uart_port *p, unsigned int offset) static void dw8250_serial_outq(struct uart_port *p, unsigned int offset, u32 value) { + if (dw8250_can_skip_reg_write(p, offset, value)) + return; + value &= 0xff; __raw_writeq(value, p->membase + (offset << p->regshift)); /* Read back to ensure register write ordering. */ @@ -248,6 +273,9 @@ static void dw8250_serial_outq(struct uart_port *p, unsigned int offset, u32 val static void dw8250_serial_out32(struct uart_port *p, unsigned int offset, u32 value) { + if (dw8250_can_skip_reg_write(p, offset, value)) + return; + writel(value, p->membase + (offset << p->regshift)); dw8250_check_lcr(p, offset, value); } @@ -261,6 +289,9 @@ static u32 dw8250_serial_in32(struct uart_port *p, unsigned int offset) static void dw8250_serial_out32be(struct uart_port *p, unsigned int offset, u32 value) { + if (dw8250_can_skip_reg_write(p, offset, value)) + return; + iowrite32be(value, p->membase + (offset << p->regshift)); dw8250_check_lcr(p, offset, value); } From 8324a54f604da18f21070702a8ad82ab2062787b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:45 +0200 Subject: [PATCH 067/352] serial: 8250: Add serial8250_handle_irq_locked() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 8250_port exports serial8250_handle_irq() to HW specific 8250 drivers. It takes port's lock within but a HW specific 8250 driver may want to take port's lock itself, do something, and then call the generic handler in 8250_port but to do that, the caller has to release port's lock for no good reason. Introduce serial8250_handle_irq_locked() which a HW specific driver can call while already holding port's lock. As this is new export, put it straight into a namespace (where all 8250 exports should eventually be moved). Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Cc: stable Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-4-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_port.c | 24 ++++++++++++++++-------- include/linux/serial_8250.h | 1 + 2 files changed, 17 insertions(+), 8 deletions(-) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index 20cf123a0540..14d6aca44551 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -1782,20 +1783,16 @@ static bool handle_rx_dma(struct uart_8250_port *up, unsigned int iir) } /* - * This handles the interrupt from one port. + * Context: port's lock must be held by the caller. */ -int serial8250_handle_irq(struct uart_port *port, unsigned int iir) +void serial8250_handle_irq_locked(struct uart_port *port, unsigned int iir) { struct uart_8250_port *up = up_to_u8250p(port); struct tty_port *tport = &port->state->port; bool skip_rx = false; - unsigned long flags; u16 status; - if (iir & UART_IIR_NO_INT) - return 0; - - uart_port_lock_irqsave(port, &flags); + lockdep_assert_held_once(&port->lock); status = serial_lsr_in(up); @@ -1828,8 +1825,19 @@ int serial8250_handle_irq(struct uart_port *port, unsigned int iir) else if (!up->dma->tx_running) __stop_tx(up); } +} +EXPORT_SYMBOL_NS_GPL(serial8250_handle_irq_locked, "SERIAL_8250"); - uart_unlock_and_check_sysrq_irqrestore(port, flags); +/* + * This handles the interrupt from one port. + */ +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); + serial8250_handle_irq_locked(port, iir); return 1; } diff --git a/include/linux/serial_8250.h b/include/linux/serial_8250.h index 01efdce0fda0..a95b2d143d24 100644 --- a/include/linux/serial_8250.h +++ b/include/linux/serial_8250.h @@ -195,6 +195,7 @@ void serial8250_do_set_mctrl(struct uart_port *port, unsigned int mctrl); void serial8250_do_set_divisor(struct uart_port *port, unsigned int baud, unsigned int quot); int fsl8250_handle_irq(struct uart_port *port); +void serial8250_handle_irq_locked(struct uart_port *port, unsigned int iir); int serial8250_handle_irq(struct uart_port *port, unsigned int iir); u16 serial8250_rx_chars(struct uart_8250_port *up, u16 lsr); void serial8250_read_char(struct uart_8250_port *up, u16 lsr); From 883c5a2bc934c165c4491d1ef7da0ac4e9765077 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:46 +0200 Subject: [PATCH 068/352] serial: 8250_dw: Rework dw8250_handle_irq() locking and IIR handling MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit dw8250_handle_irq() takes port's lock multiple times with no good reason to release it in between and calls serial8250_handle_irq() that also takes port's lock. Take port's lock only once in dw8250_handle_irq() and use serial8250_handle_irq_locked() to avoid releasing port's lock in between. As IIR_NO_INT check in serial8250_handle_irq() was outside of port's lock, it has to be done already in dw8250_handle_irq(). DW UART can, in addition to IIR_NO_INT, report BUSY_DETECT (0x7) which collided with the IIR_NO_INT (0x1) check in serial8250_handle_irq() (because & is used instead of ==) meaning that no other work is done by serial8250_handle_irq() during an BUSY_DETECT interrupt. This allows reorganizing code in dw8250_handle_irq() to do both IIR_NO_INT and BUSY_DETECT handling right at the start simplifying the logic. Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Cc: stable Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-5-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_dw.c | 37 ++++++++++++++++++------------- 1 file changed, 21 insertions(+), 16 deletions(-) diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c index f47ede4d1968..abf0814086a5 100644 --- a/drivers/tty/serial/8250/8250_dw.c +++ b/drivers/tty/serial/8250/8250_dw.c @@ -9,6 +9,9 @@ * LCR is written whilst busy. If it is, then a busy detect interrupt is * raised, the LCR needs to be rewritten and the uart status register read. */ +#include +#include +#include #include #include #include @@ -40,6 +43,8 @@ #define RZN1_UART_RDMACR 0x110 /* DMA Control Register Receive Mode */ /* DesignWare specific register fields */ +#define DW_UART_IIR_IID GENMASK(3, 0) + #define DW_UART_MCR_SIRE BIT(6) /* Renesas specific register fields */ @@ -312,7 +317,19 @@ static int dw8250_handle_irq(struct uart_port *p) bool rx_timeout = (iir & 0x3f) == UART_IIR_RX_TIMEOUT; unsigned int quirks = d->pdata->quirks; unsigned int status; - unsigned long flags; + + switch (FIELD_GET(DW_UART_IIR_IID, iir)) { + case UART_IIR_NO_INT: + return 0; + + case UART_IIR_BUSY: + /* Clear the USR */ + serial_port_in(p, d->pdata->usr_reg); + + return 1; + } + + guard(uart_port_lock_irqsave)(p); /* * There are ways to get Designware-based UARTs into a state where @@ -325,20 +342,15 @@ static int dw8250_handle_irq(struct uart_port *p) * so we limit the workaround only to non-DMA mode. */ if (!up->dma && rx_timeout) { - uart_port_lock_irqsave(p, &flags); status = serial_lsr_in(up); if (!(status & (UART_LSR_DR | UART_LSR_BI))) serial_port_in(p, UART_RX); - - uart_port_unlock_irqrestore(p, flags); } /* Manually stop the Rx DMA transfer when acting as flow controller */ if (quirks & DW_UART_QUIRK_IS_DMA_FC && up->dma && up->dma->rx_running && rx_timeout) { - uart_port_lock_irqsave(p, &flags); status = serial_lsr_in(up); - uart_port_unlock_irqrestore(p, flags); if (status & (UART_LSR_DR | UART_LSR_BI)) { dw8250_writel_ext(p, RZN1_UART_RDMACR, 0); @@ -346,17 +358,9 @@ static int dw8250_handle_irq(struct uart_port *p) } } - if (serial8250_handle_irq(p, iir)) - return 1; + serial8250_handle_irq_locked(p, iir); - if ((iir & UART_IIR_BUSY) == UART_IIR_BUSY) { - /* Clear the USR */ - serial_port_in(p, d->pdata->usr_reg); - - return 1; - } - - return 0; + return 1; } static void dw8250_clk_work_cb(struct work_struct *work) @@ -867,6 +871,7 @@ static struct platform_driver dw8250_platform_driver = { module_platform_driver(dw8250_platform_driver); +MODULE_IMPORT_NS("SERIAL_8250"); MODULE_AUTHOR("Jamie Iles"); MODULE_LICENSE("GPL"); MODULE_DESCRIPTION("Synopsys DesignWare 8250 serial port driver"); From 73a4ed8f9efaaaf8207614ccc1c9d5ca1888f23a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:47 +0200 Subject: [PATCH 069/352] serial: 8250_dw: Rework IIR_NO_INT handling to stop interrupt storm MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit INTC10EE UART can end up into an interrupt storm where it reports IIR_NO_INT (0x1). If the storm happens during active UART operation, it is promptly stopped by IIR value change due to Rx or Tx events. However, when there is no activity, either due to idle serial line or due to specific circumstances such as during shutdown that writes IER=0, there is nothing to stop the storm. During shutdown the storm is particularly problematic because serial8250_do_shutdown() calls synchronize_irq() that will hang in waiting for the storm to finish which never happens. This problem can also result in triggering a warning: irq 45: nobody cared (try booting with the "irqpoll" option) [...snip...] handlers: serial8250_interrupt Disabling IRQ #45 Normal means to reset interrupt status by reading LSR, MSR, USR, or RX register do not result in the UART deasserting the IRQ. Add a quirk to INTC10EE UARTs to enable Tx interrupts if UART's Tx is currently empty and inactive. Rework IIR_NO_INT to keep track of the number of consecutive IIR_NO_INT, and on fourth one perform the quirk. Enabling Tx interrupts should change IIR value from IIR_NO_INT to IIR_THRI which has been observed to stop the storm. Fixes: e92fad024929 ("serial: 8250_dw: Add ACPI ID for Granite Rapids-D UART") Cc: stable Reported-by: Bandal, Shankar Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-6-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_dw.c | 67 +++++++++++++++++++++++++++++-- 1 file changed, 63 insertions(+), 4 deletions(-) diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c index abf0814086a5..3d8032f752b8 100644 --- a/drivers/tty/serial/8250/8250_dw.c +++ b/drivers/tty/serial/8250/8250_dw.c @@ -61,6 +61,13 @@ #define DW_UART_QUIRK_IS_DMA_FC BIT(3) #define DW_UART_QUIRK_APMC0D08 BIT(4) #define DW_UART_QUIRK_CPR_VALUE BIT(5) +#define DW_UART_QUIRK_IER_KICK BIT(6) + +/* + * Number of consecutive IIR_NO_INT interrupts required to trigger interrupt + * storm prevention code. + */ +#define DW_UART_QUIRK_IER_KICK_THRES 4 struct dw8250_platform_data { u8 usr_reg; @@ -82,6 +89,8 @@ struct dw8250_data { unsigned int skip_autocfg:1; unsigned int uart_16550_compatible:1; + + u8 no_int_count; }; static inline struct dw8250_data *to_dw8250_data(struct dw8250_port_data *data) @@ -308,6 +317,29 @@ static u32 dw8250_serial_in32be(struct uart_port *p, unsigned int offset) return dw8250_modify_msr(p, offset, value); } +/* + * INTC10EE UART can IRQ storm while reporting IIR_NO_INT. Inducing IIR value + * change has been observed to break the storm. + * + * If Tx is empty (THRE asserted), we use here IER_THRI to cause IIR_NO_INT -> + * IIR_THRI transition. + */ +static void dw8250_quirk_ier_kick(struct uart_port *p) +{ + struct uart_8250_port *up = up_to_u8250p(p); + u32 lsr; + + if (up->ier & UART_IER_THRI) + return; + + lsr = serial_lsr_in(up); + if (!(lsr & UART_LSR_THRE)) + return; + + serial_port_out(p, UART_IER, up->ier | UART_IER_THRI); + serial_port_in(p, UART_LCR); /* safe, no side-effects */ + serial_port_out(p, UART_IER, up->ier); +} static int dw8250_handle_irq(struct uart_port *p) { @@ -318,18 +350,30 @@ static int dw8250_handle_irq(struct uart_port *p) unsigned int quirks = d->pdata->quirks; unsigned int status; + guard(uart_port_lock_irqsave)(p); + switch (FIELD_GET(DW_UART_IIR_IID, iir)) { case UART_IIR_NO_INT: + if (d->uart_16550_compatible || up->dma) + return 0; + + if (quirks & DW_UART_QUIRK_IER_KICK && + d->no_int_count == (DW_UART_QUIRK_IER_KICK_THRES - 1)) + dw8250_quirk_ier_kick(p); + d->no_int_count = (d->no_int_count + 1) % DW_UART_QUIRK_IER_KICK_THRES; + return 0; case UART_IIR_BUSY: /* Clear the USR */ serial_port_in(p, d->pdata->usr_reg); + d->no_int_count = 0; + return 1; } - guard(uart_port_lock_irqsave)(p); + d->no_int_count = 0; /* * There are ways to get Designware-based UARTs into a state where @@ -562,6 +606,14 @@ static void dw8250_reset_control_assert(void *data) reset_control_assert(data); } +static void dw8250_shutdown(struct uart_port *port) +{ + struct dw8250_data *d = to_dw8250_data(port->private_data); + + serial8250_do_shutdown(port); + d->no_int_count = 0; +} + static int dw8250_probe(struct platform_device *pdev) { struct uart_8250_port uart = {}, *up = &uart; @@ -689,10 +741,12 @@ static int dw8250_probe(struct platform_device *pdev) dw8250_quirks(p, data); /* If the Busy Functionality is not implemented, don't handle it */ - if (data->uart_16550_compatible) + if (data->uart_16550_compatible) { p->handle_irq = NULL; - else if (data->pdata) + } else if (data->pdata) { p->handle_irq = dw8250_handle_irq; + p->shutdown = dw8250_shutdown; + } dw8250_setup_dma_filter(p, data); @@ -824,6 +878,11 @@ static const struct dw8250_platform_data dw8250_skip_set_rate_data = { .quirks = DW_UART_QUIRK_SKIP_SET_RATE, }; +static const struct dw8250_platform_data dw8250_intc10ee = { + .usr_reg = DW_UART_USR, + .quirks = DW_UART_QUIRK_IER_KICK, +}; + static const struct of_device_id dw8250_of_match[] = { { .compatible = "snps,dw-apb-uart", .data = &dw8250_dw_apb }, { .compatible = "cavium,octeon-3860-uart", .data = &dw8250_octeon_3860_data }, @@ -853,7 +912,7 @@ static const struct acpi_device_id dw8250_acpi_match[] = { { "INT33C5", (kernel_ulong_t)&dw8250_dw_apb }, { "INT3434", (kernel_ulong_t)&dw8250_dw_apb }, { "INT3435", (kernel_ulong_t)&dw8250_dw_apb }, - { "INTC10EE", (kernel_ulong_t)&dw8250_dw_apb }, + { "INTC10EE", (kernel_ulong_t)&dw8250_intc10ee }, { }, }; MODULE_DEVICE_TABLE(acpi, dw8250_acpi_match); From e0a368ae79531ff92105a2692f10d83052055856 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:48 +0200 Subject: [PATCH 070/352] serial: 8250: Add late synchronize_irq() to shutdown to handle DW UART BUSY MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When DW UART is !uart_16550_compatible, it can indicate BUSY at any point (when under constant Rx pressure) unless a complex sequence of steps is performed. Any LCR write can run a foul with the condition that prevents writing LCR while the UART is BUSY, which triggers BUSY_DETECT interrupt that seems unmaskable using IER bits. Normal flow is that dw8250_handle_irq() handles BUSY_DETECT condition by reading USR register. This BUSY feature, however, breaks the assumptions made in serial8250_do_shutdown(), which runs synchronize_irq() after clearing IER and assumes no interrupts can occur after that point but then proceeds to update LCR, which on DW UART can trigger an interrupt. If serial8250_do_shutdown() releases the interrupt handler before the handler has run and processed the BUSY_DETECT condition by read the USR register, the IRQ is not deasserted resulting in interrupt storm that triggers "irq x: nobody cared" warning leading to disabling the IRQ. Add late synchronize_irq() into serial8250_do_shutdown() to ensure BUSY_DETECT from DW UART is handled before port's interrupt handler is released. Alternative would be to add DW UART specific shutdown function but it would mostly duplicate the generic code and the extra synchronize_irq() seems pretty harmless in serial8250_do_shutdown(). Fixes: 7d4008ebb1c9 ("tty: add a DesignWare 8250 driver") Cc: stable Reported-by: Bandal, Shankar Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-7-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250_port.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index 14d6aca44551..94a64b3dd1f7 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -2399,6 +2399,12 @@ void serial8250_do_shutdown(struct uart_port *port) * the IRQ chain. */ serial_port_in(port, UART_RX); + /* + * LCR writes on DW UART can trigger late (unmaskable) IRQs. + * Handle them before releasing the handler. + */ + synchronize_irq(port->irq); + serial8250_rpm_put(up); up->ops->release_irq(up); From a7b9ce39fbe4ae2919fe4f7ac16c293cb6632d30 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ilpo=20J=C3=A4rvinen?= Date: Tue, 3 Feb 2026 19:10:49 +0200 Subject: [PATCH 071/352] serial: 8250_dw: Ensure BUSY is deasserted MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit DW UART cannot write to LCR, DLL, and DLH while BUSY is asserted. Existance of BUSY depends on uart_16550_compatible, if UART HW is configured with it those registers can always be written. There currently is dw8250_force_idle() which attempts to achieve non-BUSY state by disabling FIFO, however, the solution is unreliable when Rx keeps getting more and more characters. Create a sequence of operations that ensures UART cannot keep BUSY asserted indefinitely. The new sequence relies on enabling loopback mode temporarily to prevent incoming Rx characters keeping UART BUSY. Ensure no Tx in ongoing while the UART is switches into the loopback mode (requires exporting serial8250_fifo_wait_for_lsr_thre() and adding DMA Tx pause/resume functions). According to tests performed by Adriana Nicolae , simply disabling FIFO or clearing FIFOs only once does not always ensure BUSY is deasserted but up to two tries may be needed. This could be related to ongoing Rx of a character (a guess, not known for sure). Therefore, retry FIFO clearing a few times (retry limit 4 is arbitrary number but using, e.g., p->fifosize seems overly large). Tests performed by others did not exhibit similar challenge but it does not seem harmful to leave the FIFO clearing loop in place for all DW UARTs with BUSY functionality. Use the new dw8250_idle_enter/exit() to do divisor writes and LCR writes. In case of plain LCR writes, opportunistically try to update LCR first and only invoke dw8250_idle_enter() if the write did not succeed (it has been observed that in practice most LCR writes do succeed without complications). This issue was first reported by qianfan Zhao who put lots of debugging effort into understanding the solution space. Fixes: c49436b657d0 ("serial: 8250_dw: Improve unwritable LCR workaround") Fixes: 7d4008ebb1c9 ("tty: add a DesignWare 8250 driver") Cc: stable Reported-by: qianfan Zhao Link: https://lore.kernel.org/linux-serial/289bb78a-7509-1c5c-2923-a04ed3b6487d@163.com/ Reported-by: Adriana Nicolae Link: https://lore.kernel.org/linux-serial/20250819182322.3451959-1-adriana@arista.com/ Reported-by: Bandal, Shankar Tested-by: Bandal, Shankar Tested-by: Murthy, Shanth Reviewed-by: Andy Shevchenko Signed-off-by: Ilpo Järvinen Link: https://patch.msgid.link/20260203171049.4353-8-ilpo.jarvinen@linux.intel.com Signed-off-by: Greg Kroah-Hartman --- drivers/tty/serial/8250/8250.h | 25 ++++ drivers/tty/serial/8250/8250_dw.c | 171 ++++++++++++++++++++-------- drivers/tty/serial/8250/8250_port.c | 28 ++--- 3 files changed, 165 insertions(+), 59 deletions(-) diff --git a/drivers/tty/serial/8250/8250.h b/drivers/tty/serial/8250/8250.h index 8caecfc85d93..77fe0588fd6b 100644 --- a/drivers/tty/serial/8250/8250.h +++ b/drivers/tty/serial/8250/8250.h @@ -175,7 +175,9 @@ static unsigned int __maybe_unused serial_icr_read(struct uart_8250_port *up, return value; } +void serial8250_clear_fifos(struct uart_8250_port *p); void serial8250_clear_and_reinit_fifos(struct uart_8250_port *p); +void serial8250_fifo_wait_for_lsr_thre(struct uart_8250_port *up, unsigned int count); void serial8250_rpm_get(struct uart_8250_port *p); void serial8250_rpm_put(struct uart_8250_port *p); @@ -400,6 +402,26 @@ static inline bool serial8250_tx_dma_running(struct uart_8250_port *p) return dma && dma->tx_running; } + +static inline void serial8250_tx_dma_pause(struct uart_8250_port *p) +{ + struct uart_8250_dma *dma = p->dma; + + if (!dma->tx_running) + return; + + dmaengine_pause(dma->txchan); +} + +static inline void serial8250_tx_dma_resume(struct uart_8250_port *p) +{ + struct uart_8250_dma *dma = p->dma; + + if (!dma->tx_running) + return; + + dmaengine_resume(dma->txchan); +} #else static inline int serial8250_tx_dma(struct uart_8250_port *p) { @@ -421,6 +443,9 @@ static inline bool serial8250_tx_dma_running(struct uart_8250_port *p) { return false; } + +static inline void serial8250_tx_dma_pause(struct uart_8250_port *p) { } +static inline void serial8250_tx_dma_resume(struct uart_8250_port *p) { } #endif static inline int ns16550a_goto_highspeed(struct uart_8250_port *up) diff --git a/drivers/tty/serial/8250/8250_dw.c b/drivers/tty/serial/8250/8250_dw.c index 3d8032f752b8..94beadb4024d 100644 --- a/drivers/tty/serial/8250/8250_dw.c +++ b/drivers/tty/serial/8250/8250_dw.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -47,6 +48,8 @@ #define DW_UART_MCR_SIRE BIT(6) +#define DW_UART_USR_BUSY BIT(0) + /* Renesas specific register fields */ #define RZN1_UART_xDMACR_DMA_EN BIT(0) #define RZN1_UART_xDMACR_1_WORD_BURST (0 << 1) @@ -89,6 +92,7 @@ struct dw8250_data { unsigned int skip_autocfg:1; unsigned int uart_16550_compatible:1; + unsigned int in_idle:1; u8 no_int_count; }; @@ -121,78 +125,151 @@ static inline u32 dw8250_modify_msr(struct uart_port *p, unsigned int offset, u3 return value; } -/* - * This function is being called as part of the uart_port::serial_out() - * routine. Hence, it must not call serial_port_out() or serial_out() - * against the modified registers here, i.e. LCR. - */ -static void dw8250_force_idle(struct uart_port *p) +static void dw8250_idle_exit(struct uart_port *p) { + struct dw8250_data *d = to_dw8250_data(p->private_data); struct uart_8250_port *up = up_to_u8250p(p); - unsigned int lsr; - /* - * The following call currently performs serial_out() - * against the FCR register. Because it differs to LCR - * there will be no infinite loop, but if it ever gets - * modified, we might need a new custom version of it - * that avoids infinite recursion. - */ - serial8250_clear_and_reinit_fifos(up); + if (d->uart_16550_compatible) + return; - /* - * With PSLVERR_RESP_EN parameter set to 1, the device generates an - * error response when an attempt to read an empty RBR with FIFO - * enabled. - */ - if (up->fcr & UART_FCR_ENABLE_FIFO) { - lsr = serial_port_in(p, UART_LSR); - if (!(lsr & UART_LSR_DR)) - return; + if (up->capabilities & UART_CAP_FIFO) + serial_port_out(p, UART_FCR, up->fcr); + serial_port_out(p, UART_MCR, up->mcr); + serial_port_out(p, UART_IER, up->ier); + + /* DMA Rx is restarted by IRQ handler as needed. */ + if (up->dma) + serial8250_tx_dma_resume(up); + + d->in_idle = 0; +} + +/* + * Ensure BUSY is not asserted. If DW UART is configured with + * !uart_16550_compatible, the writes to LCR, DLL, and DLH fail while + * BUSY is asserted. + * + * Context: port's lock must be held + */ +static int dw8250_idle_enter(struct uart_port *p) +{ + struct dw8250_data *d = to_dw8250_data(p->private_data); + unsigned int usr_reg = d->pdata ? d->pdata->usr_reg : DW_UART_USR; + struct uart_8250_port *up = up_to_u8250p(p); + int retries; + u32 lsr; + + lockdep_assert_held_once(&p->lock); + + if (d->uart_16550_compatible) + return 0; + + d->in_idle = 1; + + /* Prevent triggering interrupt from RBR filling */ + serial_port_out(p, UART_IER, 0); + + if (up->dma) { + serial8250_rx_dma_flush(up); + if (serial8250_tx_dma_running(up)) + serial8250_tx_dma_pause(up); } - serial_port_in(p, UART_RX); + /* + * Wait until Tx becomes empty + one extra frame time to ensure all bits + * have been sent on the wire. + * + * FIXME: frame_time delay is too long with very low baudrates. + */ + serial8250_fifo_wait_for_lsr_thre(up, p->fifosize); + ndelay(p->frame_time); + + serial_port_out(p, UART_MCR, up->mcr | UART_MCR_LOOP); + + retries = 4; /* Arbitrary limit, 2 was always enough in tests */ + do { + serial8250_clear_fifos(up); + if (!(serial_port_in(p, usr_reg) & DW_UART_USR_BUSY)) + break; + /* FIXME: frame_time delay is too long with very low baudrates. */ + ndelay(p->frame_time); + } while (--retries); + + lsr = serial_lsr_in(up); + if (lsr & UART_LSR_DR) { + serial_port_in(p, UART_RX); + up->lsr_saved_flags = 0; + } + + /* Now guaranteed to have BUSY deasserted? Just sanity check */ + if (serial_port_in(p, usr_reg) & DW_UART_USR_BUSY) { + dw8250_idle_exit(p); + return -EBUSY; + } + + return 0; +} + +static void dw8250_set_divisor(struct uart_port *p, unsigned int baud, + unsigned int quot, unsigned int quot_frac) +{ + struct uart_8250_port *up = up_to_u8250p(p); + int ret; + + ret = dw8250_idle_enter(p); + if (ret < 0) + return; + + serial_port_out(p, UART_LCR, up->lcr | UART_LCR_DLAB); + if (!(serial_port_in(p, UART_LCR) & UART_LCR_DLAB)) + goto idle_failed; + + serial_dl_write(up, quot); + serial_port_out(p, UART_LCR, up->lcr); + +idle_failed: + dw8250_idle_exit(p); } /* * This function is being called as part of the uart_port::serial_out() - * routine. Hence, it must not call serial_port_out() or serial_out() - * against the modified registers here, i.e. LCR. + * routine. Hence, special care must be taken when serial_port_out() or + * serial_out() against the modified registers here, i.e. LCR (d->in_idle is + * used to break recursion loop). */ static void dw8250_check_lcr(struct uart_port *p, unsigned int offset, u32 value) { struct dw8250_data *d = to_dw8250_data(p->private_data); - void __iomem *addr = p->membase + (offset << p->regshift); - int tries = 1000; + u32 lcr; + int ret; if (offset != UART_LCR || d->uart_16550_compatible) return; + lcr = serial_port_in(p, UART_LCR); + /* Make sure LCR write wasn't ignored */ - while (tries--) { - u32 lcr = serial_port_in(p, offset); + if ((value & ~UART_LCR_SPAR) == (lcr & ~UART_LCR_SPAR)) + return; - if ((value & ~UART_LCR_SPAR) == (lcr & ~UART_LCR_SPAR)) - return; + if (d->in_idle) + goto write_err; - dw8250_force_idle(p); + ret = dw8250_idle_enter(p); + if (ret < 0) + goto write_err; -#ifdef CONFIG_64BIT - if (p->type == PORT_OCTEON) - __raw_writeq(value & 0xff, addr); - else -#endif - if (p->iotype == UPIO_MEM32) - writel(value, addr); - else if (p->iotype == UPIO_MEM32BE) - iowrite32be(value, addr); - else - writeb(value, addr); - } + serial_port_out(p, UART_LCR, value); + dw8250_idle_exit(p); + return; + +write_err: /* * FIXME: this deadlocks if port->lock is already held * dev_err(p->dev, "Couldn't set LCR to %d\n", value); */ + return; /* Silences "label at the end of compound statement" */ } /* @@ -632,8 +709,10 @@ static int dw8250_probe(struct platform_device *pdev) p->type = PORT_8250; p->flags = UPF_FIXED_PORT; p->dev = dev; + p->set_ldisc = dw8250_set_ldisc; p->set_termios = dw8250_set_termios; + p->set_divisor = dw8250_set_divisor; data = devm_kzalloc(dev, sizeof(*data), GFP_KERNEL); if (!data) diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c index 94a64b3dd1f7..328711b5df1a 100644 --- a/drivers/tty/serial/8250/8250_port.c +++ b/drivers/tty/serial/8250/8250_port.c @@ -489,7 +489,7 @@ serial_port_out_sync(struct uart_port *p, int offset, int value) /* * FIFO support. */ -static void serial8250_clear_fifos(struct uart_8250_port *p) +void serial8250_clear_fifos(struct uart_8250_port *p) { if (p->capabilities & UART_CAP_FIFO) { serial_out(p, UART_FCR, UART_FCR_ENABLE_FIFO); @@ -498,6 +498,7 @@ static void serial8250_clear_fifos(struct uart_8250_port *p) serial_out(p, UART_FCR, 0); } } +EXPORT_SYMBOL_NS_GPL(serial8250_clear_fifos, "SERIAL_8250"); static enum hrtimer_restart serial8250_em485_handle_start_tx(struct hrtimer *t); static enum hrtimer_restart serial8250_em485_handle_stop_tx(struct hrtimer *t); @@ -3198,6 +3199,17 @@ void serial8250_set_defaults(struct uart_8250_port *up) } EXPORT_SYMBOL_GPL(serial8250_set_defaults); +void serial8250_fifo_wait_for_lsr_thre(struct uart_8250_port *up, unsigned int count) +{ + unsigned int i; + + for (i = 0; i < count; i++) { + if (wait_for_lsr(up, UART_LSR_THRE)) + return; + } +} +EXPORT_SYMBOL_NS_GPL(serial8250_fifo_wait_for_lsr_thre, "SERIAL_8250"); + #ifdef CONFIG_SERIAL_8250_CONSOLE static void serial8250_console_putchar(struct uart_port *port, unsigned char ch) @@ -3239,16 +3251,6 @@ static void serial8250_console_restore(struct uart_8250_port *up) serial8250_out_MCR(up, up->mcr | UART_MCR_DTR | UART_MCR_RTS); } -static void fifo_wait_for_lsr(struct uart_8250_port *up, unsigned int count) -{ - unsigned int i; - - for (i = 0; i < count; i++) { - if (wait_for_lsr(up, UART_LSR_THRE)) - return; - } -} - /* * Print a string to the serial port using the device FIFO * @@ -3267,7 +3269,7 @@ static void serial8250_console_fifo_write(struct uart_8250_port *up, while (s != end) { /* Allow timeout for each byte of a possibly full FIFO */ - fifo_wait_for_lsr(up, fifosize); + serial8250_fifo_wait_for_lsr_thre(up, fifosize); for (i = 0; i < fifosize && s != end; ++i) { if (*s == '\n' && !cr_sent) { @@ -3285,7 +3287,7 @@ static void serial8250_console_fifo_write(struct uart_8250_port *up, * Allow timeout for each byte written since the caller will only wait * for UART_LSR_BOTH_EMPTY using the timeout of a single character */ - fifo_wait_for_lsr(up, tx_count); + serial8250_fifo_wait_for_lsr_thre(up, tx_count); } /* From 5eb608319bb56464674a71b4a66ea65c6c435d64 Mon Sep 17 00:00:00 2001 From: Nicolas Pitre Date: Tue, 27 Jan 2026 17:56:01 -0500 Subject: [PATCH 072/352] vt: save/restore unicode screen buffer for alternate screen The alternate screen support added by commit 23743ba64709 ("vt: add support for smput/rmput escape codes") only saves and restores the regular screen buffer (vc_origin), but completely ignores the corresponding unicode screen buffer (vc_uni_lines) creating a messed-up display. Add vc_saved_uni_lines to save the unicode screen buffer when entering the alternate screen, and restore it when leaving. Also ensure proper cleanup in reset_terminal() and vc_deallocate(). Fixes: 23743ba64709 ("vt: add support for smput/rmput escape codes") Cc: stable Signed-off-by: Nicolas Pitre Link: https://patch.msgid.link/5o2p6qp3-91pq-0p17-or02-1oors4417ns7@onlyvoer.pbz Signed-off-by: Greg Kroah-Hartman --- drivers/tty/vt/vt.c | 8 ++++++++ include/linux/console_struct.h | 1 + 2 files changed, 9 insertions(+) diff --git a/drivers/tty/vt/vt.c b/drivers/tty/vt/vt.c index c1f152d8b03b..e2df99e3d458 100644 --- a/drivers/tty/vt/vt.c +++ b/drivers/tty/vt/vt.c @@ -1339,6 +1339,8 @@ struct vc_data *vc_deallocate(unsigned int currcons) kfree(vc->vc_saved_screen); vc->vc_saved_screen = NULL; } + vc_uniscr_free(vc->vc_saved_uni_lines); + vc->vc_saved_uni_lines = NULL; } return vc; } @@ -1884,6 +1886,8 @@ static void enter_alt_screen(struct vc_data *vc) vc->vc_saved_screen = kmemdup((u16 *)vc->vc_origin, size, GFP_KERNEL); if (vc->vc_saved_screen == NULL) return; + vc->vc_saved_uni_lines = vc->vc_uni_lines; + vc->vc_uni_lines = NULL; vc->vc_saved_rows = vc->vc_rows; vc->vc_saved_cols = vc->vc_cols; save_cur(vc); @@ -1905,6 +1909,8 @@ static void leave_alt_screen(struct vc_data *vc) dest = ((u16 *)vc->vc_origin) + r * vc->vc_cols; memcpy(dest, src, 2 * cols); } + vc_uniscr_set(vc, vc->vc_saved_uni_lines); + vc->vc_saved_uni_lines = NULL; restore_cur(vc); /* Update the entire screen */ if (con_should_update(vc)) @@ -2227,6 +2233,8 @@ static void reset_terminal(struct vc_data *vc, int do_clear) if (vc->vc_saved_screen != NULL) { kfree(vc->vc_saved_screen); vc->vc_saved_screen = NULL; + vc_uniscr_free(vc->vc_saved_uni_lines); + vc->vc_saved_uni_lines = NULL; vc->vc_saved_rows = 0; vc->vc_saved_cols = 0; } diff --git a/include/linux/console_struct.h b/include/linux/console_struct.h index 13b35637bd5a..d5ca855116df 100644 --- a/include/linux/console_struct.h +++ b/include/linux/console_struct.h @@ -160,6 +160,7 @@ struct vc_data { struct uni_pagedict **uni_pagedict_loc; /* [!] Location of uni_pagedict variable for this console */ u32 **vc_uni_lines; /* unicode screen content */ u16 *vc_saved_screen; + u32 **vc_saved_uni_lines; unsigned int vc_saved_cols; unsigned int vc_saved_rows; /* additional information is in vt_kern.h */ From daf8e3b253aa760ff9e96c7768a464bc1d6b3c90 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Wed, 11 Mar 2026 03:16:29 +0900 Subject: [PATCH 073/352] xfrm: Fix work re-schedule after cancel in xfrm_nat_keepalive_net_fini() After cancel_delayed_work_sync() is called from xfrm_nat_keepalive_net_fini(), xfrm_state_fini() flushes remaining states via __xfrm_state_delete(), which calls xfrm_nat_keepalive_state_updated() to re-schedule nat_keepalive_work. The following is a simple race scenario: cpu0 cpu1 cleanup_net() [Round 1] ops_undo_list() xfrm_net_exit() xfrm_nat_keepalive_net_fini() cancel_delayed_work_sync(nat_keepalive_work); xfrm_state_fini() xfrm_state_flush() xfrm_state_delete(x) __xfrm_state_delete(x) xfrm_nat_keepalive_state_updated(x) schedule_delayed_work(nat_keepalive_work); rcu_barrier(); net_complete_free(); net_passive_dec(net); llist_add(&net->defer_free_list, &defer_free_list); cleanup_net() [Round 2] rcu_barrier(); net_complete_free() kmem_cache_free(net_cachep, net); nat_keepalive_work() // on freed net To prevent this, cancel_delayed_work_sync() is replaced with disable_delayed_work_sync(). Fixes: f531d13bdfe3 ("xfrm: support sending NAT keepalives in ESP in UDP states") Signed-off-by: Hyunwoo Kim Reviewed-by: Sabrina Dubroca Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_nat_keepalive.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/xfrm/xfrm_nat_keepalive.c b/net/xfrm/xfrm_nat_keepalive.c index ebf95d48e86c..1856beee0149 100644 --- a/net/xfrm/xfrm_nat_keepalive.c +++ b/net/xfrm/xfrm_nat_keepalive.c @@ -261,7 +261,7 @@ int __net_init xfrm_nat_keepalive_net_init(struct net *net) int xfrm_nat_keepalive_net_fini(struct net *net) { - cancel_delayed_work_sync(&net->xfrm.nat_keepalive_work); + disable_delayed_work_sync(&net->xfrm.nat_keepalive_work); return 0; } From d499e9627d70b1269020d59b95ed3e18bee6b8cd Mon Sep 17 00:00:00 2001 From: Pepper Gray Date: Tue, 10 Mar 2026 14:44:28 +0100 Subject: [PATCH 074/352] arm64/scs: Fix handling of advance_loc4 DW_CFA_advance_loc4 is defined but no handler is implemented. Its CFA opcode defaults to EDYNSCS_INVALID_CFA_OPCODE triggering an error which wrongfully prevents modules from loading. Link: https://bugs.gentoo.org/971060 Signed-off-by: Pepper Gray Signed-off-by: Will Deacon --- arch/arm64/kernel/pi/patch-scs.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/arch/arm64/kernel/pi/patch-scs.c b/arch/arm64/kernel/pi/patch-scs.c index bbe7d30ed12b..dac568e4a54f 100644 --- a/arch/arm64/kernel/pi/patch-scs.c +++ b/arch/arm64/kernel/pi/patch-scs.c @@ -192,6 +192,14 @@ static int scs_handle_fde_frame(const struct eh_frame *frame, size -= 2; break; + case DW_CFA_advance_loc4: + loc += *opcode++ * code_alignment_factor; + loc += (*opcode++ << 8) * code_alignment_factor; + loc += (*opcode++ << 16) * code_alignment_factor; + loc += (*opcode++ << 24) * code_alignment_factor; + size -= 4; + break; + case DW_CFA_def_cfa: case DW_CFA_offset_extended: size = skip_xleb128(&opcode, size); From 4ad79c874e53ebb7fe3b8ae7ac6c858a2121f415 Mon Sep 17 00:00:00 2001 From: Ben Horgan Date: Fri, 6 Mar 2026 18:27:16 +0000 Subject: [PATCH 075/352] arm_mpam: Fix null pointer dereference when restoring bandwidth counters When an MSC supporting memory bandwidth monitoring is brought offline and then online, mpam_restore_mbwu_state() calls __ris_msmon_read() via ipi to restore the configuration of the bandwidth counters. It doesn't care about the value read, mbwu_arg.val, and doesn't set it leading to a null pointer dereference when __ris_msmon_read() adds to it. This results in a kernel oops with a call trace such as: Call trace: __ris_msmon_read+0x19c/0x64c (P) mpam_restore_mbwu_state+0xa0/0xe8 smp_call_on_cpu_callback+0x1c/0x38 process_one_work+0x154/0x4b4 worker_thread+0x188/0x310 kthread+0x11c/0x130 ret_from_fork+0x10/0x20 Provide a local variable for val to avoid __ris_msmon_read() dereferencing a null pointer when adding to val. Fixes: 41e8a14950e1 ("arm_mpam: Track bandwidth counter state for power management") Signed-off-by: Ben Horgan Reviewed-by: James Morse Signed-off-by: James Morse Signed-off-by: Will Deacon --- drivers/resctrl/mpam_devices.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/resctrl/mpam_devices.c b/drivers/resctrl/mpam_devices.c index 1eebc2602187..0666be6b0e88 100644 --- a/drivers/resctrl/mpam_devices.c +++ b/drivers/resctrl/mpam_devices.c @@ -1428,6 +1428,7 @@ static void mpam_reprogram_ris_partid(struct mpam_msc_ris *ris, u16 partid, static int mpam_restore_mbwu_state(void *_ris) { int i; + u64 val; struct mon_read mwbu_arg; struct mpam_msc_ris *ris = _ris; struct mpam_class *class = ris->vmsc->comp->class; @@ -1437,6 +1438,7 @@ static int mpam_restore_mbwu_state(void *_ris) mwbu_arg.ris = ris; mwbu_arg.ctx = &ris->mbwu_state[i].cfg; mwbu_arg.type = mpam_msmon_choose_counter(class); + mwbu_arg.val = &val; __ris_msmon_read(&mwbu_arg); } From c1376f1ff3f016a4b84e8030ed69df82e018d231 Mon Sep 17 00:00:00 2001 From: Ben Horgan Date: Fri, 6 Mar 2026 18:27:17 +0000 Subject: [PATCH 076/352] arm_mpam: Disable preemption when making accesses to fake MSC in kunit test Accesses to MSC must be made from a cpu that is affine to that MSC and the driver checks this in __mpam_write_reg() using smp_processor_id(). A fake in-memory MSC is used for testing. When using that, it doesn't matter which cpu we access it from but calling smp_processor_id() from a preemptible context gives warnings when running with CONFIG_DEBUG_PREEMPT. Add a test helper that wraps mpam_reset_msc_bitmap() with preemption disabled to ensure all (fake) MSC accesses are made with preemption disabled. Signed-off-by: Ben Horgan Reviewed-by: James Morse Signed-off-by: James Morse Signed-off-by: Will Deacon --- drivers/resctrl/test_mpam_devices.c | 18 +++++++++++++----- 1 file changed, 13 insertions(+), 5 deletions(-) diff --git a/drivers/resctrl/test_mpam_devices.c b/drivers/resctrl/test_mpam_devices.c index 3e8d564a0c64..75bd41bcc395 100644 --- a/drivers/resctrl/test_mpam_devices.c +++ b/drivers/resctrl/test_mpam_devices.c @@ -322,6 +322,14 @@ static void test_mpam_enable_merge_features(struct kunit *test) mutex_unlock(&mpam_list_lock); } +static void __test_mpam_reset_msc_bitmap(struct mpam_msc *msc, u16 reg, u16 wd) +{ + /* Avoid warnings when running with CONFIG_DEBUG_PREEMPT */ + guard(preempt)(); + + mpam_reset_msc_bitmap(msc, reg, wd); +} + static void test_mpam_reset_msc_bitmap(struct kunit *test) { char __iomem *buf = kunit_kzalloc(test, SZ_16K, GFP_KERNEL); @@ -341,31 +349,31 @@ static void test_mpam_reset_msc_bitmap(struct kunit *test) test_result = (u32 *)(buf + MPAMCFG_CPBM); - mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 0); + __test_mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 0); KUNIT_EXPECT_EQ(test, test_result[0], 0); KUNIT_EXPECT_EQ(test, test_result[1], 0); test_result[0] = 0; test_result[1] = 0; - mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 1); + __test_mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 1); KUNIT_EXPECT_EQ(test, test_result[0], 1); KUNIT_EXPECT_EQ(test, test_result[1], 0); test_result[0] = 0; test_result[1] = 0; - mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 16); + __test_mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 16); KUNIT_EXPECT_EQ(test, test_result[0], 0xffff); KUNIT_EXPECT_EQ(test, test_result[1], 0); test_result[0] = 0; test_result[1] = 0; - mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 32); + __test_mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 32); KUNIT_EXPECT_EQ(test, test_result[0], 0xffffffff); KUNIT_EXPECT_EQ(test, test_result[1], 0); test_result[0] = 0; test_result[1] = 0; - mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 33); + __test_mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 33); KUNIT_EXPECT_EQ(test, test_result[0], 0xffffffff); KUNIT_EXPECT_EQ(test, test_result[1], 1); test_result[0] = 0; From fb75437b447875ae97ea3a173e734dbd553a3881 Mon Sep 17 00:00:00 2001 From: Krzysztof Kozlowski Date: Fri, 6 Mar 2026 18:27:18 +0000 Subject: [PATCH 077/352] arm_mpam: Force __iomem casts Code allocates standard kernel memory to pass to the MPAM, which expects __iomem. The code is safe, because __iomem accessors should work fine on kernel mapped memory, however leads to sparse warnings: test_mpam_devices.c:327:42: warning: incorrect type in initializer (different address spaces) test_mpam_devices.c:327:42: expected char [noderef] __iomem *buf test_mpam_devices.c:327:42: got void * test_mpam_devices.c:342:24: warning: cast removes address space '__iomem' of expression Cast the pointer to memory via __force to silence them. Signed-off-by: Krzysztof Kozlowski Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202512160133.eAzPdJv2-lkp@intel.com/ Acked-by: Ben Horgan Signed-off-by: James Morse Signed-off-by: Will Deacon --- drivers/resctrl/test_mpam_devices.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/resctrl/test_mpam_devices.c b/drivers/resctrl/test_mpam_devices.c index 75bd41bcc395..31871f519729 100644 --- a/drivers/resctrl/test_mpam_devices.c +++ b/drivers/resctrl/test_mpam_devices.c @@ -332,7 +332,7 @@ static void __test_mpam_reset_msc_bitmap(struct mpam_msc *msc, u16 reg, u16 wd) static void test_mpam_reset_msc_bitmap(struct kunit *test) { - char __iomem *buf = kunit_kzalloc(test, SZ_16K, GFP_KERNEL); + char __iomem *buf = (__force char __iomem *)kunit_kzalloc(test, SZ_16K, GFP_KERNEL); struct mpam_msc fake_msc = {}; u32 *test_result; @@ -347,7 +347,7 @@ static void test_mpam_reset_msc_bitmap(struct kunit *test) mutex_init(&fake_msc.part_sel_lock); mutex_lock(&fake_msc.part_sel_lock); - test_result = (u32 *)(buf + MPAMCFG_CPBM); + test_result = (__force u32 *)(buf + MPAMCFG_CPBM); __test_mpam_reset_msc_bitmap(&fake_msc, MPAMCFG_CPBM, 0); KUNIT_EXPECT_EQ(test, test_result[0], 0); From 6922db250422a0dfee34de322f86b7a73d713d33 Mon Sep 17 00:00:00 2001 From: Stanislav Kinsburskii Date: Thu, 12 Mar 2026 16:02:53 +0000 Subject: [PATCH 078/352] mshv: Fix use-after-free in mshv_map_user_memory error path In the error path of mshv_map_user_memory(), calling vfree() directly on the region leaves the MMU notifier registered. When userspace later unmaps the memory, the notifier fires and accesses the freed region, causing a use-after-free and potential kernel panic. Replace vfree() with mshv_partition_put() to properly unregister the MMU notifier before freeing the region. Fixes: b9a66cd5ccbb9 ("mshv: Add support for movable memory regions") Signed-off-by: Stanislav Kinsburskii Signed-off-by: Wei Liu --- drivers/hv/mshv_root_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hv/mshv_root_main.c b/drivers/hv/mshv_root_main.c index e281311b6a9d..6f42423f7faa 100644 --- a/drivers/hv/mshv_root_main.c +++ b/drivers/hv/mshv_root_main.c @@ -1347,7 +1347,7 @@ mshv_map_user_memory(struct mshv_partition *partition, return 0; errout: - vfree(region); + mshv_region_put(region); return ret; } From b2ae73d954f71c7dd605eecfd817ed018dce7cc7 Mon Sep 17 00:00:00 2001 From: Saurabh Sengar Date: Thu, 12 Mar 2026 21:21:48 -0700 Subject: [PATCH 079/352] MAINTAINERS: Update maintainers for Hyper-V DRM driver Add myself, Dexuan, and Long as maintainers. Deepak is stepping down from these responsibilities. Signed-off-by: Saurabh Sengar Signed-off-by: Wei Liu --- MAINTAINERS | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 55af015174a5..2bb7eae97d2e 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -8002,7 +8002,9 @@ F: Documentation/devicetree/bindings/display/himax,hx8357.yaml F: drivers/gpu/drm/tiny/hx8357d.c DRM DRIVER FOR HYPERV SYNTHETIC VIDEO DEVICE -M: Deepak Rawat +M: Dexuan Cui +M: Long Li +M: Saurabh Sengar L: linux-hyperv@vger.kernel.org L: dri-devel@lists.freedesktop.org S: Maintained From 2c98a8fbd6aa647414c6248dacf254ebe91c79ad Mon Sep 17 00:00:00 2001 From: Helge Deller Date: Mon, 9 Mar 2026 15:16:37 +0100 Subject: [PATCH 080/352] parisc: Flush correct cache in cacheflush() syscall The assembly flush instructions were swapped for I- and D-cache flags: SYSCALL_DEFINE3(cacheflush, ...) { if (cache & DCACHE) { "fic ...\n" } if (cache & ICACHE && error == 0) { "fdc ...\n" } Fix it by using fdc for DCACHE, and fic for ICACHE flushing. Reported-by: Felix Lechner Fixes: c6d96328fecd ("parisc: Add cacheflush() syscall") Cc: # v6.5+ Signed-off-by: Helge Deller --- arch/parisc/kernel/cache.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/parisc/kernel/cache.c b/arch/parisc/kernel/cache.c index 4c5240d3a3c7..b189265785dc 100644 --- a/arch/parisc/kernel/cache.c +++ b/arch/parisc/kernel/cache.c @@ -953,7 +953,7 @@ SYSCALL_DEFINE3(cacheflush, unsigned long, addr, unsigned long, bytes, #else "1: cmpb,<<,n %0,%2,1b\n" #endif - " fic,m %3(%4,%0)\n" + " fdc,m %3(%4,%0)\n" "2: sync\n" ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 2b, "%1") : "+r" (start), "+r" (error) @@ -968,7 +968,7 @@ SYSCALL_DEFINE3(cacheflush, unsigned long, addr, unsigned long, bytes, #else "1: cmpb,<<,n %0,%2,1b\n" #endif - " fdc,m %3(%4,%0)\n" + " fic,m %3(%4,%0)\n" "2: sync\n" ASM_EXCEPTIONTABLE_ENTRY_EFAULT(1b, 2b, "%1") : "+r" (start), "+r" (error) From 1744a6ef48b9a48f017e3e1a0d05de0a6978396e Mon Sep 17 00:00:00 2001 From: Marc Zyngier Date: Thu, 12 Mar 2026 14:08:50 +0000 Subject: [PATCH 081/352] KVM: arm64: Discard PC update state on vcpu reset Our vcpu reset suffers from a particularly interesting flaw, as it does not correctly deal with state that will have an effect on the execution flow out of reset. Take the following completely random example, never seen in the wild and that never resulted in a couple of sleepless nights: /s - vcpu-A issues a PSCI_CPU_OFF using the SMC conduit - SMC being a trapped instruction (as opposed to HVC which is always normally executed), we annotate the vcpu as needing to skip the next instruction, which is the SMC itself - vcpu-A is now safely off - vcpu-B issues a PSCI_CPU_ON for vcpu-A, providing a starting PC - vcpu-A gets reset, get the new PC, and is sent on its merry way - right at the point of entering the guest, we notice that a PC increment is pending (remember the earlier SMC?) - vcpu-A skips its first instruction... What could possibly go wrong? Well, I'm glad you asked. For pKVM as a NV guest, that first instruction is extremely significant, as it indicates whether the CPU is booting or resuming. Having skipped that instruction, nothing makes any sense anymore, and CPU hotplugging fails. This is all caused by the decoupling of PC update from the handling of an exception that triggers such update, making it non-obvious what affects what when. Fix this train wreck by discarding all the PC-affecting state on vcpu reset. Fixes: f5e30680616ab ("KVM: arm64: Move __adjust_pc out of line") Cc: stable@vger.kernel.org Reviewed-by: Suzuki K Poulose Reviewed-by: Joey Gouly Link: https://patch.msgid.link/20260312140850.822968-1-maz@kernel.org Signed-off-by: Marc Zyngier --- arch/arm64/kvm/reset.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/arch/arm64/kvm/reset.c b/arch/arm64/kvm/reset.c index 959532422d3a..b963fd975aac 100644 --- a/arch/arm64/kvm/reset.c +++ b/arch/arm64/kvm/reset.c @@ -247,6 +247,20 @@ void kvm_reset_vcpu(struct kvm_vcpu *vcpu) kvm_vcpu_set_be(vcpu); *vcpu_pc(vcpu) = target_pc; + + /* + * We may come from a state where either a PC update was + * pending (SMC call resulting in PC being increpented to + * skip the SMC) or a pending exception. Make sure we get + * rid of all that, as this cannot be valid out of reset. + * + * Note that clearing the exception mask also clears PC + * updates, but that's an implementation detail, and we + * really want to make it explicit. + */ + vcpu_clear_flag(vcpu, PENDING_EXCEPTION); + vcpu_clear_flag(vcpu, EXCEPT_MASK); + vcpu_clear_flag(vcpu, INCREMENT_PC); vcpu_set_reg(vcpu, 0, reset_state.r0); } From a00da54d06f435dbbeacb84f9121dbbe6d6eda74 Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Sun, 8 Mar 2026 21:34:55 +0800 Subject: [PATCH 082/352] spi: amlogic: spifc-a4: Remove redundant clock cleanup The driver uses devm_clk_get_enabled() which enables the clock and registers a callback to automatically disable it when the device is unbound. Remove the redundant aml_sfc_disable_clk() call in the error paths and remove callback. Fixes: 4670db6f32e9 ("spi: amlogic: add driver for Amlogic SPI Flash Controller") Signed-off-by: Felix Gu Link: https://patch.msgid.link/20260308-spifc-a4-1-v1-1-77e286c26832@gmail.com Signed-off-by: Mark Brown --- drivers/spi/spi-amlogic-spifc-a4.c | 46 +++++------------------------- 1 file changed, 7 insertions(+), 39 deletions(-) diff --git a/drivers/spi/spi-amlogic-spifc-a4.c b/drivers/spi/spi-amlogic-spifc-a4.c index 3956869cfec1..1aabafa36e48 100644 --- a/drivers/spi/spi-amlogic-spifc-a4.c +++ b/drivers/spi/spi-amlogic-spifc-a4.c @@ -1083,14 +1083,6 @@ static int aml_sfc_clk_init(struct aml_sfc *sfc) return clk_set_rate(sfc->core_clk, SFC_BUS_DEFAULT_CLK); } -static int aml_sfc_disable_clk(struct aml_sfc *sfc) -{ - clk_disable_unprepare(sfc->core_clk); - clk_disable_unprepare(sfc->gate_clk); - - return 0; -} - static int aml_sfc_probe(struct platform_device *pdev) { struct device_node *np = pdev->dev.of_node; @@ -1141,16 +1133,12 @@ static int aml_sfc_probe(struct platform_device *pdev) /* Enable Amlogic flash controller spi mode */ ret = regmap_write(sfc->regmap_base, SFC_SPI_CFG, SPI_MODE_EN); - if (ret) { - dev_err(dev, "failed to enable SPI mode\n"); - goto err_out; - } + if (ret) + return dev_err_probe(dev, ret, "failed to enable SPI mode\n"); ret = dma_set_mask(sfc->dev, DMA_BIT_MASK(32)); - if (ret) { - dev_err(sfc->dev, "failed to set dma mask\n"); - goto err_out; - } + if (ret) + return dev_err_probe(sfc->dev, ret, "failed to set dma mask\n"); sfc->ecc_eng.dev = &pdev->dev; sfc->ecc_eng.integration = NAND_ECC_ENGINE_INTEGRATION_PIPELINED; @@ -1158,10 +1146,8 @@ static int aml_sfc_probe(struct platform_device *pdev) sfc->ecc_eng.priv = sfc; ret = nand_ecc_register_on_host_hw_engine(&sfc->ecc_eng); - if (ret) { - dev_err(&pdev->dev, "failed to register Aml host ecc engine.\n"); - goto err_out; - } + if (ret) + return dev_err_probe(&pdev->dev, ret, "failed to register Aml host ecc engine.\n"); ret = of_property_read_u32(np, "amlogic,rx-adj", &val); if (!ret) @@ -1177,24 +1163,7 @@ static int aml_sfc_probe(struct platform_device *pdev) ctrl->min_speed_hz = SFC_MIN_FREQUENCY; ctrl->num_chipselect = SFC_MAX_CS_NUM; - ret = devm_spi_register_controller(dev, ctrl); - if (ret) - goto err_out; - - return 0; - -err_out: - aml_sfc_disable_clk(sfc); - - return ret; -} - -static void aml_sfc_remove(struct platform_device *pdev) -{ - struct spi_controller *ctlr = platform_get_drvdata(pdev); - struct aml_sfc *sfc = spi_controller_get_devdata(ctlr); - - aml_sfc_disable_clk(sfc); + return devm_spi_register_controller(dev, ctrl); } static const struct of_device_id aml_sfc_of_match[] = { @@ -1212,7 +1181,6 @@ static struct platform_driver aml_sfc_driver = { .of_match_table = aml_sfc_of_match, }, .probe = aml_sfc_probe, - .remove = aml_sfc_remove, }; module_platform_driver(aml_sfc_driver); From b8db9552997924b750e727a625a30eaa4603bbb9 Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Sun, 8 Mar 2026 14:49:21 +0800 Subject: [PATCH 083/352] spi: amlogic-spisg: Fix memory leak in aml_spisg_probe() In aml_spisg_probe(), ctlr is allocated by spi_alloc_target()/spi_alloc_host(), but fails to call spi_controller_put() in several error paths. This leads to a memory leak whenever the driver fails to probe after the initial allocation. Convert to use devm_spi_alloc_host()/devm_spi_alloc_target() to fix the memory leak. Fixes: cef9991e04ae ("spi: Add Amlogic SPISG driver") Signed-off-by: Felix Gu Link: https://patch.msgid.link/20260308-spisg-v1-1-2cace5cafc24@gmail.com Signed-off-by: Mark Brown --- drivers/spi/spi-amlogic-spisg.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/spi/spi-amlogic-spisg.c b/drivers/spi/spi-amlogic-spisg.c index 1509df2b17ae..9d568e385f05 100644 --- a/drivers/spi/spi-amlogic-spisg.c +++ b/drivers/spi/spi-amlogic-spisg.c @@ -729,9 +729,9 @@ static int aml_spisg_probe(struct platform_device *pdev) }; if (of_property_read_bool(dev->of_node, "spi-slave")) - ctlr = spi_alloc_target(dev, sizeof(*spisg)); + ctlr = devm_spi_alloc_target(dev, sizeof(*spisg)); else - ctlr = spi_alloc_host(dev, sizeof(*spisg)); + ctlr = devm_spi_alloc_host(dev, sizeof(*spisg)); if (!ctlr) return -ENOMEM; @@ -750,10 +750,8 @@ static int aml_spisg_probe(struct platform_device *pdev) return dev_err_probe(dev, PTR_ERR(spisg->map), "regmap init failed\n"); irq = platform_get_irq(pdev, 0); - if (irq < 0) { - ret = irq; - goto out_controller; - } + if (irq < 0) + return irq; ret = device_reset_optional(dev); if (ret) @@ -817,8 +815,6 @@ static int aml_spisg_probe(struct platform_device *pdev) if (spisg->core) clk_disable_unprepare(spisg->core); clk_disable_unprepare(spisg->pclk); -out_controller: - spi_controller_put(ctlr); return ret; } From 7fc5e2f5603cfb305d0a8071f56b5bdb55161aeb Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Mon, 2 Mar 2026 01:08:40 +0800 Subject: [PATCH 084/352] spi: axiado: Fix double-free in ax_spi_probe() ctlr is allocated using devm_spi_alloc_host(), which automatically handles reference counting via the devm framework. Calling spi_controller_put() manually in the probe error path is redundant and results in a double-free. Fixes: e75a6b00ad79 ("spi: axiado: Add driver for Axiado SPI DB controller") Signed-off-by: Felix Gu Link: https://patch.msgid.link/20260302-axiado-v1-1-1132819f1cb7@gmail.com Signed-off-by: Mark Brown --- drivers/spi/spi-axiado.c | 31 +++++++++++-------------------- 1 file changed, 11 insertions(+), 20 deletions(-) diff --git a/drivers/spi/spi-axiado.c b/drivers/spi/spi-axiado.c index 8cea81432c5b..8ddcd27def22 100644 --- a/drivers/spi/spi-axiado.c +++ b/drivers/spi/spi-axiado.c @@ -765,30 +765,22 @@ static int ax_spi_probe(struct platform_device *pdev) platform_set_drvdata(pdev, ctlr); xspi->regs = devm_platform_ioremap_resource(pdev, 0); - if (IS_ERR(xspi->regs)) { - ret = PTR_ERR(xspi->regs); - goto remove_ctlr; - } + if (IS_ERR(xspi->regs)) + return PTR_ERR(xspi->regs); xspi->pclk = devm_clk_get(&pdev->dev, "pclk"); - if (IS_ERR(xspi->pclk)) { - dev_err(&pdev->dev, "pclk clock not found.\n"); - ret = PTR_ERR(xspi->pclk); - goto remove_ctlr; - } + if (IS_ERR(xspi->pclk)) + return dev_err_probe(&pdev->dev, PTR_ERR(xspi->pclk), + "pclk clock not found.\n"); xspi->ref_clk = devm_clk_get(&pdev->dev, "ref"); - if (IS_ERR(xspi->ref_clk)) { - dev_err(&pdev->dev, "ref clock not found.\n"); - ret = PTR_ERR(xspi->ref_clk); - goto remove_ctlr; - } + if (IS_ERR(xspi->ref_clk)) + return dev_err_probe(&pdev->dev, PTR_ERR(xspi->ref_clk), + "ref clock not found.\n"); ret = clk_prepare_enable(xspi->pclk); - if (ret) { - dev_err(&pdev->dev, "Unable to enable APB clock.\n"); - goto remove_ctlr; - } + if (ret) + return dev_err_probe(&pdev->dev, ret, "Unable to enable APB clock.\n"); ret = clk_prepare_enable(xspi->ref_clk); if (ret) { @@ -869,8 +861,7 @@ static int ax_spi_probe(struct platform_device *pdev) clk_disable_unprepare(xspi->ref_clk); clk_dis_apb: clk_disable_unprepare(xspi->pclk); -remove_ctlr: - spi_controller_put(ctlr); + return ret; } From 12b4c5d98cd7ca46d5035a57bcd995df614c14e1 Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Fri, 13 Mar 2026 00:03:38 -0300 Subject: [PATCH 085/352] smb: client: fix krb5 mount with username option Customer reported that some of their krb5 mounts were failing against a single server as the client was trying to mount the shares with wrong credentials. It turned out the client was reusing SMB session from first mount to try mounting the other shares, even though a different username= option had been specified to the other mounts. By using username mount option along with sec=krb5 to search for principals from keytab is supported by cifs.upcall(8) since cifs-utils-4.8. So fix this by matching username mount option in match_session() even with Kerberos. For example, the second mount below should fail with -ENOKEY as there is no 'foobar' principal in keytab (/etc/krb5.keytab). The client ends up reusing SMB session from first mount to perform the second one, which is wrong. ``` $ ktutil ktutil: add_entry -password -p testuser -k 1 -e aes256-cts Password for testuser@ZELDA.TEST: ktutil: write_kt /etc/krb5.keytab ktutil: quit $ klist -ke Keytab name: FILE:/etc/krb5.keytab KVNO Principal ---- ---------------------------------------------------------------- 1 testuser@ZELDA.TEST (aes256-cts-hmac-sha1-96) $ mount.cifs //w22-root2/scratch /mnt/1 -o sec=krb5,username=testuser $ mount.cifs //w22-root2/scratch /mnt/2 -o sec=krb5,username=foobar $ mount -t cifs | grep -Po 'username=\K\w+' testuser testuser ``` Reported-by: Oscar Santos Signed-off-by: Paulo Alcantara (Red Hat) Cc: David Howells Cc: linux-cifs@vger.kernel.org Cc: stable@vger.kernel.org Signed-off-by: Steve French --- fs/smb/client/connect.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 3bad2c5c523d..69b38f0ccf2b 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -1955,6 +1955,10 @@ static int match_session(struct cifs_ses *ses, case Kerberos: if (!uid_eq(ctx->cred_uid, ses->cred_uid)) return 0; + if (strncmp(ses->user_name ?: "", + ctx->username ?: "", + CIFS_MAX_USERNAME_LEN)) + return 0; break; case NTLMv2: case RawNTLMSSP: From ac57eb3b7d2ad649025b5a0fa207315f755ac4f6 Mon Sep 17 00:00:00 2001 From: Imre Deak Date: Mon, 9 Mar 2026 18:48:03 +0200 Subject: [PATCH 086/352] drm/i915/dmc: Fix an unlikely NULL pointer deference at probe intel_dmc_update_dc6_allowed_count() oopses when DMC hasn't been initialized, and dmc is thus NULL. That would be the case when the call path is intel_power_domains_init_hw() -> {skl,bxt,icl}_display_core_init() -> gen9_set_dc_state() -> intel_dmc_update_dc6_allowed_count(), as intel_power_domains_init_hw() is called *before* intel_dmc_init(). However, gen9_set_dc_state() calls intel_dmc_update_dc6_allowed_count() conditionally, depending on the current and target DC states. At probe, the target is disabled, but if DC6 is enabled, the function is called, and an oops follows. Apparently it's quite unlikely that DC6 is enabled at probe, as we haven't seen this failure mode before. It is also strange to have DC6 enabled at boot, since that would require the DMC firmware (loaded by BIOS); the BIOS loading the DMC firmware and the driver stopping / reprogramming the firmware is a poorly specified sequence and as such unlikely an intentional BIOS behaviour. It's more likely that BIOS is leaving an unintentionally enabled DC6 HW state behind (without actually loading the required DMC firmware for this). The tracking of the DC6 allowed counter only works if starting / stopping the counter depends on the _SW_ DC6 state vs. the current _HW_ DC6 state (since stopping the counter requires the DC5 counter captured when the counter was started). Thus, using the HW DC6 state is incorrect and it also leads to the above oops. Fix both issues by using the SW DC6 state for the tracking. This is v2 of the fix originally sent by Jani, updated based on the first Link: discussion below. Link: https://lore.kernel.org/all/3626411dc9e556452c432d0919821b76d9991217@intel.com Link: https://lore.kernel.org/all/20260228130946.50919-2-ltao@redhat.com Fixes: 88c1f9a4d36d ("drm/i915/dmc: Create debugfs entry for dc6 counter") Cc: Mohammed Thasleem Cc: Jani Nikula Cc: Tao Liu Cc: # v6.16+ Tested-by: Tao Liu Reviewed-by: Jani Nikula Signed-off-by: Imre Deak Link: https://patch.msgid.link/20260309164803.1918158-1-imre.deak@intel.com (cherry picked from commit 2344b93af8eb5da5d496b4e0529d35f0f559eaf0) Signed-off-by: Joonas Lahtinen --- drivers/gpu/drm/i915/display/intel_display_power_well.c | 2 +- drivers/gpu/drm/i915/display/intel_dmc.c | 3 +-- 2 files changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_display_power_well.c b/drivers/gpu/drm/i915/display/intel_display_power_well.c index db185a859133..fba9fa41f827 100644 --- a/drivers/gpu/drm/i915/display/intel_display_power_well.c +++ b/drivers/gpu/drm/i915/display/intel_display_power_well.c @@ -806,7 +806,7 @@ void gen9_set_dc_state(struct intel_display *display, u32 state) power_domains->dc_state, val & mask); enable_dc6 = state & DC_STATE_EN_UPTO_DC6; - dc6_was_enabled = val & DC_STATE_EN_UPTO_DC6; + dc6_was_enabled = power_domains->dc_state & DC_STATE_EN_UPTO_DC6; if (!dc6_was_enabled && enable_dc6) intel_dmc_update_dc6_allowed_count(display, true); diff --git a/drivers/gpu/drm/i915/display/intel_dmc.c b/drivers/gpu/drm/i915/display/intel_dmc.c index 1006b060c3f3..0b15cb764b1d 100644 --- a/drivers/gpu/drm/i915/display/intel_dmc.c +++ b/drivers/gpu/drm/i915/display/intel_dmc.c @@ -1599,8 +1599,7 @@ static bool intel_dmc_get_dc6_allowed_count(struct intel_display *display, u32 * return false; mutex_lock(&power_domains->lock); - dc6_enabled = intel_de_read(display, DC_STATE_EN) & - DC_STATE_EN_UPTO_DC6; + dc6_enabled = power_domains->dc_state & DC_STATE_EN_UPTO_DC6; if (dc6_enabled) intel_dmc_update_dc6_allowed_count(display, false); From ce5ae93d1a216680460040c7c0465a6e3b629dec Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Sun, 15 Mar 2026 07:24:15 +0900 Subject: [PATCH 087/352] ata: libata-core: disable LPM on ADATA SU680 SSD ADATA SU680 SSDs suffer from NCQ read and write commands timeouts or bus errors when link power management (LPM) is enabled. Flag these devices with the ATA_QUIRK_NOLPM quirk to prevent the use of LPM and avoid these command failures. Reported-by: Mohammad Khaled Bayan Closes: https://bugs.launchpad.net/ubuntu/+source/linux-hwe-6.17/+bug/2144060 Cc: stable@vger.kernel.org Tested-by: Mohammad-Khaled Bayan Signed-off-by: Damien Le Moal Reviewed-by: Martin K. Petersen Signed-off-by: Niklas Cassel --- drivers/ata/libata-core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 6c4e567b6582..374993031895 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -4188,6 +4188,9 @@ static const struct ata_dev_quirks_entry __ata_dev_quirks[] = { { "ST3320[68]13AS", "SD1[5-9]", ATA_QUIRK_NONCQ | ATA_QUIRK_FIRMWARE_WARN }, + /* ADATA devices with LPM issues. */ + { "ADATA SU680", NULL, ATA_QUIRK_NOLPM }, + /* Seagate disks with LPM issues */ { "ST1000DM010-2EP102", NULL, ATA_QUIRK_NOLPM }, { "ST2000DM008-2FR102", NULL, ATA_QUIRK_NOLPM }, From 29fe3a61bcdce398ee3955101c39f89c01a8a77e Mon Sep 17 00:00:00 2001 From: Minwoo Ra Date: Sat, 14 Mar 2026 00:58:44 +0900 Subject: [PATCH 088/352] xfrm: prevent policy_hthresh.work from racing with netns teardown A XFRM_MSG_NEWSPDINFO request can queue the per-net work item policy_hthresh.work onto the system workqueue. The queued callback, xfrm_hash_rebuild(), retrieves the enclosing struct net via container_of(). If the net namespace is torn down before that work runs, the associated struct net may already have been freed, and xfrm_hash_rebuild() may then dereference stale memory. xfrm_policy_fini() already flushes policy_hash_work during teardown, but it does not synchronize policy_hthresh.work. Synchronize policy_hthresh.work in xfrm_policy_fini() as well, so the queued work cannot outlive the net namespace teardown and access a freed struct net. Fixes: 880a6fab8f6b ("xfrm: configure policy hash table thresholds by netlink") Signed-off-by: Minwoo Ra Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_policy.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c index 66ac93e65264..2140ee7b102d 100644 --- a/net/xfrm/xfrm_policy.c +++ b/net/xfrm/xfrm_policy.c @@ -4282,6 +4282,8 @@ static void xfrm_policy_fini(struct net *net) 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); From eb2d16a7d599dc9d4df391b5e660df9949963786 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Sat, 14 Mar 2026 17:02:10 +0000 Subject: [PATCH 089/352] af_key: validate families in pfkey_send_migrate() syzbot was able to trigger a crash in skb_put() [1] Issue is that pfkey_send_migrate() does not check old/new families, and that set_ipsecrequest() @family argument was truncated, thus possibly overfilling the skb. Validate families early, do not wait set_ipsecrequest(). [1] skbuff: skb_over_panic: text:ffffffff8a752120 len:392 put:16 head:ffff88802a4ad040 data:ffff88802a4ad040 tail:0x188 end:0x180 dev: kernel BUG at net/core/skbuff.c:214 ! Call Trace: skb_over_panic net/core/skbuff.c:219 [inline] skb_put+0x159/0x210 net/core/skbuff.c:2655 skb_put_zero include/linux/skbuff.h:2788 [inline] set_ipsecrequest net/key/af_key.c:3532 [inline] pfkey_send_migrate+0x1270/0x2e50 net/key/af_key.c:3636 km_migrate+0x155/0x260 net/xfrm/xfrm_state.c:2848 xfrm_migrate+0x2140/0x2450 net/xfrm/xfrm_policy.c:4705 xfrm_do_migrate+0x8ff/0xaa0 net/xfrm/xfrm_user.c:3150 Fixes: 08de61beab8a ("[PFKEYV2]: Extension for dynamic update of endpoint address(es)") Reported-by: syzbot+b518dfc8e021988fbd55@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/69b5933c.050a0220.248e02.00f2.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Cc: Steffen Klassert Cc: Herbert Xu Signed-off-by: Steffen Klassert --- net/key/af_key.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/net/key/af_key.c b/net/key/af_key.c index 571200433aa9..bc91aeeb74bb 100644 --- a/net/key/af_key.c +++ b/net/key/af_key.c @@ -3518,7 +3518,7 @@ static int set_sadb_kmaddress(struct sk_buff *skb, const struct xfrm_kmaddress * static int set_ipsecrequest(struct sk_buff *skb, uint8_t proto, uint8_t mode, int level, - uint32_t reqid, uint8_t family, + uint32_t reqid, sa_family_t family, const xfrm_address_t *src, const xfrm_address_t *dst) { struct sadb_x_ipsecrequest *rq; @@ -3583,12 +3583,17 @@ static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type, /* ipsecrequests */ for (i = 0, mp = m; i < num_bundles; i++, mp++) { - /* old locator pair */ - size_pol += sizeof(struct sadb_x_ipsecrequest) + - pfkey_sockaddr_pair_size(mp->old_family); - /* new locator pair */ - size_pol += sizeof(struct sadb_x_ipsecrequest) + - pfkey_sockaddr_pair_size(mp->new_family); + int pair_size; + + pair_size = pfkey_sockaddr_pair_size(mp->old_family); + if (!pair_size) + return -EINVAL; + size_pol += sizeof(struct sadb_x_ipsecrequest) + pair_size; + + pair_size = pfkey_sockaddr_pair_size(mp->new_family); + if (!pair_size) + return -EINVAL; + size_pol += sizeof(struct sadb_x_ipsecrequest) + pair_size; } size += sizeof(struct sadb_msg) + size_pol; From 6f770b73d0311a5b099277653199bb6421c4fed2 Mon Sep 17 00:00:00 2001 From: Shigeru Yoshida Date: Sun, 15 Mar 2026 17:27:49 +0900 Subject: [PATCH 090/352] dma: swiotlb: add KMSAN annotations to swiotlb_bounce() When a device performs DMA to a bounce buffer, KMSAN is unaware of the write and does not mark the data as initialized. When swiotlb_bounce() later copies the bounce buffer back to the original buffer, memcpy propagates the uninitialized shadow to the original buffer, causing false positive uninit-value reports. Fix this by calling kmsan_unpoison_memory() on the bounce buffer before copying it back in the DMA_FROM_DEVICE path, so that memcpy naturally propagates initialized shadow to the destination. Suggested-by: Alexander Potapenko Link: https://lore.kernel.org/CAG_fn=WUGta-paG1BgsGRoAR+fmuCgh3xo=R3XdzOt_-DqSdHw@mail.gmail.com/ Fixes: 7ade4f10779c ("dma: kmsan: unpoison DMA mappings") Signed-off-by: Shigeru Yoshida Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260315082750.2375581-1-syoshida@redhat.com --- kernel/dma/swiotlb.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c index d8e6f1d889d5..9fd73700ddcf 100644 --- a/kernel/dma/swiotlb.c +++ b/kernel/dma/swiotlb.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -901,10 +902,19 @@ static void swiotlb_bounce(struct device *dev, phys_addr_t tlb_addr, size_t size local_irq_save(flags); page = pfn_to_page(pfn); - if (dir == DMA_TO_DEVICE) + if (dir == DMA_TO_DEVICE) { + /* + * Ideally, kmsan_check_highmem_page() + * could be used here to detect infoleaks, + * but callers may map uninitialized buffers + * that will be written by the device, + * causing false positives. + */ memcpy_from_page(vaddr, page, offset, sz); - else + } else { + kmsan_unpoison_memory(vaddr, sz); memcpy_to_page(page, offset, vaddr, sz); + } local_irq_restore(flags); size -= sz; @@ -913,8 +923,15 @@ static void swiotlb_bounce(struct device *dev, phys_addr_t tlb_addr, size_t size offset = 0; } } else if (dir == DMA_TO_DEVICE) { + /* + * Ideally, kmsan_check_memory() could be used here to detect + * infoleaks (uninitialized data being sent to device), but + * callers may map uninitialized buffers that will be written + * by the device, causing false positives. + */ memcpy(vaddr, phys_to_virt(orig_addr), size); } else { + kmsan_unpoison_memory(vaddr, size); memcpy(phys_to_virt(orig_addr), vaddr, size); } } From 81f86728a9804c7ff99df8f2cb7a7a081a270400 Mon Sep 17 00:00:00 2001 From: Leo Yan Date: Mon, 16 Mar 2026 11:11:57 +0000 Subject: [PATCH 091/352] tools headers: Skip arm64 cputype.h check Some definitions in the arm64 kernel's cputype.h are kernel specific and cause perf build failures when the header is synced into tools. Stop checking arm64's cputype.h. In the future, the header in tools will be updated manually when teaching tools about new CPUs. Signed-off-by: Leo Yan Acked-by: Mark Rutland Cc: Adrian Hunter Cc: Alexander Shishkin Cc: Ian Rogers Cc: James Clark Cc: Jiri Olsa Cc: Namhyung Kim Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/check-headers.sh | 1 - 1 file changed, 1 deletion(-) diff --git a/tools/perf/check-headers.sh b/tools/perf/check-headers.sh index da3aca87457f..31826621eebd 100755 --- a/tools/perf/check-headers.sh +++ b/tools/perf/check-headers.sh @@ -187,7 +187,6 @@ done check arch/x86/lib/memcpy_64.S '-I "^EXPORT_SYMBOL" -I "^#include " -I"^SYM_FUNC_START\(_LOCAL\)*(memcpy_\(erms\|orig\))" -I"^#include "' check arch/x86/lib/memset_64.S '-I "^EXPORT_SYMBOL" -I "^#include " -I"^SYM_FUNC_START\(_LOCAL\)*(memset_\(erms\|orig\))"' check arch/x86/include/asm/amd/ibs.h '-I "^#include .*/msr-index.h"' -check arch/arm64/include/asm/cputype.h '-I "^#include [<\"]\(asm/\)*sysreg.h"' check include/linux/unaligned.h '-I "^#include " -I "^#include " -I "^#pragma GCC diagnostic"' check include/uapi/asm-generic/mman.h '-I "^#include <\(uapi/\)*asm-generic/mman-common\(-tools\)*.h>"' check include/uapi/linux/mman.h '-I "^#include <\(uapi/\)*asm/mman.h>"' From ba3402f6c85bbeee5c11732c3aa4050a717e8f8f Mon Sep 17 00:00:00 2001 From: Maramaina Naresh Date: Mon, 16 Mar 2026 18:53:31 +0530 Subject: [PATCH 092/352] spi: geni-qcom: Fix CPHA and CPOL mode change detection setup_fifo_params computes mode_changed from spi->mode flags but tests it against SE_SPI_CPHA and SE_SPI_CPOL, which are register offsets, not SPI mode bits. This causes CPHA and CPOL updates to be skipped on mode switches, leaving the controller with stale clock phase and polarity settings. Fix this by using SPI_CPHA and SPI_CPOL to detect mode changes before updating the corresponding registers. Fixes: 781c3e71c94c ("spi: spi-geni-qcom: rework setup_fifo_params") Signed-off-by: Maramaina Naresh Reviewed-by: Konrad Dybcio Link: https://patch.msgid.link/20260316-spi-geni-cpha-cpol-fix-v1-1-4cb44c176b79@oss.qualcomm.com Signed-off-by: Mark Brown --- drivers/spi/spi-geni-qcom.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/spi/spi-geni-qcom.c b/drivers/spi/spi-geni-qcom.c index 43ce47f2454c..772b7148ba5f 100644 --- a/drivers/spi/spi-geni-qcom.c +++ b/drivers/spi/spi-geni-qcom.c @@ -359,9 +359,9 @@ static int setup_fifo_params(struct spi_device *spi_slv, writel((spi_slv->mode & SPI_LOOP) ? LOOPBACK_ENABLE : 0, se->base + SE_SPI_LOOPBACK); if (cs_changed) writel(chipselect, se->base + SE_SPI_DEMUX_SEL); - if (mode_changed & SE_SPI_CPHA) + if (mode_changed & SPI_CPHA) writel((spi_slv->mode & SPI_CPHA) ? CPHA : 0, se->base + SE_SPI_CPHA); - if (mode_changed & SE_SPI_CPOL) + if (mode_changed & SPI_CPOL) writel((spi_slv->mode & SPI_CPOL) ? CPOL : 0, se->base + SE_SPI_CPOL); if ((mode_changed & SPI_CS_HIGH) || (cs_changed && (spi_slv->mode & SPI_CS_HIGH))) writel((spi_slv->mode & SPI_CS_HIGH) ? BIT(chipselect) : 0, se->base + SE_SPI_DEMUX_OUTPUT_INV); From 5e3486e64094c28a526543f1e8aa0d5964b7f02d Mon Sep 17 00:00:00 2001 From: Luke Wang Date: Wed, 11 Mar 2026 17:50:06 +0800 Subject: [PATCH 093/352] mmc: sdhci: fix timing selection for 1-bit bus width When 1-bit bus width is used with HS200/HS400 capabilities set, mmc_select_hs200() returns 0 without actually switching. This causes mmc_select_timing() to skip mmc_select_hs(), leaving eMMC in legacy mode (26MHz) instead of High Speed SDR (52MHz). Per JEDEC eMMC spec section 5.3.2, 1-bit mode supports High Speed SDR. Drop incompatible HS200/HS400/UHS/DDR caps early so timing selection falls through to mmc_select_hs() correctly. Fixes: f2119df6b764 ("mmc: sd: add support for signal voltage switch procedure") Signed-off-by: Luke Wang Acked-by: Adrian Hunter Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index ac7e11f37af7..fec9329e1edb 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -4532,8 +4532,15 @@ int sdhci_setup_host(struct sdhci_host *host) * their platform code before calling sdhci_add_host(), and we * won't assume 8-bit width for hosts without that CAP. */ - if (!(host->quirks & SDHCI_QUIRK_FORCE_1_BIT_DATA)) + if (host->quirks & SDHCI_QUIRK_FORCE_1_BIT_DATA) { + host->caps1 &= ~(SDHCI_SUPPORT_SDR104 | SDHCI_SUPPORT_SDR50 | SDHCI_SUPPORT_DDR50); + if (host->quirks2 & SDHCI_QUIRK2_CAPS_BIT63_FOR_HS400) + host->caps1 &= ~SDHCI_SUPPORT_HS400; + mmc->caps2 &= ~(MMC_CAP2_HS200 | MMC_CAP2_HS400 | MMC_CAP2_HS400_ES); + mmc->caps &= ~(MMC_CAP_DDR | MMC_CAP_UHS); + } else { mmc->caps |= MMC_CAP_4_BIT_DATA; + } if (host->quirks2 & SDHCI_QUIRK2_HOST_NO_CMD23) mmc->caps &= ~MMC_CAP_CMD23; From c7feff27ea0a34540b4820abd0cdf0b5100516d4 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Thu, 19 Feb 2026 13:55:48 -0800 Subject: [PATCH 094/352] drm/vmwgfx: fix kernel-doc warnings in vmwgfx_drv.h Fix 45+ kernel-doc warnings in vmwgfx_drv.h: - spell a struct name correctly - don't have structs between kernel-doc and its struct - end description of struct members with ':' - start all kernel-doc lines with " *" - mark private struct member and enum value with "private:" - add kernel-doc for enum vmw_dma_map_mode - add missing struct member comments - add missing function parameter comments - convert "/**" to "/*" for non-kernel-doc comments - add missing "Returns:" comments for several functions - correct a function parameter name to eliminate kernel-doc warnings (examples): Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:128 struct vmw_bo; error: Cannot parse struct or union! Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:151 struct member 'used_prio' not described in 'vmw_resource' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:151 struct member 'mob_node' not described in 'vmw_resource' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:199 bad line: SM4 device. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:270 struct member 'private' not described in 'vmw_res_cache_entry' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:280 Enum value 'vmw_dma_alloc_coherent' not described in enum 'vmw_dma_map_mode' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:280 Enum value 'vmw_dma_map_bind' not described in enum 'vmw_dma_map_mode' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:295 struct member 'addrs' not described in 'vmw_sg_table' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:295 struct member 'mode' not described in 'vmw_sg_table' vmwgfx_drv.h:309: warning: Excess struct member 'num_regions' description in 'vmw_sg_table' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:402 struct member 'filp' not described in 'vmw_sw_context' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:732 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:742 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:762 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:887 No description found for return value of 'vmw_fifo_caps' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:901 No description found for return value of 'vmw_is_cursor_bypass3_enabled' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:906 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:961 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:996 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1082 cannot understand function prototype: 'const struct dma_buf_ops vmw_prime_dmabuf_ops;' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1303 struct member 'do_cpy' not described in 'vmw_diff_cpy' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1385 function parameter 'fmt' not described in 'VMW_DEBUG_KMS' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1389 This comment starts with '/**', but isn't a kernel-doc comment. Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1426 function parameter 'vmw' not described in 'vmw_fifo_mem_read' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1426 No description found for return value of 'vmw_fifo_mem_read' Warning: drivers/gpu/drm/vmwgfx/vmwgfx_drv.h:1441 function parameter 'fifo_reg' not described in 'vmw_fifo_mem_write' Signed-off-by: Randy Dunlap Signed-off-by: Zack Rusin Link: https://patch.msgid.link/20260219215548.470810-1-rdunlap@infradead.org --- drivers/gpu/drm/vmwgfx/vmwgfx_drv.h | 93 ++++++++++++++++++----------- 1 file changed, 57 insertions(+), 36 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h index f2abaf1bda6a..57465f69c687 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h @@ -96,12 +96,17 @@ struct vmwgfx_hash_item { struct vmw_res_func; +struct vmw_bo; +struct vmw_bo; +struct vmw_resource_dirty; + /** - * struct vmw-resource - base class for hardware resources + * struct vmw_resource - base class for hardware resources * * @kref: For refcounting. * @dev_priv: Pointer to the device private for this resource. Immutable. * @id: Device id. Protected by @dev_priv::resource_lock. + * @used_prio: Priority for this resource. * @guest_memory_size: Guest memory buffer size. Immutable. * @res_dirty: Resource contains data not yet in the guest memory buffer. * Protected by resource reserved. @@ -117,18 +122,16 @@ struct vmw_res_func; * pin-count greater than zero. It is not on the resource LRU lists and its * guest memory buffer is pinned. Hence it can't be evicted. * @func: Method vtable for this resource. Immutable. - * @mob_node; Node for the MOB guest memory rbtree. Protected by + * @mob_node: Node for the MOB guest memory rbtree. Protected by * @guest_memory_bo reserved. * @lru_head: List head for the LRU list. Protected by @dev_priv::resource_lock. * @binding_head: List head for the context binding list. Protected by * the @dev_priv::binding_mutex + * @dirty: resource's dirty tracker * @res_free: The resource destructor. * @hw_destroy: Callback to destroy the resource on the device, as part of * resource destruction. */ -struct vmw_bo; -struct vmw_bo; -struct vmw_resource_dirty; struct vmw_resource { struct kref kref; struct vmw_private *dev_priv; @@ -196,8 +199,8 @@ struct vmw_surface_offset; * @quality_level: Quality level. * @autogen_filter: Filter for automatically generated mipmaps. * @array_size: Number of array elements for a 1D/2D texture. For cubemap - texture number of faces * array_size. This should be 0 for pre - SM4 device. + * texture number of faces * array_size. This should be 0 for pre + * SM4 device. * @buffer_byte_stride: Buffer byte stride. * @num_sizes: Size of @sizes. For GB surface this should always be 1. * @base_size: Surface dimension. @@ -265,18 +268,24 @@ struct vmw_fifo_state { struct vmw_res_cache_entry { uint32_t handle; struct vmw_resource *res; + /* private: */ void *private; + /* public: */ unsigned short valid_handle; unsigned short valid; }; /** * enum vmw_dma_map_mode - indicate how to perform TTM page dma mappings. + * @vmw_dma_alloc_coherent: Use TTM coherent pages + * @vmw_dma_map_populate: Unmap from DMA just after unpopulate + * @vmw_dma_map_bind: Unmap from DMA just before unbind */ enum vmw_dma_map_mode { - vmw_dma_alloc_coherent, /* Use TTM coherent pages */ - vmw_dma_map_populate, /* Unmap from DMA just after unpopulate */ - vmw_dma_map_bind, /* Unmap from DMA just before unbind */ + vmw_dma_alloc_coherent, + vmw_dma_map_populate, + vmw_dma_map_bind, + /* private: */ vmw_dma_map_max }; @@ -284,8 +293,11 @@ enum vmw_dma_map_mode { * struct vmw_sg_table - Scatter/gather table for binding, with additional * device-specific information. * + * @mode: which page mapping mode to use + * @pages: Array of page pointers to the pages. + * @addrs: DMA addresses to the pages if coherent pages are used. * @sgt: Pointer to a struct sg_table with binding information - * @num_regions: Number of regions with device-address contiguous pages + * @num_pages: Number of @pages */ struct vmw_sg_table { enum vmw_dma_map_mode mode; @@ -353,6 +365,7 @@ struct vmw_ctx_validation_info; * than from user-space * @fp: If @kernel is false, points to the file of the client. Otherwise * NULL + * @filp: DRM state for this file * @cmd_bounce: Command bounce buffer used for command validation before * copying to fifo space * @cmd_bounce_size: Current command bounce buffer size @@ -729,7 +742,7 @@ extern void vmw_svga_disable(struct vmw_private *dev_priv); bool vmwgfx_supported(struct vmw_private *vmw); -/** +/* * GMR utilities - vmwgfx_gmr.c */ @@ -739,7 +752,7 @@ extern int vmw_gmr_bind(struct vmw_private *dev_priv, int gmr_id); extern void vmw_gmr_unbind(struct vmw_private *dev_priv, int gmr_id); -/** +/* * User handles */ struct vmw_user_object { @@ -759,7 +772,7 @@ void *vmw_user_object_map_size(struct vmw_user_object *uo, size_t size); void vmw_user_object_unmap(struct vmw_user_object *uo); bool vmw_user_object_is_mapped(struct vmw_user_object *uo); -/** +/* * Resource utilities - vmwgfx_resource.c */ struct vmw_user_resource_conv; @@ -819,7 +832,7 @@ static inline bool vmw_resource_mob_attached(const struct vmw_resource *res) return !RB_EMPTY_NODE(&res->mob_node); } -/** +/* * GEM related functionality - vmwgfx_gem.c */ struct vmw_bo_params; @@ -833,7 +846,7 @@ extern int vmw_gem_object_create_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); extern void vmw_debugfs_gem_init(struct vmw_private *vdev); -/** +/* * Misc Ioctl functionality - vmwgfx_ioctl.c */ @@ -846,7 +859,7 @@ extern int vmw_present_ioctl(struct drm_device *dev, void *data, extern int vmw_present_readback_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); -/** +/* * Fifo utilities - vmwgfx_fifo.c */ @@ -880,9 +893,11 @@ extern int vmw_cmd_flush(struct vmw_private *dev_priv, /** - * vmw_fifo_caps - Returns the capabilities of the FIFO command + * vmw_fifo_caps - Get the capabilities of the FIFO command * queue or 0 if fifo memory isn't present. * @dev_priv: The device private context + * + * Returns: capabilities of the FIFO command or %0 if fifo memory not present */ static inline uint32_t vmw_fifo_caps(const struct vmw_private *dev_priv) { @@ -893,9 +908,11 @@ static inline uint32_t vmw_fifo_caps(const struct vmw_private *dev_priv) /** - * vmw_is_cursor_bypass3_enabled - Returns TRUE iff Cursor Bypass 3 - * is enabled in the FIFO. + * vmw_is_cursor_bypass3_enabled - check Cursor Bypass 3 enabled setting + * in the FIFO. * @dev_priv: The device private context + * + * Returns: %true iff Cursor Bypass 3 is enabled in the FIFO */ static inline bool vmw_is_cursor_bypass3_enabled(const struct vmw_private *dev_priv) @@ -903,7 +920,7 @@ vmw_is_cursor_bypass3_enabled(const struct vmw_private *dev_priv) return (vmw_fifo_caps(dev_priv) & SVGA_FIFO_CAP_CURSOR_BYPASS_3) != 0; } -/** +/* * TTM buffer object driver - vmwgfx_ttm_buffer.c */ @@ -927,7 +944,7 @@ extern void vmw_piter_start(struct vmw_piter *viter, * * @viter: Pointer to the iterator to advance. * - * Returns false if past the list of pages, true otherwise. + * Returns: false if past the list of pages, true otherwise. */ static inline bool vmw_piter_next(struct vmw_piter *viter) { @@ -939,7 +956,7 @@ static inline bool vmw_piter_next(struct vmw_piter *viter) * * @viter: Pointer to the iterator * - * Returns the DMA address of the page pointed to by @viter. + * Returns: the DMA address of the page pointed to by @viter. */ static inline dma_addr_t vmw_piter_dma_addr(struct vmw_piter *viter) { @@ -951,14 +968,14 @@ static inline dma_addr_t vmw_piter_dma_addr(struct vmw_piter *viter) * * @viter: Pointer to the iterator * - * Returns the DMA address of the page pointed to by @viter. + * Returns: the DMA address of the page pointed to by @viter. */ static inline struct page *vmw_piter_page(struct vmw_piter *viter) { return viter->pages[viter->i]; } -/** +/* * Command submission - vmwgfx_execbuf.c */ @@ -993,7 +1010,7 @@ extern int vmw_execbuf_copy_fence_user(struct vmw_private *dev_priv, int32_t out_fence_fd); bool vmw_cmd_describe(const void *buf, u32 *size, char const **cmd); -/** +/* * IRQs and wating - vmwgfx_irq.c */ @@ -1016,7 +1033,7 @@ bool vmw_generic_waiter_add(struct vmw_private *dev_priv, u32 flag, bool vmw_generic_waiter_remove(struct vmw_private *dev_priv, u32 flag, int *waiter_count); -/** +/* * Kernel modesetting - vmwgfx_kms.c */ @@ -1048,7 +1065,7 @@ extern int vmw_resource_pin(struct vmw_resource *res, bool interruptible); extern void vmw_resource_unpin(struct vmw_resource *res); extern enum vmw_res_type vmw_res_type(const struct vmw_resource *res); -/** +/* * Overlay control - vmwgfx_overlay.c */ @@ -1063,20 +1080,20 @@ int vmw_overlay_unref(struct vmw_private *dev_priv, uint32_t stream_id); int vmw_overlay_num_overlays(struct vmw_private *dev_priv); int vmw_overlay_num_free_overlays(struct vmw_private *dev_priv); -/** +/* * GMR Id manager */ int vmw_gmrid_man_init(struct vmw_private *dev_priv, int type); void vmw_gmrid_man_fini(struct vmw_private *dev_priv, int type); -/** +/* * System memory manager */ int vmw_sys_man_init(struct vmw_private *dev_priv); void vmw_sys_man_fini(struct vmw_private *dev_priv); -/** +/* * Prime - vmwgfx_prime.c */ @@ -1292,7 +1309,7 @@ extern void vmw_cmdbuf_irqthread(struct vmw_cmdbuf_man *man); * @line: The current line of the blit. * @line_offset: Offset of the current line segment. * @cpp: Bytes per pixel (granularity information). - * @memcpy: Which memcpy function to use. + * @do_cpy: Which memcpy function to use. */ struct vmw_diff_cpy { struct drm_rect rect; @@ -1380,13 +1397,14 @@ vm_fault_t vmw_bo_vm_mkwrite(struct vm_fault *vmf); /** * VMW_DEBUG_KMS - Debug output for kernel mode-setting + * @fmt: format string for the args * * This macro is for debugging vmwgfx mode-setting code. */ #define VMW_DEBUG_KMS(fmt, ...) \ DRM_DEBUG_DRIVER(fmt, ##__VA_ARGS__) -/** +/* * Inline helper functions */ @@ -1417,11 +1435,13 @@ static inline void vmw_fifo_resource_dec(struct vmw_private *dev_priv) /** * vmw_fifo_mem_read - Perform a MMIO read from the fifo memory - * + * @vmw: The device private structure * @fifo_reg: The fifo register to read from * * This function is intended to be equivalent to ioread32() on * memremap'd memory, but without byteswapping. + * + * Returns: the value read */ static inline u32 vmw_fifo_mem_read(struct vmw_private *vmw, uint32 fifo_reg) { @@ -1431,8 +1451,9 @@ static inline u32 vmw_fifo_mem_read(struct vmw_private *vmw, uint32 fifo_reg) /** * vmw_fifo_mem_write - Perform a MMIO write to volatile memory - * - * @addr: The fifo register to write to + * @vmw: The device private structure + * @fifo_reg: The fifo register to write to + * @value: The value to write * * This function is intended to be equivalent to iowrite32 on * memremap'd memory, but without byteswapping. From c6cb77c474a32265e21c4871c7992468bf5e7638 Mon Sep 17 00:00:00 2001 From: Ian Forbes Date: Mon, 2 Mar 2026 14:03:30 -0600 Subject: [PATCH 095/352] drm/vmwgfx: Don't overwrite KMS surface dirty tracker MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit We were overwriting the surface's dirty tracker here causing a memory leak. Reported-by: Mika Penttilä Closes: https://lore.kernel.org/dri-devel/8c53f3c6-c6de-46fe-a8ca-d98dd52b3abe@redhat.com/ Fixes: 965544150d1c ("drm/vmwgfx: Refactor cursor handling") Signed-off-by: Ian Forbes Reviewed-by: Maaz Mombasawala Signed-off-by: Zack Rusin Link: https://patch.msgid.link/20260302200330.66763-1-ian.forbes@broadcom.com --- drivers/gpu/drm/vmwgfx/vmwgfx_kms.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index 55730e29d3ae..e7bddf840a79 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -771,7 +771,8 @@ static struct drm_framebuffer *vmw_kms_fb_create(struct drm_device *dev, ret = vmw_bo_dirty_add(bo); if (!ret && surface && surface->res.func->dirty_alloc) { surface->res.coherent = true; - ret = surface->res.func->dirty_alloc(&surface->res); + if (surface->res.dirty == NULL) + ret = surface->res.func->dirty_alloc(&surface->res); } ttm_bo_unreserve(&bo->tbo); } From b00be77302d7ec4ad0367bb236494fce7172b730 Mon Sep 17 00:00:00 2001 From: Janosch Frank Date: Wed, 4 Mar 2026 10:18:37 +0000 Subject: [PATCH 096/352] s390/mm: Add missing secure storage access fixups for donated memory There are special cases where secure storage access exceptions happen in a kernel context for pages that don't have the PG_arch_1 bit set. That bit is set for non-exported guest secure storage (memory) but is absent on storage donated to the Ultravisor since the kernel isn't allowed to export donated pages. Prior to this patch we would try to export the page by calling arch_make_folio_accessible() which would instantly return since the arch bit is absent signifying that the page was already exported and no further action is necessary. This leads to secure storage access exception loops which can never be resolved. With this patch we unconditionally try to export and if that fails we fixup. Fixes: 084ea4d611a3 ("s390/mm: add (non)secure page access exceptions handlers") Reported-by: Heiko Carstens Suggested-by: Heiko Carstens Reviewed-by: Claudio Imbrenda Tested-by: Christian Borntraeger Signed-off-by: Janosch Frank Signed-off-by: Christian Borntraeger --- arch/s390/mm/fault.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/s390/mm/fault.c b/arch/s390/mm/fault.c index a52aa7a99b6b..191cc53caead 100644 --- a/arch/s390/mm/fault.c +++ b/arch/s390/mm/fault.c @@ -441,10 +441,17 @@ void do_secure_storage_access(struct pt_regs *regs) folio = phys_to_folio(addr); if (unlikely(!folio_try_get(folio))) return; - rc = arch_make_folio_accessible(folio); + rc = uv_convert_from_secure(folio_to_phys(folio)); + if (!rc) + clear_bit(PG_arch_1, &folio->flags.f); folio_put(folio); + /* + * There are some valid fixup types for kernel + * accesses to donated secure memory. zeropad is one + * of them. + */ if (rc) - BUG(); + return handle_fault_error_nolock(regs, 0); } else { if (faulthandler_disabled()) return handle_fault_error_nolock(regs, 0); From dcf96f7ad556d84d460e5f5cf06061eb1a13c272 Mon Sep 17 00:00:00 2001 From: Janosch Frank Date: Tue, 3 Mar 2026 13:46:34 +0000 Subject: [PATCH 097/352] KVM: s390: Limit adapter indicator access to mapped page While we check the address for errors, we don't seem to check the bit offsets and since they are 32 and 64 bits a lot of memory can be reached indirectly via those offsets. Fixes: 84223598778b ("KVM: s390: irq routing for adapter interrupts.") Suggested-by: Claudio Imbrenda Reviewed-by: Christian Borntraeger Reviewed-by: Matthew Rosato Tested-by: Matthew Rosato Signed-off-by: Janosch Frank Signed-off-by: Christian Borntraeger --- arch/s390/kvm/interrupt.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index 18932a65ca68..1a702e8ef574 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -2724,6 +2724,9 @@ static unsigned long get_ind_bit(__u64 addr, unsigned long bit_nr, bool swap) bit = bit_nr + (addr % PAGE_SIZE) * 8; + /* kvm_set_routing_entry() should never allow this to happen */ + WARN_ON_ONCE(bit > (PAGE_SIZE * BITS_PER_BYTE - 1)); + return swap ? (bit ^ (BITS_PER_LONG - 1)) : bit; } @@ -2852,6 +2855,7 @@ int kvm_set_routing_entry(struct kvm *kvm, struct kvm_kernel_irq_routing_entry *e, const struct kvm_irq_routing_entry *ue) { + const struct kvm_irq_routing_s390_adapter *adapter; u64 uaddr_s, uaddr_i; int idx; @@ -2862,6 +2866,14 @@ int kvm_set_routing_entry(struct kvm *kvm, return -EINVAL; e->set = set_adapter_int; + adapter = &ue->u.adapter; + if (adapter->summary_addr + (adapter->summary_offset / 8) >= + (adapter->summary_addr & PAGE_MASK) + PAGE_SIZE) + return -EINVAL; + if (adapter->ind_addr + (adapter->ind_offset / 8) >= + (adapter->ind_addr & PAGE_MASK) + PAGE_SIZE) + return -EINVAL; + idx = srcu_read_lock(&kvm->srcu); uaddr_s = gpa_to_hva(kvm, ue->u.adapter.summary_addr); uaddr_i = gpa_to_hva(kvm, ue->u.adapter.ind_addr); From 0c6294d98a6dfadd53296d762f4a396c2f04c7c1 Mon Sep 17 00:00:00 2001 From: Janosch Frank Date: Tue, 3 Mar 2026 13:46:35 +0000 Subject: [PATCH 098/352] KVM: s390: selftests: Add IRQ routing address offset tests This test tries to setup routes which have address + offset combinations which cross a page. Reviewed-by: Matthew Rosato Tested-by: Matthew Rosato Signed-off-by: Janosch Frank Signed-off-by: Christian Borntraeger --- tools/testing/selftests/kvm/Makefile.kvm | 1 + .../testing/selftests/kvm/s390/irq_routing.c | 75 +++++++++++++++++++ 2 files changed, 76 insertions(+) create mode 100644 tools/testing/selftests/kvm/s390/irq_routing.c diff --git a/tools/testing/selftests/kvm/Makefile.kvm b/tools/testing/selftests/kvm/Makefile.kvm index fdec90e85467..271cbb63af36 100644 --- a/tools/testing/selftests/kvm/Makefile.kvm +++ b/tools/testing/selftests/kvm/Makefile.kvm @@ -205,6 +205,7 @@ TEST_GEN_PROGS_s390 += s390/ucontrol_test TEST_GEN_PROGS_s390 += s390/user_operexec TEST_GEN_PROGS_s390 += s390/keyop TEST_GEN_PROGS_s390 += rseq_test +TEST_GEN_PROGS_s390 += s390/irq_routing TEST_GEN_PROGS_riscv = $(TEST_GEN_PROGS_COMMON) TEST_GEN_PROGS_riscv += riscv/sbi_pmu_test diff --git a/tools/testing/selftests/kvm/s390/irq_routing.c b/tools/testing/selftests/kvm/s390/irq_routing.c new file mode 100644 index 000000000000..7819a0af19a8 --- /dev/null +++ b/tools/testing/selftests/kvm/s390/irq_routing.c @@ -0,0 +1,75 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * IRQ routing offset tests. + * + * Copyright IBM Corp. 2026 + * + * Authors: + * Janosch Frank + */ +#include +#include +#include +#include + +#include "test_util.h" +#include "kvm_util.h" +#include "kselftest.h" +#include "ucall_common.h" + +extern char guest_code[]; +asm("guest_code:\n" + "diag %r0,%r0,0\n" + "j .\n"); + +static void test(void) +{ + struct kvm_irq_routing *routing; + struct kvm_vcpu *vcpu; + struct kvm_vm *vm; + vm_paddr_t mem; + int ret; + + struct kvm_irq_routing_entry ue = { + .type = KVM_IRQ_ROUTING_S390_ADAPTER, + .gsi = 1, + }; + + vm = vm_create_with_one_vcpu(&vcpu, guest_code); + mem = vm_phy_pages_alloc(vm, 2, 4096 * 42, 0); + + routing = kvm_gsi_routing_create(); + routing->nr = 1; + routing->entries[0] = ue; + routing->entries[0].u.adapter.summary_addr = (uintptr_t)mem; + routing->entries[0].u.adapter.ind_addr = (uintptr_t)mem; + + routing->entries[0].u.adapter.summary_offset = 4096 * 8; + ret = __vm_ioctl(vm, KVM_SET_GSI_ROUTING, routing); + ksft_test_result(ret == -1 && errno == EINVAL, "summary offset outside of page\n"); + + routing->entries[0].u.adapter.summary_offset -= 4; + ret = __vm_ioctl(vm, KVM_SET_GSI_ROUTING, routing); + ksft_test_result(ret == 0, "summary offset inside of page\n"); + + routing->entries[0].u.adapter.ind_offset = 4096 * 8; + ret = __vm_ioctl(vm, KVM_SET_GSI_ROUTING, routing); + ksft_test_result(ret == -1 && errno == EINVAL, "ind offset outside of page\n"); + + routing->entries[0].u.adapter.ind_offset -= 4; + ret = __vm_ioctl(vm, KVM_SET_GSI_ROUTING, routing); + ksft_test_result(ret == 0, "ind offset inside of page\n"); + + kvm_vm_free(vm); +} + +int main(int argc, char *argv[]) +{ + TEST_REQUIRE(kvm_has_cap(KVM_CAP_IRQ_ROUTING)); + + ksft_print_header(); + ksft_set_plan(4); + test(); + + ksft_finished(); /* Print results and exit() accordingly */ +} From 1ca90f4ae554034d96764577196d8dd0c3bcd05e Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Fri, 6 Mar 2026 11:25:40 +0100 Subject: [PATCH 099/352] KVM: s390: log machine checks more aggressively KVM will reinject machine checks that happen during guest activity. From a host perspective this machine check is no longer visible and even for the guest, the guest might decide to only kill a userspace program or even ignore the machine check. As this can be a disruptive event nevertheless, we should log this not only in the VM debug event (that gets lost after guest shutdown) but also on the global KVM event as well as syslog. Consolidate the logging and log with loglevel 2 and higher. Signed-off-by: Christian Borntraeger Acked-by: Janosch Frank Acked-by: Hendrik Brueckner --- arch/s390/kvm/interrupt.c | 6 ++++++ arch/s390/kvm/kvm-s390.c | 1 - arch/s390/kvm/vsie.c | 1 - 3 files changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index 1a702e8ef574..7cb8ce833b62 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -2827,6 +2827,12 @@ void kvm_s390_reinject_machine_check(struct kvm_vcpu *vcpu, int rc; mci.val = mcck_info->mcic; + + /* log machine checks being reinjected on all debugs */ + VCPU_EVENT(vcpu, 2, "guest machine check %lx", mci.val); + KVM_EVENT(2, "guest machine check %lx", mci.val); + pr_info("guest machine check pid %d: %lx", current->pid, mci.val); + if (mci.sr) cr14 |= CR14_RECOVERY_SUBMASK; if (mci.dg) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index bc7d6fa66eaf..1668580008c6 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -4634,7 +4634,6 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15; if (exit_reason == -EINTR) { - VCPU_EVENT(vcpu, 3, "%s", "machine check"); sie_page = container_of(vcpu->arch.sie_block, struct sie_page, sie_block); mcck_info = &sie_page->mcck_info; diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index d249b10044eb..c0d36afd4023 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -1179,7 +1179,6 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page, struc kvm_vcpu_srcu_read_lock(vcpu); if (rc == -EINTR) { - VCPU_EVENT(vcpu, 3, "%s", "machine check"); kvm_s390_reinject_machine_check(vcpu, &vsie_page->mcck_info); return 0; } From ab5119735e984f6b724ef1b699c01479949ed1de Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Mon, 16 Mar 2026 13:13:17 +0100 Subject: [PATCH 100/352] KVM: s390: vsie: Avoid injecting machine check on signal The recent XFER_TO_GUEST_WORK change resulted in a situation, where the vsie code would interpret a signal during work as a machine check during SIE as both use the EINTR return code. The exit_reason of the sie64a function has nothing to do with the kvm_run exit_reason. Rename it and define a specific code for machine checks instead of abusing -EINTR. rename exit_reason into sie_return to avoid the naming conflict and change the code flow in vsie.c to have a separate variable for rc and sie_return. Fixes: 2bd1337a1295e ("KVM: s390: Use generic VIRT_XFER_TO_GUEST_WORK functions") Signed-off-by: Christian Borntraeger Reviewed-by: Heiko Carstens Reviewed-by: Claudio Imbrenda --- arch/s390/include/asm/kvm_host.h | 3 +++ arch/s390/include/asm/stacktrace.h | 2 +- arch/s390/kernel/asm-offsets.c | 2 +- arch/s390/kernel/entry.S | 4 ++-- arch/s390/kernel/nmi.c | 4 ++-- arch/s390/kvm/kvm-s390.c | 15 ++++++++------- arch/s390/kvm/vsie.c | 7 +++++-- 7 files changed, 22 insertions(+), 15 deletions(-) diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h index 64a50f0862aa..3039c88daa63 100644 --- a/arch/s390/include/asm/kvm_host.h +++ b/arch/s390/include/asm/kvm_host.h @@ -710,6 +710,9 @@ void kvm_arch_crypto_clear_masks(struct kvm *kvm); void kvm_arch_crypto_set_masks(struct kvm *kvm, unsigned long *apm, unsigned long *aqm, unsigned long *adm); +#define SIE64_RETURN_NORMAL 0 +#define SIE64_RETURN_MCCK 1 + int __sie64a(phys_addr_t sie_block_phys, struct kvm_s390_sie_block *sie_block, u64 *rsa, unsigned long gasce); diff --git a/arch/s390/include/asm/stacktrace.h b/arch/s390/include/asm/stacktrace.h index c9ae680a28af..ac3606c3babe 100644 --- a/arch/s390/include/asm/stacktrace.h +++ b/arch/s390/include/asm/stacktrace.h @@ -62,7 +62,7 @@ struct stack_frame { struct { unsigned long sie_control_block; unsigned long sie_savearea; - unsigned long sie_reason; + unsigned long sie_return; unsigned long sie_flags; unsigned long sie_control_block_phys; unsigned long sie_guest_asce; diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c index e1a5b5b54e4f..fbd26f3e9f96 100644 --- a/arch/s390/kernel/asm-offsets.c +++ b/arch/s390/kernel/asm-offsets.c @@ -63,7 +63,7 @@ int main(void) OFFSET(__SF_EMPTY, stack_frame, empty[0]); OFFSET(__SF_SIE_CONTROL, stack_frame, sie_control_block); OFFSET(__SF_SIE_SAVEAREA, stack_frame, sie_savearea); - OFFSET(__SF_SIE_REASON, stack_frame, sie_reason); + OFFSET(__SF_SIE_RETURN, stack_frame, sie_return); OFFSET(__SF_SIE_FLAGS, stack_frame, sie_flags); OFFSET(__SF_SIE_CONTROL_PHYS, stack_frame, sie_control_block_phys); OFFSET(__SF_SIE_GUEST_ASCE, stack_frame, sie_guest_asce); diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S index 4873fe9d891b..5817cb47b2d0 100644 --- a/arch/s390/kernel/entry.S +++ b/arch/s390/kernel/entry.S @@ -200,7 +200,7 @@ SYM_FUNC_START(__sie64a) stg %r3,__SF_SIE_CONTROL(%r15) # ...and virtual addresses stg %r4,__SF_SIE_SAVEAREA(%r15) # save guest register save area stg %r5,__SF_SIE_GUEST_ASCE(%r15) # save guest asce - xc __SF_SIE_REASON(8,%r15),__SF_SIE_REASON(%r15) # reason code = 0 + xc __SF_SIE_RETURN(8,%r15),__SF_SIE_RETURN(%r15) # return code = 0 mvc __SF_SIE_FLAGS(8,%r15),__TI_flags(%r14) # copy thread flags lmg %r0,%r13,0(%r4) # load guest gprs 0-13 mvi __TI_sie(%r14),1 @@ -237,7 +237,7 @@ SYM_INNER_LABEL(sie_exit, SYM_L_GLOBAL) xgr %r4,%r4 xgr %r5,%r5 lmg %r6,%r14,__SF_GPRS(%r15) # restore kernel registers - lg %r2,__SF_SIE_REASON(%r15) # return exit reason code + lg %r2,__SF_SIE_RETURN(%r15) # return sie return code BR_EX %r14 SYM_FUNC_END(__sie64a) EXPORT_SYMBOL(__sie64a) diff --git a/arch/s390/kernel/nmi.c b/arch/s390/kernel/nmi.c index a55abbf65333..94fbfad49f62 100644 --- a/arch/s390/kernel/nmi.c +++ b/arch/s390/kernel/nmi.c @@ -487,8 +487,8 @@ void notrace s390_do_machine_check(struct pt_regs *regs) mcck_dam_code = (mci.val & MCIC_SUBCLASS_MASK); if (test_cpu_flag(CIF_MCCK_GUEST) && (mcck_dam_code & MCCK_CODE_NO_GUEST) != mcck_dam_code) { - /* Set exit reason code for host's later handling */ - *((long *)(regs->gprs[15] + __SF_SIE_REASON)) = -EINTR; + /* Set sie return code for host's later handling */ + ((struct stack_frame *)regs->gprs[15])->sie_return = SIE64_RETURN_MCCK; } clear_cpu_flag(CIF_MCCK_GUEST); diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 1668580008c6..ebcb0ef8835e 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -4617,7 +4617,7 @@ static int vcpu_post_run_handle_fault(struct kvm_vcpu *vcpu) return 0; } -static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) +static int vcpu_post_run(struct kvm_vcpu *vcpu, int sie_return) { struct mcck_volatile_info *mcck_info; struct sie_page *sie_page; @@ -4633,13 +4633,14 @@ static int vcpu_post_run(struct kvm_vcpu *vcpu, int exit_reason) vcpu->run->s.regs.gprs[14] = vcpu->arch.sie_block->gg14; vcpu->run->s.regs.gprs[15] = vcpu->arch.sie_block->gg15; - if (exit_reason == -EINTR) { + if (sie_return == SIE64_RETURN_MCCK) { sie_page = container_of(vcpu->arch.sie_block, struct sie_page, sie_block); mcck_info = &sie_page->mcck_info; kvm_s390_reinject_machine_check(vcpu, mcck_info); return 0; } + WARN_ON_ONCE(sie_return != SIE64_RETURN_NORMAL); if (vcpu->arch.sie_block->icptcode > 0) { rc = kvm_handle_sie_intercept(vcpu); @@ -4678,7 +4679,7 @@ int noinstr kvm_s390_enter_exit_sie(struct kvm_s390_sie_block *scb, #define PSW_INT_MASK (PSW_MASK_EXT | PSW_MASK_IO | PSW_MASK_MCHECK) static int __vcpu_run(struct kvm_vcpu *vcpu) { - int rc, exit_reason; + int rc, sie_return; struct sie_page *sie_page = (struct sie_page *)vcpu->arch.sie_block; /* @@ -4718,9 +4719,9 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) guest_timing_enter_irqoff(); __disable_cpu_timer_accounting(vcpu); - exit_reason = kvm_s390_enter_exit_sie(vcpu->arch.sie_block, - vcpu->run->s.regs.gprs, - vcpu->arch.gmap->asce.val); + sie_return = kvm_s390_enter_exit_sie(vcpu->arch.sie_block, + vcpu->run->s.regs.gprs, + vcpu->arch.gmap->asce.val); __enable_cpu_timer_accounting(vcpu); guest_timing_exit_irqoff(); @@ -4743,7 +4744,7 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) } kvm_vcpu_srcu_read_lock(vcpu); - rc = vcpu_post_run(vcpu, exit_reason); + rc = vcpu_post_run(vcpu, sie_return); if (rc || guestdbg_exit_pending(vcpu)) { kvm_vcpu_srcu_read_unlock(vcpu); break; diff --git a/arch/s390/kvm/vsie.c b/arch/s390/kvm/vsie.c index c0d36afd4023..0330829b4046 100644 --- a/arch/s390/kvm/vsie.c +++ b/arch/s390/kvm/vsie.c @@ -1122,6 +1122,7 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page, struc { struct kvm_s390_sie_block *scb_s = &vsie_page->scb_s; struct kvm_s390_sie_block *scb_o = vsie_page->scb_o; + unsigned long sie_return = SIE64_RETURN_NORMAL; int guest_bp_isolation; int rc = 0; @@ -1163,7 +1164,7 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page, struc goto xfer_to_guest_mode_check; } guest_timing_enter_irqoff(); - rc = kvm_s390_enter_exit_sie(scb_s, vcpu->run->s.regs.gprs, sg->asce.val); + sie_return = kvm_s390_enter_exit_sie(scb_s, vcpu->run->s.regs.gprs, sg->asce.val); guest_timing_exit_irqoff(); local_irq_enable(); } @@ -1178,11 +1179,13 @@ static int do_vsie_run(struct kvm_vcpu *vcpu, struct vsie_page *vsie_page, struc kvm_vcpu_srcu_read_lock(vcpu); - if (rc == -EINTR) { + if (sie_return == SIE64_RETURN_MCCK) { kvm_s390_reinject_machine_check(vcpu, &vsie_page->mcck_info); return 0; } + WARN_ON_ONCE(sie_return != SIE64_RETURN_NORMAL); + if (rc > 0) rc = 0; /* we could still have an icpt */ else if (current->thread.gmap_int_code) From 8634e05b08ead636e926022f4a98416e13440df9 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 12 Mar 2026 16:18:13 +0100 Subject: [PATCH 101/352] spi: fix use-after-free on controller registration failure Make sure to deregister from driver core also in the unlikely event that per-cpu statistics allocation fails during controller registration to avoid use-after-free (of driver resources) and unclocked register accesses. Fixes: 6598b91b5ac3 ("spi: spi.c: Convert statistics to per-cpu u64_stats_t") Cc: stable@vger.kernel.org # 6.0 Cc: David Jander Signed-off-by: Johan Hovold Link: https://patch.msgid.link/20260312151817.32100-2-johan@kernel.org Signed-off-by: Mark Brown --- drivers/spi/spi.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 61f7bde8c7fb..9b2e307dc30a 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -3480,10 +3480,8 @@ int spi_register_controller(struct spi_controller *ctlr) dev_info(dev, "controller is unqueued, this is deprecated\n"); } else if (ctlr->transfer_one || ctlr->transfer_one_message) { status = spi_controller_initialize_queue(ctlr); - if (status) { - device_del(&ctlr->dev); - goto free_bus_id; - } + if (status) + goto del_ctrl; } /* Add statistics */ ctlr->pcpu_statistics = spi_alloc_pcpu_stats(dev); @@ -3506,6 +3504,8 @@ int spi_register_controller(struct spi_controller *ctlr) destroy_queue: spi_destroy_queue(ctlr); +del_ctrl: + device_del(&ctlr->dev); free_bus_id: mutex_lock(&board_lock); idr_remove(&spi_controller_idr, ctlr->bus_num); From dee0774bbb2abb172e9069ce5ffef579b12b3ae9 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Thu, 12 Mar 2026 16:18:14 +0100 Subject: [PATCH 102/352] spi: fix statistics allocation The controller per-cpu statistics is not allocated until after the controller has been registered with driver core, which leaves a window where accessing the sysfs attributes can trigger a NULL-pointer dereference. Fix this by moving the statistics allocation to controller allocation while tying its lifetime to that of the controller (rather than using implicit devres). Fixes: 6598b91b5ac3 ("spi: spi.c: Convert statistics to per-cpu u64_stats_t") Cc: stable@vger.kernel.org # 6.0 Cc: David Jander Signed-off-by: Johan Hovold Link: https://patch.msgid.link/20260312151817.32100-3-johan@kernel.org Signed-off-by: Mark Brown --- drivers/spi/spi.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 9b2e307dc30a..53dee314d76a 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -3049,6 +3049,8 @@ static void spi_controller_release(struct device *dev) struct spi_controller *ctlr; ctlr = container_of(dev, struct spi_controller, dev); + + free_percpu(ctlr->pcpu_statistics); kfree(ctlr); } @@ -3192,6 +3194,12 @@ struct spi_controller *__spi_alloc_controller(struct device *dev, if (!ctlr) return NULL; + ctlr->pcpu_statistics = spi_alloc_pcpu_stats(NULL); + if (!ctlr->pcpu_statistics) { + kfree(ctlr); + return NULL; + } + device_initialize(&ctlr->dev); INIT_LIST_HEAD(&ctlr->queue); spin_lock_init(&ctlr->queue_lock); @@ -3483,13 +3491,6 @@ int spi_register_controller(struct spi_controller *ctlr) if (status) goto del_ctrl; } - /* Add statistics */ - ctlr->pcpu_statistics = spi_alloc_pcpu_stats(dev); - if (!ctlr->pcpu_statistics) { - dev_err(dev, "Error allocating per-cpu statistics\n"); - status = -ENOMEM; - goto destroy_queue; - } mutex_lock(&board_lock); list_add_tail(&ctlr->list, &spi_controller_list); @@ -3502,8 +3503,6 @@ int spi_register_controller(struct spi_controller *ctlr) acpi_register_spi_devices(ctlr); return status; -destroy_queue: - spi_destroy_queue(ctlr); del_ctrl: device_del(&ctlr->dev); free_bus_id: From 2f2600decb3004938762a3f2d0eba3ea9e01045b Mon Sep 17 00:00:00 2001 From: Joe Lawrence Date: Tue, 10 Mar 2026 16:37:40 -0400 Subject: [PATCH 103/352] objtool/klp: fix data alignment in __clone_symbol() Commit 356e4b2f5b80 ("objtool: Fix data alignment in elf_add_data()") corrected the alignment of data within a section (honoring the section's sh_addralign). Apply the same alignment when klp-diff mode clones a symbol, adjusting the new symbol's offset for the output section's sh_addralign. Fixes: dd590d4d57eb ("objtool/klp: Introduce klp diff subcommand for diffing object files") Signed-off-by: Joe Lawrence Link: https://patch.msgid.link/20260310203751.1479229-2-joe.lawrence@redhat.com Signed-off-by: Josh Poimboeuf --- tools/objtool/klp-diff.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/tools/objtool/klp-diff.c b/tools/objtool/klp-diff.c index a3198a63c2f0..c2c4e4968bc2 100644 --- a/tools/objtool/klp-diff.c +++ b/tools/objtool/klp-diff.c @@ -14,6 +14,7 @@ #include #include +#include #include #include #include @@ -560,7 +561,7 @@ static struct symbol *__clone_symbol(struct elf *elf, struct symbol *patched_sym } if (!is_sec_sym(patched_sym)) - offset = sec_size(out_sec); + offset = ALIGN(sec_size(out_sec), out_sec->sh.sh_addralign); if (patched_sym->len || is_sec_sym(patched_sym)) { void *data = NULL; From 28e367a969b0c54c87ca655ec180715fe469fd14 Mon Sep 17 00:00:00 2001 From: Joe Lawrence Date: Tue, 10 Mar 2026 16:37:41 -0400 Subject: [PATCH 104/352] objtool/klp: fix mkstemp() failure with long paths The elf_create_file() function fails with EINVAL when the build directory path is long enough to truncate the "XXXXXX" suffix in the 256-byte tmp_name buffer. Simplify the code to remove the unnecessary dirname()/basename() split and concatenation. Instead, allocate the exact number of bytes needed for the path. Acked-by: Song Liu Signed-off-by: Joe Lawrence Link: https://patch.msgid.link/20260310203751.1479229-3-joe.lawrence@redhat.com Signed-off-by: Josh Poimboeuf --- tools/objtool/elf.c | 23 +++-------------------- 1 file changed, 3 insertions(+), 20 deletions(-) diff --git a/tools/objtool/elf.c b/tools/objtool/elf.c index 3da90686350d..2ffe3ebfbe37 100644 --- a/tools/objtool/elf.c +++ b/tools/objtool/elf.c @@ -16,7 +16,6 @@ #include #include #include -#include #include #include #include @@ -1189,7 +1188,7 @@ struct elf *elf_open_read(const char *name, int flags) struct elf *elf_create_file(GElf_Ehdr *ehdr, const char *name) { struct section *null, *symtab, *strtab, *shstrtab; - char *dir, *base, *tmp_name; + char *tmp_name; struct symbol *sym; struct elf *elf; @@ -1203,29 +1202,13 @@ struct elf *elf_create_file(GElf_Ehdr *ehdr, const char *name) INIT_LIST_HEAD(&elf->sections); - dir = strdup(name); - if (!dir) { - ERROR_GLIBC("strdup"); - return NULL; - } - - dir = dirname(dir); - - base = strdup(name); - if (!base) { - ERROR_GLIBC("strdup"); - return NULL; - } - - base = basename(base); - - tmp_name = malloc(256); + tmp_name = malloc(strlen(name) + 8); if (!tmp_name) { ERROR_GLIBC("malloc"); return NULL; } - snprintf(tmp_name, 256, "%s/%s.XXXXXX", dir, base); + sprintf(tmp_name, "%s.XXXXXX", name); elf->fd = mkstemp(tmp_name); if (elf->fd == -1) { From 6f93f7b06810d04acc6b106a7d5ecd6000f80545 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Tue, 10 Mar 2026 16:37:48 -0400 Subject: [PATCH 105/352] livepatch/klp-build: Fix inconsistent kernel version If .config hasn't been synced with auto.conf, any recent changes to CONFIG_LOCALVERSION* may not get reflected in the kernel version name. Use "make syncconfig" to force them to sync, and "make -s kernelrelease" to get the version instead of having to construct it manually. Fixes: 24ebfcd65a87 ("livepatch/klp-build: Introduce klp-build script for generating livepatch modules") Closes: https://lore.kernel.org/20260217160645.3434685-10-joe.lawrence@redhat.com Reported-by: Joe Lawrence Signed-off-by: Josh Poimboeuf Signed-off-by: Joe Lawrence Acked-by: Song Liu Link: https://patch.msgid.link/20260310203751.1479229-10-joe.lawrence@redhat.com Signed-off-by: Josh Poimboeuf --- scripts/livepatch/klp-build | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/scripts/livepatch/klp-build b/scripts/livepatch/klp-build index 809e198a561d..7b82c7503c2b 100755 --- a/scripts/livepatch/klp-build +++ b/scripts/livepatch/klp-build @@ -285,15 +285,14 @@ set_module_name() { # application from appending it with '+' due to a dirty git working tree. set_kernelversion() { local file="$SRC/scripts/setlocalversion" - local localversion + local kernelrelease stash_file "$file" - localversion="$(cd "$SRC" && make --no-print-directory kernelversion)" - localversion="$(cd "$SRC" && KERNELVERSION="$localversion" ./scripts/setlocalversion)" - [[ -z "$localversion" ]] && die "setlocalversion failed" + kernelrelease="$(cd "$SRC" && make syncconfig &>/dev/null && make -s kernelrelease)" + [[ -z "$kernelrelease" ]] && die "failed to get kernel version" - sed -i "2i echo $localversion; exit 0" scripts/setlocalversion + sed -i "2i echo $kernelrelease; exit 0" scripts/setlocalversion } get_patch_files() { From cf3287fb2c1ff74cb16e4348c6914acf140ebe30 Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 23 Feb 2026 20:15:59 +0530 Subject: [PATCH 106/352] PCI/pwrctrl: Ensure that remote endpoint node parent has supply requirement If OF graph is used in the PCI device node, the pwrctrl core creates a pwrctrl device even if the remote endpoint doesn't have power supply requirements. Since the device doesn't have any power supply requirements, there was no pwrctrl driver to probe, leading to PCI controller driver probe deferral as it waits for all pwrctrl drivers to probe before starting bus scan. This issue happens with Qcom ath12k devices with WSI interface attached to the Qcom IPQ platforms. Fix this issue by checking for the existence of at least one power supply property in the remote endpoint parent node. To consolidate all the checks, create a new helper pci_pwrctrl_is_required() and move all the checks there. Fixes: 9db826206f9b ("PCI/pwrctrl: Create pwrctrl device if graph port is found") Reported-by: Raj Kumar Bhagat Signed-off-by: Manivannan Sadhasivam Signed-off-by: Bjorn Helgaas Tested-by: Raj Kumar Bhagat Reviewed-by: Krishna Chaitanya Chundru Reviewed-by: Bartosz Golaszewski Link: https://patch.msgid.link/20260223-pwrctrl-fixes-7-0-v2-1-97566dfb1809@oss.qualcomm.com --- drivers/pci/pwrctrl/core.c | 47 +++++++++++++++++++++++++++----------- 1 file changed, 34 insertions(+), 13 deletions(-) diff --git a/drivers/pci/pwrctrl/core.c b/drivers/pci/pwrctrl/core.c index 6f7dea6746e0..8325858cc379 100644 --- a/drivers/pci/pwrctrl/core.c +++ b/drivers/pci/pwrctrl/core.c @@ -268,6 +268,39 @@ int pci_pwrctrl_power_on_devices(struct device *parent) } EXPORT_SYMBOL_GPL(pci_pwrctrl_power_on_devices); +/* + * Check whether the pwrctrl device really needs to be created or not. The + * pwrctrl device will only be created if the node satisfies below requirements: + * + * 1. Presence of compatible property to match against the pwrctrl driver (AND) + * 2. At least one of the power supplies defined in the devicetree node of the + * device (OR) in the remote endpoint parent node to indicate pwrctrl + * requirement. + */ +static bool pci_pwrctrl_is_required(struct device_node *np) +{ + struct device_node *endpoint; + + if (!of_property_present(np, "compatible")) + return false; + + if (of_pci_supply_present(np)) + return true; + + if (of_graph_is_present(np)) { + for_each_endpoint_of_node(np, endpoint) { + struct device_node *remote __free(device_node) = + of_graph_get_remote_port_parent(endpoint); + if (remote) { + if (of_pci_supply_present(remote)) + return true; + } + } + } + + return false; +} + static int pci_pwrctrl_create_device(struct device_node *np, struct device *parent) { @@ -287,19 +320,7 @@ static int pci_pwrctrl_create_device(struct device_node *np, return 0; } - /* - * Sanity check to make sure that the node has the compatible property - * to allow driver binding. - */ - if (!of_property_present(np, "compatible")) - return 0; - - /* - * Check whether the pwrctrl device really needs to be created or not. - * This is decided based on at least one of the power supplies defined - * in the devicetree node of the device or the graph property. - */ - if (!of_pci_supply_present(np) && !of_graph_is_present(np)) { + if (!pci_pwrctrl_is_required(np)) { dev_dbg(parent, "Skipping OF node: %s\n", np->name); return 0; } From ee226656cd64c1d781e6f91a38a5131106e9e094 Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 23 Feb 2026 20:16:00 +0530 Subject: [PATCH 107/352] PCI/pwrctrl: Create pwrctrl devices only for PCI device nodes A PCI host bridge node can have non-PCI child nodes (OPP tables, USB hub, etc.) as well as PCI device child nodes. Ensure that pwrctrl devices are only created for PCI device nodes by checking for the 'pci' prefix in the compatible property. Fixes: 4c4132489201 ("PCI/pwrctrl: Add APIs to create, destroy pwrctrl devices") Reported-by: Bjorn Andersson Closes: https://lore.kernel.org/all/20260212-rb3gen2-upd-gl3590-v1-1-18fb04bb32b0@oss.qualcomm.com Signed-off-by: Manivannan Sadhasivam Signed-off-by: Bjorn Helgaas Link: https://patch.msgid.link/20260223-pwrctrl-fixes-7-0-v2-2-97566dfb1809@oss.qualcomm.com --- drivers/pci/pwrctrl/core.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/drivers/pci/pwrctrl/core.c b/drivers/pci/pwrctrl/core.c index 8325858cc379..7754baed67f2 100644 --- a/drivers/pci/pwrctrl/core.c +++ b/drivers/pci/pwrctrl/core.c @@ -272,7 +272,8 @@ EXPORT_SYMBOL_GPL(pci_pwrctrl_power_on_devices); * Check whether the pwrctrl device really needs to be created or not. The * pwrctrl device will only be created if the node satisfies below requirements: * - * 1. Presence of compatible property to match against the pwrctrl driver (AND) + * 1. Presence of compatible property with "pci" prefix to match against the + * pwrctrl driver (AND) * 2. At least one of the power supplies defined in the devicetree node of the * device (OR) in the remote endpoint parent node to indicate pwrctrl * requirement. @@ -280,8 +281,14 @@ EXPORT_SYMBOL_GPL(pci_pwrctrl_power_on_devices); static bool pci_pwrctrl_is_required(struct device_node *np) { struct device_node *endpoint; + const char *compat; + int ret; - if (!of_property_present(np, "compatible")) + ret = of_property_read_string(np, "compatible", &compat); + if (ret < 0) + return false; + + if (!strstarts(compat, "pci")) return false; if (of_pci_supply_present(np)) From 4e5019216402ad0b4a84cff457b662d26803f103 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Mon, 9 Mar 2026 09:03:05 -0700 Subject: [PATCH 108/352] objtool: Fix Clang jump table detection With Clang, there can be a conditional forward jump between the load of the jump table address and the indirect branch. Fixes the following warning: vmlinux.o: warning: objtool: ___bpf_prog_run+0x1c5: sibling call from callable instruction with modified stack frame Reported-by: Arnd Bergmann Closes: https://lore.kernel.org/a426d669-58bb-4be1-9eaa-6f3d83109e2d@app.fastmail.com Link: https://patch.msgid.link/7d8600caed08901b6679767488acd639f6df9688.1773071992.git.jpoimboe@kernel.org Signed-off-by: Josh Poimboeuf --- tools/objtool/check.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/tools/objtool/check.c b/tools/objtool/check.c index 91b3ff4803cf..b6765e876507 100644 --- a/tools/objtool/check.c +++ b/tools/objtool/check.c @@ -2184,12 +2184,11 @@ static void mark_func_jump_tables(struct objtool_file *file, last = insn; /* - * Store back-pointers for unconditional forward jumps such + * Store back-pointers for forward jumps such * that find_jump_table() can back-track using those and * avoid some potentially confusing code. */ - if (insn->type == INSN_JUMP_UNCONDITIONAL && insn->jump_dest && - insn->offset > last->offset && + if (insn->jump_dest && insn->jump_dest->offset > insn->offset && !insn->jump_dest->first_jump_src) { From 75cea0776de502f2a1be5ca02d37c586dc81887e Mon Sep 17 00:00:00 2001 From: Smita Koralahalli Date: Mon, 16 Mar 2026 20:19:49 +0000 Subject: [PATCH 109/352] cxl/hdm: Avoid incorrect DVSEC fallback when HDM decoders are enabled Check the global CXL_HDM_DECODER_ENABLE bit instead of looping over per-decoder COMMITTED bits to determine whether to fall back to DVSEC range emulation. When the HDM decoder capability is globally enabled, ignore DVSEC range registers regardless of individual decoder commit state. should_emulate_decoders() currently loops over per-decoder COMMITTED bits, which leads to an incorrect DVSEC fallback when those bits are zero. One way to trigger this is to destroy a region and bounce the memdev: cxl disable-region region0 cxl destroy-region region0 cxl disable-memdev mem0 cxl enable-memdev mem0 Region teardown zeroes the HDM decoder registers including the committed bits. The subsequent memdev re-probe finds uncommitted decoders and falls back to DVSEC emulation, even though HDM remains globally enabled. Observed failures: should_emulate_decoders: cxl_port endpoint6: decoder6.0: committed: 0 base: 0x0_00000000 size: 0x0_00000000 devm_cxl_setup_hdm: cxl_port endpoint6: Fallback map 1 range register .. devm_cxl_add_region: cxl_acpi ACPI0017:00: decoder0.0: created region0 __construct_region: cxl_pci 0000:e1:00.0: mem1:decoder6.0: __construct_region region0 res: [mem 0x850000000-0x284fffffff flags 0x200] iw: 1 ig: 4096 cxl region0: pci0000:e0:port1 cxl_port_setup_targets expected iw: 1 ig: 4096 .. cxl region0: pci0000:e0:port1 cxl_port_setup_targets got iw: 1 ig: 256 state: disabled .. cxl_port endpoint6: failed to attach decoder6.0 to region0: -6 .. devm_cxl_add_region: cxl_acpi ACPI0017:00: decoder0.0: created region4 alloc_hpa: cxl region4: HPA allocation error (-34) .. Fixes: 52cc48ad2a76 ("cxl/hdm: Limit emulation to the number of range registers") Signed-off-by: Smita Koralahalli Reviewed-by: Dan Williams Link: https://patch.msgid.link/20260316201950.224567-1-Smita.KoralahalliChannabasappa@amd.com Signed-off-by: Dave Jiang --- drivers/cxl/core/hdm.c | 25 +++++++++---------------- 1 file changed, 9 insertions(+), 16 deletions(-) diff --git a/drivers/cxl/core/hdm.c b/drivers/cxl/core/hdm.c index c222e98ae736..cb5d5a047a9d 100644 --- a/drivers/cxl/core/hdm.c +++ b/drivers/cxl/core/hdm.c @@ -94,7 +94,6 @@ static bool should_emulate_decoders(struct cxl_endpoint_dvsec_info *info) struct cxl_hdm *cxlhdm; void __iomem *hdm; u32 ctrl; - int i; if (!info) return false; @@ -113,22 +112,16 @@ static bool should_emulate_decoders(struct cxl_endpoint_dvsec_info *info) return false; /* - * If any decoders are committed already, there should not be any - * emulated DVSEC decoders. + * If HDM decoders are globally enabled, do not fall back to DVSEC + * range emulation. Zeroed decoder registers after region teardown + * do not imply absence of HDM capability. + * + * Falling back to DVSEC here would treat the decoder as AUTO and + * may incorrectly latch default interleave settings. */ - for (i = 0; i < cxlhdm->decoder_count; i++) { - ctrl = readl(hdm + CXL_HDM_DECODER0_CTRL_OFFSET(i)); - dev_dbg(&info->port->dev, - "decoder%d.%d: committed: %ld base: %#x_%.8x size: %#x_%.8x\n", - info->port->id, i, - FIELD_GET(CXL_HDM_DECODER0_CTRL_COMMITTED, ctrl), - readl(hdm + CXL_HDM_DECODER0_BASE_HIGH_OFFSET(i)), - readl(hdm + CXL_HDM_DECODER0_BASE_LOW_OFFSET(i)), - readl(hdm + CXL_HDM_DECODER0_SIZE_HIGH_OFFSET(i)), - readl(hdm + CXL_HDM_DECODER0_SIZE_LOW_OFFSET(i))); - if (FIELD_GET(CXL_HDM_DECODER0_CTRL_COMMITTED, ctrl)) - return false; - } + ctrl = readl(hdm + CXL_HDM_DECODER_CTRL_OFFSET); + if (ctrl & CXL_HDM_DECODER_ENABLE) + return false; return true; } From 50bfd2a22b75a904d5900d64530ae1b69a69907c Mon Sep 17 00:00:00 2001 From: Joseph Salisbury Date: Mon, 16 Mar 2026 21:13:49 -0400 Subject: [PATCH 110/352] cifs: smb1: fix comment typo The file contains a spelling error in a source comment (resposne). Typos in comments reduce readability and make text searches less reliable for developers and maintainers. Replace 'resposne' with 'response' in the affected comment. This is a comment-only cleanup and does not change behavior. [v2: Removed Fixes: and Cc: to stable tags.] Signed-off-by: Joseph Salisbury Signed-off-by: Steve French --- fs/smb/client/smb1transport.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/smb/client/smb1transport.c b/fs/smb/client/smb1transport.c index 38d6d5538b96..53abb29fe71b 100644 --- a/fs/smb/client/smb1transport.c +++ b/fs/smb/client/smb1transport.c @@ -460,7 +460,7 @@ check_smb_hdr(struct smb_hdr *smb) return 0; /* - * Windows NT server returns error resposne (e.g. STATUS_DELETE_PENDING + * Windows NT server returns error response (e.g. STATUS_DELETE_PENDING * or STATUS_OBJECT_NAME_NOT_FOUND or ERRDOS/ERRbadfile or any other) * for some TRANS2 requests without the RESPONSE flag set in header. */ From eade54040384f54b7fb330e4b0975c5734850b3c Mon Sep 17 00:00:00 2001 From: Sheng Yong Date: Fri, 27 Feb 2026 10:30:08 +0800 Subject: [PATCH 111/352] erofs: set fileio bio failed in short read case For file-backed mount, IO requests are handled by vfs_iocb_iter_read(). However, it can be interrupted by SIGKILL, returning the number of bytes actually copied. Unused folios in bio are unexpectedly marked as uptodate. vfs_read filemap_read filemap_get_pages filemap_readahead erofs_fileio_readahead erofs_fileio_rq_submit vfs_iocb_iter_read filemap_read filemap_get_pages <= detect signal erofs_fileio_ki_complete <= set all folios uptodate This patch addresses this by setting short read bio with an error directly. Fixes: bc804a8d7e86 ("erofs: handle end of filesystem properly for file-backed mounts") Reported-by: chenguanyou Signed-off-by: Yunlei He Signed-off-by: Sheng Yong Reviewed-by: Gao Xiang Reviewed-by: Chao Yu Signed-off-by: Gao Xiang --- fs/erofs/fileio.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/erofs/fileio.c b/fs/erofs/fileio.c index abe873f01297..98cdaa1cd1a7 100644 --- a/fs/erofs/fileio.c +++ b/fs/erofs/fileio.c @@ -25,10 +25,8 @@ static void erofs_fileio_ki_complete(struct kiocb *iocb, long ret) container_of(iocb, struct erofs_fileio_rq, iocb); struct folio_iter fi; - if (ret >= 0 && ret != rq->bio.bi_iter.bi_size) { - bio_advance(&rq->bio, ret); - zero_fill_bio(&rq->bio); - } + if (ret >= 0 && ret != rq->bio.bi_iter.bi_size) + ret = -EIO; if (!rq->bio.bi_end_io) { bio_for_each_folio_all(fi, &rq->bio) { DBG_BUGON(folio_test_uptodate(fi.folio)); From b0a4dba7b623aa7cbc9efcc56b4af2ec8b274f3e Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jouni=20H=C3=B6gander?= Date: Thu, 12 Mar 2026 10:37:09 +0200 Subject: [PATCH 112/352] drm/i915/psr: Disable PSR on update_m_n and update_lrr MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit PSR/PR parameters might change based on update_m_n or update_lrr. Disable on update_m_n and update_lrr to ensure proper parameters are taken into use on next PSR enable in intel_psr_post_plane_update. Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/15771 Fixes: 2bc98c6f97af ("drm/i915/alpm: Compute ALPM parameters into crtc_state->alpm_state") Cc: # v6.19+ Signed-off-by: Jouni Högander Reviewed-by: Suraj Kandpal Link: https://patch.msgid.link/20260312083710.1593781-2-jouni.hogander@intel.com (cherry picked from commit 65852b56bfa929f99e28c96fd98b02058959da7f) Signed-off-by: Joonas Lahtinen --- drivers/gpu/drm/i915/display/intel_psr.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index b7302a32ded4..426c23319269 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -3109,6 +3109,8 @@ void intel_psr_pre_plane_update(struct intel_atomic_state *state, * - Display WA #1136: skl, bxt */ if (intel_crtc_needs_modeset(new_crtc_state) || + new_crtc_state->update_m_n || + new_crtc_state->update_lrr || !new_crtc_state->has_psr || !new_crtc_state->active_planes || new_crtc_state->has_sel_update != psr->sel_update_enabled || From 7caac659a837af9fd4cad85be851982b88859484 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jouni=20H=C3=B6gander?= Date: Thu, 12 Mar 2026 10:37:10 +0200 Subject: [PATCH 113/352] drm/i915/psr: Compute PSR entry_setup_frames into intel_crtc_state MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit PSR entry_setup_frames is currently computed directly into struct intel_dp:intel_psr:entry_setup_frames. This causes a problem if mode change gets rejected after PSR compute config: Psr_entry_setup_frames computed for this rejected state is in intel_dp:intel_psr:entry_setup_frame. Fix this by computing it into intel_crtc_state and copy the value into intel_dp:intel_psr:entry_setup_frames on PSR enable. Fixes: 2b981d57e480 ("drm/i915/display: Support PSR entry VSC packet to be transmitted one frame earlier") Cc: Mika Kahola Cc: # v6.8+ Signed-off-by: Jouni Högander Reviewed-by: Suraj Kandpal Link: https://patch.msgid.link/20260312083710.1593781-3-jouni.hogander@intel.com (cherry picked from commit 8c229b4aa00262c13787982e998c61c0783285e0) Signed-off-by: Joonas Lahtinen --- drivers/gpu/drm/i915/display/intel_display_types.h | 1 + drivers/gpu/drm/i915/display/intel_psr.c | 5 +++-- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_display_types.h b/drivers/gpu/drm/i915/display/intel_display_types.h index 6b92f333e18b..ced0e5a5989b 100644 --- a/drivers/gpu/drm/i915/display/intel_display_types.h +++ b/drivers/gpu/drm/i915/display/intel_display_types.h @@ -1186,6 +1186,7 @@ struct intel_crtc_state { u32 dc3co_exitline; u16 su_y_granularity; u8 active_non_psr_pipes; + u8 entry_setup_frames; const char *no_psr_reason; /* diff --git a/drivers/gpu/drm/i915/display/intel_psr.c b/drivers/gpu/drm/i915/display/intel_psr.c index 426c23319269..3791944389db 100644 --- a/drivers/gpu/drm/i915/display/intel_psr.c +++ b/drivers/gpu/drm/i915/display/intel_psr.c @@ -1717,7 +1717,7 @@ static bool _psr_compute_config(struct intel_dp *intel_dp, entry_setup_frames = intel_psr_entry_setup_frames(intel_dp, conn_state, adjusted_mode); if (entry_setup_frames >= 0) { - intel_dp->psr.entry_setup_frames = entry_setup_frames; + crtc_state->entry_setup_frames = entry_setup_frames; } else { crtc_state->no_psr_reason = "PSR setup timing not met"; drm_dbg_kms(display->drm, @@ -1815,7 +1815,7 @@ static bool intel_psr_needs_wa_18037818876(struct intel_dp *intel_dp, { struct intel_display *display = to_intel_display(intel_dp); - return (DISPLAY_VER(display) == 20 && intel_dp->psr.entry_setup_frames > 0 && + return (DISPLAY_VER(display) == 20 && crtc_state->entry_setup_frames > 0 && !crtc_state->has_sel_update); } @@ -2189,6 +2189,7 @@ static void intel_psr_enable_locked(struct intel_dp *intel_dp, intel_dp->psr.pkg_c_latency_used = crtc_state->pkg_c_latency_used; intel_dp->psr.io_wake_lines = crtc_state->alpm_state.io_wake_lines; intel_dp->psr.fast_wake_lines = crtc_state->alpm_state.fast_wake_lines; + intel_dp->psr.entry_setup_frames = crtc_state->entry_setup_frames; if (!psr_interrupt_error_check(intel_dp)) return; From 76f0930d6e809234904cf9f0f5f42ee6c1dc694e Mon Sep 17 00:00:00 2001 From: Felix Gu Date: Sun, 15 Mar 2026 15:17:54 +0800 Subject: [PATCH 114/352] irqchip/riscv-rpmi-sysmsi: Fix mailbox channel leak in rpmi_sysmsi_probe() When riscv_acpi_get_gsi_info() fails, the mailbox channel previously requested via mbox_request_channel() is not freed. Add the missing mbox_free_channel() call to prevent the resource leak. Fixes: 4752b0cfbc37 ("irqchip/riscv-rpmi-sysmsi: Add ACPI support") Signed-off-by: Felix Gu Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Reviewed-by: Rahul Pathak Link: https://patch.msgid.link/20260315-sysmsi-v1-1-5f090c86c2ca@gmail.com --- drivers/irqchip/irq-riscv-rpmi-sysmsi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/irqchip/irq-riscv-rpmi-sysmsi.c b/drivers/irqchip/irq-riscv-rpmi-sysmsi.c index 5c74c561ce31..612f3972f7af 100644 --- a/drivers/irqchip/irq-riscv-rpmi-sysmsi.c +++ b/drivers/irqchip/irq-riscv-rpmi-sysmsi.c @@ -250,6 +250,7 @@ static int rpmi_sysmsi_probe(struct platform_device *pdev) rc = riscv_acpi_get_gsi_info(fwnode, &priv->gsi_base, &id, &nr_irqs, NULL); if (rc) { + mbox_free_channel(priv->chan); dev_err(dev, "failed to find GSI mapping\n"); return rc; } From 50242828700f06edfa8d563f9e0acc23a59424ee Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Sun, 8 Feb 2026 19:24:16 +0000 Subject: [PATCH 115/352] btrfs: check for NULL root after calls to btrfs_extent_root() btrfs_extent_root() can return a NULL pointer in case the root we are looking for is not in the rb tree that tracks roots. So add checks to every caller that is missing such check to log a message and return an error. The same applies to callers of btrfs_block_group_root(), since it calls btrfs_extent_root(). Reported-by: Chris Mason Link: https://lore.kernel.org/linux-btrfs/20260208161657.3972997-1-clm@meta.com/ Reviewed-by: Boris Burkov Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/backref.c | 28 ++++++++++++++ fs/btrfs/block-group.c | 36 ++++++++++++++++++ fs/btrfs/disk-io.c | 16 ++++++-- fs/btrfs/extent-tree.c | 78 ++++++++++++++++++++++++++++++++++++-- fs/btrfs/free-space-tree.c | 9 ++++- fs/btrfs/qgroup.c | 8 ++++ fs/btrfs/relocation.c | 22 ++++++++++- fs/btrfs/zoned.c | 7 ++++ 8 files changed, 195 insertions(+), 9 deletions(-) diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 9bb406f7dd30..7921a926f676 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -1393,6 +1393,13 @@ static int find_parent_nodes(struct btrfs_backref_walk_ctx *ctx, .indirect_missing_keys = PREFTREE_INIT }; + if (unlikely(!root)) { + btrfs_err(ctx->fs_info, + "missing extent root for extent at bytenr %llu", + ctx->bytenr); + return -EUCLEAN; + } + /* Roots ulist is not needed when using a sharedness check context. */ if (sc) ASSERT(ctx->roots == NULL); @@ -2204,6 +2211,13 @@ int extent_from_logical(struct btrfs_fs_info *fs_info, u64 logical, struct btrfs_extent_item *ei; struct btrfs_key key; + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + logical); + return -EUCLEAN; + } + key.objectid = logical; if (btrfs_fs_incompat(fs_info, SKINNY_METADATA)) key.type = BTRFS_METADATA_ITEM_KEY; @@ -2851,6 +2865,13 @@ int btrfs_backref_iter_start(struct btrfs_backref_iter *iter, u64 bytenr) struct btrfs_key key; int ret; + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; key.type = BTRFS_METADATA_ITEM_KEY; key.offset = (u64)-1; @@ -2987,6 +3008,13 @@ int btrfs_backref_iter_next(struct btrfs_backref_iter *iter) /* We're at keyed items, there is no inline item, go to the next one */ extent_root = btrfs_extent_root(iter->fs_info, iter->bytenr); + if (unlikely(!extent_root)) { + btrfs_err(iter->fs_info, + "missing extent root for extent at bytenr %llu", + iter->bytenr); + return -EUCLEAN; + } + ret = btrfs_next_item(extent_root, iter->path); if (ret) return ret; diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 77285ade3a0e..fa55d868ecd8 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -739,6 +739,12 @@ static int load_extent_tree_free(struct btrfs_caching_control *caching_ctl) last = max_t(u64, block_group->start, BTRFS_SUPER_INFO_OFFSET); extent_root = btrfs_extent_root(fs_info, last); + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for block group at offset %llu", + block_group->start); + return -EUCLEAN; + } #ifdef CONFIG_BTRFS_DEBUG /* @@ -1061,6 +1067,11 @@ static int remove_block_group_item(struct btrfs_trans_handle *trans, int ret; root = btrfs_block_group_root(fs_info); + if (unlikely(!root)) { + btrfs_err(fs_info, "missing block group root"); + return -EUCLEAN; + } + key.objectid = block_group->start; key.type = BTRFS_BLOCK_GROUP_ITEM_KEY; key.offset = block_group->length; @@ -1349,6 +1360,11 @@ struct btrfs_trans_handle *btrfs_start_trans_remove_block_group( struct btrfs_chunk_map *map; unsigned int num_items; + if (unlikely(!root)) { + btrfs_err(fs_info, "missing block group root"); + return ERR_PTR(-EUCLEAN); + } + map = btrfs_find_chunk_map(fs_info, chunk_offset, 1); ASSERT(map != NULL); ASSERT(map->start == chunk_offset); @@ -2140,6 +2156,11 @@ static int find_first_block_group(struct btrfs_fs_info *fs_info, int ret; struct btrfs_key found_key; + if (unlikely(!root)) { + btrfs_err(fs_info, "missing block group root"); + return -EUCLEAN; + } + btrfs_for_each_slot(root, key, &found_key, path, ret) { if (found_key.objectid >= key->objectid && found_key.type == BTRFS_BLOCK_GROUP_ITEM_KEY) { @@ -2714,6 +2735,11 @@ static int insert_block_group_item(struct btrfs_trans_handle *trans, size_t size; int ret; + if (unlikely(!root)) { + btrfs_err(fs_info, "missing block group root"); + return -EUCLEAN; + } + spin_lock(&block_group->lock); btrfs_set_stack_block_group_v2_used(&bgi, block_group->used); btrfs_set_stack_block_group_v2_chunk_objectid(&bgi, block_group->global_root_id); @@ -3049,6 +3075,11 @@ int btrfs_inc_block_group_ro(struct btrfs_block_group *cache, int ret; bool dirty_bg_running; + if (unlikely(!root)) { + btrfs_err(fs_info, "missing block group root"); + return -EUCLEAN; + } + /* * This can only happen when we are doing read-only scrub on read-only * mount. @@ -3193,6 +3224,11 @@ static int update_block_group_item(struct btrfs_trans_handle *trans, u64 used, remap_bytes; u32 identity_remap_count; + if (unlikely(!root)) { + btrfs_err(fs_info, "missing block group root"); + return -EUCLEAN; + } + /* * Block group items update can be triggered out of commit transaction * critical section, thus we need a consistent view of used bytes. diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 2aeb85e3ece9..08368247668e 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1591,7 +1591,7 @@ static int find_newest_super_backup(struct btrfs_fs_info *info) * this will bump the backup pointer by one when it is * done */ -static void backup_super_roots(struct btrfs_fs_info *info) +static int backup_super_roots(struct btrfs_fs_info *info) { const int next_backup = info->backup_root_index; struct btrfs_root_backup *root_backup; @@ -1623,6 +1623,11 @@ static void backup_super_roots(struct btrfs_fs_info *info) struct btrfs_root *extent_root = btrfs_extent_root(info, 0); struct btrfs_root *csum_root = btrfs_csum_root(info, 0); + if (unlikely(!extent_root)) { + btrfs_err(info, "missing extent root for extent at bytenr 0"); + return -EUCLEAN; + } + btrfs_set_backup_extent_root(root_backup, extent_root->node->start); btrfs_set_backup_extent_root_gen(root_backup, @@ -1670,6 +1675,8 @@ static void backup_super_roots(struct btrfs_fs_info *info) memcpy(&info->super_copy->super_roots, &info->super_for_commit->super_roots, sizeof(*root_backup) * BTRFS_NUM_BACKUP_ROOTS); + + return 0; } /* @@ -4051,8 +4058,11 @@ int write_all_supers(struct btrfs_fs_info *fs_info, int max_mirrors) * not from fsync where the tree roots in fs_info have not * been consistent on disk. */ - if (max_mirrors == 0) - backup_super_roots(fs_info); + if (max_mirrors == 0) { + ret = backup_super_roots(fs_info); + if (ret < 0) + return ret; + } sb = fs_info->super_for_commit; dev_item = &sb->dev_item; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index b0d9baf5b412..7fcd83f32014 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -75,6 +75,12 @@ int btrfs_lookup_data_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len) struct btrfs_key key; BTRFS_PATH_AUTO_FREE(path); + if (unlikely(!root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", start); + return -EUCLEAN; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -131,6 +137,12 @@ int btrfs_lookup_extent_info(struct btrfs_trans_handle *trans, key.offset = offset; extent_root = btrfs_extent_root(fs_info, bytenr); + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + ret = btrfs_search_slot(NULL, extent_root, &key, path, 0, 0); if (ret < 0) return ret; @@ -436,6 +448,12 @@ static noinline int lookup_extent_data_ref(struct btrfs_trans_handle *trans, int recow; int ret; + if (unlikely(!root)) { + btrfs_err(trans->fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; if (parent) { key.type = BTRFS_SHARED_DATA_REF_KEY; @@ -510,6 +528,12 @@ static noinline int insert_extent_data_ref(struct btrfs_trans_handle *trans, u32 num_refs; int ret; + if (unlikely(!root)) { + btrfs_err(trans->fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; if (node->parent) { key.type = BTRFS_SHARED_DATA_REF_KEY; @@ -668,6 +692,12 @@ static noinline int lookup_tree_block_ref(struct btrfs_trans_handle *trans, struct btrfs_key key; int ret; + if (unlikely(!root)) { + btrfs_err(trans->fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; if (parent) { key.type = BTRFS_SHARED_BLOCK_REF_KEY; @@ -692,6 +722,12 @@ static noinline int insert_tree_block_ref(struct btrfs_trans_handle *trans, struct btrfs_key key; int ret; + if (unlikely(!root)) { + btrfs_err(trans->fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; if (node->parent) { key.type = BTRFS_SHARED_BLOCK_REF_KEY; @@ -782,6 +818,12 @@ int lookup_inline_extent_backref(struct btrfs_trans_handle *trans, bool skinny_metadata = btrfs_fs_incompat(fs_info, SKINNY_METADATA); int needed; + if (unlikely(!root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; key.offset = num_bytes; @@ -1680,6 +1722,12 @@ static int run_delayed_extent_op(struct btrfs_trans_handle *trans, } root = btrfs_extent_root(fs_info, key.objectid); + if (unlikely(!root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + key.objectid); + return -EUCLEAN; + } again: ret = btrfs_search_slot(trans, root, &key, path, 0, 1); if (ret < 0) { @@ -2379,6 +2427,12 @@ static noinline int check_committed_ref(struct btrfs_inode *inode, int type; int ret; + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } + key.objectid = bytenr; key.type = BTRFS_EXTENT_ITEM_KEY; key.offset = (u64)-1; @@ -3222,7 +3276,11 @@ static int __btrfs_free_extent(struct btrfs_trans_handle *trans, u64 delayed_ref_root = href->owning_root; extent_root = btrfs_extent_root(info, bytenr); - ASSERT(extent_root); + if (unlikely(!extent_root)) { + btrfs_err(info, + "missing extent root for extent at bytenr %llu", bytenr); + return -EUCLEAN; + } path = btrfs_alloc_path(); if (!path) @@ -4939,11 +4997,18 @@ static int alloc_reserved_file_extent(struct btrfs_trans_handle *trans, size += btrfs_extent_inline_ref_size(BTRFS_EXTENT_OWNER_REF_KEY); size += btrfs_extent_inline_ref_size(type); + extent_root = btrfs_extent_root(fs_info, ins->objectid); + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + ins->objectid); + return -EUCLEAN; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; - extent_root = btrfs_extent_root(fs_info, ins->objectid); ret = btrfs_insert_empty_item(trans, extent_root, path, ins, size); if (ret) { btrfs_free_path(path); @@ -5019,11 +5084,18 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, size += sizeof(*block_info); } + extent_root = btrfs_extent_root(fs_info, extent_key.objectid); + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + extent_key.objectid); + return -EUCLEAN; + } + path = btrfs_alloc_path(); if (!path) return -ENOMEM; - extent_root = btrfs_extent_root(fs_info, extent_key.objectid); ret = btrfs_insert_empty_item(trans, extent_root, path, &extent_key, size); if (ret) { diff --git a/fs/btrfs/free-space-tree.c b/fs/btrfs/free-space-tree.c index ecddfca92b2b..9efd1ec90f03 100644 --- a/fs/btrfs/free-space-tree.c +++ b/fs/btrfs/free-space-tree.c @@ -1073,6 +1073,14 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans, if (ret) return ret; + extent_root = btrfs_extent_root(trans->fs_info, block_group->start); + if (unlikely(!extent_root)) { + btrfs_err(trans->fs_info, + "missing extent root for block group at offset %llu", + block_group->start); + return -EUCLEAN; + } + mutex_lock(&block_group->free_space_lock); /* @@ -1086,7 +1094,6 @@ static int populate_free_space_tree(struct btrfs_trans_handle *trans, key.type = BTRFS_EXTENT_ITEM_KEY; key.offset = 0; - extent_root = btrfs_extent_root(trans->fs_info, key.objectid); ret = btrfs_search_slot_for_read(extent_root, &key, path, 1, 0); if (ret < 0) goto out_locked; diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 19edd25ff5d1..0fa66fe4fbb0 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3740,6 +3740,14 @@ static int qgroup_rescan_leaf(struct btrfs_trans_handle *trans, mutex_lock(&fs_info->qgroup_rescan_lock); extent_root = btrfs_extent_root(fs_info, fs_info->qgroup_rescan_progress.objectid); + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + fs_info->qgroup_rescan_progress.objectid); + mutex_unlock(&fs_info->qgroup_rescan_lock); + return -EUCLEAN; + } + ret = btrfs_search_slot_for_read(extent_root, &fs_info->qgroup_rescan_progress, path, 1, 0); diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index fcfbe1b1dab4..93a5ae23406d 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4954,6 +4954,12 @@ static int do_remap_reloc_trans(struct btrfs_fs_info *fs_info, struct btrfs_space_info *sinfo = src_bg->space_info; extent_root = btrfs_extent_root(fs_info, src_bg->start); + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for block group at offset %llu", + src_bg->start); + return -EUCLEAN; + } trans = btrfs_start_transaction(extent_root, 0); if (IS_ERR(trans)) @@ -5306,6 +5312,13 @@ int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start, int ret; bool bg_is_ro = false; + if (unlikely(!extent_root)) { + btrfs_err(fs_info, + "missing extent root for block group at offset %llu", + group_start); + return -EUCLEAN; + } + /* * This only gets set if we had a half-deleted snapshot on mount. We * cannot allow relocation to start while we're still trying to clean up @@ -5536,12 +5549,17 @@ int btrfs_recover_relocation(struct btrfs_fs_info *fs_info) goto out; } + rc->extent_root = btrfs_extent_root(fs_info, 0); + if (unlikely(!rc->extent_root)) { + btrfs_err(fs_info, "missing extent root for extent at bytenr 0"); + ret = -EUCLEAN; + goto out; + } + ret = reloc_chunk_start(fs_info); if (ret < 0) goto out_end; - rc->extent_root = btrfs_extent_root(fs_info, 0); - set_reloc_control(rc); trans = btrfs_join_transaction(rc->extent_root); diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 92b5ac8fac37..afb5924018ac 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -1261,6 +1261,13 @@ static int calculate_alloc_pointer(struct btrfs_block_group *cache, key.offset = 0; root = btrfs_extent_root(fs_info, key.objectid); + if (unlikely(!root)) { + btrfs_err(fs_info, + "missing extent root for extent at bytenr %llu", + key.objectid); + return -EUCLEAN; + } + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); /* We should not find the exact match */ if (unlikely(!ret)) From 2b4cb4e58f3463d142fcece5a19e0405fb82c794 Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Sun, 8 Feb 2026 19:43:01 +0000 Subject: [PATCH 116/352] btrfs: check for NULL root after calls to btrfs_csum_root() btrfs_csum_root() can return a NULL pointer in case the root we are looking for is not in the rb tree that tracks roots. So add checks to every caller that is missing such check to log a message and return an error. Reported-by: Chris Mason Link: https://lore.kernel.org/linux-btrfs/20260208161657.3972997-1-clm@meta.com/ Reviewed-by: Boris Burkov Signed-off-by: Filipe Manana Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/disk-io.c | 4 ++++ fs/btrfs/extent-tree.c | 20 ++++++++++++++++++-- fs/btrfs/file-item.c | 7 +++++++ fs/btrfs/inode.c | 18 ++++++++++++++++-- fs/btrfs/raid56.c | 12 ++++++++++-- fs/btrfs/relocation.c | 8 ++++++++ fs/btrfs/tree-log.c | 21 +++++++++++++++++++++ 7 files changed, 84 insertions(+), 6 deletions(-) diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 08368247668e..b1b53d713ee9 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1627,6 +1627,10 @@ static int backup_super_roots(struct btrfs_fs_info *info) btrfs_err(info, "missing extent root for extent at bytenr 0"); return -EUCLEAN; } + if (unlikely(!csum_root)) { + btrfs_err(info, "missing csum root for extent at bytenr 0"); + return -EUCLEAN; + } btrfs_set_backup_extent_root(root_backup, extent_root->node->start); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 7fcd83f32014..85ee5c79759d 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -1974,8 +1974,15 @@ static int cleanup_ref_head(struct btrfs_trans_handle *trans, struct btrfs_root *csum_root; csum_root = btrfs_csum_root(fs_info, head->bytenr); - ret = btrfs_del_csums(trans, csum_root, head->bytenr, - head->num_bytes); + if (unlikely(!csum_root)) { + btrfs_err(fs_info, + "missing csum root for extent at bytenr %llu", + head->bytenr); + ret = -EUCLEAN; + } else { + ret = btrfs_del_csums(trans, csum_root, head->bytenr, + head->num_bytes); + } } } @@ -3147,6 +3154,15 @@ static int do_free_extent_accounting(struct btrfs_trans_handle *trans, struct btrfs_root *csum_root; csum_root = btrfs_csum_root(trans->fs_info, bytenr); + if (unlikely(!csum_root)) { + ret = -EUCLEAN; + btrfs_abort_transaction(trans, ret); + btrfs_err(trans->fs_info, + "missing csum root for extent at bytenr %llu", + bytenr); + return ret; + } + ret = btrfs_del_csums(trans, csum_root, bytenr, num_bytes); if (unlikely(ret)) { btrfs_abort_transaction(trans, ret); diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 7bd715442f3e..f585ddfa8440 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -308,6 +308,13 @@ static int search_csum_tree(struct btrfs_fs_info *fs_info, /* Current item doesn't contain the desired range, search again */ btrfs_release_path(path); csum_root = btrfs_csum_root(fs_info, disk_bytenr); + if (unlikely(!csum_root)) { + btrfs_err(fs_info, + "missing csum root for extent at bytenr %llu", + disk_bytenr); + return -EUCLEAN; + } + item = btrfs_lookup_csum(NULL, csum_root, path, disk_bytenr, 0); if (IS_ERR(item)) { ret = PTR_ERR(item); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index b409efe1857e..ed4d19780f22 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -2012,6 +2012,13 @@ static int can_nocow_file_extent(struct btrfs_path *path, */ csum_root = btrfs_csum_root(root->fs_info, io_start); + if (unlikely(!csum_root)) { + btrfs_err(root->fs_info, + "missing csum root for extent at bytenr %llu", io_start); + ret = -EUCLEAN; + goto out; + } + ret = btrfs_lookup_csums_list(csum_root, io_start, io_start + args->file_extent.num_bytes - 1, NULL, nowait); @@ -2749,10 +2756,17 @@ static int add_pending_csums(struct btrfs_trans_handle *trans, int ret; list_for_each_entry(sum, list, list) { - trans->adding_csums = true; - if (!csum_root) + if (!csum_root) { csum_root = btrfs_csum_root(trans->fs_info, sum->logical); + if (unlikely(!csum_root)) { + btrfs_err(trans->fs_info, + "missing csum root for extent at bytenr %llu", + sum->logical); + return -EUCLEAN; + } + } + trans->adding_csums = true; ret = btrfs_csum_file_blocks(trans, csum_root, sum); trans->adding_csums = false; if (ret) diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index baadaaa189c0..230dd93dad6e 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -2295,8 +2295,7 @@ void raid56_parity_recover(struct bio *bio, struct btrfs_io_context *bioc, static void fill_data_csums(struct btrfs_raid_bio *rbio) { struct btrfs_fs_info *fs_info = rbio->bioc->fs_info; - struct btrfs_root *csum_root = btrfs_csum_root(fs_info, - rbio->bioc->full_stripe_logical); + struct btrfs_root *csum_root; const u64 start = rbio->bioc->full_stripe_logical; const u32 len = (rbio->nr_data * rbio->stripe_nsectors) << fs_info->sectorsize_bits; @@ -2329,6 +2328,15 @@ static void fill_data_csums(struct btrfs_raid_bio *rbio) goto error; } + csum_root = btrfs_csum_root(fs_info, rbio->bioc->full_stripe_logical); + if (unlikely(!csum_root)) { + btrfs_err(fs_info, + "missing csum root for extent at bytenr %llu", + rbio->bioc->full_stripe_logical); + ret = -EUCLEAN; + goto error; + } + ret = btrfs_lookup_csums_bitmap(csum_root, NULL, start, start + len - 1, rbio->csum_buf, rbio->csum_bitmap); if (ret < 0) diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 93a5ae23406d..2625a66054d0 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -5654,6 +5654,14 @@ int btrfs_reloc_clone_csums(struct btrfs_ordered_extent *ordered) LIST_HEAD(list); int ret; + if (unlikely(!csum_root)) { + btrfs_mark_ordered_extent_error(ordered); + btrfs_err(fs_info, + "missing csum root for extent at bytenr %llu", + disk_bytenr); + return -EUCLEAN; + } + ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + ordered->num_bytes - 1, &list, false); diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 37c272df52b0..9ff3933bc382 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -984,6 +984,13 @@ static noinline int replay_one_extent(struct walk_control *wc) sums = list_first_entry(&ordered_sums, struct btrfs_ordered_sum, list); csum_root = btrfs_csum_root(fs_info, sums->logical); + if (unlikely(!csum_root)) { + btrfs_err(fs_info, + "missing csum root for extent at bytenr %llu", + sums->logical); + ret = -EUCLEAN; + } + if (!ret) { ret = btrfs_del_csums(trans, csum_root, sums->logical, sums->len); @@ -4890,6 +4897,13 @@ static noinline int copy_items(struct btrfs_trans_handle *trans, } csum_root = btrfs_csum_root(trans->fs_info, disk_bytenr); + if (unlikely(!csum_root)) { + btrfs_err(trans->fs_info, + "missing csum root for extent at bytenr %llu", + disk_bytenr); + return -EUCLEAN; + } + disk_bytenr += extent_offset; ret = btrfs_lookup_csums_list(csum_root, disk_bytenr, disk_bytenr + extent_num_bytes - 1, @@ -5086,6 +5100,13 @@ static int log_extent_csums(struct btrfs_trans_handle *trans, /* block start is already adjusted for the file extent offset. */ block_start = btrfs_extent_map_block_start(em); csum_root = btrfs_csum_root(trans->fs_info, block_start); + if (unlikely(!csum_root)) { + btrfs_err(trans->fs_info, + "missing csum root for extent at bytenr %llu", + block_start); + return -EUCLEAN; + } + ret = btrfs_lookup_csums_list(csum_root, block_start + csum_offset, block_start + csum_offset + csum_len - 1, &ordered_sums, false); From f9a4e3015db1aeafbef407650eb8555445ca943e Mon Sep 17 00:00:00 2001 From: Filipe Manana Date: Fri, 13 Feb 2026 16:08:53 +0000 Subject: [PATCH 117/352] btrfs: reserve enough transaction items for qgroup ioctls Currently our qgroup ioctls don't reserve any space, they just do a transaction join, which does not reserve any space, neither for the quota tree updates nor for the delayed refs generated when updating the quota tree. The quota root uses the global block reserve, which is fine most of the time since we don't expect a lot of updates to the quota root, or to be too close to -ENOSPC such that other critical metadata updates need to resort to the global reserve. However this is not optimal, as not reserving proper space may result in a transaction abort due to not reserving space for delayed refs and then abusing the use of the global block reserve. For example, the following reproducer (which is unlikely to model any real world use case, but just to illustrate the problem), triggers such a transaction abort due to -ENOSPC when running delayed refs: $ cat test.sh #!/bin/bash DEV=/dev/nullb0 MNT=/mnt/nullb0 umount $DEV &> /dev/null # Limit device to 1G so that it's much faster to reproduce the issue. mkfs.btrfs -f -b 1G $DEV mount -o commit=600 $DEV $MNT fallocate -l 800M $MNT/filler btrfs quota enable $MNT for ((i = 1; i <= 400000; i++)); do btrfs qgroup create 1/$i $MNT done umount $MNT When running this, we can see in dmesg/syslog that a transaction abort happened: [436.490] BTRFS error (device nullb0): failed to run delayed ref for logical 30408704 num_bytes 16384 type 176 action 1 ref_mod 1: -28 [436.493] ------------[ cut here ]------------ [436.494] BTRFS: Transaction aborted (error -28) [436.495] WARNING: fs/btrfs/extent-tree.c:2247 at btrfs_run_delayed_refs+0xd9/0x110 [btrfs], CPU#4: umount/2495372 [436.497] Modules linked in: btrfs loop (...) [436.508] CPU: 4 UID: 0 PID: 2495372 Comm: umount Tainted: G W 6.19.0-rc8-btrfs-next-225+ #1 PREEMPT(full) [436.510] Tainted: [W]=WARN [436.511] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.2-0-gea1b7a073390-prebuilt.qemu.org 04/01/2014 [436.513] RIP: 0010:btrfs_run_delayed_refs+0xdf/0x110 [btrfs] [436.514] Code: 0f 82 ea (...) [436.518] RSP: 0018:ffffd511850b7d78 EFLAGS: 00010292 [436.519] RAX: 00000000ffffffe4 RBX: ffff8f120dad37e0 RCX: 0000000002040001 [436.520] RDX: 0000000000000002 RSI: 00000000ffffffe4 RDI: ffffffffc090fd80 [436.522] RBP: 0000000000000000 R08: 0000000000000001 R09: ffffffffc04d1867 [436.523] R10: ffff8f18dc1fffa8 R11: 0000000000000003 R12: ffff8f173aa89400 [436.524] R13: 0000000000000000 R14: ffff8f173aa89400 R15: 0000000000000000 [436.526] FS: 00007fe59045d840(0000) GS:ffff8f192e22e000(0000) knlGS:0000000000000000 [436.527] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [436.528] CR2: 00007fe5905ff2b0 CR3: 000000060710a002 CR4: 0000000000370ef0 [436.530] Call Trace: [436.530] [436.530] btrfs_commit_transaction+0x73/0xc00 [btrfs] [436.531] ? btrfs_attach_transaction_barrier+0x1e/0x70 [btrfs] [436.532] sync_filesystem+0x7a/0x90 [436.533] generic_shutdown_super+0x28/0x180 [436.533] kill_anon_super+0x12/0x40 [436.534] btrfs_kill_super+0x12/0x20 [btrfs] [436.534] deactivate_locked_super+0x2f/0xb0 [436.534] cleanup_mnt+0xea/0x180 [436.535] task_work_run+0x58/0xa0 [436.535] exit_to_user_mode_loop+0xed/0x480 [436.536] ? __x64_sys_umount+0x68/0x80 [436.536] do_syscall_64+0x2a5/0xf20 [436.537] entry_SYSCALL_64_after_hwframe+0x76/0x7e [436.537] RIP: 0033:0x7fe5906b6217 [436.538] Code: 0d 00 f7 (...) [436.540] RSP: 002b:00007ffcd87a61f8 EFLAGS: 00000246 ORIG_RAX: 00000000000000a6 [436.541] RAX: 0000000000000000 RBX: 00005618b9ecadc8 RCX: 00007fe5906b6217 [436.541] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 00005618b9ecb100 [436.542] RBP: 0000000000000000 R08: 00007ffcd87a4fe0 R09: 00000000ffffffff [436.544] R10: 0000000000000103 R11: 0000000000000246 R12: 00007fe59081626c [436.544] R13: 00005618b9ecb100 R14: 0000000000000000 R15: 00005618b9ecacc0 [436.545] [436.545] ---[ end trace 0000000000000000 ]--- Fix this by changing the qgroup ioctls to use start transaction instead of joining so that proper space is reserved for the delayed refs generated for the updates to the quota root. This way we don't get any transaction abort. Reviewed-by: Boris Burkov Reviewed-by: Qu Wenruo Signed-off-by: Filipe Manana Signed-off-by: David Sterba --- fs/btrfs/ioctl.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 7d86e9c8909e..7d1e3f41dc7d 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -3687,7 +3687,8 @@ static long btrfs_ioctl_qgroup_assign(struct file *file, void __user *arg) } } - trans = btrfs_join_transaction(root); + /* 2 BTRFS_QGROUP_RELATION_KEY items. */ + trans = btrfs_start_transaction(root, 2); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; @@ -3759,7 +3760,11 @@ static long btrfs_ioctl_qgroup_create(struct file *file, void __user *arg) goto out; } - trans = btrfs_join_transaction(root); + /* + * 1 BTRFS_QGROUP_INFO_KEY item. + * 1 BTRFS_QGROUP_LIMIT_KEY item. + */ + trans = btrfs_start_transaction(root, 2); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; @@ -3808,7 +3813,8 @@ static long btrfs_ioctl_qgroup_limit(struct file *file, void __user *arg) goto drop_write; } - trans = btrfs_join_transaction(root); + /* 1 BTRFS_QGROUP_LIMIT_KEY item. */ + trans = btrfs_start_transaction(root, 1); if (IS_ERR(trans)) { ret = PTR_ERR(trans); goto out; From 65ee6061388b334c341fd37c22ec9149417f6ccf Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Thu, 19 Feb 2026 18:51:11 +1030 Subject: [PATCH 118/352] btrfs: fix a bug that makes encoded write bio larger than expected [BUG] When running btrfs/284 with 64K page size and 4K fs block size, the following ASSERT() can be triggered: assertion failed: cb->bbio.bio.bi_iter.bi_size == disk_num_bytes :: 0, in inode.c:9991 ------------[ cut here ]------------ kernel BUG at inode.c:9991! Internal error: Oops - BUG: 00000000f2000800 [#1] SMP CPU: 5 UID: 0 PID: 6787 Comm: btrfs Tainted: G OE 6.19.0-rc8-custom+ #1 PREEMPT(voluntary) Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 pc : btrfs_do_encoded_write+0x9b0/0x9c0 [btrfs] lr : btrfs_do_encoded_write+0x9b0/0x9c0 [btrfs] Call trace: btrfs_do_encoded_write+0x9b0/0x9c0 [btrfs] (P) btrfs_do_write_iter+0x1d8/0x208 [btrfs] btrfs_ioctl_encoded_write+0x3c8/0x6d0 [btrfs] btrfs_ioctl+0xeb0/0x2b60 [btrfs] __arm64_sys_ioctl+0xac/0x110 invoke_syscall.constprop.0+0x64/0xe8 el0_svc_common.constprop.0+0x40/0xe8 do_el0_svc+0x24/0x38 el0_svc+0x3c/0x1b8 el0t_64_sync_handler+0xa0/0xe8 el0t_64_sync+0x1a4/0x1a8 Code: 91180021 90001080 9111a000 94039d54 (d4210000) ---[ end trace 0000000000000000 ]--- [CAUSE] After commit e1bc83f8b157 ("btrfs: get rid of compressed_folios[] usage for encoded writes"), the encoded write is changed to copy the content from the iov into a folio, and queue the folio into the compressed bio. However we always queue the full folio into the compressed bio, which can make the compressed bio larger than the on-disk extent, if the folio size is larger than the fs block size. Although we have an ASSERT() to catch such problem, for kernels without CONFIG_BTRFS_ASSERT, such larger than expected bio will just be submitted, possibly overwrite the next data extent, causing data corruption. [FIX] Instead of blindly queuing the full folio into the compressed bio, only queue the rounded up range, which is the old behavior before that offending commit. This also means we no longer need to zero the tailing range until the folio end (but still to the block boundary), as such range will not be submitted anyway. And since we're here, add a final ASSERT() into btrfs_submit_compressed_write() as the last safety net for kernels with btrfs assertions enabled Fixes: e1bc83f8b157 ("btrfs: get rid of compressed_folios[] usage for encoded writes") Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Signed-off-by: David Sterba --- fs/btrfs/compression.c | 1 + fs/btrfs/inode.c | 7 ++++--- 2 files changed, 5 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 1e7174ad32e2..ac995ec78e05 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -324,6 +324,7 @@ void btrfs_submit_compressed_write(struct btrfs_ordered_extent *ordered, cb->start = ordered->file_offset; cb->len = ordered->num_bytes; + ASSERT(cb->bbio.bio.bi_iter.bi_size == ordered->disk_num_bytes); cb->compressed_len = ordered->disk_num_bytes; cb->bbio.bio.bi_iter.bi_sector = ordered->disk_bytenr >> SECTOR_SHIFT; cb->bbio.ordered = ordered; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index ed4d19780f22..8abfe8f0f2d4 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -9889,6 +9889,7 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, int compression; size_t orig_count; const u32 min_folio_size = btrfs_min_folio_size(fs_info); + const u32 blocksize = fs_info->sectorsize; u64 start, end; u64 num_bytes, ram_bytes, disk_num_bytes; struct btrfs_key ins; @@ -9999,9 +10000,9 @@ ssize_t btrfs_do_encoded_write(struct kiocb *iocb, struct iov_iter *from, ret = -EFAULT; goto out_cb; } - if (bytes < min_folio_size) - folio_zero_range(folio, bytes, min_folio_size - bytes); - ret = bio_add_folio(&cb->bbio.bio, folio, folio_size(folio), 0); + if (!IS_ALIGNED(bytes, blocksize)) + folio_zero_range(folio, bytes, round_up(bytes, blocksize) - bytes); + ret = bio_add_folio(&cb->bbio.bio, folio, round_up(bytes, blocksize), 0); if (unlikely(!ret)) { folio_put(folio); ret = -EINVAL; From 3adf8f14152fba1cae51f9b0d3570a1da2153b16 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Thu, 19 Feb 2026 18:51:12 +1030 Subject: [PATCH 119/352] btrfs: do not touch page cache for encoded writes [BUG] When running btrfs/284, the following ASSERT() will be triggered with 64K page size and 4K fs block size: assertion failed: folio_test_writeback(folio) :: 0, in subpage.c:476 ------------[ cut here ]------------ kernel BUG at subpage.c:476! Internal error: Oops - BUG: 00000000f2000800 [#1] SMP CPU: 4 UID: 0 PID: 2313 Comm: kworker/u37:2 Tainted: G OE 6.19.0-rc8-custom+ #185 PREEMPT(voluntary) Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 Workqueue: btrfs-endio simple_end_io_work [btrfs] pc : btrfs_subpage_clear_writeback+0x148/0x160 [btrfs] lr : btrfs_subpage_clear_writeback+0x148/0x160 [btrfs] Call trace: btrfs_subpage_clear_writeback+0x148/0x160 [btrfs] (P) btrfs_folio_clamp_clear_writeback+0xb4/0xd0 [btrfs] end_compressed_writeback+0xe0/0x1e0 [btrfs] end_bbio_compressed_write+0x1e8/0x218 [btrfs] btrfs_bio_end_io+0x108/0x258 [btrfs] simple_end_io_work+0x68/0xa8 [btrfs] process_one_work+0x168/0x3f0 worker_thread+0x25c/0x398 kthread+0x154/0x250 ret_from_fork+0x10/0x20 ---[ end trace 0000000000000000 ]--- [CAUSE] The offending bio is from an encoded write, where the compressed data is directly written as a data extent, without touching the page cache. However the encoded write still utilizes the regular buffered write path for compressed data, by setting the compressed_bio::writeback flag. When that flag is set, at end_bbio_compressed_write() btrfs will go clearing the writeback flag of the folios in the page cache. However for bs < ps cases, the subpage helper has one extra check to make sure the folio has a writeback flag set in the first place. But since it's an encoded write, we never go through page cache, thus the folio has no writeback flag and triggers the ASSERT(). [FIX] Do not set compressed_bio::writeback flag for encoded writes, and change the ASSERT() in btrfs_submit_compressed_write() to make sure that flag is not set. Fixes: e1bc83f8b157 ("btrfs: get rid of compressed_folios[] usage for encoded writes") Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Signed-off-by: David Sterba --- fs/btrfs/compression.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index ac995ec78e05..dc61f7e3cbbf 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -320,7 +320,12 @@ void btrfs_submit_compressed_write(struct btrfs_ordered_extent *ordered, ASSERT(IS_ALIGNED(ordered->file_offset, fs_info->sectorsize)); ASSERT(IS_ALIGNED(ordered->num_bytes, fs_info->sectorsize)); - ASSERT(cb->writeback); + /* + * This flag determines if we should clear the writeback flag from the + * page cache. But this function is only utilized by encoded writes, it + * never goes through the page cache. + */ + ASSERT(!cb->writeback); cb->start = ordered->file_offset; cb->len = ordered->num_bytes; @@ -346,8 +351,7 @@ struct compressed_bio *btrfs_alloc_compressed_write(struct btrfs_inode *inode, cb = alloc_compressed_bio(inode, start, REQ_OP_WRITE, end_bbio_compressed_write); cb->start = start; cb->len = len; - cb->writeback = true; - + cb->writeback = false; return cb; } From 96a2d235896b53291efc6d5c3de030e570f77070 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Thu, 19 Feb 2026 18:51:13 +1030 Subject: [PATCH 120/352] btrfs: fix an incorrect ASSERT() condition inside zstd_decompress_bio() [BUG] When running btrfs/284 with 64K page size and 4K fs block size, it crashes with the following ASSERT() triggered: assertion failed: folio_size(fi.folio) == blocksize :: 0, in fs/btrfs/zstd.c:603 ------------[ cut here ]------------ kernel BUG at fs/btrfs/zstd.c:603! Internal error: Oops - BUG: 00000000f2000800 [#1] SMP CPU: 2 UID: 0 PID: 1183 Comm: kworker/u35:4 Not tainted 6.19.0-rc8-custom+ #185 PREEMPT(voluntary) Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 Workqueue: btrfs-endio simple_end_io_work [btrfs] pc : zstd_decompress_bio+0x4f0/0x508 [btrfs] lr : zstd_decompress_bio+0x4f0/0x508 [btrfs] Call trace: zstd_decompress_bio+0x4f0/0x508 [btrfs] (P) end_bbio_compressed_read+0x260/0x2c0 [btrfs] btrfs_bio_end_io+0xc4/0x258 [btrfs] btrfs_check_read_bio+0x424/0x7e0 [btrfs] simple_end_io_work+0x40/0xa8 [btrfs] process_one_work+0x168/0x3f0 worker_thread+0x25c/0x398 kthread+0x154/0x250 ret_from_fork+0x10/0x20 ---[ end trace 0000000000000000 ]--- [CAUSE] Commit 1914b94231e9 ("btrfs: zstd: use folio_iter to handle zstd_decompress_bio()") added the ASSERT() to make sure the folio size matches the fs block size. But the check is completely wrong, the original intention is to make sure for bs > ps cases, we always got a large folio that covers a full fs block. However for bs < ps cases, a folio can never be smaller than page size, and the ASSERT() gets triggered immediately. [FIX] Check the folio size against @min_folio_size instead, which will never be smaller than PAGE_SIZE, and still cover bs > ps cases. Fixes: 1914b94231e9 ("btrfs: zstd: use folio_iter to handle zstd_decompress_bio()") Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/zstd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index 32fd7f5454d3..c002d18666b7 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -600,7 +600,7 @@ int zstd_decompress_bio(struct list_head *ws, struct compressed_bio *cb) bio_first_folio(&fi, &cb->bbio.bio, 0); if (unlikely(!fi.folio)) return -EINVAL; - ASSERT(folio_size(fi.folio) == blocksize); + ASSERT(folio_size(fi.folio) == min_folio_size); stream = zstd_init_dstream( ZSTD_BTRFS_MAX_INPUT, workspace->mem, workspace->size); From 5118130e722b1261a2e92b2fb0b067463c39ecc7 Mon Sep 17 00:00:00 2001 From: Qu Wenruo Date: Thu, 19 Feb 2026 18:51:14 +1030 Subject: [PATCH 121/352] btrfs: fix an incorrect ASSERT() condition inside lzo_decompress_bio() [BUG] When running btrfs/284 with 64K page size and 4K fs block size, it crashes with the following ASSERT() triggered: BTRFS info (device dm-3): use lzo compression, level 1 assertion failed: folio_size(fi.folio) == sectorsize :: 0, in lzo.c:450 ------------[ cut here ]------------ kernel BUG at lzo.c:450! Internal error: Oops - BUG: 00000000f2000800 [#1] SMP CPU: 4 UID: 0 PID: 329 Comm: kworker/u37:2 Tainted: G OE 6.19.0-rc8-custom+ #185 PREEMPT(voluntary) Hardware name: QEMU KVM Virtual Machine, BIOS unknown 2/2/2022 Workqueue: btrfs-endio simple_end_io_work [btrfs] pc : lzo_decompress_bio+0x61c/0x630 [btrfs] lr : lzo_decompress_bio+0x61c/0x630 [btrfs] Call trace: lzo_decompress_bio+0x61c/0x630 [btrfs] (P) end_bbio_compressed_read+0x2a8/0x2c0 [btrfs] btrfs_bio_end_io+0xc4/0x258 [btrfs] btrfs_check_read_bio+0x424/0x7e0 [btrfs] simple_end_io_work+0x40/0xa8 [btrfs] process_one_work+0x168/0x3f0 worker_thread+0x25c/0x398 kthread+0x154/0x250 ret_from_fork+0x10/0x20 Code: 912a2021 b0000e00 91246000 940244e9 (d4210000) ---[ end trace 0000000000000000 ]--- [CAUSE] Commit 37cc07cab7dc ("btrfs: lzo: use folio_iter to handle lzo_decompress_bio()") added the ASSERT() to make sure the folio size matches the fs block size. But the check is completely wrong, the original intention is to make sure for bs > ps cases, we always got a large folio that covers a full fs block. However for bs < ps cases, a folio can never be smaller than page size, and the ASSERT() gets triggered immediately. [FIX] Check the folio size against @min_folio_size instead, which will never be smaller than PAGE_SIZE, and still cover bs > ps cases. Fixes: 37cc07cab7dc ("btrfs: lzo: use folio_iter to handle lzo_decompress_bio()") Reviewed-by: Filipe Manana Signed-off-by: Qu Wenruo Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/lzo.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index 8e20497afffe..971c2ea98e18 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -429,7 +429,7 @@ static void copy_compressed_segment(struct compressed_bio *cb, int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) { struct workspace *workspace = list_entry(ws, struct workspace, list); - const struct btrfs_fs_info *fs_info = cb->bbio.inode->root->fs_info; + struct btrfs_fs_info *fs_info = cb->bbio.inode->root->fs_info; const u32 sectorsize = fs_info->sectorsize; struct folio_iter fi; char *kaddr; @@ -447,7 +447,7 @@ int lzo_decompress_bio(struct list_head *ws, struct compressed_bio *cb) /* There must be a compressed folio and matches the sectorsize. */ if (unlikely(!fi.folio)) return -EINVAL; - ASSERT(folio_size(fi.folio) == sectorsize); + ASSERT(folio_size(fi.folio) == btrfs_min_folio_size(fs_info)); kaddr = kmap_local_folio(fi.folio, 0); len_in = read_compress_length(kaddr); kunmap_local(kaddr); From 057495ccc0ad381015b45d3edf995c2b6b982474 Mon Sep 17 00:00:00 2001 From: Mark Harmstone Date: Thu, 19 Feb 2026 17:03:53 +0000 Subject: [PATCH 122/352] btrfs: hold block group reference during entire move_existing_remap() There is a potential use-after-free in move_existing_remap(): we're calling btrfs_put_block_group() on dest_bg, then passing it to btrfs_add_block_group_free_space() a few lines later. Fix this by getting the BG at the start of the function and putting it near the end. This also means we're not doing a lookup twice for the same thing. Reported-by: Chris Mason Link: https://lore.kernel.org/linux-btrfs/20260125123908.2096548-1-clm@meta.com/ Fixes: bbea42dfb91f ("btrfs: move existing remaps before relocating block group") Reviewed-by: Johannes Thumshirn Signed-off-by: Mark Harmstone Reviewed-by: David Sterba Signed-off-by: David Sterba --- fs/btrfs/relocation.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 2625a66054d0..0f61bdf7f5d1 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4185,6 +4185,8 @@ static int move_existing_remap(struct btrfs_fs_info *fs_info, dest_addr = ins.objectid; dest_length = ins.offset; + dest_bg = btrfs_lookup_block_group(fs_info, dest_addr); + if (!is_data && !IS_ALIGNED(dest_length, fs_info->nodesize)) { u64 new_length = ALIGN_DOWN(dest_length, fs_info->nodesize); @@ -4295,15 +4297,12 @@ static int move_existing_remap(struct btrfs_fs_info *fs_info, if (unlikely(ret)) goto end; - dest_bg = btrfs_lookup_block_group(fs_info, dest_addr); - adjust_block_group_remap_bytes(trans, dest_bg, dest_length); mutex_lock(&dest_bg->free_space_lock); bg_needs_free_space = test_bit(BLOCK_GROUP_FLAG_NEEDS_FREE_SPACE, &dest_bg->runtime_flags); mutex_unlock(&dest_bg->free_space_lock); - btrfs_put_block_group(dest_bg); if (bg_needs_free_space) { ret = btrfs_add_block_group_free_space(trans, dest_bg); @@ -4333,13 +4332,13 @@ static int move_existing_remap(struct btrfs_fs_info *fs_info, btrfs_end_transaction(trans); } } else { - dest_bg = btrfs_lookup_block_group(fs_info, dest_addr); btrfs_free_reserved_bytes(dest_bg, dest_length, 0); - btrfs_put_block_group(dest_bg); ret = btrfs_commit_transaction(trans); } + btrfs_put_block_group(dest_bg); + return ret; } From adbb0ebacc3223a2dc2e58ef3d4c10f5e9653f09 Mon Sep 17 00:00:00 2001 From: Mark Harmstone Date: Thu, 19 Feb 2026 19:19:00 +0000 Subject: [PATCH 123/352] btrfs: check block group before marking it unused in balance_remap_chunks() Fix a potential segfault in balance_remap_chunks(): if we quit early because btrfs_inc_block_group_ro() fails, all the remaining items in the chunks list will still have their bg value set to NULL. It's thus not safe to dereference this pointer without checking first. Reported-by: Chris Mason Link: https://lore.kernel.org/linux-btrfs/20260125120717.1578828-1-clm@meta.com/ Fixes: 81e5a4551c32 ("btrfs: allow balancing remap tree") Reviewed-by: Johannes Thumshirn Signed-off-by: Mark Harmstone Signed-off-by: David Sterba --- fs/btrfs/volumes.c | 25 +++++++++++++++++-------- 1 file changed, 17 insertions(+), 8 deletions(-) diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index 7efc2f7a9bb2..8fbd736aad9f 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -4277,20 +4277,29 @@ static int balance_remap_chunks(struct btrfs_fs_info *fs_info, struct btrfs_path end: while (!list_empty(chunks)) { bool is_unused; + struct btrfs_block_group *bg; rci = list_first_entry(chunks, struct remap_chunk_info, list); - spin_lock(&rci->bg->lock); - is_unused = !btrfs_is_block_group_used(rci->bg); - spin_unlock(&rci->bg->lock); + bg = rci->bg; + if (bg) { + /* + * This is a bit racy and the 'used' status can change + * but this is not a problem as later functions will + * verify it again. + */ + spin_lock(&bg->lock); + is_unused = !btrfs_is_block_group_used(bg); + spin_unlock(&bg->lock); - if (is_unused) - btrfs_mark_bg_unused(rci->bg); + if (is_unused) + btrfs_mark_bg_unused(bg); - if (rci->made_ro) - btrfs_dec_block_group_ro(rci->bg); + if (rci->made_ro) + btrfs_dec_block_group_ro(bg); - btrfs_put_block_group(rci->bg); + btrfs_put_block_group(bg); + } list_del(&rci->list); kfree(rci); From b17b79ff896305fd74980a5f72afec370ee88ca4 Mon Sep 17 00:00:00 2001 From: ZhengYuan Huang Date: Thu, 12 Mar 2026 08:14:43 +0800 Subject: [PATCH 124/352] btrfs: reject root items with drop_progress and zero drop_level [BUG] When recovering relocation at mount time, merge_reloc_root() and btrfs_drop_snapshot() both use BUG_ON(level == 0) to guard against an impossible state: a non-zero drop_progress combined with a zero drop_level in a root_item, which can be triggered: ------------[ cut here ]------------ kernel BUG at fs/btrfs/relocation.c:1545! Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI CPU: 1 UID: 0 PID: 283 ... Tainted: 6.18.0+ #16 PREEMPT(voluntary) Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE Hardware name: QEMU Ubuntu 24.04 PC v2, BIOS 1.16.3-debian-1.16.3-2 RIP: 0010:merge_reloc_root+0x1266/0x1650 fs/btrfs/relocation.c:1545 Code: ffff0000 00004589 d7e9acfa ffffe8a1 79bafebe 02000000 Call Trace: merge_reloc_roots+0x295/0x890 fs/btrfs/relocation.c:1861 btrfs_recover_relocation+0xd6e/0x11d0 fs/btrfs/relocation.c:4195 btrfs_start_pre_rw_mount+0xa4d/0x1810 fs/btrfs/disk-io.c:3130 open_ctree+0x5824/0x5fe0 fs/btrfs/disk-io.c:3640 btrfs_fill_super fs/btrfs/super.c:987 [inline] btrfs_get_tree_super fs/btrfs/super.c:1951 [inline] btrfs_get_tree_subvol fs/btrfs/super.c:2094 [inline] btrfs_get_tree+0x111c/0x2190 fs/btrfs/super.c:2128 vfs_get_tree+0x9a/0x370 fs/super.c:1758 fc_mount fs/namespace.c:1199 [inline] do_new_mount_fc fs/namespace.c:3642 [inline] do_new_mount fs/namespace.c:3718 [inline] path_mount+0x5b8/0x1ea0 fs/namespace.c:4028 do_mount fs/namespace.c:4041 [inline] __do_sys_mount fs/namespace.c:4229 [inline] __se_sys_mount fs/namespace.c:4206 [inline] __x64_sys_mount+0x282/0x320 fs/namespace.c:4206 ... RIP: 0033:0x7f969c9a8fde Code: 0f1f4000 48c7c2b0 fffffff7 d8648902 b8ffffff ffc3660f ---[ end trace 0000000000000000 ]--- The bug is reproducible on 7.0.0-rc2-next-20260310 with our dynamic metadata fuzzing tool that corrupts btrfs metadata at runtime. [CAUSE] A non-zero drop_progress.objectid means an interrupted btrfs_drop_snapshot() left a resume point on disk, and in that case drop_level must be greater than 0 because the checkpoint is only saved at internal node levels. Although this invariant is enforced when the kernel writes the root item, it is not validated when the root item is read back from disk. That allows on-disk corruption to provide an invalid state with drop_progress.objectid != 0 and drop_level == 0. When relocation recovery later processes such a root item, merge_reloc_root() reads drop_level and hits BUG_ON(level == 0). The same invalid metadata can also trigger the corresponding BUG_ON() in btrfs_drop_snapshot(). [FIX] Fix this by validating the root_item invariant in tree-checker when reading root items from disk: if drop_progress.objectid is non-zero, drop_level must also be non-zero. Reject such malformed metadata with -EUCLEAN before it reaches merge_reloc_root() or btrfs_drop_snapshot() and triggers the BUG_ON. After the fix, the same corruption is correctly rejected by tree-checker and the BUG_ON is no longer triggered. Reviewed-by: Qu Wenruo Signed-off-by: ZhengYuan Huang Signed-off-by: David Sterba --- fs/btrfs/tree-checker.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 516ef62c8f43..b4e114efff45 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1288,6 +1288,23 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key, btrfs_root_drop_level(&ri), BTRFS_MAX_LEVEL - 1); return -EUCLEAN; } + /* + * If drop_progress.objectid is non-zero, a btrfs_drop_snapshot() was + * interrupted and the resume point was recorded in drop_progress and + * drop_level. In that case drop_level must be >= 1: level 0 is the + * leaf level and drop_snapshot never saves a checkpoint there (it + * only records checkpoints at internal node levels in DROP_REFERENCE + * stage). A zero drop_level combined with a non-zero drop_progress + * objectid indicates on-disk corruption and would cause a BUG_ON in + * merge_reloc_root() and btrfs_drop_snapshot() at mount time. + */ + if (unlikely(btrfs_disk_key_objectid(&ri.drop_progress) != 0 && + btrfs_root_drop_level(&ri) == 0)) { + generic_err(leaf, slot, + "invalid root drop_level 0 with non-zero drop_progress objectid %llu", + btrfs_disk_key_objectid(&ri.drop_progress)); + return -EUCLEAN; + } /* Flags check */ if (unlikely(btrfs_root_flags(&ri) & ~valid_root_flags)) { From d849a2f7309fc0616e79d13b008b0a47e0458b6e Mon Sep 17 00:00:00 2001 From: Paul Moses Date: Mon, 16 Mar 2026 14:56:51 +0000 Subject: [PATCH 125/352] xfrm: iptfs: only publish mode_data after clone setup iptfs_clone_state() stores x->mode_data before allocating the reorder window. If that allocation fails, the code frees the cloned state and returns -ENOMEM, leaving x->mode_data pointing at freed memory. The xfrm clone unwind later runs destroy_state() through x->mode_data, so the failed clone path tears down IPTFS state that clone_state() already freed. Keep the cloned IPTFS state private until all allocations succeed so failed clones leave x->mode_data unset. The destroy path already handles a NULL mode_data pointer. Fixes: 6be02e3e4f37 ("xfrm: iptfs: handle reordering of received packets") Cc: stable@vger.kernel.org Signed-off-by: Paul Moses Signed-off-by: Steffen Klassert --- net/xfrm/xfrm_iptfs.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/xfrm/xfrm_iptfs.c b/net/xfrm/xfrm_iptfs.c index 2c87290fe06c..7cd97c1dcd11 100644 --- a/net/xfrm/xfrm_iptfs.c +++ b/net/xfrm/xfrm_iptfs.c @@ -2664,9 +2664,6 @@ static int iptfs_clone_state(struct xfrm_state *x, struct xfrm_state *orig) if (!xtfs) return -ENOMEM; - x->mode_data = xtfs; - xtfs->x = x; - xtfs->ra_newskb = NULL; if (xtfs->cfg.reorder_win_size) { xtfs->w_saved = kcalloc(xtfs->cfg.reorder_win_size, @@ -2677,6 +2674,9 @@ static int iptfs_clone_state(struct xfrm_state *x, struct xfrm_state *orig) } } + x->mode_data = xtfs; + xtfs->x = x; + return 0; } From fe89277c9ceb0d6af0aa665bcf24a41d8b1b79cd Mon Sep 17 00:00:00 2001 From: Guanghui Feng Date: Mon, 16 Mar 2026 15:16:39 +0800 Subject: [PATCH 126/352] iommu/vt-d: Fix intel iommu iotlb sync hardlockup and retry During the qi_check_fault process after an IOMMU ITE event, requests at odd-numbered positions in the queue are set to QI_ABORT, only satisfying single-request submissions. However, qi_submit_sync now supports multiple simultaneous submissions, and can't guarantee that the wait_desc will be at an odd-numbered position. Therefore, if an item times out, IOMMU can't re-initiate the request, resulting in an infinite polling wait. This modifies the process by setting the status of all requests already fetched by IOMMU and recorded as QI_IN_USE status (including wait_desc requests) to QI_ABORT, thus enabling multiple requests to be resubmitted. Fixes: 8a1d82462540 ("iommu/vt-d: Multiple descriptors per qi_submit_sync()") Cc: stable@vger.kernel.org Signed-off-by: Guanghui Feng Tested-by: Shuai Xue Reviewed-by: Shuai Xue Reviewed-by: Samiullah Khawaja Link: https://lore.kernel.org/r/20260306101516.3885775-1-guanghuifeng@linux.alibaba.com Signed-off-by: Lu Baolu Fixes: 8a1d82462540 ("iommu/vt-d: Multiple descriptors per qi_submit_sync()") Signed-off-by: Joerg Roedel --- drivers/iommu/intel/dmar.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/iommu/intel/dmar.c b/drivers/iommu/intel/dmar.c index d68c06025cac..69222dbd2af0 100644 --- a/drivers/iommu/intel/dmar.c +++ b/drivers/iommu/intel/dmar.c @@ -1314,7 +1314,6 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index) if (fault & DMA_FSTS_ITE) { head = readl(iommu->reg + DMAR_IQH_REG); head = ((head >> shift) - 1 + QI_LENGTH) % QI_LENGTH; - head |= 1; tail = readl(iommu->reg + DMAR_IQT_REG); tail = ((tail >> shift) - 1 + QI_LENGTH) % QI_LENGTH; @@ -1331,7 +1330,7 @@ static int qi_check_fault(struct intel_iommu *iommu, int index, int wait_index) do { if (qi->desc_status[head] == QI_IN_USE) qi->desc_status[head] = QI_ABORT; - head = (head - 2 + QI_LENGTH) % QI_LENGTH; + head = (head - 1 + QI_LENGTH) % QI_LENGTH; } while (head != tail); /* From 39c20c4e83b9f78988541d829aa34668904e54a0 Mon Sep 17 00:00:00 2001 From: Lu Baolu Date: Mon, 16 Mar 2026 15:16:40 +0800 Subject: [PATCH 127/352] iommu/vt-d: Only handle IOPF for SVA when PRI is supported In intel_svm_set_dev_pasid(), the driver unconditionally manages the IOPF handling during a domain transition. However, commit a86fb7717320 ("iommu/vt-d: Allow SVA with device-specific IOPF") introduced support for SVA on devices that handle page faults internally without utilizing the PCI PRI. On such devices, the IOMMU-side IOPF infrastructure is not required. Calling iopf_for_domain_replace() on these devices is incorrect and can lead to unexpected failures during PASID attachment or unwinding. Add a check for info->pri_supported to ensure that the IOPF queue logic is only invoked for devices that actually rely on the IOMMU's PRI-based fault handling. Fixes: 17fce9d2336d ("iommu/vt-d: Put iopf enablement in domain attach path") Cc: stable@vger.kernel.org Suggested-by: Kevin Tian Reviewed-by: Kevin Tian Signed-off-by: Lu Baolu Link: https://lore.kernel.org/r/20260310075520.295104-1-baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel --- drivers/iommu/intel/svm.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/iommu/intel/svm.c b/drivers/iommu/intel/svm.c index fea10acd4f02..57cd1db7207a 100644 --- a/drivers/iommu/intel/svm.c +++ b/drivers/iommu/intel/svm.c @@ -164,9 +164,12 @@ static int intel_svm_set_dev_pasid(struct iommu_domain *domain, if (IS_ERR(dev_pasid)) return PTR_ERR(dev_pasid); - ret = iopf_for_domain_replace(domain, old, dev); - if (ret) - goto out_remove_dev_pasid; + /* SVA with non-IOMMU/PRI IOPF handling is allowed. */ + if (info->pri_supported) { + ret = iopf_for_domain_replace(domain, old, dev); + if (ret) + goto out_remove_dev_pasid; + } /* Setup the pasid table: */ sflags = cpu_feature_enabled(X86_FEATURE_LA57) ? PASID_FLAG_FL5LP : 0; @@ -181,7 +184,8 @@ static int intel_svm_set_dev_pasid(struct iommu_domain *domain, return 0; out_unwind_iopf: - iopf_for_domain_replace(old, domain, dev); + if (info->pri_supported) + iopf_for_domain_replace(old, domain, dev); out_remove_dev_pasid: domain_remove_dev_pasid(domain, dev, pasid); return ret; From 0a4d00e2e99a39a5698e4b63c394415dcbb39d90 Mon Sep 17 00:00:00 2001 From: Antheas Kapenekakis Date: Fri, 27 Feb 2026 09:06:37 +0100 Subject: [PATCH 128/352] iommu: Fix mapping check for 0x0 to avoid re-mapping it Commit 789a5913b29c ("iommu/amd: Use the generic iommu page table") introduces the shared iommu page table for AMD IOMMU. Some bioses contain an identity mapping for address 0x0, which is not parsed properly (e.g., certain Strix Halo devices). This causes the DMA components of the device to fail to initialize (e.g., the NVMe SSD controller), leading to a failed post. Specifically, on the GPD Win 5, the NVME and SSD GPU fail to mount, making collecting errors difficult. While debugging, it was found that a -EADDRINUSE error was emitted and its source was traced to iommu_iova_to_phys(). After adding some debug prints, it was found that phys_addr becomes 0, which causes the code to try to re-map the 0 address and fail, causing a cascade leading to a failed post. This is because the GPD Win 5 contains a 0x0-0x1 identity mapping for DMA devices, causing it to be repeated for each device. The cause of this failure is the following check in iommu_create_device_direct_mappings(), where address aliasing is handled via the following check: ``` phys_addr = iommu_iova_to_phys(domain, addr); if (!phys_addr) { map_size += pg_size; continue; } ```` Obviously, the iommu_iova_to_phys() signature is faulty and aliases unmapped and 0 together, causing the allocation code to try to re-allocate the 0 address per device. However, it has too many instantiations to fix. Therefore, use a ternary so that when addr is 0, the check is done for address 1 instead. Suggested-by: Robin Murphy Fixes: 789a5913b29c ("iommu/amd: Use the generic iommu page table") Signed-off-by: Antheas Kapenekakis Reviewed-by: Vasant Hegde Reviewed-by: Jason Gunthorpe Signed-off-by: Joerg Roedel --- drivers/iommu/iommu.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c index 35db51780954..50718ab810a4 100644 --- a/drivers/iommu/iommu.c +++ b/drivers/iommu/iommu.c @@ -1213,7 +1213,11 @@ static int iommu_create_device_direct_mappings(struct iommu_domain *domain, if (addr == end) goto map_end; - phys_addr = iommu_iova_to_phys(domain, addr); + /* + * Return address by iommu_iova_to_phys for 0 is + * ambiguous. Offset to address 1 if addr is 0. + */ + phys_addr = iommu_iova_to_phys(domain, addr ? addr : 1); if (!phys_addr) { map_size += pg_size; continue; From 45c6a2dc7ec8339052666b06065c521a10cc29bb Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Sun, 1 Mar 2026 16:52:14 -0800 Subject: [PATCH 129/352] iommu/io-pgtable: fix all kernel-doc warnings in io-pgtable.h Avoid kernel-doc warnings in io-pgtable.h: - use the correct struct member names or kernel-doc format - add a missing struct member description - add a missing function return comment section Warning: include/linux/io-pgtable.h:187 struct member 'coherent_walk' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:187 struct member 'arm_lpae_s1_cfg' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:187 struct member 'arm_lpae_s2_cfg' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:187 struct member 'arm_v7s_cfg' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:187 struct member 'arm_mali_lpae_cfg' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:187 struct member 'apple_dart_cfg' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:187 struct member 'amd' not described in 'io_pgtable_cfg' Warning: include/linux/io-pgtable.h:223 struct member 'read_and_clear_dirty' not described in 'io_pgtable_ops' Warning: include/linux/io-pgtable.h:237 No description found for return value of 'alloc_io_pgtable_ops' Signed-off-by: Randy Dunlap Reviewed-by: Jason Gunthorpe Signed-off-by: Joerg Roedel --- include/linux/io-pgtable.h | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/include/linux/io-pgtable.h b/include/linux/io-pgtable.h index 7a1516011ccf..e19872e37e06 100644 --- a/include/linux/io-pgtable.h +++ b/include/linux/io-pgtable.h @@ -53,7 +53,7 @@ struct iommu_flush_ops { * tables. * @ias: Input address (iova) size, in bits. * @oas: Output address (paddr) size, in bits. - * @coherent_walk A flag to indicate whether or not page table walks made + * @coherent_walk: A flag to indicate whether or not page table walks made * by the IOMMU are coherent with the CPU caches. * @tlb: TLB management callbacks for this set of tables. * @iommu_dev: The device representing the DMA configuration for the @@ -136,6 +136,7 @@ struct io_pgtable_cfg { void (*free)(void *cookie, void *pages, size_t size); /* Low-level data specific to the table format */ + /* private: */ union { struct { u64 ttbr; @@ -203,6 +204,9 @@ struct arm_lpae_io_pgtable_walk_data { * @unmap_pages: Unmap a range of virtually contiguous pages of the same size. * @iova_to_phys: Translate iova to physical address. * @pgtable_walk: (optional) Perform a page table walk for a given iova. + * @read_and_clear_dirty: Record dirty info per IOVA. If an IOVA is dirty, + * clear its dirty state from the PTE unless the + * IOMMU_DIRTY_NO_CLEAR flag is passed in. * * These functions map directly onto the iommu_ops member functions with * the same names. @@ -231,7 +235,9 @@ struct io_pgtable_ops { * the configuration actually provided by the allocator (e.g. the * pgsize_bitmap may be restricted). * @cookie: An opaque token provided by the IOMMU driver and passed back to - * the callback routines in cfg->tlb. + * the callback routines. + * + * Returns: Pointer to the &struct io_pgtable_ops for this set of page tables. */ struct io_pgtable_ops *alloc_io_pgtable_ops(enum io_pgtable_fmt fmt, struct io_pgtable_cfg *cfg, From 06e14c36e20b48171df13d51b89fe67c594ed07a Mon Sep 17 00:00:00 2001 From: Lizhi Hou Date: Wed, 4 Mar 2026 22:18:42 -0800 Subject: [PATCH 130/352] iommu/sva: Fix crash in iommu_sva_unbind_device() domain->mm->iommu_mm can be freed by iommu_domain_free(): iommu_domain_free() mmdrop() __mmdrop() mm_pasid_drop() After iommu_domain_free() returns, accessing domain->mm->iommu_mm may dereference a freed mm structure, leading to a crash. Fix this by moving the code that accesses domain->mm->iommu_mm to before the call to iommu_domain_free(). Fixes: e37d5a2d60a3 ("iommu/sva: invalidate stale IOTLB entries for kernel address space") Signed-off-by: Lizhi Hou Reviewed-by: Jason Gunthorpe Reviewed-by: Yi Liu Reviewed-by: Vasant Hegde Reviewed-by: Lu Baolu Signed-off-by: Joerg Roedel --- drivers/iommu/iommu-sva.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/iommu/iommu-sva.c b/drivers/iommu/iommu-sva.c index 07d64908a05f..bc7c7232a43e 100644 --- a/drivers/iommu/iommu-sva.c +++ b/drivers/iommu/iommu-sva.c @@ -182,13 +182,13 @@ void iommu_sva_unbind_device(struct iommu_sva *handle) iommu_detach_device_pasid(domain, dev, iommu_mm->pasid); if (--domain->users == 0) { list_del(&domain->next); - iommu_domain_free(domain); - } + if (list_empty(&iommu_mm->sva_domains)) { + list_del(&iommu_mm->mm_list_elm); + if (list_empty(&iommu_sva_mms)) + iommu_sva_present = false; + } - if (list_empty(&iommu_mm->sva_domains)) { - list_del(&iommu_mm->mm_list_elm); - if (list_empty(&iommu_sva_mms)) - iommu_sva_present = false; + iommu_domain_free(domain); } mutex_unlock(&iommu_sva_lock); From ba17de98545d07285d15ce4fe2afe98283338fb0 Mon Sep 17 00:00:00 2001 From: Joe Damato Date: Mon, 9 Mar 2026 16:52:33 -0700 Subject: [PATCH 131/352] iommu/amd: Block identity domain when SNP enabled Previously, commit 8388f7df936b ("iommu/amd: Do not support IOMMU_DOMAIN_IDENTITY after SNP is enabled") prevented users from changing the IOMMU domain to identity if SNP was enabled. This resulted in an error when writing to sysfs: # echo "identity" > /sys/kernel/iommu_groups/50/type -bash: echo: write error: Cannot allocate memory However, commit 4402f2627d30 ("iommu/amd: Implement global identity domain") changed the flow of the code, skipping the SNP guard and allowing users to change the IOMMU domain to identity after a machine has booted. Once the user does that, they will probably try to bind and the device/driver will start to do DMA which will trigger errors: iommu ivhd3: AMD-Vi: Event logged [ILLEGAL_DEV_TABLE_ENTRY device=0000:43:00.0 pasid=0x00000 address=0x3737b01000 flags=0x0020] iommu ivhd3: AMD-Vi: Control Reg : 0xc22000142148d AMD-Vi: DTE[0]: 6000000000000003 AMD-Vi: DTE[1]: 0000000000000001 AMD-Vi: DTE[2]: 2000003088b3e013 AMD-Vi: DTE[3]: 0000000000000000 bnxt_en 0000:43:00.0 (unnamed net_device) (uninitialized): Error (timeout: 500015) msg {0x0 0x0} len:0 iommu ivhd3: AMD-Vi: Event logged [ILLEGAL_DEV_TABLE_ENTRY device=0000:43:00.0 pasid=0x00000 address=0x3737b01000 flags=0x0020] iommu ivhd3: AMD-Vi: Control Reg : 0xc22000142148d AMD-Vi: DTE[0]: 6000000000000003 AMD-Vi: DTE[1]: 0000000000000001 AMD-Vi: DTE[2]: 2000003088b3e013 AMD-Vi: DTE[3]: 0000000000000000 bnxt_en 0000:43:00.0: probe with driver bnxt_en failed with error -16 To prevent this from happening, create an attach wrapper for identity_domain_ops which returns EINVAL if amd_iommu_snp_en is true. With this commit applied: # echo "identity" > /sys/kernel/iommu_groups/62/type -bash: echo: write error: Invalid argument Fixes: 4402f2627d30 ("iommu/amd: Implement global identity domain") Signed-off-by: Joe Damato Reviewed-by: Vasant Hegde Reviewed-by: Jason Gunthorpe Signed-off-by: Joerg Roedel --- drivers/iommu/amd/iommu.c | 15 ++++++++++++++- 1 file changed, 14 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c index 81c4d7733872..760d5f4623b5 100644 --- a/drivers/iommu/amd/iommu.c +++ b/drivers/iommu/amd/iommu.c @@ -2909,8 +2909,21 @@ static struct iommu_domain blocked_domain = { static struct protection_domain identity_domain; +static int amd_iommu_identity_attach(struct iommu_domain *dom, struct device *dev, + struct iommu_domain *old) +{ + /* + * Don't allow attaching a device to the identity domain if SNP is + * enabled. + */ + if (amd_iommu_snp_en) + return -EINVAL; + + return amd_iommu_attach_device(dom, dev, old); +} + static const struct iommu_domain_ops identity_domain_ops = { - .attach_dev = amd_iommu_attach_device, + .attach_dev = amd_iommu_identity_attach, }; void amd_iommu_init_identity_domain(void) From 0496acc42fb51eee040b5170cec05cec41385540 Mon Sep 17 00:00:00 2001 From: "Zenghui Yu (Huawei)" Date: Tue, 17 Mar 2026 19:57:48 +0800 Subject: [PATCH 132/352] KVM: arm64: Fix the descriptor address in __kvm_at_swap_desc() Using "(u64 __user *)hva + offset" to get the virtual addresses of S1/S2 descriptors looks really wrong, if offset is not zero. What we want to get for swapping is hva + offset, not hva + offset*8. ;-) Fix it. Fixes: f6927b41d573 ("KVM: arm64: Add helper for swapping guest descriptor") Signed-off-by: Zenghui Yu (Huawei) Link: https://patch.msgid.link/20260317115748.47332-1-zenghui.yu@linux.dev Signed-off-by: Marc Zyngier Cc: stable@vger.kernel.org --- arch/arm64/kvm/at.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm64/kvm/at.c b/arch/arm64/kvm/at.c index c5c5644b1878..a024d9a770dc 100644 --- a/arch/arm64/kvm/at.c +++ b/arch/arm64/kvm/at.c @@ -1753,7 +1753,7 @@ int __kvm_at_swap_desc(struct kvm *kvm, gpa_t ipa, u64 old, u64 new) if (!writable) return -EPERM; - ptep = (u64 __user *)hva + offset; + ptep = (void __user *)hva + offset; if (cpus_have_final_cap(ARM64_HAS_LSE_ATOMICS)) r = __lse_swap_desc(ptep, old, new); else From a55c2a5c8d680156495b7b1e2a9f5a3e313ba524 Mon Sep 17 00:00:00 2001 From: Alessio Belle Date: Mon, 9 Mar 2026 15:23:48 +0000 Subject: [PATCH 133/352] drm/imagination: Fix deadlock in soft reset sequence The soft reset sequence is currently executed from the threaded IRQ handler, hence it cannot call disable_irq() which internally waits for IRQ handlers, i.e. itself, to complete. Use disable_irq_nosync() during a soft reset instead. Fixes: cc1aeedb98ad ("drm/imagination: Implement firmware infrastructure and META FW support") Cc: stable@vger.kernel.org Signed-off-by: Alessio Belle Reviewed-by: Matt Coster Link: https://patch.msgid.link/20260309-fix-soft-reset-v1-1-121113be554f@imgtec.com Signed-off-by: Matt Coster --- drivers/gpu/drm/imagination/pvr_power.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/imagination/pvr_power.c b/drivers/gpu/drm/imagination/pvr_power.c index 0cf7393f89c6..ab10b4ac06b0 100644 --- a/drivers/gpu/drm/imagination/pvr_power.c +++ b/drivers/gpu/drm/imagination/pvr_power.c @@ -510,7 +510,16 @@ pvr_power_reset(struct pvr_device *pvr_dev, bool hard_reset) } /* Disable IRQs for the duration of the reset. */ - disable_irq(pvr_dev->irq); + if (hard_reset) { + disable_irq(pvr_dev->irq); + } else { + /* + * Soft reset is triggered as a response to a FW command to the Host and is + * processed from the threaded IRQ handler. This code cannot (nor needs to) + * wait for any IRQ processing to complete. + */ + disable_irq_nosync(pvr_dev->irq); + } do { if (hard_reset) { From 2d7f05cddf4c268cc36256a2476946041dbdd36d Mon Sep 17 00:00:00 2001 From: Alessio Belle Date: Tue, 10 Mar 2026 11:41:11 +0000 Subject: [PATCH 134/352] drm/imagination: Synchronize interrupts before suspending the GPU The runtime PM suspend callback doesn't know whether the IRQ handler is in progress on a different CPU core and doesn't wait for it to finish. Depending on timing, the IRQ handler could be running while the GPU is suspended, leading to kernel crashes when trying to access GPU registers. See example signature below. In a power off sequence initiated by the runtime PM suspend callback, wait for any IRQ handlers in progress on other CPU cores to finish, by calling synchronize_irq(). At the same time, remove the runtime PM resume/put calls in the threaded IRQ handler. On top of not being the right approach to begin with, and being at the wrong place as they should have wrapped all GPU register accesses, the driver would hit a deadlock between synchronize_irq() being called from a runtime PM suspend callback, holding the device power lock, and the resume callback requiring the same. Example crash signature on a TI AM68 SK platform: [ 337.241218] SError Interrupt on CPU0, code 0x00000000bf000000 -- SError [ 337.241239] CPU: 0 UID: 0 PID: 112 Comm: irq/234-gpu Tainted: G M 6.17.7-B2C-00005-g9c7bbe4ea16c #2 PREEMPT [ 337.241246] Tainted: [M]=MACHINE_CHECK [ 337.241249] Hardware name: Texas Instruments AM68 SK (DT) [ 337.241252] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 337.241256] pc : pvr_riscv_irq_pending+0xc/0x24 [ 337.241277] lr : pvr_device_irq_thread_handler+0x64/0x310 [ 337.241282] sp : ffff800085b0bd30 [ 337.241284] x29: ffff800085b0bd50 x28: ffff0008070d9eab x27: ffff800083a5ce10 [ 337.241291] x26: ffff000806e48f80 x25: ffff0008070d9eac x24: 0000000000000000 [ 337.241296] x23: ffff0008068e9bf0 x22: ffff0008068e9bd0 x21: ffff800085b0bd30 [ 337.241301] x20: ffff0008070d9e00 x19: ffff0008068e9000 x18: 0000000000000001 [ 337.241305] x17: 637365645f656c70 x16: 0000000000000000 x15: ffff000b7df9ff40 [ 337.241310] x14: 0000a585fe3c0d0e x13: 000000999704f060 x12: 000000000002771a [ 337.241314] x11: 00000000000000c0 x10: 0000000000000af0 x9 : ffff800085b0bd00 [ 337.241318] x8 : ffff0008071175d0 x7 : 000000000000b955 x6 : 0000000000000003 [ 337.241323] x5 : 0000000000000000 x4 : 0000000000000002 x3 : 0000000000000000 [ 337.241327] x2 : ffff800080e39d20 x1 : ffff800080e3fc48 x0 : 0000000000000000 [ 337.241333] Kernel panic - not syncing: Asynchronous SError Interrupt [ 337.241337] CPU: 0 UID: 0 PID: 112 Comm: irq/234-gpu Tainted: G M 6.17.7-B2C-00005-g9c7bbe4ea16c #2 PREEMPT [ 337.241342] Tainted: [M]=MACHINE_CHECK [ 337.241343] Hardware name: Texas Instruments AM68 SK (DT) [ 337.241345] Call trace: [ 337.241348] show_stack+0x18/0x24 (C) [ 337.241357] dump_stack_lvl+0x60/0x80 [ 337.241364] dump_stack+0x18/0x24 [ 337.241368] vpanic+0x124/0x2ec [ 337.241373] abort+0x0/0x4 [ 337.241377] add_taint+0x0/0xbc [ 337.241384] arm64_serror_panic+0x70/0x80 [ 337.241389] do_serror+0x3c/0x74 [ 337.241392] el1h_64_error_handler+0x30/0x48 [ 337.241400] el1h_64_error+0x6c/0x70 [ 337.241404] pvr_riscv_irq_pending+0xc/0x24 (P) [ 337.241410] irq_thread_fn+0x2c/0xb0 [ 337.241416] irq_thread+0x170/0x334 [ 337.241421] kthread+0x12c/0x210 [ 337.241428] ret_from_fork+0x10/0x20 [ 337.241434] SMP: stopping secondary CPUs [ 337.241451] Kernel Offset: disabled [ 337.241453] CPU features: 0x040000,02002800,20002001,0400421b [ 337.241456] Memory Limit: none [ 337.457921] ---[ end Kernel panic - not syncing: Asynchronous SError Interrupt ]--- Fixes: cc1aeedb98ad ("drm/imagination: Implement firmware infrastructure and META FW support") Fixes: 96822d38ff57 ("drm/imagination: Handle Rogue safety event IRQs") Cc: stable@vger.kernel.org # see patch description, needs adjustments for < 6.16 Signed-off-by: Alessio Belle Reviewed-by: Matt Coster Link: https://patch.msgid.link/20260310-drain-irqs-before-suspend-v1-1-bf4f9ed68e75@imgtec.com Signed-off-by: Matt Coster --- drivers/gpu/drm/imagination/pvr_device.c | 17 ----------------- drivers/gpu/drm/imagination/pvr_power.c | 11 ++++++++--- 2 files changed, 8 insertions(+), 20 deletions(-) diff --git a/drivers/gpu/drm/imagination/pvr_device.c b/drivers/gpu/drm/imagination/pvr_device.c index f58bb66a6327..dbb6f5a8ded1 100644 --- a/drivers/gpu/drm/imagination/pvr_device.c +++ b/drivers/gpu/drm/imagination/pvr_device.c @@ -225,29 +225,12 @@ static irqreturn_t pvr_device_irq_thread_handler(int irq, void *data) } if (pvr_dev->has_safety_events) { - int err; - - /* - * Ensure the GPU is powered on since some safety events (such - * as ECC faults) can happen outside of job submissions, which - * are otherwise the only time a power reference is held. - */ - err = pvr_power_get(pvr_dev); - if (err) { - drm_err_ratelimited(drm_dev, - "%s: could not take power reference (%d)\n", - __func__, err); - return ret; - } - while (pvr_device_safety_irq_pending(pvr_dev)) { pvr_device_safety_irq_clear(pvr_dev); pvr_device_handle_safety_events(pvr_dev); ret = IRQ_HANDLED; } - - pvr_power_put(pvr_dev); } return ret; diff --git a/drivers/gpu/drm/imagination/pvr_power.c b/drivers/gpu/drm/imagination/pvr_power.c index ab10b4ac06b0..cee4d16ac851 100644 --- a/drivers/gpu/drm/imagination/pvr_power.c +++ b/drivers/gpu/drm/imagination/pvr_power.c @@ -90,7 +90,7 @@ pvr_power_request_pwr_off(struct pvr_device *pvr_dev) } static int -pvr_power_fw_disable(struct pvr_device *pvr_dev, bool hard_reset) +pvr_power_fw_disable(struct pvr_device *pvr_dev, bool hard_reset, bool rpm_suspend) { if (!hard_reset) { int err; @@ -106,6 +106,11 @@ pvr_power_fw_disable(struct pvr_device *pvr_dev, bool hard_reset) return err; } + if (rpm_suspend) { + /* Wait for late processing of GPU or firmware IRQs in other cores */ + synchronize_irq(pvr_dev->irq); + } + return pvr_fw_stop(pvr_dev); } @@ -361,7 +366,7 @@ pvr_power_device_suspend(struct device *dev) return -EIO; if (pvr_dev->fw_dev.booted) { - err = pvr_power_fw_disable(pvr_dev, false); + err = pvr_power_fw_disable(pvr_dev, false, true); if (err) goto err_drm_dev_exit; } @@ -527,7 +532,7 @@ pvr_power_reset(struct pvr_device *pvr_dev, bool hard_reset) queues_disabled = true; } - err = pvr_power_fw_disable(pvr_dev, hard_reset); + err = pvr_power_fw_disable(pvr_dev, hard_reset, false); if (!err) { if (hard_reset) { pvr_dev->fw_dev.booted = false; From 74ef7844dd8c27d6b94ebc102bb4677edd3e7696 Mon Sep 17 00:00:00 2001 From: Alessio Belle Date: Tue, 10 Mar 2026 11:41:12 +0000 Subject: [PATCH 135/352] drm/imagination: Disable interrupts before suspending the GPU This is an additional safety layer to ensure no accesses to the GPU registers can be made while it is powered off. While we can disable IRQ generation from GPU, META firmware, MIPS firmware and for safety events, we cannot do the same for the RISC-V firmware. To keep a unified approach, once the firmware has completed its power off sequence, disable IRQs for the while GPU at the kernel level instead. Signed-off-by: Alessio Belle Reviewed-by: Matt Coster Link: https://patch.msgid.link/20260310-drain-irqs-before-suspend-v1-2-bf4f9ed68e75@imgtec.com Signed-off-by: Matt Coster --- drivers/gpu/drm/imagination/pvr_power.c | 33 +++++++++++++++++-------- 1 file changed, 23 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/imagination/pvr_power.c b/drivers/gpu/drm/imagination/pvr_power.c index cee4d16ac851..3ec4ec4276e4 100644 --- a/drivers/gpu/drm/imagination/pvr_power.c +++ b/drivers/gpu/drm/imagination/pvr_power.c @@ -92,9 +92,9 @@ pvr_power_request_pwr_off(struct pvr_device *pvr_dev) static int pvr_power_fw_disable(struct pvr_device *pvr_dev, bool hard_reset, bool rpm_suspend) { - if (!hard_reset) { - int err; + int err; + if (!hard_reset) { cancel_delayed_work_sync(&pvr_dev->watchdog.work); err = pvr_power_request_idle(pvr_dev); @@ -107,33 +107,46 @@ pvr_power_fw_disable(struct pvr_device *pvr_dev, bool hard_reset, bool rpm_suspe } if (rpm_suspend) { - /* Wait for late processing of GPU or firmware IRQs in other cores */ - synchronize_irq(pvr_dev->irq); + /* This also waits for late processing of GPU or firmware IRQs in other cores */ + disable_irq(pvr_dev->irq); } - return pvr_fw_stop(pvr_dev); + err = pvr_fw_stop(pvr_dev); + if (err && rpm_suspend) + enable_irq(pvr_dev->irq); + + return err; } static int -pvr_power_fw_enable(struct pvr_device *pvr_dev) +pvr_power_fw_enable(struct pvr_device *pvr_dev, bool rpm_resume) { int err; + if (rpm_resume) + enable_irq(pvr_dev->irq); + err = pvr_fw_start(pvr_dev); if (err) - return err; + goto out; err = pvr_wait_for_fw_boot(pvr_dev); if (err) { drm_err(from_pvr_device(pvr_dev), "Firmware failed to boot\n"); pvr_fw_stop(pvr_dev); - return err; + goto out; } queue_delayed_work(pvr_dev->sched_wq, &pvr_dev->watchdog.work, msecs_to_jiffies(WATCHDOG_TIME_MS)); return 0; + +out: + if (rpm_resume) + disable_irq(pvr_dev->irq); + + return err; } bool @@ -396,7 +409,7 @@ pvr_power_device_resume(struct device *dev) goto err_drm_dev_exit; if (pvr_dev->fw_dev.booted) { - err = pvr_power_fw_enable(pvr_dev); + err = pvr_power_fw_enable(pvr_dev, true); if (err) goto err_power_off; } @@ -555,7 +568,7 @@ pvr_power_reset(struct pvr_device *pvr_dev, bool hard_reset) pvr_fw_irq_clear(pvr_dev); - err = pvr_power_fw_enable(pvr_dev); + err = pvr_power_fw_enable(pvr_dev, false); } if (err && hard_reset) From 67253b28a61f0dff31f8f00dca8c9586f089b852 Mon Sep 17 00:00:00 2001 From: Jonathan Cavitt Date: Mon, 16 Mar 2026 15:15:56 +0000 Subject: [PATCH 136/352] drm/pagemap_util: Ensure proper cache lock management on free MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit For the sake of consistency, ensure that the cache lock is always unlocked after drm_pagemap_cache_fini. Spinlocks typically disable preemption and if the code-path missing the unlock is hit, preemption will remain disabled even if the lock is subsequently freed. Fixes static analysis issue. v2: - Use requested code flow (Maarten) v3: - Clear cache->dpagemap (Matt Brost, Maarten) v4: - Reword commit message (Thomas) Fixes: 77f14f2f2d73f ("drm/pagemap: Add a drm_pagemap cache and shrinker") Signed-off-by: Jonathan Cavitt Reviewed-by: Thomas Hellström Reviewed-by: Maarten Lankhorst Cc: Thomas Hellstrom Cc: Matthew Brost Cc: Maarten Lankhorst Signed-off-by: Thomas Hellström Link: https://patch.msgid.link/20260316151555.7553-2-jonathan.cavitt@intel.com --- drivers/gpu/drm/drm_pagemap_util.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/drm_pagemap_util.c b/drivers/gpu/drm/drm_pagemap_util.c index 14ddb948a32e..6111d90a38e2 100644 --- a/drivers/gpu/drm/drm_pagemap_util.c +++ b/drivers/gpu/drm/drm_pagemap_util.c @@ -65,18 +65,14 @@ static void drm_pagemap_cache_fini(void *arg) drm_dbg(cache->shrinker->drm, "Destroying dpagemap cache.\n"); spin_lock(&cache->lock); dpagemap = cache->dpagemap; - if (!dpagemap) { - spin_unlock(&cache->lock); - goto out; - } + cache->dpagemap = NULL; + if (dpagemap && !drm_pagemap_shrinker_cancel(dpagemap)) + dpagemap = NULL; + spin_unlock(&cache->lock); - if (drm_pagemap_shrinker_cancel(dpagemap)) { - cache->dpagemap = NULL; - spin_unlock(&cache->lock); + if (dpagemap) drm_pagemap_destroy(dpagemap, false); - } -out: mutex_destroy(&cache->lookup_mutex); kfree(cache); } From 21647677ba9af2cb6bc460e17d9f29a7132c40c3 Mon Sep 17 00:00:00 2001 From: Koichiro Den Date: Mon, 16 Mar 2026 23:02:25 +0900 Subject: [PATCH 137/352] PCI: endpoint: pci-epf-test: Roll back BAR mapping when subrange setup fails When the BAR subrange mapping test on DWC-based platforms fails due to insufficient free inbound iATU regions, pci_epf_test_bar_subrange_setup() returns an error (-ENOSPC) but does not restore the original BAR mapping. This causes subsequent test runs to become confusing, since the failure may leave room for the next subrange mapping test to pass. Fix this by restoring the original BAR mapping when preparation of the subrange mapping fails, so that no side effect remains regardless of the test success or failure. Fixes: 6c5e6101423b ("PCI: endpoint: pci-epf-test: Add BAR subrange mapping test support") Reported-by: Christian Bruel Closes: https://lore.kernel.org/linux-pci/b2b03ebe-9482-4a13-b22f-7b44da096eed@foss.st.com/ Signed-off-by: Koichiro Den Signed-off-by: Bjorn Helgaas Tested-by: Christian Bruel Reviewed-by: Niklas Cassel Reviewed-by: Manivannan Sadhasivam Link: https://patch.msgid.link/20260316140225.1481658-1-den@valinux.co.jp --- drivers/pci/endpoint/functions/pci-epf-test.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/pci/endpoint/functions/pci-epf-test.c b/drivers/pci/endpoint/functions/pci-epf-test.c index 582938b7b4f1..33548935765e 100644 --- a/drivers/pci/endpoint/functions/pci-epf-test.c +++ b/drivers/pci/endpoint/functions/pci-epf-test.c @@ -894,6 +894,11 @@ static void pci_epf_test_bar_subrange_setup(struct pci_epf_test *epf_test, dev_err(&epf->dev, "pci_epc_set_bar() failed: %d\n", ret); bar->submap = old_submap; bar->num_submap = old_nsub; + ret = pci_epc_set_bar(epc, epf->func_no, epf->vfunc_no, bar); + if (ret) + dev_warn(&epf->dev, "Failed to restore the original BAR mapping: %d\n", + ret); + kfree(submap); goto err; } From 4221f30e3e0a2507641b3397d21aff9e71e749f8 Mon Sep 17 00:00:00 2001 From: Hugo Villeneuve Date: Tue, 17 Mar 2026 11:23:39 -0400 Subject: [PATCH 138/352] regulator: dt-bindings: fix typos in regulator-uv-* descriptions Remove word "over". Signed-off-by: Hugo Villeneuve Link: https://patch.msgid.link/20260317152357.3473584-1-hugo@hugovil.com Signed-off-by: Mark Brown --- Documentation/devicetree/bindings/regulator/regulator.yaml | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/Documentation/devicetree/bindings/regulator/regulator.yaml b/Documentation/devicetree/bindings/regulator/regulator.yaml index 042e56396399..019aeb664cae 100644 --- a/Documentation/devicetree/bindings/regulator/regulator.yaml +++ b/Documentation/devicetree/bindings/regulator/regulator.yaml @@ -168,7 +168,7 @@ properties: offset from voltage set to regulator. regulator-uv-protection-microvolt: - description: Set over under voltage protection limit. This is a limit where + description: Set under voltage protection limit. This is a limit where hardware performs emergency shutdown. Zero can be passed to disable protection and value '1' indicates that protection should be enabled but limit setting can be omitted. Limit is given as microvolt offset from @@ -182,7 +182,7 @@ properties: is given as microvolt offset from voltage set to regulator. regulator-uv-warn-microvolt: - description: Set over under voltage warning limit. This is a limit where + description: Set under voltage warning limit. This is a limit where hardware is assumed still to be functional but approaching limit where it gets damaged. Recovery actions should be initiated. Zero can be passed to disable detection and value '1' indicates that detection should From b49814033cb5224c818cfb04dccb3260da10cc4f Mon Sep 17 00:00:00 2001 From: Alex Hung Date: Wed, 11 Mar 2026 15:18:37 -0600 Subject: [PATCH 139/352] drm/amd/display: Fix gamma 2.2 colorop TFs Use GAMMA22 for degamma/blend and GAMMA22_INV for shaper so curves match the color pipeline. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/5016 Tested-by: Xaver Hugl Reviewed-by: Melissa Wen Reviewed-by: Harry Wentland Signed-off-by: Alex Hung Signed-off-by: Alex Deucher (cherry picked from commit d8f9f42effd767ffa7bbcd7e05fbd6b20737e468) --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c index d59ba82d3d7c..aa4658867e55 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_colorop.c @@ -37,19 +37,19 @@ const u64 amdgpu_dm_supported_degam_tfs = BIT(DRM_COLOROP_1D_CURVE_SRGB_EOTF) | BIT(DRM_COLOROP_1D_CURVE_PQ_125_EOTF) | BIT(DRM_COLOROP_1D_CURVE_BT2020_INV_OETF) | - BIT(DRM_COLOROP_1D_CURVE_GAMMA22_INV); + BIT(DRM_COLOROP_1D_CURVE_GAMMA22); const u64 amdgpu_dm_supported_shaper_tfs = BIT(DRM_COLOROP_1D_CURVE_SRGB_INV_EOTF) | BIT(DRM_COLOROP_1D_CURVE_PQ_125_INV_EOTF) | BIT(DRM_COLOROP_1D_CURVE_BT2020_OETF) | - BIT(DRM_COLOROP_1D_CURVE_GAMMA22); + BIT(DRM_COLOROP_1D_CURVE_GAMMA22_INV); const u64 amdgpu_dm_supported_blnd_tfs = BIT(DRM_COLOROP_1D_CURVE_SRGB_EOTF) | BIT(DRM_COLOROP_1D_CURVE_PQ_125_EOTF) | BIT(DRM_COLOROP_1D_CURVE_BT2020_INV_OETF) | - BIT(DRM_COLOROP_1D_CURVE_GAMMA22_INV); + BIT(DRM_COLOROP_1D_CURVE_GAMMA22); #define MAX_COLOR_PIPELINE_OPS 10 From 6270b1a5dab94665d7adce3dc78bc9066ed28bdd Mon Sep 17 00:00:00 2001 From: "Jesse.Zhang" Date: Thu, 12 Mar 2026 18:06:17 +0800 Subject: [PATCH 140/352] drm/amdgpu: Limit BO list entry count to prevent resource exhaustion MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Userspace can pass an arbitrary number of BO list entries via the bo_number field. Although the previous multiplication overflow check prevents out-of-bounds allocation, a large number of entries could still cause excessive memory allocation (up to potentially gigabytes) and unnecessarily long list processing times. Introduce a hard limit of 128k entries per BO list, which is more than sufficient for any realistic use case (e.g., a single list containing all buffers in a large scene). This prevents memory exhaustion attacks and ensures predictable performance. Return -EINVAL if the requested entry count exceeds the limit Reviewed-by: Christian König Suggested-by: Christian König Signed-off-by: Jesse Zhang Signed-off-by: Alex Deucher (cherry picked from commit 688b87d39e0aa8135105b40dc167d74b5ada5332) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c index 4662bfbe70b2..43864df8af04 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c @@ -36,6 +36,7 @@ #define AMDGPU_BO_LIST_MAX_PRIORITY 32u #define AMDGPU_BO_LIST_NUM_BUCKETS (AMDGPU_BO_LIST_MAX_PRIORITY + 1) +#define AMDGPU_BO_LIST_MAX_ENTRIES (128 * 1024) static void amdgpu_bo_list_free_rcu(struct rcu_head *rcu) { @@ -188,6 +189,9 @@ int amdgpu_bo_create_list_entry_array(struct drm_amdgpu_bo_list_in *in, const uint32_t bo_number = in->bo_number; struct drm_amdgpu_bo_list_entry *info; + if (bo_number > AMDGPU_BO_LIST_MAX_ENTRIES) + return -EINVAL; + /* copy the handle array from userspace to a kernel buffer */ if (likely(info_size == bo_info_size)) { info = vmemdup_array_user(uptr, bo_number, info_size); From 10718159890bc99cbcc7b5a38dade05df335e797 Mon Sep 17 00:00:00 2001 From: Calvin Owens Date: Thu, 12 Mar 2026 10:13:34 -0700 Subject: [PATCH 141/352] drm/amd/display: Fix uninitialized variable use which breaks full LTO Commit e1b385726f7f ("drm/amd/display: Add additional checks for PSP footer size") introduced a use of an uninitialized stack variable in dm_dmub_sw_init() (region_params.bss_data_size). Interestingly, this seems to cause no issue on normal kernels. But when full LTO is enabled, it causes the compiler to "optimize" out huge swaths of amdgpu initialization code, and the driver is unusable: amdgpu 0000:03:00.0: [drm] Loading DMUB firmware via PSP: version=0x07002F00 amdgpu 0000:03:00.0: sw_init of IP block failed 5 amdgpu 0000:03:00.0: amdgpu_device_ip_init failed amdgpu 0000:03:00.0: Fatal error during GPU init It surprises me that neither gcc nor clang emit a warning about this: I only found it by bisecting the LTO breakage. Fix by using the bss_data_size field from fw_meta_info_params, as was presumably intended. Fixes: e1b385726f7f ("drm/amd/display: Add additional checks for PSP footer size") Signed-off-by: Calvin Owens Reviewed-by: Harry Wentland Reviewed-by: Nathan Chancellor Signed-off-by: Alex Deucher (cherry picked from commit b7f1402f6ad24cc6b9a01fa09ebd1c6559d787d0) --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index b3d6f2cd8ab6..0d1c772ef713 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -2554,7 +2554,7 @@ static int dm_dmub_sw_init(struct amdgpu_device *adev) fw_meta_info_params.fw_inst_const = adev->dm.dmub_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + PSP_HEADER_BYTES_256; - fw_meta_info_params.fw_bss_data = region_params.bss_data_size ? adev->dm.dmub_fw->data + + fw_meta_info_params.fw_bss_data = fw_meta_info_params.bss_data_size ? adev->dm.dmub_fw->data + le32_to_cpu(hdr->header.ucode_array_offset_bytes) + le32_to_cpu(hdr->inst_const_bytes) : NULL; fw_meta_info_params.custom_psp_footer_size = 0; From ebe82c6e75cfc547154d0fd843b0dd6cca3d548f Mon Sep 17 00:00:00 2001 From: Xi Ruoyao Date: Fri, 6 Mar 2026 14:28:03 +0800 Subject: [PATCH 142/352] drm/amd/display: Wrap dcn32_override_min_req_memclk() in DC_FP_{START, END} [Why] The dcn32_override_min_req_memclk function is in dcn32_fpu.c, which is compiled with CC_FLAGS_FPU into FP instructions. So when we call it we must use DC_FP_{START,END} to save and restore the FP context, and prepare the FP unit on architectures like LoongArch where the FP unit isn't always on. Reported-by: LiarOnce Fixes: ee7be8f3de1c ("drm/amd/display: Limit DCN32 8 channel or less parts to DPM1 for FPO") Signed-off-by: Xi Ruoyao Reviewed-by: Alex Hung Signed-off-by: Alex Deucher (cherry picked from commit 25bb1d54ba3983c064361033a8ec15474fece37e) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c index 7ebb7d1193af..c7fd604024d6 100644 --- a/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c +++ b/drivers/gpu/drm/amd/display/dc/resource/dcn32/dcn32_resource.c @@ -1785,7 +1785,10 @@ static bool dml1_validate(struct dc *dc, struct dc_state *context, enum dc_valid dc->res_pool->funcs->calculate_wm_and_dlg(dc, context, pipes, pipe_cnt, vlevel); + DC_FP_START(); dcn32_override_min_req_memclk(dc, context); + DC_FP_END(); + dcn32_override_min_req_dcfclk(dc, context); BW_VAL_TRACE_END_WATERMARKS(); From 2323b019651ad81c20a0f7f817c63392b3110652 Mon Sep 17 00:00:00 2001 From: Srinivasan Shanmugam Date: Sun, 15 Mar 2026 18:30:26 +0530 Subject: [PATCH 143/352] drm/amd/display: Fix DisplayID not-found handling in parse_edid_displayid_vrr() parse_edid_displayid_vrr() searches the EDID extension blocks for a DisplayID extension before parsing the dynamic video timing range. The code previously checked whether edid_ext was NULL after the search loop. However, edid_ext is assigned during each iteration of the loop, so it will never be NULL once the loop has executed. If no DisplayID extension is found, edid_ext ends up pointing to the last extension block, and the NULL check does not correctly detect the failure case. Instead, check whether the loop completed without finding a matching DisplayID block by testing "i == edid->extensions". This ensures the function exits early when no DisplayID extension is present and avoids parsing an unrelated EDID extension block. Also simplify the EDID validation check using "!edid || !edid->extensions". Fixes the below: drivers/gpu/drm/amd/amdgpu/../display/amdgpu_dm/amdgpu_dm.c:13079 parse_edid_displayid_vrr() warn: variable dereferenced before check 'edid_ext' (see line 13075) Fixes: a638b837d0e6 ("drm/amd/display: Fix refresh rate range for some panel") Cc: Roman Li Cc: Alex Hung Cc: Jerry Zuo Cc: Sun peng Li Cc: Tom Chung Cc: Dan Carpenter Cc: Aurabindo Pillai Signed-off-by: Srinivasan Shanmugam Reviewed-by: Tom Chung Signed-off-by: Alex Deucher (cherry picked from commit 91c7e6342e98c846b259c57273436fdea4c043f2) --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 0d1c772ef713..085cc98bd875 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -13119,7 +13119,7 @@ static void parse_edid_displayid_vrr(struct drm_connector *connector, u16 min_vfreq; u16 max_vfreq; - if (edid == NULL || edid->extensions == 0) + if (!edid || !edid->extensions) return; /* Find DisplayID extension */ @@ -13129,7 +13129,7 @@ static void parse_edid_displayid_vrr(struct drm_connector *connector, break; } - if (edid_ext == NULL) + if (i == edid->extensions) return; while (j < EDID_LENGTH) { From 39f44f54afa58661ecae9c27e15f5dbce2372892 Mon Sep 17 00:00:00 2001 From: Andy Nguyen Date: Sun, 15 Mar 2026 17:51:47 +0100 Subject: [PATCH 144/352] drm/amd: fix dcn 2.01 check The ASICREV_IS_BEIGE_GOBY_P check always took precedence, because it includes all chip revisions upto NV_UNKNOWN. Fixes: 54b822b3eac3 ("drm/amd/display: Use dce_version instead of chip_id") Signed-off-by: Andy Nguyen Signed-off-by: Alex Deucher (cherry picked from commit 9c7be0efa6f0daa949a5f3e3fdf9ea090b0713cb) --- drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c index 08d0e05a313e..d237d7b41dfd 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/clk_mgr.c @@ -255,6 +255,10 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p BREAK_TO_DEBUGGER(); return NULL; } + if (ctx->dce_version == DCN_VERSION_2_01) { + dcn201_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); + return &clk_mgr->base; + } if (ASICREV_IS_SIENNA_CICHLID_P(asic_id.hw_internal_rev)) { dcn3_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); return &clk_mgr->base; @@ -267,10 +271,6 @@ struct clk_mgr *dc_clk_mgr_create(struct dc_context *ctx, struct pp_smu_funcs *p dcn3_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); return &clk_mgr->base; } - if (ctx->dce_version == DCN_VERSION_2_01) { - dcn201_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); - return &clk_mgr->base; - } dcn20_clk_mgr_construct(ctx, clk_mgr, pp_smu, dccg); return &clk_mgr->base; } From 0b26edac4ac5535df1f63e6e8ab44c24fe1acad7 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:22:43 -0500 Subject: [PATCH 145/352] drm/amdgpu/mmhub2.0: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit e064cef4b53552602bb6ac90399c18f662f3cacd) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index a0cc8e218ca1..534cb4c544dc 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -154,14 +154,17 @@ mmhub_v2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) { case IP_VERSION(2, 0, 0): case IP_VERSION(2, 0, 2): - mmhub_cid = mmhub_client_ids_navi1x[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_navi1x) ? + mmhub_client_ids_navi1x[cid][rw] : NULL; break; case IP_VERSION(2, 1, 0): case IP_VERSION(2, 1, 1): - mmhub_cid = mmhub_client_ids_sienna_cichlid[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_sienna_cichlid) ? + mmhub_client_ids_sienna_cichlid[cid][rw] : NULL; break; case IP_VERSION(2, 1, 2): - mmhub_cid = mmhub_client_ids_beige_goby[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_beige_goby) ? + mmhub_client_ids_beige_goby[cid][rw] : NULL; break; default: mmhub_cid = NULL; From a54403a534972af5d9ba5aaa3bb6ead612500ec6 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:24:10 -0500 Subject: [PATCH 146/352] drm/amdgpu/mmhub2.3: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit 89cd90375c19fb45138990b70e9f4ba4806f05c4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c index 5eb8122e2746..ceb2f6b46de5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c @@ -94,7 +94,8 @@ mmhub_v2_3_print_l2_protection_fault_status(struct amdgpu_device *adev, case IP_VERSION(2, 3, 0): case IP_VERSION(2, 4, 0): case IP_VERSION(2, 4, 1): - mmhub_cid = mmhub_client_ids_vangogh[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_vangogh) ? + mmhub_client_ids_vangogh[cid][rw] : NULL; break; default: mmhub_cid = NULL; From 5d4e88bcfef29569a1db224ef15e28c603666c6d Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:24:35 -0500 Subject: [PATCH 147/352] drm/amdgpu/mmhub3.0.1: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit 5f76083183363c4528a4aaa593f5d38c28fe7d7b) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c index 910337dc28d1..14a742d3a99d 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_1.c @@ -117,7 +117,8 @@ mmhub_v3_0_1_print_l2_protection_fault_status(struct amdgpu_device *adev, switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) { case IP_VERSION(3, 0, 1): - mmhub_cid = mmhub_client_ids_v3_0_1[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v3_0_1) ? + mmhub_client_ids_v3_0_1[cid][rw] : NULL; break; default: mmhub_cid = NULL; From e5e6d67b1ce9764e67aef2d0eef9911af53ad99a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:25:09 -0500 Subject: [PATCH 148/352] drm/amdgpu/mmhub3.0.2: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit 1441f52c7f6ae6553664aa9e3e4562f6fc2fe8ea) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c index f0f182f033b9..e1f07f2a1852 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0_2.c @@ -108,7 +108,8 @@ mmhub_v3_0_2_print_l2_protection_fault_status(struct amdgpu_device *adev, "MMVM_L2_PROTECTION_FAULT_STATUS:0x%08X\n", status); - mmhub_cid = mmhub_client_ids_v3_0_2[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v3_0_2) ? + mmhub_client_ids_v3_0_2[cid][rw] : NULL; dev_err(adev->dev, "\t Faulty UTCL2 client ID: %s (0x%x)\n", mmhub_cid ? mmhub_cid : "unknown", cid); dev_err(adev->dev, "\t MORE_FAULTS: 0x%lx\n", From cdb82ecbeccb55fae75a3c956b605f7801a30db1 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:25:30 -0500 Subject: [PATCH 149/352] drm/amdgpu/mmhub3.0: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit f14f27bbe2a3ed7af32d5f6eaf3f417139f45253) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c index 7d5242df58a5..ab966e69a342 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v3_0.c @@ -110,7 +110,8 @@ mmhub_v3_0_print_l2_protection_fault_status(struct amdgpu_device *adev, switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) { case IP_VERSION(3, 0, 0): case IP_VERSION(3, 0, 1): - mmhub_cid = mmhub_client_ids_v3_0_0[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v3_0_0) ? + mmhub_client_ids_v3_0_0[cid][rw] : NULL; break; default: mmhub_cid = NULL; From 3cdd405831d8cc50a5eae086403402697bb98a4a Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:25:56 -0500 Subject: [PATCH 150/352] drm/amdgpu/mmhub4.1.0: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit 04f063d85090f5dd0c671010ce88ee49d9dcc8ed) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v4_1_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_1_0.c index 951998454b25..88bfe321f83a 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_1_0.c @@ -102,7 +102,8 @@ mmhub_v4_1_0_print_l2_protection_fault_status(struct amdgpu_device *adev, status); switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) { case IP_VERSION(4, 1, 0): - mmhub_cid = mmhub_client_ids_v4_1_0[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v4_1_0) ? + mmhub_client_ids_v4_1_0[cid][rw] : NULL; break; default: mmhub_cid = NULL; From 9c52f49545478aa47769378cd0b53c5005d6a846 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 4 Mar 2026 17:26:17 -0500 Subject: [PATCH 151/352] drm/amdgpu/mmhub4.2.0: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit dea5f235baf3786bfd4fd920b03c19285fdc3d9f) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c index a72770e3d0e9..2532ca80f735 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v4_2_0.c @@ -688,7 +688,8 @@ mmhub_v4_2_0_print_l2_protection_fault_status(struct amdgpu_device *adev, status); switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) { case IP_VERSION(4, 2, 0): - mmhub_cid = mmhub_client_ids_v4_2_0[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_v4_2_0) ? + mmhub_client_ids_v4_2_0[cid][rw] : NULL; break; default: mmhub_cid = NULL; From f39e1270277f4b06db0b2c6ec9405b6dd766fb13 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 16 Mar 2026 15:51:08 -0400 Subject: [PATCH 152/352] drm/amdgpu/gmc9.0: add bounds checking for cid The value should never exceed the array size as those are the only values the hardware is expected to return, but add checks anyway. Cc: Benjamin Cheng Reviewed-by: Benjamin Cheng Signed-off-by: Alex Deucher (cherry picked from commit e14d468304832bcc4a082d95849bc0a41b18ddea) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index e35ed0cc2ec6..8eba99aa0f8f 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -662,28 +662,35 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, } else { switch (amdgpu_ip_version(adev, MMHUB_HWIP, 0)) { case IP_VERSION(9, 0, 0): - mmhub_cid = mmhub_client_ids_vega10[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_vega10) ? + mmhub_client_ids_vega10[cid][rw] : NULL; break; case IP_VERSION(9, 3, 0): - mmhub_cid = mmhub_client_ids_vega12[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_vega12) ? + mmhub_client_ids_vega12[cid][rw] : NULL; break; case IP_VERSION(9, 4, 0): - mmhub_cid = mmhub_client_ids_vega20[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_vega20) ? + mmhub_client_ids_vega20[cid][rw] : NULL; break; case IP_VERSION(9, 4, 1): - mmhub_cid = mmhub_client_ids_arcturus[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_arcturus) ? + mmhub_client_ids_arcturus[cid][rw] : NULL; break; case IP_VERSION(9, 1, 0): case IP_VERSION(9, 2, 0): - mmhub_cid = mmhub_client_ids_raven[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_raven) ? + mmhub_client_ids_raven[cid][rw] : NULL; break; case IP_VERSION(1, 5, 0): case IP_VERSION(2, 4, 0): - mmhub_cid = mmhub_client_ids_renoir[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_renoir) ? + mmhub_client_ids_renoir[cid][rw] : NULL; break; case IP_VERSION(1, 8, 0): case IP_VERSION(9, 4, 2): - mmhub_cid = mmhub_client_ids_aldebaran[cid][rw]; + mmhub_cid = cid < ARRAY_SIZE(mmhub_client_ids_aldebaran) ? + mmhub_client_ids_aldebaran[cid][rw] : NULL; break; default: mmhub_cid = NULL; From 3fc4648b53b7e393b91e63600e28e6f25c8ef0c5 Mon Sep 17 00:00:00 2001 From: Pratap Nirujogi Date: Wed, 11 Mar 2026 12:15:09 -0400 Subject: [PATCH 153/352] drm/amdgpu: Fix ISP segfault issue in kernel v7.0 Add NULL pointer checks for dev->type before accessing dev->type->name in ISP genpd add/remove functions to prevent kernel crashes. This regression was introduced in v7.0 as the wakeup sources are registered using physical device instead of ACPI device. This led to adding wakeup source device as the first child of AMDGPU device without initializing dev-type variable, and resulted in segfault when accessed it in the amdgpu isp driver. Fixes: 057edc58aa59 ("ACPI: PM: Register wakeup sources under physical devices") Suggested-by: Bin Du Reviewed-by: Mario Limonciello Signed-off-by: Pratap Nirujogi Signed-off-by: Alex Deucher (cherry picked from commit c51632d1ed7ac5aed2d40dbc0718d75342c12c6a) --- drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c index b3590b33cab9..485ecdec9618 100644 --- a/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c +++ b/drivers/gpu/drm/amd/amdgpu/isp_v4_1_1.c @@ -129,7 +129,7 @@ static int isp_genpd_add_device(struct device *dev, void *data) if (!pdev) return -EINVAL; - if (!dev->type->name) { + if (!dev->type || !dev->type->name) { drm_dbg(&adev->ddev, "Invalid device type to add\n"); goto exit; } @@ -165,7 +165,7 @@ static int isp_genpd_remove_device(struct device *dev, void *data) if (!pdev) return -EINVAL; - if (!dev->type->name) { + if (!dev->type || !dev->type->name) { drm_dbg(&adev->ddev, "Invalid device type to remove\n"); goto exit; } From 6bee098b91417654703e17eb5c1822c6dfd0c01d Mon Sep 17 00:00:00 2001 From: Maarten Lankhorst Date: Fri, 13 Mar 2026 16:17:27 +0100 Subject: [PATCH 154/352] drm: Fix use-after-free on framebuffers and property blobs when calling drm_dev_unplug MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When trying to do a rather aggressive test of igt's "xe_module_load --r reload" with a full desktop environment and game running I noticed a few OOPSes when dereferencing freed pointers, related to framebuffers and property blobs after the compositor exits. Solve this by guarding the freeing in drm_file with drm_dev_enter/exit, and immediately put the references from struct drm_file objects during drm_dev_unplug(). Related warnings for framebuffers on the subtest: [ 739.713076] ------------[ cut here ]------------ WARN_ON(!list_empty(&dev->mode_config.fb_list)) [ 739.713079] WARNING: drivers/gpu/drm/drm_mode_config.c:584 at drm_mode_config_cleanup+0x30b/0x320 [drm], CPU#12: xe_module_load/13145 .... [ 739.713328] Call Trace: [ 739.713330] [ 739.713335] ? intel_pmdemand_destroy_state+0x11/0x20 [xe] [ 739.713574] ? intel_atomic_global_obj_cleanup+0xe4/0x1a0 [xe] [ 739.713794] intel_display_driver_remove_noirq+0x51/0xb0 [xe] [ 739.714041] xe_display_fini_early+0x33/0x50 [xe] [ 739.714284] devm_action_release+0xf/0x20 [ 739.714294] devres_release_all+0xad/0xf0 [ 739.714301] device_unbind_cleanup+0x12/0xa0 [ 739.714305] device_release_driver_internal+0x1b7/0x210 [ 739.714311] device_driver_detach+0x14/0x20 [ 739.714315] unbind_store+0xa6/0xb0 [ 739.714319] drv_attr_store+0x21/0x30 [ 739.714322] sysfs_kf_write+0x48/0x60 [ 739.714328] kernfs_fop_write_iter+0x16b/0x240 [ 739.714333] vfs_write+0x266/0x520 [ 739.714341] ksys_write+0x72/0xe0 [ 739.714345] __x64_sys_write+0x19/0x20 [ 739.714347] x64_sys_call+0xa15/0xa30 [ 739.714355] do_syscall_64+0xd8/0xab0 [ 739.714361] entry_SYSCALL_64_after_hwframe+0x4b/0x53 and [ 739.714459] ------------[ cut here ]------------ [ 739.714461] xe 0000:67:00.0: [drm] drm_WARN_ON(!list_empty(&fb->filp_head)) [ 739.714464] WARNING: drivers/gpu/drm/drm_framebuffer.c:833 at drm_framebuffer_free+0x6c/0x90 [drm], CPU#12: xe_module_load/13145 [ 739.714715] RIP: 0010:drm_framebuffer_free+0x7a/0x90 [drm] ... [ 739.714869] Call Trace: [ 739.714871] [ 739.714876] drm_mode_config_cleanup+0x26a/0x320 [drm] [ 739.714998] ? __drm_printfn_seq_file+0x20/0x20 [drm] [ 739.715115] ? drm_mode_config_cleanup+0x207/0x320 [drm] [ 739.715235] intel_display_driver_remove_noirq+0x51/0xb0 [xe] [ 739.715576] xe_display_fini_early+0x33/0x50 [xe] [ 739.715821] devm_action_release+0xf/0x20 [ 739.715828] devres_release_all+0xad/0xf0 [ 739.715843] device_unbind_cleanup+0x12/0xa0 [ 739.715850] device_release_driver_internal+0x1b7/0x210 [ 739.715856] device_driver_detach+0x14/0x20 [ 739.715860] unbind_store+0xa6/0xb0 [ 739.715865] drv_attr_store+0x21/0x30 [ 739.715868] sysfs_kf_write+0x48/0x60 [ 739.715873] kernfs_fop_write_iter+0x16b/0x240 [ 739.715878] vfs_write+0x266/0x520 [ 739.715886] ksys_write+0x72/0xe0 [ 739.715890] __x64_sys_write+0x19/0x20 [ 739.715893] x64_sys_call+0xa15/0xa30 [ 739.715900] do_syscall_64+0xd8/0xab0 [ 739.715905] entry_SYSCALL_64_after_hwframe+0x4b/0x53 and then finally file close blows up: [ 743.186530] Oops: general protection fault, probably for non-canonical address 0xdead000000000122: 0000 [#1] SMP [ 743.186535] CPU: 3 UID: 1000 PID: 3453 Comm: kwin_wayland Tainted: G W 7.0.0-rc1-valkyria+ #110 PREEMPT_{RT,(lazy)} [ 743.186537] Tainted: [W]=WARN [ 743.186538] Hardware name: Gigabyte Technology Co., Ltd. X299 AORUS Gaming 3/X299 AORUS Gaming 3-CF, BIOS F8n 12/06/2021 [ 743.186539] RIP: 0010:drm_framebuffer_cleanup+0x55/0xc0 [drm] [ 743.186588] Code: d8 72 73 0f b6 42 05 ff c3 39 c3 72 e8 49 8d bd 50 07 00 00 31 f6 e8 3a 80 d3 e1 49 8b 44 24 10 49 8d 7c 24 08 49 8b 54 24 08 <48> 3b 38 0f 85 95 7f 02 00 48 3b 7a 08 0f 85 8b 7f 02 00 48 89 42 [ 743.186589] RSP: 0018:ffffc900085e3cf8 EFLAGS: 00010202 [ 743.186591] RAX: dead000000000122 RBX: 0000000000000001 RCX: ffffffff8217ed03 [ 743.186592] RDX: dead000000000100 RSI: 0000000000000000 RDI: ffff88814675ba08 [ 743.186593] RBP: ffffc900085e3d10 R08: 0000000000000000 R09: 0000000000000000 [ 743.186593] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88814675ba00 [ 743.186594] R13: ffff88810d778000 R14: ffff888119f6dca0 R15: ffff88810c660bb0 [ 743.186595] FS: 00007ff377d21280(0000) GS:ffff888cec3f8000(0000) knlGS:0000000000000000 [ 743.186596] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 743.186596] CR2: 000055690b55e000 CR3: 0000000113586003 CR4: 00000000003706f0 [ 743.186597] Call Trace: [ 743.186598] [ 743.186603] intel_user_framebuffer_destroy+0x12/0x90 [xe] [ 743.186722] drm_framebuffer_free+0x3a/0x90 [drm] [ 743.186750] ? trace_hardirqs_on+0x5f/0x120 [ 743.186754] drm_mode_object_put+0x51/0x70 [drm] [ 743.186786] drm_fb_release+0x105/0x190 [drm] [ 743.186812] ? rt_mutex_slowunlock+0x3aa/0x410 [ 743.186817] ? rt_spin_lock+0xea/0x1b0 [ 743.186819] drm_file_free+0x1e0/0x2c0 [drm] [ 743.186843] drm_release_noglobal+0x91/0xf0 [drm] [ 743.186865] __fput+0x100/0x2e0 [ 743.186869] fput_close_sync+0x40/0xa0 [ 743.186870] __x64_sys_close+0x3e/0x80 [ 743.186873] x64_sys_call+0xa07/0xa30 [ 743.186879] do_syscall_64+0xd8/0xab0 [ 743.186881] entry_SYSCALL_64_after_hwframe+0x4b/0x53 [ 743.186882] RIP: 0033:0x7ff37e567732 [ 743.186884] Code: 08 0f 85 a1 38 ff ff 49 89 fb 48 89 f0 48 89 d7 48 89 ce 4c 89 c2 4d 89 ca 4c 8b 44 24 08 4c 8b 4c 24 10 4c 89 5c 24 08 0f 05 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 55 bf 01 00 [ 743.186885] RSP: 002b:00007ffc818169a8 EFLAGS: 00000246 ORIG_RAX: 0000000000000003 [ 743.186886] RAX: ffffffffffffffda RBX: 00007ffc81816a30 RCX: 00007ff37e567732 [ 743.186887] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000012 [ 743.186888] RBP: 00007ffc818169d0 R08: 0000000000000000 R09: 0000000000000000 [ 743.186889] R10: 0000000000000000 R11: 0000000000000246 R12: 000055d60a7996e0 [ 743.186889] R13: 00007ffc81816a90 R14: 00007ffc81816a90 R15: 000055d60a782a30 [ 743.186892] [ 743.186893] Modules linked in: rfcomm snd_hrtimer xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_reject_ipv4 xt_tcpudp xt_addrtype nft_compat x_tables nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nf_tables overlay cfg80211 bnep mtd_intel_dg snd_hda_codec_intelhdmi mtd snd_hda_codec_hdmi nls_utf8 mxm_wmi intel_wmi_thunderbolt gigabyte_wmi wmi_bmof xe drm_gpuvm drm_gpusvm_helper i2c_algo_bit drm_buddy drm_ttm_helper ttm video drm_suballoc_helper gpu_sched drm_client_lib drm_exec drm_display_helper cec drm_kunit_helpers drm_kms_helper kunit x86_pkg_temp_thermal intel_powerclamp coretemp snd_hda_codec_alc882 snd_hda_codec_realtek_lib snd_hda_codec_generic snd_hda_intel snd_soc_avs snd_soc_hda_codec snd_hda_ext_core snd_hda_codec snd_hwdep snd_hda_core snd_intel_dspcfg snd_soc_core snd_compress ac97_bus snd_pcm snd_seq snd_seq_device snd_timer i2c_i801 i2c_mux snd i2c_smbus btusb btrtl btbcm btmtk btintel bluetooth ecdh_generic rfkill ecc mei_me mei ioatdma dca wmi nfsd drm i2c_dev fuse nfnetlink [ 743.186938] ---[ end trace 0000000000000000 ]--- And for property blobs: void drm_mode_config_cleanup(struct drm_device *dev) { ... list_for_each_entry_safe(blob, bt, &dev->mode_config.property_blob_list, head_global) { drm_property_blob_put(blob); } Resulting in: [ 371.072940] BUG: unable to handle page fault for address: 000001ffffffffff [ 371.072944] #PF: supervisor read access in kernel mode [ 371.072945] #PF: error_code(0x0000) - not-present page [ 371.072947] PGD 0 P4D 0 [ 371.072950] Oops: Oops: 0000 [#1] SMP [ 371.072953] CPU: 0 UID: 1000 PID: 3693 Comm: kwin_wayland Not tainted 7.0.0-rc1-valkyria+ #111 PREEMPT_{RT,(lazy)} [ 371.072956] Hardware name: Gigabyte Technology Co., Ltd. X299 AORUS Gaming 3/X299 AORUS Gaming 3-CF, BIOS F8n 12/06/2021 [ 371.072957] RIP: 0010:drm_property_destroy_user_blobs+0x3b/0x90 [drm] [ 371.073019] Code: 00 00 48 83 ec 10 48 8b 86 30 01 00 00 48 39 c3 74 59 48 89 c2 48 8d 48 c8 48 8b 00 4c 8d 60 c8 eb 04 4c 8d 60 c8 48 8b 71 40 <48> 39 16 0f 85 39 32 01 00 48 3b 50 08 0f 85 2f 32 01 00 48 89 70 [ 371.073021] RSP: 0018:ffffc90006a73de8 EFLAGS: 00010293 [ 371.073022] RAX: 000001ffffffffff RBX: ffff888118a1a930 RCX: ffff8881b92355c0 [ 371.073024] RDX: ffff8881b92355f8 RSI: 000001ffffffffff RDI: ffff888118be4000 [ 371.073025] RBP: ffffc90006a73e08 R08: ffff8881009b7300 R09: ffff888cecc5b000 [ 371.073026] R10: ffffc90006a73e90 R11: 0000000000000002 R12: 000001ffffffffc7 [ 371.073027] R13: ffff888118a1a980 R14: ffff88810b366d20 R15: ffff888118a1a970 [ 371.073028] FS: 00007f1faccbb280(0000) GS:ffff888cec2db000(0000) knlGS:0000000000000000 [ 371.073029] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 371.073030] CR2: 000001ffffffffff CR3: 000000010655c001 CR4: 00000000003706f0 [ 371.073031] Call Trace: [ 371.073033] [ 371.073036] drm_file_free+0x1df/0x2a0 [drm] [ 371.073077] drm_release_noglobal+0x7a/0xe0 [drm] [ 371.073113] __fput+0xe2/0x2b0 [ 371.073118] fput_close_sync+0x40/0xa0 [ 371.073119] __x64_sys_close+0x3e/0x80 [ 371.073122] x64_sys_call+0xa07/0xa30 [ 371.073126] do_syscall_64+0xc0/0x840 [ 371.073130] entry_SYSCALL_64_after_hwframe+0x4b/0x53 [ 371.073132] RIP: 0033:0x7f1fb3501732 [ 371.073133] Code: 08 0f 85 a1 38 ff ff 49 89 fb 48 89 f0 48 89 d7 48 89 ce 4c 89 c2 4d 89 ca 4c 8b 44 24 08 4c 8b 4c 24 10 4c 89 5c 24 08 0f 05 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 55 bf 01 00 [ 371.073135] RSP: 002b:00007ffe8e6f0278 EFLAGS: 00000246 ORIG_RAX: 0000000000000003 [ 371.073136] RAX: ffffffffffffffda RBX: 00007ffe8e6f0300 RCX: 00007f1fb3501732 [ 371.073137] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000012 [ 371.073138] RBP: 00007ffe8e6f02a0 R08: 0000000000000000 R09: 0000000000000000 [ 371.073139] R10: 0000000000000000 R11: 0000000000000246 R12: 00005585ba46eea0 [ 371.073140] R13: 00007ffe8e6f0360 R14: 00007ffe8e6f0360 R15: 00005585ba458a30 [ 371.073143] [ 371.073144] Modules linked in: rfcomm snd_hrtimer xt_addrtype xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_reject_ipv4 xt_tcpudp nft_compat x_tables nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nf_tables overlay cfg80211 bnep snd_hda_codec_intelhdmi snd_hda_codec_hdmi mtd_intel_dg mtd nls_utf8 wmi_bmof mxm_wmi gigabyte_wmi intel_wmi_thunderbolt xe drm_gpuvm drm_gpusvm_helper i2c_algo_bit drm_buddy drm_ttm_helper ttm video drm_suballoc_helper gpu_sched drm_client_lib drm_exec drm_display_helper cec drm_kunit_helpers drm_kms_helper kunit x86_pkg_temp_thermal intel_powerclamp coretemp snd_hda_codec_alc882 snd_hda_codec_realtek_lib snd_hda_codec_generic snd_hda_intel snd_soc_avs snd_soc_hda_codec snd_hda_ext_core snd_hda_codec snd_hwdep snd_hda_core snd_intel_dspcfg snd_soc_core snd_compress ac97_bus snd_pcm snd_seq snd_seq_device snd_timer i2c_i801 btusb i2c_mux i2c_smbus btrtl snd btbcm btmtk btintel bluetooth ecdh_generic rfkill ecc mei_me mei ioatdma dca wmi nfsd drm i2c_dev fuse nfnetlink [ 371.073198] CR2: 000001ffffffffff [ 371.073199] ---[ end trace 0000000000000000 ]--- Add a guard around file close, and ensure the warnings from drm_mode_config do not trigger. Fix those by allowing an open reference to the file descriptor and cleaning up the file linked list entry in drm_mode_config_cleanup(). Cc: # v4.18+ Fixes: bee330f3d672 ("drm: Use srcu to protect drm_device.unplugged") Cc: Thomas Hellström Reviewed-by: Thomas Hellström Link: https://patch.msgid.link/20260313151728.14990-4-dev@lankhorst.se Signed-off-by: Maarten Lankhorst --- drivers/gpu/drm/drm_file.c | 5 ++++- drivers/gpu/drm/drm_mode_config.c | 9 ++++++--- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/drm_file.c b/drivers/gpu/drm/drm_file.c index ec820686b302..f52141f842a1 100644 --- a/drivers/gpu/drm/drm_file.c +++ b/drivers/gpu/drm/drm_file.c @@ -233,6 +233,7 @@ static void drm_events_release(struct drm_file *file_priv) void drm_file_free(struct drm_file *file) { struct drm_device *dev; + int idx; if (!file) return; @@ -249,9 +250,11 @@ void drm_file_free(struct drm_file *file) drm_events_release(file); - if (drm_core_check_feature(dev, DRIVER_MODESET)) { + if (drm_core_check_feature(dev, DRIVER_MODESET) && + drm_dev_enter(dev, &idx)) { drm_fb_release(file); drm_property_destroy_user_blobs(dev, file); + drm_dev_exit(idx); } if (drm_core_check_feature(dev, DRIVER_SYNCOBJ)) diff --git a/drivers/gpu/drm/drm_mode_config.c b/drivers/gpu/drm/drm_mode_config.c index d12db9b0bab8..802bc4608abf 100644 --- a/drivers/gpu/drm/drm_mode_config.c +++ b/drivers/gpu/drm/drm_mode_config.c @@ -577,10 +577,13 @@ void drm_mode_config_cleanup(struct drm_device *dev) */ WARN_ON(!list_empty(&dev->mode_config.fb_list)); list_for_each_entry_safe(fb, fbt, &dev->mode_config.fb_list, head) { - struct drm_printer p = drm_dbg_printer(dev, DRM_UT_KMS, "[leaked fb]"); + if (list_empty(&fb->filp_head) || drm_framebuffer_read_refcount(fb) > 1) { + struct drm_printer p = drm_dbg_printer(dev, DRM_UT_KMS, "[leaked fb]"); - drm_printf(&p, "framebuffer[%u]:\n", fb->base.id); - drm_framebuffer_print_info(&p, 1, fb); + drm_printf(&p, "framebuffer[%u]:\n", fb->base.id); + drm_framebuffer_print_info(&p, 1, fb); + } + list_del_init(&fb->filp_head); drm_framebuffer_free(&fb->base.refcount); } From 8c89a077ca796a2fe248c584e9d7e66cff0388c8 Mon Sep 17 00:00:00 2001 From: Praveen Talari Date: Fri, 13 Mar 2026 21:49:01 +0530 Subject: [PATCH 155/352] spi: geni-qcom: Check DMA interrupts early in ISR The current interrupt handler only checks the GENI main IRQ status (m_irq) before deciding to return IRQ_NONE. This can lead to spurious IRQ_NONE returns when DMA interrupts are pending but m_irq is zero. Move the DMA TX/RX status register reads to the beginning of the ISR, right after reading m_irq. Update the early return condition to check all three status registers (m_irq, dma_tx_status, dma_rx_status) before returning IRQ_NONE. Signed-off-by: Praveen Talari Reviewed-by: Konrad Dybcio Link: https://patch.msgid.link/20260313-spi-geni-qcom-fix-dma-irq-handling-v1-1-0bd122589e02@oss.qualcomm.com Signed-off-by: Mark Brown --- drivers/spi/spi-geni-qcom.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/spi/spi-geni-qcom.c b/drivers/spi/spi-geni-qcom.c index 772b7148ba5f..d5fb0edc8e0c 100644 --- a/drivers/spi/spi-geni-qcom.c +++ b/drivers/spi/spi-geni-qcom.c @@ -906,10 +906,13 @@ static irqreturn_t geni_spi_isr(int irq, void *data) struct spi_controller *spi = data; struct spi_geni_master *mas = spi_controller_get_devdata(spi); struct geni_se *se = &mas->se; - u32 m_irq; + u32 m_irq, dma_tx_status, dma_rx_status; m_irq = readl(se->base + SE_GENI_M_IRQ_STATUS); - if (!m_irq) + dma_tx_status = readl_relaxed(se->base + SE_DMA_TX_IRQ_STAT); + dma_rx_status = readl_relaxed(se->base + SE_DMA_RX_IRQ_STAT); + + if (!m_irq && !dma_tx_status && !dma_rx_status) return IRQ_NONE; if (m_irq & (M_CMD_OVERRUN_EN | M_ILLEGAL_CMD_EN | M_CMD_FAILURE_EN | @@ -957,8 +960,6 @@ static irqreturn_t geni_spi_isr(int irq, void *data) } } else if (mas->cur_xfer_mode == GENI_SE_DMA) { const struct spi_transfer *xfer = mas->cur_xfer; - u32 dma_tx_status = readl_relaxed(se->base + SE_DMA_TX_IRQ_STAT); - u32 dma_rx_status = readl_relaxed(se->base + SE_DMA_RX_IRQ_STAT); if (dma_tx_status) writel(dma_tx_status, se->base + SE_DMA_TX_IRQ_CLR); From cffcb42c57686e9a801dfcf37a3d0c62e51c1c3e Mon Sep 17 00:00:00 2001 From: Jonas Karlman Date: Sat, 28 Feb 2026 11:28:22 +0000 Subject: [PATCH 156/352] drm/bridge: dw-hdmi-qp: fix multi-channel audio output Channel Allocation (PB4) and Level Shift Information (PB5) are configured with values from PB1 and PB2 due to the wrong offset being used. This results in missing audio channels or incorrect speaker placement when playing multi-channel audio. Use the correct offset to fix multi-channel audio output. Fixes: fd0141d1a8a2 ("drm/bridge: synopsys: Add audio support for dw-hdmi-qp") Reported-by: Christian Hewitt Signed-off-by: Jonas Karlman Signed-off-by: Christian Hewitt Reviewed-by: Cristian Ciocaltea Link: https://patch.msgid.link/20260228112822.4056354-1-christianshewitt@gmail.com Signed-off-by: Luca Ceresoli --- drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c index ab7fed6214e0..facfb7526928 100644 --- a/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c +++ b/drivers/gpu/drm/bridge/synopsys/dw-hdmi-qp.c @@ -848,7 +848,7 @@ static int dw_hdmi_qp_config_audio_infoframe(struct dw_hdmi_qp *hdmi, regmap_bulk_write(hdmi->regm, PKT_AUDI_CONTENTS0, &header_bytes, 1); regmap_bulk_write(hdmi->regm, PKT_AUDI_CONTENTS1, &buffer[3], 1); - regmap_bulk_write(hdmi->regm, PKT_AUDI_CONTENTS2, &buffer[4], 1); + regmap_bulk_write(hdmi->regm, PKT_AUDI_CONTENTS2, &buffer[7], 1); /* Enable ACR, AUDI, AMD */ dw_hdmi_qp_mod(hdmi, From cb3d1049f4ea77d5ad93f17d8ac1f2ed4da70501 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Tue, 3 Mar 2026 12:53:18 +0100 Subject: [PATCH 157/352] driver core: generalize driver_override in struct device Currently, there are 12 busses (including platform and PCI) that duplicate the driver_override logic for their individual devices. All of them seem to be prone to the bug described in [1]. While this could be solved for every bus individually using a separate lock, solving this in the driver-core generically results in less (and cleaner) changes overall. Thus, move driver_override to struct device, provide corresponding accessors for busses and handle locking with a separate lock internally. In particular, add device_set_driver_override(), device_has_driver_override(), device_match_driver_override() and generalize the sysfs store() and show() callbacks via a driver_override feature flag in struct bus_type. Until all busses have migrated, keep driver_set_override() in place. Note that we can't use the device lock for the reasons described in [2]. Link: https://bugzilla.kernel.org/show_bug.cgi?id=220789 [1] Link: https://lore.kernel.org/driver-core/DGRGTIRHA62X.3RY09D9SOK77P@kernel.org/ [2] Tested-by: Gui-Dong Han Co-developed-by: Gui-Dong Han Signed-off-by: Gui-Dong Han Reviewed-by: Greg Kroah-Hartman Link: https://patch.msgid.link/20260303115720.48783-2-dakr@kernel.org [ Use dev->bus instead of sp->bus for consistency; fix commit message to refer to the struct bus_type's driver_override feature flag. - Danilo ] Signed-off-by: Danilo Krummrich --- drivers/base/bus.c | 43 ++++++++++++++++++++++++++- drivers/base/core.c | 2 ++ drivers/base/dd.c | 60 ++++++++++++++++++++++++++++++++++++++ include/linux/device.h | 54 ++++++++++++++++++++++++++++++++++ include/linux/device/bus.h | 4 +++ 5 files changed, 162 insertions(+), 1 deletion(-) diff --git a/drivers/base/bus.c b/drivers/base/bus.c index bb61d8adbab1..8b6722ff8590 100644 --- a/drivers/base/bus.c +++ b/drivers/base/bus.c @@ -504,6 +504,36 @@ int bus_for_each_drv(const struct bus_type *bus, struct device_driver *start, } EXPORT_SYMBOL_GPL(bus_for_each_drv); +static ssize_t driver_override_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + int ret; + + ret = __device_set_driver_override(dev, buf, count); + if (ret) + return ret; + + return count; +} + +static ssize_t driver_override_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + guard(spinlock)(&dev->driver_override.lock); + return sysfs_emit(buf, "%s\n", dev->driver_override.name); +} +static DEVICE_ATTR_RW(driver_override); + +static struct attribute *driver_override_dev_attrs[] = { + &dev_attr_driver_override.attr, + NULL, +}; + +static const struct attribute_group driver_override_dev_group = { + .attrs = driver_override_dev_attrs, +}; + /** * bus_add_device - add device to bus * @dev: device being added @@ -537,9 +567,15 @@ int bus_add_device(struct device *dev) if (error) goto out_put; + if (dev->bus->driver_override) { + error = device_add_group(dev, &driver_override_dev_group); + if (error) + goto out_groups; + } + error = sysfs_create_link(&sp->devices_kset->kobj, &dev->kobj, dev_name(dev)); if (error) - goto out_groups; + goto out_override; error = sysfs_create_link(&dev->kobj, &sp->subsys.kobj, "subsystem"); if (error) @@ -550,6 +586,9 @@ int bus_add_device(struct device *dev) out_subsys: sysfs_remove_link(&sp->devices_kset->kobj, dev_name(dev)); +out_override: + if (dev->bus->driver_override) + device_remove_group(dev, &driver_override_dev_group); out_groups: device_remove_groups(dev, sp->bus->dev_groups); out_put: @@ -607,6 +646,8 @@ void bus_remove_device(struct device *dev) sysfs_remove_link(&dev->kobj, "subsystem"); sysfs_remove_link(&sp->devices_kset->kobj, dev_name(dev)); + if (dev->bus->driver_override) + device_remove_group(dev, &driver_override_dev_group); device_remove_groups(dev, dev->bus->dev_groups); if (klist_node_attached(&dev->p->knode_bus)) klist_del(&dev->p->knode_bus); diff --git a/drivers/base/core.c b/drivers/base/core.c index 791f9e444df8..09b98f02f559 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -2556,6 +2556,7 @@ static void device_release(struct kobject *kobj) devres_release_all(dev); kfree(dev->dma_range_map); + kfree(dev->driver_override.name); if (dev->release) dev->release(dev); @@ -3159,6 +3160,7 @@ void device_initialize(struct device *dev) kobject_init(&dev->kobj, &device_ktype); INIT_LIST_HEAD(&dev->dma_pools); mutex_init(&dev->mutex); + spin_lock_init(&dev->driver_override.lock); lockdep_set_novalidate_class(&dev->mutex); spin_lock_init(&dev->devres_lock); INIT_LIST_HEAD(&dev->devres_head); diff --git a/drivers/base/dd.c b/drivers/base/dd.c index bea8da5f8a3a..37c7e54e0e4c 100644 --- a/drivers/base/dd.c +++ b/drivers/base/dd.c @@ -381,6 +381,66 @@ static void __exit deferred_probe_exit(void) } __exitcall(deferred_probe_exit); +int __device_set_driver_override(struct device *dev, const char *s, size_t len) +{ + const char *new, *old; + char *cp; + + if (!s) + return -EINVAL; + + /* + * The stored value will be used in sysfs show callback (sysfs_emit()), + * which has a length limit of PAGE_SIZE and adds a trailing newline. + * Thus we can store one character less to avoid truncation during sysfs + * show. + */ + if (len >= (PAGE_SIZE - 1)) + return -EINVAL; + + /* + * Compute the real length of the string in case userspace sends us a + * bunch of \0 characters like python likes to do. + */ + len = strlen(s); + + if (!len) { + /* Empty string passed - clear override */ + spin_lock(&dev->driver_override.lock); + old = dev->driver_override.name; + dev->driver_override.name = NULL; + spin_unlock(&dev->driver_override.lock); + kfree(old); + + return 0; + } + + cp = strnchr(s, len, '\n'); + if (cp) + len = cp - s; + + new = kstrndup(s, len, GFP_KERNEL); + if (!new) + return -ENOMEM; + + spin_lock(&dev->driver_override.lock); + old = dev->driver_override.name; + if (cp != s) { + dev->driver_override.name = new; + spin_unlock(&dev->driver_override.lock); + } else { + /* "\n" passed - clear override */ + dev->driver_override.name = NULL; + spin_unlock(&dev->driver_override.lock); + + kfree(new); + } + kfree(old); + + return 0; +} +EXPORT_SYMBOL_GPL(__device_set_driver_override); + /** * device_is_bound() - Check if device is bound to a driver * @dev: device to check diff --git a/include/linux/device.h b/include/linux/device.h index 0be95294b6e6..e65d564f01cd 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -483,6 +483,8 @@ struct device_physical_location { * on. This shrinks the "Board Support Packages" (BSPs) and * minimizes board-specific #ifdefs in drivers. * @driver_data: Private pointer for driver specific info. + * @driver_override: Driver name to force a match. Do not touch directly; use + * device_set_driver_override() instead. * @links: Links to suppliers and consumers of this device. * @power: For device power management. * See Documentation/driver-api/pm/devices.rst for details. @@ -576,6 +578,10 @@ struct device { core doesn't touch it */ void *driver_data; /* Driver data, set and get with dev_set_drvdata/dev_get_drvdata */ + struct { + const char *name; + spinlock_t lock; + } driver_override; struct mutex mutex; /* mutex to synchronize calls to * its driver. */ @@ -701,6 +707,54 @@ struct device_link { #define kobj_to_dev(__kobj) container_of_const(__kobj, struct device, kobj) +int __device_set_driver_override(struct device *dev, const char *s, size_t len); + +/** + * device_set_driver_override() - Helper to set or clear driver override. + * @dev: Device to change + * @s: NUL-terminated string, new driver name to force a match, pass empty + * string to clear it ("" or "\n", where the latter is only for sysfs + * interface). + * + * Helper to set or clear driver override of a device. + * + * Returns: 0 on success or a negative error code on failure. + */ +static inline int device_set_driver_override(struct device *dev, const char *s) +{ + return __device_set_driver_override(dev, s, s ? strlen(s) : 0); +} + +/** + * device_has_driver_override() - Check if a driver override has been set. + * @dev: device to check + * + * Returns true if a driver override has been set for this device. + */ +static inline bool device_has_driver_override(struct device *dev) +{ + guard(spinlock)(&dev->driver_override.lock); + return !!dev->driver_override.name; +} + +/** + * device_match_driver_override() - Match a driver against the device's driver_override. + * @dev: device to check + * @drv: driver to match against + * + * Returns > 0 if a driver override is set and matches the given driver, 0 if a + * driver override is set but does not match, or < 0 if a driver override is not + * set at all. + */ +static inline int device_match_driver_override(struct device *dev, + const struct device_driver *drv) +{ + guard(spinlock)(&dev->driver_override.lock); + if (dev->driver_override.name) + return !strcmp(dev->driver_override.name, drv->name); + return -1; +} + /** * device_iommu_mapped - Returns true when the device DMA is translated * by an IOMMU diff --git a/include/linux/device/bus.h b/include/linux/device/bus.h index 63de5f053c33..c1b463cd6464 100644 --- a/include/linux/device/bus.h +++ b/include/linux/device/bus.h @@ -65,6 +65,9 @@ struct fwnode_handle; * this bus. * @pm: Power management operations of this bus, callback the specific * device driver's pm-ops. + * @driver_override: Set to true if this bus supports the driver_override + * mechanism, which allows userspace to force a specific + * driver to bind to a device via a sysfs attribute. * @need_parent_lock: When probing or removing a device on this bus, the * device core should lock the device's parent. * @@ -106,6 +109,7 @@ struct bus_type { const struct dev_pm_ops *pm; + bool driver_override; bool need_parent_lock; }; From bcd085d5c76f687f5b6df049f7c415ae63a9b857 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Tue, 3 Mar 2026 12:53:19 +0100 Subject: [PATCH 158/352] docs: driver-model: document driver_override Now that we support driver_override as a driver-core feature through struct device and struct bus_type, add some documentation in the context of how a device / driver binding is established. Reviewed-by: Greg Kroah-Hartman Link: https://patch.msgid.link/20260303115720.48783-3-dakr@kernel.org Signed-off-by: Danilo Krummrich --- .../driver-api/driver-model/binding.rst | 48 +++++++++++++++++++ 1 file changed, 48 insertions(+) diff --git a/Documentation/driver-api/driver-model/binding.rst b/Documentation/driver-api/driver-model/binding.rst index d1d311a4011f..fa0888c2b3b9 100644 --- a/Documentation/driver-api/driver-model/binding.rst +++ b/Documentation/driver-api/driver-model/binding.rst @@ -99,3 +99,51 @@ of the driver is decremented. All symlinks between the two are removed. When a driver is removed, the list of devices that it supports is iterated over, and the driver's remove callback is called for each one. The device is removed from that list and the symlinks removed. + + +Driver Override +~~~~~~~~~~~~~~~ + +Userspace may override the standard matching by writing a driver name to +a device's ``driver_override`` sysfs attribute. When set, only a driver +whose name matches the override will be considered during binding. This +bypasses all bus-specific matching (OF, ACPI, ID tables, etc.). + +The override may be cleared by writing an empty string, which returns +the device to standard matching rules. Writing to ``driver_override`` +does not automatically unbind the device from its current driver or +make any attempt to load the specified driver. + +Buses opt into this mechanism by setting the ``driver_override`` flag in +their ``struct bus_type``:: + + const struct bus_type example_bus_type = { + ... + .driver_override = true, + }; + +When the flag is set, the driver core automatically creates the +``driver_override`` sysfs attribute for every device on that bus. + +The bus's ``match()`` callback should check the override before performing +its own matching, using ``device_match_driver_override()``:: + + static int example_match(struct device *dev, const struct device_driver *drv) + { + int ret; + + ret = device_match_driver_override(dev, drv); + if (ret >= 0) + return ret; + + /* Fall through to bus-specific matching... */ + } + +``device_match_driver_override()`` returns > 0 if the override matches +the given driver, 0 if the override is set but does not match, or < 0 if +no override is set at all. + +Additional helpers are available: + +- ``device_set_driver_override()`` - set or clear the override from kernel code. +- ``device_has_driver_override()`` - check whether an override is set. From 813bbc4d33d2ca5b0da63e70ae13b60874f20d37 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Tue, 3 Mar 2026 12:53:20 +0100 Subject: [PATCH 159/352] hwmon: axi-fan: don't use driver_override as IRQ name MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Do not use driver_override as IRQ name, as it is not guaranteed to point to a valid string; use NULL instead (which makes the devm IRQ helpers use dev_name()). Fixes: 8412b410fa5e ("hwmon: Support ADI Fan Control IP") Reviewed-by: Nuno Sá Acked-by: Guenter Roeck Reviewed-by: Greg Kroah-Hartman Link: https://patch.msgid.link/20260303115720.48783-4-dakr@kernel.org Signed-off-by: Danilo Krummrich --- drivers/hwmon/axi-fan-control.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hwmon/axi-fan-control.c b/drivers/hwmon/axi-fan-control.c index b7bb325c3ad9..01590dfa55e6 100644 --- a/drivers/hwmon/axi-fan-control.c +++ b/drivers/hwmon/axi-fan-control.c @@ -507,7 +507,7 @@ static int axi_fan_control_probe(struct platform_device *pdev) ret = devm_request_threaded_irq(&pdev->dev, ctl->irq, NULL, axi_fan_control_irq_handler, IRQF_ONESHOT | IRQF_TRIGGER_HIGH, - pdev->driver_override, ctl); + NULL, ctl); if (ret) return dev_err_probe(&pdev->dev, ret, "failed to request an irq\n"); From c5f60e3f07b6609562d21efda878e83ce8860728 Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Tue, 17 Mar 2026 00:37:15 +0100 Subject: [PATCH 160/352] sh: platform_early: remove pdev->driver_override check In commit 507fd01d5333 ("drivers: move the early platform device support to arch/sh") platform_match() was copied over to the sh platform_early code, accidentally including the driver_override check. This check does not make sense for platform_early, as sysfs is not even available in first place at this point in the boot process, hence remove the check. Reviewed-by: Greg Kroah-Hartman Reviewed-by: Geert Uytterhoeven Fixes: 507fd01d5333 ("drivers: move the early platform device support to arch/sh") Link: https://lore.kernel.org/all/DH4M3DJ4P58T.1BGVAVXN71Z09@kernel.org/ Signed-off-by: Danilo Krummrich --- arch/sh/drivers/platform_early.c | 4 ---- 1 file changed, 4 deletions(-) diff --git a/arch/sh/drivers/platform_early.c b/arch/sh/drivers/platform_early.c index 143747c45206..48ddbc547bd9 100644 --- a/arch/sh/drivers/platform_early.c +++ b/arch/sh/drivers/platform_early.c @@ -26,10 +26,6 @@ static int platform_match(struct device *dev, struct device_driver *drv) struct platform_device *pdev = to_platform_device(dev); struct platform_driver *pdrv = to_platform_driver(drv); - /* When driver_override is set, only bind to the matching driver */ - if (pdev->driver_override) - return !strcmp(pdev->driver_override, drv->name); - /* Then try to match against the id table */ if (pdrv->id_table) return platform_match_id(pdrv->id_table, pdev) != NULL; From 2b38efc05bf7a8568ec74bfffea0f5cfa62bc01d Mon Sep 17 00:00:00 2001 From: Danilo Krummrich Date: Tue, 3 Mar 2026 12:53:21 +0100 Subject: [PATCH 161/352] driver core: platform: use generic driver_override infrastructure When a driver is probed through __driver_attach(), the bus' match() callback is called without the device lock held, thus accessing the driver_override field without a lock, which can cause a UAF. Fix this by using the driver-core driver_override infrastructure taking care of proper locking internally. Note that calling match() from __driver_attach() without the device lock held is intentional. [1] Link: https://lore.kernel.org/driver-core/DGRGTIRHA62X.3RY09D9SOK77P@kernel.org/ [1] Reported-by: Gui-Dong Han Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220789 Fixes: 3d713e0e382e ("driver core: platform: add device binding path 'driver_override'") Reviewed-by: Greg Kroah-Hartman Link: https://patch.msgid.link/20260303115720.48783-5-dakr@kernel.org Signed-off-by: Danilo Krummrich --- drivers/base/platform.c | 37 +++++---------------------------- drivers/bus/simple-pm-bus.c | 4 ++-- drivers/clk/imx/clk-scu.c | 3 +-- drivers/slimbus/qcom-ngd-ctrl.c | 6 ++---- include/linux/platform_device.h | 5 ----- sound/soc/samsung/i2s.c | 6 +++--- 6 files changed, 13 insertions(+), 48 deletions(-) diff --git a/drivers/base/platform.c b/drivers/base/platform.c index b45d41b018ca..d44591d52e36 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -603,7 +603,6 @@ static void platform_device_release(struct device *dev) kfree(pa->pdev.dev.platform_data); kfree(pa->pdev.mfd_cell); kfree(pa->pdev.resource); - kfree(pa->pdev.driver_override); kfree(pa); } @@ -1306,38 +1305,9 @@ static ssize_t numa_node_show(struct device *dev, } static DEVICE_ATTR_RO(numa_node); -static ssize_t driver_override_show(struct device *dev, - struct device_attribute *attr, char *buf) -{ - struct platform_device *pdev = to_platform_device(dev); - ssize_t len; - - device_lock(dev); - len = sysfs_emit(buf, "%s\n", pdev->driver_override); - device_unlock(dev); - - return len; -} - -static ssize_t driver_override_store(struct device *dev, - struct device_attribute *attr, - const char *buf, size_t count) -{ - struct platform_device *pdev = to_platform_device(dev); - int ret; - - ret = driver_set_override(dev, &pdev->driver_override, buf, count); - if (ret) - return ret; - - return count; -} -static DEVICE_ATTR_RW(driver_override); - static struct attribute *platform_dev_attrs[] = { &dev_attr_modalias.attr, &dev_attr_numa_node.attr, - &dev_attr_driver_override.attr, NULL, }; @@ -1377,10 +1347,12 @@ static int platform_match(struct device *dev, const struct device_driver *drv) { struct platform_device *pdev = to_platform_device(dev); struct platform_driver *pdrv = to_platform_driver(drv); + int ret; /* When driver_override is set, only bind to the matching driver */ - if (pdev->driver_override) - return !strcmp(pdev->driver_override, drv->name); + ret = device_match_driver_override(dev, drv); + if (ret >= 0) + return ret; /* Attempt an OF style match first */ if (of_driver_match_device(dev, drv)) @@ -1516,6 +1488,7 @@ static const struct dev_pm_ops platform_dev_pm_ops = { const struct bus_type platform_bus_type = { .name = "platform", .dev_groups = platform_dev_groups, + .driver_override = true, .match = platform_match, .uevent = platform_uevent, .probe = platform_probe, diff --git a/drivers/bus/simple-pm-bus.c b/drivers/bus/simple-pm-bus.c index 3f00d953fb9a..c920bd6fbaaf 100644 --- a/drivers/bus/simple-pm-bus.c +++ b/drivers/bus/simple-pm-bus.c @@ -36,7 +36,7 @@ static int simple_pm_bus_probe(struct platform_device *pdev) * that's not listed in simple_pm_bus_of_match. We don't want to do any * of the simple-pm-bus tasks for these devices, so return early. */ - if (pdev->driver_override) + if (device_has_driver_override(&pdev->dev)) return 0; match = of_match_device(dev->driver->of_match_table, dev); @@ -78,7 +78,7 @@ static void simple_pm_bus_remove(struct platform_device *pdev) { const void *data = of_device_get_match_data(&pdev->dev); - if (pdev->driver_override || data) + if (device_has_driver_override(&pdev->dev) || data) return; dev_dbg(&pdev->dev, "%s\n", __func__); diff --git a/drivers/clk/imx/clk-scu.c b/drivers/clk/imx/clk-scu.c index a85ec48a798b..9b33df9967ec 100644 --- a/drivers/clk/imx/clk-scu.c +++ b/drivers/clk/imx/clk-scu.c @@ -706,8 +706,7 @@ struct clk_hw *imx_clk_scu_alloc_dev(const char *name, if (ret) goto put_device; - ret = driver_set_override(&pdev->dev, &pdev->driver_override, - "imx-scu-clk", strlen("imx-scu-clk")); + ret = device_set_driver_override(&pdev->dev, "imx-scu-clk"); if (ret) goto put_device; diff --git a/drivers/slimbus/qcom-ngd-ctrl.c b/drivers/slimbus/qcom-ngd-ctrl.c index 9aa7218b4e8d..1ed6be6e85d2 100644 --- a/drivers/slimbus/qcom-ngd-ctrl.c +++ b/drivers/slimbus/qcom-ngd-ctrl.c @@ -1535,10 +1535,8 @@ static int of_qcom_slim_ngd_register(struct device *parent, ngd->id = id; ngd->pdev->dev.parent = parent; - ret = driver_set_override(&ngd->pdev->dev, - &ngd->pdev->driver_override, - QCOM_SLIM_NGD_DRV_NAME, - strlen(QCOM_SLIM_NGD_DRV_NAME)); + ret = device_set_driver_override(&ngd->pdev->dev, + QCOM_SLIM_NGD_DRV_NAME); if (ret) { platform_device_put(ngd->pdev); kfree(ngd); diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h index 813da101b5bf..ed1d50d1c3c1 100644 --- a/include/linux/platform_device.h +++ b/include/linux/platform_device.h @@ -31,11 +31,6 @@ struct platform_device { struct resource *resource; const struct platform_device_id *id_entry; - /* - * Driver name to force a match. Do not set directly, because core - * frees it. Use driver_set_override() to set or clear it. - */ - const char *driver_override; /* MFD cell pointer */ struct mfd_cell *mfd_cell; diff --git a/sound/soc/samsung/i2s.c b/sound/soc/samsung/i2s.c index e9964f0e010a..140907a41a70 100644 --- a/sound/soc/samsung/i2s.c +++ b/sound/soc/samsung/i2s.c @@ -1360,10 +1360,10 @@ static int i2s_create_secondary_device(struct samsung_i2s_priv *priv) if (!pdev_sec) return -ENOMEM; - pdev_sec->driver_override = kstrdup("samsung-i2s", GFP_KERNEL); - if (!pdev_sec->driver_override) { + ret = device_set_driver_override(&pdev_sec->dev, "samsung-i2s"); + if (ret) { platform_device_put(pdev_sec); - return -ENOMEM; + return ret; } ret = platform_device_add(pdev_sec); From a68ed2df72131447d131531a08fe4dfcf4fa4653 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Sun, 15 Mar 2026 09:03:03 -0600 Subject: [PATCH 162/352] io_uring/poll: fix multishot recv missing EOF on wakeup race When a socket send and shutdown() happen back-to-back, both fire wake-ups before the receiver's task_work has a chance to run. The first wake gets poll ownership (poll_refs=1), and the second bumps it to 2. When io_poll_check_events() runs, it calls io_poll_issue() which does a recv that reads the data and returns IOU_RETRY. The loop then drains all accumulated refs (atomic_sub_return(2) -> 0) and exits, even though only the first event was consumed. Since the shutdown is a persistent state change, no further wakeups will happen, and the multishot recv can hang forever. Check specifically for HUP in the poll loop, and ensure that another loop is done to check for status if more than a single poll activation is pending. This ensures we don't lose the shutdown event. Cc: stable@vger.kernel.org Fixes: dbc2564cfe0f ("io_uring: let fast poll support multishot") Reported-by: Francis Brosseau Link: https://github.com/axboe/liburing/issues/1549 Signed-off-by: Jens Axboe --- io_uring/poll.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-) diff --git a/io_uring/poll.c b/io_uring/poll.c index aac4b3b881fb..488c08593b64 100644 --- a/io_uring/poll.c +++ b/io_uring/poll.c @@ -272,6 +272,7 @@ static int io_poll_check_events(struct io_kiocb *req, io_tw_token_t tw) atomic_andnot(IO_POLL_RETRY_FLAG, &req->poll_refs); v &= ~IO_POLL_RETRY_FLAG; } + v &= IO_POLL_REF_MASK; } /* the mask was stashed in __io_poll_execute */ @@ -304,8 +305,13 @@ static int io_poll_check_events(struct io_kiocb *req, io_tw_token_t tw) return IOU_POLL_REMOVE_POLL_USE_RES; } } else { - int ret = io_poll_issue(req, tw); + int ret; + /* multiple refs and HUP, ensure we loop once more */ + if ((req->cqe.res & (POLLHUP | POLLRDHUP)) && v != 1) + v--; + + ret = io_poll_issue(req, tw); if (ret == IOU_COMPLETE) return IOU_POLL_REMOVE_POLL_USE_RES; else if (ret == IOU_REQUEUE) @@ -321,7 +327,6 @@ static int io_poll_check_events(struct io_kiocb *req, io_tw_token_t tw) * Release all references, retry if someone tried to restart * task_work while we were executing it. */ - v &= IO_POLL_REF_MASK; } while (atomic_sub_return(v, &req->poll_refs) & IO_POLL_REF_MASK); io_napi_add(req); From 8dd1d9a335321d0829aeb85d8e1a897248d0da29 Mon Sep 17 00:00:00 2001 From: Ian Rogers Date: Fri, 6 Feb 2026 16:49:56 -0800 Subject: [PATCH 163/352] perf metricgroup: Fix metricgroup__has_metric_or_groups() Use metricgroup__for_each_metric() rather than pmu_metrics_table__for_each_metric() that combines the default metric table with, a potentially empty, CPUID table. Fixes: cee275edcdb1acfd ("perf metricgroup: Don't early exit if no CPUID table exists") Reviewed-by: Leo Yan Signed-off-by: Ian Rogers Tested-by: Arnaldo Carvalho de Melo Tested-by: Leo Yan Cc: Ian Rogers Signed-off-by: Namhyung Kim Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/metricgroup.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf/util/metricgroup.c b/tools/perf/util/metricgroup.c index 46bf4dfeebc8..7e39d469111b 100644 --- a/tools/perf/util/metricgroup.c +++ b/tools/perf/util/metricgroup.c @@ -1605,9 +1605,9 @@ bool metricgroup__has_metric_or_groups(const char *pmu, const char *metric_or_gr .metric_or_groups = metric_or_groups, }; - return pmu_metrics_table__for_each_metric(table, - metricgroup__has_metric_or_groups_callback, - &data) + return metricgroup__for_each_metric(table, + metricgroup__has_metric_or_groups_callback, + &data) ? true : false; } From 72a8b9c060d3188ff29e2a3f3ea47b1f2a67e005 Mon Sep 17 00:00:00 2001 From: Thomas Richter Date: Tue, 17 Feb 2026 14:14:56 +0100 Subject: [PATCH 164/352] perf parse-events: Fix big-endian 'overwrite' by writing correct union member The "Read backward ring buffer" test crashes on big-endian (e.g. s390x) due to a NULL dereference when the backward mmap path isn't enabled. Reproducer: # ./perf test -F 'Read backward ring buffer' Segmentation fault (core dumped) # uname -m s390x # Root cause: get_config_terms() stores into evsel_config_term::val.val (u64) while later code reads boolean fields such as evsel_config_term::val.overwrite. On big-endian the 1-byte boolean is left-aligned, so writing evsel_config_term::val.val = 1 is read back as evsel_config_term::val.overwrite = 0, leaving backward mmap disabled and a NULL map being used. Store values in the union member that matches the term type, e.g.: /* for OVERWRITE */ new_term->val.overwrite = 1; /* not new_term->val.val = 1 */ to fix this. Improve add_config_term() and add two more parameters for string and value. Function add_config_term() now creates a complete node element of type evsel_config_term and handles all evsel_config_term::val union members. Impact: Enables backward mmap on big-endian and prevents the crash. No change on little-endian. Output after: # ./perf test -Fv 44 --- start --- Using CPUID IBM,9175,705,ME1,3.8,002f mmap size 1052672B mmap size 8192B ---- end ---- 44: Read backward ring buffer : Ok # Fixes: 159ca97cd97ce8cc ("perf parse-events: Refactor get_config_terms() to remove macros") Reviewed-by: James Clark Reviewed-by: Jan Polensky Signed-off-by: Thomas Richter Acked-by: Ian Rogers Cc: James Clark Signed-off-by: Namhyung Kim Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/parse-events.c | 82 +++++++++++++++++++++++++++------- 1 file changed, 65 insertions(+), 17 deletions(-) diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c index b9efb296bba5..7b4629625b1e 100644 --- a/tools/perf/util/parse-events.c +++ b/tools/perf/util/parse-events.c @@ -1117,7 +1117,7 @@ static int config_attr(struct perf_event_attr *attr, static struct evsel_config_term *add_config_term(enum evsel_term_type type, struct list_head *head_terms, - bool weak) + bool weak, char *str, u64 val) { struct evsel_config_term *t; @@ -1128,8 +1128,62 @@ static struct evsel_config_term *add_config_term(enum evsel_term_type type, INIT_LIST_HEAD(&t->list); t->type = type; t->weak = weak; - list_add_tail(&t->list, head_terms); + switch (type) { + case EVSEL__CONFIG_TERM_PERIOD: + case EVSEL__CONFIG_TERM_FREQ: + case EVSEL__CONFIG_TERM_STACK_USER: + case EVSEL__CONFIG_TERM_USR_CHG_CONFIG: + case EVSEL__CONFIG_TERM_USR_CHG_CONFIG1: + case EVSEL__CONFIG_TERM_USR_CHG_CONFIG2: + case EVSEL__CONFIG_TERM_USR_CHG_CONFIG3: + case EVSEL__CONFIG_TERM_USR_CHG_CONFIG4: + t->val.val = val; + break; + case EVSEL__CONFIG_TERM_TIME: + t->val.time = val; + break; + case EVSEL__CONFIG_TERM_INHERIT: + t->val.inherit = val; + break; + case EVSEL__CONFIG_TERM_OVERWRITE: + t->val.overwrite = val; + break; + case EVSEL__CONFIG_TERM_MAX_STACK: + t->val.max_stack = val; + break; + case EVSEL__CONFIG_TERM_MAX_EVENTS: + t->val.max_events = val; + break; + case EVSEL__CONFIG_TERM_PERCORE: + t->val.percore = val; + break; + case EVSEL__CONFIG_TERM_AUX_OUTPUT: + t->val.aux_output = val; + break; + case EVSEL__CONFIG_TERM_AUX_SAMPLE_SIZE: + t->val.aux_sample_size = val; + break; + case EVSEL__CONFIG_TERM_CALLGRAPH: + case EVSEL__CONFIG_TERM_BRANCH: + case EVSEL__CONFIG_TERM_DRV_CFG: + case EVSEL__CONFIG_TERM_RATIO_TO_PREV: + case EVSEL__CONFIG_TERM_AUX_ACTION: + if (str) { + t->val.str = strdup(str); + if (!t->val.str) { + zfree(&t); + return NULL; + } + t->free_str = true; + } + break; + default: + t->val.val = val; + break; + } + + list_add_tail(&t->list, head_terms); return t; } @@ -1142,7 +1196,7 @@ static int get_config_terms(const struct parse_events_terms *head_config, struct evsel_config_term *new_term; enum evsel_term_type new_type; bool str_type = false; - u64 val; + u64 val = 0; switch (term->type_term) { case PARSE_EVENTS__TERM_TYPE_SAMPLE_PERIOD: @@ -1234,20 +1288,15 @@ static int get_config_terms(const struct parse_events_terms *head_config, continue; } - new_term = add_config_term(new_type, head_terms, term->weak); + /* + * Note: Members evsel_config_term::val and + * parse_events_term::val are unions and endianness needs + * to be taken into account when changing such union members. + */ + new_term = add_config_term(new_type, head_terms, term->weak, + str_type ? term->val.str : NULL, val); if (!new_term) return -ENOMEM; - - if (str_type) { - new_term->val.str = strdup(term->val.str); - if (!new_term->val.str) { - zfree(&new_term); - return -ENOMEM; - } - new_term->free_str = true; - } else { - new_term->val.val = val; - } } return 0; } @@ -1277,10 +1326,9 @@ static int add_cfg_chg(const struct perf_pmu *pmu, if (bits) { struct evsel_config_term *new_term; - new_term = add_config_term(new_term_type, head_terms, false); + new_term = add_config_term(new_term_type, head_terms, false, NULL, bits); if (!new_term) return -ENOMEM; - new_term->val.cfg_chg = bits; } return 0; From 563d39928db602c58d24301769e25e33a48a65ab Mon Sep 17 00:00:00 2001 From: Leo Yan Date: Sun, 1 Mar 2026 17:43:25 +0000 Subject: [PATCH 165/352] perf kvm stat: Fix relative paths for including headers Add an extra "../" to the relative paths so that the uAPI headers provided by tools can be found correctly. Fixes: a724a8fce5e25b45 ("perf kvm stat: Fix build error") Reported-by: Namhyung Kim Suggested-by: Ian Rogers Reviewed-by: Ian Rogers Signed-off-by: Leo Yan Signed-off-by: Namhyung Kim Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/util/kvm-stat-arch/kvm-stat-x86.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/tools/perf/util/kvm-stat-arch/kvm-stat-x86.c b/tools/perf/util/kvm-stat-arch/kvm-stat-x86.c index 43275d25b6cb..0f626db3a439 100644 --- a/tools/perf/util/kvm-stat-arch/kvm-stat-x86.c +++ b/tools/perf/util/kvm-stat-arch/kvm-stat-x86.c @@ -4,9 +4,9 @@ #include "../kvm-stat.h" #include "../evsel.h" #include "../env.h" -#include "../../arch/x86/include/uapi/asm/svm.h" -#include "../../arch/x86/include/uapi/asm/vmx.h" -#include "../../arch/x86/include/uapi/asm/kvm.h" +#include "../../../arch/x86/include/uapi/asm/svm.h" +#include "../../../arch/x86/include/uapi/asm/vmx.h" +#include "../../../arch/x86/include/uapi/asm/kvm.h" #include define_exit_reasons_table(vmx_exit_reasons, VMX_EXIT_REASONS); From e9f58ff991dd4be13fd7a651bbf64329c090af09 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 16 Mar 2026 11:04:46 -0400 Subject: [PATCH 166/352] drm/amdgpu: rework how we handle TLB fences MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a new VM flag to indicate whether or not we need a TLB fence. Userqs (KFD or KGD) require a TLB fence. A TLB fence is not strictly required for kernel queues, but it shouldn't hurt. That said, enabling this unconditionally should be fine, but it seems to tickle some issues in KIQ/MES. Only enable them for KFD, or when KGD userq queues are enabled (currently via module parameter). Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4798 Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4749 Fixes: f3854e04b708 ("drm/amdgpu: attach tlb fence to the PTs update") Cc: Christian König Cc: Prike Liang Reviewed-by: Prike Liang Signed-off-by: Alex Deucher (cherry picked from commit 69c5fbd2b93b5ced77c6e79afe83371bca84c788) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 7 ++++++- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h | 2 ++ 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index f2beb980e3c3..c60cbce356cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -1069,7 +1069,10 @@ amdgpu_vm_tlb_flush(struct amdgpu_vm_update_params *params, } /* Prepare a TLB flush fence to be attached to PTs */ - if (!params->unlocked) { + /* The check for need_tlb_fence should be dropped once we + * sort out the issues with KIQ/MES TLB invalidation timeouts. + */ + if (!params->unlocked && vm->need_tlb_fence) { amdgpu_vm_tlb_fence_create(params->adev, vm, fence); /* Makes sure no PD/PT is freed before the flush */ @@ -2602,6 +2605,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm, ttm_lru_bulk_move_init(&vm->lru_bulk_move); vm->is_compute_context = false; + vm->need_tlb_fence = amdgpu_userq_enabled(&adev->ddev); vm->use_cpu_for_update = !!(adev->vm_manager.vm_update_mode & AMDGPU_VM_USE_CPU_FOR_GFX); @@ -2739,6 +2743,7 @@ int amdgpu_vm_make_compute(struct amdgpu_device *adev, struct amdgpu_vm *vm) dma_fence_put(vm->last_update); vm->last_update = dma_fence_get_stub(); vm->is_compute_context = true; + vm->need_tlb_fence = true; unreserve_bo: amdgpu_bo_unreserve(vm->root.bo); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h index 806d62ed61ef..bb276c0ad06d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h @@ -441,6 +441,8 @@ struct amdgpu_vm { struct ttm_lru_bulk_move lru_bulk_move; /* Flag to indicate if VM is used for compute */ bool is_compute_context; + /* Flag to indicate if VM needs a TLB fence (KFD or KGD) */ + bool need_tlb_fence; /* Memory partition number, -1 means any partition */ int8_t mem_id; From 9787f7da186ee8143b7b6d914cfa0b6e7fee2648 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 25 Sep 2023 10:44:07 -0400 Subject: [PATCH 167/352] drm/amdgpu: apply state adjust rules to some additional HAINAN vairants They need a similar workaround. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/1839 Signed-off-by: Alex Deucher (cherry picked from commit 0de31d92a173d3d94f28051b0b80a6c98913aed4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) 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 61b1c5aa74cb..36942467d4ad 100644 --- a/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c +++ b/drivers/gpu/drm/amd/pm/legacy-dpm/si_dpm.c @@ -3454,9 +3454,11 @@ static void si_apply_state_adjust_rules(struct amdgpu_device *adev, if (adev->asic_type == CHIP_HAINAN) { if ((adev->pdev->revision == 0x81) || (adev->pdev->revision == 0xC3) || + (adev->pdev->device == 0x6660) || (adev->pdev->device == 0x6664) || (adev->pdev->device == 0x6665) || - (adev->pdev->device == 0x6667)) { + (adev->pdev->device == 0x6667) || + (adev->pdev->device == 0x666F)) { max_sclk = 75000; } if ((adev->pdev->revision == 0xC3) || From 86650ee2241ff84207eaa298ab318533f3c21a38 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Mon, 25 Sep 2023 10:44:06 -0400 Subject: [PATCH 168/352] drm/radeon: apply state adjust rules to some additional HAINAN vairants They need a similar workaround. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/1839 Signed-off-by: Alex Deucher (cherry picked from commit 87327658c848f56eac166cb382b57b83bf06c5ac) Cc: stable@vger.kernel.org --- drivers/gpu/drm/radeon/si_dpm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c index b4aa49b1ac63..4b10715f951c 100644 --- a/drivers/gpu/drm/radeon/si_dpm.c +++ b/drivers/gpu/drm/radeon/si_dpm.c @@ -2915,9 +2915,11 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev, if (rdev->family == CHIP_HAINAN) { if ((rdev->pdev->revision == 0x81) || (rdev->pdev->revision == 0xC3) || + (rdev->pdev->device == 0x6660) || (rdev->pdev->device == 0x6664) || (rdev->pdev->device == 0x6665) || - (rdev->pdev->device == 0x6667)) { + (rdev->pdev->device == 0x6667) || + (rdev->pdev->device == 0x666F)) { max_sclk = 75000; } if ((rdev->pdev->revision == 0xC3) || From c925fccc4f8fae4354d98b2af606bd4747d3738d Mon Sep 17 00:00:00 2001 From: Sanman Pradhan Date: Tue, 17 Mar 2026 17:36:53 +0000 Subject: [PATCH 169/352] hwmon: (pmbus/hac300s) Add error check for pmbus_read_word_data() return value hac300s_read_word_data() passes the return value of pmbus_read_word_data() directly to FIELD_GET() without checking for errors. If the I2C transaction fails, a negative error code is sign-extended and passed to FIELD_GET(), which silently produces garbage data instead of propagating the error. Add the missing error check before using the return value in the FIELD_GET() macro. Fixes: 669cf162f7a1 ("hwmon: Add support for HiTRON HAC300S PSU") Cc: stable@vger.kernel.org Signed-off-by: Sanman Pradhan Link: https://lore.kernel.org/r/20260317173308.382545-2-sanman.pradhan@hpe.com Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/hac300s.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hwmon/pmbus/hac300s.c b/drivers/hwmon/pmbus/hac300s.c index 0a1d52cae91e..a073db1cfe2e 100644 --- a/drivers/hwmon/pmbus/hac300s.c +++ b/drivers/hwmon/pmbus/hac300s.c @@ -58,6 +58,8 @@ static int hac300s_read_word_data(struct i2c_client *client, int page, case PMBUS_MFR_VOUT_MIN: case PMBUS_READ_VOUT: rv = pmbus_read_word_data(client, page, phase, reg); + if (rv < 0) + return rv; return FIELD_GET(LINEAR11_MANTISSA_MASK, rv); default: return -ENODATA; From 19d4b9c8a136704d5f2544e7ac550f27918a5004 Mon Sep 17 00:00:00 2001 From: Sanman Pradhan Date: Tue, 17 Mar 2026 17:37:17 +0000 Subject: [PATCH 170/352] hwmon: (pmbus/mp2975) Add error check for pmbus_read_word_data() return value mp2973_read_word_data() XORs the return value of pmbus_read_word_data() with PB_STATUS_POWER_GOOD_N without first checking for errors. If the I2C transaction fails, a negative error code is XORed with the constant, producing a corrupted value that is returned as valid status data instead of propagating the error. Add the missing error check before modifying the return value. Fixes: acda945afb465 ("hwmon: (pmbus/mp2975) Fix PGOOD in READ_STATUS_WORD") Cc: stable@vger.kernel.org Signed-off-by: Sanman Pradhan Link: https://lore.kernel.org/r/20260317173308.382545-3-sanman.pradhan@hpe.com Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/mp2975.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hwmon/pmbus/mp2975.c b/drivers/hwmon/pmbus/mp2975.c index c31982d85196..d0bc47b12cb0 100644 --- a/drivers/hwmon/pmbus/mp2975.c +++ b/drivers/hwmon/pmbus/mp2975.c @@ -313,6 +313,8 @@ static int mp2973_read_word_data(struct i2c_client *client, int page, case PMBUS_STATUS_WORD: /* MP2973 & MP2971 return PGOOD instead of PB_STATUS_POWER_GOOD_N. */ ret = pmbus_read_word_data(client, page, phase, reg); + if (ret < 0) + return ret; ret ^= PB_STATUS_POWER_GOOD_N; break; case PMBUS_OT_FAULT_LIMIT: From c6f45ed26b6eb4766db06f21ff28a97ed485bcbb Mon Sep 17 00:00:00 2001 From: Sanman Pradhan Date: Tue, 17 Mar 2026 17:37:41 +0000 Subject: [PATCH 171/352] hwmon: (pmbus/mp2869) Check pmbus_read_byte_data() before using its return value In mp2869_read_byte_data() and mp2869_read_word_data(), the return value of pmbus_read_byte_data() for PMBUS_STATUS_MFR_SPECIFIC is used directly inside FIELD_GET() macro arguments without error checking. If the I2C transaction fails, a negative error code is passed to FIELD_GET() and FIELD_PREP(), silently corrupting the status register bits being constructed. Extract the nested pmbus_read_byte_data() calls into a separate variable and check for errors before use. This also eliminates a redundant duplicate read of the same register in the PMBUS_STATUS_TEMPERATURE case. Fixes: a3a2923aaf7f2 ("hwmon: add MP2869,MP29608,MP29612 and MP29816 series driver") Cc: stable@vger.kernel.org Signed-off-by: Sanman Pradhan Link: https://lore.kernel.org/r/20260317173308.382545-4-sanman.pradhan@hpe.com Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/mp2869.c | 35 +++++++++++++++++++++-------------- 1 file changed, 21 insertions(+), 14 deletions(-) diff --git a/drivers/hwmon/pmbus/mp2869.c b/drivers/hwmon/pmbus/mp2869.c index cc69a1e91dfe..4647892e5112 100644 --- a/drivers/hwmon/pmbus/mp2869.c +++ b/drivers/hwmon/pmbus/mp2869.c @@ -165,7 +165,7 @@ static int mp2869_read_byte_data(struct i2c_client *client, int page, int reg) { const struct pmbus_driver_info *info = pmbus_get_driver_info(client); struct mp2869_data *data = to_mp2869_data(info); - int ret; + int ret, mfr; switch (reg) { case PMBUS_VOUT_MODE: @@ -188,11 +188,14 @@ static int mp2869_read_byte_data(struct i2c_client *client, int page, int reg) if (ret < 0) return ret; + mfr = pmbus_read_byte_data(client, page, + PMBUS_STATUS_MFR_SPECIFIC); + if (mfr < 0) + return mfr; + ret = (ret & ~GENMASK(2, 2)) | FIELD_PREP(GENMASK(2, 2), - FIELD_GET(GENMASK(1, 1), - pmbus_read_byte_data(client, page, - PMBUS_STATUS_MFR_SPECIFIC))); + FIELD_GET(GENMASK(1, 1), mfr)); break; case PMBUS_STATUS_TEMPERATURE: /* @@ -207,15 +210,16 @@ static int mp2869_read_byte_data(struct i2c_client *client, int page, int reg) if (ret < 0) return ret; + mfr = pmbus_read_byte_data(client, page, + PMBUS_STATUS_MFR_SPECIFIC); + if (mfr < 0) + return mfr; + ret = (ret & ~GENMASK(7, 6)) | FIELD_PREP(GENMASK(6, 6), - FIELD_GET(GENMASK(1, 1), - pmbus_read_byte_data(client, page, - PMBUS_STATUS_MFR_SPECIFIC))) | + FIELD_GET(GENMASK(1, 1), mfr)) | FIELD_PREP(GENMASK(7, 7), - FIELD_GET(GENMASK(1, 1), - pmbus_read_byte_data(client, page, - PMBUS_STATUS_MFR_SPECIFIC))); + FIELD_GET(GENMASK(1, 1), mfr)); break; default: ret = -ENODATA; @@ -230,7 +234,7 @@ static int mp2869_read_word_data(struct i2c_client *client, int page, int phase, { const struct pmbus_driver_info *info = pmbus_get_driver_info(client); struct mp2869_data *data = to_mp2869_data(info); - int ret; + int ret, mfr; switch (reg) { case PMBUS_STATUS_WORD: @@ -246,11 +250,14 @@ static int mp2869_read_word_data(struct i2c_client *client, int page, int phase, if (ret < 0) return ret; + mfr = pmbus_read_byte_data(client, page, + PMBUS_STATUS_MFR_SPECIFIC); + if (mfr < 0) + return mfr; + ret = (ret & ~GENMASK(2, 2)) | FIELD_PREP(GENMASK(2, 2), - FIELD_GET(GENMASK(1, 1), - pmbus_read_byte_data(client, page, - PMBUS_STATUS_MFR_SPECIFIC))); + FIELD_GET(GENMASK(1, 1), mfr)); break; case PMBUS_READ_VIN: /* From 32f59301b9898c0ab5e72908556d553e2d481945 Mon Sep 17 00:00:00 2001 From: Sanman Pradhan Date: Tue, 17 Mar 2026 17:46:31 +0000 Subject: [PATCH 172/352] hwmon: (pmbus/ina233) Add error check for pmbus_read_word_data() return value ina233_read_word_data() uses the return value of pmbus_read_word_data() directly in a DIV_ROUND_CLOSEST() computation without first checking for errors. If the underlying I2C transaction fails, a negative error code is used in the arithmetic, producing a garbage sensor value instead of propagating the error. Add the missing error check before using the return value. Fixes: b64b6cb163f16 ("hwmon: Add driver for TI INA233 Current and Power Monitor") Cc: stable@vger.kernel.org Signed-off-by: Sanman Pradhan Link: https://lore.kernel.org/r/20260317174553.385567-1-sanman.pradhan@hpe.com Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/ina233.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/hwmon/pmbus/ina233.c b/drivers/hwmon/pmbus/ina233.c index dde1e1678394..2d8b5a5347ed 100644 --- a/drivers/hwmon/pmbus/ina233.c +++ b/drivers/hwmon/pmbus/ina233.c @@ -67,6 +67,8 @@ static int ina233_read_word_data(struct i2c_client *client, int page, switch (reg) { case PMBUS_VIRT_READ_VMON: ret = pmbus_read_word_data(client, 0, 0xff, MFR_READ_VSHUNT); + if (ret < 0) + return ret; /* Adjust returned value to match VIN coefficients */ /* VIN: 1.25 mV VSHUNT: 2.5 uV LSB */ From 5258572aa5fd5a7ed01b123b28241e0281b6fb9b Mon Sep 17 00:00:00 2001 From: Nicholas Carlini Date: Mon, 9 Mar 2026 08:29:49 +0900 Subject: [PATCH 173/352] ksmbd: fix share_conf UAF in tree_conn disconnect __ksmbd_tree_conn_disconnect() drops the share_conf reference before checking tree_conn->refcount. When someone uses SMB3 multichannel and binds two connections to one session, a SESSION_LOGOFF on connection A calls ksmbd_conn_wait_idle(conn) which only drains connection A's request counter, not connection B's. This means there's a race condition: requests already dispatched on connection B hold tree_conn references via work->tcon. The disconnect path frees share_conf while those requests are still walking work->tcon->share_conf, causing a use-after-free. This fix combines the share_conf put with the tree_conn free so it only happens when the last reference is dropped. Fixes: b39a1833cc4a ("ksmbd: fix use-after-free in ksmbd_tree_connect_put under concurrency") Signed-off-by: Nicholas Carlini Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/mgmt/tree_connect.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/fs/smb/server/mgmt/tree_connect.c b/fs/smb/server/mgmt/tree_connect.c index a72d7e42a6c2..58e5b8592da4 100644 --- a/fs/smb/server/mgmt/tree_connect.c +++ b/fs/smb/server/mgmt/tree_connect.c @@ -102,8 +102,10 @@ ksmbd_tree_conn_connect(struct ksmbd_work *work, const char *share_name) void ksmbd_tree_connect_put(struct ksmbd_tree_connect *tcon) { - if (atomic_dec_and_test(&tcon->refcount)) + if (atomic_dec_and_test(&tcon->refcount)) { + ksmbd_share_config_put(tcon->share_conf); kfree(tcon); + } } static int __ksmbd_tree_conn_disconnect(struct ksmbd_session *sess, @@ -113,10 +115,11 @@ static int __ksmbd_tree_conn_disconnect(struct ksmbd_session *sess, ret = ksmbd_ipc_tree_disconnect_request(sess->id, tree_conn->id); ksmbd_release_tree_conn_id(sess, tree_conn->id); - ksmbd_share_config_put(tree_conn->share_conf); ksmbd_counter_dec(KSMBD_COUNTER_TREE_CONNS); - if (atomic_dec_and_test(&tree_conn->refcount)) + if (atomic_dec_and_test(&tree_conn->refcount)) { + ksmbd_share_config_put(tree_conn->share_conf); kfree(tree_conn); + } return ret; } From 282343cf8a4a5a3603b1cb0e17a7083e4a593b03 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Fri, 13 Mar 2026 10:00:58 +0900 Subject: [PATCH 174/352] ksmbd: unset conn->binding on failed binding request When a multichannel SMB2_SESSION_SETUP request with SMB2_SESSION_REQ_FLAG_BINDING fails ksmbd sets conn->binding = true but never clears it on the error path. This leaves the connection in a binding state where all subsequent ksmbd_session_lookup_all() calls fall back to the global sessions table. This fix it by clearing conn->binding = false in the error path. Cc: stable@vger.kernel.org Reported-by: Hyunwoo Kim Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 1 + 1 file changed, 1 insertion(+) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 9f7ff7491e9a..78d2d79d09b4 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -1948,6 +1948,7 @@ int smb2_sess_setup(struct ksmbd_work *work) } } smb2_set_err_rsp(work); + conn->binding = false; } else { unsigned int iov_len; From 3a64125730cabc34fccfbc230c2667c2e14f7308 Mon Sep 17 00:00:00 2001 From: Namjae Jeon Date: Fri, 13 Mar 2026 10:01:29 +0900 Subject: [PATCH 175/352] ksmbd: use volume UUID in FS_OBJECT_ID_INFORMATION Use sb->s_uuid for a proper volume identifier as the primary choice. For filesystems that do not provide a UUID, fall back to stfs.f_fsid obtained from vfs_statfs(). Cc: stable@vger.kernel.org Reported-by: Hyunwoo Kim Signed-off-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 78d2d79d09b4..796cc2413817 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -5453,7 +5453,6 @@ static int smb2_get_info_filesystem(struct ksmbd_work *work, struct smb2_query_info_req *req, struct smb2_query_info_rsp *rsp) { - struct ksmbd_session *sess = work->sess; struct ksmbd_conn *conn = work->conn; struct ksmbd_share_config *share = work->tcon->share_conf; int fsinfoclass = 0; @@ -5590,10 +5589,11 @@ static int smb2_get_info_filesystem(struct ksmbd_work *work, info = (struct object_id_info *)(rsp->Buffer); - if (!user_guest(sess->user)) - memcpy(info->objid, user_passkey(sess->user), 16); + if (path.mnt->mnt_sb->s_uuid_len == 16) + memcpy(info->objid, path.mnt->mnt_sb->s_uuid.b, + path.mnt->mnt_sb->s_uuid_len); else - memset(info->objid, 0, 16); + memcpy(info->objid, &stfs.f_fsid, sizeof(stfs.f_fsid)); info->extended_info.magic = cpu_to_le32(EXTENDED_INFO_MAGIC); info->extended_info.version = cpu_to_le32(1); From c33615f995aee80657b9fdfbc4ee7f49c2bd733d Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Thu, 12 Mar 2026 17:17:02 +0900 Subject: [PATCH 176/352] ksmbd: fix use-after-free of share_conf in compound request smb2_get_ksmbd_tcon() reuses work->tcon in compound requests without validating tcon->t_state. ksmbd_tree_conn_lookup() checks t_state == TREE_CONNECTED on the initial lookup path, but the compound reuse path bypasses this check entirely. If a prior command in the compound (SMB2_TREE_DISCONNECT) sets t_state to TREE_DISCONNECTED and frees share_conf via ksmbd_share_config_put(), subsequent commands dereference the freed share_conf through work->tcon->share_conf. KASAN report: [ 4.144653] ================================================================== [ 4.145059] BUG: KASAN: slab-use-after-free in smb2_write+0xc74/0xe70 [ 4.145415] Read of size 4 at addr ffff88810430c194 by task kworker/1:1/44 [ 4.145772] [ 4.145867] CPU: 1 UID: 0 PID: 44 Comm: kworker/1:1 Not tainted 7.0.0-rc3+ #60 PREEMPTLAZY [ 4.145871] Hardware name: QEMU Ubuntu 24.04 PC v2 (i440FX + PIIX, arch_caps fix, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 [ 4.145875] Workqueue: ksmbd-io handle_ksmbd_work [ 4.145888] Call Trace: [ 4.145892] [ 4.145894] dump_stack_lvl+0x64/0x80 [ 4.145910] print_report+0xce/0x660 [ 4.145919] ? __pfx__raw_spin_lock_irqsave+0x10/0x10 [ 4.145928] ? smb2_write+0xc74/0xe70 [ 4.145931] kasan_report+0xce/0x100 [ 4.145934] ? smb2_write+0xc74/0xe70 [ 4.145937] smb2_write+0xc74/0xe70 [ 4.145939] ? __pfx_smb2_write+0x10/0x10 [ 4.145942] ? _raw_spin_unlock+0xe/0x30 [ 4.145945] ? ksmbd_smb2_check_message+0xeb2/0x24c0 [ 4.145948] ? smb2_tree_disconnect+0x31c/0x480 [ 4.145951] handle_ksmbd_work+0x40f/0x1080 [ 4.145953] process_one_work+0x5fa/0xef0 [ 4.145962] ? assign_work+0x122/0x3e0 [ 4.145964] worker_thread+0x54b/0xf70 [ 4.145967] ? __pfx_worker_thread+0x10/0x10 [ 4.145970] kthread+0x346/0x470 [ 4.145976] ? recalc_sigpending+0x19b/0x230 [ 4.145980] ? __pfx_kthread+0x10/0x10 [ 4.145984] ret_from_fork+0x4fb/0x6c0 [ 4.145992] ? __pfx_ret_from_fork+0x10/0x10 [ 4.145995] ? __switch_to+0x36c/0xbe0 [ 4.145999] ? __pfx_kthread+0x10/0x10 [ 4.146003] ret_from_fork_asm+0x1a/0x30 [ 4.146013] [ 4.146014] [ 4.149858] Allocated by task 44: [ 4.149953] kasan_save_stack+0x33/0x60 [ 4.150061] kasan_save_track+0x14/0x30 [ 4.150169] __kasan_kmalloc+0x8f/0xa0 [ 4.150274] ksmbd_share_config_get+0x1dd/0xdd0 [ 4.150401] ksmbd_tree_conn_connect+0x7e/0x600 [ 4.150529] smb2_tree_connect+0x2e6/0x1000 [ 4.150645] handle_ksmbd_work+0x40f/0x1080 [ 4.150761] process_one_work+0x5fa/0xef0 [ 4.150873] worker_thread+0x54b/0xf70 [ 4.150978] kthread+0x346/0x470 [ 4.151071] ret_from_fork+0x4fb/0x6c0 [ 4.151176] ret_from_fork_asm+0x1a/0x30 [ 4.151286] [ 4.151332] Freed by task 44: [ 4.151418] kasan_save_stack+0x33/0x60 [ 4.151526] kasan_save_track+0x14/0x30 [ 4.151634] kasan_save_free_info+0x3b/0x60 [ 4.151751] __kasan_slab_free+0x43/0x70 [ 4.151861] kfree+0x1ca/0x430 [ 4.151952] __ksmbd_tree_conn_disconnect+0xc8/0x190 [ 4.152088] smb2_tree_disconnect+0x1cd/0x480 [ 4.152211] handle_ksmbd_work+0x40f/0x1080 [ 4.152326] process_one_work+0x5fa/0xef0 [ 4.152438] worker_thread+0x54b/0xf70 [ 4.152545] kthread+0x346/0x470 [ 4.152638] ret_from_fork+0x4fb/0x6c0 [ 4.152743] ret_from_fork_asm+0x1a/0x30 [ 4.152853] [ 4.152900] The buggy address belongs to the object at ffff88810430c180 [ 4.152900] which belongs to the cache kmalloc-96 of size 96 [ 4.153226] The buggy address is located 20 bytes inside of [ 4.153226] freed 96-byte region [ffff88810430c180, ffff88810430c1e0) [ 4.153549] [ 4.153596] The buggy address belongs to the physical page: [ 4.153750] page: refcount:0 mapcount:0 mapping:0000000000000000 index:0xffff88810430ce80 pfn:0x10430c [ 4.154000] flags: 0x100000000000200(workingset|node=0|zone=2) [ 4.154160] page_type: f5(slab) [ 4.154251] raw: 0100000000000200 ffff888100041280 ffff888100040110 ffff888100040110 [ 4.154461] raw: ffff88810430ce80 0000000800200009 00000000f5000000 0000000000000000 [ 4.154668] page dumped because: kasan: bad access detected [ 4.154820] [ 4.154866] Memory state around the buggy address: [ 4.155002] ffff88810430c080: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 4.155196] ffff88810430c100: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 4.155391] >ffff88810430c180: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc [ 4.155587] ^ [ 4.155693] ffff88810430c200: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 4.155891] ffff88810430c280: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 4.156087] ================================================================== Add the same t_state validation to the compound reuse path, consistent with ksmbd_tree_conn_lookup(). Fixes: 5005bcb42191 ("ksmbd: validate session id and tree id in the compound request") Signed-off-by: Hyunwoo Kim Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 796cc2413817..0854f238b278 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -126,6 +126,8 @@ int smb2_get_ksmbd_tcon(struct ksmbd_work *work) pr_err("The first operation in the compound does not have tcon\n"); return -EINVAL; } + if (work->tcon->t_state != TREE_CONNECTED) + return -ENOENT; if (tree_id != UINT_MAX && work->tcon->id != tree_id) { pr_err("tree id(%u) is different with id(%u) in first operation\n", tree_id, work->tcon->id); From b425e4d0eb321a1116ddbf39636333181675d8f4 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Thu, 12 Mar 2026 17:15:51 +0900 Subject: [PATCH 177/352] ksmbd: fix use-after-free in durable v2 replay of active file handles parse_durable_handle_context() unconditionally assigns dh_info->fp->conn to the current connection when handling a DURABLE_REQ_V2 context with SMB2_FLAGS_REPLAY_OPERATION. ksmbd_lookup_fd_cguid() does not filter by fp->conn, so it returns file handles that are already actively connected. The unconditional overwrite replaces fp->conn, and when the overwriting connection is subsequently freed, __ksmbd_close_fd() dereferences the stale fp->conn via spin_lock(&fp->conn->llist_lock), causing a use-after-free. KASAN report: [ 7.349357] ================================================================== [ 7.349607] BUG: KASAN: slab-use-after-free in _raw_spin_lock+0x75/0xe0 [ 7.349811] Write of size 4 at addr ffff8881056ac18c by task kworker/1:2/108 [ 7.350010] [ 7.350064] CPU: 1 UID: 0 PID: 108 Comm: kworker/1:2 Not tainted 7.0.0-rc3+ #58 PREEMPTLAZY [ 7.350068] Hardware name: QEMU Ubuntu 24.04 PC v2 (i440FX + PIIX, arch_caps fix, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 [ 7.350070] Workqueue: ksmbd-io handle_ksmbd_work [ 7.350083] Call Trace: [ 7.350087] [ 7.350087] dump_stack_lvl+0x64/0x80 [ 7.350094] print_report+0xce/0x660 [ 7.350100] ? __pfx__raw_spin_lock_irqsave+0x10/0x10 [ 7.350101] ? __pfx___mod_timer+0x10/0x10 [ 7.350106] ? _raw_spin_lock+0x75/0xe0 [ 7.350108] kasan_report+0xce/0x100 [ 7.350109] ? _raw_spin_lock+0x75/0xe0 [ 7.350114] kasan_check_range+0x105/0x1b0 [ 7.350116] _raw_spin_lock+0x75/0xe0 [ 7.350118] ? __pfx__raw_spin_lock+0x10/0x10 [ 7.350119] ? __call_rcu_common.constprop.0+0x25e/0x780 [ 7.350125] ? close_id_del_oplock+0x2cc/0x4e0 [ 7.350128] __ksmbd_close_fd+0x27f/0xaf0 [ 7.350131] ksmbd_close_fd+0x135/0x1b0 [ 7.350133] smb2_close+0xb19/0x15b0 [ 7.350142] ? __pfx_smb2_close+0x10/0x10 [ 7.350143] ? xas_load+0x18/0x270 [ 7.350146] ? _raw_spin_lock+0x84/0xe0 [ 7.350148] ? __pfx__raw_spin_lock+0x10/0x10 [ 7.350150] ? _raw_spin_unlock+0xe/0x30 [ 7.350151] ? ksmbd_smb2_check_message+0xeb2/0x24c0 [ 7.350153] ? ksmbd_tree_conn_lookup+0xcd/0xf0 [ 7.350154] handle_ksmbd_work+0x40f/0x1080 [ 7.350156] process_one_work+0x5fa/0xef0 [ 7.350162] ? assign_work+0x122/0x3e0 [ 7.350163] worker_thread+0x54b/0xf70 [ 7.350165] ? __pfx_worker_thread+0x10/0x10 [ 7.350166] kthread+0x346/0x470 [ 7.350170] ? recalc_sigpending+0x19b/0x230 [ 7.350176] ? __pfx_kthread+0x10/0x10 [ 7.350178] ret_from_fork+0x4fb/0x6c0 [ 7.350183] ? __pfx_ret_from_fork+0x10/0x10 [ 7.350185] ? __switch_to+0x36c/0xbe0 [ 7.350188] ? __pfx_kthread+0x10/0x10 [ 7.350190] ret_from_fork_asm+0x1a/0x30 [ 7.350197] [ 7.350197] [ 7.355160] Allocated by task 123: [ 7.355261] kasan_save_stack+0x33/0x60 [ 7.355373] kasan_save_track+0x14/0x30 [ 7.355484] __kasan_kmalloc+0x8f/0xa0 [ 7.355593] ksmbd_conn_alloc+0x44/0x6d0 [ 7.355711] ksmbd_kthread_fn+0x243/0xd70 [ 7.355839] kthread+0x346/0x470 [ 7.355942] ret_from_fork+0x4fb/0x6c0 [ 7.356051] ret_from_fork_asm+0x1a/0x30 [ 7.356164] [ 7.356214] Freed by task 134: [ 7.356305] kasan_save_stack+0x33/0x60 [ 7.356416] kasan_save_track+0x14/0x30 [ 7.356527] kasan_save_free_info+0x3b/0x60 [ 7.356646] __kasan_slab_free+0x43/0x70 [ 7.356761] kfree+0x1ca/0x430 [ 7.356862] ksmbd_tcp_disconnect+0x59/0xe0 [ 7.356993] ksmbd_conn_handler_loop+0x77e/0xd40 [ 7.357138] kthread+0x346/0x470 [ 7.357240] ret_from_fork+0x4fb/0x6c0 [ 7.357350] ret_from_fork_asm+0x1a/0x30 [ 7.357463] [ 7.357513] The buggy address belongs to the object at ffff8881056ac000 [ 7.357513] which belongs to the cache kmalloc-1k of size 1024 [ 7.357857] The buggy address is located 396 bytes inside of [ 7.357857] freed 1024-byte region [ffff8881056ac000, ffff8881056ac400) Fix by removing the unconditional fp->conn assignment and rejecting the replay when fp->conn is non-NULL. This is consistent with ksmbd_lookup_durable_fd(), which also rejects file handles with a non-NULL fp->conn. For disconnected file handles (fp->conn == NULL), ksmbd_reopen_durable_fd() handles setting fp->conn. Fixes: c8efcc786146 ("ksmbd: add support for durable handles v1/v2") Signed-off-by: Hyunwoo Kim Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 0854f238b278..9c44e71e3c3b 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -2831,7 +2831,11 @@ static int parse_durable_handle_context(struct ksmbd_work *work, goto out; } - dh_info->fp->conn = conn; + if (dh_info->fp->conn) { + ksmbd_put_durable_fd(dh_info->fp); + err = -EBADF; + goto out; + } dh_info->reconnected = true; goto out; } From 0162ab3220bac870e43e229e6e3024d1a21c3f26 Mon Sep 17 00:00:00 2001 From: Rahul Bukte Date: Tue, 3 Feb 2026 10:18:39 +0530 Subject: [PATCH 178/352] drm/i915/gt: Check set_default_submission() before deferencing When the i915 driver firmware binaries are not present, the set_default_submission pointer is not set. This pointer is dereferenced during suspend anyways. Add a check to make sure it is set before dereferencing. [ 23.289926] PM: suspend entry (deep) [ 23.293558] Filesystems sync: 0.000 seconds [ 23.298010] Freezing user space processes [ 23.302771] Freezing user space processes completed (elapsed 0.000 seconds) [ 23.309766] OOM killer disabled. [ 23.313027] Freezing remaining freezable tasks [ 23.318540] Freezing remaining freezable tasks completed (elapsed 0.001 seconds) [ 23.342038] serial 00:05: disabled [ 23.345719] serial 00:02: disabled [ 23.349342] serial 00:01: disabled [ 23.353782] sd 0:0:0:0: [sda] Synchronizing SCSI cache [ 23.358993] sd 1:0:0:0: [sdb] Synchronizing SCSI cache [ 23.361635] ata1.00: Entering standby power mode [ 23.368863] ata2.00: Entering standby power mode [ 23.445187] BUG: kernel NULL pointer dereference, address: 0000000000000000 [ 23.452194] #PF: supervisor instruction fetch in kernel mode [ 23.457896] #PF: error_code(0x0010) - not-present page [ 23.463065] PGD 0 P4D 0 [ 23.465640] Oops: Oops: 0010 [#1] SMP NOPTI [ 23.469869] CPU: 8 UID: 0 PID: 211 Comm: kworker/u48:18 Tainted: G S W 6.19.0-rc4-00020-gf0b9d8eb98df #10 PREEMPT(voluntary) [ 23.482512] Tainted: [S]=CPU_OUT_OF_SPEC, [W]=WARN [ 23.496511] Workqueue: async async_run_entry_fn [ 23.501087] RIP: 0010:0x0 [ 23.503755] Code: Unable to access opcode bytes at 0xffffffffffffffd6. [ 23.510324] RSP: 0018:ffffb4a60065fca8 EFLAGS: 00010246 [ 23.515592] RAX: 0000000000000000 RBX: ffff9f428290e000 RCX: 000000000000000f [ 23.522765] RDX: 0000000000000000 RSI: 0000000000000282 RDI: ffff9f428290e000 [ 23.529937] RBP: ffff9f4282907070 R08: ffff9f4281130428 R09: 00000000ffffffff [ 23.537111] R10: 0000000000000000 R11: 0000000000000001 R12: ffff9f42829070f8 [ 23.544284] R13: ffff9f4282906028 R14: ffff9f4282900000 R15: ffff9f4282906b68 [ 23.551457] FS: 0000000000000000(0000) GS:ffff9f466b2cf000(0000) knlGS:0000000000000000 [ 23.559588] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 23.565365] CR2: ffffffffffffffd6 CR3: 000000031c230001 CR4: 0000000000f70ef0 [ 23.572539] PKRU: 55555554 [ 23.575281] Call Trace: [ 23.577770] [ 23.579905] intel_engines_reset_default_submission+0x42/0x60 [ 23.585695] __intel_gt_unset_wedged+0x191/0x200 [ 23.590360] intel_gt_unset_wedged+0x20/0x40 [ 23.594675] gt_sanitize+0x15e/0x170 [ 23.598290] i915_gem_suspend_late+0x6b/0x180 [ 23.602692] i915_drm_suspend_late+0x35/0xf0 [ 23.607008] ? __pfx_pci_pm_suspend_late+0x10/0x10 [ 23.611843] dpm_run_callback+0x78/0x1c0 [ 23.615817] device_suspend_late+0xde/0x2e0 [ 23.620037] async_suspend_late+0x18/0x30 [ 23.624082] async_run_entry_fn+0x25/0xa0 [ 23.628129] process_one_work+0x15b/0x380 [ 23.632182] worker_thread+0x2a5/0x3c0 [ 23.635973] ? __pfx_worker_thread+0x10/0x10 [ 23.640279] kthread+0xf6/0x1f0 [ 23.643464] ? __pfx_kthread+0x10/0x10 [ 23.647263] ? __pfx_kthread+0x10/0x10 [ 23.651045] ret_from_fork+0x131/0x190 [ 23.654837] ? __pfx_kthread+0x10/0x10 [ 23.658634] ret_from_fork_asm+0x1a/0x30 [ 23.662597] [ 23.664826] Modules linked in: [ 23.667914] CR2: 0000000000000000 [ 23.671271] ------------[ cut here ]------------ Signed-off-by: Rahul Bukte Reviewed-by: Suraj Kandpal Signed-off-by: Suraj Kandpal Link: https://patch.msgid.link/20260203044839.1555147-1-suraj.kandpal@intel.com (cherry picked from commit daa199abc3d3d1740c9e3a2c3e9216ae5b447cad) Fixes: ff44ad51ebf8 ("drm/i915: Move engine->submit_request selection to a vfunc") Signed-off-by: Joonas Lahtinen --- drivers/gpu/drm/i915/gt/intel_engine_cs.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c b/drivers/gpu/drm/i915/gt/intel_engine_cs.c index d37966ec7a92..54c9571327e7 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c @@ -1967,7 +1967,8 @@ void intel_engines_reset_default_submission(struct intel_gt *gt) if (engine->sanitize) engine->sanitize(engine); - engine->set_default_submission(engine); + if (engine->set_default_submission) + engine->set_default_submission(engine); } } From 9232fa05921bc0ea0bc71947440ec9a50b3ad26e Mon Sep 17 00:00:00 2001 From: Michael Riesch Date: Mon, 16 Feb 2026 13:28:47 +0100 Subject: [PATCH 179/352] media: synopsys: csi2rx: fix out-of-bounds check for formats array The out-of-bounds check for the format array is off by one. Fix the check. Fixes: 355a11004066 ("media: synopsys: add driver for the designware mipi csi-2 receiver") Cc: stable@kernel.org Suggested-by: Dan Carpenter Signed-off-by: Michael Riesch Reviewed-by: Frank Li Signed-off-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/synopsys/dw-mipi-csi2rx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/platform/synopsys/dw-mipi-csi2rx.c b/drivers/media/platform/synopsys/dw-mipi-csi2rx.c index 170346ae1a59..4d96171a650b 100644 --- a/drivers/media/platform/synopsys/dw-mipi-csi2rx.c +++ b/drivers/media/platform/synopsys/dw-mipi-csi2rx.c @@ -301,7 +301,7 @@ dw_mipi_csi2rx_enum_mbus_code(struct v4l2_subdev *sd, return 0; case DW_MIPI_CSI2RX_PAD_SINK: - if (code->index > csi2->formats_num) + if (code->index >= csi2->formats_num) return -EINVAL; code->code = csi2->formats[code->index].code; From ac62a20035ecc18e6d365c6c792f5965ce1da77c Mon Sep 17 00:00:00 2001 From: Michael Riesch Date: Mon, 16 Feb 2026 13:28:48 +0100 Subject: [PATCH 180/352] media: synopsys: csi2rx: add missing kconfig dependency Fix "ERROR: modpost: "phy_mipi_dphy_get_default_config_for_hsclk" [drivers/media/platform/synopsys/dw-mipi-csi2rx.ko] undefined!" by selecting GENERIC_PHY_MIPI_DPHY in the Kconfig entry. Fixes: 355a11004066 ("media: synopsys: add driver for the designware mipi csi-2 receiver") Cc: stable@kernel.org Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202602130253.BZnVd4jh-lkp@intel.com/ Signed-off-by: Michael Riesch Reviewed-by: Frank Li Signed-off-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/synopsys/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/media/platform/synopsys/Kconfig b/drivers/media/platform/synopsys/Kconfig index e798ec00b189..bf2ac092fbb3 100644 --- a/drivers/media/platform/synopsys/Kconfig +++ b/drivers/media/platform/synopsys/Kconfig @@ -7,6 +7,7 @@ config VIDEO_DW_MIPI_CSI2RX depends on VIDEO_DEV depends on V4L_PLATFORM_DRIVERS depends on PM && COMMON_CLK + select GENERIC_PHY_MIPI_DPHY select MEDIA_CONTROLLER select V4L2_FWNODE select VIDEO_V4L2_SUBDEV_API From bef4f4a88b73e4cc550d25f665b8a9952af22773 Mon Sep 17 00:00:00 2001 From: Yuchan Nam Date: Fri, 6 Mar 2026 21:52:23 +0900 Subject: [PATCH 181/352] media: mc, v4l2: serialize REINIT and REQBUFS with req_queue_mutex MEDIA_REQUEST_IOC_REINIT can run concurrently with VIDIOC_REQBUFS(0) queue teardown paths. This can race request object cleanup against vb2 queue cancellation and lead to use-after-free reports. We already serialize request queueing against STREAMON/OFF with req_queue_mutex. Extend that serialization to REQBUFS, and also take the same mutex in media_request_ioctl_reinit() so REINIT is in the same exclusion domain. This keeps request cleanup and queue cancellation from running in parallel for request-capable devices. Fixes: 6093d3002eab ("media: vb2: keep a reference to the request until dqbuf") Cc: stable@vger.kernel.org Signed-off-by: Yuchan Nam Signed-off-by: Sakari Ailus Signed-off-by: Mauro Carvalho Chehab --- drivers/media/mc/mc-request.c | 5 +++++ drivers/media/v4l2-core/v4l2-ioctl.c | 5 +++-- 2 files changed, 8 insertions(+), 2 deletions(-) diff --git a/drivers/media/mc/mc-request.c b/drivers/media/mc/mc-request.c index c7c7d4a86c6b..13e77648807c 100644 --- a/drivers/media/mc/mc-request.c +++ b/drivers/media/mc/mc-request.c @@ -192,6 +192,8 @@ static long media_request_ioctl_reinit(struct media_request *req) struct media_device *mdev = req->mdev; unsigned long flags; + mutex_lock(&mdev->req_queue_mutex); + spin_lock_irqsave(&req->lock, flags); if (req->state != MEDIA_REQUEST_STATE_IDLE && req->state != MEDIA_REQUEST_STATE_COMPLETE) { @@ -199,6 +201,7 @@ static long media_request_ioctl_reinit(struct media_request *req) "request: %s not in idle or complete state, cannot reinit\n", req->debug_str); spin_unlock_irqrestore(&req->lock, flags); + mutex_unlock(&mdev->req_queue_mutex); return -EBUSY; } if (req->access_count) { @@ -206,6 +209,7 @@ static long media_request_ioctl_reinit(struct media_request *req) "request: %s is being accessed, cannot reinit\n", req->debug_str); spin_unlock_irqrestore(&req->lock, flags); + mutex_unlock(&mdev->req_queue_mutex); return -EBUSY; } req->state = MEDIA_REQUEST_STATE_CLEANING; @@ -216,6 +220,7 @@ static long media_request_ioctl_reinit(struct media_request *req) spin_lock_irqsave(&req->lock, flags); req->state = MEDIA_REQUEST_STATE_IDLE; spin_unlock_irqrestore(&req->lock, flags); + mutex_unlock(&mdev->req_queue_mutex); return 0; } diff --git a/drivers/media/v4l2-core/v4l2-ioctl.c b/drivers/media/v4l2-core/v4l2-ioctl.c index 37d33d4a363d..a2b650f4ec3c 100644 --- a/drivers/media/v4l2-core/v4l2-ioctl.c +++ b/drivers/media/v4l2-core/v4l2-ioctl.c @@ -3082,13 +3082,14 @@ static long __video_do_ioctl(struct file *file, } /* - * We need to serialize streamon/off with queueing new requests. + * We need to serialize streamon/off/reqbufs with queueing new requests. * These ioctls may trigger the cancellation of a streaming * operation, and that should not be mixed with queueing a new * request at the same time. */ if (v4l2_device_supports_requests(vfd->v4l2_dev) && - (cmd == VIDIOC_STREAMON || cmd == VIDIOC_STREAMOFF)) { + (cmd == VIDIOC_STREAMON || cmd == VIDIOC_STREAMOFF || + cmd == VIDIOC_REQBUFS)) { req_queue_lock = &vfd->v4l2_dev->mdev->req_queue_mutex; if (mutex_lock_interruptible(req_queue_lock)) From daa87ca42652af0d6791ef875e3c4d724b099f22 Mon Sep 17 00:00:00 2001 From: Detlev Casanova Date: Fri, 23 Jan 2026 14:22:44 -0500 Subject: [PATCH 182/352] media: rkvdec: Improve handling missing short/long term RPS The values of ext_sps_st_rps and ext_sps_lt_rps in struct rkvdec_hevc_run are not initialized when the respective controls are not set by userspace. When this is the case, set them to NULL so the rkvdec_hevc_run_preamble function that parses controls does not access garbage data which leads to a panic on unaccessible memory. Fixes: c9a59dc2acc7 ("media: rkvdec: Add HEVC support for the VDPU381 variant") Reported-by: Christian Hewitt Suggested-by: Jonas Karlman Signed-off-by: Detlev Casanova Tested-by: Christian Hewitt Reviewed-by: Nicolas Dufresne Signed-off-by: Nicolas Dufresne Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/rockchip/rkvdec/rkvdec-hevc-common.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec-hevc-common.c b/drivers/media/platform/rockchip/rkvdec/rkvdec-hevc-common.c index 28267ee30190..3119f3bc9f98 100644 --- a/drivers/media/platform/rockchip/rkvdec/rkvdec-hevc-common.c +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec-hevc-common.c @@ -500,11 +500,15 @@ void rkvdec_hevc_run_preamble(struct rkvdec_ctx *ctx, ctrl = v4l2_ctrl_find(&ctx->ctrl_hdl, V4L2_CID_STATELESS_HEVC_EXT_SPS_ST_RPS); run->ext_sps_st_rps = ctrl ? ctrl->p_cur.p : NULL; + } else { + run->ext_sps_st_rps = NULL; } if (ctx->has_sps_lt_rps) { ctrl = v4l2_ctrl_find(&ctx->ctrl_hdl, V4L2_CID_STATELESS_HEVC_EXT_SPS_LT_RPS); run->ext_sps_lt_rps = ctrl ? ctrl->p_cur.p : NULL; + } else { + run->ext_sps_lt_rps = NULL; } rkvdec_run_preamble(ctx, &run->base); From 446c6a25a4494e137ec42e886da04e29efc2dc39 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 2 Feb 2026 10:47:50 +0100 Subject: [PATCH 183/352] media: rkvdec: reduce excessive stack usage in assemble_hw_pps() The rkvdec_pps had a large set of bitfields, all of which as misaligned. This causes clang-21 and likely other versions to produce absolutely awful object code and a warning about very large stack usage, on targets without unaligned access: drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c:966:12: error: stack frame size (1472) exceeds limit (1280) in 'rkvdec_vp9_start' [-Werror,-Wframe-larger-than] Part of the problem here is how all the bitfield accesses are inlined into a function that already has large structures on the stack. Mark set_field_order_cnt() as noinline_for_stack, and split out the following accesses in assemble_hw_pps() into another noinline function, both of which now using around 800 bytes of stack in the same configuration. There is clearly still something wrong with clang here, but splitting it into multiple functions reduces the risk of stack overflow. Fixes: fde24907570d ("media: rkvdec: Add H264 support for the VDPU383 variant") Link: https://godbolt.org/z/acP1eKeq9 Signed-off-by: Arnd Bergmann Reviewed-by: Nicolas Dufresne Signed-off-by: Nicolas Dufresne Signed-off-by: Mauro Carvalho Chehab --- .../rockchip/rkvdec/rkvdec-vdpu383-h264.c | 50 ++++++++++--------- 1 file changed, 27 insertions(+), 23 deletions(-) diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec-vdpu383-h264.c b/drivers/media/platform/rockchip/rkvdec/rkvdec-vdpu383-h264.c index 97f1efde2e47..fb4f849d7366 100644 --- a/drivers/media/platform/rockchip/rkvdec/rkvdec-vdpu383-h264.c +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec-vdpu383-h264.c @@ -130,7 +130,7 @@ struct rkvdec_h264_ctx { struct vdpu383_regs_h26x regs; }; -static void set_field_order_cnt(struct rkvdec_pps *pps, const struct v4l2_h264_dpb_entry *dpb) +static noinline_for_stack void set_field_order_cnt(struct rkvdec_pps *pps, const struct v4l2_h264_dpb_entry *dpb) { pps->top_field_order_cnt0 = dpb[0].top_field_order_cnt; pps->bot_field_order_cnt0 = dpb[0].bottom_field_order_cnt; @@ -166,6 +166,31 @@ static void set_field_order_cnt(struct rkvdec_pps *pps, const struct v4l2_h264_d pps->bot_field_order_cnt15 = dpb[15].bottom_field_order_cnt; } +static noinline_for_stack void set_dec_params(struct rkvdec_pps *pps, const struct v4l2_ctrl_h264_decode_params *dec_params) +{ + const struct v4l2_h264_dpb_entry *dpb = dec_params->dpb; + + for (int i = 0; i < ARRAY_SIZE(dec_params->dpb); i++) { + if (dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_LONG_TERM) + pps->is_longterm |= (1 << i); + pps->ref_field_flags |= + (!!(dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_FIELD)) << i; + pps->ref_colmv_use_flag |= + (!!(dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE)) << i; + pps->ref_topfield_used |= + (!!(dpb[i].fields & V4L2_H264_TOP_FIELD_REF)) << i; + pps->ref_botfield_used |= + (!!(dpb[i].fields & V4L2_H264_BOTTOM_FIELD_REF)) << i; + } + pps->pic_field_flag = + !!(dec_params->flags & V4L2_H264_DECODE_PARAM_FLAG_FIELD_PIC); + pps->pic_associated_flag = + !!(dec_params->flags & V4L2_H264_DECODE_PARAM_FLAG_BOTTOM_FIELD); + + pps->cur_top_field = dec_params->top_field_order_cnt; + pps->cur_bot_field = dec_params->bottom_field_order_cnt; +} + static void assemble_hw_pps(struct rkvdec_ctx *ctx, struct rkvdec_h264_run *run) { @@ -177,7 +202,6 @@ static void assemble_hw_pps(struct rkvdec_ctx *ctx, struct rkvdec_h264_priv_tbl *priv_tbl = h264_ctx->priv_tbl.cpu; struct rkvdec_sps_pps *hw_ps; u32 pic_width, pic_height; - u32 i; /* * HW read the SPS/PPS information from PPS packet index by PPS id. @@ -261,28 +285,8 @@ static void assemble_hw_pps(struct rkvdec_ctx *ctx, !!(pps->flags & V4L2_H264_PPS_FLAG_SCALING_MATRIX_PRESENT); set_field_order_cnt(&hw_ps->pps, dpb); + set_dec_params(&hw_ps->pps, dec_params); - for (i = 0; i < ARRAY_SIZE(dec_params->dpb); i++) { - if (dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_LONG_TERM) - hw_ps->pps.is_longterm |= (1 << i); - - hw_ps->pps.ref_field_flags |= - (!!(dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_FIELD)) << i; - hw_ps->pps.ref_colmv_use_flag |= - (!!(dpb[i].flags & V4L2_H264_DPB_ENTRY_FLAG_ACTIVE)) << i; - hw_ps->pps.ref_topfield_used |= - (!!(dpb[i].fields & V4L2_H264_TOP_FIELD_REF)) << i; - hw_ps->pps.ref_botfield_used |= - (!!(dpb[i].fields & V4L2_H264_BOTTOM_FIELD_REF)) << i; - } - - hw_ps->pps.pic_field_flag = - !!(dec_params->flags & V4L2_H264_DECODE_PARAM_FLAG_FIELD_PIC); - hw_ps->pps.pic_associated_flag = - !!(dec_params->flags & V4L2_H264_DECODE_PARAM_FLAG_BOTTOM_FIELD); - - hw_ps->pps.cur_top_field = dec_params->top_field_order_cnt; - hw_ps->pps.cur_bot_field = dec_params->bottom_field_order_cnt; } static void rkvdec_write_regs(struct rkvdec_ctx *ctx) From c03b7dec3c4ddc97872fa12bfca75bae9cb46510 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 2 Feb 2026 10:47:51 +0100 Subject: [PATCH 184/352] media: rkvdec: reduce stack usage in rkvdec_init_v4l2_vp9_count_tbl() The deeply nested loop in rkvdec_init_v4l2_vp9_count_tbl() needs a lot of registers, so when the clang register allocator runs out, it ends up spilling countless temporaries to the stack: drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c:966:12: error: stack frame size (1472) exceeds limit (1280) in 'rkvdec_vp9_start' [-Werror,-Wframe-larger-than] Marking this function as noinline_for_stack keeps it out of rkvdec_vp9_start(), giving the compiler more room for optimization. The resulting code is good enough that both the total stack usage and the loop get enough better to stay under the warning limit, though it's still slow, and would need a larger rework if this function ends up being called in a fast path. Signed-off-by: Arnd Bergmann Reviewed-by: Nicolas Dufresne Signed-off-by: Nicolas Dufresne Signed-off-by: Mauro Carvalho Chehab --- drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c b/drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c index e4cdd2122873..2751f5396ee8 100644 --- a/drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c +++ b/drivers/media/platform/rockchip/rkvdec/rkvdec-vp9.c @@ -893,7 +893,8 @@ static void rkvdec_vp9_done(struct rkvdec_ctx *ctx, update_ctx_last_info(vp9_ctx); } -static void rkvdec_init_v4l2_vp9_count_tbl(struct rkvdec_ctx *ctx) +static noinline_for_stack void +rkvdec_init_v4l2_vp9_count_tbl(struct rkvdec_ctx *ctx) { struct rkvdec_vp9_ctx *vp9_ctx = ctx->priv; struct rkvdec_vp9_intra_frame_symbol_counts *intra_cnts = vp9_ctx->count_tbl.cpu; From e8d97c270cb46a2a88739019d0f8547adc7d97da Mon Sep 17 00:00:00 2001 From: Ming Qian Date: Fri, 6 Mar 2026 11:10:57 +0800 Subject: [PATCH 185/352] media: verisilicon: Fix kernel panic due to __initconst misuse Fix a kernel panic when probing the driver as a module: Unable to handle kernel paging request at virtual address ffffd9c18eb05000 of_find_matching_node_and_match+0x5c/0x1a0 hantro_probe+0x2f4/0x7d0 [hantro_vpu] The imx8mq_vpu_shared_resources array is referenced by variant structures through their shared_devices field. When built as a module, __initconst causes this data to be freed after module init, but it's later accessed during probe, causing a page fault. The imx8mq_vpu_shared_resources is referenced from non-init code, so keeping __initconst or __initconst_or_module here is wrong. Drop the __initconst annotation and let it live in the normal .rodata section. A bug of __initconst called from regular non-init probe code leading to bugs during probe deferrals or during unbind-bind cycles. Reported-by: Krzysztof Kozlowski Closes: https://lore.kernel.org/all/68ef934f-baa0-4bf6-93d8-834bbc441e66@kernel.org/ Reported-by: Franz Schnyder Closes: https://lore.kernel.org/all/n3qmcb62tepxltoskpf7ws6yiirc2so62ia23b42rj3wlmpl67@rvkbuirx7kkp/ Fixes: e0203ddf9af7 ("media: verisilicon: Avoid G2 bus error while decoding H.264 and HEVC") Suggested-by: Krzysztof Kozlowski Suggested-by: Marco Felsch Reviewed-by: Marco Felsch Signed-off-by: Ming Qian Reviewed-by: Frank Li Reviewed-by: Krzysztof Kozlowski Cc: stable@kernel.org Signed-off-by: Nicolas Dufresne Signed-off-by: Hans Verkuil --- drivers/media/platform/verisilicon/imx8m_vpu_hw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/media/platform/verisilicon/imx8m_vpu_hw.c b/drivers/media/platform/verisilicon/imx8m_vpu_hw.c index 6f8e43b7f157..fa4224de4b99 100644 --- a/drivers/media/platform/verisilicon/imx8m_vpu_hw.c +++ b/drivers/media/platform/verisilicon/imx8m_vpu_hw.c @@ -343,7 +343,7 @@ const struct hantro_variant imx8mq_vpu_variant = { .num_regs = ARRAY_SIZE(imx8mq_reg_names) }; -static const struct of_device_id imx8mq_vpu_shared_resources[] __initconst = { +static const struct of_device_id imx8mq_vpu_shared_resources[] = { { .compatible = "nxp,imx8mq-vpu-g1", }, { .compatible = "nxp,imx8mq-vpu-g2", }, { /* sentinel */ } From 9a6a2091324ab6525951651b3700e3bea0fe9a89 Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Sun, 1 Mar 2026 14:17:39 -0800 Subject: [PATCH 186/352] cxl/mbox: Use proper endpoint validity check upon sanitize Fuzzying CXL triggered: BUG: KASAN: null-ptr-deref in cxl_num_decoders_committed+0x3e/0x80 drivers/cxl/core/port.c:49 Read of size 4 at addr 0000000000000642 by task syz.0.97/2282 CPU: 2 UID: 0 PID: 2282 Comm: syz.0.97 Not tainted 7.0.0-rc1-gebd11be59f74-dirty #494 PREEMPT(full) Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.17.0-0-gb52ca86e094d-prebuilt.qemu.org 04/01/2014 Call Trace: __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120 kasan_report+0xe0/0x110 mm/kasan/report.c:595 cxl_num_decoders_committed+0x3e/0x80 drivers/cxl/core/port.c:49 cxl_mem_sanitize+0x141/0x170 drivers/cxl/core/mbox.c:1304 security_sanitize_store+0xb0/0x120 drivers/cxl/core/memdev.c:173 dev_attr_store+0x46/0x70 drivers/base/core.c:2437 sysfs_kf_write+0x95/0xb0 fs/sysfs/file.c:142 kernfs_fop_write_iter+0x276/0x330 fs/kernfs/file.c:352 new_sync_write fs/read_write.c:595 [inline] vfs_write+0x5df/0xaa0 fs/read_write.c:688 ksys_write+0x103/0x1f0 fs/read_write.c:740 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x111/0x680 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f60a584ba79 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f60a42a7038 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 00007f60a5ab5fa0 RCX: 00007f60a584ba79 RDX: 0000000000000002 RSI: 00002000000001c0 RDI: 0000000000000003 RBP: 00007f60a58a49df R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 00007f60a5ab6038 R14: 00007f60a5ab5fa0 R15: 00007ffe58fad8b8 This goes away using the correct check instead of abusing cxlmd->endpoint, which is unusable (ENXIO) until the driver has probed. During that window the memdev sysfs attributes are already visible, as soon as device_add() completes. Fixes: 29317f8dc6ed ("cxl/mem: Introduce cxl_memdev_attach for CXL-dependent operation") Signed-off-by: Davidlohr Bueso Reviewed-by: Jonathan Cameron Reviewed-by: Gregory Price Link: https://patch.msgid.link/20260301221739.1726722-1-dave@stgolabs.net Signed-off-by: Dave Jiang --- drivers/cxl/core/mbox.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cxl/core/mbox.c b/drivers/cxl/core/mbox.c index e7a6452bf544..12386d912705 100644 --- a/drivers/cxl/core/mbox.c +++ b/drivers/cxl/core/mbox.c @@ -1301,7 +1301,7 @@ int cxl_mem_sanitize(struct cxl_memdev *cxlmd, u16 cmd) * Require an endpoint to be safe otherwise the driver can not * be sure that the device is unmapped. */ - if (endpoint && cxl_num_decoders_committed(endpoint) == 0) + if (cxlmd->dev.driver && cxl_num_decoders_committed(endpoint) == 0) return __cxl_mem_sanitize(mds, cmd); return -EBUSY; From c0e296f257671ba10249630fe58026f29e4804d9 Mon Sep 17 00:00:00 2001 From: Stanislav Kinsburskii Date: Tue, 17 Mar 2026 15:04:55 +0000 Subject: [PATCH 187/352] mshv: Fix error handling in mshv_region_pin The current error handling has two issues: First, pin_user_pages_fast() can return a short pin count (less than requested but greater than zero) when it cannot pin all requested pages. This is treated as success, leading to partially pinned regions being used, which causes memory corruption. Second, when an error occurs mid-loop, already pinned pages from the current batch are not properly accounted for before calling mshv_region_invalidate_pages(), causing a page reference leak. Treat short pins as errors and fix partial batch accounting before cleanup. Signed-off-by: Stanislav Kinsburskii Reviewed-by: Michael Kelley Signed-off-by: Wei Liu --- drivers/hv/mshv_regions.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/hv/mshv_regions.c b/drivers/hv/mshv_regions.c index c28aac0726de..fdffd4f002f6 100644 --- a/drivers/hv/mshv_regions.c +++ b/drivers/hv/mshv_regions.c @@ -314,15 +314,17 @@ int mshv_region_pin(struct mshv_mem_region *region) ret = pin_user_pages_fast(userspace_addr, nr_pages, FOLL_WRITE | FOLL_LONGTERM, pages); - if (ret < 0) + if (ret != nr_pages) goto release_pages; } return 0; release_pages: + if (ret > 0) + done_count += ret; mshv_region_invalidate_pages(region, 0, done_count); - return ret; + return ret < 0 ? ret : -ENOMEM; } static int mshv_region_chunk_unmap(struct mshv_mem_region *region, From c23df30915f83e7257c8625b690a1cece94142a0 Mon Sep 17 00:00:00 2001 From: Jiucheng Xu Date: Wed, 11 Mar 2026 17:11:31 +0800 Subject: [PATCH 188/352] erofs: add GFP_NOIO in the bio completion if needed The bio completion path in the process context (e.g. dm-verity) will directly call into decompression rather than trigger another workqueue context for minimal scheduling latencies, which can then call vm_map_ram() with GFP_KERNEL. Due to insufficient memory, vm_map_ram() may generate memory swapping I/O, which can cause submit_bio_wait to deadlock in some scenarios. Trimmed down the call stack, as follows: f2fs_submit_read_io submit_bio //bio_list is initialized. mmc_blk_mq_recovery z_erofs_endio vm_map_ram __pte_alloc_kernel __alloc_pages_direct_reclaim shrink_folio_list __swap_writepage submit_bio_wait //bio_list is non-NULL, hang!!! Use memalloc_noio_{save,restore}() to wrap up this path. Reviewed-by: Gao Xiang Signed-off-by: Jiucheng Xu Reviewed-by: Chao Yu Signed-off-by: Gao Xiang --- fs/erofs/zdata.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 3977e42b9516..fe8121df9ef2 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -1445,6 +1445,7 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, int bios) { struct erofs_sb_info *const sbi = EROFS_SB(io->sb); + int gfp_flag; /* wake up the caller thread for sync decompression */ if (io->sync) { @@ -1477,7 +1478,9 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, sbi->sync_decompress = EROFS_SYNC_DECOMPRESS_FORCE_ON; return; } + gfp_flag = memalloc_noio_save(); z_erofs_decompressqueue_work(&io->u.work); + memalloc_noio_restore(gfp_flag); } static void z_erofs_fill_bio_vec(struct bio_vec *bvec, From ac512cd351f7e4ab4569f6a52c116f4ab3a239cc Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Tue, 17 Mar 2026 11:18:42 +0100 Subject: [PATCH 189/352] mtd: spi-nor: Fix RDCR controller capability core check Commit 5008c3ec3f89 ("mtd: spi-nor: core: Check read CR support") adds a controller check to make sure the core will not use CR reads on controllers not supporting them. The approach is valid but the fix is incorrect. Unfortunately, the author could not catch it, because the expected behavior was met. The patch indeed drops the RDCR capability, but it does it for all controllers! The issue comes from the use of spi_nor_spimem_check_op() which is an internal helper dedicated to check read/write operations only, despite its generic name. This helper looks for the biggest number of address bytes that can be used for a page operation and tries 4 then 3. It then calls the usual spi-mem helpers to do the checks. These will always fail because there is now an inconsistency: the address cycles are forced to 4 (then 3) bytes, but the bus width during the address cycles rightfully remains 0. There is a non-zero address length but a zero address bus width, which is an invalid combination. The correct check in this case is to directly call spi_mem_supports_op() which doesn't messes up with the operation content. Fixes: 5008c3ec3f89 ("mtd: spi-nor: core: Check read CR support") Cc: stable@vger.kernel.org Acked-by: Tudor Ambarus Acked-by: Takahiro Kuwano Reviewed-by: Pratyush Yadav Signed-off-by: Miquel Raynal --- drivers/mtd/spi-nor/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c index 8ffeb41c3e08..13201908a69f 100644 --- a/drivers/mtd/spi-nor/core.c +++ b/drivers/mtd/spi-nor/core.c @@ -2466,7 +2466,7 @@ spi_nor_spimem_adjust_hwcaps(struct spi_nor *nor, u32 *hwcaps) spi_nor_spimem_setup_op(nor, &op, nor->reg_proto); - if (spi_nor_spimem_check_op(nor, &op)) + if (!spi_mem_supports_op(nor->spimem, &op)) nor->flags |= SNOR_F_NO_READ_CR; } } From 16dec014db0f4ac6f8090dea0bdfcb1ecebc12ca Mon Sep 17 00:00:00 2001 From: Miquel Raynal Date: Tue, 17 Mar 2026 18:17:22 +0100 Subject: [PATCH 190/352] mtd: spi-nor: Rename spi_nor_spimem_check_op() This helper really is just a little helper for internal purposes, and is I/O operation oriented, despite its name. It has already been misused in commit 5008c3ec3f89 ("mtd: spi-nor: core: Check read CR support"), so rename it to clarify its purpose: it is only useful for reads and page programs. Signed-off-by: Miquel Raynal --- drivers/mtd/spi-nor/core.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/mtd/spi-nor/core.c b/drivers/mtd/spi-nor/core.c index 13201908a69f..1eee519c01e5 100644 --- a/drivers/mtd/spi-nor/core.c +++ b/drivers/mtd/spi-nor/core.c @@ -2345,15 +2345,15 @@ int spi_nor_hwcaps_pp2cmd(u32 hwcaps) } /** - * spi_nor_spimem_check_op - check if the operation is supported - * by controller + * spi_nor_spimem_check_read_pp_op - check if a read or a page program operation is + * supported by controller *@nor: pointer to a 'struct spi_nor' *@op: pointer to op template to be checked * * Returns 0 if operation is supported, -EOPNOTSUPP otherwise. */ -static int spi_nor_spimem_check_op(struct spi_nor *nor, - struct spi_mem_op *op) +static int spi_nor_spimem_check_read_pp_op(struct spi_nor *nor, + struct spi_mem_op *op) { /* * First test with 4 address bytes. The opcode itself might @@ -2396,7 +2396,7 @@ static int spi_nor_spimem_check_readop(struct spi_nor *nor, if (spi_nor_protocol_is_dtr(nor->read_proto)) op.dummy.nbytes *= 2; - return spi_nor_spimem_check_op(nor, &op); + return spi_nor_spimem_check_read_pp_op(nor, &op); } /** @@ -2414,7 +2414,7 @@ static int spi_nor_spimem_check_pp(struct spi_nor *nor, spi_nor_spimem_setup_op(nor, &op, pp->proto); - return spi_nor_spimem_check_op(nor, &op); + return spi_nor_spimem_check_read_pp_op(nor, &op); } /** From b9465b04de4b90228de03db9a1e0d56b00814366 Mon Sep 17 00:00:00 2001 From: Olivier Sobrie Date: Tue, 17 Mar 2026 18:18:07 +0100 Subject: [PATCH 191/352] mtd: rawnand: pl353: make sure optimal timings are applied Timings of the nand are adjusted by pl35x_nfc_setup_interface() but actually applied by the pl35x_nand_select_target() function. If there is only one nand chip, the pl35x_nand_select_target() will only apply the timings once since the test at its beginning will always be true after the first call to this function. As a result, the hardware will keep using the default timings set at boot to detect the nand chip, not the optimal ones. With this patch, we program directly the new timings when pl35x_nfc_setup_interface() is called. Fixes: 08d8c62164a3 ("mtd: rawnand: pl353: Add support for the ARM PL353 SMC NAND controller") Signed-off-by: Olivier Sobrie Cc: stable@vger.kernel.org Signed-off-by: Miquel Raynal --- drivers/mtd/nand/raw/pl35x-nand-controller.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/mtd/nand/raw/pl35x-nand-controller.c b/drivers/mtd/nand/raw/pl35x-nand-controller.c index 947fd86ac5fa..f2c65eb7a8d9 100644 --- a/drivers/mtd/nand/raw/pl35x-nand-controller.c +++ b/drivers/mtd/nand/raw/pl35x-nand-controller.c @@ -862,6 +862,9 @@ static int pl35x_nfc_setup_interface(struct nand_chip *chip, int cs, PL35X_SMC_NAND_TAR_CYCLES(tmgs.t_ar) | PL35X_SMC_NAND_TRR_CYCLES(tmgs.t_rr); + writel(plnand->timings, nfc->conf_regs + PL35X_SMC_CYCLES); + pl35x_smc_update_regs(nfc); + return 0; } From b826d2c0b0ecb844c84431ba6b502e744f5d919a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Tue, 17 Mar 2026 19:41:49 -0300 Subject: [PATCH 192/352] pmdomain: bcm: bcm2835-power: Increase ASB control timeout MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The bcm2835_asb_control() function uses a tight polling loop to wait for the ASB bridge to acknowledge a request. During intensive workloads, this handshake intermittently fails for V3D's master ASB on BCM2711, resulting in "Failed to disable ASB master for v3d" errors during runtime PM suspend. As a consequence, the failed power-off leaves V3D in a broken state, leading to bus faults or system hangs on later accesses. As the timeout is insufficient in some scenarios, increase the polling timeout from 1us to 5us, which is still negligible in the context of a power domain transition. Also, replace the open-coded ktime_get_ns()/ cpu_relax() polling loop with readl_poll_timeout_atomic(). Cc: stable@vger.kernel.org Fixes: 670c672608a1 ("soc: bcm: bcm2835-pm: Add support for power domains under a new binding.") Signed-off-by: Maíra Canal Reviewed-by: Stefan Wahren Signed-off-by: Ulf Hansson --- drivers/pmdomain/bcm/bcm2835-power.c | 12 ++++-------- 1 file changed, 4 insertions(+), 8 deletions(-) diff --git a/drivers/pmdomain/bcm/bcm2835-power.c b/drivers/pmdomain/bcm/bcm2835-power.c index 0450202bbee2..eee87a300532 100644 --- a/drivers/pmdomain/bcm/bcm2835-power.c +++ b/drivers/pmdomain/bcm/bcm2835-power.c @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -153,7 +154,6 @@ struct bcm2835_power { static int bcm2835_asb_control(struct bcm2835_power *power, u32 reg, bool enable) { void __iomem *base = power->asb; - u64 start; u32 val; switch (reg) { @@ -166,8 +166,6 @@ static int bcm2835_asb_control(struct bcm2835_power *power, u32 reg, bool enable break; } - start = ktime_get_ns(); - /* Enable the module's async AXI bridges. */ if (enable) { val = readl(base + reg) & ~ASB_REQ_STOP; @@ -176,11 +174,9 @@ static int bcm2835_asb_control(struct bcm2835_power *power, u32 reg, bool enable } writel(PM_PASSWORD | val, base + reg); - while (!!(readl(base + reg) & ASB_ACK) == enable) { - cpu_relax(); - if (ktime_get_ns() - start >= 1000) - return -ETIMEDOUT; - } + if (readl_poll_timeout_atomic(base + reg, val, + !!(val & ASB_ACK) != enable, 0, 5)) + return -ETIMEDOUT; return 0; } From c4192754e836e0ffed95833509b6ada975b74418 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Wed, 18 Mar 2026 11:36:29 -0700 Subject: [PATCH 193/352] fs/tests: exec: Remove bad test vector Drop an unusable test in the bprm stack limits. Reported-by: Guenter Roeck Closes: https://lore.kernel.org/all/a3e9b1c2-40c1-45df-9fa2-14ee6a7b3fe2@roeck-us.net Fixes: 60371f43e56b ("exec: Add KUnit test for bprm_stack_limits()") Signed-off-by: Kees Cook --- fs/tests/exec_kunit.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/fs/tests/exec_kunit.c b/fs/tests/exec_kunit.c index f412d1a0f6bb..1c32cac098cf 100644 --- a/fs/tests/exec_kunit.c +++ b/fs/tests/exec_kunit.c @@ -94,9 +94,6 @@ static const struct bprm_stack_limits_result bprm_stack_limits_results[] = { { { .p = ULONG_MAX, .rlim_stack.rlim_cur = 4 * (_STK_LIM / 4 * 3 + sizeof(void *)), .argc = 0, .envc = 0 }, .expected_argmin = ULONG_MAX - (_STK_LIM / 4 * 3) + sizeof(void *) }, - { { .p = ULONG_MAX, .rlim_stack.rlim_cur = 4 * (_STK_LIM / 4 * + sizeof(void *)), - .argc = 0, .envc = 0 }, - .expected_argmin = ULONG_MAX - (_STK_LIM / 4 * 3) + sizeof(void *) }, { { .p = ULONG_MAX, .rlim_stack.rlim_cur = 4 * _STK_LIM, .argc = 0, .envc = 0 }, .expected_argmin = ULONG_MAX - (_STK_LIM / 4 * 3) + sizeof(void *) }, From 86259558e422b250aa6aa57163a6d759074573f5 Mon Sep 17 00:00:00 2001 From: Sanman Pradhan Date: Wed, 18 Mar 2026 19:40:19 +0000 Subject: [PATCH 194/352] hwmon: (pmbus/isl68137) Fix unchecked return value and use sysfs_emit() isl68137_avs_enable_show_page() uses the return value of pmbus_read_byte_data() without checking for errors. If the I2C transaction fails, a negative error code is passed through bitwise operations, producing incorrect output. Add an error check to propagate the return value if it is negative. Additionally, modernize the callback by replacing sprintf() with sysfs_emit(). Fixes: 038a9c3d1e424 ("hwmon: (pmbus/isl68137) Add driver for Intersil ISL68137 PWM Controller") Cc: stable@vger.kernel.org Signed-off-by: Sanman Pradhan Link: https://lore.kernel.org/r/20260318193952.47908-2-sanman.pradhan@hpe.com Signed-off-by: Guenter Roeck --- drivers/hwmon/pmbus/isl68137.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/hwmon/pmbus/isl68137.c b/drivers/hwmon/pmbus/isl68137.c index 97b61836f53a..e7dac26b5be6 100644 --- a/drivers/hwmon/pmbus/isl68137.c +++ b/drivers/hwmon/pmbus/isl68137.c @@ -98,8 +98,11 @@ static ssize_t isl68137_avs_enable_show_page(struct i2c_client *client, { int val = pmbus_read_byte_data(client, page, PMBUS_OPERATION); - return sprintf(buf, "%d\n", - (val & ISL68137_VOUT_AVS) == ISL68137_VOUT_AVS ? 1 : 0); + if (val < 0) + return val; + + return sysfs_emit(buf, "%d\n", + (val & ISL68137_VOUT_AVS) == ISL68137_VOUT_AVS); } static ssize_t isl68137_avs_enable_store_page(struct i2c_client *client, From 201bc182ad6333468013f1af0719ffe125826b6a Mon Sep 17 00:00:00 2001 From: William Roche Date: Tue, 17 Mar 2026 10:38:10 +0000 Subject: [PATCH 195/352] x86/mce/amd: Check SMCA feature bit before accessing SMCA MSRs People do effort to inject MCEs into guests in order to simulate/test handling of hardware errors. The real use case behind it is testing the handling of SIGBUS which the memory failure code sends to the process. If that process is QEMU, instead of killing the whole guest, the MCE can be injected into the guest kernel so that latter can attempt proper handling and kill the user *process* in the guest, instead, which caused the MCE. The assumption being here that the whole injection flow can supply enough information that the guest kernel can pinpoint the right process. But that's a different topic... Regardless of virtualization or not, access to SMCA-specific registers like MCA_DESTAT should only be done after having checked the smca feature bit. And there are AMD machines like Bulldozer (the one before Zen1) which do support deferred errors but are not SMCA machines. Therefore, properly check the feature bit before accessing related MSRs. [ bp: Rewrite commit message. ] Fixes: 7cb735d7c0cb ("x86/mce: Unify AMD DFR handler with MCA Polling") Signed-off-by: William Roche Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Yazen Ghannam Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20260218163025.1316501-1-william.roche@oracle.com --- arch/x86/kernel/cpu/mce/amd.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/arch/x86/kernel/cpu/mce/amd.c b/arch/x86/kernel/cpu/mce/amd.c index da13c1e37f87..a030ee4cecc2 100644 --- a/arch/x86/kernel/cpu/mce/amd.c +++ b/arch/x86/kernel/cpu/mce/amd.c @@ -875,13 +875,18 @@ void amd_clear_bank(struct mce *m) { amd_reset_thr_limit(m->bank); - /* Clear MCA_DESTAT for all deferred errors even those logged in MCA_STATUS. */ - if (m->status & MCI_STATUS_DEFERRED) - mce_wrmsrq(MSR_AMD64_SMCA_MCx_DESTAT(m->bank), 0); + if (mce_flags.smca) { + /* + * Clear MCA_DESTAT for all deferred errors even those + * logged in MCA_STATUS. + */ + if (m->status & MCI_STATUS_DEFERRED) + mce_wrmsrq(MSR_AMD64_SMCA_MCx_DESTAT(m->bank), 0); - /* Don't clear MCA_STATUS if MCA_DESTAT was used exclusively. */ - if (m->kflags & MCE_CHECK_DFR_REGS) - return; + /* Don't clear MCA_STATUS if MCA_DESTAT was used exclusively. */ + if (m->kflags & MCE_CHECK_DFR_REGS) + return; + } mce_wrmsrq(mca_msr_reg(m->bank, MCA_STATUS), 0); } From bb288d7d869e86d382f35a0e26242c5ccb05ca82 Mon Sep 17 00:00:00 2001 From: Josh Law Date: Thu, 19 Mar 2026 08:43:05 +0900 Subject: [PATCH 196/352] lib/bootconfig: check xbc_init_node() return in override path The ':=' override path in xbc_parse_kv() calls xbc_init_node() to re-initialize an existing value node but does not check the return value. If xbc_init_node() fails (data offset out of range), parsing silently continues with stale node data. Add the missing error check to match the xbc_add_node() call path which already checks for failure. In practice, a bootconfig using ':=' to override a value near the 32KB data limit could silently retain the old value, meaning a security-relevant boot parameter override (e.g., a trace filter or debug setting) would not take effect as intended. Link: https://lore.kernel.org/all/20260318155847.78065-2-objecting@objecting.org/ Fixes: e5efaeb8a8f5 ("bootconfig: Support mixing a value and subkeys under a key") Signed-off-by: Josh Law Signed-off-by: Masami Hiramatsu (Google) --- lib/bootconfig.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/lib/bootconfig.c b/lib/bootconfig.c index 2da049216fe0..e88d0221a826 100644 --- a/lib/bootconfig.c +++ b/lib/bootconfig.c @@ -723,7 +723,8 @@ static int __init xbc_parse_kv(char **k, char *v, int op) if (op == ':') { unsigned short nidx = child->next; - xbc_init_node(child, v, XBC_VALUE); + if (xbc_init_node(child, v, XBC_VALUE) < 0) + return xbc_parse_error("Failed to override value", v); child->next = nidx; /* keep subkeys */ goto array; } From 3b2c2ab4ceb82af484310c3087541eab00ea288b Mon Sep 17 00:00:00 2001 From: Josh Law Date: Thu, 19 Mar 2026 08:43:06 +0900 Subject: [PATCH 197/352] tools/bootconfig: fix fd leak in load_xbc_file() on fstat failure If fstat() fails after open() succeeds, the function returns without closing the file descriptor. Also preserve errno across close(), since close() may overwrite it before the error is returned. Link: https://lore.kernel.org/all/20260318155847.78065-3-objecting@objecting.org/ Fixes: 950313ebf79c ("tools: bootconfig: Add bootconfig command") Signed-off-by: Josh Law Signed-off-by: Masami Hiramatsu (Google) --- tools/bootconfig/main.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/tools/bootconfig/main.c b/tools/bootconfig/main.c index 55d59ed507d5..643f707b8f1d 100644 --- a/tools/bootconfig/main.c +++ b/tools/bootconfig/main.c @@ -162,8 +162,11 @@ static int load_xbc_file(const char *path, char **buf) if (fd < 0) return -errno; ret = fstat(fd, &stat); - if (ret < 0) - return -errno; + if (ret < 0) { + ret = -errno; + close(fd); + return ret; + } ret = load_xbc_fd(fd, buf, stat.st_size); From 146bd2a87a65aa407bb17fac70d8d583d19aba06 Mon Sep 17 00:00:00 2001 From: Kumar Kartikeya Dwivedi Date: Thu, 12 Mar 2026 13:53:07 -0700 Subject: [PATCH 198/352] bpf: Release module BTF IDR before module unload Gregory reported in [0] that the global_map_resize test when run in repeatedly ends up failing during program load. This stems from the fact that BTF reference has not dropped to zero after the previous run's module is unloaded, and the older module's BTF is still discoverable and visible. Later, in libbpf, load_module_btfs() will find the ID for this stale BTF, open its fd, and then it will be used during program load where later steps taking module reference using btf_try_get_module() fail since the underlying module for the BTF is gone. Logically, once a module is unloaded, it's associated BTF artifacts should become hidden. The BTF object inside the kernel may still remain alive as long its reference counts are alive, but it should no longer be discoverable. To fix this, let us call btf_free_id() from the MODULE_STATE_GOING case for the module unload to free the BTF associated IDR entry, and disable its discovery once module unload returns to user space. If a race happens during unload, the outcome is non-deterministic anyway. However, user space should be able to rely on the guarantee that once it has synchronously established a successful module unload, no more stale artifacts associated with this module can be obtained subsequently. Note that we must be careful to not invoke btf_free_id() in btf_put() when btf_is_module() is true now. There could be a window where the module unload drops a non-terminal reference, frees the IDR, but the same ID gets reused and the second unconditional btf_free_id() ends up releasing an unrelated entry. To avoid a special case for btf_is_module() case, set btf->id to zero to make btf_free_id() idempotent, such that we can unconditionally invoke it from btf_put(), and also from the MODULE_STATE_GOING case. Since zero is an invalid IDR, the idr_remove() should be a noop. Note that we can be sure that by the time we reach final btf_put() for btf_is_module() case, the btf_free_id() is already done, since the module itself holds the BTF reference, and it will call this function for the BTF before dropping its own reference. [0]: https://lore.kernel.org/bpf/cover.1773170190.git.grbell@redhat.com Fixes: 36e68442d1af ("bpf: Load and verify kernel module BTFs") Acked-by: Martin KaFai Lau Suggested-by: Martin KaFai Lau Reported-by: Gregory Bell Reviewed-by: Emil Tsalapatis Signed-off-by: Kumar Kartikeya Dwivedi Link: https://lore.kernel.org/r/20260312205307.1346991-1-memxor@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/btf.c | 24 ++++++++++++++++++++---- 1 file changed, 20 insertions(+), 4 deletions(-) diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index 4872d2a6c42d..71f9143fe90f 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -1787,7 +1787,16 @@ static void btf_free_id(struct btf *btf) * of the _bh() version. */ spin_lock_irqsave(&btf_idr_lock, flags); - idr_remove(&btf_idr, btf->id); + if (btf->id) { + idr_remove(&btf_idr, btf->id); + /* + * Clear the id here to make this function idempotent, since it will get + * called a couple of times for module BTFs: on module unload, and then + * the final btf_put(). btf_alloc_id() starts IDs with 1, so we can use + * 0 as sentinel value. + */ + WRITE_ONCE(btf->id, 0); + } spin_unlock_irqrestore(&btf_idr_lock, flags); } @@ -8115,7 +8124,7 @@ static void bpf_btf_show_fdinfo(struct seq_file *m, struct file *filp) { const struct btf *btf = filp->private_data; - seq_printf(m, "btf_id:\t%u\n", btf->id); + seq_printf(m, "btf_id:\t%u\n", READ_ONCE(btf->id)); } #endif @@ -8197,7 +8206,7 @@ int btf_get_info_by_fd(const struct btf *btf, if (copy_from_user(&info, uinfo, info_copy)) return -EFAULT; - info.id = btf->id; + info.id = READ_ONCE(btf->id); ubtf = u64_to_user_ptr(info.btf); btf_copy = min_t(u32, btf->data_size, info.btf_size); if (copy_to_user(ubtf, btf->data, btf_copy)) @@ -8260,7 +8269,7 @@ int btf_get_fd_by_id(u32 id) u32 btf_obj_id(const struct btf *btf) { - return btf->id; + return READ_ONCE(btf->id); } bool btf_is_kernel(const struct btf *btf) @@ -8382,6 +8391,13 @@ static int btf_module_notify(struct notifier_block *nb, unsigned long op, if (btf_mod->module != module) continue; + /* + * For modules, we do the freeing of BTF IDR as soon as + * module goes away to disable BTF discovery, since the + * btf_try_get_module() on such BTFs will fail. This may + * be called again on btf_put(), but it's ok to do so. + */ + btf_free_id(btf_mod->btf); list_del(&btf_mod->list); if (btf_mod->sysfs_attr) sysfs_remove_bin_file(btf_kobj, btf_mod->sysfs_attr); From 8c6e9b60f5c7985a9fe41320556a92d7a33451df Mon Sep 17 00:00:00 2001 From: Suzuki K Poulose Date: Mon, 16 Mar 2026 16:19:01 +0000 Subject: [PATCH 199/352] arm64: realm: Fix PTE_NS_SHARED for 52bit PA support With LPA/LPA2, the top bits of the PFN (Bits[51:48]) end up in the lower bits of the PTE. So, simply creating a mask of the "top IPA bit" doesn't work well for these configurations to set the "top" bit at the output of Stage1 translation. Fix this by using the __phys_to_pte_val() to do the right thing for all configurations. Tested using, kvmtool, placing the memory at a higher address (-m @). e.g: # lkvm run --realm -c 4 -m 512M@@128T -k Image --console serial sh-5.0# dmesg | grep "LPA2\|RSI" [ 0.000000] RME: Using RSI version 1.0 [ 0.000000] CPU features: detected: 52-bit Virtual Addressing (LPA2) [ 0.777354] CPU features: detected: 52-bit Virtual Addressing for KVM (LPA2) Fixes: 399306954996 ("arm64: realm: Query IPA size from the RMM") Cc: Catalin Marinas Cc: Steven Price Cc: Will Deacon Signed-off-by: Suzuki K Poulose Reviewed-by: Steven Price Reviewed-by: Catalin Marinas Signed-off-by: Will Deacon --- arch/arm64/kernel/rsi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kernel/rsi.c b/arch/arm64/kernel/rsi.c index c64a06f58c0b..9e846ce4ef9c 100644 --- a/arch/arm64/kernel/rsi.c +++ b/arch/arm64/kernel/rsi.c @@ -12,6 +12,7 @@ #include #include +#include #include static struct realm_config config; @@ -146,7 +147,7 @@ void __init arm64_rsi_init(void) return; if (WARN_ON(rsi_get_realm_config(&config))) return; - prot_ns_shared = BIT(config.ipa_bits - 1); + prot_ns_shared = __phys_to_pte_val(BIT(config.ipa_bits - 1)); if (arm64_ioremap_prot_hook_register(realm_ioremap_hook)) return; From 9b72283ec9b8685acdb3467de8fbc3352fdb70bb Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Mon, 2 Mar 2026 16:17:33 -0800 Subject: [PATCH 200/352] drm/xe/guc: Fail immediately on GuC load error MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit By using the same variable for both the return of poll_timeout_us and the return of the polled function guc_wait_ucode, the return value of the latter is overwritten and lost after exiting the polling loop. Since guc_wait_ucode returns -1 on GuC load failure, we lose that information and always continue as if the GuC had been loaded correctly. This is fixed by simply using 2 separate variables. Fixes: a4916b4da448 ("drm/xe/guc: Refactor GuC load to use poll_timeout_us()") Signed-off-by: Daniele Ceraolo Spurio Reviewed-by: Matthew Brost Signed-off-by: Vinay Belgaumkar Link: https://patch.msgid.link/20260303001732.2540493-2-daniele.ceraolospurio@intel.com (cherry picked from commit c85ec5c5753a46b5c2aea1292536487be9470ffe) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_guc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c index 6df7c3f260e5..b0a3cddaa2ea 100644 --- a/drivers/gpu/drm/xe/xe_guc.c +++ b/drivers/gpu/drm/xe/xe_guc.c @@ -1124,14 +1124,14 @@ static int guc_wait_ucode(struct xe_guc *guc) struct xe_guc_pc *guc_pc = >->uc.guc.pc; u32 before_freq, act_freq, cur_freq; u32 status = 0, tries = 0; + int load_result, ret; ktime_t before; u64 delta_ms; - int ret; before_freq = xe_guc_pc_get_act_freq(guc_pc); before = ktime_get(); - ret = poll_timeout_us(ret = guc_load_done(gt, &status, &tries), ret, + ret = poll_timeout_us(load_result = guc_load_done(gt, &status, &tries), load_result, 10 * USEC_PER_MSEC, GUC_LOAD_TIMEOUT_SEC * USEC_PER_SEC, false); @@ -1139,7 +1139,7 @@ static int guc_wait_ucode(struct xe_guc *guc) act_freq = xe_guc_pc_get_act_freq(guc_pc); cur_freq = xe_guc_pc_get_cur_freq_fw(guc_pc); - if (ret) { + if (ret || load_result <= 0) { xe_gt_err(gt, "load failed: status = 0x%08X, time = %lldms, freq = %dMHz (req %dMHz)\n", status, delta_ms, xe_guc_pc_get_act_freq(guc_pc), xe_guc_pc_get_cur_freq_fw(guc_pc)); From 26c638d5602e329e0b26281a74c6ec69dee12f23 Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Tue, 10 Mar 2026 18:50:33 -0400 Subject: [PATCH 201/352] drm/xe: Always kill exec queues in xe_guc_submit_pause_abort MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit xe_guc_submit_pause_abort is intended to be called after something disastrous occurs (e.g., VF migration fails, device wedging, or driver unload) and should immediately trigger the teardown of remaining submission state. With that, kill any remaining queues in this function. Fixes: 7c4b7e34c83b ("drm/xe/vf: Abort VF post migration recovery on failure") Cc: stable@vger.kernel.org Signed-off-by: Zhanjun Dong Reviewed-by: Stuart Summers Signed-off-by: Matthew Brost Link: https://patch.msgid.link/20260310225039.1320161-2-zhanjun.dong@intel.com (cherry picked from commit 78f3bf00be4f15daead02ba32d4737129419c902) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_guc_submit.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 799ef9f48003..2d68f5317887 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -2695,8 +2695,7 @@ void xe_guc_submit_pause_abort(struct xe_guc *guc) continue; xe_sched_submission_start(sched); - if (exec_queue_killed_or_banned_or_wedged(q)) - xe_guc_exec_queue_trigger_cleanup(q); + guc_exec_queue_kill(q); } mutex_unlock(&guc->submission_state.lock); } From fb3738693cbdce104bf12615e980a6a37ff9087d Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Tue, 10 Mar 2026 18:50:34 -0400 Subject: [PATCH 202/352] drm/xe: Forcefully tear down exec queues in GuC submit fini MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In GuC submit fini, forcefully tear down any exec queues by disabling CTs, stopping the scheduler (which cleans up lost G2H), killing all remaining queues, and resuming scheduling to allow any remaining cleanup actions to complete and signal any remaining fences. Split guc_submit_fini into device related and software only part. Using device-managed and drm-managed action guarantees the correct ordering of cleanup. Fixes: dd08ebf6c352 ("drm/xe: Introduce a new DRM driver for Intel GPUs") Cc: stable@vger.kernel.org Reviewed-by: Zhanjun Dong Signed-off-by: Matthew Brost Link: https://patch.msgid.link/20260310225039.1320161-3-zhanjun.dong@intel.com (cherry picked from commit a6ab444a111a59924bd9d0c1e0613a75a0a40b89) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_guc.c | 26 ++++++++++++++-- drivers/gpu/drm/xe/xe_guc.h | 1 + drivers/gpu/drm/xe/xe_guc_submit.c | 48 +++++++++++++++++++++++------- 3 files changed, 63 insertions(+), 12 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c index b0a3cddaa2ea..4ab65cae8743 100644 --- a/drivers/gpu/drm/xe/xe_guc.c +++ b/drivers/gpu/drm/xe/xe_guc.c @@ -1347,15 +1347,37 @@ int xe_guc_enable_communication(struct xe_guc *guc) return 0; } -int xe_guc_suspend(struct xe_guc *guc) +/** + * xe_guc_softreset() - Soft reset GuC + * @guc: The GuC object + * + * Send soft reset command to GuC through mmio send. + * + * Return: 0 if success, otherwise error code + */ +int xe_guc_softreset(struct xe_guc *guc) { - struct xe_gt *gt = guc_to_gt(guc); u32 action[] = { XE_GUC_ACTION_CLIENT_SOFT_RESET, }; int ret; + if (!xe_uc_fw_is_running(&guc->fw)) + return 0; + ret = xe_guc_mmio_send(guc, action, ARRAY_SIZE(action)); + if (ret) + return ret; + + return 0; +} + +int xe_guc_suspend(struct xe_guc *guc) +{ + struct xe_gt *gt = guc_to_gt(guc); + int ret; + + ret = xe_guc_softreset(guc); if (ret) { xe_gt_err(gt, "GuC suspend failed: %pe\n", ERR_PTR(ret)); return ret; diff --git a/drivers/gpu/drm/xe/xe_guc.h b/drivers/gpu/drm/xe/xe_guc.h index 66e7edc70ed9..02514914f404 100644 --- a/drivers/gpu/drm/xe/xe_guc.h +++ b/drivers/gpu/drm/xe/xe_guc.h @@ -44,6 +44,7 @@ int xe_guc_opt_in_features_enable(struct xe_guc *guc); void xe_guc_runtime_suspend(struct xe_guc *guc); void xe_guc_runtime_resume(struct xe_guc *guc); int xe_guc_suspend(struct xe_guc *guc); +int xe_guc_softreset(struct xe_guc *guc); void xe_guc_notify(struct xe_guc *guc); int xe_guc_auth_huc(struct xe_guc *guc, u32 rsa_addr); int xe_guc_mmio_send(struct xe_guc *guc, const u32 *request, u32 len); diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 2d68f5317887..ef4d37b5c73c 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -48,6 +48,8 @@ #define XE_GUC_EXEC_QUEUE_CGP_CONTEXT_ERROR_LEN 6 +static int guc_submit_reset_prepare(struct xe_guc *guc); + static struct xe_guc * exec_queue_to_guc(struct xe_exec_queue *q) { @@ -239,7 +241,7 @@ static bool exec_queue_killed_or_banned_or_wedged(struct xe_exec_queue *q) EXEC_QUEUE_STATE_BANNED)); } -static void guc_submit_fini(struct drm_device *drm, void *arg) +static void guc_submit_sw_fini(struct drm_device *drm, void *arg) { struct xe_guc *guc = arg; struct xe_device *xe = guc_to_xe(guc); @@ -257,6 +259,19 @@ static void guc_submit_fini(struct drm_device *drm, void *arg) xa_destroy(&guc->submission_state.exec_queue_lookup); } +static void guc_submit_fini(void *arg) +{ + struct xe_guc *guc = arg; + + /* Forcefully kill any remaining exec queues */ + xe_guc_ct_stop(&guc->ct); + guc_submit_reset_prepare(guc); + xe_guc_softreset(guc); + xe_guc_submit_stop(guc); + xe_uc_fw_sanitize(&guc->fw); + xe_guc_submit_pause_abort(guc); +} + static void guc_submit_wedged_fini(void *arg) { struct xe_guc *guc = arg; @@ -326,7 +341,11 @@ int xe_guc_submit_init(struct xe_guc *guc, unsigned int num_ids) guc->submission_state.initialized = true; - return drmm_add_action_or_reset(&xe->drm, guc_submit_fini, guc); + err = drmm_add_action_or_reset(&xe->drm, guc_submit_sw_fini, guc); + if (err) + return err; + + return devm_add_action_or_reset(xe->drm.dev, guc_submit_fini, guc); } /* @@ -2230,6 +2249,7 @@ static const struct xe_exec_queue_ops guc_exec_queue_ops = { static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q) { struct xe_gpu_scheduler *sched = &q->guc->sched; + bool do_destroy = false; /* Stop scheduling + flush any DRM scheduler operations */ xe_sched_submission_stop(sched); @@ -2237,7 +2257,7 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q) /* Clean up lost G2H + reset engine state */ if (exec_queue_registered(q)) { if (exec_queue_destroyed(q)) - __guc_exec_queue_destroy(guc, q); + do_destroy = true; } if (q->guc->suspend_pending) { set_exec_queue_suspended(q); @@ -2273,18 +2293,15 @@ static void guc_exec_queue_stop(struct xe_guc *guc, struct xe_exec_queue *q) xe_guc_exec_queue_trigger_cleanup(q); } } + + if (do_destroy) + __guc_exec_queue_destroy(guc, q); } -int xe_guc_submit_reset_prepare(struct xe_guc *guc) +static int guc_submit_reset_prepare(struct xe_guc *guc) { int ret; - if (xe_gt_WARN_ON(guc_to_gt(guc), vf_recovery(guc))) - return 0; - - if (!guc->submission_state.initialized) - return 0; - /* * Using an atomic here rather than submission_state.lock as this * function can be called while holding the CT lock (engine reset @@ -2299,6 +2316,17 @@ int xe_guc_submit_reset_prepare(struct xe_guc *guc) return ret; } +int xe_guc_submit_reset_prepare(struct xe_guc *guc) +{ + if (xe_gt_WARN_ON(guc_to_gt(guc), vf_recovery(guc))) + return 0; + + if (!guc->submission_state.initialized) + return 0; + + return guc_submit_reset_prepare(guc); +} + void xe_guc_submit_reset_wait(struct xe_guc *guc) { wait_event(guc->ct.wq, xe_device_wedged(guc_to_xe(guc)) || From e0f82655df6fbb15b318e9d56724cd54b1cfb04d Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Tue, 10 Mar 2026 18:50:35 -0400 Subject: [PATCH 203/352] drm/xe: Trigger queue cleanup if not in wedged mode 2 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The intent of wedging a device is to allow queues to continue running only in wedged mode 2. In other modes, queues should initiate cleanup and signal all remaining fences. Fix xe_guc_submit_wedge to correctly clean up queues when wedge mode != 2. Fixes: 7dbe8af13c18 ("drm/xe: Wedge the entire device") Cc: stable@vger.kernel.org Reviewed-by: Zhanjun Dong Signed-off-by: Matthew Brost Link: https://patch.msgid.link/20260310225039.1320161-4-zhanjun.dong@intel.com (cherry picked from commit e25ba41c8227c5393c16e4aab398076014bd345f) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_guc_submit.c | 35 +++++++++++++++++++----------- 1 file changed, 22 insertions(+), 13 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index ef4d37b5c73c..fc4f99d46763 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -1271,6 +1271,7 @@ static void disable_scheduling_deregister(struct xe_guc *guc, */ void xe_guc_submit_wedge(struct xe_guc *guc) { + struct xe_device *xe = guc_to_xe(guc); struct xe_gt *gt = guc_to_gt(guc); struct xe_exec_queue *q; unsigned long index; @@ -1285,20 +1286,28 @@ void xe_guc_submit_wedge(struct xe_guc *guc) if (!guc->submission_state.initialized) return; - err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev, - guc_submit_wedged_fini, guc); - if (err) { - xe_gt_err(gt, "Failed to register clean-up in wedged.mode=%s; " - "Although device is wedged.\n", - xe_wedged_mode_to_string(XE_WEDGED_MODE_UPON_ANY_HANG_NO_RESET)); - return; - } + if (xe->wedged.mode == 2) { + err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev, + guc_submit_wedged_fini, guc); + if (err) { + xe_gt_err(gt, "Failed to register clean-up on wedged.mode=2; " + "Although device is wedged.\n"); + return; + } - mutex_lock(&guc->submission_state.lock); - xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) - if (xe_exec_queue_get_unless_zero(q)) - set_exec_queue_wedged(q); - mutex_unlock(&guc->submission_state.lock); + mutex_lock(&guc->submission_state.lock); + xa_for_each(&guc->submission_state.exec_queue_lookup, index, q) + if (xe_exec_queue_get_unless_zero(q)) + set_exec_queue_wedged(q); + mutex_unlock(&guc->submission_state.lock); + } else { + /* Forcefully kill any remaining exec queues, signal fences */ + guc_submit_reset_prepare(guc); + xe_guc_submit_stop(guc); + xe_guc_softreset(guc); + xe_uc_fw_sanitize(&guc->fw); + xe_guc_submit_pause_abort(guc); + } } static bool guc_submit_hint_wedged(struct xe_guc *guc) From 7838dd8367419e9fc43b79c038321cb3c04de2a2 Mon Sep 17 00:00:00 2001 From: Zhanjun Dong Date: Tue, 10 Mar 2026 18:50:37 -0400 Subject: [PATCH 204/352] drm/xe/guc: Ensure CT state transitions via STOP before DISABLED MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The GuC CT state transition requires moving to the STOP state before entering the DISABLED state. Update the driver teardown sequence to make the proper state machine transitions. Fixes: ee4b32220a6b ("drm/xe/guc: Add devm release action to safely tear down CT") Cc: stable@vger.kernel.org Signed-off-by: Zhanjun Dong Reviewed-by: Matthew Brost Signed-off-by: Matthew Brost Link: https://patch.msgid.link/20260310225039.1320161-6-zhanjun.dong@intel.com (cherry picked from commit dace8cb0032f57ea67c87b3b92ad73c89dd2db44) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_guc_ct.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/xe/xe_guc_ct.c b/drivers/gpu/drm/xe/xe_guc_ct.c index d04589140b77..c80082b4c876 100644 --- a/drivers/gpu/drm/xe/xe_guc_ct.c +++ b/drivers/gpu/drm/xe/xe_guc_ct.c @@ -345,6 +345,7 @@ static void guc_action_disable_ct(void *arg) { struct xe_guc_ct *ct = arg; + xe_guc_ct_stop(ct); guc_ct_change_state(ct, XE_GUC_CT_STATE_DISABLED); } From 38b8dcde231641f00eee977d245dbfe5f6b06e11 Mon Sep 17 00:00:00 2001 From: Brian Nguyen Date: Thu, 5 Mar 2026 17:15:48 +0000 Subject: [PATCH 205/352] drm/xe: Skip over non leaf pte for PRL generation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The check using xe_child->base.children was insufficient in determining if a pte was a leaf node. So explicitly skip over every non-leaf pt and conditionally abort if there is a scenario where a non-leaf pt is interleaved between leaf pt, which results in the page walker skipping over some leaf pt. Note that the behavior being targeted for abort is PD[0] = 2M PTE PD[1] = PT -> 512 4K PTEs PD[2] = 2M PTE results in abort, page walker won't descend PD[1]. With new abort, ensuring valid PRL before handling a second abort. v2: - Revert to previous assert. - Revised non-leaf handling for interleaf child pt and leaf pte. - Update comments to specifications. (Stuart) - Remove unnecessary XE_PTE_PS64. (Matthew B) v3: - Modify secondary abort to only check non-leaf PTEs. (Matthew B) Fixes: b912138df299 ("drm/xe: Create page reclaim list on unbind") Signed-off-by: Brian Nguyen Reviewed-by: Matthew Brost Cc: Stuart Summers Link: https://patch.msgid.link/20260305171546.67691-6-brian3.nguyen@intel.com Signed-off-by: Matt Roper (cherry picked from commit 1d123587525db86cc8f0d2beb35d9e33ca3ade83) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_pt.c | 38 +++++++++++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 9 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_pt.c b/drivers/gpu/drm/xe/xe_pt.c index 13b355fadd58..2d9ce2c4cb4f 100644 --- a/drivers/gpu/drm/xe/xe_pt.c +++ b/drivers/gpu/drm/xe/xe_pt.c @@ -1655,14 +1655,35 @@ static int xe_pt_stage_unbind_entry(struct xe_ptw *parent, pgoff_t offset, XE_WARN_ON(!level); /* Check for leaf node */ if (xe_walk->prl && xe_page_reclaim_list_valid(xe_walk->prl) && - (!xe_child->base.children || !xe_child->base.children[first])) { + xe_child->level <= MAX_HUGEPTE_LEVEL) { struct iosys_map *leaf_map = &xe_child->bo->vmap; pgoff_t count = xe_pt_num_entries(addr, next, xe_child->level, walk); for (pgoff_t i = 0; i < count; i++) { - u64 pte = xe_map_rd(xe, leaf_map, (first + i) * sizeof(u64), u64); + u64 pte; int ret; + /* + * If not a leaf pt, skip unless non-leaf pt is interleaved between + * leaf ptes which causes the page walk to skip over the child leaves + */ + if (xe_child->base.children && xe_child->base.children[first + i]) { + u64 pt_size = 1ULL << walk->shifts[xe_child->level]; + bool edge_pt = (i == 0 && !IS_ALIGNED(addr, pt_size)) || + (i == count - 1 && !IS_ALIGNED(next, pt_size)); + + if (!edge_pt) { + xe_page_reclaim_list_abort(xe_walk->tile->primary_gt, + xe_walk->prl, + "PT is skipped by walk at level=%u offset=%lu", + xe_child->level, first + i); + break; + } + continue; + } + + pte = xe_map_rd(xe, leaf_map, (first + i) * sizeof(u64), u64); + /* * In rare scenarios, pte may not be written yet due to racy conditions. * In such cases, invalidate the PRL and fallback to full PPC invalidation. @@ -1674,9 +1695,8 @@ static int xe_pt_stage_unbind_entry(struct xe_ptw *parent, pgoff_t offset, } /* Ensure it is a defined page */ - xe_tile_assert(xe_walk->tile, - xe_child->level == 0 || - (pte & (XE_PTE_PS64 | XE_PDE_PS_2M | XE_PDPE_PS_1G))); + xe_tile_assert(xe_walk->tile, xe_child->level == 0 || + (pte & (XE_PDE_PS_2M | XE_PDPE_PS_1G))); /* An entry should be added for 64KB but contigious 4K have XE_PTE_PS64 */ if (pte & XE_PTE_PS64) @@ -1701,11 +1721,11 @@ static int xe_pt_stage_unbind_entry(struct xe_ptw *parent, pgoff_t offset, killed = xe_pt_check_kill(addr, next, level - 1, xe_child, action, walk); /* - * Verify PRL is active and if entry is not a leaf pte (base.children conditions), - * there is a potential need to invalidate the PRL if any PTE (num_live) are dropped. + * Verify if any PTE are potentially dropped at non-leaf levels, either from being + * killed or the page walk covers the region. */ - if (xe_walk->prl && level > 1 && xe_child->num_live && - xe_child->base.children && xe_child->base.children[first]) { + if (xe_walk->prl && xe_page_reclaim_list_valid(xe_walk->prl) && + xe_child->level > MAX_HUGEPTE_LEVEL && xe_child->num_live) { bool covered = xe_pt_covers(addr, next, xe_child->level, &xe_walk->base); /* From 9be6fd9fbd2032b683e51374497768af9aaa228a Mon Sep 17 00:00:00 2001 From: Ashutosh Dixit Date: Thu, 12 Mar 2026 22:36:30 -0700 Subject: [PATCH 206/352] drm/xe/oa: Allow reading after disabling OA stream MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Some OA data might be present in the OA buffer when OA stream is disabled. Allow UMD's to retrieve this data, so that all data till the point when OA stream is disabled can be retrieved. v2: Update tail pointer after disable (Umesh) Fixes: efb315d0a013 ("drm/xe/oa/uapi: Read file_operation") Cc: stable@vger.kernel.org Signed-off-by: Ashutosh Dixit Reviewed-by: Umesh Nerlige Ramappa Link: https://patch.msgid.link/20260313053630.3176100-1-ashutosh.dixit@intel.com (cherry picked from commit 4ff57c5e8dbba23b5457be12f9709d5c016da16e) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_oa.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c index 4dd3f29933cf..fa90441d3052 100644 --- a/drivers/gpu/drm/xe/xe_oa.c +++ b/drivers/gpu/drm/xe/xe_oa.c @@ -543,8 +543,7 @@ static ssize_t xe_oa_read(struct file *file, char __user *buf, size_t offset = 0; int ret; - /* Can't read from disabled streams */ - if (!stream->enabled || !stream->sample) + if (!stream->sample) return -EINVAL; if (!(file->f_flags & O_NONBLOCK)) { @@ -1460,6 +1459,10 @@ static void xe_oa_stream_disable(struct xe_oa_stream *stream) if (stream->sample) hrtimer_cancel(&stream->poll_check_timer); + + /* Update stream->oa_buffer.tail to allow any final reports to be read */ + if (xe_oa_buffer_check_unlocked(stream)) + wake_up(&stream->poll_wq); } static int xe_oa_enable_preempt_timeslice(struct xe_oa_stream *stream) From e6e3ea52bf07a0b7b9dff189616f189b83ee397a Mon Sep 17 00:00:00 2001 From: Umesh Nerlige Ramappa Date: Thu, 12 Mar 2026 05:53:09 -0700 Subject: [PATCH 207/352] drm/xe/lrc: Fix uninitialized new_ts when capturing context timestamp MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Getting engine specific CTX TIMESTAMP register can fail. In that case, if the context is active, new_ts is uninitialized. Fix that case by initializing new_ts to the last value that was sampled in SW - lrc->ctx_timestamp. Flagged by static analysis. v2: Fix new_ts initialization (Ashutosh) Fixes: bb63e7257e63 ("drm/xe: Avoid toggling schedule state to check LRC timestamp in TDR") Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Ashutosh Dixit Link: https://patch.msgid.link/20260312125308.3126607-2-umesh.nerlige.ramappa@intel.com (cherry picked from commit 466e75d48038af252187855058a7a9312db9d2f8) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_lrc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_lrc.c b/drivers/gpu/drm/xe/xe_lrc.c index b0f037bc227f..7b70cc01fdb3 100644 --- a/drivers/gpu/drm/xe/xe_lrc.c +++ b/drivers/gpu/drm/xe/xe_lrc.c @@ -2413,14 +2413,14 @@ static int get_ctx_timestamp(struct xe_lrc *lrc, u32 engine_id, u64 *reg_ctx_ts) * @lrc: Pointer to the lrc. * * Return latest ctx timestamp. With support for active contexts, the - * calculation may bb slightly racy, so follow a read-again logic to ensure that + * calculation may be slightly racy, so follow a read-again logic to ensure that * the context is still active before returning the right timestamp. * * Returns: New ctx timestamp value */ u64 xe_lrc_timestamp(struct xe_lrc *lrc) { - u64 lrc_ts, reg_ts, new_ts; + u64 lrc_ts, reg_ts, new_ts = lrc->ctx_timestamp; u32 engine_id; lrc_ts = xe_lrc_ctx_timestamp(lrc); From cadf6019231b614ebbd9ec2a16e5997ecbd8d016 Mon Sep 17 00:00:00 2001 From: Marc Kleine-Budde Date: Tue, 10 Mar 2026 13:48:03 +0100 Subject: [PATCH 208/352] can: netlink: can_changelink(): add missing error handling to call can_ctrlmode_changelink() In commit e1a5cd9d6665 ("can: netlink: add can_ctrlmode_changelink()") the CAN Control Mode (IFLA_CAN_CTRLMODE) handling was factored out into the can_ctrlmode_changelink() function. But the call to can_ctrlmode_changelink() is missing the error handling. Add the missing error handling and propagation to the call can_ctrlmode_changelink(). Cc: stable@vger.kernel.org Fixes: e1a5cd9d6665 ("can: netlink: add can_ctrlmode_changelink()") Link: https://patch.msgid.link/20260310-can_ctrlmode_changelink-add-error-handling-v1-1-0daf63d85922@pengutronix.de Signed-off-by: Marc Kleine-Budde --- drivers/net/can/dev/netlink.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/can/dev/netlink.c b/drivers/net/can/dev/netlink.c index 0498198a4696..766d455950f5 100644 --- a/drivers/net/can/dev/netlink.c +++ b/drivers/net/can/dev/netlink.c @@ -601,7 +601,9 @@ static int can_changelink(struct net_device *dev, struct nlattr *tb[], /* We need synchronization with dev->stop() */ ASSERT_RTNL(); - can_ctrlmode_changelink(dev, data, extack); + err = can_ctrlmode_changelink(dev, data, extack); + if (err) + return err; if (data[IFLA_CAN_BITTIMING]) { struct can_bittiming bt; From 7a57354756c7df223abe2c33774235ad70cb4231 Mon Sep 17 00:00:00 2001 From: Wenyuan Li <2063309626@qq.com> Date: Mon, 16 Mar 2026 00:00:22 +0800 Subject: [PATCH 209/352] can: mcp251x: add error handling for power enable in open and resume Add missing error handling for mcp251x_power_enable() calls in both mcp251x_open() and mcp251x_can_resume() functions. In mcp251x_open(), if power enable fails, jump to error path to close candev without attempting to disable power again. In mcp251x_can_resume(), properly check return values of power enable calls for both power and transceiver regulators. If any fails, return the error code to the PM framework and log the failure. This ensures the driver properly handles power control failures and maintains correct device state. Signed-off-by: Wenyuan Li <2063309626@qq.com> Link: https://patch.msgid.link/tencent_F3EFC5D7738AC548857B91657715E2D3AA06@qq.com [mkl: fix patch description] [mkl: mcp251x_can_resume(): replace goto by return] Signed-off-by: Marc Kleine-Budde --- drivers/net/can/spi/mcp251x.c | 29 ++++++++++++++++++++++++----- 1 file changed, 24 insertions(+), 5 deletions(-) diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c index bb7782582f40..0d0190ae094a 100644 --- a/drivers/net/can/spi/mcp251x.c +++ b/drivers/net/can/spi/mcp251x.c @@ -1225,7 +1225,11 @@ static int mcp251x_open(struct net_device *net) } mutex_lock(&priv->mcp_lock); - mcp251x_power_enable(priv->transceiver, 1); + ret = mcp251x_power_enable(priv->transceiver, 1); + if (ret) { + dev_err(&spi->dev, "failed to enable transceiver power: %pe\n", ERR_PTR(ret)); + goto out_close_candev; + } priv->force_quit = 0; priv->tx_skb = NULL; @@ -1272,6 +1276,7 @@ static int mcp251x_open(struct net_device *net) mcp251x_hw_sleep(spi); out_close: mcp251x_power_enable(priv->transceiver, 0); +out_close_candev: close_candev(net); mutex_unlock(&priv->mcp_lock); if (release_irq) @@ -1516,11 +1521,25 @@ static int __maybe_unused mcp251x_can_resume(struct device *dev) { struct spi_device *spi = to_spi_device(dev); struct mcp251x_priv *priv = spi_get_drvdata(spi); + int ret = 0; - if (priv->after_suspend & AFTER_SUSPEND_POWER) - mcp251x_power_enable(priv->power, 1); - if (priv->after_suspend & AFTER_SUSPEND_UP) - mcp251x_power_enable(priv->transceiver, 1); + if (priv->after_suspend & AFTER_SUSPEND_POWER) { + ret = mcp251x_power_enable(priv->power, 1); + if (ret) { + dev_err(dev, "failed to restore power: %pe\n", ERR_PTR(ret)); + return ret; + } + } + + if (priv->after_suspend & AFTER_SUSPEND_UP) { + ret = mcp251x_power_enable(priv->transceiver, 1); + if (ret) { + dev_err(dev, "failed to restore transceiver power: %pe\n", ERR_PTR(ret)); + if (priv->after_suspend & AFTER_SUSPEND_POWER) + mcp251x_power_enable(priv->power, 0); + return ret; + } + } if (priv->after_suspend & (AFTER_SUSPEND_POWER | AFTER_SUSPEND_UP)) queue_work(priv->wq, &priv->restart_work); From 46eee1661aa9b49966e6c43d07126fe408edda57 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Wed, 18 Mar 2026 18:34:13 +0100 Subject: [PATCH 210/352] can: statistics: add missing atomic access in hot path Commit 80b5f90158d1 ("can: statistics: use atomic access in hot path") fixed a KCSAN issue in can_receive() but missed to convert the 'matches' variable used in can_rcv_filter(). Fixes: 80b5f90158d1 ("can: statistics: use atomic access in hot path") Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20260318173413.28235-1-socketcan@hartkopp.net Signed-off-by: Marc Kleine-Budde --- net/can/af_can.c | 4 ++-- net/can/af_can.h | 2 +- net/can/proc.c | 3 ++- 3 files changed, 5 insertions(+), 4 deletions(-) diff --git a/net/can/af_can.c b/net/can/af_can.c index f70e2ba0aadc..7bc86b176b4d 100644 --- a/net/can/af_can.c +++ b/net/can/af_can.c @@ -469,7 +469,7 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id, rcv->can_id = can_id; rcv->mask = mask; - rcv->matches = 0; + atomic_long_set(&rcv->matches, 0); rcv->func = func; rcv->data = data; rcv->ident = ident; @@ -573,7 +573,7 @@ EXPORT_SYMBOL(can_rx_unregister); static inline void deliver(struct sk_buff *skb, struct receiver *rcv) { rcv->func(skb, rcv->data); - rcv->matches++; + atomic_long_inc(&rcv->matches); } static int can_rcv_filter(struct can_dev_rcv_lists *dev_rcv_lists, struct sk_buff *skb) diff --git a/net/can/af_can.h b/net/can/af_can.h index 22f3352c77fe..87887014f562 100644 --- a/net/can/af_can.h +++ b/net/can/af_can.h @@ -52,7 +52,7 @@ struct receiver { struct hlist_node list; canid_t can_id; canid_t mask; - unsigned long matches; + atomic_long_t matches; void (*func)(struct sk_buff *skb, void *data); void *data; char *ident; diff --git a/net/can/proc.c b/net/can/proc.c index 0938bf7dd646..de4d05ae3459 100644 --- a/net/can/proc.c +++ b/net/can/proc.c @@ -196,7 +196,8 @@ static void can_print_rcvlist(struct seq_file *m, struct hlist_head *rx_list, " %-5s %03x %08x %pK %pK %8ld %s\n"; seq_printf(m, fmt, DNAME(dev), r->can_id, r->mask, - r->func, r->data, r->matches, r->ident); + r->func, r->data, atomic_long_read(&r->matches), + r->ident); } } From c8cfeb4b9dda2cdfce79519aee4aaff16310a7b6 Mon Sep 17 00:00:00 2001 From: Amelie Delaunay Date: Tue, 17 Mar 2026 11:06:54 +0100 Subject: [PATCH 211/352] pinctrl: stm32: fix HDP driver dependency on GPIO_GENERIC The HDP driver uses the generic GPIO chip API, but this configuration may not be enabled. Ensure it is enabled by selecting the appropriate option. Fixes: 4bcff9c05b9d ("pinctrl: stm32: use new generic GPIO chip API") Signed-off-by: Amelie Delaunay Signed-off-by: Linus Walleij --- drivers/pinctrl/stm32/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/pinctrl/stm32/Kconfig b/drivers/pinctrl/stm32/Kconfig index 5f67e1ee66dd..d6a171523012 100644 --- a/drivers/pinctrl/stm32/Kconfig +++ b/drivers/pinctrl/stm32/Kconfig @@ -65,6 +65,7 @@ config PINCTRL_STM32_HDP select PINMUX select GENERIC_PINCONF select GPIOLIB + select GPIO_GENERIC help The Hardware Debug Port allows the observation of internal signals. It uses configurable multiplexer to route signals in a dedicated observation register. From 01f2557aa684e514005541e71a3d01f4cd45c170 Mon Sep 17 00:00:00 2001 From: Matthew Brost Date: Tue, 10 Mar 2026 18:50:39 -0400 Subject: [PATCH 212/352] drm/xe: Open-code GGTT MMIO access protection MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit GGTT MMIO access is currently protected by hotplug (drm_dev_enter), which works correctly when the driver loads successfully and is later unbound or unloaded. However, if driver load fails, this protection is insufficient because drm_dev_unplug() is never called. Additionally, devm release functions cannot guarantee that all BOs with GGTT mappings are destroyed before the GGTT MMIO region is removed, as some BOs may be freed asynchronously by worker threads. To address this, introduce an open-coded flag, protected by the GGTT lock, that guards GGTT MMIO access. The flag is cleared during the dev_fini_ggtt devm release function to ensure MMIO access is disabled once teardown begins. Cc: stable@vger.kernel.org Fixes: 919bb54e989c ("drm/xe: Fix missing runtime outer protection for ggtt_remove_node") Reviewed-by: Zhanjun Dong Signed-off-by: Matthew Brost Link: https://patch.msgid.link/20260310225039.1320161-8-zhanjun.dong@intel.com (cherry picked from commit 4f3a998a173b4325c2efd90bdadc6ccd3ad9a431) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_ggtt.c | 10 ++++------ drivers/gpu/drm/xe/xe_ggtt_types.h | 5 ++++- 2 files changed, 8 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_ggtt.c b/drivers/gpu/drm/xe/xe_ggtt.c index 2bda426a6986..d1561ebe4e56 100644 --- a/drivers/gpu/drm/xe/xe_ggtt.c +++ b/drivers/gpu/drm/xe/xe_ggtt.c @@ -313,6 +313,8 @@ static void dev_fini_ggtt(void *arg) { struct xe_ggtt *ggtt = arg; + scoped_guard(mutex, &ggtt->lock) + ggtt->flags &= ~XE_GGTT_FLAGS_ONLINE; drain_workqueue(ggtt->wq); } @@ -377,6 +379,7 @@ int xe_ggtt_init_early(struct xe_ggtt *ggtt) if (err) return err; + ggtt->flags |= XE_GGTT_FLAGS_ONLINE; err = devm_add_action_or_reset(xe->drm.dev, dev_fini_ggtt, ggtt); if (err) return err; @@ -410,13 +413,10 @@ static void xe_ggtt_initial_clear(struct xe_ggtt *ggtt) static void ggtt_node_remove(struct xe_ggtt_node *node) { struct xe_ggtt *ggtt = node->ggtt; - struct xe_device *xe = tile_to_xe(ggtt->tile); bool bound; - int idx; - - bound = drm_dev_enter(&xe->drm, &idx); mutex_lock(&ggtt->lock); + bound = ggtt->flags & XE_GGTT_FLAGS_ONLINE; if (bound) xe_ggtt_clear(ggtt, node->base.start, node->base.size); drm_mm_remove_node(&node->base); @@ -429,8 +429,6 @@ static void ggtt_node_remove(struct xe_ggtt_node *node) if (node->invalidate_on_remove) xe_ggtt_invalidate(ggtt); - drm_dev_exit(idx); - free_node: xe_ggtt_node_fini(node); } diff --git a/drivers/gpu/drm/xe/xe_ggtt_types.h b/drivers/gpu/drm/xe/xe_ggtt_types.h index d82b71a198bc..c002857bb761 100644 --- a/drivers/gpu/drm/xe/xe_ggtt_types.h +++ b/drivers/gpu/drm/xe/xe_ggtt_types.h @@ -28,11 +28,14 @@ struct xe_ggtt { /** @size: Total usable size of this GGTT */ u64 size; -#define XE_GGTT_FLAGS_64K BIT(0) +#define XE_GGTT_FLAGS_64K BIT(0) +#define XE_GGTT_FLAGS_ONLINE BIT(1) /** * @flags: Flags for this GGTT * Acceptable flags: * - %XE_GGTT_FLAGS_64K - if PTE size is 64K. Otherwise, regular is 4K. + * - %XE_GGTT_FLAGS_ONLINE - is GGTT online, protected by ggtt->lock + * after init */ unsigned int flags; /** @scratch: Internal object allocation used as a scratch page */ From b9c310d72783cc2f30d103eed83920a5a29c671a Mon Sep 17 00:00:00 2001 From: Ali Norouzi Date: Thu, 19 Mar 2026 16:47:44 +0100 Subject: [PATCH 213/352] can: gw: fix OOB heap access in cgw_csum_crc8_rel() cgw_csum_crc8_rel() correctly computes bounds-safe indices via calc_idx(): int from = calc_idx(crc8->from_idx, cf->len); int to = calc_idx(crc8->to_idx, cf->len); int res = calc_idx(crc8->result_idx, cf->len); if (from < 0 || to < 0 || res < 0) return; However, the loop and the result write then use the raw s8 fields directly instead of the computed variables: for (i = crc8->from_idx; ...) /* BUG: raw negative index */ cf->data[crc8->result_idx] = ...; /* BUG: raw negative index */ With from_idx = to_idx = result_idx = -64 on a 64-byte CAN FD frame, calc_idx(-64, 64) = 0 so the guard passes, but the loop iterates with i = -64, reading cf->data[-64], and the write goes to cf->data[-64]. This write might end up to 56 (7.0-rc) or 40 (<= 6.19) bytes before the start of the canfd_frame on the heap. The companion function cgw_csum_xor_rel() uses `from`/`to`/`res` correctly throughout; fix cgw_csum_crc8_rel() to match. Confirmed with KASAN on linux-7.0-rc2: BUG: KASAN: slab-out-of-bounds in cgw_csum_crc8_rel+0x515/0x5b0 Read of size 1 at addr ffff8880076619c8 by task poc_cgw_oob/62 To configure the can-gw crc8 checksums CAP_NET_ADMIN is needed. Fixes: 456a8a646b25 ("can: gw: add support for CAN FD frames") Cc: stable@vger.kernel.org Reported-by: Ali Norouzi Reviewed-by: Oliver Hartkopp Acked-by: Oliver Hartkopp Signed-off-by: Ali Norouzi Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20260319-fix-can-gw-and-can-isotp-v2-1-c45d52c6d2d8@pengutronix.de Signed-off-by: Marc Kleine-Budde --- net/can/gw.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/can/gw.c b/net/can/gw.c index 8ee4d67a07d3..0ec99f68aa45 100644 --- a/net/can/gw.c +++ b/net/can/gw.c @@ -375,10 +375,10 @@ static void cgw_csum_crc8_rel(struct canfd_frame *cf, return; if (from <= to) { - for (i = crc8->from_idx; i <= crc8->to_idx; i++) + for (i = from; i <= to; i++) crc = crc8->crctab[crc ^ cf->data[i]]; } else { - for (i = crc8->from_idx; i >= crc8->to_idx; i--) + for (i = from; i >= to; i--) crc = crc8->crctab[crc ^ cf->data[i]]; } @@ -397,7 +397,7 @@ static void cgw_csum_crc8_rel(struct canfd_frame *cf, break; } - cf->data[crc8->result_idx] = crc ^ crc8->final_xor_val; + cf->data[res] = crc ^ crc8->final_xor_val; } static void cgw_csum_crc8_pos(struct canfd_frame *cf, From 424e95d62110cdbc8fd12b40918f37e408e35a92 Mon Sep 17 00:00:00 2001 From: Oliver Hartkopp Date: Thu, 19 Mar 2026 16:47:45 +0100 Subject: [PATCH 214/352] can: isotp: fix tx.buf use-after-free in isotp_sendmsg() isotp_sendmsg() uses only cmpxchg() on so->tx.state to serialize access to so->tx.buf. isotp_release() waits for ISOTP_IDLE via wait_event_interruptible() and then calls kfree(so->tx.buf). If a signal interrupts the wait_event_interruptible() inside close() while tx.state is ISOTP_SENDING, the loop exits early and release proceeds to force ISOTP_SHUTDOWN and continues to kfree(so->tx.buf) while sendmsg may still be reading so->tx.buf for the final CAN frame in isotp_fill_dataframe(). The so->tx.buf can be allocated once when the standard tx.buf length needs to be extended. Move the kfree() of this potentially extended tx.buf to sk_destruct time when either isotp_sendmsg() and isotp_release() are done. Fixes: 96d1c81e6a04 ("can: isotp: add module parameter for maximum pdu size") Cc: stable@vger.kernel.org Reported-by: Ali Norouzi Co-developed-by: Ali Norouzi Signed-off-by: Ali Norouzi Signed-off-by: Oliver Hartkopp Link: https://patch.msgid.link/20260319-fix-can-gw-and-can-isotp-v2-2-c45d52c6d2d8@pengutronix.de Signed-off-by: Marc Kleine-Budde --- net/can/isotp.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) diff --git a/net/can/isotp.c b/net/can/isotp.c index da3b72e7afcc..2770f43f4951 100644 --- a/net/can/isotp.c +++ b/net/can/isotp.c @@ -1248,12 +1248,6 @@ static int isotp_release(struct socket *sock) so->ifindex = 0; so->bound = 0; - if (so->rx.buf != so->rx.sbuf) - kfree(so->rx.buf); - - if (so->tx.buf != so->tx.sbuf) - kfree(so->tx.buf); - sock_orphan(sk); sock->sk = NULL; @@ -1622,6 +1616,21 @@ static int isotp_notifier(struct notifier_block *nb, unsigned long msg, return NOTIFY_DONE; } +static void isotp_sock_destruct(struct sock *sk) +{ + struct isotp_sock *so = isotp_sk(sk); + + /* do the standard CAN sock destruct work */ + can_sock_destruct(sk); + + /* free potential extended PDU buffers */ + if (so->rx.buf != so->rx.sbuf) + kfree(so->rx.buf); + + if (so->tx.buf != so->tx.sbuf) + kfree(so->tx.buf); +} + static int isotp_init(struct sock *sk) { struct isotp_sock *so = isotp_sk(sk); @@ -1666,6 +1675,9 @@ static int isotp_init(struct sock *sk) list_add_tail(&so->notifier, &isotp_notifier_list); spin_unlock(&isotp_notifier_lock); + /* re-assign default can_sock_destruct() reference */ + sk->sk_destruct = isotp_sock_destruct; + return 0; } From 65d046b2d8e0d6d855379a981869005fd6b6a41b Mon Sep 17 00:00:00 2001 From: Sanjay Yadav Date: Fri, 13 Mar 2026 12:46:09 +0530 Subject: [PATCH 215/352] drm/xe: Fix missing runtime PM reference in ccs_mode_store MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ccs_mode_store() calls xe_gt_reset() which internally invokes xe_pm_runtime_get_noresume(). That function requires the caller to already hold an outer runtime PM reference and warns if none is held: [46.891177] xe 0000:03:00.0: [drm] Missing outer runtime PM protection [46.891178] WARNING: drivers/gpu/drm/xe/xe_pm.c:885 at xe_pm_runtime_get_noresume+0x8b/0xc0 Fix this by protecting xe_gt_reset() with the scope-based guard(xe_pm_runtime)(xe), which is the preferred form when the reference lifetime matches a single scope. v2: - Use scope-based guard(xe_pm_runtime)(xe) (Shuicheng) - Update commit message accordingly Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/7593 Fixes: 480b358e7d8e ("drm/xe: Do not wake device during a GT reset") Cc: # v6.19+ Cc: Thomas Hellström Cc: Matthew Brost Cc: Rodrigo Vivi Cc: Shuicheng Lin Suggested-by: Matthew Auld Signed-off-by: Sanjay Yadav Reviewed-by: Shuicheng Lin Reviewed-by: Matthew Auld Signed-off-by: Matthew Auld Link: https://patch.msgid.link/20260313071608.3459480-2-sanjay.kumar.yadav@intel.com (cherry picked from commit 7937ea733f79b3f25e802a0c8360bf7423856f36) Signed-off-by: Thomas Hellström --- drivers/gpu/drm/xe/xe_gt_ccs_mode.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c index fe944687728c..03c1862ba497 100644 --- a/drivers/gpu/drm/xe/xe_gt_ccs_mode.c +++ b/drivers/gpu/drm/xe/xe_gt_ccs_mode.c @@ -12,6 +12,7 @@ #include "xe_gt_printk.h" #include "xe_gt_sysfs.h" #include "xe_mmio.h" +#include "xe_pm.h" #include "xe_sriov.h" static void __xe_gt_apply_ccs_mode(struct xe_gt *gt, u32 num_engines) @@ -150,6 +151,7 @@ ccs_mode_store(struct device *kdev, struct device_attribute *attr, xe_gt_info(gt, "Setting compute mode to %d\n", num_engines); gt->ccs_mode = num_engines; xe_gt_record_user_engines(gt); + guard(xe_pm_runtime)(xe); xe_gt_reset(gt); } From 9d87cb22195b2c67405f5485d525190747ad5493 Mon Sep 17 00:00:00 2001 From: Minseo Park Date: Sun, 15 Mar 2026 22:14:37 +0900 Subject: [PATCH 216/352] Bluetooth: L2CAP: Fix stack-out-of-bounds read in l2cap_ecred_conn_req Syzbot reported a KASAN stack-out-of-bounds read in l2cap_build_cmd() that is triggered by a malformed Enhanced Credit Based Connection Request. The vulnerability stems from l2cap_ecred_conn_req(). The function allocates a local stack buffer (`pdu`) designed to hold a maximum of 5 Source Channel IDs (SCIDs), totaling 18 bytes. When an attacker sends a request with more than 5 SCIDs, the function calculates `rsp_len` based on this unvalidated `cmd_len` before checking if the number of SCIDs exceeds L2CAP_ECRED_MAX_CID. If the SCID count is too high, the function correctly jumps to the `response` label to reject the packet, but `rsp_len` retains the attacker's oversized value. Consequently, l2cap_send_cmd() is instructed to read past the end of the 18-byte `pdu` buffer, triggering a KASAN panic. Fix this by moving the assignment of `rsp_len` to after the `num_scid` boundary check. If the packet is rejected, `rsp_len` will safely remain 0, and the error response will only read the 8-byte base header from the stack. Fixes: c28d2bff7044 ("Bluetooth: L2CAP: Fix result of L2CAP_ECRED_CONN_RSP when MTU is too short") Reported-by: syzbot+b7f3e7d9a596bf6a63e3@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=b7f3e7d9a596bf6a63e3 Tested-by: syzbot+b7f3e7d9a596bf6a63e3@syzkaller.appspotmail.com Signed-off-by: Minseo Park Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 5deb6c4f1e41..0882b5ac2ecc 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5081,14 +5081,14 @@ static inline int l2cap_ecred_conn_req(struct l2cap_conn *conn, cmd_len -= sizeof(*req); num_scid = cmd_len / sizeof(u16); - /* Always respond with the same number of scids as in the request */ - rsp_len = cmd_len; - if (num_scid > L2CAP_ECRED_MAX_CID) { result = L2CAP_CR_LE_INVALID_PARAMS; goto response; } + /* Always respond with the same number of scids as in the request */ + rsp_len = cmd_len; + mtu = __le16_to_cpu(req->mtu); mps = __le16_to_cpu(req->mps); From c65bd945d1c08c3db756821b6bf9f1c4a77b29c6 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Fri, 13 Mar 2026 05:22:39 +0900 Subject: [PATCH 217/352] Bluetooth: L2CAP: Validate PDU length before reading SDU length in l2cap_ecred_data_rcv() l2cap_ecred_data_rcv() reads the SDU length field from skb->data using get_unaligned_le16() without first verifying that skb contains at least L2CAP_SDULEN_SIZE (2) bytes. When skb->len is less than 2, this reads past the valid data in the skb. The ERTM reassembly path correctly calls pskb_may_pull() before reading the SDU length (l2cap_reassemble_sdu, L2CAP_SAR_START case). Apply the same validation to the Enhanced Credit Based Flow Control data path. Fixes: aac23bf63659 ("Bluetooth: Implement LE L2CAP reassembly") Signed-off-by: Hyunwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 0882b5ac2ecc..30fd6848938e 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -6690,6 +6690,11 @@ static int l2cap_ecred_data_rcv(struct l2cap_chan *chan, struct sk_buff *skb) if (!chan->sdu) { u16 sdu_len; + if (!pskb_may_pull(skb, L2CAP_SDULEN_SIZE)) { + err = -EINVAL; + goto failed; + } + sdu_len = get_unaligned_le16(skb->data); skb_pull(skb, L2CAP_SDULEN_SIZE); From 598dbba9919c5e36c54fe1709b557d64120cb94b Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Fri, 13 Mar 2026 05:26:16 +0900 Subject: [PATCH 218/352] Bluetooth: SCO: Fix use-after-free in sco_recv_frame() due to missing sock_hold sco_recv_frame() reads conn->sk under sco_conn_lock() but immediately releases the lock without holding a reference to the socket. A concurrent close() can free the socket between the lock release and the subsequent sk->sk_state access, resulting in a use-after-free. Other functions in the same file (sco_sock_timeout(), sco_conn_del()) correctly use sco_sock_hold() to safely hold a reference under the lock. Fix by using sco_sock_hold() to take a reference before releasing the lock, and adding sock_put() on all exit paths. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Hyunwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index e7db50165879..584e059de20a 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -401,7 +401,7 @@ static void sco_recv_frame(struct sco_conn *conn, struct sk_buff *skb) struct sock *sk; sco_conn_lock(conn); - sk = conn->sk; + sk = sco_sock_hold(conn); sco_conn_unlock(conn); if (!sk) @@ -410,11 +410,15 @@ static void sco_recv_frame(struct sco_conn *conn, struct sk_buff *skb) BT_DBG("sk %p len %u", 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 5f5fa4cd35f707344f65ce9e225b6528691dbbaa Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 16 Mar 2026 15:03:27 -0400 Subject: [PATCH 219/352] Bluetooth: MGMT: Fix dangling pointer on mgmt_add_adv_patterns_monitor_complete This fixes the condition checking so mgmt_pending_valid is executed whenever status != -ECANCELED otherwise calling mgmt_pending_free(cmd) would kfree(cmd) without unlinking it from the list first, leaving a dangling pointer. Any subsequent list traversal (e.g., mgmt_pending_foreach during __mgmt_power_off, or another mgmt_pending_valid call) would dereference freed memory. Link: https://lore.kernel.org/linux-bluetooth/20260315132013.75ab40c5@kernel.org/T/#m1418f9c82eeff8510c1beaa21cf53af20db96c06 Fixes: 302a1f674c00 ("Bluetooth: MGMT: Fix possible UAFs") Signed-off-by: Luiz Augusto von Dentz Reviewed-by: Paul Menzel --- net/bluetooth/mgmt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index d52238ce6a9a..e5f9287fb826 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -5355,7 +5355,7 @@ static void mgmt_add_adv_patterns_monitor_complete(struct hci_dev *hdev, * hci_adv_monitors_clear is about to be called which will take care of * freeing the adv_monitor instances. */ - if (status == -ECANCELED && !mgmt_pending_valid(hdev, cmd)) + if (status == -ECANCELED || !mgmt_pending_valid(hdev, cmd)) return; monitor = cmd->user_data; From b6807cfc195ef99e1ac37b2e1e60df40295daa8c Mon Sep 17 00:00:00 2001 From: Cen Zhang Date: Sun, 15 Mar 2026 20:07:26 +0800 Subject: [PATCH 220/352] Bluetooth: hci_sync: annotate data-races around hdev->req_status __hci_cmd_sync_sk() sets hdev->req_status under hdev->req_lock: hdev->req_status = HCI_REQ_PEND; However, several other functions read or write hdev->req_status without holding any lock: - hci_send_cmd_sync() reads req_status in hci_cmd_work (workqueue) - hci_cmd_sync_complete() reads/writes from HCI event completion - hci_cmd_sync_cancel() / hci_cmd_sync_cancel_sync() read/write - hci_abort_conn() reads in connection abort path Since __hci_cmd_sync_sk() runs on hdev->req_workqueue while hci_send_cmd_sync() runs on hdev->workqueue, these are different workqueues that can execute concurrently on different CPUs. The plain C accesses constitute a data race. Add READ_ONCE()/WRITE_ONCE() annotations on all concurrent accesses to hdev->req_status to prevent potential compiler optimizations that could affect correctness (e.g., load fusing in the wait_event condition or store reordering). Signed-off-by: Cen Zhang Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_conn.c | 2 +- net/bluetooth/hci_core.c | 2 +- net/bluetooth/hci_sync.c | 20 ++++++++++---------- 3 files changed, 12 insertions(+), 12 deletions(-) diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 6eb59e9f2aa8..e6393f17576b 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -3095,7 +3095,7 @@ int hci_abort_conn(struct hci_conn *conn, u8 reason) * hci_connect_le serializes the connection attempts so only one * connection can be in BT_CONNECT at time. */ - if (conn->state == BT_CONNECT && hdev->req_status == HCI_REQ_PEND) { + if (conn->state == BT_CONNECT && READ_ONCE(hdev->req_status) == HCI_REQ_PEND) { switch (hci_skb_event(hdev->sent_cmd)) { case HCI_EV_CONN_COMPLETE: case HCI_EV_LE_CONN_COMPLETE: diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c index 31308c1de4ec..01f8ceeb1c0c 100644 --- a/net/bluetooth/hci_core.c +++ b/net/bluetooth/hci_core.c @@ -4126,7 +4126,7 @@ static int hci_send_cmd_sync(struct hci_dev *hdev, struct sk_buff *skb) kfree_skb(skb); } - if (hdev->req_status == HCI_REQ_PEND && + if (READ_ONCE(hdev->req_status) == HCI_REQ_PEND && !hci_dev_test_and_set_flag(hdev, HCI_CMD_PENDING)) { kfree_skb(hdev->req_skb); hdev->req_skb = skb_clone(hdev->sent_cmd, GFP_KERNEL); diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 3166914b0d6c..45d16639874a 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -25,11 +25,11 @@ static void hci_cmd_sync_complete(struct hci_dev *hdev, u8 result, u16 opcode, { bt_dev_dbg(hdev, "result 0x%2.2x", result); - if (hdev->req_status != HCI_REQ_PEND) + if (READ_ONCE(hdev->req_status) != HCI_REQ_PEND) return; hdev->req_result = result; - hdev->req_status = HCI_REQ_DONE; + WRITE_ONCE(hdev->req_status, HCI_REQ_DONE); /* Free the request command so it is not used as response */ kfree_skb(hdev->req_skb); @@ -167,20 +167,20 @@ struct sk_buff *__hci_cmd_sync_sk(struct hci_dev *hdev, u16 opcode, u32 plen, hci_cmd_sync_add(&req, opcode, plen, param, event, sk); - hdev->req_status = HCI_REQ_PEND; + WRITE_ONCE(hdev->req_status, HCI_REQ_PEND); err = hci_req_sync_run(&req); if (err < 0) return ERR_PTR(err); err = wait_event_interruptible_timeout(hdev->req_wait_q, - hdev->req_status != HCI_REQ_PEND, + READ_ONCE(hdev->req_status) != HCI_REQ_PEND, timeout); if (err == -ERESTARTSYS) return ERR_PTR(-EINTR); - switch (hdev->req_status) { + switch (READ_ONCE(hdev->req_status)) { case HCI_REQ_DONE: err = -bt_to_errno(hdev->req_result); break; @@ -194,7 +194,7 @@ struct sk_buff *__hci_cmd_sync_sk(struct hci_dev *hdev, u16 opcode, u32 plen, break; } - hdev->req_status = 0; + WRITE_ONCE(hdev->req_status, 0); hdev->req_result = 0; skb = hdev->req_rsp; hdev->req_rsp = NULL; @@ -665,9 +665,9 @@ void hci_cmd_sync_cancel(struct hci_dev *hdev, int err) { bt_dev_dbg(hdev, "err 0x%2.2x", err); - if (hdev->req_status == HCI_REQ_PEND) { + if (READ_ONCE(hdev->req_status) == HCI_REQ_PEND) { hdev->req_result = err; - hdev->req_status = HCI_REQ_CANCELED; + WRITE_ONCE(hdev->req_status, HCI_REQ_CANCELED); queue_work(hdev->workqueue, &hdev->cmd_sync_cancel_work); } @@ -683,12 +683,12 @@ void hci_cmd_sync_cancel_sync(struct hci_dev *hdev, int err) { bt_dev_dbg(hdev, "err 0x%2.2x", err); - if (hdev->req_status == HCI_REQ_PEND) { + if (READ_ONCE(hdev->req_status) == HCI_REQ_PEND) { /* req_result is __u32 so error must be positive to be properly * propagated. */ hdev->req_result = err < 0 ? -err : err; - hdev->req_status = HCI_REQ_CANCELED; + WRITE_ONCE(hdev->req_status, HCI_REQ_CANCELED); wake_up_interruptible(&hdev->req_wait_q); } From 31148a7be723aa9f2e8fbd62424825ab8d577973 Mon Sep 17 00:00:00 2001 From: Anas Iqbal Date: Sun, 15 Mar 2026 10:51:37 +0000 Subject: [PATCH 221/352] Bluetooth: hci_ll: Fix firmware leak on error path Smatch reports: drivers/bluetooth/hci_ll.c:587 download_firmware() warn: 'fw' from request_firmware() not released on lines: 544. In download_firmware(), if request_firmware() succeeds but the returned firmware content is invalid (no data or zero size), the function returns without releasing the firmware, resulting in a resource leak. Fix this by calling release_firmware() before returning when request_firmware() succeeded but the firmware content is invalid. Fixes: 371805522f87 ("bluetooth: hci_uart: add LL protocol serdev driver support") Reviewed-by: Paul Menzel Signed-off-by: Anas Iqbal Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/hci_ll.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/bluetooth/hci_ll.c b/drivers/bluetooth/hci_ll.c index 91acf24f1ef5..91c96ad12342 100644 --- a/drivers/bluetooth/hci_ll.c +++ b/drivers/bluetooth/hci_ll.c @@ -541,6 +541,8 @@ static int download_firmware(struct ll_device *lldev) if (err || !fw->data || !fw->size) { bt_dev_err(lldev->hu.hdev, "request_firmware failed(errno %d) for %s", err, bts_scr_name); + if (!err) + release_firmware(fw); return -EINVAL; } ptr = (void *)fw->data; From b6552e0503973daf6f23bd6ed9273ef131ee364f Mon Sep 17 00:00:00 2001 From: Helen Koike Date: Thu, 19 Mar 2026 08:58:01 -0300 Subject: [PATCH 222/352] Bluetooth: L2CAP: Fix null-ptr-deref on l2cap_sock_ready_cb Before using sk pointer, check if it is null. Fix the following: KASAN: null-ptr-deref in range [0x0000000000000260-0x0000000000000267] CPU: 0 UID: 0 PID: 5985 Comm: kworker/0:5 Not tainted 7.0.0-rc4-00029-ga989fde763f4 #1 PREEMPT(full) Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.17.0-9.fc43 06/10/2025 Workqueue: events l2cap_info_timeout RIP: 0010:kasan_byte_accessible+0x12/0x30 Code: 79 ff ff ff 0f 1f 40 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 40 d6 48 c1 ef 03 48 b8 00 00 00 00 00 fc ff df <0f> b6 04 07 3c 08 0f 92 c0 c3 cc cce veth0_macvtap: entered promiscuous mode RSP: 0018:ffffc90006e0f808 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: ffffffff89746018 RCX: 0000000080000001 RDX: 0000000000000000 RSI: ffffffff89746018 RDI: 000000000000004c RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 R10: dffffc0000000000 R11: ffffffff8aae3e70 R12: 0000000000000000 R13: 0000000000000260 R14: 0000000000000260 R15: 0000000000000001 FS: 0000000000000000(0000) GS:ffff8880983c2000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00005582615a5008 CR3: 000000007007e000 CR4: 0000000000752ef0 PKRU: 55555554 Call Trace: __kasan_check_byte+0x12/0x40 lock_acquire+0x79/0x2e0 lock_sock_nested+0x48/0x100 ? l2cap_sock_ready_cb+0x46/0x160 l2cap_sock_ready_cb+0x46/0x160 l2cap_conn_start+0x779/0xff0 ? __pfx_l2cap_conn_start+0x10/0x10 ? l2cap_info_timeout+0x60/0xa0 ? __pfx___mutex_lock+0x10/0x10 l2cap_info_timeout+0x68/0xa0 ? process_scheduled_works+0xa8d/0x18c0 process_scheduled_works+0xb6e/0x18c0 ? __pfx_process_scheduled_works+0x10/0x10 ? assign_work+0x3d5/0x5e0 worker_thread+0xa53/0xfc0 kthread+0x388/0x470 ? __pfx_worker_thread+0x10/0x10 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x51e/0xb90 ? __pfx_ret_from_fork+0x10/0x10 veth1_macvtap: entered promiscuous mode ? __switch_to+0xc7d/0x1450 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 Modules linked in: ---[ end trace 0000000000000000 ]--- batman_adv: batadv0: Interface activated: batadv_slave_0 batman_adv: batadv0: Interface activated: batadv_slave_1 netdevsim netdevsim7 netdevsim0: set [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim7 netdevsim1: set [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim7 netdevsim2: set [1, 0] type 2 family 0 port 6081 - 0 netdevsim netdevsim7 netdevsim3: set [1, 0] type 2 family 0 port 6081 - 0 RIP: 0010:kasan_byte_accessible+0x12/0x30 Code: 79 ff ff ff 0f 1f 40 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 40 d6 48 c1 ef 03 48 b8 00 00 00 00 00 fc ff df <0f> b6 04 07 3c 08 0f 92 c0 c3 cc cce ieee80211 phy39: Selected rate control algorithm 'minstrel_ht' RSP: 0018:ffffc90006e0f808 EFLAGS: 00010202 RAX: dffffc0000000000 RBX: ffffffff89746018 RCX: 0000000080000001 RDX: 0000000000000000 RSI: ffffffff89746018 RDI: 000000000000004c RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000 R10: dffffc0000000000 R11: ffffffff8aae3e70 R12: 0000000000000000 R13: 0000000000000260 R14: 0000000000000260 R15: 0000000000000001 FS: 0000000000000000(0000) GS:ffff8880983c2000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f7e16139e9c CR3: 000000000e74e000 CR4: 0000000000752ef0 PKRU: 55555554 Kernel panic - not syncing: Fatal exception Fixes: 54a59aa2b562 ("Bluetooth: Add l2cap_chan->ops->ready()") Signed-off-by: Helen Koike Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_sock.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/bluetooth/l2cap_sock.c b/net/bluetooth/l2cap_sock.c index 597686790371..71e8c1b45bce 100644 --- a/net/bluetooth/l2cap_sock.c +++ b/net/bluetooth/l2cap_sock.c @@ -1698,6 +1698,9 @@ static void l2cap_sock_ready_cb(struct l2cap_chan *chan) struct sock *sk = chan->data; struct sock *parent; + if (!sk) + return; + lock_sock(sk); parent = bt_sk(sk)->parent; From 761fb8ec8778f0caf2bba5a41e3cff1ea86974f3 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 17 Mar 2026 11:54:01 -0400 Subject: [PATCH 223/352] Bluetooth: L2CAP: Fix regressions caused by reusing ident This attempt to fix regressions caused by reusing ident which apparently is not handled well on certain stacks causing the stack to not respond to requests, so instead of simple returning the first unallocated id this stores the last used tx_ident and then attempt to use the next until all available ids are exausted and then cycle starting over to 1. Link: https://bugzilla.kernel.org/show_bug.cgi?id=221120 Link: https://bugzilla.kernel.org/show_bug.cgi?id=221177 Fixes: 6c3ea155e5ee ("Bluetooth: L2CAP: Fix not tracking outstanding TX ident") Signed-off-by: Luiz Augusto von Dentz Tested-by: Christian Eggers --- include/net/bluetooth/l2cap.h | 1 + net/bluetooth/l2cap_core.c | 29 ++++++++++++++++++++++++++--- 2 files changed, 27 insertions(+), 3 deletions(-) diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 010f1a8fd15f..5172afee5494 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -658,6 +658,7 @@ struct l2cap_conn { struct sk_buff *rx_skb; __u32 rx_len; struct ida tx_ida; + __u8 tx_ident; struct sk_buff_head pending_rx; struct work_struct pending_rx_work; diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 30fd6848938e..3de3e3c8e966 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -926,16 +926,39 @@ int l2cap_chan_check_security(struct l2cap_chan *chan, bool initiator) static int l2cap_get_ident(struct l2cap_conn *conn) { + u8 max; + int ident; + /* LE link does not support tools like l2ping so use the full range */ if (conn->hcon->type == LE_LINK) - return ida_alloc_range(&conn->tx_ida, 1, 255, GFP_ATOMIC); - + max = 255; /* Get next available identificator. * 1 - 128 are used by kernel. * 129 - 199 are reserved. * 200 - 254 are used by utilities like l2ping, etc. */ - return ida_alloc_range(&conn->tx_ida, 1, 128, GFP_ATOMIC); + else + max = 128; + + /* Allocate ident using min as last used + 1 (cyclic) */ + ident = ida_alloc_range(&conn->tx_ida, READ_ONCE(conn->tx_ident) + 1, + max, GFP_ATOMIC); + /* Force min 1 to start over */ + if (ident <= 0) { + ident = ida_alloc_range(&conn->tx_ida, 1, max, GFP_ATOMIC); + if (ident <= 0) { + /* If all idents are in use, log an error, this is + * extremely unlikely to happen and would indicate a bug + * in the code that idents are not being freed properly. + */ + BT_ERR("Unable to allocate ident: %d", ident); + return 0; + } + } + + WRITE_ONCE(conn->tx_ident, ident); + + return ident; } static void l2cap_send_acl(struct l2cap_conn *conn, struct sk_buff *skb, From 8f9f64c8f90dca07d3b9f1d7ce5d34ccd246c9dd Mon Sep 17 00:00:00 2001 From: Luca Leonardo Scorcia Date: Tue, 17 Mar 2026 11:02:06 +0000 Subject: [PATCH 224/352] pinctrl: mediatek: common: Fix probe failure for devices without EINT Some pinctrl devices like mt6397 or mt6392 don't support EINT at all, but the mtk_eint_init function is always called and returns -ENODEV, which then bubbles up and causes probe failure. To address this only call mtk_eint_init if EINT pins are present. Tested on Xiaomi Mi Smart Clock x04g (mt6392). Fixes: e46df235b4e6 ("pinctrl: mediatek: refactor EINT related code for all MediaTek pinctrl can fit") Signed-off-by: Luca Leonardo Scorcia Reviewed-by: AngeloGioacchino Del Regno Signed-off-by: Linus Walleij --- drivers/pinctrl/mediatek/pinctrl-mtk-common.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c index d6a46fe0cda8..3f518dce6d23 100644 --- a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c +++ b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c @@ -1135,9 +1135,12 @@ int mtk_pctrl_init(struct platform_device *pdev, goto chip_error; } - ret = mtk_eint_init(pctl, pdev); - if (ret) - goto chip_error; + /* Only initialize EINT if we have EINT pins */ + if (data->eint_hw.ap_num > 0) { + ret = mtk_eint_init(pctl, pdev); + if (ret) + goto chip_error; + } return 0; From 23b5df09c27aec13962b30d32a4167ebdd043f8e Mon Sep 17 00:00:00 2001 From: Paulo Alcantara Date: Wed, 18 Mar 2026 21:43:51 -0300 Subject: [PATCH 225/352] smb: client: fix generic/694 due to wrong ->i_blocks When updating ->i_size, make sure to always update ->i_blocks as well until we query new allocation size from the server. generic/694 was failing because smb3_simple_falloc() was missing the update of ->i_blocks after calling cifs_setsize(). So, fix this by updating ->i_blocks directly in cifs_setsize(), so all places that call it doesn't need to worry about updating ->i_blocks later. Reported-by: Shyam Prasad N Closes: https://lore.kernel.org/r/CANT5p=rqgRwaADB=b_PhJkqXjtfq3SFv41SSTXSVEHnuh871pA@mail.gmail.com Signed-off-by: Paulo Alcantara (Red Hat) Cc: David Howells Cc: linux-cifs@vger.kernel.org Signed-off-by: Steve French --- fs/smb/client/cifsglob.h | 6 ++++++ fs/smb/client/file.c | 1 - fs/smb/client/inode.c | 21 ++++++--------------- fs/smb/client/smb2ops.c | 20 ++++---------------- 4 files changed, 16 insertions(+), 32 deletions(-) diff --git a/fs/smb/client/cifsglob.h b/fs/smb/client/cifsglob.h index 7877d327dbb0..709e96e07791 100644 --- a/fs/smb/client/cifsglob.h +++ b/fs/smb/client/cifsglob.h @@ -2386,4 +2386,10 @@ static inline int cifs_open_create_options(unsigned int oflags, int opts) return opts; } +/* + * The number of blocks is not related to (i_size / i_blksize), but instead + * 512 byte (2**9) size is required for calculating num blocks. + */ +#define CIFS_INO_BLOCKS(size) DIV_ROUND_UP_ULL((u64)(size), 512) + #endif /* _CIFS_GLOB_H */ diff --git a/fs/smb/client/file.c b/fs/smb/client/file.c index 27f61fe7e4e2..a69e05f86d7e 100644 --- a/fs/smb/client/file.c +++ b/fs/smb/client/file.c @@ -993,7 +993,6 @@ static int cifs_do_truncate(const unsigned int xid, struct dentry *dentry) if (!rc) { netfs_resize_file(&cinode->netfs, 0, true); cifs_setsize(inode, 0); - inode->i_blocks = 0; } } if (cfile) diff --git a/fs/smb/client/inode.c b/fs/smb/client/inode.c index 143fa2e665ed..888f9e35f14b 100644 --- a/fs/smb/client/inode.c +++ b/fs/smb/client/inode.c @@ -219,13 +219,7 @@ cifs_fattr_to_inode(struct inode *inode, struct cifs_fattr *fattr, */ if (is_size_safe_to_change(cifs_i, fattr->cf_eof, from_readdir)) { i_size_write(inode, fattr->cf_eof); - - /* - * i_blocks is not related to (i_size / i_blksize), - * but instead 512 byte (2**9) size is required for - * calculating num blocks. - */ - inode->i_blocks = (512 - 1 + fattr->cf_bytes) >> 9; + inode->i_blocks = CIFS_INO_BLOCKS(fattr->cf_bytes); } if (S_ISLNK(fattr->cf_mode) && fattr->cf_symlink_target) { @@ -3015,6 +3009,11 @@ void cifs_setsize(struct inode *inode, loff_t offset) { spin_lock(&inode->i_lock); i_size_write(inode, offset); + /* + * Until we can query the server for actual allocation size, + * this is best estimate we have for blocks allocated for a file. + */ + inode->i_blocks = CIFS_INO_BLOCKS(offset); spin_unlock(&inode->i_lock); inode_set_mtime_to_ts(inode, inode_set_ctime_current(inode)); truncate_pagecache(inode, offset); @@ -3087,14 +3086,6 @@ int cifs_file_set_size(const unsigned int xid, struct dentry *dentry, if (rc == 0) { netfs_resize_file(&cifsInode->netfs, size, true); cifs_setsize(inode, size); - /* - * i_blocks is not related to (i_size / i_blksize), but instead - * 512 byte (2**9) size is required for calculating num blocks. - * Until we can query the server for actual allocation size, - * this is best estimate we have for blocks allocated for a file - * Number of blocks must be rounded up so size 1 is not 0 blocks - */ - inode->i_blocks = (512 - 1 + size) >> 9; } return rc; diff --git a/fs/smb/client/smb2ops.c b/fs/smb/client/smb2ops.c index 98ac4e86bf99..509fcea28a42 100644 --- a/fs/smb/client/smb2ops.c +++ b/fs/smb/client/smb2ops.c @@ -1497,6 +1497,7 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, { struct smb2_file_network_open_info file_inf; struct inode *inode; + u64 asize; int rc; rc = __SMB2_close(xid, tcon, cfile->fid.persistent_fid, @@ -1520,14 +1521,9 @@ smb2_close_getattr(const unsigned int xid, struct cifs_tcon *tcon, inode_set_atime_to_ts(inode, cifs_NTtimeToUnix(file_inf.LastAccessTime)); - /* - * i_blocks is not related to (i_size / i_blksize), - * but instead 512 byte (2**9) size is required for - * calculating num blocks. - */ - if (le64_to_cpu(file_inf.AllocationSize) > 4096) - inode->i_blocks = - (512 - 1 + le64_to_cpu(file_inf.AllocationSize)) >> 9; + asize = le64_to_cpu(file_inf.AllocationSize); + if (asize > 4096) + inode->i_blocks = CIFS_INO_BLOCKS(asize); /* End of file and Attributes should not have to be updated on close */ spin_unlock(&inode->i_lock); @@ -2204,14 +2200,6 @@ smb2_duplicate_extents(const unsigned int xid, rc = smb2_set_file_size(xid, tcon, trgtfile, dest_off + len, false); if (rc) goto duplicate_extents_out; - - /* - * Although also could set plausible allocation size (i_blocks) - * here in addition to setting the file size, in reflink - * it is likely that the target file is sparse. Its allocation - * size will be queried on next revalidate, but it is important - * to make sure that file's cached size is updated immediately - */ netfs_resize_file(netfs_inode(inode), dest_off + len, true); cifs_setsize(inode, dest_off + len); } From 3ecd3e03144b38a21a3b70254f1b9d2e16629b09 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 19 Mar 2026 14:29:09 -0600 Subject: [PATCH 226/352] io_uring/kbuf: fix missing BUF_MORE for incremental buffers at EOF For a zero length transfer, io_kbuf_inc_commit() is called with !len. Since we never enter the while loop to consume the buffers, io_kbuf_inc_commit() ends up returning true, consuming the buffer. But if no data was consumed, by definition it cannot have consumed the buffer. Return false for that case. Reported-by: Martin Michaelis Cc: stable@vger.kernel.org Fixes: ae98dbf43d75 ("io_uring/kbuf: add support for incremental buffer consumption") Link: https://github.com/axboe/liburing/issues/1553 Signed-off-by: Jens Axboe --- io_uring/kbuf.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/io_uring/kbuf.c b/io_uring/kbuf.c index e7f444953dfb..a4cb6752b7aa 100644 --- a/io_uring/kbuf.c +++ b/io_uring/kbuf.c @@ -34,6 +34,10 @@ struct io_provide_buf { static bool io_kbuf_inc_commit(struct io_buffer_list *bl, int len) { + /* No data consumed, return false early to avoid consuming the buffer */ + if (!len) + return false; + while (len) { struct io_uring_buf *buf; u32 buf_len, this_len; From 418eab7a6f3c002d8e64d6e95ec27118017019af Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 19 Mar 2026 14:29:20 -0600 Subject: [PATCH 227/352] io_uring/kbuf: propagate BUF_MORE through early buffer commit path When io_should_commit() returns true (eg for non-pollable files), buffer commit happens at buffer selection time and sel->buf_list is set to NULL. When __io_put_kbufs() generates CQE flags at completion time, it calls __io_put_kbuf_ring() which finds a NULL buffer_list and hence cannot determine whether the buffer was consumed or not. This means that IORING_CQE_F_BUF_MORE is never set for non-pollable input with incrementally consumed buffers. Likewise for io_buffers_select(), which always commits upfront and discards the return value of io_kbuf_commit(). Add REQ_F_BUF_MORE to store the result of io_kbuf_commit() during early commit. Then __io_put_kbuf_ring() can check this flag and set IORING_F_BUF_MORE accordingy. Reported-by: Martin Michaelis Cc: stable@vger.kernel.org Fixes: ae98dbf43d75 ("io_uring/kbuf: add support for incremental buffer consumption") Link: https://github.com/axboe/liburing/issues/1553 Signed-off-by: Jens Axboe --- include/linux/io_uring_types.h | 3 +++ io_uring/kbuf.c | 10 +++++++--- 2 files changed, 10 insertions(+), 3 deletions(-) diff --git a/include/linux/io_uring_types.h b/include/linux/io_uring_types.h index dd1420bfcb73..214fdbd49052 100644 --- a/include/linux/io_uring_types.h +++ b/include/linux/io_uring_types.h @@ -541,6 +541,7 @@ enum { REQ_F_BL_NO_RECYCLE_BIT, REQ_F_BUFFERS_COMMIT_BIT, REQ_F_BUF_NODE_BIT, + REQ_F_BUF_MORE_BIT, REQ_F_HAS_METADATA_BIT, REQ_F_IMPORT_BUFFER_BIT, REQ_F_SQE_COPIED_BIT, @@ -626,6 +627,8 @@ enum { REQ_F_BUFFERS_COMMIT = IO_REQ_FLAG(REQ_F_BUFFERS_COMMIT_BIT), /* buf node is valid */ REQ_F_BUF_NODE = IO_REQ_FLAG(REQ_F_BUF_NODE_BIT), + /* incremental buffer consumption, more space available */ + REQ_F_BUF_MORE = IO_REQ_FLAG(REQ_F_BUF_MORE_BIT), /* request has read/write metadata assigned */ REQ_F_HAS_METADATA = IO_REQ_FLAG(REQ_F_HAS_METADATA_BIT), /* diff --git a/io_uring/kbuf.c b/io_uring/kbuf.c index a4cb6752b7aa..f72f38d22d2b 100644 --- a/io_uring/kbuf.c +++ b/io_uring/kbuf.c @@ -216,7 +216,8 @@ static struct io_br_sel io_ring_buffer_select(struct io_kiocb *req, size_t *len, sel.addr = u64_to_user_ptr(READ_ONCE(buf->addr)); if (io_should_commit(req, issue_flags)) { - io_kbuf_commit(req, sel.buf_list, *len, 1); + if (!io_kbuf_commit(req, sel.buf_list, *len, 1)) + req->flags |= REQ_F_BUF_MORE; sel.buf_list = NULL; } return sel; @@ -349,7 +350,8 @@ int io_buffers_select(struct io_kiocb *req, struct buf_sel_arg *arg, */ if (ret > 0) { req->flags |= REQ_F_BUFFERS_COMMIT | REQ_F_BL_NO_RECYCLE; - io_kbuf_commit(req, sel->buf_list, arg->out_len, ret); + if (!io_kbuf_commit(req, sel->buf_list, arg->out_len, ret)) + req->flags |= REQ_F_BUF_MORE; } } else { ret = io_provided_buffers_select(req, &arg->out_len, sel->buf_list, arg->iovs); @@ -395,8 +397,10 @@ static inline bool __io_put_kbuf_ring(struct io_kiocb *req, if (bl) ret = io_kbuf_commit(req, bl, len, nr); + if (ret && (req->flags & REQ_F_BUF_MORE)) + ret = false; - req->flags &= ~REQ_F_BUFFER_RING; + req->flags &= ~(REQ_F_BUFFER_RING | REQ_F_BUF_MORE); return ret; } From be5c5280cf2b20e363dc8e2a424dd200a29b1c77 Mon Sep 17 00:00:00 2001 From: Cui Chao Date: Thu, 19 Mar 2026 15:45:35 +0800 Subject: [PATCH 228/352] cxl: Adjust the startup priority of cxl_pmem to be higher than that of cxl_acpi During the cxl_acpi probe process, it checks whether the cxl_nvb device and driver have been attached. Currently, the startup priority of the cxl_pmem driver is lower than that of the cxl_acpi driver. At this point, the cxl_nvb driver has not yet been registered on the cxl_bus, causing the attachment check to fail. This results in a failure to add the root nvdimm bridge, leading to a cxl_acpi probe failure and ultimately affecting the subsequent loading of cxl drivers. As a consequence, only one mem device object exists on the cxl_bus, while the cxl_port device objects and decoder device objects are missing. The solution is to raise the startup priority of cxl_pmem to be higher than that of cxl_acpi, ensuring that the cxl_pmem driver is registered before the aforementioned attachment check occurs. Co-developed-by: Wang Yinfeng Signed-off-by: Wang Yinfeng Signed-off-by: Cui Chao Fixes: e7e222ad73d9 ("cxl: Move devm_cxl_add_nvdimm_bridge() to cxl_pmem.ko") Reviewed-by: Dan Williams Link: https://patch.msgid.link/20260319074535.1709250-1-cuichao1753@phytium.com.cn Signed-off-by: Dave Jiang --- drivers/cxl/pmem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cxl/pmem.c b/drivers/cxl/pmem.c index 082ec0f1c3a0..261dff7ced9f 100644 --- a/drivers/cxl/pmem.c +++ b/drivers/cxl/pmem.c @@ -554,7 +554,7 @@ static __exit void cxl_pmem_exit(void) MODULE_DESCRIPTION("CXL PMEM: Persistent Memory Support"); MODULE_LICENSE("GPL v2"); -module_init(cxl_pmem_init); +subsys_initcall(cxl_pmem_init); module_exit(cxl_pmem_exit); MODULE_IMPORT_NS("CXL"); MODULE_ALIAS_CXL(CXL_DEVICE_NVDIMM_BRIDGE); From aa79f996eb41e95aed85a1bd7f56bcd6a3842008 Mon Sep 17 00:00:00 2001 From: Johan Hovold Date: Mon, 9 Mar 2026 08:50:16 +0100 Subject: [PATCH 229/352] i2c: cp2615: fix serial string NULL-deref at probe MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The cp2615 driver uses the USB device serial string as the i2c adapter name but does not make sure that the string exists. Verify that the device has a serial number before accessing it to avoid triggering a NULL-pointer dereference (e.g. with malicious devices). Fixes: 4a7695429ead ("i2c: cp2615: add i2c driver for Silicon Labs' CP2615 Digital Audio Bridge") Cc: stable@vger.kernel.org # 5.13 Cc: Bence Csókás Signed-off-by: Johan Hovold Reviewed-by: Bence Csókás Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/20260309075016.25612-1-johan@kernel.org --- drivers/i2c/busses/i2c-cp2615.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/i2c/busses/i2c-cp2615.c b/drivers/i2c/busses/i2c-cp2615.c index e2d7cd2390fc..8212875700e1 100644 --- a/drivers/i2c/busses/i2c-cp2615.c +++ b/drivers/i2c/busses/i2c-cp2615.c @@ -298,6 +298,9 @@ cp2615_i2c_probe(struct usb_interface *usbif, const struct usb_device_id *id) if (!adap) return -ENOMEM; + if (!usbdev->serial) + return -EINVAL; + strscpy(adap->name, usbdev->serial, sizeof(adap->name)); adap->owner = THIS_MODULE; adap->dev.parent = &usbif->dev; From be627abcc0d5dbd5882873bd85fbc18aa3d189ed Mon Sep 17 00:00:00 2001 From: Christophe JAILLET Date: Sun, 1 Mar 2026 17:21:01 +0100 Subject: [PATCH 230/352] i2c: fsi: Fix a potential leak in fsi_i2c_probe() In the commit in Fixes:, when the code has been updated to use an explicit for loop, instead of for_each_available_child_of_node(), the assumption that a reference to a device_node structure would be released at each iteration has been broken. Now, an explicit of_node_put() is needed to release the reference. Fixes: 095561f476ab ("i2c: fsi: Create busses for all ports") Signed-off-by: Christophe JAILLET Cc: # v5.3+ Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/fd805c39f8de51edf303856103d782138a1633c8.1772382022.git.christophe.jaillet@wanadoo.fr --- drivers/i2c/busses/i2c-fsi.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/i2c/busses/i2c-fsi.c b/drivers/i2c/busses/i2c-fsi.c index 82c87e04ac6f..b2dc5ae1d0e4 100644 --- a/drivers/i2c/busses/i2c-fsi.c +++ b/drivers/i2c/busses/i2c-fsi.c @@ -729,6 +729,7 @@ static int fsi_i2c_probe(struct fsi_device *fsi_dev) rc = i2c_add_adapter(&port->adapter); if (rc < 0) { dev_err(dev, "Failed to register adapter: %d\n", rc); + of_node_put(np); kfree(port); continue; } From cbcb3cfcdc436d6f91a3d95ecfa9c831abe14aed Mon Sep 17 00:00:00 2001 From: Mohammad Heib Date: Tue, 17 Mar 2026 19:08:06 +0200 Subject: [PATCH 231/352] ionic: fix persistent MAC address override on PF The use of IONIC_CMD_LIF_SETATTR in the MAC address update path causes the ionic firmware to update the LIF's identity in its persistent state. Since the firmware state is maintained across host warm boots and driver reloads, any MAC change on the Physical Function (PF) becomes "sticky. This is problematic because it causes ethtool -P to report the user-configured MAC as the permanent factory address, which breaks system management tools that rely on a stable hardware identity. While Virtual Functions (VFs) need this hardware-level programming to properly handle MAC assignments in guest environments, the PF should maintain standard transient behavior. This patch gates the ionic_program_mac call using is_virtfn so that PF MAC changes remain local to the netdev filters and do not overwrite the firmware's permanent identity block. Fixes: 19058be7c48c ("ionic: VF initial random MAC address if no assigned mac") Signed-off-by: Mohammad Heib Reviewed-by: Simon Horman Reviewed-by: Brett Creeley Link: https://patch.msgid.link/20260317170806.35390-1-mheib@redhat.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/pensando/ionic/ionic_lif.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/drivers/net/ethernet/pensando/ionic/ionic_lif.c b/drivers/net/ethernet/pensando/ionic/ionic_lif.c index 8d040e611d5a..637e635bbf03 100644 --- a/drivers/net/ethernet/pensando/ionic/ionic_lif.c +++ b/drivers/net/ethernet/pensando/ionic/ionic_lif.c @@ -1719,13 +1719,18 @@ static int ionic_set_mac_address(struct net_device *netdev, void *sa) if (ether_addr_equal(netdev->dev_addr, mac)) return 0; - err = ionic_program_mac(lif, mac); - if (err < 0) - return err; + /* Only program macs for virtual functions to avoid losing the permanent + * Mac across warm reset/reboot. + */ + if (lif->ionic->pdev->is_virtfn) { + err = ionic_program_mac(lif, mac); + if (err < 0) + return err; - if (err > 0) - netdev_dbg(netdev, "%s: SET and GET ATTR Mac are not equal-due to old FW running\n", - __func__); + if (err > 0) + netdev_dbg(netdev, "%s: SET and GET ATTR Mac are not equal-due to old FW running\n", + __func__); + } err = eth_prepare_mac_addr_change(netdev, addr); if (err) From 78a6ee14f8b9e1c8f7c77612122444f3be8dc8cc Mon Sep 17 00:00:00 2001 From: Gabor Juhos Date: Thu, 26 Feb 2026 14:11:27 +0100 Subject: [PATCH 232/352] i2c: pxa: defer reset on Armada 3700 when recovery is used The I2C communication is completely broken on the Armada 3700 platform since commit 0b01392c18b9 ("i2c: pxa: move to generic GPIO recovery"). For example, on the Methode uDPU board, probing of the two onboard temperature sensors fails ... [ 7.271713] i2c i2c-0: using pinctrl states for GPIO recovery [ 7.277503] i2c i2c-0: PXA I2C adapter [ 7.282199] i2c i2c-1: using pinctrl states for GPIO recovery [ 7.288241] i2c i2c-1: PXA I2C adapter [ 7.292947] sfp sfp-eth1: Host maximum power 3.0W [ 7.299614] sfp sfp-eth0: Host maximum power 3.0W [ 7.308178] lm75 1-0048: supply vs not found, using dummy regulator [ 32.489631] lm75 1-0048: probe with driver lm75 failed with error -121 [ 32.496833] lm75 1-0049: supply vs not found, using dummy regulator [ 82.890614] lm75 1-0049: probe with driver lm75 failed with error -121 ... and accessing the plugged-in SFP modules also does not work: [ 511.298537] sfp sfp-eth1: please wait, module slow to respond [ 536.488530] sfp sfp-eth0: please wait, module slow to respond ... [ 1065.688536] sfp sfp-eth1: failed to read EEPROM: -EREMOTEIO [ 1090.888532] sfp sfp-eth0: failed to read EEPROM: -EREMOTEIO After a discussion [1], there was an attempt to fix the problem by reverting the offending change by commit 7b211c767121 ("Revert "i2c: pxa: move to generic GPIO recovery""), but that only helped to fix the issue in the 6.1.y stable tree. The reason behind the partial succes is that there was another change in commit 20cb3fce4d60 ("i2c: Set i2c pinctrl recovery info from it's device pinctrl") in the 6.3-rc1 cycle which broke things further. The cause of the problem is the same in case of both offending commits mentioned above. Namely, the I2C core code changes the pinctrl state to GPIO while running the recovery initialization code. Although the PXA specific initialization also does this, but the key difference is that it happens before the controller is getting enabled in i2c_pxa_reset(), whereas in the case of the generic initialization it happens after that. Change the code to reset the controller only before the first transfer instead of before registering the controller. This ensures that the controller is not enabled at the time when the generic recovery code performs the pinctrl state changes, thus avoids the problem described above. As the result this change restores the original behaviour, which in turn makes the I2C communication to work again as it can be seen from the following log: [ 7.363250] i2c i2c-0: using pinctrl states for GPIO recovery [ 7.369041] i2c i2c-0: PXA I2C adapter [ 7.373673] i2c i2c-1: using pinctrl states for GPIO recovery [ 7.379742] i2c i2c-1: PXA I2C adapter [ 7.384506] sfp sfp-eth1: Host maximum power 3.0W [ 7.393013] sfp sfp-eth0: Host maximum power 3.0W [ 7.399266] lm75 1-0048: supply vs not found, using dummy regulator [ 7.407257] hwmon hwmon0: temp1_input not attached to any thermal zone [ 7.413863] lm75 1-0048: hwmon0: sensor 'tmp75c' [ 7.418746] lm75 1-0049: supply vs not found, using dummy regulator [ 7.426371] hwmon hwmon1: temp1_input not attached to any thermal zone [ 7.432972] lm75 1-0049: hwmon1: sensor 'tmp75c' [ 7.755092] sfp sfp-eth1: module MENTECHOPTO POS22-LDCC-KR rev 1.0 sn MNC208U90009 dc 200828 [ 7.764997] mvneta d0040000.ethernet eth1: unsupported SFP module: no common interface modes [ 7.785362] sfp sfp-eth0: module Mikrotik S-RJ01 rev 1.0 sn 61B103C55C58 dc 201022 [ 7.803426] hwmon hwmon2: temp1_input not attached to any thermal zone Link: https://lore.kernel.org/r/20230926160255.330417-1-robert.marko@sartura.hr #1 Cc: stable@vger.kernel.org # 6.3+ Fixes: 20cb3fce4d60 ("i2c: Set i2c pinctrl recovery info from it's device pinctrl") Signed-off-by: Gabor Juhos Tested-by: Robert Marko Reviewed-by: Linus Walleij Signed-off-by: Andi Shyti Link: https://lore.kernel.org/r/20260226-i2c-pxa-fix-i2c-communication-v4-1-797a091dae87@gmail.com --- drivers/i2c/busses/i2c-pxa.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) diff --git a/drivers/i2c/busses/i2c-pxa.c b/drivers/i2c/busses/i2c-pxa.c index 09af3b3625f1..f55840b2eb9a 100644 --- a/drivers/i2c/busses/i2c-pxa.c +++ b/drivers/i2c/busses/i2c-pxa.c @@ -268,6 +268,7 @@ struct pxa_i2c { struct pinctrl *pinctrl; struct pinctrl_state *pinctrl_default; struct pinctrl_state *pinctrl_recovery; + bool reset_before_xfer; }; #define _IBMR(i2c) ((i2c)->reg_ibmr) @@ -1144,6 +1145,11 @@ static int i2c_pxa_xfer(struct i2c_adapter *adap, { struct pxa_i2c *i2c = adap->algo_data; + if (i2c->reset_before_xfer) { + i2c_pxa_reset(i2c); + i2c->reset_before_xfer = false; + } + return i2c_pxa_internal_xfer(i2c, msgs, num, i2c_pxa_do_xfer); } @@ -1521,7 +1527,16 @@ static int i2c_pxa_probe(struct platform_device *dev) } } - i2c_pxa_reset(i2c); + /* + * Skip reset on Armada 3700 when recovery is used to avoid + * controller hang due to the pinctrl state changes done by + * the generic recovery initialization code. The reset will + * be performed later, prior to the first transfer. + */ + if (i2c_type == REGS_A3700 && i2c->adap.bus_recovery_info) + i2c->reset_before_xfer = true; + else + i2c_pxa_reset(i2c); ret = i2c_add_numbered_adapter(&i2c->adap); if (ret < 0) From 4527025d440ce84bf56e75ce1df2e84cb8178616 Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Tue, 17 Mar 2026 12:33:34 -0700 Subject: [PATCH 233/352] nfc: nci: fix circular locking dependency in nci_close_device nci_close_device() flushes rx_wq and tx_wq while holding req_lock. This causes a circular locking dependency because nci_rx_work() running on rx_wq can end up taking req_lock too: nci_rx_work -> nci_rx_data_packet -> nci_data_exchange_complete -> __sk_destruct -> rawsock_destruct -> nfc_deactivate_target -> nci_deactivate_target -> nci_request -> mutex_lock(&ndev->req_lock) Move the flush of rx_wq after req_lock has been released. This should safe (I think) because NCI_UP has already been cleared and the transport is closed, so the work will see it and return -ENETDOWN. NIPA has been hitting this running the nci selftest with a debug kernel on roughly 4% of the runs. Fixes: 6a2968aaf50c ("NFC: basic NCI protocol implementation") Reviewed-by: Ian Ray Link: https://patch.msgid.link/20260317193334.988609-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- net/nfc/nci/core.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/net/nfc/nci/core.c b/net/nfc/nci/core.c index 43d871525dbc..5f46c4b5720f 100644 --- a/net/nfc/nci/core.c +++ b/net/nfc/nci/core.c @@ -579,8 +579,7 @@ static int nci_close_device(struct nci_dev *ndev) skb_queue_purge(&ndev->rx_q); skb_queue_purge(&ndev->tx_q); - /* Flush RX and TX wq */ - flush_workqueue(ndev->rx_wq); + /* Flush TX wq, RX wq flush can't be under the lock */ flush_workqueue(ndev->tx_wq); /* Reset device */ @@ -592,13 +591,13 @@ static int nci_close_device(struct nci_dev *ndev) msecs_to_jiffies(NCI_RESET_TIMEOUT)); /* After this point our queues are empty - * and no works are scheduled. + * rx work may be running but will see that NCI_UP was cleared */ ndev->ops->close(ndev); clear_bit(NCI_INIT, &ndev->flags); - /* Flush cmd wq */ + /* Flush cmd and tx wq */ flush_workqueue(ndev->cmd_wq); timer_delete_sync(&ndev->cmd_timer); @@ -613,6 +612,9 @@ static int nci_close_device(struct nci_dev *ndev) mutex_unlock(&ndev->req_lock); + /* rx_work may take req_lock via nci_deactivate_target */ + flush_workqueue(ndev->rx_wq); + return 0; } From 7c770dadfda5cbbde6aa3c4363ed513f1d212bf8 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Toke=20H=C3=B8iland-J=C3=B8rgensen?= Date: Wed, 18 Mar 2026 16:55:51 +0100 Subject: [PATCH 234/352] net: openvswitch: Avoid releasing netdev before teardown completes MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The patch cited in the Fixes tag below changed the teardown code for OVS ports to no longer unconditionally take the RTNL. After this change, the netdev_destroy() callback can proceed immediately to the call_rcu() invocation if the IFF_OVS_DATAPATH flag is already cleared on the netdev. The ovs_netdev_detach_dev() function clears the flag before completing the unregistration, and if it gets preempted after clearing the flag (as can happen on an -rt kernel), netdev_destroy() can complete and the device can be freed before the unregistration completes. This leads to a splat like: [ 998.393867] Oops: general protection fault, probably for non-canonical address 0xff00000001000239: 0000 [#1] SMP PTI [ 998.393877] CPU: 42 UID: 0 PID: 55177 Comm: ip Kdump: loaded Not tainted 6.12.0-211.1.1.el10_2.x86_64+rt #1 PREEMPT_RT [ 998.393886] Hardware name: Dell Inc. PowerEdge R740/0JMK61, BIOS 2.24.0 03/27/2025 [ 998.393889] RIP: 0010:dev_set_promiscuity+0x8d/0xa0 [ 998.393901] Code: 00 00 75 d8 48 8b 53 08 48 83 ba b0 02 00 00 00 75 ca 48 83 c4 08 5b c3 cc cc cc cc 48 83 bf 48 09 00 00 00 75 91 48 8b 47 08 <48> 83 b8 b0 02 00 00 00 74 97 eb 81 0f 1f 80 00 00 00 00 90 90 90 [ 998.393906] RSP: 0018:ffffce5864a5f6a0 EFLAGS: 00010246 [ 998.393912] RAX: ff00000000ffff89 RBX: ffff894d0adf5a05 RCX: 0000000000000000 [ 998.393917] RDX: 0000000000000000 RSI: 00000000ffffffff RDI: ffff894d0adf5a05 [ 998.393921] RBP: ffff894d19252000 R08: ffff894d19252000 R09: 0000000000000000 [ 998.393924] R10: ffff894d19252000 R11: ffff894d192521b8 R12: 0000000000000006 [ 998.393927] R13: ffffce5864a5f738 R14: 00000000ffffffe2 R15: 0000000000000000 [ 998.393931] FS: 00007fad61971800(0000) GS:ffff894cc0140000(0000) knlGS:0000000000000000 [ 998.393936] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 998.393940] CR2: 000055df0a2a6e40 CR3: 000000011c7fe003 CR4: 00000000007726f0 [ 998.393944] PKRU: 55555554 [ 998.393946] Call Trace: [ 998.393949] [ 998.393952] ? show_trace_log_lvl+0x1b0/0x2f0 [ 998.393961] ? show_trace_log_lvl+0x1b0/0x2f0 [ 998.393975] ? dp_device_event+0x41/0x80 [openvswitch] [ 998.394009] ? __die_body.cold+0x8/0x12 [ 998.394016] ? die_addr+0x3c/0x60 [ 998.394027] ? exc_general_protection+0x16d/0x390 [ 998.394042] ? asm_exc_general_protection+0x26/0x30 [ 998.394058] ? dev_set_promiscuity+0x8d/0xa0 [ 998.394066] ? ovs_netdev_detach_dev+0x3a/0x80 [openvswitch] [ 998.394092] dp_device_event+0x41/0x80 [openvswitch] [ 998.394102] notifier_call_chain+0x5a/0xd0 [ 998.394106] unregister_netdevice_many_notify+0x51b/0xa60 [ 998.394110] rtnl_dellink+0x169/0x3e0 [ 998.394121] ? rt_mutex_slowlock.constprop.0+0x95/0xd0 [ 998.394125] rtnetlink_rcv_msg+0x142/0x3f0 [ 998.394128] ? avc_has_perm_noaudit+0x69/0xf0 [ 998.394130] ? __pfx_rtnetlink_rcv_msg+0x10/0x10 [ 998.394132] netlink_rcv_skb+0x50/0x100 [ 998.394138] netlink_unicast+0x292/0x3f0 [ 998.394141] netlink_sendmsg+0x21b/0x470 [ 998.394145] ____sys_sendmsg+0x39d/0x3d0 [ 998.394149] ___sys_sendmsg+0x9a/0xe0 [ 998.394156] __sys_sendmsg+0x7a/0xd0 [ 998.394160] do_syscall_64+0x7f/0x170 [ 998.394162] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 998.394165] RIP: 0033:0x7fad61bf4724 [ 998.394188] Code: 89 02 b8 ff ff ff ff eb bb 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 80 3d c5 e9 0c 00 00 74 13 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 48 83 ec 28 89 54 24 1c 48 89 [ 998.394189] RSP: 002b:00007ffd7e2f7cb8 EFLAGS: 00000202 ORIG_RAX: 000000000000002e [ 998.394191] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007fad61bf4724 [ 998.394193] RDX: 0000000000000000 RSI: 00007ffd7e2f7d20 RDI: 0000000000000003 [ 998.394194] RBP: 00007ffd7e2f7d90 R08: 0000000000000010 R09: 000000000000003f [ 998.394195] R10: 000055df11558010 R11: 0000000000000202 R12: 00007ffd7e2f8380 [ 998.394196] R13: 0000000069b233d7 R14: 000055df0a256040 R15: 0000000000000000 [ 998.394200] To fix this, reorder the operations in ovs_netdev_detach_dev() to only clear the flag after completing the other operations, and introduce an smp_wmb() to make the ordering requirement explicit. The smp_wmb() is paired with a full smp_mb() in netdev_destroy() to make sure the call_rcu() invocation does not happen before the unregister operations are visible. Reported-by: Minxi Hou Tested-by: Minxi Hou Fixes: 549822767630 ("net: openvswitch: Avoid needlessly taking the RTNL on vport destroy") Signed-off-by: Toke Høiland-Jørgensen Link: https://patch.msgid.link/20260318155554.1133405-1-toke@redhat.com Signed-off-by: Jakub Kicinski --- net/openvswitch/vport-netdev.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index 6574f9bcdc02..c688dee96503 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -151,11 +151,15 @@ static void vport_netdev_free(struct rcu_head *rcu) void ovs_netdev_detach_dev(struct vport *vport) { ASSERT_RTNL(); - vport->dev->priv_flags &= ~IFF_OVS_DATAPATH; netdev_rx_handler_unregister(vport->dev); netdev_upper_dev_unlink(vport->dev, netdev_master_upper_dev_get(vport->dev)); dev_set_promiscuity(vport->dev, -1); + + /* paired with smp_mb() in netdev_destroy() */ + smp_wmb(); + + vport->dev->priv_flags &= ~IFF_OVS_DATAPATH; } static void netdev_destroy(struct vport *vport) @@ -174,6 +178,9 @@ static void netdev_destroy(struct vport *vport) rtnl_unlock(); } + /* paired with smp_wmb() in ovs_netdev_detach_dev() */ + smp_mb(); + call_rcu(&vport->rcu, vport_netdev_free); } From e6d7eba23b666d85cacee0643be280d6ce1ebffc Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Fri, 20 Mar 2026 12:48:01 +0900 Subject: [PATCH 235/352] ata: libata-scsi: report correct sense field pointer in ata_scsiop_maint_in() Commit 4ab7bb976343 ("ata: libata-scsi: Refactor ata_scsiop_maint_in()") modified ata_scsiop_maint_in() to directly call ata_scsi_set_invalid_field() to set the field pointer of the sense data of a failed MAINTENANCE IN command. However, in the case of an invalid command format, the sense data field incorrectly indicates byte 1 of the CDB. Fix this to indicate byte 2 of the command. Reported-by: Guenter Roeck Fixes: 4ab7bb976343 ("ata: libata-scsi: Refactor ata_scsiop_maint_in()") Cc: stable@vger.kernel.org Signed-off-by: Damien Le Moal Signed-off-by: Niklas Cassel --- drivers/ata/libata-scsi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index ad798e5246b4..3b65df914ebb 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -3600,7 +3600,7 @@ static unsigned int ata_scsiop_maint_in(struct ata_device *dev, if (cdb[2] != 1 && cdb[2] != 3) { ata_dev_warn(dev, "invalid command format %d\n", cdb[2]); - ata_scsi_set_invalid_field(dev, cmd, 1, 0xff); + ata_scsi_set_invalid_field(dev, cmd, 2, 0xff); return 0; } From eca58535b154e6951327319afda94ac80eae7dc3 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:45 +0200 Subject: [PATCH 236/352] dma-debug: Allow multiple invocations of overlapping entries Repeated DMA mappings with DMA_ATTR_CPU_CACHE_CLEAN trigger the following splat. This prevents using the attribute in cases where a DMA region is shared and reused more than seven times. ------------[ cut here ]------------ DMA-API: exceeded 7 overlapping mappings of cacheline 0x000000000438c440 WARNING: kernel/dma/debug.c:467 at add_dma_entry+0x219/0x280, CPU#4: ibv_rc_pingpong/1644 Modules linked in: xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink iptable_nat nf_nat xt_addrtype br_netfilter rpcsec_gss_krb5 auth_rpcgss oid_registry overlay mlx5_fwctl zram zsmalloc mlx5_ib fuse rpcrdma rdma_ucm ib_uverbs ib_iser libiscsi scsi_transport_iscsi ib_umad rdma_cm ib_ipoib iw_cm ib_cm mlx5_core ib_core CPU: 4 UID: 2733 PID: 1644 Comm: ibv_rc_pingpong Not tainted 6.19.0+ #129 PREEMPT Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:add_dma_entry+0x221/0x280 Code: c0 0f 84 f2 fe ff ff 83 e8 01 89 05 6d 99 11 01 e9 e4 fe ff ff 0f 8e 1f ff ff ff 48 8d 3d 07 ef 2d 01 be 07 00 00 00 48 89 e2 <67> 48 0f b9 3a e9 06 ff ff ff 48 c7 c7 98 05 2b 82 c6 05 72 92 28 RSP: 0018:ff1100010e657970 EFLAGS: 00010002 RAX: 0000000000000007 RBX: ff1100010234eb00 RCX: 0000000000000000 RDX: ff1100010e657970 RSI: 0000000000000007 RDI: ffffffff82678660 RBP: 000000000438c440 R08: 0000000000000228 R09: 0000000000000000 R10: 00000000000001be R11: 000000000000089d R12: 0000000000000800 R13: 00000000ffffffef R14: 0000000000000202 R15: ff1100010234eb00 FS: 00007fb15f3f6740(0000) GS:ff110008dcc19000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fb15f32d3a0 CR3: 0000000116f59001 CR4: 0000000000373eb0 Call Trace: debug_dma_map_sg+0x1b4/0x390 __dma_map_sg_attrs+0x6d/0x1a0 dma_map_sgtable+0x19/0x30 ib_umem_get+0x284/0x3b0 [ib_uverbs] mlx5_ib_reg_user_mr+0x68/0x2a0 [mlx5_ib] ib_uverbs_reg_mr+0x17f/0x2a0 [ib_uverbs] ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0xc2/0x130 [ib_uverbs] ib_uverbs_cmd_verbs+0xa0b/0xae0 [ib_uverbs] ? ib_uverbs_handler_UVERBS_METHOD_QUERY_PORT_SPEED+0xe0/0xe0 [ib_uverbs] ? mmap_region+0x7a/0xb0 ? do_mmap+0x3b8/0x5c0 ib_uverbs_ioctl+0xa7/0x110 [ib_uverbs] __x64_sys_ioctl+0x14f/0x8b0 ? ksys_mmap_pgoff+0xc5/0x190 do_syscall_64+0x8c/0xbf0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 RIP: 0033:0x7fb15f5e4eed Code: 04 25 28 00 00 00 48 89 45 c8 31 c0 48 8d 45 10 c7 45 b0 10 00 00 00 48 89 45 b8 48 8d 45 d0 48 89 45 c0 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1a 48 8b 45 c8 64 48 2b 04 25 28 00 00 00 RSP: 002b:00007ffe09a5c540 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007ffe09a5c5d0 RCX: 00007fb15f5e4eed RDX: 00007ffe09a5c5f0 RSI: 00000000c0181b01 RDI: 0000000000000003 RBP: 00007ffe09a5c590 R08: 0000000000000028 R09: 00007ffe09a5c794 R10: 0000000000000001 R11: 0000000000000246 R12: 00007ffe09a5c794 R13: 000000000000000c R14: 0000000025a49170 R15: 000000000000000c ---[ end trace 0000000000000000 ]--- Fixes: 61868dc55a11 ("dma-mapping: add DMA_ATTR_CPU_CACHE_CLEAN") Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-1-1dde90a7f08b@nvidia.com --- kernel/dma/debug.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/kernel/dma/debug.c b/kernel/dma/debug.c index 86f87e43438c..be207be74996 100644 --- a/kernel/dma/debug.c +++ b/kernel/dma/debug.c @@ -453,7 +453,7 @@ static int active_cacheline_set_overlap(phys_addr_t cln, int overlap) return overlap; } -static void active_cacheline_inc_overlap(phys_addr_t cln) +static void active_cacheline_inc_overlap(phys_addr_t cln, bool is_cache_clean) { int overlap = active_cacheline_read_overlap(cln); @@ -462,7 +462,7 @@ static void active_cacheline_inc_overlap(phys_addr_t cln) /* If we overflowed the overlap counter then we're potentially * leaking dma-mappings. */ - WARN_ONCE(overlap > ACTIVE_CACHELINE_MAX_OVERLAP, + WARN_ONCE(!is_cache_clean && overlap > ACTIVE_CACHELINE_MAX_OVERLAP, pr_fmt("exceeded %d overlapping mappings of cacheline %pa\n"), ACTIVE_CACHELINE_MAX_OVERLAP, &cln); } @@ -495,7 +495,7 @@ static int active_cacheline_insert(struct dma_debug_entry *entry, if (rc == -EEXIST) { struct dma_debug_entry *existing; - active_cacheline_inc_overlap(cln); + active_cacheline_inc_overlap(cln, entry->is_cache_clean); existing = radix_tree_lookup(&dma_active_cacheline, cln); /* A lookup failure here after we got -EEXIST is unexpected. */ WARN_ON(!existing); From 6f45b1604cf43945ef472ae4ef30354025307c19 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:46 +0200 Subject: [PATCH 237/352] dma-mapping: handle DMA_ATTR_CPU_CACHE_CLEAN in trace output Tracing prints decoded DMA attribute flags, but it does not yet include the recently added DMA_ATTR_CPU_CACHE_CLEAN. Add support for decoding and displaying this attribute in the trace output. Fixes: 61868dc55a11 ("dma-mapping: add DMA_ATTR_CPU_CACHE_CLEAN") Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-2-1dde90a7f08b@nvidia.com --- include/trace/events/dma.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/include/trace/events/dma.h b/include/trace/events/dma.h index 33e99e792f1a..69cb3805ee81 100644 --- a/include/trace/events/dma.h +++ b/include/trace/events/dma.h @@ -32,7 +32,8 @@ TRACE_DEFINE_ENUM(DMA_NONE); { DMA_ATTR_ALLOC_SINGLE_PAGES, "ALLOC_SINGLE_PAGES" }, \ { DMA_ATTR_NO_WARN, "NO_WARN" }, \ { DMA_ATTR_PRIVILEGED, "PRIVILEGED" }, \ - { DMA_ATTR_MMIO, "MMIO" }) + { DMA_ATTR_MMIO, "MMIO" }, \ + { DMA_ATTR_CPU_CACHE_CLEAN, "CACHE_CLEAN" }) DECLARE_EVENT_CLASS(dma_map, TP_PROTO(struct device *dev, phys_addr_t phys_addr, dma_addr_t dma_addr, From 9bb0a4d6a4433b75274204b083dac8e515d2007d Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:47 +0200 Subject: [PATCH 238/352] dma-mapping: Clarify valid conditions for CPU cache line overlap Rename the DMA_ATTR_CPU_CACHE_CLEAN attribute to better reflect that it is debugging aid to inform DMA core code that CPU cache line overlaps are allowed, and refine the documentation describing its use. Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-3-1dde90a7f08b@nvidia.com --- Documentation/core-api/dma-attributes.rst | 20 +++++++++++++------- drivers/virtio/virtio_ring.c | 10 +++++----- include/linux/dma-mapping.h | 8 ++++---- include/trace/events/dma.h | 2 +- kernel/dma/debug.c | 2 +- 5 files changed, 24 insertions(+), 18 deletions(-) diff --git a/Documentation/core-api/dma-attributes.rst b/Documentation/core-api/dma-attributes.rst index 1d7bfad73b1c..48cfe86cc06d 100644 --- a/Documentation/core-api/dma-attributes.rst +++ b/Documentation/core-api/dma-attributes.rst @@ -149,11 +149,17 @@ For architectures that require cache flushing for DMA coherence DMA_ATTR_MMIO will not perform any cache flushing. The address provided must never be mapped cacheable into the CPU. -DMA_ATTR_CPU_CACHE_CLEAN ------------------------- +DMA_ATTR_DEBUGGING_IGNORE_CACHELINES +------------------------------------ -This attribute indicates the CPU will not dirty any cacheline overlapping this -DMA_FROM_DEVICE/DMA_BIDIRECTIONAL buffer while it is mapped. This allows -multiple small buffers to safely share a cacheline without risk of data -corruption, suppressing DMA debug warnings about overlapping mappings. -All mappings sharing a cacheline should have this attribute. +This attribute indicates that CPU cache lines may overlap for buffers mapped +with DMA_FROM_DEVICE or DMA_BIDIRECTIONAL. + +Such overlap may occur when callers map multiple small buffers that reside +within the same cache line. In this case, callers must guarantee that the CPU +will not dirty these cache lines after the mappings are established. When this +condition is met, multiple buffers can safely share a cache line without risking +data corruption. + +All mappings that share a cache line must set this attribute to suppress DMA +debug warnings about overlapping mappings. diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 335692d41617..fbca7ce1c6bf 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -2912,10 +2912,10 @@ EXPORT_SYMBOL_GPL(virtqueue_add_inbuf); * @data: the token identifying the buffer. * @gfp: how to do memory allocations (if necessary). * - * Same as virtqueue_add_inbuf but passes DMA_ATTR_CPU_CACHE_CLEAN to indicate - * that the CPU will not dirty any cacheline overlapping this buffer while it - * is available, and to suppress overlapping cacheline warnings in DMA debug - * builds. + * Same as virtqueue_add_inbuf but passes DMA_ATTR_DEBUGGING_IGNORE_CACHELINES + * to indicate that the CPU will not dirty any cacheline overlapping this buffer + * while it is available, and to suppress overlapping cacheline warnings in DMA + * debug builds. * * Caller must ensure we don't call this with other virtqueue operations * at the same time (except where noted). @@ -2928,7 +2928,7 @@ int virtqueue_add_inbuf_cache_clean(struct virtqueue *vq, gfp_t gfp) { return virtqueue_add(vq, &sg, num, 0, 1, data, NULL, false, gfp, - DMA_ATTR_CPU_CACHE_CLEAN); + DMA_ATTR_DEBUGGING_IGNORE_CACHELINES); } EXPORT_SYMBOL_GPL(virtqueue_add_inbuf_cache_clean); diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 29973baa0581..da44394b3a1a 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -80,11 +80,11 @@ #define DMA_ATTR_MMIO (1UL << 10) /* - * DMA_ATTR_CPU_CACHE_CLEAN: Indicates the CPU will not dirty any cacheline - * overlapping this buffer while it is mapped for DMA. All mappings sharing - * a cacheline must have this attribute for this to be considered safe. + * DMA_ATTR_DEBUGGING_IGNORE_CACHELINES: Indicates the CPU cache line can be + * overlapped. All mappings sharing a cacheline must have this attribute for + * this to be considered safe. */ -#define DMA_ATTR_CPU_CACHE_CLEAN (1UL << 11) +#define DMA_ATTR_DEBUGGING_IGNORE_CACHELINES (1UL << 11) /* * A dma_addr_t can hold any valid DMA or bus address for the platform. It can diff --git a/include/trace/events/dma.h b/include/trace/events/dma.h index 69cb3805ee81..8c64bc0721fe 100644 --- a/include/trace/events/dma.h +++ b/include/trace/events/dma.h @@ -33,7 +33,7 @@ TRACE_DEFINE_ENUM(DMA_NONE); { DMA_ATTR_NO_WARN, "NO_WARN" }, \ { DMA_ATTR_PRIVILEGED, "PRIVILEGED" }, \ { DMA_ATTR_MMIO, "MMIO" }, \ - { DMA_ATTR_CPU_CACHE_CLEAN, "CACHE_CLEAN" }) + { DMA_ATTR_DEBUGGING_IGNORE_CACHELINES, "CACHELINES_OVERLAP" }) DECLARE_EVENT_CLASS(dma_map, TP_PROTO(struct device *dev, phys_addr_t phys_addr, dma_addr_t dma_addr, diff --git a/kernel/dma/debug.c b/kernel/dma/debug.c index be207be74996..83e1cfe05f08 100644 --- a/kernel/dma/debug.c +++ b/kernel/dma/debug.c @@ -601,7 +601,7 @@ static void add_dma_entry(struct dma_debug_entry *entry, unsigned long attrs) unsigned long flags; int rc; - entry->is_cache_clean = !!(attrs & DMA_ATTR_CPU_CACHE_CLEAN); + entry->is_cache_clean = attrs & DMA_ATTR_DEBUGGING_IGNORE_CACHELINES; bucket = get_hash_bucket(entry, &flags); hash_bucket_add(bucket, entry); From e6a58fa2556203a7f6731b4071705dc81cca5ca5 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:48 +0200 Subject: [PATCH 239/352] dma-mapping: Introduce DMA require coherency attribute The mapping buffers which carry this attribute require DMA coherent system. This means that they can't take SWIOTLB path, can perform CPU cache overlap and doesn't perform cache flushing. Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-4-1dde90a7f08b@nvidia.com --- Documentation/core-api/dma-attributes.rst | 16 ++++++++++++++++ include/linux/dma-mapping.h | 7 +++++++ include/trace/events/dma.h | 3 ++- kernel/dma/debug.c | 3 ++- kernel/dma/mapping.c | 6 ++++++ 5 files changed, 33 insertions(+), 2 deletions(-) diff --git a/Documentation/core-api/dma-attributes.rst b/Documentation/core-api/dma-attributes.rst index 48cfe86cc06d..123c8468d58f 100644 --- a/Documentation/core-api/dma-attributes.rst +++ b/Documentation/core-api/dma-attributes.rst @@ -163,3 +163,19 @@ data corruption. All mappings that share a cache line must set this attribute to suppress DMA debug warnings about overlapping mappings. + +DMA_ATTR_REQUIRE_COHERENT +------------------------- + +DMA mapping requests with the DMA_ATTR_REQUIRE_COHERENT fail on any +system where SWIOTLB or cache management is required. This should only +be used to support uAPI designs that require continuous HW DMA +coherence with userspace processes, for example RDMA and DRM. At a +minimum the memory being mapped must be userspace memory from +pin_user_pages() or similar. + +Drivers should consider using dma_mmap_pages() instead of this +interface when building their uAPIs, when possible. + +It must never be used in an in-kernel driver that only works with +kernel memory. diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index da44394b3a1a..482b919f040f 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -86,6 +86,13 @@ */ #define DMA_ATTR_DEBUGGING_IGNORE_CACHELINES (1UL << 11) +/* + * DMA_ATTR_REQUIRE_COHERENT: Indicates that DMA coherency is required. + * All mappings that carry this attribute can't work with SWIOTLB and cache + * flushing. + */ +#define DMA_ATTR_REQUIRE_COHERENT (1UL << 12) + /* * A dma_addr_t can hold any valid DMA or bus address for the platform. It can * be given to a device to use as a DMA source or target. It is specific to a diff --git a/include/trace/events/dma.h b/include/trace/events/dma.h index 8c64bc0721fe..63597b004424 100644 --- a/include/trace/events/dma.h +++ b/include/trace/events/dma.h @@ -33,7 +33,8 @@ TRACE_DEFINE_ENUM(DMA_NONE); { DMA_ATTR_NO_WARN, "NO_WARN" }, \ { DMA_ATTR_PRIVILEGED, "PRIVILEGED" }, \ { DMA_ATTR_MMIO, "MMIO" }, \ - { DMA_ATTR_DEBUGGING_IGNORE_CACHELINES, "CACHELINES_OVERLAP" }) + { DMA_ATTR_DEBUGGING_IGNORE_CACHELINES, "CACHELINES_OVERLAP" }, \ + { DMA_ATTR_REQUIRE_COHERENT, "REQUIRE_COHERENT" }) DECLARE_EVENT_CLASS(dma_map, TP_PROTO(struct device *dev, phys_addr_t phys_addr, dma_addr_t dma_addr, diff --git a/kernel/dma/debug.c b/kernel/dma/debug.c index 83e1cfe05f08..0677918f06a8 100644 --- a/kernel/dma/debug.c +++ b/kernel/dma/debug.c @@ -601,7 +601,8 @@ static void add_dma_entry(struct dma_debug_entry *entry, unsigned long attrs) unsigned long flags; int rc; - entry->is_cache_clean = attrs & DMA_ATTR_DEBUGGING_IGNORE_CACHELINES; + entry->is_cache_clean = attrs & (DMA_ATTR_DEBUGGING_IGNORE_CACHELINES | + DMA_ATTR_REQUIRE_COHERENT); bucket = get_hash_bucket(entry, &flags); hash_bucket_add(bucket, entry); diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c index 3928a509c44c..6d3dd0bd3a88 100644 --- a/kernel/dma/mapping.c +++ b/kernel/dma/mapping.c @@ -164,6 +164,9 @@ dma_addr_t dma_map_phys(struct device *dev, phys_addr_t phys, size_t size, if (WARN_ON_ONCE(!dev->dma_mask)) return DMA_MAPPING_ERROR; + if (!dev_is_dma_coherent(dev) && (attrs & DMA_ATTR_REQUIRE_COHERENT)) + return DMA_MAPPING_ERROR; + if (dma_map_direct(dev, ops) || (!is_mmio && arch_dma_map_phys_direct(dev, phys + size))) addr = dma_direct_map_phys(dev, phys, size, dir, attrs); @@ -235,6 +238,9 @@ static int __dma_map_sg_attrs(struct device *dev, struct scatterlist *sg, BUG_ON(!valid_dma_direction(dir)); + if (!dev_is_dma_coherent(dev) && (attrs & DMA_ATTR_REQUIRE_COHERENT)) + return -EOPNOTSUPP; + if (WARN_ON_ONCE(!dev->dma_mask)) return 0; From 453b8fb68f3641fea970db88b7d9a153ed2a37e8 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Thu, 9 Oct 2025 16:54:58 +0200 Subject: [PATCH 240/352] xen/privcmd: restrict usage in unprivileged domU The Xen privcmd driver allows to issue arbitrary hypercalls from user space processes. This is normally no problem, as access is usually limited to root and the hypervisor will deny any hypercalls affecting other domains. In case the guest is booted using secure boot, however, the privcmd driver would be enabling a root user process to modify e.g. kernel memory contents, thus breaking the secure boot feature. The only known case where an unprivileged domU is really needing to use the privcmd driver is the case when it is acting as the device model for another guest. In this case all hypercalls issued via the privcmd driver will target that other guest. Fortunately the privcmd driver can already be locked down to allow only hypercalls targeting a specific domain, but this mode can be activated from user land only today. The target domain can be obtained from Xenstore, so when not running in dom0 restrict the privcmd driver to that target domain from the beginning, resolving the potential problem of breaking secure boot. This is XSA-482 Reported-by: Teddy Astie Fixes: 1c5de1939c20 ("xen: add privcmd driver") Signed-off-by: Juergen Gross --- V2: - defer reading from Xenstore if Xenstore isn't ready yet (Jan Beulich) - wait in open() if target domain isn't known yet - issue message in case no target domain found (Jan Beulich) --- drivers/xen/privcmd.c | 60 ++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 57 insertions(+), 3 deletions(-) diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c index 1759cc18753f..a83bad69f4f2 100644 --- a/drivers/xen/privcmd.c +++ b/drivers/xen/privcmd.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -30,7 +31,9 @@ #include #include #include +#include #include +#include #include #include @@ -46,6 +49,7 @@ #include #include #include +#include #ifdef CONFIG_XEN_ACPI #include #endif @@ -72,6 +76,11 @@ struct privcmd_data { domid_t domid; }; +/* DOMID_INVALID implies no restriction */ +static domid_t target_domain = DOMID_INVALID; +static bool restrict_wait; +static DECLARE_WAIT_QUEUE_HEAD(restrict_wait_wq); + static int privcmd_vma_range_is_mapped( struct vm_area_struct *vma, unsigned long addr, @@ -1563,13 +1572,16 @@ static long privcmd_ioctl(struct file *file, static int privcmd_open(struct inode *ino, struct file *file) { - struct privcmd_data *data = kzalloc_obj(*data); + struct privcmd_data *data; + if (wait_event_interruptible(restrict_wait_wq, !restrict_wait) < 0) + return -EINTR; + + data = kzalloc_obj(*data); if (!data) return -ENOMEM; - /* DOMID_INVALID implies no restriction */ - data->domid = DOMID_INVALID; + data->domid = target_domain; file->private_data = data; return 0; @@ -1662,6 +1674,45 @@ static struct miscdevice privcmd_dev = { .fops = &xen_privcmd_fops, }; +static int init_restrict(struct notifier_block *notifier, + unsigned long event, + void *data) +{ + char *target; + unsigned int domid; + + /* Default to an guaranteed unused domain-id. */ + target_domain = DOMID_IDLE; + + target = xenbus_read(XBT_NIL, "target", "", NULL); + if (IS_ERR(target) || kstrtouint(target, 10, &domid)) { + pr_err("No target domain found, blocking all hypercalls\n"); + goto out; + } + + target_domain = domid; + + out: + if (!IS_ERR(target)) + kfree(target); + + restrict_wait = false; + wake_up_all(&restrict_wait_wq); + + return NOTIFY_DONE; +} + +static struct notifier_block xenstore_notifier = { + .notifier_call = init_restrict, +}; + +static void __init restrict_driver(void) +{ + restrict_wait = true; + + register_xenstore_notifier(&xenstore_notifier); +} + static int __init privcmd_init(void) { int err; @@ -1669,6 +1720,9 @@ static int __init privcmd_init(void) if (!xen_domain()) return -ENODEV; + if (!xen_initial_domain()) + restrict_driver(); + err = misc_register(&privcmd_dev); if (err != 0) { pr_err("Could not register Xen privcmd device\n"); From 2536617f20ddc7c2f4cef59b549aa45d166b03b1 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:49 +0200 Subject: [PATCH 241/352] dma-direct: prevent SWIOTLB path when DMA_ATTR_REQUIRE_COHERENT is set DMA_ATTR_REQUIRE_COHERENT indicates that SWIOTLB must not be used. Ensure the SWIOTLB path is declined whenever the DMA direct path is selected. Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-5-1dde90a7f08b@nvidia.com --- kernel/dma/direct.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/kernel/dma/direct.h b/kernel/dma/direct.h index e89f175e9c2d..6184ff303f08 100644 --- a/kernel/dma/direct.h +++ b/kernel/dma/direct.h @@ -84,7 +84,7 @@ static inline dma_addr_t dma_direct_map_phys(struct device *dev, dma_addr_t dma_addr; if (is_swiotlb_force_bounce(dev)) { - if (attrs & DMA_ATTR_MMIO) + if (attrs & (DMA_ATTR_MMIO | DMA_ATTR_REQUIRE_COHERENT)) return DMA_MAPPING_ERROR; return swiotlb_map(dev, phys, size, dir, attrs); @@ -98,7 +98,8 @@ static inline dma_addr_t dma_direct_map_phys(struct device *dev, dma_addr = phys_to_dma(dev, phys); if (unlikely(!dma_capable(dev, dma_addr, size, true)) || dma_kmalloc_needs_bounce(dev, size, dir)) { - if (is_swiotlb_active(dev)) + if (is_swiotlb_active(dev) && + !(attrs & DMA_ATTR_REQUIRE_COHERENT)) return swiotlb_map(dev, phys, size, dir, attrs); goto err_overflow; @@ -123,7 +124,7 @@ static inline void dma_direct_unmap_phys(struct device *dev, dma_addr_t addr, { phys_addr_t phys; - if (attrs & DMA_ATTR_MMIO) + if (attrs & (DMA_ATTR_MMIO | DMA_ATTR_REQUIRE_COHERENT)) /* nothing to do: uncached and no swiotlb */ return; From 636e6572e848339d2ae591949fe81de2cef00563 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:50 +0200 Subject: [PATCH 242/352] iommu/dma: add support for DMA_ATTR_REQUIRE_COHERENT attribute Add support for the DMA_ATTR_REQUIRE_COHERENT attribute to the exported functions. This attribute indicates that the SWIOTLB path must not be used and that no sync operations should be performed. Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-6-1dde90a7f08b@nvidia.com --- drivers/iommu/dma-iommu.c | 21 +++++++++++++++++---- 1 file changed, 17 insertions(+), 4 deletions(-) diff --git a/drivers/iommu/dma-iommu.c b/drivers/iommu/dma-iommu.c index 5dac64be61bb..94d514169642 100644 --- a/drivers/iommu/dma-iommu.c +++ b/drivers/iommu/dma-iommu.c @@ -1211,7 +1211,7 @@ dma_addr_t iommu_dma_map_phys(struct device *dev, phys_addr_t phys, size_t size, */ if (dev_use_swiotlb(dev, size, dir) && iova_unaligned(iovad, phys, size)) { - if (attrs & DMA_ATTR_MMIO) + if (attrs & (DMA_ATTR_MMIO | DMA_ATTR_REQUIRE_COHERENT)) return DMA_MAPPING_ERROR; phys = iommu_dma_map_swiotlb(dev, phys, size, dir, attrs); @@ -1223,7 +1223,8 @@ dma_addr_t iommu_dma_map_phys(struct device *dev, phys_addr_t phys, size_t size, arch_sync_dma_for_device(phys, size, dir); iova = __iommu_dma_map(dev, phys, size, prot, dma_mask); - if (iova == DMA_MAPPING_ERROR && !(attrs & DMA_ATTR_MMIO)) + if (iova == DMA_MAPPING_ERROR && + !(attrs & (DMA_ATTR_MMIO | DMA_ATTR_REQUIRE_COHERENT))) swiotlb_tbl_unmap_single(dev, phys, size, dir, attrs); return iova; } @@ -1233,7 +1234,7 @@ void iommu_dma_unmap_phys(struct device *dev, dma_addr_t dma_handle, { phys_addr_t phys; - if (attrs & DMA_ATTR_MMIO) { + if (attrs & (DMA_ATTR_MMIO | DMA_ATTR_REQUIRE_COHERENT)) { __iommu_dma_unmap(dev, dma_handle, size); return; } @@ -1945,9 +1946,21 @@ int dma_iova_link(struct device *dev, struct dma_iova_state *state, if (WARN_ON_ONCE(iova_start_pad && offset > 0)) return -EIO; + /* + * DMA_IOVA_USE_SWIOTLB is set on state after some entry + * took SWIOTLB path, which we were supposed to prevent + * for DMA_ATTR_REQUIRE_COHERENT attribute. + */ + if (WARN_ON_ONCE((state->__size & DMA_IOVA_USE_SWIOTLB) && + (attrs & DMA_ATTR_REQUIRE_COHERENT))) + return -EOPNOTSUPP; + + if (!dev_is_dma_coherent(dev) && (attrs & DMA_ATTR_REQUIRE_COHERENT)) + return -EOPNOTSUPP; + if (dev_use_swiotlb(dev, size, dir) && iova_unaligned(iovad, phys, size)) { - if (attrs & DMA_ATTR_MMIO) + if (attrs & (DMA_ATTR_MMIO | DMA_ATTR_REQUIRE_COHERENT)) return -EPERM; return iommu_dma_iova_link_swiotlb(dev, state, phys, offset, From d9d43a3f5c48d5a3d1da922f46c4a30d94d61ba5 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:51 +0200 Subject: [PATCH 243/352] RDMA/umem: Tell DMA mapping that UMEM requires coherency MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The RDMA subsystem exposes DMA regions through the verbs interface, which assumes a coherent system. Use the DMA_ATTR_REQUIRE_COHERENCE attribute to ensure coherency and avoid taking the SWIOTLB path. The RDMA verbs programming model resembles HMM and assumes concurrent DMA and CPU access to userspace memory. The hardware and programming model support "one-sided" operations initiated remotely without any local CPU involvement or notification. These include ATOMIC compare/swap, READ, and WRITE. A remote CPU can use these operations to traverse data structures, manipulate locks, and perform similar tasks without the host CPU’s awareness. If SWIOTLB substitutes memory or DMA is not cache coherent, these use cases break entirely. In-kernel RDMA is fine with incoherent mappings because kernel users do not rely on one-sided operations in ways that would expose these issues. A given region may also be exported multiple times, which can trigger warnings about cacheline overlaps. These warnings are suppressed when the new attribute is used. infiniband rocep8s0f0: mlx5_ib_reg_user_mr:1592:(pid 5812): start 0x2b28c000, iova 0x2b28c000, length 0x1000, access_flags 0x1 infiniband rocep8s0f0: mlx5_ib_reg_user_mr:1592:(pid 5812): start 0x2b28c001, iova 0x2b28c001, length 0xfff, access_flags 0x1 ------------[ cut here ]------------ DMA-API: mlx5_core 0000:08:00.0: cacheline tracking EEXIST, overlapping mappings aren't supported WARNING: kernel/dma/debug.c:620 at add_dma_entry+0x1bb/0x280, CPU#6: ibv_rc_pingpong/5812 Modules linked in: veth xt_conntrack xt_MASQUERADE nf_conntrack_netlink nfnetlink iptable_nat nf_nat xt_addrtype br_netfilter rpcsec_gss_krb5 auth_rpcgss oid_registry overlay mlx5_fwctl zram zsmalloc mlx5_ib fuse rpcrdma rdma_ucm ib_uverbs ib_iser libiscsi scsi_transport_iscsi ib_umad rdma_cm ib_ipoib iw_cm ib_cm mlx5_core ib_core CPU: 6 UID: 2733 PID: 5812 Comm: ibv_rc_pingpong Tainted: G W 6.19.0+ #129 PREEMPT Tainted: [W]=WARN Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014 RIP: 0010:add_dma_entry+0x1be/0x280 Code: 8b 7b 10 48 85 ff 0f 84 c3 00 00 00 48 8b 6f 50 48 85 ed 75 03 48 8b 2f e8 ff 8e 6a 00 48 89 c6 48 8d 3d 55 ef 2d 01 48 89 ea <67> 48 0f b9 3a 48 85 db 74 1a 48 c7 c7 b0 00 2b 82 e8 9c 25 fd ff RSP: 0018:ff11000138717978 EFLAGS: 00010286 RAX: ffffffffa02d7831 RBX: ff1100010246de00 RCX: 0000000000000000 RDX: ff110001036fac30 RSI: ffffffffa02d7831 RDI: ffffffff82678650 RBP: ff110001036fac30 R08: ff11000110dcb4a0 R09: ff11000110dcb478 R10: 0000000000000000 R11: ffffffff824b30a8 R12: 0000000000000000 R13: 00000000ffffffef R14: 0000000000000202 R15: ff1100010246de00 FS: 00007f59b411c740(0000) GS:ff110008dcc99000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007ffe538f7000 CR3: 000000010e066005 CR4: 0000000000373eb0 Call Trace: debug_dma_map_sg+0x1b4/0x390 __dma_map_sg_attrs+0x6d/0x1a0 dma_map_sgtable+0x19/0x30 ib_umem_get+0x254/0x380 [ib_uverbs] mlx5_ib_reg_user_mr+0x68/0x2a0 [mlx5_ib] ib_uverbs_reg_mr+0x17f/0x2a0 [ib_uverbs] ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0xc2/0x130 [ib_uverbs] ib_uverbs_cmd_verbs+0xa0b/0xae0 [ib_uverbs] ? ib_uverbs_handler_UVERBS_METHOD_QUERY_PORT_SPEED+0xe0/0xe0 [ib_uverbs] ? mmap_region+0x7a/0xb0 ? do_mmap+0x3b8/0x5c0 ib_uverbs_ioctl+0xa7/0x110 [ib_uverbs] __x64_sys_ioctl+0x14f/0x8b0 ? ksys_mmap_pgoff+0xc5/0x190 do_syscall_64+0x8c/0xbf0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 RIP: 0033:0x7f59b430aeed Code: 04 25 28 00 00 00 48 89 45 c8 31 c0 48 8d 45 10 c7 45 b0 10 00 00 00 48 89 45 b8 48 8d 45 d0 48 89 45 c0 b8 10 00 00 00 0f 05 <89> c2 3d 00 f0 ff ff 77 1a 48 8b 45 c8 64 48 2b 04 25 28 00 00 00 RSP: 002b:00007ffe538f9430 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007ffe538f94c0 RCX: 00007f59b430aeed RDX: 00007ffe538f94e0 RSI: 00000000c0181b01 RDI: 0000000000000003 RBP: 00007ffe538f9480 R08: 0000000000000028 R09: 00007ffe538f9684 R10: 0000000000000001 R11: 0000000000000246 R12: 00007ffe538f9684 R13: 000000000000000c R14: 000000002b28d170 R15: 000000000000000c ---[ end trace 0000000000000000 ]--- Reviewed-by: Jason Gunthorpe Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-7-1dde90a7f08b@nvidia.com --- drivers/infiniband/core/umem.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index cff4fcca2c34..edc34c69f0f2 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -55,7 +55,8 @@ static void __ib_umem_release(struct ib_device *dev, struct ib_umem *umem, int d if (dirty) ib_dma_unmap_sgtable_attrs(dev, &umem->sgt_append.sgt, - DMA_BIDIRECTIONAL, 0); + DMA_BIDIRECTIONAL, + DMA_ATTR_REQUIRE_COHERENT); for_each_sgtable_sg(&umem->sgt_append.sgt, sg, i) { unpin_user_page_range_dirty_lock(sg_page(sg), @@ -169,7 +170,7 @@ struct ib_umem *ib_umem_get(struct ib_device *device, unsigned long addr, unsigned long lock_limit; unsigned long new_pinned; unsigned long cur_base; - unsigned long dma_attr = 0; + unsigned long dma_attr = DMA_ATTR_REQUIRE_COHERENT; struct mm_struct *mm; unsigned long npages; int pinned, ret; From f5ebf241c407dbf629fcf515015e139fcea2c2f0 Mon Sep 17 00:00:00 2001 From: Leon Romanovsky Date: Mon, 16 Mar 2026 21:06:52 +0200 Subject: [PATCH 244/352] mm/hmm: Indicate that HMM requires DMA coherency MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit HMM is fundamentally about allowing a sophisticated device to perform DMA directly to a process’s memory while the CPU accesses that same memory at the same time. It is similar to SVA but does not rely on IOMMU support. Because the entire model depends on concurrent access to shared memory, it fails as a uAPI if SWIOTLB substitutes the memory or if the CPU caches are not coherent with DMA. Until now, there has been no reliable way to report this, and various approximations have been used: int hmm_dma_map_alloc(struct device *dev, struct hmm_dma_map *map, size_t nr_entries, size_t dma_entry_size) { <...> /* * The HMM API violates our normal DMA buffer ownership rules and can't * transfer buffer ownership. The dma_addressing_limited() check is a * best approximation to ensure no swiotlb buffering happens. */ dma_need_sync = !dev->dma_skip_sync; if (dma_need_sync || dma_addressing_limited(dev)) return -EOPNOTSUPP; So let's mark mapped buffers with DMA_ATTR_REQUIRE_COHERENT attribute to prevent silent data corruption if someone tries to use hmm in a system with swiotlb or incoherent DMA Reviewed-by: Jason Gunthorpe Signed-off-by: Leon Romanovsky Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260316-dma-debug-overlap-v3-8-1dde90a7f08b@nvidia.com --- mm/hmm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/hmm.c b/mm/hmm.c index f6c4ddff4bd6..5955f2f0c83d 100644 --- a/mm/hmm.c +++ b/mm/hmm.c @@ -778,7 +778,7 @@ dma_addr_t hmm_dma_map_pfn(struct device *dev, struct hmm_dma_map *map, struct page *page = hmm_pfn_to_page(pfns[idx]); phys_addr_t paddr = hmm_pfn_to_phys(pfns[idx]); size_t offset = idx * map->dma_entry_size; - unsigned long attrs = 0; + unsigned long attrs = DMA_ATTR_REQUIRE_COHERENT; dma_addr_t dma_addr; int ret; @@ -871,7 +871,7 @@ bool hmm_dma_unmap_pfn(struct device *dev, struct hmm_dma_map *map, size_t idx) struct dma_iova_state *state = &map->state; dma_addr_t *dma_addrs = map->dma_list; unsigned long *pfns = map->pfn_list; - unsigned long attrs = 0; + unsigned long attrs = DMA_ATTR_REQUIRE_COHERENT; if ((pfns[idx] & valid_dma) != valid_dma) return false; From 1613462be621ad5103ec338a7b0ca0746ec4e5f1 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Tue, 14 Oct 2025 13:28:15 +0200 Subject: [PATCH 245/352] xen/privcmd: add boot control for restricted usage in domU When running in an unprivileged domU under Xen, the privcmd driver is restricted to allow only hypercalls against a target domain, for which the current domU is acting as a device model. Add a boot parameter "unrestricted" to allow all hypercalls (the hypervisor will still refuse destructive hypercalls affecting other guests). Make this new parameter effective only in case the domU wasn't started using secure boot, as otherwise hypercalls targeting the domU itself might result in violating the secure boot functionality. This is achieved by adding another lockdown reason, which can be tested to not being set when applying the "unrestricted" option. This is part of XSA-482 Signed-off-by: Juergen Gross --- V2: - new patch --- drivers/xen/privcmd.c | 13 +++++++++++++ include/linux/security.h | 1 + security/security.c | 1 + 3 files changed, 15 insertions(+) diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c index a83bad69f4f2..bbf9ee21306c 100644 --- a/drivers/xen/privcmd.c +++ b/drivers/xen/privcmd.c @@ -32,6 +32,7 @@ #include #include #include +#include #include #include @@ -72,6 +73,11 @@ module_param_named(dm_op_buf_max_size, privcmd_dm_op_buf_max_size, uint, MODULE_PARM_DESC(dm_op_buf_max_size, "Maximum size of a dm_op hypercall buffer"); +static bool unrestricted; +module_param(unrestricted, bool, 0); +MODULE_PARM_DESC(unrestricted, + "Don't restrict hypercalls to target domain if running in a domU"); + struct privcmd_data { domid_t domid; }; @@ -1708,6 +1714,13 @@ static struct notifier_block xenstore_notifier = { static void __init restrict_driver(void) { + if (unrestricted) { + if (security_locked_down(LOCKDOWN_XEN_USER_ACTIONS)) + pr_warn("Kernel is locked down, parameter \"unrestricted\" ignored\n"); + else + return; + } + restrict_wait = true; register_xenstore_notifier(&xenstore_notifier); diff --git a/include/linux/security.h b/include/linux/security.h index 83a646d72f6f..ee88dd2d2d1f 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -145,6 +145,7 @@ enum lockdown_reason { LOCKDOWN_BPF_WRITE_USER, LOCKDOWN_DBG_WRITE_KERNEL, LOCKDOWN_RTAS_ERROR_INJECTION, + LOCKDOWN_XEN_USER_ACTIONS, LOCKDOWN_INTEGRITY_MAX, LOCKDOWN_KCORE, LOCKDOWN_KPROBES, diff --git a/security/security.c b/security/security.c index 67af9228c4e9..a26c1474e2e4 100644 --- a/security/security.c +++ b/security/security.c @@ -61,6 +61,7 @@ const char *const lockdown_reasons[LOCKDOWN_CONFIDENTIALITY_MAX + 1] = { [LOCKDOWN_BPF_WRITE_USER] = "use of bpf to write user RAM", [LOCKDOWN_DBG_WRITE_KERNEL] = "use of kgdb/kdb to write kernel RAM", [LOCKDOWN_RTAS_ERROR_INJECTION] = "RTAS error injection", + [LOCKDOWN_XEN_USER_ACTIONS] = "Xen guest user action", [LOCKDOWN_INTEGRITY_MAX] = "integrity", [LOCKDOWN_KCORE] = "/proc/kcore access", [LOCKDOWN_KPROBES] = "use of kprobes", From 63f8b60151dc00895137bfadc987f258060ab264 Mon Sep 17 00:00:00 2001 From: Vladimir Oltean Date: Fri, 20 Mar 2026 17:06:53 +0100 Subject: [PATCH 246/352] x86/entry/vdso: Fix path of included gettimeofday.c Commit in Fixes forgot to convert one include path to be relative to the kernel source directory after adding latter to flags-y. Fix it. [ bp: Rewrite commit message. ] Fixes: 693c819fedcd ("x86/entry/vdso: Refactor the vdso build") Signed-off-by: Vladimir Oltean Signed-off-by: Borislav Petkov (AMD) Link: https://lore.kernel.org/r/20260307174406.1808981-1-vladimir.oltean@nxp.com --- arch/x86/entry/vdso/common/vclock_gettime.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/entry/vdso/common/vclock_gettime.c b/arch/x86/entry/vdso/common/vclock_gettime.c index 027b7e88d753..57066f346b3f 100644 --- a/arch/x86/entry/vdso/common/vclock_gettime.c +++ b/arch/x86/entry/vdso/common/vclock_gettime.c @@ -13,7 +13,7 @@ #include #include -#include "../../../../lib/vdso/gettimeofday.c" +#include "lib/vdso/gettimeofday.c" int __vdso_gettimeofday(struct __kernel_old_timeval *tv, struct timezone *tz) { From 1c7bbaeed110b0fd9e65e173fb4d612f64a20d93 Mon Sep 17 00:00:00 2001 From: Kees Cook Date: Fri, 20 Mar 2026 10:51:18 -0700 Subject: [PATCH 247/352] coccinelle: kmalloc_obj: Remove default GFP_KERNEL arg Remove any GFP_KERNEL arguments found in the new kmalloc_obj-family helpers. This captures the script used in commit 189f164e573e ("Convert remaining multi-line kmalloc_obj/flex GFP_KERNEL uses"). Link: https://patch.msgid.link/20260320175113.work.016-kees@kernel.org Signed-off-by: Kees Cook --- scripts/coccinelle/api/kmalloc_objs.cocci | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/scripts/coccinelle/api/kmalloc_objs.cocci b/scripts/coccinelle/api/kmalloc_objs.cocci index db12b7be7247..e9a415b7b6f4 100644 --- a/scripts/coccinelle/api/kmalloc_objs.cocci +++ b/scripts/coccinelle/api/kmalloc_objs.cocci @@ -122,3 +122,14 @@ fresh identifier ALLOC_OBJS = script:python(ALLOC_ARRAY) { alloc_array(ALLOC_ARR - ALLOC(struct_size_t(TYPE, FLEX, COUNT), GFP) + ALLOC_FLEX(TYPE, FLEX, COUNT, GFP) ) + +@drop_gfp_kernel depends on patch && !(file in "tools") && !(file in "samples")@ +identifier ALLOC = {kmalloc_obj,kmalloc_objs,kmalloc_flex, + kzalloc_obj,kzalloc_objs,kzalloc_flex, + kvmalloc_obj,kvmalloc_objs,kvmalloc_flex, + kvzalloc_obj,kvzalloc_objs,kvzalloc_flex}; +@@ + + ALLOC(... +- , GFP_KERNEL + ) From 1f6aa5bbf1d0f81a8a2aafc16136e7dd9a609ff3 Mon Sep 17 00:00:00 2001 From: Kyle Meyer Date: Fri, 20 Mar 2026 12:19:20 -0500 Subject: [PATCH 248/352] x86/platform/uv: Handle deconfigured sockets When a socket is deconfigured, it's mapped to SOCK_EMPTY (0xffff). This causes a panic while allocating UV hub info structures. Fix this by using NUMA_NO_NODE, allowing UV hub info structures to be allocated on valid nodes. Fixes: 8a50c5851927 ("x86/platform/uv: UV support for sub-NUMA clustering") Signed-off-by: Kyle Meyer Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Steve Wahl Cc: stable@vger.kernel.org Link: https://patch.msgid.link/ab2BmGL0ehVkkjKk@hpe.com --- arch/x86/kernel/apic/x2apic_uv_x.c | 18 ++++++++++++++++-- 1 file changed, 16 insertions(+), 2 deletions(-) diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 15209f220e1f..42568ceec481 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -1708,8 +1708,22 @@ static void __init uv_system_init_hub(void) struct uv_hub_info_s *new_hub; /* Allocate & fill new per hub info list */ - new_hub = (bid == 0) ? &uv_hub_info_node0 - : kzalloc_node(bytes, GFP_KERNEL, uv_blade_to_node(bid)); + if (bid == 0) { + new_hub = &uv_hub_info_node0; + } else { + int nid; + + /* + * Deconfigured sockets are mapped to SOCK_EMPTY. Use + * NUMA_NO_NODE to allocate on a valid node. + */ + nid = uv_blade_to_node(bid); + if (nid == SOCK_EMPTY) + nid = NUMA_NO_NODE; + + new_hub = kzalloc_node(bytes, GFP_KERNEL, nid); + } + if (WARN_ON_ONCE(!new_hub)) { /* do not kfree() bid 0, which is statically allocated */ while (--bid > 0) From 7a618ca9b9c4769fc5adf7344bb1dd98f823da22 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Wed, 18 Mar 2026 17:20:18 -0700 Subject: [PATCH 249/352] init/Kconfig: Require a release version of clang-22 for CC_HAS_COUNTED_BY_PTR Commit 150a04d817d8 ("compiler_types.h: Attributes: Add __counted_by_ptr macro") used Clang 22.0.0 as a minimum supported version for __counted_by_ptr, which made sense while 22.0.0 was the version of LLVM's main branch to allow developers to easily test and develop uses of __counted_by_ptr in their code. However, __counted_by_ptr requires a change [1] merged towards the end of the 22 development cycle to avoid errors when applied to void pointers. In file included from fs/xfs/xfs_attr_inactive.c:18: fs/xfs/libxfs/xfs_attr.h:59:2: error: 'counted_by' cannot be applied to a pointer with pointee of unknown size because 'void' is an incomplete type 59 | void *buffer __counted_by_ptr(bufsize); | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ This is disruptive for deployed prerelease clang-22 builds (such as Android LLVM) or when bisecting between llvmorg-21-init and the fix. Require a released version of clang-22 (i.e., 21.1.0 or newer) to enabled __counted_by_ptr to ensure all fixes needed for proper support are present. Fixes: 150a04d817d8 ("compiler_types.h: Attributes: Add __counted_by_ptr macro") Link: https://github.com/llvm/llvm-project/commit/f29955a594aedf5943d492a999b83e8c6b8fafae [1] Signed-off-by: Nathan Chancellor Link: https://patch.msgid.link/20260318-counted_by_ptr-release-clang-22-v1-1-e017da246df0@kernel.org Signed-off-by: Kees Cook --- init/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/init/Kconfig b/init/Kconfig index b55deae9256c..399c0ed3b4a9 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -146,7 +146,7 @@ config CC_HAS_COUNTED_BY config CC_HAS_COUNTED_BY_PTR bool # supported since clang 22 - default y if CC_IS_CLANG && CLANG_VERSION >= 220000 + default y if CC_IS_CLANG && CLANG_VERSION >= 220100 # supported since gcc 16.0.0 default y if CC_IS_GCC && GCC_VERSION >= 160000 From e7bae9a7a5e1251ab414291f4e9304d702bb9221 Mon Sep 17 00:00:00 2001 From: Guenter Roeck Date: Thu, 19 Mar 2026 21:31:22 -0700 Subject: [PATCH 250/352] hwmon: (max6639) Fix pulses-per-revolution implementation The valid range for the pulses-per-revolution devicetree property is 1..4. The current code checks for a range of 1..5. Fix it. Declare the variable used to retrieve pulses per revolution from devicetree as u32 (unsigned) to match the of_property_read_u32() API. The current code uses a postfix decrement when writing the pulses per resolution into the chip. This has no effect since the value is evaluated before it is decremented. Fix it by decrementing before evaluating the value. Fixes: 7506ebcd662b ("hwmon: (max6639) : Configure based on DT property") Cc: Naresh Solanki Signed-off-by: Guenter Roeck --- drivers/hwmon/max6639.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/hwmon/max6639.c b/drivers/hwmon/max6639.c index 9a3c515efe2e..163d31f17bd4 100644 --- a/drivers/hwmon/max6639.c +++ b/drivers/hwmon/max6639.c @@ -232,7 +232,7 @@ static int max6639_read_fan(struct device *dev, u32 attr, int channel, static int max6639_set_ppr(struct max6639_data *data, int channel, u8 ppr) { /* Decrement the PPR value and shift left by 6 to match the register format */ - return regmap_write(data->regmap, MAX6639_REG_FAN_PPR(channel), ppr-- << 6); + return regmap_write(data->regmap, MAX6639_REG_FAN_PPR(channel), --ppr << 6); } static int max6639_write_fan(struct device *dev, u32 attr, int channel, @@ -524,8 +524,8 @@ static int max6639_probe_child_from_dt(struct i2c_client *client, { struct device *dev = &client->dev; - u32 i; - int err, val; + u32 i, val; + int err; err = of_property_read_u32(child, "reg", &i); if (err) { @@ -540,8 +540,8 @@ static int max6639_probe_child_from_dt(struct i2c_client *client, err = of_property_read_u32(child, "pulses-per-revolution", &val); if (!err) { - if (val < 1 || val > 5) { - dev_err(dev, "invalid pulses-per-revolution %d of %pOFn\n", val, child); + if (val < 1 || val > 4) { + dev_err(dev, "invalid pulses-per-revolution %u of %pOFn\n", val, child); return -EINVAL; } data->ppr[i] = val; From 317e49358ebbf6390fa439ef3c142f9239dd25fb Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Wed, 18 Mar 2026 14:36:58 +0800 Subject: [PATCH 251/352] net: macb: Move devm_{free,request}_irq() out of spin lock area MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The devm_free_irq() and devm_request_irq() functions should not be executed in an atomic context. During device suspend, all userspace processes and most kernel threads are frozen. Additionally, we flush all tx/rx status, disable all macb interrupts, and halt rx operations. Therefore, it is safe to split the region protected by bp->lock into two independent sections, allowing devm_free_irq() and devm_request_irq() to run in a non-atomic context. This modification resolves the following lockdep warning: BUG: sleeping function called from invalid context at kernel/locking/mutex.c:591 in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 501, name: rtcwake preempt_count: 1, expected: 0 RCU nest depth: 1, expected: 0 7 locks held by rtcwake/501: #0: ffff0008038c3408 (sb_writers#5){.+.+}-{0:0}, at: vfs_write+0xf8/0x368 #1: ffff0008049a5e88 (&of->mutex#2){+.+.}-{4:4}, at: kernfs_fop_write_iter+0xbc/0x1c8 #2: ffff00080098d588 (kn->active#70){.+.+}-{0:0}, at: kernfs_fop_write_iter+0xcc/0x1c8 #3: ffff800081c84888 (system_transition_mutex){+.+.}-{4:4}, at: pm_suspend+0x1ec/0x290 #4: ffff0008009ba0f8 (&dev->mutex){....}-{4:4}, at: device_suspend+0x118/0x4f0 #5: ffff800081d00458 (rcu_read_lock){....}-{1:3}, at: rcu_lock_acquire+0x4/0x48 #6: ffff0008031fb9e0 (&bp->lock){-.-.}-{3:3}, at: macb_suspend+0x144/0x558 irq event stamp: 8682 hardirqs last enabled at (8681): [] _raw_spin_unlock_irqrestore+0x44/0x88 hardirqs last disabled at (8682): [] _raw_spin_lock_irqsave+0x38/0x98 softirqs last enabled at (7322): [] handle_softirqs+0x52c/0x588 softirqs last disabled at (7317): [] __do_softirq+0x20/0x2c CPU: 1 UID: 0 PID: 501 Comm: rtcwake Not tainted 7.0.0-rc3-next-20260310-yocto-standard+ #125 PREEMPT Hardware name: ZynqMP ZCU102 Rev1.1 (DT) Call trace: show_stack+0x24/0x38 (C) __dump_stack+0x28/0x38 dump_stack_lvl+0x64/0x88 dump_stack+0x18/0x24 __might_resched+0x200/0x218 __might_sleep+0x38/0x98 __mutex_lock_common+0x7c/0x1378 mutex_lock_nested+0x38/0x50 free_irq+0x68/0x2b0 devm_irq_release+0x24/0x38 devres_release+0x40/0x80 devm_free_irq+0x48/0x88 macb_suspend+0x298/0x558 device_suspend+0x218/0x4f0 dpm_suspend+0x244/0x3a0 dpm_suspend_start+0x50/0x78 suspend_devices_and_enter+0xec/0x560 pm_suspend+0x194/0x290 state_store+0x110/0x158 kobj_attr_store+0x1c/0x30 sysfs_kf_write+0xa8/0xd0 kernfs_fop_write_iter+0x11c/0x1c8 vfs_write+0x248/0x368 ksys_write+0x7c/0xf8 __arm64_sys_write+0x28/0x40 invoke_syscall+0x4c/0xe8 el0_svc_common+0x98/0xf0 do_el0_svc+0x28/0x40 el0_svc+0x54/0x1e0 el0t_64_sync_handler+0x84/0x130 el0t_64_sync+0x198/0x1a0 Fixes: 558e35ccfe95 ("net: macb: WoL support for GEM type of Ethernet controller") Cc: stable@vger.kernel.org Reviewed-by: Théo Lebrun Signed-off-by: Kevin Hao Link: https://patch.msgid.link/20260318-macb-irq-v2-1-f1179768ab24@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 12 +++++++----- 1 file changed, 7 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index c16ac9c76aa3..1b95226ae696 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -5835,6 +5835,7 @@ static int __maybe_unused macb_suspend(struct device *dev) /* write IP address into register */ tmp |= MACB_BFEXT(IP, be32_to_cpu(ifa->ifa_local)); } + spin_unlock_irqrestore(&bp->lock, flags); /* Change interrupt handler and * Enable WoL IRQ on queue 0 @@ -5847,11 +5848,12 @@ static int __maybe_unused macb_suspend(struct device *dev) dev_err(dev, "Unable to request IRQ %d (error %d)\n", bp->queues[0].irq, err); - spin_unlock_irqrestore(&bp->lock, flags); return err; } + spin_lock_irqsave(&bp->lock, flags); queue_writel(bp->queues, IER, GEM_BIT(WOL)); gem_writel(bp, WOL, tmp); + spin_unlock_irqrestore(&bp->lock, flags); } else { err = devm_request_irq(dev, bp->queues[0].irq, macb_wol_interrupt, IRQF_SHARED, netdev->name, bp->queues); @@ -5859,13 +5861,13 @@ static int __maybe_unused macb_suspend(struct device *dev) dev_err(dev, "Unable to request IRQ %d (error %d)\n", bp->queues[0].irq, err); - spin_unlock_irqrestore(&bp->lock, flags); return err; } + spin_lock_irqsave(&bp->lock, flags); queue_writel(bp->queues, IER, MACB_BIT(WOL)); macb_writel(bp, WOL, tmp); + spin_unlock_irqrestore(&bp->lock, flags); } - spin_unlock_irqrestore(&bp->lock, flags); enable_irq_wake(bp->queues[0].irq); } @@ -5932,6 +5934,8 @@ static int __maybe_unused macb_resume(struct device *dev) queue_readl(bp->queues, ISR); if (bp->caps & MACB_CAPS_ISR_CLEAR_ON_WRITE) queue_writel(bp->queues, ISR, -1); + spin_unlock_irqrestore(&bp->lock, flags); + /* Replace interrupt handler on queue 0 */ devm_free_irq(dev, bp->queues[0].irq, bp->queues); err = devm_request_irq(dev, bp->queues[0].irq, macb_interrupt, @@ -5940,10 +5944,8 @@ static int __maybe_unused macb_resume(struct device *dev) dev_err(dev, "Unable to request IRQ %d (error %d)\n", bp->queues[0].irq, err); - spin_unlock_irqrestore(&bp->lock, flags); return err; } - spin_unlock_irqrestore(&bp->lock, flags); disable_irq_wake(bp->queues[0].irq); From baa35a698cea26930679a20a7550bbb4c8319725 Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Wed, 18 Mar 2026 14:36:59 +0800 Subject: [PATCH 252/352] net: macb: Protect access to net_device::ip_ptr with RCU lock MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Access to net_device::ip_ptr and its associated members must be protected by an RCU lock. Since we are modifying this piece of code, let's also move it to execute only when WAKE_ARP is enabled. To minimize the duration of the RCU lock, a local variable is used to temporarily store the IP address. This change resolves the following RCU check warning: WARNING: suspicious RCU usage 7.0.0-rc3-next-20260310-yocto-standard+ #122 Not tainted ----------------------------- drivers/net/ethernet/cadence/macb_main.c:5944 suspicious rcu_dereference_check() usage! other info that might help us debug this: rcu_scheduler_active = 2, debug_locks = 1 5 locks held by rtcwake/518: #0: ffff000803ab1408 (sb_writers#5){.+.+}-{0:0}, at: vfs_write+0xf8/0x368 #1: ffff0008090bf088 (&of->mutex#2){+.+.}-{4:4}, at: kernfs_fop_write_iter+0xbc/0x1c8 #2: ffff00080098d588 (kn->active#70){.+.+}-{0:0}, at: kernfs_fop_write_iter+0xcc/0x1c8 #3: ffff800081c84888 (system_transition_mutex){+.+.}-{4:4}, at: pm_suspend+0x1ec/0x290 #4: ffff0008009ba0f8 (&dev->mutex){....}-{4:4}, at: device_suspend+0x118/0x4f0 stack backtrace: CPU: 3 UID: 0 PID: 518 Comm: rtcwake Not tainted 7.0.0-rc3-next-20260310-yocto-standard+ #122 PREEMPT Hardware name: ZynqMP ZCU102 Rev1.1 (DT) Call trace: show_stack+0x24/0x38 (C) __dump_stack+0x28/0x38 dump_stack_lvl+0x64/0x88 dump_stack+0x18/0x24 lockdep_rcu_suspicious+0x134/0x1d8 macb_suspend+0xd8/0x4c0 device_suspend+0x218/0x4f0 dpm_suspend+0x244/0x3a0 dpm_suspend_start+0x50/0x78 suspend_devices_and_enter+0xec/0x560 pm_suspend+0x194/0x290 state_store+0x110/0x158 kobj_attr_store+0x1c/0x30 sysfs_kf_write+0xa8/0xd0 kernfs_fop_write_iter+0x11c/0x1c8 vfs_write+0x248/0x368 ksys_write+0x7c/0xf8 __arm64_sys_write+0x28/0x40 invoke_syscall+0x4c/0xe8 el0_svc_common+0x98/0xf0 do_el0_svc+0x28/0x40 el0_svc+0x54/0x1e0 el0t_64_sync_handler+0x84/0x130 el0t_64_sync+0x198/0x1a0 Fixes: 0cb8de39a776 ("net: macb: Add ARP support to WOL") Signed-off-by: Kevin Hao Cc: stable@vger.kernel.org Reviewed-by: Théo Lebrun Link: https://patch.msgid.link/20260318-macb-irq-v2-2-f1179768ab24@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/cadence/macb_main.c | 25 +++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 1b95226ae696..944aab71795c 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -5776,9 +5776,9 @@ static int __maybe_unused macb_suspend(struct device *dev) struct macb_queue *queue; struct in_device *idev; unsigned long flags; + u32 tmp, ifa_local; unsigned int q; int err; - u32 tmp; if (!device_may_wakeup(&bp->dev->dev)) phy_exit(bp->phy); @@ -5787,14 +5787,21 @@ static int __maybe_unused macb_suspend(struct device *dev) return 0; if (bp->wol & MACB_WOL_ENABLED) { - /* Check for IP address in WOL ARP mode */ - idev = __in_dev_get_rcu(bp->dev); - if (idev) - ifa = rcu_dereference(idev->ifa_list); - if ((bp->wolopts & WAKE_ARP) && !ifa) { - netdev_err(netdev, "IP address not assigned as required by WoL walk ARP\n"); - return -EOPNOTSUPP; + if (bp->wolopts & WAKE_ARP) { + /* Check for IP address in WOL ARP mode */ + rcu_read_lock(); + idev = __in_dev_get_rcu(bp->dev); + if (idev) + ifa = rcu_dereference(idev->ifa_list); + if (!ifa) { + rcu_read_unlock(); + netdev_err(netdev, "IP address not assigned as required by WoL walk ARP\n"); + return -EOPNOTSUPP; + } + ifa_local = be32_to_cpu(ifa->ifa_local); + rcu_read_unlock(); } + spin_lock_irqsave(&bp->lock, flags); /* Disable Tx and Rx engines before disabling the queues, @@ -5833,7 +5840,7 @@ static int __maybe_unused macb_suspend(struct device *dev) if (bp->wolopts & WAKE_ARP) { tmp |= MACB_BIT(ARP); /* write IP address into register */ - tmp |= MACB_BFEXT(IP, be32_to_cpu(ifa->ifa_local)); + tmp |= MACB_BFEXT(IP, ifa_local); } spin_unlock_irqrestore(&bp->lock, flags); From 6931d21f87bc6d657f145798fad0bf077b82486c Mon Sep 17 00:00:00 2001 From: Yang Yang Date: Thu, 19 Mar 2026 07:42:41 +0000 Subject: [PATCH 253/352] openvswitch: defer tunnel netdev_put to RCU release ovs_netdev_tunnel_destroy() may run after NETDEV_UNREGISTER already detached the device. Dropping the netdev reference in destroy can race with concurrent readers that still observe vport->dev. Do not release vport->dev in ovs_netdev_tunnel_destroy(). Instead, let vport_netdev_free() drop the reference from the RCU callback, matching the non-tunnel destroy path and avoiding additional synchronization under RTNL. Fixes: a9020fde67a6 ("openvswitch: Move tunnel destroy function to oppenvswitch module.") Reported-by: Yifan Wu Reported-by: Juefei Pu Tested-by: Ao Zhou Co-developed-by: Yuan Tan Signed-off-by: Yuan Tan Suggested-by: Xin Liu Signed-off-by: Yang Yang Reviewed-by: Ilya Maximets Link: https://patch.msgid.link/20260319074241.3405262-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski --- net/openvswitch/vport-netdev.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/net/openvswitch/vport-netdev.c b/net/openvswitch/vport-netdev.c index c688dee96503..12055af832dc 100644 --- a/net/openvswitch/vport-netdev.c +++ b/net/openvswitch/vport-netdev.c @@ -196,8 +196,6 @@ void ovs_netdev_tunnel_destroy(struct vport *vport) */ if (vport->dev->reg_state == NETREG_REGISTERED) rtnl_delete_link(vport->dev, 0, NULL); - netdev_put(vport->dev, &vport->dev_tracker); - vport->dev = NULL; rtnl_unlock(); call_rcu(&vport->rcu, vport_netdev_free); From 546b68ac893595877ffbd7751e5c55fd1c43ede6 Mon Sep 17 00:00:00 2001 From: Yang Yang Date: Thu, 19 Mar 2026 08:02:27 +0000 Subject: [PATCH 254/352] openvswitch: validate MPLS set/set_masked payload length validate_set() accepted OVS_KEY_ATTR_MPLS as variable-sized payload for SET/SET_MASKED actions. In action handling, OVS expects fixed-size MPLS key data (struct ovs_key_mpls). Use the already normalized key_len (masked case included) and reject non-matching MPLS action key sizes. Reject invalid MPLS action payload lengths early. Fixes: fbdcdd78da7c ("Change in Openvswitch to support MPLS label depth of 3 in ingress direction") Reported-by: Yifan Wu Reported-by: Juefei Pu Tested-by: Ao Zhou Co-developed-by: Yuan Tan Signed-off-by: Yuan Tan Suggested-by: Xin Liu Signed-off-by: Yang Yang Reviewed-by: Ilya Maximets Link: https://patch.msgid.link/20260319080228.3423307-1-n05ec@lzu.edu.cn Signed-off-by: Jakub Kicinski --- net/openvswitch/flow_netlink.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/openvswitch/flow_netlink.c b/net/openvswitch/flow_netlink.c index 67fbf6e48a30..13052408a132 100644 --- a/net/openvswitch/flow_netlink.c +++ b/net/openvswitch/flow_netlink.c @@ -2953,6 +2953,8 @@ static int validate_set(const struct nlattr *a, case OVS_KEY_ATTR_MPLS: if (!eth_p_mpls(eth_type)) return -EINVAL; + if (key_len != sizeof(struct ovs_key_mpls)) + return -EINVAL; break; case OVS_KEY_ATTR_SCTP: From 24dd586bb4cbba1889a50abe74143817a095c1c9 Mon Sep 17 00:00:00 2001 From: Qi Tang Date: Wed, 18 Mar 2026 14:48:47 +0800 Subject: [PATCH 255/352] net/smc: fix double-free of smc_spd_priv when tee() duplicates splice pipe buffer smc_rx_splice() allocates one smc_spd_priv per pipe_buffer and stores the pointer in pipe_buffer.private. The pipe_buf_operations for these buffers used .get = generic_pipe_buf_get, which only increments the page reference count when tee(2) duplicates a pipe buffer. The smc_spd_priv pointer itself was not handled, so after tee() both the original and the cloned pipe_buffer share the same smc_spd_priv *. When both pipes are subsequently released, smc_rx_pipe_buf_release() is called twice against the same object: 1st call: kfree(priv) sock_put(sk) smc_rx_update_cons() [correct] 2nd call: kfree(priv) sock_put(sk) smc_rx_update_cons() [UAF] KASAN reports a slab-use-after-free in smc_rx_pipe_buf_release(), which then escalates to a NULL-pointer dereference and kernel panic via smc_rx_update_consumer() when it chases the freed priv->smc pointer: BUG: KASAN: slab-use-after-free in smc_rx_pipe_buf_release+0x78/0x2a0 Read of size 8 at addr ffff888004a45740 by task smc_splice_tee_/74 Call Trace: dump_stack_lvl+0x53/0x70 print_report+0xce/0x650 kasan_report+0xc6/0x100 smc_rx_pipe_buf_release+0x78/0x2a0 free_pipe_info+0xd4/0x130 pipe_release+0x142/0x160 __fput+0x1c6/0x490 __x64_sys_close+0x4f/0x90 do_syscall_64+0xa6/0x1a0 entry_SYSCALL_64_after_hwframe+0x77/0x7f BUG: kernel NULL pointer dereference, address: 0000000000000020 RIP: 0010:smc_rx_update_consumer+0x8d/0x350 Call Trace: smc_rx_pipe_buf_release+0x121/0x2a0 free_pipe_info+0xd4/0x130 pipe_release+0x142/0x160 __fput+0x1c6/0x490 __x64_sys_close+0x4f/0x90 do_syscall_64+0xa6/0x1a0 entry_SYSCALL_64_after_hwframe+0x77/0x7f Kernel panic - not syncing: Fatal exception Beyond the memory-safety problem, duplicating an SMC splice buffer is semantically questionable: smc_rx_update_cons() would advance the consumer cursor twice for the same data, corrupting receive-window accounting. A refcount on smc_spd_priv could fix the double-free, but the cursor-accounting issue would still need to be addressed separately. The .get callback is invoked by both tee(2) and splice_pipe_to_pipe() for partial transfers; both will now return -EFAULT. Users who need to duplicate SMC socket data must use a copy-based read path. Fixes: 9014db202cb7 ("smc: add support for splice()") Signed-off-by: Qi Tang Link: https://patch.msgid.link/20260318064847.23341-1-tpluszz77@gmail.com Signed-off-by: Jakub Kicinski --- net/smc/smc_rx.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/smc/smc_rx.c b/net/smc/smc_rx.c index d833e36f7fd4..c1d9b923938d 100644 --- a/net/smc/smc_rx.c +++ b/net/smc/smc_rx.c @@ -135,9 +135,16 @@ static void smc_rx_pipe_buf_release(struct pipe_inode_info *pipe, sock_put(sk); } +static bool smc_rx_pipe_buf_get(struct pipe_inode_info *pipe, + struct pipe_buffer *buf) +{ + /* smc_spd_priv in buf->private is not shareable; disallow cloning. */ + return false; +} + static const struct pipe_buf_operations smc_pipe_ops = { .release = smc_rx_pipe_buf_release, - .get = generic_pipe_buf_get + .get = smc_rx_pipe_buf_get, }; static void smc_rx_spd_release(struct splice_pipe_desc *spd, From 52501989c76206462d9b11a8485beef40ef41821 Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 20 Mar 2026 00:02:52 +0100 Subject: [PATCH 256/352] rtnetlink: count IFLA_PARENT_DEV_{NAME,BUS_NAME} in if_nlmsg_size Commit 00e77ed8e64d ("rtnetlink: add IFLA_PARENT_[DEV|DEV_BUS]_NAME") added those attributes to rtnl_fill_ifinfo, but forgot to extend if_nlmsg_size. Fixes: 00e77ed8e64d ("rtnetlink: add IFLA_PARENT_[DEV|DEV_BUS]_NAME") Signed-off-by: Sabrina Dubroca Link: https://patch.msgid.link/0b849da95562af45487080528d60f578636aba5c.1773919462.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index dad4b1054955..0fe279432d82 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -1267,6 +1267,21 @@ static size_t rtnl_dpll_pin_size(const struct net_device *dev) return size; } +static size_t rtnl_dev_parent_size(const struct net_device *dev) +{ + size_t size = 0; + + /* IFLA_PARENT_DEV_NAME */ + if (dev->dev.parent) + size += nla_total_size(strlen(dev_name(dev->dev.parent)) + 1); + + /* IFLA_PARENT_DEV_BUS_NAME */ + if (dev->dev.parent && dev->dev.parent->bus) + size += nla_total_size(strlen(dev->dev.parent->bus->name) + 1); + + return size; +} + static noinline size_t if_nlmsg_size(const struct net_device *dev, u32 ext_filter_mask) { @@ -1328,6 +1343,7 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev, + nla_total_size(8) /* IFLA_MAX_PACING_OFFLOAD_HORIZON */ + nla_total_size(2) /* IFLA_HEADROOM */ + nla_total_size(2) /* IFLA_TAILROOM */ + + rtnl_dev_parent_size(dev) + 0; if (!(ext_filter_mask & RTEXT_FILTER_SKIP_STATS)) From ee00a12593ffb69db4dd1a1c00ecb0253376874a Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Fri, 20 Mar 2026 00:02:53 +0100 Subject: [PATCH 257/352] rtnetlink: count IFLA_INFO_SLAVE_KIND in if_nlmsg_size rtnl_link_get_slave_info_data_size counts IFLA_INFO_SLAVE_DATA, but rtnl_link_slave_info_fill adds both IFLA_INFO_SLAVE_DATA and IFLA_INFO_SLAVE_KIND. Fixes: ba7d49b1f0f8 ("rtnetlink: provide api for getting and setting slave info") Reviewed-by: Jiri Pirko Signed-off-by: Sabrina Dubroca Link: https://patch.msgid.link/049843b532e23cde7ddba263c0bbe35ba6f0d26d.1773919462.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 0fe279432d82..4a2278614250 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -707,11 +707,14 @@ static size_t rtnl_link_get_slave_info_data_size(const struct net_device *dev) goto out; ops = master_dev->rtnl_link_ops; - if (!ops || !ops->get_slave_size) + if (!ops) + goto out; + size += nla_total_size(strlen(ops->kind) + 1); /* IFLA_INFO_SLAVE_KIND */ + if (!ops->get_slave_size) goto out; /* IFLA_INFO_SLAVE_DATA + nested data */ - size = nla_total_size(sizeof(struct nlattr)) + - ops->get_slave_size(master_dev, dev); + size += nla_total_size(sizeof(struct nlattr)) + + ops->get_slave_size(master_dev, dev); out: rcu_read_unlock(); From cbfa5be2bf64511d49b854a0f9fd6d0b5118621f Mon Sep 17 00:00:00 2001 From: Justin Chen Date: Thu, 19 Mar 2026 16:48:12 -0700 Subject: [PATCH 258/352] net: bcmasp: fix double free of WoL irq We do not need to free wol_irq since it was instantiated with devm_request_irq(). So devres will free for us. Fixes: a2f0751206b0 ("net: bcmasp: Add support for WoL magic packet") Signed-off-by: Justin Chen Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20260319234813.1937315-2-justin.chen@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/asp2/bcmasp.c | 8 -------- 1 file changed, 8 deletions(-) diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp.c b/drivers/net/ethernet/broadcom/asp2/bcmasp.c index aa6d8606849f..2034a1593db7 100644 --- a/drivers/net/ethernet/broadcom/asp2/bcmasp.c +++ b/drivers/net/ethernet/broadcom/asp2/bcmasp.c @@ -1152,12 +1152,6 @@ void bcmasp_enable_wol(struct bcmasp_intf *intf, bool en) } } -static void bcmasp_wol_irq_destroy(struct bcmasp_priv *priv) -{ - if (priv->wol_irq > 0) - free_irq(priv->wol_irq, priv); -} - static void bcmasp_eee_fixup(struct bcmasp_intf *intf, bool en) { u32 reg, phy_lpi_overwrite; @@ -1363,7 +1357,6 @@ static int bcmasp_probe(struct platform_device *pdev) return ret; err_cleanup: - bcmasp_wol_irq_destroy(priv); bcmasp_remove_intfs(priv); return ret; @@ -1376,7 +1369,6 @@ static void bcmasp_remove(struct platform_device *pdev) if (!priv) return; - bcmasp_wol_irq_destroy(priv); bcmasp_remove_intfs(priv); } From 27dfe9030acbc601c260b42ecdbb4e5858a97b53 Mon Sep 17 00:00:00 2001 From: Justin Chen Date: Thu, 19 Mar 2026 16:48:13 -0700 Subject: [PATCH 259/352] net: bcmasp: fix double disable of clk Switch to devm_clk_get_optional() so we can manage the clock ourselves. We dynamically control the clocks depending on the state of the interface for power savings. The default state is clock disabled, so unbinding the driver causes a double disable. Fixes: 490cb412007d ("net: bcmasp: Add support for ASP2.0 Ethernet controller") Signed-off-by: Justin Chen Reviewed-by: Florian Fainelli Link: https://patch.msgid.link/20260319234813.1937315-3-justin.chen@broadcom.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/asp2/bcmasp.c | 33 ++++++++++++++------- 1 file changed, 23 insertions(+), 10 deletions(-) diff --git a/drivers/net/ethernet/broadcom/asp2/bcmasp.c b/drivers/net/ethernet/broadcom/asp2/bcmasp.c index 2034a1593db7..972474893a6b 100644 --- a/drivers/net/ethernet/broadcom/asp2/bcmasp.c +++ b/drivers/net/ethernet/broadcom/asp2/bcmasp.c @@ -1249,7 +1249,7 @@ static int bcmasp_probe(struct platform_device *pdev) if (priv->irq <= 0) return -EINVAL; - priv->clk = devm_clk_get_optional_enabled(dev, "sw_asp"); + priv->clk = devm_clk_get_optional(dev, "sw_asp"); if (IS_ERR(priv->clk)) return dev_err_probe(dev, PTR_ERR(priv->clk), "failed to request clock\n"); @@ -1277,6 +1277,10 @@ static int bcmasp_probe(struct platform_device *pdev) bcmasp_set_pdata(priv, pdata); + ret = clk_prepare_enable(priv->clk); + if (ret) + return dev_err_probe(dev, ret, "failed to start clock\n"); + /* Enable all clocks to ensure successful probing */ bcmasp_core_clock_set(priv, ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE, 0); @@ -1288,8 +1292,10 @@ static int bcmasp_probe(struct platform_device *pdev) ret = devm_request_irq(&pdev->dev, priv->irq, bcmasp_isr, 0, pdev->name, priv); - if (ret) - return dev_err_probe(dev, ret, "failed to request ASP interrupt: %d", ret); + if (ret) { + dev_err(dev, "Failed to request ASP interrupt: %d", ret); + goto err_clock_disable; + } /* Register mdio child nodes */ of_platform_populate(dev->of_node, bcmasp_mdio_of_match, NULL, dev); @@ -1301,13 +1307,17 @@ static int bcmasp_probe(struct platform_device *pdev) priv->mda_filters = devm_kcalloc(dev, priv->num_mda_filters, sizeof(*priv->mda_filters), GFP_KERNEL); - if (!priv->mda_filters) - return -ENOMEM; + if (!priv->mda_filters) { + ret = -ENOMEM; + goto err_clock_disable; + } priv->net_filters = devm_kcalloc(dev, priv->num_net_filters, sizeof(*priv->net_filters), GFP_KERNEL); - if (!priv->net_filters) - return -ENOMEM; + if (!priv->net_filters) { + ret = -ENOMEM; + goto err_clock_disable; + } bcmasp_core_init_filters(priv); @@ -1316,7 +1326,8 @@ static int bcmasp_probe(struct platform_device *pdev) ports_node = of_find_node_by_name(dev->of_node, "ethernet-ports"); if (!ports_node) { dev_warn(dev, "No ports found\n"); - return -EINVAL; + ret = -EINVAL; + goto err_clock_disable; } i = 0; @@ -1338,8 +1349,6 @@ static int bcmasp_probe(struct platform_device *pdev) */ bcmasp_core_clock_set(priv, 0, ASP_CTRL_CLOCK_CTRL_ASP_ALL_DISABLE); - clk_disable_unprepare(priv->clk); - /* Now do the registration of the network ports which will take care * of managing the clock properly. */ @@ -1352,12 +1361,16 @@ static int bcmasp_probe(struct platform_device *pdev) count++; } + clk_disable_unprepare(priv->clk); + dev_info(dev, "Initialized %d port(s)\n", count); return ret; err_cleanup: bcmasp_remove_intfs(priv); +err_clock_disable: + clk_disable_unprepare(priv->clk); return ret; } From 6c2128505f61b504c79a20b89596feba61388112 Mon Sep 17 00:00:00 2001 From: Ihor Solodrai Date: Thu, 19 Mar 2026 17:08:08 -0700 Subject: [PATCH 260/352] bpf: Fix exception exit lock checking for subprogs process_bpf_exit_full() passes check_lock = !curframe to check_resource_leak(), which is false in cases when bpf_throw() is called from a static subprog. This makes check_resource_leak() to skip validation of active_rcu_locks, active_preempt_locks, and active_irq_id on exception exits from subprogs. At runtime bpf_throw() unwinds the stack via ORC without releasing any user-acquired locks, which may cause various issues as the result. Fix by setting check_lock = true for exception exits regardless of curframe, since exceptions bypass all intermediate frame cleanup. Update the error message prefix to "bpf_throw" for exception exits to distinguish them from normal BPF_EXIT. Fix reject_subprog_with_rcu_read_lock test which was previously passing for the wrong reason. Test program returned directly from the subprog call without closing the RCU section, so the error was triggered by the unclosed RCU lock on normal exit, not by bpf_throw. Update __msg annotations for affected tests to match the new "bpf_throw" error prefix. The spin_lock case is not affected because they are already checked [1] at the call site in do_check_insn() before bpf_throw can run. [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/kernel/bpf/verifier.c?h=v7.0-rc4#n21098 Assisted-by: Claude:claude-opus-4-6 Fixes: f18b03fabaa9 ("bpf: Implement BPF exceptions") Signed-off-by: Ihor Solodrai Acked-by: Yonghong Song Acked-by: Kumar Kartikeya Dwivedi Link: https://lore.kernel.org/r/20260320000809.643798-1-ihor.solodrai@linux.dev Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 3 ++- tools/testing/selftests/bpf/progs/exceptions_fail.c | 9 ++++++--- 2 files changed, 8 insertions(+), 4 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index df22bfc572e2..5c0e6809024f 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -20911,7 +20911,8 @@ static int process_bpf_exit_full(struct bpf_verifier_env *env, * state when it exits. */ int err = check_resource_leak(env, exception_exit, - !env->cur_state->curframe, + exception_exit || !env->cur_state->curframe, + exception_exit ? "bpf_throw" : "BPF_EXIT instruction in main prog"); if (err) return err; diff --git a/tools/testing/selftests/bpf/progs/exceptions_fail.c b/tools/testing/selftests/bpf/progs/exceptions_fail.c index 8a0fdff89927..d7f1c492e3dd 100644 --- a/tools/testing/selftests/bpf/progs/exceptions_fail.c +++ b/tools/testing/selftests/bpf/progs/exceptions_fail.c @@ -8,6 +8,7 @@ #include "bpf_experimental.h" extern void bpf_rcu_read_lock(void) __ksym; +extern void bpf_rcu_read_unlock(void) __ksym; #define private(name) SEC(".bss." #name) __hidden __attribute__((aligned(8))) @@ -131,7 +132,7 @@ int reject_subprog_with_lock(void *ctx) } SEC("?tc") -__failure __msg("BPF_EXIT instruction in main prog cannot be used inside bpf_rcu_read_lock-ed region") +__failure __msg("bpf_throw cannot be used inside bpf_rcu_read_lock-ed region") int reject_with_rcu_read_lock(void *ctx) { bpf_rcu_read_lock(); @@ -147,11 +148,13 @@ __noinline static int throwing_subprog(struct __sk_buff *ctx) } SEC("?tc") -__failure __msg("BPF_EXIT instruction in main prog cannot be used inside bpf_rcu_read_lock-ed region") +__failure __msg("bpf_throw cannot be used inside bpf_rcu_read_lock-ed region") int reject_subprog_with_rcu_read_lock(void *ctx) { bpf_rcu_read_lock(); - return throwing_subprog(ctx); + throwing_subprog(ctx); + bpf_rcu_read_unlock(); + return 0; } static bool rbless(struct bpf_rb_node *n1, const struct bpf_rb_node *n2) From a1e5c46eaed3151be93e1aec9af0d8f8db79b8f6 Mon Sep 17 00:00:00 2001 From: Ihor Solodrai Date: Thu, 19 Mar 2026 17:08:09 -0700 Subject: [PATCH 261/352] selftests/bpf: Add tests for bpf_throw lock leak from subprogs Add test cases to ensure the verifier correctly rejects bpf_throw from subprogs when RCU, preempt, or IRQ locks are held: * reject_subprog_rcu_lock_throw: subprog acquires bpf_rcu_read_lock and then calls bpf_throw * reject_subprog_throw_preempt_lock: always-throwing subprog called while caller holds bpf_preempt_disable * reject_subprog_throw_irq_lock: always-throwing subprog called while caller holds bpf_local_irq_save Assisted-by: Claude:claude-opus-4-6 Signed-off-by: Ihor Solodrai Acked-by: Yonghong Song Acked-by: Kumar Kartikeya Dwivedi Link: https://lore.kernel.org/r/20260320000809.643798-2-ihor.solodrai@linux.dev Signed-off-by: Alexei Starovoitov --- .../selftests/bpf/progs/exceptions_fail.c | 47 +++++++++++++++++++ 1 file changed, 47 insertions(+) diff --git a/tools/testing/selftests/bpf/progs/exceptions_fail.c b/tools/testing/selftests/bpf/progs/exceptions_fail.c index d7f1c492e3dd..9ea1353488d7 100644 --- a/tools/testing/selftests/bpf/progs/exceptions_fail.c +++ b/tools/testing/selftests/bpf/progs/exceptions_fail.c @@ -9,6 +9,10 @@ extern void bpf_rcu_read_lock(void) __ksym; extern void bpf_rcu_read_unlock(void) __ksym; +extern void bpf_preempt_disable(void) __ksym; +extern void bpf_preempt_enable(void) __ksym; +extern void bpf_local_irq_save(unsigned long *) __ksym; +extern void bpf_local_irq_restore(unsigned long *) __ksym; #define private(name) SEC(".bss." #name) __hidden __attribute__((aligned(8))) @@ -349,4 +353,47 @@ int reject_exception_throw_cb_diff(struct __sk_buff *ctx) return 0; } +__noinline static int always_throws(void) +{ + bpf_throw(0); + return 0; +} + +__noinline static int rcu_lock_then_throw(void) +{ + bpf_rcu_read_lock(); + bpf_throw(0); + return 0; +} + +SEC("?tc") +__failure __msg("bpf_throw cannot be used inside bpf_rcu_read_lock-ed region") +int reject_subprog_rcu_lock_throw(void *ctx) +{ + rcu_lock_then_throw(); + return 0; +} + +SEC("?tc") +__failure __msg("bpf_throw cannot be used inside bpf_preempt_disable-ed region") +int reject_subprog_throw_preempt_lock(void *ctx) +{ + bpf_preempt_disable(); + always_throws(); + bpf_preempt_enable(); + return 0; +} + +SEC("?tc") +__failure __msg("bpf_throw cannot be used inside bpf_local_irq_save-ed region") +int reject_subprog_throw_irq_lock(void *ctx) +{ + unsigned long flags; + + bpf_local_irq_save(&flags); + always_throws(); + bpf_local_irq_restore(&flags); + return 0; +} + char _license[] SEC("license") = "GPL"; From c77b30bd1dcb61f66c640ff7d2757816210c7cb0 Mon Sep 17 00:00:00 2001 From: Jenny Guanni Qu Date: Wed, 11 Mar 2026 01:11:15 +0000 Subject: [PATCH 262/352] bpf: Fix undefined behavior in interpreter sdiv/smod for INT_MIN The BPF interpreter's signed 32-bit division and modulo handlers use the kernel abs() macro on s32 operands. The abs() macro documentation (include/linux/math.h) explicitly states the result is undefined when the input is the type minimum. When DST contains S32_MIN (0x80000000), abs((s32)DST) triggers undefined behavior and returns S32_MIN unchanged on arm64/x86. This value is then sign-extended to u64 as 0xFFFFFFFF80000000, causing do_div() to compute the wrong result. The verifier's abstract interpretation (scalar32_min_max_sdiv) computes the mathematically correct result for range tracking, creating a verifier/interpreter mismatch that can be exploited for out-of-bounds map value access. Introduce abs_s32() which handles S32_MIN correctly by casting to u32 before negating, avoiding signed overflow entirely. Replace all 8 abs((s32)...) call sites in the interpreter's sdiv32/smod32 handlers. s32 is the only affected case -- the s64 division/modulo handlers do not use abs(). Fixes: ec0e2da95f72 ("bpf: Support new signed div/mod instructions.") Acked-by: Yonghong Song Acked-by: Mykyta Yatsenko Signed-off-by: Jenny Guanni Qu Link: https://lore.kernel.org/r/20260311011116.2108005-2-qguanni@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/core.c | 22 ++++++++++++++-------- 1 file changed, 14 insertions(+), 8 deletions(-) diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 9e126be33755..7b675a451ec8 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -1757,6 +1757,12 @@ bool bpf_opcode_in_insntable(u8 code) } #ifndef CONFIG_BPF_JIT_ALWAYS_ON +/* Absolute value of s32 without undefined behavior for S32_MIN */ +static u32 abs_s32(s32 x) +{ + return x >= 0 ? (u32)x : -(u32)x; +} + /** * ___bpf_prog_run - run eBPF program on a given context * @regs: is the array of MAX_BPF_EXT_REG eBPF pseudo-registers @@ -1921,8 +1927,8 @@ static u64 ___bpf_prog_run(u64 *regs, const struct bpf_insn *insn) DST = do_div(AX, (u32) SRC); break; case 1: - AX = abs((s32)DST); - AX = do_div(AX, abs((s32)SRC)); + AX = abs_s32((s32)DST); + AX = do_div(AX, abs_s32((s32)SRC)); if ((s32)DST < 0) DST = (u32)-AX; else @@ -1949,8 +1955,8 @@ static u64 ___bpf_prog_run(u64 *regs, const struct bpf_insn *insn) DST = do_div(AX, (u32) IMM); break; case 1: - AX = abs((s32)DST); - AX = do_div(AX, abs((s32)IMM)); + AX = abs_s32((s32)DST); + AX = do_div(AX, abs_s32((s32)IMM)); if ((s32)DST < 0) DST = (u32)-AX; else @@ -1976,8 +1982,8 @@ static u64 ___bpf_prog_run(u64 *regs, const struct bpf_insn *insn) DST = (u32) AX; break; case 1: - AX = abs((s32)DST); - do_div(AX, abs((s32)SRC)); + AX = abs_s32((s32)DST); + do_div(AX, abs_s32((s32)SRC)); if (((s32)DST < 0) == ((s32)SRC < 0)) DST = (u32)AX; else @@ -2003,8 +2009,8 @@ static u64 ___bpf_prog_run(u64 *regs, const struct bpf_insn *insn) DST = (u32) AX; break; case 1: - AX = abs((s32)DST); - do_div(AX, abs((s32)IMM)); + AX = abs_s32((s32)DST); + do_div(AX, abs_s32((s32)IMM)); if (((s32)DST < 0) == ((s32)IMM < 0)) DST = (u32)AX; else From 4ac95c65efeaf0c010199b2b2f5f78b06f28dab6 Mon Sep 17 00:00:00 2001 From: Jenny Guanni Qu Date: Wed, 11 Mar 2026 01:11:16 +0000 Subject: [PATCH 263/352] selftests/bpf: Add tests for sdiv32/smod32 with INT_MIN dividend Add tests to verify that signed 32-bit division and modulo operations produce correct results when the dividend is INT_MIN (0x80000000). The bug fixed in the previous commit only affects the BPF interpreter path. When JIT is enabled (the default on most architectures), the native CPU division instruction produces the correct result and these tests pass regardless. With bpf_jit_enable=0, the interpreter is used and without the previous fix, INT_MIN / 2 incorrectly returns 0x40000000 instead of 0xC0000000 due to abs(S32_MIN) undefined behavior, causing these tests to fail. Test cases: - SDIV32 INT_MIN / 2 = -1073741824 (imm and reg divisor) - SMOD32 INT_MIN % 2 = 0 (positive and negative divisor) Reviewed-by: Jiayuan Chen Acked-by: Yonghong Song Signed-off-by: Jenny Guanni Qu Link: https://lore.kernel.org/r/20260311011116.2108005-3-qguanni@gmail.com Signed-off-by: Alexei Starovoitov --- .../selftests/bpf/progs/verifier_sdiv.c | 58 +++++++++++++++++++ 1 file changed, 58 insertions(+) diff --git a/tools/testing/selftests/bpf/progs/verifier_sdiv.c b/tools/testing/selftests/bpf/progs/verifier_sdiv.c index 148d2299e5b4..fd59d57e8e37 100644 --- a/tools/testing/selftests/bpf/progs/verifier_sdiv.c +++ b/tools/testing/selftests/bpf/progs/verifier_sdiv.c @@ -1209,6 +1209,64 @@ __naked void smod32_ri_divisor_neg_1(void) : __clobber_all); } +SEC("socket") +__description("SDIV32, INT_MIN divided by 2, imm") +__success __success_unpriv __retval(-1073741824) +__naked void sdiv32_int_min_div_2_imm(void) +{ + asm volatile (" \ + w0 = %[int_min]; \ + w0 s/= 2; \ + exit; \ +" : + : __imm_const(int_min, INT_MIN) + : __clobber_all); +} + +SEC("socket") +__description("SDIV32, INT_MIN divided by 2, reg") +__success __success_unpriv __retval(-1073741824) +__naked void sdiv32_int_min_div_2_reg(void) +{ + asm volatile (" \ + w0 = %[int_min]; \ + w1 = 2; \ + w0 s/= w1; \ + exit; \ +" : + : __imm_const(int_min, INT_MIN) + : __clobber_all); +} + +SEC("socket") +__description("SMOD32, INT_MIN modulo 2, imm") +__success __success_unpriv __retval(0) +__naked void smod32_int_min_mod_2_imm(void) +{ + asm volatile (" \ + w0 = %[int_min]; \ + w0 s%%= 2; \ + exit; \ +" : + : __imm_const(int_min, INT_MIN) + : __clobber_all); +} + +SEC("socket") +__description("SMOD32, INT_MIN modulo -2, imm") +__success __success_unpriv __retval(0) +__naked void smod32_int_min_mod_neg2_imm(void) +{ + asm volatile (" \ + w0 = %[int_min]; \ + w0 s%%= -2; \ + exit; \ +" : + : __imm_const(int_min, INT_MIN) + : __clobber_all); +} + + #else SEC("socket") From c845894ebd6fb43226b3118d6b017942550910c5 Mon Sep 17 00:00:00 2001 From: Daniel Wade Date: Sat, 14 Mar 2026 13:15:20 +1100 Subject: [PATCH 264/352] bpf: Fix unsound scalar forking in maybe_fork_scalars() for BPF_OR maybe_fork_scalars() is called for both BPF_AND and BPF_OR when the source operand is a constant. When dst has signed range [-1, 0], it forks the verifier state: the pushed path gets dst = 0, the current path gets dst = -1. For BPF_AND this is correct: 0 & K == 0. For BPF_OR this is wrong: 0 | K == K, not 0. The pushed path therefore tracks dst as 0 when the runtime value is K, producing an exploitable verifier/runtime divergence that allows out-of-bounds map access. Fix this by passing env->insn_idx (instead of env->insn_idx + 1) to push_stack(), so the pushed path re-executes the ALU instruction with dst = 0 and naturally computes the correct result for any opcode. Fixes: bffacdb80b93 ("bpf: Recognize special arithmetic shift in the verifier") Signed-off-by: Daniel Wade Reviewed-by: Amery Hung Acked-by: Eduard Zingerman Link: https://lore.kernel.org/r/20260314021521.128361-2-danjwade95@gmail.com Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 5c0e6809024f..62377bcb66fd 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -15999,7 +15999,7 @@ static int maybe_fork_scalars(struct bpf_verifier_env *env, struct bpf_insn *ins else return 0; - branch = push_stack(env, env->insn_idx + 1, env->insn_idx, false); + branch = push_stack(env, env->insn_idx, env->insn_idx, false); if (IS_ERR(branch)) return PTR_ERR(branch); From 0ad1734cc5598d3ddb6126a8960efe85f0f807d7 Mon Sep 17 00:00:00 2001 From: Daniel Wade Date: Sat, 14 Mar 2026 13:15:21 +1100 Subject: [PATCH 265/352] selftests/bpf: Add tests for maybe_fork_scalars() OR vs AND handling Add three test cases to verifier_bounds.c to verify that maybe_fork_scalars() correctly tracks register values for BPF_OR operations with constant source operands: 1. or_scalar_fork_rejects_oob: After ARSH 63 + OR 8, the pushed path should have dst = 8. With value_size = 8, accessing map_value + 8 is out of bounds and must be rejected. 2. and_scalar_fork_still_works: Regression test ensuring AND forking continues to work. ARSH 63 + AND 4 produces pushed dst = 0 and current dst = 4, both within value_size = 8. 3. or_scalar_fork_allows_inbounds: After ARSH 63 + OR 4, the pushed path has dst = 4, which is within value_size = 8 and should be accepted. These tests exercise the fix in the previous patch, which makes the pushed path re-execute the ALU instruction so it computes the correct result for BPF_OR. Signed-off-by: Daniel Wade Reviewed-by: Amery Hung Acked-by: Eduard Zingerman Link: https://lore.kernel.org/r/20260314021521.128361-3-danjwade95@gmail.com Signed-off-by: Alexei Starovoitov --- .../selftests/bpf/progs/verifier_bounds.c | 94 +++++++++++++++++++ 1 file changed, 94 insertions(+) diff --git a/tools/testing/selftests/bpf/progs/verifier_bounds.c b/tools/testing/selftests/bpf/progs/verifier_bounds.c index e526315c718a..79a328276805 100644 --- a/tools/testing/selftests/bpf/progs/verifier_bounds.c +++ b/tools/testing/selftests/bpf/progs/verifier_bounds.c @@ -2037,4 +2037,98 @@ __naked void signed_unsigned_intersection32_case2(void *ctx) : __clobber_all); } +SEC("socket") +__description("maybe_fork_scalars: OR with constant rejects OOB") +__failure __msg("invalid access to map value") +__naked void or_scalar_fork_rejects_oob(void) +{ + asm volatile (" \ + r1 = 0; \ + *(u64*)(r10 - 8) = r1; \ + r2 = r10; \ + r2 += -8; \ + r1 = %[map_hash_8b] ll; \ + call %[bpf_map_lookup_elem]; \ + if r0 == 0 goto l0_%=; \ + r9 = r0; \ + r6 = *(u64*)(r9 + 0); \ + r6 s>>= 63; \ + r6 |= 8; \ + /* r6 is -1 (current) or 8 (pushed) */ \ + if r6 s< 0 goto l0_%=; \ + /* pushed path: r6 = 8, OOB for value_size=8 */ \ + r9 += r6; \ + r0 = *(u8*)(r9 + 0); \ +l0_%=: r0 = 0; \ + exit; \ +" : + : __imm(bpf_map_lookup_elem), + __imm_addr(map_hash_8b) + : __clobber_all); +} + +SEC("socket") +__description("maybe_fork_scalars: AND with constant still works") +__success __retval(0) +__naked void and_scalar_fork_still_works(void) +{ + asm volatile (" \ + r1 = 0; \ + *(u64*)(r10 - 8) = r1; \ + r2 = r10; \ + r2 += -8; \ + r1 = %[map_hash_8b] ll; \ + call %[bpf_map_lookup_elem]; \ + if r0 == 0 goto l0_%=; \ + r9 = r0; \ + r6 = *(u64*)(r9 + 0); \ + r6 s>>= 63; \ + r6 &= 4; \ + /* \ + * r6 is 0 (pushed, 0&4==0) or 4 (current) \ + * both within value_size=8 \ + */ \ + if r6 s< 0 goto l0_%=; \ + r9 += r6; \ + r0 = *(u8*)(r9 + 0); \ +l0_%=: r0 = 0; \ + exit; \ +" : + : __imm(bpf_map_lookup_elem), + __imm_addr(map_hash_8b) + : __clobber_all); +} + +SEC("socket") +__description("maybe_fork_scalars: OR with constant allows in-bounds") +__success __retval(0) +__naked void or_scalar_fork_allows_inbounds(void) +{ + asm volatile (" \ + r1 = 0; \ + *(u64*)(r10 - 8) = r1; \ + r2 = r10; \ + r2 += -8; \ + r1 = %[map_hash_8b] ll; \ + call %[bpf_map_lookup_elem]; \ + if r0 == 0 goto l0_%=; \ + r9 = r0; \ + r6 = *(u64*)(r9 + 0); \ + r6 s>>= 63; \ + r6 |= 4; \ + /* \ + * r6 is -1 (current) or 4 (pushed) \ + * pushed path: r6 = 4, within value_size=8 \ + */ \ + if r6 s< 0 goto l0_%=; \ + r9 += r6; \ + r0 = *(u8*)(r9 + 0); \ +l0_%=: r0 = 0; \ + exit; \ +" : + : __imm(bpf_map_lookup_elem), + __imm_addr(map_hash_8b) + : __clobber_all); +} + char _license[] SEC("license") = "GPL"; From bc308be380c136800e1e94c6ce49cb53141d6506 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 19 Mar 2026 22:15:06 +0100 Subject: [PATCH 266/352] bpf: Fix sync_linked_regs regarding BPF_ADD_CONST32 zext propagation Jenny reported that in sync_linked_regs() the BPF_ADD_CONST32 flag is checked on known_reg (the register narrowed by a conditional branch) instead of reg (the linked target register created by an alu32 operation). Example case with reg: 1. r6 = bpf_get_prandom_u32() 2. r7 = r6 (linked, same id) 3. w7 += 5 (alu32 -- r7 gets BPF_ADD_CONST32, zero-extended by CPU) 4. if w6 < 0xFFFFFFFC goto safe (narrows r6 to [0xFFFFFFFC, 0xFFFFFFFF]) 5. sync_linked_regs() propagates to r7 but does NOT call zext_32_to_64() 6. Verifier thinks r7 is [0x100000001, 0x100000004] instead of [1, 4] Since known_reg above does not have BPF_ADD_CONST32 set above, zext_32_to_64() is never called on alu32-derived linked registers. This causes the verifier to track incorrect 64-bit bounds, while the CPU correctly zero-extends the 32-bit result. The code checking known_reg->id was correct however (see scalars_alu32_wrap selftest case), but the real fix needs to handle both directions - zext propagation should be done when either register has BPF_ADD_CONST32, since the linked relationship involves a 32-bit operation regardless of which side has the flag. Example case with known_reg (exercised also by scalars_alu32_wrap): 1. r1 = r0; w1 += 0x100 (alu32 -- r1 gets BPF_ADD_CONST32) 2. if r1 > 0x80 - known_reg = r1 (has BPF_ADD_CONST32), reg = r0 (doesn't) Hence, fix it by checking for (reg->id | known_reg->id) & BPF_ADD_CONST32. Moreover, sync_linked_regs() also has a soundness issue when two linked registers used different ALU widths: one with BPF_ADD_CONST32 and the other with BPF_ADD_CONST64. The delta relationship between linked registers assumes the same arithmetic width though. When one register went through alu32 (CPU zero-extends the 32-bit result) and the other went through alu64 (no zero-extension), the propagation produces incorrect bounds. Example: r6 = bpf_get_prandom_u32() // fully unknown if r6 >= 0x100000000 goto out // constrain r6 to [0, U32_MAX] r7 = r6 w7 += 1 // alu32: r7.id = N | BPF_ADD_CONST32 r8 = r6 r8 += 2 // alu64: r8.id = N | BPF_ADD_CONST64 if r7 < 0xFFFFFFFF goto out // narrows r7 to [0xFFFFFFFF, 0xFFFFFFFF] At the branch on r7, sync_linked_regs() runs with known_reg=r7 (BPF_ADD_CONST32) and reg=r8 (BPF_ADD_CONST64). The delta path computes: r8 = r7 + (delta_r8 - delta_r7) = 0xFFFFFFFF + (2 - 1) = 0x100000000 Then, because known_reg->id has BPF_ADD_CONST32, zext_32_to_64(r8) is called, truncating r8 to [0, 0]. But r8 used a 64-bit ALU op -- the CPU does NOT zero-extend it. The actual CPU value of r8 is 0xFFFFFFFE + 2 = 0x100000000, not 0. The verifier now underestimates r8's 64-bit bounds, which is a soundness violation. Fix sync_linked_regs() by skipping propagation when the two registers have mixed ALU widths (one BPF_ADD_CONST32, the other BPF_ADD_CONST64). Lastly, fix regsafe() used for path pruning: the existing checks used "& BPF_ADD_CONST" to test for offset linkage, which treated BPF_ADD_CONST32 and BPF_ADD_CONST64 as equivalent. Fixes: 7a433e519364 ("bpf: Support negative offsets, BPF_SUB, and alu32 for linked register tracking") Reported-by: Jenny Guanni Qu Co-developed-by: Puranjay Mohan Signed-off-by: Puranjay Mohan Signed-off-by: Daniel Borkmann Acked-by: Eduard Zingerman Link: https://lore.kernel.org/r/20260319211507.213816-1-daniel@iogearbox.net Signed-off-by: Alexei Starovoitov --- kernel/bpf/verifier.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 62377bcb66fd..f108c01ff6d0 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -17415,6 +17415,12 @@ static void sync_linked_regs(struct bpf_verifier_env *env, struct bpf_verifier_s continue; if ((reg->id & ~BPF_ADD_CONST) != (known_reg->id & ~BPF_ADD_CONST)) continue; + /* + * Skip mixed 32/64-bit links: the delta relationship doesn't + * hold across different ALU widths. + */ + if (((reg->id ^ known_reg->id) & BPF_ADD_CONST) == BPF_ADD_CONST) + continue; if ((!(reg->id & BPF_ADD_CONST) && !(known_reg->id & BPF_ADD_CONST)) || reg->off == known_reg->off) { s32 saved_subreg_def = reg->subreg_def; @@ -17442,7 +17448,7 @@ static void sync_linked_regs(struct bpf_verifier_env *env, struct bpf_verifier_s scalar32_min_max_add(reg, &fake_reg); scalar_min_max_add(reg, &fake_reg); reg->var_off = tnum_add(reg->var_off, fake_reg.var_off); - if (known_reg->id & BPF_ADD_CONST32) + if ((reg->id | known_reg->id) & BPF_ADD_CONST32) zext_32_to_64(reg); reg_bounds_sync(reg); } @@ -19870,11 +19876,14 @@ static bool regsafe(struct bpf_verifier_env *env, struct bpf_reg_state *rold, * Also verify that new value satisfies old value range knowledge. */ - /* ADD_CONST mismatch: different linking semantics */ - if ((rold->id & BPF_ADD_CONST) && !(rcur->id & BPF_ADD_CONST)) - return false; - - if (rold->id && !(rold->id & BPF_ADD_CONST) && (rcur->id & BPF_ADD_CONST)) + /* + * ADD_CONST flags must match exactly: BPF_ADD_CONST32 and + * BPF_ADD_CONST64 have different linking semantics in + * sync_linked_regs() (alu32 zero-extends, alu64 does not), + * so pruning across different flag types is unsafe. + */ + if (rold->id && + (rold->id & BPF_ADD_CONST) != (rcur->id & BPF_ADD_CONST)) return false; /* Both have offset linkage: offsets must match */ From 4a04d13576fd69149b91672b5f1dc62eca272fa5 Mon Sep 17 00:00:00 2001 From: Daniel Borkmann Date: Thu, 19 Mar 2026 22:15:07 +0100 Subject: [PATCH 267/352] selftests/bpf: Add a test cases for sync_linked_regs regarding zext propagation Add multiple test cases for linked register tracking with alu32 ops: - Add a test that checks sync_linked_regs() regarding reg->id (the linked target register) for BPF_ADD_CONST32 rather than known_reg->id (the branch register). - Add a test case for linked register tracking that exposes the cross-type sync_linked_regs() bug. One register uses alu32 (w7 += 1, BPF_ADD_CONST32) and another uses alu64 (r8 += 2, BPF_ADD_CONST64), both linked to the same base register. - Add a test case that exercises regsafe() path pruning when two execution paths reach the same program point with linked registers carrying different ADD_CONST flags (BPF_ADD_CONST32 from alu32 vs BPF_ADD_CONST64 from alu64). This particular test passes with and without the fix since the pruning will fail due to different ranges, but it would still be useful to carry this one as a regression test for the unreachable div by zero. With the fix applied all the tests pass: # LDLIBS=-static PKG_CONFIG='pkg-config --static' ./vmtest.sh -- ./test_progs -t verifier_linked_scalars [...] ./test_progs -t verifier_linked_scalars #602/1 verifier_linked_scalars/scalars: find linked scalars:OK #602/2 verifier_linked_scalars/sync_linked_regs_preserves_id:OK #602/3 verifier_linked_scalars/scalars_neg:OK #602/4 verifier_linked_scalars/scalars_neg_sub:OK #602/5 verifier_linked_scalars/scalars_neg_alu32_add:OK #602/6 verifier_linked_scalars/scalars_neg_alu32_sub:OK #602/7 verifier_linked_scalars/scalars_pos:OK #602/8 verifier_linked_scalars/scalars_sub_neg_imm:OK #602/9 verifier_linked_scalars/scalars_double_add:OK #602/10 verifier_linked_scalars/scalars_sync_delta_overflow:OK #602/11 verifier_linked_scalars/scalars_sync_delta_overflow_large_range:OK #602/12 verifier_linked_scalars/scalars_alu32_big_offset:OK #602/13 verifier_linked_scalars/scalars_alu32_basic:OK #602/14 verifier_linked_scalars/scalars_alu32_wrap:OK #602/15 verifier_linked_scalars/scalars_alu32_zext_linked_reg:OK #602/16 verifier_linked_scalars/scalars_alu32_alu64_cross_type:OK #602/17 verifier_linked_scalars/scalars_alu32_alu64_regsafe_pruning:OK #602/18 verifier_linked_scalars/alu32_negative_offset:OK #602/19 verifier_linked_scalars/spurious_precision_marks:OK #602 verifier_linked_scalars:OK Summary: 1/19 PASSED, 0 SKIPPED, 0 FAILED Co-developed-by: Puranjay Mohan Signed-off-by: Puranjay Mohan Signed-off-by: Daniel Borkmann Acked-by: Eduard Zingerman Link: https://lore.kernel.org/r/20260319211507.213816-2-daniel@iogearbox.net Signed-off-by: Alexei Starovoitov --- .../bpf/progs/verifier_linked_scalars.c | 108 ++++++++++++++++++ 1 file changed, 108 insertions(+) diff --git a/tools/testing/selftests/bpf/progs/verifier_linked_scalars.c b/tools/testing/selftests/bpf/progs/verifier_linked_scalars.c index 7bf7dbfd237d..f4f8a055af8a 100644 --- a/tools/testing/selftests/bpf/progs/verifier_linked_scalars.c +++ b/tools/testing/selftests/bpf/progs/verifier_linked_scalars.c @@ -348,6 +348,114 @@ l0_%=: \ : __clobber_all); } +/* + * Test that sync_linked_regs() checks reg->id (the linked target register) + * for BPF_ADD_CONST32 rather than known_reg->id (the branch register). + */ +SEC("socket") +__success +__naked void scalars_alu32_zext_linked_reg(void) +{ + asm volatile (" \ + call %[bpf_get_prandom_u32]; \ + w6 = w0; /* r6 in [0, 0xFFFFFFFF] */ \ + r7 = r6; /* linked: same id as r6 */ \ + w7 += 1; /* alu32: r7.id |= BPF_ADD_CONST32 */ \ + r8 = 0xFFFFffff ll; \ + if r6 < r8 goto l0_%=; \ + /* r6 in [0xFFFFFFFF, 0xFFFFFFFF] */ \ + /* sync_linked_regs: known_reg=r6, reg=r7 */ \ + /* CPU: w7 = (u32)(0xFFFFFFFF + 1) = 0, zext -> r7 = 0 */ \ + /* With fix: r7 64-bit = [0, 0] (zext applied) */ \ + /* Without fix: r7 64-bit = [0x100000000] (no zext) */ \ + r7 >>= 32; \ + if r7 == 0 goto l0_%=; \ + r0 /= 0; /* unreachable with fix */ \ +l0_%=: \ + r0 = 0; \ + exit; \ +" : + : __imm(bpf_get_prandom_u32) + : __clobber_all); +} + +/* + * Test that sync_linked_regs() skips propagation when one register used + * alu32 (BPF_ADD_CONST32) and the other used alu64 (BPF_ADD_CONST64). + * The delta relationship doesn't hold across different ALU widths. + */ +SEC("socket") +__failure __msg("div by zero") +__naked void scalars_alu32_alu64_cross_type(void) +{ + asm volatile (" \ + call %[bpf_get_prandom_u32]; \ + w6 = w0; /* r6 in [0, 0xFFFFFFFF] */ \ + r7 = r6; /* linked: same id as r6 */ \ + w7 += 1; /* alu32: BPF_ADD_CONST32, delta = 1 */ \ + r8 = r6; /* linked: same id as r6 */ \ + r8 += 2; /* alu64: BPF_ADD_CONST64, delta = 2 */ \ + r9 = 0xFFFFffff ll; \ + if r7 < r9 goto l0_%=; \ + /* r7 = 0xFFFFFFFF */ \ + /* sync: known_reg=r7 (ADD_CONST32), reg=r8 (ADD_CONST64) */ \ + /* Without fix: r8 = zext(0xFFFFFFFF + 1) = 0 */ \ + /* With fix: r8 stays [2, 0x100000001] (r8 >= 2) */ \ + if r8 > 0 goto l1_%=; \ + goto l0_%=; \ +l1_%=: \ + r0 /= 0; /* div by zero */ \ +l0_%=: \ + r0 = 0; \ + exit; \ +" : + : __imm(bpf_get_prandom_u32) + : __clobber_all); +} + +/* + * Test that regsafe() prevents pruning when two paths reach the same program + * point with linked registers carrying different ADD_CONST flags (one + * BPF_ADD_CONST32 from alu32, another BPF_ADD_CONST64 from alu64). + */ +SEC("socket") +__failure __msg("div by zero") +__flag(BPF_F_TEST_STATE_FREQ) +__naked void scalars_alu32_alu64_regsafe_pruning(void) +{ + asm volatile (" \ + call %[bpf_get_prandom_u32]; \ + w6 = w0; /* r6 in [0, 0xFFFFFFFF] */ \ + r7 = r6; /* linked: same id as r6 */ \ + /* Get another random value for the path branch */ \ + call %[bpf_get_prandom_u32]; \ + if r0 > 0 goto l_pathb_%=; \ + /* Path A: alu32 */ \ + w7 += 1; /* BPF_ADD_CONST32, delta = 1 */\ + goto l_merge_%=; \ +l_pathb_%=: \ + /* Path B: alu64 */ \ + r7 += 1; /* BPF_ADD_CONST64, delta = 1 */\ +l_merge_%=: \ + /* Merge point: regsafe() compares path B against cached path A. */ \ + /* Narrow r6 to trigger sync_linked_regs for r7 */ \ + r9 = 0xFFFFffff ll; \ + if r6 < r9 goto l0_%=; \ + /* r6 = 0xFFFFFFFF */ \ + /* sync: r7 = 0xFFFFFFFF + 1 = 0x100000000 */ \ + /* Path A: zext -> r7 = 0 */ \ + /* Path B: no zext -> r7 = 0x100000000 */ \ + r7 >>= 32; \ + if r7 == 0 goto l0_%=; \ + r0 /= 0; /* div by zero on path B */ \ +l0_%=: \ + r0 = 0; \ + exit; \ +" : + : __imm(bpf_get_prandom_u32) + : __clobber_all); +} + SEC("socket") __success void alu32_negative_offset(void) From a6f22e50c7d51aa225c392c62c33f0fae11f734d Mon Sep 17 00:00:00 2001 From: Xuewen Yan Date: Fri, 6 Mar 2026 15:59:54 +0800 Subject: [PATCH 268/352] tracing: Revert "tracing: Remove pid in task_rename tracing output" This reverts commit e3f6a42272e028c46695acc83fc7d7c42f2750ad. The commit says that the tracepoint only deals with the current task, however the following case is not current task: comm_write() { p = get_proc_task(inode); if (!p) return -ESRCH; if (same_thread_group(current, p)) set_task_comm(p, buffer); } where set_task_comm() calls __set_task_comm() which records the update of p and not current. So revert the patch to show pid. Cc: Cc: Cc: Cc: Link: https://patch.msgid.link/20260306075954.4533-1-xuewen.yan@unisoc.com Fixes: e3f6a42272e0 ("tracing: Remove pid in task_rename tracing output") Reported-by: Guohua Yan Signed-off-by: Xuewen Yan Reviewed-by: Steven Rostedt (Google) Signed-off-by: Steven Rostedt (Google) --- include/trace/events/task.h | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/include/trace/events/task.h b/include/trace/events/task.h index 4f0759634306..b9a129eb54d9 100644 --- a/include/trace/events/task.h +++ b/include/trace/events/task.h @@ -38,19 +38,22 @@ TRACE_EVENT(task_rename, TP_ARGS(task, comm), TP_STRUCT__entry( + __field( pid_t, pid) __array( char, oldcomm, TASK_COMM_LEN) __array( char, newcomm, TASK_COMM_LEN) __field( short, oom_score_adj) ), TP_fast_assign( + __entry->pid = task->pid; memcpy(entry->oldcomm, task->comm, TASK_COMM_LEN); strscpy(entry->newcomm, comm, TASK_COMM_LEN); __entry->oom_score_adj = task->signal->oom_score_adj; ), - TP_printk("oldcomm=%s newcomm=%s oom_score_adj=%hd", - __entry->oldcomm, __entry->newcomm, __entry->oom_score_adj) + TP_printk("pid=%d oldcomm=%s newcomm=%s oom_score_adj=%hd", + __entry->pid, __entry->oldcomm, + __entry->newcomm, __entry->oom_score_adj) ); /** From edca33a56297d5741ccf867669debec116681987 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Mon, 16 Mar 2026 13:07:34 -0400 Subject: [PATCH 269/352] tracing: Fix failure to read user space from system call trace events The system call trace events call trace_user_fault_read() to read the user space part of some system calls. This is done by grabbing a per-cpu buffer, disabling migration, enabling preemption, calling copy_from_user(), disabling preemption, enabling migration and checking if the task was preempted while preemption was enabled. If it was, the buffer is considered corrupted and it tries again. There's a safety mechanism that will fail out of this loop if it fails 100 times (with a warning). That warning message was triggered in some pi_futex stress tests. Enabling the sched_switch trace event and traceoff_on_warning, showed the problem: pi_mutex_hammer-1375 [006] d..21 138.981648: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981651: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981656: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981659: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981664: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981667: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981671: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981675: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981679: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981682: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981687: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981690: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981695: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981698: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981703: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981706: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981711: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981714: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981719: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981722: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981727: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981730: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 pi_mutex_hammer-1375 [006] d..21 138.981735: sched_switch: prev_comm=pi_mutex_hammer prev_pid=1375 prev_prio=95 prev_state=R+ ==> next_comm=migration/6 next_pid=47 next_prio=0 migration/6-47 [006] d..2. 138.981738: sched_switch: prev_comm=migration/6 prev_pid=47 prev_prio=0 prev_state=S ==> next_comm=pi_mutex_hammer next_pid=1375 next_prio=95 What happened was the task 1375 was flagged to be migrated. When preemption was enabled, the migration thread woke up to migrate that task, but failed because migration for that task was disabled. This caused the loop to fail to exit because the task scheduled out while trying to read user space. Every time the task enabled preemption the migration thread would schedule in, try to migrate the task, fail and let the task continue. But because the loop would only enable preemption with migration disabled, it would always fail because each time it enabled preemption to read user space, the migration thread would try to migrate it. To solve this, when the loop fails to read user space without being scheduled out, enabled and disable preemption with migration enabled. This will allow the migration task to successfully migrate the task and the next loop should succeed to read user space without being scheduled out. Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Link: https://patch.msgid.link/20260316130734.1858a998@gandalf.local.home Fixes: 64cf7d058a005 ("tracing: Have trace_marker use per-cpu data to read user space") Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index ebd996f8710e..bb4a62f4b953 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -6783,6 +6783,23 @@ char *trace_user_fault_read(struct trace_user_buf_info *tinfo, */ do { + /* + * It is possible that something is trying to migrate this + * task. What happens then, is when preemption is enabled, + * the migration thread will preempt this task, try to + * migrate it, fail, then let it run again. That will + * cause this to loop again and never succeed. + * On failures, enabled and disable preemption with + * migration enabled, to allow the migration thread to + * migrate this task. + */ + if (trys) { + preempt_enable_notrace(); + preempt_disable_notrace(); + cpu = smp_processor_id(); + buffer = per_cpu_ptr(tinfo->tbuf, cpu)->buf; + } + /* * If for some reason, copy_from_user() always causes a context * switch, this would then cause an infinite loop. From 07183aac4a6828e474f00b37c9d795d0d99e18a7 Mon Sep 17 00:00:00 2001 From: Steven Rostedt Date: Wed, 18 Mar 2026 18:55:12 -0400 Subject: [PATCH 270/352] tracing: Fix trace_marker copy link list updates When the "copy_trace_marker" option is enabled for an instance, anything written into /sys/kernel/tracing/trace_marker is also copied into that instances buffer. When the option is set, that instance's trace_array descriptor is added to the marker_copies link list. This list is protected by RCU, as all iterations uses an RCU protected list traversal. When the instance is deleted, all the flags that were enabled are cleared. This also clears the copy_trace_marker flag and removes the trace_array descriptor from the list. The issue is after the flags are called, a direct call to update_marker_trace() is performed to clear the flag. This function returns true if the state of the flag changed and false otherwise. If it returns true here, synchronize_rcu() is called to make sure all readers see that its removed from the list. But since the flag was already cleared, the state does not change and the synchronization is never called, leaving a possible UAF bug. Move the clearing of all flags below the updating of the copy_trace_marker option which then makes sure the synchronization is performed. Also use the flag for checking the state in update_marker_trace() instead of looking at if the list is empty. Cc: stable@vger.kernel.org Cc: Masami Hiramatsu Cc: Mathieu Desnoyers Link: https://patch.msgid.link/20260318185512.1b6c7db4@gandalf.local.home Fixes: 7b382efd5e8a ("tracing: Allow the top level trace_marker to write into another instances") Reported-by: Sasha Levin Closes: https://lore.kernel.org/all/20260225133122.237275-1-sashal@kernel.org/ Signed-off-by: Steven Rostedt (Google) --- kernel/trace/trace.c | 19 ++++++++++--------- 1 file changed, 10 insertions(+), 9 deletions(-) diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index bb4a62f4b953..a626211ceb9a 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -555,7 +555,7 @@ static bool update_marker_trace(struct trace_array *tr, int enabled) lockdep_assert_held(&event_mutex); if (enabled) { - if (!list_empty(&tr->marker_list)) + if (tr->trace_flags & TRACE_ITER(COPY_MARKER)) return false; list_add_rcu(&tr->marker_list, &marker_copies); @@ -563,10 +563,10 @@ static bool update_marker_trace(struct trace_array *tr, int enabled) return true; } - if (list_empty(&tr->marker_list)) + if (!(tr->trace_flags & TRACE_ITER(COPY_MARKER))) return false; - list_del_init(&tr->marker_list); + list_del_rcu(&tr->marker_list); tr->trace_flags &= ~TRACE_ITER(COPY_MARKER); return true; } @@ -9761,18 +9761,19 @@ static int __remove_instance(struct trace_array *tr) list_del(&tr->list); + if (printk_trace == tr) + update_printk_trace(&global_trace); + + /* Must be done before disabling all the flags */ + if (update_marker_trace(tr, 0)) + synchronize_rcu(); + /* Disable all the flags that were enabled coming in */ for (i = 0; i < TRACE_FLAGS_MAX_SIZE; i++) { if ((1ULL << i) & ZEROED_TRACE_FLAGS) set_tracer_flag(tr, 1ULL << i, 0); } - if (printk_trace == tr) - update_printk_trace(&global_trace); - - if (update_marker_trace(tr, 0)) - synchronize_rcu(); - tracing_set_nop(tr); clear_ftrace_function_probes(tr); event_trace_del_tracer(tr); From f35dbac6942171dc4ce9398d1d216a59224590a9 Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Thu, 19 Mar 2026 18:12:19 +0900 Subject: [PATCH 271/352] ring-buffer: Fix to update per-subbuf entries of persistent ring buffer Since the validation loop in rb_meta_validate_events() updates the same cpu_buffer->head_page->entries, the other subbuf entries are not updated. Fix to use head_page to update the entries field, since it is the cursor in this loop. Cc: stable@vger.kernel.org Cc: Mathieu Desnoyers Cc: Ian Rogers Fixes: 5f3b6e839f3c ("ring-buffer: Validate boot range memory events") Link: https://patch.msgid.link/177391153882.193994.17158784065013676533.stgit@mhiramat.tok.corp.google.com Signed-off-by: Masami Hiramatsu (Google) Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ring_buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index 17d0ea0cc3e6..170170bd83bd 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -2053,7 +2053,7 @@ static void rb_meta_validate_events(struct ring_buffer_per_cpu *cpu_buffer) entries += ret; entry_bytes += local_read(&head_page->page->commit); - local_set(&cpu_buffer->head_page->entries, ret); + local_set(&head_page->entries, ret); if (head_page == cpu_buffer->commit_page) break; From 50b35c9e50a865600344ab1d8f9a8b3384d7e63d Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 12 Mar 2026 13:37:38 +0100 Subject: [PATCH 272/352] ftrace: Use hash argument for tmp_ops in update_ftrace_direct_mod The modify logic registers temporary ftrace_ops object (tmp_ops) to trigger the slow path for all direct callers to be able to safely modify attached addresses. At the moment we use ops->func_hash for tmp_ops filter, which represents all the systems attachments. It's faster to use just the passed hash filter, which contains only the modified sites and is always a subset of the ops->func_hash. Cc: Alexei Starovoitov Cc: Daniel Borkmann Cc: Andrii Nakryiko Cc: Menglong Dong Cc: Song Liu Link: https://patch.msgid.link/20260312123738.129926-1-jolsa@kernel.org Fixes: e93672f770d7 ("ftrace: Add update_ftrace_direct_mod function") Signed-off-by: Jiri Olsa Signed-off-by: Steven Rostedt (Google) --- kernel/trace/ftrace.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 8df69e702706..413310912609 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -6606,9 +6606,9 @@ int update_ftrace_direct_mod(struct ftrace_ops *ops, struct ftrace_hash *hash, b if (!orig_hash) goto unlock; - /* Enable the tmp_ops to have the same functions as the direct ops */ + /* Enable the tmp_ops to have the same functions as the hash object. */ ftrace_ops_init(&tmp_ops); - tmp_ops.func_hash = ops->func_hash; + tmp_ops.func_hash->filter_hash = hash; err = register_ftrace_function_nolock(&tmp_ops); if (err) From 3a206a8649f83bec99a3517da5e7dac9c138875e Mon Sep 17 00:00:00 2001 From: "Lorenzo Stoakes (Oracle)" Date: Wed, 18 Mar 2026 12:26:32 +0000 Subject: [PATCH 273/352] mm/rmap: clear vma->anon_vma on error Commit 542eda1a8329 ("mm/rmap: improve anon_vma_clone(), unlink_anon_vmas() comments, add asserts") alters the way errors are handled, but overlooked one important aspect of clean up. When a VMA encounters an error state in anon_vma_clone() (that is, on attempted allocation of anon_vma_chain objects), it cleans up partially established state in cleanup_partial_anon_vmas(), before returning an error. However, this occurs prior to anon_vma->num_active_vmas being incremented, and it also fails to clear the VMA's vma->anon_vma field, which remains in place. This is immediately an inconsistent state, because anon_vma->num_active_vmas is supposed to track the number of VMAs whose vma->anon_vma field references that anon_vma, and now that count is off-by-negative-1 for each VMA for which this error state has occurred. When VMAs are unlinked from this anon_vma, unlink_anon_vmas() will eventually underflow anon_vma->num_active_vmas, which will trigger a warning. This will always eventually happen, as we unlink anon_vma's at process teardown. It could also cause maybe_reuse_anon_vma() to incorrectly permit the reuse of an anon_vma which has active VMAs attached, which will lead to a persistently invalid state. The solution is to clear the VMA's anon_vma field when we clean up partial state, as the fact we are doing so indicates clearly that the VMA is not correctly integrated into the anon_vma tree and thus this field is invalid. Link: https://lkml.kernel.org/r/20260318122632.63404-1-ljs@kernel.org Fixes: 542eda1a8329 ("mm/rmap: improve anon_vma_clone(), unlink_anon_vmas() comments, add asserts") Signed-off-by: Lorenzo Stoakes (Oracle) Reported-by: Sasha Levin Closes: https://lore.kernel.org/linux-mm/20260302151547.2389070-1-sashal@kernel.org/ Reported-by: Jiakai Xu Closes: https://lore.kernel.org/linux-mm/CAFb8wJvRhatRD-9DVmr5v5pixTMPEr3UKjYBJjCd09OfH55CKg@mail.gmail.com/ Acked-by: David Hildenbrand (Arm) Acked-by: Vlastimil Babka (SUSE) Tested-by: Jiakai Xu Acked-by: Harry Yoo Cc: Jann Horn Cc: Liam Howlett Cc: Rik van Riel Cc: Sasha Levin (Microsoft) Signed-off-by: Andrew Morton --- mm/rmap.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/mm/rmap.c b/mm/rmap.c index 391337282e3f..8f08090d7eb9 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -457,6 +457,13 @@ static void cleanup_partial_anon_vmas(struct vm_area_struct *vma) list_del(&avc->same_vma); anon_vma_chain_free(avc); } + + /* + * The anon_vma assigned to this VMA is no longer valid, as we were not + * able to correctly clone AVC state. Avoid inconsistent anon_vma tree + * state by resetting. + */ + vma->anon_vma = NULL; } /** From 26f775a054c3cda86ad465a64141894a90a9e145 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Thu, 19 Mar 2026 07:52:17 -0700 Subject: [PATCH 274/352] mm/damon/core: avoid use of half-online-committed context One major usage of damon_call() is online DAMON parameters update. It is done by calling damon_commit_ctx() inside the damon_call() callback function. damon_commit_ctx() can fail for two reasons: 1) invalid parameters and 2) internal memory allocation failures. In case of failures, the damon_ctx that attempted to be updated (commit destination) can be partially updated (or, corrupted from a perspective), and therefore shouldn't be used anymore. The function only ensures the damon_ctx object can safely deallocated using damon_destroy_ctx(). The API callers are, however, calling damon_commit_ctx() only after asserting the parameters are valid, to avoid damon_commit_ctx() fails due to invalid input parameters. But it can still theoretically fail if the internal memory allocation fails. In the case, DAMON may run with the partially updated damon_ctx. This can result in unexpected behaviors including even NULL pointer dereference in case of damos_commit_dests() failure [1]. Such allocation failure is arguably too small to fail, so the real world impact would be rare. But, given the bad consequence, this needs to be fixed. Avoid such partially-committed (maybe-corrupted) damon_ctx use by saving the damon_commit_ctx() failure on the damon_ctx object. For this, introduce damon_ctx->maybe_corrupted field. damon_commit_ctx() sets it when it is failed. kdamond_call() checks if the field is set after each damon_call_control->fn() is executed. If it is set, ignore remaining callback requests and return. All kdamond_call() callers including kdamond_fn() also check the maybe_corrupted field right after kdamond_call() invocations. If the field is set, break the kdamond_fn() main loop so that DAMON sill doesn't use the context that might be corrupted. [sj@kernel.org: let kdamond_call() with cancel regardless of maybe_corrupted] Link: https://lkml.kernel.org/r/20260320031553.2479-1-sj@kernel.org Link: https://sashiko.dev/#/patchset/20260319145218.86197-1-sj%40kernel.org Link: https://lkml.kernel.org/r/20260319145218.86197-1-sj@kernel.org Link: https://lore.kernel.org/20260319043309.97966-1-sj@kernel.org [1] Fixes: 3301f1861d34 ("mm/damon/sysfs: handle commit command using damon_call()") Signed-off-by: SeongJae Park Cc: [6.15+] Signed-off-by: Andrew Morton --- include/linux/damon.h | 6 ++++++ mm/damon/core.c | 8 ++++++++ 2 files changed, 14 insertions(+) diff --git a/include/linux/damon.h b/include/linux/damon.h index a4fea23da857..be3d198043ff 100644 --- a/include/linux/damon.h +++ b/include/linux/damon.h @@ -810,6 +810,12 @@ struct damon_ctx { struct damos_walk_control *walk_control; struct mutex walk_control_lock; + /* + * indicate if this may be corrupted. Currentonly this is set only for + * damon_commit_ctx() failure. + */ + bool maybe_corrupted; + /* Working thread of the given DAMON context */ struct task_struct *kdamond; /* Protects @kdamond field access */ diff --git a/mm/damon/core.c b/mm/damon/core.c index c1d1091d307e..3e1890d64d06 100644 --- a/mm/damon/core.c +++ b/mm/damon/core.c @@ -1252,6 +1252,7 @@ int damon_commit_ctx(struct damon_ctx *dst, struct damon_ctx *src) { int err; + dst->maybe_corrupted = true; if (!is_power_of_2(src->min_region_sz)) return -EINVAL; @@ -1277,6 +1278,7 @@ int damon_commit_ctx(struct damon_ctx *dst, struct damon_ctx *src) dst->addr_unit = src->addr_unit; dst->min_region_sz = src->min_region_sz; + dst->maybe_corrupted = false; return 0; } @@ -2678,6 +2680,8 @@ static void kdamond_call(struct damon_ctx *ctx, bool cancel) complete(&control->completion); else if (control->canceled && control->dealloc_on_cancel) kfree(control); + if (!cancel && ctx->maybe_corrupted) + break; } mutex_lock(&ctx->call_controls_lock); @@ -2707,6 +2711,8 @@ static int kdamond_wait_activation(struct damon_ctx *ctx) kdamond_usleep(min_wait_time); kdamond_call(ctx, false); + if (ctx->maybe_corrupted) + return -EINVAL; damos_walk_cancel(ctx); } return -EBUSY; @@ -2790,6 +2796,8 @@ static int kdamond_fn(void *data) * kdamond_merge_regions() if possible, to reduce overhead */ kdamond_call(ctx, false); + if (ctx->maybe_corrupted) + break; if (!list_empty(&ctx->schemes)) kdamond_apply_schemes(ctx); else From b0377ee8042985b0d91bf579afcc4ee9150db14d Mon Sep 17 00:00:00 2001 From: Sergey Senozhatsky Date: Thu, 19 Mar 2026 12:44:56 +0900 Subject: [PATCH 275/352] zram: do not slot_free() written-back slots slot_free() basically completely resets the slots by clearing all of its flags and attributes. While zram_writeback_complete() restores some of flags back (those that are necessary for async read decompression) we still lose a lot of slot's metadata. For example, slot's ac-time, or ZRAM_INCOMPRESSIBLE. More importantly, restoring flags/attrs requires extra attention as some of the flags are directly affecting zram device stats. And the original code did not pay that attention. Namely ZRAM_HUGE slots handling in zram_writeback_complete(). The call to slot_free() would decrement ->huge_pages, however when zram_writeback_complete() restored the slot's ZRAM_HUGE flag, it would not get reflected in an incremented ->huge_pages. So when the slot would finally get freed, slot_free() would decrement ->huge_pages again, leading to underflow. Fix this by open-coding the required memory free and stats updates in zram_writeback_complete(), rather than calling the destructive slot_free(). Since we now preserve the ZRAM_HUGE flag on written-back slots (for the deferred decompression path), we also update slot_free() to skip decrementing ->huge_pages if ZRAM_WB is set. Link: https://lkml.kernel.org/r/20260320023143.2372879-1-senozhatsky@chromium.org Link: https://lkml.kernel.org/r/20260319034912.1894770-1-senozhatsky@chromium.org Fixes: d38fab605c667 ("zram: introduce compressed data writeback") Signed-off-by: Sergey Senozhatsky Acked-by: Minchan Kim Cc: Brian Geffon Cc: Richard Chang Signed-off-by: Andrew Morton --- drivers/block/zram/zram_drv.c | 39 +++++++++++++---------------------- 1 file changed, 14 insertions(+), 25 deletions(-) diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index a324ede6206d..af679375b193 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -917,9 +917,8 @@ static void zram_account_writeback_submit(struct zram *zram) static int zram_writeback_complete(struct zram *zram, struct zram_wb_req *req) { - u32 size, index = req->pps->index; - int err, prio; - bool huge; + u32 index = req->pps->index; + int err; err = blk_status_to_errno(req->bio.bi_status); if (err) { @@ -946,28 +945,13 @@ static int zram_writeback_complete(struct zram *zram, struct zram_wb_req *req) goto out; } - if (zram->compressed_wb) { - /* - * ZRAM_WB slots get freed, we need to preserve data required - * for read decompression. - */ - size = get_slot_size(zram, index); - prio = get_slot_comp_priority(zram, index); - huge = test_slot_flag(zram, index, ZRAM_HUGE); - } - - slot_free(zram, index); - set_slot_flag(zram, index, ZRAM_WB); + clear_slot_flag(zram, index, ZRAM_IDLE); + if (test_slot_flag(zram, index, ZRAM_HUGE)) + atomic64_dec(&zram->stats.huge_pages); + atomic64_sub(get_slot_size(zram, index), &zram->stats.compr_data_size); + zs_free(zram->mem_pool, get_slot_handle(zram, index)); set_slot_handle(zram, index, req->blk_idx); - - if (zram->compressed_wb) { - if (huge) - set_slot_flag(zram, index, ZRAM_HUGE); - set_slot_size(zram, index, size); - set_slot_comp_priority(zram, index, prio); - } - - atomic64_inc(&zram->stats.pages_stored); + set_slot_flag(zram, index, ZRAM_WB); out: slot_unlock(zram, index); @@ -2010,8 +1994,13 @@ static void slot_free(struct zram *zram, u32 index) set_slot_comp_priority(zram, index, 0); if (test_slot_flag(zram, index, ZRAM_HUGE)) { + /* + * Writeback completion decrements ->huge_pages but keeps + * ZRAM_HUGE flag for deferred decompression path. + */ + if (!test_slot_flag(zram, index, ZRAM_WB)) + atomic64_dec(&zram->stats.huge_pages); clear_slot_flag(zram, index, ZRAM_HUGE); - atomic64_dec(&zram->stats.huge_pages); } if (test_slot_flag(zram, index, ZRAM_WB)) { From ec69c9e88315c4be70c283f18c2ff130da6320b5 Mon Sep 17 00:00:00 2001 From: Mikko Perttunen Date: Tue, 3 Mar 2026 13:32:11 +0900 Subject: [PATCH 276/352] i2c: tegra: Don't mark devices with pins as IRQ safe I2C devices with associated pinctrl states (DPAUX I2C controllers) will change pinctrl state during runtime PM. This requires taking a mutex, so these devices cannot be marked as IRQ safe. Add PINCTRL as dependency to avoid build errors. Signed-off-by: Mikko Perttunen Reported-by: Russell King Link: https://lore.kernel.org/all/E1vsNBv-00000009nfA-27ZK@rmk-PC.armlinux.org.uk/ Signed-off-by: Linus Torvalds --- drivers/i2c/busses/Kconfig | 2 ++ drivers/i2c/busses/i2c-tegra.c | 5 ++++- 2 files changed, 6 insertions(+), 1 deletion(-) diff --git a/drivers/i2c/busses/Kconfig b/drivers/i2c/busses/Kconfig index e11d50750e63..7cb6b9b864a7 100644 --- a/drivers/i2c/busses/Kconfig +++ b/drivers/i2c/busses/Kconfig @@ -1213,6 +1213,8 @@ config I2C_TEGRA tristate "NVIDIA Tegra internal I2C controller" depends on ARCH_TEGRA || (COMPILE_TEST && (ARC || ARM || ARM64 || M68K || RISCV || SUPERH || SPARC)) # COMPILE_TEST needs architectures with readsX()/writesX() primitives + depends on PINCTRL + # ARCH_TEGRA implies PINCTRL, but the COMPILE_TEST side doesn't. help If you say yes to this option, support will be included for the I2C controller embedded in NVIDIA Tegra SOCs diff --git a/drivers/i2c/busses/i2c-tegra.c b/drivers/i2c/busses/i2c-tegra.c index bec619b9af4e..4eaeb395d5db 100644 --- a/drivers/i2c/busses/i2c-tegra.c +++ b/drivers/i2c/busses/i2c-tegra.c @@ -2047,8 +2047,11 @@ static int tegra_i2c_probe(struct platform_device *pdev) * * VI I2C device shouldn't be marked as IRQ-safe because VI I2C won't * be used for atomic transfers. ACPI device is not IRQ safe also. + * + * Devices with pinctrl states cannot be marked IRQ-safe as the pinctrl + * state transitions during runtime PM require mutexes. */ - if (!IS_VI(i2c_dev) && !has_acpi_companion(i2c_dev->dev)) + if (!IS_VI(i2c_dev) && !has_acpi_companion(i2c_dev->dev) && !i2c_dev->dev->pins) pm_runtime_irq_safe(i2c_dev->dev); pm_runtime_enable(i2c_dev->dev); From 4ddd7588fae6175e748cff22c79faafb4d455d42 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 18 Mar 2026 15:42:16 -0300 Subject: [PATCH 277/352] tools arch x86: Sync the msr-index.h copy with the kernel sources To pick up the changes from these csets: 9073428bb204d921 ("x86/sev: Allow IBPB-on-Entry feature for SNP guests") That cause no changes to tooling as it doesn't include a new MSR to be captured by the tools/perf/trace/beauty/tracepoints/x86_msr.sh script. Just silences this perf build warning: Warning: Kernel ABI header differences: diff -u tools/arch/x86/include/asm/msr-index.h arch/x86/include/asm/msr-index.h Cc: Borislav Petkov (AMD) Cc: Kim Phillips Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/include/asm/msr-index.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/tools/arch/x86/include/asm/msr-index.h b/tools/arch/x86/include/asm/msr-index.h index da5275d8eda6..6673601246b3 100644 --- a/tools/arch/x86/include/asm/msr-index.h +++ b/tools/arch/x86/include/asm/msr-index.h @@ -740,7 +740,10 @@ #define MSR_AMD64_SNP_SMT_PROT BIT_ULL(MSR_AMD64_SNP_SMT_PROT_BIT) #define MSR_AMD64_SNP_SECURE_AVIC_BIT 18 #define MSR_AMD64_SNP_SECURE_AVIC BIT_ULL(MSR_AMD64_SNP_SECURE_AVIC_BIT) -#define MSR_AMD64_SNP_RESV_BIT 19 +#define MSR_AMD64_SNP_RESERVED_BITS19_22 GENMASK_ULL(22, 19) +#define MSR_AMD64_SNP_IBPB_ON_ENTRY_BIT 23 +#define MSR_AMD64_SNP_IBPB_ON_ENTRY BIT_ULL(MSR_AMD64_SNP_IBPB_ON_ENTRY_BIT) +#define MSR_AMD64_SNP_RESV_BIT 24 #define MSR_AMD64_SNP_RESERVED_MASK GENMASK_ULL(63, MSR_AMD64_SNP_RESV_BIT) #define MSR_AMD64_SAVIC_CONTROL 0xc0010138 #define MSR_AMD64_SAVIC_EN_BIT 0 From 3c71ae8ec9adde96f5ecfcbeef62ccf1d420f83f Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 18 Mar 2026 15:47:05 -0300 Subject: [PATCH 278/352] tools headers UAPI: Sync linux/kvm.h with the kernel sources To pick the changes in: da142f3d373a6dda ("KVM: Remove subtle "struct kvm_stats_desc" pseudo-overlay") That just rebuilds perf, as these patches don't add any new KVM ioctl to be harvested for the 'perf trace' ioctl syscall argument beautifiers. This addresses this perf build warning: Warning: Kernel ABI header differences: diff -u tools/include/uapi/linux/kvm.h include/uapi/linux/kvm.h Please see tools/include/uapi/README for further details. Cc: Sean Christopherson Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/uapi/linux/kvm.h | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/tools/include/uapi/linux/kvm.h b/tools/include/uapi/linux/kvm.h index 65500f5db379..80364d4dbebb 100644 --- a/tools/include/uapi/linux/kvm.h +++ b/tools/include/uapi/linux/kvm.h @@ -14,6 +14,10 @@ #include #include +#ifdef __KERNEL__ +#include +#endif + #define KVM_API_VERSION 12 /* @@ -1601,7 +1605,11 @@ struct kvm_stats_desc { __u16 size; __u32 offset; __u32 bucket_size; +#ifdef __KERNEL__ + char name[KVM_STATS_NAME_SIZE]; +#else char name[]; +#endif }; #define KVM_GET_STATS_FD _IO(KVMIO, 0xce) From 0a8b2a0857ede906f7b74a435b11778336770bea Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Wed, 18 Mar 2026 15:48:54 -0300 Subject: [PATCH 279/352] tools headers UAPI: Sync x86's asm/kvm.h with the kernel sources To pick the changes in: e2ffe85b6d2bb778 ("KVM: x86: Introduce KVM_X86_QUIRK_VMCS12_ALLOW_FREEZE_IN_SMM") That just rebuilds kvm-stat.c on x86, no change in functionality. This silences these perf build warning: Warning: Kernel ABI header differences: diff -u tools/arch/x86/include/uapi/asm/kvm.h arch/x86/include/uapi/asm/kvm.h Please see tools/include/uapi/README for further details. Cc: Jim Mattson Cc: Paolo Bonzini Signed-off-by: Arnaldo Carvalho de Melo --- tools/arch/x86/include/uapi/asm/kvm.h | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/arch/x86/include/uapi/asm/kvm.h b/tools/arch/x86/include/uapi/asm/kvm.h index 846a63215ce1..0d4538fa6c31 100644 --- a/tools/arch/x86/include/uapi/asm/kvm.h +++ b/tools/arch/x86/include/uapi/asm/kvm.h @@ -476,6 +476,7 @@ struct kvm_sync_regs { #define KVM_X86_QUIRK_SLOT_ZAP_ALL (1 << 7) #define KVM_X86_QUIRK_STUFF_FEATURE_MSRS (1 << 8) #define KVM_X86_QUIRK_IGNORE_GUEST_PAT (1 << 9) +#define KVM_X86_QUIRK_VMCS12_ALLOW_FREEZE_IN_SMM (1 << 10) #define KVM_STATE_NESTED_FORMAT_VMX 0 #define KVM_STATE_NESTED_FORMAT_SVM 1 From 493ad070cbcb0d62deed877d90e80e554cac7f01 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Sun, 22 Mar 2026 18:33:28 -0300 Subject: [PATCH 280/352] tools headers: Synchronize linux/build_bug.h with the kernel sources To pick up the changes in: 6ffd853b0b10e1e2 ("build_bug.h: correct function parameters names in kernel-doc") That just add some comments, addressing this perf tools build warning: Warning: Kernel ABI header differences: diff -u tools/include/linux/build_bug.h include/linux/build_bug.h Please take a look at tools/include/uapi/README for further info on this synchronization process. Cc: Andrew Morton Cc: Ian Rogers Cc: Randy Dunlap Signed-off-by: Arnaldo Carvalho de Melo --- tools/include/linux/build_bug.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/tools/include/linux/build_bug.h b/tools/include/linux/build_bug.h index ab2aa97bd8ce..406923bd4846 100644 --- a/tools/include/linux/build_bug.h +++ b/tools/include/linux/build_bug.h @@ -32,7 +32,8 @@ /** * BUILD_BUG_ON_MSG - break compile if a condition is true & emit supplied * error message. - * @condition: the condition which the compiler should know is false. + * @cond: the condition which the compiler should know is false. + * @msg: build-time error message * * See BUILD_BUG_ON for description. */ @@ -60,6 +61,7 @@ /** * static_assert - check integer constant expression at build time + * @expr: expression to be checked * * static_assert() is a wrapper for the C11 _Static_assert, with a * little macro magic to make the message optional (defaulting to the From c369299895a591d96745d6492d4888259b004a9e Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 22 Mar 2026 14:42:17 -0700 Subject: [PATCH 281/352] Linux 7.0-rc5 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index c9b7bee102e8..2294decf0afc 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 7 PATCHLEVEL = 0 SUBLEVEL = 0 -EXTRAVERSION = -rc4 +EXTRAVERSION = -rc5 NAME = Baby Opossum Posse # *DOCUMENTATION* From 93702ed64fd0c433289b3846c2cefced4a88a043 Mon Sep 17 00:00:00 2001 From: Denis Benato Date: Wed, 4 Mar 2026 15:11:02 +0100 Subject: [PATCH 282/352] MAINTAINERS: change email address of Denis Benato MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit I have been using a linux.dev email since that is hugely better than gmail. Signed-off-by: Denis Benato Signed-off-by: Denis Benato Link: https://patch.msgid.link/20260304141102.63732-1-denis.benato@linux.dev Signed-off-by: Ilpo Järvinen --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 55af015174a5..27377c282e67 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -3989,7 +3989,7 @@ F: drivers/hwmon/asus-ec-sensors.c ASUS NOTEBOOKS AND EEEPC ACPI/WMI EXTRAS DRIVERS M: Corentin Chary M: Luke D. Jones -M: Denis Benato +M: Denis Benato L: platform-driver-x86@vger.kernel.org S: Maintained W: https://asus-linux.org/ From e0836f48e19147a85ae652335e517b0385a32fd3 Mon Sep 17 00:00:00 2001 From: Denis Benato Date: Mon, 9 Mar 2026 19:35:57 +0100 Subject: [PATCH 283/352] platform/x86: asus-armoury: add support for GA503QM MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add TDP data for laptop model GA503QM. Signed-off-by: Denis Benato Link: https://patch.msgid.link/20260309183559.433555-2-denis.benato@linux.dev Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/asus-armoury.h | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/platform/x86/asus-armoury.h b/drivers/platform/x86/asus-armoury.h index 208f6fe16168..1979b5c47bc9 100644 --- a/drivers/platform/x86/asus-armoury.h +++ b/drivers/platform/x86/asus-armoury.h @@ -1080,6 +1080,20 @@ static const struct dmi_system_id power_limits[] = { .requires_fan_curve = true, }, }, + { + .matches = { + DMI_MATCH(DMI_BOARD_NAME, "GA503QM"), + }, + .driver_data = &(struct power_data) { + .ac_data = &(struct power_limits) { + .ppt_pl1_spl_min = 15, + .ppt_pl1_spl_def = 35, + .ppt_pl1_spl_max = 80, + .ppt_pl2_sppt_min = 65, + .ppt_pl2_sppt_max = 80, + }, + }, + }, { .matches = { DMI_MATCH(DMI_BOARD_NAME, "GA503QR"), From d2723918d51b238b42efcaac553697aa84f15232 Mon Sep 17 00:00:00 2001 From: Denis Benato Date: Mon, 9 Mar 2026 19:35:58 +0100 Subject: [PATCH 284/352] platform/x86: asus-armoury: add support for G614FP MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add TDP data for laptop model G614FP. Signed-off-by: Denis Benato Link: https://patch.msgid.link/20260309183559.433555-3-denis.benato@linux.dev Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/asus-armoury.h | 34 +++++++++++++++++++++++++++++ 1 file changed, 34 insertions(+) diff --git a/drivers/platform/x86/asus-armoury.h b/drivers/platform/x86/asus-armoury.h index 1979b5c47bc9..230d614a00e4 100644 --- a/drivers/platform/x86/asus-armoury.h +++ b/drivers/platform/x86/asus-armoury.h @@ -1610,6 +1610,40 @@ static const struct dmi_system_id power_limits[] = { .requires_fan_curve = true, }, }, + { + .matches = { + DMI_MATCH(DMI_BOARD_NAME, "G614FP"), + }, + .driver_data = &(struct power_data) { + .ac_data = &(struct power_limits) { + .ppt_pl1_spl_min = 30, + .ppt_pl1_spl_max = 120, + .ppt_pl2_sppt_min = 65, + .ppt_pl2_sppt_def = 140, + .ppt_pl2_sppt_max = 165, + .ppt_pl3_fppt_min = 65, + .ppt_pl3_fppt_def = 140, + .ppt_pl3_fppt_max = 165, + .nv_temp_target_min = 75, + .nv_temp_target_max = 87, + .nv_dynamic_boost_min = 5, + .nv_dynamic_boost_max = 15, + .nv_tgp_min = 50, + .nv_tgp_max = 100, + }, + .dc_data = &(struct power_limits) { + .ppt_pl1_spl_min = 25, + .ppt_pl1_spl_max = 65, + .ppt_pl2_sppt_min = 25, + .ppt_pl2_sppt_max = 65, + .ppt_pl3_fppt_min = 35, + .ppt_pl3_fppt_max = 75, + .nv_temp_target_min = 75, + .nv_temp_target_max = 87, + }, + .requires_fan_curve = true, + }, + }, { .matches = { DMI_MATCH(DMI_BOARD_NAME, "G614J"), From 55b964dfbadc3729d3606849471eab11705f578a Mon Sep 17 00:00:00 2001 From: Raed Date: Wed, 11 Mar 2026 18:43:38 +0530 Subject: [PATCH 285/352] platform/x86: hp-wmi: Add Omen 16-xf0xxx (8BCA) support MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The HP Omen 16-xf0xxx board 8BCA uses the same Victus-S fan and thermal WMI path as other recently supported Omen/Victus boards, but it requires Omen v1 thermal profile parameters for correct platform profile behavior. Add board 8BCA to victus_s_thermal_profile_boards[] and map it to omen_v1_thermal_params. Validated on HP Omen 16-xf0xxx (board 8BCA): - /sys/firmware/acpi/platform_profile exposes low-power/balanced/performance - fan RPM reporting works (fan1_input/fan2_input) - manual fan control works through hp-wmi hwmon (pwm1/pwm1_enable) Signed-off-by: Raed Link: https://patch.msgid.link/20260311131338.965249-1-youaretalkingtoraed@gmail.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/hp/hp-wmi.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/platform/x86/hp/hp-wmi.c b/drivers/platform/x86/hp/hp-wmi.c index 68ede7e5757a..0a3fc2942ed0 100644 --- a/drivers/platform/x86/hp/hp-wmi.c +++ b/drivers/platform/x86/hp/hp-wmi.c @@ -183,6 +183,10 @@ static const struct dmi_system_id victus_s_thermal_profile_boards[] __initconst .matches = { DMI_MATCH(DMI_BOARD_NAME, "8BBE") }, .driver_data = (void *)&victus_s_thermal_params, }, + { + .matches = { DMI_MATCH(DMI_BOARD_NAME, "8BCA") }, + .driver_data = (void *)&omen_v1_thermal_params, + }, { .matches = { DMI_MATCH(DMI_BOARD_NAME, "8BCD") }, .driver_data = (void *)&omen_v1_thermal_params, From 84d29bfd1929d08f092851162a3d055a2134d043 Mon Sep 17 00:00:00 2001 From: Krishna Chomal Date: Fri, 27 Feb 2026 21:11:06 +0530 Subject: [PATCH 286/352] platform/x86: hp-wmi: Add support for Omen 16-wf1xxx (8C76) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The HP Omen 16-wf1xxx (board ID: 8C76) has the same WMI interface as other Victus S boards, but requires quirks for correctly switching thermal profile (similar to board 8C78). Add the DMI board name to victus_s_thermal_profile_boards[] table and map it to omen_v1_thermal_params. Testing on board 8C76 confirmed that platform profile is registered successfully and fan RPMs are readable and controllable. Tested-by: WJ Enderlava Reported-by: WJ Enderlava Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221149 Signed-off-by: Krishna Chomal Link: https://patch.msgid.link/20260227154106.226809-1-krishna.chomal108@gmail.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/hp/hp-wmi.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/platform/x86/hp/hp-wmi.c b/drivers/platform/x86/hp/hp-wmi.c index 0a3fc2942ed0..0e5d491fd450 100644 --- a/drivers/platform/x86/hp/hp-wmi.c +++ b/drivers/platform/x86/hp/hp-wmi.c @@ -199,6 +199,10 @@ static const struct dmi_system_id victus_s_thermal_profile_boards[] __initconst .matches = { DMI_MATCH(DMI_BOARD_NAME, "8BD5") }, .driver_data = (void *)&victus_s_thermal_params, }, + { + .matches = { DMI_MATCH(DMI_BOARD_NAME, "8C76") }, + .driver_data = (void *)&omen_v1_thermal_params, + }, { .matches = { DMI_MATCH(DMI_BOARD_NAME, "8C78") }, .driver_data = (void *)&omen_v1_thermal_params, From 435da773966c80e1b2b6aea75460ef0e893fd9e9 Mon Sep 17 00:00:00 2001 From: Krishna Chomal Date: Mon, 2 Mar 2026 13:05:25 +0530 Subject: [PATCH 287/352] platform/x86: hp-wmi: Add support for Omen 16-k0xxx (8A4D) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The HP Omen 16-k0xxx (board ID: 8A4D) has the same WMI interface as other Victus S boards, but requires additional quirks for correctly switching thermal profile. Create a new quirk omen_v1_legacy_thermal_params which allows a board to use Omen V1 thermal values, but rely on the older legacy HP_OMEN_EC_THERMAL_PROFILE_OFFSET. Add the DMI board name to victus_s_thermal_profile_boards[] table and map it to the newly added quirk. Testing on board 8A4D confirmed that platform profile is registered successfully and fan RPMs are readable and controllable. Tested-by: Qinfeng Wu Reported-by: Qinfeng Wu Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221150 Signed-off-by: Krishna Chomal Link: https://patch.msgid.link/20260302073525.71037-1-krishna.chomal108@gmail.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/hp/hp-wmi.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/drivers/platform/x86/hp/hp-wmi.c b/drivers/platform/x86/hp/hp-wmi.c index 0e5d491fd450..988a0acc9622 100644 --- a/drivers/platform/x86/hp/hp-wmi.c +++ b/drivers/platform/x86/hp/hp-wmi.c @@ -120,6 +120,13 @@ static const struct thermal_profile_params omen_v1_thermal_params = { .ec_tp_offset = HP_VICTUS_S_EC_THERMAL_PROFILE_OFFSET, }; +static const struct thermal_profile_params omen_v1_legacy_thermal_params = { + .performance = HP_OMEN_V1_THERMAL_PROFILE_PERFORMANCE, + .balanced = HP_OMEN_V1_THERMAL_PROFILE_DEFAULT, + .low_power = HP_OMEN_V1_THERMAL_PROFILE_DEFAULT, + .ec_tp_offset = HP_OMEN_EC_THERMAL_PROFILE_OFFSET, +}; + /* * A generic pointer for the currently-active board's thermal profile * parameters. @@ -175,6 +182,10 @@ static const char * const victus_thermal_profile_boards[] = { /* DMI Board names of Victus 16-r and Victus 16-s laptops */ static const struct dmi_system_id victus_s_thermal_profile_boards[] __initconst = { + { + .matches = { DMI_MATCH(DMI_BOARD_NAME, "8A4D") }, + .driver_data = (void *)&omen_v1_legacy_thermal_params, + }, { .matches = { DMI_MATCH(DMI_BOARD_NAME, "8BAB") }, .driver_data = (void *)&omen_v1_thermal_params, From 9f11d9b15efb5f77e810b6dfbeb01b4650a79eae Mon Sep 17 00:00:00 2001 From: Li RongQing Date: Tue, 3 Mar 2026 02:46:35 -0500 Subject: [PATCH 288/352] platform/x86: ISST: Check HWP support before MSR access MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit On some systems, HWP can be explicitly disabled in the BIOS settings When HWP is disabled by firmware, the HWP CPUID bit is not set, and attempting to read MSR_PM_ENABLE will result in a General Protection (GP) fault. unchecked MSR access error: RDMSR from 0x770 at rIP: 0xffffffffc33db92e (disable_dynamic_sst_features+0xe/0x50 [isst_tpmi_core]) Call Trace: ? ex_handler_msr+0xf6/0x150 ? fixup_exception+0x1ad/0x340 ? gp_try_fixup_and_notify+0x1e/0xb0 ? exc_general_protection+0xc9/0x390 ? terminate_walk+0x64/0x100 ? asm_exc_general_protection+0x22/0x30 ? disable_dynamic_sst_features+0xe/0x50 [isst_tpmi_core] isst_if_def_ioctl+0xece/0x1050 [isst_tpmi_core] ? ioctl_has_perm.constprop.42+0xe0/0x130 isst_if_def_ioctl+0x10d/0x1a0 [isst_if_common] __se_sys_ioctl+0x86/0xc0 do_syscall_64+0x8a/0x100 entry_SYSCALL_64_after_hwframe+0x78/0xe2 RIP: 0033:0x7f36eaef54a7 Add a check for X86_FEATURE_HWP before accessing the MSR. If HWP is not available, return true safely. Fixes: 12a7d2cb811d ("platform/x86: ISST: Add SST-CP support via TPMI") Signed-off-by: Li RongQing Acked-by: Srinivas Pandruvada Link: https://patch.msgid.link/20260303074635.2218-1-lirongqing@baidu.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c b/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c index b8cdaa233ea9..6f46cbb2898f 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c +++ b/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c @@ -558,6 +558,9 @@ static bool disable_dynamic_sst_features(void) { u64 value; + if (!static_cpu_has(X86_FEATURE_HWP)) + return true; + rdmsrq(MSR_PM_ENABLE, value); return !(value & 0x1); } From 5a3955f3602950d1888df743a5b1889e43b5cb60 Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Fri, 13 Mar 2026 14:06:34 -0700 Subject: [PATCH 289/352] platform/x86: lenovo: wmi-gamezone: Drop gz_chain_head MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The gz_chain_head variable has been unused since the driver's initial addition to the tree. Its use was eliminated between v3 and v4 during development but due to the reference of gz_chain_head's wait_list member, the compiler could not warn that it was unused. After a (tip) commit ("locking/rwsem: Remove the list_head from struct rw_semaphore"), which removed a reference to the variable passed to __RWSEM_INITIALIZER(), certain configurations show an unused variable warning from the Lenovo wmi-gamezone driver: drivers/platform/x86/lenovo/wmi-gamezone.c:34:31: warning: 'gz_chain_head' defined but not used [-Wunused-variable] 34 | static BLOCKING_NOTIFIER_HEAD(gz_chain_head); | ^~~~~~~~~~~~~ include/linux/notifier.h:119:39: note: in definition of macro 'BLOCKING_NOTIFIER_HEAD' 119 | struct blocking_notifier_head name = \ | ^~~~ Remove the variable to prevent the warning from showing up. Fixes: 22024ac5366f ("platform/x86: Add Lenovo Gamezone WMI Driver") Signed-off-by: Nathan Chancellor Reviewed-by: Mark Pearson Link: https://patch.msgid.link/20260313-lenovo-wmi-gamezone-remove-gz_chain_head-v1-1-ce5231f0c6fa@kernel.org [ij: reorganized the changelog] Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/lenovo/wmi-gamezone.c | 2 -- 1 file changed, 2 deletions(-) diff --git a/drivers/platform/x86/lenovo/wmi-gamezone.c b/drivers/platform/x86/lenovo/wmi-gamezone.c index 381836d29a96..c7fe7e3c9f17 100644 --- a/drivers/platform/x86/lenovo/wmi-gamezone.c +++ b/drivers/platform/x86/lenovo/wmi-gamezone.c @@ -31,8 +31,6 @@ #define LWMI_GZ_METHOD_ID_SMARTFAN_SET 44 #define LWMI_GZ_METHOD_ID_SMARTFAN_GET 45 -static BLOCKING_NOTIFIER_HEAD(gz_chain_head); - struct lwmi_gz_priv { enum thermal_mode current_mode; struct notifier_block event_nb; From 2061f7b042f88d372cca79615f8425f3564c0b40 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 10 Mar 2026 06:01:35 -0700 Subject: [PATCH 290/352] platform/olpc: olpc-xo175-ec: Fix overflow error message to print inlen MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The command length check validates inlen (> 5), but the error message incorrectly printed resp_len. Print inlen so the log reflects the actual command length. Fixes: 0c3d931b3ab9e ("Platform: OLPC: Add XO-1.75 EC driver") Signed-off-by: Alok Tiwari Acked-by: Lubomir Rintel Reviewed-by: Randy Dunlap Link: https://patch.msgid.link/20260310130138.700687-1-alok.a.tiwari@oracle.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/olpc/olpc-xo175-ec.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/olpc/olpc-xo175-ec.c b/drivers/platform/olpc/olpc-xo175-ec.c index fa7b3bda688a..bee271a4fda1 100644 --- a/drivers/platform/olpc/olpc-xo175-ec.c +++ b/drivers/platform/olpc/olpc-xo175-ec.c @@ -482,7 +482,7 @@ static int olpc_xo175_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *resp, dev_dbg(dev, "CMD %x, %zd bytes expected\n", cmd, resp_len); if (inlen > 5) { - dev_err(dev, "command len %zd too big!\n", resp_len); + dev_err(dev, "command len %zd too big!\n", inlen); return -EOVERFLOW; } From 7ff61be5a4844d4aaa31732a5831dd5cd6136448 Mon Sep 17 00:00:00 2001 From: Alok Tiwari Date: Tue, 10 Mar 2026 05:53:05 -0700 Subject: [PATCH 291/352] platform/x86/amd/hsmp: Fix typo in error message MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix a typo in the HSMP error message where "tmeout" should be "timeout". Signed-off-by: Alok Tiwari Link: https://patch.msgid.link/20260310125307.700108-1-alok.a.tiwari@oracle.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/amd/hsmp/hsmp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/amd/hsmp/hsmp.c b/drivers/platform/x86/amd/hsmp/hsmp.c index 19f82c1d3090..631ffc0978d1 100644 --- a/drivers/platform/x86/amd/hsmp/hsmp.c +++ b/drivers/platform/x86/amd/hsmp/hsmp.c @@ -117,7 +117,7 @@ static int __hsmp_send_message(struct hsmp_socket *sock, struct hsmp_message *ms } if (unlikely(mbox_status == HSMP_STATUS_NOT_READY)) { - dev_err(sock->dev, "Message ID 0x%X failure : SMU tmeout (status = 0x%X)\n", + dev_err(sock->dev, "Message ID 0x%X failure : SMU timeout (status = 0x%X)\n", msg->msg_id, mbox_status); return -ETIMEDOUT; } else if (unlikely(mbox_status == HSMP_ERR_INVALID_MSG)) { From 0198d2743207d67f995cd6df89e267e1b9f5e1f1 Mon Sep 17 00:00:00 2001 From: Matthew Schwartz Date: Thu, 12 Mar 2026 14:22:46 -0700 Subject: [PATCH 292/352] platform/x86: asus-nb-wmi: add DMI quirk for ASUS ROG Flow Z13-KJP GZ302EAC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The ASUS ROG Flow Z13-KJP GZ302EAC model uses sys_vendor name ASUS rather than ASUSTeK COMPUTER INC., but it needs the same folio quirk as the other ROG Flow Z13. To keep things simple, just match on sys_vendor ASUS since it covers both. Signed-off-by: Matthew Schwartz Reviewed-by: Mario Limonciello (AMD) Reviewed-by: Denis Benato Link: https://patch.msgid.link/20260312212246.1608080-1-matthew.schwartz@linux.dev Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/asus-nb-wmi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/asus-nb-wmi.c b/drivers/platform/x86/asus-nb-wmi.c index a38a65f5c550..b4677c5bba5b 100644 --- a/drivers/platform/x86/asus-nb-wmi.c +++ b/drivers/platform/x86/asus-nb-wmi.c @@ -548,7 +548,7 @@ static const struct dmi_system_id asus_quirks[] = { .callback = dmi_matched, .ident = "ASUS ROG Z13", .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "ASUSTeK COMPUTER INC."), + DMI_MATCH(DMI_SYS_VENDOR, "ASUS"), DMI_MATCH(DMI_PRODUCT_NAME, "ROG Flow Z13"), }, .driver_data = &quirk_asus_z13, From 8a243d972aff2e0b0141048a3feaf3b13d78985c Mon Sep 17 00:00:00 2001 From: Matthew Schwartz Date: Thu, 12 Mar 2026 17:49:39 -0700 Subject: [PATCH 293/352] platform/x86: asus-armoury: add support for GZ302EA and GZ302EAC MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add TDP data for tablet models GZ302EA and GZ302EAC. Signed-off-by: Matthew Schwartz Reviewed-by: Denis Benato Link: https://patch.msgid.link/20260313004939.4103835-1-matthew.schwartz@linux.dev Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/asus-armoury.h | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+) diff --git a/drivers/platform/x86/asus-armoury.h b/drivers/platform/x86/asus-armoury.h index 230d614a00e4..569743746347 100644 --- a/drivers/platform/x86/asus-armoury.h +++ b/drivers/platform/x86/asus-armoury.h @@ -1532,6 +1532,35 @@ static const struct dmi_system_id power_limits[] = { }, }, }, + { + .matches = { + DMI_MATCH(DMI_BOARD_NAME, "GZ302EA"), + }, + .driver_data = &(struct power_data) { + .ac_data = &(struct power_limits) { + .ppt_pl1_spl_min = 28, + .ppt_pl1_spl_def = 60, + .ppt_pl1_spl_max = 80, + .ppt_pl2_sppt_min = 32, + .ppt_pl2_sppt_def = 75, + .ppt_pl2_sppt_max = 92, + .ppt_pl3_fppt_min = 45, + .ppt_pl3_fppt_def = 86, + .ppt_pl3_fppt_max = 93, + }, + .dc_data = &(struct power_limits) { + .ppt_pl1_spl_min = 28, + .ppt_pl1_spl_def = 45, + .ppt_pl1_spl_max = 80, + .ppt_pl2_sppt_min = 32, + .ppt_pl2_sppt_def = 52, + .ppt_pl2_sppt_max = 92, + .ppt_pl3_fppt_min = 45, + .ppt_pl3_fppt_def = 71, + .ppt_pl3_fppt_max = 93, + }, + }, + }, { .matches = { DMI_MATCH(DMI_BOARD_NAME, "G513I"), From e02ea3ae8ee40d5835a845884c7b161a27c10bcb Mon Sep 17 00:00:00 2001 From: David McFarland Date: Thu, 5 Feb 2026 19:16:24 -0400 Subject: [PATCH 294/352] platform/x86: intel-hid: disable wakeup_mode during hibernation MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Add a freeze handler which clears wakeup_mode. This fixes aborted hibernation on Dell Precision 3880. Wakeup event detected during hibernation, rolling back This system sends power button events during hibernation, even when triggered by software. Closes: https://bugzilla.kernel.org/show_bug.cgi?id=218634 Fixes: 0c4cae1bc00d ("PM: hibernate: Avoid missing wakeup events during hibernation") Signed-off-by: David McFarland Link: https://patch.msgid.link/20260205231629.1336348-1-corngood@gmail.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/intel/hid.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/platform/x86/intel/hid.c b/drivers/platform/x86/intel/hid.c index 95c405c8bac0..2ddd8af8c1ce 100644 --- a/drivers/platform/x86/intel/hid.c +++ b/drivers/platform/x86/intel/hid.c @@ -438,6 +438,14 @@ static int intel_hid_pl_suspend_handler(struct device *device) return 0; } +static int intel_hid_pl_freeze_handler(struct device *device) +{ + struct intel_hid_priv *priv = dev_get_drvdata(device); + + priv->wakeup_mode = false; + return intel_hid_pl_suspend_handler(device); +} + static int intel_hid_pl_resume_handler(struct device *device) { intel_hid_pm_complete(device); @@ -452,7 +460,7 @@ static int intel_hid_pl_resume_handler(struct device *device) static const struct dev_pm_ops intel_hid_pl_pm_ops = { .prepare = intel_hid_pm_prepare, .complete = intel_hid_pm_complete, - .freeze = intel_hid_pl_suspend_handler, + .freeze = intel_hid_pl_freeze_handler, .thaw = intel_hid_pl_resume_handler, .restore = intel_hid_pl_resume_handler, .suspend = intel_hid_pl_suspend_handler, From 42e06688c6cb7217578133bed67e9e3c1f061a58 Mon Sep 17 00:00:00 2001 From: Andre Przywara Date: Fri, 20 Mar 2026 18:52:30 +0100 Subject: [PATCH 295/352] pinctrl: sunxi: pass down flags to pinctrl routines Recent changes in the Allwinner pinctrl/GPIO IP made us add some quirks, which the new SoCs (A523 family) need to use. We now have a comfortable "flags" field on the per-SoC setup side, to tag those quirks we need, but were translating those flag bits into specific fields for runtime use, in the init routine. Now the newest Allwinner GPIO IP adds even more quirks and exceptions, some of a boolean nature. To avoid inventing various new boolean flags for the runtime struct sunxi_pinctrl, let's just directly pass on the flags variable used by the setup code, so runtime can check for those various quirk bits directly. Rename the "variant" member to "flags", and directly copy the value from the setup code into there. Move the variant masking from the init routine to the functions which actually use the "variant" value. This mostly paves the way for the new A733 IP generation, which needs more quirks to be checked at runtime. Reviewed-by: Chen-Yu Tsai Signed-off-by: Andre Przywara Signed-off-by: Michal Piekos Signed-off-by: Linus Walleij --- drivers/pinctrl/sunxi/pinctrl-sunxi.c | 23 ++++++++++++++--------- drivers/pinctrl/sunxi/pinctrl-sunxi.h | 2 +- 2 files changed, 15 insertions(+), 10 deletions(-) diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.c b/drivers/pinctrl/sunxi/pinctrl-sunxi.c index c990b6118172..685b79fc0bf8 100644 --- a/drivers/pinctrl/sunxi/pinctrl-sunxi.c +++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.c @@ -157,6 +157,7 @@ sunxi_pinctrl_desc_find_function_by_name(struct sunxi_pinctrl *pctl, const char *pin_name, const char *func_name) { + unsigned long variant = pctl->flags & SUNXI_PINCTRL_VARIANT_MASK; int i; for (i = 0; i < pctl->desc->npins; i++) { @@ -168,7 +169,7 @@ sunxi_pinctrl_desc_find_function_by_name(struct sunxi_pinctrl *pctl, while (func->name) { if (!strcmp(func->name, func_name) && (!func->variant || - func->variant & pctl->variant)) + func->variant & variant)) return func; func++; @@ -209,6 +210,8 @@ sunxi_pinctrl_desc_find_function_by_pin_and_mux(struct sunxi_pinctrl *pctl, const u16 pin_num, const u8 muxval) { + unsigned long variant = pctl->flags & SUNXI_PINCTRL_VARIANT_MASK; + for (unsigned int i = 0; i < pctl->desc->npins; i++) { const struct sunxi_desc_pin *pin = pctl->desc->pins + i; struct sunxi_desc_function *func = pin->functions; @@ -216,7 +219,7 @@ sunxi_pinctrl_desc_find_function_by_pin_and_mux(struct sunxi_pinctrl *pctl, if (pin->pin.number != pin_num) continue; - if (pin->variant && !(pctl->variant & pin->variant)) + if (pin->variant && !(variant & pin->variant)) continue; while (func->name) { @@ -1338,6 +1341,7 @@ static int sunxi_pinctrl_add_function(struct sunxi_pinctrl *pctl, static int sunxi_pinctrl_build_state(struct platform_device *pdev) { struct sunxi_pinctrl *pctl = platform_get_drvdata(pdev); + unsigned long variant = pctl->flags & SUNXI_PINCTRL_VARIANT_MASK; void *ptr; int i; @@ -1362,7 +1366,7 @@ static int sunxi_pinctrl_build_state(struct platform_device *pdev) const struct sunxi_desc_pin *pin = pctl->desc->pins + i; struct sunxi_pinctrl_group *group = pctl->groups + pctl->ngroups; - if (pin->variant && !(pctl->variant & pin->variant)) + if (pin->variant && !(variant & pin->variant)) continue; group->name = pin->pin.name; @@ -1387,11 +1391,11 @@ static int sunxi_pinctrl_build_state(struct platform_device *pdev) const struct sunxi_desc_pin *pin = pctl->desc->pins + i; struct sunxi_desc_function *func; - if (pin->variant && !(pctl->variant & pin->variant)) + if (pin->variant && !(variant & pin->variant)) continue; for (func = pin->functions; func->name; func++) { - if (func->variant && !(pctl->variant & func->variant)) + if (func->variant && !(variant & func->variant)) continue; /* Create interrupt mapping while we're at it */ @@ -1419,14 +1423,14 @@ static int sunxi_pinctrl_build_state(struct platform_device *pdev) const struct sunxi_desc_pin *pin = pctl->desc->pins + i; struct sunxi_desc_function *func; - if (pin->variant && !(pctl->variant & pin->variant)) + if (pin->variant && !(variant & pin->variant)) continue; for (func = pin->functions; func->name; func++) { struct sunxi_pinctrl_function *func_item; const char **func_grp; - if (func->variant && !(pctl->variant & func->variant)) + if (func->variant && !(variant & func->variant)) continue; func_item = sunxi_pinctrl_find_function_by_name(pctl, @@ -1568,7 +1572,7 @@ int sunxi_pinctrl_init_with_flags(struct platform_device *pdev, pctl->dev = &pdev->dev; pctl->desc = desc; - pctl->variant = flags & SUNXI_PINCTRL_VARIANT_MASK; + pctl->flags = flags; if (flags & SUNXI_PINCTRL_NEW_REG_LAYOUT) { pctl->bank_mem_size = D1_BANK_MEM_SIZE; pctl->pull_regs_offset = D1_PULL_REGS_OFFSET; @@ -1604,8 +1608,9 @@ int sunxi_pinctrl_init_with_flags(struct platform_device *pdev, for (i = 0, pin_idx = 0; i < pctl->desc->npins; i++) { const struct sunxi_desc_pin *pin = pctl->desc->pins + i; + unsigned long variant = pctl->flags & SUNXI_PINCTRL_VARIANT_MASK; - if (pin->variant && !(pctl->variant & pin->variant)) + if (pin->variant && !(variant & pin->variant)) continue; pins[pin_idx++] = pin->pin; diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.h b/drivers/pinctrl/sunxi/pinctrl-sunxi.h index ad26e4de16a8..22bffac1c3f0 100644 --- a/drivers/pinctrl/sunxi/pinctrl-sunxi.h +++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.h @@ -174,7 +174,7 @@ struct sunxi_pinctrl { unsigned *irq_array; raw_spinlock_t lock; struct pinctrl_dev *pctl_dev; - unsigned long variant; + unsigned long flags; u32 bank_mem_size; u32 pull_regs_offset; u32 dlevel_field_width; From 70f8915ea4e909826306a8567c7fa46959e278db Mon Sep 17 00:00:00 2001 From: Michal Piekos Date: Fri, 20 Mar 2026 18:52:31 +0100 Subject: [PATCH 296/352] pinctrl: sunxi: fix gpiochip_lock_as_irq() failure when pinmux is unknown Fixes kernel hang during boot due to inability to set up IRQ on AXP313a. The issue is caused by gpiochip_lock_as_irq() which is failing when gpio is in uninitialized state. Solution is to set pinmux to GPIO INPUT in sunxi_pinctrl_irq_request_resources() if it wasn't initialized earlier. Tested on Orange Pi Zero 3. Fixes: 01e10d0272b9 ("pinctrl: sunxi: Implement gpiochip::get_direction()") Reviewed-by: Andre Przywara Reviewed-by: Chen-Yu Tsai Signed-off-by: Michal Piekos Signed-off-by: Linus Walleij --- drivers/pinctrl/sunxi/pinctrl-sunxi.c | 20 ++++++++++++++++++-- drivers/pinctrl/sunxi/pinctrl-sunxi.h | 2 ++ 2 files changed, 20 insertions(+), 2 deletions(-) diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.c b/drivers/pinctrl/sunxi/pinctrl-sunxi.c index 685b79fc0bf8..d3042e0c9712 100644 --- a/drivers/pinctrl/sunxi/pinctrl-sunxi.c +++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.c @@ -1092,6 +1092,9 @@ static int sunxi_pinctrl_irq_request_resources(struct irq_data *d) { struct sunxi_pinctrl *pctl = irq_data_get_irq_chip_data(d); struct sunxi_desc_function *func; + unsigned int offset; + u32 reg, shift, mask; + u8 disabled_mux, muxval; int ret; func = sunxi_pinctrl_desc_find_function_by_pin(pctl, @@ -1099,8 +1102,21 @@ static int sunxi_pinctrl_irq_request_resources(struct irq_data *d) if (!func) return -EINVAL; - ret = gpiochip_lock_as_irq(pctl->chip, - pctl->irq_array[d->hwirq] - pctl->desc->pin_base); + offset = pctl->irq_array[d->hwirq] - pctl->desc->pin_base; + sunxi_mux_reg(pctl, offset, ®, &shift, &mask); + muxval = (readl(pctl->membase + reg) & mask) >> shift; + + /* Change muxing to GPIO INPUT mode if at reset value */ + if (pctl->flags & SUNXI_PINCTRL_NEW_REG_LAYOUT) + disabled_mux = SUN4I_FUNC_DISABLED_NEW; + else + disabled_mux = SUN4I_FUNC_DISABLED_OLD; + + if (muxval == disabled_mux) + sunxi_pmx_set(pctl->pctl_dev, pctl->irq_array[d->hwirq], + SUN4I_FUNC_INPUT); + + ret = gpiochip_lock_as_irq(pctl->chip, offset); if (ret) { dev_err(pctl->dev, "unable to lock HW IRQ %lu for IRQ\n", irqd_to_hwirq(d)); diff --git a/drivers/pinctrl/sunxi/pinctrl-sunxi.h b/drivers/pinctrl/sunxi/pinctrl-sunxi.h index 22bffac1c3f0..0daf7600e2fb 100644 --- a/drivers/pinctrl/sunxi/pinctrl-sunxi.h +++ b/drivers/pinctrl/sunxi/pinctrl-sunxi.h @@ -86,6 +86,8 @@ #define SUN4I_FUNC_INPUT 0 #define SUN4I_FUNC_IRQ 6 +#define SUN4I_FUNC_DISABLED_OLD 7 +#define SUN4I_FUNC_DISABLED_NEW 15 #define SUNXI_PINCTRL_VARIANT_MASK GENMASK(7, 0) #define SUNXI_PINCTRL_NEW_REG_LAYOUT BIT(8) From 38dfd294e24c0f397413799c2e5633aedb2058bf Mon Sep 17 00:00:00 2001 From: Muhammad Usama Anjum Date: Tue, 10 Mar 2026 17:17:39 +0000 Subject: [PATCH 297/352] mailmap: update email address for Muhammad Usama Anjum Add updated email address. Link: https://lkml.kernel.org/r/20260310171757.3970390-1-usama.anjum@arm.com Signed-off-by: Muhammad Usama Anjum Cc: Arnd Bergmann Cc: Carlos Bilbao Cc: Hans Verkuil Cc: Jakub Kacinski Cc: Martin Kepplinger Cc: Shannon Nelson Signed-off-by: Andrew Morton --- .mailmap | 1 + 1 file changed, 1 insertion(+) diff --git a/.mailmap b/.mailmap index 40b4db2b2d60..7d14504daf24 100644 --- a/.mailmap +++ b/.mailmap @@ -587,6 +587,7 @@ Morten Welinder Morten Welinder Morten Welinder Morten Welinder +Muhammad Usama Anjum Mukesh Ojha Muna Sinada Murali Nalajala From 631c1111501f34980649242751e93cfdadfd1f1c Mon Sep 17 00:00:00 2001 From: "Lorenzo Stoakes (Oracle)" Date: Mon, 16 Mar 2026 14:01:22 +0000 Subject: [PATCH 298/352] mm/zswap: add missing kunmap_local() Commit e2c3b6b21c77 ("mm: zswap: use SG list decompression APIs from zsmalloc") updated zswap_decompress() to use the scatterwalk API to copy data for uncompressed pages. In doing so, it mapped kernel memory locally for 32-bit kernels using kmap_local_folio(), however it never unmapped this memory. This resulted in the linked syzbot report where a BUG_ON() is triggered due to leaking the kmap slot. This patch fixes the issue by explicitly unmapping the established kmap. Also, add flush_dcache_folio() after the kunmap_local() call I had assumed that a new folio here combined with the flush that is done at the point of setting the PTE would suffice, but it doesn't seem that's actually the case, as update_mmu_cache() will in many archtectures only actually flush entries where a dcache flush was done on a range previously. I had also wondered whether kunmap_local() might suffice, but it doesn't seem to be the case. Some arches do seem to actually dcache flush on unmap, parisc does it if CONFIG_HIGHMEM is not set by setting ARCH_HAS_FLUSH_ON_KUNMAP and calling kunmap_flush_on_unmap() from __kunmap_local(), otherwise non-CONFIG_HIGHMEM callers do nothing here. Otherwise arch_kmap_local_pre_unmap() is called which does: * sparc - flush_cache_all() * arm - if VIVT, __cpuc_flush_dcache_area() * otherwise - nothing Also arch_kmap_local_post_unmap() is called which does: * arm - local_flush_tlb_kernel_page() * csky - kmap_flush_tlb() * microblaze, ppc - local_flush_tlb_page() * mips - local_flush_tlb_one() * sparc - flush_tlb_all() (again) * x86 - arch_flush_lazy_mmu_mode() * otherwise - nothing But this is only if it's high memory, and doesn't cover all architectures, so is presumably intended to handle other cache consistency concerns. In any case, VIPT is problematic here whether low or high memory (in spite of what the documentation claims, see [0] - 'the kernel did write to a page that is in the page cache page and / or in high memory'), because dirty cache lines may exist at the set indexed by the kernel direct mapping, which won't exist in the set indexed by any subsequent userland mapping, meaning userland might read stale data from L2 cache. Even if the documentation is correct and low memory is fine not to be flushed here, we can't be sure as to whether the memory is low or high (kmap_local_folio() will be a no-op if low), and this call should be harmless if it is low. VIVT would require more work if the memory were shared and already mapped, but this isn't the case here, and would anyway be handled by the dcache flush call. In any case, we definitely need this flush as far as I can tell. And we should probably consider updating the documentation unless it turns out there's somehow dcache synchronisation that happens for low memory/64-bit kernels elsewhere? [ljs@kernel.org: add flush_dcache_folio() after the kunmap_local() call] Link: https://lkml.kernel.org/r/13e09a99-181f-45ac-a18d-057faf94bccb@lucifer.local Link: https://lkml.kernel.org/r/20260316140122.339697-1-ljs@kernel.org Link: https://docs.kernel.org/core-api/cachetlb.html [0] Fixes: e2c3b6b21c77 ("mm: zswap: use SG list decompression APIs from zsmalloc") Signed-off-by: Lorenzo Stoakes (Oracle) Reported-by: syzbot+fe426bef95363177631d@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/69b75e2c.050a0220.12d28.015a.GAE@google.com Acked-by: Yosry Ahmed Acked-by: Johannes Weiner Reviewed-by: SeongJae Park Acked-by: Yosry Ahmed Acked-by: Nhat Pham Cc: Chengming Zhou Signed-off-by: Andrew Morton --- mm/zswap.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/mm/zswap.c b/mm/zswap.c index e6ec3295bdb0..16b2ef7223e1 100644 --- a/mm/zswap.c +++ b/mm/zswap.c @@ -942,9 +942,15 @@ static bool zswap_decompress(struct zswap_entry *entry, struct folio *folio) /* zswap entries of length PAGE_SIZE are not compressed. */ if (entry->length == PAGE_SIZE) { + void *dst; + WARN_ON_ONCE(input->length != PAGE_SIZE); - memcpy_from_sglist(kmap_local_folio(folio, 0), input, 0, PAGE_SIZE); + + dst = kmap_local_folio(folio, 0); + memcpy_from_sglist(dst, input, 0, PAGE_SIZE); dlen = PAGE_SIZE; + kunmap_local(dst); + flush_dcache_folio(folio); } else { sg_init_table(&output, 1); sg_set_folio(&output, folio, PAGE_SIZE, 0); From 84481e705ab07ed46e56587fe846af194acacafe Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Mon, 16 Mar 2026 16:51:17 -0700 Subject: [PATCH 299/352] mm/damon/stat: monitor all System RAM resources DAMON_STAT usage document (Documentation/admin-guide/mm/damon/stat.rst) says it monitors the system's entire physical memory. But, it is monitoring only the biggest System RAM resource of the system. When there are multiple System RAM resources, this results in monitoring only an unexpectedly small fraction of the physical memory. For example, suppose the system has a 500 GiB System RAM, 10 MiB non-System RAM, and 500 GiB System RAM resources in order on the physical address space. DAMON_STAT will monitor only the first 500 GiB System RAM. This situation is particularly common on NUMA systems. Select a physical address range that covers all System RAM areas of the system, to fix this issue and make it work as documented. [sj@kernel.org: return error if monitoring target region is invalid] Link: https://lkml.kernel.org/r/20260317053631.87907-1-sj@kernel.org Link: https://lkml.kernel.org/r/20260316235118.873-1-sj@kernel.org Fixes: 369c415e6073 ("mm/damon: introduce DAMON_STAT module") Signed-off-by: SeongJae Park Cc: [6.17+] Signed-off-by: Andrew Morton --- mm/damon/stat.c | 53 ++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 50 insertions(+), 3 deletions(-) diff --git a/mm/damon/stat.c b/mm/damon/stat.c index 25fb44ccf99d..cf2c5a541eee 100644 --- a/mm/damon/stat.c +++ b/mm/damon/stat.c @@ -145,12 +145,59 @@ static int damon_stat_damon_call_fn(void *data) return 0; } +struct damon_stat_system_ram_range_walk_arg { + bool walked; + struct resource res; +}; + +static int damon_stat_system_ram_walk_fn(struct resource *res, void *arg) +{ + struct damon_stat_system_ram_range_walk_arg *a = arg; + + if (!a->walked) { + a->walked = true; + a->res.start = res->start; + } + a->res.end = res->end; + return 0; +} + +static unsigned long damon_stat_res_to_core_addr(resource_size_t ra, + unsigned long addr_unit) +{ + /* + * Use div_u64() for avoiding linking errors related with __udivdi3, + * __aeabi_uldivmod, or similar problems. This should also improve the + * performance optimization (read div_u64() comment for the detail). + */ + if (sizeof(ra) == 8 && sizeof(addr_unit) == 4) + return div_u64(ra, addr_unit); + return ra / addr_unit; +} + +static int damon_stat_set_monitoring_region(struct damon_target *t, + unsigned long addr_unit, unsigned long min_region_sz) +{ + struct damon_addr_range addr_range; + struct damon_stat_system_ram_range_walk_arg arg = {}; + + walk_system_ram_res(0, -1, &arg, damon_stat_system_ram_walk_fn); + if (!arg.walked) + return -EINVAL; + addr_range.start = damon_stat_res_to_core_addr( + arg.res.start, addr_unit); + addr_range.end = damon_stat_res_to_core_addr( + arg.res.end + 1, addr_unit); + if (addr_range.end <= addr_range.start) + return -EINVAL; + return damon_set_regions(t, &addr_range, 1, min_region_sz); +} + static struct damon_ctx *damon_stat_build_ctx(void) { struct damon_ctx *ctx; struct damon_attrs attrs; struct damon_target *target; - unsigned long start = 0, end = 0; ctx = damon_new_ctx(); if (!ctx) @@ -180,8 +227,8 @@ static struct damon_ctx *damon_stat_build_ctx(void) if (!target) goto free_out; damon_add_target(ctx, target); - if (damon_set_region_biggest_system_ram_default(target, &start, &end, - ctx->min_region_sz)) + if (damon_stat_set_monitoring_region(target, ctx->addr_unit, + ctx->min_region_sz)) goto free_out; return ctx; free_out: From c7fcd269e1e07b2aa4bb37ffce7543c340796433 Mon Sep 17 00:00:00 2001 From: Michal Swiatkowski Date: Mon, 23 Feb 2026 13:51:57 +0100 Subject: [PATCH 300/352] ice: set max queues in alloc_etherdev_mqs() When allocating netdevice using alloc_etherdev_mqs() the maximum supported queues number should be passed. The vsi->alloc_txq/rxq is storing current number of queues, not the maximum ones. Use the same function for getting max Tx and Rx queues which is used during ethtool -l call to set maximum number of queues during netdev allocation. Reproduction steps: $ethtool -l $pf # says current 16, max 64 $ethtool -S $pf # fine $ethtool -L $pf combined 40 # crash [491187.472594] Call Trace: [491187.472829] [491187.473067] netif_set_xps_queue+0x26/0x40 [491187.473305] ice_vsi_cfg_txq+0x265/0x3d0 [ice] [491187.473619] ice_vsi_cfg_lan_txqs+0x68/0xa0 [ice] [491187.473918] ice_vsi_cfg_lan+0x2b/0xa0 [ice] [491187.474202] ice_vsi_open+0x71/0x170 [ice] [491187.474484] ice_vsi_recfg_qs+0x17f/0x230 [ice] [491187.474759] ? dev_get_min_mp_channel_count+0xab/0xd0 [491187.474987] ice_set_channels+0x185/0x3d0 [ice] [491187.475278] ethnl_set_channels+0x26f/0x340 Fixes: ee13aa1a2c5a ("ice: use netif_get_num_default_rss_queues()") Reviewed-by: Przemek Kitszel Signed-off-by: Michal Swiatkowski Reviewed-by: Simon Horman Reviewed-by: Paul Menzel Tested-by: Alexander Nowlin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice.h | 22 ++++++++++++++++++++ drivers/net/ethernet/intel/ice/ice_ethtool.c | 18 ---------------- drivers/net/ethernet/intel/ice/ice_main.c | 4 ++-- 3 files changed, 24 insertions(+), 20 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index 2b2b22af42be..eb3a48330cc1 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -839,6 +839,28 @@ static inline void ice_tx_xsk_pool(struct ice_vsi *vsi, u16 qid) WRITE_ONCE(ring->xsk_pool, ice_get_xp_from_qid(vsi, qid)); } +/** + * ice_get_max_txq - return the maximum number of Tx queues for in a PF + * @pf: PF structure + * + * Return: maximum number of Tx queues + */ +static inline int ice_get_max_txq(struct ice_pf *pf) +{ + return min(num_online_cpus(), pf->hw.func_caps.common_cap.num_txq); +} + +/** + * ice_get_max_rxq - return the maximum number of Rx queues for in a PF + * @pf: PF structure + * + * Return: maximum number of Rx queues + */ +static inline int ice_get_max_rxq(struct ice_pf *pf) +{ + return min(num_online_cpus(), pf->hw.func_caps.common_cap.num_rxq); +} + /** * ice_get_main_vsi - Get the PF VSI * @pf: PF instance diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index 301947d53ede..e4e1eab05bd5 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -3773,24 +3773,6 @@ ice_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info) return 0; } -/** - * ice_get_max_txq - return the maximum number of Tx queues for in a PF - * @pf: PF structure - */ -static int ice_get_max_txq(struct ice_pf *pf) -{ - return min(num_online_cpus(), pf->hw.func_caps.common_cap.num_txq); -} - -/** - * ice_get_max_rxq - return the maximum number of Rx queues for in a PF - * @pf: PF structure - */ -static int ice_get_max_rxq(struct ice_pf *pf) -{ - return min(num_online_cpus(), pf->hw.func_caps.common_cap.num_rxq); -} - /** * ice_get_combined_cnt - return the current number of combined channels * @vsi: PF VSI pointer diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index e7308e381e2f..3c36e3641b9e 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -4699,8 +4699,8 @@ static int ice_cfg_netdev(struct ice_vsi *vsi) struct net_device *netdev; u8 mac_addr[ETH_ALEN]; - netdev = alloc_etherdev_mqs(sizeof(*np), vsi->alloc_txq, - vsi->alloc_rxq); + netdev = alloc_etherdev_mqs(sizeof(*np), ice_get_max_txq(vsi->back), + ice_get_max_rxq(vsi->back)); if (!netdev) return -ENOMEM; From ad85de0fc09eb3236e73df5acb2bc257625103f5 Mon Sep 17 00:00:00 2001 From: Petr Oros Date: Thu, 12 Feb 2026 08:53:10 +0100 Subject: [PATCH 301/352] ice: fix inverted ready check for VF representors Commit 0f00a897c9fcbd ("ice: check if SF is ready in ethtool ops") refactored the VF readiness check into a generic repr->ops.ready() callback but implemented ice_repr_ready_vf() with inverted logic: return !ice_check_vf_ready_for_cfg(repr->vf); ice_check_vf_ready_for_cfg() returns 0 on success, so the negation makes ready() return non-zero when the VF is ready. All callers treat non-zero as "not ready, skip", causing ndo_get_stats64, get_drvinfo, get_strings and get_ethtool_stats to always bail out in switchdev mode. Remove the erroneous negation. The SF variant ice_repr_ready_sf() is already correct (returns !active, i.e. non-zero when not active). Fixes: 0f00a897c9fcbd ("ice: check if SF is ready in ethtool ops") Signed-off-by: Petr Oros Reviewed-by: Aleksandr Loktionov Reviewed-by: Michal Swiatkowski Tested-by: Patryk Holda Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_repr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/ice/ice_repr.c b/drivers/net/ethernet/intel/ice/ice_repr.c index 90f99443a922..68cdd04dd3aa 100644 --- a/drivers/net/ethernet/intel/ice/ice_repr.c +++ b/drivers/net/ethernet/intel/ice/ice_repr.c @@ -315,7 +315,7 @@ ice_repr_reg_netdev(struct net_device *netdev, const struct net_device_ops *ops) static int ice_repr_ready_vf(struct ice_repr *repr) { - return !ice_check_vf_ready_for_cfg(repr->vf); + return ice_check_vf_ready_for_cfg(repr->vf); } static int ice_repr_ready_sf(struct ice_repr *repr) From 2526e440df2725e7328d59b835a164826f179b93 Mon Sep 17 00:00:00 2001 From: Petr Oros Date: Thu, 12 Feb 2026 08:53:11 +0100 Subject: [PATCH 302/352] ice: use ice_update_eth_stats() for representor stats ice_repr_get_stats64() and __ice_get_ethtool_stats() call ice_update_vsi_stats() on the VF's src_vsi. This always returns early because ICE_VSI_DOWN is permanently set for VF VSIs - ice_up() is never called on them since queues are managed by iavf through virtchnl. In __ice_get_ethtool_stats() the original code called ice_update_vsi_stats() for all VSIs including representors, iterated over ice_gstrings_vsi_stats[] to populate the data, and then bailed out with an early return before the per-queue ring stats section. That early return was necessary because representor VSIs have no rings on the PF side - the rings belong to the VF driver (iavf), so accessing per-queue stats would be invalid. Move the representor handling to the top of __ice_get_ethtool_stats() and call ice_update_eth_stats() directly to read the hardware GLV_* counters. This matches ice_get_vf_stats() which already uses ice_update_eth_stats() for the same VF VSI in legacy mode. Apply the same fix to ice_repr_get_stats64(). Note that ice_gstrings_vsi_stats[] contains five software ring counters (rx_buf_failed, rx_page_failed, tx_linearize, tx_busy, tx_restart) that are always zero for representors since the PF never processes packets on VF rings. This is pre-existing behavior unchanged by this patch. Fixes: 7aae80cef7ba ("ice: add port representor ethtool ops and stats") Signed-off-by: Petr Oros Reviewed-by: Aleksandr Loktionov Tested-by: Patryk Holda Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 14 +++++++++++--- drivers/net/ethernet/intel/ice/ice_repr.c | 3 ++- 2 files changed, 13 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c index e4e1eab05bd5..e6a20af6f63d 100644 --- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -1930,6 +1930,17 @@ __ice_get_ethtool_stats(struct net_device *netdev, int i = 0; char *p; + if (ice_is_port_repr_netdev(netdev)) { + ice_update_eth_stats(vsi); + + for (j = 0; j < ICE_VSI_STATS_LEN; j++) { + p = (char *)vsi + ice_gstrings_vsi_stats[j].stat_offset; + data[i++] = (ice_gstrings_vsi_stats[j].sizeof_stat == + sizeof(u64)) ? *(u64 *)p : *(u32 *)p; + } + return; + } + ice_update_pf_stats(pf); ice_update_vsi_stats(vsi); @@ -1939,9 +1950,6 @@ __ice_get_ethtool_stats(struct net_device *netdev, sizeof(u64)) ? *(u64 *)p : *(u32 *)p; } - if (ice_is_port_repr_netdev(netdev)) - return; - /* populate per queue stats */ rcu_read_lock(); diff --git a/drivers/net/ethernet/intel/ice/ice_repr.c b/drivers/net/ethernet/intel/ice/ice_repr.c index 68cdd04dd3aa..096566c697f4 100644 --- a/drivers/net/ethernet/intel/ice/ice_repr.c +++ b/drivers/net/ethernet/intel/ice/ice_repr.c @@ -2,6 +2,7 @@ /* Copyright (C) 2019-2021, Intel Corporation. */ #include "ice.h" +#include "ice_lib.h" #include "ice_eswitch.h" #include "devlink/devlink.h" #include "devlink/port.h" @@ -67,7 +68,7 @@ ice_repr_get_stats64(struct net_device *netdev, struct rtnl_link_stats64 *stats) return; vsi = repr->src_vsi; - ice_update_vsi_stats(vsi); + ice_update_eth_stats(vsi); eth_stats = &vsi->eth_stats; stats->tx_packets = eth_stats->tx_unicast + eth_stats->tx_broadcast + From fecacfc95f195b99c71c579a472120d0b4ed65fa Mon Sep 17 00:00:00 2001 From: Kohei Enju Date: Sat, 14 Feb 2026 19:14:25 +0000 Subject: [PATCH 303/352] iavf: fix out-of-bounds writes in iavf_get_ethtool_stats() iavf incorrectly uses real_num_tx_queues for ETH_SS_STATS. Since the value could change in runtime, we should use num_tx_queues instead. Moreover iavf_get_ethtool_stats() uses num_active_queues while iavf_get_sset_count() and iavf_get_stat_strings() use real_num_tx_queues, which triggers out-of-bounds writes when we do "ethtool -L" and "ethtool -S" simultaneously [1]. For example when we change channels from 1 to 8, Thread 3 could be scheduled before Thread 2, and out-of-bounds writes could be triggered in Thread 3: Thread 1 (ethtool -L) Thread 2 (work) Thread 3 (ethtool -S) iavf_set_channels() ... iavf_alloc_queues() -> num_active_queues = 8 iavf_schedule_finish_config() iavf_get_sset_count() real_num_tx_queues: 1 -> buffer for 1 queue iavf_get_ethtool_stats() num_active_queues: 8 -> out-of-bounds! iavf_finish_config() -> real_num_tx_queues = 8 Use immutable num_tx_queues in all related functions to avoid the issue. [1] BUG: KASAN: vmalloc-out-of-bounds in iavf_add_one_ethtool_stat+0x200/0x270 Write of size 8 at addr ffffc900031c9080 by task ethtool/5800 CPU: 1 UID: 0 PID: 5800 Comm: ethtool Not tainted 6.19.0-enjuk-08403-g8137e3db7f1c #241 PREEMPT(full) Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 Call Trace: dump_stack_lvl+0x6f/0xb0 print_report+0x170/0x4f3 kasan_report+0xe1/0x180 iavf_add_one_ethtool_stat+0x200/0x270 iavf_get_ethtool_stats+0x14c/0x2e0 __dev_ethtool+0x3d0c/0x5830 dev_ethtool+0x12d/0x270 dev_ioctl+0x53c/0xe30 sock_do_ioctl+0x1a9/0x270 sock_ioctl+0x3d4/0x5e0 __x64_sys_ioctl+0x137/0x1c0 do_syscall_64+0xf3/0x690 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f7da0e6e36d ... The buggy address belongs to a 1-page vmalloc region starting at 0xffffc900031c9000 allocated at __dev_ethtool+0x3cc9/0x5830 The buggy address belongs to the physical page: page: refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88813a013de0 pfn:0x13a013 flags: 0x200000000000000(node=0|zone=2) raw: 0200000000000000 0000000000000000 dead000000000122 0000000000000000 raw: ffff88813a013de0 0000000000000000 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffffc900031c8f80: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ffffc900031c9000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >ffffc900031c9080: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ^ ffffc900031c9100: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ffffc900031c9180: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 Fixes: 64430f70ba6f ("iavf: Fix displaying queue statistics shown by ethtool") Signed-off-by: Kohei Enju Reviewed-by: Simon Horman Reviewed-by: Przemek Kitszel Reviewed-by: Paul Menzel Tested-by: Rafal Romanowski Signed-off-by: Tony Nguyen --- .../net/ethernet/intel/iavf/iavf_ethtool.c | 31 +++++++++---------- 1 file changed, 15 insertions(+), 16 deletions(-) diff --git a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c index ab67c709d5a0..1cd1f3f2930a 100644 --- a/drivers/net/ethernet/intel/iavf/iavf_ethtool.c +++ b/drivers/net/ethernet/intel/iavf/iavf_ethtool.c @@ -313,14 +313,13 @@ static int iavf_get_sset_count(struct net_device *netdev, int sset) { /* Report the maximum number queues, even if not every queue is * currently configured. Since allocation of queues is in pairs, - * use netdev->real_num_tx_queues * 2. The real_num_tx_queues is set - * at device creation and never changes. + * use netdev->num_tx_queues * 2. The num_tx_queues is set at + * device creation and never changes. */ if (sset == ETH_SS_STATS) return IAVF_STATS_LEN + - (IAVF_QUEUE_STATS_LEN * 2 * - netdev->real_num_tx_queues); + (IAVF_QUEUE_STATS_LEN * 2 * netdev->num_tx_queues); else return -EINVAL; } @@ -345,19 +344,19 @@ static void iavf_get_ethtool_stats(struct net_device *netdev, iavf_add_ethtool_stats(&data, adapter, iavf_gstrings_stats); rcu_read_lock(); - /* As num_active_queues describe both tx and rx queues, we can use - * it to iterate over rings' stats. + /* Use num_tx_queues to report stats for the maximum number of queues. + * Queues beyond num_active_queues will report zero. */ - for (i = 0; i < adapter->num_active_queues; i++) { - struct iavf_ring *ring; + for (i = 0; i < netdev->num_tx_queues; i++) { + struct iavf_ring *tx_ring = NULL, *rx_ring = NULL; - /* Tx rings stats */ - ring = &adapter->tx_rings[i]; - iavf_add_queue_stats(&data, ring); + if (i < adapter->num_active_queues) { + tx_ring = &adapter->tx_rings[i]; + rx_ring = &adapter->rx_rings[i]; + } - /* Rx rings stats */ - ring = &adapter->rx_rings[i]; - iavf_add_queue_stats(&data, ring); + iavf_add_queue_stats(&data, tx_ring); + iavf_add_queue_stats(&data, rx_ring); } rcu_read_unlock(); } @@ -376,9 +375,9 @@ static void iavf_get_stat_strings(struct net_device *netdev, u8 *data) iavf_add_stat_strings(&data, iavf_gstrings_stats); /* Queues are always allocated in pairs, so we just use - * real_num_tx_queues for both Tx and Rx queues. + * num_tx_queues for both Tx and Rx queues. */ - for (i = 0; i < netdev->real_num_tx_queues; i++) { + for (i = 0; i < netdev->num_tx_queues; i++) { iavf_add_stat_strings(&data, iavf_gstrings_queue_stats, "tx", i); iavf_add_stat_strings(&data, iavf_gstrings_queue_stats, From 1eb0db7e39da3d20ff6dfb8d359655329ea6f839 Mon Sep 17 00:00:00 2001 From: Joshua Hay Date: Mon, 2 Mar 2026 17:28:31 -0800 Subject: [PATCH 304/352] idpf: clear stale cdev_info ptr Deinit calls idpf_idc_deinit_core_aux_device to free the cdev_info memory, but leaves the adapter->cdev_info field with a stale pointer value. This will bypass subsequent "if (!cdev_info)" checks if cdev_info is not reallocated. For example, if idc_init fails after a reset, cdev_info will already have been freed during the reset handling, but it will not have been reallocated. The next reset or rmmod will result in a crash. [ +0.000008] BUG: kernel NULL pointer dereference, address: 00000000000000d0 [ +0.000033] #PF: supervisor read access in kernel mode [ +0.000020] #PF: error_code(0x0000) - not-present page [ +0.000017] PGD 2097dfa067 P4D 0 [ +0.000017] Oops: Oops: 0000 [#1] SMP NOPTI ... [ +0.000018] RIP: 0010:device_del+0x3e/0x3d0 [ +0.000010] Call Trace: [ +0.000010] [ +0.000012] idpf_idc_deinit_core_aux_device+0x36/0x70 [idpf] [ +0.000034] idpf_vc_core_deinit+0x3e/0x180 [idpf] [ +0.000035] idpf_remove+0x40/0x1d0 [idpf] [ +0.000035] pci_device_remove+0x42/0xb0 [ +0.000020] device_release_driver_internal+0x19c/0x200 [ +0.000024] driver_detach+0x48/0x90 [ +0.000018] bus_remove_driver+0x6d/0x100 [ +0.000023] pci_unregister_driver+0x2e/0xb0 [ +0.000022] __do_sys_delete_module.isra.0+0x18c/0x2b0 [ +0.000025] ? kmem_cache_free+0x2c2/0x390 [ +0.000023] do_syscall_64+0x107/0x7d0 [ +0.000023] entry_SYSCALL_64_after_hwframe+0x76/0x7e Pass the adapter struct into idpf_idc_deinit_core_aux_device instead and clear the cdev_info ptr. Fixes: f4312e6bfa2a ("idpf: implement core RDMA auxiliary dev create, init, and destroy") Signed-off-by: Joshua Hay Reviewed-by: Przemek Kitszel Reviewed-by: Aleksandr Loktionov Reviewed-by: Simon Horman Tested-by: Samuel Salin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/idpf/idpf.h | 2 +- drivers/net/ethernet/intel/idpf/idpf_idc.c | 6 ++++-- drivers/net/ethernet/intel/idpf/idpf_virtchnl.c | 2 +- 3 files changed, 6 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf.h b/drivers/net/ethernet/intel/idpf/idpf.h index b206fba092c8..ec1b75f039bb 100644 --- a/drivers/net/ethernet/intel/idpf/idpf.h +++ b/drivers/net/ethernet/intel/idpf/idpf.h @@ -1066,7 +1066,7 @@ bool idpf_vport_set_hsplit(const struct idpf_vport *vport, u8 val); int idpf_idc_init(struct idpf_adapter *adapter); int idpf_idc_init_aux_core_dev(struct idpf_adapter *adapter, enum iidc_function_type ftype); -void idpf_idc_deinit_core_aux_device(struct iidc_rdma_core_dev_info *cdev_info); +void idpf_idc_deinit_core_aux_device(struct idpf_adapter *adapter); void idpf_idc_deinit_vport_aux_device(struct iidc_rdma_vport_dev_info *vdev_info); void idpf_idc_issue_reset_event(struct iidc_rdma_core_dev_info *cdev_info); void idpf_idc_vdev_mtu_event(struct iidc_rdma_vport_dev_info *vdev_info, diff --git a/drivers/net/ethernet/intel/idpf/idpf_idc.c b/drivers/net/ethernet/intel/idpf/idpf_idc.c index bd4785fb8d3e..7e4f4ac92653 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_idc.c +++ b/drivers/net/ethernet/intel/idpf/idpf_idc.c @@ -470,10 +470,11 @@ int idpf_idc_init_aux_core_dev(struct idpf_adapter *adapter, /** * idpf_idc_deinit_core_aux_device - de-initialize Auxiliary Device(s) - * @cdev_info: IDC core device info pointer + * @adapter: driver private data structure */ -void idpf_idc_deinit_core_aux_device(struct iidc_rdma_core_dev_info *cdev_info) +void idpf_idc_deinit_core_aux_device(struct idpf_adapter *adapter) { + struct iidc_rdma_core_dev_info *cdev_info = adapter->cdev_info; struct iidc_rdma_priv_dev_info *privd; if (!cdev_info) @@ -485,6 +486,7 @@ void idpf_idc_deinit_core_aux_device(struct iidc_rdma_core_dev_info *cdev_info) kfree(privd->mapped_mem_regions); kfree(privd); kfree(cdev_info); + adapter->cdev_info = NULL; } /** diff --git a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c index d5a877e1fef8..113ecfc16dd7 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c +++ b/drivers/net/ethernet/intel/idpf/idpf_virtchnl.c @@ -3668,7 +3668,7 @@ void idpf_vc_core_deinit(struct idpf_adapter *adapter) idpf_ptp_release(adapter); idpf_deinit_task(adapter); - idpf_idc_deinit_core_aux_device(adapter->cdev_info); + idpf_idc_deinit_core_aux_device(adapter); idpf_rel_rx_pt_lkup(adapter); idpf_intr_rel(adapter); From b5e5797e3cd1fd1561b212b2b94f8865d07f2890 Mon Sep 17 00:00:00 2001 From: Joshua Hay Date: Fri, 6 Mar 2026 18:12:47 -0800 Subject: [PATCH 305/352] idpf: only assign num refillqs if allocation was successful As reported by AI review [1], if the refillqs allocation fails, refillqs will be NULL but num_refillqs will be non-zero. The release function will then dereference refillqs since it thinks the refillqs are present, resulting in a NULL ptr dereference. Only assign the num refillqs if the allocation was successful. This will prevent the release function from entering the loop and accessing refillqs. [1] https://lore.kernel.org/netdev/20260227035625.2632753-1-kuba@kernel.org/ Fixes: 95af467d9a4e3 ("idpf: configure resources for RX queues") Signed-off-by: Joshua Hay Reviewed-by: Madhu Chittim Reviewed-by: Aleksandr Loktionov Tested-by: Samuel Salin Signed-off-by: Tony Nguyen --- drivers/net/ethernet/intel/idpf/idpf_txrx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c index 252259993022..f6b3b15364ff 100644 --- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c +++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c @@ -1860,13 +1860,13 @@ static int idpf_rxq_group_alloc(struct idpf_vport *vport, idpf_queue_assign(HSPLIT_EN, q, hs); idpf_queue_assign(RSC_EN, q, rsc); - bufq_set->num_refillqs = num_rxq; bufq_set->refillqs = kcalloc(num_rxq, swq_size, GFP_KERNEL); if (!bufq_set->refillqs) { err = -ENOMEM; goto err_alloc; } + bufq_set->num_refillqs = num_rxq; for (unsigned int k = 0; k < bufq_set->num_refillqs; k++) { struct idpf_sw_queue *refillq = &bufq_set->refillqs[k]; From 6af51e9f31336632263c4680b2a3712295103e1f Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 20 Mar 2026 07:22:59 +0000 Subject: [PATCH 306/352] ipv6: Remove permanent routes from tb6_gc_hlist when all exceptions expire. Commit 5eb902b8e719 ("net/ipv6: Remove expired routes with a separated list of routes.") introduced a per-table GC list and changed GC to iterate over that list instead of traversing the entire route table. However, it forgot to add permanent routes to tb6_gc_hlist when exception routes are added. Commit cfe82469a00f ("ipv6: add exception routes to GC list in rt6_insert_exception") fixed that issue but introduced another one. Even after all exception routes expire, the permanent routes remain in tb6_gc_hlist, potentially negating the performance benefits intended by the initial change. Let's count gc_args->more before and after rt6_age_exceptions() and remove the permanent route when the delta is 0. Note that the next patch will reuse fib6_age_exceptions(). Fixes: cfe82469a00f ("ipv6: add exception routes to GC list in rt6_insert_exception") Signed-off-by: Kuniyuki Iwashima Reviewed-by: Xin Long Reviewed-by: David Ahern Link: https://patch.msgid.link/20260320072317.2561779-2-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv6/ip6_fib.c | 13 ++++++++++++- 1 file changed, 12 insertions(+), 1 deletion(-) diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index 9058e71241dc..fadfca49d6b1 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -2348,6 +2348,17 @@ static void fib6_flush_trees(struct net *net) /* * Garbage collection */ +static void fib6_age_exceptions(struct fib6_info *rt, struct fib6_gc_args *gc_args, + unsigned long now) +{ + bool may_expire = rt->fib6_flags & RTF_EXPIRES && rt->expires; + int old_more = gc_args->more; + + rt6_age_exceptions(rt, gc_args, now); + + if (!may_expire && old_more == gc_args->more) + fib6_remove_gc_list(rt); +} static int fib6_age(struct fib6_info *rt, struct fib6_gc_args *gc_args) { @@ -2370,7 +2381,7 @@ static int fib6_age(struct fib6_info *rt, struct fib6_gc_args *gc_args) * Note, that clones are aged out * only if they are not in use now. */ - rt6_age_exceptions(rt, gc_args, now); + fib6_age_exceptions(rt, gc_args, now); return 0; } From 4be7b99c253f0c85a255cc1db7127ba3232dfa30 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 20 Mar 2026 07:23:00 +0000 Subject: [PATCH 307/352] ipv6: Don't remove permanent routes with exceptions from tb6_gc_hlist. The cited commit mechanically put fib6_remove_gc_list() just after every fib6_clean_expires() call. When a temporary route is promoted to a permanent route, there may already be exception routes tied to it. If fib6_remove_gc_list() removes the route from tb6_gc_hlist, such exception routes will no longer be aged. Let's replace fib6_remove_gc_list() with a new helper fib6_may_remove_gc_list() and use fib6_age_exceptions() there. Note that net->ipv6 is only compiled when CONFIG_IPV6 is enabled, so fib6_{add,remove,may_remove}_gc_list() are guarded. Fixes: 5eb902b8e719 ("net/ipv6: Remove expired routes with a separated list of routes.") Signed-off-by: Kuniyuki Iwashima Reviewed-by: David Ahern Link: https://patch.msgid.link/20260320072317.2561779-3-kuniyu@google.com Signed-off-by: Jakub Kicinski --- include/net/ip6_fib.h | 21 ++++++++++++++++++++- net/ipv6/addrconf.c | 4 ++-- net/ipv6/ip6_fib.c | 6 +++--- net/ipv6/route.c | 2 +- 4 files changed, 26 insertions(+), 7 deletions(-) diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h index 88b0dd4d8e09..9f8b6814a96a 100644 --- a/include/net/ip6_fib.h +++ b/include/net/ip6_fib.h @@ -507,12 +507,14 @@ void fib6_rt_update(struct net *net, struct fib6_info *rt, void inet6_rt_notify(int event, struct fib6_info *rt, struct nl_info *info, unsigned int flags); +void fib6_age_exceptions(struct fib6_info *rt, struct fib6_gc_args *gc_args, + unsigned long now); void fib6_run_gc(unsigned long expires, struct net *net, bool force); - void fib6_gc_cleanup(void); int fib6_init(void); +#if IS_ENABLED(CONFIG_IPV6) /* Add the route to the gc list if it is not already there * * The callers should hold f6i->fib6_table->tb6_lock. @@ -545,6 +547,23 @@ static inline void fib6_remove_gc_list(struct fib6_info *f6i) hlist_del_init(&f6i->gc_link); } +static inline void fib6_may_remove_gc_list(struct net *net, + struct fib6_info *f6i) +{ + struct fib6_gc_args gc_args; + + if (hlist_unhashed(&f6i->gc_link)) + return; + + gc_args.timeout = READ_ONCE(net->ipv6.sysctl.ip6_rt_gc_interval); + gc_args.more = 0; + + rcu_read_lock(); + fib6_age_exceptions(f6i, &gc_args, jiffies); + rcu_read_unlock(); +} +#endif + struct ipv6_route_iter { struct seq_net_private p; struct fib6_walker w; diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 0e55f139e05d..f4e23b543585 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2862,7 +2862,7 @@ void addrconf_prefix_rcv(struct net_device *dev, u8 *opt, int len, bool sllao) fib6_add_gc_list(rt); } else { fib6_clean_expires(rt); - fib6_remove_gc_list(rt); + fib6_may_remove_gc_list(net, rt); } spin_unlock_bh(&table->tb6_lock); @@ -4840,7 +4840,7 @@ static int modify_prefix_route(struct net *net, struct inet6_ifaddr *ifp, if (!(flags & RTF_EXPIRES)) { fib6_clean_expires(f6i); - fib6_remove_gc_list(f6i); + fib6_may_remove_gc_list(net, f6i); } else { fib6_set_expires(f6i, expires); fib6_add_gc_list(f6i); diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c index fadfca49d6b1..dd26657b6a4a 100644 --- a/net/ipv6/ip6_fib.c +++ b/net/ipv6/ip6_fib.c @@ -1133,7 +1133,7 @@ static int fib6_add_rt2node(struct fib6_node *fn, struct fib6_info *rt, return -EEXIST; if (!(rt->fib6_flags & RTF_EXPIRES)) { fib6_clean_expires(iter); - fib6_remove_gc_list(iter); + fib6_may_remove_gc_list(info->nl_net, iter); } else { fib6_set_expires(iter, rt->expires); fib6_add_gc_list(iter); @@ -2348,8 +2348,8 @@ static void fib6_flush_trees(struct net *net) /* * Garbage collection */ -static void fib6_age_exceptions(struct fib6_info *rt, struct fib6_gc_args *gc_args, - unsigned long now) +void fib6_age_exceptions(struct fib6_info *rt, struct fib6_gc_args *gc_args, + unsigned long now) { bool may_expire = rt->fib6_flags & RTF_EXPIRES && rt->expires; int old_more = gc_args->more; diff --git a/net/ipv6/route.c b/net/ipv6/route.c index 08cd86f49bf9..cb521700cee7 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -1033,7 +1033,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len, if (!addrconf_finite_timeout(lifetime)) { fib6_clean_expires(rt); - fib6_remove_gc_list(rt); + fib6_may_remove_gc_list(net, rt); } else { fib6_set_expires(rt, jiffies + HZ * lifetime); fib6_add_gc_list(rt); From 3e9e84e92c9c2eec396ee62a2e47b85781520c57 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 20 Mar 2026 07:23:01 +0000 Subject: [PATCH 308/352] selftest: net: Add GC test for temporary routes with exceptions. Without the prior commit, IPv6 GC cannot track exceptions tied to permanent routes if they were originally added as temporary routes. Let's add a test case for the issue. 1. Add temporary routes 2. Create exceptions for the temporary routes 3. Promote the routes to permanent routes 4. Check if GC can find and purge the exceptions A few notes: + At step 4, unlike other test cases, we cannot wait for $GC_WAIT_TIME. While the exceptions are always iterable via netlink (since it traverses the entire fib tree instead of tb6_gc_hlist), rt6_nh_dump_exceptions() skips expired entries. If we waited for the expiration time, we would be unable to distinguish whether the exceptions were truly purged by GC or just hidden due to being expired. + For the same reason, at step 2, we use ICMPv6 redirect message instead of Packet Too Big message. This is because MTU exceptions always have RTF_EXPIRES, and rt6_age_examine_exception() does not respect the period specified by net.ipv6.route.flush=1. + We add a neighbour entry for the redirect target with NTF_ROUTER. Without this, the exceptions would be removed at step 3 when the fib6_may_remove_gc_list() is called. Without the fix, the exceptions remain even after GC is triggered by sysctl -wq net.ipv6.route.flush=1. FAIL: Expected 0 routes, got 5 TEST: ipv6 route garbage collection (promote to permanent routes) [FAIL] With the fix, GC purges the exceptions properly. TEST: ipv6 route garbage collection (promote to permanent routes) [ OK ] Signed-off-by: Kuniyuki Iwashima Reviewed-by: David Ahern Link: https://patch.msgid.link/20260320072317.2561779-4-kuniyu@google.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/fib_tests.sh | 61 ++++++++++++++++++++++-- 1 file changed, 58 insertions(+), 3 deletions(-) diff --git a/tools/testing/selftests/net/fib_tests.sh b/tools/testing/selftests/net/fib_tests.sh index c5694cc4ddd2..829f72c8ee07 100755 --- a/tools/testing/selftests/net/fib_tests.sh +++ b/tools/testing/selftests/net/fib_tests.sh @@ -868,6 +868,64 @@ fib6_gc_test() check_rt_num 5 $($IP -6 route list |grep -v expires|grep 2001:20::|wc -l) log_test $ret 0 "ipv6 route garbage collection (replace with permanent)" + # Delete dummy_10 and remove all routes + $IP link del dev dummy_10 + + # rd6 is required for the next test. (ipv6toolkit) + if [ ! -x "$(command -v rd6)" ]; then + echo "SKIP: rd6 not found." + set +e + cleanup &> /dev/null + return + fi + + setup_ns ns2 + $IP link add veth1 type veth peer veth2 netns $ns2 + $IP link set veth1 up + ip -netns $ns2 link set veth2 up + $IP addr add fe80:dead::1/64 dev veth1 + ip -netns $ns2 addr add fe80:dead::2/64 dev veth2 + + # Add NTF_ROUTER neighbour to prevent rt6_age_examine_exception() + # from removing not-yet-expired exceptions. + ip -netns $ns2 link set veth2 address 00:11:22:33:44:55 + $IP neigh add fe80:dead::3 lladdr 00:11:22:33:44:55 dev veth1 router + + $NS_EXEC sysctl -wq net.ipv6.conf.veth1.accept_redirects=1 + $NS_EXEC sysctl -wq net.ipv6.conf.veth1.forwarding=0 + + # Temporary routes + for i in $(seq 1 5); do + # Expire route after $EXPIRE seconds + $IP -6 route add 2001:10::$i \ + via fe80:dead::2 dev veth1 expires $EXPIRE + + ip netns exec $ns2 rd6 -i veth2 \ + -s fe80:dead::2 -d fe80:dead::1 \ + -r 2001:10::$i -t fe80:dead::3 -p ICMP6 + done + + check_rt_num 5 $($IP -6 route list | grep expires | grep 2001:10:: | wc -l) + + # Promote to permanent routes by "prepend" (w/o NLM_F_EXCL and NLM_F_REPLACE) + for i in $(seq 1 5); do + # -EEXIST, but the temporary route becomes the permanent route. + $IP -6 route append 2001:10::$i \ + via fe80:dead::2 dev veth1 2>/dev/null || true + done + + check_rt_num 5 $($IP -6 route list | grep -v expires | grep 2001:10:: | wc -l) + check_rt_num 5 $($IP -6 route list cache | grep 2001:10:: | wc -l) + + # Trigger GC instead of waiting $GC_WAIT_TIME. + # rt6_nh_dump_exceptions() just skips expired exceptions. + $NS_EXEC sysctl -wq net.ipv6.route.flush=1 + check_rt_num 0 $($IP -6 route list cache | grep 2001:10:: | wc -l) + log_test $ret 0 "ipv6 route garbage collection (promote to permanent routes)" + + $IP neigh del fe80:dead::3 lladdr 00:11:22:33:44:55 dev veth1 router + $IP link del veth1 + # ra6 is required for the next test. (ipv6toolkit) if [ ! -x "$(command -v ra6)" ]; then echo "SKIP: ra6 not found." @@ -876,9 +934,6 @@ fib6_gc_test() return fi - # Delete dummy_10 and remove all routes - $IP link del dev dummy_10 - # Create a pair of veth devices to send a RA message from one # device to another. $IP link add veth1 type veth peer name veth2 From 42156f93d123436f2a27c468f18c966b7e5db796 Mon Sep 17 00:00:00 2001 From: Yochai Eisenrich Date: Thu, 19 Mar 2026 22:06:10 +0200 Subject: [PATCH 309/352] net: fix fanout UAF in packet_release() via NETDEV_UP race `packet_release()` has a race window where `NETDEV_UP` can re-register a socket into a fanout group's `arr[]` array. The re-registration is not cleaned up by `fanout_release()`, leaving a dangling pointer in the fanout array. `packet_release()` does NOT zero `po->num` in its `bind_lock` section. After releasing `bind_lock`, `po->num` is still non-zero and `po->ifindex` still matches the bound device. A concurrent `packet_notifier(NETDEV_UP)` that already found the socket in `sklist` can re-register the hook. For fanout sockets, this re-registration calls `__fanout_link(sk, po)` which adds the socket back into `f->arr[]` and increments `f->num_members`, but does NOT increment `f->sk_ref`. The fix sets `po->num` to zero in `packet_release` while `bind_lock` is held to prevent NETDEV_UP from linking, preventing the race window. This bug was found following an additional audit with Claude Code based on CVE-2025-38617. Fixes: ce06b03e60fc ("packet: Add helpers to register/unregister ->prot_hook") Link: https://blog.calif.io/p/a-race-within-a-race-exploiting-cve Signed-off-by: Yochai Eisenrich Reviewed-by: Willem de Bruijn Link: https://patch.msgid.link/20260319200610.25101-1-echelonh@gmail.com Signed-off-by: Jakub Kicinski --- net/packet/af_packet.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 72d0935139f0..bb2d88205e5a 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -3135,6 +3135,7 @@ static int packet_release(struct socket *sock) spin_lock(&po->bind_lock); unregister_prot_hook(sk, false); + WRITE_ONCE(po->num, 0); packet_cached_dev_reset(po); if (po->prot_hook.dev) { From 1065913dedfd3a8269816835bfe810b6e2c28579 Mon Sep 17 00:00:00 2001 From: Qingfang Deng Date: Fri, 20 Mar 2026 17:43:15 +0800 Subject: [PATCH 310/352] net: airoha: add RCU lock around dev_fill_forward_path Since 0417adf367a0 ("ppp: fix race conditions in ppp_fill_forward_path") dev_fill_forward_path() should be called with RCU read lock held. This fix was applied to net, while the Airoha flowtable commit was applied to net-next, so it hadn't been an issue until net was merged into net-next. Fixes: a8bdd935d1dd ("net: airoha: Add wlan flowtable TX offload") Signed-off-by: Qingfang Deng Acked-by: Lorenzo Bianconi Link: https://patch.msgid.link/20260320094315.525126-1-dqfext@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_ppe.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/airoha/airoha_ppe.c b/drivers/net/ethernet/airoha/airoha_ppe.c index 42dbe8f93231..5724f8f2defd 100644 --- a/drivers/net/ethernet/airoha/airoha_ppe.c +++ b/drivers/net/ethernet/airoha/airoha_ppe.c @@ -227,7 +227,9 @@ static int airoha_ppe_get_wdma_info(struct net_device *dev, const u8 *addr, if (!dev) return -ENODEV; + rcu_read_lock(); err = dev_fill_forward_path(dev, addr, &stack); + rcu_read_unlock(); if (err) return err; From 3f0f591b44b04a77ff561676ae53fcfd7532a54c Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Fri, 20 Mar 2026 16:49:19 +0100 Subject: [PATCH 311/352] net: b44: always select CONFIG_FIXED_PHY When CONFIG_FIXED_PHY=m but CONFIG_B44=y, the kernel fails to link: ld.lld: error: undefined symbol: fixed_phy_unregister >>> referenced by b44.c >>> drivers/net/ethernet/broadcom/b44.o:(b44_remove_one) in archive vmlinux.a ld.lld: error: undefined symbol: fixed_phy_register_100fd >>> referenced by b44.c >>> drivers/net/ethernet/broadcom/b44.o:(b44_register_phy_one) in archive vmlinux.a The fixed phy support is small enough that just always enabling it for b44 is the simplest solution, and it avoids adding ugly #ifdef checks. Fixes: 10d2f15afba2 ("net: b44: register a fixed phy using fixed_phy_register_100fd if needed") Signed-off-by: Arnd Bergmann Link: https://patch.msgid.link/20260320154927.674555-1-arnd@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/broadcom/Kconfig b/drivers/net/ethernet/broadcom/Kconfig index cd7dddeb91dd..9787c1857e13 100644 --- a/drivers/net/ethernet/broadcom/Kconfig +++ b/drivers/net/ethernet/broadcom/Kconfig @@ -25,7 +25,7 @@ config B44 select SSB select MII select PHYLIB - select FIXED_PHY if BCM47XX + select FIXED_PHY help If you have a network (Ethernet) controller of this type, say Y or M here. From e537dd15d0d4ad989d56a1021290f0c674dd8b28 Mon Sep 17 00:00:00 2001 From: Martin KaFai Lau Date: Thu, 19 Mar 2026 11:18:17 -0700 Subject: [PATCH 312/352] udp: Fix wildcard bind conflict check when using hash2 When binding a udp_sock to a local address and port, UDP uses two hashes (udptable->hash and udptable->hash2) for collision detection. The current code switches to "hash2" when hslot->count > 10. "hash2" is keyed by local address and local port. "hash" is keyed by local port only. The issue can be shown in the following bind sequence (pseudo code): bind(fd1, "[fd00::1]:8888") bind(fd2, "[fd00::2]:8888") bind(fd3, "[fd00::3]:8888") bind(fd4, "[fd00::4]:8888") bind(fd5, "[fd00::5]:8888") bind(fd6, "[fd00::6]:8888") bind(fd7, "[fd00::7]:8888") bind(fd8, "[fd00::8]:8888") bind(fd9, "[fd00::9]:8888") bind(fd10, "[fd00::10]:8888") /* Correctly return -EADDRINUSE because "hash" is used * instead of "hash2". udp_lib_lport_inuse() detects the * conflict. */ bind(fail_fd, "[::]:8888") /* After one more socket is bound to "[fd00::11]:8888", * hslot->count exceeds 10 and "hash2" is used instead. */ bind(fd11, "[fd00::11]:8888") bind(fail_fd, "[::]:8888") /* succeeds unexpectedly */ The same issue applies to the IPv4 wildcard address "0.0.0.0" and the IPv4-mapped wildcard address "::ffff:0.0.0.0". For example, if there are existing sockets bound to "192.168.1.[1-11]:8888", then binding "0.0.0.0:8888" or "[::ffff:0.0.0.0]:8888" can also miss the conflict when hslot->count > 10. TCP inet_csk_get_port() already has the correct check in inet_use_bhash2_on_bind(). Rename it to inet_use_hash2_on_bind() and move it to inet_hashtables.h so udp.c can reuse it in this fix. Fixes: 30fff9231fad ("udp: bind() optimisation") Reported-by: Andrew Onyshchuk Signed-off-by: Martin KaFai Lau Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260319181817.1901357-1-martin.lau@linux.dev Signed-off-by: Jakub Kicinski --- include/net/inet_hashtables.h | 14 ++++++++++++++ net/ipv4/inet_connection_sock.c | 20 +++----------------- net/ipv4/udp.c | 2 +- 3 files changed, 18 insertions(+), 18 deletions(-) diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h index 5a979dcab538..6d936e9f2fd3 100644 --- a/include/net/inet_hashtables.h +++ b/include/net/inet_hashtables.h @@ -264,6 +264,20 @@ inet_bhashfn_portaddr(const struct inet_hashinfo *hinfo, const struct sock *sk, return &hinfo->bhash2[hash & (hinfo->bhash_size - 1)]; } +static inline bool inet_use_hash2_on_bind(const struct sock *sk) +{ +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) { + if (ipv6_addr_any(&sk->sk_v6_rcv_saddr)) + return false; + + if (!ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr)) + return true; + } +#endif + return sk->sk_rcv_saddr != htonl(INADDR_ANY); +} + struct inet_bind_hashbucket * inet_bhash2_addr_any_hashbucket(const struct sock *sk, const struct net *net, int port); diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 5dfac6ce1110..e961936b6be7 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -154,20 +154,6 @@ bool inet_sk_get_local_port_range(const struct sock *sk, int *low, int *high) } EXPORT_SYMBOL(inet_sk_get_local_port_range); -static bool inet_use_bhash2_on_bind(const struct sock *sk) -{ -#if IS_ENABLED(CONFIG_IPV6) - if (sk->sk_family == AF_INET6) { - if (ipv6_addr_any(&sk->sk_v6_rcv_saddr)) - return false; - - if (!ipv6_addr_v4mapped(&sk->sk_v6_rcv_saddr)) - return true; - } -#endif - return sk->sk_rcv_saddr != htonl(INADDR_ANY); -} - static bool inet_bind_conflict(const struct sock *sk, struct sock *sk2, kuid_t uid, bool relax, bool reuseport_cb_ok, bool reuseport_ok) @@ -259,7 +245,7 @@ static int inet_csk_bind_conflict(const struct sock *sk, * checks separately because their spinlocks have to be acquired/released * independently of each other, to prevent possible deadlocks */ - if (inet_use_bhash2_on_bind(sk)) + if (inet_use_hash2_on_bind(sk)) return tb2 && inet_bhash2_conflict(sk, tb2, uid, relax, reuseport_cb_ok, reuseport_ok); @@ -376,7 +362,7 @@ inet_csk_find_open_port(const struct sock *sk, struct inet_bind_bucket **tb_ret, head = &hinfo->bhash[inet_bhashfn(net, port, hinfo->bhash_size)]; spin_lock_bh(&head->lock); - if (inet_use_bhash2_on_bind(sk)) { + if (inet_use_hash2_on_bind(sk)) { if (inet_bhash2_addr_any_conflict(sk, port, l3mdev, relax, false)) goto next_port; } @@ -562,7 +548,7 @@ int inet_csk_get_port(struct sock *sk, unsigned short snum) check_bind_conflict = false; } - if (check_bind_conflict && inet_use_bhash2_on_bind(sk)) { + if (check_bind_conflict && inet_use_hash2_on_bind(sk)) { if (inet_bhash2_addr_any_conflict(sk, port, l3mdev, true, true)) goto fail_unlock; } diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index b60fad393e18..cb99a3c27053 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -287,7 +287,7 @@ int udp_lib_get_port(struct sock *sk, unsigned short snum, } else { hslot = udp_hashslot(udptable, net, snum); spin_lock_bh(&hslot->lock); - if (hslot->count > 10) { + if (inet_use_hash2_on_bind(sk) && hslot->count > 10) { int exist; unsigned int slot2 = udp_sk(sk)->udp_portaddr_hash ^ snum; From 70b439bf06f6a12e491f827fa81a9887a11501f9 Mon Sep 17 00:00:00 2001 From: Wei Fang Date: Fri, 20 Mar 2026 17:42:22 +0800 Subject: [PATCH 313/352] net: enetc: fix the output issue of 'ethtool --show-ring' Currently, enetc_get_ringparam() only provides rx_pending and tx_pending, but 'ethtool --show-ring' no longer displays these fields. Because the ringparam retrieval path has moved to the new netlink interface, where rings_fill_reply() emits the *x_pending only if the *x_max_pending values are non-zero. So rx_max_pending and tx_max_pending to are added to enetc_get_ringparam() to fix the issue. Note that the maximum tx/rx ring size of hardware is 64K, but we haven't added set_ringparam() to make the ring size configurable. To avoid users mistakenly believing that the ring size can be increased, so set the *x_max_pending to priv->*x_bd_count. Fixes: e4a1717b677c ("ethtool: provide ring sizes with RINGS_GET request") Signed-off-by: Wei Fang Link: https://patch.msgid.link/20260320094222.706339-1-wei.fang@nxp.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c index fed89d4f1e1d..2fe140ddebb2 100644 --- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c @@ -813,6 +813,8 @@ static void enetc_get_ringparam(struct net_device *ndev, { struct enetc_ndev_priv *priv = netdev_priv(ndev); + ring->rx_max_pending = priv->rx_bd_count; + ring->tx_max_pending = priv->tx_bd_count; ring->rx_pending = priv->rx_bd_count; ring->tx_pending = priv->tx_bd_count; From 38ec410b99a5ee6566f75650ce3d4fd632940fd0 Mon Sep 17 00:00:00 2001 From: Xuan Zhuo Date: Fri, 20 Mar 2026 10:18:17 +0800 Subject: [PATCH 314/352] virtio-net: correct hdr_len handling for VIRTIO_NET_F_GUEST_HDRLEN The commit be50da3e9d4a ("net: virtio_net: implement exact header length guest feature") introduces support for the VIRTIO_NET_F_GUEST_HDRLEN feature in virtio-net. This feature requires virtio-net to set hdr_len to the actual header length of the packet when transmitting, the number of bytes from the start of the packet to the beginning of the transport-layer payload. However, in practice, hdr_len was being set using skb_headlen(skb), which is clearly incorrect. This commit fixes that issue. Fixes: be50da3e9d4a ("net: virtio_net: implement exact header length guest feature") Signed-off-by: Xuan Zhuo Link: https://patch.msgid.link/20260320021818.111741-2-xuanzhuo@linux.alibaba.com Acked-by: Michael S. Tsirkin Signed-off-by: Paolo Abeni --- drivers/net/tun_vnet.h | 2 +- drivers/net/virtio_net.c | 6 +++++- include/linux/virtio_net.h | 34 ++++++++++++++++++++++++++++++---- 3 files changed, 36 insertions(+), 6 deletions(-) diff --git a/drivers/net/tun_vnet.h b/drivers/net/tun_vnet.h index a5f93b6c4482..fa5cab9d3e55 100644 --- a/drivers/net/tun_vnet.h +++ b/drivers/net/tun_vnet.h @@ -244,7 +244,7 @@ tun_vnet_hdr_tnl_from_skb(unsigned int flags, if (virtio_net_hdr_tnl_from_skb(skb, tnl_hdr, has_tnl_offload, tun_vnet_is_little_endian(flags), - vlan_hlen, true)) { + vlan_hlen, true, false)) { struct virtio_net_hdr_v1 *hdr = &tnl_hdr->hash_hdr.hdr; struct skb_shared_info *sinfo = skb_shinfo(skb); diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 72d6a9c6a5a2..7106333ef904 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -3267,8 +3267,12 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb, bool orphan) struct virtio_net_hdr_v1_hash_tunnel *hdr; int num_sg; unsigned hdr_len = vi->hdr_len; + bool feature_hdrlen; bool can_push; + feature_hdrlen = virtio_has_feature(vi->vdev, + VIRTIO_NET_F_GUEST_HDRLEN); + pr_debug("%s: xmit %p %pM\n", vi->dev->name, skb, dest); /* Make sure it's safe to cast between formats */ @@ -3288,7 +3292,7 @@ static int xmit_skb(struct send_queue *sq, struct sk_buff *skb, bool orphan) if (virtio_net_hdr_tnl_from_skb(skb, hdr, vi->tx_tnl, virtio_is_little_endian(vi->vdev), 0, - false)) + false, feature_hdrlen)) return -EPROTO; if (vi->mergeable_rx_bufs) diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h index 75dabb763c65..361b60c8be68 100644 --- a/include/linux/virtio_net.h +++ b/include/linux/virtio_net.h @@ -207,6 +207,23 @@ static inline int virtio_net_hdr_to_skb(struct sk_buff *skb, return __virtio_net_hdr_to_skb(skb, hdr, little_endian, hdr->gso_type); } +/* This function must be called after virtio_net_hdr_from_skb(). */ +static inline void __virtio_net_set_hdrlen(const struct sk_buff *skb, + struct virtio_net_hdr *hdr, + bool little_endian) +{ + u16 hdr_len; + + hdr_len = skb_transport_offset(skb); + + if (hdr->gso_type == VIRTIO_NET_HDR_GSO_UDP_L4) + hdr_len += sizeof(struct udphdr); + else + hdr_len += tcp_hdrlen(skb); + + hdr->hdr_len = __cpu_to_virtio16(little_endian, hdr_len); +} + static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, struct virtio_net_hdr *hdr, bool little_endian, @@ -385,7 +402,8 @@ virtio_net_hdr_tnl_from_skb(const struct sk_buff *skb, bool tnl_hdr_negotiated, bool little_endian, int vlan_hlen, - bool has_data_valid) + bool has_data_valid, + bool feature_hdrlen) { struct virtio_net_hdr *hdr = (struct virtio_net_hdr *)vhdr; unsigned int inner_nh, outer_th; @@ -394,9 +412,17 @@ virtio_net_hdr_tnl_from_skb(const struct sk_buff *skb, tnl_gso_type = skb_shinfo(skb)->gso_type & (SKB_GSO_UDP_TUNNEL | SKB_GSO_UDP_TUNNEL_CSUM); - if (!tnl_gso_type) - return virtio_net_hdr_from_skb(skb, hdr, little_endian, - has_data_valid, vlan_hlen); + if (!tnl_gso_type) { + ret = virtio_net_hdr_from_skb(skb, hdr, little_endian, + has_data_valid, vlan_hlen); + if (ret) + return ret; + + if (feature_hdrlen && hdr->hdr_len) + __virtio_net_set_hdrlen(skb, hdr, little_endian); + + return ret; + } /* Tunnel support not negotiated but skb ask for it. */ if (!tnl_hdr_negotiated) From 6c860dc02a8e60b438e26940227dfa641fcdb66a Mon Sep 17 00:00:00 2001 From: Xuan Zhuo Date: Fri, 20 Mar 2026 10:18:18 +0800 Subject: [PATCH 315/352] virtio-net: correct hdr_len handling for tunnel gso The commit a2fb4bc4e2a6a03 ("net: implement virtio helpers to handle UDP GSO tunneling.") introduces support for the UDP GSO tunnel feature in virtio-net. The virtio spec says: If the \field{gso_type} has the VIRTIO_NET_HDR_GSO_UDP_TUNNEL_IPV4 bit or VIRTIO_NET_HDR_GSO_UDP_TUNNEL_IPV6 bit set, \field{hdr_len} accounts for all the headers up to and including the inner transport. The commit did not update the hdr_len to include the inner transport. I observed that the "hdr_len" is 116 for this packet: 17:36:18.241105 52:55:00:d1:27:0a > 2e:2c:df:46:a9:e1, ethertype IPv4 (0x0800), length 2912: (tos 0x0, ttl 64, id 45197, offset 0, flags [none], proto UDP (17), length 2898) 192.168.122.100.50613 > 192.168.122.1.4789: [bad udp cksum 0x8106 -> 0x26a0!] VXLAN, flags [I] (0x08), vni 1 fa:c3:ba:82:05:ee > ce:85:0c:31:77:e5, ethertype IPv4 (0x0800), length 2862: (tos 0x0, ttl 64, id 14678, offset 0, flags [DF], proto TCP (6), length 2848) 192.168.3.1.49880 > 192.168.3.2.9898: Flags [P.], cksum 0x9266 (incorrect -> 0xaa20), seq 515667:518463, ack 1, win 64, options [nop,nop,TS val 2990048824 ecr 2798801412], length 2796 116 = 14(mac) + 20(ip) + 8(udp) + 8(vxlan) + 14(inner mac) + 20(inner ip) + 32(innner tcp) Fixes: a2fb4bc4e2a6a03 ("net: implement virtio helpers to handle UDP GSO tunneling.") Signed-off-by: Xuan Zhuo Link: https://patch.msgid.link/20260320021818.111741-3-xuanzhuo@linux.alibaba.com Acked-by: Michael S. Tsirkin Signed-off-by: Paolo Abeni --- include/linux/virtio_net.h | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h index 361b60c8be68..f36d21b5bc19 100644 --- a/include/linux/virtio_net.h +++ b/include/linux/virtio_net.h @@ -224,6 +224,22 @@ static inline void __virtio_net_set_hdrlen(const struct sk_buff *skb, hdr->hdr_len = __cpu_to_virtio16(little_endian, hdr_len); } +/* This function must be called after virtio_net_hdr_from_skb(). */ +static inline void __virtio_net_set_tnl_hdrlen(const struct sk_buff *skb, + struct virtio_net_hdr *hdr) +{ + u16 hdr_len; + + hdr_len = skb_inner_transport_offset(skb); + + if (hdr->gso_type == VIRTIO_NET_HDR_GSO_UDP_L4) + hdr_len += sizeof(struct udphdr); + else + hdr_len += inner_tcp_hdrlen(skb); + + hdr->hdr_len = __cpu_to_virtio16(true, hdr_len); +} + static inline int virtio_net_hdr_from_skb(const struct sk_buff *skb, struct virtio_net_hdr *hdr, bool little_endian, @@ -440,6 +456,9 @@ virtio_net_hdr_tnl_from_skb(const struct sk_buff *skb, if (ret) return ret; + if (feature_hdrlen && hdr->hdr_len) + __virtio_net_set_tnl_hdrlen(skb, hdr); + if (skb->protocol == htons(ETH_P_IPV6)) hdr->gso_type |= VIRTIO_NET_HDR_GSO_UDP_TUNNEL_IPV6; else From 425000dbf17373a4ab8be9428f5dc055ef870a56 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Fri, 20 Mar 2026 15:21:26 +0800 Subject: [PATCH 316/352] team: fix header_ops type confusion with non-Ethernet ports Similar to commit 950803f72547 ("bonding: fix type confusion in bond_setup_by_slave()") team has the same class of header_ops type confusion. For non-Ethernet ports, team_setup_by_port() copies port_dev->header_ops directly. When the team device later calls dev_hard_header() or dev_parse_header(), these callbacks can run with the team net_device instead of the real lower device, so netdev_priv(dev) is interpreted as the wrong private type and can crash. The syzbot report shows a crash in bond_header_create(), but the root cause is in team: the topology is gre -> bond -> team, and team calls the inherited header_ops with its own net_device instead of the lower device, so bond_header_create() receives a team device and interprets netdev_priv() as bonding private data, causing a type confusion crash. Fix this by introducing team header_ops wrappers for create/parse, selecting a team port under RCU, and calling the lower device callbacks with port->dev, so each callback always sees the correct net_device context. Also pass the selected lower device to the lower parse callback, so recursion is bounded in stacked non-Ethernet topologies and parse callbacks always run with the correct device context. Fixes: 1d76efe1577b ("team: add support for non-ethernet devices") Reported-by: syzbot+3d8bc31c45e11450f24c@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/69b46af7.050a0220.36eb34.000e.GAE@google.com/T/ Cc: Jiayuan Chen Signed-off-by: Jiayuan Chen Link: https://patch.msgid.link/20260320072139.134249-2-jiayuan.chen@linux.dev Signed-off-by: Paolo Abeni --- drivers/net/team/team_core.c | 65 +++++++++++++++++++++++++++++++++++- 1 file changed, 64 insertions(+), 1 deletion(-) diff --git a/drivers/net/team/team_core.c b/drivers/net/team/team_core.c index b7282f5c9632..120aeb539d9f 100644 --- a/drivers/net/team/team_core.c +++ b/drivers/net/team/team_core.c @@ -2058,6 +2058,68 @@ static const struct ethtool_ops team_ethtool_ops = { * rt netlink interface ***********************/ +/* For tx path we need a linkup && enabled port and for parse any port + * suffices. + */ +static struct team_port *team_header_port_get_rcu(struct team *team, + bool txable) +{ + struct team_port *port; + + list_for_each_entry_rcu(port, &team->port_list, list) { + if (!txable || team_port_txable(port)) + return port; + } + + return NULL; +} + +static int team_header_create(struct sk_buff *skb, struct net_device *team_dev, + unsigned short type, const void *daddr, + const void *saddr, unsigned int len) +{ + struct team *team = netdev_priv(team_dev); + const struct header_ops *port_ops; + struct team_port *port; + int ret = 0; + + rcu_read_lock(); + port = team_header_port_get_rcu(team, true); + if (port) { + port_ops = READ_ONCE(port->dev->header_ops); + if (port_ops && port_ops->create) + ret = port_ops->create(skb, port->dev, + type, daddr, saddr, len); + } + rcu_read_unlock(); + return ret; +} + +static int team_header_parse(const struct sk_buff *skb, + const struct net_device *team_dev, + unsigned char *haddr) +{ + struct team *team = netdev_priv(team_dev); + const struct header_ops *port_ops; + struct team_port *port; + int ret = 0; + + rcu_read_lock(); + port = team_header_port_get_rcu(team, false); + if (port) { + port_ops = READ_ONCE(port->dev->header_ops); + if (port_ops && port_ops->parse) + ret = port_ops->parse(skb, port->dev, haddr); + } + rcu_read_unlock(); + return ret; +} + +static const struct header_ops team_header_ops = { + .create = team_header_create, + .parse = team_header_parse, +}; + static void team_setup_by_port(struct net_device *dev, struct net_device *port_dev) { @@ -2066,7 +2128,8 @@ static void team_setup_by_port(struct net_device *dev, if (port_dev->type == ARPHRD_ETHER) dev->header_ops = team->header_ops_cache; else - dev->header_ops = port_dev->header_ops; + dev->header_ops = port_dev->header_ops ? + &team_header_ops : NULL; dev->type = port_dev->type; dev->hard_header_len = port_dev->hard_header_len; dev->needed_headroom = port_dev->needed_headroom; From 56063823b9f0e2acdca4d621face5c6a7a1f4c99 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Fri, 20 Mar 2026 15:21:27 +0800 Subject: [PATCH 317/352] selftests: team: add non-Ethernet header_ops reproducer Add a team selftest that sets up: g0 (gre) -> b0 (bond) -> t0 (team) and triggers IPv6 traffic on t0. This reproduces the non-Ethernet header_ops confusion scenario and protects against regressions in stacked team/bond/gre configurations. Using this script, the panic reported by syzkaller can be reproduced [1]. After the fix: # ./non_ether_header_ops.sh PASS: non-Ethernet header_ops stacking did not crash [1] https://syzkaller.appspot.com/bug?extid=3d8bc31c45e11450f24c Cc: Jiayuan Chen Signed-off-by: Jiayuan Chen Link: https://patch.msgid.link/20260320072139.134249-3-jiayuan.chen@linux.dev Signed-off-by: Paolo Abeni --- .../selftests/drivers/net/team/Makefile | 1 + .../testing/selftests/drivers/net/team/config | 2 + .../drivers/net/team/non_ether_header_ops.sh | 41 +++++++++++++++++++ 3 files changed, 44 insertions(+) create mode 100755 tools/testing/selftests/drivers/net/team/non_ether_header_ops.sh diff --git a/tools/testing/selftests/drivers/net/team/Makefile b/tools/testing/selftests/drivers/net/team/Makefile index 45a3e7ad3dcb..02d6f51d5a06 100644 --- a/tools/testing/selftests/drivers/net/team/Makefile +++ b/tools/testing/selftests/drivers/net/team/Makefile @@ -3,6 +3,7 @@ TEST_PROGS := \ dev_addr_lists.sh \ + non_ether_header_ops.sh \ options.sh \ propagation.sh \ refleak.sh \ diff --git a/tools/testing/selftests/drivers/net/team/config b/tools/testing/selftests/drivers/net/team/config index 558e1d0cf565..5d36a22ef080 100644 --- a/tools/testing/selftests/drivers/net/team/config +++ b/tools/testing/selftests/drivers/net/team/config @@ -1,7 +1,9 @@ +CONFIG_BONDING=y CONFIG_DUMMY=y CONFIG_IPV6=y CONFIG_MACVLAN=y CONFIG_NETDEVSIM=m +CONFIG_NET_IPGRE=y CONFIG_NET_TEAM=y CONFIG_NET_TEAM_MODE_ACTIVEBACKUP=y CONFIG_NET_TEAM_MODE_LOADBALANCE=y diff --git a/tools/testing/selftests/drivers/net/team/non_ether_header_ops.sh b/tools/testing/selftests/drivers/net/team/non_ether_header_ops.sh new file mode 100755 index 000000000000..948a43576bdc --- /dev/null +++ b/tools/testing/selftests/drivers/net/team/non_ether_header_ops.sh @@ -0,0 +1,41 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# shellcheck disable=SC2154 +# +# Reproduce the non-Ethernet header_ops confusion scenario with: +# g0 (gre) -> b0 (bond) -> t0 (team) +# +# Before the fix, direct header_ops inheritance in this stack could call +# callbacks with the wrong net_device context and crash. + +lib_dir=$(dirname "$0") +source "$lib_dir"/../../../net/lib.sh + +trap cleanup_all_ns EXIT + +setup_ns ns1 + +ip -n "$ns1" link add d0 type dummy +ip -n "$ns1" addr add 10.10.10.1/24 dev d0 +ip -n "$ns1" link set d0 up + +ip -n "$ns1" link add g0 type gre local 10.10.10.1 +ip -n "$ns1" link add b0 type bond mode active-backup +ip -n "$ns1" link add t0 type team + +ip -n "$ns1" link set g0 master b0 +ip -n "$ns1" link set b0 master t0 + +ip -n "$ns1" link set g0 up +ip -n "$ns1" link set b0 up +ip -n "$ns1" link set t0 up + +# IPv6 address assignment triggers MLD join reports that call +# dev_hard_header() on t0, exercising the inherited header_ops path. +ip -n "$ns1" -6 addr add 2001:db8:1::1/64 dev t0 nodad +for i in $(seq 1 20); do + ip netns exec "$ns1" ping -6 -I t0 ff02::1 -c1 -W1 &>/dev/null || true +done + +echo "PASS: non-Ethernet header_ops stacking did not crash" +exit "$EXIT_STATUS" From fbddf68d7b4e1e6da7a78dd7fbd8ec376536584a Mon Sep 17 00:00:00 2001 From: Srinivas Pandruvada Date: Mon, 23 Mar 2026 08:36:35 -0700 Subject: [PATCH 318/352] platform/x86: ISST: Correct locked bit width MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit SST-PP locked bit width is set to three bits. It should be only one bit. Use SST_PP_LOCK_WIDTH define instead of SST_PP_LEVEL_WIDTH. Fixes: ea009e4769fa ("platform/x86: ISST: Add SST-PP support via TPMI") Signed-off-by: Srinivas Pandruvada Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260323153635.3263828-1-srinivas.pandruvada@linux.intel.com Reviewed-by: Ilpo Järvinen Signed-off-by: Ilpo Järvinen --- drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c b/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c index 6f46cbb2898f..e238c3105c78 100644 --- a/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c +++ b/drivers/platform/x86/intel/speed_select_if/isst_tpmi_core.c @@ -872,7 +872,7 @@ static int isst_if_get_perf_level(void __user *argp) _read_pp_info("current_level", perf_level.current_level, SST_PP_STATUS_OFFSET, SST_PP_LEVEL_START, SST_PP_LEVEL_WIDTH, SST_MUL_FACTOR_NONE) _read_pp_info("locked", perf_level.locked, SST_PP_STATUS_OFFSET, - SST_PP_LOCK_START, SST_PP_LEVEL_WIDTH, SST_MUL_FACTOR_NONE) + SST_PP_LOCK_START, SST_PP_LOCK_WIDTH, SST_MUL_FACTOR_NONE) _read_pp_info("feature_state", perf_level.feature_state, SST_PP_STATUS_OFFSET, SST_PP_FEATURE_STATE_START, SST_PP_FEATURE_STATE_WIDTH, SST_MUL_FACTOR_NONE) perf_level.enabled = !!(power_domain_info->sst_header.cap_mask & BIT(1)); From eb8c426c9803beb171f89d15fea17505eb517714 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Fri, 20 Mar 2026 17:44:39 +0000 Subject: [PATCH 319/352] net: ti: icssg-prueth: fix use-after-free of CPPI descriptor in RX path cppi5_hdesc_get_psdata() returns a pointer into the CPPI descriptor. In both emac_rx_packet() and emac_rx_packet_zc(), the descriptor is freed via k3_cppi_desc_pool_free() before the psdata pointer is used by emac_rx_timestamp(), which dereferences psdata[0] and psdata[1]. This constitutes a use-after-free on every received packet that goes through the timestamp path. Defer the descriptor free until after all accesses through the psdata pointer are complete. For emac_rx_packet(), move the free into the requeue label so both early-exit and success paths free the descriptor after all accesses are done. For emac_rx_packet_zc(), move the free to the end of the loop body after emac_dispatch_skb_zc() (which calls emac_rx_timestamp()) has returned. Fixes: 46eeb90f03e0 ("net: ti: icssg-prueth: Use page_pool API for RX buffer allocation") Signed-off-by: David Carlier Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260320174439.41080-1-devnexen@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/ti/icssg/icssg_common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/ti/icssg/icssg_common.c b/drivers/net/ethernet/ti/icssg/icssg_common.c index 0a3cf2f848a5..fd4e7622f123 100644 --- a/drivers/net/ethernet/ti/icssg/icssg_common.c +++ b/drivers/net/ethernet/ti/icssg/icssg_common.c @@ -962,7 +962,6 @@ static int emac_rx_packet_zc(struct prueth_emac *emac, u32 flow_id, pkt_len -= 4; cppi5_desc_get_tags_ids(&desc_rx->hdr, &port_id, NULL); psdata = cppi5_hdesc_get_psdata(desc_rx); - k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); count++; xsk_buff_set_size(xdp, pkt_len); xsk_buff_dma_sync_for_cpu(xdp); @@ -988,6 +987,7 @@ static int emac_rx_packet_zc(struct prueth_emac *emac, u32 flow_id, emac_dispatch_skb_zc(emac, xdp, psdata); xsk_buff_free(xdp); } + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); } if (xdp_status & ICSSG_XDP_REDIR) @@ -1057,7 +1057,6 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) /* firmware adds 4 CRC bytes, strip them */ pkt_len -= 4; cppi5_desc_get_tags_ids(&desc_rx->hdr, &port_id, NULL); - k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); /* if allocation fails we drop the packet but push the * descriptor back to the ring with old page to prevent a stall @@ -1115,6 +1114,7 @@ static int emac_rx_packet(struct prueth_emac *emac, u32 flow_id, u32 *xdp_state) ndev->stats.rx_packets++; requeue: + k3_cppi_desc_pool_free(rx_chn->desc_pool, desc_rx); /* queue another RX DMA */ ret = prueth_dma_rx_push_mapped(emac, &emac->rx_chns, new_page, PRUETH_MAX_PKT_SIZE); From c4336a07eb6b2526dc2b62928b5104b41a7f81f5 Mon Sep 17 00:00:00 2001 From: Willem de Bruijn Date: Fri, 20 Mar 2026 15:01:46 -0400 Subject: [PATCH 320/352] net: correctly handle tunneled traffic on IPV6_CSUM GSO fallback NETIF_F_IPV6_CSUM only advertises support for checksum offload of packets without IPv6 extension headers. Packets with extension headers must fall back onto software checksumming. Since TSO depends on checksum offload, those must revert to GSO. The below commit introduces that fallback. It always checks network header length. For tunneled packets, the inner header length must be checked instead. Extend the check accordingly. A special case is tunneled packets without inner IP protocol. Such as RFC 6951 SCTP in UDP. Those are not standard IPv6 followed by transport header either, so also must revert to the software GSO path. Cc: stable@vger.kernel.org Fixes: 864e3396976e ("net: gso: Forbid IPv6 TSO with extensions on devices with only IPV6_CSUM") Reported-by: Tangxin Xie Closes: https://lore.kernel.org/netdev/0414e7e2-9a1c-4d7c-a99d-b9039cf68f40@yeah.net/ Suggested-by: Paolo Abeni Signed-off-by: Willem de Bruijn Link: https://patch.msgid.link/20260320190148.2409107-1-willemdebruijn.kernel@gmail.com Signed-off-by: Paolo Abeni --- net/core/dev.c | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-) diff --git a/net/core/dev.c b/net/core/dev.c index 14a83f2035b9..fc5557062414 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -3769,6 +3769,22 @@ static netdev_features_t dflt_features_check(struct sk_buff *skb, return vlan_features_check(skb, features); } +static bool skb_gso_has_extension_hdr(const struct sk_buff *skb) +{ + if (!skb->encapsulation) + return ((skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6 || + (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4 && + vlan_get_protocol(skb) == htons(ETH_P_IPV6))) && + skb_transport_header_was_set(skb) && + skb_network_header_len(skb) != sizeof(struct ipv6hdr)); + else + return (!skb_inner_network_header_was_set(skb) || + ((skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6 || + (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4 && + inner_ip_hdr(skb)->version == 6)) && + skb_inner_network_header_len(skb) != sizeof(struct ipv6hdr))); +} + static netdev_features_t gso_features_check(const struct sk_buff *skb, struct net_device *dev, netdev_features_t features) @@ -3816,11 +3832,7 @@ static netdev_features_t gso_features_check(const struct sk_buff *skb, * so neither does TSO that depends on it. */ if (features & NETIF_F_IPV6_CSUM && - (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6 || - (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4 && - vlan_get_protocol(skb) == htons(ETH_P_IPV6))) && - skb_transport_header_was_set(skb) && - skb_network_header_len(skb) != sizeof(struct ipv6hdr)) + skb_gso_has_extension_hdr(skb)) features &= ~(NETIF_F_IPV6_CSUM | NETIF_F_TSO6 | NETIF_F_GSO_UDP_L4); return features; From 647b8a2fe474474704110db6bd07f7a139e621eb Mon Sep 17 00:00:00 2001 From: Kevin Hao Date: Sat, 21 Mar 2026 22:04:41 +0800 Subject: [PATCH 321/352] net: macb: Use dev_consume_skb_any() to free TX SKBs The napi_consume_skb() function is not intended to be called in an IRQ disabled context. However, after commit 6bc8a5098bf4 ("net: macb: Fix tx_ptr_lock locking"), the freeing of TX SKBs is performed with IRQs disabled. To resolve the following call trace, use dev_consume_skb_any() for freeing TX SKBs: WARNING: kernel/softirq.c:430 at __local_bh_enable_ip+0x174/0x188, CPU#0: ksoftirqd/0/15 Modules linked in: CPU: 0 UID: 0 PID: 15 Comm: ksoftirqd/0 Not tainted 7.0.0-rc4-next-20260319-yocto-standard-dirty #37 PREEMPT Hardware name: ZynqMP ZCU102 Rev1.1 (DT) pstate: 200000c5 (nzCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : __local_bh_enable_ip+0x174/0x188 lr : local_bh_enable+0x24/0x38 sp : ffff800082b3bb10 x29: ffff800082b3bb10 x28: ffff0008031f3c00 x27: 000000000011ede0 x26: ffff000800a7ff00 x25: ffff800083937ce8 x24: 0000000000017a80 x23: ffff000803243a78 x22: 0000000000000040 x21: 0000000000000000 x20: ffff000800394c80 x19: 0000000000000200 x18: 0000000000000001 x17: 0000000000000001 x16: ffff000803240000 x15: 0000000000000000 x14: ffffffffffffffff x13: 0000000000000028 x12: ffff000800395650 x11: ffff8000821d1528 x10: ffff800081c2bc08 x9 : ffff800081c1e258 x8 : 0000000100000301 x7 : ffff8000810426ec x6 : 0000000000000000 x5 : 0000000000000001 x4 : 0000000000000001 x3 : 0000000000000000 x2 : 0000000000000008 x1 : 0000000000000200 x0 : ffff8000810428dc Call trace: __local_bh_enable_ip+0x174/0x188 (P) local_bh_enable+0x24/0x38 skb_attempt_defer_free+0x190/0x1d8 napi_consume_skb+0x58/0x108 macb_tx_poll+0x1a4/0x558 __napi_poll+0x50/0x198 net_rx_action+0x1f4/0x3d8 handle_softirqs+0x16c/0x560 run_ksoftirqd+0x44/0x80 smpboot_thread_fn+0x1d8/0x338 kthread+0x120/0x150 ret_from_fork+0x10/0x20 irq event stamp: 29751 hardirqs last enabled at (29750): [] _raw_spin_unlock_irqrestore+0x44/0x88 hardirqs last disabled at (29751): [] _raw_spin_lock_irqsave+0x38/0x98 softirqs last enabled at (29150): [] handle_softirqs+0x504/0x560 softirqs last disabled at (29153): [] run_ksoftirqd+0x44/0x80 Fixes: 6bc8a5098bf4 ("net: macb: Fix tx_ptr_lock locking") Signed-off-by: Kevin Hao Cc: stable@vger.kernel.org Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260321-macb-tx-v1-1-b383a58dd4e6@gmail.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/cadence/macb_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index 944aab71795c..fc4dd0757646 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -1071,7 +1071,7 @@ static void macb_tx_unmap(struct macb *bp, struct macb_tx_skb *tx_skb, int budge } if (tx_skb->skb) { - napi_consume_skb(tx_skb->skb, budget); + dev_consume_skb_any(tx_skb->skb); tx_skb->skb = NULL; } } From 938c418422c4b08523ae39aebbd828428dcfefd2 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Mon, 23 Mar 2026 17:48:57 +0800 Subject: [PATCH 322/352] erofs: update the Kconfig description Refine the description to better highlight its features and use cases. In addition, add instructions for building it as a module and clarify the compression option. Reviewed-by: Chao Yu Signed-off-by: Gao Xiang --- fs/erofs/Kconfig | 43 +++++++++++++++++++++++++++++-------------- 1 file changed, 29 insertions(+), 14 deletions(-) diff --git a/fs/erofs/Kconfig b/fs/erofs/Kconfig index a9f645f57bb2..97c48ebe8458 100644 --- a/fs/erofs/Kconfig +++ b/fs/erofs/Kconfig @@ -16,22 +16,36 @@ config EROFS_FS select ZLIB_INFLATE if EROFS_FS_ZIP_DEFLATE select ZSTD_DECOMPRESS if EROFS_FS_ZIP_ZSTD help - EROFS (Enhanced Read-Only File System) is a lightweight read-only - file system with modern designs (e.g. no buffer heads, inline - xattrs/data, chunk-based deduplication, multiple devices, etc.) for - scenarios which need high-performance read-only solutions, e.g. - smartphones with Android OS, LiveCDs and high-density hosts with - numerous containers; + EROFS (Enhanced Read-Only File System) is a modern, lightweight, + secure read-only filesystem for various use cases, such as immutable + system images, container images, application sandboxes, and datasets. - It also provides transparent compression and deduplication support to - improve storage density and maintain relatively high compression - ratios, and it implements in-place decompression to temporarily reuse - page cache for compressed data using proper strategies, which is - quite useful for ensuring guaranteed end-to-end runtime decompression + EROFS uses a flexible, hierarchical on-disk design so that features + can be enabled on demand: the core on-disk format is block-aligned in + order to perform optimally on all kinds of devices, including block + and memory-backed devices; the format is easy to parse and has zero + metadata redundancy, unlike generic filesystems, making it ideal for + filesystem auditing and remote access; inline data, random-access + friendly directory data, inline/shared extended attributes and + chunk-based deduplication ensure space efficiency while maintaining + high performance. + + Optionally, it supports multiple devices to reference external data, + enabling data sharing for container images. + + It also has advanced encoded on-disk layouts, particularly for data + compression and fine-grained deduplication. It utilizes fixed-size + output compression to improve storage density while keeping relatively + high compression ratios. Furthermore, it implements in-place + decompression to reuse file pages to keep compressed data temporarily + with proper strategies, which ensures guaranteed end-to-end runtime performance under extreme memory pressure without extra cost. - See the documentation at - and the web pages at for more details. + For more details, see the web pages at + and the documentation at . + + To compile EROFS filesystem support as a module, choose M here. The + module will be called erofs. If unsure, say N. @@ -105,7 +119,8 @@ config EROFS_FS_ZIP depends on EROFS_FS default y help - Enable transparent compression support for EROFS file systems. + Enable EROFS compression layouts so that filesystems containing + compressed files can be parsed by the kernel. If you don't want to enable compression feature, say N. From 2f0407ed923b7eb363424033fc12fe253da139c4 Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Tue, 24 Mar 2026 23:54:07 +0800 Subject: [PATCH 323/352] erofs: fix .fadvise() for page cache sharing Currently, .fadvise() doesn't work well if page cache sharing is on since shared inodes belong to a pseudo fs generated with init_pseudo(), and sb->s_bdi is the default one &noop_backing_dev_info. Then, generic_fadvise() will just behave as a no-op if sb->s_bdi is &noop_backing_dev_info, but as the bdev fs (the bdev fs changes inode_to_bdi() instead), it's actually NOT a pure memfs. Let's generate a real bdi for erofs_ishare_mnt instead. Fixes: d86d7817c042 ("erofs: implement .fadvise for page cache share") Reviewed-by: Hongbo Li Signed-off-by: Gao Xiang --- fs/erofs/ishare.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-) diff --git a/fs/erofs/ishare.c b/fs/erofs/ishare.c index 829d50d5c717..ec433bacc592 100644 --- a/fs/erofs/ishare.c +++ b/fs/erofs/ishare.c @@ -200,8 +200,19 @@ struct inode *erofs_real_inode(struct inode *inode, bool *need_iput) int __init erofs_init_ishare(void) { - erofs_ishare_mnt = kern_mount(&erofs_anon_fs_type); - return PTR_ERR_OR_ZERO(erofs_ishare_mnt); + struct vfsmount *mnt; + int ret; + + mnt = kern_mount(&erofs_anon_fs_type); + if (IS_ERR(mnt)) + return PTR_ERR(mnt); + /* generic_fadvise() doesn't work if s_bdi == &noop_backing_dev_info */ + ret = super_setup_bdi(mnt->mnt_sb); + if (ret) + kern_unmount(mnt); + else + erofs_ishare_mnt = mnt; + return ret; } void erofs_exit_ishare(void) From ba8bda9a0896746053aa97ac6c3e08168729172c Mon Sep 17 00:00:00 2001 From: xietangxin Date: Thu, 12 Mar 2026 10:54:06 +0800 Subject: [PATCH 324/352] virtio_net: Fix UAF on dst_ops when IFF_XMIT_DST_RELEASE is cleared and napi_tx is false A UAF issue occurs when the virtio_net driver is configured with napi_tx=N and the device's IFF_XMIT_DST_RELEASE flag is cleared (e.g., during the configuration of tc route filter rules). When IFF_XMIT_DST_RELEASE is removed from the net_device, the network stack expects the driver to hold the reference to skb->dst until the packet is fully transmitted and freed. In virtio_net with napi_tx=N, skbs may remain in the virtio transmit ring for an extended period. If the network namespace is destroyed while these skbs are still pending, the corresponding dst_ops structure has freed. When a subsequent packet is transmitted, free_old_xmit() is triggered to clean up old skbs. It then calls dst_release() on the skb associated with the stale dst_entry. Since the dst_ops (referenced by the dst_entry) has already been freed, a UAF kernel paging request occurs. fix it by adds skb_dst_drop(skb) in start_xmit to explicitly release the dst reference before the skb is queued in virtio_net. Call Trace: Unable to handle kernel paging request at virtual address ffff80007e150000 CPU: 2 UID: 0 PID: 6236 Comm: ping Kdump: loaded Not tainted 7.0.0-rc1+ #6 PREEMPT ... percpu_counter_add_batch+0x3c/0x158 lib/percpu_counter.c:98 (P) dst_release+0xe0/0x110 net/core/dst.c:177 skb_release_head_state+0xe8/0x108 net/core/skbuff.c:1177 sk_skb_reason_drop+0x54/0x2d8 net/core/skbuff.c:1255 dev_kfree_skb_any_reason+0x64/0x78 net/core/dev.c:3469 napi_consume_skb+0x1c4/0x3a0 net/core/skbuff.c:1527 __free_old_xmit+0x164/0x230 drivers/net/virtio_net.c:611 [virtio_net] free_old_xmit drivers/net/virtio_net.c:1081 [virtio_net] start_xmit+0x7c/0x530 drivers/net/virtio_net.c:3329 [virtio_net] ... Reproduction Steps: NETDEV="enp3s0" config_qdisc_route_filter() { tc qdisc del dev $NETDEV root tc qdisc add dev $NETDEV root handle 1: prio tc filter add dev $NETDEV parent 1:0 \ protocol ip prio 100 route to 100 flowid 1:1 ip route add 192.168.1.100/32 dev $NETDEV realm 100 } test_ns() { ip netns add testns ip link set $NETDEV netns testns ip netns exec testns ifconfig $NETDEV 10.0.32.46/24 ip netns exec testns ping -c 1 10.0.32.1 ip netns del testns } config_qdisc_route_filter test_ns sleep 2 test_ns Fixes: f2fc6a54585a ("[NETNS][IPV6] route6 - move ip6_dst_ops inside the network namespace") Cc: stable@vger.kernel.org Signed-off-by: xietangxin Reviewed-by: Xuan Zhuo Fixes: 0287587884b1 ("net: better IFF_XMIT_DST_RELEASE support") Link: https://patch.msgid.link/20260312025406.15641-1-xietangxin@yeah.net Signed-off-by: Jakub Kicinski --- drivers/net/virtio_net.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c index 7106333ef904..ab2108ee206a 100644 --- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -3355,6 +3355,7 @@ static netdev_tx_t start_xmit(struct sk_buff *skb, struct net_device *dev) /* Don't wait up for transmitted skbs to be freed. */ if (!use_napi) { skb_orphan(skb); + skb_dst_drop(skb); nf_reset_ct(skb); } From 71399707876b93240f236f48b8062f3423a5fe97 Mon Sep 17 00:00:00 2001 From: Thangaraj Samynathan Date: Mon, 23 Mar 2026 12:23:45 +0530 Subject: [PATCH 325/352] net: lan743x: fix duplex configuration in mac_link_up The driver does not explicitly configure the MAC duplex mode when bringing the link up. As a result, the MAC may retain a stale duplex setting from a previous link state, leading to duplex mismatches with the link partner and degraded network performance. Update lan743x_phylink_mac_link_up() to set or clear the MAC_CR_DPX_ bit according to the negotiated duplex mode. This ensures the MAC configuration is consistent with the phylink resolved state. Fixes: a5f199a8d8a03 ("net: lan743x: Migrate phylib to phylink") Signed-off-by: Thangaraj Samynathan Reviewed-by: Russell King (Oracle) Link: https://patch.msgid.link/20260323065345.144915-1-thangaraj.s@microchip.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microchip/lan743x_main.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index a3845edf0e48..f0b5dd752f08 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -3053,6 +3053,11 @@ static void lan743x_phylink_mac_link_up(struct phylink_config *config, else if (speed == SPEED_100) mac_cr |= MAC_CR_CFG_L_; + if (duplex == DUPLEX_FULL) + mac_cr |= MAC_CR_DPX_; + else + mac_cr &= ~MAC_CR_DPX_; + lan743x_csr_write(adapter, MAC_CR, mac_cr); lan743x_ptp_update_latency(adapter, speed); From 09474055f2619be9445ba4245e4013741ed01a5e Mon Sep 17 00:00:00 2001 From: Sabrina Dubroca Date: Mon, 23 Mar 2026 16:19:43 +0100 Subject: [PATCH 326/352] rtnetlink: fix leak of SRCU struct in rtnl_link_register Commit 6b57ff21a310 ("rtnetlink: Protect link_ops by mutex.") swapped the EEXIST check with the init_srcu_struct, but didn't add cleanup of the SRCU struct we just allocated in case of error. Fixes: 6b57ff21a310 ("rtnetlink: Protect link_ops by mutex.") Signed-off-by: Sabrina Dubroca Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/e77fe499f9a58c547b33b5212b3596dad417cec6.1774025341.git.sd@queasysnail.net Signed-off-by: Jakub Kicinski --- net/core/rtnetlink.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index 4a2278614250..fae8034efbff 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -629,6 +629,9 @@ int rtnl_link_register(struct rtnl_link_ops *ops) unlock: mutex_unlock(&link_ops_mutex); + if (err) + cleanup_srcu_struct(&ops->srcu); + return err; } EXPORT_SYMBOL_GPL(rtnl_link_register); From 815980fe6dbb01ad4007e8b260a45617f598b76d Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jonas=20K=C3=B6ppeler?= Date: Mon, 23 Mar 2026 18:49:20 +0100 Subject: [PATCH 327/352] net_sched: codel: fix stale state for empty flows in fq_codel MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When codel_dequeue() finds an empty queue, it resets vars->dropping but does not reset vars->first_above_time. The reference CoDel algorithm (Nichols & Jacobson, ACM Queue 2012) resets both: dodeque_result codel_queue_t::dodeque(time_t now) { ... if (r.p == NULL) { first_above_time = 0; // <-- Linux omits this } ... } Note that codel_should_drop() does reset first_above_time when called with a NULL skb, but codel_dequeue() returns early before ever calling codel_should_drop() in the empty-queue case. The post-drop code paths do reach codel_should_drop(NULL) and correctly reset the timer, so a dropped packet breaks the cycle -- but the next delivered packet re-arms first_above_time and the cycle repeats. For sparse flows such as ICMP ping (one packet every 200ms-1s), the first packet arms first_above_time, the flow goes empty, and the second packet arrives after the interval has elapsed and gets dropped. The pattern repeats, producing sustained loss on flows that are not actually congested. Test: veth pair, fq_codel, BQL disabled, 30000 iptables rules in the consumer namespace (NAPI-64 cycle ~14ms, well above fq_codel's 5ms target), ping at 5 pps under UDP flood: Before fix: 26% ping packet loss After fix: 0% ping packet loss Fix by resetting first_above_time to zero in the empty-queue path of codel_dequeue(), matching the reference algorithm. Fixes: 76e3cc126bb2 ("codel: Controlled Delay AQM") Fixes: d068ca2ae2e6 ("codel: split into multiple files") Co-developed-by: Jesper Dangaard Brouer Signed-off-by: Jesper Dangaard Brouer Signed-off-by: Jonas Köppeler Reported-by: Chris Arges Tested-by: Jonas Köppeler Reviewed-by: Toke Høiland-Jørgensen Link: https://lore.kernel.org/all/20260318134826.1281205-7-hawk@kernel.org/ Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260323174920.253526-1-hawk@kernel.org Signed-off-by: Jakub Kicinski --- include/net/codel_impl.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/net/codel_impl.h b/include/net/codel_impl.h index 78a27ac73070..b2c359c6dd1b 100644 --- a/include/net/codel_impl.h +++ b/include/net/codel_impl.h @@ -158,6 +158,7 @@ static struct sk_buff *codel_dequeue(void *ctx, bool drop; if (!skb) { + vars->first_above_time = 0; vars->dropping = false; return skb; } From c4ea7d8907cf72b259bf70bd8c2e791e1c4ff70f Mon Sep 17 00:00:00 2001 From: Guangshuo Li Date: Tue, 24 Mar 2026 00:57:30 +0800 Subject: [PATCH 328/352] net: mana: fix use-after-free in add_adev() error path If auxiliary_device_add() fails, add_adev() jumps to add_fail and calls auxiliary_device_uninit(adev). The auxiliary device has its release callback set to adev_release(), which frees the containing struct mana_adev. Since adev is embedded in struct mana_adev, the subsequent fall-through to init_fail and access to adev->id may result in a use-after-free. Fix this by saving the allocated auxiliary device id in a local variable before calling auxiliary_device_add(), and use that saved id in the cleanup path after auxiliary_device_uninit(). Fixes: a69839d4327d ("net: mana: Add support for auxiliary device") Cc: stable@vger.kernel.org Reviewed-by: Long Li Signed-off-by: Guangshuo Li Link: https://patch.msgid.link/20260323165730.945365-1-lgs201920130244@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microsoft/mana/mana_en.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c index 9017e806ecda..dca62fb9a3a9 100644 --- a/drivers/net/ethernet/microsoft/mana/mana_en.c +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c @@ -3425,6 +3425,7 @@ static int add_adev(struct gdma_dev *gd, const char *name) struct auxiliary_device *adev; struct mana_adev *madev; int ret; + int id; madev = kzalloc_obj(*madev); if (!madev) @@ -3434,7 +3435,8 @@ static int add_adev(struct gdma_dev *gd, const char *name) ret = mana_adev_idx_alloc(); if (ret < 0) goto idx_fail; - adev->id = ret; + id = ret; + adev->id = id; adev->name = name; adev->dev.parent = gd->gdma_context->dev; @@ -3460,7 +3462,7 @@ static int add_adev(struct gdma_dev *gd, const char *name) auxiliary_device_uninit(adev); init_fail: - mana_adev_idx_free(adev->id); + mana_adev_idx_free(id); idx_fail: kfree(madev); From 2cdaff22ed26f1e619aa2b43f27bb84f2c6ef8f8 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Wed, 25 Mar 2026 02:55:48 +0100 Subject: [PATCH 329/352] dma-mapping: add missing `inline` for `dma_free_attrs` Under an UML build for an upcoming series [1], I got `-Wstatic-in-inline` for `dma_free_attrs`: BINDGEN rust/bindings/bindings_generated.rs - due to target missing In file included from rust/helpers/helpers.c:59: rust/helpers/dma.c:17:2: warning: static function 'dma_free_attrs' is used in an inline function with external linkage [-Wstatic-in-inline] 17 | dma_free_attrs(dev, size, cpu_addr, dma_handle, attrs); | ^ rust/helpers/dma.c:12:1: note: use 'static' to give inline function 'rust_helper_dma_free_attrs' internal linkage 12 | __rust_helper void rust_helper_dma_free_attrs(struct device *dev, size_t size, | ^ | static The issue is that `dma_free_attrs` was not marked `inline` when it was introduced alongside the rest of the stubs. Thus mark it. Fixes: ed6ccf10f24b ("dma-mapping: properly stub out the DMA API for !CONFIG_HAS_DMA") Closes: https://lore.kernel.org/rust-for-linux/20260322194616.89847-1-ojeda@kernel.org/ [1] Signed-off-by: Miguel Ojeda Signed-off-by: Marek Szyprowski Link: https://lore.kernel.org/r/20260325015548.70912-1-ojeda@kernel.org --- include/linux/dma-mapping.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h index 482b919f040f..99ef042ecdb4 100644 --- a/include/linux/dma-mapping.h +++ b/include/linux/dma-mapping.h @@ -255,8 +255,8 @@ static inline void *dma_alloc_attrs(struct device *dev, size_t size, { return NULL; } -static void dma_free_attrs(struct device *dev, size_t size, void *cpu_addr, - dma_addr_t dma_handle, unsigned long attrs) +static inline void dma_free_attrs(struct device *dev, size_t size, + void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs) { } static inline void *dmam_alloc_attrs(struct device *dev, size_t size, From 175b45ed343a9c547b5f45293d3ea08d38a7b6f4 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Sat, 14 Mar 2026 04:12:58 -0700 Subject: [PATCH 330/352] srcu: Use raw spinlocks so call_srcu() can be used under preempt_disable() Tree SRCU has used non-raw spinlocks for many years, motivated by a desire to avoid unnecessary real-time latency and the absence of any reason to use raw spinlocks. However, the recent use of SRCU in tracing as the underlying implementation of RCU Tasks Trace means that call_srcu() is invoked from preemption-disabled regions of code, which in turn requires that any locks acquired by call_srcu() or its callees must be raw spinlocks. This commit therefore converts SRCU's spinlocks to raw spinlocks. [boqun: Add Fixes tag] Reported-by: Kumar Kartikeya Dwivedi Fixes: c27cea4416a3 ("rcu: Re-implement RCU Tasks Trace in terms of SRCU-fast") Signed-off-by: Paul E. McKenney Signed-off-by: Boqun Feng Cc: Sebastian Andrzej Siewior --- include/linux/srcutree.h | 8 +- kernel/rcu/rcu.h | 9 ++ kernel/rcu/srcutree.c | 174 ++++++++++++++++----------------------- 3 files changed, 82 insertions(+), 109 deletions(-) diff --git a/include/linux/srcutree.h b/include/linux/srcutree.h index 958cb7ef41cb..dfb31d11ff05 100644 --- a/include/linux/srcutree.h +++ b/include/linux/srcutree.h @@ -34,7 +34,7 @@ struct srcu_data { /* Values: SRCU_READ_FLAVOR_.* */ /* Update-side state. */ - spinlock_t __private lock ____cacheline_internodealigned_in_smp; + raw_spinlock_t __private lock ____cacheline_internodealigned_in_smp; struct rcu_segcblist srcu_cblist; /* List of callbacks.*/ unsigned long srcu_gp_seq_needed; /* Furthest future GP needed. */ unsigned long srcu_gp_seq_needed_exp; /* Furthest future exp GP. */ @@ -55,7 +55,7 @@ struct srcu_data { * Node in SRCU combining tree, similar in function to rcu_data. */ struct srcu_node { - spinlock_t __private lock; + raw_spinlock_t __private lock; unsigned long srcu_have_cbs[4]; /* GP seq for children having CBs, but only */ /* if greater than ->srcu_gp_seq. */ unsigned long srcu_data_have_cbs[4]; /* Which srcu_data structs have CBs for given GP? */ @@ -74,7 +74,7 @@ struct srcu_usage { /* First node at each level. */ int srcu_size_state; /* Small-to-big transition state. */ struct mutex srcu_cb_mutex; /* Serialize CB preparation. */ - spinlock_t __private lock; /* Protect counters and size state. */ + raw_spinlock_t __private lock; /* Protect counters and size state. */ struct mutex srcu_gp_mutex; /* Serialize GP work. */ unsigned long srcu_gp_seq; /* Grace-period seq #. */ unsigned long srcu_gp_seq_needed; /* Latest gp_seq needed. */ @@ -156,7 +156,7 @@ struct srcu_struct { #define __SRCU_USAGE_INIT(name) \ { \ - .lock = __SPIN_LOCK_UNLOCKED(name.lock), \ + .lock = __RAW_SPIN_LOCK_UNLOCKED(name.lock), \ .srcu_gp_seq = SRCU_GP_SEQ_INITIAL_VAL, \ .srcu_gp_seq_needed = SRCU_GP_SEQ_INITIAL_VAL_WITH_STATE, \ .srcu_gp_seq_needed_exp = SRCU_GP_SEQ_INITIAL_VAL, \ diff --git a/kernel/rcu/rcu.h b/kernel/rcu/rcu.h index dc5d614b372c..9b10b57b79ad 100644 --- a/kernel/rcu/rcu.h +++ b/kernel/rcu/rcu.h @@ -502,6 +502,15 @@ do { \ ___locked; \ }) +#define raw_spin_trylock_irqsave_rcu_node(p, flags) \ +({ \ + bool ___locked = raw_spin_trylock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ + \ + if (___locked) \ + smp_mb__after_unlock_lock(); \ + ___locked; \ +}) + #define raw_lockdep_assert_held_rcu_node(p) \ lockdep_assert_held(&ACCESS_PRIVATE(p, lock)) diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index aef8e91ad33e..2328827f8775 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -77,42 +77,6 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay); static void process_srcu(struct work_struct *work); static void srcu_delay_timer(struct timer_list *t); -/* Wrappers for lock acquisition and release, see raw_spin_lock_rcu_node(). */ -#define spin_lock_rcu_node(p) \ -do { \ - spin_lock(&ACCESS_PRIVATE(p, lock)); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define spin_unlock_rcu_node(p) spin_unlock(&ACCESS_PRIVATE(p, lock)) - -#define spin_lock_irq_rcu_node(p) \ -do { \ - spin_lock_irq(&ACCESS_PRIVATE(p, lock)); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define spin_unlock_irq_rcu_node(p) \ - spin_unlock_irq(&ACCESS_PRIVATE(p, lock)) - -#define spin_lock_irqsave_rcu_node(p, flags) \ -do { \ - spin_lock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ - smp_mb__after_unlock_lock(); \ -} while (0) - -#define spin_trylock_irqsave_rcu_node(p, flags) \ -({ \ - bool ___locked = spin_trylock_irqsave(&ACCESS_PRIVATE(p, lock), flags); \ - \ - if (___locked) \ - smp_mb__after_unlock_lock(); \ - ___locked; \ -}) - -#define spin_unlock_irqrestore_rcu_node(p, flags) \ - spin_unlock_irqrestore(&ACCESS_PRIVATE(p, lock), flags) \ - /* * Initialize SRCU per-CPU data. Note that statically allocated * srcu_struct structures might already have srcu_read_lock() and @@ -131,7 +95,7 @@ static void init_srcu_struct_data(struct srcu_struct *ssp) */ for_each_possible_cpu(cpu) { sdp = per_cpu_ptr(ssp->sda, cpu); - spin_lock_init(&ACCESS_PRIVATE(sdp, lock)); + raw_spin_lock_init(&ACCESS_PRIVATE(sdp, lock)); rcu_segcblist_init(&sdp->srcu_cblist); sdp->srcu_cblist_invoking = false; sdp->srcu_gp_seq_needed = ssp->srcu_sup->srcu_gp_seq; @@ -186,7 +150,7 @@ static bool init_srcu_struct_nodes(struct srcu_struct *ssp, gfp_t gfp_flags) /* Each pass through this loop initializes one srcu_node structure. */ srcu_for_each_node_breadth_first(ssp, snp) { - spin_lock_init(&ACCESS_PRIVATE(snp, lock)); + raw_spin_lock_init(&ACCESS_PRIVATE(snp, lock)); BUILD_BUG_ON(ARRAY_SIZE(snp->srcu_have_cbs) != ARRAY_SIZE(snp->srcu_data_have_cbs)); for (i = 0; i < ARRAY_SIZE(snp->srcu_have_cbs); i++) { @@ -242,7 +206,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) if (!ssp->srcu_sup) return -ENOMEM; if (!is_static) - spin_lock_init(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); + raw_spin_lock_init(&ACCESS_PRIVATE(ssp->srcu_sup, lock)); ssp->srcu_sup->srcu_size_state = SRCU_SIZE_SMALL; ssp->srcu_sup->node = NULL; mutex_init(&ssp->srcu_sup->srcu_cb_mutex); @@ -394,20 +358,20 @@ static void srcu_transition_to_big(struct srcu_struct *ssp) /* Double-checked locking on ->srcu_size-state. */ if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) return; - spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); if (smp_load_acquire(&ssp->srcu_sup->srcu_size_state) != SRCU_SIZE_SMALL) { - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } __srcu_transition_to_big(ssp); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* * Check to see if the just-encountered contention event justifies * a transition to SRCU_SIZE_BIG. */ -static void spin_lock_irqsave_check_contention(struct srcu_struct *ssp) +static void raw_spin_lock_irqsave_check_contention(struct srcu_struct *ssp) { unsigned long j; @@ -429,16 +393,16 @@ static void spin_lock_irqsave_check_contention(struct srcu_struct *ssp) * to SRCU_SIZE_BIG. But only if the srcutree.convert_to_big module * parameter permits this. */ -static void spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned long *flags) +static void raw_spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned long *flags) { struct srcu_struct *ssp = sdp->ssp; - if (spin_trylock_irqsave_rcu_node(sdp, *flags)) + if (raw_spin_trylock_irqsave_rcu_node(sdp, *flags)) return; - spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); - spin_lock_irqsave_check_contention(ssp); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, *flags); - spin_lock_irqsave_rcu_node(sdp, *flags); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); + raw_spin_lock_irqsave_check_contention(ssp); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, *flags); + raw_spin_lock_irqsave_rcu_node(sdp, *flags); } /* @@ -447,12 +411,12 @@ static void spin_lock_irqsave_sdp_contention(struct srcu_data *sdp, unsigned lon * to SRCU_SIZE_BIG. But only if the srcutree.convert_to_big module * parameter permits this. */ -static void spin_lock_irqsave_ssp_contention(struct srcu_struct *ssp, unsigned long *flags) +static void raw_spin_lock_irqsave_ssp_contention(struct srcu_struct *ssp, unsigned long *flags) { - if (spin_trylock_irqsave_rcu_node(ssp->srcu_sup, *flags)) + if (raw_spin_trylock_irqsave_rcu_node(ssp->srcu_sup, *flags)) return; - spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); - spin_lock_irqsave_check_contention(ssp); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, *flags); + raw_spin_lock_irqsave_check_contention(ssp); } /* @@ -470,13 +434,13 @@ static void check_init_srcu_struct(struct srcu_struct *ssp) /* The smp_load_acquire() pairs with the smp_store_release(). */ if (!rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq_needed))) /*^^^*/ return; /* Already initialized. */ - spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); if (!rcu_seq_state(ssp->srcu_sup->srcu_gp_seq_needed)) { - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); return; } init_srcu_struct_fields(ssp, true); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -742,9 +706,9 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) unsigned long delay; struct srcu_usage *sup = ssp->srcu_sup; - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); delay = srcu_get_delay(ssp); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (WARN_ON(!delay)) return; /* Just leak it! */ if (WARN_ON(srcu_readers_active(ssp))) @@ -960,7 +924,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) mutex_lock(&sup->srcu_cb_mutex); /* End the current grace period. */ - spin_lock_irq_rcu_node(sup); + raw_spin_lock_irq_rcu_node(sup); idx = rcu_seq_state(sup->srcu_gp_seq); WARN_ON_ONCE(idx != SRCU_STATE_SCAN2); if (srcu_gp_is_expedited(ssp)) @@ -971,7 +935,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) gpseq = rcu_seq_current(&sup->srcu_gp_seq); if (ULONG_CMP_LT(sup->srcu_gp_seq_needed_exp, gpseq)) WRITE_ONCE(sup->srcu_gp_seq_needed_exp, gpseq); - spin_unlock_irq_rcu_node(sup); + raw_spin_unlock_irq_rcu_node(sup); mutex_unlock(&sup->srcu_gp_mutex); /* A new grace period can start at this point. But only one. */ @@ -983,7 +947,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) } else { idx = rcu_seq_ctr(gpseq) % ARRAY_SIZE(snp->srcu_have_cbs); srcu_for_each_node_breadth_first(ssp, snp) { - spin_lock_irq_rcu_node(snp); + raw_spin_lock_irq_rcu_node(snp); cbs = false; last_lvl = snp >= sup->level[rcu_num_lvls - 1]; if (last_lvl) @@ -998,7 +962,7 @@ static void srcu_gp_end(struct srcu_struct *ssp) else mask = snp->srcu_data_have_cbs[idx]; snp->srcu_data_have_cbs[idx] = 0; - spin_unlock_irq_rcu_node(snp); + raw_spin_unlock_irq_rcu_node(snp); if (cbs) srcu_schedule_cbs_snp(ssp, snp, mask, cbdelay); } @@ -1008,27 +972,27 @@ static void srcu_gp_end(struct srcu_struct *ssp) if (!(gpseq & counter_wrap_check)) for_each_possible_cpu(cpu) { sdp = per_cpu_ptr(ssp->sda, cpu); - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); if (ULONG_CMP_GE(gpseq, sdp->srcu_gp_seq_needed + 100)) sdp->srcu_gp_seq_needed = gpseq; if (ULONG_CMP_GE(gpseq, sdp->srcu_gp_seq_needed_exp + 100)) sdp->srcu_gp_seq_needed_exp = gpseq; - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); } /* Callback initiation done, allow grace periods after next. */ mutex_unlock(&sup->srcu_cb_mutex); /* Start a new grace period if needed. */ - spin_lock_irq_rcu_node(sup); + raw_spin_lock_irq_rcu_node(sup); gpseq = rcu_seq_current(&sup->srcu_gp_seq); if (!rcu_seq_state(gpseq) && ULONG_CMP_LT(gpseq, sup->srcu_gp_seq_needed)) { srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(sup); + raw_spin_unlock_irq_rcu_node(sup); srcu_reschedule(ssp, 0); } else { - spin_unlock_irq_rcu_node(sup); + raw_spin_unlock_irq_rcu_node(sup); } /* Transition to big if needed. */ @@ -1059,19 +1023,19 @@ static void srcu_funnel_exp_start(struct srcu_struct *ssp, struct srcu_node *snp if (WARN_ON_ONCE(rcu_seq_done(&ssp->srcu_sup->srcu_gp_seq, s)) || (!srcu_invl_snp_seq(sgsne) && ULONG_CMP_GE(sgsne, s))) return; - spin_lock_irqsave_rcu_node(snp, flags); + raw_spin_lock_irqsave_rcu_node(snp, flags); sgsne = snp->srcu_gp_seq_needed_exp; if (!srcu_invl_snp_seq(sgsne) && ULONG_CMP_GE(sgsne, s)) { - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); return; } WRITE_ONCE(snp->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); } - spin_lock_irqsave_ssp_contention(ssp, &flags); + raw_spin_lock_irqsave_ssp_contention(ssp, &flags); if (ULONG_CMP_LT(ssp->srcu_sup->srcu_gp_seq_needed_exp, s)) WRITE_ONCE(ssp->srcu_sup->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); } /* @@ -1109,12 +1073,12 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, for (snp = snp_leaf; snp != NULL; snp = snp->srcu_parent) { if (WARN_ON_ONCE(rcu_seq_done(&sup->srcu_gp_seq, s)) && snp != snp_leaf) return; /* GP already done and CBs recorded. */ - spin_lock_irqsave_rcu_node(snp, flags); + raw_spin_lock_irqsave_rcu_node(snp, flags); snp_seq = snp->srcu_have_cbs[idx]; if (!srcu_invl_snp_seq(snp_seq) && ULONG_CMP_GE(snp_seq, s)) { if (snp == snp_leaf && snp_seq == s) snp->srcu_data_have_cbs[idx] |= sdp->grpmask; - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); if (snp == snp_leaf && snp_seq != s) { srcu_schedule_cbs_sdp(sdp, do_norm ? SRCU_INTERVAL : 0); return; @@ -1129,11 +1093,11 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, sgsne = snp->srcu_gp_seq_needed_exp; if (!do_norm && (srcu_invl_snp_seq(sgsne) || ULONG_CMP_LT(sgsne, s))) WRITE_ONCE(snp->srcu_gp_seq_needed_exp, s); - spin_unlock_irqrestore_rcu_node(snp, flags); + raw_spin_unlock_irqrestore_rcu_node(snp, flags); } /* Top of tree, must ensure the grace period will be started. */ - spin_lock_irqsave_ssp_contention(ssp, &flags); + raw_spin_lock_irqsave_ssp_contention(ssp, &flags); if (ULONG_CMP_LT(sup->srcu_gp_seq_needed, s)) { /* * Record need for grace period s. Pair with load @@ -1160,7 +1124,7 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, else if (list_empty(&sup->work.work.entry)) list_add(&sup->work.work.entry, &srcu_boot_list); } - spin_unlock_irqrestore_rcu_node(sup, flags); + raw_spin_unlock_irqrestore_rcu_node(sup, flags); } /* @@ -1172,9 +1136,9 @@ static bool try_check_zero(struct srcu_struct *ssp, int idx, int trycount) { unsigned long curdelay; - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); curdelay = !srcu_get_delay(ssp); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); for (;;) { if (srcu_readers_active_idx_check(ssp, idx)) @@ -1285,12 +1249,12 @@ static bool srcu_should_expedite(struct srcu_struct *ssp) return false; /* If the local srcu_data structure has callbacks, not idle. */ sdp = raw_cpu_ptr(ssp->sda); - spin_lock_irqsave_rcu_node(sdp, flags); + raw_spin_lock_irqsave_rcu_node(sdp, flags); if (rcu_segcblist_pend_cbs(&sdp->srcu_cblist)) { - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); return false; /* Callbacks already present, so not idle. */ } - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); /* * No local callbacks, so probabilistically probe global state. @@ -1350,7 +1314,7 @@ static unsigned long srcu_gp_start_if_needed(struct srcu_struct *ssp, sdp = per_cpu_ptr(ssp->sda, get_boot_cpu_id()); else sdp = raw_cpu_ptr(ssp->sda); - spin_lock_irqsave_sdp_contention(sdp, &flags); + raw_spin_lock_irqsave_sdp_contention(sdp, &flags); if (rhp) rcu_segcblist_enqueue(&sdp->srcu_cblist, rhp); /* @@ -1410,7 +1374,7 @@ static unsigned long srcu_gp_start_if_needed(struct srcu_struct *ssp, sdp->srcu_gp_seq_needed_exp = s; needexp = true; } - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); /* Ensure that snp node tree is fully initialized before traversing it */ if (ss_state < SRCU_SIZE_WAIT_BARRIER) @@ -1522,7 +1486,7 @@ static void __synchronize_srcu(struct srcu_struct *ssp, bool do_norm) /* * Make sure that later code is ordered after the SRCU grace - * period. This pairs with the spin_lock_irq_rcu_node() + * period. This pairs with the raw_spin_lock_irq_rcu_node() * in srcu_invoke_callbacks(). Unlike Tree RCU, this is needed * because the current CPU might have been totally uninvolved with * (and thus unordered against) that grace period. @@ -1701,7 +1665,7 @@ static void srcu_barrier_cb(struct rcu_head *rhp) */ static void srcu_barrier_one_cpu(struct srcu_struct *ssp, struct srcu_data *sdp) { - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); atomic_inc(&ssp->srcu_sup->srcu_barrier_cpu_cnt); sdp->srcu_barrier_head.func = srcu_barrier_cb; debug_rcu_head_queue(&sdp->srcu_barrier_head); @@ -1710,7 +1674,7 @@ static void srcu_barrier_one_cpu(struct srcu_struct *ssp, struct srcu_data *sdp) debug_rcu_head_unqueue(&sdp->srcu_barrier_head); atomic_dec(&ssp->srcu_sup->srcu_barrier_cpu_cnt); } - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); } /** @@ -1761,7 +1725,7 @@ static void srcu_expedite_current_cb(struct rcu_head *rhp) bool needcb = false; struct srcu_data *sdp = container_of(rhp, struct srcu_data, srcu_ec_head); - spin_lock_irqsave_sdp_contention(sdp, &flags); + raw_spin_lock_irqsave_sdp_contention(sdp, &flags); if (sdp->srcu_ec_state == SRCU_EC_IDLE) { WARN_ON_ONCE(1); } else if (sdp->srcu_ec_state == SRCU_EC_PENDING) { @@ -1771,7 +1735,7 @@ static void srcu_expedite_current_cb(struct rcu_head *rhp) sdp->srcu_ec_state = SRCU_EC_PENDING; needcb = true; } - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); // If needed, requeue ourselves as an expedited SRCU callback. if (needcb) __call_srcu(sdp->ssp, &sdp->srcu_ec_head, srcu_expedite_current_cb, false); @@ -1795,7 +1759,7 @@ void srcu_expedite_current(struct srcu_struct *ssp) migrate_disable(); sdp = this_cpu_ptr(ssp->sda); - spin_lock_irqsave_sdp_contention(sdp, &flags); + raw_spin_lock_irqsave_sdp_contention(sdp, &flags); if (sdp->srcu_ec_state == SRCU_EC_IDLE) { sdp->srcu_ec_state = SRCU_EC_PENDING; needcb = true; @@ -1804,7 +1768,7 @@ void srcu_expedite_current(struct srcu_struct *ssp) } else { WARN_ON_ONCE(sdp->srcu_ec_state != SRCU_EC_REPOST); } - spin_unlock_irqrestore_rcu_node(sdp, flags); + raw_spin_unlock_irqrestore_rcu_node(sdp, flags); // If needed, queue an expedited SRCU callback. if (needcb) __call_srcu(ssp, &sdp->srcu_ec_head, srcu_expedite_current_cb, false); @@ -1848,17 +1812,17 @@ static void srcu_advance_state(struct srcu_struct *ssp) */ idx = rcu_seq_state(smp_load_acquire(&ssp->srcu_sup->srcu_gp_seq)); /* ^^^ */ if (idx == SRCU_STATE_IDLE) { - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); if (ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)) { WARN_ON_ONCE(rcu_seq_state(ssp->srcu_sup->srcu_gp_seq)); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; } idx = rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)); if (idx == SRCU_STATE_IDLE) srcu_gp_start(ssp); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (idx != SRCU_STATE_IDLE) { mutex_unlock(&ssp->srcu_sup->srcu_gp_mutex); return; /* Someone else started the grace period. */ @@ -1872,10 +1836,10 @@ static void srcu_advance_state(struct srcu_struct *ssp) return; /* readers present, retry later. */ } srcu_flip(ssp); - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); rcu_seq_set_state(&ssp->srcu_sup->srcu_gp_seq, SRCU_STATE_SCAN2); ssp->srcu_sup->srcu_n_exp_nodelay = 0; - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); } if (rcu_seq_state(READ_ONCE(ssp->srcu_sup->srcu_gp_seq)) == SRCU_STATE_SCAN2) { @@ -1913,7 +1877,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) ssp = sdp->ssp; rcu_cblist_init(&ready_cbs); - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); WARN_ON_ONCE(!rcu_segcblist_segempty(&sdp->srcu_cblist, RCU_NEXT_TAIL)); rcu_segcblist_advance(&sdp->srcu_cblist, rcu_seq_current(&ssp->srcu_sup->srcu_gp_seq)); @@ -1924,7 +1888,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) */ if (sdp->srcu_cblist_invoking || !rcu_segcblist_ready_cbs(&sdp->srcu_cblist)) { - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); return; /* Someone else on the job or nothing to do. */ } @@ -1932,7 +1896,7 @@ static void srcu_invoke_callbacks(struct work_struct *work) sdp->srcu_cblist_invoking = true; rcu_segcblist_extract_done_cbs(&sdp->srcu_cblist, &ready_cbs); len = ready_cbs.len; - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); rhp = rcu_cblist_dequeue(&ready_cbs); for (; rhp != NULL; rhp = rcu_cblist_dequeue(&ready_cbs)) { debug_rcu_head_unqueue(rhp); @@ -1947,11 +1911,11 @@ static void srcu_invoke_callbacks(struct work_struct *work) * Update counts, accelerate new callbacks, and if needed, * schedule another round of callback invocation. */ - spin_lock_irq_rcu_node(sdp); + raw_spin_lock_irq_rcu_node(sdp); rcu_segcblist_add_len(&sdp->srcu_cblist, -len); sdp->srcu_cblist_invoking = false; more = rcu_segcblist_ready_cbs(&sdp->srcu_cblist); - spin_unlock_irq_rcu_node(sdp); + raw_spin_unlock_irq_rcu_node(sdp); /* An SRCU barrier or callbacks from previous nesting work pending */ if (more) srcu_schedule_cbs_sdp(sdp, 0); @@ -1965,7 +1929,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) { bool pushgp = true; - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); if (ULONG_CMP_GE(ssp->srcu_sup->srcu_gp_seq, ssp->srcu_sup->srcu_gp_seq_needed)) { if (!WARN_ON_ONCE(rcu_seq_state(ssp->srcu_sup->srcu_gp_seq))) { /* All requests fulfilled, time to go idle. */ @@ -1975,7 +1939,7 @@ static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay) /* Outstanding request and no GP. Start one. */ srcu_gp_start(ssp); } - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (pushgp) queue_delayed_work(rcu_gp_wq, &ssp->srcu_sup->work, delay); @@ -1995,9 +1959,9 @@ static void process_srcu(struct work_struct *work) ssp = sup->srcu_ssp; srcu_advance_state(ssp); - spin_lock_irq_rcu_node(ssp->srcu_sup); + raw_spin_lock_irq_rcu_node(ssp->srcu_sup); curdelay = srcu_get_delay(ssp); - spin_unlock_irq_rcu_node(ssp->srcu_sup); + raw_spin_unlock_irq_rcu_node(ssp->srcu_sup); if (curdelay) { WRITE_ONCE(sup->reschedule_count, 0); } else { From 61bbcfb50514a8a94e035a7349697a3790ab4783 Mon Sep 17 00:00:00 2001 From: "Paul E. McKenney" Date: Fri, 20 Mar 2026 20:29:20 -0700 Subject: [PATCH 331/352] srcu: Push srcu_node allocation to GP when non-preemptible When the srcutree.convert_to_big and srcutree.big_cpu_lim kernel boot parameters specify initialization-time allocation of the srcu_node tree for statically allocated srcu_struct structures (for example, in DEFINE_SRCU() at build time instead of init_srcu_struct() at runtime), init_srcu_struct_nodes() will attempt to dynamically allocate this tree at the first run-time update-side use of this srcu_struct structure, but while holding a raw spinlock. Because the memory allocator can acquire non-raw spinlocks, this can result in lockdep splats. This commit therefore uses the same SRCU_SIZE_ALLOC trick that is used when the first run-time update-side use of this srcu_struct structure happens before srcu_init() is called. The actual allocation then takes place from workqueue context at the ends of upcoming SRCU grace periods. [boqun: Adjust the sha1 of the Fixes tag] Fixes: 175b45ed343a ("srcu: Use raw spinlocks so call_srcu() can be used under preempt_disable()") Signed-off-by: Paul E. McKenney Signed-off-by: Boqun Feng --- kernel/rcu/srcutree.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 2328827f8775..678bd9a73875 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -227,9 +227,12 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) ssp->srcu_sup->srcu_gp_seq_needed_exp = SRCU_GP_SEQ_INITIAL_VAL; ssp->srcu_sup->srcu_last_gp_end = ktime_get_mono_fast_ns(); if (READ_ONCE(ssp->srcu_sup->srcu_size_state) == SRCU_SIZE_SMALL && SRCU_SIZING_IS_INIT()) { - if (!init_srcu_struct_nodes(ssp, is_static ? GFP_ATOMIC : GFP_KERNEL)) + if (!preemptible()) + WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_ALLOC); + else if (init_srcu_struct_nodes(ssp, GFP_KERNEL)) + WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); + else goto err_free_sda; - WRITE_ONCE(ssp->srcu_sup->srcu_size_state, SRCU_SIZE_BIG); } ssp->srcu_sup->srcu_ssp = ssp; smp_store_release(&ssp->srcu_sup->srcu_gp_seq_needed, From 7c405fb3279b39244b260b54f1bd6488689ae235 Mon Sep 17 00:00:00 2001 From: Boqun Feng Date: Wed, 18 Mar 2026 17:56:21 -0700 Subject: [PATCH 332/352] rcu: Use an intermediate irq_work to start process_srcu() Since commit c27cea4416a3 ("rcu: Re-implement RCU Tasks Trace in terms of SRCU-fast") we switched to SRCU in BPF. However as BPF instrument can happen basically everywhere (including where a scheduler lock is held), call_srcu() now needs to avoid acquiring scheduler lock because otherwise it could cause deadlock [1]. Fix this by following what the previous RCU Tasks Trace did: using an irq_work to delay the queuing of the work to start process_srcu(). [boqun: Apply Joel's feedback] [boqun: Apply Andrea's test feedback] Reported-by: Andrea Righi Closes: https://lore.kernel.org/all/abjzvz_tL_siV17s@gpd4/ Fixes: commit c27cea4416a3 ("rcu: Re-implement RCU Tasks Trace in terms of SRCU-fast") Link: https://lore.kernel.org/rcu/3c4c5a29-24ea-492d-aeee-e0d9605b4183@nvidia.com/ [1] Suggested-by: Zqiang Tested-by: Andrea Righi Tested-by: Paul E. McKenney Tested-by: Joel Fernandes Signed-off-by: Boqun Feng --- include/linux/srcutree.h | 1 + kernel/rcu/srcutree.c | 30 ++++++++++++++++++++++++++++-- 2 files changed, 29 insertions(+), 2 deletions(-) diff --git a/include/linux/srcutree.h b/include/linux/srcutree.h index dfb31d11ff05..be76fa4fc170 100644 --- a/include/linux/srcutree.h +++ b/include/linux/srcutree.h @@ -95,6 +95,7 @@ struct srcu_usage { unsigned long reschedule_jiffies; unsigned long reschedule_count; struct delayed_work work; + struct irq_work irq_work; struct srcu_struct *srcu_ssp; }; diff --git a/kernel/rcu/srcutree.c b/kernel/rcu/srcutree.c index 678bd9a73875..0d01cd8c4b4a 100644 --- a/kernel/rcu/srcutree.c +++ b/kernel/rcu/srcutree.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include #include @@ -75,6 +76,7 @@ static bool __read_mostly srcu_init_done; static void srcu_invoke_callbacks(struct work_struct *work); static void srcu_reschedule(struct srcu_struct *ssp, unsigned long delay); static void process_srcu(struct work_struct *work); +static void srcu_irq_work(struct irq_work *work); static void srcu_delay_timer(struct timer_list *t); /* @@ -216,6 +218,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp, bool is_static) mutex_init(&ssp->srcu_sup->srcu_barrier_mutex); atomic_set(&ssp->srcu_sup->srcu_barrier_cpu_cnt, 0); INIT_DELAYED_WORK(&ssp->srcu_sup->work, process_srcu); + init_irq_work(&ssp->srcu_sup->irq_work, srcu_irq_work); ssp->srcu_sup->sda_is_static = is_static; if (!is_static) { ssp->sda = alloc_percpu(struct srcu_data); @@ -716,6 +719,8 @@ void cleanup_srcu_struct(struct srcu_struct *ssp) return; /* Just leak it! */ if (WARN_ON(srcu_readers_active(ssp))) return; /* Just leak it! */ + /* Wait for irq_work to finish first as it may queue a new work. */ + irq_work_sync(&sup->irq_work); flush_delayed_work(&sup->work); for_each_possible_cpu(cpu) { struct srcu_data *sdp = per_cpu_ptr(ssp->sda, cpu); @@ -1121,9 +1126,13 @@ static void srcu_funnel_gp_start(struct srcu_struct *ssp, struct srcu_data *sdp, // it isn't. And it does not have to be. After all, it // can only be executed during early boot when there is only // the one boot CPU running with interrupts still disabled. + // + // Use an irq_work here to avoid acquiring runqueue lock with + // srcu rcu_node::lock held. BPF instrument could introduce the + // opposite dependency, hence we need to break the possible + // locking dependency here. if (likely(srcu_init_done)) - queue_delayed_work(rcu_gp_wq, &sup->work, - !!srcu_get_delay(ssp)); + irq_work_queue(&sup->irq_work); else if (list_empty(&sup->work.work.entry)) list_add(&sup->work.work.entry, &srcu_boot_list); } @@ -1982,6 +1991,23 @@ static void process_srcu(struct work_struct *work) srcu_reschedule(ssp, curdelay); } +static void srcu_irq_work(struct irq_work *work) +{ + struct srcu_struct *ssp; + struct srcu_usage *sup; + unsigned long delay; + unsigned long flags; + + sup = container_of(work, struct srcu_usage, irq_work); + ssp = sup->srcu_ssp; + + raw_spin_lock_irqsave_rcu_node(ssp->srcu_sup, flags); + delay = srcu_get_delay(ssp); + raw_spin_unlock_irqrestore_rcu_node(ssp->srcu_sup, flags); + + queue_delayed_work(rcu_gp_wq, &sup->work, !!delay); +} + void srcutorture_get_gp_data(struct srcu_struct *ssp, int *flags, unsigned long *gp_seq) { From a6fc88b22bc8d12ad52e8412c667ec0f5bf055af Mon Sep 17 00:00:00 2001 From: Joel Fernandes Date: Mon, 23 Mar 2026 20:14:18 -0400 Subject: [PATCH 333/352] srcu: Use irq_work to start GP in tiny SRCU Tiny SRCU's srcu_gp_start_if_needed() directly calls schedule_work(), which acquires the workqueue pool->lock. This causes a lockdep splat when call_srcu() is called with a scheduler lock held, due to: call_srcu() [holding pi_lock] srcu_gp_start_if_needed() schedule_work() -> pool->lock workqueue_init() / create_worker() [holding pool->lock] wake_up_process() -> try_to_wake_up() -> pi_lock Also add irq_work_sync() to cleanup_srcu_struct() to prevent a use-after-free if a queued irq_work fires after cleanup begins. Tested with rcutorture SRCU-T and no lockdep warnings. [ Thanks to Boqun for similar fix in patch "rcu: Use an intermediate irq_work to start process_srcu()" ] Signed-off-by: Joel Fernandes Reviewed-by: Paul E. McKenney Signed-off-by: Boqun Feng --- include/linux/srcutiny.h | 4 ++++ kernel/rcu/srcutiny.c | 19 ++++++++++++++++++- 2 files changed, 22 insertions(+), 1 deletion(-) diff --git a/include/linux/srcutiny.h b/include/linux/srcutiny.h index dec7cbe015aa..905b629e8fa3 100644 --- a/include/linux/srcutiny.h +++ b/include/linux/srcutiny.h @@ -11,6 +11,7 @@ #ifndef _LINUX_SRCU_TINY_H #define _LINUX_SRCU_TINY_H +#include #include struct srcu_struct { @@ -24,18 +25,21 @@ struct srcu_struct { struct rcu_head *srcu_cb_head; /* Pending callbacks: Head. */ struct rcu_head **srcu_cb_tail; /* Pending callbacks: Tail. */ struct work_struct srcu_work; /* For driving grace periods. */ + struct irq_work srcu_irq_work; /* Defer schedule_work() to irq work. */ #ifdef CONFIG_DEBUG_LOCK_ALLOC struct lockdep_map dep_map; #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */ }; void srcu_drive_gp(struct work_struct *wp); +void srcu_tiny_irq_work(struct irq_work *irq_work); #define __SRCU_STRUCT_INIT(name, __ignored, ___ignored, ____ignored) \ { \ .srcu_wq = __SWAIT_QUEUE_HEAD_INITIALIZER(name.srcu_wq), \ .srcu_cb_tail = &name.srcu_cb_head, \ .srcu_work = __WORK_INITIALIZER(name.srcu_work, srcu_drive_gp), \ + .srcu_irq_work = { .func = srcu_tiny_irq_work }, \ __SRCU_DEP_MAP_INIT(name) \ } diff --git a/kernel/rcu/srcutiny.c b/kernel/rcu/srcutiny.c index 3450c3751ef7..a2e2d516e51b 100644 --- a/kernel/rcu/srcutiny.c +++ b/kernel/rcu/srcutiny.c @@ -9,6 +9,7 @@ */ #include +#include #include #include #include @@ -41,6 +42,7 @@ static int init_srcu_struct_fields(struct srcu_struct *ssp) ssp->srcu_idx_max = 0; INIT_WORK(&ssp->srcu_work, srcu_drive_gp); INIT_LIST_HEAD(&ssp->srcu_work.entry); + init_irq_work(&ssp->srcu_irq_work, srcu_tiny_irq_work); return 0; } @@ -84,6 +86,7 @@ EXPORT_SYMBOL_GPL(init_srcu_struct); void cleanup_srcu_struct(struct srcu_struct *ssp) { WARN_ON(ssp->srcu_lock_nesting[0] || ssp->srcu_lock_nesting[1]); + irq_work_sync(&ssp->srcu_irq_work); flush_work(&ssp->srcu_work); WARN_ON(ssp->srcu_gp_running); WARN_ON(ssp->srcu_gp_waiting); @@ -177,6 +180,20 @@ void srcu_drive_gp(struct work_struct *wp) } EXPORT_SYMBOL_GPL(srcu_drive_gp); +/* + * Use an irq_work to defer schedule_work() to avoid acquiring the workqueue + * pool->lock while the caller might hold scheduler locks, causing lockdep + * splats due to workqueue_init() doing a wakeup. + */ +void srcu_tiny_irq_work(struct irq_work *irq_work) +{ + struct srcu_struct *ssp; + + ssp = container_of(irq_work, struct srcu_struct, srcu_irq_work); + schedule_work(&ssp->srcu_work); +} +EXPORT_SYMBOL_GPL(srcu_tiny_irq_work); + static void srcu_gp_start_if_needed(struct srcu_struct *ssp) { unsigned long cookie; @@ -189,7 +206,7 @@ static void srcu_gp_start_if_needed(struct srcu_struct *ssp) WRITE_ONCE(ssp->srcu_idx_max, cookie); if (!READ_ONCE(ssp->srcu_gp_running)) { if (likely(srcu_init_done)) - schedule_work(&ssp->srcu_work); + irq_work_queue(&ssp->srcu_irq_work); else if (list_empty(&ssp->srcu_work.entry)) list_add(&ssp->srcu_work.entry, &srcu_boot_list); } From f39f905e55f529b036321220af1ba4f4085564a5 Mon Sep 17 00:00:00 2001 From: Zhang Chen Date: Thu, 19 Mar 2026 17:32:11 +0800 Subject: [PATCH 334/352] Bluetooth: L2CAP: Fix send LE flow credits in ACL link When the L2CAP channel mode is L2CAP_MODE_ERTM/L2CAP_MODE_STREAMING, l2cap_publish_rx_avail will be called and le flow credits will be sent in l2cap_chan_rx_avail, even though the link type is ACL. The logs in question as follows: > ACL Data RX: Handle 129 flags 0x02 dlen 12 L2CAP: Unknown (0x16) ident 4 len 4 40 00 ed 05 < ACL Data TX: Handle 129 flags 0x00 dlen 10 L2CAP: Command Reject (0x01) ident 4 len 2 Reason: Command not understood (0x0000) Bluetooth: Unknown BR/EDR signaling command 0x16 Bluetooth: Wrong link type (-22) Fixes: ce60b9231b66 ("Bluetooth: compute LE flow credits based on recvbuf space") Signed-off-by: Zhang Chen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 3de3e3c8e966..6fd884203dc6 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -6630,6 +6630,10 @@ static void l2cap_chan_le_send_credits(struct l2cap_chan *chan) struct l2cap_le_credits pkt; u16 return_credits = l2cap_le_rx_credits(chan); + if (chan->mode != L2CAP_MODE_LE_FLOWCTL && + chan->mode != L2CAP_MODE_EXT_FLOWCTL) + return; + if (chan->rx_credits >= return_credits) return; From 94d8e6fe5d0818e9300e514e095a200bd5ff93ae Mon Sep 17 00:00:00 2001 From: Cen Zhang Date: Wed, 18 Mar 2026 20:54:03 +0800 Subject: [PATCH 335/352] Bluetooth: btintel: serialize btintel_hw_error() with hci_req_sync_lock btintel_hw_error() issues two __hci_cmd_sync() calls (HCI_OP_RESET and Intel exception-info retrieval) without holding hci_req_sync_lock(). This lets it race against hci_dev_do_close() -> btintel_shutdown_combined(), which also runs __hci_cmd_sync() under the same lock. When both paths manipulate hdev->req_status/req_rsp concurrently, the close path may free the response skb first, and the still-running hw_error path hits a slab-use-after-free in kfree_skb(). Wrap the whole recovery sequence in hci_req_sync_lock/unlock so it is serialized with every other synchronous HCI command issuer. Below is the data race report and the kasan report: BUG: data-race in __hci_cmd_sync_sk / btintel_shutdown_combined read of hdev->req_rsp at net/bluetooth/hci_sync.c:199 by task kworker/u17:1/83: __hci_cmd_sync_sk+0x12f2/0x1c30 net/bluetooth/hci_sync.c:200 __hci_cmd_sync+0x55/0x80 net/bluetooth/hci_sync.c:223 btintel_hw_error+0x114/0x670 drivers/bluetooth/btintel.c:254 hci_error_reset+0x348/0xa30 net/bluetooth/hci_core.c:1030 write/free by task ioctl/22580: btintel_shutdown_combined+0xd0/0x360 drivers/bluetooth/btintel.c:3648 hci_dev_close_sync+0x9ae/0x2c10 net/bluetooth/hci_sync.c:5246 hci_dev_do_close+0x232/0x460 net/bluetooth/hci_core.c:526 BUG: KASAN: slab-use-after-free in sk_skb_reason_drop+0x43/0x380 net/core/skbuff.c:1202 Read of size 4 at addr ffff888144a738dc by task kworker/u17:1/83: __hci_cmd_sync_sk+0x12f2/0x1c30 net/bluetooth/hci_sync.c:200 __hci_cmd_sync+0x55/0x80 net/bluetooth/hci_sync.c:223 btintel_hw_error+0x186/0x670 drivers/bluetooth/btintel.c:260 Fixes: 973bb97e5aee ("Bluetooth: btintel: Add generic function for handling hardware errors") Signed-off-by: Cen Zhang Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btintel.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/drivers/bluetooth/btintel.c b/drivers/bluetooth/btintel.c index 246b6205c5e0..ab146894ba4e 100644 --- a/drivers/bluetooth/btintel.c +++ b/drivers/bluetooth/btintel.c @@ -251,11 +251,13 @@ void btintel_hw_error(struct hci_dev *hdev, u8 code) bt_dev_err(hdev, "Hardware error 0x%2.2x", code); + hci_req_sync_lock(hdev); + skb = __hci_cmd_sync(hdev, HCI_OP_RESET, 0, NULL, HCI_INIT_TIMEOUT); if (IS_ERR(skb)) { bt_dev_err(hdev, "Reset after hardware error failed (%ld)", PTR_ERR(skb)); - return; + goto unlock; } kfree_skb(skb); @@ -263,18 +265,21 @@ void btintel_hw_error(struct hci_dev *hdev, u8 code) if (IS_ERR(skb)) { bt_dev_err(hdev, "Retrieving Intel exception info failed (%ld)", PTR_ERR(skb)); - return; + goto unlock; } if (skb->len != 13) { bt_dev_err(hdev, "Exception info size mismatch"); kfree_skb(skb); - return; + goto unlock; } bt_dev_err(hdev, "Exception info %s", (char *)(skb->data + 1)); kfree_skb(skb); + +unlock: + hci_req_sync_unlock(hdev); } EXPORT_SYMBOL_GPL(btintel_hw_error); From 00fdebbbc557a2fc21321ff2eaa22fd70c078608 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Fri, 20 Mar 2026 20:01:26 +0900 Subject: [PATCH 336/352] Bluetooth: L2CAP: Fix deadlock in l2cap_conn_del() l2cap_conn_del() calls cancel_delayed_work_sync() for both info_timer and id_addr_timer while holding conn->lock. However, the work functions l2cap_info_timeout() and l2cap_conn_update_id_addr() both acquire conn->lock, creating a potential AB-BA deadlock if the work is already executing when l2cap_conn_del() takes the lock. Move the work cancellations before acquiring conn->lock and use disable_delayed_work_sync() to additionally prevent the works from being rearmed after cancellation, consistent with the pattern used in hci_conn_del(). Fixes: ab4eedb790ca ("Bluetooth: L2CAP: Fix corrupted list in hci_chan_del") Signed-off-by: Hyunwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 6fd884203dc6..2603c98d7ed1 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -1771,6 +1771,9 @@ static void l2cap_conn_del(struct hci_conn *hcon, int err) BT_DBG("hcon %p conn %p, err %d", hcon, conn, err); + disable_delayed_work_sync(&conn->info_timer); + disable_delayed_work_sync(&conn->id_addr_timer); + mutex_lock(&conn->lock); kfree_skb(conn->rx_skb); @@ -1786,8 +1789,6 @@ static void l2cap_conn_del(struct hci_conn *hcon, int err) ida_destroy(&conn->tx_ida); - cancel_delayed_work_sync(&conn->id_addr_timer); - l2cap_unregister_all_users(conn); /* Force the connection to be immediately dropped */ @@ -1806,9 +1807,6 @@ static void l2cap_conn_del(struct hci_conn *hcon, int err) l2cap_chan_put(chan); } - if (conn->info_state & L2CAP_INFO_FEAT_MASK_REQ_SENT) - cancel_delayed_work_sync(&conn->info_timer); - hci_chan_del(conn->hchan); conn->hchan = NULL; From 25f420a0d4cfd61d3d23ec4b9c56d9f443d91377 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Fri, 20 Mar 2026 20:23:10 +0900 Subject: [PATCH 337/352] Bluetooth: L2CAP: Fix ERTM re-init and zero pdu_len infinite loop l2cap_config_req() processes CONFIG_REQ for channels in BT_CONNECTED state to support L2CAP reconfiguration (e.g. MTU changes). However, since both CONF_INPUT_DONE and CONF_OUTPUT_DONE are already set from the initial configuration, the reconfiguration path falls through to l2cap_ertm_init(), which re-initializes tx_q, srej_q, srej_list, and retrans_list without freeing the previous allocations and sets chan->sdu to NULL without freeing the existing skb. This leaks all previously allocated ERTM resources. Additionally, l2cap_parse_conf_req() does not validate the minimum value of remote_mps derived from the RFC max_pdu_size option. A zero value propagates to l2cap_segment_sdu() where pdu_len becomes zero, causing the while loop to never terminate since len is never decremented, exhausting all available memory. Fix the double-init by skipping l2cap_ertm_init() and l2cap_chan_ready() when the channel is already in BT_CONNECTED state, while still allowing the reconfiguration parameters to be updated through l2cap_parse_conf_req(). Also add a pdu_len zero check in l2cap_segment_sdu() as a safeguard. Fixes: 96298f640104 ("Bluetooth: L2CAP: handle l2cap config request during open state") Signed-off-by: Hyunwoo Kim Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/l2cap_core.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 2603c98d7ed1..95c65fece39b 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -2398,6 +2398,9 @@ static int l2cap_segment_sdu(struct l2cap_chan *chan, /* Remote device may have requested smaller PDUs */ pdu_len = min_t(size_t, pdu_len, chan->remote_mps); + if (!pdu_len) + return -EINVAL; + if (len <= pdu_len) { sar = L2CAP_SAR_UNSEGMENTED; sdu_len = 0; @@ -4333,14 +4336,16 @@ static inline int l2cap_config_req(struct l2cap_conn *conn, if (test_bit(CONF_INPUT_DONE, &chan->conf_state)) { set_default_fcs(chan); - if (chan->mode == L2CAP_MODE_ERTM || - chan->mode == L2CAP_MODE_STREAMING) - err = l2cap_ertm_init(chan); + if (chan->state != BT_CONNECTED) { + if (chan->mode == L2CAP_MODE_ERTM || + chan->mode == L2CAP_MODE_STREAMING) + err = l2cap_ertm_init(chan); - if (err < 0) - l2cap_send_disconn_req(chan, -err); - else - l2cap_chan_ready(chan); + if (err < 0) + l2cap_send_disconn_req(chan, -err); + else + l2cap_chan_ready(chan); + } goto unlock; } From 129fa608b6ad08b8ab7178eeb2ec272c993aaccc Mon Sep 17 00:00:00 2001 From: Pengpeng Hou Date: Wed, 25 Mar 2026 08:42:45 +0800 Subject: [PATCH 338/352] Bluetooth: btusb: clamp SCO altsetting table indices btusb_work() maps the number of active SCO links to USB alternate settings through a three-entry lookup table when CVSD traffic uses transparent voice settings. The lookup currently indexes alts[] with data->sco_num - 1 without first constraining sco_num to the number of available table entries. While the table only defines alternate settings for up to three SCO links, data->sco_num comes from hci_conn_num() and is used directly. Cap the lookup to the last table entry before indexing it so the driver keeps selecting the highest supported alternate setting without reading past alts[]. Fixes: baac6276c0a9 ("Bluetooth: btusb: handle mSBC audio over USB Endpoints") Signed-off-by: Pengpeng Hou Signed-off-by: Luiz Augusto von Dentz --- drivers/bluetooth/btusb.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c index a1c5eb993e47..5c535f3ab722 100644 --- a/drivers/bluetooth/btusb.c +++ b/drivers/bluetooth/btusb.c @@ -2376,8 +2376,11 @@ static void btusb_work(struct work_struct *work) if (data->air_mode == HCI_NOTIFY_ENABLE_SCO_CVSD) { if (hdev->voice_setting & 0x0020) { static const int alts[3] = { 2, 4, 5 }; + unsigned int sco_idx; - new_alts = alts[data->sco_num - 1]; + sco_idx = min_t(unsigned int, data->sco_num - 1, + ARRAY_SIZE(alts) - 1); + new_alts = alts[sco_idx]; } else { new_alts = data->sco_num; } From d3c0037ffe1273fa1961e779ff6906234d6cf53c Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 25 Mar 2026 14:10:55 +0100 Subject: [PATCH 339/352] netfilter: nft_set_pipapo_avx2: don't return non-matching entry on expiry New test case fails unexpectedly when avx2 matching functions are used. The test first loads a ranomly generated pipapo set with 'ipv4 . port' key, i.e. nft -f foo. This works. Then, it reloads the set after a flush: (echo flush set t s; cat foo) | nft -f - This is expected to work, because its the same set after all and it was already loaded once. But with avx2, this fails: nft reports a clashing element. The reported clash is of following form: We successfully re-inserted a . b c . d Then we try to insert a . d avx2 finds the already existing a . d, which (due to 'flush set') is marked as invalid in the new generation. It skips the element and moves to next. Due to incorrect masking, the skip-step finds the next matching element *only considering the first field*, i.e. we return the already reinserted "a . b", even though the last field is different and the entry should not have been matched. No such error is reported for the generic c implementation (no avx2) or when the last field has to use the 'nft_pipapo_avx2_lookup_slow' fallback. Bisection points to 7711f4bb4b36 ("netfilter: nft_set_pipapo: fix range overlap detection") but that fix merely uncovers this bug. Before this commit, the wrong element is returned, but erronously reported as a full, identical duplicate. The root-cause is too early return in the avx2 match functions. When we process the last field, we should continue to process data until the entire input size has been consumed to make sure no stale bits remain in the map. Link: https://lore.kernel.org/netfilter-devel/20260321152506.037f68c0@elisabeth/ Signed-off-by: Florian Westphal Reviewed-by: Stefano Brivio Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_pipapo_avx2.c | 20 ++++++++++---------- 1 file changed, 10 insertions(+), 10 deletions(-) diff --git a/net/netfilter/nft_set_pipapo_avx2.c b/net/netfilter/nft_set_pipapo_avx2.c index 7ff90325c97f..6395982e4d95 100644 --- a/net/netfilter/nft_set_pipapo_avx2.c +++ b/net/netfilter/nft_set_pipapo_avx2.c @@ -242,7 +242,7 @@ static int nft_pipapo_avx2_lookup_4b_2(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -319,7 +319,7 @@ static int nft_pipapo_avx2_lookup_4b_4(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -414,7 +414,7 @@ static int nft_pipapo_avx2_lookup_4b_8(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -505,7 +505,7 @@ static int nft_pipapo_avx2_lookup_4b_12(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -641,7 +641,7 @@ static int nft_pipapo_avx2_lookup_4b_32(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -699,7 +699,7 @@ static int nft_pipapo_avx2_lookup_8b_1(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -764,7 +764,7 @@ static int nft_pipapo_avx2_lookup_8b_2(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -839,7 +839,7 @@ static int nft_pipapo_avx2_lookup_8b_4(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -925,7 +925,7 @@ static int nft_pipapo_avx2_lookup_8b_6(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; @@ -1019,7 +1019,7 @@ static int nft_pipapo_avx2_lookup_8b_16(unsigned long *map, unsigned long *fill, b = nft_pipapo_avx2_refill(i_ul, &map[i_ul], fill, f->mt, last); if (last) - return b; + ret = b; if (unlikely(ret == -1)) ret = b / XSAVE_YMM_SIZE; From 6caefcd9491c408a4d161f7b60c8bb3d956526dd Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 25 Mar 2026 14:10:56 +0100 Subject: [PATCH 340/352] selftests: netfilter: nft_concat_range.sh: add check for flush+reload bug This test will fail without the preceding commit ("netfilter: nft_set_pipapo_avx2: fix match retart if found element is expired"): reject overlapping range on add 0s [ OK ] reload with flush /dev/stdin:59:32-52: Error: Could not process rule: File exists add element inet filter test { 10.0.0.29 . 10.0.2.29 } Reviewed-by: Stefano Brivio Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- .../net/netfilter/nft_concat_range.sh | 70 ++++++++++++++++++- 1 file changed, 69 insertions(+), 1 deletion(-) diff --git a/tools/testing/selftests/net/netfilter/nft_concat_range.sh b/tools/testing/selftests/net/netfilter/nft_concat_range.sh index 394166f224a4..ffdc6ccc6511 100755 --- a/tools/testing/selftests/net/netfilter/nft_concat_range.sh +++ b/tools/testing/selftests/net/netfilter/nft_concat_range.sh @@ -29,7 +29,8 @@ TYPES="net_port port_net net6_port port_proto net6_port_mac net6_port_mac_proto net6_port_net6_port net_port_mac_proto_net" # Reported bugs, also described by TYPE_ variables below -BUGS="flush_remove_add reload net_port_proto_match avx2_mismatch doublecreate insert_overlap" +BUGS="flush_remove_add reload net_port_proto_match avx2_mismatch doublecreate + insert_overlap load_flush_load4 load_flush_load8" # List of possible paths to pktgen script from kernel tree for performance tests PKTGEN_SCRIPT_PATHS=" @@ -432,6 +433,30 @@ race_repeat 0 perf_duration 0 " +TYPE_load_flush_load4=" +display reload with flush, 4bit groups +type_spec ipv4_addr . ipv4_addr +chain_spec ip saddr . ip daddr +dst addr4 +proto icmp + +race_repeat 0 + +perf_duration 0 +" + +TYPE_load_flush_load8=" +display reload with flush, 8bit groups +type_spec ipv4_addr . ipv4_addr +chain_spec ip saddr . ip daddr +dst addr4 +proto icmp + +race_repeat 0 + +perf_duration 0 +" + # Set template for all tests, types and rules are filled in depending on test set_template=' flush ruleset @@ -1997,6 +2022,49 @@ test_bug_insert_overlap() return 0 } +test_bug_load_flush_load4() +{ + local i + + setup veth send_"${proto}" set || return ${ksft_skip} + + for i in $(seq 0 255); do + local addelem="add element inet filter test" + local j + + for j in $(seq 0 20); do + echo "$addelem { 10.$j.0.$i . 10.$j.1.$i }" + echo "$addelem { 10.$j.0.$i . 10.$j.2.$i }" + done + done > "$tmp" + + nft -f "$tmp" || return 1 + + ( echo "flush set inet filter test";cat "$tmp") | nft -f - + [ $? -eq 0 ] || return 1 + + return 0 +} + +test_bug_load_flush_load8() +{ + local i + + setup veth send_"${proto}" set || return ${ksft_skip} + + for i in $(seq 1 100); do + echo "add element inet filter test { 10.0.0.$i . 10.0.1.$i }" + echo "add element inet filter test { 10.0.0.$i . 10.0.2.$i }" + done > "$tmp" + + nft -f "$tmp" || return 1 + + ( echo "flush set inet filter test";cat "$tmp") | nft -f - + [ $? -eq 0 ] || return 1 + + return 0 +} + test_reported_issues() { eval test_bug_"${subtest}" } From 84a8335d8300576f1b377ae24abca1d9f197807f Mon Sep 17 00:00:00 2001 From: Chuck Lever Date: Tue, 24 Mar 2026 08:53:23 -0400 Subject: [PATCH 341/352] tls: Purge async_hold in tls_decrypt_async_wait() The async_hold queue pins encrypted input skbs while the AEAD engine references their scatterlist data. Once tls_decrypt_async_wait() returns, every AEAD operation has completed and the engine no longer references those skbs, so they can be freed unconditionally. A subsequent patch adds batch async decryption to tls_sw_read_sock(), introducing a new call site that must drain pending AEAD operations and release held skbs. Move __skb_queue_purge(&ctx->async_hold) into tls_decrypt_async_wait() so the purge is centralized and every caller -- recvmsg's drain path, the -EBUSY fallback in tls_do_decryption(), and the new read_sock batch path -- releases held skbs on synchronization without each site managing the purge independently. This fixes a leak when tls_strp_msg_hold() fails part-way through, after having added some cloned skbs to the async_hold queue. tls_decrypt_sg() will then call tls_decrypt_async_wait() to process all pending decrypts, and drop back to synchronous mode, but tls_sw_recvmsg() only flushes the async_hold queue when one record has been processed in "fully-async" mode, which may not be the case here. Signed-off-by: Chuck Lever Reported-by: Yiming Qian Fixes: b8a6ff84abbc ("tls: wait for pending async decryptions if tls_strp_msg_hold fails") Link: https://patch.msgid.link/20260324-tls-read-sock-v5-1-5408befe5774@oracle.com [pabeni@redhat.com: added leak comment] Signed-off-by: Paolo Abeni --- net/tls/tls_sw.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c index 5fe07f110fe8..dd9dda759bbb 100644 --- a/net/tls/tls_sw.c +++ b/net/tls/tls_sw.c @@ -246,6 +246,7 @@ static int tls_decrypt_async_wait(struct tls_sw_context_rx *ctx) crypto_wait_req(-EINPROGRESS, &ctx->async_wait); atomic_inc(&ctx->decrypt_pending); + __skb_queue_purge(&ctx->async_hold); return ctx->async_wait.err; } @@ -2225,7 +2226,6 @@ int tls_sw_recvmsg(struct sock *sk, /* Wait for all previously submitted records to be decrypted */ ret = tls_decrypt_async_wait(ctx); - __skb_queue_purge(&ctx->async_hold); if (ret) { if (err >= 0 || err == -EINPROGRESS) From 52025ebaa29f4eb4ed8bf92ce83a68f24ab7fdf7 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Wed, 25 Mar 2026 14:10:58 +0100 Subject: [PATCH 342/352] netfilter: nfnetlink_log: fix uninitialized padding leak in NFULA_PAYLOAD __build_packet_message() manually constructs the NFULA_PAYLOAD netlink attribute using skb_put() and skb_copy_bits(), bypassing the standard nla_reserve()/nla_put() helpers. While nla_total_size(data_len) bytes are allocated (including NLA alignment padding), only data_len bytes of actual packet data are copied. The trailing nla_padlen(data_len) bytes (1-3 when data_len is not 4-byte aligned) are never initialized, leaking stale heap contents to userspace via the NFLOG netlink socket. Replace the manual attribute construction with nla_reserve(), which handles the tailroom check, header setup, and padding zeroing via __nla_reserve(). The subsequent skb_copy_bits() fills in the payload data on top of the properly initialized attribute. Fixes: df6fb868d611 ("[NETFILTER]: nfnetlink: convert to generic netlink attribute functions") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nfnetlink_log.c | 8 ++------ 1 file changed, 2 insertions(+), 6 deletions(-) diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index b35a90955e2e..fcbe54940b2e 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -647,15 +647,11 @@ __build_packet_message(struct nfnl_log_net *log, if (data_len) { struct nlattr *nla; - int size = nla_attr_size(data_len); - if (skb_tailroom(inst->skb) < nla_total_size(data_len)) + nla = nla_reserve(inst->skb, NFULA_PAYLOAD, data_len); + if (!nla) goto nla_put_failure; - nla = skb_put(inst->skb, nla_total_size(data_len)); - nla->nla_type = NFULA_PAYLOAD; - nla->nla_len = size; - if (skb_copy_bits(skb, 0, nla_data(nla), data_len)) BUG(); } From 9d3f027327c2fa265f7f85ead41294792c3296ed Mon Sep 17 00:00:00 2001 From: Ren Wei Date: Wed, 25 Mar 2026 14:11:00 +0100 Subject: [PATCH 343/352] netfilter: ip6t_rt: reject oversized addrnr in rt_mt6_check() Reject rt match rules whose addrnr exceeds IP6T_RT_HOPS. rt_mt6() expects addrnr to stay within the bounds of rtinfo->addrs[]. Validate addrnr during rule installation so malformed rules are rejected before the match logic can use an out-of-range value. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Yifan Wu Reported-by: Juefei Pu Co-developed-by: Yuan Tan Signed-off-by: Yuan Tan Suggested-by: Xin Liu Tested-by: Yuhang Zheng Signed-off-by: Ren Wei Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv6/netfilter/ip6t_rt.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv6/netfilter/ip6t_rt.c b/net/ipv6/netfilter/ip6t_rt.c index 4ad8b2032f1f..5561bd9cea81 100644 --- a/net/ipv6/netfilter/ip6t_rt.c +++ b/net/ipv6/netfilter/ip6t_rt.c @@ -157,6 +157,10 @@ static int rt_mt6_check(const struct xt_mtchk_param *par) pr_debug("unknown flags %X\n", rtinfo->invflags); return -EINVAL; } + if (rtinfo->addrnr > IP6T_RT_HOPS) { + pr_debug("too many addresses specified\n"); + return -EINVAL; + } if ((rtinfo->flags & (IP6T_RT_RES | IP6T_RT_FST_MASK)) && (!(rtinfo->flags & IP6T_RT_TYP) || (rtinfo->rt_type != 0) || From fafdd92b9e30fe057740c5bb5cd4f92ecea9bf26 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Mar 2026 14:11:01 +0100 Subject: [PATCH 344/352] netfilter: nft_set_rbtree: revisit array resize logic Chris Arges reports high memory consumption with thousands of containers, this patch revisits the array allocation logic. For anonymous sets, start by 16 slots (which takes 256 bytes on x86_64). Expand it by x2 until threshold of 512 slots is reached, over that threshold, expand it by x1.5. For non-anonymous set, start by 1024 slots in the array (which takes 16 Kbytes initially on x86_64). Expand it by x1.5. Use set->ndeact to subtract deactivated elements when calculating the number of the slots in the array, otherwise the array size array gets increased artifically. Add special case shrink logic to deal with flush set too. The shrink logic is skipped by anonymous sets. Use check_add_overflow() to calculate the new array size. Add a WARN_ON_ONCE check to make sure elements fit into the new array size. Reported-by: Chris Arges Fixes: 7e43e0a1141d ("netfilter: nft_set_rbtree: translate rbtree to array for binary search") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_set_rbtree.c | 92 +++++++++++++++++++++++++++------- 1 file changed, 75 insertions(+), 17 deletions(-) diff --git a/net/netfilter/nft_set_rbtree.c b/net/netfilter/nft_set_rbtree.c index fe8bd497d74a..737c339decd0 100644 --- a/net/netfilter/nft_set_rbtree.c +++ b/net/netfilter/nft_set_rbtree.c @@ -572,14 +572,12 @@ static struct nft_array *nft_array_alloc(u32 max_intervals) return array; } -#define NFT_ARRAY_EXTRA_SIZE 10240 - /* Similar to nft_rbtree_{u,k}size to hide details to userspace, but consider * packed representation coming from userspace for anonymous sets too. */ static u32 nft_array_elems(const struct nft_set *set) { - u32 nelems = atomic_read(&set->nelems); + u32 nelems = atomic_read(&set->nelems) - set->ndeact; /* Adjacent intervals are represented with a single start element in * anonymous sets, use the current element counter as is. @@ -595,27 +593,87 @@ static u32 nft_array_elems(const struct nft_set *set) return (nelems / 2) + 2; } -static int nft_array_may_resize(const struct nft_set *set) +#define NFT_ARRAY_INITIAL_SIZE 1024 +#define NFT_ARRAY_INITIAL_ANON_SIZE 16 +#define NFT_ARRAY_INITIAL_ANON_THRESH (8192U / sizeof(struct nft_array_interval)) + +static int nft_array_may_resize(const struct nft_set *set, bool flush) { - u32 nelems = nft_array_elems(set), new_max_intervals; + u32 initial_intervals, max_intervals, new_max_intervals, delta; + u32 shrinked_max_intervals, nelems = nft_array_elems(set); struct nft_rbtree *priv = nft_set_priv(set); struct nft_array *array; - if (!priv->array_next) { - array = nft_array_alloc(nelems + NFT_ARRAY_EXTRA_SIZE); + if (nft_set_is_anonymous(set)) + initial_intervals = NFT_ARRAY_INITIAL_ANON_SIZE; + else + initial_intervals = NFT_ARRAY_INITIAL_SIZE; + + if (priv->array_next) { + max_intervals = priv->array_next->max_intervals; + new_max_intervals = priv->array_next->max_intervals; + } else { + if (priv->array) { + max_intervals = priv->array->max_intervals; + new_max_intervals = priv->array->max_intervals; + } else { + max_intervals = 0; + new_max_intervals = initial_intervals; + } + } + + if (nft_set_is_anonymous(set)) + goto maybe_grow; + + if (flush) { + /* Set flush just started, nelems still report elements.*/ + nelems = 0; + new_max_intervals = NFT_ARRAY_INITIAL_SIZE; + goto realloc_array; + } + + if (check_add_overflow(new_max_intervals, new_max_intervals, + &shrinked_max_intervals)) + return -EOVERFLOW; + + shrinked_max_intervals = DIV_ROUND_UP(shrinked_max_intervals, 3); + + if (shrinked_max_intervals > NFT_ARRAY_INITIAL_SIZE && + nelems < shrinked_max_intervals) { + new_max_intervals = shrinked_max_intervals; + goto realloc_array; + } +maybe_grow: + if (nelems > new_max_intervals) { + if (nft_set_is_anonymous(set) && + new_max_intervals < NFT_ARRAY_INITIAL_ANON_THRESH) { + new_max_intervals <<= 1; + } else { + delta = new_max_intervals >> 1; + if (check_add_overflow(new_max_intervals, delta, + &new_max_intervals)) + return -EOVERFLOW; + } + } + +realloc_array: + if (WARN_ON_ONCE(nelems > new_max_intervals)) + return -ENOMEM; + + if (priv->array_next) { + if (max_intervals == new_max_intervals) + return 0; + + if (nft_array_intervals_alloc(priv->array_next, new_max_intervals) < 0) + return -ENOMEM; + } else { + array = nft_array_alloc(new_max_intervals); if (!array) return -ENOMEM; priv->array_next = array; } - if (nelems < priv->array_next->max_intervals) - return 0; - - new_max_intervals = priv->array_next->max_intervals + NFT_ARRAY_EXTRA_SIZE; - if (nft_array_intervals_alloc(priv->array_next, new_max_intervals) < 0) - return -ENOMEM; - return 0; } @@ -630,7 +688,7 @@ static int nft_rbtree_insert(const struct net *net, const struct nft_set *set, nft_rbtree_maybe_reset_start_cookie(priv, tstamp); - if (nft_array_may_resize(set) < 0) + if (nft_array_may_resize(set, false) < 0) return -ENOMEM; do { @@ -741,7 +799,7 @@ nft_rbtree_deactivate(const struct net *net, const struct nft_set *set, nft_rbtree_interval_null(set, this)) priv->start_rbe_cookie = 0; - if (nft_array_may_resize(set) < 0) + if (nft_array_may_resize(set, false) < 0) return NULL; while (parent != NULL) { @@ -811,7 +869,7 @@ static void nft_rbtree_walk(const struct nft_ctx *ctx, switch (iter->type) { case NFT_ITER_UPDATE_CLONE: - if (nft_array_may_resize(set) < 0) { + if (nft_array_may_resize(set, true) < 0) { iter->err = -ENOMEM; break; } From 9c42bc9db90a154bc61ae337a070465f3393485a Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Mar 2026 14:11:02 +0100 Subject: [PATCH 345/352] netfilter: nf_conntrack_expect: honor expectation helper field The expectation helper field is mostly unused. As a result, the netfilter codebase relies on accessing the helper through exp->master. Always set on the expectation helper field so it can be used to reach the helper. nf_ct_expect_init() is called from packet path where the skb owns the ct object, therefore accessing exp->master for the newly created expectation is safe. This saves a lot of updates in all callsites to pass the ct object as parameter to nf_ct_expect_init(). This is a preparation patches for follow up fixes. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_expect.h | 2 +- net/netfilter/nf_conntrack_broadcast.c | 2 +- net/netfilter/nf_conntrack_expect.c | 14 +++++++++++++- net/netfilter/nf_conntrack_h323_main.c | 12 ++++++------ net/netfilter/nf_conntrack_helper.c | 7 ++++++- net/netfilter/nf_conntrack_netlink.c | 2 +- net/netfilter/nf_conntrack_sip.c | 2 +- 7 files changed, 29 insertions(+), 12 deletions(-) diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 165e7a03b8e9..1b01400b10bd 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -40,7 +40,7 @@ struct nf_conntrack_expect { struct nf_conntrack_expect *this); /* Helper to assign to new connection */ - struct nf_conntrack_helper *helper; + struct nf_conntrack_helper __rcu *helper; /* The conntrack of the master connection */ struct nf_conn *master; diff --git a/net/netfilter/nf_conntrack_broadcast.c b/net/netfilter/nf_conntrack_broadcast.c index a7552a46d6ac..1964c596c646 100644 --- a/net/netfilter/nf_conntrack_broadcast.c +++ b/net/netfilter/nf_conntrack_broadcast.c @@ -70,7 +70,7 @@ int nf_conntrack_broadcast_help(struct sk_buff *skb, exp->expectfn = NULL; exp->flags = NF_CT_EXPECT_PERMANENT; exp->class = NF_CT_EXPECT_CLASS_DEFAULT; - exp->helper = NULL; + rcu_assign_pointer(exp->helper, helper); nf_ct_expect_related(exp, 0); nf_ct_expect_put(exp); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index cfc2daa3fc7f..841e316240da 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -309,12 +309,19 @@ struct nf_conntrack_expect *nf_ct_expect_alloc(struct nf_conn *me) } EXPORT_SYMBOL_GPL(nf_ct_expect_alloc); +/* This function can only be used from packet path, where accessing + * master's helper is safe, because the packet holds a reference on + * the conntrack object. Never use it from control plane. + */ void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, u_int8_t family, const union nf_inet_addr *saddr, const union nf_inet_addr *daddr, u_int8_t proto, const __be16 *src, const __be16 *dst) { + struct nf_conntrack_helper *helper = NULL; + struct nf_conn *ct = exp->master; + struct nf_conn_help *help; int len; if (family == AF_INET) @@ -325,7 +332,12 @@ void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, exp->flags = 0; exp->class = class; exp->expectfn = NULL; - exp->helper = NULL; + + help = nfct_help(ct); + if (help) + helper = rcu_dereference(help->helper); + + rcu_assign_pointer(exp->helper, helper); exp->tuple.src.l3num = family; exp->tuple.dst.protonum = proto; diff --git a/net/netfilter/nf_conntrack_h323_main.c b/net/netfilter/nf_conntrack_h323_main.c index a2a0e22ccee1..3f5c50455b71 100644 --- a/net/netfilter/nf_conntrack_h323_main.c +++ b/net/netfilter/nf_conntrack_h323_main.c @@ -643,7 +643,7 @@ static int expect_h245(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); - exp->helper = &nf_conntrack_helper_h245; + rcu_assign_pointer(exp->helper, &nf_conntrack_helper_h245); nathook = rcu_dereference(nfct_h323_nat_hook); if (memcmp(&ct->tuplehash[dir].tuple.src.u3, @@ -767,7 +767,7 @@ static int expect_callforwarding(struct sk_buff *skb, nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); - exp->helper = nf_conntrack_helper_q931; + rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); nathook = rcu_dereference(nfct_h323_nat_hook); if (memcmp(&ct->tuplehash[dir].tuple.src.u3, @@ -1234,7 +1234,7 @@ static int expect_q931(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3 : NULL, &ct->tuplehash[!dir].tuple.dst.u3, IPPROTO_TCP, NULL, &port); - exp->helper = nf_conntrack_helper_q931; + rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); exp->flags = NF_CT_EXPECT_PERMANENT; /* Accept multiple calls */ nathook = rcu_dereference(nfct_h323_nat_hook); @@ -1306,7 +1306,7 @@ static int process_gcf(struct sk_buff *skb, struct nf_conn *ct, nf_ct_expect_init(exp, NF_CT_EXPECT_CLASS_DEFAULT, nf_ct_l3num(ct), &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_UDP, NULL, &port); - exp->helper = nf_conntrack_helper_ras; + rcu_assign_pointer(exp->helper, nf_conntrack_helper_ras); if (nf_ct_expect_related(exp, 0) == 0) { pr_debug("nf_ct_ras: expect RAS "); @@ -1523,7 +1523,7 @@ static int process_acf(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; - exp->helper = nf_conntrack_helper_q931; + rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); if (nf_ct_expect_related(exp, 0) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); @@ -1577,7 +1577,7 @@ static int process_lcf(struct sk_buff *skb, struct nf_conn *ct, &ct->tuplehash[!dir].tuple.src.u3, &addr, IPPROTO_TCP, NULL, &port); exp->flags = NF_CT_EXPECT_PERMANENT; - exp->helper = nf_conntrack_helper_q931; + rcu_assign_pointer(exp->helper, nf_conntrack_helper_q931); if (nf_ct_expect_related(exp, 0) == 0) { pr_debug("nf_ct_ras: expect Q.931 "); diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index ceb48c3ca0a4..294a6ffcbccd 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -399,7 +399,7 @@ static bool expect_iter_me(struct nf_conntrack_expect *exp, void *data) const struct nf_conntrack_helper *me = data; const struct nf_conntrack_helper *this; - if (exp->helper == me) + if (rcu_access_pointer(exp->helper) == me) return true; this = rcu_dereference_protected(help->helper, @@ -421,6 +421,11 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) nf_ct_expect_iterate_destroy(expect_iter_me, NULL); nf_ct_iterate_destroy(unhelp, me); + + /* nf_ct_iterate_destroy() does an unconditional synchronize_rcu() as + * last step, this ensures rcu readers of exp->helper are done. + * No need for another synchronize_rcu() here. + */ } EXPORT_SYMBOL_GPL(nf_conntrack_helper_unregister); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index c156574e1273..a42d14290786 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -3573,7 +3573,7 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, exp->class = class; exp->master = ct; - exp->helper = helper; + rcu_assign_pointer(exp->helper, helper); exp->tuple = *tuple; exp->mask.src.u3 = mask->src.u3; exp->mask.src.u.all = mask->src.u.all; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 4ab5ef71d96d..106b2f419e19 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1297,7 +1297,7 @@ static int process_register_request(struct sk_buff *skb, unsigned int protoff, nf_ct_expect_init(exp, SIP_EXPECT_SIGNALLING, nf_ct_l3num(ct), saddr, &daddr, proto, NULL, &port); exp->timeout.expires = sip_timeout * HZ; - exp->helper = helper; + rcu_assign_pointer(exp->helper, helper); exp->flags = NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE; hooks = rcu_dereference(nf_nat_sip_hooks); From f01794106042ee27e54af6fdf5b319a2fe3df94d Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Mar 2026 14:11:03 +0100 Subject: [PATCH 346/352] netfilter: nf_conntrack_expect: use expect->helper Use expect->helper in ctnetlink and /proc to dump the helper name. Using nfct_help() without holding a reference to the master conntrack is unsafe. Use exp->master->helper in ctnetlink path if userspace does not provide an explicit helper when creating an expectation to retain the existing behaviour. The ctnetlink expectation path holds the reference on the master conntrack and nf_conntrack_expect lock and the nfnetlink glue path refers to the master ct that is attached to the skb. Reported-by: Hyunwoo Kim Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_expect.c | 2 +- net/netfilter/nf_conntrack_helper.c | 6 +----- net/netfilter/nf_conntrack_netlink.c | 24 ++++++++++-------------- net/netfilter/nf_conntrack_sip.c | 2 +- 4 files changed, 13 insertions(+), 21 deletions(-) diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 841e316240da..64977db12b1d 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -666,7 +666,7 @@ static int exp_seq_show(struct seq_file *s, void *v) if (expect->flags & NF_CT_EXPECT_USERSPACE) seq_printf(s, "%sUSERSPACE", delim); - helper = rcu_dereference(nfct_help(expect->master)->helper); + helper = rcu_dereference(expect->helper); if (helper) { seq_printf(s, "%s%s", expect->flags ? " " : "", helper->name); if (helper->expect_policy[expect->class].name[0]) diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 294a6ffcbccd..1b330ba6613b 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -395,14 +395,10 @@ EXPORT_SYMBOL_GPL(nf_conntrack_helper_register); static bool expect_iter_me(struct nf_conntrack_expect *exp, void *data) { - struct nf_conn_help *help = nfct_help(exp->master); const struct nf_conntrack_helper *me = data; const struct nf_conntrack_helper *this; - if (rcu_access_pointer(exp->helper) == me) - return true; - - this = rcu_dereference_protected(help->helper, + this = rcu_dereference_protected(exp->helper, lockdep_is_held(&nf_conntrack_expect_lock)); return this == me; } diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index a42d14290786..8477c3736432 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -3012,7 +3012,7 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, { struct nf_conn *master = exp->master; long timeout = ((long)exp->timeout.expires - (long)jiffies) / HZ; - struct nf_conn_help *help; + struct nf_conntrack_helper *helper; #if IS_ENABLED(CONFIG_NF_NAT) struct nlattr *nest_parms; struct nf_conntrack_tuple nat_tuple = {}; @@ -3057,15 +3057,12 @@ ctnetlink_exp_dump_expect(struct sk_buff *skb, nla_put_be32(skb, CTA_EXPECT_FLAGS, htonl(exp->flags)) || nla_put_be32(skb, CTA_EXPECT_CLASS, htonl(exp->class))) goto nla_put_failure; - help = nfct_help(master); - if (help) { - struct nf_conntrack_helper *helper; - helper = rcu_dereference(help->helper); - if (helper && - nla_put_string(skb, CTA_EXPECT_HELP_NAME, helper->name)) - goto nla_put_failure; - } + helper = rcu_dereference(exp->helper); + if (helper && + nla_put_string(skb, CTA_EXPECT_HELP_NAME, helper->name)) + goto nla_put_failure; + expfn = nf_ct_helper_expectfn_find_by_symbol(exp->expectfn); if (expfn != NULL && nla_put_string(skb, CTA_EXPECT_FN, expfn->name)) @@ -3394,12 +3391,9 @@ static int ctnetlink_get_expect(struct sk_buff *skb, static bool expect_iter_name(struct nf_conntrack_expect *exp, void *data) { struct nf_conntrack_helper *helper; - const struct nf_conn_help *m_help; const char *name = data; - m_help = nfct_help(exp->master); - - helper = rcu_dereference(m_help->helper); + helper = rcu_dereference(exp->helper); if (!helper) return false; @@ -3534,9 +3528,9 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple *mask) { - u_int32_t class = 0; struct nf_conntrack_expect *exp; struct nf_conn_help *help; + u32 class = 0; int err; help = nfct_help(ct); @@ -3573,6 +3567,8 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, exp->class = class; exp->master = ct; + if (!helper) + helper = rcu_dereference(help->helper); rcu_assign_pointer(exp->helper, helper); exp->tuple = *tuple; exp->mask.src.u3 = mask->src.u3; diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 106b2f419e19..20e57cf5c83a 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -924,7 +924,7 @@ static int set_expected_rtp_rtcp(struct sk_buff *skb, unsigned int protoff, exp = __nf_ct_expect_find(net, nf_ct_zone(ct), &tuple); if (!exp || exp->master == ct || - nfct_help(exp->master)->helper != nfct_help(ct)->helper || + exp->helper != nfct_help(ct)->helper || exp->class != class) break; #if IS_ENABLED(CONFIG_NF_NAT) From bffcaad9afdfe45d7fc777397d3b83c1e3ebffe5 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Mar 2026 14:11:04 +0100 Subject: [PATCH 347/352] netfilter: ctnetlink: ensure safe access to master conntrack Holding reference on the expectation is not sufficient, the master conntrack object can just go away, making exp->master invalid. To access exp->master safely: - Grab the nf_conntrack_expect_lock, this gets serialized with clean_from_lists() which also holds this lock when the master conntrack goes away. - Hold reference on master conntrack via nf_conntrack_find_get(). Not so easy since the master tuple to look up for the master conntrack is not available in the existing problematic paths. This patch goes for extending the nf_conntrack_expect_lock section to address this issue for simplicity, in the cases that are described below this is just slightly extending the lock section. The add expectation command already holds a reference to the master conntrack from ctnetlink_create_expect(). However, the delete expectation command needs to grab the spinlock before looking up for the expectation. Expand the existing spinlock section to address this to cover the expectation lookup. Note that, the nf_ct_expect_iterate_net() calls already grabs the spinlock while iterating over the expectation table, which is correct. The get expectation command needs to grab the spinlock to ensure master conntrack does not go away. This also expands the existing spinlock section to cover the expectation lookup too. I needed to move the netlink skb allocation out of the spinlock to keep it GFP_KERNEL. For the expectation events, the IPEXP_DESTROY event is already delivered under the spinlock, just move the delivery of IPEXP_NEW under the spinlock too because the master conntrack event cache is reached through exp->master. While at it, add lockdep notations to help identify what codepaths need to grab the spinlock. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_core.h | 5 ++++ net/netfilter/nf_conntrack_ecache.c | 2 ++ net/netfilter/nf_conntrack_expect.c | 10 +++++++- net/netfilter/nf_conntrack_netlink.c | 28 +++++++++++++++-------- 4 files changed, 35 insertions(+), 10 deletions(-) diff --git a/include/net/netfilter/nf_conntrack_core.h b/include/net/netfilter/nf_conntrack_core.h index 3384859a8921..8883575adcc1 100644 --- a/include/net/netfilter/nf_conntrack_core.h +++ b/include/net/netfilter/nf_conntrack_core.h @@ -83,6 +83,11 @@ void nf_conntrack_lock(spinlock_t *lock); extern spinlock_t nf_conntrack_expect_lock; +static inline void lockdep_nfct_expect_lock_held(void) +{ + lockdep_assert_held(&nf_conntrack_expect_lock); +} + /* ctnetlink code shared by both ctnetlink and nf_conntrack_bpf */ static inline void __nf_ct_set_timeout(struct nf_conn *ct, u64 timeout) diff --git a/net/netfilter/nf_conntrack_ecache.c b/net/netfilter/nf_conntrack_ecache.c index 81baf2082604..9df159448b89 100644 --- a/net/netfilter/nf_conntrack_ecache.c +++ b/net/netfilter/nf_conntrack_ecache.c @@ -247,6 +247,8 @@ void nf_ct_expect_event_report(enum ip_conntrack_expect_events event, struct nf_ct_event_notifier *notify; struct nf_conntrack_ecache *e; + lockdep_nfct_expect_lock_held(); + rcu_read_lock(); notify = rcu_dereference(net->ct.nf_conntrack_event_cb); if (!notify) diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 64977db12b1d..1cbe5f1108c2 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -51,6 +51,7 @@ void nf_ct_unlink_expect_report(struct nf_conntrack_expect *exp, struct net *net = nf_ct_exp_net(exp); struct nf_conntrack_net *cnet; + lockdep_nfct_expect_lock_held(); WARN_ON(!master_help); WARN_ON(timer_pending(&exp->timeout)); @@ -118,6 +119,8 @@ nf_ct_exp_equal(const struct nf_conntrack_tuple *tuple, bool nf_ct_remove_expect(struct nf_conntrack_expect *exp) { + lockdep_nfct_expect_lock_held(); + if (timer_delete(&exp->timeout)) { nf_ct_unlink_expect(exp); nf_ct_expect_put(exp); @@ -177,6 +180,8 @@ nf_ct_find_expectation(struct net *net, struct nf_conntrack_expect *i, *exp = NULL; unsigned int h; + lockdep_nfct_expect_lock_held(); + if (!cnet->expect_count) return NULL; @@ -454,6 +459,8 @@ static inline int __nf_ct_expect_check(struct nf_conntrack_expect *expect, unsigned int h; int ret = 0; + lockdep_nfct_expect_lock_held(); + if (!master_help) { ret = -ESHUTDOWN; goto out; @@ -510,8 +517,9 @@ int nf_ct_expect_related_report(struct nf_conntrack_expect *expect, nf_ct_expect_insert(expect); - spin_unlock_bh(&nf_conntrack_expect_lock); nf_ct_expect_event_report(IPEXP_NEW, expect, portid, report); + spin_unlock_bh(&nf_conntrack_expect_lock); + return 0; out: spin_unlock_bh(&nf_conntrack_expect_lock); diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 8477c3736432..89540112d165 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -3355,31 +3355,37 @@ static int ctnetlink_get_expect(struct sk_buff *skb, if (err < 0) return err; + skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); + if (!skb2) + return -ENOMEM; + + spin_lock_bh(&nf_conntrack_expect_lock); exp = nf_ct_expect_find_get(info->net, &zone, &tuple); - if (!exp) + if (!exp) { + spin_unlock_bh(&nf_conntrack_expect_lock); + kfree_skb(skb2); return -ENOENT; + } if (cda[CTA_EXPECT_ID]) { __be32 id = nla_get_be32(cda[CTA_EXPECT_ID]); if (id != nf_expect_get_id(exp)) { nf_ct_expect_put(exp); + spin_unlock_bh(&nf_conntrack_expect_lock); + kfree_skb(skb2); return -ENOENT; } } - skb2 = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL); - if (!skb2) { - nf_ct_expect_put(exp); - return -ENOMEM; - } - rcu_read_lock(); err = ctnetlink_exp_fill_info(skb2, NETLINK_CB(skb).portid, info->nlh->nlmsg_seq, IPCTNL_MSG_EXP_NEW, exp); rcu_read_unlock(); nf_ct_expect_put(exp); + spin_unlock_bh(&nf_conntrack_expect_lock); + if (err <= 0) { kfree_skb(skb2); return -ENOMEM; @@ -3426,22 +3432,26 @@ static int ctnetlink_del_expect(struct sk_buff *skb, if (err < 0) return err; + spin_lock_bh(&nf_conntrack_expect_lock); + /* bump usage count to 2 */ exp = nf_ct_expect_find_get(info->net, &zone, &tuple); - if (!exp) + if (!exp) { + spin_unlock_bh(&nf_conntrack_expect_lock); return -ENOENT; + } if (cda[CTA_EXPECT_ID]) { __be32 id = nla_get_be32(cda[CTA_EXPECT_ID]); if (id != nf_expect_get_id(exp)) { nf_ct_expect_put(exp); + spin_unlock_bh(&nf_conntrack_expect_lock); return -ENOENT; } } /* after list removal, usage count == 1 */ - spin_lock_bh(&nf_conntrack_expect_lock); if (timer_delete(&exp->timeout)) { nf_ct_unlink_expect_report(exp, NETLINK_CB(skb).portid, nlmsg_report(info->nlh)); From 02a3231b6d82efe750da6554ebf280e4a6f78756 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Mar 2026 22:39:55 +0100 Subject: [PATCH 348/352] netfilter: nf_conntrack_expect: store netns and zone in expectation __nf_ct_expect_find() and nf_ct_expect_find_get() are called under rcu_read_lock() but they dereference the master conntrack via exp->master. Since the expectation does not hold a reference on the master conntrack, this could be dying conntrack or different recycled conntrack than the real master due to SLAB_TYPESAFE_RCU. Store the netns, the master_tuple and the zone in struct nf_conntrack_expect as a safety measure. This patch is required by the follow up fix not to dump expectations that do not belong to this netns. Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/netfilter/nf_conntrack_expect.h | 18 +++++++++++++++++- net/netfilter/nf_conntrack_broadcast.c | 6 +++++- net/netfilter/nf_conntrack_expect.c | 9 +++++++-- net/netfilter/nf_conntrack_netlink.c | 5 +++++ 4 files changed, 34 insertions(+), 4 deletions(-) diff --git a/include/net/netfilter/nf_conntrack_expect.h b/include/net/netfilter/nf_conntrack_expect.h index 1b01400b10bd..e9a8350e7ccf 100644 --- a/include/net/netfilter/nf_conntrack_expect.h +++ b/include/net/netfilter/nf_conntrack_expect.h @@ -22,10 +22,16 @@ struct nf_conntrack_expect { /* Hash member */ struct hlist_node hnode; + /* Network namespace */ + possible_net_t net; + /* We expect this tuple, with the following mask */ struct nf_conntrack_tuple tuple; struct nf_conntrack_tuple_mask mask; +#ifdef CONFIG_NF_CONNTRACK_ZONES + struct nf_conntrack_zone zone; +#endif /* Usage count. */ refcount_t use; @@ -62,7 +68,17 @@ struct nf_conntrack_expect { static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp) { - return nf_ct_net(exp->master); + return read_pnet(&exp->net); +} + +static inline bool nf_ct_exp_zone_equal_any(const struct nf_conntrack_expect *a, + const struct nf_conntrack_zone *b) +{ +#ifdef CONFIG_NF_CONNTRACK_ZONES + return a->zone.id == b->id; +#else + return true; +#endif } #define NF_CT_EXP_POLICY_NAME_LEN 16 diff --git a/net/netfilter/nf_conntrack_broadcast.c b/net/netfilter/nf_conntrack_broadcast.c index 1964c596c646..4f39bf7c843f 100644 --- a/net/netfilter/nf_conntrack_broadcast.c +++ b/net/netfilter/nf_conntrack_broadcast.c @@ -21,6 +21,7 @@ int nf_conntrack_broadcast_help(struct sk_buff *skb, unsigned int timeout) { const struct nf_conntrack_helper *helper; + struct net *net = read_pnet(&ct->ct_net); struct nf_conntrack_expect *exp; struct iphdr *iph = ip_hdr(skb); struct rtable *rt = skb_rtable(skb); @@ -71,7 +72,10 @@ int nf_conntrack_broadcast_help(struct sk_buff *skb, exp->flags = NF_CT_EXPECT_PERMANENT; exp->class = NF_CT_EXPECT_CLASS_DEFAULT; rcu_assign_pointer(exp->helper, helper); - + write_pnet(&exp->net, net); +#ifdef CONFIG_NF_CONNTRACK_ZONES + exp->zone = ct->zone; +#endif nf_ct_expect_related(exp, 0); nf_ct_expect_put(exp); diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index 1cbe5f1108c2..db28801b1688 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -113,8 +113,8 @@ nf_ct_exp_equal(const struct nf_conntrack_tuple *tuple, const struct net *net) { return nf_ct_tuple_mask_cmp(tuple, &i->tuple, &i->mask) && - net_eq(net, nf_ct_net(i->master)) && - nf_ct_zone_equal_any(i->master, zone); + net_eq(net, read_pnet(&i->net)) && + nf_ct_exp_zone_equal_any(i, zone); } bool nf_ct_remove_expect(struct nf_conntrack_expect *exp) @@ -326,6 +326,7 @@ void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, { struct nf_conntrack_helper *helper = NULL; struct nf_conn *ct = exp->master; + struct net *net = read_pnet(&ct->ct_net); struct nf_conn_help *help; int len; @@ -343,6 +344,10 @@ void nf_ct_expect_init(struct nf_conntrack_expect *exp, unsigned int class, helper = rcu_dereference(help->helper); rcu_assign_pointer(exp->helper, helper); + write_pnet(&exp->net, net); +#ifdef CONFIG_NF_CONNTRACK_ZONES + exp->zone = ct->zone; +#endif exp->tuple.src.l3num = family; exp->tuple.dst.protonum = proto; diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 89540112d165..6e6aeb0ab0a1 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -3538,6 +3538,7 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, struct nf_conntrack_tuple *tuple, struct nf_conntrack_tuple *mask) { + struct net *net = read_pnet(&ct->ct_net); struct nf_conntrack_expect *exp; struct nf_conn_help *help; u32 class = 0; @@ -3577,6 +3578,10 @@ ctnetlink_alloc_expect(const struct nlattr * const cda[], struct nf_conn *ct, exp->class = class; exp->master = ct; + write_pnet(&exp->net, net); +#ifdef CONFIG_NF_CONNTRACK_ZONES + exp->zone = ct->zone; +#endif if (!helper) helper = rcu_dereference(help->helper); rcu_assign_pointer(exp->helper, helper); From 3db5647984de03d9cae0dcddb509b058351f0ee4 Mon Sep 17 00:00:00 2001 From: Pablo Neira Ayuso Date: Wed, 25 Mar 2026 14:11:06 +0100 Subject: [PATCH 349/352] netfilter: nf_conntrack_expect: skip expectations in other netns via proc Skip expectations that do not reside in this netns. Similar to e77e6ff502ea ("netfilter: conntrack: do not dump other netns's conntrack entries via proc"). Fixes: 9b03f38d0487 ("netfilter: netns nf_conntrack: per-netns expectations") Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_expect.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/netfilter/nf_conntrack_expect.c b/net/netfilter/nf_conntrack_expect.c index db28801b1688..24d0576d84b7 100644 --- a/net/netfilter/nf_conntrack_expect.c +++ b/net/netfilter/nf_conntrack_expect.c @@ -652,11 +652,15 @@ static int exp_seq_show(struct seq_file *s, void *v) { struct nf_conntrack_expect *expect; struct nf_conntrack_helper *helper; + struct net *net = seq_file_net(s); struct hlist_node *n = v; char *delim = ""; expect = hlist_entry(n, struct nf_conntrack_expect, hnode); + if (!net_eq(nf_ct_exp_net(expect), net)) + return 0; + if (expect->timeout.function) seq_printf(s, "%ld ", timer_pending(&expect->timeout) ? (long)(expect->timeout.expires - jiffies)/HZ : 0); From 6a2b724460cb67caed500c508c2ae5cf012e4db4 Mon Sep 17 00:00:00 2001 From: Weiming Shi Date: Wed, 25 Mar 2026 14:11:07 +0100 Subject: [PATCH 350/352] netfilter: nf_conntrack_sip: fix use of uninitialized rtp_addr in process_sdp process_sdp() declares union nf_inet_addr rtp_addr on the stack and passes it to the nf_nat_sip sdp_session hook after walking the SDP media descriptions. However rtp_addr is only initialized inside the media loop when a recognized media type with a non-zero port is found. If the SDP body contains no m= lines, only inactive media sections (m=audio 0 ...) or only unrecognized media types, rtp_addr is never assigned. Despite that, the function still calls hooks->sdp_session() with &rtp_addr, causing nf_nat_sdp_session() to format the stale stack value as an IP address and rewrite the SDP session owner and connection lines with it. With CONFIG_INIT_STACK_ALL_ZERO (default on most distributions) this results in the session-level o= and c= addresses being rewritten to 0.0.0.0 for inactive SDP sessions. Without stack auto-init the rewritten address is whatever happened to be on the stack. Fix this by pre-initializing rtp_addr from the session-level connection address (caddr) when available, and tracking via a have_rtp_addr flag whether any valid address was established. Skip the sdp_session hook entirely when no valid address exists. Fixes: 4ab9e64e5e3c ("[NETFILTER]: nf_nat_sip: split up SDP mangling") Reported-by: Xiang Mei Signed-off-by: Weiming Shi Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_sip.c | 14 ++++++++++---- 1 file changed, 10 insertions(+), 4 deletions(-) diff --git a/net/netfilter/nf_conntrack_sip.c b/net/netfilter/nf_conntrack_sip.c index 20e57cf5c83a..939502ff7c87 100644 --- a/net/netfilter/nf_conntrack_sip.c +++ b/net/netfilter/nf_conntrack_sip.c @@ -1040,6 +1040,7 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff, unsigned int port; const struct sdp_media_type *t; int ret = NF_ACCEPT; + bool have_rtp_addr = false; hooks = rcu_dereference(nf_nat_sip_hooks); @@ -1056,8 +1057,11 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff, caddr_len = 0; if (ct_sip_parse_sdp_addr(ct, *dptr, sdpoff, *datalen, SDP_HDR_CONNECTION, SDP_HDR_MEDIA, - &matchoff, &matchlen, &caddr) > 0) + &matchoff, &matchlen, &caddr) > 0) { caddr_len = matchlen; + memcpy(&rtp_addr, &caddr, sizeof(rtp_addr)); + have_rtp_addr = true; + } mediaoff = sdpoff; for (i = 0; i < ARRAY_SIZE(sdp_media_types); ) { @@ -1091,9 +1095,11 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff, &matchoff, &matchlen, &maddr) > 0) { maddr_len = matchlen; memcpy(&rtp_addr, &maddr, sizeof(rtp_addr)); - } else if (caddr_len) + have_rtp_addr = true; + } else if (caddr_len) { memcpy(&rtp_addr, &caddr, sizeof(rtp_addr)); - else { + have_rtp_addr = true; + } else { nf_ct_helper_log(skb, ct, "cannot parse SDP message"); return NF_DROP; } @@ -1125,7 +1131,7 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff, /* Update session connection and owner addresses */ hooks = rcu_dereference(nf_nat_sip_hooks); - if (hooks && ct->status & IPS_NAT_MASK) + if (hooks && ct->status & IPS_NAT_MASK && have_rtp_addr) ret = hooks->sdp_session(skb, protoff, dataoff, dptr, datalen, sdpoff, &rtp_addr); From 8f15b5071b4548b0aafc03b366eb45c9c6566704 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Wed, 25 Mar 2026 14:11:08 +0100 Subject: [PATCH 351/352] netfilter: ctnetlink: use netlink policy range checks Replace manual range and mask validations with netlink policy annotations in ctnetlink code paths, so that the netlink core rejects invalid values early and can generate extack errors. - CTA_PROTOINFO_TCP_STATE: reject values > TCP_CONNTRACK_SYN_SENT2 at policy level, removing the manual >= TCP_CONNTRACK_MAX check. - CTA_PROTOINFO_TCP_WSCALE_ORIGINAL/REPLY: reject values > TCP_MAX_WSCALE (14). The normal TCP option parsing path already clamps to this value, but the ctnetlink path accepted 0-255, causing undefined behavior when used as a u32 shift count. - CTA_FILTER_ORIG_FLAGS/REPLY_FLAGS: use NLA_POLICY_MASK with CTA_FILTER_F_ALL, removing the manual mask checks. - CTA_EXPECT_FLAGS: use NLA_POLICY_MASK with NF_CT_EXPECT_MASK, adding a new mask define grouping all valid expect flags. Extracted from a broader nf-next patch by Florian Westphal, scoped to ctnetlink for the fixes tree. Fixes: c8e2078cfe41 ("[NETFILTER]: ctnetlink: add support for internal tcp connection tracking flags handling") Signed-off-by: David Carlier Co-developed-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- .../uapi/linux/netfilter/nf_conntrack_common.h | 4 ++++ net/netfilter/nf_conntrack_netlink.c | 16 +++++----------- net/netfilter/nf_conntrack_proto_tcp.c | 10 +++------- 3 files changed, 12 insertions(+), 18 deletions(-) diff --git a/include/uapi/linux/netfilter/nf_conntrack_common.h b/include/uapi/linux/netfilter/nf_conntrack_common.h index 26071021e986..56b6b60a814f 100644 --- a/include/uapi/linux/netfilter/nf_conntrack_common.h +++ b/include/uapi/linux/netfilter/nf_conntrack_common.h @@ -159,5 +159,9 @@ enum ip_conntrack_expect_events { #define NF_CT_EXPECT_INACTIVE 0x2 #define NF_CT_EXPECT_USERSPACE 0x4 +#ifdef __KERNEL__ +#define NF_CT_EXPECT_MASK (NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE | \ + NF_CT_EXPECT_USERSPACE) +#endif #endif /* _UAPI_NF_CONNTRACK_COMMON_H */ diff --git a/net/netfilter/nf_conntrack_netlink.c b/net/netfilter/nf_conntrack_netlink.c index 6e6aeb0ab0a1..3f408f3713bb 100644 --- a/net/netfilter/nf_conntrack_netlink.c +++ b/net/netfilter/nf_conntrack_netlink.c @@ -910,8 +910,8 @@ struct ctnetlink_filter { }; static const struct nla_policy cta_filter_nla_policy[CTA_FILTER_MAX + 1] = { - [CTA_FILTER_ORIG_FLAGS] = { .type = NLA_U32 }, - [CTA_FILTER_REPLY_FLAGS] = { .type = NLA_U32 }, + [CTA_FILTER_ORIG_FLAGS] = NLA_POLICY_MASK(NLA_U32, CTA_FILTER_F_ALL), + [CTA_FILTER_REPLY_FLAGS] = NLA_POLICY_MASK(NLA_U32, CTA_FILTER_F_ALL), }; static int ctnetlink_parse_filter(const struct nlattr *attr, @@ -925,17 +925,11 @@ static int ctnetlink_parse_filter(const struct nlattr *attr, if (ret) return ret; - if (tb[CTA_FILTER_ORIG_FLAGS]) { + if (tb[CTA_FILTER_ORIG_FLAGS]) filter->orig_flags = nla_get_u32(tb[CTA_FILTER_ORIG_FLAGS]); - if (filter->orig_flags & ~CTA_FILTER_F_ALL) - return -EOPNOTSUPP; - } - if (tb[CTA_FILTER_REPLY_FLAGS]) { + if (tb[CTA_FILTER_REPLY_FLAGS]) filter->reply_flags = nla_get_u32(tb[CTA_FILTER_REPLY_FLAGS]); - if (filter->reply_flags & ~CTA_FILTER_F_ALL) - return -EOPNOTSUPP; - } return 0; } @@ -2634,7 +2628,7 @@ static const struct nla_policy exp_nla_policy[CTA_EXPECT_MAX+1] = { [CTA_EXPECT_HELP_NAME] = { .type = NLA_NUL_STRING, .len = NF_CT_HELPER_NAME_LEN - 1 }, [CTA_EXPECT_ZONE] = { .type = NLA_U16 }, - [CTA_EXPECT_FLAGS] = { .type = NLA_U32 }, + [CTA_EXPECT_FLAGS] = NLA_POLICY_MASK(NLA_BE32, NF_CT_EXPECT_MASK), [CTA_EXPECT_CLASS] = { .type = NLA_U32 }, [CTA_EXPECT_NAT] = { .type = NLA_NESTED }, [CTA_EXPECT_FN] = { .type = NLA_NUL_STRING }, diff --git a/net/netfilter/nf_conntrack_proto_tcp.c b/net/netfilter/nf_conntrack_proto_tcp.c index 0c1d086e96cb..b67426c2189b 100644 --- a/net/netfilter/nf_conntrack_proto_tcp.c +++ b/net/netfilter/nf_conntrack_proto_tcp.c @@ -1385,9 +1385,9 @@ static int tcp_to_nlattr(struct sk_buff *skb, struct nlattr *nla, } static const struct nla_policy tcp_nla_policy[CTA_PROTOINFO_TCP_MAX+1] = { - [CTA_PROTOINFO_TCP_STATE] = { .type = NLA_U8 }, - [CTA_PROTOINFO_TCP_WSCALE_ORIGINAL] = { .type = NLA_U8 }, - [CTA_PROTOINFO_TCP_WSCALE_REPLY] = { .type = NLA_U8 }, + [CTA_PROTOINFO_TCP_STATE] = NLA_POLICY_MAX(NLA_U8, TCP_CONNTRACK_SYN_SENT2), + [CTA_PROTOINFO_TCP_WSCALE_ORIGINAL] = NLA_POLICY_MAX(NLA_U8, TCP_MAX_WSCALE), + [CTA_PROTOINFO_TCP_WSCALE_REPLY] = NLA_POLICY_MAX(NLA_U8, TCP_MAX_WSCALE), [CTA_PROTOINFO_TCP_FLAGS_ORIGINAL] = { .len = sizeof(struct nf_ct_tcp_flags) }, [CTA_PROTOINFO_TCP_FLAGS_REPLY] = { .len = sizeof(struct nf_ct_tcp_flags) }, }; @@ -1414,10 +1414,6 @@ static int nlattr_to_tcp(struct nlattr *cda[], struct nf_conn *ct) if (err < 0) return err; - if (tb[CTA_PROTOINFO_TCP_STATE] && - nla_get_u8(tb[CTA_PROTOINFO_TCP_STATE]) >= TCP_CONNTRACK_MAX) - return -EINVAL; - spin_lock_bh(&ct->lock); if (tb[CTA_PROTOINFO_TCP_STATE]) ct->proto.tcp.state = nla_get_u8(tb[CTA_PROTOINFO_TCP_STATE]); From 72d96e4e24bbefdcfbc68bdb9341a05d8f5cb6e5 Mon Sep 17 00:00:00 2001 From: Paolo Valerio Date: Mon, 23 Mar 2026 20:16:34 +0100 Subject: [PATCH 352/352] net: macb: use the current queue number for stats There's a potential mismatch between the memory reserved for statistics and the amount of memory written. gem_get_sset_count() correctly computes the number of stats based on the active queues, whereas gem_get_ethtool_stats() indiscriminately copies data using the maximum number of queues, and in the case the number of active queues is less than MACB_MAX_QUEUES, this results in a OOB write as observed in the KASAN splat. ================================================================== BUG: KASAN: vmalloc-out-of-bounds in gem_get_ethtool_stats+0x54/0x78 [macb] Write of size 760 at addr ffff80008080b000 by task ethtool/1027 CPU: [...] Tainted: [E]=UNSIGNED_MODULE Hardware name: raspberrypi rpi/rpi, BIOS 2025.10 10/01/2025 Call trace: show_stack+0x20/0x38 (C) dump_stack_lvl+0x80/0xf8 print_report+0x384/0x5e0 kasan_report+0xa0/0xf0 kasan_check_range+0xe8/0x190 __asan_memcpy+0x54/0x98 gem_get_ethtool_stats+0x54/0x78 [macb 926c13f3af83b0c6fe64badb21ec87d5e93fcf65] dev_ethtool+0x1220/0x38c0 dev_ioctl+0x4ac/0xca8 sock_do_ioctl+0x170/0x1d8 sock_ioctl+0x484/0x5d8 __arm64_sys_ioctl+0x12c/0x1b8 invoke_syscall+0xd4/0x258 el0_svc_common.constprop.0+0xb4/0x240 do_el0_svc+0x48/0x68 el0_svc+0x40/0xf8 el0t_64_sync_handler+0xa0/0xe8 el0t_64_sync+0x1b0/0x1b8 The buggy address belongs to a 1-page vmalloc region starting at 0xffff80008080b000 allocated at dev_ethtool+0x11f0/0x38c0 The buggy address belongs to the physical page: page: refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff00000a333000 pfn:0xa333 flags: 0x7fffc000000000(node=0|zone=0|lastcpupid=0x1ffff) raw: 007fffc000000000 0000000000000000 dead000000000122 0000000000000000 raw: ffff00000a333000 0000000000000000 00000001ffffffff 0000000000000000 page dumped because: kasan: bad access detected Memory state around the buggy address: ffff80008080b080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ffff80008080b100: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 >ffff80008080b180: 00 00 00 00 00 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ^ ffff80008080b200: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ffff80008080b280: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 ================================================================== Fix it by making sure the copied size only considers the active number of queues. Fixes: 512286bbd4b7 ("net: macb: Added some queue statistics") Signed-off-by: Paolo Valerio Reviewed-by: Nicolai Buchwitz Link: https://patch.msgid.link/20260323191634.2185840-1-pvalerio@redhat.com Signed-off-by: Paolo Abeni --- drivers/net/ethernet/cadence/macb_main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c index fc4dd0757646..99e7d5cf3786 100644 --- a/drivers/net/ethernet/cadence/macb_main.c +++ b/drivers/net/ethernet/cadence/macb_main.c @@ -3224,7 +3224,7 @@ static void gem_get_ethtool_stats(struct net_device *dev, spin_lock_irq(&bp->stats_lock); gem_update_stats(bp); memcpy(data, &bp->ethtool_stats, sizeof(u64) - * (GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES)); + * (GEM_STATS_LEN + QUEUE_STATS_LEN * bp->num_queues)); spin_unlock_irq(&bp->stats_lock); }