From 387a926ee166814611acecb960207fe2f3c4fd3e Mon Sep 17 00:00:00 2001 From: Amirreza Zarrabi Date: Mon, 16 Feb 2026 14:24:06 -0800 Subject: [PATCH 001/329] tee: optee: prevent use-after-free when the client exits before the supplicant Commit 70b0d6b0a199 ("tee: optee: Fix supplicant wait loop") made the client wait as killable so it can be interrupted during shutdown or after a supplicant crash. This changes the original lifetime expectations: the client task can now terminate while the supplicant is still processing its request. If the client exits first it removes the request from its queue and kfree()s it, while the request ID remains in supp->idr. A subsequent lookup on the supplicant path then dereferences freed memory, leading to a use-after-free. Serialise access to the request with supp->mutex: * Hold supp->mutex in optee_supp_recv() and optee_supp_send() while looking up and touching the request. * Let optee_supp_thrd_req() notice that the client has terminated and signal optee_supp_send() accordingly. With these changes the request cannot be freed while the supplicant still has a reference, eliminating the race. Fixes: 70b0d6b0a199 ("tee: optee: Fix supplicant wait loop") Signed-off-by: Amirreza Zarrabi Tested-by: Ox Yeh Reviewed-by: Sumit Garg Signed-off-by: Jens Wiklander --- drivers/tee/optee/supp.c | 107 +++++++++++++++++++++++++++------------ 1 file changed, 74 insertions(+), 33 deletions(-) diff --git a/drivers/tee/optee/supp.c b/drivers/tee/optee/supp.c index a3d11b1f90fa..06747e90c230 100644 --- a/drivers/tee/optee/supp.c +++ b/drivers/tee/optee/supp.c @@ -10,7 +10,11 @@ struct optee_supp_req { struct list_head link; + int id; + bool in_queue; + bool processed; + u32 func; u32 ret; size_t num_params; @@ -19,6 +23,9 @@ struct optee_supp_req { struct completion c; }; +/* It is temporary request used for revoked pending request in supp->idr. */ +#define INVALID_REQ_PTR ((struct optee_supp_req *)ERR_PTR(-EBADF)) + void optee_supp_init(struct optee_supp *supp) { memset(supp, 0, sizeof(*supp)); @@ -39,21 +46,23 @@ void optee_supp_release(struct optee_supp *supp) { int id; struct optee_supp_req *req; - struct optee_supp_req *req_tmp; mutex_lock(&supp->mutex); - /* Abort all request retrieved by supplicant */ + /* Abort all request */ idr_for_each_entry(&supp->idr, req, id) { idr_remove(&supp->idr, id); - req->ret = TEEC_ERROR_COMMUNICATION; - complete(&req->c); - } + /* Skip if request was already marked invalid */ + if (IS_ERR(req)) + continue; - /* Abort all queued requests */ - list_for_each_entry_safe(req, req_tmp, &supp->reqs, link) { - list_del(&req->link); - req->in_queue = false; + /* For queued requests where supplicant has not seen it */ + if (req->in_queue) { + list_del(&req->link); + req->in_queue = false; + } + + req->processed = true; req->ret = TEEC_ERROR_COMMUNICATION; complete(&req->c); } @@ -100,8 +109,16 @@ u32 optee_supp_thrd_req(struct tee_context *ctx, u32 func, size_t num_params, /* Insert the request in the request list */ mutex_lock(&supp->mutex); + req->id = idr_alloc(&supp->idr, req, 1, 0, GFP_KERNEL); + if (req->id < 0) { + mutex_unlock(&supp->mutex); + kfree(req); + return TEEC_ERROR_OUT_OF_MEMORY; + } + list_add_tail(&req->link, &supp->reqs); req->in_queue = true; + req->processed = false; mutex_unlock(&supp->mutex); /* Tell an eventual waiter there's a new request */ @@ -117,21 +134,43 @@ u32 optee_supp_thrd_req(struct tee_context *ctx, u32 func, size_t num_params, if (wait_for_completion_killable(&req->c)) { mutex_lock(&supp->mutex); if (req->in_queue) { + /* Supplicant has not seen this request yet. */ + idr_remove(&supp->idr, req->id); list_del(&req->link); req->in_queue = false; + + ret = TEEC_ERROR_COMMUNICATION; + } else if (req->processed) { + /* + * Supplicant has processed this request. Ignore the + * kill signal for now and submit the result. req is not + * in supp->reqs (removed by supp_pop_entry()) nor in + * supp->idr (removed by supp_pop_req()). + */ + ret = req->ret; + } else { + /* + * Supplicant is in the middle of processing this + * request. Replace req with INVALID_REQ_PTR so that + * the ID remains busy, causing optee_supp_send() to + * fail on the next call to supp_pop_req() with this ID. + */ + idr_replace(&supp->idr, INVALID_REQ_PTR, req->id); + ret = TEEC_ERROR_COMMUNICATION; } + mutex_unlock(&supp->mutex); - req->ret = TEEC_ERROR_COMMUNICATION; + } else { + ret = req->ret; } - ret = req->ret; kfree(req); return ret; } static struct optee_supp_req *supp_pop_entry(struct optee_supp *supp, - int num_params, int *id) + int num_params) { struct optee_supp_req *req; @@ -153,10 +192,6 @@ static struct optee_supp_req *supp_pop_entry(struct optee_supp *supp, return ERR_PTR(-EINVAL); } - *id = idr_alloc(&supp->idr, req, 1, 0, GFP_KERNEL); - if (*id < 0) - return ERR_PTR(-ENOMEM); - list_del(&req->link); req->in_queue = false; @@ -214,7 +249,6 @@ int optee_supp_recv(struct tee_context *ctx, u32 *func, u32 *num_params, struct optee *optee = tee_get_drvdata(teedev); struct optee_supp *supp = &optee->supp; struct optee_supp_req *req = NULL; - int id; size_t num_meta; int rc; @@ -224,15 +258,11 @@ int optee_supp_recv(struct tee_context *ctx, u32 *func, u32 *num_params, while (true) { mutex_lock(&supp->mutex); - req = supp_pop_entry(supp, *num_params - num_meta, &id); + req = supp_pop_entry(supp, *num_params - num_meta); + if (req) + break; /* Keep mutex held. */ mutex_unlock(&supp->mutex); - if (req) { - if (IS_ERR(req)) - return PTR_ERR(req); - break; - } - /* * If we didn't get a request we'll block in * wait_for_completion() to avoid needless spinning. @@ -245,6 +275,13 @@ int optee_supp_recv(struct tee_context *ctx, u32 *func, u32 *num_params, return -ERESTARTSYS; } + /* supp->mutex held and req != NULL. */ + + if (IS_ERR(req)) { + mutex_unlock(&supp->mutex); + return PTR_ERR(req); + } + if (num_meta) { /* * tee-supplicant support meta parameters -> requsts can be @@ -252,13 +289,11 @@ int optee_supp_recv(struct tee_context *ctx, u32 *func, u32 *num_params, */ param->attr = TEE_IOCTL_PARAM_ATTR_TYPE_VALUE_INOUT | TEE_IOCTL_PARAM_ATTR_META; - param->u.value.a = id; + param->u.value.a = req->id; param->u.value.b = 0; param->u.value.c = 0; } else { - mutex_lock(&supp->mutex); - supp->req_id = id; - mutex_unlock(&supp->mutex); + supp->req_id = req->id; } *func = req->func; @@ -266,6 +301,7 @@ int optee_supp_recv(struct tee_context *ctx, u32 *func, u32 *num_params, memcpy(param + num_meta, req->param, sizeof(struct tee_param) * req->num_params); + mutex_unlock(&supp->mutex); return 0; } @@ -297,12 +333,17 @@ static struct optee_supp_req *supp_pop_req(struct optee_supp *supp, if (!req) return ERR_PTR(-ENOENT); + /* optee_supp_thrd_req() already returned to optee. */ + if (IS_ERR(req)) + goto failed_req; + if ((num_params - nm) != req->num_params) return ERR_PTR(-EINVAL); + *num_meta = nm; +failed_req: idr_remove(&supp->idr, id); supp->req_id = -1; - *num_meta = nm; return req; } @@ -328,10 +369,9 @@ int optee_supp_send(struct tee_context *ctx, u32 ret, u32 num_params, mutex_lock(&supp->mutex); req = supp_pop_req(supp, num_params, param, &num_meta); - mutex_unlock(&supp->mutex); - if (IS_ERR(req)) { - /* Something is wrong, let supplicant restart. */ + mutex_unlock(&supp->mutex); + /* Something is wrong, let supplicant handel it. */ return PTR_ERR(req); } @@ -355,9 +395,10 @@ int optee_supp_send(struct tee_context *ctx, u32 ret, u32 num_params, } } req->ret = ret; - + req->processed = true; /* Let the requesting thread continue */ complete(&req->c); + mutex_unlock(&supp->mutex); return 0; } From 48fcc895403cc97aa6c776cb65e6aa11290c0b44 Mon Sep 17 00:00:00 2001 From: Wei Liu Date: Thu, 23 Apr 2026 17:26:26 +0000 Subject: [PATCH 002/329] mshv: add a missing padding field MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit That was missed when importing the header. Reported-by: Doru Blânzeanu Reported-by: Magnus Kulke Fixes: e68bda71a2384 ("hyperv: Add new Hyper-V headers in include/hyperv") Cc: stable@kernel.org Reviewed-by: Easwar Hariharan Signed-off-by: Wei Liu --- include/hyperv/hvhdk.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/hyperv/hvhdk.h b/include/hyperv/hvhdk.h index 5e83d3714966..0c89c62c9706 100644 --- a/include/hyperv/hvhdk.h +++ b/include/hyperv/hvhdk.h @@ -79,6 +79,7 @@ struct hv_vp_register_page { u64 registers[18]; }; + u8 reserved[8]; /* Volatile XMM registers (HV_X64_REGISTER_CLASS_XMM) */ union { struct { From 3fcf923302a8f5c0dc3af3d2ca2657cb5fae4297 Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Tue, 14 Apr 2026 13:10:08 +0200 Subject: [PATCH 003/329] hv: utils: handle and propagate errors in kvp_register Make kvp_register() return an error code instead of silently ignoring failures, and propagate the error from kvp_handle_handshake() instead of returning success. This propagates both kzalloc_obj() and hvutil_transport_send() failures to kvp_handle_handshake() and thus to kvp_on_msg(). Fixes: 245ba56a52a3 ("Staging: hv: Implement key/value pair (KVP)") Cc: stable@vger.kernel.org Signed-off-by: Thorsten Blum Reviewed-by: Long Li Signed-off-by: Wei Liu --- drivers/hv/hv_kvp.c | 25 +++++++++++++------------ 1 file changed, 13 insertions(+), 12 deletions(-) diff --git a/drivers/hv/hv_kvp.c b/drivers/hv/hv_kvp.c index 0d73daf745a7..6180ebe040ff 100644 --- a/drivers/hv/hv_kvp.c +++ b/drivers/hv/hv_kvp.c @@ -93,7 +93,7 @@ static void kvp_send_key(struct work_struct *dummy); static void kvp_respond_to_host(struct hv_kvp_msg *msg, int error); static void kvp_timeout_func(struct work_struct *dummy); static void kvp_host_handshake_func(struct work_struct *dummy); -static void kvp_register(int); +static int kvp_register(int); static DECLARE_DELAYED_WORK(kvp_timeout_work, kvp_timeout_func); static DECLARE_DELAYED_WORK(kvp_host_handshake_work, kvp_host_handshake_func); @@ -127,24 +127,26 @@ static void kvp_register_done(void) hv_poll_channel(kvp_transaction.recv_channel, kvp_poll_wrapper); } -static void +static int kvp_register(int reg_value) { struct hv_kvp_msg *kvp_msg; char *version; + int ret; kvp_msg = kzalloc_obj(*kvp_msg); + if (!kvp_msg) + return -ENOMEM; - if (kvp_msg) { - version = kvp_msg->body.kvp_register.version; - kvp_msg->kvp_hdr.operation = reg_value; - strcpy(version, HV_DRV_VERSION); + version = kvp_msg->body.kvp_register.version; + kvp_msg->kvp_hdr.operation = reg_value; + strcpy(version, HV_DRV_VERSION); - hvutil_transport_send(hvt, kvp_msg, sizeof(*kvp_msg), - kvp_register_done); - kfree(kvp_msg); - } + ret = hvutil_transport_send(hvt, kvp_msg, sizeof(*kvp_msg), + kvp_register_done); + kfree(kvp_msg); + return ret; } static void kvp_timeout_func(struct work_struct *dummy) @@ -186,9 +188,8 @@ static int kvp_handle_handshake(struct hv_kvp_msg *msg) */ pr_debug("KVP: userspace daemon ver. %d connected\n", msg->kvp_hdr.operation); - kvp_register(dm_reg_value); - return 0; + return kvp_register(dm_reg_value); } From 3f8c8497b4fc249e27cb335c627114d8412e584d Mon Sep 17 00:00:00 2001 From: Thorsten Blum Date: Tue, 28 Apr 2026 19:11:05 +0200 Subject: [PATCH 004/329] hv: utils: replace deprecated strcpy with strscpy in kvp_register strcpy() has been deprecated [1] because it performs no bounds checking on the destination buffer, which can lead to buffer overflows. While the current code works correctly, replace strcpy() with the safer strscpy() to follow secure coding best practices. Use ->body.kvp_register.version directly as the destination buffer and remove the local variable. [1] https://www.kernel.org/doc/html/latest/process/deprecated.html#strcpy Signed-off-by: Thorsten Blum Signed-off-by: Wei Liu --- drivers/hv/hv_kvp.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/hv/hv_kvp.c b/drivers/hv/hv_kvp.c index 6180ebe040ff..336b278b2182 100644 --- a/drivers/hv/hv_kvp.c +++ b/drivers/hv/hv_kvp.c @@ -27,6 +27,7 @@ #include #include #include +#include #include #include "hyperv_vmbus.h" @@ -130,18 +131,15 @@ static void kvp_register_done(void) static int kvp_register(int reg_value) { - struct hv_kvp_msg *kvp_msg; - char *version; int ret; kvp_msg = kzalloc_obj(*kvp_msg); if (!kvp_msg) return -ENOMEM; - version = kvp_msg->body.kvp_register.version; kvp_msg->kvp_hdr.operation = reg_value; - strcpy(version, HV_DRV_VERSION); + strscpy(kvp_msg->body.kvp_register.version, HV_DRV_VERSION); ret = hvutil_transport_send(hvt, kvp_msg, sizeof(*kvp_msg), kvp_register_done); From f1a9e67c11388965802a61922c313bfc43272afe Mon Sep 17 00:00:00 2001 From: Jork Loeser Date: Mon, 27 Apr 2026 14:38:52 -0700 Subject: [PATCH 005/329] mshv: limit SynIC management to MSHV-owned resources The SynIC is shared between VMBus and MSHV. VMBus owns the message page (SIMP), event flags page (SIEFP), global enable (SCONTROL), and SINT2. MSHV adds SINT0, SINT5, and the event ring page (SIRBP). Currently mshv_synic_cpu_init() redundantly enables SIMP, SIEFP, and SCONTROL that VMBus already configured, and mshv_synic_cpu_exit() disables all of them. This is wrong because MSHV can be torn down while VMBus is still active. In particular, a kexec reboot notifier tears down MSHV first. Disabling SCONTROL, SIMP, and SIEFP out from under VMBus causes its later cleanup to write SynIC MSRs while SynIC is disabled, which the hypervisor does not tolerate. Restrict MSHV to managing only the resources it owns: - SINT0, SINT5: mask on cleanup, unmask on init - SIRBP: enable/disable as before - SIMP, SIEFP, SCONTROL: leave to VMBus when it is active (L1VH and nested root partition); on a non-nested root partition VMBus does not run, so MSHV must enable/disable them While here, fix the SIEFP and SIRBP memremap() and virt_to_phys() calls to use HV_HYP_PAGE_SHIFT/HV_HYP_PAGE_SIZE instead of PAGE_SHIFT/PAGE_SIZE. The hypervisor always uses 4K pages for SynIC register GPAs regardless of the kernel page size, so using PAGE_SHIFT produces wrong addresses on ARM64 with 64K pages. Note that initialization order matters - VMBUS first, MSHV second, and the reverse on de-init. Ideally, we would want a dedicated SYNIC driver that replaces the cross-dependencies with a clear API and dynamic tracking. Such refactor should go into its own dedicated series, outside of this kexec fix series. Signed-off-by: Jork Loeser Reviewed-by: Anirudh Rayabharam (Microsoft) Reviewed-by: Stanislav Kinsburskii Signed-off-by: Wei Liu --- drivers/hv/hv.c | 3 + drivers/hv/mshv_synic.c | 152 ++++++++++++++++++++++++++-------------- 2 files changed, 104 insertions(+), 51 deletions(-) diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c index ae60fd542292..ef4b1b03395d 100644 --- a/drivers/hv/hv.c +++ b/drivers/hv/hv.c @@ -272,6 +272,9 @@ void hv_synic_free(void) /* * hv_hyp_synic_enable_regs - Initialize the Synthetic Interrupt Controller * with the hypervisor. + * + * Note: When MSHV is present, mshv_synic_cpu_init() intializes further + * registers later. */ void hv_hyp_synic_enable_regs(unsigned int cpu) { diff --git a/drivers/hv/mshv_synic.c b/drivers/hv/mshv_synic.c index e2288a726fec..2db3b0192eac 100644 --- a/drivers/hv/mshv_synic.c +++ b/drivers/hv/mshv_synic.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include #include @@ -456,46 +457,75 @@ static int mshv_synic_cpu_init(unsigned int cpu) union hv_synic_siefp siefp; union hv_synic_sirbp sirbp; union hv_synic_sint sint; - 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; struct hv_synic_event_flags_page **event_flags_page = &spages->synic_event_flags_page; struct hv_synic_event_ring_page **event_ring_page = &spages->synic_event_ring_page; + /* + * VMBus owns SIMP/SIEFP/SCONTROL when it is active. + * See hv_hyp_synic_enable_regs() for that initialization. + */ + bool vmbus_active = hv_vmbus_exists(); - /* Setup the Synic's message page */ + /* + * Map the SYNIC message page. When VMBus is not active the + * hypervisor pre-provisions the SIMP GPA but may not set + * simp_enabled — enable it here. + */ simp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIMP); - simp.simp_enabled = true; + if (!vmbus_active) { + simp.simp_enabled = true; + hv_set_non_nested_msr(HV_MSR_SIMP, simp.as_uint64); + } *msg_page = memremap(simp.base_simp_gpa << HV_HYP_PAGE_SHIFT, HV_HYP_PAGE_SIZE, MEMREMAP_WB); if (!(*msg_page)) - return -EFAULT; + goto cleanup_simp; - hv_set_non_nested_msr(HV_MSR_SIMP, simp.as_uint64); - - /* Setup the Synic's event flags page */ + /* + * Map the event flags page. Same as SIMP: enable when + * VMBus is not active, already enabled by VMBus otherwise. + */ siefp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIEFP); - siefp.siefp_enabled = true; - *event_flags_page = memremap(siefp.base_siefp_gpa << PAGE_SHIFT, - PAGE_SIZE, MEMREMAP_WB); + if (!vmbus_active) { + siefp.siefp_enabled = true; + hv_set_non_nested_msr(HV_MSR_SIEFP, siefp.as_uint64); + } + *event_flags_page = memremap(siefp.base_siefp_gpa << HV_HYP_PAGE_SHIFT, + HV_HYP_PAGE_SIZE, MEMREMAP_WB); if (!(*event_flags_page)) - goto cleanup; - - hv_set_non_nested_msr(HV_MSR_SIEFP, siefp.as_uint64); + goto cleanup_siefp; /* Setup the Synic's event ring page */ sirbp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIRBP); + + if (hv_root_partition()) { + *event_ring_page = memremap(sirbp.base_sirbp_gpa << HV_HYP_PAGE_SHIFT, + HV_HYP_PAGE_SIZE, MEMREMAP_WB); + + if (!(*event_ring_page)) + goto cleanup_siefp; + } else { + /* + * On L1VH the hypervisor does not provide a SIRBP page. + * Allocate one and program its GPA into the MSR. + */ + *event_ring_page = (struct hv_synic_event_ring_page *) + get_zeroed_page(GFP_KERNEL); + + if (!(*event_ring_page)) + goto cleanup_siefp; + + sirbp.base_sirbp_gpa = virt_to_phys(*event_ring_page) + >> HV_HYP_PAGE_SHIFT; + } + sirbp.sirbp_enabled = true; - *event_ring_page = memremap(sirbp.base_sirbp_gpa << PAGE_SHIFT, - PAGE_SIZE, MEMREMAP_WB); - - if (!(*event_ring_page)) - goto cleanup; - hv_set_non_nested_msr(HV_MSR_SIRBP, sirbp.as_uint64); if (mshv_sint_irq != -1) @@ -518,28 +548,30 @@ static int mshv_synic_cpu_init(unsigned int cpu) hv_set_non_nested_msr(HV_MSR_SINT0 + HV_SYNIC_DOORBELL_SINT_INDEX, sint.as_uint64); - /* Enable global synic bit */ - sctrl.as_uint64 = hv_get_non_nested_msr(HV_MSR_SCONTROL); - sctrl.enable = 1; - hv_set_non_nested_msr(HV_MSR_SCONTROL, sctrl.as_uint64); + /* When VMBus is active it already enabled SCONTROL. */ + if (!vmbus_active) { + union hv_synic_scontrol sctrl; + + sctrl.as_uint64 = hv_get_non_nested_msr(HV_MSR_SCONTROL); + sctrl.enable = 1; + hv_set_non_nested_msr(HV_MSR_SCONTROL, sctrl.as_uint64); + } return 0; -cleanup: - if (*event_ring_page) { - sirbp.sirbp_enabled = false; - hv_set_non_nested_msr(HV_MSR_SIRBP, sirbp.as_uint64); - memunmap(*event_ring_page); - } - if (*event_flags_page) { +cleanup_siefp: + if (*event_flags_page) + memunmap(*event_flags_page); + if (!vmbus_active) { siefp.siefp_enabled = false; hv_set_non_nested_msr(HV_MSR_SIEFP, siefp.as_uint64); - memunmap(*event_flags_page); } - if (*msg_page) { +cleanup_simp: + if (*msg_page) + memunmap(*msg_page); + if (!vmbus_active) { simp.simp_enabled = false; hv_set_non_nested_msr(HV_MSR_SIMP, simp.as_uint64); - memunmap(*msg_page); } return -EFAULT; @@ -548,16 +580,15 @@ static int mshv_synic_cpu_init(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(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; struct hv_synic_event_ring_page **event_ring_page = &spages->synic_event_ring_page; + /* VMBus owns SIMP/SIEFP/SCONTROL when it is active */ + bool vmbus_active = hv_vmbus_exists(); /* Disable the interrupt */ sint.as_uint64 = hv_get_non_nested_msr(HV_MSR_SINT0 + HV_SYNIC_INTERCEPTION_SINT_INDEX); @@ -574,28 +605,47 @@ static int mshv_synic_cpu_exit(unsigned int cpu) if (mshv_sint_irq != -1) disable_percpu_irq(mshv_sint_irq); - /* Disable Synic's event ring page */ + /* Disable SYNIC event ring page owned by MSHV */ sirbp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIRBP); sirbp.sirbp_enabled = false; - hv_set_non_nested_msr(HV_MSR_SIRBP, sirbp.as_uint64); - memunmap(*event_ring_page); - /* Disable Synic's event flags page */ - siefp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIEFP); - siefp.siefp_enabled = false; - hv_set_non_nested_msr(HV_MSR_SIEFP, siefp.as_uint64); + if (hv_root_partition()) { + hv_set_non_nested_msr(HV_MSR_SIRBP, sirbp.as_uint64); + memunmap(*event_ring_page); + } else { + sirbp.base_sirbp_gpa = 0; + hv_set_non_nested_msr(HV_MSR_SIRBP, sirbp.as_uint64); + free_page((unsigned long)*event_ring_page); + } + + /* + * Release our mappings of the message and event flags pages. + * When VMBus is not active, we enabled SIMP/SIEFP — disable + * them. Otherwise VMBus owns the MSRs — leave them. + */ memunmap(*event_flags_page); + if (!vmbus_active) { + union hv_synic_simp simp; + union hv_synic_siefp siefp; - /* Disable Synic's message page */ - simp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIMP); - simp.simp_enabled = false; - hv_set_non_nested_msr(HV_MSR_SIMP, simp.as_uint64); + siefp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIEFP); + siefp.siefp_enabled = false; + hv_set_non_nested_msr(HV_MSR_SIEFP, siefp.as_uint64); + + simp.as_uint64 = hv_get_non_nested_msr(HV_MSR_SIMP); + simp.simp_enabled = false; + hv_set_non_nested_msr(HV_MSR_SIMP, simp.as_uint64); + } memunmap(*msg_page); - /* Disable global synic bit */ - sctrl.as_uint64 = hv_get_non_nested_msr(HV_MSR_SCONTROL); - sctrl.enable = 0; - hv_set_non_nested_msr(HV_MSR_SCONTROL, sctrl.as_uint64); + /* When VMBus is active it owns SCONTROL — leave it. */ + if (!vmbus_active) { + union hv_synic_scontrol sctrl; + + sctrl.as_uint64 = hv_get_non_nested_msr(HV_MSR_SCONTROL); + sctrl.enable = 0; + hv_set_non_nested_msr(HV_MSR_SCONTROL, sctrl.as_uint64); + } return 0; } From efe0fb8c3fe2b996522f7418fd311eeff43c1148 Mon Sep 17 00:00:00 2001 From: Jork Loeser Date: Mon, 27 Apr 2026 14:38:53 -0700 Subject: [PATCH 006/329] mshv: clean up SynIC state on kexec for L1VH The reboot notifier that tears down the SynIC cpuhp state guards the cleanup with hv_root_partition(), so on L1VH (where hv_root_partition() is false) SINT0, SINT5, and SIRBP are never cleaned up before kexec. The kexec'd kernel then inherits stale unmasked SINTs and an enabled SIRBP pointing to freed memory. Remove the hv_root_partition() guard so the cleanup runs for all parent partitions. Signed-off-by: Jork Loeser Reviewed-by: Stanislav Kinsburskii Reviewed-by: Anirudh Rayabharam (Microsoft) Signed-off-by: Wei Liu --- drivers/hv/mshv_synic.c | 3 --- 1 file changed, 3 deletions(-) diff --git a/drivers/hv/mshv_synic.c b/drivers/hv/mshv_synic.c index 2db3b0192eac..978a1cace341 100644 --- a/drivers/hv/mshv_synic.c +++ b/drivers/hv/mshv_synic.c @@ -723,9 +723,6 @@ mshv_unregister_doorbell(u64 partition_id, int 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; } From 6e55a8d9f21bcb238596bdd8092c92e74c698a3d Mon Sep 17 00:00:00 2001 From: Jork Loeser Date: Mon, 27 Apr 2026 14:38:54 -0700 Subject: [PATCH 007/329] mshv: unmap debugfs stats pages on kexec On L1VH, debugfs stats pages are overlay pages: the kernel allocates them and registers the GPAs with the hypervisor via HVCALL_MAP_STATS_PAGE2. These overlay mappings persist in the hypervisor across kexec. If the kexec'd kernel reuses those physical pages, the hypervisor's overlay semantics cause a machine check exception. Fix this by calling mshv_debugfs_exit() from the reboot notifier, which issues HVCALL_UNMAP_STATS_PAGE for each mapped stats page before kexec. This releases the overlay bindings so the physical pages can be safely reused. Guard mshv_debugfs_exit() against being called when init failed. Signed-off-by: Jork Loeser Reviewed-by: Anirudh Rayabharam (Microsoft) Reviewed-by: Stanislav Kinsburskii Signed-off-by: Wei Liu --- drivers/hv/mshv_debugfs.c | 7 ++++++- drivers/hv/mshv_synic.c | 1 + 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/hv/mshv_debugfs.c b/drivers/hv/mshv_debugfs.c index 418b6dc8f3c2..3c3e02237ae9 100644 --- a/drivers/hv/mshv_debugfs.c +++ b/drivers/hv/mshv_debugfs.c @@ -674,8 +674,10 @@ int __init mshv_debugfs_init(void) mshv_debugfs = debugfs_create_dir("mshv", NULL); if (IS_ERR(mshv_debugfs)) { + err = PTR_ERR(mshv_debugfs); + mshv_debugfs = NULL; pr_err("%s: failed to create debugfs directory\n", __func__); - return PTR_ERR(mshv_debugfs); + return err; } if (hv_root_partition()) { @@ -710,6 +712,9 @@ int __init mshv_debugfs_init(void) void mshv_debugfs_exit(void) { + if (!mshv_debugfs) + return; + mshv_debugfs_parent_partition_remove(); if (hv_root_partition()) { diff --git a/drivers/hv/mshv_synic.c b/drivers/hv/mshv_synic.c index 978a1cace341..88170ce6b83f 100644 --- a/drivers/hv/mshv_synic.c +++ b/drivers/hv/mshv_synic.c @@ -723,6 +723,7 @@ mshv_unregister_doorbell(u64 partition_id, int doorbell_portid) static int mshv_synic_reboot_notify(struct notifier_block *nb, unsigned long code, void *unused) { + mshv_debugfs_exit(); cpuhp_remove_state(synic_cpuhp_online); return 0; } From 754d60ad1c91895be0bc7d771fbf9fb3c9448640 Mon Sep 17 00:00:00 2001 From: Alexander Dahl Date: Wed, 29 Apr 2026 14:59:30 +0200 Subject: [PATCH 008/329] memory: atmel-ebi: Allow deferred probing After removing of_platform_default_populate() calls the atmel-ebi driver was affected by deferred probing. platform_driver_probe() is incompatible with deferred probing. This led to atmel-ebi driver eventually not being probed on at91 sam9x60-curiosity and other sam9x60 based boards. Subsequently the nand-controller driver (nand-controller being a child node of ebi) on that platform was not probed and thus raw NAND flash was inaccessible, preventing devices to boot with rootfs on raw NAND flash (e.g. with UBI/UBIFS). Fixes: 0b0f7e6539a7 ("ARM: at91: remove unnecessary of_platform_default_populate calls") Cc: stable@vger.kernel.org Suggested-by: Miquel Raynal Signed-off-by: Alexander Dahl Link: https://patch.msgid.link/20260429125930.844790-1-ada@thorsis.com Signed-off-by: Krzysztof Kozlowski --- drivers/memory/atmel-ebi.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/memory/atmel-ebi.c b/drivers/memory/atmel-ebi.c index 8db970da9af9..1e8e8aba2542 100644 --- a/drivers/memory/atmel-ebi.c +++ b/drivers/memory/atmel-ebi.c @@ -628,10 +628,11 @@ static __maybe_unused int atmel_ebi_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(atmel_ebi_pm_ops, NULL, atmel_ebi_resume); static struct platform_driver atmel_ebi_driver = { + .probe = atmel_ebi_probe, .driver = { .name = "atmel-ebi", .of_match_table = atmel_ebi_id_table, .pm = &atmel_ebi_pm_ops, }, }; -builtin_platform_driver_probe(atmel_ebi_driver, atmel_ebi_probe); +builtin_platform_driver(atmel_ebi_driver); From d25e5cbac4e2287c843fc8b45c50dd8e57e3a696 Mon Sep 17 00:00:00 2001 From: Andy Shevchenko Date: Mon, 4 May 2026 12:00:28 +0200 Subject: [PATCH 009/329] auxdisplay: max6959: use regmap_assign_bits() in max6959_enable() Replace the ternary with a direct call to the regmap_assign_bits() helper and save a couple lines of code. Signed-off-by: Andy Shevchenko --- drivers/auxdisplay/max6959.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/auxdisplay/max6959.c b/drivers/auxdisplay/max6959.c index 6bbc8d48fb1b..3bdef099a225 100644 --- a/drivers/auxdisplay/max6959.c +++ b/drivers/auxdisplay/max6959.c @@ -86,10 +86,7 @@ static const struct linedisp_ops max6959_linedisp_ops = { static int max6959_enable(struct max6959_priv *priv, bool enable) { - u8 mask = REG_CONFIGURATION_S_BIT; - u8 value = enable ? mask : 0; - - return regmap_update_bits(priv->regmap, REG_CONFIGURATION, mask, value); + return regmap_assign_bits(priv->regmap, REG_CONFIGURATION, REG_CONFIGURATION_S_BIT, enable); } static void max6959_power_off(void *priv) From 2c6821657ce3b3c85f92719ea81ec9f9ff27df11 Mon Sep 17 00:00:00 2001 From: Peng Fan Date: Mon, 27 Apr 2026 09:01:48 +0800 Subject: [PATCH 010/329] soc: imx8m: Fix match data lookup for soc device The i.MX8M soc device is registered via platform_device_register_simple(), so it is not associated with a Device Tree node and the imx8m_soc_driver has no of_match_table. As a result, device_get_match_data() always returns NULL when probing the soc device. Retrieve the match data directly from the machine compatible using of_machine_get_match_data(imx8_soc_match), which provides the correct SoC data. Fixes: 2524b293a59e5 ("soc: imx8m: don't access of_root directly") Signed-off-by: Peng Fan Reviewed-by: Lucas Stach Signed-off-by: Frank Li --- drivers/soc/imx/soc-imx8m.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soc/imx/soc-imx8m.c b/drivers/soc/imx/soc-imx8m.c index 77763a107edb..fc080e56f50d 100644 --- a/drivers/soc/imx/soc-imx8m.c +++ b/drivers/soc/imx/soc-imx8m.c @@ -247,7 +247,7 @@ static int imx8m_soc_probe(struct platform_device *pdev) if (ret) return ret; - data = device_get_match_data(dev); + data = of_machine_get_match_data(imx8_soc_match); if (data) { soc_dev_attr->soc_id = data->name; ret = imx8m_soc_prepare(pdev, data->ocotp_compatible); From e27264daac7d9ce892a2a5b4a864d6d9a3c9276a Mon Sep 17 00:00:00 2001 From: Harshal Dev Date: Thu, 16 Apr 2026 17:29:18 +0530 Subject: [PATCH 011/329] dt-bindings: crypto: qcom,ice: Fix missing power-domain and iface clk The DT bindings for inline-crypto engine do not specify the UFS_PHY_GDSC power-domain and iface clock. Without enabling the iface clock and the associated power-domain the ICE hardware cannot function correctly and leads to unclocked hardware accesses being observed during probe. Fix the DT bindings for inline-crypto engine to require the UFS_PHY_GDSC power-domain and iface clock for new devices (Eliza and Milos) introduced in the current release (7.1) with yet-to-stabilize ABI, while preserving backward compatibility for older devices. Fixes: 618195a7ac3df ("dt-bindings: crypto: qcom,inline-crypto-engine: Document the Eliza ICE") Fixes: 85faec1e85555 ("dt-bindings: crypto: qcom,inline-crypto-engine: document the Milos ICE") Reviewed-by: Kuldeep Singh Reviewed-by: Krzysztof Kozlowski Signed-off-by: Harshal Dev Link: https://lore.kernel.org/r/20260416-qcom_ice_power_and_clk_vote-v5-1-5ccf5d7e2846@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- .../crypto/qcom,inline-crypto-engine.yaml | 35 ++++++++++++++++++- 1 file changed, 34 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml index 876bf90ed96e..ccb6b8dd8e11 100644 --- a/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml +++ b/Documentation/devicetree/bindings/crypto/qcom,inline-crypto-engine.yaml @@ -30,6 +30,16 @@ properties: maxItems: 1 clocks: + minItems: 1 + maxItems: 2 + + clock-names: + minItems: 1 + items: + - const: core + - const: iface + + power-domains: maxItems: 1 operating-points-v2: true @@ -44,6 +54,25 @@ required: additionalProperties: false +allOf: + - if: + properties: + compatible: + contains: + enum: + - qcom,eliza-inline-crypto-engine + - qcom,milos-inline-crypto-engine + + then: + required: + - power-domains + - clock-names + properties: + clocks: + minItems: 2 + clock-names: + minItems: 2 + examples: - | #include @@ -52,7 +81,11 @@ examples: compatible = "qcom,sm8550-inline-crypto-engine", "qcom,inline-crypto-engine"; reg = <0x01d88000 0x8000>; - clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>; + clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>, + <&gcc GCC_UFS_PHY_AHB_CLK>; + clock-names = "core", + "iface"; + power-domains = <&gcc UFS_PHY_GDSC>; operating-points-v2 = <&ice_opp_table>; From 0d5dc5818191b55e4364d04b1b898a14a2ccac38 Mon Sep 17 00:00:00 2001 From: Harshal Dev Date: Thu, 16 Apr 2026 17:29:19 +0530 Subject: [PATCH 012/329] soc: qcom: ice: Allow explicit votes on 'iface' clock for ICE Since Qualcomm inline-crypto engine (ICE) is now a dedicated driver de-coupled from the QCOM UFS driver, it explicitly votes for its required clocks during probe. For scenarios where the 'clk_ignore_unused' flag is not passed on the kernel command line, to avoid potential unclocked ICE hardware register access during probe the ICE driver should additionally vote on the 'iface' clock. Also update the suspend and resume callbacks to handle un-voting and voting on the 'iface' clock. Fixes: 2afbf43a4aec6 ("soc: qcom: Make the Qualcomm UFS/SDCC ICE a dedicated driver") Reviewed-by: Manivannan Sadhasivam Reviewed-by: Kuldeep Singh Reviewed-by: Konrad Dybcio Signed-off-by: Harshal Dev Link: https://lore.kernel.org/r/20260416-qcom_ice_power_and_clk_vote-v5-2-5ccf5d7e2846@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/soc/qcom/ice.c | 17 +++++++++++++++-- 1 file changed, 15 insertions(+), 2 deletions(-) diff --git a/drivers/soc/qcom/ice.c b/drivers/soc/qcom/ice.c index b203bc685cad..bf4ab2d9e5c0 100644 --- a/drivers/soc/qcom/ice.c +++ b/drivers/soc/qcom/ice.c @@ -108,6 +108,7 @@ struct qcom_ice { void __iomem *base; struct clk *core_clk; + struct clk *iface_clk; bool use_hwkm; bool hwkm_init_complete; u8 hwkm_version; @@ -312,8 +313,13 @@ int qcom_ice_resume(struct qcom_ice *ice) err = clk_prepare_enable(ice->core_clk); if (err) { - dev_err(dev, "failed to enable core clock (%d)\n", - err); + dev_err(dev, "Failed to enable core clock: %d\n", err); + return err; + } + + err = clk_prepare_enable(ice->iface_clk); + if (err) { + dev_err(dev, "Failed to enable iface clock: %d\n", err); return err; } qcom_ice_hwkm_init(ice); @@ -323,6 +329,7 @@ EXPORT_SYMBOL_GPL(qcom_ice_resume); int qcom_ice_suspend(struct qcom_ice *ice) { + clk_disable_unprepare(ice->iface_clk); clk_disable_unprepare(ice->core_clk); ice->hwkm_init_complete = false; @@ -579,11 +586,17 @@ static struct qcom_ice *qcom_ice_create(struct device *dev, engine->core_clk = devm_clk_get_optional_enabled(dev, "ice_core_clk"); if (!engine->core_clk) engine->core_clk = devm_clk_get_optional_enabled(dev, "ice"); + if (!engine->core_clk) + engine->core_clk = devm_clk_get_optional_enabled(dev, "core"); if (!engine->core_clk) engine->core_clk = devm_clk_get_enabled(dev, NULL); if (IS_ERR(engine->core_clk)) return ERR_CAST(engine->core_clk); + engine->iface_clk = devm_clk_get_optional_enabled(dev, "iface"); + if (IS_ERR(engine->iface_clk)) + return ERR_CAST(engine->iface_clk); + if (!qcom_ice_check_supported(engine)) return ERR_PTR(-EOPNOTSUPP); From 5a52c5701a67d5176eb1afbf1bdaf7d6dfeec597 Mon Sep 17 00:00:00 2001 From: Kamal Dasu Date: Thu, 23 Apr 2026 15:18:55 -0400 Subject: [PATCH 013/329] mmc: core: Fix host controller programming for fixed driver type When using the fixed-emmc-driver-type device tree property, the MMC core correctly selects the driver strength for the card but fails to program the host controller accordingly. This causes a mismatch where the card uses the specified driver type while the host controller defaults to Type B (since ios->drv_type remains zero). Split the driver type programming logic to handle both fixed and dynamic driver type selection paths. For fixed driver types, program the host controller with the selected drive_strength value. For dynamic selection, use the existing drv_type as before. This ensures both the eMMC device and host controller use matching driver strengths, preventing potential signal integrity issues. Fixes: 6186d06c519e ("mmc: parse new binding for eMMC fixed driver type") Signed-off-by: Kamal Dasu Reviewed-by: Shawn Lin Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/core/mmc.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c index 8846550a8892..05444ecf3909 100644 --- a/drivers/mmc/core/mmc.c +++ b/drivers/mmc/core/mmc.c @@ -1371,7 +1371,9 @@ static void mmc_select_driver_type(struct mmc_card *card) card->drive_strength = drive_strength; - if (drv_type) + if (fixed_drv_type >= 0 && drive_strength) + mmc_set_driver_type(card->host, drive_strength); + else if (drv_type) mmc_set_driver_type(card->host, drv_type); } From 12c97d1c15f926cd430bf5cdf8ffe878cb478165 Mon Sep 17 00:00:00 2001 From: Abel Vesa Date: Tue, 14 Apr 2026 20:05:51 +0300 Subject: [PATCH 014/329] arm64: dts: qcom: glymur: Drop RPMh CXO clocks from QMP PHYs On Glymur, all QMP PHYs except the one used by USB SS0 take their reference clock from the TCSR clock controller. Since these TCSR clocks already derive from RPMH_CXO_CLK as their sole parent, there is no need to provide an extra `clkref` clock to the PHY nodes. Drop the extra RPMh CXO clock inputs and use the TCSR clocks as the PHY reference clocks instead. This also fixes the devicetree schema validation, as the bindings do not allow a separate `clkref` clock. Fixes: 4eee57dd4df9 ("arm64: dts: qcom: glymur: Add USB related nodes") Reported-by: Krzysztof Kozlowski Reported-by: Rob Herring Closes: https://lore.kernel.org/r/20260410145205.GA554754-robh@kernel.org/ Signed-off-by: Abel Vesa Reviewed-by: Konrad Dybcio Reviewed-by: Dmitry Baryshkov Link: https://lore.kernel.org/r/20260414-dts-glymur-drop-rpmh-cxo-clk-from-qmpphys-v1-1-ab12d77c4aec@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- arch/arm64/boot/dts/qcom/glymur.dtsi | 20 ++++++-------------- 1 file changed, 6 insertions(+), 14 deletions(-) diff --git a/arch/arm64/boot/dts/qcom/glymur.dtsi b/arch/arm64/boot/dts/qcom/glymur.dtsi index f23cf81ddb77..82436984485d 100644 --- a/arch/arm64/boot/dts/qcom/glymur.dtsi +++ b/arch/arm64/boot/dts/qcom/glymur.dtsi @@ -2314,11 +2314,9 @@ usb_mp_qmpphy0: phy@fa3000 { clocks = <&gcc GCC_USB3_MP_PHY_AUX_CLK>, <&tcsr TCSR_USB3_0_CLKREF_EN>, - <&rpmhcc RPMH_CXO_CLK>, <&gcc GCC_USB3_MP_PHY_COM_AUX_CLK>, <&gcc GCC_USB3_MP_PHY_PIPE_0_CLK>; clock-names = "aux", - "clkref", "ref", "com_aux", "pipe"; @@ -2343,11 +2341,9 @@ usb_mp_qmpphy1: phy@fa5000 { clocks = <&gcc GCC_USB3_MP_PHY_AUX_CLK>, <&tcsr TCSR_USB3_1_CLKREF_EN>, - <&rpmhcc RPMH_CXO_CLK>, <&gcc GCC_USB3_MP_PHY_COM_AUX_CLK>, <&gcc GCC_USB3_MP_PHY_PIPE_1_CLK>; clock-names = "aux", - "clkref", "ref", "com_aux", "pipe"; @@ -2482,15 +2478,13 @@ usb_1_qmpphy: phy@fde000 { reg = <0x0 0x00fde000 0x0 0x8000>; clocks = <&gcc GCC_USB3_SEC_PHY_AUX_CLK>, - <&rpmhcc RPMH_CXO_CLK>, + <&tcsr TCSR_USB4_1_CLKREF_EN>, <&gcc GCC_USB3_SEC_PHY_COM_AUX_CLK>, - <&gcc GCC_USB3_SEC_PHY_PIPE_CLK>, - <&tcsr TCSR_USB4_1_CLKREF_EN>; + <&gcc GCC_USB3_SEC_PHY_PIPE_CLK>; clock-names = "aux", "ref", "com_aux", - "usb3_pipe", - "clkref"; + "usb3_pipe"; power-domains = <&gcc GCC_USB_1_PHY_GDSC>; @@ -3750,15 +3744,13 @@ usb_2_qmpphy: phy@88e1000 { reg = <0x0 0x088e1000 0x0 0x8000>; clocks = <&gcc GCC_USB3_TERT_PHY_AUX_CLK>, - <&rpmhcc RPMH_CXO_CLK>, + <&tcsr TCSR_USB4_2_CLKREF_EN>, <&gcc GCC_USB3_TERT_PHY_COM_AUX_CLK>, - <&gcc GCC_USB3_TERT_PHY_PIPE_CLK>, - <&tcsr TCSR_USB4_2_CLKREF_EN>; + <&gcc GCC_USB3_TERT_PHY_PIPE_CLK>; clock-names = "aux", "ref", "com_aux", - "usb3_pipe", - "clkref"; + "usb3_pipe"; power-domains = <&gcc GCC_USB_2_PHY_GDSC>; From 4b15b03166cc5d28e9912287b1f9b6607c8710ec Mon Sep 17 00:00:00 2001 From: Val Packett Date: Wed, 11 Mar 2026 21:53:37 -0300 Subject: [PATCH 015/329] arm64: dts: qcom: x1-dell-thena: remove i2c20 (battery SMBus) and reserve its pins i2c20 is used by the battmgr service on the ADSP to communicate with the SBS interface of the battery. Initializing it from Linux would break the battmgr functionality when booted in EL2. Mark those pins as reserved. Fixes: e7733b42111c ("arm64: dts: qcom: Add support for Dell Inspiron 7441 / Latitude 7455") Reviewed-by: Konrad Dybcio Reviewed-by: Abel Vesa Signed-off-by: Val Packett Link: https://lore.kernel.org/r/20260312005731.12488-2-val@packett.cool Signed-off-by: Bjorn Andersson --- arch/arm64/boot/dts/qcom/x1-dell-thena.dtsi | 7 +------ 1 file changed, 1 insertion(+), 6 deletions(-) diff --git a/arch/arm64/boot/dts/qcom/x1-dell-thena.dtsi b/arch/arm64/boot/dts/qcom/x1-dell-thena.dtsi index 0d9a324cc6cc..db291730130c 100644 --- a/arch/arm64/boot/dts/qcom/x1-dell-thena.dtsi +++ b/arch/arm64/boot/dts/qcom/x1-dell-thena.dtsi @@ -982,12 +982,6 @@ &i2c8 { status = "okay"; }; -&i2c20 { - clock-frequency = <400000>; - - status = "okay"; -}; - &lpass_tlmm { spkr_01_sd_n_active: spkr-01-sd-n-active-state { pins = "gpio12"; @@ -1308,6 +1302,7 @@ right_tweeter: speaker@0,1 { &tlmm { gpio-reserved-ranges = <44 4>, /* SPI11 (TPM) */ <76 4>, /* SPI19 (TZ Protected) */ + <80 2>, /* I2C20 (Battery SMBus) */ <238 1>; /* UFS Reset */ cam_rgb_default: cam-rgb-default-state { From c5c3ef8d49e15d2fc1cec4ad7c91d81b99977440 Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Tue, 17 Feb 2026 10:23:34 -0800 Subject: [PATCH 016/329] Drivers: hv: vmbus: Provide option to skip VMBus unload on panic Currently, VMBus code initiates a VMBus unload in the panic path so that if a kdump kernel is loaded, it can start fresh in setting up its own VMBus connection. However, a driver for the VMBus virtual frame buffer may need to flush dirty portions of the frame buffer back to the Hyper-V host so that panic information is visible in the graphics console. To support such flushing, provide exported functions for the frame buffer driver to specify that the VMBus unload should not be done by the VMBus driver, and to initiate the VMBus unload itself. Together these allow a frame buffer driver to delay the VMBus unload until after it has completed the flush. Ideally, the VMBus driver could use its own panic-path callback to do the unload after all frame buffer drivers have finished. But DRM frame buffer drivers use the kmsg dump callback, and there are no callbacks after that in the panic path. Hence this somewhat messy approach to properly sequencing the frame buffer flush and the VMBus unload. Fixes: 3671f3777758 ("drm/hyperv: Add support for drm_panic") Signed-off-by: Michael Kelley Reviewed-by: Long Li Signed-off-by: Wei Liu --- drivers/hv/channel_mgmt.c | 1 + drivers/hv/hyperv_vmbus.h | 1 - drivers/hv/vmbus_drv.c | 25 ++++++++++++++++++------- include/linux/hyperv.h | 3 +++ 4 files changed, 22 insertions(+), 8 deletions(-) diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c index 84eb0a6a0b54..89d214dda360 100644 --- a/drivers/hv/channel_mgmt.c +++ b/drivers/hv/channel_mgmt.c @@ -952,6 +952,7 @@ void vmbus_initiate_unload(bool crash) else vmbus_wait_for_unload(); } +EXPORT_SYMBOL_GPL(vmbus_initiate_unload); static void vmbus_setup_channel_state(struct vmbus_channel *channel, struct vmbus_channel_offer_channel *offer) diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h index 05a36854389a..eb8bdd8bb1f5 100644 --- a/drivers/hv/hyperv_vmbus.h +++ b/drivers/hv/hyperv_vmbus.h @@ -441,7 +441,6 @@ void hv_vss_deinit(void); int hv_vss_pre_suspend(void); int hv_vss_pre_resume(void); void hv_vss_onchannelcallback(void *context); -void vmbus_initiate_unload(bool crash); static inline void hv_poll_channel(struct vmbus_channel *channel, void (*cb)(void *)) diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index d28ff45d4cfd..c9eeb2ec365d 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -69,19 +69,29 @@ bool vmbus_is_confidential(void) } EXPORT_SYMBOL_GPL(vmbus_is_confidential); +static bool skip_vmbus_unload; + +/* + * Allow a VMBus framebuffer driver to specify that in the case of a panic, + * it will do the VMbus unload operation once it has flushed any dirty + * portions of the framebuffer to the Hyper-V host. + */ +void vmbus_set_skip_unload(bool skip) +{ + skip_vmbus_unload = skip; +} +EXPORT_SYMBOL_GPL(vmbus_set_skip_unload); + /* * The panic notifier below is responsible solely for unloading the * vmbus connection, which is necessary in a panic event. - * - * Notice an intrincate relation of this notifier with Hyper-V - * framebuffer panic notifier exists - we need vmbus connection alive - * there in order to succeed, so we need to order both with each other - * [see hvfb_on_panic()] - this is done using notifiers' priorities. */ static int hv_panic_vmbus_unload(struct notifier_block *nb, unsigned long val, void *args) { - vmbus_initiate_unload(true); + if (!skip_vmbus_unload) + vmbus_initiate_unload(true); + return NOTIFY_DONE; } static struct notifier_block hyperv_panic_vmbus_unload_block = { @@ -2897,7 +2907,8 @@ static void hv_crash_handler(struct pt_regs *regs) { int cpu; - vmbus_initiate_unload(true); + if (!skip_vmbus_unload) + vmbus_initiate_unload(true); /* * In crash handler we can't schedule synic cleanup for all CPUs, * doing the cleanup for current CPU only. This should be sufficient diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 964f1be8150c..41a3d82f0722 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -1336,6 +1336,9 @@ int vmbus_allocate_mmio(struct resource **new, struct hv_device *device_obj, bool fb_overlap_ok); void vmbus_free_mmio(resource_size_t start, resource_size_t size); +void vmbus_initiate_unload(bool crash); +void vmbus_set_skip_unload(bool skip); + /* * GUID definitions of various offer types - services offered to the guest. */ From 8b35874f56ded0cc1a90a25b87411249a86246cd Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Tue, 17 Feb 2026 10:23:35 -0800 Subject: [PATCH 017/329] drm/hyperv: During panic do VMBus unload after frame buffer is flushed In a VM, Linux panic information (reason for the panic, stack trace, etc.) may be written to a serial console and/or a virtual frame buffer for a graphics console. The latter may need to be flushed back to the host hypervisor for display. The current Hyper-V DRM driver for the frame buffer does the flushing *after* the VMBus connection has been unloaded, such that panic messages are not displayed on the graphics console. A user with a Hyper-V graphics console is left with just a hung empty screen after a panic. The enhanced control that DRM provides over the panic display in the graphics console is similarly non-functional. Commit 3671f3777758 ("drm/hyperv: Add support for drm_panic") added the Hyper-V DRM driver support to flush the virtual frame buffer. It provided necessary functionality but did not handle the sequencing problem with VMBus unload. Fix the full problem by using VMBus functions to suppress the VMBus unload that is normally done by the VMBus driver in the panic path. Then after the frame buffer has been flushed, do the VMBus unload so that a kdump kernel can start cleanly. As expected, CONFIG_DRM_PANIC must be selected for these changes to have effect. As a side benefit, the enhanced features of the DRM panic path are also functional. Fixes: 3671f3777758 ("drm/hyperv: Add support for drm_panic") Signed-off-by: Michael Kelley Reviewed-by: Jocelyn Falempe Signed-off-by: Wei Liu --- drivers/gpu/drm/hyperv/hyperv_drm_drv.c | 5 +++++ drivers/gpu/drm/hyperv/hyperv_drm_modeset.c | 15 ++++++++------- 2 files changed, 13 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_drv.c b/drivers/gpu/drm/hyperv/hyperv_drm_drv.c index 06b5d96e6eaf..b6bf6412ae34 100644 --- a/drivers/gpu/drm/hyperv/hyperv_drm_drv.c +++ b/drivers/gpu/drm/hyperv/hyperv_drm_drv.c @@ -150,6 +150,10 @@ static int hyperv_vmbus_probe(struct hv_device *hdev, goto err_free_mmio; } + /* If DRM panic path is stubbed out VMBus code must do the unload */ + if (IS_ENABLED(CONFIG_DRM_PANIC)) + vmbus_set_skip_unload(true); + drm_client_setup(dev, NULL); return 0; @@ -169,6 +173,7 @@ static void hyperv_vmbus_remove(struct hv_device *hdev) struct drm_device *dev = hv_get_drvdata(hdev); struct hyperv_drm_device *hv = to_hv(dev); + vmbus_set_skip_unload(false); drm_dev_unplug(dev); drm_atomic_helper_shutdown(dev); vmbus_close(hdev->channel); diff --git a/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c index 7978f8c8108c..d48ca6c23b7c 100644 --- a/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c +++ b/drivers/gpu/drm/hyperv/hyperv_drm_modeset.c @@ -212,15 +212,16 @@ static void hyperv_plane_panic_flush(struct drm_plane *plane) struct hyperv_drm_device *hv = to_hv(plane->dev); struct drm_rect rect; - if (!plane->state || !plane->state->fb) - return; + if (plane->state && plane->state->fb) { + rect.x1 = 0; + rect.y1 = 0; + rect.x2 = plane->state->fb->width; + rect.y2 = plane->state->fb->height; - rect.x1 = 0; - rect.y1 = 0; - rect.x2 = plane->state->fb->width; - rect.y2 = plane->state->fb->height; + hyperv_update_dirt(hv->hdev, &rect); + } - hyperv_update_dirt(hv->hdev, &rect); + vmbus_initiate_unload(true); } static const struct drm_plane_helper_funcs hyperv_plane_helper_funcs = { From f133bd4b5daf71bccdde0ad1a4f47fac76a6bfb1 Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Tue, 5 May 2026 13:12:58 +0000 Subject: [PATCH 018/329] firmware: samsung: acpm: Fix cross-thread RX length corruption Sashiko identified a cross-thread RX length corruption bug when reviewing the thermal addition to ACPM [1]. When multiple threads concurrently send IPC requests, the ACPM polling mechanism can encounter responses belonging to other threads. To drain the queue, the driver saves these concurrent responses into an internal cache (`rx_data->cmd`) to be retrieved later by the owning thread. Previously, the driver incorrectly used `xfer->rxcnt` (the expected receive length of the *current* polling thread) when copying data for *other* threads into this cache. If the threads expected responses of different lengths, this resulted in buffer underflows (leading to reads of uninitialized memory) or potential buffer overflows. Fix this by replacing the boolean `response` flag in `struct acpm_rx_data` with `rxcnt`, caching the exact expected receive length for each specific transaction during transfer preparation. Use this cached length when saving concurrent responses. Consequently, ensure that `xfer->rxcnt` is explicitly zeroed in driver helpers (e.g., `acpm_dvfs_set_xfer`) for fire-and-forget messages to prevent uninitialized stack garbage from being interpreted as a massive expected receive length. Cc: stable@vger.kernel.org Fixes: a88927b534ba ("firmware: add Exynos ACPM protocol driver") Closes: https://sashiko.dev/#/patchset/20260420-acpm-tmu-v3-0-3dc8e93f0b26%40linaro.org [1] Reported-by: Titouan Ameline de Cadeville Closes: https://lore.kernel.org/r/20260426210255.73674-1-titouan.ameline@gmail.com/ Signed-off-by: Tudor Ambarus Link: https://patch.msgid.link/20260505-acpm-fixes-sashiko-reports-v5-1-43b5ee7f1674@linaro.org Signed-off-by: Krzysztof Kozlowski --- drivers/firmware/samsung/exynos-acpm-dvfs.c | 3 +++ drivers/firmware/samsung/exynos-acpm.c | 15 ++++++++------- 2 files changed, 11 insertions(+), 7 deletions(-) diff --git a/drivers/firmware/samsung/exynos-acpm-dvfs.c b/drivers/firmware/samsung/exynos-acpm-dvfs.c index 06bdf62dea1f..fdea7aa24ca0 100644 --- a/drivers/firmware/samsung/exynos-acpm-dvfs.c +++ b/drivers/firmware/samsung/exynos-acpm-dvfs.c @@ -31,6 +31,9 @@ static void acpm_dvfs_set_xfer(struct acpm_xfer *xfer, u32 *cmd, size_t cmdlen, if (response) { xfer->rxcnt = cmdlen; xfer->rxd = cmd; + } else { + xfer->rxcnt = 0; + xfer->rxd = NULL; } } diff --git a/drivers/firmware/samsung/exynos-acpm.c b/drivers/firmware/samsung/exynos-acpm.c index 16c46ed60837..e95edc350efa 100644 --- a/drivers/firmware/samsung/exynos-acpm.c +++ b/drivers/firmware/samsung/exynos-acpm.c @@ -104,12 +104,12 @@ struct acpm_queue { * * @cmd: pointer to where the data shall be saved. * @n_cmd: number of 32-bit commands. - * @response: true if the client expects the RX data. + * @rxcnt: expected length of the response in 32-bit words. */ struct acpm_rx_data { u32 *cmd; size_t n_cmd; - bool response; + size_t rxcnt; }; #define ACPM_SEQNUM_MAX 64 @@ -199,7 +199,7 @@ static void acpm_get_saved_rx(struct acpm_chan *achan, const struct acpm_rx_data *rx_data = &achan->rx_data[tx_seqnum - 1]; u32 rx_seqnum; - if (!rx_data->response) + if (!rx_data->rxcnt) return; rx_seqnum = FIELD_GET(ACPM_PROTOCOL_SEQNUM, rx_data->cmd[0]); @@ -256,7 +256,7 @@ static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) seqnum = rx_seqnum - 1; rx_data = &achan->rx_data[seqnum]; - if (rx_data->response) { + if (rx_data->rxcnt) { if (rx_seqnum == tx_seqnum) { __ioread32_copy(xfer->rxd, addr, xfer->rxcnt); rx_set = true; @@ -268,7 +268,8 @@ static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) * clear yet the bitmap. It will be cleared * after the response is copied to the request. */ - __ioread32_copy(rx_data->cmd, addr, xfer->rxcnt); + __ioread32_copy(rx_data->cmd, addr, + rx_data->rxcnt); } } else { clear_bit(seqnum, achan->bitmap_seqnum); @@ -380,8 +381,8 @@ static void acpm_prepare_xfer(struct acpm_chan *achan, /* Clear data for upcoming responses */ rx_data = &achan->rx_data[achan->seqnum - 1]; memset(rx_data->cmd, 0, sizeof(*rx_data->cmd) * rx_data->n_cmd); - if (xfer->rxd) - rx_data->response = true; + /* zero means no response expected */ + rx_data->rxcnt = xfer->rxcnt; /* Flag the index based on seqnum. (seqnum: 1~63, bitmap: 0~62) */ set_bit(achan->seqnum - 1, achan->bitmap_seqnum); From b66829b17f6385cc9ffbcbe2476d532d2e3121ad Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Tue, 5 May 2026 13:12:59 +0000 Subject: [PATCH 019/329] firmware: samsung: acpm: Fix mailbox channel leak on probe error Sashiko identified the leak at [1]. The ACPM driver allocates hardware mailbox channels using `mbox_request_channel()` during `acpm_channels_init()`. However, the driver lacked a `.remove` callback and did not free these channels on subsequent error paths inside `acpm_probe()`. Additionally, if `acpm_achan_alloc_cmds()` failed during the channel initialization loop, the function returned immediately, bypassing the manual cleanup and permanently leaking any channels successfully requested in previous loop iterations. Fix this by modifying `acpm_free_mbox_chans()` to match the `devres` action signature and registering it via `devm_add_action_or_reset()`. Cc: stable@vger.kernel.org Fixes: a88927b534ba ("firmware: add Exynos ACPM protocol driver") Closes: https://sashiko.dev/#/patchset/20260420-acpm-tmu-v3-0-3dc8e93f0b26%40linaro.org [1] Signed-off-by: Tudor Ambarus Link: https://patch.msgid.link/20260505-acpm-fixes-sashiko-reports-v5-2-43b5ee7f1674@linaro.org Signed-off-by: Krzysztof Kozlowski --- drivers/firmware/samsung/exynos-acpm.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/firmware/samsung/exynos-acpm.c b/drivers/firmware/samsung/exynos-acpm.c index e95edc350efa..9766425a44ab 100644 --- a/drivers/firmware/samsung/exynos-acpm.c +++ b/drivers/firmware/samsung/exynos-acpm.c @@ -527,10 +527,11 @@ static int acpm_achan_alloc_cmds(struct acpm_chan *achan) /** * acpm_free_mbox_chans() - free mailbox channels. - * @acpm: pointer to driver data. + * @data: pointer to driver data. */ -static void acpm_free_mbox_chans(struct acpm_info *acpm) +static void acpm_free_mbox_chans(void *data) { + struct acpm_info *acpm = data; int i; for (i = 0; i < acpm->num_chans; i++) @@ -558,6 +559,10 @@ static int acpm_channels_init(struct acpm_info *acpm) if (!acpm->chans) return -ENOMEM; + ret = devm_add_action_or_reset(dev, acpm_free_mbox_chans, acpm); + if (ret) + return dev_err_probe(dev, ret, "Failed to add mbox free action.\n"); + chans_shmem = acpm->sram_base + readl(&shmem->chans); for (i = 0; i < acpm->num_chans; i++) { @@ -579,10 +584,8 @@ static int acpm_channels_init(struct acpm_info *acpm) cl->dev = dev; achan->chan = mbox_request_channel(cl, 0); - if (IS_ERR(achan->chan)) { - acpm_free_mbox_chans(acpm); + if (IS_ERR(achan->chan)) return PTR_ERR(achan->chan); - } } return 0; From 84a0f7caafc679f763d3868635837e22bb89651a Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Mon, 11 May 2026 17:02:44 +0900 Subject: [PATCH 020/329] ARM: Do not select HAVE_RUST when KASAN is enabled When KASAN is enabled, such as with allmodconfig, the build fails when building the Rust code with: error: kernel-address sanitizer is not supported for this target error: aborting due to 1 previous error make[4]: *** [rust/Makefile:654: rust/core.o] Error 1 The arm-unknown-linux-gnueabi target does not support KASAN, so avoid saying Rust is supported when it is enabled. Cc: stable@vger.kernel.org Fixes: ccb8ce526807 ("ARM: 9441/1: rust: Enable Rust support for ARMv7") Link: https://github.com/Rust-for-Linux/linux/issues/1234 Signed-off-by: Nathan Chancellor Reviewed-by: Christian Schrefl Link: https://patch.msgid.link/20260511-arm-avoid-rust-with-kasan-v1-1-24d55f4a900b@kernel.org Signed-off-by: Miguel Ojeda --- arch/arm/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 71fc5dd4123f..73e6647bea46 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -136,7 +136,7 @@ config ARM select MMU_GATHER_RCU_TABLE_FREE if SMP && ARM_LPAE select HAVE_REGS_AND_STACK_ACCESS_API select HAVE_RSEQ - select HAVE_RUST if CPU_LITTLE_ENDIAN && CPU_32v7 + select HAVE_RUST if CPU_LITTLE_ENDIAN && CPU_32v7 && !KASAN select HAVE_STACKPROTECTOR select HAVE_SYSCALL_TRACEPOINTS select HAVE_UID16 From 765aaba18413a66f6c8fe8416336ca9b3dd98a79 Mon Sep 17 00:00:00 2001 From: Mihai Sain Date: Mon, 9 Mar 2026 09:53:29 +0200 Subject: [PATCH 021/329] ARM: dts: microchip: sam9x7: fix GMAC clock configuration The GMAC node incorrectly listed four clocks, including a separate tx_clk and a TSU GCK clock sourced from ID 67. According to the SAM9X7 clocking scheme, the GMAC uses only three clocks: HCLK, PCLK, and the TSU GCK derived from the GMAC peripheral clock (ID 24). Remove the unused tx_clk, update the clock-names accordingly, and correct the assigned clock to use GCK 24 instead of GCK 67. This aligns the device tree with the actual hardware clock topology and prevents misconfiguration of the GMAC clock tree. [root@SAM9X75 ~]$ cat /sys/kernel/debug/clk/clk_summary | grep gmac gmac_gclk 1 1 1 266666666 0 0 50000 Y f802c000.ethernet tsu_clk f802c000.ethernet tsu_clk gmac_clk 2 2 0 266666666 0 0 50000 Y f802c000.ethernet hclk f802c000.ethernet pclk Fixes: 41af45af8bc3 ("ARM: dts: at91: sam9x7: add device tree for SoC") Signed-off-by: Mihai Sain Link: https://lore.kernel.org/r/20260309075329.1528-5-mihai.sain@microchip.com [claudiu.beznea: massaged the patch description] Signed-off-by: Claudiu Beznea --- arch/arm/boot/dts/microchip/sam9x7.dtsi | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/arm/boot/dts/microchip/sam9x7.dtsi b/arch/arm/boot/dts/microchip/sam9x7.dtsi index d242d7a934d0..c680a5033b6b 100644 --- a/arch/arm/boot/dts/microchip/sam9x7.dtsi +++ b/arch/arm/boot/dts/microchip/sam9x7.dtsi @@ -990,9 +990,9 @@ gmac: ethernet@f802c000 { <62 IRQ_TYPE_LEVEL_HIGH 3>, /* Queue 3 */ <63 IRQ_TYPE_LEVEL_HIGH 3>, /* Queue 4 */ <64 IRQ_TYPE_LEVEL_HIGH 3>; /* Queue 5 */ - clocks = <&pmc PMC_TYPE_PERIPHERAL 24>, <&pmc PMC_TYPE_PERIPHERAL 24>, <&pmc PMC_TYPE_GCK 24>, <&pmc PMC_TYPE_GCK 67>; - clock-names = "hclk", "pclk", "tx_clk", "tsu_clk"; - assigned-clocks = <&pmc PMC_TYPE_GCK 67>; + clocks = <&pmc PMC_TYPE_PERIPHERAL 24>, <&pmc PMC_TYPE_PERIPHERAL 24>, <&pmc PMC_TYPE_GCK 24>; + clock-names = "hclk", "pclk", "tsu_clk"; + assigned-clocks = <&pmc PMC_TYPE_GCK 24>; assigned-clock-rates = <266666666>; status = "disabled"; }; From a7511dcd9dd4bc55d123f9b800c8a4ed2662e5c6 Mon Sep 17 00:00:00 2001 From: Stepan Ionichev Date: Thu, 14 May 2026 22:43:42 +0500 Subject: [PATCH 022/329] auxdisplay: line-display: fix OOB read on zero-length message_store() linedisp_display() unconditionally reads msg[count - 1] before checking whether count is zero, so a write of zero bytes to the message sysfs attribute hits msg[-1]: write(fd, "", 0); -> message_store(..., buf, count=0) -> linedisp_display(linedisp, buf, count=0) -> msg[count - 1] == '\n' ; OOB read The kernfs write buffer for that store is a 1-byte allocation (kernfs_fop_write_iter() does kmalloc(len + 1) with len == 0), so msg[-1] is a 1-byte read before the slab object. On a KASAN-enabled kernel this trips an out-of-bounds report and panics; on stock kernels it silently reads adjacent slab data and, if that byte happens to be '\n', the following count-- wraps ssize_t 0 to -1 and is then passed to kmemdup_nul(). linedisp_display() is reached from the message_store() sysfs callback (drivers/auxdisplay/line-display.c message attribute, mode 0644) and from the in-tree initial-message setup with count == -1, so the OOB path is only userspace-triggerable via zero-byte writes; vfs_write() does not short-circuit on count == 0 and kernfs_fop_write_iter() dispatches the store callback regardless. Guard the trailing-newline trim with a count check. The existing if (!count) block then takes the clear-display path unchanged. Affects every auxdisplay driver that registers via linedisp_register() / linedisp_attach(): ht16k33, max6959, img-ascii-lcd, seg-led-gpio. Fixes: 7e76aece6f03 ("auxdisplay: Extract character line display core support") Signed-off-by: Stepan Ionichev Signed-off-by: Andy Shevchenko --- drivers/auxdisplay/line-display.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/auxdisplay/line-display.c b/drivers/auxdisplay/line-display.c index fb6d9294140d..915eb5cd96b2 100644 --- a/drivers/auxdisplay/line-display.c +++ b/drivers/auxdisplay/line-display.c @@ -173,7 +173,7 @@ static int linedisp_display(struct linedisp *linedisp, const char *msg, count = strlen(msg); /* if the string ends with a newline, trim it */ - if (msg[count - 1] == '\n') + if (count && msg[count - 1] == '\n') count--; if (!count) { From 4dc76c305a73bacaf330bebf723a181427bb4540 Mon Sep 17 00:00:00 2001 From: Stepan Ionichev Date: Fri, 15 May 2026 18:30:04 +0500 Subject: [PATCH 023/329] auxdisplay: Kconfig: drop unneeded quotes in PANEL_BOOT_MESSAGE dep The PANEL_BOOT_MESSAGE dependency uses a quoted-string comparison against the PANEL_CHANGE_MESSAGE bool symbol: depends on PANEL_CHANGE_MESSAGE="y" This is the only such pattern under drivers/auxdisplay/ (grep shows no other Kconfig file in the tree uses depends on FOO="y" with quotes for a plain bool symbol). The quoted form is parsed by Kconfig but is not idiomatic; the common form for the same intent is the unquoted tristate-style dependency: depends on PANEL_CHANGE_MESSAGE which evaluates true when PANEL_CHANGE_MESSAGE is y or m. Since PANEL_CHANGE_MESSAGE is declared as bool (not tristate), there is no behaviour change in practice: y is the only enabled value either form can match. Drop the quoted comparison so the dependency matches the prevailing kernel Kconfig style and so it is obvious to readers that the comparison works. Suggested-by: Andy Shevchenko Link: https://lore.kernel.org/r/CAHp75VfsA_LsbEKjxoeMdbhPbWj7OHZ7=0SYNA3c=ZLj_M94Bw@mail.gmail.com Signed-off-by: Stepan Ionichev Signed-off-by: Andy Shevchenko --- drivers/auxdisplay/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/auxdisplay/Kconfig b/drivers/auxdisplay/Kconfig index bedc6133f970..1ea7c039160c 100644 --- a/drivers/auxdisplay/Kconfig +++ b/drivers/auxdisplay/Kconfig @@ -327,7 +327,7 @@ config PANEL_CHANGE_MESSAGE say 'N' and keep the default message with the version. config PANEL_BOOT_MESSAGE - depends on PANEL_CHANGE_MESSAGE="y" + depends on PANEL_CHANGE_MESSAGE string "New initialization message" default "" help From 1711b6ed6953cee5940ca4c3a6e77f1b3798cee2 Mon Sep 17 00:00:00 2001 From: Nirmoy Das Date: Thu, 14 May 2026 07:42:57 -0700 Subject: [PATCH 024/329] ovl: keep err zero after successful ovl_cache_get() ovl_iterate_merged() stores PTR_ERR(cache) in err before checking IS_ERR(cache). On success err holds the truncated cache pointer and can be returned as a bogus non-zero error. The syzbot reproducer reaches this through overlay-on-overlay readdir: getdents64 iterate_dir(outer overlay file) ovl_iterate_merged() ovl_cache_get() ovl_dir_read_merged() ovl_dir_read() iterate_dir(inner overlay file) ovl_iterate_merged() Only compute PTR_ERR(cache) on the error path. Fixes: d25e4b739f83 ("ovl: refactor ovl_iterate() and port to cred guard") Reported-by: syzbot+a16fb0cce329a320661c@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=a16fb0cce329a320661c Cc: stable@vger.kernel.org Signed-off-by: Nirmoy Das Link: https://patch.msgid.link/20260514144258.3068715-1-nirmoyd@nvidia.com Signed-off-by: Christian Brauner --- fs/overlayfs/readdir.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/fs/overlayfs/readdir.c b/fs/overlayfs/readdir.c index 1dcc75b3a90f..e7fe29cb6028 100644 --- a/fs/overlayfs/readdir.c +++ b/fs/overlayfs/readdir.c @@ -838,15 +838,14 @@ static int ovl_iterate_merged(struct file *file, struct dir_context *ctx) struct ovl_dir_file *od = file->private_data; struct dentry *dentry = file->f_path.dentry; struct ovl_cache_entry *p; - int err = 0; + int err; if (!od->cache) { struct ovl_dir_cache *cache; cache = ovl_cache_get(dentry); - err = PTR_ERR(cache); if (IS_ERR(cache)) - return err; + return PTR_ERR(cache); od->cache = cache; ovl_seek_cursor(od, ctx->pos); @@ -869,7 +868,7 @@ static int ovl_iterate_merged(struct file *file, struct dir_context *ctx) od->cursor = p->l_node.next; ctx->pos++; } - return err; + return 0; } static bool ovl_need_adjust_d_ino(struct file *file) From d922113ef91e6e7e8065e9070f349365341ba32e Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 18 May 2026 19:22:17 +0530 Subject: [PATCH 025/329] soc: qcom: ice: Fix race between qcom_ice_probe() and of_qcom_ice_get() The current platform driver design causes probe ordering races with consumers (UFS, eMMC) due to ICE's dependency on SCM firmware calls. If ICE probe fails (missing ICE SCM or DT registers), devm_of_qcom_ice_get() loops with -EPROBE_DEFER, leaving consumers non-functional even when ICE should be gracefully disabled. devm_of_qcom_ice_get() doesn't know if the ICE driver probe has failed due to above reasons or it is waiting for the SCM driver. Moreover, there is no devlink dependency between ICE and consumer drivers as 'qcom,ice' is not considered as a DT 'supplier'. So the consumer drivers have no idea of when the ICE driver is going to probe. To address these issues, store the error pointer in a global xarray with ice node phandle as a key during probe in addition to the valid ice pointer and synchronize both qcom_ice_probe() and of_qcom_ice_get() using a mutex. If the xarray entry is NULL, then it implies that the driver is not probed yet, so return -EPROBE_DEFER. If it has any error pointer, return that error pointer directly. Otherwise, add the devlink as usual and return the valid pointer to the consumer. Xarray is used instead of platform drvdata, since driver core frees the drvdata during probe failure. So it cannot be used to pass the error pointer to the consumers. Note that this change only fixes the standalone ICE DT node bindings and not the ones with 'ice' range embedded in the consumer nodes, where there is no issue. Fixes: 2afbf43a4aec ("soc: qcom: Make the Qualcomm UFS/SDCC ICE a dedicated driver") Reported-by: Sumit Garg Tested-by: Sumit Garg # OP-TEE as TZ Acked-by: Sumit Garg Cc: stable@vger.kernel.org # 6.4 Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260518-qcom-ice-fix-v7-1-2a595382185b@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/soc/qcom/ice.c | 38 +++++++++++++++++++++++++++++++------- 1 file changed, 31 insertions(+), 7 deletions(-) diff --git a/drivers/soc/qcom/ice.c b/drivers/soc/qcom/ice.c index b203bc685cad..91991864b4a3 100644 --- a/drivers/soc/qcom/ice.c +++ b/drivers/soc/qcom/ice.c @@ -16,6 +16,7 @@ #include #include #include +#include #include @@ -113,6 +114,9 @@ struct qcom_ice { u8 hwkm_version; }; +static DEFINE_XARRAY(ice_handles); +static DEFINE_MUTEX(ice_mutex); + static bool qcom_ice_check_supported(struct qcom_ice *ice) { u32 regval = qcom_ice_readl(ice, QCOM_ICE_REG_VERSION); @@ -631,6 +635,8 @@ static struct qcom_ice *of_qcom_ice_get(struct device *dev) return qcom_ice_create(&pdev->dev, base); } + guard(mutex)(&ice_mutex); + /* * If the consumer node does not provider an 'ice' reg range * (legacy DT binding), then it must at least provide a phandle @@ -647,12 +653,13 @@ static struct qcom_ice *of_qcom_ice_get(struct device *dev) return ERR_PTR(-EPROBE_DEFER); } - ice = platform_get_drvdata(pdev); - if (!ice) { - dev_err(dev, "Cannot get ice instance from %s\n", - dev_name(&pdev->dev)); + ice = xa_load(&ice_handles, pdev->dev.of_node->phandle); + if (IS_ERR_OR_NULL(ice)) { platform_device_put(pdev); - return ERR_PTR(-EPROBE_DEFER); + if (!ice) + return ERR_PTR(-EPROBE_DEFER); + else + return ice; } link = device_link_add(dev, &pdev->dev, DL_FLAG_AUTOREMOVE_SUPPLIER); @@ -716,24 +723,40 @@ EXPORT_SYMBOL_GPL(devm_of_qcom_ice_get); static int qcom_ice_probe(struct platform_device *pdev) { + unsigned long phandle = pdev->dev.of_node->phandle; struct qcom_ice *engine; void __iomem *base; + guard(mutex)(&ice_mutex); + base = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(base)) { dev_warn(&pdev->dev, "ICE registers not found\n"); + /* Store the error pointer for devm_of_qcom_ice_get() */ + xa_store(&ice_handles, phandle, (__force void *)base, GFP_KERNEL); return PTR_ERR(base); } engine = qcom_ice_create(&pdev->dev, base); - if (IS_ERR(engine)) + if (IS_ERR(engine)) { + /* Store the error pointer for devm_of_qcom_ice_get() */ + xa_store(&ice_handles, phandle, engine, GFP_KERNEL); return PTR_ERR(engine); + } - platform_set_drvdata(pdev, engine); + xa_store(&ice_handles, phandle, engine, GFP_KERNEL); return 0; } +static void qcom_ice_remove(struct platform_device *pdev) +{ + unsigned long phandle = pdev->dev.of_node->phandle; + + guard(mutex)(&ice_mutex); + xa_store(&ice_handles, phandle, NULL, GFP_KERNEL); +} + static const struct of_device_id qcom_ice_of_match_table[] = { { .compatible = "qcom,inline-crypto-engine" }, { }, @@ -742,6 +765,7 @@ MODULE_DEVICE_TABLE(of, qcom_ice_of_match_table); static struct platform_driver qcom_ice_driver = { .probe = qcom_ice_probe, + .remove = qcom_ice_remove, .driver = { .name = "qcom-ice", .of_match_table = qcom_ice_of_match_table, From 5a4dc805a80e6fe303d6a4748cd451ea15987ffd Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 18 May 2026 19:22:18 +0530 Subject: [PATCH 026/329] soc: qcom: ice: Return -ENODEV if the ICE platform device is not found By the time the consumer driver calls devm_of_qcom_ice_get(), all the platform devices for ICE nodes would've been created by of_platform_default_populate(). So for the absence of any platform device, -ENODEV should not returned, not -EPROBE_DEFER. Fixes: 2afbf43a4aec ("soc: qcom: Make the Qualcomm UFS/SDCC ICE a dedicated driver") Tested-by: Sumit Garg # OP-TEE as TZ Acked-by: Sumit Garg Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260518-qcom-ice-fix-v7-2-2a595382185b@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/soc/qcom/ice.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soc/qcom/ice.c b/drivers/soc/qcom/ice.c index 91991864b4a3..85deb9ea4a68 100644 --- a/drivers/soc/qcom/ice.c +++ b/drivers/soc/qcom/ice.c @@ -650,7 +650,7 @@ static struct qcom_ice *of_qcom_ice_get(struct device *dev) pdev = of_find_device_by_node(node); if (!pdev) { dev_err(dev, "Cannot find device node %s\n", node->name); - return ERR_PTR(-EPROBE_DEFER); + return ERR_PTR(-ENODEV); } ice = xa_load(&ice_handles, pdev->dev.of_node->phandle); From b9ab7217dd7d567c50311afa94d6d6746cb77e04 Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 18 May 2026 19:22:19 +0530 Subject: [PATCH 027/329] soc: qcom: ice: Return proper error codes from devm_of_qcom_ice_get() instead of NULL devm_of_qcom_ice_get() currently returns NULL if ICE SCM is not available or "qcom,ice" property is not found in DT. But this confuses the clients since NULL doesn't convey the reason for failure. So return proper error codes instead of NULL. Reported-by: Sumit Garg Reviewed-by: Konrad Dybcio Tested-by: Sumit Garg # OP-TEE as TZ Acked-by: Sumit Garg Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260518-qcom-ice-fix-v7-3-2a595382185b@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/soc/qcom/ice.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/drivers/soc/qcom/ice.c b/drivers/soc/qcom/ice.c index 85deb9ea4a68..2b592aa42941 100644 --- a/drivers/soc/qcom/ice.c +++ b/drivers/soc/qcom/ice.c @@ -563,7 +563,7 @@ static struct qcom_ice *qcom_ice_create(struct device *dev, if (!qcom_scm_ice_available()) { dev_warn(dev, "ICE SCM interface not found\n"); - return NULL; + return ERR_PTR(-EOPNOTSUPP); } engine = devm_kzalloc(dev, sizeof(*engine), GFP_KERNEL); @@ -645,7 +645,7 @@ static struct qcom_ice *of_qcom_ice_get(struct device *dev) struct device_node *node __free(device_node) = of_parse_phandle(dev->of_node, "qcom,ice", 0); if (!node) - return NULL; + return ERR_PTR(-ENODEV); pdev = of_find_device_by_node(node); if (!pdev) { @@ -698,8 +698,7 @@ static void devm_of_qcom_ice_put(struct device *dev, void *res) * phandle via 'qcom,ice' property to an ICE DT, the ICE instance will already * be created and so this function will return that instead. * - * Return: ICE pointer on success, NULL if there is no ICE data provided by the - * consumer or ERR_PTR() on error. + * Return: ICE pointer on success, ERR_PTR() on error. */ struct qcom_ice *devm_of_qcom_ice_get(struct device *dev) { @@ -710,7 +709,7 @@ struct qcom_ice *devm_of_qcom_ice_get(struct device *dev) return ERR_PTR(-ENOMEM); ice = of_qcom_ice_get(dev); - if (!IS_ERR_OR_NULL(ice)) { + if (!IS_ERR(ice)) { *dr = ice; devres_add(dev, dr); } else { From 2ccbb3fa5cf47d05849cf6722aad1b4cc14df6d9 Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 18 May 2026 19:22:20 +0530 Subject: [PATCH 028/329] mmc: sdhci-msm: Remove NULL check from devm_of_qcom_ice_get() Now since the devm_of_qcom_ice_get() API never returns NULL, remove the NULL check and also simplify the error handling. Reviewed-by: Konrad Dybcio Acked-by: Ulf Hansson Acked-by: Adrian Hunter Tested-by: Sumit Garg # OP-TEE as TZ Acked-by: Sumit Garg Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260518-qcom-ice-fix-v7-4-2a595382185b@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/mmc/host/sdhci-msm.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/mmc/host/sdhci-msm.c b/drivers/mmc/host/sdhci-msm.c index 633462c0be5f..0882ce74e0c9 100644 --- a/drivers/mmc/host/sdhci-msm.c +++ b/drivers/mmc/host/sdhci-msm.c @@ -1918,13 +1918,13 @@ static int sdhci_msm_ice_init(struct sdhci_msm_host *msm_host, return 0; ice = devm_of_qcom_ice_get(dev); - if (ice == ERR_PTR(-EOPNOTSUPP)) { - dev_warn(dev, "Disabling inline encryption support\n"); - ice = NULL; - } + if (IS_ERR(ice)) { + if (ice != ERR_PTR(-EOPNOTSUPP)) + return PTR_ERR(ice); - if (IS_ERR_OR_NULL(ice)) - return PTR_ERR_OR_ZERO(ice); + dev_warn(dev, "Disabling inline encryption support\n"); + return 0; + } msm_host->ice = ice; From 4ac19b36bf4108706238cbc4f300b17dba8b881e Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Mon, 18 May 2026 19:22:21 +0530 Subject: [PATCH 029/329] scsi: ufs: ufs-qcom: Remove NULL check from devm_of_qcom_ice_get() Now since the devm_of_qcom_ice_get() API never returns NULL, remove the NULL check and also simplify the error handling. Reviewed-by: Konrad Dybcio Acked-by: Martin K. Petersen # UFS Tested-by: Sumit Garg # OP-TEE as TZ Acked-by: Sumit Garg Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260518-qcom-ice-fix-v7-5-2a595382185b@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/ufs/host/ufs-qcom.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/ufs/host/ufs-qcom.c b/drivers/ufs/host/ufs-qcom.c index bc037db46624..9c0973a7ffc3 100644 --- a/drivers/ufs/host/ufs-qcom.c +++ b/drivers/ufs/host/ufs-qcom.c @@ -177,13 +177,13 @@ static int ufs_qcom_ice_init(struct ufs_qcom_host *host) int i; ice = devm_of_qcom_ice_get(dev); - if (ice == ERR_PTR(-EOPNOTSUPP)) { - dev_warn(dev, "Disabling inline encryption support\n"); - ice = NULL; - } + if (IS_ERR(ice)) { + if (ice != ERR_PTR(-EOPNOTSUPP)) + return PTR_ERR(ice); - if (IS_ERR_OR_NULL(ice)) - return PTR_ERR_OR_ZERO(ice); + dev_warn(dev, "Disabling inline encryption support\n"); + return 0; + } host->ice = ice; From b73953af9bbd5c721c9d92b805a8aea8b0db74b1 Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Thu, 14 May 2026 12:20:17 +0530 Subject: [PATCH 030/329] arm64: defconfig: Enable PCI M.2 power sequencing driver POWER_SEQUENCING_PCIE_M2 driver handles power supply to the PCIe M.2 connectors and is required on wide variety of ARM64 platforms such as Qcom Snapdragon X Elite laptops and Mediatek Dojo Chromebooks. Reviewed-by: Dmitry Baryshkov Reviewed-by: Krzysztof Kozlowski Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260514065017.11305-1-manivannan.sadhasivam@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- arch/arm64/configs/defconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/configs/defconfig b/arch/arm64/configs/defconfig index d905a0777f93..96ce783f24e7 100644 --- a/arch/arm64/configs/defconfig +++ b/arch/arm64/configs/defconfig @@ -260,6 +260,7 @@ CONFIG_PCI_ENDPOINT=y CONFIG_PCI_ENDPOINT_CONFIGFS=y CONFIG_PCI_EPF_TEST=m CONFIG_PCI_PWRCTRL_GENERIC=m +CONFIG_POWER_SEQUENCING_PCIE_M2=m CONFIG_DEVTMPFS=y CONFIG_DEVTMPFS_MOUNT=y CONFIG_FW_LOADER_USER_HELPER=y From 62c4d31d78294bd61cf3403626b789e854357177 Mon Sep 17 00:00:00 2001 From: Christian Brauner Date: Mon, 18 May 2026 10:32:11 +0200 Subject: [PATCH 031/329] pidfd: refuse access to tasks that have started exiting harder The recent ptrace fix closed a hole where someone could rely on task->mm becoming NULL during do_exit() to bypass dumpability checks. This api here leans on on the very same check and so inherits the fix. But there is no good reason to let it succeed at all once the target has entered do_exit(). PF_EXITING is set by exit_signals() at the very top of do_exit(), before exit_mm() and exit_files() run. Once we observe it, the task is committed to dying and exit_files() will release the fdtable shortly. Fixes: 8649c322f75c ("pid: Implement pidfd_getfd syscall") Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260518-obgleich-petersilie-2d77ccccf9b9@brauner Signed-off-by: Christian Brauner (Amutable) --- kernel/pid.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/kernel/pid.c b/kernel/pid.c index fd5c2d4aa349..f55189a3d07d 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -885,10 +885,12 @@ static struct file *__pidfd_fget(struct task_struct *task, int fd) if (ret) return ERR_PTR(ret); - if (ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS)) - file = fget_task(task, fd); - else + if (!ptrace_may_access(task, PTRACE_MODE_ATTACH_REALCREDS)) file = ERR_PTR(-EPERM); + else if (task->flags & PF_EXITING) + file = ERR_PTR(-ESRCH); + else + file = fget_task(task, fd); up_read(&task->signal->exec_update_lock); From c15d7a2a11ea055bcecc0b538ae8ba79475637f9 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Thu, 4 Dec 2025 11:17:23 +0100 Subject: [PATCH 032/329] tee: fix tee_ioctl_object_invoke_arg padding The tee_ioctl_object_invoke_arg structure has padding on some architectures but not on x86-32 and a few others: include/linux/tee.h:474:32: error: padding struct to align 'params' [-Werror=padded] I expect that all current users of this are on architectures that do have implicit padding here (arm64, arm, x86, riscv), so make the padding explicit in order to avoid surprises if this later gets used elsewhere. Fixes: d5b8b0fa1775 ("tee: add TEE_IOCTL_PARAM_ATTR_TYPE_OBJREF") Signed-off-by: Arnd Bergmann Reviewed-by: Jens Wiklander Tested-by: Harshal Dev Reviewed-by: Sumit Garg Signed-off-by: Jens Wiklander --- include/uapi/linux/tee.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/uapi/linux/tee.h b/include/uapi/linux/tee.h index cab5cadca8ef..5203977ed35d 100644 --- a/include/uapi/linux/tee.h +++ b/include/uapi/linux/tee.h @@ -470,6 +470,7 @@ struct tee_ioctl_object_invoke_arg { __u32 op; __u32 ret; __u32 num_params; + __u32 :32; /* num_params tells the actual number of element in params */ struct tee_ioctl_param params[]; }; From 26682f5efc276e3ad96d102019472bfbf03833b2 Mon Sep 17 00:00:00 2001 From: Georgiy Osokin Date: Wed, 8 Apr 2026 18:52:03 +0300 Subject: [PATCH 033/329] tee: shm: fix shm leak in register_shm_helper() register_shm_helper() allocates shm before calling iov_iter_npages(). If iov_iter_npages() returns 0, the function jumps to err_ctx_put and leaks shm. This can be triggered by TEE_IOC_SHM_REGISTER with struct tee_ioctl_shm_register_data where length is 0. Jump to err_free_shm instead. Fixes: 7bdee4157591 ("tee: Use iov_iter to better support shared buffer registration") Cc: stable@vger.kernel.org Cc: lvc-project@linuxtesting.org Signed-off-by: Georgiy Osokin Reviewed-by: Sumit Garg Signed-off-by: Jens Wiklander --- drivers/tee/tee_shm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tee/tee_shm.c b/drivers/tee/tee_shm.c index e9ea9f80cfd9..6742b3579c86 100644 --- a/drivers/tee/tee_shm.c +++ b/drivers/tee/tee_shm.c @@ -435,7 +435,7 @@ register_shm_helper(struct tee_context *ctx, struct iov_iter *iter, u32 flags, num_pages = iov_iter_npages(iter, INT_MAX); if (!num_pages) { ret = ERR_PTR(-ENOMEM); - goto err_ctx_put; + goto err_free_shm; } shm->pages = kzalloc_objs(*shm->pages, num_pages); From 6fa9b543f6b4ed15ff72af266b29f316643de289 Mon Sep 17 00:00:00 2001 From: Qihang Date: Thu, 7 May 2026 23:39:17 +0800 Subject: [PATCH 034/329] tee: fix params_from_user() error path in tee_ioctl_supp_recv params_from_user() may acquire tee_shm references for MEMREF parameters before failing after partially processing the supplied parameter array. In tee_ioctl_supp_recv(), those references are currently not released on that error path. Fix this by freeing MEMREF references before returning when params_from_user() fails. Keep the final cleanup path in tee_ioctl_supp_recv() unchanged since supp_recv() may consume and replace the supplied parameters, unlike the other TEE ioctl callback paths. Signed-off-by: Qihang Signed-off-by: Jens Wiklander --- drivers/tee/tee_core.c | 56 +++++++++++++++++++----------------------- 1 file changed, 25 insertions(+), 31 deletions(-) diff --git a/drivers/tee/tee_core.c b/drivers/tee/tee_core.c index ef9642d72672..1aac50c7c1de 100644 --- a/drivers/tee/tee_core.c +++ b/drivers/tee/tee_core.c @@ -530,11 +530,24 @@ static int params_to_user(struct tee_ioctl_param __user *uparams, return 0; } +static void free_params(struct tee_param *params, size_t num_params) +{ + size_t n; + + if (!params) + return; + + for (n = 0; n < num_params; n++) + if (tee_param_is_memref(params + n) && params[n].u.memref.shm) + tee_shm_put(params[n].u.memref.shm); + + kfree(params); +} + static int tee_ioctl_open_session(struct tee_context *ctx, struct tee_ioctl_buf_data __user *ubuf) { int rc; - size_t n; struct tee_ioctl_buf_data buf; struct tee_ioctl_open_session_arg __user *uarg; struct tee_ioctl_open_session_arg arg; @@ -595,16 +608,7 @@ static int tee_ioctl_open_session(struct tee_context *ctx, */ if (rc && have_session && ctx->teedev->desc->ops->close_session) ctx->teedev->desc->ops->close_session(ctx, arg.session); - - if (params) { - /* Decrease ref count for all valid shared memory pointers */ - for (n = 0; n < arg.num_params; n++) - if (tee_param_is_memref(params + n) && - params[n].u.memref.shm) - tee_shm_put(params[n].u.memref.shm); - kfree(params); - } - + free_params(params, arg.num_params); return rc; } @@ -612,7 +616,6 @@ static int tee_ioctl_invoke(struct tee_context *ctx, struct tee_ioctl_buf_data __user *ubuf) { int rc; - size_t n; struct tee_ioctl_buf_data buf; struct tee_ioctl_invoke_arg __user *uarg; struct tee_ioctl_invoke_arg arg; @@ -657,14 +660,7 @@ static int tee_ioctl_invoke(struct tee_context *ctx, } rc = params_to_user(uparams, arg.num_params, params); out: - if (params) { - /* Decrease ref count for all valid shared memory pointers */ - for (n = 0; n < arg.num_params; n++) - if (tee_param_is_memref(params + n) && - params[n].u.memref.shm) - tee_shm_put(params[n].u.memref.shm); - kfree(params); - } + free_params(params, arg.num_params); return rc; } @@ -672,7 +668,6 @@ static int tee_ioctl_object_invoke(struct tee_context *ctx, struct tee_ioctl_buf_data __user *ubuf) { int rc; - size_t n; struct tee_ioctl_buf_data buf; struct tee_ioctl_object_invoke_arg __user *uarg; struct tee_ioctl_object_invoke_arg arg; @@ -716,14 +711,7 @@ static int tee_ioctl_object_invoke(struct tee_context *ctx, } rc = params_to_user(uparams, arg.num_params, params); out: - if (params) { - /* Decrease ref count for all valid shared memory pointers */ - for (n = 0; n < arg.num_params; n++) - if (tee_param_is_memref(params + n) && - params[n].u.memref.shm) - tee_shm_put(params[n].u.memref.shm); - kfree(params); - } + free_params(params, arg.num_params); return rc; } @@ -846,9 +834,15 @@ static int tee_ioctl_supp_recv(struct tee_context *ctx, return -ENOMEM; rc = params_from_user(ctx, params, num_params, uarg->params); - if (rc) - goto out; + if (rc) { + free_params(params, num_params); + return rc; + } + /* + * supp_recv() may consume and replace the supplied parameters, so the + * final cleanup cannot use free_params() like the other ioctl paths. + */ rc = ctx->teedev->desc->ops->supp_recv(ctx, &func, &num_params, params); if (rc) goto out; From 471c18323dfdfe7844e193b896a9267ae23a1026 Mon Sep 17 00:00:00 2001 From: Robertus Diawan Chris Date: Tue, 19 May 2026 09:05:28 +0700 Subject: [PATCH 035/329] tee: qcomtee: add missing va_end in early return qcomtee_object_user_init() qcomtee_object_user_init() is a variadic function and when the function return because there's no dispatch callback in QCOMTEE_OBJECT_TYPE_CB case, there's no va_end to cleanup "ap" object initialized by va_start and that can cause undefined behavior. So make sure to use va_end before returning the error code when there's no dispatch callback. This is reported by Coverity Scan as "Missing varargs init or cleanup". Fixes: d6e290837e50 ("tee: add Qualcomm TEE driver") Signed-off-by: Robertus Diawan Chris Reviewed-by: Amirreza Zarrabi Signed-off-by: Jens Wiklander --- drivers/tee/qcomtee/core.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/tee/qcomtee/core.c b/drivers/tee/qcomtee/core.c index b1cb50e434f0..60fe3b5776e3 100644 --- a/drivers/tee/qcomtee/core.c +++ b/drivers/tee/qcomtee/core.c @@ -306,8 +306,10 @@ int qcomtee_object_user_init(struct qcomtee_object *object, break; case QCOMTEE_OBJECT_TYPE_CB: object->ops = ops; - if (!object->ops->dispatch) - return -EINVAL; + if (!object->ops->dispatch) { + ret = -EINVAL; + break; + } /* If failed, "no-name". */ object->name = kvasprintf_const(GFP_KERNEL, fmt, ap); From 0eb307d61317b42b120ab02099b597226318358a Mon Sep 17 00:00:00 2001 From: Geliang Tang Date: Mon, 20 Apr 2026 17:45:28 +0800 Subject: [PATCH 036/329] selftests: harness: fix pidfd leak in __wait_for_test MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Fix the pidfd leak in kselftest_harness.h's __wait_for_test() where childfd = syscall(__NR_pidfd_open, t->pid, 0) is never closed. Fixes: 73a3cde97677 ("selftests: harness: Implement test timeouts through pidfd") Signed-off-by: Geliang Tang Link: https://patch.msgid.link/a82e275ccfb2609a1984d90ab559fa3af78f1e81.1776678050.git.tanggeliang@kylinos.cn Reviewed-by: Thomas Weißschuh Signed-off-by: Christian Brauner (Amutable) --- tools/testing/selftests/kselftest_harness.h | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/testing/selftests/kselftest_harness.h b/tools/testing/selftests/kselftest_harness.h index cfdce9cd252e..261e4df94d9d 100644 --- a/tools/testing/selftests/kselftest_harness.h +++ b/tools/testing/selftests/kselftest_harness.h @@ -996,6 +996,7 @@ static void __wait_for_test(struct __test_metadata *t) poll_child.fd = childfd; poll_child.events = POLLIN; ret = poll(&poll_child, 1, t->timeout * 1000); + close(childfd); if (ret == -1) { t->exit_code = KSFT_FAIL; fprintf(TH_LOG_STREAM, From b7c9047f851e80b580aba485b61785c7554b992c Mon Sep 17 00:00:00 2001 From: Harshal Dev Date: Thu, 16 Apr 2026 17:29:29 +0530 Subject: [PATCH 037/329] arm64: dts: qcom: milos: Add power-domain and iface clk for ice node Qualcomm in-line crypto engine (ICE) platform driver specifies and votes for its own resources. Before accessing ICE hardware during probe, to avoid potential unclocked register access issues (when clk_ignore_unused is not passed on the kernel command line), in addition to the 'core' clock the 'iface' clock should also be turned on by the driver. This can only be done if the UFS_PHY_GDSC power domain is enabled. Specify both the UFS_PHY_GDSC power domain and the 'iface' clock in the ICE node for milos. Fixes: 04bb37433330e ("arm64: dts: qcom: milos: Add UFS nodes") Signed-off-by: Harshal Dev Reviewed-by: Konrad Dybcio Reviewed-by: Kuldeep Singh Link: https://lore.kernel.org/r/20260416-qcom_ice_power_and_clk_vote-v5-12-5ccf5d7e2846@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- arch/arm64/boot/dts/qcom/milos.dtsi | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/qcom/milos.dtsi b/arch/arm64/boot/dts/qcom/milos.dtsi index 4a64a98a434b..a6e463f3885d 100644 --- a/arch/arm64/boot/dts/qcom/milos.dtsi +++ b/arch/arm64/boot/dts/qcom/milos.dtsi @@ -1275,7 +1275,11 @@ ice: crypto@1d88000 { "qcom,inline-crypto-engine"; reg = <0x0 0x01d88000 0x0 0x18000>; - clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>; + clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>, + <&gcc GCC_UFS_PHY_AHB_CLK>; + clock-names = "core", + "iface"; + power-domains = <&gcc UFS_PHY_GDSC>; }; tcsr_mutex: hwlock@1f40000 { From 90825ab392ac15a51f62e3f561ad77e0226a1cfc Mon Sep 17 00:00:00 2001 From: Harshal Dev Date: Thu, 16 Apr 2026 17:29:30 +0530 Subject: [PATCH 038/329] arm64: dts: qcom: eliza: Add power-domain and iface clk for ice node Qualcomm in-line crypto engine (ICE) platform driver specifies and votes for its own resources. Before accessing ICE hardware during probe, to avoid potential unclocked register access issues (when clk_ignore_unused is not passed on the kernel command line), in addition to the 'core' clock the 'iface' clock should also be turned on by the driver. This can only be done if the GCC_UFS_PHY_GDSC power domain is enabled. Specify both the GCC_UFS_PHY_GDSC power domain and the 'iface' clock in the ICE node for eliza. Fixes: af20af39fc09b ("arm64: dts: qcom: Introduce Eliza Soc base dtsi") Signed-off-by: Harshal Dev Reviewed-by: Konrad Dybcio Fixes: 54a4f0239f2e ("KVM: MMU: make kvm_mmu_zap_page() return the Reviewed-by: Kuldeep Singh Link: https://lore.kernel.org/r/20260416-qcom_ice_power_and_clk_vote-v5-13-5ccf5d7e2846@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- arch/arm64/boot/dts/qcom/eliza.dtsi | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/arch/arm64/boot/dts/qcom/eliza.dtsi b/arch/arm64/boot/dts/qcom/eliza.dtsi index 4a7a0ac40ce6..7e97361a5dc5 100644 --- a/arch/arm64/boot/dts/qcom/eliza.dtsi +++ b/arch/arm64/boot/dts/qcom/eliza.dtsi @@ -843,7 +843,11 @@ ice: crypto@1d88000 { "qcom,inline-crypto-engine"; reg = <0x0 0x01d88000 0x0 0x18000>; - clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>; + clocks = <&gcc GCC_UFS_PHY_ICE_CORE_CLK>, + <&gcc GCC_UFS_PHY_AHB_CLK>; + clock-names = "core", + "iface"; + power-domains = <&gcc GCC_UFS_PHY_GDSC>; }; tcsr_mutex: hwlock@1f40000 { From 462a85f9f887a4fef36550bb76c7f7d7a0fa296c Mon Sep 17 00:00:00 2001 From: Manivannan Sadhasivam Date: Wed, 20 May 2026 21:27:04 +0530 Subject: [PATCH 039/329] soc: qcom: ice: Fix the error code when 'qcom,ice' property is not found When both 'ice' reg entry and 'qcom,ice' property are not found in DT, then it implies that ICE is not supported. So return -EOPNOTSUPP instead of -ENODEV to client drivers to specify ICE functionality is not supported. Fixes: b9ab7217dd7d ("soc: qcom: ice: Return proper error codes from devm_of_qcom_ice_get() instead of NULL") Reported-by: Marek Szyprowski Closes: https://lore.kernel.org/linux-arm-msm/8bac0358-9da0-4cbb-98ee-333b85ba4908@samsung.com Signed-off-by: Manivannan Sadhasivam Link: https://lore.kernel.org/r/20260520155704.130803-1-manivannan.sadhasivam@oss.qualcomm.com Signed-off-by: Bjorn Andersson --- drivers/soc/qcom/ice.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soc/qcom/ice.c b/drivers/soc/qcom/ice.c index e26d5a64763c..5f20108aa03e 100644 --- a/drivers/soc/qcom/ice.c +++ b/drivers/soc/qcom/ice.c @@ -658,7 +658,7 @@ static struct qcom_ice *of_qcom_ice_get(struct device *dev) struct device_node *node __free(device_node) = of_parse_phandle(dev->of_node, "qcom,ice", 0); if (!node) - return ERR_PTR(-ENODEV); + return ERR_PTR(-EOPNOTSUPP); pdev = of_find_device_by_node(node); if (!pdev) { From 4e3d1b2c48ca6c55f1e9ca7f8dccc76f120f276c Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Tue, 19 May 2026 16:40:34 +0200 Subject: [PATCH 040/329] fuse: limit FUSE_NOTIFY_RETRIEVE to uptodate folios FUSE_NOTIFY_RETRIEVE must be limited to uptodate folios; !uptodate folios can contain uninitialized data. Since FUSE_NOTIFY_RETRIEVE is intended to only return data that is already in the page cache and not wait for data from the FUSE daemon, treat !uptodate folios as if they weren't present. This only has security impact on systems that don't enable automatic zero-initialization of all page allocations via CONFIG_INIT_ON_ALLOC_DEFAULT_ON or init_on_alloc=1. Cc: stable@kernel.org Fixes: 2d45ba381a74 ("fuse: add retrieve request") Signed-off-by: Jann Horn Link: https://patch.msgid.link/20260519-fuse-retrieve-uptodate-v1-1-a7a1912a37f9@google.com Acked-by: Miklos Szeredi Signed-off-by: Christian Brauner (Amutable) --- fs/fuse/dev.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 5dda7080f4a9..08d364ed7d6c 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1912,6 +1912,10 @@ static int fuse_retrieve(struct fuse_mount *fm, struct inode *inode, folio = filemap_get_folio(mapping, index); if (IS_ERR(folio)) break; + if (!folio_test_uptodate(folio)) { + folio_put(folio); + break; + } folio_offset = offset_in_folio(folio, pos); nr_bytes = min(folio_size(folio) - folio_offset, num); From 9c954499d43aefac01c5dfb57a82b13d2dcf4b94 Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Tue, 19 May 2026 16:29:38 +0200 Subject: [PATCH 041/329] fuse: reject fuse_notify() pagecache ops on directories The operations FUSE_NOTIFY_STORE and FUSE_NOTIFY_RETRIEVE allow the FUSE daemon to actively write/read pagecache contents. For directories with FOPEN_CACHE_DIR, the pagecache is used as kernel-internal cache storage, and userspace is not supposed to have direct access to this cache - in particular, fuse_parse_cache() will hit WARN_ON() if the cache contains bogus data. Reject FUSE_NOTIFY_STORE and FUSE_NOTIFY_RETRIEVE on anything other than regular files with -EINVAL. Fixes: 5d7bc7e8680c ("fuse: allow using readdir cache") Cc: stable@vger.kernel.org Signed-off-by: Jann Horn Link: https://patch.msgid.link/20260519-fuse-dir-pagecache-v2-1-5428fa48e175@google.com Acked-by: Miklos Szeredi Signed-off-by: Christian Brauner (Amutable) --- fs/fuse/dev.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 08d364ed7d6c..c105aaf9ff5d 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1793,6 +1793,10 @@ static int fuse_notify_store(struct fuse_conn *fc, unsigned int size, inode = fuse_ilookup(fc, nodeid, NULL); if (!inode) goto out_up_killsb; + if (!S_ISREG(inode->i_mode)) { + err = -EINVAL; + goto out_iput; + } mapping = inode->i_mapping; file_size = i_size_read(inode); @@ -1970,7 +1974,10 @@ static int fuse_notify_retrieve(struct fuse_conn *fc, unsigned int size, inode = fuse_ilookup(fc, nodeid, &fm); if (inode) { - err = fuse_retrieve(fm, inode, &outarg); + if (!S_ISREG(inode->i_mode)) + err = -EINVAL; + else + err = fuse_retrieve(fm, inode, &outarg); iput(inode); } up_read(&fc->killsb); From 90918794a4e2c3b440f8fcf3847765a8b1d81b25 Mon Sep 17 00:00:00 2001 From: Aleksandr Nogikh Date: Thu, 21 May 2026 16:22:40 +0200 Subject: [PATCH 042/329] signal: clear JOBCTL_PENDING_MASK for caller in zap_other_threads() When a multi-threaded process receives a stop signal (e.g., SIGSTOP), do_signal_stop() sets JOBCTL_STOP_PENDING and JOBCTL_STOP_CONSUME on all threads and sets signal->group_stop_count to the number of threads. If one of the threads concurrently calls execve(), de_thread() invokes zap_other_threads() to kill all other threads. zap_other_threads() aborts the pending group stop by resetting signal->group_stop_count to 0 and clears the JOBCTL_PENDING_MASK for all other threads. However, it fails to clear the job control flags for the calling thread. When execve() completes, the calling thread returns to user mode and checks for pending signals. Seeing the stale JOBCTL_STOP_PENDING flag, it calls do_signal_stop(), which invokes task_participate_group_stop(). Since JOBCTL_STOP_CONSUME is still set, it attempts to decrement the already-zero signal->group_stop_count, triggering a warning: sig->group_stop_count == 0 WARNING: CPU: 1 PID: 6475 at kernel/signal.c:373 task_participate_group_stop+0x215/0x2d0 Call Trace: do_signal_stop+0x3be/0x5c0 kernel/signal.c:2619 get_signal+0xa8c/0x1330 kernel/signal.c:2884 arch_do_signal_or_restart+0xbc/0x840 arch/x86/kernel/signal.c:337 exit_to_user_mode_loop+0x8c/0x4d0 kernel/entry/common.c:98 do_syscall_64+0x33e/0xf80 arch/x86/entry/syscall_64.c:100 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fix this race condition by clearing the JOBCTL_PENDING_MASK for the calling thread in zap_other_threads(), ensuring it does not retain any stale job control state after the thread group is destroyed. This aligns with other functions that tear down a thread group and abort group stops, such as zap_process() and complete_signal(), which correctly clear these flags for all threads including the current one. Fixes: 39efa3ef3a37 ("signal: Use GROUP_STOP_PENDING to stop once for a single group stop") Assisted-by: Gemini:gemini-3.1-pro-preview Gemini:gemini-3-flash-preview syzbot Reported-by: syzbot+b109633ea805cac54a61@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=b109633ea805cac54a61 Link: https://syzkaller.appspot.com/ai_job?id=d70208cc-862b-4fe3-bf02-3031e10cd0b3 Signed-off-by: Aleksandr Nogikh Link: https://patch.msgid.link/20260521142240.2973022-1-nogikh@google.com Signed-off-by: Christian Brauner (Amutable) --- kernel/signal.c | 1 + 1 file changed, 1 insertion(+) diff --git a/kernel/signal.c b/kernel/signal.c index 2d102e025883..9c2b32c4d755 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1338,6 +1338,7 @@ int zap_other_threads(struct task_struct *p) int count = 0; p->signal->group_stop_count = 0; + task_clear_jobctl_pending(p, JOBCTL_PENDING_MASK); for_other_threads(p, t) { task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK); From 539b7cf1f77a5135241f410777b3c780c6e23994 Mon Sep 17 00:00:00 2001 From: Niklas Schnelle Date: Thu, 21 May 2026 13:10:44 +0200 Subject: [PATCH 043/329] s390/configs: Enable IOMMUFD and VFIO cdev in defconfigs Enable IOMMUFD and VFIO cdev such that PCI pass-through to QEMU/KVM can optionally utilize native IOMMUFD. Note that because the defconfigs do not enable IOMMUFD_VFIO_CONTAINER the default PCI pass-through using VFIO with the existing container interface is not affected. Signed-off-by: Niklas Schnelle Acked-by: Heiko Carstens Acked-by: Matthew Rosato Signed-off-by: Alexander Gordeev --- arch/s390/configs/debug_defconfig | 4 ++++ arch/s390/configs/defconfig | 4 ++++ 2 files changed, 8 insertions(+) diff --git a/arch/s390/configs/debug_defconfig b/arch/s390/configs/debug_defconfig index c28f9a7d0bd8..730c90b4a876 100644 --- a/arch/s390/configs/debug_defconfig +++ b/arch/s390/configs/debug_defconfig @@ -56,6 +56,10 @@ CONFIG_EXPOLINE_AUTO=y CONFIG_CHSC_SCH=y CONFIG_VFIO_CCW=m CONFIG_VFIO_AP=m +CONFIG_VFIO_DEVICE_CDEV=y +CONFIG_IOMMUFD_DRIVER=y +CONFIG_IOMMUFD_DRIVER_CORE=y +CONFIG_IOMMUFD=y CONFIG_CMM=m CONFIG_APPLDATA_BASE=y CONFIG_S390_HYPFS_FS=y diff --git a/arch/s390/configs/defconfig b/arch/s390/configs/defconfig index d89c988f33ea..dd5fc1426c88 100644 --- a/arch/s390/configs/defconfig +++ b/arch/s390/configs/defconfig @@ -54,6 +54,10 @@ CONFIG_EXPOLINE_AUTO=y CONFIG_CHSC_SCH=y CONFIG_VFIO_CCW=m CONFIG_VFIO_AP=m +CONFIG_VFIO_DEVICE_CDEV=y +CONFIG_IOMMUFD_DRIVER=y +CONFIG_IOMMUFD_DRIVER_CORE=y +CONFIG_IOMMUFD=y CONFIG_CMM=m CONFIG_APPLDATA_BASE=y CONFIG_S390_HYPFS_FS=y From 4a44b17406cb5a93f90af3df9392b3a45eb336fb Mon Sep 17 00:00:00 2001 From: Alice Ryhl Date: Thu, 7 May 2026 11:14:42 +0000 Subject: [PATCH 044/329] rust: kasan/kbuild: fix rustc-option when cross-compiling The Makefile version of rustc-option currently checks whether the option exists for the host target instead of the target actually being compiled for. It was done this way in commit 46e24a545cdb ("rust: kasan/kbuild: fix missing flags on first build") to avoid a circular dependency on target.json. However, because of this, rustc-option currently does not function when cross-compiling from x86_64 to aarch64 if CONFIG_SHADOW_CALL_STACK is enabled. This is because KBUILD_RUSTFLAGS contains -Zfixed-x18 under this configuration. Since that flag does not exist on the host target, rustc-option runs into a compilation failure every time, leading to all flags being rejected as unsupported. To fix this, update rustc-option to pass a --target parameter so that the host target is not used. For targets using target.json, use a built-in target that is as close as possible to the target created with target.json to avoid the circular dependency on target.json. One scenario where this causes a boot failure: * Cross-compiled from x86_64 to aarch64. * With CONFIG_SHADOW_CALL_STACK=y * With CONFIG_KASAN_SW_TAGS=y * With CONFIG_KASAN_INLINE=n Then the resulting kernel image will fail to boot when it first calls into Rust code with a crash along the lines of "Unable to handle kernel paging request at virtual address 0ffffffc08541796". This is because the call threshold is not specified, so rustc will inline kasan operations, but the kasan shadow offset is not specified, which leads to the inlined kasan instructions being incorrect. Note that the -Zsanitizer=kernel-hwaddress parameter itself does not lead to a rustc-option failure despite being aarch64-specific because RUSTFLAGS_KASAN has not yet been added to KBUILD_RUSTFLAGS when rustc-option is evaluated by the kasan Makefile. Cc: stable@vger.kernel.org Fixes: 46e24a545cdb ("rust: kasan/kbuild: fix missing flags on first build") Signed-off-by: Alice Ryhl Link: https://patch.msgid.link/20260507-rustc-option-cross-v2-1-2f650a49c2b5@google.com [ Edited slightly: - Reset variable to avoid using the environment. - Use a simply expanded variable flavor for simplicity. - Export variable so that behavior in sub-`make`s is consistent. This matches other variables. - Miguel ] Signed-off-by: Miguel Ojeda --- Makefile | 3 ++- arch/x86/Makefile | 4 ++++ arch/x86/Makefile.um | 8 ++++++++ scripts/Makefile.compiler | 2 +- 4 files changed, 15 insertions(+), 2 deletions(-) diff --git a/Makefile b/Makefile index b7b80e84e1eb..1f205a004437 100644 --- a/Makefile +++ b/Makefile @@ -607,6 +607,7 @@ KBUILD_RUSTFLAGS := $(rust_common_flags) \ -Crelocation-model=static \ -Zfunction-sections=n \ -Wclippy::float_arithmetic +KBUILD_RUSTFLAGS_OPTION_CHKS := KBUILD_AFLAGS_KERNEL := KBUILD_CFLAGS_KERNEL := @@ -643,7 +644,7 @@ export KBUILD_USERCFLAGS KBUILD_USERLDFLAGS export KBUILD_CPPFLAGS NOSTDINC_FLAGS LINUXINCLUDE OBJCOPYFLAGS KBUILD_LDFLAGS export KBUILD_CFLAGS CFLAGS_KERNEL CFLAGS_MODULE -export KBUILD_RUSTFLAGS RUSTFLAGS_KERNEL RUSTFLAGS_MODULE +export KBUILD_RUSTFLAGS RUSTFLAGS_KERNEL RUSTFLAGS_MODULE KBUILD_RUSTFLAGS_OPTION_CHKS export KBUILD_AFLAGS AFLAGS_KERNEL AFLAGS_MODULE export KBUILD_AFLAGS_MODULE KBUILD_CFLAGS_MODULE KBUILD_RUSTFLAGS_MODULE KBUILD_LDFLAGS_MODULE export KBUILD_AFLAGS_KERNEL KBUILD_CFLAGS_KERNEL KBUILD_RUSTFLAGS_KERNEL diff --git a/arch/x86/Makefile b/arch/x86/Makefile index 46fec0b08487..1d526a5d2a83 100644 --- a/arch/x86/Makefile +++ b/arch/x86/Makefile @@ -77,6 +77,10 @@ KBUILD_CFLAGS += -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -mno-avx -mno-sse4a KBUILD_RUSTFLAGS += --target=$(objtree)/scripts/target.json KBUILD_RUSTFLAGS += -Ctarget-feature=-sse,-sse2,-sse3,-ssse3,-sse4.1,-sse4.2,-avx,-avx2 +# The target.json file is not available when invoking rustc-option, so use the +# built-in target when checking whether flags are supported instead. +KBUILD_RUSTFLAGS_OPTION_CHKS += --target=x86_64-unknown-none + # # CFLAGS for compiling floating point code inside the kernel. # diff --git a/arch/x86/Makefile.um b/arch/x86/Makefile.um index 19c13afa474e..9adecd65639f 100644 --- a/arch/x86/Makefile.um +++ b/arch/x86/Makefile.um @@ -14,6 +14,14 @@ endif KBUILD_RUSTFLAGS += --target=$(objtree)/scripts/target.json +# The target.json file is not available when invoking rustc-option, so use the +# built-in target when checking whether flags are supported instead. +ifeq ($(CONFIG_X86_32),y) +KBUILD_RUSTFLAGS_OPTION_CHKS += --target=i686-unknown-linux-gnu +else +KBUILD_RUSTFLAGS_OPTION_CHKS += --target=x86_64-unknown-linux-gnu +endif + ifeq ($(CONFIG_X86_32),y) START := 0x8048000 diff --git a/scripts/Makefile.compiler b/scripts/Makefile.compiler index ef91910de265..06bbe29c846c 100644 --- a/scripts/Makefile.compiler +++ b/scripts/Makefile.compiler @@ -80,7 +80,7 @@ ld-option = $(call try-run, $(LD) $(KBUILD_LDFLAGS) $(1) -v,$(1),$(2),$(3)) # TODO: remove RUSTC_BOOTSTRAP=1 when we raise the minimum GNU Make version to 4.4 __rustc-option = $(call try-run,\ echo '$(pound)![allow(missing_docs)]$(pound)![feature(no_core)]$(pound)![no_core]' | RUSTC_BOOTSTRAP=1\ - $(1) --sysroot=/dev/null $(filter-out --sysroot=/dev/null --target=%,$(2)) $(3)\ + $(1) --sysroot=/dev/null $(KBUILD_RUSTFLAGS_OPTION_CHKS) $(filter-out --sysroot=/dev/null --target=%target.json,$(2)) $(3)\ --crate-type=rlib --out-dir=$(TMPOUT) --emit=obj=- - >/dev/null,$(3),$(4)) # rustc-option From 29d87434cb91b7689de2917830ca82acfd2770f5 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Tue, 26 May 2026 12:19:37 +0200 Subject: [PATCH 045/329] regulator: mt6363: select CONFIG_IRQ_DOMAIN When build-testing this driver without CONFIG_IRQ_DOMAIN causes a compile-time error: drivers/regulator/mt6363-regulator.c: In function 'mt6363_regulator_probe': drivers/regulator/mt6363-regulator.c:884:18: error: implicit declaration of function 'irq_find_host' [-Wimplicit-function-declaration] 884 | domain = irq_find_host(interrupt_parent); | ^~~~~~~~~~~~~ drivers/regulator/mt6363-regulator.c:884:16: error: assignment to 'struct irq_domain *' from 'int' makes pointer from integer without a cast [-Wint-conversion] 884 | domain = irq_find_host(interrupt_parent); | ^ drivers/regulator/mt6363-regulator.c:896:30: error: implicit declaration of function 'irq_create_fwspec_mapping'; did you mean 'irq_create_of_mapping'? [-Wimplicit-function-declaration] 896 | info->virq = irq_create_fwspec_mapping(&fwspec); | ^~~~~~~~~~~~~~~~~~~~~~~~~ | irq_create_of_mapping This is rather hard to trigger because so many other drivers enable IRQ_DOMAIN already, but I ran into this on an s390 randconfig build. Ensure this is always enabled using a Kconfig 'select IRQ_DOMAIN' entry, as we do for all other users of this. Fixes: 3c36965df808 ("regulator: Add support for MediaTek MT6363 SPMI PMIC Regulators") Signed-off-by: Arnd Bergmann Reviewed-by: AngeloGioacchino Del Regno Link: https://patch.msgid.link/20260526102003.2527570-1-arnd@kernel.org Signed-off-by: Mark Brown --- drivers/regulator/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/regulator/Kconfig b/drivers/regulator/Kconfig index 78076ac6eac4..87554ab92801 100644 --- a/drivers/regulator/Kconfig +++ b/drivers/regulator/Kconfig @@ -977,6 +977,7 @@ config REGULATOR_MT6363 tristate "MT6363 SPMI PMIC regulator driver" depends on SPMI select REGMAP_SPMI + select IRQ_DOMAIN help Say Y here to enable support for regulators found in the MediaTek MT6363 SPMI PMIC. From 63838c323924fe4a78b2323bd45aa1030f72ca60 Mon Sep 17 00:00:00 2001 From: Yuho Choi Date: Sun, 24 May 2026 22:47:09 -0400 Subject: [PATCH 046/329] ARM: socfpga: Fix OF node refcount leak in SMP setup socfpga_smp_prepare_cpus() looks up the Cortex-A9 SCU node with of_find_compatible_node(), which returns a node reference that must be released with of_node_put(). The function maps the SCU registers and then returns without dropping that reference, leaking the node on both the success path and the of_iomap() failure path. Drop the reference once the mapping attempt is complete. The returned MMIO mapping does not depend on keeping the device node reference held. Fixes: 122694a0c712 ("ARM: socfpga: use of_iomap to map the SCU") Cc: stable@vger.kernel.org Signed-off-by: Yuho Choi Signed-off-by: Dinh Nguyen --- arch/arm/mach-socfpga/platsmp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm/mach-socfpga/platsmp.c b/arch/arm/mach-socfpga/platsmp.c index 201191cf68f3..349e6c54518e 100644 --- a/arch/arm/mach-socfpga/platsmp.c +++ b/arch/arm/mach-socfpga/platsmp.c @@ -78,6 +78,7 @@ static void __init socfpga_smp_prepare_cpus(unsigned int max_cpus) } socfpga_scu_base_addr = of_iomap(np, 0); + of_node_put(np); if (!socfpga_scu_base_addr) return; scu_enable(socfpga_scu_base_addr); From 3cd07ee35a66038fd1a643632bfc057645e07c9a Mon Sep 17 00:00:00 2001 From: Zhan Xusheng Date: Tue, 26 May 2026 10:21:31 +0800 Subject: [PATCH 047/329] cpufreq/amd-pstate: drop stale @epp_cached kdoc Commit 4e16c1175238 ("cpufreq/amd-pstate: Stop caching EPP") removed the epp_cached field from struct amd_cpudata in favour of always reading from cppc_req_cached, but the kdoc above the struct still documents @epp_cached. Drop the now-stale @epp_cached entry. Reviewed-by: Mario Limonciello (AMD) Fixes: 4e16c1175238 ("cpufreq/amd-pstate: Stop caching EPP") Signed-off-by: Zhan Xusheng Link: https://lore.kernel.org/r/20260526022131.1302373-1-zhanxusheng@xiaomi.com Signed-off-by: Mario Limonciello (AMD) --- drivers/cpufreq/amd-pstate.h | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/cpufreq/amd-pstate.h b/drivers/cpufreq/amd-pstate.h index e4722e54387b..23e8baa05849 100644 --- a/drivers/cpufreq/amd-pstate.h +++ b/drivers/cpufreq/amd-pstate.h @@ -84,7 +84,6 @@ struct amd_aperf_mperf { * @hw_prefcore: check whether HW supports preferred core featue. * Only when hw_prefcore and early prefcore param are true, * AMD P-State driver supports preferred core featue. - * @epp_cached: Cached CPPC energy-performance preference value * @policy: Cpufreq policy value * @suspended: If CPU core if offlined * @epp_default_ac: Default EPP value for AC power source From 880fa3a1e5c493d0deafe9153f8c2bed427b9428 Mon Sep 17 00:00:00 2001 From: Shivam Kalra Date: Sat, 23 May 2026 00:24:32 +0530 Subject: [PATCH 048/329] rust: helpers: add is_vmalloc_addr wrapper for NOMMU builds Commit 47ac2a4b5cd8 ("rust: kvec: implement shrink_to for KVVec") introduced a call to bindings::is_vmalloc_addr(). However, this fails to compile on architectures where CONFIG_MMU is disabled, resulting in the following build error: error[E0425]: cannot find function `is_vmalloc_addr` in crate `bindings` --> rust/kernel/alloc/kvec.rs:781:32 | 781 | if !unsafe { bindings::is_vmalloc_addr(self.ptr.as_ptr().cast()) } { | ^^^^^^^^^^^^^^^ not found in `bindings` When CONFIG_MMU is not set, is_vmalloc_addr() is defined as a static inline function in that unconditionally returns false. Because bindgen skips static inline functions when generating bindings, the symbol is completely missing from the Rust bindings crate. Fix this by providing a C helper wrapper, rust_helper_is_vmalloc_addr(), in rust/helpers/vmalloc.c. This ensures the function is reliably exposed to Rust regardless of the MMU configuration. On NOMMU builds, this allows KVVec::shrink_to() to successfully compile and correctly route all allocations through the kmalloc realloc path. Fixes: 47ac2a4b5cd8 ("rust: kvec: implement shrink_to for KVVec") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202605220811.LRplxeBR-lkp@intel.com/ Signed-off-by: Shivam Kalra Reviewed-by: Danilo Krummrich Reviewed-by: Alice Ryhl Link: https://patch.msgid.link/20260523-is-vmalloc-addr-build-fix-v1-1-73c919440c41@zohomail.in [ Pasted exact compiler output and expanded it. - Miguel ] Signed-off-by: Miguel Ojeda --- rust/helpers/vmalloc.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/rust/helpers/vmalloc.c b/rust/helpers/vmalloc.c index 326b030487a2..6aed13292313 100644 --- a/rust/helpers/vmalloc.c +++ b/rust/helpers/vmalloc.c @@ -1,5 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 +#include #include __rust_helper void *__must_check __realloc_size(2) @@ -8,3 +9,8 @@ rust_helper_vrealloc_node_align(const void *p, size_t size, unsigned long align, { return vrealloc_node_align(p, size, align, flags, node); } + +__rust_helper bool rust_helper_is_vmalloc_addr(const void *x) +{ + return is_vmalloc_addr(x); +} From 0a68853de27b522bca2b9934127277185374a24f Mon Sep 17 00:00:00 2001 From: Sun Shaojie Date: Wed, 27 May 2026 14:43:28 +0800 Subject: [PATCH 049/329] cgroup/cpuset: Use effective_xcpus in partcmd_update add/del mask calculation When sibling CPU exclusion occurs, a partition's user_xcpus may contain CPUs that were never actually granted to it. These CPUs are present in user_xcpus(cs) but not in cs->effective_xcpus. The partcmd_update path in update_parent_effective_cpumask() uses user_xcpus(cs) (via the local variable xcpus) to compute the addmask (CPUs to return to parent) and delmask (CPUs to request from parent). This is incorrect: 1) When newmask removes a CPU that was previously excluded by a sibling, addmask incorrectly includes that CPU and tries to return it to the parent even though the partition never actually owned it, causing CPU overlap with sibling partitions and triggering warnings in generate_sched_domains(). 2) When newmask adds a previously excluded CPU that is now available, delmask fails to request it from the parent because user_xcpus(cs) already includes it. Fix this by using cs->effective_xcpus instead of user_xcpus(cs) in all partcmd_update paths that calculate addmask or delmask, including the PERR_NOCPUS error handling paths. Reproducers: Example 1 - Removing a sibling-excluded CPU incorrectly returns it: # cd /sys/fs/cgroup # echo "0-1" > a1/cpuset.cpus # echo "root" > a1/cpuset.cpus.partition # echo "0-2" > b1/cpuset.cpus # echo "root" > b1/cpuset.cpus.partition # echo "2" > b1/cpuset.cpus # cat cpuset.cpus.effective # Actual: 0-1,3 Expected: 3 Example 2 - Expanding to a previously excluded CPU fails to request it: # cd /sys/fs/cgroup # echo "0-1" > a1/cpuset.cpus # echo "root" > a1/cpuset.cpus.partition # echo "0-2" > b1/cpuset.cpus # echo "root" > b1/cpuset.cpus.partition # echo "member" > a1/cpuset.cpus.partition # echo "1-2" > b1/cpuset.cpus # cat cpuset.cpus.effective # Actual: 0-1,3 Expected: 0,3 Fixes: 2a3602030d80 ("cgroup/cpuset: Don't invalidate sibling partitions on cpuset.cpus conflict") Cc: stable@vger.kernel.org # v7.0+ Suggested-by: Zhang Guopeng Signed-off-by: Sun Shaojie Reviewed-by: Waiman Long Signed-off-by: Tejun Heo --- kernel/cgroup/cpuset.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 5c33ab20cc20..c9e14fda3d6f 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -1811,9 +1811,9 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd, * Compute add/delete mask to/from effective_cpus * * For valid partition: - * addmask = exclusive_cpus & ~newmask + * addmask = effective_xcpus & ~newmask * & parent->effective_xcpus - * delmask = newmask & ~exclusive_cpus + * delmask = newmask & ~effective_xcpus * & parent->effective_xcpus * * For invalid partition: @@ -1825,11 +1825,11 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd, deleting = cpumask_and(tmp->delmask, newmask, parent->effective_xcpus); } else { - cpumask_andnot(tmp->addmask, xcpus, newmask); + cpumask_andnot(tmp->addmask, cs->effective_xcpus, newmask); adding = cpumask_and(tmp->addmask, tmp->addmask, parent->effective_xcpus); - cpumask_andnot(tmp->delmask, newmask, xcpus); + cpumask_andnot(tmp->delmask, newmask, cs->effective_xcpus); deleting = cpumask_and(tmp->delmask, tmp->delmask, parent->effective_xcpus); } @@ -1868,7 +1868,7 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd, part_error = PERR_NOCPUS; deleting = false; adding = cpumask_and(tmp->addmask, - xcpus, parent->effective_xcpus); + cs->effective_xcpus, parent->effective_xcpus); } } else { /* @@ -1890,7 +1890,8 @@ static int update_parent_effective_cpumask(struct cpuset *cs, int cmd, part_error = PERR_NOCPUS; if (is_partition_valid(cs)) adding = cpumask_and(tmp->addmask, - xcpus, parent->effective_xcpus); + cs->effective_xcpus, + parent->effective_xcpus); } else if (is_partition_invalid(cs) && !cpumask_empty(xcpus) && cpumask_subset(xcpus, parent->effective_xcpus)) { struct cgroup_subsys_state *css; From 645c3b7ef1a7eed9627664bd11d7a8eb4519ee15 Mon Sep 17 00:00:00 2001 From: Sun Shaojie Date: Wed, 27 May 2026 15:05:09 +0800 Subject: [PATCH 050/329] cgroup/cpuset: Add test cases for sibling CPU exclusion on partition update When sibling CPU exclusion occurs, a partition's effective_xcpus may be a subset of its user_xcpus. The partcmd_update path must use effective_xcpus instead of user_xcpus when calculating CPUs to return to or request from the parent. Add two test cases to verify this behavior: 1) Narrowing cpuset.cpus to only the sibling-excluded CPUs should not return CPUs to parent that the partition never actually owned. 2) Expanding cpuset.cpus after a sibling becomes a member should correctly request the additional CPUs from parent. Co-developed-by: Zhang Guopeng Signed-off-by: Zhang Guopeng Signed-off-by: Sun Shaojie Reviewed-by: Waiman Long Signed-off-by: Tejun Heo --- tools/testing/selftests/cgroup/test_cpuset_prs.sh | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/tools/testing/selftests/cgroup/test_cpuset_prs.sh b/tools/testing/selftests/cgroup/test_cpuset_prs.sh index a56f4153c64d..683b05062810 100755 --- a/tools/testing/selftests/cgroup/test_cpuset_prs.sh +++ b/tools/testing/selftests/cgroup/test_cpuset_prs.sh @@ -492,6 +492,16 @@ REMOTE_TEST_MATRIX=( " C1-5:P1 . C1-4:P1 C2-3 . . \ . . . P1 . . p1:5|c11:1-4|c12:5 \ p1:P1|c11:P1|c12:P-1" + # Narrowing cpuset.cpus to previously sibling-excluded CPUs should + # not return CPUs that were never actually owned. + " C1-4:P1 . C1-2:P1 C1-3:P2 . . \ + . . . C3 . . p1:4|c11:1-2|c12:3 \ + p1:P1|c11:P1|c12:P2 3" + # Expanding cpuset.cpus to include a previously sibling-excluded CPU + # after the sibling has become a member should correctly request it. + " C1-4:P1 . C1-2:P1 C1-3:P2 . . \ + . . P0 C2-3 . . p1:1,4|c11:1|c12:2-3 \ + p1:P1|c11:P0|c12:P2 2-3" ) # From b565a73baec275a3f4b49c1300ab396daf4a748e Mon Sep 17 00:00:00 2001 From: Zicheng Qu Date: Wed, 27 May 2026 17:38:50 +0800 Subject: [PATCH 051/329] tools/sched_ext: Fix scx_show_state per-scheduler state reads scx_show_state.py still reads scx_aborting and scx_bypass_depth as global symbols. Those symbols no longer exist after the state was moved into struct scx_sched, so the drgn script fails when it reaches either field. Read aborting and bypass_depth from scx_root instead. This preserves the script's current root-scheduler view: with sub-scheduler support, the reported values are for the root scheduler and sub-schedulers are not enumerated. Fixes: 5c8d98a1b4de ("sched_ext: Move bypass state into scx_sched") Fixes: c1743da43cf5 ("sched_ext: Move aborting flag to per-scheduler field") Signed-off-by: Zicheng Qu Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo --- tools/sched_ext/scx_show_state.py | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-) diff --git a/tools/sched_ext/scx_show_state.py b/tools/sched_ext/scx_show_state.py index 02e43c184d43..446d82807f90 100644 --- a/tools/sched_ext/scx_show_state.py +++ b/tools/sched_ext/scx_show_state.py @@ -27,18 +27,25 @@ def read_static_key(name): def state_str(state): return prog['scx_enable_state_str'][state].string_().decode() +def read_root_ops_name(): + if root: + return root.ops.name.string_().decode() + return '' + +def read_root_field(name, default): + if root: + return getattr(root, name).value_() + return default + root = prog['scx_root'] enable_state = read_atomic("scx_enable_state_var") -if root: - print(f'ops : {root.ops.name.string_().decode()}') -else: - print('ops : ') +print(f'ops : {read_root_ops_name()}') print(f'enabled : {read_static_key("__scx_enabled")}') print(f'switching_all : {read_int("scx_switching_all")}') print(f'switched_all : {read_static_key("__scx_switched_all")}') print(f'enable_state : {state_str(enable_state)} ({enable_state})') -print(f'aborting : {prog["scx_aborting"].value_()}') -print(f'bypass_depth : {prog["scx_bypass_depth"].value_()}') +print(f'aborting : {read_root_field("aborting", False)}') +print(f'bypass_depth : {read_root_field("bypass_depth", 0)}') print(f'nr_rejected : {read_atomic("scx_nr_rejected")}') print(f'enable_seq : {read_atomic("scx_enable_seq")}') From 83eb00f31eb1b10735d48e469df72cc2b0e06f6d Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Wed, 27 May 2026 12:21:01 -0700 Subject: [PATCH 052/329] hyperv: Clean up and fix the guest ID comment in hvgdk.h Change the "64 bit" to "64-bit", and the "Os" to "OS". Remove the obsolete paragraph since the guideline has been published in the Hypervisor Top Level Functional Specification for many years. The "OS Type" is 0x1 for Linux, not 0x100. No functional change. Fixes: 83ba0c4f3f31 ("Drivers: hv: Cleanup the guest ID computation") Signed-off-by: Dexuan Cui Reviewed-by: Hamza Mahfooz Signed-off-by: Wei Liu --- include/hyperv/hvgdk.h | 10 ++-------- 1 file changed, 2 insertions(+), 8 deletions(-) diff --git a/include/hyperv/hvgdk.h b/include/hyperv/hvgdk.h index 384c3f3ff4a5..f538144280ca 100644 --- a/include/hyperv/hvgdk.h +++ b/include/hyperv/hvgdk.h @@ -10,18 +10,12 @@ /* * The guest OS needs to register the guest ID with the hypervisor. - * The guest ID is a 64 bit entity and the structure of this ID is + * The guest ID is a 64-bit entity and the structure of this ID is * specified in the Hyper-V TLFS specification. * - * While the current guideline does not specify how Linux guest ID(s) - * need to be generated, our plan is to publish the guidelines for - * Linux and other guest operating systems that currently are hosted - * on Hyper-V. The implementation here conforms to this yet - * unpublished guidelines. - * * Bit(s) * 63 - Indicates if the OS is Open Source or not; 1 is Open Source - * 62:56 - Os Type; Linux is 0x100 + * 62:56 - OS Type; Linux is 0x1 * 55:48 - Distro specific identification * 47:16 - Linux kernel version number * 15:0 - Distro specific identification From 840b740a35bf969734e0f2e44c21289fdd03079e Mon Sep 17 00:00:00 2001 From: Michael Kelley Date: Tue, 26 May 2026 07:13:04 -0700 Subject: [PATCH 053/329] mshv: Add conditional VMBus dependency When the VMBus driver is not part of the kernel (CONFIG_HYPERV_VMBUS=n), the MSHV root driver fails to link: ERROR: modpost: "hv_vmbus_exists" [drivers/hv/mshv_root.ko] undefined! Fix this while meeting these requirements: * It must be possible to include the MSHV root driver without the VMBus driver. In such case, the MSHV root driver can be built-in to the kernel image, or it can be built as a separate module. * If both the MSHV root driver and the VMBus driver are present, the MSHV root driver and VMBus driver can both be built-in, or they can both be separate modules. Or the MSHV root driver can be a module while the VMBus driver can be built-in, but the reverse is disallowed. Regardless of the build choices, the VMBus driver must be loaded before the MSHV driver in order for the SynIC to be managed properly (see comments in the MSHV SynIC code). The fix has two parts: * Add a Kconfig entry for MSHV_ROOT to depend on HYPERV_VMBUS if HYPERV_VMBUS is present. The entry disallows MSHV_ROOT being built-in when HYPERV_VMBUS is a module, but without requiring that HYPERV_VMBUS be built. * Add a stub implementation of hv_vmbus_exists() for when the VMBus driver is not present so that the MSHV root driver has no module dependency on VMBus. When the VMBus driver *is* present, the module dependency ensures that the VMBus driver loads first when both are built as modules. Existing code ensures that the VMBus driver loads first if it is built-in. The VMBus driver uses subsys_initcall(), which is initcall level 4. The MSHV root driver uses module_init(), which becomes device_init() when built-in, and device_init() is initcall level 6. Reported-by: Arnd Bergmann Closes: https://lore.kernel.org/all/20260520074044.923728-1-arnd@kernel.org/ Signed-off-by: Michael Kelley Acked-by: Arnd Bergmann Reviewed-by: Jork Loeser Reviewed-by: Hardik Garg Signed-off-by: Wei Liu --- drivers/hv/Kconfig | 1 + include/linux/hyperv.h | 4 ++++ 2 files changed, 5 insertions(+) diff --git a/drivers/hv/Kconfig b/drivers/hv/Kconfig index 2d0b3fcb0ff8..aa11bcefddf2 100644 --- a/drivers/hv/Kconfig +++ b/drivers/hv/Kconfig @@ -74,6 +74,7 @@ config MSHV_ROOT # e.g. When withdrawing memory, the hypervisor gives back 4k pages in # no particular order, making it impossible to reassemble larger pages depends on PAGE_SIZE_4KB + depends on HYPERV_VMBUS if HYPERV_VMBUS select EVENTFD select VIRT_XFER_TO_GUEST_WORK select HMM_MIRROR diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h index 41a3d82f0722..734b7ef98f4d 100644 --- a/include/linux/hyperv.h +++ b/include/linux/hyperv.h @@ -1304,7 +1304,11 @@ static inline void *hv_get_drvdata(struct hv_device *dev) struct device *hv_get_vmbus_root_device(void); +#if IS_ENABLED(CONFIG_HYPERV_VMBUS) bool hv_vmbus_exists(void); +#else +static inline bool hv_vmbus_exists(void) { return false; } +#endif struct hv_ring_buffer_debug_info { u32 current_interrupt_mask; From c53763aa2b9c96fbabee68ebe1e13074cb09bfb2 Mon Sep 17 00:00:00 2001 From: Can Peng Date: Wed, 20 May 2026 15:16:32 +0800 Subject: [PATCH 054/329] mshv: use kmalloc_array in mshv_root_scheduler_init Replace kmalloc() with kmalloc_array() to prevent potential overflow, as recommended in Documentation/process/deprecated.rst. No functional change. Signed-off-by: Can Peng 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 bd1359eb58dd..146726cc4e9b 100644 --- a/drivers/hv/mshv_root_main.c +++ b/drivers/hv/mshv_root_main.c @@ -2241,7 +2241,7 @@ static int mshv_root_scheduler_init(unsigned int cpu) outputarg = (void **)this_cpu_ptr(root_scheduler_output); /* Allocate two consecutive pages. One for input, one for output. */ - p = kmalloc(2 * HV_HYP_PAGE_SIZE, GFP_KERNEL); + p = kmalloc_array(2, HV_HYP_PAGE_SIZE, GFP_KERNEL); if (!p) return -ENOMEM; From 016a25e4b0df4d77e7c258edee4aaf982e4ee809 Mon Sep 17 00:00:00 2001 From: Dexuan Cui Date: Thu, 7 May 2026 14:28:38 -0700 Subject: [PATCH 055/329] Drivers: hv: vmbus: Improve the logic of reserving fb_mmio on Gen2 VMs If vmbus_reserve_fb() in the kdump/kexec kernel fails to properly reserve the framebuffer MMIO range (which is below 4GB) due to a Gen2 VM's screen.lfb_base being zero [1], there is an MMIO conflict between the drivers hyperv-drm and pci-hyperv: when the driver pci-hyperv's hv_allocate_config_window() calls vmbus_allocate_mmio() to get an MMIO range, typically it gets a 32-bit MMIO range that overlaps with the framebuffer MMIO range, and later hv_pci_enter_d0() fails with an error message "PCI Pass-through VSP failed D0 Entry with status" since the host thinks that PCI devices must not use MMIO space that the host has assigned to the framebuffer. This is especially an issue if pci-hyperv is built-in and hyperv-drm is built as a module. Consequently, the kdump/kexec kernel fails to detect PCI devices via pci-hyperv, and may fail to mount the root file system, which may reside in a NVMe disk. The issue described here has existed for SR-IOV VF NICs since day one of the pci-hyperv driver, and has been worked around on x64 when possible. With the recent introduction of ARM64 VMs that boot from NVMe, there is no workaround, so we need a formal fix. On Gen2 VMs, if the screen.lfb_base is 0 in the kdump/kexec kernel [1], fall back to the low MMIO base, which should be equal to the framebuffer MMIO base [2] (the statement is true according to my testing on x64 Windows Server 2016, and on x64 and ARM64 Windows Server 2025 and on Azure. I checked with the Hyper-V team and they said the statement should continue to be true for Gen2 VMs). In the first kernel, screen.lfb_base is not 0; if the user specifies a very high resolution, it's not enough to only reserve 8MB: let's always reserve half of the space below 4GB, but cap the reservation to 128MB, which is the required framebuffer size of the highest resolution 7680*4320 supported by Hyper-V. While at it, fix the comparison "end > VTPM_BASE_ADDRESS" by changing the > to >=. Here the 'end' is an inclusive end (typically, it's 0xFFFF_FFFF for the low MMIO range). Note: vmbus_reserve_fb() now also reserves an MMIO range at the beginning of the low MMIO range on CVMs, which have no framebuffers (the 'screen.lfb_base' in vmbus_reserve_fb() is 0 for CVMs), just in case the host might treat the beginning of the low MMIO range specially [3]. BTW, the OpenHCL kernel is not affected by the change, because that kernel boots with DeviceTree rather than ACPI (so vmbus_reserve_fb() won't run there), and there is no framebuffer device for that kernel. Note: normally Gen1 VMs don't have the MMIO conflict issue because the framebuffer MMIO range (which is hardcoded to base=4GB-128MB and size=64MB for Gen1 VMs by the host) is always reported via the legacy PCI graphics device's BAR, so the kdump/kexec kernel can reserve the 64MB MMIO range; however, if the VM is configured to use a very high resolution and the required framebuffer size exceeds 64MB (AFAIK, in practice, this isn't a typical configuration by users), the hyperv-drm driver may need to allocate an MMIO range above 4GB and change the framebuffer MMIO location to the allocated MMIO range -- in this case, there can still be issues [4] which can't be easily fixed: any possible affected Gen1 users would have to use a resolution whose framebuffer size is <= 64MB, or switch to Gen2 VMs. [1] https://lore.kernel.org/all/SA1PR21MB692176C1BC53BFC9EAE5CF8EBF51A@SA1PR21MB6921.namprd21.prod.outlook.com/ [2] https://lore.kernel.org/all/SA1PR21MB69218F955B62DFF62E3E88D2BF222@SA1PR21MB6921.namprd21.prod.outlook.com/ [3] https://lore.kernel.org/all/SN6PR02MB415726B17D5A6027CD1717E8D4342@SN6PR02MB4157.namprd02.prod.outlook.com/ [4] https://lore.kernel.org/all/SA1PR21MB69213486F821CA5A2C793C81BF342@SA1PR21MB6921.namprd21.prod.outlook.com/ Fixes: 4daace0d8ce8 ("PCI: hv: Add paravirtual PCI front-end for Microsoft Hyper-V VMs") CC: stable@vger.kernel.org Reviewed-by: Michael Kelley Tested-by: Krister Johansen Tested-by: Matthew Ruffell Signed-off-by: Dexuan Cui Signed-off-by: Wei Liu --- drivers/hv/vmbus_drv.c | 29 ++++++++++++++++++++++++++--- 1 file changed, 26 insertions(+), 3 deletions(-) diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c index c9eeb2ec365d..b80a35c778ab 100644 --- a/drivers/hv/vmbus_drv.c +++ b/drivers/hv/vmbus_drv.c @@ -2327,8 +2327,8 @@ static acpi_status vmbus_walk_resources(struct acpi_resource *res, void *ctx) return AE_NO_MEMORY; /* If this range overlaps the virtual TPM, truncate it. */ - if (end > VTPM_BASE_ADDRESS && start < VTPM_BASE_ADDRESS) - end = VTPM_BASE_ADDRESS; + if (end >= VTPM_BASE_ADDRESS && start < VTPM_BASE_ADDRESS) + end = VTPM_BASE_ADDRESS - 1; new_res->name = "hyperv mmio"; new_res->flags = IORESOURCE_MEM; @@ -2395,6 +2395,7 @@ static void vmbus_mmio_remove(void) static void __maybe_unused vmbus_reserve_fb(void) { resource_size_t start = 0, size; + resource_size_t low_mmio_base; struct pci_dev *pdev; if (efi_enabled(EFI_BOOT)) { @@ -2402,6 +2403,24 @@ static void __maybe_unused vmbus_reserve_fb(void) if (IS_ENABLED(CONFIG_SYSFB)) { start = sysfb_primary_display.screen.lfb_base; size = max_t(__u32, sysfb_primary_display.screen.lfb_size, 0x800000); + + low_mmio_base = hyperv_mmio->start; + if (!low_mmio_base || upper_32_bits(low_mmio_base) || + (start && start < low_mmio_base)) { + pr_warn("Unexpected low mmio base %pa\n", &low_mmio_base); + } else { + /* + * If the kdump/kexec or CVM kernel's lfb_base + * is 0, fall back to the low mmio base. + */ + if (!start) + start = low_mmio_base; + /* + * Reserve half of the space below 4GB for high + * resolutions, but cap the reservation to 128MB. + */ + size = min((SZ_4G - start) / 2, SZ_128M); + } } } else { /* Gen1 VM: get FB base from PCI */ @@ -2422,8 +2441,10 @@ static void __maybe_unused vmbus_reserve_fb(void) pci_dev_put(pdev); } - if (!start) + if (!start) { + pr_warn("Unexpected framebuffer mmio base of zero\n"); return; + } /* * Make a claim for the frame buffer in the resource tree under the @@ -2433,6 +2454,8 @@ static void __maybe_unused vmbus_reserve_fb(void) */ for (; !fb_mmio && (size >= 0x100000); size >>= 1) fb_mmio = __request_region(hyperv_mmio, start, size, fb_mmio_name, 0); + + pr_info("hv_mmio=%pR,%pR fb=%pR\n", hyperv_mmio, hyperv_mmio->sibling, fb_mmio); } /** From 98e0fc32e53dd62cd38a0d67eaf5846ae20078cc Mon Sep 17 00:00:00 2001 From: "Anirudh Rayabharam (Microsoft)" Date: Wed, 13 May 2026 13:25:56 +0000 Subject: [PATCH 056/329] mshv: support 1G hugepages by passing them as 2M-aligned chunks The hypervisor's map GPA hypercall coalesces contiguous 2M-aligned chunks into 1G mappings when alignment permits, so the driver can support 1G hugepages by feeding them in as 2M chunks. Note that this is the only way to make 1G mappings; there is no way to directly map a 1G hugepage using the hypercall. Always emit a 2M (PMD_ORDER) stride for the huge-page case. The hypercall has no 1G stride, so 1G folios are processed as a sequence of 2M chunks. Folios whose order is less than PMD_ORDER (e.g. mTHP) fall back to single-page stride; mapping them as 2M would fail in the hypervisor anyway. Assisted-by: Copilot-CLI:claude-opus-4.7 Signed-off-by: Anirudh Rayabharam (Microsoft) Acked-by: Stanislav Kinsburskii Reviewed-by: Michael Kelley Signed-off-by: Wei Liu --- drivers/hv/mshv_regions.c | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) diff --git a/drivers/hv/mshv_regions.c b/drivers/hv/mshv_regions.c index fdffd4f002f6..6d65e5b42152 100644 --- a/drivers/hv/mshv_regions.c +++ b/drivers/hv/mshv_regions.c @@ -29,29 +29,27 @@ * Uses huge page stride if the backing page is huge and the guest mapping * is properly aligned; otherwise falls back to single page stride. * - * Return: Stride in pages, or -EINVAL if page order is unsupported. + * Return: Stride in pages. */ -static int mshv_chunk_stride(struct page *page, - u64 gfn, u64 page_count) +static unsigned int mshv_chunk_stride(struct page *page, u64 gfn, + u64 page_count) { - unsigned int page_order; + unsigned int page_order = folio_order(page_folio(page)); /* * Use single page stride by default. For huge page stride, the - * page must be compound and point to the head of the compound - * page, and both gfn and page_count must be huge-page aligned. + * folio order must be at least PMD_ORDER, the page's PFN must be + * 2M-aligned (so that a 2M-aligned tail page of a larger folio is + * acceptable), and both gfn and page_count must be 2M-aligned. */ - if (!PageCompound(page) || !PageHead(page) || + if (page_order < PMD_ORDER || + !IS_ALIGNED(page_to_pfn(page), PTRS_PER_PMD) || !IS_ALIGNED(gfn, PTRS_PER_PMD) || !IS_ALIGNED(page_count, PTRS_PER_PMD)) return 1; - page_order = folio_order(page_folio(page)); - /* The hypervisor only supports 2M huge page */ - if (page_order != PMD_ORDER) - return -EINVAL; - - return 1 << page_order; + /* Use 2M stride always i.e. process 1G folios as 2M chunks */ + return 1 << PMD_ORDER; } /** @@ -86,15 +84,14 @@ static long mshv_region_process_chunk(struct mshv_mem_region *region, u64 gfn = region->start_gfn + page_offset; u64 count; struct page *page; - int stride, ret; + unsigned int stride; + int ret; page = region->mreg_pages[page_offset]; if (!page) return -EINVAL; stride = mshv_chunk_stride(page, gfn, page_count); - if (stride < 0) - return stride; /* Start at stride since the first stride is validated */ for (count = stride; count < page_count; count += stride) { From 3daad7f60aa92d0307fa2b2edd38c886a09902f2 Mon Sep 17 00:00:00 2001 From: Jan Polensky Date: Thu, 21 May 2026 14:01:32 +0200 Subject: [PATCH 057/329] s390/bug: Always emit format word in __BUG_ENTRY When CONFIG_DEBUG_BUGVERBOSE is disabled, the s390 __BUG_ENTRY() macro omits the format string pointer, so the generated __bug_table entry no longer matches struct bug_entry. With HAVE_ARCH_BUG_FORMAT enabled, the generic BUG infrastructure reads bug_entry::format via bug_get_format(). If the format word is missing, subsequent fields are read from the wrong offset, which may: - Misinterpret flags (BUG vs WARN classification errors) - Fault when dereferencing a misread format pointer The root cause is that __BUG_ENTRY() delegates format word emission to __BUG_ENTRY_VERBOSE(), which is conditional on CONFIG_DEBUG_BUGVERBOSE. Fix this by moving the format field emission directly into __BUG_ENTRY() so it is always emitted unconditionally. Remove the format parameter from __BUG_ENTRY_VERBOSE() and keep only file/line emission conditional on CONFIG_DEBUG_BUGVERBOSE. Fixes: 2b71b8ab9718 ("s390/bug: Use BUG_FORMAT for DEBUG_BUGVERBOSE_DETAILED") Signed-off-by: Jan Polensky Reviewed-by: Heiko Carstens Signed-off-by: Alexander Gordeev --- arch/s390/include/asm/bug.h | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/s390/include/asm/bug.h b/arch/s390/include/asm/bug.h index 59017fd3d935..50a270edb020 100644 --- a/arch/s390/include/asm/bug.h +++ b/arch/s390/include/asm/bug.h @@ -12,12 +12,11 @@ #if defined(CONFIG_BUG) && defined(CONFIG_CC_HAS_ASM_IMMEDIATE_STRINGS) #ifdef CONFIG_DEBUG_BUGVERBOSE -#define __BUG_ENTRY_VERBOSE(format, file, line) \ - " .long " format " - . # bug_entry::format\n" \ +#define __BUG_ENTRY_VERBOSE(file, line) \ " .long " file " - . # bug_entry::file\n" \ " .short " line " # bug_entry::line\n" #else -#define __BUG_ENTRY_VERBOSE(format, file, line) +#define __BUG_ENTRY_VERBOSE(file, line) #endif #ifdef CONFIG_DEBUG_BUGVERBOSE_DETAILED @@ -28,9 +27,10 @@ #define __BUG_ENTRY(format, file, line, flags, size) \ " .section __bug_table,\"aw\"\n" \ - "1: .long 0b - . # bug_entry::bug_addr\n" \ - __BUG_ENTRY_VERBOSE(format, file, line) \ - " .short "flags" # bug_entry::flags\n" \ + "1: .long 0b - . # bug_entry::bug_addr\n"\ + " .long " format " - . # bug_entry::format\n" \ + __BUG_ENTRY_VERBOSE(file, line) \ + " .short "flags" # bug_entry::flags\n" \ " .org 1b+"size"\n" \ " .previous" From e824bbd4d224cce4b5fb59cc9dcd3447fe0b7e44 Mon Sep 17 00:00:00 2001 From: NeilBrown Date: Mon, 25 May 2026 16:23:45 +1000 Subject: [PATCH 058/329] VFS: fix possible failure to unlock in nfsd4_create_file() atomic_create() in fs/namei.c drops the reference to the dentry when it returns an error. This behaviour was imported into dentry_create() so that it will drop the reference if an error is returned from atomic_create(), though not if vfs_create() returns an error (in the case where ->atomic_create is not supported). The caller - nfsd4_create_file() - is made aware of this by checking path->dentry, which will either be a counted reference to a dentry, or an error pointer. However the change to use start_creating()/end_creating() (which landed shortly before the dentry_create() change landed, though was likely developed around the same time) means that nfsd4_create_file() *needs* a valid dentry so that it can unlock the parent. The net result is that if NFSD exports a filesystem which uses ->atomic_create, and if a call to ->atomic_create returns an error, then nfsd4_create_file() will pass an error pointer to end_creating() and the parent will not be unlocked. Fix this by changing dentry_create() to make sure path->dentry is always a valid dentry, never an error-pointer. The actual error is already returned a different way. Note that if ->atomic_create() returns a different dentry (which may not be possible in practice) we are guaranteed (because it is only ever provided by d_spliace_alias()) that it will have the same d_parent and so it will have the same effect when passed to end_creating(). Fixes: 64a989dbd144 ("VFS/knfsd: Teach dentry_create() to use atomic_open()") Signed-off-by: NeilBrown Link: https://patch.msgid.link/177969022571.3379282.16448744624428323496@noble.neil.brown.name Reviewed-by: Jeff Layton Reviewed-by: Benjamin Coddington Reviewed-by: Jori Koolstra Signed-off-by: Christian Brauner (Amutable) --- fs/namei.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/fs/namei.c b/fs/namei.c index c7fac83c9a85..4787244ca4a7 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -5024,6 +5024,7 @@ struct file *dentry_create(struct path *path, int flags, umode_t mode, { struct file *file __free(fput) = NULL; struct dentry *dentry = path->dentry; + struct dentry *orig_dentry = dentry; struct dentry *dir = dentry->d_parent; struct inode *dir_inode = d_inode(dir); struct mnt_idmap *idmap; @@ -5043,9 +5044,18 @@ struct file *dentry_create(struct path *path, int flags, umode_t mode, if (create_error) flags &= ~O_CREAT; + /* atomic_open will dput(dentry) on error */ + dget(orig_dentry); dentry = atomic_open(path, dentry, file, flags, mode); error = PTR_ERR_OR_ZERO(dentry); + if (IS_ERR(dentry)) + /* keep the original */ + dentry = orig_dentry; + else + /* Drop the extra reference */ + dput(orig_dentry); + if (unlikely(create_error) && error == -ENOENT) error = create_error; From 89c4a1167f3a0a0efd2ec3e1801036d2eb65ae1a Mon Sep 17 00:00:00 2001 From: Arpith Kalaginanavoor Date: Tue, 26 May 2026 05:38:58 -0700 Subject: [PATCH 059/329] fs/qnx6: fix pointer arithmetic in directory iteration The conversion to qnx6_get_folio() in commit b2aa61556fcf ("qnx6: Convert qnx6_get_page() to qnx6_get_folio()") introduced a regression in directory iteration. The pointer 'de' and the 'limit' address were calculated using byte offsets from a char pointer without scaling by the size of a QNX6 directory entry. This causes the driver to read from incorrect memory offsets, leading to "invalid direntry size" errors and premature termination of directory scans. Fix this by casting 'kaddr' to 'struct qnx6_dir_entry *' before applying the offset and last_entry(...) increments. This allows the compiler to correctly scale the pointer arithmetic by the 32-byte stride of the directory entry structure. Fixes: b2aa61556fcf ("qnx6: Convert qnx6_get_page() to qnx6_get_folio()") Cc: stable@vger.kernel.org Signed-off-by: Arpith Kalaginanavoor Link: https://patch.msgid.link/20260526123858.1683035-1-arpithk@nvidia.com Signed-off-by: Christian Brauner (Amutable) --- fs/qnx6/dir.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/qnx6/dir.c b/fs/qnx6/dir.c index 135fb42f6936..56bbaffe4b44 100644 --- a/fs/qnx6/dir.c +++ b/fs/qnx6/dir.c @@ -132,16 +132,16 @@ static int qnx6_readdir(struct file *file, struct dir_context *ctx) struct qnx6_dir_entry *de; struct folio *folio; char *kaddr = qnx6_get_folio(inode, n, &folio); - char *limit; + struct qnx6_dir_entry *limit; if (IS_ERR(kaddr)) { pr_err("%s(): read failed\n", __func__); ctx->pos = (n + 1) << PAGE_SHIFT; return PTR_ERR(kaddr); } - de = (struct qnx6_dir_entry *)(kaddr + offset); - limit = kaddr + last_entry(inode, n); - for (; (char *)de < limit; de++, ctx->pos += QNX6_DIR_ENTRY_SIZE) { + de = (struct qnx6_dir_entry *)kaddr + offset; + limit = (struct qnx6_dir_entry *)kaddr + last_entry(inode, n); + for (; de < limit; de++, ctx->pos += QNX6_DIR_ENTRY_SIZE) { int size = de->de_size; u32 no_inode = fs32_to_cpu(sbi, de->de_inode); From 66ac2df408ede627aaae588d4ce7e611dd25b4f9 Mon Sep 17 00:00:00 2001 From: Linus Walleij Date: Thu, 28 May 2026 10:25:26 +0200 Subject: [PATCH 060/329] ARM: dts: gemini: Fix partition offsets These FIS partition offsets were never right: the comment clearly states the FIS index is at 0xfe0000 and 0x7f * 0x200000 is 0xfe0000. Tested on the iTian SQ201. Fixes: d88b11ef91b1 ("ARM: dts: Fix up SQ201 flash access") Fixes: b5a923f8c739 ("ARM: dts: gemini: Switch to redboot partition parsing") Signed-off-by: Linus Walleij Signed-off-by: Arnd Bergmann --- arch/arm/boot/dts/gemini/gemini-sl93512r.dts | 2 +- arch/arm/boot/dts/gemini/gemini-sq201.dts | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm/boot/dts/gemini/gemini-sl93512r.dts b/arch/arm/boot/dts/gemini/gemini-sl93512r.dts index 4992ec276de9..341dec9b636a 100644 --- a/arch/arm/boot/dts/gemini/gemini-sl93512r.dts +++ b/arch/arm/boot/dts/gemini/gemini-sl93512r.dts @@ -146,7 +146,7 @@ flash@30000000 { partitions { compatible = "redboot-fis"; /* Eraseblock at 0xfe0000 */ - fis-index-block = <0x1fc>; + fis-index-block = <0x7f>; }; }; diff --git a/arch/arm/boot/dts/gemini/gemini-sq201.dts b/arch/arm/boot/dts/gemini/gemini-sq201.dts index f8c6f6e5cdea..bfd1e8581ad6 100644 --- a/arch/arm/boot/dts/gemini/gemini-sq201.dts +++ b/arch/arm/boot/dts/gemini/gemini-sq201.dts @@ -134,7 +134,7 @@ flash@30000000 { partitions { compatible = "redboot-fis"; /* Eraseblock at 0xfe0000 */ - fis-index-block = <0x1fc>; + fis-index-block = <0x7f>; }; }; From b4aea43cd37afad714b5684fe9fdfcb0e78dba26 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Wed, 13 May 2026 09:56:58 +0100 Subject: [PATCH 061/329] mm/hugetlb: avoid false positive lockdep assertion Commit 081056dc00a2 ("mm/hugetlb: unshare page tables during VMA split, not before") changed the locking model around hugetlbfs PMD unsharing on VMA split, but did not update the function which asserts the locks, hugetlb_vma_assert_locked(). This function asserts that either the hugetlb VMA lock is held (if a shared mapping) or that the reservation map lock is held (if private). If you get an unfortunate race between something which results in one of these locks being released and a hugetlb VMA split and you have CONFIG_LOCKDEP enabled, you can therefore see a false positive assertion arise when there is in fact no issue. Since this change introduced a new take_locks parameter to hugetlb_unshare_pmds(), which, when set to false, indicates that locking is sufficient, simply pass this to the unsharing logic and predicate the lock assertions on this. This is safe, as we already asserted the file rmap lock and the VMA write lock prior to this (implying exclusive mmap write lock), so we cannot be raced by either rmap or page fault page table walkers which the asserted locks are intended to protect against (we don't mind GUP-fast). Separate out huge_pmd_unshare() into __huge_pmd_unshare() to add a check_locks parameter, and update hugetlb_unshare_pmds() to pass this parameter to it. This leaves all other callers of huge_pmd_unshare() still correctly asserting the locks. The below reproducer will trigger the assert in a kernel with CONFIG_LOCKDEP enabled by racing process teardown (which will release the hugetlb lock) against a hugetlb split. void execute_one(void) { void *ptr; pid_t pid; /* * Create a hugetlb mapping spanning a PUD entry. * * We force the hugetlb page allocation with populate and * noreserve. * * |---------------------| * | | * |---------------------| * 0 PUD boundary */ ptr = mmap(0, PUD_SIZE, PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED | MAP_ANON | MAP_NORESERVE | MAP_HUGETLB | MAP_POPULATE, -1, 0); if (ptr == MAP_FAILED) { perror("mmap"); exit(EXIT_FAILURE); } /* * Fork but with a bogus stack pointer so we try to execute code in * a non-VM_EXEC VMA, causing segfault + teardown via exit_mmap(). * * The clone will cause PMD page table sharing between the * processes first via: * copy_process() -> ... -> huge_pte_alloc() -> huge_pmd_share() * * Then tear down and release the hugetlb 'VMA' lock via: * exit_mmap() -> ... -> vma_close() -> hugetlb_vma_lock_free() */ pid = syscall(__NR_clone, 0, 2 * PMD_SIZE, 0, 0, 0); if (pid < 0) { perror("clone"); exit(EXIT_FAILURE); } if (pid == 0) { /* Pop stack... */ return; } /* * We are the parent process. * * Race the child process's teardown with a PMD unshare. * * We do this by triggering: * * __split_vma() -> hugetlb_split() -> hugetlb_unshare_pmds() * * Which, importantly, doesn't hold the hugetlb VMA lock (nor can * it), meaning we assert in hugetlb_vma_assert_locked(). * * . * |----------.----------| * | . | * |----------.----------| * 0 . PUD boundary */ mmap(0, PUD_SIZE / 2, PROT_READ | PROT_WRITE, MAP_FIXED | MAP_ANON | MAP_PRIVATE, -1, 0); } int main(void) { int i; /* Kick off fork children. */ for (i = 0; i < NUM_FORKS; i++) { pid_t pid = fork(); if (pid < 0) { perror("fork"); exit(EXIT_FAILURE); } /* Fork children do their work and exit. */ if (!pid) { int j; for (j = 0; j < NUM_ITERS; j++) execute_one(); return EXIT_SUCCESS; } } /* If we succeeded, wait on children. */ for (i = 0; i < NUM_FORKS; i++) wait(NULL); return EXIT_SUCCESS; } [ljs@kernel.org: account for the !CONFIG_HUGETLB_PMD_PAGE_TABLE_SHARING case] Link: https://lore.kernel.org/agWZsPGYid08uU6O@lucifer Link: https://lore.kernel.org/20260513085658.45264-1-ljs@kernel.org Fixes: 081056dc00a2 ("mm/hugetlb: unshare page tables during VMA split, not before") Signed-off-by: Lorenzo Stoakes Acked-by: David Hildenbrand (Arm) Acked-by: Oscar Salvador Cc: Jann Horn Cc: Muchun Song Cc: Signed-off-by: Andrew Morton --- mm/hugetlb.c | 56 ++++++++++++++++++++++++++++++++++------------------ 1 file changed, 37 insertions(+), 19 deletions(-) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 4b80b167cc9c..ece86d9339ce 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -118,6 +118,9 @@ static int hugetlb_acct_memory(struct hstate *h, long delta); static void hugetlb_vma_lock_free(struct vm_area_struct *vma); static void hugetlb_vma_lock_alloc(struct vm_area_struct *vma); static void __hugetlb_vma_unlock_write_free(struct vm_area_struct *vma); +static int __huge_pmd_unshare(struct mmu_gather *tlb, + struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, + bool check_locks); static void hugetlb_unshare_pmds(struct vm_area_struct *vma, unsigned long start, unsigned long end, bool take_locks); static struct resv_map *vma_resv_map(struct vm_area_struct *vma); @@ -6891,6 +6894,31 @@ pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma, return pte; } +static int __huge_pmd_unshare(struct mmu_gather *tlb, + struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, + bool check_locks) +{ + unsigned long sz = huge_page_size(hstate_vma(vma)); + struct mm_struct *mm = vma->vm_mm; + pgd_t *pgd = pgd_offset(mm, addr); + p4d_t *p4d = p4d_offset(pgd, addr); + pud_t *pud = pud_offset(p4d, addr); + + if (sz != PMD_SIZE) + return 0; + if (!ptdesc_pmd_is_shared(virt_to_ptdesc(ptep))) + return 0; + i_mmap_assert_write_locked(vma->vm_file->f_mapping); + if (check_locks) + hugetlb_vma_assert_locked(vma); + pud_clear(pud); + + tlb_unshare_pmd_ptdesc(tlb, virt_to_ptdesc(ptep), addr); + + mm_dec_nr_pmds(mm); + return 1; +} + /** * huge_pmd_unshare - Unmap a pmd table if it is shared by multiple users * @tlb: the current mmu_gather. @@ -6910,24 +6938,7 @@ pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma, int huge_pmd_unshare(struct mmu_gather *tlb, struct vm_area_struct *vma, unsigned long addr, pte_t *ptep) { - unsigned long sz = huge_page_size(hstate_vma(vma)); - struct mm_struct *mm = vma->vm_mm; - pgd_t *pgd = pgd_offset(mm, addr); - p4d_t *p4d = p4d_offset(pgd, addr); - pud_t *pud = pud_offset(p4d, addr); - - if (sz != PMD_SIZE) - return 0; - if (!ptdesc_pmd_is_shared(virt_to_ptdesc(ptep))) - return 0; - i_mmap_assert_write_locked(vma->vm_file->f_mapping); - hugetlb_vma_assert_locked(vma); - pud_clear(pud); - - tlb_unshare_pmd_ptdesc(tlb, virt_to_ptdesc(ptep), addr); - - mm_dec_nr_pmds(mm); - return 1; + return __huge_pmd_unshare(tlb, vma, addr, ptep, /*check_locks=*/true); } /* @@ -6961,6 +6972,13 @@ pte_t *huge_pmd_share(struct mm_struct *mm, struct vm_area_struct *vma, return NULL; } +static int __huge_pmd_unshare(struct mmu_gather *tlb, + struct vm_area_struct *vma, unsigned long addr, pte_t *ptep, + bool check_locks) +{ + return 0; +} + int huge_pmd_unshare(struct mmu_gather *tlb, struct vm_area_struct *vma, unsigned long addr, pte_t *ptep) { @@ -7269,7 +7287,7 @@ static void hugetlb_unshare_pmds(struct vm_area_struct *vma, if (!ptep) continue; ptl = huge_pte_lock(h, mm, ptep); - huge_pmd_unshare(&tlb, vma, address, ptep); + __huge_pmd_unshare(&tlb, vma, address, ptep, take_locks); spin_unlock(ptl); } huge_pmd_unshare_flush(&tlb, vma); From c0cafe24d3f6534294c4b2bc2d47734ff7cbd313 Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Thu, 21 May 2026 15:37:51 -0700 Subject: [PATCH 062/329] memcg: use round-robin victim selection in refill_stock Harry Yoo reported that get_random_u32_below() is not safe to call in the nmi context and memcg charge draining can happen in nmi context. More specifically get_random_u32_below() is neither reentrant- nor NMI-safe: it acquires a per-cpu local_lock via local_lock_irqsave() on the batched_entropy_u32 state. An NMI that lands on a CPU mid-update of the ChaCha batch state and recurses into the random subsystem would corrupt that state. The memcg_stock local_trylock prevents re-entry on the percpu stock itself, but cannot protect an unrelated subsystem's per-cpu lock. Replace the random pick with a per-cpu round-robin counter stored in memcg_stock_pcp and serialized by the same local_trylock that already guards cached[] and nr_pages[]. No atomics, no random calls, no extra locks needed. Link: https://lore.kernel.org/20260521223751.3794625-1-shakeel.butt@linux.dev Fixes: f735eebe55f8f ("memcg: multi-memcg percpu charge cache") Signed-off-by: Shakeel Butt Reported-by: Harry Yoo Closes: https://lore.kernel.org/4e20f643-6983-4b6e-b12d-c6c4eb20ae0c@kernel.org/ Acked-by: Harry Yoo (Oracle) Acked-by: Michal Hocko Cc: Johannes Weiner Cc: Muchun Song Cc: Roman Gushchin Cc: Signed-off-by: Andrew Morton --- mm/memcontrol.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 177732fef010..1a4fd2504bcd 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2011,6 +2011,7 @@ struct memcg_stock_pcp { struct work_struct work; unsigned long flags; + uint8_t drain_idx; }; static DEFINE_PER_CPU_ALIGNED(struct memcg_stock_pcp, memcg_stock) = { @@ -2194,7 +2195,9 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages) if (!success) { i = empty_slot; if (i == -1) { - i = get_random_u32_below(NR_MEMCG_STOCK); + i = stock->drain_idx++; + if (stock->drain_idx == NR_MEMCG_STOCK) + stock->drain_idx = 0; drain_stock(stock, i); } css_get(&memcg->css); From c0ca59beb5252ea2bd4fdaef009d003dedc2030e Mon Sep 17 00:00:00 2001 From: Muchun Song Date: Wed, 20 May 2026 14:10:25 +0800 Subject: [PATCH 063/329] mm/cma_debug: fix invalid accesses for inactive CMA areas cma_activate_area() can fail after allocating range bitmaps. Its cleanup path frees those bitmaps, but only clears cma->count and cma->available_count. It leaves cma->nranges and each range's count in place, so cma_debugfs_init() can still register debugfs files for an area that never activated successfully. That exposes two problems. Reading the bitmap file can make debugfs walk a freed range bitmap and trigger an invalid memory access. Reading maxchunk can also take cma->lock even though that lock is initialized only on the successful activation path. Fix this by creating debugfs entries only for CMA areas that reached CMA_ACTIVATED. c009da4258f9 introduced the invalid access to bitmap file. 2e32b947606d introduced the invalid access to cma->lock. This change applies to both issues. So I added two Fixes tags. Link: https://lore.kernel.org/20260520061025.3971821-1-songmuchun@bytedance.com Fixes: c009da4258f9 ("mm, cma: support multiple contiguous ranges, if requested") Fixes: 2e32b947606d ("mm: cma: add functions to get region pages counters") Signed-off-by: Muchun Song Acked-by: Mike Rapoport (Microsoft) Acked-by: Oscar Salvador (SUSE) Acked-by: David Hildenbrand (Arm) Cc: Dmitry Safonov <0x7f454c46@gmail.com> Cc: Frank van der Linden Cc: Liam R. Howlett Cc: Lorenzo Stoakes Cc: Michal Hocko Cc: Michal Nazarewicz Cc: Stefan Strogin Cc: Suren Baghdasaryan Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- mm/cma_debug.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/mm/cma_debug.c b/mm/cma_debug.c index 5ae38f5abbcc..523ba4a0f9f7 100644 --- a/mm/cma_debug.c +++ b/mm/cma_debug.c @@ -205,7 +205,8 @@ static int __init cma_debugfs_init(void) cma_debugfs_root = debugfs_create_dir("cma", NULL); for (i = 0; i < cma_area_count; i++) - cma_debugfs_add_one(&cma_areas[i], cma_debugfs_root); + if (test_bit(CMA_ACTIVATED, &cma_areas[i].flags)) + cma_debugfs_add_one(&cma_areas[i], cma_debugfs_root); return 0; } From 40c81856e622a9dc59294a90d169ac07ea25b0b0 Mon Sep 17 00:00:00 2001 From: David Carlier Date: Wed, 20 May 2026 05:49:12 +0100 Subject: [PATCH 064/329] mm/hugetlb: restore reservation on error in hugetlb folio copy paths Two sites in mm/hugetlb.c allocate a hugetlb folio via alloc_hugetlb_folio() (consuming a VMA reservation) and then call copy_user_large_folio(), which became int-returning in commit 1cb9dc4b475c ("mm: hwpoison: support recovery from HugePage copy-on-write faults") and can now fail (e.g. -EHWPOISON on a hwpoisoned source page). On the failure path, folio_put() restores the global hugetlb pool count through free_huge_folio(), but the per-VMA reservation map entry is left marked consumed: - hugetlb_mfill_atomic_pte() resubmission path (UFFDIO_COPY) - copy_hugetlb_page_range() fork-time CoW path when hugetlb_try_dup_anon_rmap() fails (rare: pinned hugetlb anon folio under fork) User-visible effect: on UFFDIO_COPY into a private hugetlb VMA where the resubmission copy fails, the reservation for that address is leaked from the VMA's reserve map. A subsequent fault at the same address takes the no-reservation path, and under hugetlb pool pressure the task is SIGBUSed at an address it had previously reserved. The fork-time CoW path leaks the same way in the child VMA's reserve map, though it requires the much rarer combination of pinned hugetlb anon page + hwpoisoned source. Add the missing restore_reserve_on_error() call before folio_put() on both error paths. Link: https://lore.kernel.org/20260520044912.6751-1-devnexen@gmail.com Fixes: 1cb9dc4b475c ("mm: hwpoison: support recovery from HugePage copy-on-write faults") Signed-off-by: David Carlier Reviewed-by: Muchun Song Cc: David Hildenbrand Cc: Mina Almasry Cc: Muchun Song Cc: Oscar Salvador Cc: yuehaibing Cc: Signed-off-by: Andrew Morton --- mm/hugetlb.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/hugetlb.c b/mm/hugetlb.c index ece86d9339ce..1b1d4f87a3a4 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -4977,6 +4977,7 @@ int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src, addr, dst_vma); folio_put(pte_folio); if (ret) { + restore_reserve_on_error(h, dst_vma, addr, new_folio); folio_put(new_folio); break; } @@ -6273,6 +6274,7 @@ int hugetlb_mfill_atomic_pte(pte_t *dst_pte, folio_put(*foliop); *foliop = NULL; if (ret) { + restore_reserve_on_error(h, dst_vma, dst_addr, folio); folio_put(folio); goto out; } From 3c2d42b8ee345b17a4ba56b0f6492d1ff4c1178e Mon Sep 17 00:00:00 2001 From: Wupeng Ma Date: Fri, 22 May 2026 09:03:05 +0800 Subject: [PATCH 065/329] mm/memory-failure: fix hugetlb_lock AA deadlock in get_huge_page_for_hwpoison Two concurrent madvise(MADV_HWPOISON) calls on the same hugetlb page can trigger a recursive spinlock self-deadlock (AA deadlock) on hugetlb_lock when racing with a concurrent unmap: thread#0 thread#1 -------- -------- madvise(folio, MADV_HWPOISON) -> poisons the folio successfully madvise(folio, MADV_HWPOISON) unmap(folio) try_memory_failure_hugetlb get_huge_page_for_hwpoison spin_lock_irq(&hugetlb_lock) <- held __get_huge_page_for_hwpoison hugetlb_update_hwpoison() -> MF_HUGETLB_FOLIO_PRE_POISONED goto out: folio_put() refcount: 1 -> 0 free_huge_folio() spin_lock_irqsave(&hugetlb_lock) -> AA DEADLOCK! The out: path in __get_huge_page_for_hwpoison() calls folio_put() to drop the GUP reference while the hugetlb_lock is still held by the hugetlb.c wrapper get_huge_page_for_hwpoison(). If concurrent unmap has released the page table mapping reference, folio_put() drops the folio refcount to zero, triggering free_huge_folio() which attempts to re-acquire the non-recursive hugetlb_lock. Fix this by moving hugetlb_lock acquisition from the hugetlb.c wrapper into get_huge_page_for_hwpoison(). Place spin_unlock_irq() before the folio_put() at the out: label so the folio is always released outside the lock. [akpm@linux-foundation.org: fix race, rename label per Miaohe] Link: https://sashiko.dev/#/patchset/20260522010305.4099834-1-mawupeng1@huawei.com Link: https://lore.kernel.org/f39f405e-4b4b-8f79-70fe-a2b5b62114eb@huawei.com Link: https://lore.kernel.org/20260522010305.4099834-1-mawupeng1@huawei.com Fixes: 405ce051236c ("mm/hwpoison: fix race between hugetlb free/demotion and memory_failure_hugetlb()") Signed-off-by: Wupeng Ma Acked-by: Oscar Salvador (SUSE) Acked-by: Muchun Song Reviewed-by: Kefeng Wang Acked-by: Miaohe Lin Cc: David Hildenbrand Cc: Liam Howlett Cc: Lorenzo Stoakes Cc: Michal Hocko Cc: Mike Rapoport Cc: Naoya Horiguchi Cc: Suren Baghdasaryan Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- include/linux/hugetlb.h | 8 -------- include/linux/mm.h | 8 -------- mm/hugetlb.c | 11 ----------- mm/memory-failure.c | 19 ++++++++++--------- 4 files changed, 10 insertions(+), 36 deletions(-) diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h index 5957bc25efa8..2abaf99321e9 100644 --- a/include/linux/hugetlb.h +++ b/include/linux/hugetlb.h @@ -153,8 +153,6 @@ long hugetlb_unreserve_pages(struct inode *inode, long start, long end, long freed); bool folio_isolate_hugetlb(struct folio *folio, struct list_head *list); int get_hwpoison_hugetlb_folio(struct folio *folio, bool *hugetlb, bool unpoison); -int get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared); void folio_putback_hugetlb(struct folio *folio); void move_hugetlb_state(struct folio *old_folio, struct folio *new_folio, int reason); void hugetlb_fix_reserve_counts(struct inode *inode); @@ -421,12 +419,6 @@ static inline int get_hwpoison_hugetlb_folio(struct folio *folio, bool *hugetlb, return 0; } -static inline int get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) -{ - return 0; -} - static inline void folio_putback_hugetlb(struct folio *folio) { } diff --git a/include/linux/mm.h b/include/linux/mm.h index 06bbe9eba636..fc2acedf0b76 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -4975,8 +4975,6 @@ extern int soft_offline_page(unsigned long pfn, int flags); */ extern const struct attribute_group memory_failure_attr_group; extern void memory_failure_queue(unsigned long pfn, int flags); -extern int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared); void num_poisoned_pages_inc(unsigned long pfn); void num_poisoned_pages_sub(unsigned long pfn, long i); #else @@ -4984,12 +4982,6 @@ static inline void memory_failure_queue(unsigned long pfn, int flags) { } -static inline int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) -{ - return 0; -} - static inline void num_poisoned_pages_inc(unsigned long pfn) { } diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 1b1d4f87a3a4..c921287489de 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -7161,17 +7161,6 @@ int get_hwpoison_hugetlb_folio(struct folio *folio, bool *hugetlb, bool unpoison return ret; } -int get_huge_page_for_hwpoison(unsigned long pfn, int flags, - bool *migratable_cleared) -{ - int ret; - - spin_lock_irq(&hugetlb_lock); - ret = __get_huge_page_for_hwpoison(pfn, flags, migratable_cleared); - spin_unlock_irq(&hugetlb_lock); - return ret; -} - /** * folio_putback_hugetlb - unisolate a hugetlb folio * @folio: the isolated hugetlb folio diff --git a/mm/memory-failure.c b/mm/memory-failure.c index ee42d4361309..d47aef256a32 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1966,20 +1966,19 @@ void folio_clear_hugetlb_hwpoison(struct folio *folio) folio_free_raw_hwp(folio, true); } -/* - * Called from hugetlb code with hugetlb_lock held. - */ -int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, +static int get_huge_page_for_hwpoison(unsigned long pfn, int flags, bool *migratable_cleared) { struct page *page = pfn_to_page(pfn); - struct folio *folio = page_folio(page); + struct folio *folio; bool count_increased = false; int ret, rc; + spin_lock_irq(&hugetlb_lock); + folio = page_folio(page); if (!folio_test_hugetlb(folio)) { ret = MF_HUGETLB_NON_HUGEPAGE; - goto out; + goto out_unlock; } else if (flags & MF_COUNT_INCREASED) { ret = MF_HUGETLB_IN_USED; count_increased = true; @@ -1995,13 +1994,13 @@ int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, } else { ret = MF_HUGETLB_RETRY; if (!(flags & MF_NO_RETRY)) - goto out; + goto out_unlock; } rc = hugetlb_update_hwpoison(folio, page); if (rc >= MF_HUGETLB_FOLIO_PRE_POISONED) { ret = rc; - goto out; + goto out_unlock; } /* @@ -2013,8 +2012,10 @@ int __get_huge_page_for_hwpoison(unsigned long pfn, int flags, *migratable_cleared = true; } + spin_unlock_irq(&hugetlb_lock); return ret; -out: +out_unlock: + spin_unlock_irq(&hugetlb_lock); if (count_increased) folio_put(folio); return ret; From 00739e4dd46dde2b39dd9dd19a27e3c8af4ca0d0 Mon Sep 17 00:00:00 2001 From: Muchun Song Date: Sat, 23 May 2026 14:01:23 +0800 Subject: [PATCH 066/329] mm/cma: fix reserved page leak on activation failure If cma_activate_area() fails after allocating only part of the range bitmaps, the cleanup path still has to release the reserved pages when CMA_RESERVE_PAGES_ON_ERROR is clear. That is still worth doing even in this __init path. A bitmap_zalloc() failure does not necessarily mean the system cannot make further progress: freeing the reserved CMA pages can return a substantial amount of memory to the buddy allocator and may relieve the temporary memory shortage that caused the allocation failure in the first place. However, the cleanup path currently uses the bitmap-freeing bound for page release as well. That is only correct for ranges whose bitmap allocation already succeeded. The failed range and all later ranges still keep their reserved pages, so a partial bitmap allocation failure can permanently leak them. Fix this by releasing reserved pages for all ranges. Use the saved early_pfn[] value for ranges whose bitmap allocation already succeeded and for the failed range, and use cmr->early_pfn for later ranges whose bitmap allocation was never attempted. Link: https://lore.kernel.org/20260523060123.2207992-1-songmuchun@bytedance.com Fixes: c009da4258f9 ("mm, cma: support multiple contiguous ranges, if requested") Signed-off-by: Muchun Song Reviewed-by: Oscar Salvador (SUSE) Acked-by: Usama Arif Cc: David Hildenbrand Cc: Frank van der Linden Cc: Liam R. Howlett Cc: Lorenzo Stoakes Cc: Michal Hocko Cc: Mike Rapoport Cc: Suren Baghdasaryan Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- mm/cma.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/mm/cma.c b/mm/cma.c index c7ca567f4c5c..a13ce4999b39 100644 --- a/mm/cma.c +++ b/mm/cma.c @@ -188,10 +188,13 @@ static void __init cma_activate_area(struct cma *cma) /* Expose all pages to the buddy, they are useless for CMA. */ if (!test_bit(CMA_RESERVE_PAGES_ON_ERROR, &cma->flags)) { - for (r = 0; r < allocrange; r++) { + for (r = 0; r < cma->nranges; r++) { + unsigned long start_pfn; + cmr = &cma->ranges[r]; + start_pfn = r <= allocrange ? early_pfn[r] : cmr->early_pfn; end_pfn = cmr->base_pfn + cmr->count; - for (pfn = early_pfn[r]; pfn < end_pfn; pfn++) + for (pfn = start_pfn; pfn < end_pfn; pfn++) free_reserved_page(pfn_to_page(pfn)); } } From 1aee05e814d292064bf5fa15733741040cdc48ba Mon Sep 17 00:00:00 2001 From: Gao Xiang Date: Fri, 22 May 2026 16:27:16 +0800 Subject: [PATCH 067/329] erofs: fix use-after-free on sbi->sync_decompress z_erofs_decompress_kickoff() can race with filesystem unmount, causing a use-after-free on sbi->sync_decompress. When I/O completes, z_erofs_endio() calls z_erofs_decompress_kickoff() to queue z_erofs_decompressqueue_work() asynchronously. Then, after all folios are unlocked, unmount workflow can proceed and sbi will be freed before accessing to sbi->sync_decompress. Thread (unmount) I/O completion kworker queue_work z_erofs_decompressqueue_work (all folios are unlocked) cleanup_mnt .. erofs_kill_sb erofs_sb_free kfree(sbi) access sbi->sync_decompress // UAF!! Fixes: 40452ffca3c1 ("erofs: add sysfs node to control sync decompression strategy") Reported-by: syzbot+52bae5c495dbe261a0bc@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=52bae5c495dbe261a0bc Reviewed-by: Chao Yu Reviewed-by: Jianan Huang Signed-off-by: Gao Xiang --- fs/erofs/zdata.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 27ab7bd844ec..c6240dccbb0f 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -1455,6 +1455,9 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, if (atomic_add_return(bios, &io->pending_bios)) return; if (z_erofs_in_atomic()) { + /* See `sync_decompress` in sysfs-fs-erofs for more details */ + if (sbi->sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO) + sbi->sync_decompress = EROFS_SYNC_DECOMPRESS_FORCE_ON; #ifdef CONFIG_EROFS_FS_PCPU_KTHREAD struct kthread_worker *worker; @@ -1471,9 +1474,6 @@ static void z_erofs_decompress_kickoff(struct z_erofs_decompressqueue *io, #else queue_work(z_erofs_workqueue, &io->u.work); #endif - /* See `sync_decompress` in sysfs-fs-erofs for more details */ - if (sbi->sync_decompress == EROFS_SYNC_DECOMPRESS_AUTO) - sbi->sync_decompress = EROFS_SYNC_DECOMPRESS_FORCE_ON; return; } gfp_flag = memalloc_noio_save(); From ecf3edd349dfabee9bc8a46c5ff91c9ebd858d48 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Wed, 20 May 2026 09:38:44 +0200 Subject: [PATCH 068/329] crypto: s390 - add select CRYPTO_AEAD for aes The aes driver registers both skcipher and aead algorithms, but when aead is not enabled this causes a link failure: s390-linux-ld: arch/s390/crypto/aes_s390.o: in function `aes_s390_fini': arch/s390/crypto/aes_s390.c:969:(.text+0x115e): undefined reference to `crypto_unregister_aead' s390-linux-ld: arch/s390/crypto/aes_s390.o: in function `aes_s390_init': arch/s390/crypto/aes_s390.c:1028:(.init.text+0x294): undefined reference to `crypto_register_aead' Add the missing 'select' statement. Fixes: bf7fa038707c ("s390/crypto: add s390 platform specific aes gcm support.") Signed-off-by: Arnd Bergmann Reviewed-by: Harald Freudenberger Signed-off-by: Herbert Xu --- arch/s390/crypto/Kconfig | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/s390/crypto/Kconfig b/arch/s390/crypto/Kconfig index 00051d27db95..228570a1b233 100644 --- a/arch/s390/crypto/Kconfig +++ b/arch/s390/crypto/Kconfig @@ -4,6 +4,7 @@ menu "Accelerated Cryptographic Algorithms for CPU (s390)" config CRYPTO_AES_S390 tristate "Ciphers: AES, modes: ECB, CBC, CTR, XTS, GCM" + select CRYPTO_AEAD select CRYPTO_SKCIPHER help AEAD cipher: AES with GCM From c889b146478885344a220dd468e5a08de088cbc5 Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Tue, 5 May 2026 13:13:02 +0000 Subject: [PATCH 069/329] firmware: samsung: acpm: Fix false timeouts and Use-After-Free in polling Sashiko identified severe races in the polling state machine [1]. In the ACPM driver's polling mode, threads waited for responses by monitoring the globally shared 'bitmap_seqnum'. This caused false timeouts because if a thread processed its response and freed the sequence number, a concurrent TX thread could immediately reallocate it before the polling thread woke up. Additionally, the driver suffered from a cross-thread Use-After-Free (UAF) preemption race. Previously, acpm_get_rx() cleared the sequence number of whichever RX message it drained from the hardware queue. This meant Thread A could globally free Thread B's sequence slot while Thread B was asleep. A new Thread C could then steal the slot, overwrite the buffer, and leave Thread B to wake up to corrupted state or a timeout. Fix this by rewriting the polling state machine: 1. Decouple polling from the global allocator by introducing a per-slot 'completed' flag, synchronized via smp_store_release() and smp_load_acquire(). 2. Strip acpm_get_saved_rx() out of acpm_get_rx() to make it a pure queue-draining function. Introduce a 'native_match' boolean argument which evaluates to true only if the thread natively processed its own sequence number during the call. This explicitly informs the polling loop whether it must retrieve its payload from the cross-thread cache. 3. Centralize the cache fallback and sequence number free (clear_bit) inside the polling loop. Crucially, the free operation now strictly targets the thread's own TX sequence number (xfer->txd[0]), rather than the drained RX sequence number. This enforces strict ownership: a thread only ever frees its own allocated sequence slot, and only at the exact moment it completes its poll, eliminating the UAF window. Furthermore, explicitly guard the 'native_match' assignment with an if (rx_seqnum == tx_seqnum) check, even for zero-length (no payload) responses. While an unguarded assignment wouldn't crash (because the cache fallback acpm_get_saved_rx() safely returns early on zero-length transfers) doing so would "lie" to the state machine. If a thread drained the queue and found another thread's zero-length message, setting native_match = true would falsely convince the polling loop that it natively handled its own response. Maintaining a rigorous state machine requires that native_match is only set when a thread explicitly processes its own sequence number. Cc: stable@vger.kernel.org Fixes: a88927b534ba ("firmware: add Exynos ACPM protocol driver") Closes: https://sashiko.dev/#/patchset/20260429-acpm-fixes-sashiko-reports-v3-0-47cf74ab09ad%40linaro.org [1] Signed-off-by: Tudor Ambarus Link: https://patch.msgid.link/20260505-acpm-fixes-sashiko-reports-v5-5-43b5ee7f1674@linaro.org Signed-off-by: Krzysztof Kozlowski --- drivers/firmware/samsung/exynos-acpm.c | 68 ++++++++++++++++++-------- 1 file changed, 48 insertions(+), 20 deletions(-) diff --git a/drivers/firmware/samsung/exynos-acpm.c b/drivers/firmware/samsung/exynos-acpm.c index 9766425a44ab..620880d8820a 100644 --- a/drivers/firmware/samsung/exynos-acpm.c +++ b/drivers/firmware/samsung/exynos-acpm.c @@ -105,11 +105,14 @@ struct acpm_queue { * @cmd: pointer to where the data shall be saved. * @n_cmd: number of 32-bit commands. * @rxcnt: expected length of the response in 32-bit words. + * @completed: flag indicating if the firmware response has been fully + * processed. */ struct acpm_rx_data { u32 *cmd; size_t n_cmd; size_t rxcnt; + bool completed; }; #define ACPM_SEQNUM_MAX 64 @@ -204,26 +207,28 @@ static void acpm_get_saved_rx(struct acpm_chan *achan, rx_seqnum = FIELD_GET(ACPM_PROTOCOL_SEQNUM, rx_data->cmd[0]); - if (rx_seqnum == tx_seqnum) { + if (rx_seqnum == tx_seqnum) memcpy(xfer->rxd, rx_data->cmd, xfer->rxcnt * sizeof(*xfer->rxd)); - clear_bit(rx_seqnum - 1, achan->bitmap_seqnum); - } } /** * acpm_get_rx() - get response from RX queue. * @achan: ACPM channel info. * @xfer: reference to the transfer to get response for. + * @native_match: pointer to a boolean set to true if the thread natively + * processed its own sequence number during this call. * * Return: 0 on success, -errno otherwise. */ -static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) +static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer, + bool *native_match) { u32 rx_front, rx_seqnum, tx_seqnum, seqnum; const void __iomem *base, *addr; struct acpm_rx_data *rx_data; u32 i, val, mlen; - bool rx_set = false; + + *native_match = false; guard(mutex)(&achan->rx_lock); @@ -232,10 +237,8 @@ static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) tx_seqnum = FIELD_GET(ACPM_PROTOCOL_SEQNUM, xfer->txd[0]); - if (i == rx_front) { - acpm_get_saved_rx(achan, xfer, tx_seqnum); + if (i == rx_front) return 0; - } base = achan->rx.base; mlen = achan->mlen; @@ -259,8 +262,13 @@ static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) if (rx_data->rxcnt) { if (rx_seqnum == tx_seqnum) { __ioread32_copy(xfer->rxd, addr, xfer->rxcnt); - rx_set = true; - clear_bit(seqnum, achan->bitmap_seqnum); + /* + * Signal completion to the polling thread. + * Pairs with smp_load_acquire() in polling + * loop. + */ + smp_store_release(&rx_data->completed, true); + *native_match = true; } else { /* * The RX data corresponds to another request. @@ -270,9 +278,21 @@ static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) */ __ioread32_copy(rx_data->cmd, addr, rx_data->rxcnt); + /* + * Signal completion to the polling thread. + * Pairs with smp_load_acquire() in polling + * loop. + */ + smp_store_release(&rx_data->completed, true); } } else { - clear_bit(seqnum, achan->bitmap_seqnum); + /* + * Signal completion to the polling thread. + * Pairs with smp_load_acquire() in polling loop. + */ + smp_store_release(&rx_data->completed, true); + if (rx_seqnum == tx_seqnum) + *native_match = true; } i = (i + 1) % achan->qlen; @@ -281,13 +301,6 @@ static int acpm_get_rx(struct acpm_chan *achan, const struct acpm_xfer *xfer) /* We saved all responses, mark RX empty. */ writel(rx_front, achan->rx.rear); - /* - * If the response was not in this iteration of the queue, check if the - * RX data was previously saved. - */ - if (!rx_set) - acpm_get_saved_rx(achan, xfer, tx_seqnum); - return 0; } @@ -302,6 +315,7 @@ static int acpm_dequeue_by_polling(struct acpm_chan *achan, const struct acpm_xfer *xfer) { struct device *dev = achan->acpm->dev; + bool native_match; ktime_t timeout; u32 seqnum; int ret; @@ -310,12 +324,25 @@ static int acpm_dequeue_by_polling(struct acpm_chan *achan, timeout = ktime_add_us(ktime_get(), ACPM_POLL_TIMEOUT_US); do { - ret = acpm_get_rx(achan, xfer); + ret = acpm_get_rx(achan, xfer, &native_match); if (ret) return ret; - if (!test_bit(seqnum - 1, achan->bitmap_seqnum)) + /* + * Safely check if our specific transaction has been processed. + * smp_load_acquire prevents the CPU from speculatively + * executing subsequent instructions before the transaction is + * synchronized. + */ + if (smp_load_acquire(&achan->rx_data[seqnum - 1].completed)) { + /* Retrieve payload if another thread cached it for us */ + if (!native_match) + acpm_get_saved_rx(achan, xfer, seqnum); + + /* Relinquish ownership of the sequence slot */ + clear_bit(seqnum - 1, achan->bitmap_seqnum); return 0; + } /* Determined experimentally. */ udelay(20); @@ -380,6 +407,7 @@ static void acpm_prepare_xfer(struct acpm_chan *achan, /* Clear data for upcoming responses */ rx_data = &achan->rx_data[achan->seqnum - 1]; + rx_data->completed = false; memset(rx_data->cmd, 0, sizeof(*rx_data->cmd) * rx_data->n_cmd); /* zero means no response expected */ rx_data->rxcnt = xfer->rxcnt; From bf296f83a3ddab1ab875edc4e8862cb10553064f Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Tue, 5 May 2026 13:13:03 +0000 Subject: [PATCH 070/329] firmware: samsung: acpm: Fix missing LKMM barriers in sequence allocator Sashiko identified memory ordering races in [1]. The ACPM driver uses a globally shared 'bitmap_seqnum' to track available sequence numbers. Even though threads now strictly free their own sequence numbers, the allocation and freeing of these bits across concurrent threads are effectively lockless operations and require explicit LKMM memory barriers. Previously, the driver used plain bitwise operators (test_bit, set_bit, clear_bit), which lack ordering guarantees. This creates two race conditions on weakly ordered architectures like ARM64: 1. Polling Release Violation: The polling thread copies its payload and calls clear_bit(). Without a release barrier, the CPU can reorder the memory operations, making the cleared bit globally visible before the payload reads have fully completed. 2. TX Acquire Violation: The TX thread loops on test_bit(), calls set_bit(), and then wipes the payload buffer via memset(). Without an acquire barrier, the CPU can speculatively execute the memset() before the bit is safely and formally claimed. If these reorderings overlap, a new TX thread can claim the sequence number and overwrite the buffer while the original polling thread is still actively reading from it. Fix this by upgrading the bitwise operators. Wrap the TX allocation in test_and_set_bit_lock() to establish formal LKMM Acquire semantics, and pair it with clear_bit_unlock() in the polling path to enforce Release semantics. Cc: stable@vger.kernel.org Fixes: a88927b534ba ("firmware: add Exynos ACPM protocol driver") Closes: https://sashiko.dev/#/patchset/20260423-acpm-fixes-sashiko-reports-v1-0-2217b790925e%40linaro.org [1] Signed-off-by: Tudor Ambarus Link: https://patch.msgid.link/20260505-acpm-fixes-sashiko-reports-v5-6-43b5ee7f1674@linaro.org Signed-off-by: Krzysztof Kozlowski --- drivers/firmware/samsung/exynos-acpm.c | 18 +++++++++++------- 1 file changed, 11 insertions(+), 7 deletions(-) diff --git a/drivers/firmware/samsung/exynos-acpm.c b/drivers/firmware/samsung/exynos-acpm.c index 620880d8820a..3fa9fe283be4 100644 --- a/drivers/firmware/samsung/exynos-acpm.c +++ b/drivers/firmware/samsung/exynos-acpm.c @@ -7,7 +7,7 @@ #include #include -#include +#include #include #include #include @@ -340,7 +340,7 @@ static int acpm_dequeue_by_polling(struct acpm_chan *achan, acpm_get_saved_rx(achan, xfer, seqnum); /* Relinquish ownership of the sequence slot */ - clear_bit(seqnum - 1, achan->bitmap_seqnum); + clear_bit_unlock(seqnum - 1, achan->bitmap_seqnum); return 0; } @@ -397,11 +397,18 @@ static void acpm_prepare_xfer(struct acpm_chan *achan, struct acpm_rx_data *rx_data; u32 *txd = (u32 *)xfer->txd; - /* Prevent chan->seqnum from being re-used */ + /* + * Prevent chan->seqnum from being re-used. + * test_and_set_bit_lock() provides formal LKMM Acquire semantics. + * It pairs with the RX thread's clear_bit_unlock() to ensure the CPU + * does not speculatively execute the rx_data buffer wipe (memset) + * before the sequence number is safely claimed. + */ do { if (++achan->seqnum == ACPM_SEQNUM_MAX) achan->seqnum = 1; - } while (test_bit(achan->seqnum - 1, achan->bitmap_seqnum)); + /* Flag the index based on seqnum. (seqnum: 1~63, bitmap: 0~62) */ + } while (test_and_set_bit_lock(achan->seqnum - 1, achan->bitmap_seqnum)); txd[0] |= FIELD_PREP(ACPM_PROTOCOL_SEQNUM, achan->seqnum); @@ -411,9 +418,6 @@ static void acpm_prepare_xfer(struct acpm_chan *achan, memset(rx_data->cmd, 0, sizeof(*rx_data->cmd) * rx_data->n_cmd); /* zero means no response expected */ rx_data->rxcnt = xfer->rxcnt; - - /* Flag the index based on seqnum. (seqnum: 1~63, bitmap: 0~62) */ - set_bit(achan->seqnum - 1, achan->bitmap_seqnum); } /** From 7fe40c32a33905302341797b5d12c541729dd08d Mon Sep 17 00:00:00 2001 From: Tudor Ambarus Date: Tue, 5 May 2026 13:13:04 +0000 Subject: [PATCH 071/329] firmware: samsung: acpm: Fix infinite loop on sequence number exhaustion Sashiko identified a possible infinite loop [1]. ACPM IPC sequence numbers are tracked via a 64-bit bitmap. Previously, acpm_prepare_xfer() used a do...while loop to search for a free sequence number. If all 63 available sequence numbers are leaked due to transient hardware timeouts or mailbox failures, the bitmap becomes full. The next call to acpm_prepare_xfer() would enter an infinite loop. Fix this by utilizing the kernel's optimized bitmap search functions (find_next_zero_bit / find_first_zero_bit). If the pool is completely exhausted, log the failure and return -EBUSY to allow the kernel to fail gracefully instead of hanging. Furthermore, drop the allocation loop entirely. Because acpm_prepare_xfer() is strictly called under the 'tx_lock' mutex, sequence number allocations are perfectly serialized. If find_next_zero_bit() locates a free bit, a single test_and_set_bit_lock() is mathematically guaranteed to succeed. To enforce this locking invariant, wrap the allocation in a WARN_ON_ONCE. If the atomic set fails, it indicates the driver's mutex serialization is fundamentally broken. The warning generates a stack trace for debugging, while returning -EIO immediately aborts the transfer to prevent silent payload corruption. Cc: stable@vger.kernel.org Fixes: a88927b534ba ("firmware: add Exynos ACPM protocol driver") Closes: https://sashiko.dev/#/patchset/20260420-acpm-tmu-v3-0-3dc8e93f0b26%40linaro.org [1] Signed-off-by: Tudor Ambarus Link: https://patch.msgid.link/20260505-acpm-fixes-sashiko-reports-v5-7-43b5ee7f1674@linaro.org Signed-off-by: Krzysztof Kozlowski --- drivers/firmware/samsung/exynos-acpm.c | 45 ++++++++++++++++++-------- 1 file changed, 31 insertions(+), 14 deletions(-) diff --git a/drivers/firmware/samsung/exynos-acpm.c b/drivers/firmware/samsung/exynos-acpm.c index 3fa9fe283be4..19db3674a28f 100644 --- a/drivers/firmware/samsung/exynos-acpm.c +++ b/drivers/firmware/samsung/exynos-acpm.c @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -390,34 +391,48 @@ static int acpm_wait_for_queue_slots(struct acpm_chan *achan, u32 next_tx_front) * TX queue. * @achan: ACPM channel info. * @xfer: reference to the transfer being prepared. + * + * Return: 0 on success, -errno otherwise. */ -static void acpm_prepare_xfer(struct acpm_chan *achan, - const struct acpm_xfer *xfer) +static int acpm_prepare_xfer(struct acpm_chan *achan, + const struct acpm_xfer *xfer) { struct acpm_rx_data *rx_data; u32 *txd = (u32 *)xfer->txd; + unsigned long size = ACPM_SEQNUM_MAX - 1; + unsigned long bit = achan->seqnum; + + bit = find_next_zero_bit(achan->bitmap_seqnum, size, bit); + if (bit >= size) { + bit = find_first_zero_bit(achan->bitmap_seqnum, size); + if (bit >= size) { + dev_err_ratelimited(achan->acpm->dev, + "ACPM sequence number pool exhausted\n"); + return -EBUSY; + } + } /* - * Prevent chan->seqnum from being re-used. - * test_and_set_bit_lock() provides formal LKMM Acquire semantics. - * It pairs with the RX thread's clear_bit_unlock() to ensure the CPU - * does not speculatively execute the rx_data buffer wipe (memset) - * before the sequence number is safely claimed. + * Execute the atomic set to formally claim the bit and establish + * LKMM Acquire semantics against the RX thread's clear_bit_unlock(). + * A loop is unnecessary because allocations are strictly serialized + * by tx_lock. */ - do { - if (++achan->seqnum == ACPM_SEQNUM_MAX) - achan->seqnum = 1; - /* Flag the index based on seqnum. (seqnum: 1~63, bitmap: 0~62) */ - } while (test_and_set_bit_lock(achan->seqnum - 1, achan->bitmap_seqnum)); + if (WARN_ON_ONCE(test_and_set_bit_lock(bit, achan->bitmap_seqnum))) + return -EIO; + /* Flag the index based on seqnum. (seqnum: 1~63, bitmap: 0~62) */ + achan->seqnum = bit + 1; txd[0] |= FIELD_PREP(ACPM_PROTOCOL_SEQNUM, achan->seqnum); /* Clear data for upcoming responses */ - rx_data = &achan->rx_data[achan->seqnum - 1]; + rx_data = &achan->rx_data[bit]; rx_data->completed = false; memset(rx_data->cmd, 0, sizeof(*rx_data->cmd) * rx_data->n_cmd); /* zero means no response expected */ rx_data->rxcnt = xfer->rxcnt; + + return 0; } /** @@ -477,7 +492,9 @@ int acpm_do_xfer(struct acpm_handle *handle, const struct acpm_xfer *xfer) if (ret) return ret; - acpm_prepare_xfer(achan, xfer); + ret = acpm_prepare_xfer(achan, xfer); + if (ret) + return ret; /* Write TX command. */ __iowrite32_copy(achan->tx.base + achan->mlen * tx_front, From 8d4ae34e997062076a9098602eaca43353665bd9 Mon Sep 17 00:00:00 2001 From: Huan He Date: Sat, 9 May 2026 16:49:07 +0800 Subject: [PATCH 072/329] mmc: sdhci-of-dwcmshc: Fix reset, clk, and SDIO support for Eswin EIC7700 The EIC7700 code in sdhci-of-dwcmshc uses host->mmc->caps2 to select different configuration paths for different card types. The current logic distinguishes eMMC and SD, but does not handle SDIO separately. Update the EIC7700 card-type checks so that eMMC, SD and SDIO are distinguished explicitly. Switch the reset path to dwcmshc_reset() so that pending interrupt state is cleared consistently, and use sdhci_enable_clk() so the clock enable sequence follows the standard SDHCI flow. Fixes: 32b2633219d3 ("mmc: sdhci-of-dwcmshc: Add support for Eswin EIC7700") Signed-off-by: Huan He Acked-by: Adrian Hunter Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci-of-dwcmshc.c | 44 ++++++++++++++--------------- 1 file changed, 22 insertions(+), 22 deletions(-) diff --git a/drivers/mmc/host/sdhci-of-dwcmshc.c b/drivers/mmc/host/sdhci-of-dwcmshc.c index 0b2158a7e409..b9ecd91f44ad 100644 --- a/drivers/mmc/host/sdhci-of-dwcmshc.c +++ b/drivers/mmc/host/sdhci-of-dwcmshc.c @@ -277,6 +277,7 @@ #define PHY_DELAY_CODE_MAX 0x7f #define PHY_DELAY_CODE_EMMC 0x17 #define PHY_DELAY_CODE_SD 0x55 +#define PHY_DELAY_CODE_SDIO 0x29 struct rk35xx_priv { struct reset_control *reset; @@ -1433,10 +1434,7 @@ static void sdhci_eic7700_set_clock(struct sdhci_host *host, unsigned int clock) clk_set_rate(pltfm_host->clk, clock); clk = sdhci_readw(host, SDHCI_CLOCK_CONTROL); - clk |= SDHCI_CLOCK_INT_EN; - sdhci_writew(host, clk, SDHCI_CLOCK_CONTROL); - - dwcmshc_enable_card_clk(host); + sdhci_enable_clk(host, clk); } static void sdhci_eic7700_config_phy_delay(struct sdhci_host *host, int delay) @@ -1497,7 +1495,7 @@ static void sdhci_eic7700_config_phy(struct sdhci_host *host) static void sdhci_eic7700_reset(struct sdhci_host *host, u8 mask) { - sdhci_reset(host, mask); + dwcmshc_reset(host, mask); /* after reset all, the phy's config will be clear */ if (mask == SDHCI_RESET_ALL) @@ -1594,18 +1592,17 @@ static int sdhci_eic7700_phase_code_tuning(struct sdhci_host *host, u32 opcode) { struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host); struct dwcmshc_priv *priv = sdhci_pltfm_priv(pltfm_host); - u32 sd_caps = MMC_CAP2_NO_MMC | MMC_CAP2_NO_SDIO; + u32 emmc_caps = MMC_CAP2_NO_SD | MMC_CAP2_NO_SDIO; int phase_code = -1; int code_range = -1; - bool is_sd = false; int code_min = -1; int code_max = -1; int cmd_error = 0; + bool is_emmc; int ret = 0; int i = 0; - if ((host->mmc->caps2 & sd_caps) == sd_caps) - is_sd = true; + is_emmc = (host->mmc->caps2 & emmc_caps) == emmc_caps; for (i = 0; i <= MAX_PHASE_CODE; i++) { /* Centered Phase code */ @@ -1614,8 +1611,8 @@ static int sdhci_eic7700_phase_code_tuning(struct sdhci_host *host, u32 opcode) host->ops->reset(host, SDHCI_RESET_CMD | SDHCI_RESET_DATA); if (ret) { - /* SD specific range tracking */ - if (is_sd && code_min != -1 && code_max != -1) { + /* SD/SDIO specific range tracking */ + if (!is_emmc && code_min != -1 && code_max != -1) { if (code_max - code_min > code_range) { code_range = code_max - code_min; phase_code = (code_min + code_max) / 2; @@ -1626,17 +1623,17 @@ static int sdhci_eic7700_phase_code_tuning(struct sdhci_host *host, u32 opcode) code_max = -1; } /* EMMC breaks after first valid range */ - if (!is_sd && code_min != -1 && code_max != -1) + if (is_emmc && code_min != -1 && code_max != -1) break; } else { /* Track valid phase code range */ if (code_min == -1) { code_min = i; - if (!is_sd) + if (is_emmc) continue; } code_max = i; - if (is_sd && i == MAX_PHASE_CODE) { + if (!is_emmc && i == MAX_PHASE_CODE) { if (code_max - code_min > code_range) { code_range = code_max - code_min; phase_code = (code_min + code_max) / 2; @@ -1646,19 +1643,19 @@ static int sdhci_eic7700_phase_code_tuning(struct sdhci_host *host, u32 opcode) } /* Handle tuning failure case */ - if ((is_sd && phase_code == -1) || - (!is_sd && code_min == -1 && code_max == -1)) { + if ((!is_emmc && phase_code == -1) || + (is_emmc && code_min == -1 && code_max == -1)) { pr_err("%s: phase code tuning failed!\n", mmc_hostname(host->mmc)); sdhci_writew(host, 0, priv->vendor_specific_area1 + DWCMSHC_AT_STAT); return -EIO; } - if (!is_sd) + if (is_emmc) phase_code = (code_min + code_max) / 2; sdhci_writew(host, phase_code, priv->vendor_specific_area1 + DWCMSHC_AT_STAT); - /* SD specific final verification */ - if (is_sd) { + /* SD/SDIO specific final verification */ + if (!is_emmc) { ret = mmc_send_tuning(host->mmc, opcode, &cmd_error); host->ops->reset(host, SDHCI_RESET_CMD | SDHCI_RESET_DATA); if (ret) { @@ -1756,9 +1753,9 @@ static void sdhci_eic7700_set_uhs_signaling(struct sdhci_host *host, unsigned in static void sdhci_eic7700_set_uhs_wrapper(struct sdhci_host *host, unsigned int timing) { - u32 sd_caps = MMC_CAP2_NO_MMC | MMC_CAP2_NO_SDIO; + u32 emmc_caps = MMC_CAP2_NO_SD | MMC_CAP2_NO_SDIO; - if ((host->mmc->caps2 & sd_caps) == sd_caps) + if ((host->mmc->caps2 & emmc_caps) != emmc_caps) sdhci_set_uhs_signaling(host, timing); else sdhci_eic7700_set_uhs_signaling(host, timing); @@ -1767,6 +1764,7 @@ static void sdhci_eic7700_set_uhs_wrapper(struct sdhci_host *host, unsigned int static int eic7700_init(struct device *dev, struct sdhci_host *host, struct dwcmshc_priv *dwc_priv) { u32 emmc_caps = MMC_CAP2_NO_SD | MMC_CAP2_NO_SDIO; + u32 sd_caps = MMC_CAP2_NO_MMC | MMC_CAP2_NO_SDIO; unsigned int val, hsp_int_status, hsp_pwr_ctrl; static const char * const clk_ids[] = {"axi"}; struct of_phandle_args args; @@ -1821,8 +1819,10 @@ static int eic7700_init(struct device *dev, struct sdhci_host *host, struct dwcm if ((host->mmc->caps2 & emmc_caps) == emmc_caps) dwc_priv->delay_line = PHY_DELAY_CODE_EMMC; - else + else if ((host->mmc->caps2 & sd_caps) == sd_caps) dwc_priv->delay_line = PHY_DELAY_CODE_SD; + else + dwc_priv->delay_line = PHY_DELAY_CODE_SDIO; if (!of_property_read_u32(dev->of_node, "eswin,drive-impedance-ohms", &val)) priv->drive_impedance = eic7700_convert_drive_impedance_ohm(dev, val); From f48ee49726ee4ab545fd2dc644f169c0809b19b3 Mon Sep 17 00:00:00 2001 From: Lad Prabhakar Date: Tue, 19 May 2026 14:53:40 +0100 Subject: [PATCH 073/329] mmc: renesas_sdhi: Add OF entry for RZ/G2H SoC The RZ/G2H (R8A774E1) SoC was previously handled via the generic "renesas,rcar-gen3-sdhi" fallback compatible string. However, because the SDHI IP on RZ/G2H is identical with the R-Car H3-N (R8A77951), it requires the specific quirks and configuration defined in `of_r8a7795_compatible` rather than the generic Gen3 data. Add the explicit "renesas,sdhi-r8a774e1" match entry to map it correctly. Note that the DT binding file renesas,sdhi.yaml does not need an update as the entry for this SoC is already present. Fixes: 31941342888d ("arm64: dts: renesas: r8a774e1: Add SDHI nodes") Cc: stable@vger.kernel.org Signed-off-by: Lad Prabhakar Reviewed-by: Wolfram Sang Reviewed-by: Geert Uytterhoeven Signed-off-by: Ulf Hansson --- drivers/mmc/host/renesas_sdhi_internal_dmac.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mmc/host/renesas_sdhi_internal_dmac.c b/drivers/mmc/host/renesas_sdhi_internal_dmac.c index f6ebb7bc7ede..838248bf8dd6 100644 --- a/drivers/mmc/host/renesas_sdhi_internal_dmac.c +++ b/drivers/mmc/host/renesas_sdhi_internal_dmac.c @@ -279,6 +279,7 @@ static const struct renesas_sdhi_of_data_with_quirks of_rza2_compatible = { static const struct of_device_id renesas_sdhi_internal_dmac_of_match[] = { { .compatible = "renesas,sdhi-r7s9210", .data = &of_rza2_compatible, }, { .compatible = "renesas,sdhi-mmc-r8a77470", .data = &of_rcar_gen3_compatible, }, + { .compatible = "renesas,sdhi-r8a774e1", .data = &of_r8a7795_compatible, }, { .compatible = "renesas,sdhi-r8a7795", .data = &of_r8a7795_compatible, }, { .compatible = "renesas,sdhi-r8a77961", .data = &of_r8a77961_compatible, }, { .compatible = "renesas,sdhi-r8a77965", .data = &of_r8a77965_compatible, }, From b837e38c255dd9f8b53511d52e87f1fda32b3dfe Mon Sep 17 00:00:00 2001 From: Inochi Amaoto Date: Thu, 21 May 2026 15:21:20 +0800 Subject: [PATCH 074/329] mmc: litex_mmc: Use DIV_ROUND_UP for more accurate clock calculation The previous clock uses roundup_pow_of_two() to calculate the core clock frequency. It does not meet the actual hardware meaning. The actual frequency is calculated by "ref_clk / ((div >> 1) << 1)". Fix the clock divider calculation. Fixes: 92e099104729 ("mmc: Add driver for LiteX's LiteSDCard interface") Signed-off-by: Inochi Amaoto Reviewed-by: Gabriel Somlo Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/host/litex_mmc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/mmc/host/litex_mmc.c b/drivers/mmc/host/litex_mmc.c index d2f19c2dc673..52571bb17c61 100644 --- a/drivers/mmc/host/litex_mmc.c +++ b/drivers/mmc/host/litex_mmc.c @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -436,11 +437,10 @@ static void litex_mmc_setclk(struct litex_mmc_host *host, unsigned int freq) struct device *dev = mmc_dev(host->mmc); u32 div; - div = freq ? host->ref_clk / freq : 256U; - div = roundup_pow_of_two(div); + div = freq ? DIV_ROUND_UP(host->ref_clk, freq) : 256U; div = clamp(div, 2U, 256U); dev_dbg(dev, "sd_clk_freq=%d: set to %d via div=%d\n", - freq, host->ref_clk / div, div); + freq, host->ref_clk / ((div + 1) & ~1U), div); litex_write16(host->sdphy + LITEX_PHY_CLOCKERDIV, div); host->sd_clk = freq; } From 99982b743e5ba72bd1f5de0e03e3b96ae70b1e51 Mon Sep 17 00:00:00 2001 From: Inochi Amaoto Date: Thu, 21 May 2026 15:21:21 +0800 Subject: [PATCH 075/329] mmc: litex_mmc: Set mandatory idle clocks before CMD0 The litex_mmc driver assumes the card is already probed in the BIOS and skip the phy initialization. This will cause the command fail like the following when the old card is unplugged and then insert a new card: [ 62.923593] litex-mmc f0004000.mmc: Command (cmd 8) error, status -110 [ 62.949717] litex-mmc f0004000.mmc: Command (cmd 55) error, status -110 [ 62.976606] litex-mmc f0004000.mmc: Command (cmd 55) error, status -110 [ 63.002516] litex-mmc f0004000.mmc: Command (cmd 55) error, status -110 [ 63.028442] litex-mmc f0004000.mmc: Command (cmd 55) error, status -110 Add required clock settings and initialization for the CMD 0, so it can probe the new card. Fixes: 92e099104729 ("mmc: Add driver for LiteX's LiteSDCard interface") Signed-off-by: Inochi Amaoto Reviewed-by: Gabriel Somlo Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/host/litex_mmc.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+) diff --git a/drivers/mmc/host/litex_mmc.c b/drivers/mmc/host/litex_mmc.c index 52571bb17c61..3655542ca998 100644 --- a/drivers/mmc/host/litex_mmc.c +++ b/drivers/mmc/host/litex_mmc.c @@ -69,6 +69,9 @@ #define SD_SLEEP_US 5 #define SD_TIMEOUT_US 20000 +#define SD_INIT_DELAY_US 1000 +#define SD_INIT_CLK_HZ 400000 + #define SDIRQ_CARD_DETECT 1 #define SDIRQ_SD_TO_MEM_DONE 2 #define SDIRQ_MEM_TO_SD_DONE 4 @@ -449,6 +452,17 @@ static void litex_mmc_set_ios(struct mmc_host *mmc, struct mmc_ios *ios) { struct litex_mmc_host *host = mmc_priv(mmc); + /* + * The SD specification requires at least 74 idle clocks before CMD0. + * These dummy cycles is generated by writing LITEX_PHY_INITIALIZE. + */ + if (ios->chip_select == MMC_CS_HIGH) { + litex_mmc_setclk(host, SD_INIT_CLK_HZ); + litex_write8(host->sdphy + LITEX_PHY_INITIALIZE, 1); + fsleep(SD_INIT_DELAY_US); + return; + } + /* * NOTE: Ignore any ios->bus_width updates; they occur right after * the mmc core sends its own acmd6 bus-width change notification, From 1e9a4850afa0ceb63984fb1a9f3e86d0fc4fd18f Mon Sep 17 00:00:00 2001 From: Heiko Stuebner Date: Fri, 22 May 2026 20:43:07 +0200 Subject: [PATCH 076/329] mmc: dw_mmc-rockchip: Add missing private data for very old controllers The really old controllers (rk2928, rk3066, rk3188) do not support UHS speeds at all, and thus never handled phase data. For that reason it never had a parse_dt callback and no driver private data at all. Commit ff6f0286c896 ("mmc: dw_mmc-rockchip: Add memory clock auto-gating support") makes the private data sort of mandatory, because the init function checks whether phases are configured internally or through the clock controller. This results in the old SoCs then experiencing NULL-pointer dereferences when they try to access that private-data struct. While we could have if (priv) conditionals in all places, it's way less cluttery to just give the old types their private-data struct. Fixes: ff6f0286c896 ("mmc: dw_mmc-rockchip: Add memory clock auto-gating support") Cc: stable@vger.kernel.org Signed-off-by: Heiko Stuebner Acked-by: Shawn Lin Signed-off-by: Ulf Hansson --- drivers/mmc/host/dw_mmc-rockchip.c | 17 +++++++++++++++++ 1 file changed, 17 insertions(+) diff --git a/drivers/mmc/host/dw_mmc-rockchip.c b/drivers/mmc/host/dw_mmc-rockchip.c index c6eece4ec3fd..75c82ff20f17 100644 --- a/drivers/mmc/host/dw_mmc-rockchip.c +++ b/drivers/mmc/host/dw_mmc-rockchip.c @@ -441,6 +441,22 @@ static int dw_mci_common_parse_dt(struct dw_mci *host) return 0; } +static int dw_mci_rk2928_parse_dt(struct dw_mci *host) +{ + struct dw_mci_rockchip_priv_data *priv; + int err; + + err = dw_mci_common_parse_dt(host); + if (err) + return err; + + priv = host->priv; + + priv->internal_phase = false; + + return 0; +} + static int dw_mci_rk3288_parse_dt(struct dw_mci *host) { struct dw_mci_rockchip_priv_data *priv; @@ -514,6 +530,7 @@ static int dw_mci_rockchip_init(struct dw_mci *host) static const struct dw_mci_drv_data rk2928_drv_data = { .init = dw_mci_rockchip_init, + .parse_dt = dw_mci_rk2928_parse_dt, }; static const struct dw_mci_drv_data rk3288_drv_data = { From f595e8e77a51eee35e331f69321766593a845ef2 Mon Sep 17 00:00:00 2001 From: Jisheng Zhang Date: Sun, 24 May 2026 10:34:55 +0800 Subject: [PATCH 077/329] mmc: sdhci: add signal voltage switch in sdhci_resume_host I met one suspend/resume issue with sdr104 capable sdio wifi card (with "keep-power-in-suspend" set in DT property): After resuming from suspend to ram, the sdio wifi card stops working. Further debug shows that although ios shows the sdio card is at sdr104 mode, the voltage is still at 3V3. This is due to missing the calling of ->start_signal_voltage_switch() in sdhci_resume_host(). Fix this issue by adding ->start_signal_voltage_switch() in sdhci_resume_host(). This also matches what we do for sdhci_runtime_resume_host(). Then the question is: why this issue hasn't reported and fixed for so long time. IMHO, several reasons: Some host controllers just kick off the runtime resume for system resume, so they benefit from the well supported runtime pm code; Some platforms just use the old sdio wifi card which doesn't need signal voltage switch at all, the default voltage is 3v3 after resuming. Fixes: 6308d2905bd3 ("mmc: sdhci: add quirk for keeping card power during suspend") Signed-off-by: Jisheng Zhang Acked-by: Adrian Hunter Cc: stable@vger.kernel.org Signed-off-by: Ulf Hansson --- drivers/mmc/host/sdhci.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c index 605be55f8d2d..e3bf901b10aa 100644 --- a/drivers/mmc/host/sdhci.c +++ b/drivers/mmc/host/sdhci.c @@ -3836,6 +3836,7 @@ int sdhci_resume_host(struct sdhci_host *host) host->pwr = 0; host->clock = 0; host->reinit_uhs = true; + mmc->ops->start_signal_voltage_switch(mmc, &mmc->ios); mmc->ops->set_ios(mmc, &mmc->ios); } else { sdhci_init(host, (mmc->pm_flags & MMC_PM_KEEP_POWER)); From bc95fa240a1b8ae64d3dabe87cbe103b912afc45 Mon Sep 17 00:00:00 2001 From: Hans Holmberg Date: Mon, 18 May 2026 08:52:24 +0200 Subject: [PATCH 078/329] xfs: handle racing deletions in xfs_zone_gc_iter_irec Under heavy garbage collection pressure from RocksDB workloads, filesystem shutdowns can occur in xfs_zone_gc_iter_irec when xfs_iget() returns -EINVAL for deleted files. Fix this by handling -EINVAL just like we handle -ENOENT, allowing zone GC to safely ignore stale mappings. Fixes: 080d01c41d44 ("xfs: implement zoned garbage collection") Signed-off-by: Hans Holmberg Reviewed-by: Christoph Hellwig Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_zone_gc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/xfs/xfs_zone_gc.c b/fs/xfs/xfs_zone_gc.c index c8a1d5c0332c..f03211e4354a 100644 --- a/fs/xfs/xfs_zone_gc.c +++ b/fs/xfs/xfs_zone_gc.c @@ -400,7 +400,7 @@ xfs_zone_gc_iter_irec( /* * If the inode was already deleted, skip over it. */ - if (error == -ENOENT) { + if (error == -ENOENT || error == -EINVAL) { iter->rec_idx++; goto retry; } From 97bdbf2613eb89428b739ce09d1a6e1c8435a286 Mon Sep 17 00:00:00 2001 From: Dai Ngo Date: Tue, 19 May 2026 17:32:58 -0700 Subject: [PATCH 079/329] xfs: fix use of uninitialized imap in xfs_fs_map_blocks error path xfs_fs_map_blocks() acquires the data map lock and then calls xfs_bmapi_read(). If xfs_bmapi_read() fails, the function currently still falls through to xfs_bmbt_to_iomap(), which consumes an uninitialized imap record and may return invalid data to the caller. Fix this by releasing the data map lock and returning immediately when xfs_bmapi_read() reports an error. This prevents xfs_bmbt_to_iomap() from being called with an uninitialized xfs_bmbt_irec. Fixes: 527851124d10f ("xfs: implement pNFS export operations") Signed-off-by: Dai Ngo Reviewed-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_pnfs.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_pnfs.c b/fs/xfs/xfs_pnfs.c index 221e55887a2a..b792e066b403 100644 --- a/fs/xfs/xfs_pnfs.c +++ b/fs/xfs/xfs_pnfs.c @@ -174,12 +174,15 @@ xfs_fs_map_blocks( lock_flags = xfs_ilock_data_map_shared(ip); error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap, &nimaps, bmapi_flags); + if (error) { + xfs_iunlock(ip, lock_flags); + goto out_unlock; + } seq = xfs_iomap_inode_sequence(ip, 0); ASSERT(!nimaps || imap.br_startblock != DELAYSTARTBLOCK); - if (!error && write && - (!nimaps || imap.br_startblock == HOLESTARTBLOCK)) { + if (write && (!nimaps || imap.br_startblock == HOLESTARTBLOCK)) { if (offset + length > XFS_ISIZE(ip)) end_fsb = xfs_iomap_eof_align_last_fsb(ip, end_fsb); else if (nimaps && imap.br_startblock == HOLESTARTBLOCK) From 36ca6f11424a5b6d92b88df37c40bf2fe825d5a0 Mon Sep 17 00:00:00 2001 From: Dai Ngo Date: Tue, 19 May 2026 17:32:59 -0700 Subject: [PATCH 080/329] xfs: fix overlapping extents returned for pNFS LAYOUTGET xfs_fs_map_blocks() currently passes XFS_BMAPI_ENTIRE to xfs_bmapi_read(), which causes the bmap code to expand the mapping to cover the entire extent rather than the requested range. A single LAYOUTGET request from the client can cause the server to issue multiple calls to xfs_fs_map_blocks() for different offsets within the same extent. Because the use of XFS_BMAPI_ENTIRE flag, these calls can produce overlapping mappings. As a result, the LAYOUTGET reply sent to the NFS client may contain overlapping extents. This creates ambiguity in extent selection for a given file range, which can lead to incorrect device selection, inconsistent handling of datastate, and ultimately data corruption or protocol violations on the client side. Problem discovered with xfstest generic/075 test using NFSv4.2 mount with SCSI layout. Fix this by replacing the XFS_BMAPI_ENTIRE flag with '0' so that xfs_bmapi_read() returns only the mapping for the requested range. Fixes: cc6c40e09d7b1 ("NFSD/blocklayout: Support multiple extents per LAYOUTGET"). Signed-off-by: Dai Ngo Reviewed-by: Carlos Maiolino Reviewed-by: Christoph Hellwig Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_pnfs.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_pnfs.c b/fs/xfs/xfs_pnfs.c index b792e066b403..d92993367ab6 100644 --- a/fs/xfs/xfs_pnfs.c +++ b/fs/xfs/xfs_pnfs.c @@ -118,7 +118,6 @@ xfs_fs_map_blocks( struct xfs_bmbt_irec imap; xfs_fileoff_t offset_fsb, end_fsb; loff_t limit; - int bmapi_flags = XFS_BMAPI_ENTIRE; int nimaps = 1; uint lock_flags; int error = 0; @@ -172,8 +171,9 @@ xfs_fs_map_blocks( offset_fsb = XFS_B_TO_FSBT(mp, offset); lock_flags = xfs_ilock_data_map_shared(ip); + /* request mappings for the specified range only */ error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, - &imap, &nimaps, bmapi_flags); + &imap, &nimaps, 0); if (error) { xfs_iunlock(ip, lock_flags); goto out_unlock; From fcf4faba9f986b3bb528da11913c9ec5d6e8f689 Mon Sep 17 00:00:00 2001 From: Yingjie Gao Date: Wed, 27 May 2026 12:31:33 +0800 Subject: [PATCH 081/329] xfs: fix error returns in CoW fork repair xrep_cow_find_bad() returns success after the cleanup labels even if AG setup, btree queries, or bitmap updates failed. This can make repair continue with an incomplete bad-file-offset bitmap instead of stopping at the original error. The force-rebuild path has a related cleanup problem. If xrep_cow_mark_file_range() fails, the function returns directly and skips the scrub AG context and perag cleanup. Let the force-rebuild path fall through to the existing cleanup code and return the saved error after cleanup. Fixes: dbbdbd008632 ("xfs: repair problems in CoW forks") Cc: # v6.8 Signed-off-by: Yingjie Gao Reviewed-by: "Darrick J. Wong" Signed-off-by: Carlos Maiolino --- fs/xfs/scrub/cow_repair.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/fs/xfs/scrub/cow_repair.c b/fs/xfs/scrub/cow_repair.c index bffc4666ce60..a6ff09ace43d 100644 --- a/fs/xfs/scrub/cow_repair.c +++ b/fs/xfs/scrub/cow_repair.c @@ -300,18 +300,15 @@ xrep_cow_find_bad( * on the debugging knob, replace everything in the CoW fork. */ if ((sc->sm->sm_flags & XFS_SCRUB_IFLAG_FORCE_REBUILD) || - XFS_TEST_ERROR(sc->mp, XFS_ERRTAG_FORCE_SCRUB_REPAIR)) { + XFS_TEST_ERROR(sc->mp, XFS_ERRTAG_FORCE_SCRUB_REPAIR)) error = xrep_cow_mark_file_range(xc, xc->irec.br_startblock, xc->irec.br_blockcount); - if (error) - return error; - } out_sa: xchk_ag_free(sc, &sc->sa); out_pag: xfs_perag_put(pag); - return 0; + return error; } /* From c3e073894379532c00cca7ba5762e18fafe29da8 Mon Sep 17 00:00:00 2001 From: Yingjie Gao Date: Wed, 27 May 2026 12:31:34 +0800 Subject: [PATCH 082/329] xfs: fix rtgroup cleanup in CoW fork repair xrep_cow_find_bad_rt() initializes scrub rtgroup state before the force-rebuild path calls xrep_cow_mark_file_range(). If that call fails, the code jumps directly to out_rtg, which skips the scrub rtgroup cleanup and only drops the local rtgroup reference. Remove the unnecessary jump so the function falls through to out_sr, ensuring the realtime cursors, lock state, and sr->rtg reference are released before returning. Fixes: fd97fe111208 ("xfs: fix CoW forks for realtime files") Cc: # v6.14 Signed-off-by: Yingjie Gao Reviewed-by: "Darrick J. Wong" Signed-off-by: Carlos Maiolino --- fs/xfs/scrub/cow_repair.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/fs/xfs/scrub/cow_repair.c b/fs/xfs/scrub/cow_repair.c index a6ff09ace43d..c25716fc4fee 100644 --- a/fs/xfs/scrub/cow_repair.c +++ b/fs/xfs/scrub/cow_repair.c @@ -382,12 +382,9 @@ xrep_cow_find_bad_rt( * CoW fork and then scan for staging extents in the refcountbt. */ if ((sc->sm->sm_flags & XFS_SCRUB_IFLAG_FORCE_REBUILD) || - XFS_TEST_ERROR(sc->mp, XFS_ERRTAG_FORCE_SCRUB_REPAIR)) { + XFS_TEST_ERROR(sc->mp, XFS_ERRTAG_FORCE_SCRUB_REPAIR)) error = xrep_cow_mark_file_range(xc, xc->irec.br_startblock, xc->irec.br_blockcount); - if (error) - goto out_rtg; - } out_sr: xchk_rtgroup_btcur_free(&sc->sr); From c316bb56bf4dda9bd3456a163c3a3856ca1b9498 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 27 May 2026 14:39:28 +0200 Subject: [PATCH 083/329] xfs: drop the RTG reference later in xfs_ioc_rtgroup_geometry Keep the rtgroup reference until after reporting the write pointer, as that uses it. Right now this is not a major issue as we don't support shrinking file systems in a way that makes RTGs go away, but let's stick to the proper reference counting to prepare for that. Fixes: c6ce65cb17aa ("xfs: add write pointer to xfs_rtgroup_geometry") Signed-off-by: Christoph Hellwig Reviewed-by: Carlos Maiolino Reviewed-by: Wilfred Mallawa Reviewed-by: Damien Le Moal Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_ioctl.c | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 46e234863644..737afcb5652a 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -433,9 +433,8 @@ xfs_ioc_rtgroup_geometry( return -EINVAL; error = xfs_rtgroup_get_geometry(rtg, &rgeo); - xfs_rtgroup_put(rtg); if (error) - return error; + goto out_put_rtg; if (xfs_has_zoned(mp)) { xfs_rtgroup_lock(rtg, XFS_RTGLOCK_RMAP); @@ -453,8 +452,10 @@ xfs_ioc_rtgroup_geometry( } if (copy_to_user(arg, &rgeo, sizeof(rgeo))) - return -EFAULT; - return 0; + error = -EFAULT; +out_put_rtg: + xfs_rtgroup_put(rtg); + return error; } /* From 2fa29c1a167739fdef0fda9742ddeb15d9a6c7e4 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 27 May 2026 14:39:29 +0200 Subject: [PATCH 084/329] xfs: factor rtgroup geom write pointer reporting into a helper Sticks out a bit better if we add a separate helper for it. Signed-off-by: Christoph Hellwig Reviewed-by: Carlos Maiolino Reviewed-by: Wilfred Mallawa Reviewed-by: Damien Le Moal Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_ioctl.c | 38 ++++++++++++++++++++++---------------- 1 file changed, 22 insertions(+), 16 deletions(-) diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 737afcb5652a..96af6b62ce39 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -409,6 +409,26 @@ xfs_ioc_ag_geometry( return 0; } +static void +xfs_rtgroup_report_write_pointer( + struct xfs_rtgroup *rtg, + struct xfs_rtgroup_geometry *rgeo) +{ + xfs_rtgroup_lock(rtg, XFS_RTGLOCK_RMAP); + if (rtg->rtg_open_zone) { + rgeo->rg_writepointer = rtg->rtg_open_zone->oz_allocated; + } else { + xfs_rgblock_t highest_rgbno = xfs_rtrmap_highest_rgbno(rtg); + + if (highest_rgbno == NULLRGBLOCK) + rgeo->rg_writepointer = 0; + else + rgeo->rg_writepointer = highest_rgbno + 1; + } + xfs_rtgroup_unlock(rtg, XFS_RTGLOCK_RMAP); + rgeo->rg_flags |= XFS_RTGROUP_GEOM_WRITEPOINTER; +} + STATIC int xfs_ioc_rtgroup_geometry( struct xfs_mount *mp, @@ -416,7 +436,6 @@ xfs_ioc_rtgroup_geometry( { struct xfs_rtgroup *rtg; struct xfs_rtgroup_geometry rgeo; - xfs_rgblock_t highest_rgbno; int error; if (copy_from_user(&rgeo, arg, sizeof(rgeo))) @@ -435,21 +454,8 @@ xfs_ioc_rtgroup_geometry( error = xfs_rtgroup_get_geometry(rtg, &rgeo); if (error) goto out_put_rtg; - - if (xfs_has_zoned(mp)) { - xfs_rtgroup_lock(rtg, XFS_RTGLOCK_RMAP); - if (rtg->rtg_open_zone) { - rgeo.rg_writepointer = rtg->rtg_open_zone->oz_allocated; - } else { - highest_rgbno = xfs_rtrmap_highest_rgbno(rtg); - if (highest_rgbno == NULLRGBLOCK) - rgeo.rg_writepointer = 0; - else - rgeo.rg_writepointer = highest_rgbno + 1; - } - xfs_rtgroup_unlock(rtg, XFS_RTGLOCK_RMAP); - rgeo.rg_flags |= XFS_RTGROUP_GEOM_WRITEPOINTER; - } + if (xfs_has_zoned(mp)) + xfs_rtgroup_report_write_pointer(rtg, &rgeo); if (copy_to_user(arg, &rgeo, sizeof(rgeo))) error = -EFAULT; From 8070a50b88971f9fada55a3ae3330f529bef9adc Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Wed, 27 May 2026 14:32:37 +0200 Subject: [PATCH 085/329] xfs: abort mount if xfs_fs_reserve_ag_blocks fails xfs_mountfs currently ignores all errors from xfs_fs_reserve_ag_blocks, which can lead to the mount path continuing on corruption errors. Fix the check to only ignore -ENOSPC as in other callers, and unwind for all other errors. Fixes: 81ed94751b15 ("xfs: fix log intent recovery ENOSPC shutdowns when inactivating inodes") Signed-off-by: Christoph Hellwig Reviewed-by: "Darrick J. Wong" Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_mount.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index b24195f570cd..7aa51826b1ca 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -1149,9 +1149,12 @@ xfs_mountfs( * blocks. */ error = xfs_fs_reserve_ag_blocks(mp); - if (error && error == -ENOSPC) + if (error) { + if (error != -ENOSPC) + goto out_rtunmount; xfs_warn(mp, - "ENOSPC reserving per-AG metadata pool, log recovery may fail."); +"ENOSPC reserving per-AG metadata pool, log recovery may fail."); + } error = xfs_log_mount_finish(mp); xfs_fs_unreserve_ag_blocks(mp); if (error) { From dabfaca8140f64535a79020f0f86ea56f5db5bb2 Mon Sep 17 00:00:00 2001 From: "Matthew Wilcox (Oracle)" Date: Tue, 26 May 2026 20:53:59 +0100 Subject: [PATCH 086/329] xfs: Remove mention of PageWriteback Update a comment to refer to folios instead of pages. Signed-off-by: Matthew Wilcox (Oracle) Reviewed-by: Christoph Hellwig Signed-off-by: Carlos Maiolino --- fs/xfs/xfs_reflink.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index db23a0f231d6..251dec48f0e3 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -949,16 +949,16 @@ xfs_reflink_end_cow( * repeatedly cycles the ILOCK to allocate one transaction per remapped * extent. * - * If we're being called by writeback then the pages will still - * have PageWriteback set, which prevents races with reflink remapping - * and truncate. Reflink remapping prevents races with writeback by - * taking the iolock and mmaplock before flushing the pages and - * remapping, which means there won't be any further writeback or page - * cache dirtying until the reflink completes. + * If we're being called by writeback then the folios will still + * have the writeback flag set, which prevents races with reflink + * remapping and truncate. Reflink remapping prevents races with + * writeback by taking the iolock and mmaplock before flushing + * the folios and remapping, which means there won't be any further + * writeback or page cache dirtying until the reflink completes. * * We should never have two threads issuing writeback for the same file * region. There are also have post-eof checks in the writeback - * preparation code so that we don't bother writing out pages that are + * preparation code so that we don't bother writing out folios that are * about to be truncated. * * If we're being called as part of directio write completion, the dio From ac35b5580ace12e5d0a0b5e61e36d2c4e1ffa29c Mon Sep 17 00:00:00 2001 From: Alice Ryhl Date: Wed, 27 May 2026 18:18:07 +0000 Subject: [PATCH 087/329] rust: arm64: set uwtable llvm module flag for CONFIG_UNWIND_TABLES Due to a rustc bug [1] the -Cforce-unwind-tables=y flag only emits the uwtable annotation for functions, but not for the module. This means that compiler-generated functions such as 'asan.module_ctor' do not receive the uwtable annotation. When CONFIG_UNWIND_PATCH_PAC_INTO_SCS is enabled, this leads to boot failures because the dwarf information emitted for the kasan constructors is wrong, which causes the SCS boot patching code to patch the constructor in an illegal manner. Specifically, the paciasp instruction is patched, but the autiasp instruction is not. This mismatch leads to a crash when the constructor is called during boot. ================================================================== BUG: KASAN: global-out-of-bounds in do_basic_setup+0x4c/0x90 Read of size 8 at addr ffffffe3cc7eb488 by task swapper/0/1 Specifically the faulting instruction is the (*fn)() to invoke the constructor in do_ctors() of the init/main.c file. Once the fix lands in rustc, this flag can be made conditional on the rustc version. Note that passing the flag on a rustc with the fix present has no effect. [ The fix [1] has landed for Rust 1.98.0 (expected release on 2026-08-20). Thus add a version check as discussed. - Miguel ] Fixes: d077242d68a3 ("rust: support for shadow call stack sanitizer") Cc: stable@kernel.org Link: https://github.com/rust-lang/rust/pull/156973 [1] Reported-by: Bo Ye Debugged-by: Isaac Manjarres Debugged-by: Sami Tolvanen Tested-by: Isaac Manjarres Signed-off-by: Alice Ryhl Link: https://patch.msgid.link/20260527-uwtable-module-flag-v1-1-caa41342be4b@google.com [ Adjusted link and comment. - Miguel ] Signed-off-by: Miguel Ojeda --- arch/arm64/Makefile | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile index 73a10f65ce8b..6b005c8fef70 100644 --- a/arch/arm64/Makefile +++ b/arch/arm64/Makefile @@ -63,6 +63,9 @@ else KBUILD_CFLAGS += -fasynchronous-unwind-tables KBUILD_AFLAGS += -fasynchronous-unwind-tables KBUILD_RUSTFLAGS += -Cforce-unwind-tables=y -Zuse-sync-unwind=n +# Work around rustc bug on compilers without +# https://github.com/rust-lang/rust/pull/156973. +KBUILD_RUSTFLAGS += $(if $(call rustc-min-version,109800),,-Zllvm_module_flag=uwtable:u32:2:max) endif ifeq ($(CONFIG_STACKPROTECTOR_PER_TASK),y) From 7c6535c37dbc03c1c35926b7420d66fb122b513a Mon Sep 17 00:00:00 2001 From: Marco Elver Date: Thu, 21 May 2026 14:27:23 +0200 Subject: [PATCH 088/329] s390: Implement _THIS_IP_ using inline asm Both GCC [1] and Clang [2] consider the generic version of _THIS_IP_ to be broken: #define _THIS_IP_ ({ __label__ __here; __here: (unsigned long)&&__here; }) In particular, the address of a label is only expected to be used with a computed goto. While the generic version more or less works today, it is known to be brittle and may break with current and future optimizations. For example, Clang -O2 always returns 1 when this function is inlined: static inline unsigned long get_ip(void) { return ({ __label__ __here; __here: (unsigned long)&&__here; }); } Fix it by overriding _THIS_IP_ in (which is included by ) using an architecture-specific inline asm version. Additionally, avoiding taking the address of a label prevents compilers from emitting spurious indirect branch targets (e.g. ENDBR or BTI) under control-flow integrity schemes. Link: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120071 [1] Link: https://github.com/llvm/llvm-project/issues/138272 [2] Signed-off-by: Marco Elver Reviewed-by: Heiko Carstens Signed-off-by: Alexander Gordeev --- arch/s390/include/asm/linkage.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/s390/include/asm/linkage.h b/arch/s390/include/asm/linkage.h index df3fb7d8227b..1b3ac553a642 100644 --- a/arch/s390/include/asm/linkage.h +++ b/arch/s390/include/asm/linkage.h @@ -7,4 +7,6 @@ #define __ALIGN .balign CONFIG_FUNCTION_ALIGNMENT, 0x07 #define __ALIGN_STR __stringify(__ALIGN) +#define _THIS_IP_ ({ unsigned long __ip; asm volatile("larl %0, ." : "=d" (__ip)); __ip; }) + #endif From 905b06d32a52afe32fcf5f30cf298c9ea6359f11 Mon Sep 17 00:00:00 2001 From: Miguel Ojeda Date: Sat, 30 May 2026 13:49:25 +0200 Subject: [PATCH 089/329] rust: x86: support Rust >= 1.98.0 target spec Starting with Rust 1.98.0 (expected 2026-08-20), the target spec will not support `x86-softfloat` anymore [1]. Instead, `softfloat` should be used, which is an alias. Otherwise, one gets: error: error loading target specification: rustc-abi: invalid rustc abi: 'x86-softfloat'. allowed values: 'x86-sse2', 'softfloat' at line 3 column 32 | = help: run `rustc --print target-list` for a list of built-in targets Thus conditionally use one or the other depending on the version. The alias has existed since Rust 1.95.0 (released 2026-04-16) [2], but use the newer version instead to avoid changing how the build works for existing compilers, at least until more testing takes place. Cc: Ralf Jung Cc: stable@vger.kernel.org # Needed in 6.12.y and later (Rust is pinned in older LTSs). Link: https://github.com/rust-lang/rust/pull/157151 [1] Link: https://github.com/rust-lang/rust/pull/151154 [2] Reviewed-by: Alice Ryhl Link: https://patch.msgid.link/20260530114925.260754-1-ojeda@kernel.org Signed-off-by: Miguel Ojeda --- scripts/generate_rust_target.rs | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/scripts/generate_rust_target.rs b/scripts/generate_rust_target.rs index 38b3416bb979..16f7e855e012 100644 --- a/scripts/generate_rust_target.rs +++ b/scripts/generate_rust_target.rs @@ -196,7 +196,9 @@ fn main() { } } else if cfg.has("X86_64") { ts.push("arch", "x86_64"); - if cfg.rustc_version_atleast(1, 86, 0) { + if cfg.rustc_version_atleast(1, 98, 0) { + ts.push("rustc-abi", "softfloat"); + } else if cfg.rustc_version_atleast(1, 86, 0) { ts.push("rustc-abi", "x86-softfloat"); } ts.push( @@ -236,7 +238,9 @@ fn main() { panic!("32-bit x86 only works under UML"); } ts.push("arch", "x86"); - if cfg.rustc_version_atleast(1, 86, 0) { + if cfg.rustc_version_atleast(1, 98, 0) { + ts.push("rustc-abi", "softfloat"); + } else if cfg.rustc_version_atleast(1, 86, 0) { ts.push("rustc-abi", "x86-softfloat"); } ts.push( From 85e0f27dd1396307913ffc5745b0c05137e9beac Mon Sep 17 00:00:00 2001 From: "Masami Hiramatsu (Google)" Date: Mon, 25 May 2026 11:21:14 +0900 Subject: [PATCH 090/329] tracing/probes: Point the error offset correctly for eprobe argument error Fix to point the error offset correctly for eprobe argument error. In the cleanup commit 1b8b0cd754cd ("tracing/probes: Move event parameter fetching code to common parser"), due to incorrect backward compatibility aimed at conforming to the test specifications, the error location was set to 0 when a non-existent formal parameter was specified for Eprobe. However, this should be corrected in both the test and the implementation to point correct error position. Link: https://lore.kernel.org/all/177967567399.209006.1451571244515632097.stgit@devnote2/ Fixes: 1b8b0cd754cd ("tracing/probes: Move event parameter fetching code to common parser") Cc: stable@vger.kernel.org Signed-off-by: Masami Hiramatsu (Google) Reviewed-by: Steven Rostedt --- kernel/trace/trace_probe.c | 2 -- .../selftests/ftrace/test.d/dynevent/eprobes_syntax_errors.tc | 2 +- 2 files changed, 1 insertion(+), 3 deletions(-) diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index e0d3a0da26af..44c22d4e7881 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -962,8 +962,6 @@ static int parse_probe_vars(char *orig_arg, const struct fetch_type *t, code->op = FETCH_OP_COMM; return 0; } - /* backward compatibility */ - ctx->offset = 0; goto inval; } diff --git a/tools/testing/selftests/ftrace/test.d/dynevent/eprobes_syntax_errors.tc b/tools/testing/selftests/ftrace/test.d/dynevent/eprobes_syntax_errors.tc index 4f5e8c665156..2a680c086047 100644 --- a/tools/testing/selftests/ftrace/test.d/dynevent/eprobes_syntax_errors.tc +++ b/tools/testing/selftests/ftrace/test.d/dynevent/eprobes_syntax_errors.tc @@ -20,7 +20,7 @@ check_error 'e:foo/^123456789012345678901234567890123456789012345678901234567890 check_error 'e:foo/^bar.1 syscalls/sys_enter_openat' # BAD_EVENT_NAME check_error 'e:foo/bar syscalls/sys_enter_openat arg=^dfd' # BAD_FETCH_ARG -check_error 'e:foo/bar syscalls/sys_enter_openat ^arg=$foo' # BAD_ATTACH_ARG +check_error 'e:foo/bar syscalls/sys_enter_openat arg=^$foo' # BAD_ATTACH_ARG if grep -q '\..*\[if \]' README; then check_error 'e:foo/bar syscalls/sys_enter_openat if ^' # NO_EP_FILTER From 0eaa1f245ac03ed0c6394159360532726f666811 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Fri, 29 May 2026 08:57:05 +0300 Subject: [PATCH 091/329] wifi: iwlwifi: mvm: don't support the reset handshake for old firmwares -77.ucode doesn't contain the fixes for this flow it seems. Don't use the firmware reset handshake even if the firmware claims support for it. Fixes: 906d4eb84408 ("iwlwifi: support firmware reset handshake") Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220600 Signed-off-by: Emmanuel Grumbach Reviewed-by: Johannes Berg Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20260529085453.9307b81d9b02.I21bba9e649f4cd0e35d3ea6cd97a03258be5832f@changeid --- drivers/net/wireless/intel/iwlwifi/mvm/ops.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c index ae177477b201..384bed95835d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c @@ -1416,6 +1416,12 @@ iwl_op_mode_mvm_start(struct iwl_trans *trans, const struct iwl_rf_cfg *cfg, fw_has_capa(&mvm->fw->ucode_capa, IWL_UCODE_TLV_CAPA_FW_RESET_HANDSHAKE); + /* Those firmware versions claim to support the fw_reset_handshake + * but they are buggy. + */ + if (IWL_UCODE_MAJOR(mvm->fw->ucode_ver) <= 77) + trans->conf.fw_reset_handshake = false; + trans->conf.queue_alloc_cmd_ver = iwl_fw_lookup_cmd_ver(mvm->fw, WIDE_ID(DATA_PATH_GROUP, From 9bf1b409afc7c4a1f0340f3975846c4f3278643a Mon Sep 17 00:00:00 2001 From: Pagadala Yesu Anjaneyulu Date: Fri, 29 May 2026 08:57:06 +0300 Subject: [PATCH 092/329] wifi: iwlwifi: mld: send tx power constraints before link activation TX power constraints must be sent to the firmware before link activation. If not, the firmware will use default power values. Fix this by moving the iwl_mld_send_ap_tx_power_constraint_cmd() call from iwl_mld_start_ap_ibss() to iwl_mld_assign_vif_chanctx(), before iwl_mld_activate_link() for AP interfaces. Also update the guard in the function to allow it to run before link activation for AP interfaces. Signed-off-by: Pagadala Yesu Anjaneyulu Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20260529085453.06c94b01efd2.Id43bdfe5eb030061c23348779687ba71b5f58182@changeid --- drivers/net/wireless/intel/iwlwifi/mld/ap.c | 4 ---- drivers/net/wireless/intel/iwlwifi/mld/mac80211.c | 7 +++++++ drivers/net/wireless/intel/iwlwifi/mld/power.c | 2 +- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mld/ap.c b/drivers/net/wireless/intel/iwlwifi/mld/ap.c index 5c59acc8c4c5..6598d9333333 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/ap.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/ap.c @@ -9,7 +9,6 @@ #include "ap.h" #include "hcmd.h" #include "tx.h" -#include "power.h" #include "key.h" #include "phy.h" #include "iwl-utils.h" @@ -273,9 +272,6 @@ int iwl_mld_start_ap_ibss(struct ieee80211_hw *hw, struct ieee80211_chanctx_conf *ctx; int ret; - if (vif->type == NL80211_IFTYPE_AP) - iwl_mld_send_ap_tx_power_constraint_cmd(mld, vif, link); - ret = iwl_mld_update_beacon_template(mld, vif, link); if (ret) return ret; diff --git a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c index da6fd7471568..3c8daddc0bcb 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/mac80211.c @@ -1150,6 +1150,13 @@ int iwl_mld_assign_vif_chanctx(struct ieee80211_hw *hw, if (iwl_mld_can_activate_link(mld, vif, link)) { iwl_mld_tlc_update_phy(mld, vif, link); + /* FW requires AP_TX_POWER_CONSTRAINTS_CMD before link + * activation for AP and after link activation for STA, + * for an unknown reason. + */ + if (vif->type == NL80211_IFTYPE_AP) + iwl_mld_send_ap_tx_power_constraint_cmd(mld, vif, link); + ret = iwl_mld_activate_link(mld, link); if (ret) goto err; diff --git a/drivers/net/wireless/intel/iwlwifi/mld/power.c b/drivers/net/wireless/intel/iwlwifi/mld/power.c index 49b0d9f8f865..266fe16bb95d 100644 --- a/drivers/net/wireless/intel/iwlwifi/mld/power.c +++ b/drivers/net/wireless/intel/iwlwifi/mld/power.c @@ -366,7 +366,7 @@ iwl_mld_send_ap_tx_power_constraint_cmd(struct iwl_mld *mld, lockdep_assert_wiphy(mld->wiphy); - if (!mld_link->active) + if (!mld_link->active && vif->type != NL80211_IFTYPE_AP) return; if (link->chanreq.oper.chan->band != NL80211_BAND_6GHZ) From e0c121d545134af886b28c4c26d91abf5dd39c17 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Fri, 29 May 2026 08:57:07 +0300 Subject: [PATCH 093/329] wifi: iwlwifi: mvm: avoid oversized UATS command copy MCC_ALLOWED_AP_TYPE_CMD exceeds the fixed copied host-command buffer and triggers warnings in the gen2 enqueue path when command 0xc05 is sent. Use IWL_HCMD_DFL_NOCOPY as it was done before the offending commit. Fixes: 078df640ef05 ("wifi: iwlwifi: mld: add support for iwl_mcc_allowed_ap_type_cmd v2") Signed-off-by: Emmanuel Grumbach Signed-off-by: Miri Korenblit Link: https://patch.msgid.link/20260529085453.9af349ab459b.I348df3980764c15efce0099a35fe8a88fb2a6ee2@changeid --- drivers/net/wireless/intel/iwlwifi/mvm/fw.c | 28 ++++++++++++++++----- 1 file changed, 22 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c index f05df3a3300e..6e507d6dcdd2 100644 --- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c +++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c @@ -1,6 +1,6 @@ // SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause /* - * Copyright (C) 2012-2014, 2018-2025 Intel Corporation + * Copyright (C) 2012-2014, 2018-2026 Intel Corporation * Copyright (C) 2013-2015 Intel Mobile Communications GmbH * Copyright (C) 2016-2017 Intel Deutschland GmbH */ @@ -459,9 +459,14 @@ static void iwl_mvm_phy_filter_init(struct iwl_mvm *mvm, static void iwl_mvm_uats_init(struct iwl_mvm *mvm) { + struct iwl_mcc_allowed_ap_type_cmd_v1 *cmd __free(kfree) = NULL; int cmd_id = WIDE_ID(REGULATORY_AND_NVM_GROUP, MCC_ALLOWED_AP_TYPE_CMD); - struct iwl_mcc_allowed_ap_type_cmd_v1 cmd = {}; + struct iwl_host_cmd hcmd = { + .id = cmd_id, + .len[0] = sizeof(*cmd), + .dataflags[0] = IWL_HCMD_DFL_NOCOPY, + }; u8 cmd_ver; int ret; @@ -485,14 +490,25 @@ static void iwl_mvm_uats_init(struct iwl_mvm *mvm) if (!mvm->fwrt.ap_type_cmd_valid) return; - BUILD_BUG_ON(sizeof(mvm->fwrt.ap_type_cmd.mcc_to_ap_type_map) != - sizeof(cmd.mcc_to_ap_type_map)); + /* Since we free the command immediately after iwl_mvm_send_cmd, we + * must send this command in SYNC mode. + */ + lockdep_assert_held(&mvm->mutex); - memcpy(cmd.mcc_to_ap_type_map, + cmd = kzalloc_obj(*cmd); + if (!cmd) + return; + + BUILD_BUG_ON(sizeof(mvm->fwrt.ap_type_cmd.mcc_to_ap_type_map) != + sizeof(cmd->mcc_to_ap_type_map)); + + memcpy(cmd->mcc_to_ap_type_map, mvm->fwrt.ap_type_cmd.mcc_to_ap_type_map, sizeof(mvm->fwrt.ap_type_cmd.mcc_to_ap_type_map)); - ret = iwl_mvm_send_cmd_pdu(mvm, cmd_id, 0, sizeof(cmd), &cmd); + hcmd.data[0] = cmd; + + ret = iwl_mvm_send_cmd(mvm, &hcmd); if (ret < 0) IWL_ERR(mvm, "failed to send MCC_ALLOWED_AP_TYPE_CMD (%d)\n", ret); From 093305d801fae6ff9b8bb531fd78b579794c4f80 Mon Sep 17 00:00:00 2001 From: Emmanuel Grumbach Date: Sun, 31 May 2026 13:30:19 +0300 Subject: [PATCH 094/329] wifi: iwlwifi: pcie: simplify the resume flow if fast resume is not used In most distributions, NetworkManager shuts the device down before entering system suspend, so fast suspend is typically not used. On older devices, resume currently tries to grab NIC access to infer whether the device was powered off while suspended. That probe is only meaningful for the fast-suspend path where the device is expected to remain alive. Unfortunately, for unclear reasons, grabbing NIC access was harmful as reported in the bugzilla ticket below. Workaround this issue by simply not grabbing NIC access if fast suspend is not used. Cc: stable@vger.kernel.org Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221501 Assisted-by: GitHub Copilot:gpt-5.3-codex Signed-off-by: Emmanuel Grumbach Link: https://patch.msgid.link/20260531133005.e2ed9e0cd44f.If283625983a843933e0c01561a421daff184e9e9@changeid Signed-off-by: Miri Korenblit --- drivers/net/wireless/intel/iwlwifi/pcie/drv.c | 56 +++++++++++-------- 1 file changed, 32 insertions(+), 24 deletions(-) diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c index dc99e7ac4726..eb3c5a6dd088 100644 --- a/drivers/net/wireless/intel/iwlwifi/pcie/drv.c +++ b/drivers/net/wireless/intel/iwlwifi/pcie/drv.c @@ -1225,33 +1225,41 @@ static int _iwl_pci_resume(struct device *device, bool restore) if (!trans->op_mode) return 0; - /* - * Scratch value was altered, this means the device was powered off, we - * need to reset it completely. - * Note: MAC (bits 0:7) will be cleared upon suspend even with wowlan, - * but not bits [15:8]. So if we have bits set in lower word, assume - * the device is alive. - * Alternatively, if the scratch value is 0xFFFFFFFF, then we no longer - * have access to the device and consider it powered off. - * For older devices, just try silently to grab the NIC. - */ - if (trans->mac_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) { - u32 scratch = iwl_read32(trans, CSR_FUNC_SCRATCH); - - if (!(scratch & CSR_FUNC_SCRATCH_POWER_OFF_MASK) || - scratch == ~0U) - device_was_powered_off = true; - } else { + if (test_bit(STATUS_DEVICE_ENABLED, &trans->status)) { /* - * bh are re-enabled by iwl_trans_pcie_release_nic_access, - * so re-enable them if _iwl_trans_pcie_grab_nic_access fails. + * Scratch value was altered, this means the device was powered + * off, we need to reset it completely. + * Note: MAC (bits 0:7) will be cleared upon suspend even with + * wowlan, but not bits [15:8]. So if we have bits set in lower + * word, assume the device is alive. + * Alternatively, if the scratch value is 0xFFFFFFFF, then we + * no longer have access to the device and consider it powered + * off. + * For older devices, just try silently to grab the NIC. */ - local_bh_disable(); - if (_iwl_trans_pcie_grab_nic_access(trans, true)) { - iwl_trans_pcie_release_nic_access(trans); + if (trans->mac_cfg->device_family >= IWL_DEVICE_FAMILY_BZ) { + u32 scratch = iwl_read32(trans, CSR_FUNC_SCRATCH); + + if (!(scratch & CSR_FUNC_SCRATCH_POWER_OFF_MASK) || + scratch == ~0U) { + IWL_DEBUG_WOWLAN(trans, + "Scratch 0x%08x indicates device was powered off\n", + scratch); + device_was_powered_off = true; + } } else { - device_was_powered_off = true; - local_bh_enable(); + /* + * bh are re-enabled by iwl_trans_pcie_release_nic_access, + * so re-enable them if _iwl_trans_pcie_grab_nic_access + * fails. + */ + local_bh_disable(); + if (_iwl_trans_pcie_grab_nic_access(trans, true)) { + iwl_trans_pcie_release_nic_access(trans); + } else { + device_was_powered_off = true; + local_bh_enable(); + } } } From aa2f4addab44407c7aa742321de5dc1914ab5762 Mon Sep 17 00:00:00 2001 From: Rong Zhang Date: Sun, 31 May 2026 03:52:49 +0800 Subject: [PATCH 095/329] ALSA: usb-audio: Set the value of potential sticky mixers to maximum It makes no sense to restore the saved value for a sticky mixer, since setting any value is a no-op. However, in some rare cases, SET_CUR is effective despite GET_CUR always returns a constant value. These mixers are not sticky, but there's no way to distinguish them. Without any additional information, the best thing we can do is to set the mixer value to the maximum before bailing out, so that a soft mixer can still reach the maximum hardware volume if the mixer turns out to be non-sticky. Meanwhile, all channels must be synchronized to prevent imbalance volume. Fixes: 86aa1ea1f15c ("ALSA: usb-audio: Do not expose sticky mixers") Signed-off-by: Rong Zhang Link: https://patch.msgid.link/20260531-uac-sticky-error-path-v1-1-12c2329d17ef@rong.moe Signed-off-by: Takashi Iwai --- sound/usb/mixer.c | 33 +++++++++++++++++++++++++++++---- 1 file changed, 29 insertions(+), 4 deletions(-) diff --git a/sound/usb/mixer.c b/sound/usb/mixer.c index 5fba456eb4a9..fb37bb8ad9a9 100644 --- a/sound/usb/mixer.c +++ b/sound/usb/mixer.c @@ -1371,10 +1371,8 @@ static int get_min_max_with_quirks(struct usb_mixer_elem_info *cval, goto no_checks; ret = check_sticky_volume_control(cval, minchn, saved); - if (ret < 0) { - snd_usb_set_cur_mix_value(cval, minchn, 0, saved); - return ret; - } + if (ret < 0) + goto sticky; if (cval->min + cval->res < cval->max) check_volume_control_res(cval, minchn, saved); @@ -1431,6 +1429,33 @@ static int get_min_max_with_quirks(struct usb_mixer_elem_info *cval, } return 0; + +sticky: + /* + * It makes no sense to restore the saved value for a sticky mixer, + * since setting any value is a no-op. + * + * However, in some rare cases, SET_CUR is effective despite GET_CUR + * always returns a constant value. These mixers are not sticky, but + * there's no way to distinguish them. Without any additional + * information, the best thing we can do is to set the mixer value to + * the maximum before bailing out, so that a soft mixer can still reach + * the maximum hardware volume if the mixer turns out to be non-sticky. + * Meanwhile, all channels must be synchronized to prevent imbalance + * volume. + */ + if (!cval->cmask) { + snd_usb_set_cur_mix_value(cval, 0, 0, cval->max); + } else { + for (i = 0; i < MAX_CHANNELS; i++) { + idx = 0; + if (cval->cmask & BIT(i)) { + snd_usb_set_cur_mix_value(cval, i + 1, idx, cval->max); + idx++; + } + } + } + return ret; } #define get_min_max(cval, def) get_min_max_with_quirks(cval, def, NULL) From b003086d76968298f22e7cf62239833b5a3a06b1 Mon Sep 17 00:00:00 2001 From: Gil Portnoy Date: Thu, 28 May 2026 00:00:00 +0000 Subject: [PATCH 096/329] ksmbd: fix NULL-deref of opinfo->conn in oplock/lease break notifiers smb2_oplock_break_noti() and smb2_lease_break_noti() read opinfo->conn into a local with neither READ_ONCE() nor a NULL check. Both run from oplock_break() after opinfo_get_list() has dropped ci->m_lock, so a concurrent SMB2 LOGOFF (session_fd_check()) can set op->conn = NULL under ci->m_lock within that window. ksmbd_conn_r_count_inc(conn) then writes through NULL at offset 0xc4 -- a remotely triggerable oops. Guard both reads the way compare_guid_key() already does: read opinfo->conn with READ_ONCE() and return early if it is NULL, before allocating the work struct so nothing leaks. A NULL conn means the client is gone and the break is moot, so return 0; oplock_break() treats that as success and runs the normal teardown. Fixes: c8efcc786146 ("ksmbd: add support for durable handles v1/v2") Assisted-by: Henry (Claude):claude-opus-4 Signed-off-by: Gil Portnoy Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/oplock.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index 0f5c18520eff..b193dde4810d 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -711,11 +711,16 @@ static void __smb2_oplock_break_noti(struct work_struct *wk) */ static int smb2_oplock_break_noti(struct oplock_info *opinfo) { - struct ksmbd_conn *conn = opinfo->conn; + struct ksmbd_conn *conn; struct oplock_break_info *br_info; int ret = 0; - struct ksmbd_work *work = ksmbd_alloc_work_struct(); + struct ksmbd_work *work; + conn = READ_ONCE(opinfo->conn); + if (!conn) + return 0; + + work = ksmbd_alloc_work_struct(); if (!work) return -ENOMEM; @@ -815,11 +820,15 @@ static void __smb2_lease_break_noti(struct work_struct *wk) */ static int smb2_lease_break_noti(struct oplock_info *opinfo) { - struct ksmbd_conn *conn = opinfo->conn; + struct ksmbd_conn *conn; struct ksmbd_work *work; struct lease_break_info *br_info; struct lease *lease = opinfo->o_lease; + conn = READ_ONCE(opinfo->conn); + if (!conn) + return 0; + work = ksmbd_alloc_work_struct(); if (!work) return -ENOMEM; From 7ce4fc40018de07f05f3035241122d992610dbfb Mon Sep 17 00:00:00 2001 From: Gil Portnoy Date: Thu, 28 May 2026 00:00:00 +0000 Subject: [PATCH 097/329] ksmbd: fix durable reconnect double-bind race in ksmbd_reopen_durable_fd Two concurrent same-user DHnC reconnects can both observe fp->conn == NULL before either sets it. ksmbd_reopen_durable_fd() checks fp->conn to guard against a handle already being reconnected, but the check and the binding assignment are not atomic: both threads pass the guard, both call ksmbd_conn_get() on the same fp, and both eventually reach kfree(fp->owner.name) -- a double-free of the owner.name slab object. The double-bound ksmbd_file also causes a write-UAF on the 344-byte ksmbd_file_cache object when a concurrent smb2_close() spins on fp->f_lock after the object has been freed by the losing reconnect path. KASAN on 7.1-rc5 (48-thread concurrent reconnect, 3000 cycles): BUG: KASAN: double-free in ksmbd_reopen_durable_fd+0x268/0x308 BUG: KASAN: slab-use-after-free in _raw_spin_lock+0xac/0x150 Write of size 4 at offset 24 into freed ksmbd_file_cache object Five double-bind windows observed; 63 total KASAN reports triggered. Fix: validate and claim fp->conn under write_lock(&global_ft.lock) so the check-and-claim is atomic. ksmbd_lookup_durable_fd() already treats fp->conn != NULL as "in use" and skips such an fp; setting fp->conn before dropping the lock closes the race. ksmbd_conn_get() is a non-sleeping refcount increment, safe under the rwlock. The rollback path on __open_id() failure also clears fp->conn/tcon under the lock so concurrent readers see a consistent state. Fixes: b1f1e80620de ("ksmbd: centralize ksmbd_conn final release to plug transport leak") Assisted-by: Henry (Claude):claude-opus-4 Signed-off-by: Gil Portnoy Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/vfs_cache.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index 4d2d33df6231..ba3355a6057a 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -1390,19 +1390,19 @@ int ksmbd_reopen_durable_fd(struct ksmbd_work *work, struct ksmbd_file *fp) struct ksmbd_lock *smb_lock; unsigned int old_f_state; + write_lock(&global_ft.lock); if (!fp->is_durable || fp->conn || fp->tcon) { + write_unlock(&global_ft.lock); pr_err("Invalid durable fd [%p:%p]\n", fp->conn, fp->tcon); return -EBADF; } if (has_file_id(fp->volatile_id)) { + write_unlock(&global_ft.lock); pr_err("Still in use durable fd: %llu\n", fp->volatile_id); return -EBADF; } - old_f_state = fp->f_state; - fp->f_state = FP_NEW; - /* * Initialize fp's connection binding before publishing fp into the * session's file table. If __open_id() is ordered first, a @@ -1413,11 +1413,17 @@ int ksmbd_reopen_durable_fd(struct ksmbd_work *work, struct ksmbd_file *fp) */ fp->conn = ksmbd_conn_get(conn); fp->tcon = work->tcon; + write_unlock(&global_ft.lock); + + old_f_state = fp->f_state; + fp->f_state = FP_NEW; __open_id(&work->sess->file_table, fp, OPEN_ID_TYPE_VOLATILE_ID); if (!has_file_id(fp->volatile_id)) { + write_lock(&global_ft.lock); fp->conn = NULL; fp->tcon = NULL; + write_unlock(&global_ft.lock); ksmbd_conn_put(conn); fp->f_state = old_f_state; return -EBADF; From f580d27e8928828693df44ba2db0fffdbe11dfea Mon Sep 17 00:00:00 2001 From: Gil Portnoy Date: Mon, 1 Jun 2026 08:27:56 +0900 Subject: [PATCH 098/329] ksmbd: fix use-after-free of a deferred file_lock on double SMB2_CANCEL A deferred byte-range lock (an SMB2_LOCK that blocks) registers an async work on conn->async_requests via setup_async_work(), with cancel_fn = smb2_remove_blocked_lock and cancel_argv[0] pointing at the struct file_lock. When the request is cancelled, the worker frees the file_lock with locks_free_lock() and takes the cancelled early-exit, which "goto out"s and never reaches release_async_work() -- the only site that unlinks the work from conn->async_requests and clears cancel_fn/cancel_argv. The work therefore stays matchable on async_requests with a live cancel_fn pointing at the freed file_lock, until connection teardown finally runs release_async_work(). smb2_cancel() fires cancel_fn unconditionally with no state guard, so a second SMB2_CANCEL for the same AsyncId, arriving in that window, re-runs smb2_remove_blocked_lock() on the freed file_lock -- a slab use-after-free: BUG: KASAN: slab-use-after-free in __locks_delete_block __locks_delete_block locks_delete_block ksmbd_vfs_posix_lock_unblock smb2_remove_blocked_lock smb2_cancel <- 2nd SMB2_CANCEL fires cancel_fn handle_ksmbd_work Allocated by ...: locks_alloc_lock <- smb2_lock Freed by ...: locks_free_lock <- smb2_lock (cancelled branch) ... cache file_lock_cache of size 192 Reproduced on mainline with KASAN by an authenticated SMB client. Skip a work whose state is already KSMBD_WORK_CANCELLED so its cancel callback cannot be fired a second time. Cc: stable@vger.kernel.org Signed-off-by: Gil Portnoy Acked-by: Namjae Jeon Signed-off-by: Steve French --- fs/smb/server/smb2pdu.c | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/fs/smb/server/smb2pdu.c b/fs/smb/server/smb2pdu.c index 620bcfbbfd92..3eb3b1711acb 100644 --- a/fs/smb/server/smb2pdu.c +++ b/fs/smb/server/smb2pdu.c @@ -7322,6 +7322,17 @@ int smb2_cancel(struct ksmbd_work *work) le64_to_cpu(hdr->Id.AsyncId)) continue; + /* + * A cancelled deferred byte-range lock frees its + * file_lock and takes the smb2_lock() early-exit that + * skips release_async_work(), so the work stays on + * conn->async_requests with a live cancel_fn pointing + * at the freed file_lock. Re-firing it on a second + * SMB2_CANCEL is a use-after-free. + */ + if (iter->state == KSMBD_WORK_CANCELLED) + break; + ksmbd_debug(SMB, "smb2 with AsyncId %llu cancelled command = 0x%x\n", le64_to_cpu(hdr->Id.AsyncId), From d6b8b02a27b3dd09ec12144322b3dac46d9bc9ef Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Mon, 25 May 2026 09:22:55 -0700 Subject: [PATCH 099/329] mm/damon/ops-common: call folio_test_lru() after folio_get() damon_get_folio() speculatively calls folio_test_lru() before folio_try_get(). The folio can get freed and reallocated to a tail page. In the case, VM_BUG_ON_PGFLAGS() in const_folio_flags() can be triggered. Remove the speculative call. Also mark folio_test_lru() check right after folio_try_get() success as no more unlikely. The race should be rare. Also the problem can happen only if the kernel has enabled CONFIG_DEBUG_VM_PGFLAGS. No real world report of this issue has been made so far. This fix is based on only theoretical analysis. That said, a bug is a bug. A similar issue was also fixed via commit 3203b3ab0fcf ("mm/filemap: don't call folio_test_locked() without a reference in next_uptodate_folio()"). I don't expect this change will make a meaningful impact to DAMON performance in the real world, though I will be happy to be corrected from the real world reports. The issue was discovered [1] by Sashiko. Link: https://lore.kernel.org/20260525162256.8317-1-sj@kernel.org Link: https://lore.kernel.org/20260517234112.89245-1-sj@kernel.org [1] Fixes: 3f49584b262c ("mm/damon: implement primitives for the virtual memory address spaces") Signed-off-by: SeongJae Park Cc: Fernand Sieber Cc: Leonard Foerster Cc: Shakeel Butt Cc: # 5.15.x Signed-off-by: Andrew Morton --- mm/damon/ops-common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/mm/damon/ops-common.c b/mm/damon/ops-common.c index 8c6d613425c1..c3e4c871b0bb 100644 --- a/mm/damon/ops-common.c +++ b/mm/damon/ops-common.c @@ -32,9 +32,9 @@ struct folio *damon_get_folio(unsigned long pfn) return NULL; folio = page_folio(page); - if (!folio_test_lru(folio) || !folio_try_get(folio)) + if (!folio_try_get(folio)) return NULL; - if (unlikely(page_folio(page) != folio || !folio_test_lru(folio))) { + if (unlikely(page_folio(page) != folio) || !folio_test_lru(folio)) { folio_put(folio); folio = NULL; } From c7bde43f6daf70e05a64fbca7efdf6fa93e057dc Mon Sep 17 00:00:00 2001 From: Muchun Song Date: Mon, 25 May 2026 10:52:13 +0800 Subject: [PATCH 100/329] mm/hugetlb_vmemmap: fix incorrect vmemmap restore in rollback vmemmap_restore_pte() rebuilds restored vmemmap pages from a tail-page template derived from compound_head(). This is wrong when the current PTE already maps a page whose contents are not tail-page metadata. In the rollback path of vmemmap_remap_free(), the first restored PTE is backed by vmemmap_head and contains head-page metadata. Reconstructing that page from a tail-page template overwrites the head-page state and corrupts the restored vmemmap page. Fix this by copying the full page from the page currently mapped by the PTE. Also pass vmemmap_tail to the rollback walk so only PTEs backed by the shared tail page are restored, while the head PTE remains mapped to vmemmap_head. Add VM_WARN_ON_ONCE() checks for unexpected cases. Link: https://lore.kernel.org/20260525025213.2229628-1-songmuchun@bytedance.com Fixes: c0b495b91a47 ("mm/hugetlb: refactor code around vmemmap_walk") Signed-off-by: Muchun Song Acked-by: Kiryl Shutsemau Acked-by: Oscar Salvador (SUSE) Cc: David Hildenbrand Cc: Signed-off-by: Andrew Morton --- mm/hugetlb_vmemmap.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c index 4a077d231d3a..133b46dfb09f 100644 --- a/mm/hugetlb_vmemmap.c +++ b/mm/hugetlb_vmemmap.c @@ -207,6 +207,8 @@ static void vmemmap_remap_pte(pte_t *pte, unsigned long addr, /* Remapping the head page requires r/w */ if (unlikely(walk->nr_walked == 0 && walk->vmemmap_head)) { + VM_WARN_ON_ONCE(!PageHead((const struct page *)addr)); + list_del(&walk->vmemmap_head->lru); /* @@ -218,6 +220,8 @@ static void vmemmap_remap_pte(pte_t *pte, unsigned long addr, entry = mk_pte(walk->vmemmap_head, PAGE_KERNEL); } else { + VM_WARN_ON_ONCE(!PageTail((const struct page *)addr)); + /* * Remap the tail pages as read-only to catch illegal write * operation to the tail pages. @@ -232,33 +236,28 @@ static void vmemmap_remap_pte(pte_t *pte, unsigned long addr, static void vmemmap_restore_pte(pte_t *pte, unsigned long addr, struct vmemmap_remap_walk *walk) { - struct page *page; - struct page *from, *to; - - page = list_first_entry(walk->vmemmap_pages, struct page, lru); - list_del(&page->lru); + struct page *src = pte_page(ptep_get(pte)), *dst; /* - * Initialize tail pages in the newly allocated vmemmap page. - * - * There is folio-scope metadata that is encoded in the first few - * tail pages. - * - * Use the value last tail page in the page with the head page - * to initialize the rest of tail pages. + * When rolling back vmemmap_remap_free(), keep the copied head page + * mapping and restore only PTEs currently pointing at the shared tail + * page. */ - from = compound_head((struct page *)addr) + - PAGE_SIZE / sizeof(struct page) - 1; - to = page_to_virt(page); - for (int i = 0; i < PAGE_SIZE / sizeof(struct page); i++, to++) - *to = *from; + if (walk->vmemmap_tail && walk->vmemmap_tail != src) + return; + + VM_WARN_ON_ONCE(PageHead((const struct page *)addr)); + + dst = list_first_entry(walk->vmemmap_pages, struct page, lru); + list_del(&dst->lru); + copy_page(page_to_virt(dst), page_to_virt(src)); /* * Makes sure that preceding stores to the page contents become visible * before the set_pte_at() write. */ smp_wmb(); - set_pte_at(&init_mm, addr, pte, mk_pte(page, PAGE_KERNEL)); + set_pte_at(&init_mm, addr, pte, mk_pte(dst, PAGE_KERNEL)); } /** @@ -324,6 +323,7 @@ static int vmemmap_remap_free(unsigned long start, unsigned long end, */ walk = (struct vmemmap_remap_walk) { .remap_pte = vmemmap_restore_pte, + .vmemmap_tail = vmemmap_tail, .vmemmap_pages = vmemmap_pages, .flags = 0, }; From 40990c87a26e371594475acdc560c93cfae308a1 Mon Sep 17 00:00:00 2001 From: Yin Tirui Date: Tue, 26 May 2026 18:13:55 +0800 Subject: [PATCH 101/329] mm/huge_memory: update file PUD counter before folio_put() __split_huge_pud_locked() updates the file/shmem RSS counter after dropping the PUD mapping's folio reference. If folio_put() drops the last reference, mm_counter_file() can later read freed folio state via folio_test_swapbacked(). Move the counter update before folio_put(). Link: https://lore.kernel.org/20260526101355.1984244-1-yintirui@huawei.com Fixes: dbe54153296d ("mm/huge_memory: add vmf_insert_folio_pud()") Signed-off-by: Yin Tirui Reviewed-by: Lorenzo Stoakes Acked-by: David Hildenbrand (arm) Reviewed-by: Lance Yang Reviewed-by: Dev Jain Cc: Alistair Popple Cc: Baolin Wang Cc: Barry Song Cc: Chen Jun Cc: Kefeng Wang Cc: Liam R. Howlett Cc: Nico Pache Cc: Ryan Roberts Cc: Zi Yan Cc: Signed-off-by: Andrew Morton --- mm/huge_memory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 970e077019b7..1f78b73a0ca4 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3015,9 +3015,9 @@ static void __split_huge_pud_locked(struct vm_area_struct *vma, pud_t *pud, if (!folio_test_referenced(folio) && pud_young(old_pud)) folio_set_referenced(folio); folio_remove_rmap_pud(folio, page, vma); - folio_put(folio); add_mm_counter(vma->vm_mm, mm_counter_file(folio), -HPAGE_PUD_NR); + folio_put(folio); } void __split_huge_pud(struct vm_area_struct *vma, pud_t *pud, From 8d878059924f12c1bc24556a92ec56add74de3c8 Mon Sep 17 00:00:00 2001 From: Yin Tirui Date: Tue, 26 May 2026 18:13:37 +0800 Subject: [PATCH 102/329] mm/huge_memory: update file PMD counter before folio_put() __split_huge_pmd_locked() updates the file/shmem RSS counter after dropping the PMD mapping's folio reference. If folio_put() drops the last reference, mm_counter_file() can later read freed folio state via folio_test_swapbacked(). Move the counter update before folio_put(). Link: https://lore.kernel.org/20260526101337.1984081-1-yintirui@huawei.com Fixes: fadae2953072 ("thp: use mm_file_counter to determine update which rss counter") Signed-off-by: Yin Tirui Reviewed-by: Lorenzo Stoakes Acked-by: David Hildenbrand (arm) Reviewed-by: Lance Yang Reviewed-by: Dev Jain Cc: Baolin Wang Cc: Barry Song Cc: Chen Jun Cc: Kefeng Wang Cc: Liam R. Howlett Cc: Nico Pache Cc: Ryan Roberts Cc: Vlastimil Babka Cc: Yang Shi Cc: Zi Yan Cc: Signed-off-by: Andrew Morton --- mm/huge_memory.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 1f78b73a0ca4..653f2dc03403 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -3133,7 +3133,9 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd, if (!folio_test_referenced(folio) && pmd_young(old_pmd)) folio_set_referenced(folio); folio_remove_rmap_pmd(folio, page, vma); + add_mm_counter(mm, mm_counter_file(folio), -HPAGE_PMD_NR); folio_put(folio); + return; } add_mm_counter(mm, mm_counter_file(folio), -HPAGE_PMD_NR); return; From 85668fda932a5b8f15f649cf06411525a0e4c8ec Mon Sep 17 00:00:00 2001 From: "Mike Rapoport (Microsoft)" Date: Wed, 27 May 2026 21:47:49 +0300 Subject: [PATCH 103/329] userfaultfd: verify VMA state across UFFDIO_COPY retry Patch series "userfaultfd: verify VMA state across UFFDIO_COPY retry", v2. ... and two more small fixes. This patch (of 3): mfill_copy_folio_retry() drops the VMA lock for copy_from_user() and reacquires it afterwards. The destination VMA can be replaced during that window. The existing check compares vma_uffd_ops() before and after the retry, but if a shmem VMA with MAP_SHARED is replaced with a shmem VMA with MAP_PRIVATE (or vice versa) the replacement goes undetected. The change from MAP_PRIVATE to MAP_SHARED will treat the folio allocated with shmem_alloc_folio() as anonymous and this will cause BUG() when mfill_atomic_install_pte() will try to folio_add_new_anon_rmap(). The change from MAP_SHARED to MAP_PRIVATE allows injection of folios into the page cache of the original VMA. There is no need to change for hugetlb because it never uses mfill_copy_folio_retry(). Introduce helpers for more comprehensive comparison of VMA state: - mfill_retry_state_save() to save the relevant VMA state into a struct mfill_retry_state (original uffd_ops, relevant VMA flags, vm_file and pgoff) before dropping the lock - mfill_retry_state_changed() to compare the saved state with the state of the VMA acquired after retaking the locks - mfill_retry_state_put() to release vm_file pinning. Use DEFINE_FREE() cleanup to wrap mfill_retry_state_put() to avoid complicating error handling paths in mfill_copy_folio_retry(). Link: https://lore.kernel.org/20260527184751.4147364-1-rppt@kernel.org Link: https://lore.kernel.org/20260527184751.4147364-2-rppt@kernel.org Fixes: 292411fda25b ("mm/userfaultfd: detect VMA type change after copy retry in mfill_copy_folio_retry()") Fixes: 6ab703034f14 ("userfaultfd: mfill_atomic(): remove retry logic") Co-developed-by: Michael Bommarito Signed-off-by: Michael Bommarito Signed-off-by: Mike Rapoport (Microsoft) Suggested-by: Peter Xu Co-developed-by: David Carlier Signed-off-by: David Carlier Reviewed-by: Lorenzo Stoakes Cc: David Hildenbrand Cc: Liam R. Howlett Signed-off-by: Andrew Morton --- mm/userfaultfd.c | 85 +++++++++++++++++++++++++++++++++++++++++------- 1 file changed, 73 insertions(+), 12 deletions(-) diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index 180bad42fc79..e5d2fb3ce2c1 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -14,6 +14,8 @@ #include #include #include +#include +#include #include #include #include "internal.h" @@ -443,16 +445,80 @@ static int mfill_copy_folio_locked(struct folio *folio, unsigned long src_addr) return ret; } -static int mfill_copy_folio_retry(struct mfill_state *state, +#define MFILL_RETRY_STATE_VMA_FLAGS \ + append_vma_flags(__VMA_UFFD_FLAGS, VMA_SHARED_BIT) + +/* + * VMA state saved before dropping the locks in mfill_copy_folio_retry(). + * Used to detect VMA replacement or incompatible changes after reacquiring the + * locks. + */ +struct mfill_retry_state { + const struct vm_uffd_ops *ops; + struct file *file; + vma_flags_t flags; + pgoff_t pgoff; +}; + +static void mfill_retry_state_save(struct mfill_retry_state *s, + struct vm_area_struct *vma) +{ + s->flags = vma_flags_and_mask(&vma->flags, MFILL_RETRY_STATE_VMA_FLAGS); + s->ops = vma_uffd_ops(vma); + s->pgoff = vma->vm_pgoff; + + if (vma->vm_file) + s->file = get_file(vma->vm_file); +} + +static bool mfill_retry_state_changed(struct mfill_retry_state *state, + struct vm_area_struct *vma) +{ + vma_flags_t flags = vma_flags_and_mask(&vma->flags, + MFILL_RETRY_STATE_VMA_FLAGS); + + /* Have any UFFD flags (missing, WP, minor) changed? */ + if (!vma_flags_same_pair(&state->flags, &flags)) + return true; + + /* VMA type or effective uffd_ops changed while the lock was dropped */ + if (state->ops != vma_uffd_ops(vma)) + return true; + + /* VMA was anonymous before; changed only if it no longer is */ + if (!state->file) + return !vma_is_anonymous(vma); + + /* VMA was file backed, but file, inode or offset has changed */ + if (!vma->vm_file || vma->vm_file->f_inode != state->file->f_inode || + state->file != vma->vm_file || vma->vm_pgoff != state->pgoff) + return true; + + return false; +} + +static void mfill_retry_state_put(struct mfill_retry_state *s) +{ + if (s->file) + fput(s->file); +} + +DEFINE_FREE(retry_put, struct mfill_retry_state *, + if (_T) mfill_retry_state_put(_T)); + +static int mfill_copy_folio_retry(struct mfill_state *mfill_state, struct folio *folio) { - const struct vm_uffd_ops *orig_ops = vma_uffd_ops(state->vma); - unsigned long src_addr = state->src_addr; + struct mfill_retry_state retry_state = { 0 }; + struct mfill_retry_state *for_free __free(retry_put) = &retry_state; + unsigned long src_addr = mfill_state->src_addr; void *kaddr; int err; + mfill_retry_state_save(&retry_state, mfill_state->vma); + /* retry copying with mm_lock dropped */ - mfill_put_vma(state); + mfill_put_vma(mfill_state); kaddr = kmap_local_folio(folio, 0); err = copy_from_user(kaddr, (const void __user *) src_addr, PAGE_SIZE); @@ -463,19 +529,14 @@ static int mfill_copy_folio_retry(struct mfill_state *state, flush_dcache_folio(folio); /* reget VMA and PMD, they could change underneath us */ - err = mfill_get_vma(state); + err = mfill_get_vma(mfill_state); if (err) return err; - /* - * The VMA type may have changed while the lock was dropped - * (e.g. replaced with a hugetlb mapping), making the caller's - * ops pointer stale. - */ - if (vma_uffd_ops(state->vma) != orig_ops) + if (mfill_retry_state_changed(&retry_state, mfill_state->vma)) return -EAGAIN; - err = mfill_establish_pmd(state); + err = mfill_establish_pmd(mfill_state); if (err) return err; From df3ee3b3bbc327f570c5451666bbaf6cf8b4436a Mon Sep 17 00:00:00 2001 From: "Mike Rapoport (Microsoft)" Date: Wed, 27 May 2026 21:47:50 +0300 Subject: [PATCH 104/329] userfaultfd: refuse to __mfill_atomic_pte() for unsupported VMAs __mfill_atomic_pte() unconditionally dereferences ops because there is an assumption that VMAs that can undergo mfill_* operations are vetted on registration and must have valid vm_uffd_ops. Add a guard against potential bugs and make sure __mfill_atomic_pte() bails out if ops is NULL. Link: https://lore.kernel.org/20260527184751.4147364-3-rppt@kernel.org Fixes: ad9ac3081332 ("userfaultfd: introduce vm_uffd_ops->alloc_folio()") Signed-off-by: Mike Rapoport (Microsoft) Suggested-by: Lorenzo Stoakes Reviewed-by: Lorenzo Stoakes Reviewed-by: David CARLIER Cc: David Hildenbrand Cc: Liam R. Howlett Cc: Michael Bommarito Cc: Peter Xu Signed-off-by: Andrew Morton --- mm/userfaultfd.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index e5d2fb3ce2c1..2872c71bbf36 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -552,6 +552,11 @@ static int __mfill_atomic_pte(struct mfill_state *state, struct folio *folio; int ret; + if (!ops) { + VM_WARN_ONCE(1, "UFFDIO_COPY for unsupported VMA"); + return -EOPNOTSUPP; + } + folio = ops->alloc_folio(state->vma, state->dst_addr); if (!folio) return -ENOMEM; From 9d7bea186ba5a002456471edf36cc9b69f809397 Mon Sep 17 00:00:00 2001 From: "Mike Rapoport (Microsoft)" Date: Wed, 27 May 2026 21:47:51 +0300 Subject: [PATCH 105/329] userfaultfd: remove redundant check in vm_uffd_ops() Lorenzo says: static const struct vm_uffd_ops *vma_uffd_ops(struct vm_area_struct *vma) { if (vma_is_anonymous(vma)) return &anon_uffd_ops; return vma->vm_ops ? vma->vm_ops->uffd_ops : NULL; } This is doing a redundant check _and_ making life confusing, as if !vma->vm_ops is a condition that can be reached there, it can't, as vma_is_anonymous() is literally a !vma->vm_ops check :) Remove the redundant check. Link: https://lore.kernel.org/20260527184751.4147364-4-rppt@kernel.org Fixes: 0f48947c4232 ("userfaultfd: introduce vm_uffd_ops") Signed-off-by: Mike Rapoport (Microsoft) Suggested-by: Lorenzo Stoakes Reviewed-by: Lorenzo Stoakes Cc: David Hildenbrand Cc: Liam R. Howlett Cc: Peter Xu Cc: David Carlier Cc: Michael Bommarito Signed-off-by: Andrew Morton --- mm/userfaultfd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c index 2872c71bbf36..80cc8be5725f 100644 --- a/mm/userfaultfd.c +++ b/mm/userfaultfd.c @@ -68,7 +68,7 @@ static const struct vm_uffd_ops *vma_uffd_ops(struct vm_area_struct *vma) { if (vma_is_anonymous(vma)) return &anon_uffd_ops; - return vma->vm_ops ? vma->vm_ops->uffd_ops : NULL; + return vma->vm_ops->uffd_ops; } static __always_inline From c6c5327dd18bec1e1bbf139b2cf5ae53608a9d30 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Fri, 22 May 2026 12:47:17 +0200 Subject: [PATCH 106/329] netfilter: xt_NFQUEUE: prefer raw_smp_processor_id With PREEMPT_RCU this triggers a splat because smp_processor_id() can be preempted while inside a RCU critical section. If xt_NFQUEUE target is invoked via nft_compat_eval() path, we are inside a RCU critical section. Just use the raw version instead. Fixes: 0ca743a55991 ("netfilter: nf_tables: add compatibility layer for x_tables") Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/xt_NFQUEUE.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/xt_NFQUEUE.c b/net/netfilter/xt_NFQUEUE.c index 466da23e36ff..b32d153e3a18 100644 --- a/net/netfilter/xt_NFQUEUE.c +++ b/net/netfilter/xt_NFQUEUE.c @@ -91,7 +91,7 @@ nfqueue_tg_v3(struct sk_buff *skb, const struct xt_action_param *par) if (info->queues_total > 1) { if (info->flags & NFQ_FLAG_CPU_FANOUT) { - int cpu = smp_processor_id(); + int cpu = raw_smp_processor_id(); queue = info->queuenum + cpu % info->queues_total; } else { From 193989cc6d80dd8e0460fb3992e69fa03bf0ff9b Mon Sep 17 00:00:00 2001 From: Julian Anastasov Date: Mon, 25 May 2026 07:07:44 +0300 Subject: [PATCH 107/329] ipvs: clear the svc scheduler ptr early on edit ip_vs_edit_service() while unbinding the old scheduler clears the svc->scheduler ptr after the scheduler module initiates RCU callbacks. This can cause packets to use the old scheduler at the time when svc->sched_data is already freed after RCU grace period. Fix it by clearing the ptr early in ip_vs_unbind_scheduler(), before the done_service method schedules any RCU callbacks. Also, if the new scheduler fails to initialize when replacing the old scheduler, try to restore the old scheduler while still returning the error code. Link: https://sashiko.dev/#/patchset/20260519015506.634185-1-rosenp%40gmail.com Fixes: 05f00505a89a ("ipvs: fix crash if scheduler is changed") Signed-off-by: Julian Anastasov Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- include/net/ip_vs.h | 3 +-- net/netfilter/ipvs/ip_vs_ctl.c | 13 ++++++++----- net/netfilter/ipvs/ip_vs_sched.c | 14 +++++++------- 3 files changed, 16 insertions(+), 14 deletions(-) diff --git a/include/net/ip_vs.h b/include/net/ip_vs.h index a02e569813d2..e517eaaa177b 100644 --- a/include/net/ip_vs.h +++ b/include/net/ip_vs.h @@ -1824,8 +1824,7 @@ int register_ip_vs_scheduler(struct ip_vs_scheduler *scheduler); int unregister_ip_vs_scheduler(struct ip_vs_scheduler *scheduler); int ip_vs_bind_scheduler(struct ip_vs_service *svc, struct ip_vs_scheduler *scheduler); -void ip_vs_unbind_scheduler(struct ip_vs_service *svc, - struct ip_vs_scheduler *sched); +void ip_vs_unbind_scheduler(struct ip_vs_service *svc); struct ip_vs_scheduler *ip_vs_scheduler_get(const char *sched_name); void ip_vs_scheduler_put(struct ip_vs_scheduler *scheduler); struct ip_vs_conn * diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c index bd9cae44d214..16daba8cac83 100644 --- a/net/netfilter/ipvs/ip_vs_ctl.c +++ b/net/netfilter/ipvs/ip_vs_ctl.c @@ -1898,7 +1898,7 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u, if (ret_hooks >= 0) ip_vs_unregister_hooks(ipvs, u->af); if (svc != NULL) { - ip_vs_unbind_scheduler(svc, sched); + ip_vs_unbind_scheduler(svc); ip_vs_service_free(svc); } ip_vs_scheduler_put(sched); @@ -1962,9 +1962,8 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) old_sched = rcu_dereference_protected(svc->scheduler, 1); if (sched != old_sched) { if (old_sched) { - ip_vs_unbind_scheduler(svc, old_sched); - RCU_INIT_POINTER(svc->scheduler, NULL); - /* Wait all svc->sched_data users */ + ip_vs_unbind_scheduler(svc); + /* Wait all svc->scheduler/sched_data users */ synchronize_rcu(); } /* Bind the new scheduler */ @@ -1972,6 +1971,10 @@ ip_vs_edit_service(struct ip_vs_service *svc, struct ip_vs_service_user_kern *u) ret = ip_vs_bind_scheduler(svc, sched); if (ret) { ip_vs_scheduler_put(sched); + /* Try to restore the old_sched */ + if (old_sched && + !ip_vs_bind_scheduler(svc, old_sched)) + old_sched = NULL; goto out; } } @@ -2027,7 +2030,7 @@ static void __ip_vs_del_service(struct ip_vs_service *svc, bool cleanup) /* Unbind scheduler */ old_sched = rcu_dereference_protected(svc->scheduler, 1); - ip_vs_unbind_scheduler(svc, old_sched); + ip_vs_unbind_scheduler(svc); ip_vs_scheduler_put(old_sched); /* Unbind persistence engine, keep svc->pe */ diff --git a/net/netfilter/ipvs/ip_vs_sched.c b/net/netfilter/ipvs/ip_vs_sched.c index c6e421c4e299..24adc38942a0 100644 --- a/net/netfilter/ipvs/ip_vs_sched.c +++ b/net/netfilter/ipvs/ip_vs_sched.c @@ -56,19 +56,19 @@ int ip_vs_bind_scheduler(struct ip_vs_service *svc, /* * Unbind a service with its scheduler */ -void ip_vs_unbind_scheduler(struct ip_vs_service *svc, - struct ip_vs_scheduler *sched) +void ip_vs_unbind_scheduler(struct ip_vs_service *svc) { - struct ip_vs_scheduler *cur_sched; + struct ip_vs_scheduler *sched; - cur_sched = rcu_dereference_protected(svc->scheduler, 1); - /* This check proves that old 'sched' was installed */ - if (!cur_sched) + sched = rcu_dereference_protected(svc->scheduler, 1); + if (!sched) return; + /* Reset the scheduler before initiating any RCU callbacks */ + rcu_assign_pointer(svc->scheduler, NULL); + smp_wmb(); /* paired with smp_rmb() in ip_vs_schedule() */ if (sched->done_service) sched->done_service(svc); - /* svc->scheduler can be set to NULL only by caller */ } From 36d29ceec32c8206a12dc2810cf65fd394e45baa Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Tue, 26 May 2026 10:02:27 +0800 Subject: [PATCH 108/329] netfilter: nft_fib_ipv6: bail out of sibling walk if rt got unlinked This was reported by Sashiko [1]. The RCU walk over rt->fib6_siblings can spin forever if rt is unlinked mid-iteration: rt->fib6_siblings.next still points into the old ring, so the loop never meets &rt->fib6_siblings as its terminator. fib6_purge_rt() always does WRITE_ONCE(rt->fib6_nsiblings, 0) before list_del_rcu(), so readers can use rt->fib6_nsiblings == 0 as the detach signal. The same pattern is used in fib6_info_uses_dev() and rt6_nlmsg_size(). [1]: https://sashiko.dev/#/patchset/20260520023411.391233-1-jiayuan.chen%40linux.dev Suggested-by: Florian Westphal Fixes: 1c32b24c234b ("netfilter: nft_fib_ipv6: switch to fib6_lookup") Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/ipv6/netfilter/nft_fib_ipv6.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/ipv6/netfilter/nft_fib_ipv6.c b/net/ipv6/netfilter/nft_fib_ipv6.c index c0a0075e2590..2dbe44715df3 100644 --- a/net/ipv6/netfilter/nft_fib_ipv6.c +++ b/net/ipv6/netfilter/nft_fib_ipv6.c @@ -191,6 +191,9 @@ static bool nft_fib6_info_nh_uses_dev(struct fib6_info *rt, if (nft_fib6_info_nh_dev_match(nh_dev, dev)) return true; + + if (!READ_ONCE(rt->fib6_nsiblings)) + return false; } return false; From 2fcba19caaeb2a33017459d3430f057967bb91b6 Mon Sep 17 00:00:00 2001 From: Fernando Fernandez Mancera Date: Tue, 26 May 2026 23:58:31 +0200 Subject: [PATCH 109/329] netfilter: synproxy: add mutex to guard hook reference counting As the synproxy infrastructure register netfilter hooks on-demand when a user adds the first iptables target or nftables expression, if done concurrently they can race each other. Introduce a mutex to serialize the refcount control blocks access from both frontends. While a per namespace mutex might be more efficient, it is not needed for target/expression like SYNPROXY. Fixes: ad49d86e07a4 ("netfilter: nf_tables: Add synproxy support") Signed-off-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_synproxy_core.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) diff --git a/net/netfilter/nf_synproxy_core.c b/net/netfilter/nf_synproxy_core.c index 036c8586f49b..ed00114f65f3 100644 --- a/net/netfilter/nf_synproxy_core.c +++ b/net/netfilter/nf_synproxy_core.c @@ -22,6 +22,8 @@ #include #include +static DEFINE_MUTEX(synproxy_mutex); + unsigned int synproxy_net_id; EXPORT_SYMBOL_GPL(synproxy_net_id); @@ -769,26 +771,31 @@ static const struct nf_hook_ops ipv4_synproxy_ops[] = { int nf_synproxy_ipv4_init(struct synproxy_net *snet, struct net *net) { - int err; + int err = 0; + mutex_lock(&synproxy_mutex); if (snet->hook_ref4 == 0) { err = nf_register_net_hooks(net, ipv4_synproxy_ops, ARRAY_SIZE(ipv4_synproxy_ops)); if (err) - return err; + goto out; } snet->hook_ref4++; - return 0; +out: + mutex_unlock(&synproxy_mutex); + return err; } EXPORT_SYMBOL_GPL(nf_synproxy_ipv4_init); void nf_synproxy_ipv4_fini(struct synproxy_net *snet, struct net *net) { + mutex_lock(&synproxy_mutex); snet->hook_ref4--; if (snet->hook_ref4 == 0) nf_unregister_net_hooks(net, ipv4_synproxy_ops, ARRAY_SIZE(ipv4_synproxy_ops)); + mutex_unlock(&synproxy_mutex); } EXPORT_SYMBOL_GPL(nf_synproxy_ipv4_fini); @@ -1193,27 +1200,32 @@ static const struct nf_hook_ops ipv6_synproxy_ops[] = { int nf_synproxy_ipv6_init(struct synproxy_net *snet, struct net *net) { - int err; + int err = 0; + mutex_lock(&synproxy_mutex); if (snet->hook_ref6 == 0) { err = nf_register_net_hooks(net, ipv6_synproxy_ops, ARRAY_SIZE(ipv6_synproxy_ops)); if (err) - return err; + goto out; } snet->hook_ref6++; - return 0; +out: + mutex_unlock(&synproxy_mutex); + return err; } EXPORT_SYMBOL_GPL(nf_synproxy_ipv6_init); void nf_synproxy_ipv6_fini(struct synproxy_net *snet, struct net *net) { + mutex_lock(&synproxy_mutex); snet->hook_ref6--; if (snet->hook_ref6 == 0) nf_unregister_net_hooks(net, ipv6_synproxy_ops, ARRAY_SIZE(ipv6_synproxy_ops)); + mutex_unlock(&synproxy_mutex); } EXPORT_SYMBOL_GPL(nf_synproxy_ipv6_fini); #endif /* CONFIG_IPV6 */ From 66eba0ffce3b7e11449946b4cbbef8ea36112f56 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Wed, 27 May 2026 12:20:19 +0200 Subject: [PATCH 110/329] netfilter: conntrack_irc: fix possible out-of-bounds read When parsing fails after we've matched the command string we should bail out instead of trying to match a different command. This helper should be deprecated, given prevalence of TLS I doubt it has any relevance in 2026. Fixes: 869f37d8e48f ("[NETFILTER]: nf_conntrack/nf_nat: add IRC helper port") Closes: https://sashiko.dev/#/patchset/20260525182924.28456-1-fw%40strlen.de Signed-off-by: Florian Westphal Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nf_conntrack_irc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/netfilter/nf_conntrack_irc.c b/net/netfilter/nf_conntrack_irc.c index 522183b9a604..2ebe4cb47cf6 100644 --- a/net/netfilter/nf_conntrack_irc.c +++ b/net/netfilter/nf_conntrack_irc.c @@ -203,7 +203,7 @@ static int help(struct sk_buff *skb, unsigned int protoff, if (parse_dcc(data, data_limit, &dcc_ip, &dcc_port, &addr_beg_p, &addr_end_p)) { pr_debug("unable to parse dcc command\n"); - continue; + goto out; } pr_debug("DCC bound ip/port: %pI4:%u\n", @@ -217,7 +217,7 @@ static int help(struct sk_buff *skb, unsigned int protoff, net_warn_ratelimited("Forged DCC command from %pI4: %pI4:%u\n", &tuple->src.u3.ip, &dcc_ip, dcc_port); - continue; + goto out; } exp = nf_ct_expect_alloc(ct); From c32b26aaa2f9216520a38b3f4bfeec846eb3eb8a Mon Sep 17 00:00:00 2001 From: Tristan Madani Date: Wed, 27 May 2026 13:57:50 +0000 Subject: [PATCH 111/329] netfilter: nft_tunnel: fix use-after-free on object destroy nft_tunnel_obj_destroy() calls metadata_dst_free() which directly kfree()s the metadata_dst, ignoring the dst_entry refcount. Packets that took a reference via dst_hold() in nft_tunnel_obj_eval() and are still queued (e.g. in a netem qdisc) are left with a dangling pointer. When these packets are eventually dequeued, dst_release() operates on freed memory. Replace metadata_dst_free() with dst_release() so the metadata_dst is freed only after all references are dropped. The dst subsystem already handles metadata_dst cleanup in dst_destroy() when DST_METADATA is set. Fixes: af308b94a2a4 ("netfilter: nf_tables: add tunnel support") Cc: stable@vger.kernel.org Signed-off-by: Tristan Madani Reviewed-by: Fernando Fernandez Mancera Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_tunnel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nft_tunnel.c b/net/netfilter/nft_tunnel.c index 0b987bc2132a..68f7cfbbee06 100644 --- a/net/netfilter/nft_tunnel.c +++ b/net/netfilter/nft_tunnel.c @@ -676,7 +676,7 @@ static void nft_tunnel_obj_destroy(const struct nft_ctx *ctx, { struct nft_tunnel_obj *priv = nft_obj_data(obj); - metadata_dst_free(priv->md); + dst_release(&priv->md->dst); } static struct nft_object_type nft_tunnel_obj_type; From 3027ecbdb5fdf9200251c21d4818e4c447ef78e1 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Thu, 28 May 2026 19:09:19 +0800 Subject: [PATCH 112/329] netfilter: nft_ct: bail out on template ct in get eval I noticed this issue while looking at a historic syzbot report [1]. A rule like the one below is enough to trigger the bug: table ip t { chain pre { type filter hook prerouting priority raw; ct zone set 1 ct original saddr 1.2.3.4 accept } } The first expression attaches a per-cpu template ct via nft_ct_set_zone_eval() (nf_ct_tmpl_alloc -> kzalloc, tuple is all zero, nf_ct_l3num(ct) == 0). The next expression then calls nft_ct_get_eval() on the same skb, treats the template as a real ct and hits the 16-byte memcpy path. With dreg at NFT_REG32_15 this overflows past struct nft_regs on the kernel stack; with smaller dreg values it silently clobbers adjacent registers. Reject template ct at the eval entry and in nft_ct_get_fast_eval(), mirroring the check nft_ct_set_eval() already has. Additionally, bound the address copy in NFT_CT_SRC / NFT_CT_DST by priv->len instead of by nf_ct_l3num(ct): nf_ct_get_tuple() zeroes the tuple before pkt_to_tuple() fills in only the protocol-relevant leading bytes, so the trailing bytes of tuple->{src,dst}.u3.all are well-defined zero. priv->len is validated at rule load, so the copy size is now bounded by the destination register rather than by an untrusted field on the conntrack. [1]: https://syzkaller.appspot.com/bug?id=389cf09cb72926114fce90dc85a2c3231dcb647c Fixes: 45d9bcda21f4 ("netfilter: nf_tables: validate len in nft_validate_data_load()") Suggested-by: Florian Westphal Signed-off-by: Jiayuan Chen Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_ct.c | 8 +++----- net/netfilter/nft_ct_fast.c | 2 +- 2 files changed, 4 insertions(+), 6 deletions(-) diff --git a/net/netfilter/nft_ct.c b/net/netfilter/nft_ct.c index fa2cc556331c..357513c6dcea 100644 --- a/net/netfilter/nft_ct.c +++ b/net/netfilter/nft_ct.c @@ -78,7 +78,7 @@ static void nft_ct_get_eval(const struct nft_expr *expr, break; } - if (ct == NULL) + if (!ct || nf_ct_is_template(ct)) goto err; switch (priv->key) { @@ -180,12 +180,10 @@ static void nft_ct_get_eval(const struct nft_expr *expr, tuple = &ct->tuplehash[priv->dir].tuple; switch (priv->key) { case NFT_CT_SRC: - memcpy(dest, tuple->src.u3.all, - nf_ct_l3num(ct) == NFPROTO_IPV4 ? 4 : 16); + memcpy(dest, tuple->src.u3.all, priv->len); return; case NFT_CT_DST: - memcpy(dest, tuple->dst.u3.all, - nf_ct_l3num(ct) == NFPROTO_IPV4 ? 4 : 16); + memcpy(dest, tuple->dst.u3.all, priv->len); return; case NFT_CT_PROTO_SRC: nft_reg_store16(dest, (__force u16)tuple->src.u.all); diff --git a/net/netfilter/nft_ct_fast.c b/net/netfilter/nft_ct_fast.c index e684c8a91848..ecf7b3a404be 100644 --- a/net/netfilter/nft_ct_fast.c +++ b/net/netfilter/nft_ct_fast.c @@ -30,7 +30,7 @@ void nft_ct_get_fast_eval(const struct nft_expr *expr, break; } - if (!ct) { + if (!ct || nf_ct_is_template(ct)) { regs->verdict.code = NFT_BREAK; return; } From 67ba971ae02514d85818fe0c32549ab4bfa3bf49 Mon Sep 17 00:00:00 2001 From: Yiming Qian Date: Sat, 23 May 2026 12:29:10 +0000 Subject: [PATCH 113/329] netfilter: bridge: make ebt_snat ARP rewrite writable The ebtables SNAT target keeps the Ethernet source address rewrite behind skb_ensure_writable(skb, 0). This is intentional: at the bridge ebtables hooks the Ethernet header is addressed through skb_mac_header()/eth_hdr(), while skb->data points at the Ethernet payload. Asking skb_ensure_writable() for ETH_HLEN bytes would check the payload, not the Ethernet header, and would reintroduce the small packet regression fixed by commit 63137bc5882a. However, the optional ARP sender hardware address rewrite is different. It writes through skb_store_bits() at an offset relative to skb->data: skb_store_bits(skb, sizeof(struct arphdr), info->mac, ETH_ALEN) skb_header_pointer() only safely reads the ARP header; it does not make the later sender hardware address range writable. If that range is still held in a nonlinear skb fragment backed by a splice-imported file page, skb_store_bits() maps the frag page and copies the new MAC address directly into it. Ensure the ARP SHA range is writable before reading the ARP header and before calling skb_store_bits(). Fixes: 63137bc5882a ("netfilter: ebtables: Fixes dropping of small packets in bridge nat") Reported-by: Yiming Qian Signed-off-by: Yiming Qian Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/bridge/netfilter/ebt_snat.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/bridge/netfilter/ebt_snat.c b/net/bridge/netfilter/ebt_snat.c index 7dfbcdfc30e5..c9e229af0366 100644 --- a/net/bridge/netfilter/ebt_snat.c +++ b/net/bridge/netfilter/ebt_snat.c @@ -31,6 +31,9 @@ ebt_snat_tg(struct sk_buff *skb, const struct xt_action_param *par) const struct arphdr *ap; struct arphdr _ah; + if (skb_ensure_writable(skb, sizeof(_ah) + ETH_ALEN)) + return EBT_DROP; + ap = skb_header_pointer(skb, 0, sizeof(_ah), &_ah); if (ap == NULL) return EBT_DROP; From bb061d3de41707415269be75ebf700efb03ec212 Mon Sep 17 00:00:00 2001 From: Florian Westphal Date: Tue, 12 May 2026 15:36:14 +0200 Subject: [PATCH 114/329] netfilter: nft_byteorder: remove multi-register support 64bit byteorder conversion is broken when several registers need to be converted because the source register array advances in steps for 4 bytes instead of 8: for (i = ... src64 = nft_reg_load64(&src[i]); ~~~~~ u32 *src nft_reg_store64(&dst64[i], Remove the multi-register support, it has other issues as well: Pablo points out that commit caf3ef7468f7 ("netfilter: nf_tables: prevent OOB access in nft_byteorder_eval") alters semantics: before the loop operated on registers, i.e. for ( ... ) dst32[i] = htons((u16)src32[i]) .. but after the patch it will operate on bytes, which makes this useless to convert e.g. concatenations, which store each compound in its own register. Multi-convert of u32 has one theoretical application: ct mark . meta mark . tcp dport @intervalset Because ct mark and meta mark are host byte order, use with intervals has to convert the byteorder for ct/meta mark value to network byte order (bigendian). nftables emits this: [ meta load mark => reg 1 ] [ byteorder reg 1 = hton(reg 1, 4, 4) ] [ ct load mark => reg 9 ] [ byteorder reg 9 = hton(reg 9, 4, 4) ] ... I.e. two separate calls. Theoretically it could be changed to do: [ meta load mark => reg 1 ] [ ct load mark => reg 9 ] [ byteorder reg 1 = htonl(reg 1, 4, 8) ] ... But then all it would take to change the set to meta mark . tcp dport . ct mark ... and we'd be back to two "byteorder" calls. IOW, support to convert a range of registers is both dysfunctional and dubious. Simplify this: remove the feature. Pablo Neira Ayuso points out that nftables before 1.1.0 can generate incorrect byteorder conversions, see 9fe58952c45a, "evaluate: skip byteorder conversion for selector smaller than 2 bytes" in nftables.git). Affected rulesets fail to load with this change and old userspace due to 'len != size' check. Fixes: c301f0981fdd ("netfilter: nf_tables: fix pointer math issue in nft_byteorder_eval()") Cc: # may break rule load with old nftables versions Reported-by: Michal Kubecek Link: https://lore.kernel.org/netfilter-devel/20240206104336.ctigqpkunom2ufmn@lion.mk-sys.cz/ Signed-off-by: Florian Westphal Signed-off-by: Pablo Neira Ayuso --- net/netfilter/nft_byteorder.c | 51 ++++++++++++++--------------------- 1 file changed, 20 insertions(+), 31 deletions(-) diff --git a/net/netfilter/nft_byteorder.c b/net/netfilter/nft_byteorder.c index 2316c77f4228..dfd41fc8d9b8 100644 --- a/net/netfilter/nft_byteorder.c +++ b/net/netfilter/nft_byteorder.c @@ -19,7 +19,6 @@ struct nft_byteorder { u8 sreg; u8 dreg; enum nft_byteorder_ops op:8; - u8 len; u8 size; }; @@ -28,13 +27,8 @@ void nft_byteorder_eval(const struct nft_expr *expr, const struct nft_pktinfo *pkt) { const struct nft_byteorder *priv = nft_expr_priv(expr); - u32 *src = ®s->data[priv->sreg]; + const u32 *src = ®s->data[priv->sreg]; u32 *dst = ®s->data[priv->dreg]; - u16 *s16, *d16; - unsigned int i; - - s16 = (void *)src; - d16 = (void *)dst; switch (priv->size) { case 8: { @@ -43,18 +37,14 @@ void nft_byteorder_eval(const struct nft_expr *expr, switch (priv->op) { case NFT_BYTEORDER_NTOH: - for (i = 0; i < priv->len / 8; i++) { - src64 = nft_reg_load64(&src[i]); - nft_reg_store64(&dst64[i], - be64_to_cpu((__force __be64)src64)); - } + src64 = nft_reg_load64(src); + + nft_reg_store64(dst64, be64_to_cpu((__force __be64)src64)); break; case NFT_BYTEORDER_HTON: - for (i = 0; i < priv->len / 8; i++) { - src64 = (__force __u64) - cpu_to_be64(nft_reg_load64(&src[i])); - nft_reg_store64(&dst64[i], src64); - } + src64 = (__force __u64)cpu_to_be64(nft_reg_load64(src)); + + nft_reg_store64(dst64, src64); break; } break; @@ -62,24 +52,20 @@ void nft_byteorder_eval(const struct nft_expr *expr, case 4: switch (priv->op) { case NFT_BYTEORDER_NTOH: - for (i = 0; i < priv->len / 4; i++) - dst[i] = ntohl((__force __be32)src[i]); + *dst = ntohl((__force __be32)*src); break; case NFT_BYTEORDER_HTON: - for (i = 0; i < priv->len / 4; i++) - dst[i] = (__force __u32)htonl(src[i]); + *dst = (__force __u32)htonl(*src); break; } break; case 2: switch (priv->op) { case NFT_BYTEORDER_NTOH: - for (i = 0; i < priv->len / 2; i++) - d16[i] = ntohs((__force __be16)s16[i]); + nft_reg_store16(dst, ntohs(nft_reg_load_be16(src))); break; case NFT_BYTEORDER_HTON: - for (i = 0; i < priv->len / 2; i++) - d16[i] = (__force __u16)htons(s16[i]); + nft_reg_store_be16(dst, htons(nft_reg_load16(src))); break; } break; @@ -137,20 +123,22 @@ static int nft_byteorder_init(const struct nft_ctx *ctx, if (err < 0) return err; - priv->len = len; + /* no longer support multi-reg conversions */ + if (len != size) + return -EOPNOTSUPP; err = nft_parse_register_load(ctx, tb[NFTA_BYTEORDER_SREG], &priv->sreg, - priv->len); + len); if (err < 0) return err; err = nft_parse_register_store(ctx, tb[NFTA_BYTEORDER_DREG], &priv->dreg, NULL, NFT_DATA_VALUE, - priv->len); + len); if (err < 0) return err; - if (nft_reg_overlap(priv->sreg, priv->dreg, priv->len)) + if (nft_reg_overlap(priv->sreg, priv->dreg, len)) return -EINVAL; return 0; @@ -167,10 +155,11 @@ static int nft_byteorder_dump(struct sk_buff *skb, goto nla_put_failure; if (nla_put_be32(skb, NFTA_BYTEORDER_OP, htonl(priv->op))) goto nla_put_failure; - if (nla_put_be32(skb, NFTA_BYTEORDER_LEN, htonl(priv->len))) - goto nla_put_failure; if (nla_put_be32(skb, NFTA_BYTEORDER_SIZE, htonl(priv->size))) goto nla_put_failure; + /* compatibility for old userspace which permitted size != len */ + if (nla_put_be32(skb, NFTA_BYTEORDER_LEN, htonl(priv->size))) + goto nla_put_failure; return 0; nla_put_failure: From 02896a7fa4cd3ec61d60ba30136841e4f04bdeac Mon Sep 17 00:00:00 2001 From: Nikolay Kuratov Date: Tue, 26 May 2026 19:29:32 +0300 Subject: [PATCH 115/329] net/mlx5: Reorder completion before putting command entry in cmd_work_handler Assuming callback != NULL && !page_queue, cmd_work_handler takes command entry with refcnt == 1 from mlx5_cmd_invoke. If either semaphore timeout or index allocation error happens, it does final cmd_ent_put(ent). To avoid access to freed memory, notify slotted completion before cmd_ent_put. This is theoretical issue found by Svace static analyser. Cc: stable@vger.kernel.org Fixes: 485d65e135712 ("net/mlx5: Add a timeout to acquire the command queue semaphore") Fixes: 0e2909c6bec90 ("net/mlx5: Fix variable not being completed when function returns") Signed-off-by: Nikolay Kuratov Reviewed-by: Md Haris Iqbal Reviewed-by: Moshe Shemesh Acked-by: Tariq Toukan Link: https://patch.msgid.link/20260526162932.501584-1-kniv@yandex-team.ru Signed-off-by: Paolo Abeni --- drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c index c89417c1a1f9..e2895972cc82 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c @@ -1002,12 +1002,13 @@ static void cmd_work_handler(struct work_struct *work) ent->callback(-EBUSY, ent->context); mlx5_free_cmd_msg(dev, ent->out); free_msg(dev, ent->in); + complete(&ent->slotted); cmd_ent_put(ent); } else { ent->ret = -EBUSY; complete(&ent->done); + complete(&ent->slotted); } - complete(&ent->slotted); return; } alloc_ret = cmd_alloc_index(cmd, ent); @@ -1017,13 +1018,14 @@ static void cmd_work_handler(struct work_struct *work) ent->callback(-EAGAIN, ent->context); mlx5_free_cmd_msg(dev, ent->out); free_msg(dev, ent->in); + complete(&ent->slotted); cmd_ent_put(ent); } else { ent->ret = -EAGAIN; complete(&ent->done); + complete(&ent->slotted); } up(&cmd->vars.sem); - complete(&ent->slotted); return; } } else { From d3f0a606b9f278ece8a0df626ded9c4044071235 Mon Sep 17 00:00:00 2001 From: Guangshuo Li Date: Fri, 29 May 2026 23:57:45 +0800 Subject: [PATCH 116/329] dm cache policy smq: check allocation under invalidate lock commit 2d1f7b65f5de ("dm cache policy smq: fix missing locks in invalidating cache blocks") added mq->lock around the destructive part of smq_invalidate_mapping(), but left the e->allocated check outside the critical section. That leaves a check-then-act race. Two concurrent invalidators can both observe e->allocated as true before either of them takes mq->lock. The first invalidator that acquires the lock removes the entry from the queues and hash table and then calls free_entry(), which clears e->allocated and puts the entry back on the free list. The second invalidator can then acquire mq->lock and continue with the stale result of the unlocked check. This can corrupt the SMQ queues or hash table by deleting an entry that is no longer on those structures. It can also hit the allocation check in free_entry() when the same entry is freed again. Move the allocation check under mq->lock so the predicate and the destructive operations are serialized by the same lock. Fixes: 2d1f7b65f5de ("dm cache policy smq: fix missing locks in invalidating cache blocks") Signed-off-by: Guangshuo Li Signed-off-by: Mikulas Patocka --- drivers/md/dm-cache-policy-smq.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/md/dm-cache-policy-smq.c b/drivers/md/dm-cache-policy-smq.c index dd77a93fd68d..1ae304c2f573 100644 --- a/drivers/md/dm-cache-policy-smq.c +++ b/drivers/md/dm-cache-policy-smq.c @@ -1590,18 +1590,22 @@ static int smq_invalidate_mapping(struct dm_cache_policy *p, dm_cblock_t cblock) struct smq_policy *mq = to_smq_policy(p); struct entry *e = get_entry(&mq->cache_alloc, from_cblock(cblock)); unsigned long flags; - - if (!e->allocated) - return -ENODATA; + int r = 0; spin_lock_irqsave(&mq->lock, flags); + if (!e->allocated) { + r = -ENODATA; + goto out; + } // FIXME: what if this block has pending background work? del_queue(mq, e); h_remove(&mq->table, e); free_entry(&mq->cache_alloc, e); + +out: spin_unlock_irqrestore(&mq->lock, flags); - return 0; + return r; } static uint32_t smq_get_hint(struct dm_cache_policy *p, dm_cblock_t cblock) From fc12cf16df9af867e1455b011cdd9310474a612e Mon Sep 17 00:00:00 2001 From: Jackie Dong Date: Wed, 27 May 2026 18:20:05 +0800 Subject: [PATCH 117/329] ASoC: amd: acp: Add DMI quirk for Lenovo Yoga Pro 7 15ASH11 Lenovo Yoga Pro 7 15ASH11 with AMD RYZEN AI MAX+ 388 (Strix Halo, ACP 7.0) uses Realtek ALC287 series codec and no any DMIC connected by ACP. All DMICs directly connet with ALC codec. Without this quirk, Input Device of Gnome Sound settings shows Internal Stereo Microphone and Digital Microphone by default. In fact, Digital Microphone of ACP doesn't work due to no connecting with ALC287 codec, the Internal Stereo Microphone as analog device based on snd_hda_intel driver can work well. Add a DMI quirk to override the flag to 0, consistent with the existing entry for the HN7306EA. Signed-off-by: Jackie Dong Link: https://patch.msgid.link/20260527102005.58528-1-xy-jackie@139.com Signed-off-by: Mark Brown --- sound/soc/amd/acp-config.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/amd/acp-config.c b/sound/soc/amd/acp-config.c index 309dc9ed6e0d..0d977f4f758d 100644 --- a/sound/soc/amd/acp-config.c +++ b/sound/soc/amd/acp-config.c @@ -37,6 +37,13 @@ static const struct dmi_system_id acp70_acpi_flag_override_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Zenbook S16 UM5606GA"), }, }, + { + /* Lenovo Yoga Pro 7 15ASH11 (Strix Halo, ACP 7.0) */ + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_NAME, "83W5"), + }, + }, {} }; From c9c64820a48dcc739311fd7a9138511888539d02 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Jo=C3=A3o=20Miguel?= Date: Sat, 23 May 2026 22:35:48 +0100 Subject: [PATCH 118/329] ASoC: amd: yc: Enable internal mic on MSI Bravo 17 C7VF MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit The MSI Bravo 17 C7VF routes its internal digital microphone through the ACP6x. The machine driver only enables the DMIC for boards present in the DMI quirk table, so on this model the internal mic is never detected and no capture device is created. Add a quirk entry matching the board's DMI identifiers so the DMIC is enabled and the internal microphone works. Signed-off-by: João Miguel Link: https://patch.msgid.link/20260523213548.5219-1-jmiguel.ghp@gmail.com Signed-off-by: Mark Brown --- sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c index 7a637d6b5576..a93a7fd4877e 100644 --- a/sound/soc/amd/yc/acp6x-mach.c +++ b/sound/soc/amd/yc/acp6x-mach.c @@ -794,6 +794,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "M7601RM"), } }, + { + .driver_data = &acp6x_card, + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "Micro-Star International Co., Ltd."), + DMI_MATCH(DMI_BOARD_NAME, "MS-17LN"), + } + }, {} }; From 42445de1765547f56f48d107c0b8f3482c98458e Mon Sep 17 00:00:00 2001 From: Daniele Ceraolo Spurio Date: Fri, 29 May 2026 12:36:02 -0700 Subject: [PATCH 119/329] Revert "drm/xe/nvls: Define GuC firmware for NVL-S" This reverts commit 4e88de313ff4d1c67b644b1f39f9fb4089711b71. The early GuC FW definition meant for our CI branch was accidentally merged to the drm-xe-next branch instead. This GuC FW will never be released to linux-firmware, so we do not want the definition to be available in the mainline Linux codebase. Fixes: 4e88de313ff4 ("drm/xe/nvls: Define GuC firmware for NVL-S") Signed-off-by: Daniele Ceraolo Spurio Cc: Julia Filipchuk Cc: Rodrigo Vivi Cc: Matt Roper Cc: stable@vger.kernel.org # v7.0+ Reviewed-by: Rodrigo Vivi Link: https://patch.msgid.link/20260529193558.185436-11-daniele.ceraolospurio@intel.com Signed-off-by: Rodrigo Vivi (cherry picked from commit 65b8e0ac86e48cfc9128c04dfc53ea3395d030dd) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_uc_fw.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_uc_fw.c b/drivers/gpu/drm/xe/xe_uc_fw.c index 9cebb2490245..18ebefd444fe 100644 --- a/drivers/gpu/drm/xe/xe_uc_fw.c +++ b/drivers/gpu/drm/xe/xe_uc_fw.c @@ -115,7 +115,6 @@ struct fw_blobs_by_type { #define XE_GT_TYPE_ANY XE_GT_TYPE_UNINITIALIZED #define XE_GUC_FIRMWARE_DEFS(fw_def, mmp_ver, major_ver) \ - fw_def(NOVALAKE_S, GT_TYPE_ANY, mmp_ver(xe, guc, nvl, 70, 55, 4)) \ fw_def(PANTHERLAKE, GT_TYPE_ANY, major_ver(xe, guc, ptl, 70, 54, 0)) \ fw_def(BATTLEMAGE, GT_TYPE_ANY, major_ver(xe, guc, bmg, 70, 54, 0)) \ fw_def(LUNARLAKE, GT_TYPE_ANY, major_ver(xe, guc, lnl, 70, 53, 0)) \ From b5f53e6d3d32f2884d71500bb6d773c7bfe21f2e Mon Sep 17 00:00:00 2001 From: Pratik Vishwakarma Date: Sat, 30 May 2026 06:18:18 +0000 Subject: [PATCH 120/329] x86/CPU/AMD: Add more Zen6 models Family 0x1a, models 0xd0 - 0xef are Zen6, so add them to the range which sets X86_FEATURE_ZEN6. [ bp: Massage commit message. ] Signed-off-by: Pratik Vishwakarma Signed-off-by: Borislav Petkov (AMD) Link: https://patch.msgid.link/20260530061819.9721-1-Pratik.Vishwakarma@amd.com --- arch/x86/kernel/cpu/amd.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 2f8e8ff2d000..31f01e9c7114 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -518,7 +518,7 @@ static void bsp_init_amd(struct cpuinfo_x86 *c) break; case 0x50 ... 0x5f: case 0x80 ... 0xaf: - case 0xc0 ... 0xcf: + case 0xc0 ... 0xef: setup_force_cpu_cap(X86_FEATURE_ZEN6); break; default: From 27f2d085bd72abe4235689d34d8654cfc876d568 Mon Sep 17 00:00:00 2001 From: Zhan Xusheng Date: Mon, 1 Jun 2026 16:51:36 +0800 Subject: [PATCH 121/329] erofs: fix EFSCORRUPTED on multi-algorithm images in z_erofs_map_sanity_check() Commit a5242d37c83a ("erofs: error out obviously illegal extents in advance") changed the per-extent algorithm presence check from "is the bit set" to "is the only bit set": - !(sbi->available_compr_algs & (1 << map->m_algorithmformat)) + (sbi->available_compr_algs ^ BIT(map->m_algorithmformat)) `available_compr_algs` is a bitmap of every compression algorithm available in the image (z_erofs_parse_cfgs() iterates it with for_each_set_bit()), so an image that enables more than one algorithm has multiple bits set. XOR is zero only when the bitmap is exactly BIT(map->m_algorithmformat); for any image with two or more algorithms the test is non-zero for every extent and the read fails with -EFSCORRUPTED ("inconsistent algorithmtype %u"). Reproducer (mkfs.erofs from erofs-utils 1.7.1): $ mkdir src $ yes A | head -c 100K > src/a $ head -c 64K /dev/zero > src/b $ mkfs.erofs -zlz4:deflate multi.erofs src $ mount -t erofs -o loop multi.erofs /mnt $ cat /mnt/a >/dev/null cat: /mnt/a: Structure needs cleaning $ dmesg | tail erofs (device loop0): inconsistent algorithmtype 0 for nid 46 erofs (device loop0): read error -117 @ 0 of nid 46 The erofs on-disk format (Z_EROFS_COMPRESSION_MAX = 4 with LZ4, LZMA, DEFLATE, ZSTD) and the kernel parser explicitly support multi-algorithm images, and erofs-utils 1.7.1 generates them via the "-z X:Y" syntax. Restore the original per-bit presence check. Fixes: a5242d37c83a ("erofs: error out obviously illegal extents in advance") Signed-off-by: Zhan Xusheng Reviewed-by: Gao Xiang Signed-off-by: Gao Xiang --- fs/erofs/zmap.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index a72db36096ca..e1a02a2c8406 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -716,7 +716,7 @@ static int z_erofs_map_sanity_check(struct inode *inode, } if (map->m_algorithmformat < Z_EROFS_COMPRESSION_MAX) { - if (sbi->available_compr_algs ^ BIT(map->m_algorithmformat)) { + if (!(sbi->available_compr_algs & BIT(map->m_algorithmformat))) { erofs_err(inode->i_sb, "inconsistent algorithmtype %u for nid %llu", map->m_algorithmformat, EROFS_I(inode)->nid); return -EFSCORRUPTED; From abf888b03a9805a3bc37948a0df443553b1c0910 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Sat, 30 May 2026 15:37:42 -0300 Subject: [PATCH 122/329] drm/v3d: Wait for pending L2T flush before cleaning caches MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit v3d_clean_caches() starts the cache-clean sequence by writing V3D_L2TCACTL_TMUWCF to V3D_CTL_L2TCACTL and then polling for that bit to clear. It does not, however, check for an L2T flush (L2TFLS) that may still be in flight from a previous operation. On pre-V3D 7.1 hardware, kicking off the TMU write-combiner flush while an L2T flush is still pending can clobber bits in L2TCACTL and cause cache inconsistencies. Poll for L2TFLS to clear before writing L2TCACTL on V3D < 7.1, ensuring any pending flush has completed before a new clean is issued. Cc: stable@vger.kernel.org Fixes: d223f98f0209 ("drm/v3d: Add support for compute shader dispatch.") Link: https://patch.msgid.link/20260530-v3d-fix-rpi4-freezes-v1-1-c2c8307da6ce@igalia.com Signed-off-by: Maíra Canal Reviewed-by: Iago Toral Quiroga --- drivers/gpu/drm/v3d/v3d_gem.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c index 75d9eccd7966..dd7da419702f 100644 --- a/drivers/gpu/drm/v3d/v3d_gem.c +++ b/drivers/gpu/drm/v3d/v3d_gem.c @@ -213,6 +213,14 @@ v3d_clean_caches(struct v3d_dev *v3d) trace_v3d_cache_clean_begin(dev); + /* GFXH-1897: Ensure pending flushes complete before writing L2TCACTL */ + if (v3d->ver < V3D_GEN_71) { + if (wait_for(!(V3D_CORE_READ(core, V3D_CTL_L2TCACTL) & + V3D_L2TCACTL_L2TFLS), 100)) { + drm_err(dev, "Timeout waiting for L2T clean\n"); + } + } + V3D_CORE_WRITE(core, V3D_CTL_L2TCACTL, V3D_L2TCACTL_TMUWCF); if (wait_for(!(V3D_CORE_READ(core, V3D_CTL_L2TCACTL) & V3D_L2TCACTL_TMUWCF), 100)) { From 5057e1aca011e51ef51498c940ef96f3d3e8a305 Mon Sep 17 00:00:00 2001 From: Jamal Hadi Salim Date: Sun, 31 May 2026 12:08:12 -0400 Subject: [PATCH 123/329] net/sched: act_api: use RCU with deferred freeing for action lifecycle When NEWTFILTER and DELFILTER are run concurrently it is possible to create a race with an associated action. Let's illustrate with CPU0 running NEWTFILTER and CPU1 running DELFILTER: 0: mutex_lock() <-- holds the idr lock 0: rcu_read_lock() 0: p = idr_find(idr, index) <-- action p is valid (RCU protects IDR) 0: mutex_unlock() <-- releases the idr lock 1: refcount_dec_and_mutex_lock() <-- refcnt 1->0, mutex held 1: idr_remove(idr, index) <-- Action removed from IDR 1: mutex_unlock() <-- mutex released allowing us to delete the action 1: tcf_action_cleanup(p); kfree(p) <-- Kfrees p immediately, no deferral 0: refcount_inc_not_zero(&p->tcfa_refcnt) <-- ouch, UAF p points to freed memory This patch fixes the race condition between NEWTFILTER and DELFILTER by adding struct rcu_head to tc_action used in the deferral and introducing a call_rcu() in the delete path to defer the final kfree(). Note: this is a revert of commit d7fb60b9cafb ("net_sched: get rid of tcfa_rcu") but also modernization/simplification to directly use kfree_rcu(). Let's illustrate the new restored code path: 0: rcu_read_lock() 1: refcount_dec_and_mutex_lock() <-- refcnt 1->0, mutex held 1: idr_remove(idr, index) 1: mutex_unlock() 1: call_rcu(&p->tcfa_rcu, tcf_action_rcu_free) <-- defer kfree after grace period 0: p = idr_find(idr, index) 0: refcount_inc_not_zero(&p->tcfa_refcnt) <-- fails, refcnt already 0 1: rcu_read_unlock() <-- release so freeing can run after grace period After CPU1 calls idr_remove(), the object is no longer reachable through the IDR. CPU0's subsequent idr_find() will return NULL, and even if it still held a stale pointer, the immediate kfree() is now deferred until after the RCU grace period, so no UAF can occur. Fixes: d7fb60b9cafb ("net_sched: get rid of tcfa_rcu") Suggested-by: Jakub Kicinski Reported-by: Kyle Zeng Tested-by: Victor Nogueira Tested-by: syzbot@syzkaller.appspotmail.com Signed-off-by: Jamal Hadi Salim Tested-by: Kyle Zeng Reviewed-by: Pedro Tammela Reviewed-by: Eric Dumazet Reviewed-by: Victor Nogueira Link: https://patch.msgid.link/20260531160812.68020-1-jhs@mojatatu.com Signed-off-by: Jakub Kicinski --- include/net/act_api.h | 1 + net/sched/act_api.c | 7 +------ 2 files changed, 2 insertions(+), 6 deletions(-) diff --git a/include/net/act_api.h b/include/net/act_api.h index d11b79107930..fd2967ee08f7 100644 --- a/include/net/act_api.h +++ b/include/net/act_api.h @@ -45,6 +45,7 @@ struct tc_action { struct tc_cookie __rcu *user_cookie; struct tcf_chain __rcu *goto_chain; u32 tcfa_flags; + struct rcu_head tcfa_rcu; u8 hw_stats; u8 used_hw_stats; bool used_hw_stats_valid; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 332fd9695e54..04ea11c90e03 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -112,11 +112,6 @@ struct tcf_chain *tcf_action_set_ctrlact(struct tc_action *a, int action, } EXPORT_SYMBOL(tcf_action_set_ctrlact); -/* XXX: For standalone actions, we don't need a RCU grace period either, because - * actions are always connected to filters and filters are already destroyed in - * RCU callbacks, so after a RCU grace period actions are already disconnected - * from filters. Readers later can not find us. - */ static void free_tcf(struct tc_action *p) { struct tcf_chain *chain = rcu_dereference_protected(p->goto_chain, 1); @@ -129,7 +124,7 @@ static void free_tcf(struct tc_action *p) if (chain) tcf_chain_put_by_act(chain); - kfree(p); + kfree_rcu(p, tcfa_rcu); } static void offload_action_hw_count_set(struct tc_action *act, From 4a1b1ac2744694a2ecd66a84bdb1445f4ef24bee Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Tue, 26 May 2026 12:40:25 -0300 Subject: [PATCH 124/329] RDMA/core: Validate the passed in fops for ib_get_ucaps() Sashiko pointed out it is not safe to rely only on the devt because char/block alias so if the user finds a block device with the same dev_t it can masquerade as a ucap cdev fd. Test the f_ops to only accept authentic cdevs. Link: https://patch.msgid.link/r/0-v1-fd9482545e37+1e25-ib_ucaps_fd_ops_jgg@nvidia.com Cc: stable@vger.kernel.org Fixes: 61e51682816d ("RDMA/uverbs: Introduce UCAP (User CAPabilities) API") Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/ucaps.c | 8 +++----- 1 file changed, 3 insertions(+), 5 deletions(-) diff --git a/drivers/infiniband/core/ucaps.c b/drivers/infiniband/core/ucaps.c index 948093260dbd..5155ff0e538e 100644 --- a/drivers/infiniband/core/ucaps.c +++ b/drivers/infiniband/core/ucaps.c @@ -82,14 +82,12 @@ static int get_ucap_from_devt(dev_t devt, u64 *idx_mask) static int get_devt_from_fd(unsigned int fd, dev_t *ret_dev) { - struct file *file; + CLASS(fd, f)(fd); - file = fget(fd); - if (!file) + if (fd_empty(f) || fd_file(f)->f_op != &ucaps_cdev_fops) return -EBADF; - *ret_dev = file_inode(file)->i_rdev; - fput(file); + *ret_dev = file_inode(fd_file(f))->i_rdev; return 0; } From 2a58899d11009bffc7b4b32a571858f381121837 Mon Sep 17 00:00:00 2001 From: Yizhou Zhao Date: Wed, 27 May 2026 16:18:01 +0800 Subject: [PATCH 125/329] 6lowpan: fix off-by-one in multicast context address compression The second memcpy in lowpan_iphc_mcast_ctx_addr_compress() uses &data[1] as destination and &ipaddr->s6_addr[11] as source, but both should be offset by one: &data[2] and &ipaddr->s6_addr[12] respectively. This off-by-one has two consequences: 1. data[1] is overwritten with s6_addr[11], corrupting the RIID field in the compressed multicast address 2. data[5] is never written, so uninitialized kernel stack memory is transmitted over the network via lowpan_push_hc_data(), leaking kernel stack contents The correct inline data layout must match what the decompression function lowpan_uncompress_multicast_ctx_daddr() expects: data[0..1] = s6_addr[1..2] (flags/scope + RIID) data[2..5] = s6_addr[12..15] (group ID) Also zero-initialize the data array as a defensive measure against similar bugs in the future. Fixes: 5609c185f24d ("6lowpan: iphc: add support for stateful compression") Reported-by: Yizhou Zhao Reported-by: Yuxiang Yang Reported-by: Ao Wang Reported-by: Xuewei Feng Reported-by: Qi Li Reported-by: Ke Xu Signed-off-by: Yizhou Zhao Acked-by: Alexander Aring Link: https://patch.msgid.link/20260527081806.42747-1-zhaoyz24@mails.tsinghua.edu.cn Signed-off-by: Jakub Kicinski --- net/6lowpan/iphc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/net/6lowpan/iphc.c b/net/6lowpan/iphc.c index e116d308a8df..37eaff3f7b69 100644 --- a/net/6lowpan/iphc.c +++ b/net/6lowpan/iphc.c @@ -1086,12 +1086,12 @@ static u8 lowpan_iphc_mcast_ctx_addr_compress(u8 **hc_ptr, const struct lowpan_iphc_ctx *ctx, const struct in6_addr *ipaddr) { - u8 data[6]; + u8 data[6] = {}; /* flags/scope, reserved (RIID) */ memcpy(data, &ipaddr->s6_addr[1], 2); /* group ID */ - memcpy(&data[1], &ipaddr->s6_addr[11], 4); + memcpy(&data[2], &ipaddr->s6_addr[12], 4); lowpan_push_hc_data(hc_ptr, data, 6); return LOWPAN_IPHC_DAM_00; From a213a8950414c684999dcf03edeea6c46ede172e Mon Sep 17 00:00:00 2001 From: Lee Jones Date: Wed, 27 May 2026 13:36:29 +0000 Subject: [PATCH 126/329] l2tp: pppol2tp: hold reference to session in pppol2tp_ioctl() pppol2tp_ioctl() read sock->sk->sk_user_data directly without any locks or reference counting. If a controllable sleep was induced during copy_from_user() (e.g. via a userfaultfd page fault sleep), a concurrent socket close could trigger pppol2tp_session_close() asynchronously. This frees the l2tp_session structure via the l2tp_session_del_work workqueue. Upon resuming, the ioctl thread dereferences the stale session pointer, resulting in a Use-After-Free (UAF). Fix this by securely fetching the session reference using the RCU-safe, refcounted helper pppol2tp_sock_to_session(sk) on entry. This locks the session's refcount across the sleep. We structured the function to exit via standard err breaks, guaranteeing that l2tp_session_put() is cleanly called on all return paths to drop the reference. To preserve existing behavior we validate the session and its magic signature only for the specific L2TP commands that require it. This ensures that generic/unknown ioctls called on an unconnected socket still return -ENOIOCTLCMD and correctly fall back to generic handlers (e.g. in sock_do_ioctl()). Signed-off-by: Lee Jones Fixes: fd558d186df2 ("l2tp: Split pppol2tp patch into separate l2tp and ppp parts") Link: https://patch.msgid.link/20260527133630.2120612-1-lee@kernel.org Signed-off-by: Jakub Kicinski --- net/l2tp/l2tp_ppp.c | 92 +++++++++++++++++++++++++++------------------ 1 file changed, 55 insertions(+), 37 deletions(-) diff --git a/net/l2tp/l2tp_ppp.c b/net/l2tp/l2tp_ppp.c index 99d6582f41de..e0b1915be1a6 100644 --- a/net/l2tp/l2tp_ppp.c +++ b/net/l2tp/l2tp_ppp.c @@ -1045,64 +1045,76 @@ static int pppol2tp_ioctl(struct socket *sock, unsigned int cmd, { struct pppol2tp_ioc_stats stats; struct l2tp_session *session; + int err = 0; + + session = pppol2tp_sock_to_session(sock->sk); + + /* Validate session presence and magic integrity ONLY for commands + * that belong to L2TP and require a valid session. + */ + switch (cmd) { + case PPPIOCGMRU: + case PPPIOCGFLAGS: + case PPPIOCSMRU: + case PPPIOCSFLAGS: + case PPPIOCGL2TPSTATS: + if (!session) + return -ENOTCONN; + + if (session->magic != L2TP_SESSION_MAGIC) { + l2tp_session_put(session); + return -EBADF; + } + break; + default: + break; + } switch (cmd) { case PPPIOCGMRU: case PPPIOCGFLAGS: - session = sock->sk->sk_user_data; - if (!session) - return -ENOTCONN; - - if (WARN_ON(session->magic != L2TP_SESSION_MAGIC)) - return -EBADF; - /* Not defined for tunnels */ - if (!session->session_id && !session->peer_session_id) - return -ENOSYS; + if (!session->session_id && !session->peer_session_id) { + err = -ENOSYS; + break; + } - if (put_user(0, (int __user *)arg)) - return -EFAULT; + if (put_user(0, (int __user *)arg)) { + err = -EFAULT; + break; + } break; case PPPIOCSMRU: case PPPIOCSFLAGS: - session = sock->sk->sk_user_data; - if (!session) - return -ENOTCONN; - - if (WARN_ON(session->magic != L2TP_SESSION_MAGIC)) - return -EBADF; - /* Not defined for tunnels */ - if (!session->session_id && !session->peer_session_id) - return -ENOSYS; + if (!session->session_id && !session->peer_session_id) { + err = -ENOSYS; + break; + } - if (!access_ok((int __user *)arg, sizeof(int))) - return -EFAULT; + if (!access_ok((int __user *)arg, sizeof(int))) { + err = -EFAULT; + break; + } break; case PPPIOCGL2TPSTATS: - session = sock->sk->sk_user_data; - if (!session) - return -ENOTCONN; - - if (WARN_ON(session->magic != L2TP_SESSION_MAGIC)) - return -EBADF; - /* Session 0 represents the parent tunnel */ if (!session->session_id && !session->peer_session_id) { u32 session_id; - int err; if (copy_from_user(&stats, (void __user *)arg, - sizeof(stats))) - return -EFAULT; + sizeof(stats))) { + err = -EFAULT; + break; + } session_id = stats.session_id; err = pppol2tp_tunnel_copy_stats(&stats, session->tunnel); if (err < 0) - return err; + break; stats.session_id = session_id; } else { @@ -1112,15 +1124,21 @@ static int pppol2tp_ioctl(struct socket *sock, unsigned int cmd, stats.tunnel_id = session->tunnel->tunnel_id; stats.using_ipsec = l2tp_tunnel_uses_xfrm(session->tunnel); - if (copy_to_user((void __user *)arg, &stats, sizeof(stats))) - return -EFAULT; + if (copy_to_user((void __user *)arg, &stats, sizeof(stats))) { + err = -EFAULT; + break; + } break; default: - return -ENOIOCTLCMD; + err = -ENOIOCTLCMD; + break; } - return 0; + if (session) + l2tp_session_put(session); + + return err; } /***************************************************************************** From 3522b21fd7e1863d0734537737bd59f1b90d0190 Mon Sep 17 00:00:00 2001 From: Mark Bloch Date: Thu, 28 May 2026 22:14:10 +0300 Subject: [PATCH 127/329] devlink: Release nested relation on devlink free devlink relation state is normally released from devl_unregister(), which calls devlink_rel_put(). This misses devlink instances that get a nested relation before registration and then fail probe before devl_register() is reached. That flow can happen for SFs. The child devlink gets linked to its parent before registration, then a later probe error calls devlink_free() directly. Since the instance was never registered, devl_unregister() is not called and devlink->rel is leaked. Release any pending relation from devlink_free() as well. The registered path is unchanged because devl_unregister() already clears devlink->rel before devlink_free() runs. Fixes: c137743bce02 ("devlink: introduce object and nested devlink relationship infra") Signed-off-by: Mark Bloch Reviewed-by: Jiri Pirko Link: https://patch.msgid.link/20260528191411.3270532-1-mbloch@nvidia.com Signed-off-by: Jakub Kicinski --- net/devlink/core.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/devlink/core.c b/net/devlink/core.c index eeb6a71f5f56..fe9f6a0a67d5 100644 --- a/net/devlink/core.c +++ b/net/devlink/core.c @@ -518,6 +518,8 @@ void devlink_free(struct devlink *devlink) { ASSERT_DEVLINK_NOT_REGISTERED(devlink); + devlink_rel_put(devlink); + WARN_ON(!list_empty(&devlink->trap_policer_list)); WARN_ON(!list_empty(&devlink->trap_group_list)); WARN_ON(!list_empty(&devlink->trap_list)); From 1d0b597facdd3c0239c88e8797c1014e1ea0ef15 Mon Sep 17 00:00:00 2001 From: Andrzej Kacprowski Date: Fri, 29 May 2026 14:08:53 +0200 Subject: [PATCH 128/329] accel/ivpu: Add bounds check for firmware runtime memory Validate that the firmware runtime memory specified in the image header is properly aligned and sized to hold the firmware image. This prevents errors during memory allocation and image transfer. Fixes: 2007e210b6a1 ("accel/ivpu: Split FW runtime and global memory buffers") Cc: stable@vger.kernel.org # v7.0+ Signed-off-by: Andrzej Kacprowski Reviewed-by: Karol Wachowski Signed-off-by: Karol Wachowski Link: https://patch.msgid.link/20260529120853.135876-1-andrzej.kacprowski@linux.intel.com --- drivers/accel/ivpu/ivpu_fw.c | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_fw.c b/drivers/accel/ivpu/ivpu_fw.c index 107f8ad31050..33c50779c06b 100644 --- a/drivers/accel/ivpu/ivpu_fw.c +++ b/drivers/accel/ivpu/ivpu_fw.c @@ -259,6 +259,22 @@ static int ivpu_fw_parse(struct ivpu_device *vdev) return -EINVAL; } + if (!PAGE_ALIGNED(runtime_addr)) { + ivpu_err(vdev, "Runtime address 0x%llx not page aligned\n", runtime_addr); + return -EINVAL; + } + + if (!PAGE_ALIGNED(runtime_size)) { + ivpu_err(vdev, "Runtime size %llu not page aligned\n", runtime_size); + return -EINVAL; + } + + if (runtime_size < image_size) { + ivpu_err(vdev, "Runtime size too small: %llu, image size: %llu\n", + runtime_size, image_size); + return -EINVAL; + } + if (!ivpu_is_within_range(image_load_addr, image_size, &vdev->hw->ranges.runtime)) { ivpu_err(vdev, "Invalid firmware load address: 0x%llx and size %llu\n", image_load_addr, image_size); From dd1311bcf0e62f0c515115f46a3813370f4a4bb1 Mon Sep 17 00:00:00 2001 From: Andrzej Kacprowski Date: Fri, 29 May 2026 13:58:42 +0200 Subject: [PATCH 129/329] accel/ivpu: Add bounds checks for firmware log indices Add validation that read and write indices in the firmware log buffer are within valid bounds (< data_size) before using them. If out-of-bounds indices are encountered (from firmware), clamp them to safe values instead of proceeding with invalid offsets. This prevents potential out-of-bounds buffer access when firmware supplies invalid log indices. Fixes: 1fc1251149a7 ("accel/ivpu: Refactor functions in ivpu_fw_log.c") Cc: stable@vger.kernel.org # v6.18+ Signed-off-by: Andrzej Kacprowski Reviewed-by: Karol Wachowski Signed-off-by: Karol Wachowski Link: https://patch.msgid.link/20260529115842.135378-1-andrzej.kacprowski@linux.intel.com --- drivers/accel/ivpu/ivpu_fw_log.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_fw_log.c b/drivers/accel/ivpu/ivpu_fw_log.c index 337c906b0210..275baf844b56 100644 --- a/drivers/accel/ivpu/ivpu_fw_log.c +++ b/drivers/accel/ivpu/ivpu_fw_log.c @@ -98,6 +98,11 @@ static void fw_log_print_buffer(struct vpu_tracing_buffer_header *log, const cha u32 log_start = only_new_msgs ? READ_ONCE(log->read_index) : 0; u32 log_end = READ_ONCE(log->write_index); + if (log_start >= data_size) + log_start = 0; + if (log_end > data_size) + log_end = data_size; + if (log->wrap_count == log->read_wrap_count) { if (log_end <= log_start) { drm_printf(p, "==== %s \"%s\" log empty ====\n", prefix, log->name); From fb176425837693f50c5c9fc8db6fbb04af22bd0a Mon Sep 17 00:00:00 2001 From: Andrzej Kacprowski Date: Fri, 29 May 2026 14:08:41 +0200 Subject: [PATCH 130/329] accel/ivpu: Add buffer overflow check in MS get_info_ioctl Add validation that the info size returned from the metric stream info query is not exceeded when checked against the allocated buffer size. If the firmware returns a size larger than the buffer, reject the operation with -EOVERFLOW instead of proceeding with an incorrect buffer copy. Fixes: cdfad4db7756 ("accel/ivpu: Add NPU profiling support") Cc: stable@vger.kernel.org # v6.18+ Signed-off-by: Andrzej Kacprowski Reviewed-by: Karol Wachowski Signed-off-by: Karol Wachowski Link: https://patch.msgid.link/20260529120841.135852-1-andrzej.kacprowski@linux.intel.com --- drivers/accel/ivpu/ivpu_ms.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/accel/ivpu/ivpu_ms.c b/drivers/accel/ivpu/ivpu_ms.c index be43851f5f32..cd176e77b9a0 100644 --- a/drivers/accel/ivpu/ivpu_ms.c +++ b/drivers/accel/ivpu/ivpu_ms.c @@ -291,6 +291,13 @@ int ivpu_ms_get_info_ioctl(struct drm_device *dev, void *data, struct drm_file * if (ret) goto unlock; + if (info_size > ivpu_bo_size(bo)) { + ivpu_warn_ratelimited(vdev, "MS info overflow: %#llx > %#zx\n", + info_size, ivpu_bo_size(bo)); + ret = -EOVERFLOW; + goto unlock; + } + if (args->buffer_size < info_size) { ret = -ENOSPC; goto unlock; From ae0383e5a9a4b12d68c76c4769857def4665deff Mon Sep 17 00:00:00 2001 From: Yicong Hui Date: Mon, 6 Apr 2026 19:00:13 +0100 Subject: [PATCH 131/329] drm/imx: Fix three kernel-doc warnings in dcss-scaler.c Fix the following W=1 kerneldoc warnings by adding the missing parameter descriptions for @phase0_identity and @nn_interpolation in dcss_scaler_filter_design() and @phase0_identity in dcss_scaler_gaussian_filter() Warning: drivers/gpu/drm/imx/dcss/dcss-scaler.c:173 function parameter 'phase0_identity' not described in 'dcss_scaler_gaussian_filter' Warning: drivers/gpu/drm/imx/dcss/dcss-scaler.c:270 function parameter 'phase0_identity' not described in 'dcss_scaler_filter_design' Warning: drivers/gpu/drm/imx/dcss/dcss-scaler.c:270 function parameter 'nn_interpolation' not described in 'dcss_scaler_filter_design' Fixes: 9021c317b770 ("drm/imx: Add initial support for DCSS on iMX8MQ") Signed-off-by: Yicong Hui Reviewed-by: Laurentiu Palcu Link: https://patch.msgid.link/20260406180013.2442096-1-yiconghui@gmail.com Signed-off-by: Liu Ying --- drivers/gpu/drm/imx/dcss/dcss-scaler.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/imx/dcss/dcss-scaler.c b/drivers/gpu/drm/imx/dcss/dcss-scaler.c index 32c3f46b21da..5c7f8d952ec1 100644 --- a/drivers/gpu/drm/imx/dcss/dcss-scaler.c +++ b/drivers/gpu/drm/imx/dcss/dcss-scaler.c @@ -166,6 +166,7 @@ static int exp_approx_q(int x) * dcss_scaler_gaussian_filter() - Generate gaussian prototype filter. * @fc_q: fixed-point cutoff frequency normalized to range [0, 1] * @use_5_taps: indicates whether to use 5 taps or 7 taps + * @phase0_identity: whether to override phase 0 coefficients with identity filter * @coef: output filter coefficients */ static void dcss_scaler_gaussian_filter(int fc_q, bool use_5_taps, @@ -262,7 +263,9 @@ static void dcss_scaler_nearest_neighbor_filter(bool use_5_taps, * @src_length: length of input * @dst_length: length of output * @use_5_taps: 0 for 7 taps per phase, 1 for 5 taps + * @phase0_identity: whether to override phase 0 coefficients with identity filter * @coef: output coefficients + * @nn_interpolation: whether to use nearest neighbor instead of gaussian filter */ static void dcss_scaler_filter_design(int src_length, int dst_length, bool use_5_taps, bool phase0_identity, From 317d5146fb399ad1e87b310ee7d018fe648d40ba Mon Sep 17 00:00:00 2001 From: Dave Jones Date: Mon, 18 May 2026 14:46:28 -0400 Subject: [PATCH 132/329] NFS: write_completion: dereference loop-local req, not hdr->req 5d3869a41f36 ("NFS: fix writeback in presence of errors") introduced a dereference of hdr->req->wb_lock_context in nfs_write_completion's per-request loop. hdr->req is set once at nfs_pgheader_init() time and is not refcount-protected for the lifetime of the loop; when hdr aggregates requests from multiple page groups (common under heavy NFSv3 writeback), a parallel COMMIT on hdr->req's group can drop the last reference and free it while the outer loop is still iterating requests from other groups. KASAN catches this as an 8-byte read at offset +24 of a freed nfs_page slab object (wb_lock_context). All requests in a given pgio share the same open_context, so reading the loop-local req's wb_lock_context yields the same value and is safe -- req is still on hdr->pages and holds its writeback kref through the commit branch. Caught with kasan: BUG: KASAN: slab-use-after-free in nfs_write_completion+0x8f8/0xa50 [nfs] Read of size 8 at addr ffff888118af2058 by task kworker/u16:16/122062 CPU: 2 UID: 0 PID: 122062 Comm: kworker/u16:16 Kdump: loaded Not tainted 7.1.0-rc4+ #ge05a759574b2 PREEMPT Workqueue: nfsiod rpc_async_release Call Trace: dump_stack_lvl+0xaf/0x100 ? nfs_write_completion+0x8f8/0xa50 [nfs] print_report+0x157/0x4a1 ? __virt_addr_valid+0x1fb/0x400 ? nfs_write_completion+0x8f8/0xa50 [nfs] kasan_report+0xc2/0x190 ? nfs_write_completion+0x8f8/0xa50 [nfs] nfs_write_completion+0x8f8/0xa50 [nfs] ? nfs_commit_release_pages+0xbd0/0xbd0 [nfs] ? lock_acquire+0x182/0x2e0 ? process_one_work+0x937/0x1890 ? nfs_pgio_header_alloc+0xd0/0xd0 [nfs] rpc_free_task+0xee/0x160 rpc_async_release+0x5d/0xb0 process_one_work+0x9b0/0x1890 ? pwq_dec_nr_in_flight+0xed0/0xed0 ? rpc_final_put_task+0x140/0x140 worker_thread+0x75a/0x10a0 ? process_one_work+0x1890/0x1890 ? kthread+0x1af/0x4d0 ? process_one_work+0x1890/0x1890 kthread+0x3d3/0x4d0 ? kthread_affine_node+0x2c0/0x2c0 ret_from_fork+0x669/0xa50 ? native_tss_update_io_bitmap+0x660/0x660 ? __switch_to+0x9dd/0x1310 ? kthread_affine_node+0x2c0/0x2c0 ret_from_fork_asm+0x11/0x20 Allocated by task 121997 on cpu 3 at 31643.290294s: kasan_save_stack+0x1e/0x40 kasan_save_track+0x13/0x60 __kasan_slab_alloc+0x62/0x70 kmem_cache_alloc_noprof+0x1ab/0x4e0 nfs_page_create+0x152/0x460 [nfs] nfs_page_create_from_folio+0x7e/0x210 [nfs] nfs_update_folio+0x7a9/0x32a0 [nfs] nfs_write_end+0x290/0xc60 [nfs] generic_perform_write+0x4ce/0x990 nfs_file_write+0x6b3/0xce0 [nfs] vfs_write+0x63c/0xfa0 ksys_write+0x122/0x240 do_syscall_64+0xc3/0x13f0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 Freed by task 122046 on cpu 0 at 31647.037964s: kasan_save_stack+0x1e/0x40 kasan_save_track+0x13/0x60 kasan_save_free_info+0x37/0x60 __kasan_slab_free+0x3b/0x60 kmem_cache_free+0x11b/0x5a0 nfs_page_group_destroy+0x13a/0x210 [nfs] nfs_unlock_and_release_request+0x64/0x90 [nfs] nfs_commit_release_pages+0x339/0xbd0 [nfs] nfs_commit_release+0x51/0xb0 [nfs] rpc_free_task+0xee/0x160 rpc_async_release+0x5d/0xb0 process_one_work+0x9b0/0x1890 worker_thread+0x75a/0x10a0 kthread+0x3d3/0x4d0 ret_from_fork+0x669/0xa50 ret_from_fork_asm+0x11/0x20 The buggy address belongs to the object at ffff888118af2040\x0a which belongs to the cache nfs_page of size 96 The buggy address is located 24 bytes inside of\x0a freed 96-byte region [ffff888118af2040, ffff888118af20a0) The buggy address belongs to the physical page: page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x118af2 head: order:1 mapcount:0 entire_mapcount:0 nr_pages_mapped:0 pincount:0 flags: 0x4000000000000040(head|zone=2) page_type: f5(slab) raw: 4000000000000040 ffff88818cf2c4c0 ffffea000e61b990 ffffea0004e7d110 raw: 0000000000000000 0000000800190019 00000000f5000000 0000000000000000 head: 4000000000000040 ffff88818cf2c4c0 ffffea000e61b990 ffffea0004e7d110 head: 0000000000000000 0000000800190019 00000000f5000000 0000000000000000 head: 4000000000000001 ffffffffffffff81 00000000ffffffff 00000000ffffffff head: ffffffffffffffff 0000000000000000 00000000ffffffff 0000000000000002 page dumped because: kasan: bad access detected page_owner tracks the page as allocated page last allocated via order 1, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 121997, tgid 121997 (rsync), ts 31643290274577, free_ts 31642154777182 post_alloc_hook+0xd1/0x100 get_page_from_freelist+0xbad/0x2910 __alloc_frozen_pages_noprof+0x1c6/0x4a0 allocate_slab+0x330/0x620 ___slab_alloc+0xe9/0x930 kmem_cache_alloc_noprof+0x35b/0x4e0 nfs_page_create+0x152/0x460 [nfs] nfs_page_create_from_folio+0x7e/0x210 [nfs] nfs_update_folio+0x7a9/0x32a0 [nfs] nfs_write_end+0x290/0xc60 [nfs] generic_perform_write+0x4ce/0x990 nfs_file_write+0x6b3/0xce0 [nfs] vfs_write+0x63c/0xfa0 ksys_write+0x122/0x240 do_syscall_64+0xc3/0x13f0 entry_SYSCALL_64_after_hwframe+0x4b/0x53 page last free pid 122202 tgid 122202 stack trace: __free_frozen_pages+0x6da/0xf30 qlist_free_all+0x53/0x130 kasan_quarantine_reduce+0x198/0x1f0 __kasan_slab_alloc+0x46/0x70 kmem_cache_alloc_noprof+0x1ab/0x4e0 __alloc_object+0x2f/0x230 __create_object+0x22/0x80 kmem_cache_alloc_node_noprof+0x416/0x4d0 __alloc_skb+0x146/0x6e0 tcp_stream_alloc_skb+0x35/0x660 tcp_sendmsg_locked+0x1746/0x4260 tcp_sendmsg+0x2f/0x40 inet_sendmsg+0x9e/0xe0 __sock_sendmsg+0xd9/0x180 sock_sendmsg+0x122/0x200 xprt_sock_sendmsg+0x4ff/0x9a0 Memory state around the buggy address: ffff888118af1f00: 00 00 00 00 00 00 00 00 00 00 00 00 00 fc fc fc ffff888118af1f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc >ffff888118af2000: fc fc fc fc fc fc fc fc fa fb fb fb fb fb fb fb ^ ffff888118af2080: fb fb fb fb fc fc fc fc fc fc fc fc fc fc fc fc ffff888118af2100: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc ================================================================== Reviewed-by Jeff Layton Fixes: 5d3869a41f36 ("NFS: fix writeback in presence of errors") Cc: Olga Kornievskaia Cc: Trond Myklebust Cc: Anna Schumaker Cc: linux-nfs@vger.kernel.org Assisted-by: Claude:claude-opus-4-7 Signed-off-by: Dave Jones Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 3134bb17f3e3..d7c399763ad9 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -927,7 +927,7 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr) } if (nfs_write_need_commit(hdr)) { struct nfs_open_context *ctx = - hdr->req->wb_lock_context->open_context; + req->wb_lock_context->open_context; /* Reset wb_nio, since the write was successful. */ req->wb_nio = 0; From 6d99479799c69c3cb588fcda19c81d8f61d64ecd Mon Sep 17 00:00:00 2001 From: Qing Wang Date: Tue, 2 Jun 2026 11:08:54 +0800 Subject: [PATCH 133/329] rseq: Fix using an uninitialized stack variable in rseq_exit_user_update() There is an bug in which an uninitialized stack variable is used in rseq_exit_user_update() as reported by syzbot: BUG: KMSAN: kernel-infoleak in rseq_set_ids_get_csaddr include/linux/rseq_entry.h:502 [inline] The local variable: struct rseq_ids ids = { .cpu_id = task_cpu(t), .mm_cid = task_mm_cid(t), .node_id = cpu_to_node(ids.cpu_id), }; According to the C standard, the evaluation order of expressions in an initializer list is indeterminately sequenced. The compiler (Clang, in this KMSAN build) evaluates `cpu_to_node(ids.cpu_id)` *before* `ids.cpu_id` is initialized with `task_cpu(t)`. This is fixed by moving the assignment of ids.node_id outside the structure initialization. Fixes: 82f572449cfe ("rseq: Implement read only ABI enforcement for optimized RSEQ V2 mode") Closes: https://syzkaller.appspot.com/bug?extid=185a631927096f9da2fc Reported-by: syzbot+185a631927096f9da2fc@syzkaller.appspotmail.com Signed-off-by: Qing Wang Signed-off-by: Peter Zijlstra (Intel) Acked-by: Mark Rutland Link: https://patch.msgid.link/20260602030854.574038-1-wangqing7171@gmail.com --- include/linux/rseq_entry.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/linux/rseq_entry.h b/include/linux/rseq_entry.h index 63bc72086e75..ed9da6e41a2a 100644 --- a/include/linux/rseq_entry.h +++ b/include/linux/rseq_entry.h @@ -635,10 +635,11 @@ static __always_inline bool rseq_exit_user_update(struct pt_regs *regs, struct t return true; } + int cpu = task_cpu(t); struct rseq_ids ids = { - .cpu_id = task_cpu(t), + .cpu_id = cpu, .mm_cid = task_mm_cid(t), - .node_id = cpu_to_node(ids.cpu_id), + .node_id = cpu_to_node(cpu), }; return rseq_update_usr(t, regs, &ids); From 4cd92957e8f8cc4ebfe8a5d4203c14c592fde6b1 Mon Sep 17 00:00:00 2001 From: Yuqi Xu Date: Fri, 29 May 2026 23:25:37 +0800 Subject: [PATCH 134/329] wifi: nl80211: reject oversized EMA RNR lists nl80211_parse_rnr_elems() stores the parsed element count in a u8-backed cfg80211_rnr_elems::cnt field and uses that count to size the flexible array allocation. Reject nested NL80211_ATTR_EMA_RNR_ELEMS input once the count reaches 255, before incrementing it again. This keeps the parser aligned with the data structure it fills and matches the existing bound check used by nl80211_parse_mbssid_elems(). Fixes: dbbb27e183b1 ("cfg80211: support RNR for EMA AP") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Zhengchuan Liang Reported-by: Xin Liu Assisted-by: Codex:gpt-5.4 Signed-off-by: Yuqi Xu Signed-off-by: Ren Wei Link: https://patch.msgid.link/20260529152542.1412734-1-n05ec@lzu.edu.cn Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 7db9cd433801..dac2e8643c49 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -6366,6 +6366,9 @@ nl80211_parse_rnr_elems(struct wiphy *wiphy, struct nlattr *attrs, if (ret) return ERR_PTR(ret); + if (num_elems >= 255) + return ERR_PTR(-EINVAL); + num_elems++; } From 6c0cf89f36ac0c0fd8687a4ccdce2efb23a9c663 Mon Sep 17 00:00:00 2001 From: Deepanshu Kartikey Date: Sun, 31 May 2026 06:47:21 +0530 Subject: [PATCH 135/329] wifi: mac80211: limit injected antenna index in ieee80211_parse_tx_radiotap When parsing the radiotap header of an injected frame, ieee80211_parse_tx_radiotap() uses the IEEE80211_RADIOTAP_ANTENNA value directly as a shift count: info->control.antennas |= BIT(*iterator.this_arg); *iterator.this_arg is an 8-bit value taken straight from the frame supplied by userspace, so BIT() can be asked to shift by up to 255. That is undefined behaviour on the unsigned long and is reported by UBSAN: UBSAN: shift-out-of-bounds in net/mac80211/tx.c:2174:30 shift exponent 235 is too large for 64-bit type 'unsigned long' Call Trace: ieee80211_parse_tx_radiotap+0xadb/0x1950 net/mac80211/tx.c:2174 ieee80211_monitor_start_xmit+0xb1f/0x1250 net/mac80211/tx.c:2451 ... packet_sendmsg+0x3eb6/0x50f0 net/packet/af_packet.c:3109 info->control.antennas is a 2-bit bitmap (u8 antennas:2), so only antenna indices 0 and 1 can ever be represented. Ignore any larger value instead of shifting out of bounds. Reported-by: syzbot+8e0622f6d9446420271f@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=8e0622f6d9446420271f Fixes: ef246a1480cc ("wifi: mac80211: support antenna control in injection") Signed-off-by: Deepanshu Kartikey Link: https://patch.msgid.link/20260531011721.102941-1-kartikey406@gmail.com Signed-off-by: Johannes Berg --- net/mac80211/tx.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/mac80211/tx.c b/net/mac80211/tx.c index b487d2330f25..ea7f63e1fc17 100644 --- a/net/mac80211/tx.c +++ b/net/mac80211/tx.c @@ -2181,7 +2181,9 @@ bool ieee80211_parse_tx_radiotap(struct sk_buff *skb, case IEEE80211_RADIOTAP_ANTENNA: /* this can appear multiple times, keep a bitmap */ - info->control.antennas |= BIT(*iterator.this_arg); + /* control.antennas is only a 2-bit bitmap */ + if (*iterator.this_arg < 2) + info->control.antennas |= BIT(*iterator.this_arg); break; case IEEE80211_RADIOTAP_DATA_RETRIES: From 29e8751c1dd278262fb4cd234e8909287d4189d4 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:47 +0200 Subject: [PATCH 136/329] KVM: s390: Fix _gmap_unmap_crste() In _gmap_unmap_crste(), the crste to be unmapped is zapped calling gmap_crstep_xchg_atomic() exactly once, and expecting it to succeed. This is a reasonable sanity check, since kvm->mmu_lock is being held in write mode, and thus no races should be possible. An upcoming patch will change the behaviour of gmap_crstep_xchg_atomic() to return false and clear the vsie_notif bit if the operation triggers an unshadow operation. With the new behaviour, an unmap operation that triggers an unshadow would cause the VM to be killed. Prepare for the change by checking if the vsie_notif bit was set in the old crste if gmap_crstep_xchg_atomic() fails the first time, and try a second time. The second time no failures are allowed. Fixes: b827ef02f409 ("KVM: s390: Remove non-atomic dat_crstep_xchg()") Fixes: a2c17f9270cc ("KVM: s390: New gmap code") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-2-imbrenda@linux.ibm.com> --- arch/s390/kvm/gmap.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) diff --git a/arch/s390/kvm/gmap.c b/arch/s390/kvm/gmap.c index 957126ab991c..52d55ddea8d4 100644 --- a/arch/s390/kvm/gmap.c +++ b/arch/s390/kvm/gmap.c @@ -395,15 +395,28 @@ static long _gmap_unmap_crste(union crste *crstep, gfn_t gfn, gfn_t next, struct struct gmap_unmap_priv *priv = walk->priv; struct folio *folio = NULL; union crste old = *crstep; + bool ok; if (!old.h.fc) return 0; if (old.s.fc1.pr && test_bit(GMAP_FLAG_EXPORT_ON_UNMAP, &priv->gmap->flags)) folio = phys_to_folio(crste_origin_large(old)); - /* No races should happen because kvm->mmu_lock is held in write mode */ - KVM_BUG_ON(!gmap_crstep_xchg_atomic(priv->gmap, crstep, old, _CRSTE_EMPTY(old.h.tt), gfn), - priv->gmap->kvm); + /* + * No races should happen because kvm->mmu_lock is held in write mode, + * but the unmap operation could have triggered an unshadow, which + * causes gmap_crstep_xchg_atomic() to return false and clear the + * vsie_notif bit. Allow the operation to fail once, if the old crste + * had the vsie_notif bit set. A second failure is not allowed, for + * the reasons above. + */ + ok = gmap_crstep_xchg_atomic(priv->gmap, crstep, old, _CRSTE_EMPTY(old.h.tt), gfn); + if (!ok) { + KVM_BUG_ON(!old.s.fc1.vsie_notif, priv->gmap->kvm); + old.s.fc1.vsie_notif = 0; + ok = gmap_crstep_xchg_atomic(priv->gmap, crstep, old, _CRSTE_EMPTY(old.h.tt), gfn); + KVM_BUG_ON(!ok, priv->gmap->kvm); + } if (folio) uv_convert_from_secure_folio(folio); From d1adc098ce08893c92fce3db63f7bb750fbb4c30 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:48 +0200 Subject: [PATCH 137/329] KVM: s390: Fix _gmap_crstep_xchg_atomic() The previous incorrect behaviour cleared the vsie_notif bit without returning false, which allowed shadow crstes to be installed without the vsie_notif bit. Return false and do not perform the operation if an unshadow event has been triggered, but still attempt to clear the vsie_notif bit from the existing crste. This will prevent the installation of shadow crstes without vsie_notif bit and will also prevent the caller from looping forever if it was not checking for the sg->invalidated flag. Fixes: b827ef02f409 ("KVM: s390: Remove non-atomic dat_crstep_xchg()") Fixes: a2c17f9270cc ("KVM: s390: New gmap code") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-3-imbrenda@linux.ibm.com> --- arch/s390/kvm/gmap.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/arch/s390/kvm/gmap.h b/arch/s390/kvm/gmap.h index 742e42a31744..5374f21aaf8d 100644 --- a/arch/s390/kvm/gmap.h +++ b/arch/s390/kvm/gmap.h @@ -273,11 +273,14 @@ static inline bool __must_check _gmap_crstep_xchg_atomic(struct gmap *gmap, unio gmap_unmap_prefix(gmap, gfn, gfn + align); } if (crste_leaf(oldcrste) && crste_needs_unshadow(oldcrste, newcrste)) { + newcrste = oldcrste; newcrste.s.fc1.vsie_notif = 0; if (needs_lock) gmap_handle_vsie_unshadow_event(gmap, gfn); else _gmap_handle_vsie_unshadow_event(gmap, gfn); + dat_crstep_xchg_atomic(crstep, oldcrste, newcrste, gfn, gmap->asce); + return false; } if (!oldcrste.s.fc1.d && newcrste.s.fc1.d && !newcrste.s.fc1.s) SetPageDirty(phys_to_page(crste_origin_large(newcrste))); From 89fa757931dc0bcd64ef22b28d1d5ad00c5d02f4 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:49 +0200 Subject: [PATCH 138/329] KVM: s390: Avoid potentially sleeping while atomic when zapping pages Factor out try_get_locked_pte(), which behaves similarly to get_locked_pte(), but does not attempt to allocate missing tables and performs a spin_trylock() instead of blocking. The new function is also exported, since it will be used in other patches. If intermediate entries are missing, there can be no pte swap entry to free, so it's safe to ignore them. This avoids potentially sleeping while atomic. Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-4-imbrenda@linux.ibm.com> --- arch/s390/include/asm/gmap_helpers.h | 1 + arch/s390/mm/gmap_helpers.c | 117 ++++++++++++++++----------- 2 files changed, 73 insertions(+), 45 deletions(-) diff --git a/arch/s390/include/asm/gmap_helpers.h b/arch/s390/include/asm/gmap_helpers.h index 2d3ae421077e..d2b616604a46 100644 --- a/arch/s390/include/asm/gmap_helpers.h +++ b/arch/s390/include/asm/gmap_helpers.h @@ -12,5 +12,6 @@ void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr); void gmap_helper_discard(struct mm_struct *mm, unsigned long vmaddr, unsigned long end); int gmap_helper_disable_cow_sharing(void); void gmap_helper_try_set_pte_unused(struct mm_struct *mm, unsigned long vmaddr); +pte_t *try_get_locked_pte(struct mm_struct *mm, unsigned long addr, spinlock_t **ptl); #endif /* _ASM_S390_GMAP_HELPERS_H */ diff --git a/arch/s390/mm/gmap_helpers.c b/arch/s390/mm/gmap_helpers.c index f8789ffcc05c..396207163ca6 100644 --- a/arch/s390/mm/gmap_helpers.c +++ b/arch/s390/mm/gmap_helpers.c @@ -34,6 +34,70 @@ static void ptep_zap_softleaf_entry(struct mm_struct *mm, softleaf_t entry) swap_put_entries_direct(entry, 1); } +/** + * try_get_locked_pte() - like get_locked_pte(), but atomic and with trylock + * @mm: the mm + * @vmaddr: the userspace virtual address whose pte is to be found + * @ptl: will be set to the pointer to the lock used to lock the pte in case + * of success. + * + * This function returns the pointer to the pte corresponding to @addr in @mm, + * similarly to get_locked_pte(). Unlike get_locked_pte(), no attempt is made + * to allocate missing page tables. If a missing or large entry is found, the + * function will return NULL. If the ptl lock is contended, %-EAGAIN is + * returned. + * + * In case of success, *@ptl will point to the locked pte lock for the returned + * pte, like get_locked_pte() does. + * + * Context: mmap_lock or vma lock for read or for write needs to be held. + * Return: + * * %NULL if the pte cannot be reached. + * * %-EAGAIN if the pte can be reached, but cannot be locked. + * * the pointer to the pte corresponding to @addr in @mm, if it can be reached + * and locked. + */ +pte_t *try_get_locked_pte(struct mm_struct *mm, unsigned long vmaddr, spinlock_t **ptl) +{ + pmd_t *pmdp, pmd, pmdval; + pud_t *pudp, pud; + p4d_t *p4dp, p4d; + pgd_t *pgdp, pgd; + pte_t *ptep; + + pgdp = pgd_offset(mm, vmaddr); + pgd = pgdp_get(pgdp); + if (pgd_none(pgd) || !pgd_present(pgd)) + return NULL; + p4dp = p4d_offset(pgdp, vmaddr); + p4d = p4dp_get(p4dp); + if (p4d_none(p4d) || !p4d_present(p4d)) + return NULL; + pudp = pud_offset(p4dp, vmaddr); + pud = pudp_get(pudp); + if (pud_none(pud) || pud_leaf(pud) || !pud_present(pud)) + return NULL; + pmdp = pmd_offset(pudp, vmaddr); + pmd = pmdp_get_lockless(pmdp); + if (pmd_none(pmd) || pmd_leaf(pmd) || !pmd_present(pmd)) + return NULL; + ptep = pte_offset_map_rw_nolock(mm, pmdp, vmaddr, &pmdval, ptl); + if (!ptep) + return NULL; + + if (spin_trylock(*ptl)) { + if (unlikely(!pmd_same(pmdval, pmdp_get_lockless(pmdp)))) { + pte_unmap_unlock(ptep, *ptl); + return ERR_PTR(-EAGAIN); + } + return ptep; + } + + pte_unmap(ptep); + return ERR_PTR(-EAGAIN); +} +EXPORT_SYMBOL_GPL(try_get_locked_pte); + /** * gmap_helper_zap_one_page() - discard a page if it was swapped. * @mm: the mm @@ -46,7 +110,7 @@ static void ptep_zap_softleaf_entry(struct mm_struct *mm, softleaf_t entry) void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr) { struct vm_area_struct *vma; - spinlock_t *ptl; + spinlock_t *ptl; /* Lock for the host (userspace) page table */ pte_t *ptep; mmap_assert_locked(mm); @@ -57,8 +121,8 @@ void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr) return; /* Get pointer to the page table entry */ - ptep = get_locked_pte(mm, vmaddr, &ptl); - if (unlikely(!ptep)) + ptep = try_get_locked_pte(mm, vmaddr, &ptl); + if (IS_ERR_OR_NULL(ptep)) return; if (pte_swap(*ptep)) { ptep_zap_softleaf_entry(mm, softleaf_from_pte(*ptep)); @@ -113,37 +177,9 @@ EXPORT_SYMBOL_GPL(gmap_helper_discard); */ void gmap_helper_try_set_pte_unused(struct mm_struct *mm, unsigned long vmaddr) { - pmd_t *pmdp, pmd, pmdval; - pud_t *pudp, pud; - p4d_t *p4dp, p4d; - pgd_t *pgdp, pgd; spinlock_t *ptl; /* Lock for the host (userspace) page table */ pte_t *ptep; - pgdp = pgd_offset(mm, vmaddr); - pgd = pgdp_get(pgdp); - if (pgd_none(pgd) || !pgd_present(pgd)) - return; - - p4dp = p4d_offset(pgdp, vmaddr); - p4d = p4dp_get(p4dp); - if (p4d_none(p4d) || !p4d_present(p4d)) - return; - - pudp = pud_offset(p4dp, vmaddr); - pud = pudp_get(pudp); - if (pud_none(pud) || pud_leaf(pud) || !pud_present(pud)) - return; - - pmdp = pmd_offset(pudp, vmaddr); - pmd = pmdp_get_lockless(pmdp); - if (pmd_none(pmd) || pmd_leaf(pmd) || !pmd_present(pmd)) - return; - - ptep = pte_offset_map_rw_nolock(mm, pmdp, vmaddr, &pmdval, &ptl); - if (!ptep) - return; - /* * Several paths exists that takes the ptl lock and then call the * mmu_notifier, which takes the mmu_lock. The unmap path, instead, @@ -156,21 +192,12 @@ void gmap_helper_try_set_pte_unused(struct mm_struct *mm, unsigned long vmaddr) * If the lock is contended the bit is not set and the deadlock is * avoided. */ - if (spin_trylock(ptl)) { - /* - * Make sure the pte we are touching is still the correct - * one. In theory this check should not be needed, but - * better safe than sorry. - * Disabling interrupts or holding the mmap lock is enough to - * guarantee that no concurrent updates to the page tables - * are possible. - */ - if (likely(pmd_same(pmdval, pmdp_get_lockless(pmdp)))) - __atomic64_or(_PAGE_UNUSED, (long *)ptep); - spin_unlock(ptl); - } + ptep = try_get_locked_pte(mm, vmaddr, &ptl); + if (IS_ERR_OR_NULL(ptep)) + return; - pte_unmap(ptep); + __atomic64_or(_PAGE_UNUSED, (long *)ptep); + pte_unmap_unlock(ptep, ptl); } EXPORT_SYMBOL_GPL(gmap_helper_try_set_pte_unused); From ca42c16638d5570c44842ff68a772c62e6dd0124 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:50 +0200 Subject: [PATCH 139/329] KVM: s390: Fix guest / virtual address confusion in _essa_clear_cbrl() Until now, gmap_helper_zap_one_page() was being called with the guest absolute address, but it expects a userspace virtual address. This meant that in the best case the requested pages were not being discarded, and in the worst case that the wrong pages were being discarded. Fix this by converting the guest absolute address to host virtual before passing it to gmap_helper_zap_one_page(). Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-5-imbrenda@linux.ibm.com> --- arch/s390/kvm/priv.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index cc0553da14cb..447ec7ed423d 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -1188,6 +1188,7 @@ static void _essa_clear_cbrl(struct kvm_vcpu *vcpu, unsigned long *cbrl, int len union crste *crstep; union pgste pgste; union pte *ptep; + hva_t hva; int i; lockdep_assert_held(&vcpu->kvm->mmu_lock); @@ -1199,8 +1200,11 @@ static void _essa_clear_cbrl(struct kvm_vcpu *vcpu, unsigned long *cbrl, int len if (!ptep || ptep->s.pr) continue; pgste = pgste_get_lock(ptep); - if (pgste.usage == PGSTE_GPS_USAGE_UNUSED || pgste.zero) - gmap_helper_zap_one_page(vcpu->kvm->mm, cbrl[i]); + if (pgste.usage == PGSTE_GPS_USAGE_UNUSED || pgste.zero) { + hva = gpa_to_hva(vcpu->kvm, cbrl[i]); + if (!kvm_is_error_hva(hva)) + gmap_helper_zap_one_page(vcpu->kvm->mm, hva); + } pgste_set_unlock(ptep, pgste); } } From 6af5563e827913d3e14dbf12eefd5af4aa592739 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:51 +0200 Subject: [PATCH 140/329] KVM: s390: vsie: Fix rmap handling in _do_shadow_crste() Fix _do_shadow_crste() to also apply a mask on the reverse address, to prevent spurious entries from being created, like already done in gmap_protect_rmap(). Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-6-imbrenda@linux.ibm.com> --- arch/s390/kvm/gaccess.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 4f8d5592c9a9..20e28b183c1a 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -1466,15 +1466,17 @@ static int _do_shadow_crste(struct gmap *sg, gpa_t raddr, union crste *host, uni struct guest_fault *f, bool p) { union crste newcrste, oldcrste; - gfn_t gfn; + unsigned long mask; + gfn_t r_gfn; int rc; lockdep_assert_held(&sg->kvm->mmu_lock); lockdep_assert_held(&sg->parent->children_lock); - gfn = f->gfn & (is_pmd(*table) ? _SEGMENT_FR_MASK : _REGION3_FR_MASK); + mask = is_pmd(*table) ? _SEGMENT_FR_MASK : _REGION3_FR_MASK; + r_gfn = gpa_to_gfn(raddr) & mask; scoped_guard(spinlock, &sg->host_to_rmap_lock) - rc = gmap_insert_rmap(sg, gfn, gpa_to_gfn(raddr), host->h.tt); + rc = gmap_insert_rmap(sg, f->gfn & mask, r_gfn, host->h.tt); if (rc) return rc; @@ -1497,8 +1499,7 @@ static int _do_shadow_crste(struct gmap *sg, gpa_t raddr, union crste *host, uni return -EAGAIN; newcrste = _crste_fc1(f->pfn, oldcrste.h.tt, 0, !p); - gfn = gpa_to_gfn(raddr); - while (!dat_crstep_xchg_atomic(table, READ_ONCE(*table), newcrste, gfn, sg->asce)) + while (!dat_crstep_xchg_atomic(table, READ_ONCE(*table), newcrste, r_gfn, sg->asce)) ; return 0; } From 9a1dfbbd3506c4f7159feab5ef27434e80256fa5 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:52 +0200 Subject: [PATCH 141/329] KVM: s390: Fix fault-in code Fix the fault-in code so that it does not return success if a concurrent unmap event invalidated the fault-in process between the best-effort lockless check and the proper check with lock. The new behaviour is to retry, like the best-effort lockless check already did. This prevents the fault-in handler from returning success without having actually faulted in the requested page. Fixes: e907ae530133 ("KVM: s390: Add helper functions for fault handling") Reviewed-by: Steffen Eiden Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-7-imbrenda@linux.ibm.com> --- arch/s390/kvm/faultin.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/arch/s390/kvm/faultin.c b/arch/s390/kvm/faultin.c index ddf0ca71f374..cf542b0a7e8e 100644 --- a/arch/s390/kvm/faultin.c +++ b/arch/s390/kvm/faultin.c @@ -36,7 +36,8 @@ int kvm_s390_faultin_gfn(struct kvm_vcpu *vcpu, struct kvm *kvm, struct guest_fa struct kvm_s390_mmu_cache *mc = NULL; struct kvm_memory_slot *slot; unsigned long inv_seq; - int foll, rc = 0; + int rc = -EAGAIN; + int foll; foll = f->write_attempt ? FOLL_WRITE : 0; foll |= f->attempt_pfault ? FOLL_NOWAIT : 0; @@ -53,7 +54,7 @@ int kvm_s390_faultin_gfn(struct kvm_vcpu *vcpu, struct kvm *kvm, struct guest_fa return 0; } - while (1) { + while (rc == -EAGAIN) { f->valid = false; inv_seq = kvm->mmu_invalidate_seq; /* Pairs with the smp_wmb() in kvm_mmu_invalidate_end(). */ @@ -110,20 +111,19 @@ int kvm_s390_faultin_gfn(struct kvm_vcpu *vcpu, struct kvm *kvm, struct guest_fa if (!mmu_invalidate_retry_gfn(kvm, inv_seq, f->gfn)) { f->valid = true; rc = gmap_link(mc, kvm->arch.gmap, f, slot); - kvm_release_faultin_page(kvm, f->page, !!rc, f->write_attempt); - f->page = NULL; } + kvm_release_faultin_page(kvm, f->page, !!rc, f->write_attempt); } - kvm_release_faultin_page(kvm, f->page, true, false); if (rc == -ENOMEM) { rc = kvm_s390_mmu_cache_topup(mc); if (rc) return rc; - } else if (rc != -EAGAIN) { - return rc; + rc = -EAGAIN; } } + + return rc; } int kvm_s390_get_guest_page(struct kvm *kvm, struct guest_fault *f, gfn_t gfn, bool w) From 42546fc642e929de07459ff839a9f43a653ffb4e Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:53 +0200 Subject: [PATCH 142/329] KVM: s390: Lock pte when making page secure Make sure _kvm_s390_pv_make_secure() takes the pte lock for the given address when attempting to make the page secure. One of the steps in making the page secure is freezing the folio using folio_ref_freeze(), which temporarily sets the reference count to 0. Any attempt to get such a folio while frozen will fail and cause a warning to be printed. Other users of folio_ref_freeze() make sure that the page is not mapped while it's being frozen, thus preventing gup functions from being able to access it. For _kvm_s390_pv_make_secure(), this is not possible, because the page needs to be mapped in order for the import to succeed. By taking the pte lock, gup functions will be blocked until the import operation is done, thus avoiding the race. In theory this does not completely solve the issue: if a page is mapped through multiple mappings, locking one pte does not protect from calling gup on it through the other mapping. In practice this does not happen and it is a decent stopgap solution until a more correct solution is available. Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-8-imbrenda@linux.ibm.com> --- arch/s390/kvm/pv.c | 21 ++++++++++++++++++++- 1 file changed, 20 insertions(+), 1 deletion(-) diff --git a/arch/s390/kvm/pv.c b/arch/s390/kvm/pv.c index c2dafd812a3b..4b865e75351c 100644 --- a/arch/s390/kvm/pv.c +++ b/arch/s390/kvm/pv.c @@ -17,6 +17,7 @@ #include #include #include +#include #include "kvm-s390.h" #include "dat.h" #include "gaccess.h" @@ -73,6 +74,7 @@ static bool should_export_before_import(struct uv_cb_header *uvcb, struct mm_str struct pv_make_secure { void *uvcb; struct folio *folio; + struct kvm *kvm; int rc; bool needs_export; }; @@ -103,9 +105,21 @@ static void _kvm_s390_pv_make_secure(struct guest_fault *f) { struct pv_make_secure *priv = f->priv; struct folio *folio; + spinlock_t *ptl; /* pte lock from try_get_locked_pte() */ + pte_t *ptep; folio = pfn_folio(f->pfn); priv->rc = -EAGAIN; + + if (!mmap_read_trylock(priv->kvm->mm)) + return; + + ptep = try_get_locked_pte(priv->kvm->mm, gfn_to_hva(priv->kvm, f->gfn), &ptl); + if (IS_ERR_VALUE(ptep)) { + priv->rc = PTR_ERR(ptep); + goto out; + } + if (folio_trylock(folio)) { priv->rc = __kvm_s390_pv_make_secure(f, folio); if (priv->rc == -E2BIG || priv->rc == -EBUSY) { @@ -114,6 +128,11 @@ static void _kvm_s390_pv_make_secure(struct guest_fault *f) } folio_unlock(folio); } + + if (ptep) + pte_unmap_unlock(ptep, ptl); +out: + mmap_read_unlock(priv->kvm->mm); } /** @@ -127,7 +146,7 @@ static void _kvm_s390_pv_make_secure(struct guest_fault *f) */ int kvm_s390_pv_make_secure(struct kvm *kvm, unsigned long gaddr, void *uvcb) { - struct pv_make_secure priv = { .uvcb = uvcb }; + struct pv_make_secure priv = { .uvcb = uvcb, .kvm = kvm, }; struct guest_fault f = { .write_attempt = true, .gfn = gpa_to_gfn(gaddr), From 7b53b4801884d48551f26a17f00cdc0ae4640d64 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:54 +0200 Subject: [PATCH 143/329] KVM: s390: Prevent memslots outside the ASCE range With KVM_S390_VM_MEM_LIMIT_SIZE, userspace can set the highest address allowed for the VM. Creating a memslot that lies over the maximum address does not make sense and is only a potential source of bugs. Prevent creation of memslots over the maximum address, and prevent the maximum address from being reduced below the end of existing memslots. Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-9-imbrenda@linux.ibm.com> --- arch/s390/kvm/kvm-s390.c | 33 ++++++++++++++++++++++++++++----- 1 file changed, 28 insertions(+), 5 deletions(-) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index e09960c2e6ed..ffb20a64d328 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -999,7 +999,10 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att break; } case KVM_S390_VM_MEM_LIMIT_SIZE: { + struct kvm_memslots *slots; + struct kvm_memory_slot *ms; unsigned long new_limit; + int bkt; if (kvm_is_ucontrol(kvm)) return -EINVAL; @@ -1007,6 +1010,9 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att if (get_user(new_limit, (u64 __user *)attr->addr)) return -EFAULT; + guard(mutex)(&kvm->lock); + + new_limit = ALIGN(new_limit, HPAGE_SIZE); if (kvm->arch.mem_limit != KVM_S390_NO_MEM_LIMIT && new_limit > kvm->arch.mem_limit) return -E2BIG; @@ -1014,12 +1020,27 @@ static int kvm_s390_set_mem_control(struct kvm *kvm, struct kvm_device_attr *att if (!new_limit) return -EINVAL; - ret = -EBUSY; - if (!kvm->created_vcpus) - ret = gmap_set_limit(kvm->arch.gmap, gpa_to_gfn(new_limit)); + if (kvm->created_vcpus) + return -EBUSY; + + ret = 0; + scoped_guard(mutex, &kvm->slots_lock) { + slots = kvm_memslots(kvm); + if (slots && !kvm_memslots_empty(slots)) { + kvm_for_each_memslot(ms, bkt, slots) { + if (gpa_to_gfn(new_limit) < ms->base_gfn + ms->npages) { + ret = -EBUSY; + break; + } + } + } + if (!ret) + ret = gmap_set_limit(kvm->arch.gmap, gpa_to_gfn(new_limit)); + } + if (ret) + break; VM_EVENT(kvm, 3, "SET: max guest address: %lu", new_limit); - VM_EVENT(kvm, 3, "New guest asce: 0x%p", - (void *)kvm->arch.gmap->asce.val); + VM_EVENT(kvm, 3, "New guest asce: 0x%p", (void *)kvm->arch.gmap->asce.val); break; } default: @@ -5672,6 +5693,8 @@ int kvm_arch_prepare_memory_region(struct kvm *kvm, return -EINVAL; if ((new->base_gfn + new->npages) * PAGE_SIZE > kvm->arch.mem_limit) return -EINVAL; + if (!asce_contains_gfn(kvm->arch.gmap->asce, new->base_gfn + new->npages - 1)) + return -EINVAL; } if (!kvm->arch.migration_mode) From 6ae67dcac742529210a23dac6c9a7de1acfb52a3 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:55 +0200 Subject: [PATCH 144/329] KVM: s390: Fix possible reference leak in fault-in code If kvm_s390_new_mmu_cache() fails, kvm_s390_faultin_gfn() returns without releasing the faulted page. Fix this by moving the allocation of the memory cache outside of the loop. There is no reason to check at every iteration. Opportunistically fix a comment. Fixes: e907ae530133 ("KVM: s390: Add helper functions for fault handling") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-10-imbrenda@linux.ibm.com> --- arch/s390/kvm/faultin.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/arch/s390/kvm/faultin.c b/arch/s390/kvm/faultin.c index cf542b0a7e8e..fee80047bd94 100644 --- a/arch/s390/kvm/faultin.c +++ b/arch/s390/kvm/faultin.c @@ -54,6 +54,13 @@ int kvm_s390_faultin_gfn(struct kvm_vcpu *vcpu, struct kvm *kvm, struct guest_fa return 0; } + if (!mc) { + local_mc = kvm_s390_new_mmu_cache(); + if (!local_mc) + return -ENOMEM; + mc = local_mc; + } + while (rc == -EAGAIN) { f->valid = false; inv_seq = kvm->mmu_invalidate_seq; @@ -94,14 +101,7 @@ int kvm_s390_faultin_gfn(struct kvm_vcpu *vcpu, struct kvm *kvm, struct guest_fa if (is_error_pfn(f->pfn)) return -EFAULT; - if (!mc) { - local_mc = kvm_s390_new_mmu_cache(); - if (!local_mc) - return -ENOMEM; - mc = local_mc; - } - - /* Loop, will automatically release the faulted page. */ + /* Loop, release the faulted page. */ if (mmu_invalidate_retry_gfn_unsafe(kvm, inv_seq, f->gfn)) { kvm_release_faultin_page(kvm, f->page, true, false); continue; From c1edda54a0f713412f5914f9c9080856694bddca Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Tue, 2 Jun 2026 16:23:56 +0200 Subject: [PATCH 145/329] KVM: s390: Remove ptep_zap_softleaf_entry() Migration entries do not need to be removed. The swap subsystem has been (and still is being) heavily reworked. The current implementation of ptep_zap_softleaf_entry() has been slowly modified and is now wrong, since it unconditionally calls swap_put_entries_direct() for both swap and migration entries. Remove ptep_zap_softleaf_entry() altogether, merge the path for proper swap entries directly in the only caller, and ignore migration entries. Fixes: 200197908dc4 ("KVM: s390: Refactor and split some gmap helpers") Signed-off-by: Claudio Imbrenda Message-ID: <20260602142356.169458-11-imbrenda@linux.ibm.com> --- arch/s390/mm/gmap_helpers.c | 25 +++++-------------------- 1 file changed, 5 insertions(+), 20 deletions(-) diff --git a/arch/s390/mm/gmap_helpers.c b/arch/s390/mm/gmap_helpers.c index 396207163ca6..1cfe4724fbe2 100644 --- a/arch/s390/mm/gmap_helpers.c +++ b/arch/s390/mm/gmap_helpers.c @@ -16,24 +16,6 @@ #include #include -/** - * ptep_zap_softleaf_entry() - discard a software leaf entry. - * @mm: the mm - * @entry: the software leaf entry that needs to be zapped - * - * Discards the given software leaf entry. If the leaf entry was an actual - * swap entry (and not a migration entry, for example), the actual swapped - * page is also discarded from swap. - */ -static void ptep_zap_softleaf_entry(struct mm_struct *mm, softleaf_t entry) -{ - if (softleaf_is_swap(entry)) - dec_mm_counter(mm, MM_SWAPENTS); - else if (softleaf_is_migration(entry)) - dec_mm_counter(mm, mm_counter(softleaf_to_folio(entry))); - swap_put_entries_direct(entry, 1); -} - /** * try_get_locked_pte() - like get_locked_pte(), but atomic and with trylock * @mm: the mm @@ -111,6 +93,7 @@ void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr) { struct vm_area_struct *vma; spinlock_t *ptl; /* Lock for the host (userspace) page table */ + softleaf_t sl; pte_t *ptep; mmap_assert_locked(mm); @@ -124,8 +107,10 @@ void gmap_helper_zap_one_page(struct mm_struct *mm, unsigned long vmaddr) ptep = try_get_locked_pte(mm, vmaddr, &ptl); if (IS_ERR_OR_NULL(ptep)) return; - if (pte_swap(*ptep)) { - ptep_zap_softleaf_entry(mm, softleaf_from_pte(*ptep)); + sl = softleaf_from_pte(*ptep); + if (pte_swap(*ptep) && softleaf_is_swap(sl)) { + dec_mm_counter(mm, MM_SWAPENTS); + swap_put_entries_direct(sl, 1); pte_clear(mm, vmaddr, ptep); } pte_unmap_unlock(ptep, ptl); From 5629eec1a2829871d496f3042884cdc267612f6a Mon Sep 17 00:00:00 2001 From: "Mario Limonciello (AMD)" Date: Sat, 30 May 2026 17:04:34 +0200 Subject: [PATCH 146/329] cpufreq/amd-pstate: Fix setting EPP in performance mode EPP 0 is the only supported value in the performance policy. commit 798c47593cca ("cpufreq/amd-pstate: Add support for platform profile class") changed this while adding platform profile support to the dynamic EPP feature, but this actually wasn't necessary since platform profile writes disable manual EPP writes. Restore allowing writing EPP of 0 when in performance mode. Reviewed-by: Marco Scardovi Tested-by: Marco Scardovi Reported-by: Stuart Meckle Closes: https://bugzilla.kernel.org/show_bug.cgi?id=221473 Closes: https://gitlab.freedesktop.org/upower/power-profiles-daemon/-/work_items/190 Fixes: 798c47593cca ("cpufreq/amd-pstate: Add support for platform profile class") Signed-off-by: Mario Limonciello (AMD) --- drivers/cpufreq/amd-pstate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/cpufreq/amd-pstate.c b/drivers/cpufreq/amd-pstate.c index 62b5d995281d..72df461e7b39 100644 --- a/drivers/cpufreq/amd-pstate.c +++ b/drivers/cpufreq/amd-pstate.c @@ -1428,7 +1428,7 @@ ssize_t store_energy_performance_preference(struct cpufreq_policy *policy, epp = cpudata->epp_default_dc; } - if (cpudata->policy == CPUFREQ_POLICY_PERFORMANCE) { + if (epp > 0 && cpudata->policy == CPUFREQ_POLICY_PERFORMANCE) { pr_debug("EPP cannot be set under performance policy\n"); return -EBUSY; } From b230b57bd6f242aaac3b8e62c7d18c69e1e30392 Mon Sep 17 00:00:00 2001 From: Yonatan Nachum Date: Tue, 26 May 2026 08:15:36 +0000 Subject: [PATCH 147/329] RDMA/efa: Validate SQ ring size against max LLQ size Validate the SQ ring size against the device's max LLQ size. This ensures that when using 128-byte WQEs, userspace cannot exceed the queue limits. On create QP, userspace provides the SQ ring size (depth x WQE size) which is validated against the max LLQ size. Fixes: 40909f664d27 ("RDMA/efa: Add EFA verbs implementation") Link: https://patch.msgid.link/r/20260526081536.1203553-1-ynachum@amazon.com Reviewed-by: Michael Margolin Signed-off-by: Yonatan Nachum Signed-off-by: Jason Gunthorpe --- drivers/infiniband/hw/efa/efa_verbs.c | 27 ++++++++++++++++++--------- 1 file changed, 18 insertions(+), 9 deletions(-) diff --git a/drivers/infiniband/hw/efa/efa_verbs.c b/drivers/infiniband/hw/efa/efa_verbs.c index 7bd0838ebc99..9b2b652800e4 100644 --- a/drivers/infiniband/hw/efa/efa_verbs.c +++ b/drivers/infiniband/hw/efa/efa_verbs.c @@ -613,7 +613,8 @@ static int qp_mmap_entries_setup(struct efa_qp *qp, } static int efa_qp_validate_cap(struct efa_dev *dev, - struct ib_qp_init_attr *init_attr) + struct ib_qp_init_attr *init_attr, + u32 sq_ring_size) { if (init_attr->cap.max_send_wr > dev->dev_attr.max_sq_depth) { ibdev_dbg(&dev->ibdev, @@ -622,6 +623,14 @@ static int efa_qp_validate_cap(struct efa_dev *dev, dev->dev_attr.max_sq_depth); return -EINVAL; } + + if (sq_ring_size > dev->dev_attr.max_llq_size) { + ibdev_dbg(&dev->ibdev, + "qp: requested sq ring size[%u] exceeds the max[%u]\n", + sq_ring_size, dev->dev_attr.max_llq_size); + return -EINVAL; + } + if (init_attr->cap.max_recv_wr > dev->dev_attr.max_rq_depth) { ibdev_dbg(&dev->ibdev, "qp: requested receive wr[%u] exceeds the max[%u]\n", @@ -691,14 +700,6 @@ int efa_create_qp(struct ib_qp *ibqp, struct ib_qp_init_attr *init_attr, ucontext = rdma_udata_to_drv_context(udata, struct efa_ucontext, ibucontext); - err = efa_qp_validate_cap(dev, init_attr); - if (err) - goto err_out; - - err = efa_qp_validate_attr(dev, init_attr); - if (err) - goto err_out; - err = ib_copy_validate_udata_in_cm(udata, cmd, driver_qp_type, 0); if (err) goto err_out; @@ -720,6 +721,14 @@ int efa_create_qp(struct ib_qp *ibqp, struct ib_qp_init_attr *init_attr, goto err_out; } + err = efa_qp_validate_cap(dev, init_attr, cmd.sq_ring_size); + if (err) + goto err_out; + + err = efa_qp_validate_attr(dev, init_attr); + if (err) + goto err_out; + create_qp_params.uarn = ucontext->uarn; create_qp_params.pd = to_epd(ibqp->pd)->pdn; From 73bf3cca7de6a73f53b6a52dc3b1c82ae5667a4d Mon Sep 17 00:00:00 2001 From: Oscar Maes Date: Thu, 28 May 2026 16:03:20 +0200 Subject: [PATCH 148/329] pcnet32: stop holding device spin lock during napi_complete_done napi_complete_done may call gro_flush_normal (though not currently, as GRO is unsupported at the moment), which may result in packet TX. This will eventually result in calling pcnet32_start_xmit - resulting in a deadlock while trying to re-acquire the already locked spin lock. It is safe to split the spinlock block into two, because the hardware registers are still protected from concurrent access, and the two blocks perform unrelated operations that don't need to happen atomically. Fixes: 5b2ec6f2be51 ("pcnet32: use napi_complete_done()") Reviewed-by: Andrew Lunn Signed-off-by: Oscar Maes Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260528140320.5556-1-oscmaes92@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/amd/pcnet32.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/drivers/net/ethernet/amd/pcnet32.c b/drivers/net/ethernet/amd/pcnet32.c index 911808ab13a7..4f3076d4ea34 100644 --- a/drivers/net/ethernet/amd/pcnet32.c +++ b/drivers/net/ethernet/amd/pcnet32.c @@ -1407,8 +1407,10 @@ static int pcnet32_poll(struct napi_struct *napi, int budget) pcnet32_restart(dev, CSR0_START); netif_wake_queue(dev); } + spin_unlock_irqrestore(&lp->lock, flags); if (work_done < budget && napi_complete_done(napi, work_done)) { + spin_lock_irqsave(&lp->lock, flags); /* clear interrupt masks */ val = lp->a->read_csr(ioaddr, CSR3); val &= 0x00ff; @@ -1416,9 +1418,9 @@ static int pcnet32_poll(struct napi_struct *napi, int budget) /* Set interrupt enable. */ lp->a->write_csr(ioaddr, CSR0, CSR0_INTEN); + spin_unlock_irqrestore(&lp->lock, flags); } - spin_unlock_irqrestore(&lp->lock, flags); return work_done; } From b748765019fe9e9234660327090fc1a9665cdbdd Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Fri, 29 May 2026 19:39:23 +0000 Subject: [PATCH 149/329] net: Annotate sk->sk_write_space() for UDP SOCKMAP. UDP TX skb->destructor() is sock_wfree(), and UDP holds lock_sock() only for UDP_CORK / MSG_MORE sendmsg(). Otherwise, sk->sk_write_space() may be read locklessly while SOCKMAP rewrites sk->sk_write_space(). Let's use WRITE_ONCE() and READ_ONCE() for sk->sk_write_space(). Note that the write side is annotated by commit 2ef2b20cf4e0 ("net: annotate data-races around sk->sk_{data_ready,write_space}"). Fixes: 7b98cd42b049 ("bpf: sockmap: Add UDP support") Signed-off-by: Kuniyuki Iwashima Reviewed-by: Jakub Sitnicki Link: https://patch.msgid.link/20260529193941.3897256-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/core/sock.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/core/sock.c b/net/core/sock.c index b37b664b6eb9..d097025c116a 100644 --- a/net/core/sock.c +++ b/net/core/sock.c @@ -2676,8 +2676,12 @@ void sock_wfree(struct sk_buff *skb) int old; if (!sock_flag(sk, SOCK_USE_WRITE_QUEUE)) { + void (*sk_write_space)(struct sock *sk); + + sk_write_space = READ_ONCE(sk->sk_write_space); + if (sock_flag(sk, SOCK_RCU_FREE) && - sk->sk_write_space == sock_def_write_space) { + sk_write_space == sock_def_write_space) { rcu_read_lock(); free = __refcount_sub_and_test(len, &sk->sk_wmem_alloc, &old); @@ -2693,7 +2697,7 @@ void sock_wfree(struct sk_buff *skb) * after sk_write_space() call */ WARN_ON(refcount_sub_and_test(len - 1, &sk->sk_wmem_alloc)); - sk->sk_write_space(sk); + sk_write_space(sk); len = 1; } /* From e10902df24488ca722303133acfc82490f7d59ad Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Mon, 1 Jun 2026 18:20:55 +0000 Subject: [PATCH 150/329] tcp: Add preempt_{disable,enable}_nested() in reqsk_queue_hash_req(). syzbot reported a weird reqsk->rsk_refcnt underflow in __inet_csk_reqsk_queue_drop(). The captured reqsk_put() in __inet_csk_reqsk_queue_drop() is called only when it successfully removes reqsk from ehash. Moreover, reqsk_timer_handler() calls another reqsk_put() after that. This indicates that the reqsk was missing both refcnts for ehash and the timer itself. Since all the syzbot reports had PREEMPT_RT enabled, the only possible scenario is that reqsk_queue_hash_req() is preempted after mod_timer() and before refcount_set(), and then the timer triggered after 1s aborts the reqsk due to its listener's close(). Let's wrap mod_timer() and refcount_set() with preempt_disable_nested() and preempt_enable_nested(). Note that inet_ehash_insert() holds the normal spin_lock() (mutex in PREEMPT_RT), so it must be called outside of preempt_disable_nested(), but this is fine. The lookup path just ignores 0 sk_refcnt entries in ehash and tries to create another reqsk, but this will fail at inet_ehash_insert(). [0]: refcount_t: underflow; use-after-free. WARNING: lib/refcount.c:28 at refcount_warn_saturate+0xb2/0x110 lib/refcount.c:28, CPU#0: ktimers/0/16 Modules linked in: CPU: 0 UID: 0 PID: 16 Comm: ktimers/0 Tainted: G L syzkaller #0 PREEMPT_{RT,(full)} Tainted: [L]=SOFTLOCKUP Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/18/2026 RIP: 0010:refcount_warn_saturate+0xb2/0x110 lib/refcount.c:28 Code: e4 7d d1 0a 67 48 0f b9 3a eb 4a e8 38 3d 23 fd 48 8d 3d e1 7d d1 0a 67 48 0f b9 3a eb 37 e8 25 3d 23 fd 48 8d 3d de 7d d1 0a <67> 48 0f b9 3a eb 24 e8 12 3d 23 fd 48 8d 3d db 7d d1 0a 67 48 0f RSP: 0000:ffffc90000157948 EFLAGS: 00010246 RAX: ffffffff84a1301b RBX: 0000000000000003 RCX: ffff88801ca98000 RDX: 0000000000000100 RSI: 0000000000000000 RDI: ffffffff8f72ae00 RBP: ffffffff99ae3b01 R08: ffff88801ca98000 R09: 0000000000000005 R10: 0000000000000100 R11: 0000000000000004 R12: ffff8880425ef568 R13: ffff8880425ef4f8 R14: ffff8880425ef578 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffff888126386000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f7b46710e9c CR3: 000000000dbb6000 CR4: 00000000003526f0 Call Trace: __refcount_sub_and_test include/linux/refcount.h:400 [inline] __refcount_dec_and_test include/linux/refcount.h:432 [inline] refcount_dec_and_test include/linux/refcount.h:450 [inline] reqsk_put include/net/request_sock.h:136 [inline] __inet_csk_reqsk_queue_drop+0x3ce/0x440 net/ipv4/inet_connection_sock.c:1007 reqsk_timer_handler+0x651/0xdf0 net/ipv4/inet_connection_sock.c:1137 call_timer_fn+0x192/0x5e0 kernel/time/timer.c:1748 expire_timers kernel/time/timer.c:1799 [inline] __run_timers kernel/time/timer.c:2374 [inline] __run_timer_base+0x6a3/0x9f0 kernel/time/timer.c:2386 run_timer_base kernel/time/timer.c:2395 [inline] run_timer_softirq+0x67/0x170 kernel/time/timer.c:2403 handle_softirqs+0x1de/0x6d0 kernel/softirq.c:622 __do_softirq kernel/softirq.c:656 [inline] run_ktimerd+0x69/0x100 kernel/softirq.c:1151 smpboot_thread_fn+0x541/0xa50 kernel/smpboot.c:160 kthread+0x388/0x470 kernel/kthread.c:436 ret_from_fork+0x514/0xb70 arch/x86/kernel/process.c:158 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:245 Fixes: d2d6422f8bd1 ("x86: Allow to enable PREEMPT_RT.") Reported-by: syzbot+e809069bc15f26300526@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/6a1a7bcf.0a9e871e.332604.000b.GAE@google.com/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Eric Dumazet Reviewed-by: Sebastian Andrzej Siewior Link: https://patch.msgid.link/20260601182101.3183993-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/inet_connection_sock.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index dbcd37dfdc15..5b934ce8d98a 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -1148,6 +1148,9 @@ static bool reqsk_queue_hash_req(struct request_sock *req) /* The timer needs to be setup after a successful insertion. */ req->timeout = tcp_timeout_init((struct sock *)req); timer_setup(&req->rsk_timer, reqsk_timer_handler, TIMER_PINNED); + + preempt_disable_nested(); + mod_timer(&req->rsk_timer, jiffies + req->timeout); /* before letting lookups find us, make sure all req fields @@ -1155,6 +1158,9 @@ static bool reqsk_queue_hash_req(struct request_sock *req) */ smp_wmb(); refcount_set(&req->rsk_refcnt, 2 + 1); + + preempt_enable_nested(); + return true; } From afd0f17ca46258cec3a5cc48b8df9327fe772490 Mon Sep 17 00:00:00 2001 From: Kuniyuki Iwashima Date: Sat, 30 May 2026 06:42:58 +0000 Subject: [PATCH 151/329] hsr: Remove WARN_ONCE() in hsr_addr_is_self(). syzbot reported the warning [0] in hsr_addr_is_self(), whose assumption is simply wrong. hsr->self_node is cleared in hsr_del_self_node(), which is called from hsr_dellink(). Since dev->rtnl_link_ops->dellink() is called before unregister_netdevice_many(), there is a window when user can find the device but without hsr->self_node. Let's remove WARN_ONCE() in hsr_addr_is_self(). [0]: HSR: No self node WARNING: net/hsr/hsr_framereg.c:39 at hsr_addr_is_self+0x211/0x3f0 net/hsr/hsr_framereg.c:39, CPU#0: syz.4.16848/17220 Modules linked in: CPU: 0 UID: 0 PID: 17220 Comm: syz.4.16848 Tainted: G L syzkaller #0 PREEMPT_{RT,(full)} Tainted: [L]=SOFTLOCKUP Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/18/2026 RIP: 0010:hsr_addr_is_self+0x211/0x3f0 net/hsr/hsr_framereg.c:39 Code: 33 2f 41 0f b7 dd 89 ee 09 de 31 ff e8 c8 b4 c6 f6 09 dd 74 54 e8 0f b0 c6 f6 31 ed eb 53 e8 06 b0 c6 f6 48 8d 3d 2f 50 9c 04 <67> 48 0f b9 3a 31 ed eb 42 e8 c1 13 1f 00 89 c5 31 ff 89 c6 e8 96 RSP: 0018:ffffc900041c70e0 EFLAGS: 00010283 RAX: ffffffff8afdc6ca RBX: ffffffff8afdc4e6 RCX: 0000000000080000 RDX: ffffc90010493000 RSI: 0000000000000948 RDI: ffffffff8f9a1700 RBP: 0000000000000001 R08: 0000000000000000 R09: 0000000000000000 R10: ffffc900041c71e8 R11: fffff52000838e3f R12: dffffc0000000000 R13: ffff888041f9e3c0 R14: ffff888086ee3802 R15: 0000000000000000 FS: 00007f6fe985d6c0(0000) GS:ffff888126176000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f80bd437dac CR3: 0000000025096000 CR4: 00000000003526f0 DR0: ffffffffffffffff DR1: 00000000000001f8 DR2: 0000000000000002 DR3: ffffffffefffff15 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Call Trace: check_local_dest net/hsr/hsr_forward.c:592 [inline] fill_frame_info net/hsr/hsr_forward.c:728 [inline] hsr_forward_skb+0xa11/0x2a80 net/hsr/hsr_forward.c:739 hsr_dev_xmit+0x253/0x370 net/hsr/hsr_device.c:236 __netdev_start_xmit include/linux/netdevice.h:5368 [inline] netdev_start_xmit include/linux/netdevice.h:5377 [inline] xmit_one net/core/dev.c:3888 [inline] dev_hard_start_xmit+0x2df/0x860 net/core/dev.c:3904 __dev_queue_xmit+0x1428/0x3900 net/core/dev.c:4870 neigh_output include/net/neighbour.h:556 [inline] ip_finish_output2+0xcec/0x10b0 net/ipv4/ip_output.c:237 ip_send_skb net/ipv4/ip_output.c:1510 [inline] ip_push_pending_frames+0x8b/0x110 net/ipv4/ip_output.c:1530 raw_sendmsg+0x1547/0x1a50 net/ipv4/raw.c:659 sock_sendmsg_nosec net/socket.c:787 [inline] __sock_sendmsg net/socket.c:802 [inline] ____sys_sendmsg+0x7da/0x9c0 net/socket.c:2698 ___sys_sendmsg+0x2a5/0x360 net/socket.c:2752 __sys_sendmsg net/socket.c:2784 [inline] __do_sys_sendmsg net/socket.c:2789 [inline] __se_sys_sendmsg net/socket.c:2787 [inline] __x64_sys_sendmsg+0x1c3/0x2a0 net/socket.c:2787 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x15f/0xf80 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f6feb62ce59 Code: ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 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 e8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f6fe985d028 EFLAGS: 00000246 ORIG_RAX: 000000000000002e RAX: ffffffffffffffda RBX: 00007f6feb8a6090 RCX: 00007f6feb62ce59 RDX: 0000000000000000 RSI: 0000200000000000 RDI: 0000000000000004 RBP: 00007f6feb6c2d6f R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 00007f6feb8a6128 R14: 00007f6feb8a6090 R15: 00007ffcf01cc488 Fixes: f266a683a480 ("net/hsr: Better frame dispatch") Reported-by: syzbot+652670cf249077eb498b@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/6a1a861e.b111c304.35cd64.0016.GAE@google.com/ Signed-off-by: Kuniyuki Iwashima Reviewed-by: Fernando Fernandez Mancera Link: https://patch.msgid.link/20260530064300.340793-1-kuniyu@google.com Signed-off-by: Jakub Kicinski --- net/hsr/hsr_framereg.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/net/hsr/hsr_framereg.c b/net/hsr/hsr_framereg.c index b514e43766ef..a28dfd8490c5 100644 --- a/net/hsr/hsr_framereg.c +++ b/net/hsr/hsr_framereg.c @@ -35,10 +35,8 @@ bool hsr_addr_is_self(struct hsr_priv *hsr, unsigned char *addr) rcu_read_lock(); sn = rcu_dereference(hsr->self_node); - if (!sn) { - WARN_ONCE(1, "HSR: No self node\n"); + if (!sn) goto out; - } if (ether_addr_equal(addr, sn->macaddress_A) || ether_addr_equal(addr, sn->macaddress_B)) From c1ca14ca227e92101c7ae597213275b60f4212c6 Mon Sep 17 00:00:00 2001 From: Naveen Kumar Chaudhary Date: Mon, 1 Jun 2026 09:27:46 +0530 Subject: [PATCH 152/329] clockevents: Fix duplicate type specifier in stub function parameter The stub for arch_inlined_clockevent_set_next_coupled() has 'u64 u64 cycles' in its parameter list. Since u64 is a typedef, the compiler parses the second 'u64' as the parameter name, making 'cycles' an unused token. Remove the duplicate so the parameter is correctly named. Fixes: 89f951a1e8ad ("clockevents: Provide support for clocksource coupled comparators") Signed-off-by: Naveen Kumar Chaudhary Signed-off-by: Thomas Gleixner Link: https://patch.msgid.link/7tostpvxzdn6tobmyow63a5rweatls5kux3scqp2vzhe7mv6uq@ecr746b4hyhf --- kernel/time/clockevents.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 5e22697b098d..0014d163f989 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -301,7 +301,7 @@ static int clockevents_program_min_delta(struct clock_event_device *dev) #include #else static __always_inline void -arch_inlined_clockevent_set_next_coupled(u64 u64 cycles, struct clock_event_device *dev) { } +arch_inlined_clockevent_set_next_coupled(u64 cycles, struct clock_event_device *dev) { } #endif static inline bool clockevent_set_next_coupled(struct clock_event_device *dev, ktime_t expires) From ce4abda5e12622f33450159e76c8f56d28d7f03d Mon Sep 17 00:00:00 2001 From: Naveen Kumar Chaudhary Date: Tue, 2 Jun 2026 23:37:37 +0530 Subject: [PATCH 153/329] time: Fix off-by-one in settimeofday() usec validation The validation check uses '>' instead of '>=' when comparing tv_usec against USEC_PER_SEC, allowing the value 1000000 through. After conversion to nanoseconds (*= 1000), this produces tv_nsec == NSEC_PER_SEC, violating the timespec invariant that tv_nsec must be less than NSEC_PER_SEC. Use '>=' to reject tv_usec values that are not in the valid range of 0 to 999999. Fixes: 5e0fb1b57bea ("y2038: time: avoid timespec usage in settimeofday()") Signed-off-by: Naveen Kumar Chaudhary Signed-off-by: Thomas Gleixner Acked-by: John Stultz Link: https://patch.msgid.link/4rikk44zew3s6577dugmx4jyblz7o5c57niuap6ct3td5yfm6w@gh7pcumg7qor --- kernel/time/time.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/time/time.c b/kernel/time/time.c index 0d832317d576..771cef87ad3b 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -207,7 +207,7 @@ SYSCALL_DEFINE2(settimeofday, struct __kernel_old_timeval __user *, tv, get_user(new_ts.tv_nsec, &tv->tv_usec)) return -EFAULT; - if (new_ts.tv_nsec > USEC_PER_SEC || new_ts.tv_nsec < 0) + if (new_ts.tv_nsec >= USEC_PER_SEC || new_ts.tv_nsec < 0) return -EINVAL; new_ts.tv_nsec *= NSEC_PER_USEC; From 16e408e607a94b646fb14a2a98422c6877ae4b3c Mon Sep 17 00:00:00 2001 From: Yizhou Zhao Date: Wed, 27 May 2026 16:31:58 +0800 Subject: [PATCH 154/329] net: garp: fix unsigned integer underflow in garp_pdu_parse_attr The receive-side GARP attribute parser computes dlen with reversed operands: dlen = sizeof(*ga) - ga->len; ga->len is the on-wire attribute length and includes the GARP attribute header. For normal attributes with data, ga->len is larger than sizeof(*ga), so the subtraction underflows in unsigned arithmetic. The resulting value is later passed to garp_attr_lookup(), whose length argument is u8. After truncation, the parsed data length usually no longer matches the length stored for locally registered attributes, so received Join/Leave events are ignored. This breaks the GARP receive path for common attributes, such as GVRP VLAN registration attributes. Compute the data length as the attribute length minus the header length. Fixes: eca9ebac651f ("net: Add GARP applicant-only participant") Reported-by: Yizhou Zhao Reported-by: Yuxiang Yang Reported-by: Ao Wang Reported-by: Xuewei Feng Reported-by: Qi Li Reported-by: Ke Xu Signed-off-by: Yizhou Zhao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260527083200.42861-1-zhaoyz24@mails.tsinghua.edu.cn Signed-off-by: Jakub Kicinski --- net/802/garp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/802/garp.c b/net/802/garp.c index 6f563b6797d9..c7a39f298ad6 100644 --- a/net/802/garp.c +++ b/net/802/garp.c @@ -453,7 +453,7 @@ static int garp_pdu_parse_attr(struct garp_applicant *app, struct sk_buff *skb, if (!pskb_may_pull(skb, ga->len)) return -1; skb_pull(skb, ga->len); - dlen = sizeof(*ga) - ga->len; + dlen = ga->len - sizeof(*ga); if (attrtype > app->app->maxattr) return 0; From 3f7fbde4cdd4a25c199a60849897459a63907ca5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Thomas=20Wei=C3=9Fschuh?= Date: Wed, 13 May 2026 08:32:46 +0200 Subject: [PATCH 155/329] vdso/datastore: Mark vdso_k_*_data pointers as __ro_after_init MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit These pointers are only modified once in vdso_setup_data_pages(), during the init phase. Make them read-only after that. Drop __refdata as that would conflict with __ro_after_init. Modpost does accept the reference from a __ro_after_init symbol to an __init one. Fixes: 05988dba1179 ("vdso/datastore: Allocate data pages dynamically") Signed-off-by: Thomas Weißschuh Signed-off-by: Thomas Gleixner Link: https://patch.msgid.link/20260513-vdso-ro-after-init-v1-1-4b51f74015a4@linutronix.de --- lib/vdso/datastore.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/vdso/datastore.c b/lib/vdso/datastore.c index cf5d784a4a5a..17d37b82ebc6 100644 --- a/lib/vdso/datastore.c +++ b/lib/vdso/datastore.c @@ -11,21 +11,21 @@ static u8 vdso_initdata[VDSO_NR_PAGES * PAGE_SIZE] __aligned(PAGE_SIZE) __initdata = {}; #ifdef CONFIG_GENERIC_GETTIMEOFDAY -struct vdso_time_data *vdso_k_time_data __refdata = +struct vdso_time_data *vdso_k_time_data __ro_after_init = (void *)&vdso_initdata[VDSO_TIME_PAGE_OFFSET * PAGE_SIZE]; static_assert(sizeof(struct vdso_time_data) <= PAGE_SIZE); #endif /* CONFIG_GENERIC_GETTIMEOFDAY */ #ifdef CONFIG_VDSO_GETRANDOM -struct vdso_rng_data *vdso_k_rng_data __refdata = +struct vdso_rng_data *vdso_k_rng_data __ro_after_init = (void *)&vdso_initdata[VDSO_RNG_PAGE_OFFSET * PAGE_SIZE]; static_assert(sizeof(struct vdso_rng_data) <= PAGE_SIZE); #endif /* CONFIG_VDSO_GETRANDOM */ #ifdef CONFIG_ARCH_HAS_VDSO_ARCH_DATA -struct vdso_arch_data *vdso_k_arch_data __refdata = +struct vdso_arch_data *vdso_k_arch_data __ro_after_init = (void *)&vdso_initdata[VDSO_ARCH_PAGES_START * PAGE_SIZE]; #endif /* CONFIG_ARCH_HAS_VDSO_ARCH_DATA */ From 74e144274af39935b0f410c0ee4d2b91c3730414 Mon Sep 17 00:00:00 2001 From: Ji'an Zhou Date: Tue, 2 Jun 2026 09:12:04 +0000 Subject: [PATCH 156/329] futex/requeue: Prevent NULL pointer dereference in remove_waiter() on self-deadlock When FUTEX_CMP_REQUEUE_PI requeues a non-top waiter that already owns the target PI futex, task_blocks_on_rt_mutex() returns -EDEADLK before setting waiter->task. The subsequent remove_waiter() in rt_mutex_start_proxy_lock() dereferences the NULL waiter->task, causing a kernel crash. Add a self-deadlock check for non-top waiters before calling rt_mutex_start_proxy_lock(), analogous to the top-waiter check in futex_lock_pi_atomic(). Fixes: 3bfdc63936dd4773109b7b8c280c0f3b5ae7d349 ("rtmutex: Use waiter::task instead of current in remove_waiter()") Signed-off-by: Ji'an Zhou Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org --- kernel/futex/requeue.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/kernel/futex/requeue.c b/kernel/futex/requeue.c index b597cb3d17fc..1d99a84dc9ad 100644 --- a/kernel/futex/requeue.c +++ b/kernel/futex/requeue.c @@ -643,6 +643,12 @@ int futex_requeue(u32 __user *uaddr1, unsigned int flags1, continue; } + /* Self-deadlock: non-top waiter already owns the PI futex. */ + if (rt_mutex_owner(&pi_state->pi_mutex) == this->task) { + ret = -EDEADLK; + break; + } + ret = rt_mutex_start_proxy_lock(&pi_state->pi_mutex, this->rt_waiter, this->task); From 20cf0fb715c41111469577e85e35d15f099473e0 Mon Sep 17 00:00:00 2001 From: Yuqi Xu Date: Fri, 29 May 2026 21:01:44 +0800 Subject: [PATCH 157/329] net: rds: clear i_sends on setup unwind The RDS IB connection teardown path is written so it can run during partial startup and on repeated shutdown attempts. It uses NULL pointers to distinguish resources that are still owned from resources that have already been released. When rds_ib_setup_qp() fails after allocating i_sends but before allocating i_recvs, the sends_out path frees i_sends without clearing the pointer. A later shutdown pass can still treat that stale pointer as a live send ring allocation. Clear i_sends after vfree() in the error unwind path so the existing shutdown logic continues to use the correct ownership state. Fixes: 3b12f73a5c29 ("rds: ib: add error handle") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Zhengchuan Liang Reported-by: Xin Liu Signed-off-by: Yuqi Xu Signed-off-by: Ren Wei Reviewed-by: Allison Henderson Link: https://patch.msgid.link/5a0f7624bb9845a7b67d26166a150b59e7f394ce.1779632468.git.xuyq21@lenovo.com Signed-off-by: Jakub Kicinski --- net/rds/ib_cm.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/rds/ib_cm.c b/net/rds/ib_cm.c index 0c64c504f79d..4001de0c4959 100644 --- a/net/rds/ib_cm.c +++ b/net/rds/ib_cm.c @@ -656,6 +656,7 @@ static int rds_ib_setup_qp(struct rds_connection *conn) sends_out: vfree(ic->i_sends); + ic->i_sends = NULL; ack_dma_out: rds_dma_hdr_free(rds_ibdev->dev, ic->i_ack, ic->i_ack_dma, From 8173d22b211f615015f7b35f48ab11a6dd78dc99 Mon Sep 17 00:00:00 2001 From: David Thompson Date: Fri, 29 May 2026 21:03:00 +0000 Subject: [PATCH 158/329] net: lan743x: permit VLAN-tagged packets up to configured MTU VLAN-tagged interfaces on lan743x devices were previously unreachable via SSH and failed to respond to large ping packets (e.g. "ping -s 1469" given MTU=1500). In these scenarios, "ethtool -S" reports non-zero "RX Oversize Frame Errors". According to Microchip AN2948, the MAC_RX FSE (VLAN field size enforcement) bit determines whether frames with VLAN tags exceeding the base MTU plus tag length are discarded. The driver must set the MAC_RX.FSE bit before setting MAC_RX.RXEN to allow VLAN-tagged frames up to the interface MTU, preventing them from being treated as oversized. As a result, both the base and VLAN-tagged interfaces can use the same MTU without receive errors. Fixes: 23f0703c125b ("lan743x: Add main source files for new lan743x driver") Signed-off-by: David Thompson Reviewed-by: Thangaraj Samynathan Reviewed-by: Nicolai Buchwitz Tested-by: Nicolai Buchwitz # lan7430 on arm64 (RevPi Link: https://patch.msgid.link/20260529210300.433135-1-davthompson@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/microchip/lan743x_main.c | 32 +++++++++++++++++++ drivers/net/ethernet/microchip/lan743x_main.h | 1 + 2 files changed, 33 insertions(+) diff --git a/drivers/net/ethernet/microchip/lan743x_main.c b/drivers/net/ethernet/microchip/lan743x_main.c index f3332417162e..ffac22883e49 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.c +++ b/drivers/net/ethernet/microchip/lan743x_main.c @@ -1219,6 +1219,36 @@ static void lan743x_mac_set_address(struct lan743x_adapter *adapter, "MAC address set to %pM\n", addr); } +static void lan743x_mac_rx_enable_fse(struct lan743x_adapter *adapter) +{ + u32 mac_rx; + bool rxen; + + mac_rx = lan743x_csr_read(adapter, MAC_RX); + if (mac_rx & MAC_RX_FSE_) + return; + + rxen = mac_rx & MAC_RX_RXEN_; + if (rxen) { + mac_rx &= ~MAC_RX_RXEN_; + lan743x_csr_write(adapter, MAC_RX, mac_rx); + lan743x_csr_wait_for_bit(adapter, MAC_RX, MAC_RX_RXD_, + 1, 1000, 20000, 100); + } + + /* Per AN2948, hardware prevents modification of the FSE bit while the + * MAC receiver is enabled (RXEN bit set). Use separate register write + * to assert the FSE bit before enabling the RXEN bit in MAC_RX + */ + mac_rx |= MAC_RX_FSE_; + lan743x_csr_write(adapter, MAC_RX, mac_rx); + + if (rxen) { + mac_rx |= MAC_RX_RXEN_; + lan743x_csr_write(adapter, MAC_RX, mac_rx); + } +} + static int lan743x_mac_init(struct lan743x_adapter *adapter) { bool mac_address_valid = true; @@ -1258,6 +1288,8 @@ static int lan743x_mac_init(struct lan743x_adapter *adapter) lan743x_mac_set_address(adapter, adapter->mac_address); eth_hw_addr_set(netdev, adapter->mac_address); + lan743x_mac_rx_enable_fse(adapter); + return 0; } diff --git a/drivers/net/ethernet/microchip/lan743x_main.h b/drivers/net/ethernet/microchip/lan743x_main.h index 160d94a7cee6..1573c8f9c993 100644 --- a/drivers/net/ethernet/microchip/lan743x_main.h +++ b/drivers/net/ethernet/microchip/lan743x_main.h @@ -182,6 +182,7 @@ #define MAC_RX (0x104) #define MAC_RX_MAX_SIZE_SHIFT_ (16) #define MAC_RX_MAX_SIZE_MASK_ (0x3FFF0000) +#define MAC_RX_FSE_ BIT(2) #define MAC_RX_RXD_ BIT(1) #define MAC_RX_RXEN_ BIT(0) From b455410146bf723c7ebcb49ecd5becc0d6611482 Mon Sep 17 00:00:00 2001 From: Tapio Reijonen Date: Fri, 29 May 2026 06:18:57 +0000 Subject: [PATCH 159/329] net: fec: fix pinctrl default state restore order on resume In fec_resume(), fec_enet_clk_enable() is called before pinctrl_pm_select_default_state() in the non-WoL path, inverting the ordering used in fec_suspend() which correctly switches to the sleep pinctrl state before disabling clocks. For PHYs with the PHY_RST_AFTER_CLK_EN flag (e.g. TI DP83848 or SMSC LAN87xx), fec_enet_clk_enable() triggers a hardware reset pulse via the phy-reset GPIO. With the GPIO pin still in sleep pinctrl state at that point, the GPIO write has no physical effect and the PHY never receives the required reset after clock enable, leading to unreliable link establishment after system resume. Fix by restoring the default pinctrl state before enabling clocks, making resume the proper mirror of suspend. The call is made unconditionally: fec_suspend() only switches to the sleep pinctrl state on the non-WoL path and leaves the pins in the default state when WoL is enabled, so on a WoL resume the device is already in the default state and pinctrl_pm_select_default_state() is a no-op. Fixes: de40ed31b3c5 ("net: fec: add Wake-on-LAN support") Signed-off-by: Tapio Reijonen Reviewed-by: Wei Fang Link: https://patch.msgid.link/20260529-b4-fec-resume-pinctrl-order-v3-1-6eda0f592fca@vaisala.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/freescale/fec_main.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c index f89aa94ce020..6ebde65d7f1b 100644 --- a/drivers/net/ethernet/freescale/fec_main.c +++ b/drivers/net/ethernet/freescale/fec_main.c @@ -5594,6 +5594,7 @@ static int fec_resume(struct device *dev) if (fep->rpm_active) pm_runtime_force_resume(dev); + pinctrl_pm_select_default_state(&fep->pdev->dev); ret = fec_enet_clk_enable(ndev, true); if (ret) { rtnl_unlock(); @@ -5610,8 +5611,6 @@ static int fec_resume(struct device *dev) val &= ~(FEC_ECR_MAGICEN | FEC_ECR_SLEEP); writel(val, fep->hwp + FEC_ECNTRL); fep->wol_flag &= ~FEC_WOL_FLAG_SLEEP_ON; - } else { - pinctrl_pm_select_default_state(&fep->pdev->dev); } fec_restart(ndev); netif_tx_lock_bh(ndev); From 5eba3e48d78edd7551b992cb7ba687019b3a78da Mon Sep 17 00:00:00 2001 From: Zhao Zhang Date: Sat, 30 May 2026 23:57:14 +0800 Subject: [PATCH 160/329] sctp: diag: reject stale associations in dump_one path The SCTP exact sock_diag lookup can hold a transport reference, block on lock_sock(sk), and then resume after sctp_association_free() has marked the association dead and freed its bind address list. When that happens, inet_assoc_attr_size() and inet_diag_msg_sctpasoc_fill() can still dereference association state that is no longer valid for reporting. In particular, inet_diag_msg_sctpasoc_fill() may read an empty bind-address list as a real sctp_sockaddr_entry and trigger an out-of-bounds read from unrelated association memory. Reject the association after taking the socket lock if it has been reaped or detached from the endpoint, and report the lookup as stale. This keeps the exact dump-one path from formatting torn association state. Fixes: 8f840e47f190 ("sctp: add the sctp_diag.c file") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Zhengchuan Liang Reported-by: Xin Liu Signed-off-by: Zhao Zhang Signed-off-by: Ren Wei Acked-by: Xin Long Link: https://patch.msgid.link/fac6043fa20a2ff68e12958c431836f692c51268.1780113823.git.zzhan461@ucr.edu Signed-off-by: Jakub Kicinski --- net/sctp/diag.c | 17 +++++++++-------- 1 file changed, 9 insertions(+), 8 deletions(-) diff --git a/net/sctp/diag.c b/net/sctp/diag.c index 2afb376299fe..d758f5c3e06e 100644 --- a/net/sctp/diag.c +++ b/net/sctp/diag.c @@ -266,15 +266,15 @@ static int sctp_sock_dump_one(struct sctp_endpoint *ep, struct sctp_transport *t lock_sock(sk); - rep = nlmsg_new(inet_assoc_attr_size(sk, assoc), GFP_KERNEL); - if (!rep) { - release_sock(sk); - return -ENOMEM; + if (ep != assoc->ep || assoc->base.dead) { + err = -ESTALE; + goto out_unlock; } - if (ep != assoc->ep) { - err = -EAGAIN; - goto out; + rep = nlmsg_new(inet_assoc_attr_size(sk, assoc), GFP_KERNEL); + if (!rep) { + err = -ENOMEM; + goto out_unlock; } err = inet_sctp_diag_fill(sk, assoc, rep, req, sk_user_ns(NETLINK_CB(skb).sk), @@ -289,8 +289,9 @@ static int sctp_sock_dump_one(struct sctp_endpoint *ep, struct sctp_transport *t return nlmsg_unicast(sock_net(skb->sk)->diag_nlsk, rep, NETLINK_CB(skb).portid); out: - release_sock(sk); kfree_skb(rep); +out_unlock: + release_sock(sk); return err; } From 02e545c4297a26dbbc41df81b831e7f605bcd306 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 1 Jun 2026 09:22:37 -1000 Subject: [PATCH 161/329] sched_ext: Don't warn on NULL cgrp_moving_from in scx_cgroup_move_task() A WARN fires when systemd's user manager writes "+cpu +memory +pids" to its own subtree_control while a sched_ext scheduler is loaded: WARNING: at kernel/sched/ext.c:3227 scx_cgroup_move_task+0xa8/0xb0 scx_cgroup_move_task+0xa8/0xb0 sched_move_task+0x134/0x290 cpu_cgroup_attach+0x39/0x70 cgroup_migrate_execute+0x37d/0x450 cgroup_update_dfl_csses+0x1e3/0x270 cgroup_subtree_control_write+0x3e7/0x440 scx_cgroup_can_attach() arms cgrp_moving_from only when a task's cpu cgroup changes. It can still be NULL when scx_cgroup_move_task() runs, through this sequence: Step Result --------------------------------- ---------------------------------- 1. cpu enabled on cgroup G cpu css = A 2. cpu toggled off then on for G A killed, B created (same cgroup) 3. an exiting task keeps A alive migration skips it, A now stale 4. +memory migrates G stale A vs current B pulls cpu in 5. cpu attach runs for all tasks hits a live, cpu-unchanged task 6. scx_cgroup_move_task() on it cgrp_moving_from NULL -> WARN The mismatch is that scx_cgroup_can_attach() keys on cgroup identity while migration drives the move on css identity, so a NULL cgrp_moving_from here is a legitimate css-only migration, not a missing prep. The call is already gated on cgrp_moving_from, so just drop the warning. ops.cgroup_prep_move() and ops.cgroup_move() stay paired. Fixes: 819513666966 ("sched_ext: Add cgroup support") Cc: stable@vger.kernel.org # v6.12+ Reported-by: Matt Fleming Closes: https://lore.kernel.org/all/20260601124156.2205704-1-mfleming@cloudflare.com/ Signed-off-by: Tejun Heo Reviewed-by: Andrea Righi --- kernel/sched/ext.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index c1762420cc35..8e88a25bc602 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -4402,11 +4402,13 @@ void scx_cgroup_move_task(struct task_struct *p) return; /* - * @p must have ops.cgroup_prep_move() called on it and thus - * cgrp_moving_from set. + * scx_cgroup_can_attach() sets cgrp_moving_from only when the task's + * cgroup changes. Migration keys off css rather than cgroup identity, + * so it can hand an unchanged-cgroup task here with cgrp_moving_from + * NULL. Nothing to report to the BPF scheduler then, so skip it and + * keep prep_move and move paired. */ - if (SCX_HAS_OP(sch, cgroup_move) && - !WARN_ON_ONCE(!p->scx.cgrp_moving_from)) + if (SCX_HAS_OP(sch, cgroup_move) && p->scx.cgrp_moving_from) SCX_CALL_OP_TASK(sch, cgroup_move, task_rq(p), p, p->scx.cgrp_moving_from, tg_cgrp(task_group(p))); From 57aff991119693e09b414aff3267c0eae5e81da0 Mon Sep 17 00:00:00 2001 From: Ridong Chen Date: Tue, 2 Jun 2026 17:10:38 +0800 Subject: [PATCH 162/329] cgroup/cpuset: Change Ridong's email The chenridong@huaweicloud.com is no longer a valid email, replace it with the personal email ridong.chen@linux.dev Signed-off-by: Ridong Chen Acked-by: Waiman Long Signed-off-by: Tejun Heo --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index 74c86cf9bc65..634eb67acd06 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -6526,7 +6526,7 @@ F: include/linux/blk-cgroup.h CONTROL GROUP - CPUSET M: Waiman Long -R: Chen Ridong +R: Ridong Chen L: cgroups@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git From ad0979fe053e9f2db82da82188256ef6eb41095a Mon Sep 17 00:00:00 2001 From: Zeyu WANG Date: Wed, 3 Jun 2026 01:09:09 +0800 Subject: [PATCH 163/329] Input: atkbd - add DMI quirk for Lenovo Yoga Air 14 (83QK) The Lenovo Yoga Air 14 (83QK) laptop keyboard becomes unresponsive after the standard atkbd init sequence. Controlled testing on the actual hardware shows the F5 (ATKBD_CMD_RESET_DIS / deactivate) command specifically corrupts the EC state, causing zero IRQ1 interrupts after init. Skipping only the deactivate command (while keeping F4 ENABLE) resolves the issue completely: both keystroke input and CapsLock LED toggle work correctly. The reverse test - skipping only F4 while keeping F5 - makes the problem worse (zero keystroke interrupts), confirming F5 is the sole culprit. Add a DMI quirk entry for LENOVO/83QK using the existing atkbd_deactivate_fixup callback, consistent with the existing entries for LG Electronics and HONOR FMB-P that address the same EC F5 deactivate issue. Signed-off-by: Zeyu WANG Link: https://patch.msgid.link/20260602170909.14725-1-zeyu.thomas.wang@gmail.com Cc: stable@vger.kernel.org Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/atkbd.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c index c8ad55f26ea8..217e66ee36a1 100644 --- a/drivers/input/keyboard/atkbd.c +++ b/drivers/input/keyboard/atkbd.c @@ -1923,6 +1923,14 @@ static const struct dmi_system_id atkbd_dmi_quirk_table[] __initconst = { }, .callback = atkbd_deactivate_fixup, }, + { + /* Lenovo Yoga Air 14 (83QK) */ + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"), + DMI_MATCH(DMI_PRODUCT_NAME, "83QK"), + }, + .callback = atkbd_deactivate_fixup, + }, { } }; From f723ccaff2fb72b71ae8a9fd283f0dee4d9ae7a3 Mon Sep 17 00:00:00 2001 From: Jiayuan Chen Date: Fri, 29 May 2026 23:22:18 +0800 Subject: [PATCH 164/329] ipv6: anycast: insert aca into global hash under idev->lock syzbot reported a splat [1]: a slab-use-after-free in ipv6_chk_acast_addr(), which walks the global inet6_acaddr_lst[] hash under RCU and dereferences a struct ifacaddr6 that has already been freed while still linked in the hash, so a later reader walks into a dangling node. In __ipv6_dev_ac_inc() the aca is allocated with refcount 1, then aca_get() bumps it to 2 to keep it alive across the unlocked region. It is published to idev->ac_list under idev->lock, but ipv6_add_acaddr_hash() runs after write_unlock_bh(). A concurrent teardown (ipv6_ac_destroy_dev() from addrconf_ifdown(), under RTNL) can slip into that window: CPU0 __ipv6_dev_ac_inc CPU1 ipv6_ac_destroy_dev (RTNL) ------------------------------ ------------------------------------ aca_alloc() refcnt 1 aca_get() refcnt 2 write_lock_bh(idev->lock) add aca to ac_list write_unlock_bh(idev->lock) write_lock_bh(idev->lock) pull aca off ac_list write_unlock_bh(idev->lock) ipv6_del_acaddr_hash(aca) hlist_del_init_rcu() is a no-op, aca is not in the hash yet aca_put() refcnt 2->1 ipv6_add_acaddr_hash(aca) aca now inserted into the hash aca_put() refcnt 1->0 call_rcu(aca_free_rcu) -> kfree(aca) The hash removal becomes a no-op because the insertion has not happened yet, so once CPU0 inserts and drops the last reference, the aca is freed while still linked in inet6_acaddr_lst[], and readers dereference freed memory after the slab slot is reused. This window opened once RTNL stopped serializing the join path against device teardown. Move ipv6_add_acaddr_hash() inside the idev->lock section so the ac_list and hash insertions are atomic with respect to teardown: a racing remover now either misses the aca entirely or finds it in both lists. acaddr_hash_lock is now nested under idev->lock, which is acquired in softirq context, so switch all acaddr_hash_lock sites to spin_lock_bh() to avoid the irq lock inversion reported in [2]. [1] https://syzkaller.appspot.com/bug?extid=a01df04303c131efbf3a [2] https://lore.kernel.org/netdev/6a194ef7.ba3b1513.1890b4.0000.GAE@google.com/ Reported-by: syzbot+819eb928d120d2bdad0e@syzkaller.appspotmail.com Closes: https://lore.kernel.org/all/6a191f87.ce022c6e.138e56.0003.GAE@google.com/T/ Reviewed-by: Kuniyuki Iwashima Fixes: eb1ac9ff6c4a ("ipv6: anycast: Don't hold RTNL for IPV6_JOIN_ANYCAST.") Signed-off-by: Jiayuan Chen Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260529152219.235475-1-jiayuan.chen@linux.dev Signed-off-by: Jakub Kicinski --- net/ipv6/anycast.c | 16 ++++++++-------- 1 file changed, 8 insertions(+), 8 deletions(-) diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 67a42e01dfc3..be6dac8a8566 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -243,16 +243,16 @@ static void ipv6_add_acaddr_hash(struct net *net, struct ifacaddr6 *aca) { unsigned int hash = inet6_acaddr_hash(net, &aca->aca_addr); - spin_lock(&acaddr_hash_lock); + spin_lock_bh(&acaddr_hash_lock); hlist_add_head_rcu(&aca->aca_addr_lst, &inet6_acaddr_lst[hash]); - spin_unlock(&acaddr_hash_lock); + spin_unlock_bh(&acaddr_hash_lock); } static void ipv6_del_acaddr_hash(struct ifacaddr6 *aca) { - spin_lock(&acaddr_hash_lock); + spin_lock_bh(&acaddr_hash_lock); hlist_del_init_rcu(&aca->aca_addr_lst); - spin_unlock(&acaddr_hash_lock); + spin_unlock_bh(&acaddr_hash_lock); } static void aca_get(struct ifacaddr6 *aca) @@ -371,10 +371,10 @@ int __ipv6_dev_ac_inc(struct inet6_dev *idev, const struct in6_addr *addr) aca->aca_next = idev->ac_list; rcu_assign_pointer(idev->ac_list, aca); - write_unlock_bh(&idev->lock); - ipv6_add_acaddr_hash(net, aca); + write_unlock_bh(&idev->lock); + ip6_ins_rt(net, f6i); addrconf_join_solict(idev->dev, &aca->aca_addr); @@ -649,8 +649,8 @@ void ipv6_anycast_cleanup(void) { int i; - spin_lock(&acaddr_hash_lock); + spin_lock_bh(&acaddr_hash_lock); for (i = 0; i < IN6_ADDR_HSIZE; i++) WARN_ON(!hlist_empty(&inet6_acaddr_lst[i])); - spin_unlock(&acaddr_hash_lock); + spin_unlock_bh(&acaddr_hash_lock); } From e8694f7cc29287e843648d1075177b9a2000d957 Mon Sep 17 00:00:00 2001 From: Fedor Pchelkin Date: Mon, 1 Jun 2026 12:41:56 +0300 Subject: [PATCH 165/329] wifi: fix leak if split 6 GHz scanning fails rdev->int_scan_req is leaked if cfg80211_scan() fails. Note that it's supposed to be released at ___cfg80211_scan_done() but this doesn't happen as rdev->scan_req is NULL at that point, too, leading to the early return from the freeing function. unreferenced object 0xffff8881161d0800 (size 512): comm "wpa_supplicant", pid 379, jiffies 4294749765 hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 00 00 00 00 00 00 00 00 f0 81 13 16 81 88 ff ff ................ backtrace (crc c867fdb6): kmemleak_alloc+0x89/0x90 __kmalloc_noprof+0x2fd/0x410 cfg80211_scan+0x133/0x730 nl80211_trigger_scan+0xc69/0x1cc0 genl_family_rcv_msg_doit+0x204/0x2f0 genl_rcv_msg+0x431/0x6b0 netlink_rcv_skb+0x143/0x3f0 genl_rcv+0x27/0x40 netlink_unicast+0x4f6/0x820 netlink_sendmsg+0x797/0xce0 __sock_sendmsg+0xc4/0x160 ____sys_sendmsg+0x5e4/0x890 ___sys_sendmsg+0xf8/0x180 __sys_sendmsg+0x136/0x1e0 __x64_sys_sendmsg+0x76/0xc0 x64_sys_call+0x13f0/0x17d0 Found by Linux Verification Center (linuxtesting.org). Fixes: c8cb5b854b40 ("nl80211/cfg80211: support 6 GHz scanning") Signed-off-by: Fedor Pchelkin Link: https://patch.msgid.link/20260601094157.92703-1-pchelkin@ispras.ru Signed-off-by: Johannes Berg --- net/wireless/scan.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 358cbc9e43d8..27a56ee2e8f0 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -1071,6 +1071,7 @@ int cfg80211_scan(struct cfg80211_registered_device *rdev) struct cfg80211_scan_request_int *request; struct cfg80211_scan_request_int *rdev_req = rdev->scan_req; u32 n_channels = 0, idx, i; + int err; if (!(rdev->wiphy.flags & WIPHY_FLAG_SPLIT_SCAN_6GHZ)) { rdev_req->req.first_part = true; @@ -1100,8 +1101,14 @@ int cfg80211_scan(struct cfg80211_registered_device *rdev) rdev_req->req.scan_6ghz = false; rdev_req->req.first_part = true; + err = rdev_scan(rdev, request); + if (err) { + kfree(request); + return err; + } + rdev->int_scan_req = request; - return rdev_scan(rdev, request); + return 0; } void ___cfg80211_scan_done(struct cfg80211_registered_device *rdev, From 26eb7c0a7ab09d83eec833db6a5a2bc60b9d4d9a Mon Sep 17 00:00:00 2001 From: Chaitanya Kumar Borah Date: Mon, 1 Jun 2026 13:59:53 +0530 Subject: [PATCH 166/329] drm/i915: Fix color blob reference handling in intel_plane_state Take proper references for hw color blobs (degamma_lut, gamma_lut, ctm, lut_3d) in intel_plane_duplicate_state() and drop them in intel_plane_destroy_state(). v2: - handle blobs in hw state clear Cc: #v6.19+ Fixes: 3b7476e786c2 ("drm/i915/color: Add framework to program PRE/POST CSC LUT") Fixes: a78f1b6baf4d ("drm/i915/color: Add framework to program CSC") Fixes: 65db7a1f9cf7 ("drm/i915/color: Add 3D LUT to color pipeline") Reviewed-by: Pranay Samala #v1 Reviewed-by: Uma Shankar Signed-off-by: Chaitanya Kumar Borah Signed-off-by: Uma Shankar Link: https://patch.msgid.link/20260601082953.128539-4-chaitanya.kumar.borah@intel.com (cherry picked from commit c6eea1925154b6697fe22b217faab9bb30635e6b) Signed-off-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/display/intel_plane.c | 27 ++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/drivers/gpu/drm/i915/display/intel_plane.c b/drivers/gpu/drm/i915/display/intel_plane.c index 82f445c83158..07eae4176dad 100644 --- a/drivers/gpu/drm/i915/display/intel_plane.c +++ b/drivers/gpu/drm/i915/display/intel_plane.c @@ -144,6 +144,15 @@ intel_plane_duplicate_state(struct drm_plane *plane) if (intel_state->hw.fb) drm_framebuffer_get(intel_state->hw.fb); + if (intel_state->hw.degamma_lut) + drm_property_blob_get(intel_state->hw.degamma_lut); + if (intel_state->hw.gamma_lut) + drm_property_blob_get(intel_state->hw.gamma_lut); + if (intel_state->hw.ctm) + drm_property_blob_get(intel_state->hw.ctm); + if (intel_state->hw.lut_3d) + drm_property_blob_get(intel_state->hw.lut_3d); + return &intel_state->uapi; } @@ -167,6 +176,16 @@ intel_plane_destroy_state(struct drm_plane *plane, __drm_atomic_helper_plane_destroy_state(&plane_state->uapi); if (plane_state->hw.fb) drm_framebuffer_put(plane_state->hw.fb); + + if (plane_state->hw.degamma_lut) + drm_property_blob_put(plane_state->hw.degamma_lut); + if (plane_state->hw.gamma_lut) + drm_property_blob_put(plane_state->hw.gamma_lut); + if (plane_state->hw.ctm) + drm_property_blob_put(plane_state->hw.ctm); + if (plane_state->hw.lut_3d) + drm_property_blob_put(plane_state->hw.lut_3d); + kfree(plane_state); } @@ -317,6 +336,14 @@ static void intel_plane_clear_hw_state(struct intel_plane_state *plane_state) { if (plane_state->hw.fb) drm_framebuffer_put(plane_state->hw.fb); + if (plane_state->hw.degamma_lut) + drm_property_blob_put(plane_state->hw.degamma_lut); + if (plane_state->hw.gamma_lut) + drm_property_blob_put(plane_state->hw.gamma_lut); + if (plane_state->hw.ctm) + drm_property_blob_put(plane_state->hw.ctm); + if (plane_state->hw.lut_3d) + drm_property_blob_put(plane_state->hw.lut_3d); memset(&plane_state->hw, 0, sizeof(plane_state->hw)); } From 4793e8a6e20e4c17ed4e755ea40e5912cc3539af Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:28 +0200 Subject: [PATCH 167/329] rv: Fix __user specifier usage in extract_params() The attributes variables extracted from syscalls in the helper are both defined with the __user specifier although only the actual pointer to user data should be marked. Remove the __user specifier from attr. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202604150820.Ny143u6X-lkp@intel.com Fixes: b133207deb72 ("rv: Add nomiss deadline monitor") Reviewed-by: Wen Yang Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-2-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- kernel/trace/rv/monitors/deadline/deadline.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/trace/rv/monitors/deadline/deadline.h b/kernel/trace/rv/monitors/deadline/deadline.h index 0bbfd2543329..78fca873d61e 100644 --- a/kernel/trace/rv/monitors/deadline/deadline.h +++ b/kernel/trace/rv/monitors/deadline/deadline.h @@ -95,7 +95,8 @@ static inline u8 get_server_type(struct task_struct *tsk) static inline int extract_params(struct pt_regs *regs, long id, pid_t *pid_out) { size_t size = offsetofend(struct sched_attr, sched_flags); - struct sched_attr __user *uattr, attr; + struct sched_attr __user *uattr; + struct sched_attr attr; int new_policy = -1, ret; unsigned long args[6]; From e32d7f404d7d9dac307c1cd9a1fe132fa62ab6d6 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:29 +0200 Subject: [PATCH 168/329] rv: Reset per-task DA monitors before releasing the slot Per-task monitors use task_mon_slot to determine which slot in the array to use for the monitor. During destruction, this slot is returned but this is done before resetting the monitor. As a result, the monitor's reset is in fact resetting a slot that is outside of the array (RV_PER_TASK_MONITOR_INIT). Release the slot only after the reset to avoid out-of-bound memory access. Fixes: f5587d1b6ec93 ("rv: Add Hybrid Automata monitor type") Cc: stable@vger.kernel.org Suggested-by: Wen Yang Reviewed-by: Wen Yang Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-3-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index 39765ff6f098..1459fb3dfee6 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -309,10 +309,11 @@ static inline void da_monitor_destroy(void) WARN_ONCE(1, "Disabling a disabled monitor: " __stringify(MONITOR_NAME)); return; } - rv_put_task_monitor_slot(task_mon_slot); - task_mon_slot = RV_PER_TASK_MONITOR_INIT; da_monitor_reset_all(); + + rv_put_task_monitor_slot(task_mon_slot); + task_mon_slot = RV_PER_TASK_MONITOR_INIT; } #elif RV_MON_TYPE == RV_MON_PER_OBJ From c3d016ea823a9941ab8cbcef01a500821ff0cf16 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:30 +0200 Subject: [PATCH 169/329] rv: Prevent in-flight per-task handlers from using invalid slots Per-task monitors use a slot in the task_struct->rv[] array and store that locally (e.g. task_mon_slot), this slot is returned during the destruction process but currently hanlers can be running while that slot is returning and this race may lead to accessing an invalid slot. Synchronise with all in-flight tracepoint handlers using tracepoint_synchronize_unregister() before returning the slot. Fixes: f5587d1b6ec9 ("rv: Add Hybrid Automata monitor type") Fixes: a9769a5b9878 ("rv: Add support for LTL monitors") Suggested-by: Wen Yang Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-4-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 4 ++++ include/rv/ltl_monitor.h | 1 + 2 files changed, 5 insertions(+) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index 1459fb3dfee6..cc97cc5dfbfd 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -302,6 +302,9 @@ static int da_monitor_init(void) /* * da_monitor_destroy - return the allocated slot + * + * Wait for all in-flight handlers before returning the slot to avoid + * out-of-bound accesses. */ static inline void da_monitor_destroy(void) { @@ -310,6 +313,7 @@ static inline void da_monitor_destroy(void) return; } + tracepoint_synchronize_unregister(); da_monitor_reset_all(); rv_put_task_monitor_slot(task_mon_slot); diff --git a/include/rv/ltl_monitor.h b/include/rv/ltl_monitor.h index eff60cd61106..38e792401f76 100644 --- a/include/rv/ltl_monitor.h +++ b/include/rv/ltl_monitor.h @@ -77,6 +77,7 @@ static void ltl_monitor_destroy(void) { rv_detach_trace_probe(name, task_newtask, handle_task_newtask); + tracepoint_synchronize_unregister(); rv_put_task_monitor_slot(ltl_monitor_slot); ltl_monitor_slot = RV_PER_TASK_MONITOR_INIT; } From 32171d828ab964dc1f05f2056a81477522a3361b Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:31 +0200 Subject: [PATCH 170/329] rv: Ensure all pending probes terminate on per-obj monitor destroy The monitor disable/destroy sequence detaches all probes and resets the monitor's data, however it doesn't wait for pending probes. This is an issue with per-object monitors, which free the monitor storage. Call tracepoint_synchronize_unregister() to make sure to wait for all pending probes before destroying the monitor storage. Fixes: 4a24127bd6cb ("rv: Add support for per-object monitors in DA/HA") Reviewed-by: Wen Yang Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-5-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index cc97cc5dfbfd..a7e103654406 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -511,9 +511,10 @@ static inline void da_monitor_destroy(void) struct hlist_node *tmp; int bkt; + tracepoint_synchronize_unregister(); /* - * This function is called after all probes are disabled, we need only - * worry about concurrency against old events. + * This function is called after all probes are disabled and no longer + * pending, we can safely assume no concurrent user. */ synchronize_rcu(); hash_for_each_safe(da_monitor_ht, bkt, tmp, mon_storage, node) { From 2b9a8c27417eeef01967c6297def426a44776c04 Mon Sep 17 00:00:00 2001 From: Wen Yang Date: Mon, 1 Jun 2026 17:38:32 +0200 Subject: [PATCH 171/329] rv: Fix monitor start ordering and memory ordering for monitoring flag da_monitor_start() set monitoring=1 before calling da_monitor_init_hook(), may racing with the sched_switch handler: da_monitor_start() sched_switch handler ------------------------- --------------------------------- da_mon->monitoring = 1; if (da_monitoring(da_mon)) /* true */ ha_start_timer_ns(...); /* hrtimer->base == NULL, crash */ da_monitor_init_hook(da_mon); /* hrtimer_setup() sets base */ Fix the ordering and pair with release/acquire semantics: da_monitor_init_hook(da_mon); smp_store_release(&da_mon->monitoring, 1); /* da_monitor_start() */ return smp_load_acquire(&da_mon->monitoring); /* da_monitoring() */ On ARM64 a plain STR + LDR does not form a release-acquire pair, so the load can observe monitoring=1 while hrtimer->base is still NULL. The plain accesses are also data races under KCSAN. Use WRITE_ONCE for the monitoring=0 store in da_monitor_reset() to cover the reset path. Fixes: 792575348ff7 ("rv/include: Add deterministic automata monitor definition via C macros") Signed-off-by: Wen Yang Reviewed-by: Gabriele Monaco Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-6-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index a7e103654406..60dc39f2620a 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -82,7 +82,7 @@ static void react(enum states curr_state, enum events event) static inline void da_monitor_reset(struct da_monitor *da_mon) { da_monitor_reset_hook(da_mon); - da_mon->monitoring = 0; + WRITE_ONCE(da_mon->monitoring, 0); da_mon->curr_state = model_get_initial_state(); } @@ -95,8 +95,9 @@ static inline void da_monitor_reset(struct da_monitor *da_mon) static inline void da_monitor_start(struct da_monitor *da_mon) { da_mon->curr_state = model_get_initial_state(); - da_mon->monitoring = 1; da_monitor_init_hook(da_mon); + /* Pairs with smp_load_acquire in da_monitoring(). */ + smp_store_release(&da_mon->monitoring, 1); } /* @@ -104,7 +105,8 @@ static inline void da_monitor_start(struct da_monitor *da_mon) */ static inline bool da_monitoring(struct da_monitor *da_mon) { - return da_mon->monitoring; + /* Pairs with smp_store_release in da_monitor_start(). */ + return smp_load_acquire(&da_mon->monitoring); } /* From 713c9e1ea7623846f8e0bb657f1b9d211d1a61e6 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:33 +0200 Subject: [PATCH 172/329] rv: Do not rely on clean monitor when initialising HA Hybrid Automata monitors hook into the DA implementation when doing da_monitor_reset(). This function is called both on initialisation and teardown, HA monitors try to cancel a timer only when it's initialised relying on the da_mon->monitoring flag. This flag could however be corrupted during initialisation. This happens for instance on per-task monitors that share the same storage with different type of monitors like LTL or in case of races during a previous teardown. Stop relying on the monitoring flag during initialisation, assume that can have any value, so use a separate da_reset_state() skiping timer cancellation. New monitors (e.g. new tasks) are always zero-initialised so it is safe to rely on the monitoring flag for those. Reported-by: Wen Yang Closes: https://lore.kernel.org/lkml/d02c656aada7d071f083460a5c9a454363669b61.1778522945.git.wen.yang@linux.dev Suggested-by: Nam Cao Fixes: f5587d1b6ec9 ("rv: Add Hybrid Automata monitor type") Reviewed-by: Wen Yang Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-7-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 91 +++++++++++++++++++++++++++++++++-------- include/rv/ha_monitor.h | 2 +- 2 files changed, 76 insertions(+), 17 deletions(-) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index 60dc39f2620a..ec9bc88bd4c4 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -76,14 +76,22 @@ static void react(enum states curr_state, enum events event) model_get_state_name(curr_state)); } +/* + * da_monitor_reset_state - reset a monitor and setting it to init state + */ +static inline void da_monitor_reset_state(struct da_monitor *da_mon) +{ + WRITE_ONCE(da_mon->monitoring, 0); + da_mon->curr_state = model_get_initial_state(); +} + /* * da_monitor_reset - reset a monitor and setting it to init state */ static inline void da_monitor_reset(struct da_monitor *da_mon) { da_monitor_reset_hook(da_mon); - WRITE_ONCE(da_mon->monitoring, 0); - da_mon->curr_state = model_get_initial_state(); + da_monitor_reset_state(da_mon); } /* @@ -158,12 +166,28 @@ static struct da_monitor *da_get_monitor(void) return &DA_MON_NAME; } +/* + * __da_monitor_reset_all - reset the single monitor + */ +static void __da_monitor_reset_all(void (*reset)(struct da_monitor *)) +{ + reset(da_get_monitor()); +} + /* * da_monitor_reset_all - reset the single monitor */ static void da_monitor_reset_all(void) { - da_monitor_reset(da_get_monitor()); + __da_monitor_reset_all(da_monitor_reset); +} + +/* + * da_monitor_reset_state_all - reset the single monitor + */ +static inline void da_monitor_reset_state_all(void) +{ + __da_monitor_reset_all(da_monitor_reset_state); } /* @@ -171,7 +195,7 @@ static void da_monitor_reset_all(void) */ static inline int da_monitor_init(void) { - da_monitor_reset_all(); + da_monitor_reset_state_all(); return 0; } @@ -202,25 +226,41 @@ static struct da_monitor *da_get_monitor(void) } /* - * da_monitor_reset_all - reset all CPUs' monitor + * __da_monitor_reset_all - reset all CPUs' monitor */ -static void da_monitor_reset_all(void) +static void __da_monitor_reset_all(void (*reset)(struct da_monitor *)) { struct da_monitor *da_mon; int cpu; for_each_cpu(cpu, cpu_online_mask) { da_mon = per_cpu_ptr(&DA_MON_NAME, cpu); - da_monitor_reset(da_mon); + reset(da_mon); } } +/* + * da_monitor_reset_all - reset all CPUs' monitor + */ +static void da_monitor_reset_all(void) +{ + __da_monitor_reset_all(da_monitor_reset); +} + +/* + * da_monitor_reset_state_all - reset all CPUs' monitor + */ +static inline void da_monitor_reset_state_all(void) +{ + __da_monitor_reset_all(da_monitor_reset_state); +} + /* * da_monitor_init - initialize all CPUs' monitor */ static inline int da_monitor_init(void) { - da_monitor_reset_all(); + da_monitor_reset_state_all(); return 0; } @@ -269,19 +309,29 @@ static inline da_id_type da_get_id(struct da_monitor *da_mon) return da_get_target(da_mon)->pid; } -static void da_monitor_reset_all(void) +static void __da_monitor_reset_all(void (*reset)(struct da_monitor *)) { struct task_struct *g, *p; int cpu; read_lock(&tasklist_lock); for_each_process_thread(g, p) - da_monitor_reset(da_get_monitor(p)); + reset(da_get_monitor(p)); for_each_present_cpu(cpu) - da_monitor_reset(da_get_monitor(idle_task(cpu))); + reset(da_get_monitor(idle_task(cpu))); read_unlock(&tasklist_lock); } +static void da_monitor_reset_all(void) +{ + __da_monitor_reset_all(da_monitor_reset); +} + +static inline void da_monitor_reset_state_all(void) +{ + __da_monitor_reset_all(da_monitor_reset_state); +} + /* * da_monitor_init - initialize the per-task monitor * @@ -298,7 +348,7 @@ static int da_monitor_init(void) task_mon_slot = slot; - da_monitor_reset_all(); + da_monitor_reset_state_all(); return 0; } @@ -490,15 +540,24 @@ static inline void da_destroy_storage(da_id_type id) kfree_rcu(mon_storage, rcu); } -static void da_monitor_reset_all(void) +static void __da_monitor_reset_all(void (*reset)(struct da_monitor *)) { struct da_monitor_storage *mon_storage; int bkt; - rcu_read_lock(); + guard(rcu)(); hash_for_each_rcu(da_monitor_ht, bkt, mon_storage, node) - da_monitor_reset(&mon_storage->rv.da_mon); - rcu_read_unlock(); + reset(&mon_storage->rv.da_mon); +} + +static void da_monitor_reset_all(void) +{ + __da_monitor_reset_all(da_monitor_reset); +} + +static inline void da_monitor_reset_state_all(void) +{ + __da_monitor_reset_all(da_monitor_reset_state); } static inline int da_monitor_init(void) diff --git a/include/rv/ha_monitor.h b/include/rv/ha_monitor.h index d59507e8cb30..bd87055567cc 100644 --- a/include/rv/ha_monitor.h +++ b/include/rv/ha_monitor.h @@ -153,12 +153,12 @@ static inline void ha_monitor_init_env(struct da_monitor *da_mon) * Called from a hook in the DA reset functions, it supplies the da_mon * corresponding to the current ha_mon. * Not all hybrid automata require the timer, still clear it for simplicity. + * Monitors that never started have their timer uninitialized, do not stop those. */ static inline void ha_monitor_reset_env(struct da_monitor *da_mon) { struct ha_monitor *ha_mon = to_ha_monitor(da_mon); - /* Initialisation resets the monitor before initialising the timer */ if (likely(da_monitoring(da_mon))) ha_cancel_timer(ha_mon); } From 700782ec8f6589c5792b323efd6e004dd183328b Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:34 +0200 Subject: [PATCH 173/329] rv: Add automatic cleanup handlers for per-task HA monitors Hybrid automata monitors may start timers, depending on the model, these may remain active on an exiting task and cause false positives or even access freed memory. Add an enable/disable hook in the HA code, currently only populated by the per-task handler for registration and deregistration. This hooks to the sched_process_exit event and ensures the timer is stopped for every exiting task. The handler is enabled automatically but may be disabled, for instance if the monitor uses the event for another purpose (but should still manually ensure timers are stopped). Fixes: f5587d1b6ec9 ("rv: Add Hybrid Automata monitor type") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-8-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/ha_monitor.h | 60 +++++++++++++++++++ kernel/trace/rv/monitors/nomiss/nomiss.c | 4 +- kernel/trace/rv/monitors/opid/opid.c | 4 +- kernel/trace/rv/monitors/stall/stall.c | 4 +- .../rvgen/rvgen/templates/dot2k/main.c | 4 +- 5 files changed, 68 insertions(+), 8 deletions(-) diff --git a/include/rv/ha_monitor.h b/include/rv/ha_monitor.h index bd87055567cc..4002b5247c46 100644 --- a/include/rv/ha_monitor.h +++ b/include/rv/ha_monitor.h @@ -28,6 +28,7 @@ static inline void ha_monitor_init_env(struct da_monitor *da_mon); static inline void ha_monitor_reset_env(struct da_monitor *da_mon); static inline void ha_setup_timer(struct ha_monitor *ha_mon); static inline bool ha_cancel_timer(struct ha_monitor *ha_mon); +static inline void ha_cancel_timer_sync(struct ha_monitor *ha_mon); static bool ha_monitor_handle_constraint(struct da_monitor *da_mon, enum states curr_state, enum events event, @@ -37,6 +38,26 @@ static bool ha_monitor_handle_constraint(struct da_monitor *da_mon, #define da_monitor_init_hook ha_monitor_init_env #define da_monitor_reset_hook ha_monitor_reset_env +#if !defined(HA_SKIP_AUTO_CLEANUP) && RV_MON_TYPE == RV_MON_PER_TASK +/* + * Automatic cleanup handlers for per-task HA monitors, only skip if you know + * what you are doing (e.g. you want to implement cleanup manually in another + * handler doing more things). + */ +static void ha_handle_sched_process_exit(void *data, struct task_struct *p, + bool group_dead); + +#define ha_monitor_enable_hook() \ + rv_attach_trace_probe(__stringify(MONITOR_NAME), sched_process_exit, \ + ha_handle_sched_process_exit) +#define ha_monitor_disable_hook() \ + rv_detach_trace_probe(__stringify(MONITOR_NAME), sched_process_exit, \ + ha_handle_sched_process_exit) +#else +#define ha_monitor_enable_hook() ((void)0) +#define ha_monitor_disable_hook() ((void)0) +#endif + #include #include @@ -115,6 +136,22 @@ static enum hrtimer_restart ha_monitor_timer_callback(struct hrtimer *hrtimer); #define ha_get_ns() 0 #endif /* HA_CLK_NS */ +static int ha_monitor_init(void) +{ + int ret; + + ret = da_monitor_init(); + if (ret == 0) + ha_monitor_enable_hook(); + return ret; +} + +static void ha_monitor_destroy(void) +{ + ha_monitor_disable_hook(); + da_monitor_destroy(); +} + /* Should be supplied by the monitor */ static u64 ha_get_env(struct ha_monitor *ha_mon, enum envs env, u64 time_ns); static bool ha_verify_constraint(struct ha_monitor *ha_mon, @@ -200,6 +237,20 @@ static inline void ha_trace_error_env(struct ha_monitor *ha_mon, { CONCATENATE(trace_error_env_, MONITOR_NAME)(id, curr_state, event, env); } + +#if !defined(HA_SKIP_AUTO_CLEANUP) && RV_MON_TYPE == RV_MON_PER_TASK +static void ha_handle_sched_process_exit(void *data, struct task_struct *p, + bool group_dead) +{ + struct da_monitor *da_mon = da_get_monitor(p); + + if (likely(da_monitoring(da_mon))) { + da_monitor_reset(da_mon); + ha_cancel_timer_sync(to_ha_monitor(da_mon)); + } +} +#endif + #endif /* RV_MON_TYPE */ /* @@ -412,6 +463,10 @@ static inline bool ha_cancel_timer(struct ha_monitor *ha_mon) { return timer_delete(&ha_mon->timer); } +static inline void ha_cancel_timer_sync(struct ha_monitor *ha_mon) +{ + timer_delete_sync(&ha_mon->timer); +} #elif HA_TIMER_TYPE == HA_TIMER_HRTIMER /* * Helper functions to handle the monitor timer. @@ -463,6 +518,10 @@ static inline bool ha_cancel_timer(struct ha_monitor *ha_mon) { return hrtimer_try_to_cancel(&ha_mon->hrtimer) == 1; } +static inline void ha_cancel_timer_sync(struct ha_monitor *ha_mon) +{ + hrtimer_cancel(&ha_mon->hrtimer); +} #else /* HA_TIMER_NONE */ /* * Start function is intentionally not defined, monitors using timers must @@ -473,6 +532,7 @@ static inline bool ha_cancel_timer(struct ha_monitor *ha_mon) { return false; } +static inline void ha_cancel_timer_sync(struct ha_monitor *ha_mon) { } #endif #endif diff --git a/kernel/trace/rv/monitors/nomiss/nomiss.c b/kernel/trace/rv/monitors/nomiss/nomiss.c index 31f90f3638d8..8ead8783c29f 100644 --- a/kernel/trace/rv/monitors/nomiss/nomiss.c +++ b/kernel/trace/rv/monitors/nomiss/nomiss.c @@ -227,7 +227,7 @@ static int enable_nomiss(void) { int retval; - retval = da_monitor_init(); + retval = ha_monitor_init(); if (retval) return retval; @@ -263,7 +263,7 @@ static void disable_nomiss(void) rv_detach_trace_probe("nomiss", sched_switch, handle_sched_switch); rv_detach_trace_probe("nomiss", sched_wakeup, handle_sched_wakeup); - da_monitor_destroy(); + ha_monitor_destroy(); } static struct rv_monitor rv_this = { diff --git a/kernel/trace/rv/monitors/opid/opid.c b/kernel/trace/rv/monitors/opid/opid.c index 4594c7c46601..2922318c6112 100644 --- a/kernel/trace/rv/monitors/opid/opid.c +++ b/kernel/trace/rv/monitors/opid/opid.c @@ -73,7 +73,7 @@ static int enable_opid(void) { int retval; - retval = da_monitor_init(); + retval = ha_monitor_init(); if (retval) return retval; @@ -90,7 +90,7 @@ static void disable_opid(void) rv_detach_trace_probe("opid", sched_set_need_resched_tp, handle_sched_need_resched); rv_detach_trace_probe("opid", sched_waking, handle_sched_waking); - da_monitor_destroy(); + ha_monitor_destroy(); } /* diff --git a/kernel/trace/rv/monitors/stall/stall.c b/kernel/trace/rv/monitors/stall/stall.c index 9ccfda6b0e73..3c38fb1a0159 100644 --- a/kernel/trace/rv/monitors/stall/stall.c +++ b/kernel/trace/rv/monitors/stall/stall.c @@ -103,7 +103,7 @@ static int enable_stall(void) { int retval; - retval = da_monitor_init(); + retval = ha_monitor_init(); if (retval) return retval; @@ -120,7 +120,7 @@ static void disable_stall(void) rv_detach_trace_probe("stall", sched_switch, handle_sched_switch); rv_detach_trace_probe("stall", sched_wakeup, handle_sched_wakeup); - da_monitor_destroy(); + ha_monitor_destroy(); } static struct rv_monitor rv_this = { diff --git a/tools/verification/rvgen/rvgen/templates/dot2k/main.c b/tools/verification/rvgen/rvgen/templates/dot2k/main.c index bf0999f6657a..889446760e3c 100644 --- a/tools/verification/rvgen/rvgen/templates/dot2k/main.c +++ b/tools/verification/rvgen/rvgen/templates/dot2k/main.c @@ -35,7 +35,7 @@ static int enable_%%MODEL_NAME%%(void) { int retval; - retval = da_monitor_init(); + retval = %%MONITOR_CLASS%%_monitor_init(); if (retval) return retval; @@ -50,7 +50,7 @@ static void disable_%%MODEL_NAME%%(void) %%TRACEPOINT_DETACH%% - da_monitor_destroy(); + %%MONITOR_CLASS%%_monitor_destroy(); } /* From 74e17bd6fc8ed7e30363bb78d4c50b38cfd71efe Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:35 +0200 Subject: [PATCH 174/329] rv: Ensure synchronous cleanup for HA monitors HA monitors may start timers, all cleanup functions currently stop the timers asynchronously to avoid sleeping in the wrong context. Nothing makes sure running callbacks terminate on cleanup. Run the entire HA timer callback in an RCU read-side critical section, this way we can simply synchronize_rcu() with any pending timer and are sure any cleanup using kfree_rcu() runs after callbacks terminated. Additionally make sure any unlikely callback running late won't run any code if the monitor is marked as disabled or if destruction started. Use memory barriers to serialise with racing resets. Fixes: f5587d1b6ec9 ("rv: Add Hybrid Automata monitor type") Fixes: 4a24127bd6cb ("rv: Add support for per-object monitors in DA/HA") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-9-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 19 ++++++++++++++++--- include/rv/ha_monitor.h | 27 +++++++++++++++++++++++++-- 2 files changed, 41 insertions(+), 5 deletions(-) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index ec9bc88bd4c4..1f440c7818e6 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -57,6 +57,15 @@ static struct rv_monitor rv_this; #define da_monitor_reset_hook(da_mon) #endif +/* + * Hook to allow the implementation of hybrid automata: define it with a + * function that waits for the termination of all monitors background + * activities (e.g. all timers). This hook can sleep. + */ +#ifndef da_monitor_sync_hook +#define da_monitor_sync_hook() +#endif + /* * Type for the target id, default to int but can be overridden. * A long type can work as hash table key (PER_OBJ) but will be downgraded to @@ -82,7 +91,8 @@ static void react(enum states curr_state, enum events event) static inline void da_monitor_reset_state(struct da_monitor *da_mon) { WRITE_ONCE(da_mon->monitoring, 0); - da_mon->curr_state = model_get_initial_state(); + /* Pair with load in __ha_monitor_timer_callback */ + smp_store_release(&da_mon->curr_state, model_get_initial_state()); } /* @@ -205,6 +215,7 @@ static inline int da_monitor_init(void) static inline void da_monitor_destroy(void) { da_monitor_reset_all(); + da_monitor_sync_hook(); } #elif RV_MON_TYPE == RV_MON_PER_CPU @@ -270,6 +281,7 @@ static inline int da_monitor_init(void) static inline void da_monitor_destroy(void) { da_monitor_reset_all(); + da_monitor_sync_hook(); } #elif RV_MON_TYPE == RV_MON_PER_TASK @@ -367,6 +379,7 @@ static inline void da_monitor_destroy(void) tracepoint_synchronize_unregister(); da_monitor_reset_all(); + da_monitor_sync_hook(); rv_put_task_monitor_slot(task_mon_slot); task_mon_slot = RV_PER_TASK_MONITOR_INIT; @@ -573,13 +586,13 @@ static inline void da_monitor_destroy(void) int bkt; tracepoint_synchronize_unregister(); + da_monitor_reset_all(); + da_monitor_sync_hook(); /* * This function is called after all probes are disabled and no longer * pending, we can safely assume no concurrent user. */ - synchronize_rcu(); hash_for_each_safe(da_monitor_ht, bkt, tmp, mon_storage, node) { - da_monitor_reset_hook(&mon_storage->rv.da_mon); hash_del_rcu(&mon_storage->node); kfree(mon_storage); } diff --git a/include/rv/ha_monitor.h b/include/rv/ha_monitor.h index 4002b5247c46..28d3c74cabfc 100644 --- a/include/rv/ha_monitor.h +++ b/include/rv/ha_monitor.h @@ -37,6 +37,7 @@ static bool ha_monitor_handle_constraint(struct da_monitor *da_mon, #define da_monitor_event_hook ha_monitor_handle_constraint #define da_monitor_init_hook ha_monitor_init_env #define da_monitor_reset_hook ha_monitor_reset_env +#define da_monitor_sync_hook() synchronize_rcu() #if !defined(HA_SKIP_AUTO_CLEANUP) && RV_MON_TYPE == RV_MON_PER_TASK /* @@ -136,10 +137,13 @@ static enum hrtimer_restart ha_monitor_timer_callback(struct hrtimer *hrtimer); #define ha_get_ns() 0 #endif /* HA_CLK_NS */ +static bool ha_mon_destroying; + static int ha_monitor_init(void) { int ret; + WRITE_ONCE(ha_mon_destroying, false); ret = da_monitor_init(); if (ret == 0) ha_monitor_enable_hook(); @@ -148,6 +152,7 @@ static int ha_monitor_init(void) static void ha_monitor_destroy(void) { + WRITE_ONCE(ha_mon_destroying, true); ha_monitor_disable_hook(); da_monitor_destroy(); } @@ -288,12 +293,30 @@ static bool ha_monitor_handle_constraint(struct da_monitor *da_mon, return false; } +/* + * __ha_monitor_timer_callback - generic callback representation + * + * This callback runs in an RCU read-side critical section to allow the + * destruction sequence to easily synchronize_rcu() with all pending timers + * after asynchronously disabling them. The ha_mon_destroying check ensures + * any callback entering the RCU section after synchronize_rcu() completes + * will see the flag and bail out immediately. + */ static inline void __ha_monitor_timer_callback(struct ha_monitor *ha_mon) { - enum states curr_state = READ_ONCE(ha_mon->da_mon.curr_state); DECLARE_SEQ_BUF(env_string, ENV_BUFFER_SIZE); - u64 time_ns = ha_get_ns(); + enum states curr_state; + u64 time_ns; + guard(rcu)(); + if (unlikely(READ_ONCE(ha_mon_destroying))) + return; + /* Ensure consistent curr_state if we race with da_monitor_reset */ + curr_state = smp_load_acquire(&ha_mon->da_mon.curr_state); + if (unlikely(!da_monitor_handling_event(&ha_mon->da_mon))) + return; + + time_ns = ha_get_ns(); ha_get_env_string(&env_string, ha_mon, time_ns); ha_react(curr_state, EVENT_NONE, env_string.buffer); ha_trace_error_env(ha_mon, model_get_state_name(curr_state), From 7c147fae71f3b3542ba3292d2099ef7237cfc0da Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:36 +0200 Subject: [PATCH 175/329] rv: Prevent task migration while handling per-CPU events Tracepoint handlers are fully preemptible after a46023d5616 ("tracing: Guard __DECLARE_TRACE() use of __DO_TRACE_CALL() with SRCU-fast"). When a per-CPU monitor handles an event, it retrieves the monitor state using a per-CPU pointer. If the event itself doesn't disable preemption, the task can migrate to a different CPU and we risk updating the wrong monitor. Mitigate this by explicitly disabling task migration before acquiring the monitor pointer. This cannot guarantee the monitor runs on the correct CPU but reduces the race condition window and prevents warnings. Reviewed-by: Wen Yang Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-10-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- include/rv/da_monitor.h | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/include/rv/da_monitor.h b/include/rv/da_monitor.h index 1f440c7818e6..34b8fba9ecd4 100644 --- a/include/rv/da_monitor.h +++ b/include/rv/da_monitor.h @@ -218,6 +218,10 @@ static inline void da_monitor_destroy(void) da_monitor_sync_hook(); } +#ifndef da_implicit_guard +#define da_implicit_guard() +#endif + #elif RV_MON_TYPE == RV_MON_PER_CPU /* * Functions to define, init and get a per-cpu monitor. @@ -284,6 +288,10 @@ static inline void da_monitor_destroy(void) da_monitor_sync_hook(); } +#ifndef da_implicit_guard +#define da_implicit_guard() guard(migrate)() +#endif + #elif RV_MON_TYPE == RV_MON_PER_TASK /* * Functions to define, init and get a per-task monitor. @@ -756,6 +764,7 @@ static inline bool __da_handle_start_run_event(struct da_monitor *da_mon, */ static inline void da_handle_event(enum events event) { + da_implicit_guard(); __da_handle_event(da_get_monitor(), event, 0); } @@ -771,6 +780,7 @@ static inline void da_handle_event(enum events event) */ static inline bool da_handle_start_event(enum events event) { + da_implicit_guard(); return __da_handle_start_event(da_get_monitor(), event, 0); } @@ -782,6 +792,7 @@ static inline bool da_handle_start_event(enum events event) */ static inline bool da_handle_start_run_event(enum events event) { + da_implicit_guard(); return __da_handle_start_run_event(da_get_monitor(), event, 0); } From d9022172c1abea9e220d13ee453813dafec5b9e4 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:37 +0200 Subject: [PATCH 176/329] rv: Use 0 to check preemption enabled in opid Tracepoint handlers no longer run with preemption disabled by default since a46023d5616 ("tracing: Guard __DECLARE_TRACE() use of __DO_TRACE_CALL() with SRCU-fast"), the opid monitor should now count 1 in the preemption count as preemption disabled. Change the rule for preempt_off to preempt > 0. Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-11-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- kernel/trace/rv/monitors/opid/opid.c | 8 +------- 1 file changed, 1 insertion(+), 7 deletions(-) diff --git a/kernel/trace/rv/monitors/opid/opid.c b/kernel/trace/rv/monitors/opid/opid.c index 2922318c6112..3b6a85e815b8 100644 --- a/kernel/trace/rv/monitors/opid/opid.c +++ b/kernel/trace/rv/monitors/opid/opid.c @@ -22,14 +22,8 @@ static u64 ha_get_env(struct ha_monitor *ha_mon, enum envs_opid env, u64 time_ns if (env == irq_off_opid) return irqs_disabled(); else if (env == preempt_off_opid) { - /* - * If CONFIG_PREEMPTION is enabled, then the tracepoint itself disables - * preemption (adding one to the preempt_count). Since we are - * interested in the preempt_count at the time the tracepoint was - * hit, we consider 1 as still enabled. - */ if (IS_ENABLED(CONFIG_PREEMPTION)) - return (preempt_count() & PREEMPT_MASK) > 1; + return (preempt_count() & PREEMPT_MASK) > 0; return true; } return ENV_INVALID_VALUE; From ae7676952790f421c40918e2586a2c9f12a682b6 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Tue, 2 Jun 2026 14:50:14 -0300 Subject: [PATCH 177/329] drm/v3d: Fix vaddr leak when indirect CSD has zeroed workgroups MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit v3d_rewrite_csd_job_wg_counts_from_indirect() maps both the indirect buffer and the workgroup buffer and is expected to release them before returning. When any of the workgroup counts read from the buffer is zero, the function bailed out early and skipped the cleanup, leaking the vaddr mappings of both BOs. Jump to the cleanup path instead of returning directly, so the mappings are always dropped. Cc: stable@vger.kernel.org Fixes: 18b8413b25b7 ("drm/v3d: Create a CPU job extension for a indirect CSD job") Suggested-by: Jose Maria Casanova Crespo Reviewed-by: Iago Toral Quiroga Link: https://patch.msgid.link/20260602-v3d-fix-indirect-csd-v4-1-654309e32bc0@igalia.com Signed-off-by: Maíra Canal --- drivers/gpu/drm/v3d/v3d_sched.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c index 94bf628dc91c..47f83936cd73 100644 --- a/drivers/gpu/drm/v3d/v3d_sched.c +++ b/drivers/gpu/drm/v3d/v3d_sched.c @@ -403,7 +403,7 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job) wg_counts = (uint32_t *)(bo->vaddr + indirect_csd->offset); if (wg_counts[0] == 0 || wg_counts[1] == 0 || wg_counts[2] == 0) - return; + goto unmap_bo; args->cfg[0] = wg_counts[0] << V3D_CSD_CFG012_WG_COUNT_SHIFT; args->cfg[1] = wg_counts[1] << V3D_CSD_CFG012_WG_COUNT_SHIFT; @@ -428,6 +428,7 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job) } } +unmap_bo: v3d_put_bo_vaddr(indirect); v3d_put_bo_vaddr(bo); } From 7f93fad5ea0affc9e1505dd0f7596c0fdb496213 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Tue, 2 Jun 2026 14:50:15 -0300 Subject: [PATCH 178/329] drm/v3d: Skip CSD when it has zeroed workgroups MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit A compute shader dispatch encodes its workgroup counts in the CFG0..CFG2 registers. Kicking off a dispatch with a zero count in any of the three dimensions is invalid. First, the hardware will process 0 as 65536, while the user-space driver exposes a maximum of 65535. Over that, a submission with a zeroed workgroup dimension should be a no-op. These zeroed counts can reach the dispatch path through an indirect CSD job, whose workgroup counts are only known once the indirect buffer is read and may legitimately be zero, but such scenario should only result in a no-op. Overwrite the indirect CSD job workgroup counts with the indirect BO ones, even if they are zeroed, and don't submit the job to the hardware when any of the workgroup counts is zero, so the job completes immediately instead of running the shader. Cc: stable@vger.kernel.org Fixes: d223f98f0209 ("drm/v3d: Add support for compute shader dispatch.") Suggested-by: Jose Maria Casanova Crespo Reviewed-by: Iago Toral Quiroga Link: https://patch.msgid.link/20260602-v3d-fix-indirect-csd-v4-2-654309e32bc0@igalia.com Signed-off-by: Maíra Canal --- drivers/gpu/drm/v3d/v3d_sched.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/v3d/v3d_sched.c b/drivers/gpu/drm/v3d/v3d_sched.c index 47f83936cd73..8a635a9ec046 100644 --- a/drivers/gpu/drm/v3d/v3d_sched.c +++ b/drivers/gpu/drm/v3d/v3d_sched.c @@ -352,6 +352,16 @@ v3d_csd_job_run(struct drm_sched_job *sched_job) return NULL; } + /* The HW interprets a workgroup size of 0 as 65536; however, the + * user-space driver exposes a maximum of 65535. Therefore, a 0 in + * any dimension means that we have no workgroups and the compute + * shader should not be dispatched. + */ + if (!V3D_GET_FIELD(job->args.cfg[0], V3D_CSD_QUEUED_CFG0_NUM_WGS_X) || + !V3D_GET_FIELD(job->args.cfg[1], V3D_CSD_QUEUED_CFG1_NUM_WGS_Y) || + !V3D_GET_FIELD(job->args.cfg[2], V3D_CSD_QUEUED_CFG2_NUM_WGS_Z)) + return NULL; + v3d->queue[V3D_CSD].active_job = &job->base; v3d_invalidate_caches(v3d); @@ -402,13 +412,13 @@ v3d_rewrite_csd_job_wg_counts_from_indirect(struct v3d_cpu_job *job) wg_counts = (uint32_t *)(bo->vaddr + indirect_csd->offset); - if (wg_counts[0] == 0 || wg_counts[1] == 0 || wg_counts[2] == 0) - goto unmap_bo; - args->cfg[0] = wg_counts[0] << V3D_CSD_CFG012_WG_COUNT_SHIFT; args->cfg[1] = wg_counts[1] << V3D_CSD_CFG012_WG_COUNT_SHIFT; args->cfg[2] = wg_counts[2] << V3D_CSD_CFG012_WG_COUNT_SHIFT; + if (wg_counts[0] == 0 || wg_counts[1] == 0 || wg_counts[2] == 0) + goto unmap_bo; + num_batches = DIV_ROUND_UP(indirect_csd->wg_size, 16) * (wg_counts[0] * wg_counts[1] * wg_counts[2]); From cb9959ab5f99611d27a06586add84811fe8102dc Mon Sep 17 00:00:00 2001 From: Johannes Berg Date: Wed, 3 Jun 2026 11:18:11 +0200 Subject: [PATCH 179/329] wifi: cfg80211: enforce HE/EHT cap/oper consistency Xiang Mei reports that mac80211 could crash if eht_cap is set but eht_oper isn't. Rather than fixing that for the individual user(s), enforce that both HE/EHT have consistent elements. Reported-by: Xiang Mei Fixes: 22c64f37e1d4 ("wifi: mac80211: Update MCS15 support in link_conf") Link: https://patch.msgid.link/20260603091812.101894-2-johannes@sipsolutions.net Signed-off-by: Johannes Berg --- net/wireless/nl80211.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index dac2e8643c49..76c537a6e8b5 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -6714,6 +6714,12 @@ static int nl80211_calculate_ap_params(struct cfg80211_ap_settings *params) return -EINVAL; } + if (!!params->he_cap != !!params->he_oper) + return -EINVAL; + + if (!!params->eht_cap != !!params->eht_oper) + return -EINVAL; + return 0; } From 15fe76e23615f502d051ef0768f86babaf08746c Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Mon, 1 Jun 2026 13:52:31 -0300 Subject: [PATCH 180/329] RDMA/umem: Fix truncation for block sizes >= 4G When the iommu is used the linearization of the mapping can give a single block that is very large split across multiple SG entries. When __rdma_block_iter_next() reassembles the split SG entries it is overflowing the 32 bit stack values and computed the wrong DMA addresses for blocks after the truncation. Use the right types to hold DMA addresses. Link: https://patch.msgid.link/r/1-v1-88303e9e509f+f7-ib_umem_types_jgg@nvidia.com Cc: stable@vger.kernel.org Fixes: a808273a495c ("RDMA/verbs: Add a DMA iterator to return aligned contiguous memory blocks") Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/iter.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/core/iter.c b/drivers/infiniband/core/iter.c index 8e543d100657..3ed351e8fcf6 100644 --- a/drivers/infiniband/core/iter.c +++ b/drivers/infiniband/core/iter.c @@ -19,8 +19,8 @@ EXPORT_SYMBOL(__rdma_block_iter_start); bool __rdma_block_iter_next(struct ib_block_iter *biter) { - unsigned int block_offset; - unsigned int delta; + dma_addr_t block_offset; + dma_addr_t delta; if (!biter->__sg_nents || !biter->__sg) return false; From 43c441edacf953b39517a44f5e5e10a93618b226 Mon Sep 17 00:00:00 2001 From: Zhang Cen Date: Thu, 28 May 2026 15:56:41 +0800 Subject: [PATCH 181/329] Bluetooth: RFCOMM: hold listener socket in rfcomm_connect_ind() rfcomm_get_sock_by_channel() scans rfcomm_sk_list under the list lock, but returns the selected listener after dropping that lock without taking a reference. rfcomm_connect_ind() then locks the listener, queues a child socket on it, and may notify it after unlocking it. The buggy scenario involves two paths, with each column showing the order within that path: rfcomm_connect_ind(): listener close: 1. Find parent in 1. close() enters rfcomm_get_sock_by_channel() rfcomm_sock_release(). 2. Drop rfcomm_sk_list.lock 2. rfcomm_sock_shutdown() without pinning parent. closes the listener. 3. Call lock_sock(parent) and 3. rfcomm_sock_kill() bt_accept_enqueue(parent, unlinks and puts parent. sk, true). 4. Read parent flags and may 4. parent can be freed. call sk_state_change(). If close wins the race, parent can be freed before rfcomm_connect_ind() reaches lock_sock(), bt_accept_enqueue(), or the deferred-setup callback. Take a reference on the listener before leaving rfcomm_sk_list.lock. After lock_sock() succeeds, recheck that it is still in BT_LISTEN before queueing a child, cache the deferred-setup bit while the parent is locked, and drop the reference after the last parent use. KASAN reported a slab-use-after-free in lock_sock_nested() from rfcomm_connect_ind(), with the freeing stack going through rfcomm_sock_kill() and rfcomm_sock_release(). Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Zhang Cen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/rfcomm/sock.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-) diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c index bd7d959c6e9e..805ed5d28ed6 100644 --- a/net/bluetooth/rfcomm/sock.c +++ b/net/bluetooth/rfcomm/sock.c @@ -122,7 +122,7 @@ static struct sock *__rfcomm_get_listen_sock_by_addr(u8 channel, bdaddr_t *src) } /* Find socket with channel and source bdaddr. - * Returns closest match. + * Returns closest match with an extra reference held. */ static struct sock *rfcomm_get_sock_by_channel(int state, u8 channel, bdaddr_t *src) { @@ -136,15 +136,25 @@ static struct sock *rfcomm_get_sock_by_channel(int state, u8 channel, bdaddr_t * if (rfcomm_pi(sk)->channel == channel) { /* Exact match. */ - if (!bacmp(&rfcomm_pi(sk)->src, src)) + if (!bacmp(&rfcomm_pi(sk)->src, src)) { + sock_hold(sk); break; + } /* Closest match */ - if (!bacmp(&rfcomm_pi(sk)->src, BDADDR_ANY)) + if (!bacmp(&rfcomm_pi(sk)->src, BDADDR_ANY)) { + if (sk1) + sock_put(sk1); + sk1 = sk; + sock_hold(sk1); + } } } + if (sk && sk1) + sock_put(sk1); + read_unlock(&rfcomm_sk_list.lock); return sk ? sk : sk1; @@ -941,6 +951,7 @@ int rfcomm_connect_ind(struct rfcomm_session *s, u8 channel, struct rfcomm_dlc * { struct sock *sk, *parent; bdaddr_t src, dst; + bool defer_setup = false; int result = 0; BT_DBG("session %p channel %d", s, channel); @@ -954,6 +965,11 @@ int rfcomm_connect_ind(struct rfcomm_session *s, u8 channel, struct rfcomm_dlc * lock_sock(parent); + if (parent->sk_state != BT_LISTEN) + goto done; + + defer_setup = test_bit(BT_SK_DEFER_SETUP, &bt_sk(parent)->flags); + /* Check for backlog size */ if (sk_acceptq_is_full(parent)) { BT_DBG("backlog full %d", parent->sk_ack_backlog); @@ -981,9 +997,11 @@ int rfcomm_connect_ind(struct rfcomm_session *s, u8 channel, struct rfcomm_dlc * done: release_sock(parent); - if (test_bit(BT_SK_DEFER_SETUP, &bt_sk(parent)->flags)) + if (defer_setup) parent->sk_state_change(parent); + sock_put(parent); + return result; } From de23fb62259aa01d294f77238ae3b835eb674413 Mon Sep 17 00:00:00 2001 From: Zhang Cen Date: Thu, 28 May 2026 17:45:06 +0800 Subject: [PATCH 182/329] Bluetooth: MGMT: validate advertising TLV before type checks tlv_data_is_valid() reads each advertising data field length from data[i], then inspects data[i + 1] for managed EIR types before checking that the current field still fits inside the supplied buffer. A malformed field whose length byte is the last byte of the buffer can therefore make the parser read one byte past the advertising data. KASAN reported the following when a malformed MGMT_OP_ADD_ADVERTISING request reached that path: BUG: KASAN: vmalloc-out-of-bounds in tlv_data_is_valid() Read of size 1 Call trace: tlv_data_is_valid() add_advertising() hci_mgmt_cmd() hci_sock_sendmsg() Move the existing element-length check before any type-octet inspection so each non-empty element is proven to contain its type byte before the parser looks at data[i + 1]. Fixes: 2bb36870e8cb ("Bluetooth: Unify advertising instance flags check") Reviewed-by: Paul Menzel Signed-off-by: Zhang Cen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index de5bd6b637b2..027b266ccc74 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -8638,6 +8638,12 @@ static bool tlv_data_is_valid(struct hci_dev *hdev, u32 adv_flags, u8 *data, if (!cur_len) continue; + /* If the current field length would exceed the total data + * length, then it's invalid. + */ + if (i + cur_len >= len) + return false; + if (data[i + 1] == EIR_FLAGS && (!is_adv_data || flags_managed(adv_flags))) return false; @@ -8654,12 +8660,6 @@ static bool tlv_data_is_valid(struct hci_dev *hdev, u32 adv_flags, u8 *data, if (data[i + 1] == EIR_APPEARANCE && appearance_managed(adv_flags)) return false; - - /* If the current field length would exceed the total data - * length, then it's invalid. - */ - if (i + cur_len >= len) - return false; } return true; From 23882b828c3c8c51d0c946446a396b10abb3b16b Mon Sep 17 00:00:00 2001 From: SeungJu Cheon Date: Mon, 25 May 2026 20:04:43 +0900 Subject: [PATCH 183/329] Bluetooth: RFCOMM: validate skb length in MCC handlers The RFCOMM MCC handlers cast skb->data to protocol-specific structs without validating skb->len first. A malicious remote device can send truncated MCC frames and trigger out-of-bounds reads in these handlers. Fix this by using skb_pull_data() to validate and access the required data before dereferencing it. rfcomm_recv_rpn() requires special handling since ETSI TS 07.10 allows 1-byte RPN requests. Handle this by validating only the DLCI byte first, and validating the full struct only when len > 1. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Suggested-by: Muhammad Bilal Signed-off-by: SeungJu Cheon Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/rfcomm/core.c | 69 +++++++++++++++++++++++++++---------- 1 file changed, 50 insertions(+), 19 deletions(-) diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index d11bd5337d57..364b9381c2dc 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -1431,10 +1431,15 @@ static int rfcomm_apply_pn(struct rfcomm_dlc *d, int cr, struct rfcomm_pn *pn) static int rfcomm_recv_pn(struct rfcomm_session *s, int cr, struct sk_buff *skb) { - struct rfcomm_pn *pn = (void *) skb->data; + struct rfcomm_pn *pn; struct rfcomm_dlc *d; - u8 dlci = pn->dlci; + u8 dlci; + pn = skb_pull_data(skb, sizeof(*pn)); + if (!pn) + return -EILSEQ; + + dlci = pn->dlci; BT_DBG("session %p state %ld dlci %d", s, s->state, dlci); if (!dlci) @@ -1483,8 +1488,8 @@ static int rfcomm_recv_pn(struct rfcomm_session *s, int cr, struct sk_buff *skb) static int rfcomm_recv_rpn(struct rfcomm_session *s, int cr, int len, struct sk_buff *skb) { - struct rfcomm_rpn *rpn = (void *) skb->data; - u8 dlci = __get_dlci(rpn->dlci); + struct rfcomm_rpn *rpn; + u8 dlci; u8 bit_rate = 0; u8 data_bits = 0; @@ -1495,15 +1500,16 @@ static int rfcomm_recv_rpn(struct rfcomm_session *s, int cr, int len, struct sk_ u8 xoff_char = 0; u16 rpn_mask = RFCOMM_RPN_PM_ALL; - BT_DBG("dlci %d cr %d len 0x%x bitr 0x%x line 0x%x flow 0x%x xonc 0x%x xoffc 0x%x pm 0x%x", - dlci, cr, len, rpn->bit_rate, rpn->line_settings, rpn->flow_ctrl, - rpn->xon_char, rpn->xoff_char, rpn->param_mask); - - if (!cr) - return 0; - if (len == 1) { - /* This is a request, return default (according to ETSI TS 07.10) settings */ + rpn = skb_pull_data(skb, 1); + if (!rpn) + return -EILSEQ; + + dlci = __get_dlci(rpn->dlci); + + if (!cr) + return 0; + bit_rate = RFCOMM_RPN_BR_9600; data_bits = RFCOMM_RPN_DATA_8; stop_bits = RFCOMM_RPN_STOP_1; @@ -1514,6 +1520,19 @@ static int rfcomm_recv_rpn(struct rfcomm_session *s, int cr, int len, struct sk_ goto rpn_out; } + rpn = skb_pull_data(skb, sizeof(*rpn)); + if (!rpn) + return -EILSEQ; + + dlci = __get_dlci(rpn->dlci); + + BT_DBG("dlci %d cr %d len 0x%x bitr 0x%x line 0x%x flow 0x%x xonc 0x%x xoffc 0x%x pm 0x%x", + dlci, cr, len, rpn->bit_rate, rpn->line_settings, rpn->flow_ctrl, + rpn->xon_char, rpn->xoff_char, rpn->param_mask); + + if (!cr) + return 0; + /* Check for sane values, ignore/accept bit_rate, 8 bits, 1 stop bit, * no parity, no flow control lines, normal XON/XOFF chars */ @@ -1589,9 +1608,14 @@ static int rfcomm_recv_rpn(struct rfcomm_session *s, int cr, int len, struct sk_ static int rfcomm_recv_rls(struct rfcomm_session *s, int cr, struct sk_buff *skb) { - struct rfcomm_rls *rls = (void *) skb->data; - u8 dlci = __get_dlci(rls->dlci); + struct rfcomm_rls *rls; + u8 dlci; + rls = skb_pull_data(skb, sizeof(*rls)); + if (!rls) + return -EILSEQ; + + dlci = __get_dlci(rls->dlci); BT_DBG("dlci %d cr %d status 0x%x", dlci, cr, rls->status); if (!cr) @@ -1608,10 +1632,15 @@ static int rfcomm_recv_rls(struct rfcomm_session *s, int cr, struct sk_buff *skb static int rfcomm_recv_msc(struct rfcomm_session *s, int cr, struct sk_buff *skb) { - struct rfcomm_msc *msc = (void *) skb->data; + struct rfcomm_msc *msc; struct rfcomm_dlc *d; - u8 dlci = __get_dlci(msc->dlci); + u8 dlci; + msc = skb_pull_data(skb, sizeof(*msc)); + if (!msc) + return -EILSEQ; + + dlci = __get_dlci(msc->dlci); BT_DBG("dlci %d cr %d v24 0x%x", dlci, cr, msc->v24_sig); d = rfcomm_dlc_get(s, dlci); @@ -1644,17 +1673,19 @@ static int rfcomm_recv_msc(struct rfcomm_session *s, int cr, struct sk_buff *skb static int rfcomm_recv_mcc(struct rfcomm_session *s, struct sk_buff *skb) { - struct rfcomm_mcc *mcc = (void *) skb->data; + struct rfcomm_mcc *mcc; u8 type, cr, len; + mcc = skb_pull_data(skb, sizeof(*mcc)); + if (!mcc) + return -EILSEQ; + cr = __test_cr(mcc->type); type = __get_mcc_type(mcc->type); len = __get_mcc_len(mcc->len); BT_DBG("%p type 0x%x cr %d", s, type, cr); - skb_pull(skb, 2); - switch (type) { case RFCOMM_PN: rfcomm_recv_pn(s, cr, skb); From dd214733544427587a95f66dbf3adff072568990 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Thu, 21 May 2026 10:45:17 -0400 Subject: [PATCH 184/329] Bluetooth: L2CAP: reject BR/EDR signaling packets over MTUsig net/bluetooth/l2cap_core.c:l2cap_sig_channel() accepts BR/EDR signaling packets up to the channel MTU and dispatches each command without enforcing the signaling MTU (MTUsig). A Bluetooth BR/EDR peer within radio range can send a fixed-channel CID 0x0001 packet that is larger than MTUsig and contains many L2CAP_ECHO_REQ commands before pairing. In a real-radio stock-kernel run, one 681-byte signaling packet containing 168 zero-length ECHO_REQ commands made the target transmit 168 ECHO_RSP frames over about 220 ms. Impact: a Bluetooth BR/EDR peer within radio range, before pairing, can force 168 ECHO_RSP frames from one 681-byte fixed-channel signaling packet containing packed ECHO_REQ commands. Define Linux's BR/EDR signaling MTU as the spec minimum of 48 bytes and reject any larger signaling packet with one L2CAP_COMMAND_REJECT_RSP carrying L2CAP_REJ_MTU_EXCEEDED before any command is dispatched. The Bluetooth Core spec wording for MTUExceeded says the reject identifier shall match the first request command in the packet, and that packets containing only responses shall be silently discarded. Linux intentionally deviates from that prescription: silently discarding desynchronizes the peer because the remote stack never learns its responses were dropped, and locating the first request command requires walking command headers past MTUsig, i.e. processing bytes from a packet we have already decided is too large to process. We therefore always emit one reject and use the identifier from the first command header, a single fixed-offset byte read. The unrestricted BR/EDR signaling parser and ECHO_REQ response path both trace to the initial git import; no later introducing commit is available for a Fixes tag. Cc: stable@vger.kernel.org Suggested-by: Luiz Augusto von Dentz Link: https://lore.kernel.org/r/20260518002800.1361430-1-michael.bommarito@gmail.com Link: https://lore.kernel.org/r/20260520135034.1060859-1-michael.bommarito@gmail.com Link: https://lore.kernel.org/r/20260521000555.3712030-1-michael.bommarito@gmail.com Assisted-by: Claude:claude-opus-4-7 Assisted-by: Codex:gpt-5-5-xhigh Signed-off-by: Michael Bommarito Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/l2cap.h | 1 + net/bluetooth/l2cap_core.c | 46 +++++++++++++++++++++++++++++++++++ 2 files changed, 47 insertions(+) diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 5172afee5494..e0a1f2293679 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -33,6 +33,7 @@ /* L2CAP defaults */ #define L2CAP_DEFAULT_MTU 672 #define L2CAP_DEFAULT_MIN_MTU 48 +#define L2CAP_SIG_MTU 48 /* BR/EDR signaling MTU */ #define L2CAP_DEFAULT_FLUSH_TO 0xFFFF #define L2CAP_EFS_DEFAULT_FLUSH_TO 0xFFFFFFFF #define L2CAP_DEFAULT_TX_WINDOW 63 diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 45b175399e8d..c4ccfbda9d78 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -5643,6 +5643,15 @@ static inline void l2cap_sig_send_rej(struct l2cap_conn *conn, u16 ident) l2cap_send_cmd(conn, ident, L2CAP_COMMAND_REJ, sizeof(rej), &rej); } +static inline void l2cap_sig_send_mtu_rej(struct l2cap_conn *conn, u8 ident) +{ + struct l2cap_cmd_rej_mtu rej; + + rej.reason = cpu_to_le16(L2CAP_REJ_MTU_EXCEEDED); + rej.max_mtu = cpu_to_le16(L2CAP_SIG_MTU); + l2cap_send_cmd(conn, ident, L2CAP_COMMAND_REJ, sizeof(rej), &rej); +} + static inline void l2cap_sig_channel(struct l2cap_conn *conn, struct sk_buff *skb) { @@ -5655,6 +5664,43 @@ static inline void l2cap_sig_channel(struct l2cap_conn *conn, if (hcon->type != ACL_LINK) goto drop; + /* + * Bluetooth Core v5.4, Vol 3, Part A, Section 4: the BR/EDR + * signaling channel has a fixed signaling MTU (MTUsig) whose + * minimum and default is 48 octets. Section 4.1 says that on + * an MTUExceeded command reject the identifier "shall match + * the first request command in the L2CAP packet" and that + * packets containing only response commands "shall be + * silently discarded". + * + * Linux intentionally deviates from that prescription: + * + * 1. Silently discarding desynchronizes the peer. The + * remote stack never learns its responses were dropped, + * so any state machine waiting on a paired response + * stalls until its own timer fires. + * + * 2. Locating "the first request command" requires walking + * command headers past MTUsig, i.e. processing bytes + * from a packet we have already decided is too large to + * process. + * + * Reject every over-MTUsig signaling packet with one + * L2CAP_REJ_MTU_EXCEEDED command reject. The reject's + * reason field is what tells the peer that the whole packet + * was discarded; the identifier value is informational, so + * we use the identifier from the first command header, a + * single fixed-offset byte read. + */ + if (skb->len > L2CAP_SIG_MTU) { + u8 ident = skb->data[1]; + + BT_DBG("signaling packet exceeds MTU: %u > %u", + skb->len, L2CAP_SIG_MTU); + l2cap_sig_send_mtu_rej(conn, ident); + goto drop; + } + while (skb->len >= L2CAP_CMD_HDR_SIZE) { u16 len; From 5c65b96b549ea2dcfde497436bf9e048deb87758 Mon Sep 17 00:00:00 2001 From: Yuqi Xu Date: Fri, 29 May 2026 16:54:23 +0800 Subject: [PATCH 185/329] Bluetooth: hci_sync: reject oversized Broadcast Announcement prepend Existing advertising instances can already hold the maximum extended advertising payload. When hci_adv_bcast_annoucement() prepends the Broadcast Announcement service data to that payload, the combined data may no longer fit in the temporary buffer used to rebuild the advertising data. Reject that case before copying the existing payload and report the failure through the device log. This keeps the existing advertising data intact and avoids overrunning the temporary buffer. Fixes: 5725bc608252 ("Bluetooth: hci_sync: Fix broadcast/PA when using an existing instance") Cc: stable@kernel.org Reported-by: Yuan Tan Reported-by: Zhengchuan Liang Reported-by: Xin Liu Assisted-by: Codex:GPT-5.4 Signed-off-by: Yuqi Xu Signed-off-by: Ren Wei Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sync.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index aeccd8084cba..df23245d6ccd 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -1725,6 +1725,11 @@ static int hci_adv_bcast_annoucement(struct hci_dev *hdev, struct adv_info *adv) /* Generate Broadcast ID */ get_random_bytes(bid, sizeof(bid)); len = eir_append_service_data(ad, 0, 0x1852, bid, sizeof(bid)); + if (adv->adv_data_len > sizeof(ad) - len) { + bt_dev_err(hdev, "No room for Broadcast Announcement"); + return -EINVAL; + } + memcpy(ad + len, adv->adv_data, adv->adv_data_len); hci_set_adv_instance_data(hdev, adv->instance, len + adv->adv_data_len, ad, 0, NULL); From 6770d3a8acdf9151769180cc3710346c4cfbe6f0 Mon Sep 17 00:00:00 2001 From: Zhang Cen Date: Fri, 29 May 2026 11:22:09 +0800 Subject: [PATCH 186/329] Bluetooth: bnep: reject short frames before parsing A BNEP peer can send a short BNEP SDU. bnep_rx_frame() reads the packet type byte immediately and, for control packets, reads the control opcode and setup UUID-size byte before proving that those bytes are present. bnep_rx_control() also dereferences the control opcode without rejecting an empty control payload. Use skb_pull_data() for the fixed fields in bnep_rx_frame() so a NULL return gates each dereference. Split the control handler so the frame path can pass an opcode that has already been pulled, and keep the byte-buffer wrapper for extension control payloads. For BNEP_SETUP_CONN_REQ, name the UUID-size byte before pulling the setup payload. struct bnep_setup_conn_req carries destination and source service UUIDs after that byte, each uuid_size bytes, so the parser now documents that tuple explicitly instead of leaving the pull length as an opaque multiplication. Validation reproduced this kernel report: KASAN slab-out-of-bounds in bnep_rx_frame.isra.0+0x130c/0x1790 The buggy address belongs to the object at ffff88800c0f7908 which belongs to the cache kmalloc-8 of size 8 The buggy address is located 0 bytes to the right of allocated 1-byte region [ffff88800c0f7908, ffff88800c0f7909) Read of size 1 Call trace: dump_stack_lvl+0xb3/0x140 (?:?) print_address_description+0x57/0x3a0 (?:?) bnep_rx_frame+0x130c/0x1790 (net/bluetooth/bnep/core.c:306) print_report+0xb9/0x2b0 (?:?) __virt_addr_valid+0x1ba/0x3a0 (?:?) srso_alias_return_thunk+0x5/0xfbef5 (?:?) kasan_addr_to_slab+0x21/0x60 (?:?) kasan_report+0xe0/0x110 (?:?) process_one_work+0xfce/0x17e0 (kernel/workqueue.c:3200) worker_thread+0x65c/0xe40 (?:?) __kthread_parkme+0x184/0x230 (?:?) kthread+0x35e/0x470 (?:?) _raw_spin_unlock_irq+0x28/0x50 (?:?) ret_from_fork+0x586/0x870 (?:?) __switch_to+0x74f/0xdc0 (?:?) ret_from_fork_asm+0x1a/0x30 (?:?) Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Assisted-by: Codex:gpt-5.5 Signed-off-by: Zhang Cen Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/bnep/core.c | 59 ++++++++++++++++++++++++--------------- 1 file changed, 37 insertions(+), 22 deletions(-) diff --git a/net/bluetooth/bnep/core.c b/net/bluetooth/bnep/core.c index 0de5df690bd0..5c5f53ff30e8 100644 --- a/net/bluetooth/bnep/core.c +++ b/net/bluetooth/bnep/core.c @@ -206,14 +206,11 @@ static int bnep_ctrl_set_mcfilter(struct bnep_session *s, u8 *data, int len) return 0; } -static int bnep_rx_control(struct bnep_session *s, void *data, int len) +static int bnep_rx_control_cmd(struct bnep_session *s, u8 cmd, void *data, + int len) { - u8 cmd = *(u8 *)data; int err = 0; - data++; - len--; - switch (cmd) { case BNEP_CMD_NOT_UNDERSTOOD: case BNEP_SETUP_CONN_RSP: @@ -254,6 +251,14 @@ static int bnep_rx_control(struct bnep_session *s, void *data, int len) return err; } +static int bnep_rx_control(struct bnep_session *s, void *data, int len) +{ + if (len < 1) + return -EILSEQ; + + return bnep_rx_control_cmd(s, *(u8 *)data, data + 1, len - 1); +} + static int bnep_rx_extension(struct bnep_session *s, struct sk_buff *skb) { struct bnep_ext_hdr *h; @@ -299,19 +304,26 @@ static int bnep_rx_frame(struct bnep_session *s, struct sk_buff *skb) { struct net_device *dev = s->dev; struct sk_buff *nskb; + u8 *data; u8 type, ctrl_type; dev->stats.rx_bytes += skb->len; - type = *(u8 *) skb->data; - skb_pull(skb, 1); - ctrl_type = *(u8 *)skb->data; + data = skb_pull_data(skb, sizeof(type)); + if (!data) + goto badframe; + type = *data; if ((type & BNEP_TYPE_MASK) >= sizeof(__bnep_rx_hlen)) goto badframe; if ((type & BNEP_TYPE_MASK) == BNEP_CONTROL) { - if (bnep_rx_control(s, skb->data, skb->len) < 0) { + data = skb_pull_data(skb, sizeof(ctrl_type)); + if (!data) + goto badframe; + ctrl_type = *data; + + if (bnep_rx_control_cmd(s, ctrl_type, skb->data, skb->len) < 0) { dev->stats.tx_errors++; kfree_skb(skb); return 0; @@ -324,24 +336,27 @@ static int bnep_rx_frame(struct bnep_session *s, struct sk_buff *skb) /* Verify and pull ctrl message since it's already processed */ switch (ctrl_type) { - case BNEP_SETUP_CONN_REQ: - /* Pull: ctrl type (1 b), len (1 b), data (len bytes) */ - if (!skb_pull(skb, 2 + *(u8 *)(skb->data + 1) * 2)) - goto badframe; - break; - case BNEP_FILTER_MULTI_ADDR_SET: - case BNEP_FILTER_NET_TYPE_SET: { - u8 *hdr; + case BNEP_SETUP_CONN_REQ: { + u8 uuid_size; - /* Pull ctrl type (1 b) + len (2 b) */ - hdr = skb_pull_data(skb, 3); - if (!hdr) + /* Pull uuid_size and the dst/src service UUIDs. */ + data = skb_pull_data(skb, sizeof(uuid_size)); + if (!data) goto badframe; - /* Pull data (len bytes); length is big-endian */ - if (!skb_pull(skb, get_unaligned_be16(&hdr[1]))) + uuid_size = *data; + if (!skb_pull(skb, uuid_size + uuid_size)) goto badframe; break; } + case BNEP_FILTER_MULTI_ADDR_SET: + case BNEP_FILTER_NET_TYPE_SET: + /* Pull: len (2 b), data (len bytes) */ + data = skb_pull_data(skb, sizeof(u16)); + if (!data) + goto badframe; + if (!skb_pull(skb, get_unaligned_be16(data))) + goto badframe; + break; default: kfree_skb(skb); return 0; From 37b3009bf5976e8ab77c8b9a9bc3bbd7ff49e37f Mon Sep 17 00:00:00 2001 From: Bharath Reddy Date: Mon, 1 Jun 2026 08:54:26 +0530 Subject: [PATCH 187/329] Bluetooth: fix memory leak in error path of hci_alloc_dev() Early failures in Bluetooth HCI UART configuration leak SRCU percpu memory. When device initialization fails before hci_register_dev() completes, the HCI_UNREGISTER flag is never set. As a result, when the device reference count reaches zero, bt_host_release() evaluates this flag as false and falls back to a direct kfree(hdev). Because hci_release_dev() is bypassed, the SRCU struct initialized early in hci_alloc_dev() is never cleaned up, resulting in a leak of percpu memory. Fix the leak by explicitly calling cleanup_srcu_struct() in the fallback (unregistered) branch of bt_host_release() before freeing the device. Reported-by: syzbot+535ecc844591e50588a5@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=535ecc844591e50588a5 Tested-by: syzbot+535ecc844591e50588a5@syzkaller.appspotmail.com Fixes: 1d6123102e9f ("Bluetooth: hci_core: Fix use-after-free in vhci_flush()") Signed-off-by: Bharath Reddy Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/hci_sysfs.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c index 041ce9adc378..8957ce7c21b7 100644 --- a/net/bluetooth/hci_sysfs.c +++ b/net/bluetooth/hci_sysfs.c @@ -83,10 +83,12 @@ static void bt_host_release(struct device *dev) { struct hci_dev *hdev = to_hci_dev(dev); - if (hci_dev_test_flag(hdev, HCI_UNREGISTER)) + if (hci_dev_test_flag(hdev, HCI_UNREGISTER)) { hci_release_dev(hdev); - else + } else { + cleanup_srcu_struct(&hdev->srcu); kfree(hdev); + } module_put(THIS_MODULE); } From 5cbf290b79351971f20c7a533247e8d58a3f970c Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 1 Jun 2026 14:45:42 -0400 Subject: [PATCH 188/329] Bluetooth: ISO: Fix not releasing hdev reference on iso_conn_big_sync hci_get_route() returns a reference-counted hci_dev pointer via hci_dev_hold(). The function exits normally or with an error without ever releasing it. Fixes: 07a9342b94a9 ("Bluetooth: ISO: Send BIG Create Sync via hci_sync") Reported-by: Sashiko Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 1 + 1 file changed, 1 insertion(+) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 876649556d3c..3d707c43bdc0 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1595,6 +1595,7 @@ static void iso_conn_big_sync(struct sock *sk) release_sock(sk); hci_dev_unlock(hdev); + hci_dev_put(hdev); } static int iso_sock_recvmsg(struct socket *sock, struct msghdr *msg, From f50331f2a1441ec49988832c3a95f2edacc47322 Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Mon, 1 Jun 2026 14:52:09 -0400 Subject: [PATCH 189/329] Bluetooth: ISO: Fix a use-after-free of the hci_conn pointer In iso_sock_rebind_bc(), the bis pointer is cached, then the socket lock is dropped: bis = iso_pi(sk)->conn->hcon; /* Release the socket before lookups since that requires hci_dev_lock * which shall not be acquired while holding sock_lock for proper * ordering. */ release_sock(sk); hci_dev_lock(bis->hdev); During the unlocked window, could a concurrent close() destroy the connection and free the bis structure, causing hci_dev_lock(bis->hdev) to access memory after it is freed, fix this by using the hdev reference which was safely acquired via iso_conn_get_hdev(). Fixes: d3413703d5f8 ("Bluetooth: ISO: Add support to bind to trigger PAST") Reported-by: Sashiko Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index 3d707c43bdc0..c21ed2bb3cf8 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -1082,7 +1082,7 @@ static int iso_sock_rebind_bc(struct sock *sk, struct sockaddr_iso *sa, * ordering. */ release_sock(sk); - hci_dev_lock(bis->hdev); + hci_dev_lock(hdev); lock_sock(sk); if (!iso_pi(sk)->conn || iso_pi(sk)->conn->hcon != bis) { From 9ca7053d6215d89c33f28893bfd1625a32919d3f Mon Sep 17 00:00:00 2001 From: SeungJu Cheon Date: Mon, 1 Jun 2026 20:19:07 +0900 Subject: [PATCH 190/329] Bluetooth: ISO: Fix data-race on iso_pi fields in hci_get_route calls iso_connect_bis(), iso_connect_cis(), iso_listen_bis(), and iso_conn_big_sync() call hci_get_route() using iso_pi(sk)->dst, iso_pi(sk)->src, and iso_pi(sk)->src_type without holding lock_sock(). These fields may be modified concurrently by connect() or setsockopt() on the same socket, resulting in data-races reported by KCSAN. Fix this by snapshotting the required fields under lock_sock() before calling hci_get_route(). BUG: KCSAN: data-race in memcmp+0x45/0xb0 race at unknown origin, with read to 0xffff8880122135cf of 1 bytes by task 333 on cpu 1: memcmp+0x45/0xb0 hci_get_route+0x27e/0x490 iso_connect_cis+0x4c/0xa10 iso_sock_connect+0x60e/0xb30 __sys_connect_file+0xbd/0xe0 __sys_connect+0xe0/0x110 __x64_sys_connect+0x40/0x50 x64_sys_call+0xcad/0x1c60 do_syscall_64+0x133/0x590 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 241f51931c35 ("Bluetooth: ISO: Avoid circular locking dependency") Signed-off-by: SeungJu Cheon Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/iso.c | 60 +++++++++++++++++++++++++++++++++------------ 1 file changed, 44 insertions(+), 16 deletions(-) diff --git a/net/bluetooth/iso.c b/net/bluetooth/iso.c index c21ed2bb3cf8..3abd8111dda8 100644 --- a/net/bluetooth/iso.c +++ b/net/bluetooth/iso.c @@ -337,12 +337,20 @@ static int iso_connect_bis(struct sock *sk) struct iso_conn *conn; struct hci_conn *hcon; struct hci_dev *hdev; + bdaddr_t src, dst; + u8 src_type, bc_sid; int err; - BT_DBG("%pMR (SID 0x%2.2x)", &iso_pi(sk)->src, iso_pi(sk)->bc_sid); + lock_sock(sk); + bacpy(&src, &iso_pi(sk)->src); + bacpy(&dst, &iso_pi(sk)->dst); + src_type = iso_pi(sk)->src_type; + bc_sid = iso_pi(sk)->bc_sid; + release_sock(sk); - hdev = hci_get_route(&iso_pi(sk)->dst, &iso_pi(sk)->src, - iso_pi(sk)->src_type); + BT_DBG("%pMR (SID 0x%2.2x)", &src, bc_sid); + + hdev = hci_get_route(&dst, &src, src_type); if (!hdev) return -EHOSTUNREACH; @@ -430,12 +438,19 @@ static int iso_connect_cis(struct sock *sk) struct iso_conn *conn; struct hci_conn *hcon; struct hci_dev *hdev; + bdaddr_t src, dst; + u8 src_type; int err; - BT_DBG("%pMR -> %pMR", &iso_pi(sk)->src, &iso_pi(sk)->dst); + lock_sock(sk); + bacpy(&src, &iso_pi(sk)->src); + bacpy(&dst, &iso_pi(sk)->dst); + src_type = iso_pi(sk)->src_type; + release_sock(sk); - hdev = hci_get_route(&iso_pi(sk)->dst, &iso_pi(sk)->src, - iso_pi(sk)->src_type); + BT_DBG("%pMR -> %pMR", &src, &dst); + + hdev = hci_get_route(&dst, &src, src_type); if (!hdev) return -EHOSTUNREACH; @@ -1212,18 +1227,25 @@ static int iso_sock_connect(struct socket *sock, struct sockaddr_unsized *addr, static int iso_listen_bis(struct sock *sk) { - struct hci_dev *hdev; - int err = 0; struct iso_conn *conn; struct hci_conn *hcon; + struct hci_dev *hdev; + bdaddr_t src, dst; + u8 src_type, bc_sid; + int err = 0; - BT_DBG("%pMR -> %pMR (SID 0x%2.2x)", &iso_pi(sk)->src, - &iso_pi(sk)->dst, iso_pi(sk)->bc_sid); + lock_sock(sk); + bacpy(&src, &iso_pi(sk)->src); + bacpy(&dst, &iso_pi(sk)->dst); + src_type = iso_pi(sk)->src_type; + bc_sid = iso_pi(sk)->bc_sid; + release_sock(sk); + + BT_DBG("%pMR -> %pMR (SID 0x%2.2x)", &src, &dst, bc_sid); write_lock(&iso_sk_list.lock); - if (__iso_get_sock_listen_by_sid(&iso_pi(sk)->src, &iso_pi(sk)->dst, - iso_pi(sk)->bc_sid)) + if (__iso_get_sock_listen_by_sid(&src, &dst, bc_sid)) err = -EADDRINUSE; write_unlock(&iso_sk_list.lock); @@ -1231,8 +1253,7 @@ static int iso_listen_bis(struct sock *sk) if (err) return err; - hdev = hci_get_route(&iso_pi(sk)->dst, &iso_pi(sk)->src, - iso_pi(sk)->src_type); + hdev = hci_get_route(&dst, &src, src_type); if (!hdev) return -EHOSTUNREACH; @@ -1568,9 +1589,16 @@ static void iso_conn_big_sync(struct sock *sk) { int err; struct hci_dev *hdev; + bdaddr_t src, dst; + u8 src_type; - hdev = hci_get_route(&iso_pi(sk)->dst, &iso_pi(sk)->src, - iso_pi(sk)->src_type); + lock_sock(sk); + bacpy(&src, &iso_pi(sk)->src); + bacpy(&dst, &iso_pi(sk)->dst); + src_type = iso_pi(sk)->src_type; + release_sock(sk); + + hdev = hci_get_route(&dst, &src, src_type); if (!hdev) return; From 4847c5bca22227100ae69e96af86618b6fd2671f Mon Sep 17 00:00:00 2001 From: SeungJu Cheon Date: Mon, 1 Jun 2026 20:19:08 +0900 Subject: [PATCH 191/329] Bluetooth: SCO: Fix data-race on sco_pi fields in sco_connect sco_sock_connect() copies the destination address into sco_pi(sk)->dst under lock_sock(), then releases the lock and calls sco_connect(), which reads dst, src, setting, and codec without holding lock_sock() in hci_get_route() and hci_connect_sco(). These fields may be modified concurrently by connect(), bind(), or setsockopt() on the same socket, resulting in data-races reported by KCSAN. Fix this by snapshotting dst, src, setting, and codec under lock_sock() at the start of sco_connect() before passing them to hci_get_route() and hci_connect_sco(). BUG: KCSAN: data-race in memcmp+0x45/0xb0 race at unknown origin, with read to 0xffff88800e6b0dd0 of 1 bytes by task 315 on cpu 0: memcmp+0x45/0xb0 hci_connect_acl+0x1b7/0x6b0 hci_connect_sco+0x4d/0xb30 sco_sock_connect+0x27b/0xd60 __sys_connect_file+0xbd/0xe0 __sys_connect+0xe0/0x110 __x64_sys_connect+0x40/0x50 x64_sys_call+0xcad/0x1c60 do_syscall_64+0x133/0x590 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fixes: 9a8ec9e8ebb5 ("Bluetooth: SCO: Fix possible circular locking dependency on sco_connect_cfm") Signed-off-by: SeungJu Cheon Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/sco.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index f1799c6a6f87..140869e5b2df 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -312,11 +312,21 @@ static int sco_connect(struct sock *sk) struct sco_conn *conn; struct hci_conn *hcon; struct hci_dev *hdev; + bdaddr_t src, dst; + struct bt_codec codec; + __u16 setting; int err, type; - BT_DBG("%pMR -> %pMR", &sco_pi(sk)->src, &sco_pi(sk)->dst); + lock_sock(sk); + bacpy(&src, &sco_pi(sk)->src); + bacpy(&dst, &sco_pi(sk)->dst); + setting = sco_pi(sk)->setting; + codec = sco_pi(sk)->codec; + release_sock(sk); - hdev = hci_get_route(&sco_pi(sk)->dst, &sco_pi(sk)->src, BDADDR_BREDR); + BT_DBG("%pMR -> %pMR", &src, &dst); + + hdev = hci_get_route(&dst, &src, BDADDR_BREDR); if (!hdev) return -EHOSTUNREACH; @@ -327,7 +337,7 @@ static int sco_connect(struct sock *sk) else type = SCO_LINK; - switch (sco_pi(sk)->setting & SCO_AIRMODE_MASK) { + switch (setting & SCO_AIRMODE_MASK) { case SCO_AIRMODE_TRANSP: if (!lmp_transp_capable(hdev) || !lmp_esco_capable(hdev)) { err = -EOPNOTSUPP; @@ -336,8 +346,8 @@ static int sco_connect(struct sock *sk) break; } - hcon = hci_connect_sco(hdev, type, &sco_pi(sk)->dst, - sco_pi(sk)->setting, &sco_pi(sk)->codec, + hcon = hci_connect_sco(hdev, type, &dst, + setting, &codec, READ_ONCE(sk->sk_sndtimeo)); if (IS_ERR(hcon)) { err = PTR_ERR(hcon); From 149324fc762c2a7acef9c26790566f81f475e51f Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Tue, 2 Jun 2026 16:48:34 -0400 Subject: [PATCH 192/329] Bluetooth: MGMT: Fix backward compatibility with userspace bluetoothd has a bug with makes it send extra bytes as part of MGMT_OP_ADD_EXT_ADV_DATA which are now being checked to be the exact the expected length, relax this so only when the expected length is greater than the data length to cause an error since that would result in accessing invalid memory, otherwise just ignore the extra bytes. Link: https://lore.kernel.org/linux-bluetooth/20260602204749.210857-1-luiz.dentz@gmail.com/T/#u Fixes: d3f7d17960ed ("Bluetooth: MGMT: validate Add Extended Advertising Data length") Signed-off-by: Luiz Augusto von Dentz --- net/bluetooth/mgmt.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 027b266ccc74..f4aa814a0397 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -9114,8 +9114,9 @@ static int add_ext_adv_data(struct sock *sk, struct hci_dev *hdev, void *data, BT_DBG("%s", hdev->name); - expected_len = struct_size(cp, data, cp->adv_data_len + cp->scan_rsp_len); - if (expected_len != data_len) + expected_len = struct_size(cp, data, cp->adv_data_len + + cp->scan_rsp_len); + if (expected_len > data_len) return mgmt_cmd_status(sk, hdev->id, MGMT_OP_ADD_EXT_ADV_DATA, MGMT_STATUS_INVALID_PARAMS); From 30252e6f71ba974ecf9cd8ce395b73b9900bc378 Mon Sep 17 00:00:00 2001 From: Thomas Zimmermann Date: Tue, 2 Jun 2026 13:24:01 +0200 Subject: [PATCH 193/329] drm/dumb-buffer: Drop buffer-size limits for now The size limits break some of the CI tests. So drop them for now. Keep the other overflow tests from commit 5ab62dd3687b ("drm: prevent integer overflows in dumb buffer creation helpers") in place. There is still a pre-existing overflow check for 32-bit type limits in drm_mode_create_dumb() that will catch the really absurd size requests. Drivers that still do not use drm_mode_size_dumb() should be updated. The helper calculates dumb-buffer geometry with overflow checks. Signed-off-by: Thomas Zimmermann Fixes: 5ab62dd3687b ("drm: prevent integer overflows in dumb buffer creation helpers") Reported-by: Jani Nikula Closes: https://lore.kernel.org/dri-devel/ddf0233e50044059c85279f928661563ef6a55bf@intel.com/ Cc: Rajat Gupta Cc: Thomas Zimmermann Cc: Maarten Lankhorst Cc: Maxime Ripard Acked-by: Jani Nikula Link: https://patch.msgid.link/20260602112842.252279-1-tzimmermann@suse.de --- drivers/gpu/drm/drm_dumb_buffers.c | 7 ------- 1 file changed, 7 deletions(-) diff --git a/drivers/gpu/drm/drm_dumb_buffers.c b/drivers/gpu/drm/drm_dumb_buffers.c index cc99681a9ed0..2156dbe601c9 100644 --- a/drivers/gpu/drm/drm_dumb_buffers.c +++ b/drivers/gpu/drm/drm_dumb_buffers.c @@ -202,13 +202,6 @@ int drm_mode_create_dumb(struct drm_device *dev, if (!args->width || !args->height || !args->bpp) return -EINVAL; - /* Reject unreasonable inputs early. Dumb buffers are for software - * rendering; nothing legitimate needs more than 8192x8192 at 32bpp. - * This prevents overflows in downstream alignment helpers. - */ - if (args->width >= 8192 || args->height >= 8192 || args->bpp > 32) - return -EINVAL; - /* overflow checks for 32bit size calculations */ if (args->bpp > U32_MAX - 8) return -EINVAL; From 323c98a4ff06aa28114f2bf658fb43eb3b536bbc Mon Sep 17 00:00:00 2001 From: Yishai Hadas Date: Mon, 25 May 2026 17:21:36 +0300 Subject: [PATCH 194/329] RDMA/core: Validate cpu_id against nr_cpu_ids in DMAH alloc The cpu_id attribute supplied by user space through UVERBS_ATTR_ALLOC_DMAH_CPU_ID is passed directly to cpumask_test_cpu() without first verifying that the value is within the valid CPU range. Passing such untrusted data to cpumask_test_cpu() may lead to an out-of-bounds read of the underlying cpumask bitmap: the helper expands to a test_bit() that indexes the bitmap by cpu_id / BITS_PER_LONG with no bound check. In addition, on kernels built with CONFIG_DEBUG_PER_CPU_MAPS it trips the WARN_ON_ONCE() in cpumask_check(); combined with panic_on_warn this turns a bad user input into a machine reboot. Reject any cpu_id that is not smaller than nr_cpu_ids with -EINVAL before it is used. Reported by Smatch. Fixes: d83edab562a4 ("RDMA/core: Introduce a DMAH object and its alloc/free APIs") Link: https://patch.msgid.link/r/20260525142136.28165-1-yishaih@nvidia.com Cc: stable@vger.kernel.org Reported-by: Dan Carpenter Closes: https://lore.kernel.org/r/ag68qoAW3P04J7pT@stanley.mountain/ Signed-off-by: Yishai Hadas Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/uverbs_std_types_dmah.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/infiniband/core/uverbs_std_types_dmah.c b/drivers/infiniband/core/uverbs_std_types_dmah.c index 453ce656c6f2..97101e093826 100644 --- a/drivers/infiniband/core/uverbs_std_types_dmah.c +++ b/drivers/infiniband/core/uverbs_std_types_dmah.c @@ -47,6 +47,11 @@ static int UVERBS_HANDLER(UVERBS_METHOD_DMAH_ALLOC)( if (ret) goto err; + if (dmah->cpu_id >= nr_cpu_ids) { + ret = -EINVAL; + goto err; + } + if (!cpumask_test_cpu(dmah->cpu_id, current->cpus_ptr)) { ret = -EPERM; goto err; From 6590fe323ce2807f5d9454e7fccf3fab875d4352 Mon Sep 17 00:00:00 2001 From: Leorize Date: Wed, 27 May 2026 23:58:54 -0700 Subject: [PATCH 195/329] drm/amd/display: add missing CSC entries for BT.2020 for DCE IPs DCE-based hardware does not have the CSC matrices for BT.2020, which causes the driver to fallback to the GPU built-in matrices. This does not appear to cause any issues for RGB sinks, but causes major color artifacts for YCbCr ones (e.g. black becomes green). This commit adds the missing CSC matrices (taken from DC common) to DCE CSC tables, resolving the issue. Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/3358 Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5333 Assisted-by: oh-my-pi:GPT-5.5 Signed-off-by: Leorize Reviewed-by: Alex Hung Signed-off-by: Alex Deucher (cherry picked from commit 51e6668ab4baf55b082c376318d51ef965757196) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/dce/dce_transform.c | 10 +++++++++- .../gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c | 10 +++++++++- 2 files changed, 18 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c index c1448ae47366..0d312b40bcfa 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_transform.c @@ -110,7 +110,15 @@ static const struct out_csc_color_matrix global_color_matrix[] = { { COLOR_SPACE_YCBCR601_LIMITED, { 0xE00, 0xF447, 0xFDB9, 0x1000, 0x991, 0x12C9, 0x3A6, 0x200, 0xFB47, 0xF6B9, 0xE00, 0x1000} }, { COLOR_SPACE_YCBCR709_LIMITED, { 0xE00, 0xF349, 0xFEB7, 0x1000, 0x6CE, 0x16E3, - 0x24F, 0x200, 0xFCCB, 0xF535, 0xE00, 0x1000} } + 0x24F, 0x200, 0xFCCB, 0xF535, 0xE00, 0x1000} }, +{ COLOR_SPACE_2020_RGB_FULLRANGE, + { 0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} }, +{ COLOR_SPACE_2020_RGB_LIMITEDRANGE, + { 0x1B67, 0, 0, 0x201, 0, 0x1B67, 0, 0x201, 0, 0, 0x1B67, 0x201} }, +{ COLOR_SPACE_2020_YCBCR_LIMITED, { 0x1000, 0xF149, 0xFEB7, 0x1004, 0x0868, + 0x15B2, 0x01E6, 0x201, 0xFB88, 0xF478, 0x1000, 0x1004} }, +{ COLOR_SPACE_2020_YCBCR_FULL, { 0x1000, 0xF149, 0xFEB7, 0x1004, 0x0868, 0x15B2, + 0x01E6, 0x201, 0xFB88, 0xF478, 0x1000, 0x1004} } }; static bool setup_scaling_configuration( diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c index cf63fac82832..1ed018aaa4bb 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_opp_csc_v.c @@ -88,7 +88,15 @@ static const struct out_csc_color_matrix global_color_matrix[] = { { COLOR_SPACE_YCBCR601_LIMITED, { 0xE00, 0xF447, 0xFDB9, 0x1000, 0x991, 0x12C9, 0x3A6, 0x200, 0xFB47, 0xF6B9, 0xE00, 0x1000} }, { COLOR_SPACE_YCBCR709_LIMITED, { 0xE00, 0xF349, 0xFEB7, 0x1000, 0x6CE, 0x16E3, - 0x24F, 0x200, 0xFCCB, 0xF535, 0xE00, 0x1000} } + 0x24F, 0x200, 0xFCCB, 0xF535, 0xE00, 0x1000} }, +{ COLOR_SPACE_2020_RGB_FULLRANGE, + { 0x2000, 0, 0, 0, 0, 0x2000, 0, 0, 0, 0, 0x2000, 0} }, +{ COLOR_SPACE_2020_RGB_LIMITEDRANGE, + { 0x1B67, 0, 0, 0x201, 0, 0x1B67, 0, 0x201, 0, 0, 0x1B67, 0x201} }, +{ COLOR_SPACE_2020_YCBCR_LIMITED, { 0x1000, 0xF149, 0xFEB7, 0x1004, 0x0868, + 0x15B2, 0x01E6, 0x201, 0xFB88, 0xF478, 0x1000, 0x1004} }, +{ COLOR_SPACE_2020_YCBCR_FULL, { 0x1000, 0xF149, 0xFEB7, 0x1004, 0x0868, 0x15B2, + 0x01E6, 0x201, 0xFB88, 0xF478, 0x1000, 0x1004} } }; enum csc_color_mode { From e8b4d37eba05141ee01794fc6b7f2da808cee83b Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Tue, 5 May 2026 11:44:15 -0400 Subject: [PATCH 196/329] drm/amd/display: Fix out-of-bounds read in dp_get_eq_aux_rd_interval() [Why & How] The aux_rd_interval array in struct dc_lttpr_caps is declared with MAX_REPEATER_CNT - 1 (7) elements, indexed 0..6. However, the offset parameter passed to dp_get_eq_aux_rd_interval() can be as large as MAX_REPEATER_CNT (8) when a sink reports 8 LTTPR repeaters via DPCD. This leads to an out-of-bounds read of aux_rd_interval[7] when offset is 8. Fix this by growing aux_rd_interval to MAX_REPEATER_CNT elements to accommodate the full range of valid repeater counts defined by the DP spec. Assisted-by: GitHub Copilot:Claude claude-4-opus Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit a55a458a8df37a65ffda5cf721d554a8f74f6b04) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/dc_dp_types.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index 7fa336bf1115..7dd73eaaf940 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -1217,7 +1217,7 @@ struct dc_lttpr_caps { union dp_main_link_channel_coding_lttpr_cap main_link_channel_coding; union dp_128b_132b_supported_lttpr_link_rates supported_128b_132b_rates; union dp_alpm_lttpr_cap alpm; - uint8_t aux_rd_interval[MAX_REPEATER_CNT - 1]; + uint8_t aux_rd_interval[MAX_REPEATER_CNT]; uint8_t lttpr_ieee_oui[3]; // Always read from closest LTTPR to host uint8_t lttpr_device_id[6]; // Always read from closest LTTPR to host }; From fb0707ce00eef4e2d60c3020e1c0432739703e4a Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Mon, 4 May 2026 15:51:13 -0400 Subject: [PATCH 197/329] drm/amd/display: Clamp VBIOS HDMI retimer register count to array size [Why & How] The VBIOS integrated info tables (v1_11 and v2_1) contain HdmiRegNum and Hdmi6GRegNum fields that are used as loop bounds when copying retimer I2C register settings into fixed-size arrays (dp*_ext_hdmi_reg_settings[9] and dp*_ext_hdmi_6g_reg_settings[3]). These u8 fields are not validated before use, so a malformed VBIOS can specify values up to 255, causing an out-of-bounds heap write during driver probe. Clamp each register count to the destination array size using min_t() before the copy loops, in both get_integrated_info_v11() and get_integrated_info_v2_1(). Assisted-by: GitHub Copilot:claude-opus-4.6 Reviewed-by: Alex Hung Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 5a7f0ef90195940c54b0f5bb85b87da55f038c69) Cc: stable@vger.kernel.org --- .../drm/amd/display/dc/bios/bios_parser2.c | 48 ++++++++++++------- 1 file changed, 32 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index c51c4b2c6fae..e8d8947c552e 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -2600,14 +2600,16 @@ static enum bp_result get_integrated_info_v11( info_v11->extdispconninfo.checksum; info->dp0_ext_hdmi_slv_addr = info_v11->dp0_retimer_set.HdmiSlvAddr; - info->dp0_ext_hdmi_reg_num = info_v11->dp0_retimer_set.HdmiRegNum; + info->dp0_ext_hdmi_reg_num = min_t(u8, info_v11->dp0_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp0_ext_hdmi_reg_settings)); for (i = 0; i < info->dp0_ext_hdmi_reg_num; i++) { info->dp0_ext_hdmi_reg_settings[i].i2c_reg_index = info_v11->dp0_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp0_ext_hdmi_reg_settings[i].i2c_reg_val = info_v11->dp0_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp0_ext_hdmi_6g_reg_num = info_v11->dp0_retimer_set.Hdmi6GRegNum; + info->dp0_ext_hdmi_6g_reg_num = min_t(u8, info_v11->dp0_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp0_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp0_ext_hdmi_6g_reg_num; i++) { info->dp0_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v11->dp0_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2616,14 +2618,16 @@ static enum bp_result get_integrated_info_v11( } info->dp1_ext_hdmi_slv_addr = info_v11->dp1_retimer_set.HdmiSlvAddr; - info->dp1_ext_hdmi_reg_num = info_v11->dp1_retimer_set.HdmiRegNum; + info->dp1_ext_hdmi_reg_num = min_t(u8, info_v11->dp1_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp1_ext_hdmi_reg_settings)); for (i = 0; i < info->dp1_ext_hdmi_reg_num; i++) { info->dp1_ext_hdmi_reg_settings[i].i2c_reg_index = info_v11->dp1_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp1_ext_hdmi_reg_settings[i].i2c_reg_val = info_v11->dp1_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp1_ext_hdmi_6g_reg_num = info_v11->dp1_retimer_set.Hdmi6GRegNum; + info->dp1_ext_hdmi_6g_reg_num = min_t(u8, info_v11->dp1_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp1_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp1_ext_hdmi_6g_reg_num; i++) { info->dp1_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v11->dp1_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2632,14 +2636,16 @@ static enum bp_result get_integrated_info_v11( } info->dp2_ext_hdmi_slv_addr = info_v11->dp2_retimer_set.HdmiSlvAddr; - info->dp2_ext_hdmi_reg_num = info_v11->dp2_retimer_set.HdmiRegNum; + info->dp2_ext_hdmi_reg_num = min_t(u8, info_v11->dp2_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp2_ext_hdmi_reg_settings)); for (i = 0; i < info->dp2_ext_hdmi_reg_num; i++) { info->dp2_ext_hdmi_reg_settings[i].i2c_reg_index = info_v11->dp2_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp2_ext_hdmi_reg_settings[i].i2c_reg_val = info_v11->dp2_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp2_ext_hdmi_6g_reg_num = info_v11->dp2_retimer_set.Hdmi6GRegNum; + info->dp2_ext_hdmi_6g_reg_num = min_t(u8, info_v11->dp2_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp2_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp2_ext_hdmi_6g_reg_num; i++) { info->dp2_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v11->dp2_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2648,14 +2654,16 @@ static enum bp_result get_integrated_info_v11( } info->dp3_ext_hdmi_slv_addr = info_v11->dp3_retimer_set.HdmiSlvAddr; - info->dp3_ext_hdmi_reg_num = info_v11->dp3_retimer_set.HdmiRegNum; + info->dp3_ext_hdmi_reg_num = min_t(u8, info_v11->dp3_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp3_ext_hdmi_reg_settings)); for (i = 0; i < info->dp3_ext_hdmi_reg_num; i++) { info->dp3_ext_hdmi_reg_settings[i].i2c_reg_index = info_v11->dp3_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp3_ext_hdmi_reg_settings[i].i2c_reg_val = info_v11->dp3_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp3_ext_hdmi_6g_reg_num = info_v11->dp3_retimer_set.Hdmi6GRegNum; + info->dp3_ext_hdmi_6g_reg_num = min_t(u8, info_v11->dp3_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp3_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp3_ext_hdmi_6g_reg_num; i++) { info->dp3_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v11->dp3_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2805,14 +2813,16 @@ static enum bp_result get_integrated_info_v2_1( info->ext_disp_conn_info.checksum = info_v2_1->extdispconninfo.checksum; info->dp0_ext_hdmi_slv_addr = info_v2_1->dp0_retimer_set.HdmiSlvAddr; - info->dp0_ext_hdmi_reg_num = info_v2_1->dp0_retimer_set.HdmiRegNum; + info->dp0_ext_hdmi_reg_num = min_t(u8, info_v2_1->dp0_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp0_ext_hdmi_reg_settings)); for (i = 0; i < info->dp0_ext_hdmi_reg_num; i++) { info->dp0_ext_hdmi_reg_settings[i].i2c_reg_index = info_v2_1->dp0_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp0_ext_hdmi_reg_settings[i].i2c_reg_val = info_v2_1->dp0_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp0_ext_hdmi_6g_reg_num = info_v2_1->dp0_retimer_set.Hdmi6GRegNum; + info->dp0_ext_hdmi_6g_reg_num = min_t(u8, info_v2_1->dp0_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp0_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp0_ext_hdmi_6g_reg_num; i++) { info->dp0_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v2_1->dp0_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2820,14 +2830,16 @@ static enum bp_result get_integrated_info_v2_1( info_v2_1->dp0_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegVal; } info->dp1_ext_hdmi_slv_addr = info_v2_1->dp1_retimer_set.HdmiSlvAddr; - info->dp1_ext_hdmi_reg_num = info_v2_1->dp1_retimer_set.HdmiRegNum; + info->dp1_ext_hdmi_reg_num = min_t(u8, info_v2_1->dp1_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp1_ext_hdmi_reg_settings)); for (i = 0; i < info->dp1_ext_hdmi_reg_num; i++) { info->dp1_ext_hdmi_reg_settings[i].i2c_reg_index = info_v2_1->dp1_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp1_ext_hdmi_reg_settings[i].i2c_reg_val = info_v2_1->dp1_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp1_ext_hdmi_6g_reg_num = info_v2_1->dp1_retimer_set.Hdmi6GRegNum; + info->dp1_ext_hdmi_6g_reg_num = min_t(u8, info_v2_1->dp1_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp1_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp1_ext_hdmi_6g_reg_num; i++) { info->dp1_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v2_1->dp1_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2835,14 +2847,16 @@ static enum bp_result get_integrated_info_v2_1( info_v2_1->dp1_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegVal; } info->dp2_ext_hdmi_slv_addr = info_v2_1->dp2_retimer_set.HdmiSlvAddr; - info->dp2_ext_hdmi_reg_num = info_v2_1->dp2_retimer_set.HdmiRegNum; + info->dp2_ext_hdmi_reg_num = min_t(u8, info_v2_1->dp2_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp2_ext_hdmi_reg_settings)); for (i = 0; i < info->dp2_ext_hdmi_reg_num; i++) { info->dp2_ext_hdmi_reg_settings[i].i2c_reg_index = info_v2_1->dp2_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp2_ext_hdmi_reg_settings[i].i2c_reg_val = info_v2_1->dp2_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp2_ext_hdmi_6g_reg_num = info_v2_1->dp2_retimer_set.Hdmi6GRegNum; + info->dp2_ext_hdmi_6g_reg_num = min_t(u8, info_v2_1->dp2_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp2_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp2_ext_hdmi_6g_reg_num; i++) { info->dp2_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v2_1->dp2_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; @@ -2850,14 +2864,16 @@ static enum bp_result get_integrated_info_v2_1( info_v2_1->dp2_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegVal; } info->dp3_ext_hdmi_slv_addr = info_v2_1->dp3_retimer_set.HdmiSlvAddr; - info->dp3_ext_hdmi_reg_num = info_v2_1->dp3_retimer_set.HdmiRegNum; + info->dp3_ext_hdmi_reg_num = min_t(u8, info_v2_1->dp3_retimer_set.HdmiRegNum, + ARRAY_SIZE(info->dp3_ext_hdmi_reg_settings)); for (i = 0; i < info->dp3_ext_hdmi_reg_num; i++) { info->dp3_ext_hdmi_reg_settings[i].i2c_reg_index = info_v2_1->dp3_retimer_set.HdmiRegSetting[i].ucI2cRegIndex; info->dp3_ext_hdmi_reg_settings[i].i2c_reg_val = info_v2_1->dp3_retimer_set.HdmiRegSetting[i].ucI2cRegVal; } - info->dp3_ext_hdmi_6g_reg_num = info_v2_1->dp3_retimer_set.Hdmi6GRegNum; + info->dp3_ext_hdmi_6g_reg_num = min_t(u8, info_v2_1->dp3_retimer_set.Hdmi6GRegNum, + ARRAY_SIZE(info->dp3_ext_hdmi_6g_reg_settings)); for (i = 0; i < info->dp3_ext_hdmi_6g_reg_num; i++) { info->dp3_ext_hdmi_6g_reg_settings[i].i2c_reg_index = info_v2_1->dp3_retimer_set.Hdmi6GhzRegSetting[i].ucI2cRegIndex; From adf67034b1f61f7119295208085bfd43f85f56af Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Mon, 11 May 2026 16:46:25 -0400 Subject: [PATCH 198/329] drm/amd/display: Fix NULL deref and buffer over-read in SDP debugfs [Why & How] dp_sdp_message_debugfs_write() dereferences connector->base.state->crtc without checking for NULL. A connector can be connected but not bound to any CRTC (e.g. after hot-plug before the next atomic commit), causing a kernel crash when writing to the sdp_message debugfs node. The function also ignores the user-provided size argument and always passes 36 bytes to copy_from_user(), reading past the user buffer when size < 36. Fix both issues by: - Returning -ENODEV when connector->base.state or state->crtc is NULL - Clamping write_size to min(size, sizeof(data)) Fixes: c7ba3653e977 ("drm/amd/display: Generic SDP message access in amdgpu") Assisted-by: Copilot:claude-opus-4.6 Reviewed-by: Alex Hung Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 6ab4c36a522842ff70474a1c0af2e40e50fc8300) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 2409ac72b166..3a3d01ce0d42 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -1344,8 +1344,13 @@ static ssize_t dp_sdp_message_debugfs_write(struct file *f, const char __user *b if (size == 0) return 0; + if (!connector->base.state || !connector->base.state->crtc) + return -ENODEV; + acrtc_state = to_dm_crtc_state(connector->base.state->crtc->state); + write_size = min_t(size_t, size, sizeof(data)); + r = copy_from_user(data, buf, write_size); write_size -= r; From da48bc4461b8a5ebfb9264c9b191a701d8e99009 Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Tue, 5 May 2026 11:52:15 -0400 Subject: [PATCH 199/329] drm/amd/display: Use krealloc_array() in dal_vector_reserve() [Why & How] dal_vector_reserve() computes the allocation size as "capacity * vector->struct_size" using uint32_t arithmetic, which can silently wrap to a small value on overflow. This would cause krealloc to return a smaller buffer than expected, leading to heap overflows on subsequent vector appends. Replace krealloc() with krealloc_array() which performs an internal overflow check and returns NULL on wrap, preventing the issue. Fixes: 2004f45ef83f ("drm/amd/display: Use kernel alloc/free") Assisted-by: Copilot:claude-opus-4.6 Reviewed-by: Alex Hung Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 37668568641ccc4cc1dbca4923d0a16609dd5707) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/basics/vector.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/basics/vector.c b/drivers/gpu/drm/amd/display/dc/basics/vector.c index e8736c134b8d..60bd9ead928a 100644 --- a/drivers/gpu/drm/amd/display/dc/basics/vector.c +++ b/drivers/gpu/drm/amd/display/dc/basics/vector.c @@ -289,8 +289,8 @@ bool dal_vector_reserve(struct vector *vector, uint32_t capacity) if (capacity <= vector->capacity) return true; - new_container = krealloc(vector->container, - capacity * vector->struct_size, GFP_KERNEL); + new_container = krealloc_array(vector->container, + capacity, vector->struct_size, GFP_KERNEL); if (new_container) { vector->container = new_container; From 49c3da65961fe9857c831d47fa1989084e87514a Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Tue, 5 May 2026 11:50:07 -0400 Subject: [PATCH 200/329] drm/amd/display: Reject gpio_bitshift >= 32 in bios_parser_get_gpio_pin_info() [Why & How] gpio_bitshift is a uint8_t read directly from the VBIOS GPIO pin table. If the value is >= 32, the expression "1 << gpio_bitshift" triggers undefined behaviour in C (shift count exceeds type width). On x86 the shift is silently masked to 5 bits, producing an incorrect GPIO mask that may cause wrong MMIO register bits to be toggled. Validate gpio_bitshift before use and return BP_RESULT_BADBIOSTABLE for out-of-range values. Fixes: ae79c310b1a6 ("drm/amd/display: Add DCE12 bios parser support") Assisted-by: Copilot:claude-opus-4.6 Reviewed-by: Alex Hung Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit eadf438ab8d370b9d19acee9359918c85afeb80d) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index e8d8947c552e..4f213ea865b8 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -701,8 +701,10 @@ static enum bp_result bios_parser_get_gpio_pin_info( info->offset_en = info->offset + 1; info->offset_mask = info->offset - 1; - info->mask = (uint32_t) (1 << - header->gpio_pin[i].gpio_bitshift); + if (header->gpio_pin[i].gpio_bitshift >= 32) + return BP_RESULT_BADBIOSTABLE; + + info->mask = 1u << header->gpio_pin[i].gpio_bitshift; info->mask_y = info->mask + 2; info->mask_en = info->mask + 1; info->mask_mask = info->mask - 1; From f0f3981c43b32cadfe373d636d9e9ca522bb3702 Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Thu, 7 May 2026 15:38:37 -0400 Subject: [PATCH 201/329] drm/amd/display: Clamp HDMI HDCP2 rx_id_list read to buffer size [Why & How] During HDCP 2.x repeater authentication over HDMI, the driver reads the sink's RxStatus register and extracts a 10-bit message size field (max value 1023). This value is used as the read length for the ReceiverID list without being clamped to the size of the destination buffer rx_id_list[177]. A malicious HDMI repeater could advertise a message size larger than the buffer, causing an out-of-bounds write during the I2C read. Clamp the read length in mod_hdcp_read_rx_id_list() to the size of the rx_id_list buffer, matching the approach already used in the DP branch. Fixes: eff682f83c9c ("drm/amd/display: Add DDC handles for HDCP2.2") Assisted-by: Copilot:claude-opus-4.6 Reviewed-by: Alex Hung Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 229212219e4247d9486f8ba41ef087358490be09) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c index 0ca39873f807..324413a090bf 100644 --- a/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c +++ b/drivers/gpu/drm/amd/display/modules/hdcp/hdcp_ddc.c @@ -529,7 +529,8 @@ enum mod_hdcp_status mod_hdcp_read_rx_id_list(struct mod_hdcp *hdcp) } else { status = read(hdcp, MOD_HDCP_MESSAGE_ID_READ_REPEATER_AUTH_SEND_RECEIVERID_LIST, hdcp->auth.msg.hdcp2.rx_id_list, - hdcp->auth.msg.hdcp2.rx_id_list_size); + MIN(hdcp->auth.msg.hdcp2.rx_id_list_size, + sizeof(hdcp->auth.msg.hdcp2.rx_id_list))); } return status; } From ff287df16a1a58aca78b08d1f3ee09fc44da0351 Mon Sep 17 00:00:00 2001 From: Harry Wentland Date: Tue, 12 May 2026 15:24:22 -0400 Subject: [PATCH 202/329] drm/amd/display: Bound VBIOS record-chain walk loops [Why & How] All record-chain walk loops in bios_parser.c and bios_parser2.c use for(;;) and only terminate on a 0xFF record_type sentinel or zero record_size. A malformed VBIOS image missing the terminator record causes unbounded iteration at probe time, potentially hundreds of thousands of iterations with record_size=1. In the final iterations near the BIOS image boundary, struct casts beyond the 2-byte header validated by GET_IMAGE can also read out of bounds. Cap all 14 record-chain walk loops to BIOS_MAX_NUM_RECORD (256) iterations. The atombios.h defines up to 22 distinct record types and atomfirmware.h has 13. Assuming an average of less than 10 records per type (which is reasonable since most are connector- based) 256 is a generous upper bound. Fixes: 4562236b3bc0 ("drm/amd/dc: Add dc display driver (v2)") Assisted-by: Copilot:claude-opus-4.6 Mythos Reviewed-by: Alex Hung Signed-off-by: Harry Wentland Signed-off-by: Ray Wu Tested-by: Daniel Wheeler Signed-off-by: Alex Deucher (cherry picked from commit 95700a3d660287ed657d6892f7be9ffc0e294a93) Cc: stable@vger.kernel.org --- .../gpu/drm/amd/display/dc/bios/bios_parser.c | 15 +++++++---- .../drm/amd/display/dc/bios/bios_parser2.c | 27 ++++++++++++------- .../amd/display/dc/bios/bios_parser_helper.h | 5 ++++ 3 files changed, 33 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c index c307f42fe0b9..507b628abdb5 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser.c @@ -222,6 +222,7 @@ static enum bp_result bios_parser_get_i2c_info(struct dc_bios *dcb, ATOM_COMMON_RECORD_HEADER *header; ATOM_I2C_RECORD *record; struct bios_parser *bp = BP_FROM_DCB(dcb); + int i; if (!info) return BP_RESULT_BADINPUT; @@ -234,7 +235,7 @@ static enum bp_result bios_parser_get_i2c_info(struct dc_bios *dcb, offset = le16_to_cpu(object->usRecordOffset) + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(ATOM_COMMON_RECORD_HEADER, offset); if (!header) @@ -293,11 +294,12 @@ static enum bp_result bios_parser_get_device_tag_record( { ATOM_COMMON_RECORD_HEADER *header; uint32_t offset; + int i; offset = le16_to_cpu(object->usRecordOffset) + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(ATOM_COMMON_RECORD_HEADER, offset); if (!header) @@ -966,6 +968,7 @@ static ATOM_HPD_INT_RECORD *get_hpd_record(struct bios_parser *bp, { ATOM_COMMON_RECORD_HEADER *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -975,7 +978,7 @@ static ATOM_HPD_INT_RECORD *get_hpd_record(struct bios_parser *bp, offset = le16_to_cpu(object->usRecordOffset) + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(ATOM_COMMON_RECORD_HEADER, offset); if (!header) @@ -1670,6 +1673,7 @@ static ATOM_ENCODER_CAP_RECORD_V2 *get_encoder_cap_record( { ATOM_COMMON_RECORD_HEADER *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -1679,7 +1683,7 @@ static ATOM_ENCODER_CAP_RECORD_V2 *get_encoder_cap_record( offset = le16_to_cpu(object->usRecordOffset) + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(ATOM_COMMON_RECORD_HEADER, offset); if (!header) @@ -2769,6 +2773,7 @@ static enum bp_result update_slot_layout_info(struct dc_bios *dcb, { (void)i; unsigned int j; + unsigned int n; struct bios_parser *bp; ATOM_BRACKET_LAYOUT_RECORD *record; ATOM_COMMON_RECORD_HEADER *record_header; @@ -2778,7 +2783,7 @@ static enum bp_result update_slot_layout_info(struct dc_bios *dcb, record = NULL; record_header = NULL; - for (;;) { + for (n = 0; n < BIOS_MAX_NUM_RECORD; n++) { record_header = GET_IMAGE(ATOM_COMMON_RECORD_HEADER, record_offset); if (record_header == NULL) { diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c index 4f213ea865b8..0e1f973326ed 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser2.c @@ -396,6 +396,7 @@ static enum bp_result bios_parser_get_i2c_info(struct dc_bios *dcb, struct atom_i2c_record *record; struct atom_i2c_record dummy_record = {0}; struct bios_parser *bp = BP_FROM_DCB(dcb); + int i; if (!info) return BP_RESULT_BADINPUT; @@ -429,7 +430,7 @@ static enum bp_result bios_parser_get_i2c_info(struct dc_bios *dcb, break; } - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -534,6 +535,7 @@ static struct atom_hpd_int_record *get_hpd_record_for_path_v3(struct bios_parser { struct atom_common_record_header *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -542,7 +544,7 @@ static struct atom_hpd_int_record *get_hpd_record_for_path_v3(struct bios_parser offset = object->disp_recordoffset + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -611,6 +613,7 @@ static struct atom_hpd_int_record *get_hpd_record( { struct atom_common_record_header *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -620,7 +623,7 @@ static struct atom_hpd_int_record *get_hpd_record( offset = le16_to_cpu(object->disp_recordoffset) + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -2195,6 +2198,7 @@ static struct atom_encoder_caps_record *get_encoder_cap_record( { struct atom_common_record_header *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -2203,7 +2207,7 @@ static struct atom_encoder_caps_record *get_encoder_cap_record( offset = object->encoder_recordoffset + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -2232,6 +2236,7 @@ static struct atom_disp_connector_caps_record *get_disp_connector_caps_record( { struct atom_common_record_header *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -2240,7 +2245,7 @@ static struct atom_disp_connector_caps_record *get_disp_connector_caps_record( offset = object->disp_recordoffset + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -2268,6 +2273,7 @@ static struct atom_connector_caps_record *get_connector_caps_record(struct bios_ { struct atom_common_record_header *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -2276,7 +2282,7 @@ static struct atom_connector_caps_record *get_connector_caps_record(struct bios_ offset = object->disp_recordoffset + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -2354,6 +2360,7 @@ static struct atom_connector_speed_record *get_connector_speed_cap_record(struct { struct atom_common_record_header *header; uint32_t offset; + int i; if (!object) { BREAK_TO_DEBUGGER(); /* Invalid object */ @@ -2362,7 +2369,7 @@ static struct atom_connector_speed_record *get_connector_speed_cap_record(struct offset = object->disp_recordoffset + bp->object_info_tbl_offset; - for (;;) { + for (i = 0; i < BIOS_MAX_NUM_RECORD; i++) { header = GET_IMAGE(struct atom_common_record_header, offset); if (!header) @@ -3263,6 +3270,7 @@ static enum bp_result update_slot_layout_info( { unsigned int record_offset; unsigned int j; + unsigned int n; struct atom_display_object_path_v2 *object; struct atom_bracket_layout_record *record; struct atom_common_record_header *record_header; @@ -3284,7 +3292,7 @@ static enum bp_result update_slot_layout_info( (object->disp_recordoffset) + (unsigned int)(bp->object_info_tbl_offset); - for (;;) { + for (n = 0; n < BIOS_MAX_NUM_RECORD; n++) { record_header = (struct atom_common_record_header *) GET_IMAGE(struct atom_common_record_header, @@ -3378,6 +3386,7 @@ static enum bp_result update_slot_layout_info_v2( struct slot_layout_info *slot_layout_info) { unsigned int record_offset; + unsigned int n; struct atom_display_object_path_v3 *object; struct atom_bracket_layout_record_v2 *record; struct atom_common_record_header *record_header; @@ -3400,7 +3409,7 @@ static enum bp_result update_slot_layout_info_v2( (object->disp_recordoffset) + (unsigned int)(bp->object_info_tbl_offset); - for (;;) { + for (n = 0; n < BIOS_MAX_NUM_RECORD; n++) { record_header = (struct atom_common_record_header *) GET_IMAGE(struct atom_common_record_header, diff --git a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h index ab162f2fe577..19fd7aea18f1 100644 --- a/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h +++ b/drivers/gpu/drm/amd/display/dc/bios/bios_parser_helper.h @@ -37,4 +37,9 @@ void bios_set_scratch_critical_state(struct dc_bios *bios, bool state); #define GET_IMAGE(type, offset) ((type *) bios_get_image(&bp->base, offset, sizeof(type))) +/* Upper bound on the number of records in a VBIOS record chain. Prevents + * unbounded looping if the VBIOS image is malformed and lacks a terminator. + */ +#define BIOS_MAX_NUM_RECORD 256 + #endif From 181eda5549c5d9fad3fdb88b050fbf0844d884f8 Mon Sep 17 00:00:00 2001 From: Alysa Liu Date: Wed, 27 May 2026 11:31:35 -0400 Subject: [PATCH 203/329] drm/amdkfd: fix UAF race in destroy_queue_cpsch wait_on_destroy_queue() drops locks to wait for queue resume, allowing a concurrent destroy to free the queue. Use is_being_destroyed flag to serialize destruction. Reviewed-by: Amir Shetaia Signed-off-by: Alysa Liu Signed-off-by: Alex Deucher (cherry picked from commit ac081deaf16a639ea7dff2f285fe421a33c1ade0) --- drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 0d7296c739ed..0a408f95baac 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -2502,6 +2502,9 @@ static int wait_on_destroy_queue(struct device_queue_manager *dqm, if (pdd->qpd.is_debug) return ret; + if (q->properties.is_being_destroyed) + return -EBUSY; + q->properties.is_being_destroyed = true; if (pdd->process->debug_trap_enabled && q->properties.is_suspended) { @@ -2514,6 +2517,9 @@ static int wait_on_destroy_queue(struct device_queue_manager *dqm, dqm_lock(dqm); } + if (ret) + q->properties.is_being_destroyed = false; + return ret; } @@ -2607,7 +2613,7 @@ static int destroy_queue_cpsch(struct device_queue_manager *dqm, return retval; failed_try_destroy_debugged_queue: - + q->properties.is_being_destroyed = false; dqm_unlock(dqm); return retval; } From 1a4a55f181a9584386701f9b4183d9ecc5271b21 Mon Sep 17 00:00:00 2001 From: Alex Deucher Date: Wed, 27 May 2026 15:41:59 -0400 Subject: [PATCH 204/329] drm/amdgpu/sdma7.1: fix support for disable_kq Set the flag in the ring structure. Fixes: 80d4d3a45b86 ("drm/amdgpu/sdma7.1: add support for disable_kq") Reviewed-by: Kent Russell Signed-off-by: Alex Deucher (cherry picked from commit e0a3aa8a6750e8cf067fe2146dc618ffd296d5ef) --- drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c b/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c index 061934a2e93a..9c9bbe043a47 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v7_1.c @@ -1316,6 +1316,7 @@ static int sdma_v7_1_sw_init(struct amdgpu_ip_block *ip_block) ring->ring_obj = NULL; ring->use_doorbell = true; ring->me = i; + ring->no_user_submission = adev->sdma.no_user_submission; for (xcc_id = 0; xcc_id < fls(adev->gfx.xcc_mask); xcc_id++) { if (adev->sdma.instance[i].xcc_id == GET_INST(GC, xcc_id)) From e0153b94a1d104cf4545878c83302672573de65e Mon Sep 17 00:00:00 2001 From: Prike Liang Date: Thu, 14 May 2026 17:21:09 +0800 Subject: [PATCH 205/329] drm/amdgpu: unmap userq for evicting user queue MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit If the driver only preempts queues, there can still be inflight waves, pending dispatch state, or resume/redispatch possibility tied to the same queue. Then the VM/TTM side may proceed to move/unmap queue related BOs during evicting userq objects while shader TCP clients still need to access them. So for eviction, unmap is safer because it makes the queue nonrunnable before memory backing is invalidated. Meanwhile, for a idle queue it's more sutiable for unmapping it rather preempt and unmapping also can save more processing time than preempt. Signed-off-by: Prike Liang Reviewed-by: Christian König Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit d87c9d86727a0bcc95c3009a213a1b27a11b691e) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index cf192500800f..e937099de3e7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -887,7 +887,7 @@ amdgpu_userq_restore_all(struct amdgpu_userq_mgr *uq_mgr) continue; } - r = amdgpu_userq_restore_helper(queue); + r = amdgpu_userq_map_helper(queue); if (r) ret = r; @@ -1124,7 +1124,7 @@ amdgpu_userq_evict_all(struct amdgpu_userq_mgr *uq_mgr) /* Try to unmap all the queues in this process ctx */ xa_for_each(&uq_mgr->userq_xa, queue_id, queue) { - r = amdgpu_userq_preempt_helper(queue); + r = amdgpu_userq_unmap_helper(queue); if (r) ret = r; } From a40412285ec17a69ed728675a56b7ad479c86e36 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Timur=20Krist=C3=B3f?= Date: Mon, 25 May 2026 13:22:04 +0200 Subject: [PATCH 206/329] drm/amdgpu: Align amdgpu_gtt_mgr entries to TLB size on all SI MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit It seems that Pitcairn has the same issues as Tahiti with regards to the TLB size. This commit fixes a VCE1 FW validation timeout on suspend/resume on Pitcairn. Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5336 Signed-off-by: Timur Kristóf Signed-off-by: Alex Deucher (cherry picked from commit 629279e2e798cd161cf74f40aaebfeb16d45eb01) --- drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c index a5d26b943f6d..d23a91d029aa 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c @@ -203,7 +203,7 @@ int amdgpu_gtt_mgr_alloc_entries(struct amdgpu_gtt_mgr *mgr, int r; /* Align to TLB L2 cache entry size to work around "V bit HW bug" */ - if (adev->asic_type == CHIP_TAHITI) { + if (adev->family == AMDGPU_FAMILY_SI) { alignment = 32 * 1024 / AMDGPU_GPU_PAGE_SIZE; num_pages = ALIGN(num_pages, alignment); } From 2493d87bb4c31ec9ca7f0ef7257e33b8b175f913 Mon Sep 17 00:00:00 2001 From: Yang Wang Date: Thu, 21 May 2026 22:36:37 +0800 Subject: [PATCH 207/329] drm/amd/pm: apply SMU 13.0.10 workaround during MP1 unload On SMU v13.0.10, sending PrepareMp1ForUnload with the default parameter may leave the device in an inaccessible state. This can affect runtime power management and partial PnP flows. e.g: kexec, driver unload, boco/d3cold. Pass the required workaround parameter 0x55, when preparing MP1 for unload on SMU v13.0.10, keep the existing behavior for other SMU versions. Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5133 Signed-off-by: Yang Wang Reviewed-by: Kenneth Feng Signed-off-by: Alex Deucher (cherry picked from commit 4e8ee1afeedb8d24dd22cdd5ae9f98a6d76ebe4b) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index 0a7f5fa3c1d3..fa861ec4d700 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -2801,11 +2801,19 @@ static void smu_v13_0_0_i2c_control_fini(struct smu_context *smu) static int smu_v13_0_0_set_mp1_state(struct smu_context *smu, enum pp_mp1_state mp1_state) { + uint32_t param; int ret; switch (mp1_state) { case PP_MP1_STATE_UNLOAD: - ret = smu_cmn_set_mp1_state(smu, mp1_state); + /* + * NOTE: Param 0x55 comes from PMFW 80.31.0, ignored in older versions. + * No PMFW version check required. + */ + param = amdgpu_ip_version(smu->adev, MP1_HWIP, 0) == IP_VERSION(13, 0, 10) ? + 0x55 : 0x00; + ret = smu_cmn_send_smc_msg_with_param(smu, SMU_MSG_PrepareMp1ForUnload, + param, NULL); break; default: /* Ignore others */ From bb204f19e4a115f094a6a3c4d82fcf48862d0766 Mon Sep 17 00:00:00 2001 From: Yang Wang Date: Tue, 19 May 2026 11:18:12 +0800 Subject: [PATCH 208/329] drm/amd/pm: fix smu13 power limit default/cap calculation smu_v13_0_0_get_power_limit() and smu_v13_0_7_get_power_limit() mix runtime power_limit with PP table limits when reporting default/min/max. When current power limit query succeeds, default_power_limit was set to the runtime value instead of the PP table default, and min/max could be derived from inconsistent bases (MsgLimits/runtime), leading to incorrect cap info. Use SocketPowerLimitAc/Dc as the PP default base (pp_limit), keep current_power_limit as runtime value, and derive min/max from pp_limit with OD percentages. Closes: https://gitlab.freedesktop.org/drm/amd/-/work_items/5227 Signed-off-by: Yang Wang Reviewed-by: Kenneth Feng Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher (cherry picked from commit 1eaf26db95901ca70737503a89b831dd763c8453) Cc: stable@vger.kernel.org --- .../drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c | 32 ++++++++++--------- .../drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c | 32 +++++++++++-------- 2 files changed, 35 insertions(+), 29 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c index fa861ec4d700..7f8d4bb47d02 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_0_ppt.c @@ -2390,28 +2390,30 @@ static int smu_v13_0_0_enable_mgpu_fan_boost(struct smu_context *smu) } static int smu_v13_0_0_get_power_limit(struct smu_context *smu, - uint32_t *current_power_limit, - uint32_t *default_power_limit, - uint32_t *max_power_limit, - uint32_t *min_power_limit) + uint32_t *current_power_limit, + uint32_t *default_power_limit, + uint32_t *max_power_limit, + uint32_t *min_power_limit) { struct smu_table_context *table_context = &smu->smu_table; struct smu_13_0_0_powerplay_table *powerplay_table = (struct smu_13_0_0_powerplay_table *)table_context->power_play_table; PPTable_t *pptable = table_context->driver_pptable; SkuTable_t *skutable = &pptable->SkuTable; - uint32_t power_limit, od_percent_upper = 0, od_percent_lower = 0; - uint32_t msg_limit = skutable->MsgLimits.Power[PPT_THROTTLER_PPT0][POWER_SOURCE_AC]; - - if (smu_v13_0_get_current_power_limit(smu, &power_limit)) - power_limit = smu->adev->pm.ac_power ? + uint32_t pp_limit = smu->adev->pm.ac_power ? skutable->SocketPowerLimitAc[PPT_THROTTLER_PPT0] : skutable->SocketPowerLimitDc[PPT_THROTTLER_PPT0]; + uint32_t power_limit = 0, od_percent_upper = 0, od_percent_lower = 0; + int ret; + + if (current_power_limit) { + ret = smu_v13_0_get_current_power_limit(smu, &power_limit); + if (ret) + *current_power_limit = pp_limit; + } - if (current_power_limit) - *current_power_limit = power_limit; if (default_power_limit) - *default_power_limit = power_limit; + *default_power_limit = pp_limit; if (powerplay_table) { if (smu->od_enabled && @@ -2425,15 +2427,15 @@ static int smu_v13_0_0_get_power_limit(struct smu_context *smu, } dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n", - od_percent_upper, od_percent_lower, power_limit); + od_percent_upper, od_percent_lower, pp_limit); if (max_power_limit) { - *max_power_limit = msg_limit * (100 + od_percent_upper); + *max_power_limit = pp_limit * (100 + od_percent_upper); *max_power_limit /= 100; } if (min_power_limit) { - *min_power_limit = power_limit * (100 - od_percent_lower); + *min_power_limit = pp_limit * (100 - od_percent_lower); *min_power_limit /= 100; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c index 5abf2b0703c6..0f774b0920ce 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_7_ppt.c @@ -2372,28 +2372,32 @@ static int smu_v13_0_7_enable_mgpu_fan_boost(struct smu_context *smu) } static int smu_v13_0_7_get_power_limit(struct smu_context *smu, - uint32_t *current_power_limit, - uint32_t *default_power_limit, - uint32_t *max_power_limit, - uint32_t *min_power_limit) + uint32_t *current_power_limit, + uint32_t *default_power_limit, + uint32_t *max_power_limit, + uint32_t *min_power_limit) { struct smu_table_context *table_context = &smu->smu_table; struct smu_13_0_7_powerplay_table *powerplay_table = (struct smu_13_0_7_powerplay_table *)table_context->power_play_table; PPTable_t *pptable = table_context->driver_pptable; SkuTable_t *skutable = &pptable->SkuTable; - uint32_t power_limit, od_percent_upper = 0, od_percent_lower = 0; - uint32_t msg_limit = skutable->MsgLimits.Power[PPT_THROTTLER_PPT0][POWER_SOURCE_AC]; - - if (smu_v13_0_get_current_power_limit(smu, &power_limit)) - power_limit = smu->adev->pm.ac_power ? + uint32_t pp_limit = smu->adev->pm.ac_power ? skutable->SocketPowerLimitAc[PPT_THROTTLER_PPT0] : skutable->SocketPowerLimitDc[PPT_THROTTLER_PPT0]; + uint32_t power_limit = 0, od_percent_upper = 0, od_percent_lower = 0; + int ret; + + if (current_power_limit) { + ret = smu_v13_0_get_current_power_limit(smu, &power_limit); + if (ret) + power_limit = pp_limit; - if (current_power_limit) *current_power_limit = power_limit; + } + if (default_power_limit) - *default_power_limit = power_limit; + *default_power_limit = pp_limit; if (powerplay_table) { if (smu->od_enabled && @@ -2407,15 +2411,15 @@ static int smu_v13_0_7_get_power_limit(struct smu_context *smu, } dev_dbg(smu->adev->dev, "od percent upper:%d, od percent lower:%d (default power: %d)\n", - od_percent_upper, od_percent_lower, power_limit); + od_percent_upper, od_percent_lower, pp_limit); if (max_power_limit) { - *max_power_limit = msg_limit * (100 + od_percent_upper); + *max_power_limit = pp_limit * (100 + od_percent_upper); *max_power_limit /= 100; } if (min_power_limit) { - *min_power_limit = power_limit * (100 - od_percent_lower); + *min_power_limit = pp_limit * (100 - od_percent_lower); *min_power_limit /= 100; } From ee193c5bbd5e2b56bbeb54ef554414b43a6fc896 Mon Sep 17 00:00:00 2001 From: Yang Wang Date: Fri, 29 May 2026 11:47:31 +0800 Subject: [PATCH 209/329] drm/amd/pm: mark metrics.energy_accumulator is invalid for smu 14.0.2 EnergyAccumulator is unsupported on SMU 14.0.2, mark it invalid. Signed-off-by: Yang Wang Reviewed-by: Asad Kamal Signed-off-by: Alex Deucher (cherry picked from commit 646b05043eeed04b51c14aad22a400a8250af4b7) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c index 5ce4e982ca33..fdc1456b885c 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_2_ppt.c @@ -2152,7 +2152,6 @@ static ssize_t smu_v14_0_2_get_gpu_metrics(struct smu_context *smu, metrics->Vcn1ActivityPercentage); gpu_metrics->average_socket_power = metrics->AverageSocketPower; - gpu_metrics->energy_accumulator = metrics->EnergyAccumulator; if (metrics->AverageGfxActivity <= SMU_14_0_2_BUSY_THRESHOLD) gpu_metrics->average_gfxclk_frequency = metrics->AverageGfxclkFrequencyPostDs; From a169b326bab4a3617a9b1d34beddaaa6f798aa20 Mon Sep 17 00:00:00 2001 From: Yang Wang Date: Mon, 11 May 2026 16:33:37 +0800 Subject: [PATCH 210/329] drm/amd/pm: zero unused SMU argument registers SMU messages may use fewer arguments than the available argument registers, the previous code only wrote used registers and left the rest unchanged, so stale values from a prior message could persist. Write all argument registers for each message and zero the unused tail to keep command arguments deterministic and avoid unintended carry-over. Signed-off-by: Yang Wang Reviewed-by: Asad Kamal Signed-off-by: Alex Deucher (cherry picked from commit e03b635f61f77ebd5107ef82f48e3221cb695856) --- drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 90c7127beabf..fe97fda8bfe9 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -272,11 +272,15 @@ static void __smu_msg_v1_send(struct smu_msg_ctl *ctl, u16 index, { struct amdgpu_device *adev = ctl->smu->adev; struct smu_msg_config *cfg = &ctl->config; + u32 arg; int i; WREG32(cfg->resp_reg, 0); - for (i = 0; i < args->num_args; i++) - WREG32(cfg->arg_regs[i], args->args[i]); + for (i = 0; i < cfg->num_arg_regs; i++) { + /* NOTE: Clear unused argument registers to avoid stale values. */ + arg = i < args->num_args ? args->args[i] : 0; + WREG32(cfg->arg_regs[i], arg); + } WREG32(cfg->msg_reg, index); } From ae4e30f24d67075dc975002effa68d424c7ff7e3 Mon Sep 17 00:00:00 2001 From: Harish Kasiviswanathan Date: Tue, 28 Apr 2026 17:45:06 -0400 Subject: [PATCH 211/329] drm/amdgpu: Use asic specific pte_addr_mask For PTE creation use asic specific physical page base address mask v2: Change variable name from pa_mask to pte_addr_mask Signed-off-by: Harish Kasiviswanathan Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher (cherry picked from commit 2ea989885941a6e5607ef86dbe309e90b7191f21) --- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 2 +- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c | 4 ++++ drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c | 1 + drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 1 + 9 files changed, 12 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 3d9497d121ca..13a5acdf8da3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -170,7 +170,7 @@ int amdgpu_gmc_set_pte_pde(struct amdgpu_device *adev, void *cpu_pt_addr, /* * The following is for PTE only. GART does not have PDEs. */ - value = addr & 0x0000FFFFFFFFF000ULL; + value = addr & adev->gmc.pte_addr_mask; value |= flags; writeq(value, ptr + (gpu_page_idx * 8)); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index 6ab4c1e297fc..d03536b969b5 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -280,6 +280,7 @@ struct amdgpu_gmc { u64 real_vram_size; int vram_mtrr; u64 mc_mask; + uint64_t pte_addr_mask; const struct firmware *fw; /* MC firmware */ uint32_t fw_version; struct amdgpu_irq_src vm_fault; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index e1ace7d44ffd..f5bdfea54afa 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -847,6 +847,7 @@ static int gmc_v10_0_sw_init(struct amdgpu_ip_block *ip_block) * internal address space. */ adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */ + adev->gmc.pte_addr_mask = 0x0000FFFFFFFFF000ULL; /* 48 bit PA */ r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c index 94d6631ce0bc..807bd180b9d4 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c @@ -821,6 +821,7 @@ static int gmc_v11_0_sw_init(struct amdgpu_ip_block *ip_block) * internal address space. */ adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */ + adev->gmc.pte_addr_mask = 0x0000FFFFFFFFF000ULL; /* 48 bit PA */ r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c index e10ac9788d13..52c161c2df0a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c @@ -814,6 +814,7 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) { int r, vram_width = 0, vram_type = 0, vram_vendor = 0; struct amdgpu_device *adev = ip_block->adev; + uint64_t pte_addr_mask = 0; int i; adev->mmhub.funcs->init(adev); @@ -843,6 +844,7 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) * block size 512 (9bit) */ amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); + pte_addr_mask = 0x0000FFFFFFFFF000ULL; /* 48 bit PA */ break; case IP_VERSION(12, 1, 0): bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0), @@ -855,6 +857,7 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) * block size 512 (9bit) */ amdgpu_vm_adjust_size(adev, 128 * 1024 * 1024, 9, 4, 57); + pte_addr_mask = 0x000FFFFFFFFFF000ULL; /* 52 bit PA */ break; default: break; @@ -911,6 +914,7 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) * internal address space. */ adev->gmc.mc_mask = AMDGPU_GMC_HOLE_MASK; + adev->gmc.pte_addr_mask = pte_addr_mask; r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c index cc272a96fcef..6aa581b1c148 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v6_0.c @@ -836,6 +836,7 @@ static int gmc_v6_0_sw_init(struct amdgpu_ip_block *ip_block) amdgpu_vm_adjust_size(adev, 64, 9, 1, 40); adev->gmc.mc_mask = 0xffffffffffULL; + adev->gmc.pte_addr_mask = 0x000000FFFFFFF000ULL; r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c index bb16ba2ef6fd..2b0362c4d9eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c @@ -1016,6 +1016,7 @@ static int gmc_v7_0_sw_init(struct amdgpu_ip_block *ip_block) * internal address space. */ adev->gmc.mc_mask = 0xffffffffffULL; /* 40 bit MC */ + adev->gmc.pte_addr_mask = 0x000000FFFFFFF000ULL; /* 40 bit PA */ r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c index a59174f6bcc1..fbccfcb3d7cf 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c @@ -1131,6 +1131,7 @@ static int gmc_v8_0_sw_init(struct amdgpu_ip_block *ip_block) * internal address space. */ adev->gmc.mc_mask = 0xffffffffffULL; /* 40 bit MC */ + adev->gmc.pte_addr_mask = 0x000000FFFFFFF000ULL; /* 40 bit PA */ r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(40)); if (r) { diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index e7b78027002b..c6dbe25f2bd9 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -1994,6 +1994,7 @@ static int gmc_v9_0_sw_init(struct amdgpu_ip_block *ip_block) * internal address space. */ adev->gmc.mc_mask = 0xffffffffffffULL; /* 48 bit MC */ + adev->gmc.pte_addr_mask = 0x0000FFFFFFFFF000ULL; /* 48 bit PA */ dma_addr_bits = amdgpu_ip_version(adev, GC_HWIP, 0) >= IP_VERSION(9, 4, 2) ? From e3fa02872e223d24a925eb876b31c00d833bce7c Mon Sep 17 00:00:00 2001 From: Harish Kasiviswanathan Date: Tue, 12 May 2026 10:57:49 -0400 Subject: [PATCH 212/329] drm/amdgpu: drm/amdgpu: Set correct DMA mask for gfx12.1 Set correct DMA mask for gfx12 Signed-off-by: Harish Kasiviswanathan Reviewed-by: Lijo Lazar Signed-off-by: Alex Deucher (cherry picked from commit a2ef14ee2593b48242b8d90f229f71c1710529da) --- drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c index 52c161c2df0a..8dc9c053897b 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v12_0.c @@ -812,7 +812,7 @@ static int gmc_v12_0_gart_init(struct amdgpu_device *adev) static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) { - int r, vram_width = 0, vram_type = 0, vram_vendor = 0; + int r, vram_width = 0, vram_type = 0, vram_vendor = 0, dma_addr_bits; struct amdgpu_device *adev = ip_block->adev; uint64_t pte_addr_mask = 0; int i; @@ -845,6 +845,7 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) */ amdgpu_vm_adjust_size(adev, 256 * 1024, 9, 3, 48); pte_addr_mask = 0x0000FFFFFFFFF000ULL; /* 48 bit PA */ + dma_addr_bits = 44; break; case IP_VERSION(12, 1, 0): bitmap_set(adev->vmhubs_mask, AMDGPU_GFXHUB(0), @@ -858,9 +859,12 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) */ amdgpu_vm_adjust_size(adev, 128 * 1024 * 1024, 9, 4, 57); pte_addr_mask = 0x000FFFFFFFFFF000ULL; /* 52 bit PA */ + dma_addr_bits = 52; break; default: - break; + dev_warn(adev->dev, "Unrecognized GC IP version: 0x%08x\n", + amdgpu_ip_version(adev, GC_HWIP, 0)); + return -EINVAL; } /* This interrupt is VMC page fault.*/ @@ -916,13 +920,13 @@ static int gmc_v12_0_sw_init(struct amdgpu_ip_block *ip_block) adev->gmc.mc_mask = AMDGPU_GMC_HOLE_MASK; adev->gmc.pte_addr_mask = pte_addr_mask; - r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(44)); + r = dma_set_mask_and_coherent(adev->dev, DMA_BIT_MASK(dma_addr_bits)); if (r) { drm_warn(adev_to_drm(adev), "No suitable DMA available.\n"); return r; } - adev->need_swiotlb = drm_need_swiotlb(44); + adev->need_swiotlb = drm_need_swiotlb(dma_addr_bits); r = gmc_v12_0_mc_init(adev); if (r) From 58bafc666c484b21839a2d27e923ae1b2727a1df Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Christian=20K=C3=B6nig?= Date: Wed, 18 Feb 2026 13:05:46 +0100 Subject: [PATCH 213/329] drm/amdgpu: fix waiting for all submissions for userptrs MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Wait for all submissions when userptrs need to be invalidated by the MMU notifier, not just the one the userptr was involved into. Signed-off-by: Christian König Reviewed-by: Vitaly Prosyak Tested-by: Vitaly Prosyak Reviewed-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 91250893cbaa25c86872deca95a540d08de1f91e) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c index 5bfa5a84b09c..e452444b33b0 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_hmm.c @@ -67,6 +67,7 @@ static bool amdgpu_hmm_invalidate_gfx(struct mmu_interval_notifier *mni, { struct amdgpu_bo *bo = container_of(mni, struct amdgpu_bo, notifier); struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev); + struct amdgpu_bo *vm_root = bo->vm_bo->vm->root.bo; long r; if (!mmu_notifier_range_blockable(range)) @@ -77,8 +78,9 @@ static bool amdgpu_hmm_invalidate_gfx(struct mmu_interval_notifier *mni, mmu_interval_set_seq(mni, cur_seq); amdgpu_vm_bo_invalidate(bo, false); - r = dma_resv_wait_timeout(bo->tbo.base.resv, DMA_RESV_USAGE_BOOKKEEP, - false, MAX_SCHEDULE_TIMEOUT); + r = dma_resv_wait_timeout(vm_root->tbo.base.resv, + DMA_RESV_USAGE_BOOKKEEP, false, + MAX_SCHEDULE_TIMEOUT); mutex_unlock(&adev->notifier_lock); if (r <= 0) DRM_ERROR("(%ld) failed to wait for user bo\n", r); From 40bab7c606702e8ea3e95c6f30c99cc8295826af Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Timur=20Krist=C3=B3f?= Date: Mon, 25 May 2026 13:45:02 +0200 Subject: [PATCH 214/329] drm/amdgpu/gfxhub: Program CRASH_ON_*_FAULT bits to 0 as needed MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When the fault stop mode isn't AMDGPU_VM_FAULT_STOP_ALWAYS, these bits should be programmed to 0. Program CRASH_ON_NO_RETRY_FAULT and CRASH_ON_RETRY_FAULT always, to make sure to clear the bits when we don't want to crash. Signed-off-by: Timur Kristóf Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit d0cd99e73090700b7a942b98a3327ec966597d0a) --- drivers/gpu/drm/amd/amdgpu/gfxhub_v11_5_0.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v12_0.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v12_1.c | 14 ++++++-------- drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c | 10 ++++------ drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c | 10 ++++------ 9 files changed, 38 insertions(+), 56 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v11_5_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v11_5_0.c index f9949fedfbb9..f2fe6f5bc7f7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v11_5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v11_5_0.c @@ -449,12 +449,10 @@ static void gfxhub_v11_5_0_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, regGCVM_L2_PROTECTION_FAULT_CNTL, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_0.c index 7609b9cecae8..efcaca70c27a 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_0.c @@ -454,12 +454,10 @@ static void gfxhub_v12_0_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, regGCVM_L2_PROTECTION_FAULT_CNTL, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_1.c index 3544eb42dca6..4c2fd1e6616e 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_1.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v12_1.c @@ -633,19 +633,17 @@ static void gfxhub_v12_1_xcc_set_fault_enable_default(struct amdgpu_device *adev tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL_LO32, OTHER_CLIENT_ID_NO_RETRY_FAULT_INTERRUPT, value); - if (!value) - tmp = REG_SET_FIELD(tmp, - GCVM_L2_PROTECTION_FAULT_CNTL_LO32, - CRASH_ON_NO_RETRY_FAULT, 1); + tmp = REG_SET_FIELD(tmp, + GCVM_L2_PROTECTION_FAULT_CNTL_LO32, + CRASH_ON_NO_RETRY_FAULT, !value); WREG32_SOC15(GC, GET_INST(GC, i), regGCVM_L2_PROTECTION_FAULT_CNTL_LO32, tmp); tmp = RREG32_SOC15(GC, GET_INST(GC, i), regGCVM_L2_PROTECTION_FAULT_CNTL_HI32); - if (!value) - tmp = REG_SET_FIELD(tmp, - GCVM_L2_PROTECTION_FAULT_CNTL_HI32, - CRASH_ON_RETRY_FAULT, 1); + tmp = REG_SET_FIELD(tmp, + GCVM_L2_PROTECTION_FAULT_CNTL_HI32, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, GET_INST(GC, i), regGCVM_L2_PROTECTION_FAULT_CNTL_HI32, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index a7bfc9f41d0e..bfe247b1a333 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -403,12 +403,10 @@ static void gfxhub_v1_0_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, mmVM_L2_PROTECTION_FAULT_CNTL, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c index 6c03bf9f1ae8..fbdf46070b38 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c @@ -516,12 +516,10 @@ static void gfxhub_v1_2_xcc_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, VM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, GET_INST(GC, i), regVM_L2_PROTECTION_FAULT_CNTL, tmp); } } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index 793faf62cb07..9ea593e2c719 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -418,12 +418,10 @@ static void gfxhub_v2_0_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_CNTL, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c index aceb8447feac..30b90d35abd0 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c @@ -449,12 +449,10 @@ static void gfxhub_v2_1_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, mmGCVM_L2_PROTECTION_FAULT_CNTL, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c index abe30c8bd2ba..f089f70571aa 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0.c @@ -446,12 +446,10 @@ static void gfxhub_v3_0_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, regGCVM_L2_PROTECTION_FAULT_CNTL, tmp); } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c index b3ef6e71811f..128115a2cb45 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v3_0_3.c @@ -434,12 +434,10 @@ static void gfxhub_v3_0_3_set_fault_enable_default(struct amdgpu_device *adev, WRITE_PROTECTION_FAULT_ENABLE_DEFAULT, value); tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, EXECUTE_PROTECTION_FAULT_ENABLE_DEFAULT, value); - if (!value) { - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_NO_RETRY_FAULT, 1); - tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, - CRASH_ON_RETRY_FAULT, 1); - } + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_NO_RETRY_FAULT, !value); + tmp = REG_SET_FIELD(tmp, GCVM_L2_PROTECTION_FAULT_CNTL, + CRASH_ON_RETRY_FAULT, !value); WREG32_SOC15(GC, 0, regGCVM_L2_PROTECTION_FAULT_CNTL, tmp); } From e47b0056a08dc70430ffc44bbf62197e7d1ff8ea Mon Sep 17 00:00:00 2001 From: Vitaly Prosyak Date: Fri, 29 May 2026 13:50:38 -0400 Subject: [PATCH 215/329] drm/amdgpu: set noretry=1 as default for GFX 10.1.x (Navi10/12/14) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Problem: While developing the amd_close_race IGT test (which intentionally triggers execute permission faults by removing VM_PAGE_EXECUTABLE from GPU page table entries), we discovered that on Navi10 (GFX 10.1.x) these faults produce zero diagnostic output. The GPU simply hangs silently for ~10s until the scheduler timeout fires. There is no way to distinguish an execute permission fault from any other type of GPU hang. Root cause: GFX 10.1.x defaults to noretry=0, which sets RETRY_PERMISSION_OR_INVALID_PAGE_FAULT=1 in the GFXHUB UTCL2 registers (gfxhub_v2_0.c line 313). With this bit set, permission faults (valid PTE, wrong R/W/X bits) are handled entirely within the UTCL1/UTCL2 hardware loop: UTCL2 returns an XNACK to UTCL1, and UTCL1 re-requests the translation indefinitely, expecting software to eventually fix the permission bits (as happens in SVM/HMM recovery). No interrupt of any kind reaches the IH ring. This is different from invalid-page faults (V=0) which DO generate a retry fault interrupt that the driver can escalate to a no-retry fault. Permission faults with valid PTEs loop silently forever in hardware. GFX 10.3+ already defaults to noretry=1, which makes permission faults generate immediate L2 protection fault interrupts. GFX 10.1.x was inadvertently left out of this default. Fix: Change the noretry=1 threshold from IP_VERSION(10, 3, 0) to IP_VERSION(10, 1, 0) in amdgpu_gmc_noretry_set(). This is a one-line change that aligns GFX 10.1.x behavior with GFX 10.3+ and all newer generations. With noretry=1, the existing non-retry fault handler (gmc_v10_0_process_interrupt) already decodes and prints the full GCVM_L2_PROTECTION_FAULT_STATUS register including PERMISSION_FAULTS, faulting address, VMID, PASID, and process name. No additional logging code is needed — the fix is purely routing permission faults to the existing, fully-capable non-retry interrupt handler. v2: Dropped GFX10-specific logging from gmc_v10_0.c and kfd_int_process_v10.c (Felix Kuehling). v1 added logging in the retry fault handler, but with noretry=1 permission faults take the non-retry path — the v1 retry handler code was dead and would never execute. Tested on Navi10 (GFX 10.1.10): - Execute permission faults now produce immediate, clear output: [gfxhub] page fault (src_id:0 ring:64 vmid:4 pasid:592) Process amd_close_race pid 13380 thread amd_close_race pid 13384 in page at address 0x40001000 from client 0x1b (UTCL2) GCVM_L2_PROTECTION_FAULT_STATUS:0x00700881 PERMISSION_FAULTS: 0x8 - No regressions with properly-mapped GPU workloads Cc: Christian Koenig Cc: Alex Deucher Cc: Felix Kuehling Signed-off-by: Vitaly Prosyak Acked-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit eb21edd24c40d81066753f8ac6f23bce15745395) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 13a5acdf8da3..c076c5f06e77 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -1003,7 +1003,7 @@ void amdgpu_gmc_noretry_set(struct amdgpu_device *adev) gc_ver == IP_VERSION(9, 4, 3) || gc_ver == IP_VERSION(9, 4, 4) || gc_ver == IP_VERSION(9, 5, 0) || - gc_ver >= IP_VERSION(10, 3, 0)); + gc_ver >= IP_VERSION(10, 1, 0)); if (!amdgpu_sriov_xnack_support(adev)) gmc->noretry = 1; From 2bd550b547deabef98bd3b017ff743b7c34d3a6d Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sat, 23 May 2026 16:56:46 +0000 Subject: [PATCH 216/329] drm/amdkfd: fix NULL dereference in get_queue_ids() When usr_queue_id_array is NULL and num_queues is non-zero, get_queue_ids() returns NULL. The callers check only IS_ERR() on the return value; since IS_ERR(NULL) == false the check passes, and suspend_queues() calls q_array_invalidate() which immediately dereferences NULL while iterating num_queues times. Userspace can trigger this via kfd_ioctl_set_debug_trap() by supplying num_queues > 0 with a zero queue_array_ptr, causing a kernel panic. A NULL usr_queue_id_array with num_queues == 0 is a legitimate no-op (q_array_invalidate never executes, and resume_queues already guards all queue_ids dereferences behind a NULL check). Return ERR_PTR(-EINVAL) only when num_queues is non-zero and the pointer is absent; both callers already propagate IS_ERR() returns correctly to userspace. Fixes: a70a93fa568b ("drm/amdkfd: add debug suspend and resume process queues operation") Signed-off-by: Muhammad Bilal Signed-off-by: Alex Deucher (cherry picked from commit f165a82cdf503884bb1797771c61b2fcc72113d4) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 0a408f95baac..31187ddbb79e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -3315,7 +3315,7 @@ static void copy_context_work_handler(struct work_struct *work) static uint32_t *get_queue_ids(uint32_t num_queues, uint32_t *usr_queue_id_array) { if (!usr_queue_id_array) - return NULL; + return num_queues ? ERR_PTR(-EINVAL) : NULL; if (num_queues > KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) return ERR_PTR(-EINVAL); From 352ea59028ea48a6fff77f19ae28f98f71946a80 Mon Sep 17 00:00:00 2001 From: Andrew Martin Date: Thu, 28 May 2026 12:54:39 -0400 Subject: [PATCH 217/329] drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11 The v11 MQD manager incorrectly assigned the CP-compute variants of checkpoint_mqd/restore_mqd for KFD_MQD_TYPE_SDMA queues. These functions use sizeof(struct v11_compute_mqd) (2048 bytes) instead of sizeof(struct v11_sdma_mqd) (512 bytes), causing a 1536-byte overflow. During CRIU checkpoint of an SDMA queue on Navi3x: - checkpoint_mqd() reads 2048 bytes from a 512-byte SDMA MQD buffer, leaking 1536 bytes of adjacent GTT memory to userspace During CRIU restore: - restore_mqd() writes 2048 bytes into a 512-byte SDMA MQD buffer, corrupting 1536 bytes of adjacent GTT memory (often the ring buffer or neighboring MQDs) This is a copy-paste regression unique to v11. All other ASIC backends (cik, vi, v9, v10, v12) correctly use the SDMA-specific variants. Add checkpoint_mqd_sdma() and restore_mqd_sdma() functions that properly handle the smaller v11_sdma_mqd structure, matching the pattern used in other MQD managers. Fixes: cc009e613de6 ("drm/amdkfd: Add KFD support for soc21 v3") Assisted-by: Claude:Sonnet 4-5 Signed-off-by: Andrew Martin Acked-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 6fa41db7ffdec97d62433adf03b7b9b759af8c2c) Cc: stable@vger.kernel.org --- .../gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c | 49 ++++++++++++++++--- 1 file changed, 41 insertions(+), 8 deletions(-) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c index a1e3cf2384dd..527c531676e4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager_v11.c @@ -320,8 +320,7 @@ static void checkpoint_mqd(struct mqd_manager *mm, void *mqd, void *mqd_dst, voi static void restore_mqd(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, - struct queue_properties *qp, - const void *mqd_src, + struct queue_properties *qp, const void *mqd_src, const void *ctl_stack_src, const u32 ctl_stack_size) { uint64_t addr; @@ -337,14 +336,48 @@ static void restore_mqd(struct mqd_manager *mm, void **mqd, *gart_addr = addr; m->cp_hqd_pq_doorbell_control = - qp->doorbell_off << - CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; - pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", - m->cp_hqd_pq_doorbell_control); + qp->doorbell_off << CP_HQD_PQ_DOORBELL_CONTROL__DOORBELL_OFFSET__SHIFT; + pr_debug("cp_hqd_pq_doorbell_control 0x%x\n", m->cp_hqd_pq_doorbell_control); qp->is_active = 0; } +static void checkpoint_mqd_sdma(struct mqd_manager *mm, + void *mqd, + void *mqd_dst, + void *ctl_stack_dst) +{ + struct v11_sdma_mqd *m; + + m = get_sdma_mqd(mqd); + + memcpy(mqd_dst, m, sizeof(struct v11_sdma_mqd)); +} + +static void restore_mqd_sdma(struct mqd_manager *mm, void **mqd, + struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, + struct queue_properties *qp, + const void *mqd_src, + const void *ctl_stack_src, + const u32 ctl_stack_size) +{ + uint64_t addr; + struct v11_sdma_mqd *m; + + m = (struct v11_sdma_mqd *) mqd_mem_obj->cpu_ptr; + addr = mqd_mem_obj->gpu_addr; + + memcpy(m, mqd_src, sizeof(*m)); + + m->sdmax_rlcx_doorbell_offset = + qp->doorbell_off << SDMA0_QUEUE0_DOORBELL_OFFSET__OFFSET__SHIFT; + + *mqd = m; + if (gart_addr) + *gart_addr = addr; + + qp->is_active = 0; +} static void init_mqd_hiq(struct mqd_manager *mm, void **mqd, struct kfd_mem_obj *mqd_mem_obj, uint64_t *gart_addr, @@ -529,8 +562,8 @@ struct mqd_manager *mqd_manager_init_v11(enum KFD_MQD_TYPE type, mqd->update_mqd = update_mqd_sdma; mqd->destroy_mqd = kfd_destroy_mqd_sdma; mqd->is_occupied = kfd_is_occupied_sdma; - mqd->checkpoint_mqd = checkpoint_mqd; - mqd->restore_mqd = restore_mqd; + mqd->checkpoint_mqd = checkpoint_mqd_sdma; + mqd->restore_mqd = restore_mqd_sdma; mqd->mqd_size = sizeof(struct v11_sdma_mqd); mqd->mqd_stride = kfd_mqd_stride; #if defined(CONFIG_DEBUG_FS) From 2a07f3fa4998e2ef248e1f55cd3776348758aec0 Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Mon, 25 May 2026 13:18:00 +0530 Subject: [PATCH 218/329] drm/amdgpu/userq: remove the vital queue unmap logging MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Mesa userqueues free does not wait for the free to complete and go ahead in unmapping the vital bos while kernel is still in queue free and corresponding cleanup. So ideally we don't need the logging for that and hence remove the warn message as this is expected behaviour and functionally, we are making sure to wait for the required fences before unmap. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 758a868043dcb07eca923bc451c16da3e73dc47c) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 10 +++------- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h | 3 +-- drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c | 2 +- 3 files changed, 5 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index e937099de3e7..986ef5ca0087 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -1344,8 +1344,7 @@ int amdgpu_userq_start_sched_for_enforce_isolation(struct amdgpu_device *adev, } void amdgpu_userq_gem_va_unmap_validate(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, - uint64_t saddr) + struct amdgpu_bo_va_mapping *mapping) { u32 ip_mask = amdgpu_userq_get_supported_ip_mask(adev); struct amdgpu_bo_va *bo_va = mapping->bo_va; @@ -1354,12 +1353,9 @@ void amdgpu_userq_gem_va_unmap_validate(struct amdgpu_device *adev, if (!ip_mask) return; - dev_warn_once(adev->dev, "now unmapping a vital queue va:%llx\n", saddr); /** - * The userq VA mapping reservation should include the eviction fence, - * if the eviction fence can't signal successfully during unmapping, - * then driver will warn to flag this improper unmap of the userq VA. - * Note: The eviction fence may be attached to different BOs, and this + * The userq VA mapping reservation should include the eviction fence. + * Note: The eviction fence may be attached to different BOs and this * unmap is only for one kind of userq VAs, so at this point suppose * the eviction fence is always unsignaled. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h index 28cfc6682333..d1751febaefe 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.h @@ -182,6 +182,5 @@ int amdgpu_userq_input_va_validate(struct amdgpu_device *adev, u64 addr, u64 expected_size, u64 *va_out); void amdgpu_userq_gem_va_unmap_validate(struct amdgpu_device *adev, - struct amdgpu_bo_va_mapping *mapping, - uint64_t saddr); + struct amdgpu_bo_va_mapping *mapping); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index c9f88ecce1a7..381901bc539f 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -2006,7 +2006,7 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev, * from user space. */ if (unlikely(bo_va->userq_va_mapped)) - amdgpu_userq_gem_va_unmap_validate(adev, mapping, saddr); + amdgpu_userq_gem_va_unmap_validate(adev, mapping); list_del(&mapping->list); amdgpu_vm_it_remove(mapping, &vm->va); From 5af28a22ce834544bb22efbba30ffb837098b2b8 Mon Sep 17 00:00:00 2001 From: Prike Liang Date: Tue, 26 May 2026 10:25:26 +0800 Subject: [PATCH 219/329] drm/amdgpu: improve the userq seq BO free bit lookup MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Use find_next_zero_bit() to locate the next free seq slot bit instead of the current walk, for more efficient bitmap scanning. Signed-off-by: Prike Liang Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit ff905a9b6228de9eedd0db71ecb1bdde91fb898d) --- drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c index f4be19223588..21a225b0116a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_seq64.c @@ -173,16 +173,17 @@ void amdgpu_seq64_unmap(struct amdgpu_device *adev, struct amdgpu_fpriv *fpriv) int amdgpu_seq64_alloc(struct amdgpu_device *adev, u64 *va, u64 *gpu_addr, u64 **cpu_addr) { - unsigned long bit_pos; + unsigned long bit_pos = 0; - for (;;) { - bit_pos = find_first_zero_bit(adev->seq64.used, adev->seq64.num_sem); + do { + bit_pos = find_next_zero_bit(adev->seq64.used, + adev->seq64.num_sem, bit_pos); if (bit_pos >= adev->seq64.num_sem) return -ENOSPC; - if (!test_and_set_bit(bit_pos, adev->seq64.used)) break; - } + bit_pos++; + } while (1); *va = bit_pos * sizeof(u64) + amdgpu_seq64_get_va_base(adev); From 14ad7e1e6e2cd44f866e2dbb3f6a2b2f4a39b96d Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Mon, 25 May 2026 09:56:23 +0530 Subject: [PATCH 220/329] drm/amdgpu/userq: move wptr_obj cleanup in mqd_destroy MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In case when queue_create fails and mqd has already been allocated and hence wptr_obj is not cleaned up. So moving that cleanup part to mqd_destroy so it takes care of all the cases of clean up and during tear down of the queue. Signed-off-by: Sunil Khatri Reviewed-by: Christian König Signed-off-by: Alex Deucher (cherry picked from commit 43355f62cd2ef5386c2693df537c232ea0f2ce6c) --- drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c | 4 ---- drivers/gpu/drm/amd/amdgpu/mes_userqueue.c | 5 +++++ 2 files changed, 5 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c index 986ef5ca0087..59ffaa7b61c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_userq.c @@ -532,10 +532,6 @@ amdgpu_userq_destroy(struct amdgpu_userq_mgr *uq_mgr, struct amdgpu_usermode_que amdgpu_bo_unreserve(queue->db_obj.obj); amdgpu_bo_unref(&queue->db_obj.obj); - amdgpu_bo_reserve(queue->wptr_obj.obj, true); - amdgpu_bo_unpin(queue->wptr_obj.obj); - amdgpu_bo_unreserve(queue->wptr_obj.obj); - amdgpu_bo_unref(&queue->wptr_obj.obj); kfree(queue); pm_runtime_put_autosuspend(adev_to_drm(adev)->dev); diff --git a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c index 98aa00eeb2f4..4cbd46f53e85 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_userqueue.c @@ -467,6 +467,11 @@ static void mes_userq_mqd_destroy(struct amdgpu_usermode_queue *queue) kfree(queue->userq_prop); amdgpu_bo_free_kernel(&queue->mqd.obj, &queue->mqd.gpu_addr, &queue->mqd.cpu_ptr); + + amdgpu_bo_reserve(queue->wptr_obj.obj, true); + amdgpu_bo_unpin(queue->wptr_obj.obj); + amdgpu_bo_unreserve(queue->wptr_obj.obj); + amdgpu_bo_unref(&queue->wptr_obj.obj); } static int mes_userq_preempt(struct amdgpu_usermode_queue *queue) From ec4c462e2d8161b32038e21e7187f4a15fe1661d Mon Sep 17 00:00:00 2001 From: Donet Tom Date: Wed, 27 May 2026 18:49:31 +0530 Subject: [PATCH 221/329] drm/amdgpu: Fix incorrect VRAM GART mappings on non-4K page size systems MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When mapping VRAM pages into the GART page table, amdgpu_gart_map_vram_range() assumes that the system page size is the same as the GPU page size. On systems with non-4K page sizes, multiple GPU pages can exist within a single CPU page. As a result, the mappings are created incorrectly because fewer page table entries are programmed than required. Fix this by programming the mappings correctly for non-4K page size systems. Fixes: 237d623ae659 ("drm/amdgpu/gart: Add helper to bind VRAM pages (v2)") Reviewed-by: Timur Kristóf Reviewed-by: Christian König Signed-off-by: Donet Tom Signed-off-by: Alex Deucher (cherry picked from commit a8f0bc22388f74e0cf4ed8b7d1846c580eaf44cc) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c index b6f849d51c2e..c4c21dbbbdbf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c @@ -394,7 +394,8 @@ void amdgpu_gart_map_vram_range(struct amdgpu_device *adev, uint64_t pa, uint64_t start_page, uint64_t num_pages, uint64_t flags, void *dst) { - u32 i, idx; + u32 i, j, t, idx; + u64 page_base; /* The SYSTEM flag indicates the pages aren't in VRAM. */ WARN_ON_ONCE(flags & AMDGPU_PTE_SYSTEM); @@ -402,9 +403,12 @@ void amdgpu_gart_map_vram_range(struct amdgpu_device *adev, uint64_t pa, if (!drm_dev_enter(adev_to_drm(adev), &idx)) return; - for (i = 0; i < num_pages; ++i) { - amdgpu_gmc_set_pte_pde(adev, dst, - start_page + i, pa + AMDGPU_GPU_PAGE_SIZE * i, flags); + page_base = pa; + for (i = 0, t = 0; i < num_pages; i++) { + for (j = 0; j < AMDGPU_GPU_PAGES_IN_CPU_PAGE; j++, t++) { + amdgpu_gmc_set_pte_pde(adev, dst, start_page + t, page_base, flags); + page_base += AMDGPU_GPU_PAGE_SIZE; + } } drm_dev_exit(idx); From 03b70e0d8aa26bab89a0f1394c1c80a871925e42 Mon Sep 17 00:00:00 2001 From: Priya Hosur Date: Thu, 7 May 2026 13:31:37 +0530 Subject: [PATCH 222/329] drm/amd/pm: smu_v14_0_0: use SoftMin for gfxclk in set_soft_freq_limited_range In smu_v14_0_0_set_soft_freq_limited_range(), the gfxclk floor is programmed via SetHardMinGfxClk together with SetSoftMaxGfxClk. Under power_dpm_force_performance_level=high this pins HardMin to peak gfxclk. In PMFW arbitration HardMin has higher priority than SoftMax, so the firmware thermal/PPT throttler cannot clamp gfxclk via SoftMax once HardMin is set to peak. Replace SetHardMinGfxClk with SetSoftMinGfxclk so the driver still requests peak performance but the firmware throttler retains the ability to clamp gfxclk under thermal/PPT pressure. SoftMax handling is unchanged and no other clock domains are affected. Signed-off-by: Priya Hosur Acked-by: Alex Deucher Signed-off-by: Alex Deucher (cherry picked from commit 3ea273267fd29cbf6d83ee72329f59eb5042605b) Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c index a28624d4847a..75719c47a41e 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu14/smu_v14_0_0_ppt.c @@ -1231,7 +1231,8 @@ static int smu_v14_0_0_set_soft_freq_limited_range(struct smu_context *smu, switch (clk_type) { case SMU_GFXCLK: case SMU_SCLK: - msg_set_min = SMU_MSG_SetHardMinGfxClk; + /* SoftMin lets PMFW throttle gfxclk; HardMin would override SoftMax. */ + msg_set_min = SMU_MSG_SetSoftMinGfxclk; msg_set_max = SMU_MSG_SetSoftMaxGfxClk; break; case SMU_FCLK: From 40a25d59e85b3c8709ac2424d44f65610467871e Mon Sep 17 00:00:00 2001 From: Davidlohr Bueso Date: Thu, 7 May 2026 04:29:13 -0700 Subject: [PATCH 223/329] locking/rtmutex: Skip remove_waiter() when waiter is not enqueued syzbot triggered the following splat in remove_waiter() via FUTEX_CMP_REQUEUE_PI: KASAN: null-ptr-deref in range [0x0000000000000a88-0x0000000000000a8f] class_raw_spinlock_constructor remove_waiter+0x159/0x1200 kernel/locking/rtmutex.c:1561 rt_mutex_start_proxy_lock+0x103/0x120 futex_requeue+0x10e4/0x20d0 __x64_sys_futex+0x34f/0x4d0 task_blocks_on_rt_mutex() does not arm the waiter upon deadlock detection, leaving waiter->task nil, where 3bfdc63936dd ("rtmutex: Use waiter::task instead of current in remove_waiter()") made this fatal. Furthermore, rt_mutex_start_proxy_lock() should not be calling into remove_waiter() upon a successfully grabbing the rtmutex. 1a1fb985f2e2 ("futex: Handle early deadlock return correctly"), moved the remove_waiter() out of __rt_mutex_start_proxy_lock() (where 'ret' was only ever 0 or < 0) into the wrapper. Tighten this check to account for try_to_take_rt_mutex(). Fixes: 3bfdc63936dd ("rtmutex: Use waiter::task instead of current in remove_waiter()") Reported-by: syzbot+78147abe6c524f183ee9@syzkaller.appspotmail.com Signed-off-by: Davidlohr Bueso Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Closes: https://lore.kernel.org/all/69f114ac.050a0220.ac8b.0003.GAE@google.com/ Link: https://patch.msgid.link/20260507112913.1019537-1-dave@stgolabs.net --- kernel/locking/rtmutex.c | 3 +++ kernel/locking/rtmutex_api.c | 2 +- 2 files changed, 4 insertions(+), 1 deletion(-) diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index 4f386ea6c792..daeeeef973e2 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -1558,6 +1558,9 @@ static void __sched remove_waiter(struct rt_mutex_base *lock, lockdep_assert_held(&lock->wait_lock); + if (!waiter_task) /* never enqueued */ + return; + scoped_guard(raw_spinlock, &waiter_task->pi_lock) { rt_mutex_dequeue(lock, waiter); waiter_task->pi_blocked_on = NULL; diff --git a/kernel/locking/rtmutex_api.c b/kernel/locking/rtmutex_api.c index 124219aea46e..514fce7a4e0a 100644 --- a/kernel/locking/rtmutex_api.c +++ b/kernel/locking/rtmutex_api.c @@ -365,7 +365,7 @@ int __sched rt_mutex_start_proxy_lock(struct rt_mutex_base *lock, raw_spin_lock_irq(&lock->wait_lock); ret = __rt_mutex_start_proxy_lock(lock, waiter, task, &wake_q); - if (unlikely(ret)) + if (unlikely(ret < 0)) remove_waiter(lock, waiter); preempt_disable(); raw_spin_unlock_irq(&lock->wait_lock); From 2e5c6f4fd4001562781e99bbfc7f1f0127187542 Mon Sep 17 00:00:00 2001 From: Yilin Zhu Date: Thu, 30 Apr 2026 13:21:34 +0800 Subject: [PATCH 224/329] ipc/shm: serialize orphan cleanup with shm_nattch updates shm_destroy_orphaned() walks the shm idr under shm_ids(ns).rwsem, but that does not serialize all fields tested by shm_may_destroy(). In particular, shm_nattch is updated while holding shm_perm.lock, and attach paths can do that without holding the rwsem. Do not decide that an orphaned segment is unused before taking the object lock. Move the shm_may_destroy() check under shm_perm.lock, matching the other destroy paths, and unlock the segment when it no longer qualifies for removal. Link: https://lore.kernel.org/9d97cc1031de2d0bace0edf3a668818aa2f4eca6.1777410234.git.zylzyl2333@gmail.com Fixes: 4c677e2eefdb ("shm: optimize locking and ipc_namespace getting") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Xin Liu Signed-off-by: Yilin Zhu Signed-off-by: Ren Wei Cc: Christian Brauner Cc: Jeongjun Park Cc: Kees Cook Cc: Liam Howlett Cc: Lorenzo Stoakes Cc: Serge Hallyn Cc: Vasiliy Kulikov Cc: Davidlohr Bueso Cc: Oleg Nesterov Cc: Serge Hallyn Cc: Signed-off-by: Andrew Morton --- ipc/shm.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/ipc/shm.c b/ipc/shm.c index a95dae447707..b3e8a58e177d 100644 --- a/ipc/shm.c +++ b/ipc/shm.c @@ -418,15 +418,17 @@ static int shm_try_destroy_orphaned(int id, void *p, void *data) * We want to destroy segments without users and with already * exit'ed originating process. * - * As shp->* are changed under rwsem, it's safe to skip shp locking. + * shm_nattch can be changed under shm_perm.lock without holding the + * rwsem, so take the object lock before checking shm_may_destroy(). */ if (!list_empty(&shp->shm_clist)) return 0; - if (shm_may_destroy(shp)) { - shm_lock_by_ptr(shp); + shm_lock_by_ptr(shp); + if (shm_may_destroy(shp)) shm_destroy(ns, shp); - } + else + shm_unlock(shp); return 0; } From 0ade3a8d83a8f5bc274e1ace776c1a26aa7a0308 Mon Sep 17 00:00:00 2001 From: Kaitao Cheng Date: Fri, 22 May 2026 21:14:34 +0800 Subject: [PATCH 225/329] mm/cma_sysfs: skip inactive CMA areas in sysfs cma_activate_area() can fail after a CMA area has already been added to cma_areas[]. In that case the area is left in the global array, but it does not reach the point where CMA_ACTIVATED is set. cma_sysfs_init() currently walks all cma_area_count entries and creates sysfs files for every area, including ones that failed activation. These areas are not usable CMA areas and should not be exposed to userspace as valid CMA regions. If such an inactive area is exposed, userspace sees a CMA directory whose read-only accounting files report zeros. total_pages and available_pages report zero because the failed activation path clears cma->count and cma->available_count, while the allocation and release counters also stay at zero because the area cannot service CMA allocations. This makes the failed area look like a valid but empty CMA region and can mislead tests, monitoring, and diagnostics. Skip CMA areas that did not reach CMA_ACTIVATED when creating the sysfs objects. Since inactive entries can now be skipped, make the error unwind tolerate entries that never had cma_kobj initialized. Link: https://lore.kernel.org/20260524140420.61864-1-kaitao.cheng@linux.dev Link: https://lore.kernel.org/20260522131434.78532-1-kaitao.cheng@linux.dev Fixes: 43ca106fa8ec ("mm: cma: support sysfs") Signed-off-by: Kaitao Cheng Reported-by: David Hildenbrand (Arm) Suggested-by: David Hildenbrand (Arm) Suggested-by: Muchun Song Reported-by: Muchun Song Closes: https://lore.kernel.org/linux-mm/55481a8b-dcfc-4bef-ba59-aa0b43dca88b@kernel.org/ Acked-by: Muchun Song Cc: David Hildenbrand Cc: Dmitry Osipenko Cc: Greg Kroah-Hartman Cc: John Hubbard Cc: Liam R. Howlett Cc: Lorenzo Stoakes Cc: Michal Hocko Cc: Mike Rapoport Cc: Minchan Kim Cc: Suren Baghdasaryan Cc: Vlastimil Babka Signed-off-by: Andrew Morton --- mm/cma_sysfs.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/mm/cma_sysfs.c b/mm/cma_sysfs.c index f52b696bc46d..d5bf792c6245 100644 --- a/mm/cma_sysfs.c +++ b/mm/cma_sysfs.c @@ -117,13 +117,16 @@ static int __init cma_sysfs_init(void) return -ENOMEM; for (i = 0; i < cma_area_count; i++) { + cma = &cma_areas[i]; + if (!test_bit(CMA_ACTIVATED, &cma->flags)) + continue; + cma_kobj = kzalloc_obj(*cma_kobj); if (!cma_kobj) { err = -ENOMEM; goto out; } - cma = &cma_areas[i]; cma->cma_kobj = cma_kobj; cma_kobj->cma = cma; err = kobject_init_and_add(&cma_kobj->kobj, &cma_ktype, @@ -138,7 +141,8 @@ static int __init cma_sysfs_init(void) out: while (--i >= 0) { cma = &cma_areas[i]; - kobject_put(&cma->cma_kobj->kobj); + if (cma->cma_kobj) + kobject_put(&cma->cma_kobj->kobj); } kobject_put(cma_kobj_root); From 5921629dd625a69546692a92e1ca7ae987f03964 Mon Sep 17 00:00:00 2001 From: Wang Yaxin Date: Wed, 27 May 2026 21:35:58 +0800 Subject: [PATCH 226/329] tools headers UAPI: sync linux/taskstats.h for procacct.c MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit After commit 9b93f7e32774 ("tools/getdelays: use the static UAPI headers from tools/include/uapi"), the Makefile was changed to use -I../include/uapi/ instead of -I../../usr/include to ensure tools always use the up-to-date UAPI headers. However, only linux/taskstats.h was added to tools/include/uapi/ in commit e5bbb35a07b3 ("tools headers UAPI: sync linux/taskstats.h"), but linux/acct.h was missing. This causes procacct.c to fail to compile with: procacct.c:234:37: error: 'AGROUP' undeclared (first use in this function) gcc -I../include/uapi/ getdelays.c -o getdelays gcc -I../include/uapi/ procacct.c -o procacct procacct.c: In function `print_procacct': procacct.c:234:37: error: `AGROUP' undeclared (first use in this function) did you mean `NOGROUP'? 234 | , t->version >= 12 ? (t->ac_flag & AGROUP ? 'P' : 'T') : '?' | ^~~~~~ | NOGROUP procacct.c:234:37: note: each undeclared ident because procacct.c uses the AGROUP macro defined in linux/acct.h. Add the missing linux/acct.h to complete the static UAPI header set. Link: https://lore.kernel.org/20260527213558929EhiHHy9EDTMjmg3uuDOMi@zte.com.cn Fixes: 9b93f7e32774 ("tools/getdelays: use the static UAPI headers from tools/include/uapi") Signed-off-by: Wang Yaxin Reviewed-by: Thomas Weißschuh Cc: Fan Yu Cc: Jonathan Corbet Cc: xu xin Cc: Yang Yang Signed-off-by: Andrew Morton --- tools/include/uapi/linux/acct.h | 128 ++++++++++++++++++++++++++++++++ 1 file changed, 128 insertions(+) create mode 100644 tools/include/uapi/linux/acct.h diff --git a/tools/include/uapi/linux/acct.h b/tools/include/uapi/linux/acct.h new file mode 100644 index 000000000000..1e2382ed4fd5 --- /dev/null +++ b/tools/include/uapi/linux/acct.h @@ -0,0 +1,128 @@ +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ +/* + * BSD Process Accounting for Linux - Definitions + * + * Author: Marco van Wieringen (mvw@planets.elm.net) + * + * This header file contains the definitions needed to implement + * BSD-style process accounting. The kernel accounting code and all + * user-level programs that try to do something useful with the + * process accounting log must include this file. + * + * Copyright (C) 1995 - 1997 Marco van Wieringen - ELM Consultancy B.V. + * + */ + +#ifndef _UAPI_LINUX_ACCT_H +#define _UAPI_LINUX_ACCT_H + +#include + +#include +#include + +/* + * comp_t is a 16-bit "floating" point number with a 3-bit base 8 + * exponent and a 13-bit fraction. + * comp2_t is 24-bit with 5-bit base 2 exponent and 20 bit fraction + * (leading 1 not stored). + * See linux/kernel/acct.c for the specific encoding systems used. + */ + +typedef __u16 comp_t; +typedef __u32 comp2_t; + +/* + * accounting file record + * + * This structure contains all of the information written out to the + * process accounting file whenever a process exits. + */ + +#define ACCT_COMM 16 + +struct acct +{ + char ac_flag; /* Flags */ + char ac_version; /* Always set to ACCT_VERSION */ + /* for binary compatibility back until 2.0 */ + __u16 ac_uid16; /* LSB of Real User ID */ + __u16 ac_gid16; /* LSB of Real Group ID */ + __u16 ac_tty; /* Control Terminal */ + /* __u32 range means times from 1970 to 2106 */ + __u32 ac_btime; /* Process Creation Time */ + comp_t ac_utime; /* User Time */ + comp_t ac_stime; /* System Time */ + comp_t ac_etime; /* Elapsed Time */ + comp_t ac_mem; /* Average Memory Usage */ + comp_t ac_io; /* Chars Transferred */ + comp_t ac_rw; /* Blocks Read or Written */ + comp_t ac_minflt; /* Minor Pagefaults */ + comp_t ac_majflt; /* Major Pagefaults */ + comp_t ac_swaps; /* Number of Swaps */ +/* m68k had no padding here. */ +#if !defined(CONFIG_M68K) || !defined(__KERNEL__) + __u16 ac_ahz; /* AHZ */ +#endif + __u32 ac_exitcode; /* Exitcode */ + char ac_comm[ACCT_COMM + 1]; /* Command Name */ + __u8 ac_etime_hi; /* Elapsed Time MSB */ + __u16 ac_etime_lo; /* Elapsed Time LSB */ + __u32 ac_uid; /* Real User ID */ + __u32 ac_gid; /* Real Group ID */ +}; + +struct acct_v3 +{ + char ac_flag; /* Flags */ + char ac_version; /* Always set to ACCT_VERSION */ + __u16 ac_tty; /* Control Terminal */ + __u32 ac_exitcode; /* Exitcode */ + __u32 ac_uid; /* Real User ID */ + __u32 ac_gid; /* Real Group ID */ + __u32 ac_pid; /* Process ID */ + __u32 ac_ppid; /* Parent Process ID */ + /* __u32 range means times from 1970 to 2106 */ + __u32 ac_btime; /* Process Creation Time */ +#ifdef __KERNEL__ + __u32 ac_etime; /* Elapsed Time */ +#else + float ac_etime; /* Elapsed Time */ +#endif + comp_t ac_utime; /* User Time */ + comp_t ac_stime; /* System Time */ + comp_t ac_mem; /* Average Memory Usage */ + comp_t ac_io; /* Chars Transferred */ + comp_t ac_rw; /* Blocks Read or Written */ + comp_t ac_minflt; /* Minor Pagefaults */ + comp_t ac_majflt; /* Major Pagefaults */ + comp_t ac_swaps; /* Number of Swaps */ + char ac_comm[ACCT_COMM]; /* Command Name */ +}; + +/* + * accounting flags + */ + /* bit set when the process/task ... */ +#define AFORK 0x01 /* ... executed fork, but did not exec */ +#define ASU 0x02 /* ... used super-user privileges */ +#define ACOMPAT 0x04 /* ... used compatibility mode (VAX only not used) */ +#define ACORE 0x08 /* ... dumped core */ +#define AXSIG 0x10 /* ... was killed by a signal */ +#define AGROUP 0x20 /* ... was the last task of the process (task group) */ + +#if defined(__BYTE_ORDER) ? __BYTE_ORDER == __BIG_ENDIAN : defined(__BIG_ENDIAN) +#define ACCT_BYTEORDER 0x80 /* accounting file is big endian */ +#elif defined(__BYTE_ORDER) ? __BYTE_ORDER == __LITTLE_ENDIAN : defined(__LITTLE_ENDIAN) +#define ACCT_BYTEORDER 0x00 /* accounting file is little endian */ +#else +#error unspecified endianness +#endif + +#ifndef __KERNEL__ +#define ACCT_VERSION 2 +#define AHZ (HZ) +#endif /* __KERNEL */ + + +#endif /* _UAPI_LINUX_ACCT_H */ From 8ad974459aef3349ec5cb3cc54e2bfc1b7aebf34 Mon Sep 17 00:00:00 2001 From: Baoquan He Date: Thu, 28 May 2026 21:14:54 +0800 Subject: [PATCH 227/329] MAINTAINERS: update Baoquan He's email address I will switch to use @linux.dev mailbox, update all entries in MAINTAINERS. And map the address in .mailmap. Link: https://lore.kernel.org/20260528131454.1996752-1-baoquan.he@linux.dev Signed-off-by: Baoquan He Signed-off-by: Andrew Morton --- .mailmap | 1 + MAINTAINERS | 6 +++--- 2 files changed, 4 insertions(+), 3 deletions(-) diff --git a/.mailmap b/.mailmap index a009f73d7ea5..0b9298a55d2d 100644 --- a/.mailmap +++ b/.mailmap @@ -126,6 +126,7 @@ Baolin Wang Baolin Wang Baolin Wang Baolin Wang +Baoquan He Barry Song <21cnbao@gmail.com> Barry Song Barry Song diff --git a/MAINTAINERS b/MAINTAINERS index 461a3eed6129..fcc46165038c 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -13864,7 +13864,7 @@ F: scripts/Makefile.kcsan KDUMP M: Andrew Morton -M: Baoquan He +M: Baoquan He M: Mike Rapoport M: Pasha Tatashin M: Pratyush Yadav @@ -14183,7 +14183,7 @@ F: include/linux/kernfs.h KEXEC M: Andrew Morton -M: Baoquan He +M: Baoquan He M: Mike Rapoport M: Pasha Tatashin M: Pratyush Yadav @@ -17036,7 +17036,7 @@ M: Chris Li M: Kairui Song R: Kemeng Shi R: Nhat Pham -R: Baoquan He +R: Baoquan He R: Barry Song R: Youngjun Park L: linux-mm@kvack.org From 732fd9f0b9c1cdc6dfd77162ded60df005182cc0 Mon Sep 17 00:00:00 2001 From: Cunlong Li Date: Thu, 28 May 2026 10:48:44 +0800 Subject: [PATCH 228/329] zram: fix use-after-free in zram_bvec_write_partial() zram_read_page() picks the sync or async backing device read path based on whether the parent bio is NULL. zram_bvec_write_partial() passes its parent bio down, so for ZRAM_WB slots the read is dispatched asynchronously and zram_read_page() returns 0 while the bio is still in flight. The caller then runs memcpy_from_bvec(), zram_write_page() and __free_page() on the buffer, leaving the async read to write into a freed page. zram_bvec_read_partial() was switched to NULL in commit 4e3c87b9421d ("zram: fix synchronous reads") for the same reason; the write_partial counterpart was missed. Link: https://lore.kernel.org/20260528-zram-v3-1-cab86eef8764@gmail.com Fixes: 8e654f8fbff5 ("zram: read page from backing device") Reviewed-by: Christoph Hellwig Reviewed-by: Sergey Senozhatsky Signed-off-by: Cunlong Li Cc: Jens Axboe Cc: Minchan Kim Cc: Yisheng Xie Cc: Signed-off-by: Andrew Morton --- drivers/block/zram/zram_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c index 07111455eecf..e11ee1ed3832 100644 --- a/drivers/block/zram/zram_drv.c +++ b/drivers/block/zram/zram_drv.c @@ -2337,7 +2337,7 @@ static int zram_bvec_write_partial(struct zram *zram, struct bio_vec *bvec, if (!page) return -ENOMEM; - ret = zram_read_page(zram, page, index, bio); + ret = zram_read_page(zram, page, index, NULL); if (!ret) { memcpy_from_bvec(page_address(page) + offset, bvec); ret = zram_write_page(zram, page, index); From 7e2ed8a29427af534bf2cb9b8bc51762b8b6e654 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Thu, 28 May 2026 17:01:02 -0700 Subject: [PATCH 229/329] mm/damon/reclaim: handle ctx allocation failure Patch series "mm/damon/{reclaim,lru_sort}: handle ctx allocation failures". DAMON_RECLAIM and DAMON_LRU_SORT could dereference NULL pointers if their damon_ctx object allocations fail. The bugs are expected to happen infrequently because the allocations are arguably too small to fail on common setups. But theoretically they are possible and the consequences are bad. Fix those. The issues were discovered [1] by Sashiko. This patch (of 2): DAMON_RECLAIM allocates the damon_ctx object for its kdamond in its init function. damon_reclaim_enabled_store() wrongly assumes the allocation will always succeed once tried. If the damon_ctx allocation was failed, therefore, code execution reaches to damon_commit_ctx() while 'ctx' is NULL. As a result, it dereferences the NULL 'ctx' pointer. Avoid the NULL dereference by returning -ENOMEM if 'ctx' is NULL. Link: https://lore.kernel.org/20260529000104.7006-2-sj@kernel.org Link: https://lore.kernel.org/20260419014800.877-1-sj@kernel.org [1] Fixes: 3f7a914ab9a5 ("mm/damon/reclaim: use damon_initialized()") Signed-off-by: SeongJae Park Cc: # 6.18.x Signed-off-by: Andrew Morton --- mm/damon/reclaim.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/mm/damon/reclaim.c b/mm/damon/reclaim.c index fe7fce26cf6c..96f6dfc28eae 100644 --- a/mm/damon/reclaim.c +++ b/mm/damon/reclaim.c @@ -339,6 +339,10 @@ static int damon_reclaim_enabled_store(const char *val, if (!damon_initialized()) return 0; + /* damon_modules_new_paddr_ctx_target() in the init function failed. */ + if (!ctx) + return -ENOMEM; + return damon_reclaim_turn(enabled); } From ab04340b5ae5d52c1d46b750538febcde9d889e7 Mon Sep 17 00:00:00 2001 From: SeongJae Park Date: Thu, 28 May 2026 17:01:03 -0700 Subject: [PATCH 230/329] mm/damon/lru_sort: handle ctx allocation failure DAMON_LRU_SORT allocates the damon_ctx object for its kdamond in its init function. damon_lru_sort_enabled_store() wrongly assumes the allocation will always succeed once tried. If the damon_ctx allocation was failed, therefore, code execution reaches to damon_commit_ctx() while 'ctx' is NULL. As a result, it dereferences the NULL 'ctx' pointer. Avoid the NULL dereference by returning -ENOMEM if 'ctx' is NULL. Link: https://lore.kernel.org/20260529000104.7006-3-sj@kernel.org Fixes: c4a8e662c839 ("mm/damon/lru_sort: use damon_initialized()") Signed-off-by: SeongJae Park Cc: # 6.18.x Signed-off-by: Andrew Morton --- mm/damon/lru_sort.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/mm/damon/lru_sort.c b/mm/damon/lru_sort.c index 8494040b1ee4..8cfe7bd3dc1d 100644 --- a/mm/damon/lru_sort.c +++ b/mm/damon/lru_sort.c @@ -437,6 +437,10 @@ static int damon_lru_sort_enabled_store(const char *val, if (!damon_initialized()) return 0; + /* damon_modules_new_paddr_ctx_target() in the init function failed. */ + if (!ctx) + return -ENOMEM; + return damon_lru_sort_turn(enabled); } From 43e7f189769c512c843184a8a5892ac779a6bd90 Mon Sep 17 00:00:00 2001 From: Lorenzo Stoakes Date: Mon, 1 Jun 2026 09:30:44 +0100 Subject: [PATCH 231/329] mm/huge_memory: use correct flags for device private PMD entry Commit 65edfda6f3f2 ("mm/rmap: extend rmap and migration support device-private entries") updated set_pmd_migration_entry() to use pmdp_huge_get_and_clear() in the softleaf case, but made no further adjustments to the function itself. Therefore this function continues to incorrectly use pmd_write(), pmd_soft_dirty() and pmd_uffd_wp() to determine whether the installed migration entry should be marked writable, softdirty or uffd-wp respectively. Whilst all are incorrect, the most problematic of these is pmd_write(), as this can lead to corrupted rmap state. On x86-64 _PAGE_SWP_SOFT_DIRTY is aliased to _PAGE_RW. So calling pmd_write() on a softleaf will return the softdirty state encoded in the entry, assuming CONFIG_MEM_SOFT_DIRTY was enabled. This was observed when running the hmm.hmm_device_private.anon_write_child selftest: 1. The test faults in a range then migrates it such that a device-private THP range is established. 2. The parent then migrates it to a device-private writable PMD entry whose folio is entirely AnonExclusive with entire_mapcount=1, softdirty set (accidentally correct write state). 3. The parent forks and the PMD entries are set to device-private read only entries, entire_mapcount=2, softdirty still set. 4. [BUG] The child writes to the range then migrates to RAM - intending to install non-writable migration entries - but replacing parent and child PMD mappings with WRITABLE entries due to misinterpreting the softdirty bit. 5. In remove_migration_pmd(), if !softleaf_is_migration_read(entry) we set the RMAP_EXCLUSIVE flag when calling folio_add_anon_rmap_pmd() for both parent and child, which are therefore AnonExclusive. 6. [SPLAT] Child sets migrated folio entire_mapcount=1, parent sets entire_mapcount=2 and we end up with an AnonExclusive folio with entire_mapcount=2! Assert fires in __folio_add_anon_rmap(): VM_WARN_ON_FOLIO(folio_test_large(folio) && folio_entire_mapcount(folio) > 1 && PageAnonExclusive(cur_page), folio) This patch fixes the issue by correctly referencing the softleaf entry fields for writable, softdirty and uffd-wp in set_pmd_migration_entry(). It also only updates A/D flags if the entry is present as these are otherwise not meaningful for a softleaf entry. This patch also flips the if (!present) { ... } else { ... } logic in set_pmd_migration_entry() so it is easier to understand, and adds some comments to make things clearer. I was able to bisect this to commit 775465fd26a3 ("lib/test_hmm: add zone device private THP test infrastructure") which first exposes this bug as it was the commit that permitted test_hmm to generate the test. However commit 65edfda6f3f2 ("mm/rmap: extend rmap and migration support device-private entries") is the commit that actually enabled this behaviour. Link: https://lore.kernel.org/20260601083044.57132-1-ljs@kernel.org Fixes: 65edfda6f3f2 ("mm/rmap: extend rmap and migration support device-private entries") Signed-off-by: Lorenzo Stoakes Acked-by: David Hildenbrand (Arm) Reviewed-by: Dev Jain Reviewed-by: Balbir Singh Reviewed-by: Baolin Wang Reviewed-by: Oscar Salvador (SUSE) Reviewed-by: Barry Song Reviewed-by: Lance Yang Reviewed-by: Zi Yan Cc: Baolin Wang Cc: Liam R. Howlett Cc: Nico Pache Cc: Ryan Roberts Cc: SeongJae Park Cc: Wei Yang Cc: Signed-off-by: Andrew Morton --- mm/huge_memory.c | 45 +++++++++++++++++++++++++++++++++------------ 1 file changed, 33 insertions(+), 12 deletions(-) diff --git a/mm/huge_memory.c b/mm/huge_memory.c index 653f2dc03403..b118bcd392cb 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -4983,7 +4983,7 @@ int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw, struct vm_area_struct *vma = pvmw->vma; struct mm_struct *mm = vma->vm_mm; unsigned long address = pvmw->address; - bool anon_exclusive; + bool anon_exclusive, present, writable, softdirty, uffd_wp; pmd_t pmdval; swp_entry_t entry; pmd_t pmdswp; @@ -4991,12 +4991,26 @@ int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw, if (!(pvmw->pmd && !pvmw->pte)) return 0; - flush_cache_range(vma, address, address + HPAGE_PMD_SIZE); - if (unlikely(!pmd_present(*pvmw->pmd))) - pmdval = pmdp_huge_get_and_clear(vma->vm_mm, address, pvmw->pmd); - else + present = pmd_present(*pvmw->pmd); + if (likely(present)) { + flush_cache_range(vma, address, address + HPAGE_PMD_SIZE); + pmdval = pmdp_invalidate(vma, address, pvmw->pmd); + writable = pmd_write(pmdval); + softdirty = pmd_soft_dirty(pmdval); + uffd_wp = pmd_uffd_wp(pmdval); + } else { + softleaf_t old_entry; + + pmdval = pmdp_huge_get_and_clear(vma->vm_mm, address, pvmw->pmd); + old_entry = softleaf_from_pmd(pmdval); + + writable = softleaf_is_device_private_write(old_entry); + softdirty = pmd_swp_soft_dirty(pmdval); + uffd_wp = pmd_swp_uffd_wp(pmdval); + } + /* See folio_try_share_anon_rmap_pmd(): invalidate PMD first. */ anon_exclusive = folio_test_anon(folio) && PageAnonExclusive(page); if (anon_exclusive && folio_try_share_anon_rmap_pmd(folio, page)) { @@ -5004,24 +5018,31 @@ int set_pmd_migration_entry(struct page_vma_mapped_walk *pvmw, return -EBUSY; } - if (pmd_dirty(pmdval)) - folio_mark_dirty(folio); - if (pmd_write(pmdval)) + /* Determine type of migration entry. */ + if (writable) entry = make_writable_migration_entry(page_to_pfn(page)); else if (anon_exclusive) entry = make_readable_exclusive_migration_entry(page_to_pfn(page)); else entry = make_readable_migration_entry(page_to_pfn(page)); - if (pmd_young(pmdval)) + + /* Set A/D bits as necessary. */ + if (present && pmd_young(pmdval)) entry = make_migration_entry_young(entry); - if (pmd_dirty(pmdval)) + if (present && pmd_dirty(pmdval)) { + folio_mark_dirty(folio); entry = make_migration_entry_dirty(entry); + } + + /* Set PMD. */ pmdswp = swp_entry_to_pmd(entry); - if (pmd_soft_dirty(pmdval)) + if (softdirty) pmdswp = pmd_swp_mksoft_dirty(pmdswp); - if (pmd_uffd_wp(pmdval)) + if (uffd_wp) pmdswp = pmd_swp_mkuffd_wp(pmdswp); set_pmd_at(mm, address, pvmw->pmd, pmdswp); + + /* Migration entry installed: cleanup rmap, folio. */ folio_remove_rmap_pmd(folio, page, vma); folio_put(folio); trace_set_migration_pmd(address, pmd_val(pmdswp)); From 98733f3f0becb1ae0701d021c1748e974e5fa55c Mon Sep 17 00:00:00 2001 From: Shakeel Butt Date: Mon, 1 Jun 2026 09:15:01 -0700 Subject: [PATCH 232/329] mm/list_lru: drain before clearing xarray entry on reparent memcg_reparent_list_lrus() clears the dying memcg's xarray entry with xas_store(&xas, NULL) before reparenting its per-node lists into the parent. This opens a window where a concurrent list_lru_del() arriving for the dying memcg sees xa_load() == NULL, walks to the parent in lock_list_lru_of_memcg(), takes the parent's per-node lock, and calls list_del_init() on an item still physically linked on the dying memcg's list. If another in-flight thread holds the dying memcg's per-node lock at the same moment (another list_lru_del, or a list_lru_walk_one running an isolate callback), both threads modify ->next/->prev pointers on the same physical list under different locks. Adjacent items can corrupt each other's links. Fix it by reversing the order: reparent each per-node list and mark the child's list lru dead and then clear the xarray entry. Any concurrent list_lru op that finds the still-set xarray entry either takes the dying memcg's per-node lock (synchronizing with the drain) or sees LONG_MIN and walks to the parent, where the items now live. Link: https://lore.kernel.org/20260601161501.1444829-1-shakeel.butt@linux.dev Fixes: fb56fdf8b9a2 ("mm/list_lru: split the lock to per-cgroup scope") Signed-off-by: Shakeel Butt Reported-by: Chris Mason Reviewed-by: Kairui Song Acked-by: Muchun Song Cc: Dave Chinner Cc: Johannes Weiner Cc: Roman Gushchin Cc: Signed-off-by: Andrew Morton --- mm/list_lru.c | 21 ++++++++++++--------- 1 file changed, 12 insertions(+), 9 deletions(-) diff --git a/mm/list_lru.c b/mm/list_lru.c index dd29bcf8eb5f..9bf7f524796b 100644 --- a/mm/list_lru.c +++ b/mm/list_lru.c @@ -473,26 +473,29 @@ void memcg_reparent_list_lrus(struct mem_cgroup *memcg, struct mem_cgroup *paren mutex_lock(&list_lrus_mutex); list_for_each_entry(lru, &memcg_list_lrus, list) { struct list_lru_memcg *mlru; - XA_STATE(xas, &lru->xa, memcg->kmemcg_id); /* - * Lock the Xarray to ensure no on going list_lru_memcg - * allocation and further allocation will see css_is_dying(). + * css_is_dying() check in memcg_list_lru_alloc() avoids + * allocating a new mlru since CSS_DYING is already set for this + * memcg a rcu grace period ago. */ - xas_lock_irq(&xas); - mlru = xas_store(&xas, NULL); - xas_unlock_irq(&xas); + mlru = xa_load(&lru->xa, memcg->kmemcg_id); if (!mlru) continue; /* - * With Xarray value set to NULL, holding the lru lock below - * prevents list_lru_{add,del,isolate} from touching the lru, - * safe to reparent. + * Reparent each per-node list and mark the child dead + * (LONG_MIN) before clearing xarray entry otherwise a + * concurrent list_lru_del() may corrupt the list if it arrives + * after xarray clear but before reparenting as + * lock_list_lru_of_memcg will acquire parent's lock while the + * item is still on child's list. */ for_each_node(i) memcg_reparent_list_lru_one(lru, i, &mlru->node[i], parent); + xa_erase_irq(&lru->xa, memcg->kmemcg_id); + /* * Here all list_lrus corresponding to the cgroup are guaranteed * to remain empty, we can safely free this lru, any further From c594b83457ccdee76d458416fb3bc9348a37592f Mon Sep 17 00:00:00 2001 From: Alistair Popple Date: Thu, 21 May 2026 13:27:30 +1000 Subject: [PATCH 233/329] arm64: mm: call pagetable dtor when freeing hot-removed page tables Since 5e8eb9aeeda3 ("arm64: mm: always call PTE/PMD ctor in __create_pgd_mapping()") page-table allocation on ARM64 always calls pagetable_{pte,pmd,pud,p4d}_ctor(). This sets the page_type to PGTY_table, increments NR_PAGETABLE and possible allocates a PTL. However the matching pagetable_dtor() calls were never added. With DEBUG_VM enabled on kernel versions prior to v6.17 without 2dfcd1608f3a9 ("mm/page_alloc: let page freeing clear any set page type") this leads to the following warning when freeing these pages due to page->page_type sharing page->_mapcount: BUG: Bad page state in process ... pfn:284fbb page: refcount:0 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x284fbb flags: 0x17fffc000000000(node=0|zone=2|lastcpupid=0x1ffff) page_type: f2(table) page dumped because: nonzero mapcount Call trace: bad_page+0x13c/0x160 __free_frozen_pages+0x6cc/0x860 ___free_pages+0xf4/0x180 free_pages+0x54/0x80 free_hotplug_page_range.part.0+0x58/0x90 free_empty_tables+0x438/0x500 __remove_pgd_mapping.constprop.0+0x60/0xa8 arch_remove_memory+0x48/0x80 try_remove_memory+0x158/0x1d8 offline_and_remove_memory+0x138/0x180 It can also lead to leaking the ptl allocation if ALLOC_SPLIT_PTLOCKS is defined and incorrect NR_PAGETABLE stats. Fix this by calling pagetable_dtor() in free_hotplug_pgtable_page() prior to freeing the page to undo the effects of calling pagetable_*_ctor(). Link: https://lore.kernel.org/20260521032730.2104017-1-apopple@nvidia.com Fixes: 5e8eb9aeeda3 ("arm64: mm: always call PTE/PMD ctor in __create_pgd_mapping()") Signed-off-by: Alistair Popple Cc: Catalin Marinas Cc: David Hildenbrand Cc: Will Deacon Cc: Signed-off-by: Andrew Morton --- arch/arm64/mm/mmu.c | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c index dd85e093ffdb..8242f93f05e4 100644 --- a/arch/arm64/mm/mmu.c +++ b/arch/arm64/mm/mmu.c @@ -1441,6 +1441,7 @@ static void free_hotplug_page_range(struct page *page, size_t size, static void free_hotplug_pgtable_page(struct page *page) { + pagetable_dtor(page_ptdesc(page)); free_hotplug_page_range(page, PAGE_SIZE, NULL); } From 0c25b8734367574e21aeb8468c2e522713134da7 Mon Sep 17 00:00:00 2001 From: Usama Arif Date: Tue, 2 Jun 2026 10:22:47 -0700 Subject: [PATCH 234/329] mm/mincore: handle non-swap entries before !CONFIG_SWAP guard mincore_swap() also fields migration/hwpoison entries (and shmem swapin-error entries), which can exist on !CONFIG_SWAP builds when CONFIG_MIGRATION or CONFIG_MEMORY_FAILURE is enabled. The !IS_ENABLED(CONFIG_SWAP) guard ran before the non-swap-entry early return, so mincore_pte_range() can spuriously WARN and report these pages nonresident on !CONFIG_SWAP kernels. Move the guard below the non-swap-entry check so only true swap entries trip the WARN, and migration/hwpoison entries take the existing "uptodate / non-shmem" path. Link: https://lore.kernel.org/20260602172247.279421-1-usama.arif@linux.dev Fixes: 1f2052755c15 ("mm/mincore: use a helper for checking the swap cache") Signed-off-by: Usama Arif Reviewed-by: Pedro Falcato Reviewed-by: Kairui Song Reviewed-by: Lorenzo Stoakes Acked-by: Johannes Weiner Cc: Baoquan He Cc: Chris Li Cc: Jann Horn Cc: Liam R. Howlett Cc: Rik van Riel Cc: Shakeel Butt Cc: Vlastimil Babka Cc: Signed-off-by: Andrew Morton --- mm/mincore.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/mm/mincore.c b/mm/mincore.c index e5d13eea9234..296f2e3922b5 100644 --- a/mm/mincore.c +++ b/mm/mincore.c @@ -64,11 +64,6 @@ static unsigned char mincore_swap(swp_entry_t entry, bool shmem) struct folio *folio = NULL; unsigned char present = 0; - if (!IS_ENABLED(CONFIG_SWAP)) { - WARN_ON(1); - return 0; - } - /* * Shmem mapping may contain swapin error entries, which are * absent. Page table may contain migration or hwpoison @@ -77,6 +72,11 @@ static unsigned char mincore_swap(swp_entry_t entry, bool shmem) if (!softleaf_is_swap(entry)) return !shmem; + if (!IS_ENABLED(CONFIG_SWAP)) { + WARN_ON(1); + return 0; + } + /* * Shmem mapping lookup is lockless, so we need to grab the swap * device. mincore page table walk locks the PTL, and the swap From 22ba97ea9cc1f63a0d0244fae38057ed452b6ac7 Mon Sep 17 00:00:00 2001 From: Jason Xing Date: Sat, 30 May 2026 12:26:30 +0800 Subject: [PATCH 235/329] xsk: cache csum_start/csum_offset to fix TOCTOU in xsk_skb_metadata() The TX metadata area resides in the UMEM buffer which is memory-mapped and concurrently writable by userspace. In xsk_skb_metadata(), csum_start and csum_offset are read from shared memory for bounds validation, then read again for skb assignment. A malicious userspace application can race to overwrite these values between the two reads, bypassing the bounds check and causing out-of-bounds memory access during checksum computation in the transmit path. Fix this by reading csum_start and csum_offset into local variables once, then using the local copies for both validation and assignment. Note that other metadata fields (flags, launch_time) and the cached csum fields may be mutually inconsistent due to concurrent userspace writes, but this is benign: the only security-critical invariant is that each field's validated value is the same one used, which local caching guarantees. Closes: https://lore.kernel.org/all/20260503200927.73EA1C2BCB4@smtp.kernel.org/ Reviewed-by: Maciej Fijalkowski Signed-off-by: Jason Xing Acked-by: Stanislav Fomichev Fixes: 48eb03dd2630 ("xsk: Add TX timestamp and TX checksum offload support") Link: https://patch.msgid.link/20260530042630.80626-1-kerneljasonxing@gmail.com Signed-off-by: Jakub Kicinski --- net/xdp/xsk.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c index 5e5786cd9af5..f8c8a8c9dfba 100644 --- a/net/xdp/xsk.c +++ b/net/xdp/xsk.c @@ -802,6 +802,7 @@ static int xsk_skb_metadata(struct sk_buff *skb, void *buffer, u32 hr) { struct xsk_tx_metadata *meta = NULL; + u16 csum_start, csum_offset; if (unlikely(pool->tx_metadata_len == 0)) return -EINVAL; @@ -811,13 +812,15 @@ static int xsk_skb_metadata(struct sk_buff *skb, void *buffer, return -EINVAL; if (meta->flags & XDP_TXMD_FLAGS_CHECKSUM) { - if (unlikely(meta->request.csum_start + - meta->request.csum_offset + + csum_start = READ_ONCE(meta->request.csum_start); + csum_offset = READ_ONCE(meta->request.csum_offset); + + if (unlikely(csum_start + csum_offset + sizeof(__sum16) > desc->len)) return -EINVAL; - skb->csum_start = hr + meta->request.csum_start; - skb->csum_offset = meta->request.csum_offset; + skb->csum_start = hr + csum_start; + skb->csum_offset = csum_offset; skb->ip_summed = CHECKSUM_PARTIAL; if (unlikely(pool->tx_sw_csum)) { From 56d0885514491e5ed8f7593400879ab77c52504c Mon Sep 17 00:00:00 2001 From: Jonas Jelonek Date: Thu, 28 May 2026 20:52:40 +0000 Subject: [PATCH 236/329] net: sfp: initialize i2c_block_size at adapter configure time sfp->i2c_block_size is only assigned in sfp_sm_mod_probe(), which runs from the state machine timer after SFP_F_PRESENT has been set. Between those two points, sfp_module_eeprom() (the ethtool -m callback) gates only on SFP_F_PRESENT and can be entered with i2c_block_size still at its kzalloc'd value of 0. On a pure-I2C adapter, sfp_i2c_read() then issues an i2c_transfer() with msgs[1].len = 0 inside a loop that subtracts this_len from len each iteration; on adapters that succeed a zero-length read the loop never advances, spinning while holding rtnl_lock. This was previously addressed by initializing i2c_block_size in sfp_alloc() (commit 813c2dd78618), but the initialization was dropped when i2c_block_size was split from i2c_max_block_size. Initialize sfp->i2c_block_size from sfp->i2c_max_block_size in sfp_i2c_configure(), so the field is valid as soon as the adapter is known. sfp_sm_mod_probe() still reassigns it on each module insertion to recover from a per-module clamp to 1 (sfp_id_needs_byte_io). Fixes: 7662abf4db94 ("net: phy: sfp: Add support for SMBus module access") Cc: stable@vger.kernel.org Signed-off-by: Jonas Jelonek Link: https://patch.msgid.link/20260528205242.971410-2-jelonek.jonas@gmail.com Signed-off-by: Jakub Kicinski --- drivers/net/phy/sfp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/phy/sfp.c b/drivers/net/phy/sfp.c index bd970f753beb..b94b9c433a21 100644 --- a/drivers/net/phy/sfp.c +++ b/drivers/net/phy/sfp.c @@ -822,6 +822,7 @@ static int sfp_i2c_configure(struct sfp *sfp, struct i2c_adapter *i2c) return -EINVAL; } + sfp->i2c_block_size = sfp->i2c_max_block_size; return 0; } From 2cdeaba5a1087f0f83e56729ea5c730b498639d9 Mon Sep 17 00:00:00 2001 From: Yizhou Zhao Date: Fri, 29 May 2026 18:50:16 +0800 Subject: [PATCH 237/329] appletalk: aarp: zero-initialize aarp_entry to prevent heap info leak aarp_alloc() allocates struct aarp_entry without zeroing it, but only initializes refcnt and packet_queue. When an unresolved AARP entry is created, hwaddr[ETH_ALEN] is left uninitialized. aarp_seq_show() later prints this field with %pM when users read /proc/net/atalk/arp. This can expose 6 bytes of stale heap data for each unresolved entry. Fix this by zero-initializing struct aarp_entry at allocation time. Reported-by: Yizhou Zhao Reported-by: Yuxiang Yang Reported-by: Ao Wang Reported-by: Xuewei Feng Reported-by: Qi Li Reported-by: Ke Xu Signed-off-by: Yizhou Zhao Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260529105017.81531-1-zhaoyz24@mails.tsinghua.edu.cn Signed-off-by: Jakub Kicinski --- net/appletalk/aarp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/appletalk/aarp.c b/net/appletalk/aarp.c index 30493ea3c010..078fb7a6efa5 100644 --- a/net/appletalk/aarp.c +++ b/net/appletalk/aarp.c @@ -393,7 +393,7 @@ static void aarp_purge(void) */ static struct aarp_entry *aarp_alloc(void) { - struct aarp_entry *a = kmalloc_obj(*a, GFP_ATOMIC); + struct aarp_entry *a = kzalloc_obj(*a, GFP_ATOMIC); if (!a) return NULL; From a910fb8f7b9e4c566db363e6c2ec378dc7153995 Mon Sep 17 00:00:00 2001 From: Geetha sowjanya Date: Fri, 29 May 2026 17:07:57 +0530 Subject: [PATCH 238/329] octeontx2-pf: Fix NDC sync operation errors On system reboot "rvu_nicpf 0002:03:00.0: NDC sync operation failed" error messages are shown, even if the operations is successful. This is due to wrong if error check in ndc_syc() function. Fixes: 42c45ac1419c ("octeontx2-af: Sync NIX and NPA contexts from NDC to LLC/DRAM") Signed-off-by: Geetha sowjanya Signed-off-by: Subbaraya Sundeep Reviewed-by: Simon Horman Link: https://patch.msgid.link/1780054677-17249-1-git-send-email-sbhatta@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c index ee623476e5ff..f9fbf0c17648 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c @@ -3473,7 +3473,7 @@ static void otx2_ndc_sync(struct otx2_nic *pf) req->nix_lf_rx_sync = 1; req->npa_lf_sync = 1; - if (!otx2_sync_mbox_msg(mbox)) + if (otx2_sync_mbox_msg(mbox)) dev_err(pf->dev, "NDC sync operation failed\n"); mutex_unlock(&mbox->lock); From 9a85ec3dc28b6df246801c19e4d9bae6297a25b0 Mon Sep 17 00:00:00 2001 From: Suman Ghosh Date: Fri, 29 May 2026 17:07:05 +0530 Subject: [PATCH 239/329] octeontx2-af: Fix initialization of mcam's entry2target_pffunc field NPC mcam entry stores a mapping between mcam entry and target pcifunc. During initialization of this field, API kmalloc_array has been used which caused some junk values to array. Whereas, the array is expected to be initialized by 0. This patch fixes the same by using kcalloc instead of kmalloc_array. Fixes: 55307fcb9258 ("octeontx2-af: Add mbox messages to install and delete MCAM rules") Signed-off-by: Suman Ghosh Signed-off-by: Subbaraya Sundeep Reviewed-by: Simon Horman Link: https://patch.msgid.link/1780054625-17090-1-git-send-email-sbhatta@marvell.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c index 607d0cf1a778..6bbda0593fcd 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c @@ -2192,8 +2192,8 @@ int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr) goto free_entry_cntr_map; /* Alloc memory for saving target device of mcam rule */ - mcam->entry2target_pffunc = kmalloc_array(mcam->total_entries, - sizeof(u16), GFP_KERNEL); + mcam->entry2target_pffunc = kcalloc(mcam->total_entries, + sizeof(u16), GFP_KERNEL); if (!mcam->entry2target_pffunc) goto free_cntr_refcnt; From c1f07a7f2d47aeb9878301e7bb36bc1c2bc2be8e Mon Sep 17 00:00:00 2001 From: Jianyu Li Date: Mon, 1 Jun 2026 19:36:39 +0800 Subject: [PATCH 240/329] af_unix: Fix inq_len update problem in partial read Currently inq_len is updated only when the whole skb is consumed. If only part of the data is read, following SIOCINQ query would get value greater than what actually left. This change update inq_len timely in unix_stream_read_generic(), and adjust unix_stream_read_skb() accordingly to prevent repetitive update. Fixes: f4e1fb04c123 ("af_unix: Use cached value for SOCK_STREAM in unix_inq_len().") Signed-off-by: Jianyu Li Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260601113640.231897-2-jianyu.li@mediatek.com Signed-off-by: Jakub Kicinski --- net/unix/af_unix.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index dc71ed79be4a..0d9cd977c7b7 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -2886,7 +2886,7 @@ static int unix_stream_read_skb(struct sock *sk, skb_read_actor_t recv_actor) return -EAGAIN; } - WRITE_ONCE(u->inq_len, u->inq_len - skb->len); + WRITE_ONCE(u->inq_len, u->inq_len - unix_skb_len(skb)); #if IS_ENABLED(CONFIG_AF_UNIX_OOB) if (skb == u->oob_skb) { @@ -3063,11 +3063,12 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state, unix_detach_fds(&scm, skb); } - if (unix_skb_len(skb)) - break; - spin_lock(&sk->sk_receive_queue.lock); - WRITE_ONCE(u->inq_len, u->inq_len - skb->len); + WRITE_ONCE(u->inq_len, u->inq_len - chunk); + if (unix_skb_len(skb)) { + spin_unlock(&sk->sk_receive_queue.lock); + break; + } __skb_unlink(skb, &sk->sk_receive_queue); spin_unlock(&sk->sk_receive_queue.lock); From dd8975ad710ea1f3d7c7a36295072fd5ee59ca0a Mon Sep 17 00:00:00 2001 From: Jianyu Li Date: Mon, 1 Jun 2026 19:36:40 +0800 Subject: [PATCH 241/329] af_unix: Add test for SCM_INQ on partial read Add test to verify that when a skb is partially consumed, unix_inq_len() return correct remaining byte count. Before: # RUN scm_inq.stream.partial_read ... # scm_inq.c:165:partial_read:Expected remain (512) == *(int *)CMSG_DATA(cmsg) (768) # partial_read: Test terminated by assertion # FAIL scm_inq.stream.partial_read not ok 2 scm_inq.stream.partial_read After: # RUN scm_inq.stream.partial_read ... # OK scm_inq.stream.partial_read ok 2 scm_inq.stream.partial_read Signed-off-by: Jianyu Li Reviewed-by: Kuniyuki Iwashima Link: https://patch.msgid.link/20260601113640.231897-3-jianyu.li@mediatek.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/af_unix/scm_inq.c | 54 ++++++++++++++++++- 1 file changed, 52 insertions(+), 2 deletions(-) diff --git a/tools/testing/selftests/net/af_unix/scm_inq.c b/tools/testing/selftests/net/af_unix/scm_inq.c index 3a86be9bda17..6268b5bf50be 100644 --- a/tools/testing/selftests/net/af_unix/scm_inq.c +++ b/tools/testing/selftests/net/af_unix/scm_inq.c @@ -8,8 +8,9 @@ #include "kselftest_harness.h" -#define NR_CHUNKS 100 -#define MSG_LEN 256 +#define NR_CHUNKS 100 +#define MSG_LEN 256 +#define NR_PARTIAL_READS 3 FIXTURE(scm_inq) { @@ -120,4 +121,53 @@ TEST_F(scm_inq, basic) recv_chunks(_metadata, self); } +TEST_F(scm_inq, partial_read) +{ + char buf[MSG_LEN * NR_PARTIAL_READS] = {}; + char cmsg_buf[CMSG_SPACE(sizeof(int))]; + struct msghdr msg = {}; + struct iovec iov = {}; + struct cmsghdr *cmsg; + int err, inq, ret, i; + int remain; + + err = setsockopt(self->fd[1], SOL_SOCKET, SO_INQ, &(int){1}, sizeof(int)); + if (variant->type != SOCK_STREAM) { + ASSERT_EQ(-ENOPROTOOPT, -errno); + return; + } + ASSERT_EQ(0, err); + + ret = send(self->fd[0], buf, sizeof(buf), 0); + ASSERT_EQ(sizeof(buf), ret); + + msg.msg_iov = &iov; + msg.msg_iovlen = 1; + msg.msg_control = cmsg_buf; + msg.msg_controllen = sizeof(cmsg_buf); + + iov.iov_base = buf; + iov.iov_len = MSG_LEN; + + for (i = 0; i < NR_PARTIAL_READS; i++) { + remain = MSG_LEN * (NR_PARTIAL_READS - 1 - i); + + memset(buf, 0, MSG_LEN); + memset(cmsg_buf, 0, sizeof(cmsg_buf)); + ret = recvmsg(self->fd[1], &msg, 0); + ASSERT_EQ(MSG_LEN, ret); + + cmsg = CMSG_FIRSTHDR(&msg); + ASSERT_NE(NULL, cmsg); + ASSERT_EQ(CMSG_LEN(sizeof(int)), cmsg->cmsg_len); + ASSERT_EQ(SOL_SOCKET, cmsg->cmsg_level); + ASSERT_EQ(SCM_INQ, cmsg->cmsg_type); + ASSERT_EQ(remain, *(int *)CMSG_DATA(cmsg)); + + ret = ioctl(self->fd[1], SIOCINQ, &inq); + ASSERT_EQ(0, ret); + ASSERT_EQ(remain, inq); + } +} + TEST_HARNESS_MAIN From d3915a1f5a4bc0ac911032903c3c6ab8df9fcc7c Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 2 Jun 2026 16:15:47 +0000 Subject: [PATCH 242/329] ipv4: restrict IPOPT_SSRR and IPOPT_LSRR options This patch restricts setting Loose Source and Record Route (LSRR) and Strict Source and Record Route (SSRR) IP options to users with CAP_NET_RAW capability. This prevents unprivileged applications from forcing packets to route through attacker-controlled nodes to leak TCP ISN and possibly other protocol information. While LSRR and SSRR are commonly filtered in many network environments, they may still be supported and forwarded along some network paths. RFC 7126 (Recommendations on Filtering of IPv4 Packets Containing IPv4 Options) recommend to drop these options in 4.3 and 4.4. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Tamir Shahar Reported-by: Amit Klein Signed-off-by: Eric Dumazet Reviewed-by: David Ahern Reviewed-by: Ido Schimmel Link: https://patch.msgid.link/20260602161547.2642155-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ipv4/ip_options.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/ipv4/ip_options.c b/net/ipv4/ip_options.c index be8815ce3ac2..09d745112c15 100644 --- a/net/ipv4/ip_options.c +++ b/net/ipv4/ip_options.c @@ -530,6 +530,10 @@ int ip_options_get(struct net *net, struct ip_options_rcu **optp, kfree(opt); return -EINVAL; } + if (opt->opt.srr && !ns_capable(net->user_ns, CAP_NET_RAW)) { + kfree(opt); + return -EPERM; + } kfree(*optp); *optp = opt; return 0; From 672bd0519e27c357c43b7f8c0d653fce3817d06e Mon Sep 17 00:00:00 2001 From: Kurt Kanzenbach Date: Fri, 29 May 2026 19:11:47 +0200 Subject: [PATCH 243/329] ptp: vclock: Switch from RCU to SRCU The usage of PTP vClocks leads immediately to the following issues with ptp4l with LOCKDEP and DEBUG_ATOMIC_SLEEP enabled: "BUG: sleeping function called from invalid context". ptp_convert_timestamp() acquires a mutex_t within a RCU read section. This is illegal, because acquiring a mutex_t can result in voluntary scheduling request which is not allowed within a RCU read section. Replace the RCU usage with SRCU where sleeping is allowed. Reported-by: Florian Zeitz Closes: https://lore.kernel.org/all/00a8cce8-410e-4038-98af-49be6d93d7bd@schettke.com/ Fixes: 67d93ffc0f3c ("ptp: vclock: use mutex to fix "sleep on atomic" bug") Signed-off-by: Kurt Kanzenbach Reviewed-by: Sebastian Andrzej Siewior Link: https://patch.msgid.link/20260529-vclock_rcu-v2-1-02a5531fab92@linutronix.de Signed-off-by: Jakub Kicinski --- drivers/ptp/ptp_vclock.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/ptp/ptp_vclock.c b/drivers/ptp/ptp_vclock.c index 915a4f6defc9..84cb527f59cc 100644 --- a/drivers/ptp/ptp_vclock.c +++ b/drivers/ptp/ptp_vclock.c @@ -19,6 +19,8 @@ static DEFINE_SPINLOCK(vclock_hash_lock); static DEFINE_READ_MOSTLY_HASHTABLE(vclock_hash, 8); +DEFINE_STATIC_SRCU(vclock_srcu); + static void ptp_vclock_hash_add(struct ptp_vclock *vclock) { spin_lock(&vclock_hash_lock); @@ -37,7 +39,7 @@ static void ptp_vclock_hash_del(struct ptp_vclock *vclock) spin_unlock(&vclock_hash_lock); - synchronize_rcu(); + synchronize_srcu(&vclock_srcu); } static int ptp_vclock_adjfine(struct ptp_clock_info *ptp, long scaled_ppm) @@ -276,14 +278,16 @@ ktime_t ptp_convert_timestamp(const ktime_t *hwtstamp, int vclock_index) { unsigned int hash = vclock_index % HASH_SIZE(vclock_hash); struct ptp_vclock *vclock; - u64 ns; u64 vclock_ns = 0; + int srcu_idx; + u64 ns; ns = ktime_to_ns(*hwtstamp); - rcu_read_lock(); + srcu_idx = srcu_read_lock(&vclock_srcu); - hlist_for_each_entry_rcu(vclock, &vclock_hash[hash], vclock_hash_node) { + hlist_for_each_entry_srcu(vclock, &vclock_hash[hash], vclock_hash_node, + srcu_read_lock_held(&vclock_srcu)) { if (vclock->clock->index != vclock_index) continue; @@ -294,7 +298,7 @@ ktime_t ptp_convert_timestamp(const ktime_t *hwtstamp, int vclock_index) break; } - rcu_read_unlock(); + srcu_read_unlock(&vclock_srcu, srcu_idx); return ns_to_ktime(vclock_ns); } From 9d8d28738f24b75616d6ca7a27cb4aed88520343 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 2 Jun 2026 22:14:08 +1000 Subject: [PATCH 244/329] mptcp: fix missing wakeups in edge scenarios The mptcp_recvmsg() can fill MPTCP socket receive queue via mptcp_move_skbs(), but currently does not try to wakeup any listener, because the same process is going to check the receive queue soon. When multiple threads are reading from the same fd, the above can cause stall. Add the missing wakeup. Fixes: 6771bfd9ee24 ("mptcp: update mptcp ack sequence from work queue") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-1-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index a72a6ad6ee8b..5a20ab2789ae 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2276,6 +2276,10 @@ static bool mptcp_move_skbs(struct sock *sk) mptcp_backlog_spooled(sk, moved, &skbs); } mptcp_data_unlock(sk); + + if (enqueued && mptcp_epollin_ready(sk)) + sk->sk_data_ready(sk); + return enqueued; } From d1918b36edcaed0ec4ef6888b2358c6b1ddcff47 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 2 Jun 2026 22:14:09 +1000 Subject: [PATCH 245/329] mptcp: fix retransmission loop when csum is enabled Sashiko noted that retransmission with csum enabled can actually transmit new data, but currently the relevant code does not update accordingly snd_nxt. The may cause incoming ack drop and an endless retransmission loop. Address the issue incrementing snd_nxt as needed. Fixes: 4e14867d5e91 ("mptcp: tune re-injections for csum enabled mode") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-2-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 5a20ab2789ae..7fac5fac2097 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -2869,6 +2869,10 @@ static void __mptcp_retrans(struct sock *sk) msk->bytes_retrans += len; dfrag->already_sent = max(dfrag->already_sent, len); + /* With csum enabled retransmission can send new data. */ + if (after64(dfrag->already_sent + dfrag->data_seq, msk->snd_nxt)) + WRITE_ONCE(msk->snd_nxt, dfrag->already_sent + dfrag->data_seq); + reset_timer: mptcp_check_and_set_pending(sk); From 8ab24fdebc369c0dfb90f82c1650b1e66662bb45 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 2 Jun 2026 22:14:10 +1000 Subject: [PATCH 246/329] mptcp: close TOCTOU race while computing rcv_wnd The MPTCP output path access locklessly the MPTCP-level ack_seq in multiple times, using possibly different values for the data_ack in the DSS option and to compute the announced rcv wnd for the same packet. Refactor the cote to avoid inconsistencies which may confuse the peer. Also ensure that the MPTCP level rcv wnd is updated only when the egress packet actually contains a DSS ack. Fixes: fa3fe2b15031 ("mptcp: track window announced to peer") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-3-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/options.c | 36 ++++++++++++++++++------------------ 1 file changed, 18 insertions(+), 18 deletions(-) diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 8a1c5698983c..2d25f319f328 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -570,7 +570,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, struct mptcp_ext *mpext; unsigned int ack_size; bool ret = false; - u64 ack_seq; opts->csum_reqd = READ_ONCE(msk->csum_enabled); mpext = skb ? mptcp_get_ext(skb) : NULL; @@ -601,14 +600,11 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, return ret; } - ack_seq = READ_ONCE(msk->ack_seq); if (READ_ONCE(msk->use_64bit_ack)) { ack_size = TCPOLEN_MPTCP_DSS_ACK64; - opts->ext_copy.data_ack = ack_seq; opts->ext_copy.ack64 = 1; } else { ack_size = TCPOLEN_MPTCP_DSS_ACK32; - opts->ext_copy.data_ack32 = (uint32_t)ack_seq; opts->ext_copy.ack64 = 0; } opts->ext_copy.use_ack = 1; @@ -1297,19 +1293,14 @@ bool mptcp_incoming_options(struct sock *sk, struct sk_buff *skb) return true; } -static void mptcp_set_rwin(struct tcp_sock *tp, struct tcphdr *th) +static u64 mptcp_set_rwin(struct mptcp_sock *msk, struct tcp_sock *tp, + struct tcphdr *th, u64 ack_seq) { const struct sock *ssk = (const struct sock *)tp; - struct mptcp_subflow_context *subflow; - u64 ack_seq, rcv_wnd_old, rcv_wnd_new; - struct mptcp_sock *msk; + u64 rcv_wnd_old, rcv_wnd_new; u32 new_win; u64 win; - subflow = mptcp_subflow_ctx(ssk); - msk = mptcp_sk(subflow->conn); - - ack_seq = READ_ONCE(msk->ack_seq); rcv_wnd_new = ack_seq + tp->rcv_wnd; rcv_wnd_old = atomic64_read(&msk->rcv_wnd_sent); @@ -1362,7 +1353,7 @@ static void mptcp_set_rwin(struct tcp_sock *tp, struct tcphdr *th) update_wspace: WRITE_ONCE(msk->old_wspace, tp->rcv_wnd); - subflow->rcv_wnd_sent = rcv_wnd_new; + return rcv_wnd_new; } static void mptcp_track_rwin(struct tcp_sock *tp) @@ -1474,13 +1465,25 @@ void mptcp_write_options(struct tcphdr *th, __be32 *ptr, struct tcp_sock *tp, *ptr++ = mptcp_option(MPTCPOPT_DSS, len, 0, flags); if (mpext->use_ack) { + struct mptcp_sock *msk; + u64 ack_seq; + + /* DSS option is set only by mptcp_established_options, + * the caller is __tcp_transmit_skb() and ssk is always + * not NULL. + */ + subflow = mptcp_subflow_ctx(ssk); + msk = mptcp_sk(subflow->conn); + ack_seq = READ_ONCE(msk->ack_seq); if (mpext->ack64) { - put_unaligned_be64(mpext->data_ack, ptr); + put_unaligned_be64(ack_seq, ptr); ptr += 2; } else { - put_unaligned_be32(mpext->data_ack32, ptr); + put_unaligned_be32(ack_seq, ptr); ptr += 1; } + subflow->rcv_wnd_sent = mptcp_set_rwin(msk, tp, th, + ack_seq); } if (mpext->use_map) { @@ -1708,9 +1711,6 @@ void mptcp_write_options(struct tcphdr *th, __be32 *ptr, struct tcp_sock *tp, i += 4; } } - - if (tp) - mptcp_set_rwin(tp, th); } __be32 mptcp_get_reset_option(const struct sk_buff *skb) From da23be77e1292cd611e736c3aa17da633d7ddce7 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 2 Jun 2026 22:14:11 +1000 Subject: [PATCH 247/329] mptcp: allow subflow rcv wnd to shrink In MPTCP connection, the `window` field in the TCP header refers to the MPTCP-level rcv_nxt and it's right edge should not move backward. Such constraint is enforced at DSS option generation time. At the same time, the TCP stack ensures independently that the TCP-level rcv wnd right's edge does not move backward. That in turn causes artificial inflating of the MPTCP rcv window when the incoming data is acked at the TCP level and is OoO in the MPTCP sequence space (or lands in the backlog). As a consequence, the incoming traffic can exceed the receiver rcvbuf size even when the sender is not misbehaving. Prevent such scenario forcibly allowing the TCP subflow to shrink the TCP-level rcv wnd regardless of the current netns setting. Fixes: f3589be0c420 ("mptcp: never shrink offered window") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-4-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/options.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 2d25f319f328..51ca334678b4 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -566,6 +566,7 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); struct mptcp_sock *msk = mptcp_sk(subflow->conn); + struct tcp_sock *tp = tcp_sk(sk); unsigned int dss_size = 0; struct mptcp_ext *mpext; unsigned int ack_size; @@ -614,6 +615,12 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, if (dss_size == 0) ack_size += TCPOLEN_MPTCP_DSS_BASE; + /* The caller is __tcp_transmit_skb(), and will compute the new rcv + * wnd soon: ensure that the window can shrink. + */ + if (skb) + tp->rcv_wnd = tp->rcv_nxt - tp->rcv_wup; + dss_size += ack_size; *size = ALIGN(dss_size, 4); From 14e9fea30b68fc75b2b3d97396a7e6adb544bd2a Mon Sep 17 00:00:00 2001 From: Tao Cui Date: Tue, 2 Jun 2026 22:14:12 +1000 Subject: [PATCH 248/329] mptcp: pm: fix extra_subflows underflow on userspace PM subflow creation The userspace PM increments extra_subflows after __mptcp_subflow_connect() succeeds, but __mptcp_subflow_connect() calls mptcp_pm_close_subflow() on failure to roll back the pre-increment done by the kernel PM's fill_*() helpers. Because the userspace PM hasn't incremented yet at that point, this decrement is spurious and causes extra_subflows to underflow. Fix it by aligning the userspace PM with the kernel PM: increment extra_subflows before calling __mptcp_subflow_connect(), so the existing error path in subflow.c correctly rolls it back on failure. Also simplify the error handling by taking pm.lock only when needed for cleanup. Fixes: 77e4b94a3de6 ("mptcp: update userspace pm infos") Cc: stable@vger.kernel.org Signed-off-by: Tao Cui Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-5-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/pm_userspace.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-) diff --git a/net/mptcp/pm_userspace.c b/net/mptcp/pm_userspace.c index 8cbc1920afb4..0d3a95e676f1 100644 --- a/net/mptcp/pm_userspace.c +++ b/net/mptcp/pm_userspace.c @@ -408,19 +408,21 @@ int mptcp_pm_nl_subflow_create_doit(struct sk_buff *skb, struct genl_info *info) local.flags = entry.flags; local.ifindex = entry.ifindex; + spin_lock_bh(&msk->pm.lock); + msk->pm.extra_subflows++; + spin_unlock_bh(&msk->pm.lock); + lock_sock(sk); err = __mptcp_subflow_connect(sk, &local, &addr_r); release_sock(sk); - if (err) + if (err) { GENL_SET_ERR_MSG_FMT(info, "connect error: %d", err); - spin_lock_bh(&msk->pm.lock); - if (err) + spin_lock_bh(&msk->pm.lock); mptcp_userspace_pm_delete_local_addr(msk, &entry); - else - msk->pm.extra_subflows++; - spin_unlock_bh(&msk->pm.lock); + spin_unlock_bh(&msk->pm.lock); + } create_err: sock_put(sk); From 06fd2bec7aebf393288e4b78924482fe170caabc Mon Sep 17 00:00:00 2001 From: Tao Cui Date: Tue, 2 Jun 2026 22:14:13 +1000 Subject: [PATCH 249/329] selftests: mptcp: add test for extra_subflows underflow on userspace PM Add a test to verify that when userspace PM fails to create a subflow (e.g. using an unreachable address), the extra_subflows counter is not decremented below zero. Fixes: 77e4b94a3de6 ("mptcp: update userspace pm infos") Cc: stable@vger.kernel.org Signed-off-by: Tao Cui Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-6-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/mptcp/mptcp_join.sh | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/net/mptcp/mptcp_join.sh b/tools/testing/selftests/net/mptcp/mptcp_join.sh index 5acd12021e6e..4b3f71e66609 100755 --- a/tools/testing/selftests/net/mptcp/mptcp_join.sh +++ b/tools/testing/selftests/net/mptcp/mptcp_join.sh @@ -4100,6 +4100,10 @@ userspace_tests() chk_rm_nr 0 1 chk_mptcp_info subflows 0 subflows 0 chk_subflows_total 1 1 + # check counters are not affected by errors at creation time + userspace_pm_add_sf $ns2 10.0.12.2 10 2>/dev/null + chk_mptcp_info subflows 0 subflows 0 + chk_subflows_total 1 1 kill_events_pids mptcp_lib_kill_group_wait $tests_pid fi From 57132affbc89c02e1bf73fdf5724311bdc9a29da Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 2 Jun 2026 22:14:14 +1000 Subject: [PATCH 250/329] mptcp: sockopt: check timestamping ret value sock_set_timestamping() can fail for different reasons. The returned value should then be checked. If sock_set_timestamping() fails for at least one subflow, the first error is now reported to the userspace, similar to what is done with other socket options. Fixes: 9061f24bf82e ("mptcp: sockopt: propagate timestamp request to subflows") Cc: stable@vger.kernel.org Reported-by: Willem de Bruijn Closes: https://lore.kernel.org/willemdebruijn.kernel.178a41a53d041@gmail.com Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-7-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 87b5796d0135..91aa57f1d0fd 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -241,15 +241,19 @@ static int mptcp_setsockopt_sol_socket_timestamping(struct mptcp_sock *msk, mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); + int err; lock_sock(ssk); - sock_set_timestamping(ssk, optname, timestamping); + err = sock_set_timestamping(ssk, optname, timestamping); release_sock(ssk); + + if (err < 0 && ret == 0) + ret = err; } release_sock(sk); - return 0; + return ret; } static int mptcp_setsockopt_sol_socket_linger(struct mptcp_sock *msk, sockptr_t optval, From 7690137e70ab0fb1f8b5a30e6f087f8ee908b680 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 2 Jun 2026 22:14:15 +1000 Subject: [PATCH 251/329] mptcp: sockopt: set sockopt on all subflows The mptcp_setsockopt_all_sf(), currently used only with TCP_MAXSEG, stopped when one subflow returned an error. Even if it is not wrong, this is different from the other helpers trying to set the option on all subflows, and then returning an error if at least one of them had an issue. Follow this behaviour, for a question of uniformity. Fixes: 51c5fd09e1b4 ("mptcp: add TCP_MAXSEG sockopt support") Cc: stable@vger.kernel.org Reviewed-by: Mat Martineau Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-8-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/sockopt.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/net/mptcp/sockopt.c b/net/mptcp/sockopt.c index 91aa57f1d0fd..fcf6feb2a9eb 100644 --- a/net/mptcp/sockopt.c +++ b/net/mptcp/sockopt.c @@ -817,10 +817,11 @@ static int mptcp_setsockopt_all_sf(struct mptcp_sock *msk, int level, mptcp_for_each_subflow(msk, subflow) { struct sock *ssk = mptcp_subflow_tcp_sock(subflow); + int err; - ret = tcp_setsockopt(ssk, level, optname, optval, optlen); - if (ret) - break; + err = tcp_setsockopt(ssk, level, optname, optval, optlen); + if (err < 0 && ret == 0) + ret = err; } if (!ret) From c378b1a6f8dd3e02eb08661f4d5d50f236eead03 Mon Sep 17 00:00:00 2001 From: Gang Yan Date: Tue, 2 Jun 2026 22:14:16 +1000 Subject: [PATCH 252/329] mptcp: check desc->count in read_sock __tcp_read_sock() checks desc->count after each skb is consumed and breaks the loop when it reaches 0. The MPTCP variant lacks this check. This is a functional bug, other subsystems also rely on this check: TLS strparser sets desc->count to 0 once a full TLS record is assembled and depends on this break to stop reading. Add the same desc->count check to __mptcp_read_sock(), mirroring __tcp_read_sock(). Fixes: 250d9766a984 ("mptcp: implement .read_sock") Cc: stable@vger.kernel.org Co-developed-by: Geliang Tang Signed-off-by: Geliang Tang Signed-off-by: Gang Yan Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-9-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/protocol.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 7fac5fac2097..cb9515f505aa 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -4428,6 +4428,8 @@ static int __mptcp_read_sock(struct sock *sk, read_descriptor_t *desc, } mptcp_eat_recv_skb(sk, skb); + if (!desc->count) + break; } if (noack) From 5e939544f9d2b4d5c052a07cfcde97de44263946 Mon Sep 17 00:00:00 2001 From: Paolo Abeni Date: Tue, 2 Jun 2026 22:14:17 +1000 Subject: [PATCH 253/329] mptcp: fix uninit-value in mptcp_established_options syzbot reported the following uninit splat: BUG: KMSAN: uninit-value in mptcp_write_data_fin net/mptcp/options.c:542 [inline] BUG: KMSAN: uninit-value in mptcp_established_options_dss net/mptcp/options.c:590 [inline] BUG: KMSAN: uninit-value in mptcp_established_options+0x112f/0x3530 net/mptcp/options.c:874 mptcp_write_data_fin net/mptcp/options.c:542 [inline] mptcp_established_options_dss net/mptcp/options.c:590 [inline] mptcp_established_options+0x112f/0x3530 net/mptcp/options.c:874 tcp_established_options+0x312/0xcc0 net/ipv4/tcp_output.c:1192 __tcp_transmit_skb+0x5dc/0x5fe0 net/ipv4/tcp_output.c:1575 __tcp_send_ack+0x967/0xad0 net/ipv4/tcp_output.c:4499 tcp_send_ack+0x3d/0x60 net/ipv4/tcp_output.c:4505 mptcp_subflow_shutdown+0x164/0x690 net/mptcp/protocol.c:3137 mptcp_check_send_data_fin+0x31b/0x3d0 net/mptcp/protocol.c:3218 __mptcp_wr_shutdown net/mptcp/protocol.c:3234 [inline] __mptcp_close+0x860/0x1360 net/mptcp/protocol.c:3313 mptcp_close+0x42/0x260 net/mptcp/protocol.c:3367 inet_release+0x1ee/0x2a0 net/ipv4/af_inet.c:442 __sock_release net/socket.c:722 [inline] sock_close+0xd6/0x2f0 net/socket.c:1514 __fput+0x60e/0x1010 fs/file_table.c:510 ____fput+0x25/0x30 fs/file_table.c:538 task_work_run+0x208/0x2b0 kernel/task_work.c:233 resume_user_mode_work include/linux/resume_user_mode.h:50 [inline] __exit_to_user_mode_loop kernel/entry/common.c:67 [inline] exit_to_user_mode_loop+0x306/0x1b60 kernel/entry/common.c:98 __exit_to_user_mode_prepare include/linux/irq-entry-common.h:207 [inline] syscall_exit_to_user_mode_prepare include/linux/irq-entry-common.h:238 [inline] syscall_exit_to_user_mode include/linux/entry-common.h:318 [inline] __do_fast_syscall_32+0x2c7/0x460 arch/x86/entry/syscall_32.c:310 do_fast_syscall_32+0x37/0x80 arch/x86/entry/syscall_32.c:332 do_SYSENTER_32+0x1f/0x30 arch/x86/entry/syscall_32.c:370 entry_SYSENTER_compat_after_hwframe+0x84/0x8e Local variable opts created at: __tcp_transmit_skb+0x4d/0x5fe0 net/ipv4/tcp_output.c:1536 __tcp_send_ack+0x967/0xad0 net/ipv4/tcp_output.c:4499 The output path currently omits initializing the mptcp extension `use_map` flag in a few corner cases. Address the issue always zeroing all the extensions flags before eventually initializing the individual bits. To that extent, introduce and use a struct_group to avoid multiple bitwise operations. Fixes: cfcceb7a39fc ("tcp: shrink per-packet memset in __tcp_transmit_skb()") Cc: stable@vger.kernel.org Reported-by: syzbot+ff020673c5e3d94d9478@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=ff020673c5e3d94d9478 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts (NGI0) Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-10-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- include/net/mptcp.h | 7 +++++-- net/mptcp/options.c | 6 +++++- 2 files changed, 10 insertions(+), 3 deletions(-) diff --git a/include/net/mptcp.h b/include/net/mptcp.h index f7263fe2a2e4..ee70f597a4de 100644 --- a/include/net/mptcp.h +++ b/include/net/mptcp.h @@ -27,7 +27,9 @@ struct mptcp_ext { u32 subflow_seq; u16 data_len; __sum16 csum; - u8 use_map:1, + + struct_group(flags, + u8 use_map:1, dsn64:1, data_fin:1, use_ack:1, @@ -35,9 +37,10 @@ struct mptcp_ext { mpc_map:1, frozen:1, reset_transient:1; - u8 reset_reason:4, + u8 reset_reason:4, csum_reqd:1, infinite_map:1; + ); /* end of flags group */ }; #define MPTCPOPT_HMAC_LEN 20 diff --git a/net/mptcp/options.c b/net/mptcp/options.c index 51ca334678b4..f9f587203c35 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -572,6 +572,11 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, unsigned int ack_size; bool ret = false; + /* Zero `use_ack` and `use_map` flags with one shot. */ + BUILD_BUG_ON(sizeof_field(struct mptcp_ext, flags) != sizeof(u16)); + BUILD_BUG_ON(!IS_ALIGNED(offsetof(struct mptcp_ext, flags), + sizeof(u16))); + *(u16 *)&opts->ext_copy.flags = 0; opts->csum_reqd = READ_ONCE(msk->csum_enabled); mpext = skb ? mptcp_get_ext(skb) : NULL; @@ -595,7 +600,6 @@ static bool mptcp_established_options_dss(struct sock *sk, struct sk_buff *skb, /* passive sockets msk will set the 'can_ack' after accept(), even * if the first subflow may have the already the remote key handy */ - opts->ext_copy.use_ack = 0; if (!READ_ONCE(msk->can_ack)) { *size = ALIGN(dss_size, 4); return ret; From bd34fa0257261b76964df1c98f44b3cb4ee14620 Mon Sep 17 00:00:00 2001 From: "Matthieu Baerts (NGI0)" Date: Tue, 2 Jun 2026 22:14:18 +1000 Subject: [PATCH 254/329] mptcp: add-addr: always drop other suboptions When an ADD_ADDR needs to be sent, it could be prepared if there is enough remaining space and even if the packet is not a pure ACK. But it would be dropped soon after. Indeed, in mptcp_pm_add_addr_signal(), there is enough space to fit a DSS of 20 octets and an ADD_ADDR echo containing an IPv4 address on 8 octets for example. In this case, the packet would be prepared, the MPTCP_ADD_ADDR_ECHO bit would be removed from pm->addr_signal, but the option would be silently dropped in mptcp_established_options_add_addr() not to override DSS info in the union from 'struct mptcp_out_options', and also because mptcp_write_options() will enforce mutually exclusion with DSS. Instead, don't even try to send an ADD_ADDR if it is not a pure ACK. Retry for each new packet until a pure-ACK is emitted. That's fine to do that, because each time an ADD_ADDR (echo) is scheduled, a pure ACK is queued. This also simplifies the code, and the skb checks can be done earlier, before the lock. Note: also, since commit 6d0060f600ad ("mptcp: Write MPTCP DSS headers to outgoing data packets"), opts->ahmac would not have been set to 0 when other suboptions were not dropped, and when sending an ADD_ADDR echo. That would have resulted in sending an ADD_ADDR using garbage info, where there was not enough space, instead of an echo one without the ADD_ADDR HMAC. Fixes: 1bff1e43a30e ("mptcp: optimize out option generation") Cc: stable@vger.kernel.org Signed-off-by: Matthieu Baerts (NGI0) Link: https://patch.msgid.link/20260602-net-mptcp-misc-fixes-7-1-rc7-v2-11-856831229976@kernel.org Signed-off-by: Jakub Kicinski --- net/mptcp/options.c | 30 +++++++----------------------- net/mptcp/pm.c | 15 ++++----------- net/mptcp/protocol.h | 7 +++---- 3 files changed, 14 insertions(+), 38 deletions(-) diff --git a/net/mptcp/options.c b/net/mptcp/options.c index f9f587203c35..b3ea7854818f 100644 --- a/net/mptcp/options.c +++ b/net/mptcp/options.c @@ -665,7 +665,6 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * { struct mptcp_subflow_context *subflow = mptcp_subflow_ctx(sk); struct mptcp_sock *msk = mptcp_sk(subflow->conn); - bool drop_other_suboptions = false; unsigned int opt_size = *size; struct mptcp_addr_info addr; bool echo; @@ -676,36 +675,20 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * */ if (!mptcp_pm_should_add_signal(msk) || (opts->suboptions & (OPTION_MPTCP_MPJ_ACK | OPTION_MPTCP_MPC_ACK)) || - !mptcp_pm_add_addr_signal(msk, skb, opt_size, remaining, &addr, - &echo, &drop_other_suboptions)) + !skb || !skb_is_tcp_pure_ack(skb) || + !mptcp_pm_add_addr_signal(msk, opt_size, remaining, &addr, &echo)) return false; - /* - * Later on, mptcp_write_options() will enforce mutually exclusion with - * DSS, bail out if such option is set and we can't drop it. - */ - if (drop_other_suboptions) - remaining += opt_size; - else if (opts->suboptions & OPTION_MPTCP_DSS) - return false; + remaining += opt_size; len = mptcp_add_addr_len(addr.family, echo, !!addr.port); if (remaining < len) return false; *size = len; - if (drop_other_suboptions) { - pr_debug("drop other suboptions\n"); - opts->suboptions = 0; - - /* note that e.g. DSS could have written into the memory - * aliased by ahmac, we must reset the field here - * to avoid appending the hmac even for ADD_ADDR echo - * options - */ - opts->ahmac = 0; - *size -= opt_size; - } + pr_debug("drop other suboptions\n"); + opts->suboptions = 0; + *size -= opt_size; opts->addr = addr; opts->suboptions |= OPTION_MPTCP_ADD_ADDR; if (!echo) { @@ -715,6 +698,7 @@ static bool mptcp_established_options_add_addr(struct sock *sk, struct sk_buff * &opts->addr); } else { MPTCP_INC_STATS(sock_net(sk), MPTCP_MIB_ECHOADDTX); + opts->ahmac = 0; } pr_debug("addr_id=%d, ahmac=%llu, echo=%d, port=%d\n", opts->addr.id, opts->ahmac, echo, ntohs(opts->addr.port)); diff --git a/net/mptcp/pm.c b/net/mptcp/pm.c index 3e770c7407e1..470501470fe5 100644 --- a/net/mptcp/pm.c +++ b/net/mptcp/pm.c @@ -887,10 +887,9 @@ void mptcp_pm_mp_fail_received(struct sock *sk, u64 fail_seq) } } -bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, const struct sk_buff *skb, - unsigned int opt_size, unsigned int remaining, - struct mptcp_addr_info *addr, bool *echo, - bool *drop_other_suboptions) +bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, unsigned int opt_size, + unsigned int remaining, + struct mptcp_addr_info *addr, bool *echo) { bool skip_add_addr = false; int ret = false; @@ -908,10 +907,7 @@ bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, const struct sk_buff *skb, * plain dup-ack from TCP perspective. The other MPTCP-relevant info, * if any, will be carried by the 'original' TCP ack */ - if (skb && skb_is_tcp_pure_ack(skb)) { - remaining += opt_size; - *drop_other_suboptions = true; - } + remaining += opt_size; *echo = mptcp_pm_should_add_signal_echo(msk); if (*echo) { @@ -929,9 +925,6 @@ bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, const struct sk_buff *skb, if (remaining < mptcp_add_addr_len(family, *echo, port)) { struct net *net = sock_net((struct sock *)msk); - if (!*drop_other_suboptions) - goto out_unlock; - if (*echo) { MPTCP_INC_STATS(net, MPTCP_MIB_ECHOADDTXDROP); } else { diff --git a/net/mptcp/protocol.h b/net/mptcp/protocol.h index e4f5aba24da7..b93b878478d2 100644 --- a/net/mptcp/protocol.h +++ b/net/mptcp/protocol.h @@ -1229,10 +1229,9 @@ static inline int mptcp_rm_addr_len(const struct mptcp_rm_list *rm_list) return TCPOLEN_MPTCP_RM_ADDR_BASE + roundup(rm_list->nr - 1, 4) + 1; } -bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, const struct sk_buff *skb, - unsigned int opt_size, unsigned int remaining, - struct mptcp_addr_info *addr, bool *echo, - bool *drop_other_suboptions); +bool mptcp_pm_add_addr_signal(struct mptcp_sock *msk, unsigned int opt_size, + unsigned int remaining, + struct mptcp_addr_info *addr, bool *echo); bool mptcp_pm_rm_addr_signal(struct mptcp_sock *msk, unsigned int remaining, struct mptcp_rm_list *rm_list); int mptcp_pm_get_local_id(struct mptcp_sock *msk, struct sock_common *skc); From b38cae85d1c45ff189d7ecb6ac36f41cdc3d84d0 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 2 Jun 2026 11:21:04 +0200 Subject: [PATCH 255/329] net: airoha: Fix use-after-free in metadata dst teardown airoha_metadata_dst_free() runs metadata_dst_free() which frees the metadata_dst with kfree() immediately, bypassing the RCU grace period. In the RX path, skb_dst_set_noref() sets a non-refcounted pointer from the skb to the metadata_dst. This function requires RCU read-side protection and the dst must remain valid until all RCU readers complete. Since metadata_dst_free() calls kfree() directly, an use-after-free can occur if any skb still holds a noref pointer to the dst when the driver tears it down. Replace metadata_dst_free() with dst_release() which properly goes through the refcount path: when the refcount drops to zero, it schedules the actual free via call_rcu_hurry(), ensuring all RCU readers have completed before the memory is freed. Fixes: af3cf757d5c9 ("net: airoha: Move DSA tag in DMA descriptor") Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20260602-airoha-mtk-metadata-uaf-fix-v1-1-3aaa99d83351@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/airoha/airoha_eth.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/airoha/airoha_eth.c b/drivers/net/ethernet/airoha/airoha_eth.c index cecd66251dba..eab6a98d62b9 100644 --- a/drivers/net/ethernet/airoha/airoha_eth.c +++ b/drivers/net/ethernet/airoha/airoha_eth.c @@ -2936,7 +2936,7 @@ static void airoha_metadata_dst_free(struct airoha_gdm_port *port) if (!port->dsa_meta[i]) continue; - metadata_dst_free(port->dsa_meta[i]); + dst_release(&port->dsa_meta[i]->dst); } } From 80df409e1a483676826a6c66e693dba6ac507751 Mon Sep 17 00:00:00 2001 From: Lorenzo Bianconi Date: Tue, 2 Jun 2026 11:21:05 +0200 Subject: [PATCH 256/329] net: ethernet: mtk_eth_soc: Fix use-after-free in metadata dst teardown mtk_free_dev() calls metadata_dst_free() which frees the metadata_dst with kfree() immediately, bypassing the RCU grace period. In the RX path, skb_dst_set_noref() sets a non-refcounted pointer from the skb to the metadata_dst. This function requires RCU read-side protection and the dst must remain valid until all RCU readers complete. Since metadata_dst_free() calls kfree() directly, a use-after-free can occur if any skb still holds a noref pointer to the dst when the driver tears it down. Replace metadata_dst_free() with dst_release() which properly goes through the refcount path: when the refcount drops to zero, it schedules the actual free via call_rcu_hurry(), ensuring all RCU readers have completed before the memory is freed. Fixes: 2d7605a72906 ("net: ethernet: mtk_eth_soc: enable hardware DSA untagging") Signed-off-by: Lorenzo Bianconi Link: https://patch.msgid.link/20260602-airoha-mtk-metadata-uaf-fix-v1-2-3aaa99d83351@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/mediatek/mtk_eth_soc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c index 8d225bc9f063..7d771168b990 100644 --- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c +++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c @@ -4491,7 +4491,7 @@ static int mtk_free_dev(struct mtk_eth *eth) for (i = 0; i < ARRAY_SIZE(eth->dsa_meta); i++) { if (!eth->dsa_meta[i]) break; - metadata_dst_free(eth->dsa_meta[i]); + dst_release(ð->dsa_meta[i]->dst); } return 0; From 40ab6644b99685755f740b872c00ef40d9aa870e Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Wed, 3 Jun 2026 21:31:57 +0200 Subject: [PATCH 257/329] fhandle: fix UAF due to unlocked ->mnt_ns read in may_decode_fh() may_decode_fh() accesses mount::mnt_ns without holding any locks; that means the mount can concurrently be unmounted, and the mnt_namespace can concurrently be freed after an RCU grace period. This race can happens as follows, assuming that the mount point was created by open_tree(..., OPEN_TREE_CLONE): thread 1 thread 2 RCU __do_sys_open_by_handle_at do_handle_open handle_to_path may_decode_fh is_mounted [mount::mnt_ns access] [mount::mnt_ns access] __do_sys_close fput_close_sync __fput dissolve_on_fput umount_tree class_namespace_excl_destructor namespace_unlock free_mnt_ns mnt_ns_tree_remove call_rcu(mnt_ns_release_rcu) mnt_ns_release_rcu mnt_ns_release kfree [mnt_namespace::user_ns access] **UAF** Fix it by taking rcu_read_lock() around the mount::mnt_ns access, like in __prepend_path(). Additionally, document the semantics of mount::mnt_ns, and use WRITE_ONCE() for writers that can race with lockless readers. This bug is unreachable unless one of the following is set: - CONFIG_PREEMPTION - CONFIG_RCU_STRICT_GRACE_PERIOD because it requires an RCU grace period to happen during a syscall without an explicit preemption. This doesn't seem to have interesting security impact; worst-case, it could leak the result of an integer comparison to userspace (from the level check in cap_capable()), cause an endless loop, or crash the kernel by dereferencing an invalid address. Fixes: 620c266f3949 ("fhandle: relax open_by_handle_at() permission checks") Cc: stable@vger.kernel.org Signed-off-by: Jann Horn Link: https://patch.msgid.link/20260603-vfs-fhandle-uaf-fix-v2-1-d05db76a5084@google.com Signed-off-by: Christian Brauner (Amutable) --- fs/fhandle.c | 16 ++++++++++++++-- fs/mount.h | 10 +++++++++- fs/namespace.c | 6 +++--- 3 files changed, 26 insertions(+), 6 deletions(-) diff --git a/fs/fhandle.c b/fs/fhandle.c index 642e3d569497..1ca7eb3a6cb5 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -285,6 +285,19 @@ static int do_handle_to_path(struct file_handle *handle, struct path *path, return 0; } +static bool capable_wrt_mount(struct mount *mount) +{ + struct mnt_namespace *mnt_ns; + + /* + * For ->mnt_ns access. + * The following READ_ONCE() is semantically rcu_dereference(). + */ + guard(rcu)(); + mnt_ns = READ_ONCE(mount->mnt_ns); + return ns_capable(mnt_ns->user_ns, CAP_SYS_ADMIN); +} + static inline int may_decode_fh(struct handle_to_path_ctx *ctx, unsigned int o_flags) { @@ -320,8 +333,7 @@ static inline int may_decode_fh(struct handle_to_path_ctx *ctx, if (ns_capable(root->mnt->mnt_sb->s_user_ns, CAP_SYS_ADMIN)) ctx->flags = HANDLE_CHECK_PERMS; else if (is_mounted(root->mnt) && - ns_capable(real_mount(root->mnt)->mnt_ns->user_ns, - CAP_SYS_ADMIN) && + capable_wrt_mount(real_mount(root->mnt)) && !has_locked_children(real_mount(root->mnt), root->dentry)) ctx->flags = HANDLE_CHECK_PERMS | HANDLE_CHECK_SUBTREE; else diff --git a/fs/mount.h b/fs/mount.h index e0816c11a198..5c120f8361bd 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -71,7 +71,15 @@ struct mount { struct hlist_head mnt_slave_list;/* list of slave mounts */ struct hlist_node mnt_slave; /* slave list entry */ struct mount *mnt_master; /* slave is on master->mnt_slave_list */ - struct mnt_namespace *mnt_ns; /* containing namespace */ + /* + * Containing namespace (active or deactivating, non-refcounted). + * Normally protected by namespace_sem. + * Can also be accessed locklessly under RCU. RCU readers can't rely on + * the namespace still being active, but implicitly hold a passive + * reference (because an RCU delay happens between a namespace being + * deactivated and the corresponding passive refcount drop). + */ + struct mnt_namespace *mnt_ns; struct mountpoint *mnt_mp; /* where is it mounted */ union { struct hlist_node mnt_mp_list; /* list mounts with the same mountpoint */ diff --git a/fs/namespace.c b/fs/namespace.c index fe919abd2f01..f5905f4ec560 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -1079,7 +1079,7 @@ static void mnt_add_to_ns(struct mnt_namespace *ns, struct mount *mnt) bool mnt_first_node = true, mnt_last_node = true; WARN_ON(mnt_ns_attached(mnt)); - mnt->mnt_ns = ns; + WRITE_ONCE(mnt->mnt_ns, ns); while (*link) { parent = *link; if (mnt->mnt_id_unique < node_to_mount(parent)->mnt_id_unique) { @@ -1434,7 +1434,7 @@ EXPORT_SYMBOL(mntget); void mnt_make_shortterm(struct vfsmount *mnt) { if (mnt) - real_mount(mnt)->mnt_ns = NULL; + WRITE_ONCE(real_mount(mnt)->mnt_ns, NULL); } /** @@ -1806,7 +1806,7 @@ static void umount_tree(struct mount *mnt, enum umount_tree_flags how) ns->nr_mounts--; __touch_mnt_namespace(ns); } - p->mnt_ns = NULL; + WRITE_ONCE(p->mnt_ns, NULL); if (how & UMOUNT_SYNC) p->mnt.mnt_flags |= MNT_SYNC_UMOUNT; From 2eea7f44b9c8b42fd7d3a1a87c06a7cd1b99c327 Mon Sep 17 00:00:00 2001 From: Joanne Koong Date: Wed, 3 Jun 2026 18:18:58 -0700 Subject: [PATCH 258/329] iomap: avoid potential null folio->mapping deref during error reporting When a buffered read fails, iomap_finish_folio_read() reports the error with fserror_report_io(folio->mapping->host, ...). This is called after ifs->read_bytes_pending has been decremented by the bytes attempted to be read. For a folio split across multiple read completions, the folio is only guaranteed to stay locked while read_bytes_pending > 0. Once iomap_finish_folio_read() decrements read_bytes_pending, another in-flight read can complete and end the read on the folio, which unlocks it. This allows truncate logic to run and detach the folio (set folio->mapping to NULL). The error reporting path then can dereference a NULL folio->mapping. As reported by Sam Sun, this is the race that can occur: CPU0: failed completion CPU1: final completion CPU2: truncate ----------------------- ---------------------- -------------- read_bytes_pending -= len finished = false /* preempted before fserror_report_io() */ read_bytes_pending -= len finished = true folio_end_read() truncate clears folio->mapping fserror_report_io( folio->mapping->host, ...) ^ NULL deref Fix this by reporting the error first before decrementing ifs->read_bytes_pending. Fixes: a9d573ee88af ("iomap: report file I/O errors to the VFS") Cc: stable@vger.kernel.org Reported-by: Sam Sun Closes: https://lore.kernel.org/linux-fsdevel/CAEkJfYPhWdd59RKmuNLJg-bkypHz7xiOwaWyNVu3A8CUqQCnvg@mail.gmail.com/ Signed-off-by: Joanne Koong Link: https://patch.msgid.link/20260604011858.2297561-1-joannelkoong@gmail.com Reviewed-by: "Darrick J. Wong" Signed-off-by: Christian Brauner (Amutable) --- fs/iomap/buffered-io.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index d7b648421a70..d55b936e6986 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -400,6 +400,11 @@ void iomap_finish_folio_read(struct folio *folio, size_t off, size_t len, bool uptodate = !error; bool finished = true; + if (error) + fserror_report_io(folio->mapping->host, FSERR_BUFFERED_READ, + folio_pos(folio) + off, len, error, + GFP_ATOMIC); + if (ifs) { unsigned long flags; @@ -411,11 +416,6 @@ void iomap_finish_folio_read(struct folio *folio, size_t off, size_t len, spin_unlock_irqrestore(&ifs->state_lock, flags); } - if (error) - fserror_report_io(folio->mapping->host, FSERR_BUFFERED_READ, - folio_pos(folio) + off, len, error, - GFP_ATOMIC); - if (finished) folio_end_read(folio, uptodate); } From 1231623fd3b5aa6b41cce799ffb0d82e10914be4 Mon Sep 17 00:00:00 2001 From: Antoine Tenart Date: Fri, 29 May 2026 16:47:00 +0200 Subject: [PATCH 259/329] geneve: fix length used in GRO hint UDP checksum adjustment In geneve_post_decap_hint the length used for adjusting the UDP checksum should be 'skb->len - gro_hint->nested_tp_offset' (UDP length) instead of 'skb->len - gro_hint->nested_nh_offset' (IP length). Fixes: fd0dd796576e ("geneve: use GRO hint option in the RX path") Cc: Paolo Abeni Reported-by: Sashiko Closes: https://sashiko.dev/#/patchset/20260521131436.748832-1-jhs%40mojatatu.com Signed-off-by: Antoine Tenart Reviewed-by: Simon Horman Link: https://patch.msgid.link/20260529144713.780938-1-atenart@kernel.org Signed-off-by: Paolo Abeni --- drivers/net/geneve.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c index c6563367d382..715180c3a1b3 100644 --- a/drivers/net/geneve.c +++ b/drivers/net/geneve.c @@ -632,7 +632,7 @@ static int geneve_post_decap_hint(const struct sock *sk, struct sk_buff *skb, uh = udp_hdr(skb); uh->len = htons(skb->len - gro_hint->nested_tp_offset); if (uh->check) { - len = skb->len - gro_hint->nested_nh_offset; + len = skb->len - gro_hint->nested_tp_offset; skb_shinfo(skb)->gso_type |= SKB_GSO_UDP_TUNNEL_CSUM; if (gro_hint->nested_is_v6) uh->check = ~udp_v6_check(len, &ipv6h->saddr, From e9e41d3035032ed6053d8bad7b7077e1cb3a6540 Mon Sep 17 00:00:00 2001 From: Tomas Glozar Date: Tue, 2 Jun 2026 14:55:06 +0200 Subject: [PATCH 260/329] rtla: Fix parsing of multi-character short options A bug was reported where the parsing of multi-character short options, be it a short option with an argument specified without space (e.g. "-p100") or multiple short options in one argument (e.g. -un), ignores options specific to individual tools. Furthermore, if the rest of the option is supposed to be an argument, it gets reinterpreted as a string of options. For example, -p100 gets interpreted as -100, which is due to hackish implementation read as --no-thread --no-irq --no-irq with timerlat hist, causing rtla to error out: $ rtla timerlat hist -p100 no-irq and no-thread set, there is nothing to do here This behavior is caused by getopt_long() being called twice on each argument, once in common_parse_options(), once in [tool]_parse_args(): - common_parse_options() calls getopt_long() with an array of options common for all rtla tools, while suppressing errors (opterr = 0). - If the option fails to parse, common_parse_options() returns 0. - If 0 is returned from common_parse_options(), [tool]_parse_args() calls getopt_long() again, with its own set of options. * [tool] means one of {osnoise,timerlat}_{top,hist} At least in glibc, getopt_long() increments its internal nextchar variable even if the option is not recognized. That means that in the case of "-p100", common_parse_options() sets nextchar pointing to '1', and timerlat_hist_parse_args() sees '1', not 'p'; the same then repeats for the first and second '0'. As there is no way to restore the correct internal state of getopt_long() reliably, fix the issue by merging the common options back to the longopt array and option string of the [tool]_parse_args() functions using a macro; only the switch part is left in the original function, which is renamed to set_common_option(). Fixes: 850cd24cb6d6 ("tools/rtla: Add common_parse_options()") Reported-by: John Kacur Tested-by: John Kacur Link: https://lore.kernel.org/r/20260602125506.3325345-1-tglozar@redhat.com Signed-off-by: Tomas Glozar --- tools/tracing/rtla/src/common.c | 28 +++++--------------------- tools/tracing/rtla/src/common.h | 12 ++++++++++- tools/tracing/rtla/src/osnoise_hist.c | 7 ++++--- tools/tracing/rtla/src/osnoise_top.c | 7 ++++--- tools/tracing/rtla/src/timerlat_hist.c | 7 ++++--- tools/tracing/rtla/src/timerlat_top.c | 7 ++++--- 6 files changed, 32 insertions(+), 36 deletions(-) diff --git a/tools/tracing/rtla/src/common.c b/tools/tracing/rtla/src/common.c index 35e3d3aa922e..bc9d01ddd102 100644 --- a/tools/tracing/rtla/src/common.c +++ b/tools/tracing/rtla/src/common.c @@ -84,37 +84,20 @@ int getopt_auto(int argc, char **argv, const struct option *long_opts) } /* - * common_parse_options - parse common command line options + * set_common_option - set common options * + * @c: option character * @argc: argument count * @argv: argument vector * @common: common parameters structure * * Parse command line options that are common to all rtla tools. * - * Returns: non zero if a common option was parsed, or 0 - * if the option should be handled by tool-specific parsing. + * Returns: 1 if the option was set, 0 otherwise. */ -int common_parse_options(int argc, char **argv, struct common_params *common) +int set_common_option(int c, int argc, char **argv, struct common_params *common) { struct trace_events *tevent; - int saved_state = optind; - int c; - - static struct option long_options[] = { - {"cpus", required_argument, 0, 'c'}, - {"cgroup", optional_argument, 0, 'C'}, - {"debug", no_argument, 0, 'D'}, - {"duration", required_argument, 0, 'd'}, - {"event", required_argument, 0, 'e'}, - {"house-keeping", required_argument, 0, 'H'}, - {"priority", required_argument, 0, 'P'}, - {0, 0, 0, 0} - }; - - opterr = 0; - c = getopt_auto(argc, argv, long_options); - opterr = 1; switch (c) { case 'c': @@ -154,11 +137,10 @@ int common_parse_options(int argc, char **argv, struct common_params *common) common->set_sched = 1; break; default: - optind = saved_state; return 0; } - return c; + return 1; } /* diff --git a/tools/tracing/rtla/src/common.h b/tools/tracing/rtla/src/common.h index 51665db4ffce..8921807bda98 100644 --- a/tools/tracing/rtla/src/common.h +++ b/tools/tracing/rtla/src/common.h @@ -178,7 +178,17 @@ int osnoise_set_stop_total_us(struct osnoise_context *context, long long stop_total_us); int getopt_auto(int argc, char **argv, const struct option *long_opts); -int common_parse_options(int argc, char **argv, struct common_params *common); + +#define COMMON_OPTIONS \ + {"cpus", required_argument, 0, 'c'},\ + {"cgroup", optional_argument, 0, 'C'},\ + {"debug", no_argument, 0, 'D'},\ + {"duration", required_argument, 0, 'd'},\ + {"event", required_argument, 0, 'e'},\ + {"house-keeping", required_argument, 0, 'H'},\ + {"priority", required_argument, 0, 'P'} +int set_common_option(int c, int argc, char **argv, struct common_params *common); + int common_apply_config(struct osnoise_tool *tool, struct common_params *params); int top_main_loop(struct osnoise_tool *tool); int hist_main_loop(struct osnoise_tool *tool); diff --git a/tools/tracing/rtla/src/osnoise_hist.c b/tools/tracing/rtla/src/osnoise_hist.c index 8ad816b80265..cb4ce58c5987 100644 --- a/tools/tracing/rtla/src/osnoise_hist.c +++ b/tools/tracing/rtla/src/osnoise_hist.c @@ -475,6 +475,7 @@ static struct common_params while (1) { static struct option long_options[] = { + COMMON_OPTIONS, {"auto", required_argument, 0, 'a'}, {"bucket-size", required_argument, 0, 'b'}, {"entries", required_argument, 0, 'E'}, @@ -498,15 +499,15 @@ static struct common_params {0, 0, 0, 0} }; - if (common_parse_options(argc, argv, ¶ms->common)) - continue; - c = getopt_auto(argc, argv, long_options); /* detect the end of the options. */ if (c == -1) break; + if (set_common_option(c, argc, argv, ¶ms->common)) + continue; + switch (c) { case 'a': /* set sample stop to auto_thresh */ diff --git a/tools/tracing/rtla/src/osnoise_top.c b/tools/tracing/rtla/src/osnoise_top.c index 244bdce022ad..e65312ec26c4 100644 --- a/tools/tracing/rtla/src/osnoise_top.c +++ b/tools/tracing/rtla/src/osnoise_top.c @@ -328,6 +328,7 @@ struct common_params *osnoise_top_parse_args(int argc, char **argv) while (1) { static struct option long_options[] = { + COMMON_OPTIONS, {"auto", required_argument, 0, 'a'}, {"help", no_argument, 0, 'h'}, {"period", required_argument, 0, 'p'}, @@ -346,15 +347,15 @@ struct common_params *osnoise_top_parse_args(int argc, char **argv) {0, 0, 0, 0} }; - if (common_parse_options(argc, argv, ¶ms->common)) - continue; - c = getopt_auto(argc, argv, long_options); /* Detect the end of the options. */ if (c == -1) break; + if (set_common_option(c, argc, argv, ¶ms->common)) + continue; + switch (c) { case 'a': /* set sample stop to auto_thresh */ diff --git a/tools/tracing/rtla/src/timerlat_hist.c b/tools/tracing/rtla/src/timerlat_hist.c index 79142af4f566..4b6708e333b8 100644 --- a/tools/tracing/rtla/src/timerlat_hist.c +++ b/tools/tracing/rtla/src/timerlat_hist.c @@ -785,6 +785,7 @@ static struct common_params while (1) { static struct option long_options[] = { + COMMON_OPTIONS, {"auto", required_argument, 0, 'a'}, {"bucket-size", required_argument, 0, 'b'}, {"entries", required_argument, 0, 'E'}, @@ -819,11 +820,11 @@ static struct common_params {0, 0, 0, 0} }; - if (common_parse_options(argc, argv, ¶ms->common)) - continue; - c = getopt_auto(argc, argv, long_options); + if (set_common_option(c, argc, argv, ¶ms->common)) + continue; + /* detect the end of the options. */ if (c == -1) break; diff --git a/tools/tracing/rtla/src/timerlat_top.c b/tools/tracing/rtla/src/timerlat_top.c index 64cbdcc878b0..91f88bbebad9 100644 --- a/tools/tracing/rtla/src/timerlat_top.c +++ b/tools/tracing/rtla/src/timerlat_top.c @@ -549,6 +549,7 @@ static struct common_params while (1) { static struct option long_options[] = { + COMMON_OPTIONS, {"auto", required_argument, 0, 'a'}, {"help", no_argument, 0, 'h'}, {"irq", required_argument, 0, 'i'}, @@ -577,11 +578,11 @@ static struct common_params {0, 0, 0, 0} }; - if (common_parse_options(argc, argv, ¶ms->common)) - continue; - c = getopt_auto(argc, argv, long_options); + if (set_common_option(c, argc, argv, ¶ms->common)) + continue; + /* detect the end of the options. */ if (c == -1) break; From 0652a3daa78723f955b1ebeb621665ce72bec53e Mon Sep 17 00:00:00 2001 From: Eva Kurchatova Date: Wed, 3 Jun 2026 18:31:42 +0300 Subject: [PATCH 261/329] tracing: Fix CFI violation in probestub being called by tprobes The probestub is a function to allow tprobes to hook to a tracepoint to gain access to its parameters. The function itself is only referenced by the tracepoint structure which lives in the __tracepoint section. objtool explicitly ignores that section and when processing functions in the kernel, if it detects one that has no references it will seal it to have its ENDBR stripped on boot up. This means when a tprobe is attached to the sched_wakeup tracepoint, when it is triggered it will call __probestub_sched_wakeup and due to the missing ENDBR on a CFI-enabled machine it will take a #CP exception. Fix this by adding CFI_NOSEAL annotation to probestub declaration. Cc: stable@vger.kernel.org Acked-by: Masami Hiramatsu (Google) Link: https://patch.msgid.link/20260603153147.573589-1-eva.kurchatova@virtuozzo.com Fixes: d5173f753750 ("objtool: Exclude __tracepoints data from ENDBR checks") Signed-off-by: Eva Kurchatova [ Updated change log ] Signed-off-by: Steven Rostedt --- include/linux/tracepoint.h | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h index 763eea4d80d8..2d2b9f8cdda4 100644 --- a/include/linux/tracepoint.h +++ b/include/linux/tracepoint.h @@ -20,6 +20,7 @@ #include #include #include +#include struct module; struct tracepoint; @@ -389,6 +390,13 @@ static inline struct tracepoint *tracepoint_ptr_deref(tracepoint_ptr_t *p) void __probestub_##_name(void *__data, proto) \ { \ } \ + /* \ + * Annotate the probestub 'CFI_NOSEAL' to stop objtool from \ + * requesting the kernel remove the ENDBR, because the only \ + * references to the function are in the __tracepoint section, \ + * that objtool doesn't scan. \ + */ \ + CFI_NOSEAL(__probestub_##_name); \ DEFINE_STATIC_CALL(tp_func_##_name, __traceiter_##_name); \ DEFINE_RUST_DO_TRACE(_name, TP_PROTO(proto), TP_ARGS(args)) From a764b0e8317a863006e05732e1aefe821b9d8c2d Mon Sep 17 00:00:00 2001 From: ZhaoJinming Date: Mon, 1 Jun 2026 16:56:49 +0800 Subject: [PATCH 262/329] net: bonding: fix NULL pointer dereference in bond_do_ioctl() In bond_do_ioctl(), slave_dev is obtained via __dev_get_by_name() which can return NULL if the requested interface name does not exist. However, the subsequent slave_dbg() call is placed before the NULL check: slave_dev = __dev_get_by_name(net, ifr->ifr_slave); slave_dbg(bond_dev, slave_dev, "slave_dev=%p:\n", slave_dev); //here if (!slave_dev) return -ENODEV; The slave_dbg() macro expands to netdev_dbg(bond_dev, "(slave %s): " fmt, (slave_dev)->name, ...) which unconditionally dereferences slave_dev->name before the NULL check is performed. This results in a NULL pointer dereference kernel oops when a user calls bonding ioctl (e.g. SIOCBONDENSLAVE, SIOCBONDRELEASE, etc.) with a non-existent slave interface name. This is reachable from userspace via the bonding ioctl interface with CAP_NET_ADMIN capability, making it a potential local denial-of-service vector. Fix by moving the slave_dbg() call after the NULL check. Fixes: e2a7420df2e0 ("bonding/main: convert to using slave printk macros") Cc: stable@vger.kernel.org # v5.2+ Signed-off-by: ZhaoJinming Link: https://patch.msgid.link/20260601085649.4029067-1-zhaojinming@uniontech.com Signed-off-by: Paolo Abeni --- drivers/net/bonding/bond_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 82e779f7916b..8e75453ce0ef 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -4621,11 +4621,11 @@ static int bond_do_ioctl(struct net_device *bond_dev, struct ifreq *ifr, int cmd slave_dev = __dev_get_by_name(net, ifr->ifr_slave); - slave_dbg(bond_dev, slave_dev, "slave_dev=%p:\n", slave_dev); - if (!slave_dev) return -ENODEV; + slave_dbg(bond_dev, slave_dev, "slave_dev=%p:\n", slave_dev); + switch (cmd) { case SIOCBONDENSLAVE: res = bond_enslave(bond_dev, slave_dev, NULL); From 4790af1cc2e8871fb31f28c66e42b9a949a23992 Mon Sep 17 00:00:00 2001 From: Chancel Liu Date: Mon, 1 Jun 2026 17:33:27 +0900 Subject: [PATCH 263/329] ASoC: fsl_sai: Fix 32 slots TDM broken by integer shift UB in xMR write When configuring 32 slots TDM (channels == slots == 32), the xMR (Mask Register) write used: ~0UL - ((1 << min(channels, slots)) - 1) The literal "1" is a signed 32-bit int. Shifting it by 32 positions is undefined behaviour which may set this register to 0xFFFFFFFF, masking all 32 slots. Use GENMASK_U32() macro instead. For 32 slots this produces a zero mask: ~GENMASK_U32(31, 0) = ~0xFFFFFFFF = 0x00000000 Behaviour for fewer than 32 slots is unchanged. Fixes: 770f58d7d2c5 ("ASoC: fsl_sai: Support multiple data channel enable bits") Cc: stable@vger.kernel.org Signed-off-by: Chancel Liu Reviewed-by: Shengjiu Wang Link: https://patch.msgid.link/20260601083327.1535185-1-chancel.liu@oss.nxp.com Signed-off-by: Mark Brown --- sound/soc/fsl/fsl_sai.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/fsl/fsl_sai.c b/sound/soc/fsl/fsl_sai.c index e364552c1f47..78e953cd3e9f 100644 --- a/sound/soc/fsl/fsl_sai.c +++ b/sound/soc/fsl/fsl_sai.c @@ -793,7 +793,7 @@ static int fsl_sai_hw_params(struct snd_pcm_substream *substream, FSL_SAI_CR4_FSD_MSTR, FSL_SAI_CR4_FSD_MSTR); regmap_write(sai->regmap, FSL_SAI_xMR(tx), - ~0UL - ((1 << min(channels, slots)) - 1)); + ~GENMASK_U32(min(channels, slots) - 1, 0)); return 0; } From cad530a8371124da988608f11df30b7dd82391ef Mon Sep 17 00:00:00 2001 From: David Glushkov Date: Sun, 31 May 2026 23:45:12 +0200 Subject: [PATCH 264/329] ASoC: amd: yc: Add MSI Raider A18 HX A9WJG to quirk table The MSI Raider A18 HX A9WJG has an internal digital microphone connected through AMD ACP6x, but this machine does not expose the AcpDmicConnected ACPI property, so acp_yc_mach does not bind. Add a DMI quirk for this model. This was tested on an MSI Raider A18 HX A9WJG with board MS-182L, BIOS E182LAMS.31A, AMD ACP6x rev 0x62, and Realtek ALC274. After applying the quirk, the internal microphone appears as an acp6x DMIC capture device and records correctly. Signed-off-by: David Glushkov Link: https://patch.msgid.link/20260531214512.170716-1-david.glushkov@sntiq.com Signed-off-by: Mark Brown --- sound/soc/amd/yc/acp6x-mach.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/amd/yc/acp6x-mach.c b/sound/soc/amd/yc/acp6x-mach.c index a93a7fd4877e..b426cda529a9 100644 --- a/sound/soc/amd/yc/acp6x-mach.c +++ b/sound/soc/amd/yc/acp6x-mach.c @@ -521,6 +521,13 @@ static const struct dmi_system_id yc_acp_quirk_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "Vector A16 HX A8WHG"), } }, + { + .driver_data = &acp6x_card, + .matches = { + DMI_MATCH(DMI_BOARD_VENDOR, "Micro-Star International Co., Ltd."), + DMI_MATCH(DMI_PRODUCT_NAME, "Raider A18 HX A9WJG"), + } + }, { .driver_data = &acp6x_card, .matches = { From d2dcd85f9e09fe3566d4cdcd357856a42ac73f93 Mon Sep 17 00:00:00 2001 From: Aleksander Pshenitsyn Date: Sun, 31 May 2026 13:11:59 +0300 Subject: [PATCH 265/329] ASoC: amd: acp70: add standalone RT721 SoundWire machine The ASUS Vivobook 18 M1807GA (AMD ACP7.x, PCI 1022:15e2, subsystem 1043:3531) exposes a single Realtek RT721 SDCA codec on SoundWire link 1. The BIOS reports the ACP audio config flag as 0 (SoundWire mode), so snd_pci_ps claims the device, brings up the SoundWire managers and enumerates the RT721 peripheral (sdw:0:1:025d:0721:01); the rt721-sdca codec driver binds successfully. No sound card is created, however: acp63_sdw_machine_select() walks snd_soc_acpi_amd_acp70_sdw_machines[] and finds no entry whose declared SoundWire peripherals are all present on the bus. The only existing RT721 entry, acp70_rt721_l1u0_tas2783x2_l1u8b, additionally requires two TAS2783 amplifiers and deliberately exposes the RT721 as jack + DMIC only. This M1807GA variant has no external amplifiers - the RT721's internal AIF2 amplifier path drives the speakers - so that entry never matches and no machine device is registered. Add a standalone RT721 machine entry for link 1 exposing all three RT721 endpoints (jack/AIF1, speaker amplifier/AIF2, DMIC/AIF3), mirroring the standalone RT722 configuration. Place it after the TAS2783 combo entry so platforms that do have the external amplifiers continue to match the more specific entry first. ACPI _ADR of the codec: 0x000130025D072101 (link_id=1 version=3 mfg_id=0x025d Realtek part_id=0x0721 class=0x01). Verified on the hardware: with the entry present the amd_sdw machine binds, an "amd-soundwire" card is registered exposing the rt721-sdca AIF1 (SimpleJack) and AIF2 (SmartAmp) PCM devices, and audio plays out of the built-in speakers. Link: https://bugzilla.kernel.org/show_bug.cgi?id=221282 Signed-off-by: Aleksander Pshenitsyn Link: https://patch.msgid.link/20260531101159.14241-1-brains.fatman@gmail.com Signed-off-by: Mark Brown --- sound/soc/amd/acp/amd-acp70-acpi-match.c | 44 ++++++++++++++++++++++++ 1 file changed, 44 insertions(+) diff --git a/sound/soc/amd/acp/amd-acp70-acpi-match.c b/sound/soc/amd/acp/amd-acp70-acpi-match.c index 1ae43df5da6c..18f2918d4ada 100644 --- a/sound/soc/amd/acp/amd-acp70-acpi-match.c +++ b/sound/soc/amd/acp/amd-acp70-acpi-match.c @@ -619,6 +619,45 @@ static const struct snd_soc_acpi_link_adr acp70_rt721_l1u0_tas2783x2_l1u8b[] = { {} }; +static const struct snd_soc_acpi_endpoint rt721_endpoints[] = { + { /* Jack Playback/Capture Endpoint (AIF1) */ + .num = 0, + .aggregated = 0, + .group_position = 0, + .group_id = 0, + }, + { /* Speaker Amplifier Endpoint (AIF2, internal amp) */ + .num = 1, + .aggregated = 0, + .group_position = 0, + .group_id = 0, + }, + { /* DMIC Capture Endpoint (AIF3) */ + .num = 2, + .aggregated = 0, + .group_position = 0, + .group_id = 0, + }, +}; + +static const struct snd_soc_acpi_adr_device rt721_1_single_adr[] = { + { + .adr = 0x000130025D072101ull, + .num_endpoints = ARRAY_SIZE(rt721_endpoints), + .endpoints = rt721_endpoints, + .name_prefix = "rt721" + } +}; + +static const struct snd_soc_acpi_link_adr acp70_rt721_only[] = { + { + .mask = BIT(1), + .num_adr = ARRAY_SIZE(rt721_1_single_adr), + .adr_d = rt721_1_single_adr, + }, + {} +}; + struct snd_soc_acpi_mach snd_soc_acpi_amd_acp70_sdw_machines[] = { { .link_mask = BIT(0) | BIT(1), @@ -711,6 +750,11 @@ struct snd_soc_acpi_mach snd_soc_acpi_amd_acp70_sdw_machines[] = { .links = acp70_rt721_l1u0_tas2783x2_l1u8b, .drv_name = "amd_sdw", }, + { + .link_mask = BIT(1), + .links = acp70_rt721_only, + .drv_name = "amd_sdw", + }, {}, }; EXPORT_SYMBOL(snd_soc_acpi_amd_acp70_sdw_machines); From c05fa14db43ebef3bd862ca9d073981c0358b3f0 Mon Sep 17 00:00:00 2001 From: Raf Dickson Date: Tue, 26 May 2026 10:43:56 +0000 Subject: [PATCH 266/329] vsock/vmci: fix sk_ack_backlog leak on failed handshake When vmci_transport_recv_connecting_server() returns an error, vmci_transport_recv_listen() calls vsock_remove_pending() but never calls sk_acceptq_removed(). This leaves sk_ack_backlog incremented permanently. Repeated handshake failures (malformed packets, queue pair alloc failure, event subscribe failure) cause sk_ack_backlog to climb toward sk_max_ack_backlog. Once it reaches the limit the listener permanently refuses all new connections with -ECONNREFUSED, a silent denial of service requiring a process restart to recover. The two existing sk_acceptq_removed() calls in af_vsock.c do not cover this path: line 764 checks vsock_is_pending() which returns false after vsock_remove_pending(), and line 1889 is only reached on successful accept(). Fix by balancing sk_acceptq_added() with sk_acceptq_removed() on the error path. Fixes: d021c344051a ("VSOCK: Introduce VM Sockets") Cc: stable@vger.kernel.org Signed-off-by: Raf Dickson Acked-by: Stefano Garzarella Link: https://patch.msgid.link/20260526104356.469928-1-rafdog35@gmail.com Signed-off-by: Paolo Abeni --- net/vmw_vsock/vmci_transport.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/net/vmw_vsock/vmci_transport.c b/net/vmw_vsock/vmci_transport.c index 5c1ecd5bfdbc..91516488a742 100644 --- a/net/vmw_vsock/vmci_transport.c +++ b/net/vmw_vsock/vmci_transport.c @@ -980,8 +980,10 @@ static int vmci_transport_recv_listen(struct sock *sk, err = -EINVAL; } - if (err < 0) + if (err < 0) { vsock_remove_pending(sk, pending); + sk_acceptq_removed(sk); + } release_sock(pending); vmci_transport_release_pending(pending); From d486b4934a8e504376b85cdb3766f306d57aff5b Mon Sep 17 00:00:00 2001 From: Amit Matityahu Date: Wed, 3 Jun 2026 17:01:39 +0000 Subject: [PATCH 267/329] timers/migration: Fix livelock in tmigr_handle_remote_up() tmigr_handle_remote_cpu() skips timer_expire_remote() when cpu == smp_processor_id(), assuming the local softirq path already handled this CPU's timers. This assumption is wrong because jiffies can advance after the handling of the CPU's global timers in run_timer_base(BASE_GLOBAL) and before tmigr_handle_remote() evaluates the expiry times. As a consequence a timer which expires after the CPU local timer wheel advanced and becomes expired in the remote handling is ignored and the callback is never invoked and removed from the timer wheel. What's worse is that fetch_next_timer_interrupt_remote() keeps reporting it as expired, and the event is re-queued with expires == now on each iteration. The goto-again loop spins indefinitely. Fix this by calling timer_expire_remote() unconditionally. That's minimal overhead for the common case as __run_timer_base() returns immediately if there is nothing to expire in the local wheel. [ tglx: Amend change log and add a comment ] Fixes: 7ee988770326 ("timers: Implement the hierarchical pull model") Reported-by: Alon Kariv Signed-off-by: Amit Matityahu Signed-off-by: Thomas Gleixner Cc: stable@vger.kernel.org Link: https://patch.msgid.link/20260603170139.33628-1-amitmat@amazon.com --- kernel/time/timer_migration.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/kernel/time/timer_migration.c b/kernel/time/timer_migration.c index 1d0d3a4058d5..52c15affdbff 100644 --- a/kernel/time/timer_migration.c +++ b/kernel/time/timer_migration.c @@ -978,8 +978,12 @@ static void tmigr_handle_remote_cpu(unsigned int cpu, u64 now, /* Drop the lock to allow the remote CPU to exit idle */ raw_spin_unlock_irq(&tmc->lock); - if (cpu != smp_processor_id()) - timer_expire_remote(cpu); + /* + * This can't exclude the local CPU because jiffies might have advanced + * after the timer softirq invoked run_timer_base(BASE_GLOBAL) and the + * point where the jiffies snapshot @jif was taken in tmigr_handle_remote(). + */ + timer_expire_remote(cpu); /* * Lock ordering needs to be preserved - timer_base locks before tmigr From fa7c84726dc217ce0c183926ef9411636c7a2213 Mon Sep 17 00:00:00 2001 From: Tangudu Tilak Tirumalesh Date: Wed, 3 Jun 2026 12:22:15 +0530 Subject: [PATCH 268/329] Revert "drm/xe: Skip exec queue schedule toggle if queue is idle during suspend" This reverts commit 8533051ce92015e9cc6f75e0d52119b9d91610b6. The idle-skip optimization bypasses GuC suspend, so the GPU may not perform the context switch that flushes TLB entries for invalidated userptr VMAs. In LR/preempt-fence VM mode, this can lead to missed TLB invalidation and page faults during userptr invalidation tests. Restore unconditional schedule toggling on suspend so the context-switch TLB flush is always performed. This optimization will be reintroduced with a fix that does not skip suspend in LR/preempt-fence VM mode. Fixes: 8533051ce920 ("drm/xe: Skip exec queue schedule toggle if queue is idle during suspend") Cc: stable@vger.kernel.org # v7.0+ Suggested-by: Thomas Hellstrom Signed-off-by: Tangudu Tilak Tirumalesh Reviewed-by: Thomas Hellstrom Signed-off-by: Daniele Ceraolo Spurio Link: https://patch.msgid.link/20260603065217.3131066-2-tilak.tirumalesh.tangudu@intel.com (cherry picked from commit 6a1e7934d9a6cf46aecae00a99c2603d1295e170) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_exec_queue.h | 17 -------- drivers/gpu/drm/xe/xe_guc_submit.c | 55 ++----------------------- drivers/gpu/drm/xe/xe_hw_engine_group.c | 10 +---- 3 files changed, 5 insertions(+), 77 deletions(-) diff --git a/drivers/gpu/drm/xe/xe_exec_queue.h b/drivers/gpu/drm/xe/xe_exec_queue.h index a82d99bd77bc..0225426c57b0 100644 --- a/drivers/gpu/drm/xe/xe_exec_queue.h +++ b/drivers/gpu/drm/xe/xe_exec_queue.h @@ -162,21 +162,4 @@ int xe_exec_queue_contexts_hwsp_rebase(struct xe_exec_queue *q, void *scratch); struct xe_lrc *xe_exec_queue_lrc(struct xe_exec_queue *q); struct xe_lrc *xe_exec_queue_get_lrc(struct xe_exec_queue *q, u16 idx); -/** - * xe_exec_queue_idle_skip_suspend() - Can exec queue skip suspend - * @q: The exec_queue - * - * If an exec queue is not parallel and is idle, the suspend steps can be - * skipped in the submission backend immediatley signaling the suspend fence. - * Parallel queues cannot skip this step due to limitations in the submission - * backend. - * - * Return: True if exec queue is idle and can skip suspend steps, False - * otherwise - */ -static inline bool xe_exec_queue_idle_skip_suspend(struct xe_exec_queue *q) -{ - return !xe_exec_queue_is_parallel(q) && xe_exec_queue_is_idle(q); -} - #endif diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 912182dc7704..3db627b56e11 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -71,7 +71,6 @@ exec_queue_to_guc(struct xe_exec_queue *q) #define EXEC_QUEUE_STATE_WEDGED (1 << 8) #define EXEC_QUEUE_STATE_BANNED (1 << 9) #define EXEC_QUEUE_STATE_PENDING_RESUME (1 << 10) -#define EXEC_QUEUE_STATE_IDLE_SKIP_SUSPEND (1 << 11) static bool exec_queue_registered(struct xe_exec_queue *q) { @@ -218,21 +217,6 @@ static void clear_exec_queue_pending_resume(struct xe_exec_queue *q) atomic_and(~EXEC_QUEUE_STATE_PENDING_RESUME, &q->guc->state); } -static bool exec_queue_idle_skip_suspend(struct xe_exec_queue *q) -{ - return atomic_read(&q->guc->state) & EXEC_QUEUE_STATE_IDLE_SKIP_SUSPEND; -} - -static void set_exec_queue_idle_skip_suspend(struct xe_exec_queue *q) -{ - atomic_or(EXEC_QUEUE_STATE_IDLE_SKIP_SUSPEND, &q->guc->state); -} - -static void clear_exec_queue_idle_skip_suspend(struct xe_exec_queue *q) -{ - atomic_and(~EXEC_QUEUE_STATE_IDLE_SKIP_SUSPEND, &q->guc->state); -} - static bool exec_queue_killed_or_banned_or_wedged(struct xe_exec_queue *q) { return (atomic_read(&q->guc->state) & @@ -1153,7 +1137,7 @@ static void submit_exec_queue(struct xe_exec_queue *q, struct xe_sched_job *job) if (!job->restore_replay || job->last_replay) { if (xe_exec_queue_is_parallel(q)) wq_item_append(q); - else if (!exec_queue_idle_skip_suspend(q)) + else xe_lrc_set_ring_tail(lrc, lrc->ring.tail); job->last_replay = false; } @@ -1810,10 +1794,9 @@ static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg) { struct xe_exec_queue *q = msg->private_data; struct xe_guc *guc = exec_queue_to_guc(q); - bool idle_skip_suspend = xe_exec_queue_idle_skip_suspend(q); - if (!idle_skip_suspend && guc_exec_queue_allowed_to_change_state(q) && - !exec_queue_suspended(q) && exec_queue_enabled(q)) { + if (guc_exec_queue_allowed_to_change_state(q) && !exec_queue_suspended(q) && + exec_queue_enabled(q)) { wait_event(guc->ct.wq, vf_recovery(guc) || ((q->guc->resume_time != RESUME_PENDING || xe_guc_read_stopped(guc)) && !exec_queue_pending_disable(q))); @@ -1832,33 +1815,11 @@ static void __guc_exec_queue_process_msg_suspend(struct xe_sched_msg *msg) disable_scheduling(q, false); } } else if (q->guc->suspend_pending) { - if (idle_skip_suspend) - set_exec_queue_idle_skip_suspend(q); set_exec_queue_suspended(q); suspend_fence_signal(q); } } -static void sched_context(struct xe_exec_queue *q) -{ - struct xe_guc *guc = exec_queue_to_guc(q); - struct xe_lrc *lrc = q->lrc[0]; - u32 action[] = { - XE_GUC_ACTION_SCHED_CONTEXT, - q->guc->id, - }; - - xe_gt_assert(guc_to_gt(guc), !xe_exec_queue_is_parallel(q)); - xe_gt_assert(guc_to_gt(guc), !exec_queue_destroyed(q)); - xe_gt_assert(guc_to_gt(guc), exec_queue_registered(q)); - xe_gt_assert(guc_to_gt(guc), !exec_queue_pending_disable(q)); - - trace_xe_exec_queue_submit(q); - - xe_lrc_set_ring_tail(lrc, lrc->ring.tail); - xe_guc_ct_send(&guc->ct, action, ARRAY_SIZE(action), 0, 0); -} - static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg) { struct xe_exec_queue *q = msg->private_data; @@ -1866,22 +1827,12 @@ static void __guc_exec_queue_process_msg_resume(struct xe_sched_msg *msg) if (guc_exec_queue_allowed_to_change_state(q)) { clear_exec_queue_suspended(q); if (!exec_queue_enabled(q)) { - if (exec_queue_idle_skip_suspend(q)) { - struct xe_lrc *lrc = q->lrc[0]; - - clear_exec_queue_idle_skip_suspend(q); - xe_lrc_set_ring_tail(lrc, lrc->ring.tail); - } q->guc->resume_time = RESUME_PENDING; set_exec_queue_pending_resume(q); enable_scheduling(q); - } else if (exec_queue_idle_skip_suspend(q)) { - clear_exec_queue_idle_skip_suspend(q); - sched_context(q); } } else { clear_exec_queue_suspended(q); - clear_exec_queue_idle_skip_suspend(q); } } diff --git a/drivers/gpu/drm/xe/xe_hw_engine_group.c b/drivers/gpu/drm/xe/xe_hw_engine_group.c index 4c2b113364d3..02cf32ae5aa9 100644 --- a/drivers/gpu/drm/xe/xe_hw_engine_group.c +++ b/drivers/gpu/drm/xe/xe_hw_engine_group.c @@ -208,21 +208,15 @@ static int xe_hw_engine_group_suspend_faulting_lr_jobs(struct xe_hw_engine_group lockdep_assert_held_write(&group->mode_sem); list_for_each_entry(q, &group->exec_queue_list, hw_engine_group_link) { - bool idle_skip_suspend; if (!xe_vm_in_fault_mode(q->vm)) continue; - idle_skip_suspend = xe_exec_queue_idle_skip_suspend(q); - if (!idle_skip_suspend && has_deps) + if (has_deps) return -EAGAIN; xe_gt_stats_incr(q->gt, XE_GT_STATS_ID_HW_ENGINE_GROUP_SUSPEND_LR_QUEUE_COUNT, 1); - if (idle_skip_suspend) - xe_gt_stats_incr(q->gt, - XE_GT_STATS_ID_HW_ENGINE_GROUP_SKIP_LR_QUEUE_COUNT, 1); - - need_resume |= !idle_skip_suspend; + need_resume = true; q->ops->suspend(q); gt = q->gt; } From 54f2a0442a30fe7a0f6bc8345e81f8b2db8effbd Mon Sep 17 00:00:00 2001 From: Tangudu Tilak Tirumalesh Date: Wed, 3 Jun 2026 12:22:16 +0530 Subject: [PATCH 269/329] drm/xe: Clear pending_disable before signaling suspend fence In the schedule-disable done path for suspend, we signal the suspend fence before clearing pending_disable. That wakeup can let suspend_wait complete and resume be queued immediately. The resume path may then reach enable_scheduling() while pending_disable is still set and hit the !exec_queue_pending_disable(q) assertion. Fix this by clearing pending_disable before signaling the suspend fence, so any resumed transition observes a consistent state. Fixes: 87651f31ae4e ("drm/xe/guc_submit: fix race around suspend_pending") Cc: stable@vger.kernel.org # v7.0+ Signed-off-by: Tangudu Tilak Tirumalesh Reviewed-by: Thomas Hellstrom Signed-off-by: Daniele Ceraolo Spurio Link: https://patch.msgid.link/20260603065217.3131066-3-tilak.tirumalesh.tangudu@intel.com (cherry picked from commit 4b1ae138b0e103d753773956a84eebc2edbf62c4) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_submit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 3db627b56e11..3493dd533d6c 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -2804,8 +2804,8 @@ static void handle_sched_done(struct xe_guc *guc, struct xe_exec_queue *q, xe_gt_assert(guc_to_gt(guc), exec_queue_pending_disable(q)); if (q->guc->suspend_pending) { - suspend_fence_signal(q); clear_exec_queue_pending_disable(q); + suspend_fence_signal(q); } else { if (exec_queue_banned(q)) { smp_wmb(); From ec4cbdd163f9bb2a2bd44eb93ecf4a2fa0e912a9 Mon Sep 17 00:00:00 2001 From: Niranjana Vishwanathapura Date: Wed, 3 Jun 2026 16:39:47 -0700 Subject: [PATCH 270/329] drm/xe/multi_queue: skip submit when primary queue is suspended Return early in submit path when the multi-queue primary exec queue is suspended to avoid submitting while suspended. v2: Remove idle_skip_suspend fix as that feature is being reverted here https://patchwork.freedesktop.org/series/167262/ Fixes: bc5775c59258 ("drm/xe/multi_queue: Add GuC interface for multi queue support") Cc: stable@vger.kernel.org # v7.0+ Assisted-by: GitHub-Copilot:claude-sonnet-4.6 Reviewed-by: Daniele Ceraolo Spurio Signed-off-by: Niranjana Vishwanathapura Link: https://patch.msgid.link/20260603233946.863663-2-niranjana.vishwanathapura@intel.com (cherry picked from commit b7fb55cc3364ca128cfff9d50649ffd4327cd01e) Signed-off-by: Rodrigo Vivi --- drivers/gpu/drm/xe/xe_guc_submit.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c index 3493dd533d6c..a4a8f0d41fe8 100644 --- a/drivers/gpu/drm/xe/xe_guc_submit.c +++ b/drivers/gpu/drm/xe/xe_guc_submit.c @@ -1147,9 +1147,12 @@ static void submit_exec_queue(struct xe_exec_queue *q, struct xe_sched_job *job) /* * All queues in a multi-queue group will use the primary queue - * of the group to interface with GuC. + * of the group to interface with GuC. If primay is suspended, + * just return. Jobs will get scheduled once primary is resumed. */ q = xe_exec_queue_multi_queue_primary(q); + if (exec_queue_suspended(q)) + return; if (!exec_queue_enabled(q) && !exec_queue_suspended(q)) { action[len++] = XE_GUC_ACTION_SCHED_CONTEXT_MODE_SET; From 6bf7e2affc6e62da7add393d7f352d4040f5bc27 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Ma=C3=ADra=20Canal?= Date: Sun, 31 May 2026 17:18:55 -0300 Subject: [PATCH 271/329] drm/v3d: Fix global performance monitor reference counting MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit In the SET_GLOBAL ioctl, v3d_perfmon_find() bumps the reference count on the perfmon it returns, but v3d_perfmon_set_global_ioctl() and v3d_perfmon_delete() fail to release that reference on several paths: 1. v3d_perfmon_set_global_ioctl() leaks the reference on its error paths. 2. CLEAR_GLOBAL leaks both the find reference and the reference previously stashed in v3d->global_perfmon by the SET_GLOBAL ioctl that configured it. 3. Destroying a perfmon that is the current global perfmon leaks the reference stashed by the SET_GLOBAL ioctl. Release each of these references explicitly. Cc: stable@vger.kernel.org Fixes: c6eabbab359c ("drm/v3d: Add DRM_IOCTL_V3D_PERFMON_SET_GLOBAL") Reviewed-by: Iago Toral Quiroga Link: https://patch.msgid.link/20260531-v3d-perfmon-lifetime-v2-1-60ed4485a203@igalia.com Signed-off-by: Maíra Canal --- drivers/gpu/drm/v3d/v3d_perfmon.c | 24 +++++++++++++++++++----- 1 file changed, 19 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/v3d/v3d_perfmon.c b/drivers/gpu/drm/v3d/v3d_perfmon.c index 8e0249580bba..ecfd446ff75f 100644 --- a/drivers/gpu/drm/v3d/v3d_perfmon.c +++ b/drivers/gpu/drm/v3d/v3d_perfmon.c @@ -309,8 +309,11 @@ static void v3d_perfmon_delete(struct v3d_file_priv *v3d_priv, if (perfmon == v3d->active_perfmon) v3d_perfmon_stop(v3d, perfmon, false); - /* If the global perfmon is being destroyed, set it to NULL */ - cmpxchg(&v3d->global_perfmon, perfmon, NULL); + /* If the global perfmon is being destroyed, clean it and release + * the reference stashed in v3d_perfmon_set_global_ioctl(). + */ + if (cmpxchg(&v3d->global_perfmon, perfmon, NULL) == perfmon) + v3d_perfmon_put(perfmon); v3d_perfmon_put(perfmon); } @@ -461,16 +464,27 @@ int v3d_perfmon_set_global_ioctl(struct drm_device *dev, void *data, /* If the request is to clear the global performance monitor */ if (req->flags & DRM_V3D_PERFMON_CLEAR_GLOBAL) { - if (!v3d->global_perfmon) + struct v3d_perfmon *old; + + /* DRM_V3D_PERFMON_CLEAR_GLOBAL doesn't check if + * v3d->global_perfmon == perfmon. Therefore, there + * is no need to keep perfmon's reference. + */ + v3d_perfmon_put(perfmon); + + old = xchg(&v3d->global_perfmon, NULL); + if (!old) return -EINVAL; - xchg(&v3d->global_perfmon, NULL); + v3d_perfmon_put(old); return 0; } - if (cmpxchg(&v3d->global_perfmon, NULL, perfmon)) + if (cmpxchg(&v3d->global_perfmon, NULL, perfmon)) { + v3d_perfmon_put(perfmon); return -EBUSY; + } return 0; } From 08904765bb941f98306ae6841c33cfd299343faf Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Thu, 4 Jun 2026 14:09:45 +0200 Subject: [PATCH 272/329] tools/rv: Ensure monitor name and desc are NUL-terminated ikm_fill_monitor_definition() copies monitor name and description with strncpy(), but does not guarantee NUL termination when source strings are equal to or longer than the destination buffers. Clamp copies to sizeof(dst) - 1 and explicitly append '\0' for both fields to keep them safe for later string operations. Suggested-by: unknownbbqrx Fixes: 6d60f89691fc9 ("tools/rv: Add in-kernel monitor interface") Link: https://lore.kernel.org/r/20260604120946.90302-2-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rv/src/in_kernel.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/tools/verification/rv/src/in_kernel.c b/tools/verification/rv/src/in_kernel.c index 4bb746ea6e17..d324538249d3 100644 --- a/tools/verification/rv/src/in_kernel.c +++ b/tools/verification/rv/src/in_kernel.c @@ -215,10 +215,11 @@ static int ikm_fill_monitor_definition(char *name, struct monitor *ikm, char *co return -1; } - strncpy(ikm->name, nested_name, MAX_DA_NAME_LEN); + strncpy(ikm->name, nested_name, sizeof(ikm->name) - 1); + ikm->name[sizeof(ikm->name) - 1] = '\0'; ikm->enabled = enabled; - strncpy(ikm->desc, desc, MAX_DESCRIPTION); - + strncpy(ikm->desc, desc, sizeof(ikm->desc) - 1); + ikm->desc[sizeof(ikm->desc) - 1] = '\0'; free(desc); return 0; From a963fbf3166f2e178ac38b6c3c186a0c98092fb9 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Thu, 14 May 2026 17:20:42 +0200 Subject: [PATCH 273/329] tools/rv: Fix substring match bug in monitor name search __ikm_find_monitor_name() relies on strstr() to find a monitor by name, which fails if the target monitor is a substring of a previously listed monitor. Fix it by tokenizing the available_monitors file and matching full tokens instead. Fixes: eba321a16fc6 ("tools/rv: Add support for nested monitors") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260514152055.229162-2-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rv/src/in_kernel.c | 50 ++++++++++++++------------- 1 file changed, 26 insertions(+), 24 deletions(-) diff --git a/tools/verification/rv/src/in_kernel.c b/tools/verification/rv/src/in_kernel.c index d324538249d3..95eac9ab1484 100644 --- a/tools/verification/rv/src/in_kernel.c +++ b/tools/verification/rv/src/in_kernel.c @@ -58,38 +58,40 @@ static int __ikm_read_enable(char *monitor_name) */ static int __ikm_find_monitor_name(char *monitor_name, char *out_name) { - char *available_monitors, container[MAX_DA_NAME_LEN+1], *cursor, *end; - int retval = 1; + char *available_monitors, *cursor, *line; + int len = strlen(monitor_name); + int found = 0; available_monitors = tracefs_instance_file_read(NULL, "rv/available_monitors", NULL); if (!available_monitors) return -1; - cursor = strstr(available_monitors, monitor_name); - if (!cursor) { - retval = 0; - goto out_free; + config_is_container = 0; + cursor = available_monitors; + while ((line = strsep(&cursor, "\n"))) { + char *colon = strchr(line, ':'); + + if (strcmp(line, monitor_name) && (!colon || strcmp(colon + 1, monitor_name))) + continue; + + strncpy(out_name, line, 2 * MAX_DA_NAME_LEN); + out_name[2 * MAX_DA_NAME_LEN - 1] = '\0'; + + if (colon) { + out_name[colon - line] = '/'; + } else { + /* If there are children, they are on the next line. */ + line = strsep(&cursor, "\n"); + if (line && !strncmp(line, monitor_name, len) && line[len] == ':') + config_is_container = 1; + } + + found = 1; + break; } - for (; cursor > available_monitors; cursor--) - if (*(cursor-1) == '\n') - break; - end = strstr(cursor, "\n"); - memcpy(out_name, cursor, end-cursor); - out_name[end-cursor] = '\0'; - - cursor = strstr(out_name, ":"); - if (cursor) - *cursor = '/'; - else { - sprintf(container, "%s:", monitor_name); - if (strstr(available_monitors, container)) - config_is_container = 1; - } - -out_free: free(available_monitors); - return retval; + return found; } /* From ba0247c5aa3fcb2890a92a97a88c70fe5ce704a6 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Thu, 14 May 2026 17:20:43 +0200 Subject: [PATCH 274/329] tools/rv: Fix substring match when listing container monitors When listing monitors within a specific container (rv list ), the tool incorrectly matched monitors if the requested container name was only a prefix of the actual container (e.g., 'rv list sche' would incorrectly list monitors from 'sched:'). Fix this by ensuring the container name is an exact match and is immediately followed by the ':' separator. Fixes: eba321a16fc6 ("tools/rv: Add support for nested monitors") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260514152055.229162-3-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rv/src/in_kernel.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/tools/verification/rv/src/in_kernel.c b/tools/verification/rv/src/in_kernel.c index 95eac9ab1484..e4f35940374f 100644 --- a/tools/verification/rv/src/in_kernel.c +++ b/tools/verification/rv/src/in_kernel.c @@ -193,8 +193,12 @@ static int ikm_fill_monitor_definition(char *name, struct monitor *ikm, char *co nested_name = strstr(name, ":"); if (nested_name) { /* it belongs in container if it starts with "container:" */ - if (container && strstr(name, container) != name) - return 1; + if (container) { + int len = strlen(container); + + if (strncmp(name, container, len) || name[len] != ':') + return 1; + } *nested_name = '/'; ++nested_name; ikm->nested = 1; From 33ec2269a4155cad7e9e42c92327dcaa9aee59a7 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Thu, 14 May 2026 17:20:45 +0200 Subject: [PATCH 275/329] tools/rv: Fix cleanup after failed trace setup Currently if ikm_setup_trace_instance() fails, the tool returns without any cleanup, if rv was called with both -t and -r, this means the reactor is not going to be cleared. Jump to the cleanup label to restore the reactor if necessary. Fixes: 6d60f89691fc9 ("tools/rv: Add in-kernel monitor interface") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260514152055.229162-5-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rv/src/in_kernel.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/verification/rv/src/in_kernel.c b/tools/verification/rv/src/in_kernel.c index e4f35940374f..e6dea4040f8f 100644 --- a/tools/verification/rv/src/in_kernel.c +++ b/tools/verification/rv/src/in_kernel.c @@ -810,7 +810,7 @@ int ikm_run_monitor(char *monitor_name, int argc, char **argv) if (config_trace) { inst = ikm_setup_trace_instance(nested_name); if (!inst) - return -1; + goto out_free_instance; } retval = ikm_enable(full_name); From 85339442de941e4d7ff5d53f51ae1413905e45ec Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Mon, 1 Jun 2026 17:38:38 +0200 Subject: [PATCH 276/329] verification/rvgen: Fix suffix strip in dot2k __start_to_invariant_check() and __get_constraint_env() parse the environment variable's name from sources that have it padded with the monitor name. This is removed using rstrip(), which is not meant to strip a substring but rather a set of characters. Use removesuffix() to actually get rid of the trailing _. Fixes: a82adadb16894 ("verification/rvgen: Add support for Hybrid Automata") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260601153840.124372-12-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rvgen/rvgen/dot2k.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tools/verification/rvgen/rvgen/dot2k.py b/tools/verification/rvgen/rvgen/dot2k.py index e6f476b903b0..110cfd69e53a 100644 --- a/tools/verification/rvgen/rvgen/dot2k.py +++ b/tools/verification/rvgen/rvgen/dot2k.py @@ -215,14 +215,14 @@ class ha2k(dot2k): def __get_constraint_env(self, constr: str) -> str: """Extract the second argument from an ha_ function""" env = constr.split("(")[1].split()[1].rstrip(")").rstrip(",") - assert env.rstrip(f"_{self.name}") in self.envs + assert env.removesuffix(f"_{self.name}") in self.envs return env def __start_to_invariant_check(self, constr: str) -> str: # by default assume the timer has ns expiration env = self.__get_constraint_env(constr) clock_type = "ns" - if self.env_types.get(env.rstrip(f"_{self.name}")) == "j": + if self.env_types.get(env.removesuffix(f"_{self.name}")) == "j": clock_type = "jiffy" return f"return ha_check_invariant_{clock_type}(ha_mon, {env}, time_ns)" From 5f845ad706c0b394ae274e9a930044f78bef782e Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Thu, 14 May 2026 17:20:47 +0200 Subject: [PATCH 277/329] verification/rvgen: Fix options shared among commands After rvgen was refactored to use subparsers, the common options (-a and -D) were left in the main parser. This meant that they needed to be called /before/ the subcommand and using them without subcommand was allowed. This is not the original intent. rvgen -D "some description" container -n name Define the options as parent in the subparsers to allow them to be used from both subcommands together with other options. rvgen container -n name -D "some description" Fixes: 5270a0e3041c ("verification/dot2k: Replace is_container() hack with subparsers") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260514152055.229162-7-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rvgen/__main__.py | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/tools/verification/rvgen/__main__.py b/tools/verification/rvgen/__main__.py index 3be7f85fe37b..5c923dc10d0f 100644 --- a/tools/verification/rvgen/__main__.py +++ b/tools/verification/rvgen/__main__.py @@ -18,14 +18,16 @@ if __name__ == '__main__': import sys parser = argparse.ArgumentParser(description='Generate kernel rv monitor') - parser.add_argument("-D", "--description", dest="description", required=False) - parser.add_argument("-a", "--auto_patch", dest="auto_patch", + + parent_parser = argparse.ArgumentParser(add_help=False) + parent_parser.add_argument("-D", "--description", dest="description", required=False) + parent_parser.add_argument("-a", "--auto_patch", dest="auto_patch", action="store_true", required=False, help="Patch the kernel in place") subparsers = parser.add_subparsers(dest="subcmd", required=True) - monitor_parser = subparsers.add_parser("monitor") + monitor_parser = subparsers.add_parser("monitor", parents=[parent_parser]) monitor_parser.add_argument('-n', "--model_name", dest="model_name") monitor_parser.add_argument("-p", "--parent", dest="parent", required=False, help="Create a monitor nested to parent") @@ -36,7 +38,7 @@ if __name__ == '__main__': monitor_parser.add_argument('-t', "--monitor_type", dest="monitor_type", required=True, help=f"Available options: {', '.join(Monitor.monitor_types.keys())}") - container_parser = subparsers.add_parser("container") + container_parser = subparsers.add_parser("container", parents=[parent_parser]) container_parser.add_argument('-n', "--model_name", dest="model_name", required=True) params = parser.parse_args() From df996599cc69a9b74ff437c67751cf8a61f62e39 Mon Sep 17 00:00:00 2001 From: Gabriele Monaco Date: Thu, 14 May 2026 17:20:48 +0200 Subject: [PATCH 278/329] verification/rvgen: Fix ltl2k writing True as a literal The rvgen parser for LTL stores literal true values in the python representation (capitalised True), this doesn't build in C. The Literal class should already handle this case but ASTNode skips its strigification method and converts the value (true/false) directly. Fix by delegating ASTNode stringification to the Literal and Variable classes instead of bypassing them. Fixes: 97ffa4ce6ab32 ("verification/rvgen: Add support for linear temporal logic") Reviewed-by: Nam Cao Link: https://lore.kernel.org/r/20260514152055.229162-8-gmonaco@redhat.com Signed-off-by: Gabriele Monaco --- tools/verification/rvgen/rvgen/ltl2ba.py | 9 +++++---- 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/tools/verification/rvgen/rvgen/ltl2ba.py b/tools/verification/rvgen/rvgen/ltl2ba.py index 7f538598a868..016e7cf93bbb 100644 --- a/tools/verification/rvgen/rvgen/ltl2ba.py +++ b/tools/verification/rvgen/rvgen/ltl2ba.py @@ -122,10 +122,8 @@ class ASTNode: return self.op.expand(self, node, node_set) def __str__(self): - if isinstance(self.op, Literal): - return str(self.op.value) - if isinstance(self.op, Variable): - return self.op.name.lower() + if isinstance(self.op, (Literal, Variable)): + return str(self.op) return "val" + str(self.id) def normalize(self): @@ -382,6 +380,9 @@ class Variable: def __iter__(self): yield from () + def __str__(self): + return self.name.lower() + def negate(self): new = ASTNode(self) return NotOp(new) From 88fe2e3658726cb21ff2dcf9770bf672f9b9d31b Mon Sep 17 00:00:00 2001 From: Ji'an Zhou Date: Thu, 4 Jun 2026 14:25:59 +0000 Subject: [PATCH 279/329] ALSA: PCM: Fix wait queue list corruption in snd_pcm_drain() on linked streams snd_pcm_drain() uses init_waitqueue_entry which does not clear entry.prev/next, and add_wait_queue with a conditional remove_wait_queue that is skipped when to_check is no longer in the group after concurrent UNLINK. The orphaned wait entry remains on the unlinked substream sleep queue. On the next drain iteration, add_wait_queue adds the entry to a new queue while still linked on the old one, corrupting both lists. A subsequent wake_up dereferences NULL at the func pointer (mapped from the spinlock at offset 0 of the misinterpreted wait_queue_head_t), causing a kernel panic. Replace init_waitqueue_entry/add_wait_queue/conditional remove_wait_queue with init_wait_entry/prepare_to_wait/ finish_wait. init_wait_entry clears prev/next via INIT_LIST_HEAD on each iteration and sets autoremove_wake_function which auto-removes the entry on wake-up. finish_wait safely handles both the already-removed and still-queued cases. Fixes: 9b1dbd69ba6f ("ALSA: pcm: fix use-after-free on linked stream runtime in snd_pcm_drain") Signed-off-by: Ji'an Zhou Link: https://patch.msgid.link/20260604142559.3840881-1-eilaimemedsnaimel@gmail.com Signed-off-by: Takashi Iwai --- sound/core/pcm_native.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index a541bb235cfa..302643c1c192 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -2199,9 +2199,8 @@ static int snd_pcm_drain(struct snd_pcm_substream *substream, drain_no_period_wakeup = to_check->no_period_wakeup; drain_rate = to_check->rate; drain_bufsz = to_check->buffer_size; - init_waitqueue_entry(&wait, current); - set_current_state(TASK_INTERRUPTIBLE); - add_wait_queue(&to_check->sleep, &wait); + init_wait_entry(&wait, 0); + prepare_to_wait(&to_check->sleep, &wait, TASK_INTERRUPTIBLE); snd_pcm_stream_unlock_irq(substream); if (drain_no_period_wakeup) tout = MAX_SCHEDULE_TIMEOUT; @@ -2219,7 +2218,7 @@ static int snd_pcm_drain(struct snd_pcm_substream *substream, group = snd_pcm_stream_group_ref(substream); snd_pcm_group_for_each_entry(s, substream) { if (s->runtime == to_check) { - remove_wait_queue(&to_check->sleep, &wait); + finish_wait(&to_check->sleep, &wait); break; } } From d76b56b06b2734c6e0ff519599126919fa4f154b Mon Sep 17 00:00:00 2001 From: Lianqin Hu Date: Thu, 4 Jun 2026 12:40:28 +0000 Subject: [PATCH 280/329] ALSA: usb-audio: Add iface reset and delay quirk for AB13X USB Audio Setting up the interface when suspended/resumeing fail on this card. Adding a reset and delay quirk will eliminate this problem. usb 1-1: new full-speed USB device number 2 using xhci-hcd usb 1-1: New USB device found, idVendor=3c20, idProduct=3d21 usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3 usb 1-1: Product: AB13X USB Audio usb 1-1: Manufacturer: Generic usb 1-1: SerialNumber: 20210726905926 Signed-off-by: Lianqin Hu Signed-off-by: Takashi Iwai Link: https://patch.msgid.link/TYUPR06MB62174610061C213260E1A992D2102@TYUPR06MB6217.apcprd06.prod.outlook.com --- sound/usb/quirks.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c index 3d1b3523b020..f4e885f96e83 100644 --- a/sound/usb/quirks.c +++ b/sound/usb/quirks.c @@ -2487,6 +2487,8 @@ static const struct usb_audio_quirk_flags_table quirk_flags_table[] = { QUIRK_FLAG_MIXER_PLAYBACK_MIN_MUTE), DEVICE_FLG(0x3443, 0x930d, /* NexiGo N930W 60fps Webcam */ QUIRK_FLAG_GET_SAMPLE_RATE | QUIRK_FLAG_MIC_RES_16), + DEVICE_FLG(0x3c20, 0x3d21, /* AB13X USB Audio */ + QUIRK_FLAG_FORCE_IFACE_RESET | QUIRK_FLAG_IFACE_DELAY), DEVICE_FLG(0x413c, 0xa506, /* Dell AE515 sound bar */ QUIRK_FLAG_GET_SAMPLE_RATE), DEVICE_FLG(0x534d, 0x0021, /* MacroSilicon MS2100/MS2106 */ From 899ee91156e57784090c5565e4f31bd7dbffbc5a Mon Sep 17 00:00:00 2001 From: Rajat Gupta Date: Sun, 31 May 2026 08:32:21 -0400 Subject: [PATCH 281/329] net/sched: fix pedit partial COW leading to page cache corruption MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit tcf_pedit_act() computes the COW range for skb_ensure_writable() once before the key loop using tcfp_off_max_hint, but the hint does not account for the runtime header offset added by typed keys. This can leave part of the write region un-COW'd. Fix by moving skb_ensure_writable() inside the per-key loop where the actual write offset is known, and add overflow checking on the offset arithmetic. For negative offsets (e.g. Ethernet header edits at ingress), use skb_cow() to COW the headroom instead. Guard offset_valid() against INT_MIN, where negation is undefined. Fixes: 8b796475fd78 ("net/sched: act_pedit: really ensure the skb is writable") Reported-by: Yiming Qian Reported-by: Keenan Dong Reported-by: Han Guidong <2045gemini@gmail.com> Reported-by: Zhang Cen Reviewed-by: Han Guidong <2045gemini@gmail.com> Tested-by: Han Guidong <2045gemini@gmail.com> Reviewed-by: Davide Caratti Tested-by: Davide Caratti Reviewed-by: Toke Høiland-Jørgensen Tested-by: Toke Høiland-Jørgensen Reviewed-by: Victor Nogueira Tested-by: Victor Nogueira Acked-by: Jamal Hadi Salim Signed-off-by: Rajat Gupta Link: https://patch.msgid.link/20260531123221.48732-1-jhs@mojatatu.com Signed-off-by: Jakub Kicinski --- include/net/tc_act/tc_pedit.h | 1 - net/sched/act_pedit.c | 77 +++++++++++++++++++---------------- 2 files changed, 41 insertions(+), 37 deletions(-) diff --git a/include/net/tc_act/tc_pedit.h b/include/net/tc_act/tc_pedit.h index f58ee15cd858..cb7b82f2cbc7 100644 --- a/include/net/tc_act/tc_pedit.h +++ b/include/net/tc_act/tc_pedit.h @@ -15,7 +15,6 @@ struct tcf_pedit_parms { struct tc_pedit_key *tcfp_keys; struct tcf_pedit_key_ex *tcfp_keys_ex; int action; - u32 tcfp_off_max_hint; unsigned char tcfp_nkeys; unsigned char tcfp_flags; struct rcu_head rcu; diff --git a/net/sched/act_pedit.c b/net/sched/act_pedit.c index bc20f08a2789..bd3b1da3cd63 100644 --- a/net/sched/act_pedit.c +++ b/net/sched/act_pedit.c @@ -16,6 +16,8 @@ #include #include #include +#include +#include #include #include #include @@ -242,7 +244,6 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, goto out_free_ex; } - nparms->tcfp_off_max_hint = 0; nparms->tcfp_flags = parm->flags; nparms->tcfp_nkeys = parm->nkeys; @@ -268,14 +269,6 @@ static int tcf_pedit_init(struct net *net, struct nlattr *nla, BITS_PER_TYPE(int) - 1, nparms->tcfp_keys[i].shift); - /* The AT option can read a single byte, we can bound the actual - * value with uchar max. - */ - cur += (0xff & offmask) >> nparms->tcfp_keys[i].shift; - - /* Each key touches 4 bytes starting from the computed offset */ - nparms->tcfp_off_max_hint = - max(nparms->tcfp_off_max_hint, cur + 4); } p = to_pedit(*a); @@ -318,15 +311,12 @@ static void tcf_pedit_cleanup(struct tc_action *a) call_rcu(&parms->rcu, tcf_pedit_cleanup_rcu); } -static bool offset_valid(struct sk_buff *skb, int offset) +static bool offset_valid(struct sk_buff *skb, int offset, int len) { - if (offset > 0 && offset > skb->len) + if (offset < -(int)skb_headroom(skb)) return false; - if (offset < 0 && -offset > skb_headroom(skb)) - return false; - - return true; + return offset <= (int)skb->len - len; } static int pedit_l4_skb_offset(struct sk_buff *skb, int *hoffset, const int header_type) @@ -393,18 +383,10 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, struct tcf_pedit_key_ex *tkey_ex; struct tcf_pedit_parms *parms; struct tc_pedit_key *tkey; - u32 max_offset; int i; parms = rcu_dereference_bh(p->parms); - max_offset = (skb_transport_header_was_set(skb) ? - skb_transport_offset(skb) : - skb_network_offset(skb)) + - parms->tcfp_off_max_hint; - if (skb_ensure_writable(skb, min(skb->len, max_offset))) - goto done; - tcf_lastuse_update(&p->tcf_tm); tcf_action_update_bstats(&p->common, skb); @@ -412,10 +394,11 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, tkey_ex = parms->tcfp_keys_ex; for (i = parms->tcfp_nkeys; i > 0; i--, tkey++) { + int write_offset, write_len; int offset = tkey->off; int hoffset = 0; - u32 *ptr, hdata; - u32 val; + u32 cur_val, val; + u32 *ptr; int rc; if (tkey_ex) { @@ -433,13 +416,15 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, if (tkey->offmask) { u8 *d, _d; + int at_offset; - if (!offset_valid(skb, hoffset + tkey->at)) { + if (check_add_overflow(hoffset, (int)tkey->at, &at_offset) || + !offset_valid(skb, at_offset, sizeof(_d))) { pr_info_ratelimited("tc action pedit 'at' offset %d out of bounds\n", hoffset + tkey->at); goto bad; } - d = skb_header_pointer(skb, hoffset + tkey->at, + d = skb_header_pointer(skb, at_offset, sizeof(_d), &_d); if (!d) goto bad; @@ -451,31 +436,51 @@ TC_INDIRECT_SCOPE int tcf_pedit_act(struct sk_buff *skb, } } - if (!offset_valid(skb, hoffset + offset)) { - pr_info_ratelimited("tc action pedit offset %d out of bounds\n", hoffset + offset); + if (check_add_overflow(hoffset, offset, &write_offset)) { + pr_info_ratelimited("tc action pedit offset overflow\n"); goto bad; } - ptr = skb_header_pointer(skb, hoffset + offset, - sizeof(hdata), &hdata); - if (!ptr) + if (!offset_valid(skb, write_offset, sizeof(*ptr))) { + pr_info_ratelimited("tc action pedit offset %d out of bounds\n", + write_offset); goto bad; + } + + if (write_offset < 0) { + if (skb_cow(skb, -write_offset)) + goto bad; + if (write_offset + (int)sizeof(*ptr) > 0) { + if (skb_ensure_writable(skb, + min_t(int, skb->len, + write_offset + (int)sizeof(*ptr)))) + goto bad; + } + } else { + if (check_add_overflow(write_offset, (int)sizeof(*ptr), + &write_len)) + goto bad; + if (skb_ensure_writable(skb, min_t(int, skb->len, + write_len))) + goto bad; + } + + ptr = (u32 *)(skb->data + write_offset); + cur_val = get_unaligned(ptr); /* just do it, baby */ switch (cmd) { case TCA_PEDIT_KEY_EX_CMD_SET: val = tkey->val; break; case TCA_PEDIT_KEY_EX_CMD_ADD: - val = (*ptr + tkey->val) & ~tkey->mask; + val = (cur_val + tkey->val) & ~tkey->mask; break; default: pr_info_ratelimited("tc action pedit bad command (%d)\n", cmd); goto bad; } - *ptr = ((*ptr & tkey->mask) ^ val); - if (ptr == &hdata) - skb_store_bits(skb, hoffset + offset, ptr, 4); + put_unaligned((cur_val & tkey->mask) ^ val, ptr); } goto done; From 0861615c28de668669d748ef4eb913ea9262d13b Mon Sep 17 00:00:00 2001 From: Xin Long Date: Mon, 1 Jun 2026 21:06:06 -0400 Subject: [PATCH 282/329] sctp: validate cached peer INIT chunk length in COOKIE_ECHO processing When a listening SCTP server processes a COOKIE_ECHO chunk, the cached peer INIT chunk embedded after the cookie is parsed and its parameters are later walked by sctp_process_init() using sctp_walk_params(). However, the chunk header length of this cached INIT chunk was not validated against the remaining buffer in the COOKIE_ECHO payload. If the length field is inflated, the parameter walk can run beyond the actual received data, leading to out-of-bounds reads and potential memory corruption during later parameter handling (e.g. STATE_COOKIE processing and kmemdup() copies). Add a bounds check in sctp_unpack_cookie() to ensure the cached INIT chunk length does not exceed the available data in the COOKIE_ECHO buffer before it is used. Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Reported-by: Brian Geffon Signed-off-by: Xin Long Link: https://patch.msgid.link/eb60825fa22d6f9e663c7d4dbb69f397b5d34d42.1780362366.git.lucien.xin@gmail.com Signed-off-by: Jakub Kicinski --- net/sctp/sm_make_chunk.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index de86ac088289..85264862fb6b 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -1730,6 +1730,7 @@ struct sctp_association *sctp_unpack_cookie( struct sctp_signed_cookie *cookie; struct sk_buff *skb = chunk->skb; struct sctp_cookie *bear_cookie; + struct sctp_chunkhdr *ch; enum sctp_scope scope; unsigned int len; ktime_t kt; @@ -1759,6 +1760,10 @@ struct sctp_association *sctp_unpack_cookie( cookie = chunk->subh.cookie_hdr; bear_cookie = &cookie->c; + ch = (struct sctp_chunkhdr *)(bear_cookie + 1); + if (ntohs(ch->length) > len - fixed_size) + goto malformed; + /* Verify the cookie's MAC, if cookie authentication is enabled. */ if (sctp_sk(ep->base.sk)->cookie_auth_enable) { u8 mac[SHA256_DIGEST_SIZE]; From 1232b3104b4b2c0267f31608fe0f8a8758428f28 Mon Sep 17 00:00:00 2001 From: Zhi Li Date: Tue, 2 Jun 2026 09:45:28 +0800 Subject: [PATCH 283/329] dt-bindings: ethernet: eswin: fix hsp-sp-csr backward compatibility Commit c36069c6f46c ("dt-bindings: ethernet: eswin: add optional TXD and RXD delay register offsets") added two optional cells to eswin,hsp-sp-csr but omitted minItems: 4. As a result, dt-schema implicitly required all 6 cells, which broke backward compatibility with existing 4-cell device trees. Add minItems: 4 to preserve backward compatibility. Fixes: c36069c6f46c ("dt-bindings: ethernet: eswin: add optional TXD and RXD delay register offsets") Reported-by: Sashiko AI Closes: https://lore.kernel.org/all/20260519022334.35742C2BCB7@smtp.kernel.org/ Reviewed-by: Krzysztof Kozlowski Signed-off-by: Zhi Li Link: https://patch.msgid.link/20260602014528.2076-1-lizhi2@eswincomputing.com Signed-off-by: Jakub Kicinski --- Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml b/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml index b66ae6300faf..65882ff79d8d 100644 --- a/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml +++ b/Documentation/devicetree/bindings/net/eswin,eic7700-eth.yaml @@ -84,7 +84,8 @@ properties: This reference is provided for background information only. $ref: /schemas/types.yaml#/definitions/phandle-array items: - - items: + - minItems: 4 + items: - description: Phandle to HSP(High-Speed Peripheral) device - description: Offset of phy control register for internal or external clock selection From 1d31eb27e570daa04f5373345f9ac98c95863be9 Mon Sep 17 00:00:00 2001 From: Nithin Dabilpuram Date: Tue, 2 Jun 2026 10:28:53 +0530 Subject: [PATCH 284/329] octeontx2-af: npc: Fix CPT channel mask in npc_install_flow Use the CPT-aware NIX channel mask in the npc_install_flow path so that when the host PF installs steering rules in kernel for a VF used from userspace (e.g. DPDK), MCAM entries see the same channel mask semantics as other RX paths. Fixes: 56bcef528bd8 ("octeontx2-af: Use npc_install_flow API for promisc and broadcast entries") Cc: Naveen Mamindlapalli Signed-off-by: Nithin Dabilpuram Signed-off-by: Ratheesh Kannoth Link: https://patch.msgid.link/20260602045853.1558530-1-rkannoth@marvell.com Signed-off-by: Jakub Kicinski --- .../net/ethernet/marvell/octeontx2/af/rvu.h | 1 + .../ethernet/marvell/octeontx2/af/rvu_npc.c | 32 +++++++++---------- .../marvell/octeontx2/af/rvu_npc_fs.c | 2 +- 3 files changed, 18 insertions(+), 17 deletions(-) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h index de3fbd3d15d6..65397daae4c2 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h @@ -1145,6 +1145,7 @@ int rvu_cpt_lf_teardown(struct rvu *rvu, u16 pcifunc, int blkaddr, int lf, int slot); int rvu_cpt_ctx_flush(struct rvu *rvu, u16 pcifunc); int rvu_cpt_init(struct rvu *rvu); +u32 rvu_get_cpt_chan_mask(struct rvu *rvu); #define NDC_AF_BANK_MASK GENMASK_ULL(7, 0) #define NDC_AF_BANK_LINE_MASK GENMASK_ULL(31, 16) diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c index 6bbda0593fcd..d301a3f0f87a 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c @@ -701,6 +701,19 @@ void npc_set_mcam_action(struct rvu *rvu, struct npc_mcam *mcam, return rvu_write64(rvu, blkaddr, reg, cfg); } +u32 rvu_get_cpt_chan_mask(struct rvu *rvu) +{ + /* For cn10k the upper two bits of the channel number are + * cpt channel number. with masking out these bits in the + * mcam entry, same entry used for NIX will allow packets + * received from cpt for parsing. + */ + if (!is_rvu_otx2(rvu)) + return NIX_CHAN_CPT_X2P_MASK; + else + return 0xFFFu; +} + void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc, int nixlf, u64 chan, u8 *mac_addr) { @@ -750,7 +763,7 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc, eth_broadcast_addr((u8 *)&req.mask.dmac); req.features = BIT_ULL(NPC_DMAC); req.channel = chan; - req.chan_mask = 0xFFFU; + req.chan_mask = rvu_get_cpt_chan_mask(rvu); req.intf = pfvf->nix_rx_intf; req.op = action.op; req.hdr.pcifunc = 0; /* AF is requester */ @@ -845,11 +858,7 @@ void rvu_npc_install_promisc_entry(struct rvu *rvu, u16 pcifunc, * mcam entry, same entry used for NIX will allow packets * received from cpt for parsing. */ - if (!is_rvu_otx2(rvu)) { - req.chan_mask = NIX_CHAN_CPT_X2P_MASK; - } else { - req.chan_mask = 0xFFFU; - } + req.chan_mask = rvu_get_cpt_chan_mask(rvu); if (chan_cnt > 1) { if (!is_power_of_2(chan_cnt)) { @@ -1053,16 +1062,7 @@ void rvu_npc_install_allmulti_entry(struct rvu *rvu, u16 pcifunc, int nixlf, ether_addr_copy(req.mask.dmac, mac_addr); req.features = BIT_ULL(NPC_DMAC); - /* For cn10k the upper two bits of the channel number are - * cpt channel number. with masking out these bits in the - * mcam entry, same entry used for NIX will allow packets - * received from cpt for parsing. - */ - if (!is_rvu_otx2(rvu)) - req.chan_mask = NIX_CHAN_CPT_X2P_MASK; - else - req.chan_mask = 0xFFFU; - + req.chan_mask = rvu_get_cpt_chan_mask(rvu); req.channel = chan; req.intf = pfvf->nix_rx_intf; req.entry = index; diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c index 6ae9cdcb608b..34f1e066707b 100644 --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c @@ -1820,7 +1820,7 @@ int rvu_mbox_handler_npc_install_flow(struct rvu *rvu, /* ignore chan_mask in case pf func is not AF, revisit later */ if (!is_pffunc_af(req->hdr.pcifunc)) - req->chan_mask = 0xFFF; + req->chan_mask = rvu_get_cpt_chan_mask(rvu); err = npc_check_unsupported_flows(rvu, req->features, req->intf); if (err) { From ab1ecaabe74b7d86c38ab2ab44bd56cdcc33645a Mon Sep 17 00:00:00 2001 From: Justin Lai Date: Tue, 2 Jun 2026 19:46:59 +0800 Subject: [PATCH 285/329] rtase: Reset TX subqueue when clearing TX ring rtase_tx_clear() clears the TX ring and resets the ring indexes. However, the TX queue state and BQL accounting are not reset at the same time. This may leave __QUEUE_STATE_STACK_XOFF asserted after rtase_sw_reset(), preventing new TX packets from being scheduled. Reset the TX subqueue when clearing the TX ring so the TX queue state and BQL accounting are restored together. Fixes: 5a2a2f15244c ("rtase: Implement the rtase_down function") Cc: stable@vger.kernel.org Signed-off-by: Justin Lai Reviewed-by: Alexander Lobakin Link: https://patch.msgid.link/20260602114659.12335-1-justinlai0215@realtek.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/rtase/rtase_main.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/realtek/rtase/rtase_main.c b/drivers/net/ethernet/realtek/rtase/rtase_main.c index ef13109c49cf..6ccbefb5acf2 100644 --- a/drivers/net/ethernet/realtek/rtase/rtase_main.c +++ b/drivers/net/ethernet/realtek/rtase/rtase_main.c @@ -239,6 +239,8 @@ static void rtase_tx_clear(struct rtase_private *tp) rtase_tx_clear_range(ring, ring->dirty_idx, RTASE_NUM_DESC); ring->cur_idx = 0; ring->dirty_idx = 0; + + netdev_tx_reset_subqueue(tp->dev, i); } } From aa6ca1c5c338907817374b59f7551fd855a88754 Mon Sep 17 00:00:00 2001 From: Andy Roulin Date: Tue, 2 Jun 2026 11:51:36 -0700 Subject: [PATCH 286/329] vxlan: vnifilter: send notification on VNI add When a new VNI is added to a vxlan device with vnifilter enabled, no RTM_NEWTUNNEL notification is sent to userspace. This means 'bridge monitor vni' never shows VNI add events, even though VNI delete events are reported correctly. The bug is in vxlan_vni_add(), where the notification is guarded by 'if (changed)'. The 'changed' flag is set by vxlan_vni_update_group() only when the multicast group or remote IP is modified, but for a new VNI added without a group (e.g. in L3 VxLAN interface scenarios), the function returns early without setting changed=true. Since this is a new VNI, the notification should be sent unconditionally. The notification is not guarded by the return value of vxlan_vni_update_group() because, at this point, the VNI has already been inserted into the hash table and list with no rollback on error. The VNI will be visible in 'bridge vni show' regardless, so userspace should be informed. This is consistent with vxlan_vni_del() which also notifies unconditionally. The 'if (changed)' guard remains correct in vxlan_vni_update(), which handles the case where a VNI already exists and is being re-added -- there, we only want to notify if the group/remote actually changed. Reproducer: # ip link add vxlan100 type vxlan dstport 4789 local 10.0.0.1 \ nolearning external vnifilter # ip link set vxlan100 up # bridge monitor vni & # bridge vni add vni 1000 dev vxlan100 # no notification # bridge vni delete vni 1000 dev vxlan100 # notification received Fixes: f9c4bb0b245c ("vxlan: vni filtering support on collect metadata device") Reported-by: Chirag Shah Signed-off-by: Andy Roulin Reviewed-by: Petr Machata Link: https://patch.msgid.link/20260602185138.253265-2-aroulin@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/vxlan/vxlan_vnifilter.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/net/vxlan/vxlan_vnifilter.c b/drivers/net/vxlan/vxlan_vnifilter.c index 2042369379ff..f2a202d46892 100644 --- a/drivers/net/vxlan/vxlan_vnifilter.c +++ b/drivers/net/vxlan/vxlan_vnifilter.c @@ -759,8 +759,7 @@ static int vxlan_vni_add(struct vxlan_dev *vxlan, err = vxlan_vni_update_group(vxlan, vninode, group, true, &changed, extack); - if (changed) - vxlan_vnifilter_notify(vxlan, vninode, RTM_NEWTUNNEL); + vxlan_vnifilter_notify(vxlan, vninode, RTM_NEWTUNNEL); return err; } From 84683b5b60c7274e2c8f7f413d39d78d3db5540f Mon Sep 17 00:00:00 2001 From: Andy Roulin Date: Tue, 2 Jun 2026 11:51:37 -0700 Subject: [PATCH 287/329] vxlan: vnifilter: fix spurious notification on VNI update When a VNI is re-added with the same attributes (e.g. same group or no group), vxlan_vni_update() sends a spurious RTM_NEWTUNNEL notification even though nothing changed. The bug is that 'if (changed)' tests whether the pointer is non-NULL, not the bool value it points to. Since every caller passes a valid pointer, the condition is always true and the notification fires unconditionally. Fix by dereferencing the pointer: 'if (*changed)'. Reproducer: # ip link add vxlan100 type vxlan dstport 4789 local 10.0.0.1 \ nolearning external vnifilter # ip link set vxlan100 up # bridge monitor vni & # bridge vni add vni 1000 dev vxlan100 # bridge vni add vni 1000 dev vxlan100 # spurious notification Fixes: f9c4bb0b245c ("vxlan: vni filtering support on collect metadata device") Signed-off-by: Andy Roulin Reviewed-by: Petr Machata Link: https://patch.msgid.link/20260602185138.253265-3-aroulin@nvidia.com Signed-off-by: Jakub Kicinski --- drivers/net/vxlan/vxlan_vnifilter.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/vxlan/vxlan_vnifilter.c b/drivers/net/vxlan/vxlan_vnifilter.c index f2a202d46892..3e76f4e21094 100644 --- a/drivers/net/vxlan/vxlan_vnifilter.c +++ b/drivers/net/vxlan/vxlan_vnifilter.c @@ -661,7 +661,7 @@ static int vxlan_vni_update(struct vxlan_dev *vxlan, if (ret) return ret; - if (changed) + if (*changed) vxlan_vnifilter_notify(vxlan, vninode, RTM_NEWTUNNEL); return 0; From 8df1c84e7272a5e24b563df7e50111dc81014d4a Mon Sep 17 00:00:00 2001 From: Andy Roulin Date: Tue, 2 Jun 2026 11:51:38 -0700 Subject: [PATCH 288/329] selftests: net: add vxlan vnifilter notification test Add a selftest for VXLAN vnifilter netlink notifications that verifies RTM_NEWTUNNEL and RTM_DELTUNNEL are sent correctly when VNIs are added, deleted, or updated, and that no spurious notifications are sent when a VNI is re-added with the same attributes. Signed-off-by: Andy Roulin Acked-by: Petr Machata Link: https://patch.msgid.link/20260602185138.253265-4-aroulin@nvidia.com Signed-off-by: Jakub Kicinski --- tools/testing/selftests/net/Makefile | 1 + .../net/test_vxlan_vnifilter_notify.sh | 184 ++++++++++++++++++ 2 files changed, 185 insertions(+) create mode 100755 tools/testing/selftests/net/test_vxlan_vnifilter_notify.sh diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile index f3da38c54d27..2ed7d803eb54 100644 --- a/tools/testing/selftests/net/Makefile +++ b/tools/testing/selftests/net/Makefile @@ -109,6 +109,7 @@ TEST_PROGS := \ test_vxlan_nh.sh \ test_vxlan_nolocalbypass.sh \ test_vxlan_under_vrf.sh \ + test_vxlan_vnifilter_notify.sh \ test_vxlan_vnifiltering.sh \ tfo_passive.sh \ traceroute.sh \ diff --git a/tools/testing/selftests/net/test_vxlan_vnifilter_notify.sh b/tools/testing/selftests/net/test_vxlan_vnifilter_notify.sh new file mode 100755 index 000000000000..9d51a9e02ae0 --- /dev/null +++ b/tools/testing/selftests/net/test_vxlan_vnifilter_notify.sh @@ -0,0 +1,184 @@ +#!/bin/bash +# SPDX-License-Identifier: GPL-2.0 +# shellcheck disable=SC2034,SC2154,SC2317,SC2329 +# +# Test for VXLAN vnifilter netlink notifications (RTM_NEWTUNNEL / +# RTM_DELTUNNEL). +# +# Verifies that: +# - Adding a new VNI sends a notification +# - Adding a new VNI with a remote sends a notification +# - Deleting a VNI sends a notification +# - Re-adding an existing VNI with the same attributes does not send +# a spurious notification +# - Updating an existing VNI's remote sends a notification +# - Deleting a non-existent VNI does not send a notification + +source lib.sh + +require_command bridge + +VXLAN_DEV=vxlan100 + +ALL_TESTS=" + test_vni_add_notify + test_vni_add_remote_notify + test_vni_del_notify + test_vni_readd_no_notify + test_vni_update_remote_notify + test_vni_del_nonexistent_no_notify +" + +setup_prepare() +{ + setup_ns NS1 + defer cleanup_all_ns + + ip -n "$NS1" link add $VXLAN_DEV type vxlan dstport 4789 \ + local 10.0.0.1 nolearning external vnifilter + ip -n "$NS1" link set $VXLAN_DEV up +} + +# Run bridge monitor in the background, execute a command, then count +# the notification lines. +# Usage: vni_notify_check [args...] +# Sets: NOTIFY_COUNT with the number of notifications observed. +vni_notify_check() +{ + local tmpf cmd_ret monitor_pid + + tmpf=$(mktemp) + defer rm "$tmpf" + + defer_scope_push + ip netns exec "$NS1" bridge monitor vni > "$tmpf" 2>/dev/null & + monitor_pid=$! + defer kill_process "$monitor_pid" + + sleep 0.5 + if [ ! -e "/proc/$monitor_pid" ]; then + RET=$ksft_skip + log_test "iproute2 'bridge monitor vni' not supported" + return "$RET" + fi + + "$@" + cmd_ret=$? + sleep 0.2 + defer_scope_pop + + NOTIFY_COUNT=$(grep -c "$VXLAN_DEV" "$tmpf") + NOTIFY_COUNT=${NOTIFY_COUNT:-0} + return "$cmd_ret" +} + +# Adding a brand new VNI should produce a notification. +test_vni_add_notify() +{ + RET=0 + + vni_notify_check \ + bridge -n "$NS1" vni add vni 1000 dev "$VXLAN_DEV" + check_err $? "Failed to add VNI" + + [ "$NOTIFY_COUNT" -eq 1 ] + check_err $? "Expected 1 notification for VNI add, got $NOTIFY_COUNT" + + bridge -n "$NS1" vni delete vni 1000 dev "$VXLAN_DEV" 2>/dev/null + + log_test "VNI add sends notification" +} + +# Adding a VNI with a remote should produce a notification. +test_vni_add_remote_notify() +{ + RET=0 + + vni_notify_check \ + bridge -n "$NS1" vni add vni 4000 remote 10.0.0.2 dev "$VXLAN_DEV" + check_err $? "Failed to add VNI with remote" + + [ "$NOTIFY_COUNT" -eq 1 ] + check_err $? "Expected 1 notification for VNI add with remote, got $NOTIFY_COUNT" + + bridge -n "$NS1" vni delete vni 4000 dev "$VXLAN_DEV" + + log_test "VNI add with remote sends notification" +} + +# Deleting a VNI should produce a notification. +test_vni_del_notify() +{ + RET=0 + + bridge -n "$NS1" vni add vni 2000 dev "$VXLAN_DEV" + + vni_notify_check \ + bridge -n "$NS1" vni delete vni 2000 dev "$VXLAN_DEV" + check_err $? "Failed to delete VNI" + + [ "$NOTIFY_COUNT" -eq 1 ] + check_err $? "Expected 1 notification for VNI del, got $NOTIFY_COUNT" + + log_test "VNI delete sends notification" +} + +# Re-adding an existing VNI with the same attributes should not produce +# a notification. +test_vni_readd_no_notify() +{ + RET=0 + + bridge -n "$NS1" vni add vni 3000 dev "$VXLAN_DEV" + + vni_notify_check \ + bridge -n "$NS1" vni add vni 3000 dev "$VXLAN_DEV" + check_err $? "Failed to re-add VNI" + + [ "$NOTIFY_COUNT" -eq 0 ] + check_err $? "Expected 0 notifications for VNI re-add, got $NOTIFY_COUNT" + + bridge -n "$NS1" vni delete vni 3000 dev "$VXLAN_DEV" + + log_test "VNI re-add does not send spurious notification" +} + +# Updating an existing VNI's remote should produce a notification. +test_vni_update_remote_notify() +{ + RET=0 + + bridge -n "$NS1" vni add vni 5000 remote 10.0.0.2 dev "$VXLAN_DEV" + + vni_notify_check \ + bridge -n "$NS1" vni add vni 5000 remote 10.0.0.3 dev "$VXLAN_DEV" + check_err $? "Failed to update VNI remote" + + [ "$NOTIFY_COUNT" -eq 1 ] + check_err $? "Expected 1 notification for VNI remote update, got $NOTIFY_COUNT" + + bridge -n "$NS1" vni delete vni 5000 dev "$VXLAN_DEV" + + log_test "VNI remote update sends notification" +} + +# Deleting a non-existent VNI should not produce a notification. +test_vni_del_nonexistent_no_notify() +{ + RET=0 + + vni_notify_check \ + bridge -n "$NS1" vni delete vni 9999 dev "$VXLAN_DEV" 2>/dev/null + + [ "$NOTIFY_COUNT" -eq 0 ] + check_err $? "Expected 0 notifications for non-existent VNI del, got $NOTIFY_COUNT" + + log_test "Non-existent VNI delete does not send notification" +} + +trap defer_scopes_cleanup EXIT + +setup_prepare +tests_run + +exit "$EXIT_STATUS" From 791c91dc7a9dfb2457d5e29b8216a6484b9c4b40 Mon Sep 17 00:00:00 2001 From: Ido Schimmel Date: Wed, 3 Jun 2026 13:18:11 +0300 Subject: [PATCH 289/329] ipv6: mcast: Fix use-after-free when processing MLD queries When processing an MLD query, a pointer to the multicast group address is retrieved when initially parsing the packet. This pointer is later dereferenced without being reloaded despite the fact that the skb header might have been reallocated following the pskb_may_pull() calls, leading to a use-after-free [1]. Fix by copying the multicast group address when the packet is initially parsed. [1] BUG: KASAN: slab-use-after-free in __mld_query_work (net/ipv6/mcast.c:1512) Read of size 8 at addr ffff8881154b8e90 by task kworker/4:1/118 Workqueue: mld mld_query_work Call Trace: dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120) print_address_description.constprop.0 (mm/kasan/report.c:378) print_report (mm/kasan/report.c:482) kasan_report (mm/kasan/report.c:595) __mld_query_work (net/ipv6/mcast.c:1512) mld_query_work (net/ipv6/mcast.c:1563) process_one_work (kernel/workqueue.c:3314) worker_thread (kernel/workqueue.c:3397 kernel/workqueue.c:3478) kthread (kernel/kthread.c:436) ret_from_fork (arch/x86/kernel/process.c:158) ret_from_fork_asm (arch/x86/entry/entry_64.S:245) [...] Freed by task 118: kasan_save_stack (mm/kasan/common.c:57) kasan_save_track (mm/kasan/common.c:78) kasan_save_free_info (mm/kasan/generic.c:584) __kasan_slab_free (mm/kasan/common.c:253 mm/kasan/common.c:285) kfree (./include/linux/kasan.h:235 mm/slub.c:2689 mm/slub.c:6251 mm/slub.c:6566) pskb_expand_head (net/core/skbuff.c:2335) __pskb_pull_tail (net/core/skbuff.c:2878 (discriminator 4)) __mld_query_work (net/ipv6/mcast.c:1495 (discriminator 1)) mld_query_work (net/ipv6/mcast.c:1563) process_one_work (kernel/workqueue.c:3314) worker_thread (kernel/workqueue.c:3397 kernel/workqueue.c:3478) kthread (kernel/kthread.c:436) ret_from_fork (arch/x86/kernel/process.c:158) ret_from_fork_asm (arch/x86/entry/entry_64.S:245) Fixes: 97300b5fdfe2 ("[MCAST] IPv6: Check packet size when process Multicast") Reported-by: Leo Lin Reviewed-by: David Ahern Signed-off-by: Ido Schimmel Reviewed-by: Eric Dumazet Reviewed-by: Jiayuan Chen Link: https://patch.msgid.link/20260603101811.612594-1-idosch@nvidia.com Signed-off-by: Jakub Kicinski --- net/ipv6/mcast.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 3330adcf26db..d9b855d5191b 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1424,9 +1424,9 @@ void igmp6_event_query(struct sk_buff *skb) static void __mld_query_work(struct sk_buff *skb) { struct mld2_query *mlh2 = NULL; - const struct in6_addr *group; unsigned long max_delay; struct inet6_dev *idev; + struct in6_addr group; struct ifmcaddr6 *ma; struct mld_msg *mld; int group_type; @@ -1458,8 +1458,8 @@ static void __mld_query_work(struct sk_buff *skb) goto kfree_skb; mld = (struct mld_msg *)icmp6_hdr(skb); - group = &mld->mld_mca; - group_type = ipv6_addr_type(group); + group = mld->mld_mca; + group_type = ipv6_addr_type(&group); if (group_type != IPV6_ADDR_ANY && !(group_type&IPV6_ADDR_MULTICAST)) @@ -1509,7 +1509,7 @@ static void __mld_query_work(struct sk_buff *skb) } } else { for_each_mc_mclock(idev, ma) { - if (!ipv6_addr_equal(group, &ma->mca_addr)) + if (!ipv6_addr_equal(&group, &ma->mca_addr)) continue; if (ma->mca_flags & MAF_TIMER_RUNNING) { /* gsquery <- gsquery && mark */ From 3a5f3f7aff18bcc36a57839cf50cf0cc8de707f3 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 3 Jun 2026 07:29:55 +0000 Subject: [PATCH 290/329] ieee802154: 6lowpan: only accept IPv6 packets in lowpan_xmit() The aoe driver (or similar) generates a non-IPv6 packet (e.g., ETH_P_AOE) and queues it for transmission via dev_queue_xmit() on a 6LoWPAN interface (configured by the user or test case). Since the packet is not IPv6, the 6LoWPAN header_ops->create function (lowpan_header_create or header_create) returns early without initializing the lowpan_addr_info structure in the skb headroom. In the transmit function (lowpan_xmit), the driver calls lowpan_header (or setup_header) which unconditionally copies and uses the lowpan_addr_info from the headroom, which contains uninitialized data. Fix this by dropping non IPv6 packets. A similar fix is needed in net/bluetooth/6lowpan.c bt_xmit(). Fixes: 4dc315e267fe ("ieee802154: 6lowpan: move transmit functionality") Reported-by: syzbot+f13c19f75e1097abd116@syzkaller.appspotmail.com Closes: https://lore.kernel.org/netdev/6a1fd763.278b5b03.2bcf39.0049.GAE@google.com/T/#u Signed-off-by: Eric Dumazet Reviewed-by: Miquel Raynal Link: https://patch.msgid.link/20260603072955.4032221-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- net/ieee802154/6lowpan/tx.c | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/net/ieee802154/6lowpan/tx.c b/net/ieee802154/6lowpan/tx.c index 0c07662b44c0..4df76ff50699 100644 --- a/net/ieee802154/6lowpan/tx.c +++ b/net/ieee802154/6lowpan/tx.c @@ -255,6 +255,11 @@ netdev_tx_t lowpan_xmit(struct sk_buff *skb, struct net_device *ldev) pr_debug("package xmit\n"); + if (skb->protocol != htons(ETH_P_IPV6)) { + kfree_skb(skb); + return NET_XMIT_DROP; + } + WARN_ON_ONCE(skb->len > IPV6_MIN_MTU); /* We must take a copy of the skb before we modify/replace the ipv6 From 9fc237f8d49f06d05f0f8e80361047b718894e81 Mon Sep 17 00:00:00 2001 From: Justin Lai Date: Wed, 3 Jun 2026 14:18:16 +0800 Subject: [PATCH 291/329] rtase: Avoid sleeping in get_stats64() The .ndo_get_stats64 callback must not sleep because it can be called when reading /proc/net/dev. rtase_get_stats64() calls rtase_dump_tally_counter(), which polls the tally counter dump bit with read_poll_timeout(). This may sleep while waiting for the hardware counter dump to complete. Use read_poll_timeout_atomic() instead to avoid sleeping in the get_stats64() path. Fixes: 079600489960 ("rtase: Implement net_device_ops") Cc: stable@vger.kernel.org Signed-off-by: Justin Lai Link: https://patch.msgid.link/20260603061816.31356-1-justinlai0215@realtek.com Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/realtek/rtase/rtase_main.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/realtek/rtase/rtase_main.c b/drivers/net/ethernet/realtek/rtase/rtase_main.c index 6ccbefb5acf2..55105d34bc79 100644 --- a/drivers/net/ethernet/realtek/rtase/rtase_main.c +++ b/drivers/net/ethernet/realtek/rtase/rtase_main.c @@ -1565,8 +1565,9 @@ static void rtase_dump_tally_counter(const struct rtase_private *tp) rtase_w32(tp, RTASE_DTCCR0, cmd); rtase_w32(tp, RTASE_DTCCR0, cmd | RTASE_COUNTER_DUMP); - err = read_poll_timeout(rtase_r32, val, !(val & RTASE_COUNTER_DUMP), - 10, 250, false, tp, RTASE_DTCCR0); + err = read_poll_timeout_atomic(rtase_r32, val, + !(val & RTASE_COUNTER_DUMP), + 10, 250, false, tp, RTASE_DTCCR0); if (err == -ETIMEDOUT) netdev_err(tp->dev, "error occurred in dump tally counter\n"); From 7561c7fbc694308da73300f036719e63e42bf0b4 Mon Sep 17 00:00:00 2001 From: Yizhou Zhao Date: Wed, 3 Jun 2026 14:00:13 +0800 Subject: [PATCH 292/329] net/802/mrp: fix vector attribute parsing in mrp_pdu_parse_vecattr In mrp_pdu_parse_vecattr(), vector attribute events are encoded three per byte and valen tracks the number of events left to process. The parser decrements valen after processing the first and second events from each event byte, but not after processing the third one. When valen is exactly a multiple of three, the loop continues after the last valid event and consumes the next byte as a new event byte, applying a spurious event to the MRP applicant state. Additionally, when valen is zero the parser unconditionally consumes attrlen bytes as FirstValue and advances the offset, even though per IEEE 802.1ak a VectorAttribute with only a LeaveAllEvent has valen of zero and no FirstValue or Vector fields. This corrupts the offset for subsequent PDU parsing. Also, when valen exceeds three the loop crosses byte boundaries but the attribute value is not incremented between the last event of one byte and the first event of the next. This causes the first event of the next byte to use the same attribute value as the third event rather than the next consecutive value. Decrement valen after processing the third event, skip FirstValue consumption when valen is zero, and increment the attribute value at the end of each loop iteration. Fixes: febf018d2234 ("net/802: Implement Multiple Registration Protocol (MRP)") Reported-by: Yizhou Zhao Reported-by: Yuxiang Yang Reported-by: Ao Wang Reported-by: Xuewei Feng Reported-by: Qi Li Reported-by: Ke Xu Signed-off-by: Yizhou Zhao Link: https://patch.msgid.link/20260603060016.21522-1-zhaoyz24@mails.tsinghua.edu.cn Signed-off-by: Jakub Kicinski --- net/802/mrp.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/net/802/mrp.c b/net/802/mrp.c index ff0e80574e6b..160a3b14569c 100644 --- a/net/802/mrp.c +++ b/net/802/mrp.c @@ -703,6 +703,12 @@ static int mrp_pdu_parse_vecattr(struct mrp_applicant *app, valen = be16_to_cpu(get_unaligned(&mrp_cb(skb)->vah->lenflags) & MRP_VECATTR_HDR_LEN_MASK); + /* If valen is 0, only a LeaveAllEvent is present; FirstValue and + * Vector fields are absent per IEEE 802.1ak. + */ + if (valen == 0) + return 0; + /* The VectorAttribute structure in a PDU carries event information * about one or more attributes having consecutive values. Only the * value for the first attribute is contained in the structure. So @@ -753,6 +759,9 @@ static int mrp_pdu_parse_vecattr(struct mrp_applicant *app, vaevents %= __MRP_VECATTR_EVENT_MAX; vaevent = vaevents; mrp_pdu_parse_vecattr_event(app, skb, vaevent); + valen--; + mrp_attrvalue_inc(mrp_cb(skb)->attrvalue, + mrp_cb(skb)->mh->attrlen); } return 0; } From b47ff80f280e18ad2310f44293cc057d9b64ff11 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Wed, 3 Jun 2026 12:35:14 +0000 Subject: [PATCH 293/329] bonding: annotate data-races arcound churn variables These fields are updated asynchronously by the bonding state machine in ad_churn_machine() while holding bond->mode_lock. bond_info_show_slave() and bond_fill_slave_info() read them without bond->mode_lock being held, we need to add READ_ONCE() and WRITE_ONCE() annotations. Note that AD_CHURN_MONITOR, AD_CHURN, and AD_NO_CHURN are defined exclusively in (kernel private) include/net/bond_3ad.h header. They should be moved to include/uapi/linux/if_bonding.h or userspace tools will have to hardcode their values. Fixes: 4916f2e2f3fc ("bonding: print churn state via netlink") Fixes: 14c9551a32eb ("bonding: Implement port churn-machine (AD standard 43.4.17).") Signed-off-by: Eric Dumazet Link: https://patch.msgid.link/20260603123514.388226-1-edumazet@google.com Signed-off-by: Jakub Kicinski --- drivers/net/bonding/bond_3ad.c | 18 ++++++++++-------- drivers/net/bonding/bond_netlink.c | 4 ++-- drivers/net/bonding/bond_procfs.c | 8 ++++---- 3 files changed, 16 insertions(+), 14 deletions(-) diff --git a/drivers/net/bonding/bond_3ad.c b/drivers/net/bonding/bond_3ad.c index f0aa7d2f2171..985ef66dc333 100644 --- a/drivers/net/bonding/bond_3ad.c +++ b/drivers/net/bonding/bond_3ad.c @@ -1386,8 +1386,8 @@ static void ad_churn_machine(struct port *port) { if (port->sm_vars & AD_PORT_CHURNED) { port->sm_vars &= ~AD_PORT_CHURNED; - port->sm_churn_actor_state = AD_CHURN_MONITOR; - port->sm_churn_partner_state = AD_CHURN_MONITOR; + WRITE_ONCE(port->sm_churn_actor_state, AD_CHURN_MONITOR); + WRITE_ONCE(port->sm_churn_partner_state, AD_CHURN_MONITOR); port->sm_churn_actor_timer_counter = __ad_timer_to_ticks(AD_ACTOR_CHURN_TIMER, 0); port->sm_churn_partner_timer_counter = @@ -1398,20 +1398,22 @@ static void ad_churn_machine(struct port *port) !(--port->sm_churn_actor_timer_counter) && port->sm_churn_actor_state == AD_CHURN_MONITOR) { if (port->actor_oper_port_state & LACP_STATE_SYNCHRONIZATION) { - port->sm_churn_actor_state = AD_NO_CHURN; + WRITE_ONCE(port->sm_churn_actor_state, AD_NO_CHURN); } else { - port->churn_actor_count++; - port->sm_churn_actor_state = AD_CHURN; + WRITE_ONCE(port->churn_actor_count, + port->churn_actor_count + 1); + WRITE_ONCE(port->sm_churn_actor_state, AD_CHURN); } } if (port->sm_churn_partner_timer_counter && !(--port->sm_churn_partner_timer_counter) && port->sm_churn_partner_state == AD_CHURN_MONITOR) { if (port->partner_oper.port_state & LACP_STATE_SYNCHRONIZATION) { - port->sm_churn_partner_state = AD_NO_CHURN; + WRITE_ONCE(port->sm_churn_partner_state, AD_NO_CHURN); } else { - port->churn_partner_count++; - port->sm_churn_partner_state = AD_CHURN; + WRITE_ONCE(port->churn_partner_count, + port->churn_partner_count + 1); + WRITE_ONCE(port->sm_churn_partner_state, AD_CHURN); } } } diff --git a/drivers/net/bonding/bond_netlink.c b/drivers/net/bonding/bond_netlink.c index c7d3e0602c83..90365d3f7ebf 100644 --- a/drivers/net/bonding/bond_netlink.c +++ b/drivers/net/bonding/bond_netlink.c @@ -82,10 +82,10 @@ static int bond_fill_slave_info(struct sk_buff *skb, goto nla_put_failure_rcu; if (nla_put_u8(skb, IFLA_BOND_SLAVE_AD_CHURN_ACTOR_STATE, - ad_port->sm_churn_actor_state)) + READ_ONCE(ad_port->sm_churn_actor_state))) goto nla_put_failure_rcu; if (nla_put_u8(skb, IFLA_BOND_SLAVE_AD_CHURN_PARTNER_STATE, - ad_port->sm_churn_partner_state)) + READ_ONCE(ad_port->sm_churn_partner_state))) goto nla_put_failure_rcu; } rcu_read_unlock(); diff --git a/drivers/net/bonding/bond_procfs.c b/drivers/net/bonding/bond_procfs.c index 3714aab1a3d9..3607b62f9b63 100644 --- a/drivers/net/bonding/bond_procfs.c +++ b/drivers/net/bonding/bond_procfs.c @@ -221,13 +221,13 @@ static void bond_info_show_slave(struct seq_file *seq, seq_printf(seq, "Aggregator ID: %d\n", agg->aggregator_identifier); seq_printf(seq, "Actor Churn State: %s\n", - bond_3ad_churn_desc(port->sm_churn_actor_state)); + bond_3ad_churn_desc(READ_ONCE(port->sm_churn_actor_state))); seq_printf(seq, "Partner Churn State: %s\n", - bond_3ad_churn_desc(port->sm_churn_partner_state)); + bond_3ad_churn_desc(READ_ONCE(port->sm_churn_partner_state))); seq_printf(seq, "Actor Churned Count: %d\n", - port->churn_actor_count); + READ_ONCE(port->churn_actor_count)); seq_printf(seq, "Partner Churned Count: %d\n", - port->churn_partner_count); + READ_ONCE(port->churn_partner_count)); if (capable(CAP_NET_ADMIN)) { seq_puts(seq, "details actor lacp pdu:\n"); From e374b22e9b07b72a25909621464ff74096151bfb Mon Sep 17 00:00:00 2001 From: Xin Long Date: Wed, 3 Jun 2026 14:11:44 -0400 Subject: [PATCH 294/329] sctp: purge outqueue on stale COOKIE-ECHO handling sctp_stream_update() is only invoked when the association is moved into COOKIE_WAIT during association setup/reconfiguration. In this path, the outbound stream scheduler state (stream->out_curr) is expected to be clean, since no user data should have been transmitted yet unless the state machine has already partially progressed. However, a corner case exists in sctp_sf_do_5_2_6_stale(): when a Stale Cookie ERROR is received, the association is rolled back from COOKIE_ECHOED to COOKIE_WAIT. In this scenario, user data may already have been queued and even bundled with the COOKIE-ECHO chunk. During the rollback, sctp_stream_update() frees the old stream table and installs a new one, but it does not invalidate stream->out_curr. As a result, out_curr may still point to a freed sctp_stream_out entry from the previous stream state. Later, SCTP scheduler dequeue paths (FCFS, RR, PRIO, etc.) rely on stream->out_curr->ext, which can lead to use-after-free once the old stream state has been released via sctp_stream_free(). This results in crashes such as (reported by Yuqi): BUG: KASAN: slab-use-after-free in sctp_sched_fcfs_dequeue+0x13a/0x140 Read of size 8 at addr ff1100004d4d3208 by task mini_poc/9312 CPU: 1 UID: 1001 PID: 9312 Comm: mini_poc Not tainted 7.1.0-rc1-00305-gbd3a4795d574 #5 PREEMPT(full) sctp_sched_fcfs_dequeue+0x13a/0x140 sctp_outq_flush+0x1603/0x33e0 sctp_do_sm+0x31c9/0x5d30 sctp_assoc_bh_rcv+0x392/0x6f0 sctp_inq_push+0x1db/0x270 sctp_rcv+0x138d/0x3c10 Fix this by fully purging the association outqueue when handling the Stale Cookie case. This ensures all pending transmit and retransmit state is dropped, and any scheduler cached pointers are invalidated, making it safe to rebuild stream state during COOKIE_WAIT restart. Updating only stream->out_curr would be insufficient, since queued and retransmittable data would still reference the old stream state and trigger later use-after-free in dequeue paths. Fixes: 5bbbbe32a431 ("sctp: introduce stream scheduler foundations") Reported-by: Yuan Tan Reported-by: Yifan Wu Reported-by: Juefei Pu Reported-by: Zhengchuan Liang Reported-by: Xin Liu Reported-by: Yuqi Xu Reported-by: Ren Wei Signed-off-by: Xin Long Link: https://patch.msgid.link/94318159b9052907a6cbb7256aee8b5f8dfbfccb.1780510304.git.lucien.xin@gmail.com Signed-off-by: Jakub Kicinski --- net/sctp/sm_statefuns.c | 6 +----- 1 file changed, 1 insertion(+), 5 deletions(-) diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 8e89a870780c..9b23c11cbb9e 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -2598,11 +2598,7 @@ static enum sctp_disposition sctp_sf_do_5_2_6_stale( */ sctp_add_cmd_sf(commands, SCTP_CMD_DEL_NON_PRIMARY, SCTP_NULL()); - /* If we've sent any data bundled with COOKIE-ECHO we will need to - * resend - */ - sctp_add_cmd_sf(commands, SCTP_CMD_T1_RETRAN, - SCTP_TRANSPORT(asoc->peer.primary_path)); + sctp_add_cmd_sf(commands, SCTP_CMD_PURGE_OUTQUEUE, SCTP_NULL()); /* Cast away the const modifier, as we want to just * rerun it through as a sideffect. From 3c94f241f776562c489876ff506f366224565c21 Mon Sep 17 00:00:00 2001 From: Sechang Lim Date: Wed, 3 Jun 2026 16:27:33 +0000 Subject: [PATCH 295/329] udp: clear skb->dev before running a sockmap verdict On the UDP receive path skb->dev is repurposed as dev_scratch (the truesize/state cache set by udp_set_dev_scratch()), through the union { struct net_device *dev; unsigned long dev_scratch; } in sk_buff. When a UDP socket is in a sockmap, sk_data_ready is sk_psock_verdict_data_ready(), which calls udp_read_skb() -> recv_actor() (sk_psock_verdict_recv) to run the attached SK_SKB verdict program in softirq. If that program calls a socket-lookup helper (bpf_sk_lookup_tcp/udp, bpf_skc_lookup_tcp), bpf_skc_lookup() does: if (skb->dev) caller_net = dev_net(skb->dev); skb->dev still holds the dev_scratch value (a non-NULL integer), so dev_net() dereferences it as a struct net_device * and the kernel takes a general protection fault on a non-canonical address in softirq: Oops: general protection fault, probably for non-canonical address 0x1010000800004a0 CPU: 1 UID: 0 PID: 1406 Comm: syz.2.19 Not tainted 7.1.0-rc6 #1 PREEMPT(full) RIP: 0010:bpf_skc_lookup net/core/filter.c:7033 [inline] RIP: 0010:bpf_sk_lookup+0x45/0x160 net/core/filter.c:7047 Call Trace: bpf_prog_4675cb904b7071f8+0x12e/0x14e bpf_prog_run_pin_on_cpu+0xc6/0x1f0 sk_psock_verdict_recv+0x1ba/0x350 udp_read_skb+0x31a/0x370 sk_psock_verdict_data_ready+0x2e3/0x600 __udp_enqueue_schedule_skb+0x4c8/0x650 udpv6_queue_rcv_one_skb+0x3ec/0x740 udp6_unicast_rcv_skb+0x11d/0x140 ip6_protocol_deliver_rcu+0x61e/0x950 ip6_input_finish+0xa9/0x150 NF_HOOK+0x286/0x2f0 ip6_input+0x117/0x220 NF_HOOK+0x286/0x2f0 __netif_receive_skb+0x85/0x200 process_backlog+0x374/0x9a0 __napi_poll+0x4f/0x1c0 net_rx_action+0x3b0/0x770 handle_softirqs+0x15a/0x460 do_softirq+0x57/0x80 The rmem charge that dev_scratch accounted for is released by skb_recv_udp() on dequeue, just above, so the scratch is dead by the time recv_actor() runs. Clear skb->dev so bpf_skc_lookup() falls back to sock_net(skb->sk), which skb_set_owner_sk_safe() set just above. Fixes: 965b57b469a5 ("net: Introduce a new proto_ops ->read_skb()") Cc: stable@vger.kernel.org Signed-off-by: Sechang Lim Reviewed-by: Jiayuan Chen Reviewed-by: Eric Dumazet Link: https://patch.msgid.link/20260603162737.697215-1-rhkrqnwk98@gmail.com Signed-off-by: Jakub Kicinski --- net/ipv4/udp.c | 8 ++++++++ 1 file changed, 8 insertions(+) diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index 0ac2bf4f8759..70f6cbd4ef73 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -2011,6 +2011,14 @@ int udp_read_skb(struct sock *sk, skb_read_actor_t recv_actor) } WARN_ON_ONCE(!skb_set_owner_sk_safe(skb, sk)); + + /* + * skb->dev still aliases the UDP rx dev_scratch (its charge was freed + * on dequeue above); a sockmap verdict program may deref it via + * bpf_sk_lookup_*(), so clear it -> bpf_skc_lookup() uses skb->sk + */ + skb->dev = NULL; + return recv_actor(sk, skb); } From b6197b386677ae5268d4702e23849d9ad53051ad Mon Sep 17 00:00:00 2001 From: Jakub Kicinski Date: Wed, 3 Jun 2026 12:58:45 -0700 Subject: [PATCH 296/329] Reapply "bnxt_en: bring back rtnl_lock() in the bnxt_open() path" This reverts commit 850d9248d2eac662f869c766a598c877690c74e5. This reapplies commit 325eb217e41f ("bnxt_en: bring back rtnl_lock() in the bnxt_open() path"). Breno reports a lockdep warning in bnxt. During FW reset the driver may end up calling netif_set_real_num_tx_queues() (if queue count changes), so calls to bnxt_open() still require rtnl_lock. net/sched/sch_generic.c:1416 suspicious rcu_dereference_protected() usage! dev_qdisc_change_real_num_tx+0x54/0xe0 netif_set_real_num_tx_queues+0x4ed/0xa80 __bnxt_open_nic+0x9cb/0x3490 bnxt_open+0x1cb/0x370 bnxt_fw_reset_task+0x80d/0x1e80 process_scheduled_works+0x9c1/0x13b0 The reverted commit was just an optimization / experiment so let's go back to taking the lock. Reported-by: Breno Leitao Link: https://lore.kernel.org/ah726OtFX-Qw3U-R@gmail.com Fixes: 850d9248d2ea ("Revert "bnxt_en: bring back rtnl_lock() in the bnxt_open() path"") Acked-by: Stanislav Fomichev Reviewed-by: Michael Chan Reviewed-by: Breno Leitao Link: https://patch.msgid.link/20260603195845.2574426-1-kuba@kernel.org Signed-off-by: Jakub Kicinski --- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 36 ++++++++++++++++++----- 1 file changed, 29 insertions(+), 7 deletions(-) diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c index 008c34cff7b4..35e1f8f663c7 100644 --- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c +++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c @@ -14388,13 +14388,28 @@ static void bnxt_unlock_sp(struct bnxt *bp) netdev_unlock(bp->dev); } +/* Same as bnxt_lock_sp() with additional rtnl_lock */ +static void bnxt_rtnl_lock_sp(struct bnxt *bp) +{ + clear_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + rtnl_lock(); + netdev_lock(bp->dev); +} + +static void bnxt_rtnl_unlock_sp(struct bnxt *bp) +{ + set_bit(BNXT_STATE_IN_SP_TASK, &bp->state); + netdev_unlock(bp->dev); + rtnl_unlock(); +} + /* Only called from bnxt_sp_task() */ static void bnxt_reset(struct bnxt *bp, bool silent) { - bnxt_lock_sp(bp); + bnxt_rtnl_lock_sp(bp); if (test_bit(BNXT_STATE_OPEN, &bp->state)) bnxt_reset_task(bp, silent); - bnxt_unlock_sp(bp); + bnxt_rtnl_unlock_sp(bp); } /* Only called from bnxt_sp_task() */ @@ -14402,9 +14417,9 @@ static void bnxt_rx_ring_reset(struct bnxt *bp) { int i; - bnxt_lock_sp(bp); + bnxt_rtnl_lock_sp(bp); if (!test_bit(BNXT_STATE_OPEN, &bp->state)) { - bnxt_unlock_sp(bp); + bnxt_rtnl_unlock_sp(bp); return; } /* Disable and flush TPA before resetting the RX ring */ @@ -14443,7 +14458,7 @@ static void bnxt_rx_ring_reset(struct bnxt *bp) } if (bp->flags & BNXT_FLAG_TPA) bnxt_set_tpa(bp, true); - bnxt_unlock_sp(bp); + bnxt_rtnl_unlock_sp(bp); } static void bnxt_fw_fatal_close(struct bnxt *bp) @@ -15358,15 +15373,17 @@ static void bnxt_fw_reset_task(struct work_struct *work) bp->fw_reset_state = BNXT_FW_RESET_STATE_OPENING; fallthrough; case BNXT_FW_RESET_STATE_OPENING: - while (!netdev_trylock(bp->dev)) { + while (!rtnl_trylock()) { bnxt_queue_fw_reset_work(bp, HZ / 10); return; } + netdev_lock(bp->dev); rc = bnxt_open(bp->dev); if (rc) { netdev_err(bp->dev, "bnxt_open() failed during FW reset\n"); bnxt_fw_reset_abort(bp, rc); netdev_unlock(bp->dev); + rtnl_unlock(); goto ulp_start; } @@ -15386,6 +15403,7 @@ static void bnxt_fw_reset_task(struct work_struct *work) bnxt_dl_health_fw_status_update(bp, true); } netdev_unlock(bp->dev); + rtnl_unlock(); bnxt_ulp_start(bp); bnxt_reenable_sriov(bp); netdev_lock(bp->dev); @@ -16379,7 +16397,7 @@ static int bnxt_queue_start(struct net_device *dev, rc); napi_enable_locked(&bnapi->napi); bnxt_db_nq_arm(bp, &cpr->cp_db, cpr->cp_raw_cons); - bnxt_reset_task(bp, true); + netif_close(dev); return rc; } @@ -17230,6 +17248,7 @@ static int bnxt_resume(struct device *device) struct bnxt *bp = netdev_priv(dev); int rc = 0; + rtnl_lock(); netdev_lock(dev); rc = pci_enable_device(bp->pdev); if (rc) { @@ -17274,6 +17293,7 @@ static int bnxt_resume(struct device *device) resume_exit: netdev_unlock(bp->dev); + rtnl_unlock(); if (!rc) { bnxt_ulp_start(bp); bnxt_reenable_sriov(bp); @@ -17445,6 +17465,7 @@ static void bnxt_io_resume(struct pci_dev *pdev) int err; netdev_info(bp->dev, "PCI Slot Resume\n"); + rtnl_lock(); netdev_lock(netdev); err = bnxt_hwrm_func_qcaps(bp); @@ -17462,6 +17483,7 @@ static void bnxt_io_resume(struct pci_dev *pdev) netif_device_attach(netdev); netdev_unlock(netdev); + rtnl_unlock(); if (!err) { bnxt_ulp_start(bp); bnxt_reenable_sriov(bp); From 00f547e0dfecf83014fb32bcba587c6b684c1362 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sat, 23 May 2026 19:51:59 +0000 Subject: [PATCH 297/329] accel/ethosu: fix IFM region index out-of-bounds in command stream parser NPU_SET_IFM_REGION extracts the region index with param & 0x7f, giving a maximum value of 127. However region_size[] and output_region[] in struct ethosu_validated_cmdstream_info are both sized to NPU_BASEP_REGION_MAX (8), giving valid indices [0..7]. Every other region assignment in the same switch uses param & 0x7: NPU_SET_OFM_REGION: st.ofm.region = param & 0x7; NPU_SET_IFM2_REGION: st.ifm2.region = param & 0x7; NPU_SET_WEIGHT_REGION: st.weight[0].region = param & 0x7; NPU_SET_SCALE_REGION: st.scale[0].region = param & 0x7; The 0x7f mask on IFM is inconsistent and appears to be a typo. feat_matrix_length() and calc_sizes() use the region index directly as an array subscript into the kzalloc'd info struct: info->region_size[fm->region] = max(...); A userspace caller supplying NPU_SET_IFM_REGION with param > 7 causes a write up to 127*8 = 1016 bytes past the start of region_size[], corrupting adjacent kernel heap data. Fix by applying the same & 0x7 mask used by all other region assignments. Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Link: https://patch.msgid.link/20260523195159.55801-1-meatuni001@gmail.com Signed-off-by: Rob Herring (Arm) --- drivers/accel/ethosu/ethosu_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ethosu/ethosu_gem.c b/drivers/accel/ethosu/ethosu_gem.c index 7994e7073903..ced99cf9cdfc 100644 --- a/drivers/accel/ethosu/ethosu_gem.c +++ b/drivers/accel/ethosu/ethosu_gem.c @@ -464,7 +464,7 @@ static int ethosu_gem_cmdstream_copy_and_validate(struct drm_device *ddev, st.ifm.broadcast = param; break; case NPU_SET_IFM_REGION: - st.ifm.region = param & 0x7f; + st.ifm.region = param & 0x7; break; case NPU_SET_IFM_WIDTH0_M1: st.ifm.width0 = param; From ef911805d86a05363d3ec2fa9835a41def83bb7e Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sat, 23 May 2026 21:07:52 +0000 Subject: [PATCH 298/329] accel/ethosu: reject NPU_OP_RESIZE commands from userspace NPU_OP_RESIZE is a U85-only command that the driver does not yet implement. The existing WARN_ON(1) placeholder fires unconditionally whenever userspace submits this command via DRM_IOCTL_ETHOSU_GEM_CREATE, causing unbounded kernel log spam. If panic_on_warn is set the kernel panics, giving any unprivileged user with access to the DRM device a trivial denial-of-service primitive. Replace the WARN_ON(1) with an explicit -EINVAL return so the ioctl rejects the command before it reaches hardware. Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Link: https://patch.msgid.link/20260523210840.92039-2-meatuni001@gmail.com Signed-off-by: Rob Herring (Arm) --- drivers/accel/ethosu/ethosu_gem.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/accel/ethosu/ethosu_gem.c b/drivers/accel/ethosu/ethosu_gem.c index ced99cf9cdfc..863cdadb137a 100644 --- a/drivers/accel/ethosu/ethosu_gem.c +++ b/drivers/accel/ethosu/ethosu_gem.c @@ -431,8 +431,7 @@ static int ethosu_gem_cmdstream_copy_and_validate(struct drm_device *ddev, return ret; break; case NPU_OP_RESIZE: // U85 only - WARN_ON(1); // TODO - break; + return -EINVAL; case NPU_SET_KERNEL_WIDTH_M1: st.ifm.width = param; break; From e703843f242b28e35ac79408de571ae110c740b5 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sat, 23 May 2026 21:07:53 +0000 Subject: [PATCH 299/329] accel/ethosu: fix wrong weight index in NPU_SET_SCALE1_LENGTH on U85 On non-U65 hardware (e.g. U85), opcode 0x4093 is NPU_SET_WEIGHT2_LENGTH. The BASE handler for the same opcode correctly assigns to st.weight[2].base, but the LENGTH handler mistakenly assigns cmds[1] to st.weight[1].length instead of st.weight[2].length. This leaves weight[2].length at its initialised sentinel value of 0xffffffff and corrupts weight[1].length with the user-supplied value, breaking the software bounds-check state for both weight buffers on U85. Fix the index to match the BASE handler. Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Link: https://patch.msgid.link/20260523210840.92039-3-meatuni001@gmail.com Signed-off-by: Rob Herring (Arm) --- drivers/accel/ethosu/ethosu_gem.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/accel/ethosu/ethosu_gem.c b/drivers/accel/ethosu/ethosu_gem.c index 863cdadb137a..52b6a8752c75 100644 --- a/drivers/accel/ethosu/ethosu_gem.c +++ b/drivers/accel/ethosu/ethosu_gem.c @@ -598,7 +598,7 @@ static int ethosu_gem_cmdstream_copy_and_validate(struct drm_device *ddev, if (ethosu_is_u65(edev)) st.scale[1].length = cmds[1]; else - st.weight[1].length = cmds[1]; + st.weight[2].length = cmds[1]; break; case NPU_SET_WEIGHT3_BASE: st.weight[3].base = addr; From ee6d9b6e51626f259c6f0e38d94f91be4fd14754 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sun, 24 May 2026 10:37:10 +0000 Subject: [PATCH 300/329] accel/ethosu: fix arithmetic issues in dma_length() dma_length() derives DMA region usage from command stream values and updates region_size[]: len = ((len + stride[0]) * size0 + stride[1]) * size1 region_size[region] = max(..., len + dma->offset) Several arithmetic issues can corrupt the derived region size: - signed stride values may underflow when added to len - intermediate multiplications may overflow - len + dma->offset may overflow during region_size updates - dma_length() error returns were not validated by the caller region_size[] is later used by ethosu_job.c to validate command stream accesses against GEM buffer sizes. Arithmetic wraparound can therefore under-report region usage and bypass the bounds validation. Fix by validating signed additions, using overflow helpers for multiplications and offset updates, and propagating dma_length() failures to the caller. Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Link: https://patch.msgid.link/20260524103710.47397-1-meatuni001@gmail.com Signed-off-by: Rob Herring (Arm) --- drivers/accel/ethosu/ethosu_gem.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-) diff --git a/drivers/accel/ethosu/ethosu_gem.c b/drivers/accel/ethosu/ethosu_gem.c index 52b6a8752c75..27c5fdbe9a5d 100644 --- a/drivers/accel/ethosu/ethosu_gem.c +++ b/drivers/accel/ethosu/ethosu_gem.c @@ -2,6 +2,7 @@ /* Copyright 2025 Arm, Ltd. */ #include +#include #include #include @@ -164,16 +165,26 @@ static u64 dma_length(struct ethosu_validated_cmdstream_info *info, u64 len = dma->len; if (mode >= 1) { + if (dma->stride[0] < 0 && (u64)(-dma->stride[0]) > len) + return U64_MAX; len += dma->stride[0]; - len *= dma_st->size0; + if (check_mul_overflow(len, (u64)dma_st->size0, &len)) + return U64_MAX; } if (mode == 2) { + if (dma->stride[1] < 0 && (u64)(-dma->stride[1]) > len) + return U64_MAX; len += dma->stride[1]; - len *= dma_st->size1; + if (check_mul_overflow(len, (u64)dma_st->size1, &len)) + return U64_MAX; + } + if (dma->region >= 0) { + u64 end; + + if (check_add_overflow(len, dma->offset, &end)) + return U64_MAX; + info->region_size[dma->region] = max(info->region_size[dma->region], end); } - if (dma->region >= 0) - info->region_size[dma->region] = max(info->region_size[dma->region], - len + dma->offset); return len; } @@ -395,6 +406,8 @@ static int ethosu_gem_cmdstream_copy_and_validate(struct drm_device *ddev, case NPU_OP_DMA_START: srclen = dma_length(info, &st.dma, &st.dma.src); dstlen = dma_length(info, &st.dma, &st.dma.dst); + if (srclen == U64_MAX || dstlen == U64_MAX) + return -EINVAL; if (st.dma.dst.region >= 0) info->output_region[st.dma.dst.region] = true; From d9d021218162b6c4fe0bdf42b2b340f1aae23a12 Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sun, 24 May 2026 13:03:19 +0000 Subject: [PATCH 301/329] accel/ethosu: reject DMA commands with uninitialized length cmd_state_init() initializes the command state with memset(0xff), leaving dma->len at U64_MAX to signal missing setup. The only setter is NPU_SET_DMA0_LEN; if userspace omits this command and issues NPU_OP_DMA_START, dma->len remains U64_MAX. In dma_length(), a positive stride added to U64_MAX wraps to a small value. With size0 == 1, check_mul_overflow() does not trigger and dma_length() returns 0 instead of U64_MAX. The caller's U64_MAX check then passes, region_size[] stays 0, and the bounds check in ethosu_job.c is bypassed, allowing hardware to execute DMA with stale physical addresses. Fix by checking for U64_MAX at the start of dma_length() before any arithmetic, consistent with the sentinel value used throughout the driver to detect uninitialized fields. Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Link: https://patch.msgid.link/20260524130319.12747-1-meatuni001@gmail.com Signed-off-by: Rob Herring (Arm) --- drivers/accel/ethosu/ethosu_gem.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/accel/ethosu/ethosu_gem.c b/drivers/accel/ethosu/ethosu_gem.c index 27c5fdbe9a5d..2cb7964ddfa5 100644 --- a/drivers/accel/ethosu/ethosu_gem.c +++ b/drivers/accel/ethosu/ethosu_gem.c @@ -164,6 +164,9 @@ static u64 dma_length(struct ethosu_validated_cmdstream_info *info, s8 mode = dma_st->mode; u64 len = dma->len; + if (len == U64_MAX) + return U64_MAX; + if (mode >= 1) { if (dma->stride[0] < 0 && (u64)(-dma->stride[0]) > len) return U64_MAX; From c0837b9cf6eabbad8b8cbddaff1a46a6d0a2e29d Mon Sep 17 00:00:00 2001 From: Muhammad Bilal Date: Sat, 23 May 2026 19:08:43 +0000 Subject: [PATCH 302/329] accel/ethosu: fix OOB write in ethosu_gem_cmdstream_copy_and_validate() The command stream parsing loop increments the index variable a second time when a 64-bit command word is encountered (bit 14 set), but does not re-check the loop bound before writing the second word: for (i = 0; i < size / 4; i++) { bocmds[i] = cmds[0]; if (cmd & 0x4000) { i++; bocmds[i] = cmds[1]; /* unchecked */ } } The buffer bocmds is backed by a DMA allocation of exactly size bytes from drm_gem_dma_create(ddev, size), giving valid indices [0, size/4-1]. When i == size/4 - 1 on entry to an iteration and bit 14 of cmds[0] is set, bocmds[size/4-1] is written in bounds, i is then incremented to size/4, and bocmds[size/4] writes four bytes past the end of the allocation. Userspace controls both the buffer contents and the size argument via the ioctl, making this a userspace-triggerable heap out-of-bounds write. Fix by checking the incremented index against the buffer bound before the second write and returning -EINVAL if the buffer is too small to contain the extended command. Fixes: 5a5e9c0228e6 ("accel: Add Arm Ethos-U NPU driver") Cc: stable@vger.kernel.org Signed-off-by: Muhammad Bilal Link: https://patch.msgid.link/20260523190843.33977-1-meatuni001@gmail.com Signed-off-by: Rob Herring (Arm) --- drivers/accel/ethosu/ethosu_gem.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/accel/ethosu/ethosu_gem.c b/drivers/accel/ethosu/ethosu_gem.c index 2cb7964ddfa5..3401883e207f 100644 --- a/drivers/accel/ethosu/ethosu_gem.c +++ b/drivers/accel/ethosu/ethosu_gem.c @@ -401,6 +401,8 @@ static int ethosu_gem_cmdstream_copy_and_validate(struct drm_device *ddev, return -EFAULT; i++; + if (i >= size / 4) + return -EINVAL; bocmds[i] = cmds[1]; addr = cmd_to_addr(cmds); } From 2b5ff4db5d7aa5b981d966df02e687f79ad7b311 Mon Sep 17 00:00:00 2001 From: Kyle Zeng Date: Fri, 5 Jun 2026 01:02:04 -0700 Subject: [PATCH 303/329] ALSA: seq: dummy: fix UMP event stack overread The dummy sequencer port forwards events by copying an incoming struct snd_seq_event into a stack temporary, rewriting source and destination, and dispatching the temporary to subscribers. That legacy event storage is smaller than struct snd_seq_ump_event. When a UMP event reaches the dummy client, the copy leaves the UMP flag set but only provides legacy-sized stack storage. The subscriber delivery path then uses snd_seq_event_packet_size() and copies a UMP-sized packet from that stack object, reading past the end of the temporary. Use the existing union __snd_seq_event storage and copy the packet size reported for the incoming event before rewriting the common routing fields. This preserves the full UMP packet for UMP events while keeping legacy event handling unchanged. Fixes: 32cb23a0f911 ("ALSA: seq: dummy: Allow UMP conversion") Signed-off-by: Kyle Zeng Link: https://patch.msgid.link/20260605080204.32045-1-kylebot@openai.com Signed-off-by: Takashi Iwai --- sound/core/seq/seq_dummy.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/sound/core/seq/seq_dummy.c b/sound/core/seq/seq_dummy.c index af45f328ae99..8abe80985dad 100644 --- a/sound/core/seq/seq_dummy.c +++ b/sound/core/seq/seq_dummy.c @@ -9,6 +9,7 @@ #include #include #include "seq_clientmgr.h" +#include "seq_memory.h" #include #include @@ -81,19 +82,21 @@ dummy_input(struct snd_seq_event *ev, int direct, void *private_data, int atomic, int hop) { struct snd_seq_dummy_port *p; - struct snd_seq_event tmpev; + union __snd_seq_event tmpev; + size_t size; p = private_data; if (ev->source.client == SNDRV_SEQ_CLIENT_SYSTEM || ev->type == SNDRV_SEQ_EVENT_KERNEL_ERROR) return 0; /* ignore system messages */ - tmpev = *ev; + size = snd_seq_event_packet_size(ev); + memcpy(&tmpev, ev, size); if (p->duplex) - tmpev.source.port = p->connect; + tmpev.legacy.source.port = p->connect; else - tmpev.source.port = p->port; - tmpev.dest.client = SNDRV_SEQ_ADDRESS_SUBSCRIBERS; - return snd_seq_kernel_client_dispatch(p->client, &tmpev, atomic, hop); + tmpev.legacy.source.port = p->port; + tmpev.legacy.dest.client = SNDRV_SEQ_ADDRESS_SUBSCRIBERS; + return snd_seq_kernel_client_dispatch(p->client, &tmpev.legacy, atomic, hop); } /* From 6f6947b2387e94e405f80d472f8a189bfbf2bd6c Mon Sep 17 00:00:00 2001 From: Tony Luck Date: Thu, 4 Jun 2026 21:46:49 -0700 Subject: [PATCH 304/329] x86/resctrl: Only check Intel systems for SNC topology_num_nodes_per_package() reports values greater than one on certain AMD systems resulting in resctrl's Intel model specific SNC detection printing the confusing message: "CoD enabled system? Resctrl not supported" Add a check for Intel systems before looking at the topology. [ reinette: Add Closes tag, fix tag typos, rework changelog ] Fixes: 59674fc9d0bf ("x86/resctrl: Fix SNC detection") Reported-by: Babu Moger Signed-off-by: Tony Luck Signed-off-by: Reinette Chatre Signed-off-by: Ingo Molnar Tested-by: Babu Moger Link: https://patch.msgid.link/9849330f45ac86344cc5ac54df2d313906d70bc4.1780634584.git.reinette.chatre@intel.com Closes: https://lore.kernel.org/lkml/37ac0376-43a3-4283-a3d5-4d57b3bec578@amd.com/ --- arch/x86/kernel/cpu/resctrl/monitor.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/resctrl/monitor.c b/arch/x86/kernel/cpu/resctrl/monitor.c index 9bd87bae4983..59215fef3924 100644 --- a/arch/x86/kernel/cpu/resctrl/monitor.c +++ b/arch/x86/kernel/cpu/resctrl/monitor.c @@ -377,7 +377,12 @@ static const struct x86_cpu_id snc_cpu_ids[] __initconst = { static __init int snc_get_config(void) { - int ret = topology_num_nodes_per_package(); + int ret; + + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + return 1; + + ret = topology_num_nodes_per_package(); if (ret > 1 && !x86_match_cpu(snc_cpu_ids)) { pr_warn("CoD enabled system? Resctrl not supported\n"); From 13031fb6b8357fbbcded2a7f4cba73e4781ee594 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Mon, 1 Jun 2026 23:53:26 +0900 Subject: [PATCH 305/329] KVM: arm64: vgic-its: Drop the translation cache reference only for the erased entry vgic_its_invalidate_cache() walks the per-ITS translation cache with xa_for_each() and drops the cache's reference on each entry with vgic_put_irq(). It puts the iterated pointer, though, rather than the value returned by xa_erase(). The function is called from contexts that do not exclude one another: the ITS command handlers hold its_lock, the GITS_CTLR write path holds cmd_lock, and the path that clears EnableLPIs in a redistributor's GICR_CTLR holds neither. Two or more of them can drain the same cache concurrently, and if each one observes the same entry, erases it and then puts it, the single reference the cache holds on that entry is dropped more than once. The entry can then be freed while an ITE still maps it. xa_erase() is atomic and returns the previous entry, so put only the entry that this context actually removed. The cache reference is then dropped exactly once per entry even when the invalidations run concurrently, and the behavior is unchanged when only one context runs. Fixes: 8201d1028caa ("KVM: arm64: vgic-its: Maintain a translation cache per ITS") Signed-off-by: Hyunwoo Kim Reviewed-by: Oliver Upton Link: https://patch.msgid.link/ah2c5lu4JbUg7dj-@v4bel Signed-off-by: Marc Zyngier Cc: stable@vger.kernel.org --- arch/arm64/kvm/vgic/vgic-its.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c index 1d7e5d560af4..1e3706ac3b8e 100644 --- a/arch/arm64/kvm/vgic/vgic-its.c +++ b/arch/arm64/kvm/vgic/vgic-its.c @@ -597,8 +597,10 @@ static void vgic_its_invalidate_cache(struct vgic_its *its) unsigned long idx; xa_for_each(&its->translation_cache, idx, irq) { - xa_erase(&its->translation_cache, idx); - vgic_put_irq(kvm, irq); + /* Only the context that erases the entry drops its cache ref. */ + irq = xa_erase(&its->translation_cache, idx); + if (irq) + vgic_put_irq(kvm, irq); } } From f2ca45b50d4216c9cc7ffabf50d9ad1932209251 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Wed, 3 Jun 2026 21:09:33 +0900 Subject: [PATCH 306/329] KVM: arm64: Take the SRCU lock for page table walks in fault injection and AT emulation walk_s1() and kvm_walk_nested_s2() expect to be called while holding kvm->srcu to guard against memslot changes. While this is generally the case, __kvm_at_s12() and __kvm_find_s1_desc_level() call into the respective walkers without taking kvm->srcu. Fix by acquiring kvm->srcu prior to the table walk in both instances. Cc: stable@vger.kernel.org Fixes: 50f77dc87f13 ("KVM: arm64: Populate level on S1PTW SEA injection") Fixes: be04cebf3e78 ("KVM: arm64: nv: Add emulation of AT S12E{0,1}{R,W}") Suggested-by: Oliver Upton Signed-off-by: Hyunwoo Kim Reviewed-by: Oliver Upton Link: https://patch.msgid.link/aiAZfdeyanIvP8SD@v4bel Signed-off-by: Marc Zyngier --- arch/arm64/kvm/at.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/arch/arm64/kvm/at.c b/arch/arm64/kvm/at.c index 9f8f0ae8e86e..889c2c15d7bd 100644 --- a/arch/arm64/kvm/at.c +++ b/arch/arm64/kvm/at.c @@ -1569,7 +1569,8 @@ int __kvm_at_s12(struct kvm_vcpu *vcpu, u32 op, u64 vaddr) /* Do the stage-2 translation */ ipa = (par & GENMASK_ULL(47, 12)) | (vaddr & GENMASK_ULL(11, 0)); out.esr = 0; - ret = kvm_walk_nested_s2(vcpu, ipa, &out); + scoped_guard(srcu, &vcpu->kvm->srcu) + ret = kvm_walk_nested_s2(vcpu, ipa, &out); if (ret < 0) return ret; @@ -1665,7 +1666,8 @@ int __kvm_find_s1_desc_level(struct kvm_vcpu *vcpu, u64 va, u64 ipa, int *level) } /* Walk the guest's PT, looking for a match along the way */ - ret = walk_s1(vcpu, &wi, &wr, va); + scoped_guard(srcu, &vcpu->kvm->srcu) + ret = walk_s1(vcpu, &wi, &wr, va); switch (ret) { case -EINTR: /* We interrupted the walk on a match, return the level */ From 2914709c914101eb704e01bed2351070d4161ccf Mon Sep 17 00:00:00 2001 From: Suraj Kandpal Date: Sun, 17 May 2026 08:17:09 +0530 Subject: [PATCH 307/329] Revert "drm/i915/backlight: Remove try_vesa_interface" This reverts commit 40d2f5820951dee818d05c14677277048bd85f9f. Removing the try_vesa_interface gate caused a backlight regression on panels whose VBT correctly reports INTEL_BACKLIGHT_DISPLAY_DDI and whose PWM path is the actual backlight control, but whose DPCD optimistically advertises DP_EDP_BACKLIGHT_AUX_ENABLE_CAP / _BRIGHTNESS_AUX_SET_CAP. After the commit such panels silently bind to the VESA AUX backlight funcs; AUX writes complete but the panel ignores them, leaving brightness stuck (no-op backlight). Observed on at least KBL and TGL eDP setups. Signed-off-by: Suraj Kandpal Reviewed-by: Ankit Nautiyal Link: https://patch.msgid.link/20260517024709.1016121-1-suraj.kandpal@intel.com (cherry picked from commit f30fddb4402313aa5301a74d721638d343395269) Signed-off-by: Tvrtko Ursulin --- .../drm/i915/display/intel_dp_aux_backlight.c | 19 ++++++++++++------- 1 file changed, 12 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c index a8d56ebf06a2..7a6c07f6aaeb 100644 --- a/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c +++ b/drivers/gpu/drm/i915/display/intel_dp_aux_backlight.c @@ -691,10 +691,9 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *connector) struct intel_dp *intel_dp = intel_attached_dp(connector); struct drm_device *dev = connector->base.dev; struct intel_panel *panel = &connector->panel; - bool try_intel_interface = false; + bool try_intel_interface = false, try_vesa_interface = false; - /* - * Check the VBT and user's module parameters to figure out which + /* Check the VBT and user's module parameters to figure out which * interfaces to probe */ switch (display->params.enable_dpcd_backlight) { @@ -703,6 +702,7 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *connector) case INTEL_DP_AUX_BACKLIGHT_AUTO: switch (panel->vbt.backlight.type) { case INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE: + try_vesa_interface = true; break; case INTEL_BACKLIGHT_DISPLAY_DDI: try_intel_interface = true; @@ -715,12 +715,20 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *connector) if (panel->vbt.backlight.type != INTEL_BACKLIGHT_VESA_EDP_AUX_INTERFACE) try_intel_interface = true; + try_vesa_interface = true; + break; + case INTEL_DP_AUX_BACKLIGHT_FORCE_VESA: + try_vesa_interface = true; break; case INTEL_DP_AUX_BACKLIGHT_FORCE_INTEL: try_intel_interface = true; break; } + /* For eDP 1.5 and above we are supposed to use VESA interface for brightness control */ + if (intel_dp->edp_dpcd[0] >= DP_EDP_15) + try_vesa_interface = true; + /* * Since Intel has their own backlight control interface, the majority of machines out there * using DPCD backlight controls with Intel GPUs will be using this interface as opposed to @@ -733,9 +741,6 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *connector) * panel with Intel's OUI - which is also required for us to be able to detect Intel's * backlight interface at all. This means that the only sensible way for us to detect both * interfaces is to probe for Intel's first, and VESA's second. - * - * Also there is a chance some VBTs may advertise false Intel backlight support even if the - * TCON DPCD says otherwise. This means we keep VESA interface as fallback in that case. */ if (try_intel_interface && intel_dp->edp_dpcd[0] <= DP_EDP_14b && intel_dp_aux_supports_hdr_backlight(connector)) { @@ -745,7 +750,7 @@ int intel_dp_aux_init_backlight_funcs(struct intel_connector *connector) return 0; } - if (intel_dp_aux_supports_vesa_backlight(connector)) { + if (try_vesa_interface && intel_dp_aux_supports_vesa_backlight(connector)) { drm_dbg_kms(dev, "[CONNECTOR:%d:%s] Using VESA eDP backlight controls\n", connector->base.base.id, connector->base.name); panel->backlight.funcs = &intel_dp_vesa_bl_funcs; From cbaffe843a942c0d3102e0f9bce0e72b029b2594 Mon Sep 17 00:00:00 2001 From: Joey Gouly Date: Thu, 4 Jun 2026 11:54:34 +0100 Subject: [PATCH 308/329] KVM: arm64: Restore POR_EL0 access to host EL0 CPTR_EL2.E0POE was being cleared in __deactivate_cptr_traps_vhe(), which meant that any accesses to POR_EL0 from host EL0 would trap and be reported to userspace as an Illegal instruction. This would happen after running any VM, regardless if it used POE or not. Signed-off-by: Joey Gouly Link: https://sashiko.dev/#/patchset/20260602155430.2088142-1-maz@kernel.org?part=1 Link: https://patch.msgid.link/20260604105434.2297268-1-joey.gouly@arm.com Signed-off-by: Marc Zyngier Cc: stable@vger,kernel.org --- arch/arm64/kvm/hyp/include/hyp/switch.h | 2 ++ 1 file changed, 2 insertions(+) diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h index 320cd45d49c5..e9b36a3b27bb 100644 --- a/arch/arm64/kvm/hyp/include/hyp/switch.h +++ b/arch/arm64/kvm/hyp/include/hyp/switch.h @@ -181,6 +181,8 @@ static inline void __deactivate_cptr_traps_vhe(struct kvm_vcpu *vcpu) val |= CPACR_EL1_ZEN; if (cpus_have_final_cap(ARM64_SME)) val |= CPACR_EL1_SMEN; + if (cpus_have_final_cap(ARM64_HAS_S1POE)) + val |= CPACR_EL1_E0POE; write_sysreg(val, cpacr_el1); } From 70543358fa08e0f7cebc3447c3b70fe97ad7aaa8 Mon Sep 17 00:00:00 2001 From: Hyunwoo Kim Date: Fri, 5 Jun 2026 17:27:01 +0900 Subject: [PATCH 309/329] KVM: arm64: Reassign nested_mmus array behind mmu_lock kvm->arch.nested_mmus[] is walked under kvm->mmu_lock, including from the MMU notifier path (kvm_unmap_gfn_range() -> kvm_nested_s2_unmap()), which can run at any time. kvm_vcpu_init_nested() reallocates the array and frees the old buffer while holding only kvm->arch.config_lock, so such a walker can reference the freed array. Allocate the new array outside of mmu_lock, as the allocation can sleep. Under the lock, copy the existing entries, fix up the back pointers and reassign the array. Free the old buffer after dropping the lock, as kvfree() can sleep as well. Fixes: 4f128f8e1aaac ("KVM: arm64: nv: Support multiple nested Stage-2 mmu structures") Signed-off-by: Hyunwoo Kim Reviewed-by: Oliver Upton Link: https://patch.msgid.link/aiKIVVeIr1aAB1yp@v4bel Signed-off-by: Marc Zyngier Cc: stable@vger,kernel.org --- arch/arm64/kvm/nested.c | 33 ++++++++++++++++++++------------- 1 file changed, 20 insertions(+), 13 deletions(-) diff --git a/arch/arm64/kvm/nested.c b/arch/arm64/kvm/nested.c index 38f672e94087..6f7bc9a9992e 100644 --- a/arch/arm64/kvm/nested.c +++ b/arch/arm64/kvm/nested.c @@ -89,21 +89,28 @@ int kvm_vcpu_init_nested(struct kvm_vcpu *vcpu) * again, and there is no reason to affect the whole VM for this. */ num_mmus = atomic_read(&kvm->online_vcpus) * S2_MMU_PER_VCPU; - tmp = kvrealloc(kvm->arch.nested_mmus, - size_mul(sizeof(*kvm->arch.nested_mmus), num_mmus), - GFP_KERNEL_ACCOUNT | __GFP_ZERO); - if (!tmp) - return -ENOMEM; - swap(kvm->arch.nested_mmus, tmp); + if (num_mmus > kvm->arch.nested_mmus_size) { + tmp = kvcalloc(num_mmus, sizeof(*tmp), GFP_KERNEL_ACCOUNT); + if (!tmp) + return -ENOMEM; - /* - * If we went through a realocation, adjust the MMU back-pointers in - * the previously initialised kvm_pgtable structures. - */ - if (kvm->arch.nested_mmus != tmp) - for (int i = 0; i < kvm->arch.nested_mmus_size; i++) - kvm->arch.nested_mmus[i].pgt->mmu = &kvm->arch.nested_mmus[i]; + write_lock(&kvm->mmu_lock); + + if (kvm->arch.nested_mmus_size) { + memcpy(tmp, kvm->arch.nested_mmus, + size_mul(sizeof(*tmp), kvm->arch.nested_mmus_size)); + + for (int i = 0; i < kvm->arch.nested_mmus_size; i++) + tmp[i].pgt->mmu = &tmp[i]; + } + + swap(kvm->arch.nested_mmus, tmp); + + write_unlock(&kvm->mmu_lock); + + kvfree(tmp); + } for (int i = kvm->arch.nested_mmus_size; !ret && i < num_mmus; i++) ret = init_nested_s2_mmu(kvm, &kvm->arch.nested_mmus[i]); From ed46f39c47eb5530a9c161481a2080d3a869cfaf Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Cl=C3=A9ment=20L=C3=A9ger?= Date: Thu, 4 Jun 2026 09:07:13 -0700 Subject: [PATCH 310/329] io_uring/net: inherit IORING_CQE_F_BUF_MORE across bundle recv retries MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit When a bundle recv retries inside io_recv_finish(), the merge logic OR the saved cflags from the previous iteration with the cflags returned by the new iteration: cflags = req->cqe.flags | (cflags & CQE_F_MASK); Bits listed in CQE_F_MASK are inherited from the new iteration, and all other bits (notably IORING_CQE_F_BUFFER and the buffer ID) come from the saved cflags. Before this change CQE_F_MASK covered only IORING_CQE_F_SOCK_NONEMPTY and IORING_CQE_F_MORE. When using provided buffer rings (IOU_PBUF_RING_INC) with incremental mode, and bundle recv, io_kbuf_inc_commit() can leave the head ring entry partially consumed, __io_put_kbufs() then sets IORING_CQE_F_BUF_MORE on the returned cflags so userspace knows the buffer ID will be reused for subsequent completions. Because IORING_CQE_F_BUF_MORE was not in CQE_F_MASK, the merge above silently dropped it whenever the final retry iteration partially consumed the buffer, and the subsequent req->cqe.flags = cflags & ~CQE_F_MASK save would have left a stale IORING_CQE_F_BUF_MORE in the carried-over cflags had one been present. Userspace would then wrongfully advance it ring head past an entry the kernel still uses. Add IORING_CQE_F_BUF_MORE to CQE_F_MASK so it is both inherited from the new iteration into the user-visible CQE and stripped from the saved cflags between iterations. Cc: stable@vger.kernel.org Signed-off-by: Clément Léger Assisted-by: Claude:claude-opus-4.6 Fixes: ae98dbf43d75 ("io_uring/kbuf: add support for incremental buffer consumption") Link: https://patch.msgid.link/20260604160715.2482972-1-cleger@meta.com Signed-off-by: Jens Axboe --- io_uring/net.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/io_uring/net.c b/io_uring/net.c index 8df15b639358..ee848eb65ec9 100644 --- a/io_uring/net.c +++ b/io_uring/net.c @@ -842,7 +842,8 @@ int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) } /* bits to clear in old and inherit in new cflags on bundle retry */ -#define CQE_F_MASK (IORING_CQE_F_SOCK_NONEMPTY|IORING_CQE_F_MORE) +#define CQE_F_MASK (IORING_CQE_F_SOCK_NONEMPTY|IORING_CQE_F_MORE|\ + IORING_CQE_F_BUF_MORE) /* * Finishes io_recv and io_recvmsg. From 49b32ddb87a3a109afecea89e55d70f73956b8bc Mon Sep 17 00:00:00 2001 From: Oliver Upton Date: Tue, 2 Jun 2026 09:59:00 -0700 Subject: [PATCH 311/329] KVM: arm64: nv: Fix handling of XN[0] when !FEAT_XNX XN has already been extracted from its bitfield position so using FIELD_PREP() on the mask that clears XN[0] is completely broken, having the effect of unconditionally granting execute permissions... Fix the obvious mistake by manipulating the right bit. Cc: stable@vger.kernel.org Fixes: d93febe2ed2e ("KVM: arm64: nv: Forward FEAT_XNX permissions to the shadow stage-2") Reviewed-by: Wei-Lin Chang Signed-off-by: Oliver Upton Link: https://patch.msgid.link/20260602165901.52800-2-oupton@kernel.org Signed-off-by: Marc Zyngier --- arch/arm64/include/asm/kvm_nested.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/arm64/include/asm/kvm_nested.h b/arch/arm64/include/asm/kvm_nested.h index dc2957662ff2..cdf3e8422ea1 100644 --- a/arch/arm64/include/asm/kvm_nested.h +++ b/arch/arm64/include/asm/kvm_nested.h @@ -132,7 +132,7 @@ static inline bool kvm_s2_trans_exec_el0(struct kvm *kvm, struct kvm_s2_trans *t u8 xn = FIELD_GET(KVM_PTE_LEAF_ATTR_HI_S2_XN, trans->desc); if (!kvm_has_xnx(kvm)) - xn &= FIELD_PREP(KVM_PTE_LEAF_ATTR_HI_S2_XN, 0b10); + xn &= 0b10; switch (xn) { case 0b00: @@ -148,7 +148,7 @@ static inline bool kvm_s2_trans_exec_el1(struct kvm *kvm, struct kvm_s2_trans *t u8 xn = FIELD_GET(KVM_PTE_LEAF_ATTR_HI_S2_XN, trans->desc); if (!kvm_has_xnx(kvm)) - xn &= FIELD_PREP(KVM_PTE_LEAF_ATTR_HI_S2_XN, 0b10); + xn &= 0b10; switch (xn) { case 0b00: From 17f073f78fc43280891ecde8f8ec3f84f98bb37c Mon Sep 17 00:00:00 2001 From: Oliver Upton Date: Tue, 2 Jun 2026 09:59:01 -0700 Subject: [PATCH 312/329] KVM: arm64: Correctly identify executable PTEs at stage-2 KVM invalidates the I-cache before installing an executable PTE on implementations without DIC. Unfortunately, support for FEAT_XNX broke this check as KVM_PTE_LEAF_ATTR_HI_S2_XN was expanded to a bitfield. Fix it by reusing kvm_pgtable_stage2_pte_prot() and testing the abstract permission bits instead. Fixes: 2608563b466b ("KVM: arm64: Add support for FEAT_XNX stage-2 permissions") Reported-by: Sashiko (gemini/gemini-3.1-pro-preview) Signed-off-by: Oliver Upton Reviewed-by: Wei-Lin Chang Link: https://patch.msgid.link/20260602165901.52800-3-oupton@kernel.org Signed-off-by: Marc Zyngier Cc: stable@vger.kernel.org --- arch/arm64/kvm/hyp/pgtable.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c index 0c1defa5fb0f..91a7dfad6686 100644 --- a/arch/arm64/kvm/hyp/pgtable.c +++ b/arch/arm64/kvm/hyp/pgtable.c @@ -925,7 +925,9 @@ static bool stage2_pte_cacheable(struct kvm_pgtable *pgt, kvm_pte_t pte) static bool stage2_pte_executable(kvm_pte_t pte) { - return kvm_pte_valid(pte) && !(pte & KVM_PTE_LEAF_ATTR_HI_S2_XN); + enum kvm_pgtable_prot prot = kvm_pgtable_stage2_pte_prot(pte); + + return prot & (KVM_PGTABLE_PROT_UX | KVM_PGTABLE_PROT_PX); } static u64 stage2_map_walker_phys_addr(const struct kvm_pgtable_visit_ctx *ctx, From 65b09bfa8aa7ebe087093b591525385efb2d58b0 Mon Sep 17 00:00:00 2001 From: Zhou Yuhang Date: Wed, 20 May 2026 15:08:00 +0800 Subject: [PATCH 313/329] kconfig: Fix repeated include selftest expectation The err_repeated_inc test was added with an expected stderr fixture that does not match the diagnostic printed by kconfig. Running "make testconfig" currently fails in that test even though the parser reports the duplicated include correctly: [stderr] Kconfig.inc1:4: error: repeated inclusion of Kconfig.inc3 Kconfig.inc2:3: note: location of first inclusion of Kconfig.inc3 The fixture expects "Repeated" and "Location" with capital letters, but the diagnostic emitted by scripts/kconfig/util.c uses lowercase words. Update the fixture to match the real message. Fixes: 102d712ded3e ("kconfig: Error out on duplicated kconfig inclusion") Signed-off-by: Zhou Yuhang Tested-by: Nicolas Schier Reviewed-by: Nathan Chancellor Link: https://patch.msgid.link/20260520070800.2265479-1-zhouyuhang1010@163.com Signed-off-by: Nicolas Schier --- scripts/kconfig/tests/err_repeated_inc/expected_stderr | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/kconfig/tests/err_repeated_inc/expected_stderr b/scripts/kconfig/tests/err_repeated_inc/expected_stderr index 95d90d6a93c5..53071430ea7d 100644 --- a/scripts/kconfig/tests/err_repeated_inc/expected_stderr +++ b/scripts/kconfig/tests/err_repeated_inc/expected_stderr @@ -1,2 +1,2 @@ -Kconfig.inc1:4: error: Repeated inclusion of Kconfig.inc3 -Kconfig.inc2:3: note: Location of first inclusion of Kconfig.inc3 +Kconfig.inc1:4: error: repeated inclusion of Kconfig.inc3 +Kconfig.inc2:3: note: location of first inclusion of Kconfig.inc3 From 1a4f03d22fb655e5f192244fb2c87d8066fcfca2 Mon Sep 17 00:00:00 2001 From: Simona Vetter Date: Thu, 4 Jun 2026 21:44:37 +0200 Subject: [PATCH 314/329] drm/gem: Try to fix change_handle ioctl, attempt 4 [airlied: just added some comments on how to reenable] On-list because the cat is out of the bag and we're clearly not good enough to figure this out in private. The story thus far: 5e28b7b94408 ("drm: Set old handle to NULL before prime swap in change_handle") tried to fix a race condition between the gem_close and gem_change_handle ioctls, but got a few things wrong: - There's a confusion with the local variable handle, which is actually the new handle, and so the two-stage trick was actually applied to the wrong idr slot. 7164d78559b0 ("drm/gem: fix race between change_handle and handle_delete") tried to fix that by adding yet another code block, but forgot to add the error handling. Which meant we now have two paths, both kinda wrong. - dc366607c41c ("drm: Replace old pointer to new idr") tried to apply another fix, but inconsistently, again because of the handle confusion - this would be the right fix (kinda, somewhat, it's a mess) if we'd do the two-stage approach for the new handle. Except that wasn't the intent of the original fix. We also didn't have an igt merged for the original ioctl, which is a big no-go. This was attempted to address off-list in the original bugfix, and amd QA people claimed the bug was fixed now. Very clearly that's not the case. Here's my attempt to sort this out: - Rename the local variable to new_handle, the old aliasing with args->handle is just too dangerously confusing. - Merge the gem obj lookup with the two-stage idr_replace so that we avoid getting ourselves confused there. - This means we don't have a surplus temporary reference anymore, only an inherited from the idr. A concurrent gem_close on the new_handle could steal that. Fix that with the same two-stage approach create_tail uses. This is a bit overkill as documented in the comment, but I also don't trust my ability to understand this all correctly, so go with the established pattern we have from other ioctls instead for maximum paranoia. - Adjust error paths. I've tried to make the error and success paths common, because they are identical except for which handle is removed and on which we call idr_replace to (re)install the object again. But that made things messier to read, so I've left it at the more verbose version, which unfortunately hides the symmetry in the entire code flow a bit. - While at it, also replace the 7 space indent with 1 tab. And finally, because I flat out don't trust my abilities here at all anymore: - Disable the ioctl until we have the igt situation and everything else sorted out on-list and with full consensus. v2: Sashiko noticed that I didn't handle the error path for idr_replace correctly, it must be checked with IS_ERR_OR_NULL like in gem_handle_delete. So yeah, definitely should just the existing paths 1:1 because this is endless amounts of tricky. Also add the Fixes: line for the original ioctl, I forgot that too. Reported-by: DARKNAVY (@DarkNavyOrg) Signed-off-by: Simona Vetter Fixes: dc366607c41c ("drm: Replace old pointer to new idr") Cc: syzbot+d7c9eed171647e421013@syzkaller.appspotmail.com Cc: stable@vger.kernel.org Cc: Edward Adam Davis Cc: Dave Airlie Cc: Maarten Lankhorst Cc: Maxime Ripard Cc: Thomas Zimmermann Fixes: 5e28b7b94408 ("drm: Set old handle to NULL before prime swap in change_handle") Cc: David Francis Cc: Puttimet Thammasaeng Cc: Christian Koenig Fixes: 7164d78559b0 ("drm/gem: fix race between change_handle and handle_delete") Cc: Zhenghang Xiao Fixes: 5e28b7b94408 ("drm: Set old handle to NULL before prime swap in change_handle") Reviewed-by: David Francis Signed-off-by: Dave Airlie Link: https://patch.msgid.link/20260604194437.1725314-1-simona.vetter@ffwll.ch --- drivers/gpu/drm/drm_gem.c | 73 +++++++++++++++++-------------------- drivers/gpu/drm/drm_ioctl.c | 3 +- 2 files changed, 36 insertions(+), 40 deletions(-) diff --git a/drivers/gpu/drm/drm_gem.c b/drivers/gpu/drm/drm_gem.c index e12cdf91f4dc..3b2448a3a9de 100644 --- a/drivers/gpu/drm/drm_gem.c +++ b/drivers/gpu/drm/drm_gem.c @@ -1015,12 +1015,25 @@ drm_gem_open_ioctl(struct drm_device *dev, void *data, return ret; } +/* + * This ioctl is disabled for security reasons but also it failed + * to follow process in terms of adding testing in igt and verifying + * all the corner cases which made fixing security bugs in it even + * harder than necessary. + * + * To re-enable this ioctl + * 1. land working IGT tests in igt-gpu-tools that cover + * all corner cases and race conditions. + * 2. handle idr_preload + * 3. handle == 0 + * 4. handle == new_handle semantics definition. + */ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) { struct drm_gem_change_handle *args = data; - struct drm_gem_object *obj, *idrobj; - int handle, ret; + struct drm_gem_object *obj; + int new_handle, ret; if (!drm_core_check_feature(dev, DRIVER_GEM)) return -EOPNOTSUPP; @@ -1028,52 +1041,36 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data, /* idr_alloc() limitation. */ if (args->new_handle > INT_MAX) return -EINVAL; - handle = args->new_handle; + new_handle = args->new_handle; - obj = drm_gem_object_lookup(file_priv, args->handle); - if (!obj) - return -ENOENT; - - if (args->handle == handle) { - ret = 0; - goto out; - } + if (args->handle == new_handle) + return 0; mutex_lock(&file_priv->prime.lock); - spin_lock(&file_priv->table_lock); - - /* When create_tail allocs an obj idr, it needs to first alloc as NULL, - * then later replace with the correct object. This is not necessary - * here, because the only operations that could race are drm_prime - * bookkeeping, and we hold the prime lock. - */ - ret = idr_alloc(&file_priv->object_idr, obj, handle, handle + 1, + ret = idr_alloc(&file_priv->object_idr, NULL, new_handle, new_handle + 1, GFP_NOWAIT); - if (ret < 0) { - spin_unlock(&file_priv->table_lock); - goto out_unlock; - } + if (ret < 0) { + spin_unlock(&file_priv->table_lock); + goto out_unlock; + } - idrobj = idr_replace(&file_priv->object_idr, NULL, handle); - if (idrobj != obj) { - idr_replace(&file_priv->object_idr, idrobj, handle); - idr_remove(&file_priv->object_idr, args->new_handle); - spin_unlock(&file_priv->table_lock); - ret = -ENOENT; - goto out_unlock; - } - - idr_replace(&file_priv->object_idr, NULL, args->handle); + obj = idr_replace(&file_priv->object_idr, NULL, args->handle); + if (IS_ERR_OR_NULL(obj)) { + idr_remove(&file_priv->object_idr, new_handle); + spin_unlock(&file_priv->table_lock); + ret = -ENOENT; + goto out_unlock; + } spin_unlock(&file_priv->table_lock); if (obj->dma_buf) { ret = drm_prime_add_buf_handle(&file_priv->prime, obj->dma_buf, - handle); + new_handle); if (ret < 0) { spin_lock(&file_priv->table_lock); - idr_remove(&file_priv->object_idr, handle); + idr_remove(&file_priv->object_idr, new_handle); idr_replace(&file_priv->object_idr, obj, args->handle); spin_unlock(&file_priv->table_lock); goto out_unlock; @@ -1086,14 +1083,12 @@ int drm_gem_change_handle_ioctl(struct drm_device *dev, void *data, spin_lock(&file_priv->table_lock); idr_remove(&file_priv->object_idr, args->handle); - idrobj = idr_replace(&file_priv->object_idr, obj, handle); + obj = idr_replace(&file_priv->object_idr, obj, new_handle); spin_unlock(&file_priv->table_lock); - WARN_ON(idrobj != NULL); + WARN_ON(obj != NULL); out_unlock: mutex_unlock(&file_priv->prime.lock); -out: - drm_gem_object_put(obj); return ret; } diff --git a/drivers/gpu/drm/drm_ioctl.c b/drivers/gpu/drm/drm_ioctl.c index ff193155129e..e2df4becce62 100644 --- a/drivers/gpu/drm/drm_ioctl.c +++ b/drivers/gpu/drm/drm_ioctl.c @@ -660,7 +660,8 @@ static const struct drm_ioctl_desc drm_ioctls[] = { DRM_IOCTL_DEF(DRM_IOCTL_GEM_CLOSE, drm_gem_close_ioctl, DRM_RENDER_ALLOW), DRM_IOCTL_DEF(DRM_IOCTL_GEM_FLINK, drm_gem_flink_ioctl, DRM_AUTH), DRM_IOCTL_DEF(DRM_IOCTL_GEM_OPEN, drm_gem_open_ioctl, DRM_AUTH), - DRM_IOCTL_DEF(DRM_IOCTL_GEM_CHANGE_HANDLE, drm_gem_change_handle_ioctl, DRM_RENDER_ALLOW), + /* see drm_gem.c:drm_gem_change_handle_ioctl for why this is invalid */ + DRM_IOCTL_DEF(DRM_IOCTL_GEM_CHANGE_HANDLE, drm_invalid_op, DRM_RENDER_ALLOW), DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETRESOURCES, drm_mode_getresources, 0), From fb402386af4cdce108ff991a796386de55439735 Mon Sep 17 00:00:00 2001 From: Cryolitia PukNgae Date: Fri, 5 Jun 2026 15:27:21 +0800 Subject: [PATCH 315/329] Input: atkbd - skip deactivate for HONOR BCC-N's internal keyboard After commit 9cf6e24c9fbf17e52de9fff07f12be7565ea6d61 ("Input: atkbd - do not skip atkbd_deactivate() when skipping ATKBD_CMD_GETID"), HONOR BCC-N, aka HONOR MagicBook 14 2026's internal keyboard stops working. Adding the atkbd_deactivate_fixup quirk fixes it. DMI: HONOR BCC-N/BCC-N-PCB, BIOS 1.04 04/07/2026 Fixes: 9cf6e24c9fbf17e52de9fff07f12be7565ea6d61 ("Input: atkbd - do not skip atkbd_deactivate() when skipping ATKBD_CMD_GETID") Reported-by: Hongfei Ren Link: https://github.com/colorcube/Linux-on-Honor-Magicbook-14-Pro/issues/1#issuecomment-4562679891 Tested-by: Hongfei Ren Cc: stable@kernel.org Signed-off-by: Cryolitia PukNgae Link: https://patch.msgid.link/20260605-honor-v1-1-78e05e491193@linux.dev Signed-off-by: Dmitry Torokhov --- drivers/input/keyboard/atkbd.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/drivers/input/keyboard/atkbd.c b/drivers/input/keyboard/atkbd.c index 217e66ee36a1..8cb4dc6fb165 100644 --- a/drivers/input/keyboard/atkbd.c +++ b/drivers/input/keyboard/atkbd.c @@ -1931,6 +1931,13 @@ static const struct dmi_system_id atkbd_dmi_quirk_table[] __initconst = { }, .callback = atkbd_deactivate_fixup, }, + { + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "HONOR"), + DMI_MATCH(DMI_PRODUCT_NAME, "BCC-N"), + }, + .callback = atkbd_deactivate_fixup, + }, { } }; From 979c294509f9248fe1e7c358d582fb37dd5ca12d Mon Sep 17 00:00:00 2001 From: Nathan Chancellor Date: Thu, 4 Jun 2026 17:33:21 -0700 Subject: [PATCH 316/329] cfi: Include uaccess.h for get_kernel_nofault() After commit 0652a3daa787 ("tracing: Fix CFI violation in probestub being called by tprobes"), there are many build errors when building ARCH=arm multi_v7_defconfig + CONFIG_CFI=y like: In file included from drivers/base/devres.c:17: In file included from drivers/base/trace.h:16: In file included from include/linux/tracepoint.h:23: include/linux/cfi.h:44:6: error: call to undeclared function 'get_kernel_nofault'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration] 44 | if (get_kernel_nofault(hash, func - cfi_get_offset())) | ^ 1 error generated. get_kernel_nofault() is called in the generic version of cfi_get_func_hash() but nothing ensures uaccess.h is always included for a proper expansion and prototype. Include uaccess.h in cfi.h to clear up the errors. Cc: stable@vger.kernel.org Fixes: 0652a3daa787 ("tracing: Fix CFI violation in probestub being called by tprobes") Signed-off-by: Nathan Chancellor Acked-by: Masami Hiramatsu (Google) Reviewed-by: Sami Tolvanen Signed-off-by: Linus Torvalds --- include/linux/cfi.h | 1 + 1 file changed, 1 insertion(+) diff --git a/include/linux/cfi.h b/include/linux/cfi.h index 1fd22ea6eba4..0f220d29225c 100644 --- a/include/linux/cfi.h +++ b/include/linux/cfi.h @@ -9,6 +9,7 @@ #include #include +#include #include #ifdef CONFIG_CFI From e3573f739e3dadab57ec80488d07e05c8f6e82d3 Mon Sep 17 00:00:00 2001 From: Andreas Schwab Date: Fri, 22 May 2026 00:34:30 +0200 Subject: [PATCH 317/329] riscv/ptrace: Use USER_REGSET_NOTE_TYPE for REGSET_CFI Fixes a warning while dumping core: [54983.546369][ C7] WARNING: [!note_name] fs/binfmt_elf.c:1771 at elf_core_dump+0x910/0xf68, CPU#7: abort01/31982 Fixes: 2af7c9cf021c ("riscv/ptrace: expose riscv CFI status and state via ptrace and in core files") Signed-off-by: Andreas Schwab Link: https://patch.msgid.link/87y0hcxuh5.fsf@igel.home Signed-off-by: Paul Walmsley --- arch/riscv/kernel/ptrace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/riscv/kernel/ptrace.c b/arch/riscv/kernel/ptrace.c index 793bcee46182..f336a183667e 100644 --- a/arch/riscv/kernel/ptrace.c +++ b/arch/riscv/kernel/ptrace.c @@ -413,7 +413,7 @@ static struct user_regset riscv_user_regset[] __ro_after_init = { #endif #ifdef CONFIG_RISCV_USER_CFI [REGSET_CFI] = { - .core_note_type = NT_RISCV_USER_CFI, + USER_REGSET_NOTE_TYPE(RISCV_USER_CFI), .align = sizeof(__u64), .n = sizeof(struct user_cfi_state) / sizeof(__u64), .size = sizeof(__u64), From aad60bdd0b5a6dfa40454ab03ce658408eaf4615 Mon Sep 17 00:00:00 2001 From: Nam Cao Date: Tue, 7 Apr 2026 14:06:39 +0200 Subject: [PATCH 318/329] riscv: Fix fast_unaligned_access_speed_key not getting initialized The static key fast_unaligned_access_speed_key is supposed to be initialized after check_unaligned_access_all_cpus() has been completed. However, check_unaligned_access_all_cpus() has been moved to late_initcall while setting fast_unaligned_access_speed_key still happens at arch_initcall_sync, thus the static key does not get properly initialized. fast_unaligned_access_speed_key can still be initialized in CPU hotplug events, but that cannot be relied on. Move fast_unaligned_access_speed_key's initialization into check_unaligned_access_all_cpus() to fix this issue. This also prevent someone from moving one initcall while forgetting the other in the future. Fixes: 6455c6c11827 ("riscv: Clean up & optimize unaligned scalar access probe") Reported-by: Michael Neuling Closes: https://lore.kernel.org/linux-riscv/CAEjGV6y0=bSLp_wrS0uHFj1S2TCRtz4GKzaU5O-L1VV-EL7Nnw@mail.gmail.com/ Signed-off-by: Nam Cao Link: https://patch.msgid.link/20260407120639.4006031-1-namcao@linutronix.de Signed-off-by: Paul Walmsley --- arch/riscv/kernel/unaligned_access_speed.c | 15 ++++----------- 1 file changed, 4 insertions(+), 11 deletions(-) diff --git a/arch/riscv/kernel/unaligned_access_speed.c b/arch/riscv/kernel/unaligned_access_speed.c index 485ab1d105d3..11c781a4de73 100644 --- a/arch/riscv/kernel/unaligned_access_speed.c +++ b/arch/riscv/kernel/unaligned_access_speed.c @@ -235,17 +235,6 @@ static void set_unaligned_access_static_branches(void) modify_unaligned_access_branches(&fast_and_online, num_online_cpus()); } -static int __init lock_and_set_unaligned_access_static_branch(void) -{ - cpus_read_lock(); - set_unaligned_access_static_branches(); - cpus_read_unlock(); - - return 0; -} - -arch_initcall_sync(lock_and_set_unaligned_access_static_branch); - static int riscv_online_cpu(unsigned int cpu) { int ret = cpu_online_unaligned_access_init(cpu); @@ -440,6 +429,10 @@ static int __init check_unaligned_access_all_cpus(void) cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, "riscv:online", riscv_online_cpu_vec, NULL); + cpus_read_lock(); + set_unaligned_access_static_branches(); + cpus_read_unlock(); + return 0; } From 15b4155138505669d3d43d7692459ee8ea2a86e7 Mon Sep 17 00:00:00 2001 From: Richard Patel Date: Mon, 18 May 2026 18:39:18 +0000 Subject: [PATCH 319/329] riscv: cfi: reject unknown flags in PR_SET_CFI prctl(PR_SET_CFI,PR_CFI_BRANCH_LANDING_PADS) silently ignored unknown control values. Only PR_CFI_{ENABLE,DISABLE,LOCK} should be permitted. This changes the behavior of the uABI (fails previously accepted bits with EINVAL). Fixes: 08ee1559052b ("prctl: cfi: change the branch landing pad prctl()s to be more descriptive") Signed-off-by: Richard Patel Link: https://patch.msgid.link/20260518183918.322545-1-ripatel@wii.dev [pjw@kernel.org: change the patch description to note that although this is a uABI change, it does not break the uABI] Signed-off-by: Paul Walmsley --- arch/riscv/include/asm/usercfi.h | 1 + arch/riscv/kernel/usercfi.c | 3 +++ tools/testing/selftests/riscv/cfi/cfitests.c | 6 ++++++ 3 files changed, 10 insertions(+) diff --git a/arch/riscv/include/asm/usercfi.h b/arch/riscv/include/asm/usercfi.h index f56966edbf5c..61ee02cee297 100644 --- a/arch/riscv/include/asm/usercfi.h +++ b/arch/riscv/include/asm/usercfi.h @@ -50,6 +50,7 @@ void set_indir_lp_status(struct task_struct *task, bool enable); void set_indir_lp_lock(struct task_struct *task, bool lock); #define PR_SHADOW_STACK_SUPPORTED_STATUS_MASK (PR_SHADOW_STACK_ENABLE) +#define PR_CFI_SUPPORTED_STATUS_MASK (PR_CFI_ENABLE | PR_CFI_DISABLE | PR_CFI_LOCK) #else diff --git a/arch/riscv/kernel/usercfi.c b/arch/riscv/kernel/usercfi.c index cbfb4e495e9f..5a7113d69bad 100644 --- a/arch/riscv/kernel/usercfi.c +++ b/arch/riscv/kernel/usercfi.c @@ -467,6 +467,9 @@ int arch_prctl_set_branch_landing_pad_state(struct task_struct *t, unsigned long if (!is_user_lpad_enabled()) return -EINVAL; + if (state & ~PR_CFI_SUPPORTED_STATUS_MASK) + return -EINVAL; + /* indirect branch tracking is locked and further can't be modified by user */ if (is_indir_lp_locked(t)) return -EINVAL; diff --git a/tools/testing/selftests/riscv/cfi/cfitests.c b/tools/testing/selftests/riscv/cfi/cfitests.c index 39d097b6881f..0e3943461e7d 100644 --- a/tools/testing/selftests/riscv/cfi/cfitests.c +++ b/tools/testing/selftests/riscv/cfi/cfitests.c @@ -141,6 +141,12 @@ int main(int argc, char *argv[]) ksft_print_msg("Starting risc-v tests\n"); + /* Test unknown PR_CFI bits */ + ret = my_syscall5(__NR_prctl, PR_SET_CFI, PR_CFI_BRANCH_LANDING_PADS, + PR_CFI_ENABLE | 0xffff0, 0, 0); + if (!ret) + ksft_exit_fail_msg("PR_SET_CFI accepted reserved branch landing pad bits\n"); + /* * Landing pad test. Not a lot of kernel changes to support landing * pads for user mode except lighting up a bit in senvcfg via a prctl. From 4549871118cf616eecdd2d939f78e3b9e1dddc48 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sun, 7 Jun 2026 15:37:58 -0700 Subject: [PATCH 320/329] Linux 7.1-rc7 --- Makefile | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index e3aabcda1ff7..e156e2696efe 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ VERSION = 7 PATCHLEVEL = 1 SUBLEVEL = 0 -EXTRAVERSION = -rc6 +EXTRAVERSION = -rc7 NAME = Baby Opossum Posse # *DOCUMENTATION* From badad6fad60def1b9805559dd81dbab3d97b82aa Mon Sep 17 00:00:00 2001 From: Jason Gunthorpe Date: Thu, 4 Jun 2026 15:03:13 -0300 Subject: [PATCH 321/329] RDMA: During rereg_mr ensure that REREG_ACCESS is compatible If IB_MR_REREG_ACCESS changes from RO to RW then the umem has to be re-evaluated to ensure it is properly pinned as RW. Since the umem is hidden inside each driver's mr struct add a ib_umem_check_rereg() function that each driver has to call before processing IB_MR_REREG_ACCESS. mlx4 has to retain its duplicate ib_access_writable check because it implements IB_MR_REREG_ACCESS | IB_MR_REREG_TRANS by changing both items in place sequentially while the MR is live, so it will continue to not support this combination. Cc: stable@vger.kernel.org Fixes: b40656aa7d55 ("RDMA/umem: remove FOLL_FORCE usage") Link: https://patch.msgid.link/r/0-v1-06fb1a2d6cf5+107-rereg_access_jgg@nvidia.com Reported-by: Philip Tsukerman Signed-off-by: Jason Gunthorpe --- drivers/infiniband/core/umem.c | 16 ++++++++++++++++ drivers/infiniband/hw/hns/hns_roce_mr.c | 4 ++++ drivers/infiniband/hw/irdma/verbs.c | 4 ++++ drivers/infiniband/hw/mlx4/mr.c | 4 ++++ drivers/infiniband/hw/mlx5/mr.c | 4 ++++ drivers/infiniband/sw/rxe/rxe_verbs.c | 5 +++++ include/rdma/ib_umem.h | 8 ++++++++ 7 files changed, 45 insertions(+) diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 786fa1aa8e55..4b055712b0d0 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -332,3 +332,19 @@ int ib_umem_copy_from(void *dst, struct ib_umem *umem, size_t offset, return 0; } EXPORT_SYMBOL(ib_umem_copy_from); + +/* + * Called during rereg mr if the driver is able to re-use a umem for + * IB_MR_REREG_ACCESS. + */ +int ib_umem_check_rereg(struct ib_umem *umem, int flags, int new_access_flags) +{ + if (!umem) + return 0; + + if ((flags & IB_MR_REREG_ACCESS) && !(flags & IB_MR_REREG_TRANS)) + if (ib_access_writable(new_access_flags) && !umem->writable) + return -EACCES; + return 0; +} +EXPORT_SYMBOL(ib_umem_check_rereg); diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index 896af1828a38..25bfd3970f5b 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -300,6 +300,10 @@ struct ib_mr *hns_roce_rereg_user_mr(struct ib_mr *ibmr, int flags, u64 start, goto err_out; } + ret = ib_umem_check_rereg(mr->pbl_mtr.umem, flags, mr_access_flags); + if (ret) + goto err_out; + mailbox = hns_roce_alloc_cmd_mailbox(hr_dev); ret = PTR_ERR_OR_ZERO(mailbox); if (ret) diff --git a/drivers/infiniband/hw/irdma/verbs.c b/drivers/infiniband/hw/irdma/verbs.c index 17086048d2d7..8cd427532805 100644 --- a/drivers/infiniband/hw/irdma/verbs.c +++ b/drivers/infiniband/hw/irdma/verbs.c @@ -3803,6 +3803,10 @@ static struct ib_mr *irdma_rereg_user_mr(struct ib_mr *ib_mr, int flags, if (flags & ~(IB_MR_REREG_TRANS | IB_MR_REREG_PD | IB_MR_REREG_ACCESS)) return ERR_PTR(-EOPNOTSUPP); + ret = ib_umem_check_rereg(iwmr->region, flags, new_access); + if (ret) + return ERR_PTR(ret); + if (dmabuf_revocable) { umem_dmabuf = to_ib_umem_dmabuf(iwmr->region); diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 650b4a9121ff..6747bca30677 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -209,6 +209,10 @@ struct ib_mr *mlx4_ib_rereg_user_mr(struct ib_mr *mr, int flags, u64 start, struct mlx4_mpt_entry **pmpt_entry = &mpt_entry; int err; + err = ib_umem_check_rereg(mmr->umem, flags, mr_access_flags); + if (err) + return ERR_PTR(err); + /* Since we synchronize this call and mlx4_ib_dereg_mr via uverbs, * we assume that the calls can't run concurrently. Otherwise, a * race exists. diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 3b6da45061a5..fb40b44496f4 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -1179,6 +1179,10 @@ struct ib_mr *mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, if (flags & ~(IB_MR_REREG_TRANS | IB_MR_REREG_PD | IB_MR_REREG_ACCESS)) return ERR_PTR(-EOPNOTSUPP); + err = ib_umem_check_rereg(mr->umem, flags, new_access_flags); + if (err) + return ERR_PTR(err); + if (!(flags & IB_MR_REREG_ACCESS)) new_access_flags = mr->access_flags; if (!(flags & IB_MR_REREG_PD)) diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 4d4891dc2884..4cf04a44189c 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -1319,6 +1319,7 @@ static struct ib_mr *rxe_rereg_user_mr(struct ib_mr *ibmr, int flags, struct rxe_mr *mr = to_rmr(ibmr); struct rxe_pd *old_pd = to_rpd(ibmr->pd); struct rxe_pd *pd = to_rpd(ibpd); + int err; /* for now only support the two easy cases: * rereg_pd and rereg_access @@ -1328,6 +1329,10 @@ static struct ib_mr *rxe_rereg_user_mr(struct ib_mr *ibmr, int flags, return ERR_PTR(-EOPNOTSUPP); } + err = ib_umem_check_rereg(mr->umem, flags, access); + if (err) + return ERR_PTR(err); + if (flags & IB_MR_REREG_PD) { rxe_put(old_pd); rxe_get(pd); diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 2ad52cc1d52b..49172098a8de 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -156,6 +156,8 @@ void ib_umem_dmabuf_revoke_lock(struct ib_umem_dmabuf *umem_dmabuf); void ib_umem_dmabuf_revoke_unlock(struct ib_umem_dmabuf *umem_dmabuf); void ib_umem_dmabuf_revoke(struct ib_umem_dmabuf *umem_dmabuf); +int ib_umem_check_rereg(struct ib_umem *umem, int flags, int new_access_flags); + #else /* CONFIG_INFINIBAND_USER_MEM */ #include @@ -230,5 +232,11 @@ static inline void ib_umem_dmabuf_revoke_lock(struct ib_umem_dmabuf *umem_dmabuf static inline void ib_umem_dmabuf_revoke_unlock(struct ib_umem_dmabuf *umem_dmabuf) {} static inline void ib_umem_dmabuf_revoke(struct ib_umem_dmabuf *umem_dmabuf) {} +static inline int ib_umem_check_rereg(struct ib_umem *umem, int flags, + int new_access_flags) +{ + return -EOPNOTSUPP; +} + #endif /* CONFIG_INFINIBAND_USER_MEM */ #endif /* IB_UMEM_H */ From 29e7b925ae6df64894e82ab6419994dc25580a8a Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Tue, 2 Jun 2026 15:46:42 -0400 Subject: [PATCH 322/329] IB/isert: Reject login PDUs shorter than ISER_HEADERS_LEN In drivers/infiniband/ulp/isert/ib_isert.c, isert_login_recv_done() computes the login request payload length as wc->byte_len minus ISER_HEADERS_LEN with no lower bound, and login_req_len is a signed int. A remote iSER initiator can post a login Send work request carrying fewer than ISER_HEADERS_LEN (76) bytes, so the subtraction underflows and login_req_len becomes negative. isert_rx_login_req() then reads that negative length back into a signed int, takes size = min(rx_buflen, MAX_KEY_VALUE_PAIRS), and because the min() is signed it keeps the negative value; the value is then passed as the memcpy() length and sign-extended to a multi-gigabyte size_t. The copy into the 8192-byte login->req_buf runs far out of bounds and faults, crashing the target node. The login phase precedes iSCSI authentication, so no credentials are required to reach this path. Reject any login PDU shorter than ISER_HEADERS_LEN before the subtraction, mirroring the existing early return on a failed work completion, so login_req_len can never go negative. The upper bound was already safe: a posted login buffer cannot deliver more than ISER_RX_PAYLOAD_SIZE, so the difference stays at or below MAX_KEY_VALUE_PAIRS and the existing min() clamps it; only the missing lower bound needs to be added. Fixes: b8d26b3be8b3 ("iser-target: Add iSCSI Extensions for RDMA (iSER) target driver") Link: https://patch.msgid.link/r/20260602194642.2273217-1-michael.bommarito@gmail.com Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito Signed-off-by: Jason Gunthorpe --- drivers/infiniband/ulp/isert/ib_isert.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c index 348005e71891..1015a51f750a 100644 --- a/drivers/infiniband/ulp/isert/ib_isert.c +++ b/drivers/infiniband/ulp/isert/ib_isert.c @@ -1383,6 +1383,12 @@ isert_login_recv_done(struct ib_cq *cq, struct ib_wc *wc) ib_dma_sync_single_for_cpu(ib_dev, isert_conn->login_desc->dma_addr, ISER_RX_SIZE, DMA_FROM_DEVICE); + if (unlikely(wc->byte_len < ISER_HEADERS_LEN)) { + isert_dbg("login request length %u is too short\n", + wc->byte_len); + return; + } + isert_conn->login_req_len = wc->byte_len - ISER_HEADERS_LEN; if (isert_conn->conn) { From 13e91fd076306f5d0cdfa14f53d69e37274723c4 Mon Sep 17 00:00:00 2001 From: Michael Bommarito Date: Tue, 2 Jun 2026 18:04:57 -0400 Subject: [PATCH 323/329] RDMA/srp: bound SRP_RSP sense copy by the received length srp_process_rsp() copies sense data from rsp->data + resp_data_len, where resp_data_len is the full 32-bit value supplied by the SRP target and is never checked against the number of bytes actually received (wc->byte_len). The copy length is bounded to SCSI_SENSE_BUFFERSIZE, so at most 96 bytes are copied, but the source offset is not bounded. A malicious or compromised SRP target on the InfiniBand/RoCE fabric that the initiator has logged into can return an SRP_RSP with SRP_RSP_FLAG_SNSVALID set and a large resp_data_len. The receive buffer is allocated at the target-chosen max_ti_iu_len, so the source of the sense copy lands past the bytes actually received; with resp_data_len near 0xFFFFFFFF it is gigabytes past the buffer and the read faults. Copy the sense data only if it has not been truncated, that is, only if the response header, the response data, and the sense region fit within the bytes actually received; otherwise drop the sense and log. The in-tree iSER and NVMe-RDMA receive paths already bound their parse by wc->byte_len; this brings ib_srp into line with them. Fixes: aef9ec39c47f ("IB: Add SCSI RDMA Protocol (SRP) initiator") Link: https://patch.msgid.link/r/20260602220457.2542840-1-michael.bommarito@gmail.com Cc: stable@vger.kernel.org Assisted-by: Claude:claude-opus-4-8 Signed-off-by: Michael Bommarito Reviewed-by: Bart Van Assche Signed-off-by: Jason Gunthorpe --- drivers/infiniband/ulp/srp/ib_srp.c | 30 +++++++++++++++++++++++------ 1 file changed, 24 insertions(+), 6 deletions(-) diff --git a/drivers/infiniband/ulp/srp/ib_srp.c b/drivers/infiniband/ulp/srp/ib_srp.c index b58868e1cf11..acbd787de265 100644 --- a/drivers/infiniband/ulp/srp/ib_srp.c +++ b/drivers/infiniband/ulp/srp/ib_srp.c @@ -1932,7 +1932,8 @@ static int srp_post_recv(struct srp_rdma_ch *ch, struct srp_iu *iu) return ib_post_recv(ch->qp, &wr, NULL); } -static void srp_process_rsp(struct srp_rdma_ch *ch, struct srp_rsp *rsp) +static void srp_process_rsp(struct srp_rdma_ch *ch, struct srp_rsp *rsp, + u32 byte_len) { struct srp_target_port *target = ch->target; struct srp_request *req; @@ -1973,10 +1974,27 @@ static void srp_process_rsp(struct srp_rdma_ch *ch, struct srp_rsp *rsp) scmnd->result = rsp->status; if (rsp->flags & SRP_RSP_FLAG_SNSVALID) { - memcpy(scmnd->sense_buffer, rsp->data + - be32_to_cpu(rsp->resp_data_len), - min_t(int, be32_to_cpu(rsp->sense_data_len), - SCSI_SENSE_BUFFERSIZE)); + u32 resp_len = be32_to_cpu(rsp->resp_data_len); + u32 sense_len = be32_to_cpu(rsp->sense_data_len); + + /* + * The sense data starts resp_data_len bytes past the + * response data area; both lengths come from the + * target-controlled response. Copy the sense data + * only if it has not been truncated, that is, only if + * the full sense region fits within the bytes actually + * received. Otherwise the copy source would run past + * the receive buffer (sized to the target-chosen + * max_ti_iu_len), reading out of bounds. + */ + if (sizeof(*rsp) + (u64)resp_len + sense_len <= byte_len) + memcpy(scmnd->sense_buffer, + rsp->data + resp_len, + min(sense_len, SCSI_SENSE_BUFFERSIZE)); + else + shost_printk(KERN_ERR, target->scsi_host, + "dropping truncated sense data (resp_data_len %u sense_data_len %u, %u bytes received)\n", + resp_len, sense_len, byte_len); } if (unlikely(rsp->flags & SRP_RSP_FLAG_DIUNDER)) @@ -2086,7 +2104,7 @@ static void srp_recv_done(struct ib_cq *cq, struct ib_wc *wc) switch (opcode) { case SRP_RSP: - srp_process_rsp(ch, iu->buf); + srp_process_rsp(ch, iu->buf, wc->byte_len); break; case SRP_CRED_REQ: From 805d5a2b792819171be100c50c9ddafa0f8c2231 Mon Sep 17 00:00:00 2001 From: Jann Horn Date: Fri, 5 Jun 2026 22:27:33 +0200 Subject: [PATCH 324/329] namespace: restrict OPEN_TREE_NAMESPACE/FSMOUNT_NAMESPACE to directories open_tree(..., OPEN_TREE_NAMESPACE) and fsmount(..., FSMOUNT_NAMESPACE, ...) currently work on non-directories, like regular files. That's bad for two reasons: - It ends up mounting a regular file over the inherited namespace root, which is a directory; mounting a non-directory over a directory is normally explicitly forbidden, see for example do_move_mount() - It causes setns() on the new namespace to set the cwd to a regular file, which the rest of VFS does not expect Fix it by restricting create_new_namespace() (which is used by both of these flags) to directories. Leave the behavior for OPEN_TREE_CLONE as-is, that seems unproblematic. Fixes: 9b8a0ba68246 ("mount: add OPEN_TREE_NAMESPACE") Cc: Al Viro Cc: Christian Brauner Cc: Jan Kara Cc: stable@kernel.org Signed-off-by: Jann Horn Signed-off-by: Linus Torvalds --- fs/namespace.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/namespace.c b/fs/namespace.c index f5905f4ec560..341ddd353b3a 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -3103,6 +3103,9 @@ static struct mnt_namespace *create_new_namespace(struct path *path, unsigned int copy_flags = 0; bool locked = false, recurse = flags & MOUNT_COPY_RECURSIVE; + if (unlikely(!d_can_lookup(path->dentry))) + return ERR_PTR(-ENOTDIR); + if (user_ns != ns->user_ns) copy_flags |= CL_SLAVE; From f79413b2dde10829a1b526543e725dd5c1847e9a Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 11 Jun 2026 12:48:46 +0200 Subject: [PATCH 325/329] KVM: s390: Silence potential warnings in _gmap_crstep_xchg_atomic() While dat_crstep_xchg_atomic() is marked as __must_check, in this particular case the return value should be ignored. Silence potential compiler warnings with a pointless check, and add a comment to explain the situation. Fixes: d1adc098ce08 ("KVM: s390: Fix _gmap_crstep_xchg_atomic()") CC: stable@vger.kernel.org # 7.1 Signed-off-by: Claudio Imbrenda Message-ID: <20260611104850.110313-2-imbrenda@linux.ibm.com> --- arch/s390/kvm/gmap.h | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/s390/kvm/gmap.h b/arch/s390/kvm/gmap.h index 5374f21aaf8d..20881e3ce9d8 100644 --- a/arch/s390/kvm/gmap.h +++ b/arch/s390/kvm/gmap.h @@ -279,7 +279,16 @@ static inline bool __must_check _gmap_crstep_xchg_atomic(struct gmap *gmap, unio gmap_handle_vsie_unshadow_event(gmap, gfn); else _gmap_handle_vsie_unshadow_event(gmap, gfn); - dat_crstep_xchg_atomic(crstep, oldcrste, newcrste, gfn, gmap->asce); + if (!dat_crstep_xchg_atomic(crstep, oldcrste, newcrste, gfn, gmap->asce)) + return false; + /* + * Return false even if the swap was successful, as it only + * indicates that the best effort clearing of the vsie_notif + * bit was successful. The caller will have to try again + * regardless, since the desired value has not been set. + * This pointless check is needed to silence a potential + * __must_check warning. + */ return false; } if (!oldcrste.s.fc1.d && newcrste.s.fc1.d && !newcrste.s.fc1.s) From 5670b7f927f8d98685f3f5873dbf9f8d7a5a63f3 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 11 Jun 2026 12:48:47 +0200 Subject: [PATCH 326/329] KVM: s390: Fix unlikely race in try_get_locked_pte() Fix an unlikely race in try_get_locked_pte(), which could have happened if puds or pmds get unmapped between the p?dp_get() and p?d_offset() functions. Fixes: 89fa757931dc ("KVM: s390: Avoid potentially sleeping while atomic when zapping pages") CC: stable@vger.kernel.org # 7.1 Signed-off-by: Claudio Imbrenda Message-ID: <20260611104850.110313-3-imbrenda@linux.ibm.com> --- arch/s390/mm/gmap_helpers.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/arch/s390/mm/gmap_helpers.c b/arch/s390/mm/gmap_helpers.c index 1cfe4724fbe2..ee3f37af8aee 100644 --- a/arch/s390/mm/gmap_helpers.c +++ b/arch/s390/mm/gmap_helpers.c @@ -51,15 +51,15 @@ pte_t *try_get_locked_pte(struct mm_struct *mm, unsigned long vmaddr, spinlock_t pgd = pgdp_get(pgdp); if (pgd_none(pgd) || !pgd_present(pgd)) return NULL; - p4dp = p4d_offset(pgdp, vmaddr); + p4dp = p4d_offset_lockless(pgdp, pgd, vmaddr); p4d = p4dp_get(p4dp); if (p4d_none(p4d) || !p4d_present(p4d)) return NULL; - pudp = pud_offset(p4dp, vmaddr); + pudp = pud_offset_lockless(p4dp, p4d, vmaddr); pud = pudp_get(pudp); if (pud_none(pud) || pud_leaf(pud) || !pud_present(pud)) return NULL; - pmdp = pmd_offset(pudp, vmaddr); + pmdp = pmd_offset_lockless(pudp, pud, vmaddr); pmd = pmdp_get_lockless(pmdp); if (pmd_none(pmd) || pmd_leaf(pmd) || !pmd_present(pmd)) return NULL; From 505fcce0c64957f475f9b11fb1403821b7f33e7e Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 11 Jun 2026 12:48:48 +0200 Subject: [PATCH 327/329] KVM: s390: vsie: Fix allocation of struct vsie_rmap The allocation size for struct vsie_rmap in kvm_s390_mmu_cache_topup() was wrong due to a copy-paste error. Fix it by using the type name. Fixes: 12f2f61a9e1a ("KVM: s390: KVM page table management functions: allocation") CC: stable@vger.kernel.org # 7.1 Signed-off-by: Claudio Imbrenda Message-ID: <20260611104850.110313-4-imbrenda@linux.ibm.com> --- arch/s390/kvm/dat.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/arch/s390/kvm/dat.c b/arch/s390/kvm/dat.c index 4a41c0247ffa..a4fe664f65ee 100644 --- a/arch/s390/kvm/dat.c +++ b/arch/s390/kvm/dat.c @@ -45,7 +45,7 @@ int kvm_s390_mmu_cache_topup(struct kvm_s390_mmu_cache *mc) mc->pts[mc->n_pts] = o; } for ( ; mc->n_rmaps < KVM_S390_MMU_CACHE_N_RMAPS; mc->n_rmaps++) { - o = kzalloc_obj(*mc->rmaps[0], GFP_KERNEL_ACCOUNT); + o = kzalloc_obj(struct vsie_rmap, GFP_KERNEL_ACCOUNT); if (!o) return -ENOMEM; mc->rmaps[mc->n_rmaps] = o; From 668e70cc545e2659f3c4adad20a0883533042473 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 11 Jun 2026 12:48:49 +0200 Subject: [PATCH 328/329] KVM: s390: vsie: Add missing radix_tree_preload() in _gaccess_shadow_fault() Add missing radix_tree_preload() in _gaccess_shadow_fault() to guarantee forward progress. The core of _gaccess_shadow_fault() has been split into ___gaccess_shadow_fault() in order to simplify locking. Fixes: e38c884df921 ("KVM: s390: Switch to new gmap") CC: stable@vger.kernel.org # 7.1 Signed-off-by: Claudio Imbrenda Message-ID: <20260611104850.110313-5-imbrenda@linux.ibm.com> --- arch/s390/kvm/gaccess.c | 59 +++++++++++++++++++++++++---------------- 1 file changed, 36 insertions(+), 23 deletions(-) diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 20e28b183c1a..0584fc91606f 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -1582,35 +1582,48 @@ static int _gaccess_do_shadow(struct kvm_s390_mmu_cache *mc, struct gmap *sg, return _do_shadow_crste(sg, saddr, host, table, entries + LEVEL_MEM, w->p); } -static inline int _gaccess_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg, gpa_t saddr, - unsigned long seq, struct pgtwalk *walk) +static inline int ___gaccess_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg, gpa_t saddr, + unsigned long seq, struct pgtwalk *walk) { struct gmap *parent; int rc; + if (kvm_s390_array_needs_retry_safe(vcpu->kvm, seq, walk->raw_entries)) + return -EAGAIN; + parent = READ_ONCE(sg->parent); + if (!parent) + return -EAGAIN; + scoped_guard(spinlock, &parent->children_lock) { + if (READ_ONCE(sg->parent) != parent) + return -EAGAIN; + sg->invalidated = false; + rc = _gaccess_do_shadow(vcpu->arch.mc, sg, saddr, walk); + } + if (!rc) + kvm_s390_release_faultin_array(vcpu->kvm, walk->raw_entries, false); + return rc; +} + +static inline int _gaccess_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg, gpa_t saddr, + unsigned long seq, struct pgtwalk *walk) +{ + int rc; + if (kvm_s390_array_needs_retry_unsafe(vcpu->kvm, seq, walk->raw_entries)) return -EAGAIN; -again: - rc = kvm_s390_mmu_cache_topup(vcpu->arch.mc); - if (rc) - return rc; - scoped_guard(read_lock, &vcpu->kvm->mmu_lock) { - if (kvm_s390_array_needs_retry_safe(vcpu->kvm, seq, walk->raw_entries)) - return -EAGAIN; - parent = READ_ONCE(sg->parent); - if (!parent) - return -EAGAIN; - scoped_guard(spinlock, &parent->children_lock) { - if (READ_ONCE(sg->parent) != parent) - return -EAGAIN; - sg->invalidated = false; - rc = _gaccess_do_shadow(vcpu->arch.mc, sg, saddr, walk); - } - if (rc == -ENOMEM) - goto again; - if (!rc) - kvm_s390_release_faultin_array(vcpu->kvm, walk->raw_entries, false); - } + + do { + rc = kvm_s390_mmu_cache_topup(vcpu->arch.mc); + if (rc) + return rc; + rc = radix_tree_preload(GFP_KERNEL); + if (rc) + return rc; + scoped_guard(read_lock, &vcpu->kvm->mmu_lock) + rc = ___gaccess_shadow_fault(vcpu, sg, saddr, seq, walk); + radix_tree_preload_end(); + } while (rc == -ENOMEM); + return rc; } From abeb7eb57f1671d9185ddf11236c784f07bdb928 Mon Sep 17 00:00:00 2001 From: Claudio Imbrenda Date: Thu, 11 Jun 2026 12:48:50 +0200 Subject: [PATCH 329/329] KVM: s390: vsie: Use mmu cache to allocate rmap Use kvm_s390_mmu_cache_alloc_rmap() to allocate the rmap in gmap_insert_rmap(), instead of a normal kzalloc_obj() with GFP_ATOMIC. This guarantees forward progress. Fixes: a2c17f9270cc ("KVM: s390: New gmap code") CC: stable@vger.kernel.org # 7.1 Signed-off-by: Claudio Imbrenda Message-ID: <20260611104850.110313-6-imbrenda@linux.ibm.com> --- arch/s390/kvm/gaccess.c | 16 ++++++++-------- arch/s390/kvm/gmap.c | 7 ++++--- arch/s390/kvm/gmap.h | 3 ++- 3 files changed, 14 insertions(+), 12 deletions(-) diff --git a/arch/s390/kvm/gaccess.c b/arch/s390/kvm/gaccess.c index 0584fc91606f..36102b2727fb 100644 --- a/arch/s390/kvm/gaccess.c +++ b/arch/s390/kvm/gaccess.c @@ -1419,8 +1419,8 @@ static int walk_guest_tables(struct gmap *sg, unsigned long saddr, struct pgtwal return kvm_s390_get_guest_page(kvm, entries + LEVEL_MEM, table.pte.pfra, wr); } -static int _do_shadow_pte(struct gmap *sg, gpa_t raddr, union pte *ptep_h, union pte *ptep, - struct guest_fault *f, bool p) +static int _do_shadow_pte(struct kvm_s390_mmu_cache *mc, struct gmap *sg, gpa_t raddr, + union pte *ptep_h, union pte *ptep, struct guest_fault *f, bool p) { union pgste pgste; union pte newpte; @@ -1430,7 +1430,7 @@ static int _do_shadow_pte(struct gmap *sg, gpa_t raddr, union pte *ptep_h, union lockdep_assert_held(&sg->parent->children_lock); scoped_guard(spinlock, &sg->host_to_rmap_lock) - rc = gmap_insert_rmap(sg, f->gfn, gpa_to_gfn(raddr), TABLE_TYPE_PAGE_TABLE); + rc = gmap_insert_rmap(mc, sg, f->gfn, gpa_to_gfn(raddr), TABLE_TYPE_PAGE_TABLE); if (rc) return rc; @@ -1462,8 +1462,8 @@ static int _do_shadow_pte(struct gmap *sg, gpa_t raddr, union pte *ptep_h, union return 0; } -static int _do_shadow_crste(struct gmap *sg, gpa_t raddr, union crste *host, union crste *table, - struct guest_fault *f, bool p) +static int _do_shadow_crste(struct kvm_s390_mmu_cache *mc, struct gmap *sg, gpa_t raddr, + union crste *host, union crste *table, struct guest_fault *f, bool p) { union crste newcrste, oldcrste; unsigned long mask; @@ -1476,7 +1476,7 @@ static int _do_shadow_crste(struct gmap *sg, gpa_t raddr, union crste *host, uni mask = is_pmd(*table) ? _SEGMENT_FR_MASK : _REGION3_FR_MASK; r_gfn = gpa_to_gfn(raddr) & mask; scoped_guard(spinlock, &sg->host_to_rmap_lock) - rc = gmap_insert_rmap(sg, f->gfn & mask, r_gfn, host->h.tt); + rc = gmap_insert_rmap(mc, sg, f->gfn & mask, r_gfn, host->h.tt); if (rc) return rc; @@ -1578,8 +1578,8 @@ static int _gaccess_do_shadow(struct kvm_s390_mmu_cache *mc, struct gmap *sg, if (KVM_BUG_ON(l > TABLE_TYPE_REGION3, sg->kvm)) return -EFAULT; if (l == TABLE_TYPE_PAGE_TABLE) - return _do_shadow_pte(sg, saddr, ptep_h, ptep, entries + LEVEL_MEM, w->p); - return _do_shadow_crste(sg, saddr, host, table, entries + LEVEL_MEM, w->p); + return _do_shadow_pte(mc, sg, saddr, ptep_h, ptep, entries + LEVEL_MEM, w->p); + return _do_shadow_crste(mc, sg, saddr, host, table, entries + LEVEL_MEM, w->p); } static inline int ___gaccess_shadow_fault(struct kvm_vcpu *vcpu, struct gmap *sg, gpa_t saddr, diff --git a/arch/s390/kvm/gmap.c b/arch/s390/kvm/gmap.c index 52d55ddea8d4..1d289f8fa3b2 100644 --- a/arch/s390/kvm/gmap.c +++ b/arch/s390/kvm/gmap.c @@ -1000,7 +1000,8 @@ int gmap_pv_destroy_range(struct gmap *gmap, gfn_t start, gfn_t end, bool interr return 0; } -int gmap_insert_rmap(struct gmap *sg, gfn_t p_gfn, gfn_t r_gfn, int level) +int gmap_insert_rmap(struct kvm_s390_mmu_cache *mc, struct gmap *sg, gfn_t p_gfn, + gfn_t r_gfn, int level) { struct vsie_rmap *rmap __free(kvfree) = NULL; struct vsie_rmap *temp; @@ -1010,7 +1011,7 @@ int gmap_insert_rmap(struct gmap *sg, gfn_t p_gfn, gfn_t r_gfn, int level) KVM_BUG_ON(!is_shadow(sg), sg->kvm); lockdep_assert_held(&sg->host_to_rmap_lock); - rmap = kzalloc_obj(*rmap, GFP_ATOMIC); + rmap = kvm_s390_mmu_cache_alloc_rmap(mc); if (!rmap) return -ENOMEM; @@ -1057,7 +1058,7 @@ int gmap_protect_rmap(struct kvm_s390_mmu_cache *mc, struct gmap *sg, gfn_t p_gf if (level <= TABLE_TYPE_REGION1) { bitmask = -1UL << (8 + 11 * level); scoped_guard(spinlock, &sg->host_to_rmap_lock) - rc = gmap_insert_rmap(sg, p_gfn, r_gfn & bitmask, level); + rc = gmap_insert_rmap(mc, sg, p_gfn, r_gfn & bitmask, level); } if (rc) return rc; diff --git a/arch/s390/kvm/gmap.h b/arch/s390/kvm/gmap.h index 20881e3ce9d8..1c040472f56d 100644 --- a/arch/s390/kvm/gmap.h +++ b/arch/s390/kvm/gmap.h @@ -100,7 +100,8 @@ int gmap_ucas_map(struct gmap *gmap, gfn_t p_gfn, gfn_t c_gfn, unsigned long cou void gmap_ucas_unmap(struct gmap *gmap, gfn_t c_gfn, unsigned long count); int gmap_enable_skeys(struct gmap *gmap); int gmap_pv_destroy_range(struct gmap *gmap, gfn_t start, gfn_t end, bool interruptible); -int gmap_insert_rmap(struct gmap *sg, gfn_t p_gfn, gfn_t r_gfn, int level); +int gmap_insert_rmap(struct kvm_s390_mmu_cache *mc, struct gmap *sg, gfn_t p_gfn, + gfn_t r_gfn, int level); int gmap_protect_rmap(struct kvm_s390_mmu_cache *mc, struct gmap *sg, gfn_t p_gfn, gfn_t r_gfn, kvm_pfn_t pfn, int level, bool wr); void gmap_set_cmma_all_dirty(struct gmap *gmap);