From 93a4b36ef3cf4ce5e6a7e7a7686181de76e246a1 Mon Sep 17 00:00:00 2001 From: Nirbhay Sharma Date: Fri, 3 Oct 2025 17:15:55 +0530 Subject: cgroup: Fix seqcount lockdep assertion in cgroup freezer The commit afa3701c0e45 ("cgroup: cgroup.stat.local time accounting") introduced a seqcount to track freeze timing but initialized it as a plain seqcount_t using seqcount_init(). However, the write-side critical section in cgroup_do_freeze() holds the css_set_lock spinlock while calling write_seqcount_begin(). On PREEMPT_RT kernels, spinlocks do not disable preemption, causing the lockdep assertion for a plain seqcount_t, which checks for preemption being disabled, to fail. This triggers the following warning: WARNING: CPU: 0 PID: 9692 at include/linux/seqlock.h:221 Fix this by changing the type to seqcount_spinlock_t and initializing it with seqcount_spinlock_init() to associate css_set_lock with the seqcount. This allows lockdep to correctly validate that the spinlock is held during write operations, resolving the assertion failure on all kernel configurations. Reported-by: syzbot+27a2519eb4dad86d0156@syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=27a2519eb4dad86d0156 Fixes: afa3701c0e45 ("cgroup: cgroup.stat.local time accounting") Signed-off-by: Nirbhay Sharma Link: https://lore.kernel.org/r/20251002165510.KtY3IT--@linutronix.de/ Acked-by: Michal Koutný Signed-off-by: Tejun Heo --- include/linux/cgroup-defs.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 93318fce31f3..b760a3c470a5 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -452,7 +452,7 @@ struct cgroup_freezer_state { int nr_frozen_tasks; /* Freeze time data consistency protection */ - seqcount_t freeze_seq; + seqcount_spinlock_t freeze_seq; /* * Most recent time the cgroup was requested to freeze. -- cgit v1.2.3 From 48b77733d0dbaf8cd0a122712072f92b2d95d894 Mon Sep 17 00:00:00 2001 From: Jan Kara Date: Wed, 1 Oct 2025 15:19:07 +0200 Subject: expfs: Fix exportfs_can_encode_fh() for EXPORT_FH_FID After commit 5402c4d4d200 ("exportfs: require ->fh_to_parent() to encode connectable file handles") we will fail to create non-decodable file handles for filesystems without export operations. Fix it. Fixes: 5402c4d4d200 ("exportfs: require ->fh_to_parent() to encode connectable file handles") Reviewed-by: Christian Brauner Reviewed-by: Amir Goldstein Signed-off-by: Jan Kara --- include/linux/exportfs.h | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/exportfs.h b/include/linux/exportfs.h index d0cf10d5e0f7..f0cf2714ec52 100644 --- a/include/linux/exportfs.h +++ b/include/linux/exportfs.h @@ -320,9 +320,6 @@ static inline bool exportfs_can_decode_fh(const struct export_operations *nop) static inline bool exportfs_can_encode_fh(const struct export_operations *nop, int fh_flags) { - if (!nop) - return false; - /* * If a non-decodeable file handle was requested, we only need to make * sure that filesystem did not opt-out of encoding fid. @@ -330,6 +327,10 @@ static inline bool exportfs_can_encode_fh(const struct export_operations *nop, if (fh_flags & EXPORT_FH_FID) return exportfs_can_encode_fid(nop); + /* Normal file handles cannot be created without export ops */ + if (!nop) + return false; + /* * If a connectable file handle was requested, we need to make sure that * filesystem can also decode connected file handles. -- cgit v1.2.3 From 11fb1a82aefa6f7fea6ac82334edb5639b9927df Mon Sep 17 00:00:00 2001 From: Sudeep Holla Date: Tue, 23 Sep 2025 16:09:27 +0100 Subject: firmware: arm_ffa: Add support for IMPDEF value in the memory access descriptor FF-A v1.2 introduced 16 byte IMPLEMENTATION DEFINED value in the endpoint memory access descriptor to allow any sender could to specify an its any custom value for each receiver. Also this value must be specified by the receiver when retrieving the memory region. The sender must ensure it informs the receiver of this value via an IMPLEMENTATION DEFINED mechanism such as a partition message. So the FF-A driver can use the message interfaces to communicate the value and set the same in the ffa_mem_region_attributes structures when using the memory interfaces. The driver ensure that the size of the endpoint memory access descriptors is set correctly based on the FF-A version. Fixes: 9fac08d9d985 ("firmware: arm_ffa: Upgrade FF-A version to v1.2 in the driver") Reported-by: Lixiang Mao Tested-by: Lixiang Mao Message-Id: <20250923150927.1218364-1-sudeep.holla@arm.com> Signed-off-by: Sudeep Holla --- drivers/firmware/arm_ffa/driver.c | 37 +++++++++++++++++++++++++++---------- include/linux/arm_ffa.h | 21 +++++++++++++++++++-- 2 files changed, 46 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/drivers/firmware/arm_ffa/driver.c b/drivers/firmware/arm_ffa/driver.c index 65bf1685350a..c72ee4756585 100644 --- a/drivers/firmware/arm_ffa/driver.c +++ b/drivers/firmware/arm_ffa/driver.c @@ -649,6 +649,26 @@ static u16 ffa_memory_attributes_get(u32 func_id) return FFA_MEM_NORMAL | FFA_MEM_WRITE_BACK | FFA_MEM_INNER_SHAREABLE; } +static void ffa_emad_impdef_value_init(u32 version, void *dst, void *src) +{ + struct ffa_mem_region_attributes *ep_mem_access; + + if (FFA_EMAD_HAS_IMPDEF_FIELD(version)) + memcpy(dst, src, sizeof(ep_mem_access->impdef_val)); +} + +static void +ffa_mem_region_additional_setup(u32 version, struct ffa_mem_region *mem_region) +{ + if (!FFA_MEM_REGION_HAS_EP_MEM_OFFSET(version)) { + mem_region->ep_mem_size = 0; + } else { + mem_region->ep_mem_size = ffa_emad_size_get(version); + mem_region->ep_mem_offset = sizeof(*mem_region); + memset(mem_region->reserved, 0, 12); + } +} + static int ffa_setup_and_transmit(u32 func_id, void *buffer, u32 max_fragsize, struct ffa_mem_ops_args *args) @@ -667,27 +687,24 @@ ffa_setup_and_transmit(u32 func_id, void *buffer, u32 max_fragsize, mem_region->flags = args->flags; mem_region->sender_id = drv_info->vm_id; mem_region->attributes = ffa_memory_attributes_get(func_id); - ep_mem_access = buffer + - ffa_mem_desc_offset(buffer, 0, drv_info->version); composite_offset = ffa_mem_desc_offset(buffer, args->nattrs, drv_info->version); - for (idx = 0; idx < args->nattrs; idx++, ep_mem_access++) { + for (idx = 0; idx < args->nattrs; idx++) { + ep_mem_access = buffer + + ffa_mem_desc_offset(buffer, idx, drv_info->version); ep_mem_access->receiver = args->attrs[idx].receiver; ep_mem_access->attrs = args->attrs[idx].attrs; ep_mem_access->composite_off = composite_offset; ep_mem_access->flag = 0; ep_mem_access->reserved = 0; + ffa_emad_impdef_value_init(drv_info->version, + ep_mem_access->impdef_val, + args->attrs[idx].impdef_val); } mem_region->handle = 0; mem_region->ep_count = args->nattrs; - if (drv_info->version <= FFA_VERSION_1_0) { - mem_region->ep_mem_size = 0; - } else { - mem_region->ep_mem_size = sizeof(*ep_mem_access); - mem_region->ep_mem_offset = sizeof(*mem_region); - memset(mem_region->reserved, 0, 12); - } + ffa_mem_region_additional_setup(drv_info->version, mem_region); composite = buffer + composite_offset; composite->total_pg_cnt = ffa_get_num_pages_sg(args->sg); diff --git a/include/linux/arm_ffa.h b/include/linux/arm_ffa.h index cd7ee4df9045..81e603839c4a 100644 --- a/include/linux/arm_ffa.h +++ b/include/linux/arm_ffa.h @@ -338,6 +338,7 @@ struct ffa_mem_region_attributes { * an `struct ffa_mem_region_addr_range`. */ u32 composite_off; + u8 impdef_val[16]; u64 reserved; }; @@ -417,15 +418,31 @@ struct ffa_mem_region { #define CONSTITUENTS_OFFSET(x) \ (offsetof(struct ffa_composite_mem_region, constituents[x])) +#define FFA_EMAD_HAS_IMPDEF_FIELD(version) ((version) >= FFA_VERSION_1_2) +#define FFA_MEM_REGION_HAS_EP_MEM_OFFSET(version) ((version) > FFA_VERSION_1_0) + +static inline u32 ffa_emad_size_get(u32 ffa_version) +{ + u32 sz; + struct ffa_mem_region_attributes *ep_mem_access; + + if (FFA_EMAD_HAS_IMPDEF_FIELD(ffa_version)) + sz = sizeof(*ep_mem_access); + else + sz = sizeof(*ep_mem_access) - sizeof(ep_mem_access->impdef_val); + + return sz; +} + static inline u32 ffa_mem_desc_offset(struct ffa_mem_region *buf, int count, u32 ffa_version) { - u32 offset = count * sizeof(struct ffa_mem_region_attributes); + u32 offset = count * ffa_emad_size_get(ffa_version); /* * Earlier to v1.1, the endpoint memory descriptor array started at * offset 32(i.e. offset of ep_mem_offset in the current structure) */ - if (ffa_version <= FFA_VERSION_1_0) + if (!FFA_MEM_REGION_HAS_EP_MEM_OFFSET(ffa_version)) offset += offsetof(struct ffa_mem_region, ep_mem_offset); else offset += sizeof(struct ffa_mem_region); -- cgit v1.2.3 From c97513cddcfc235f2522617980838e500af21d01 Mon Sep 17 00:00:00 2001 From: Lance Yang Date: Tue, 9 Sep 2025 22:52:43 +0800 Subject: hung_task: fix warnings caused by unaligned lock pointers The blocker tracking mechanism assumes that lock pointers are at least 4-byte aligned to use their lower bits for type encoding. However, as reported by Eero Tamminen, some architectures like m68k only guarantee 2-byte alignment of 32-bit values. This breaks the assumption and causes two related WARN_ON_ONCE checks to trigger. To fix this, the runtime checks are adjusted to silently ignore any lock that is not 4-byte aligned, effectively disabling the feature in such cases and avoiding the related warnings. Thanks to Geert Uytterhoeven for bisecting! Link: https://lkml.kernel.org/r/20250909145243.17119-1-lance.yang@linux.dev Fixes: e711faaafbe5 ("hung_task: replace blocker_mutex with encoded blocker") Signed-off-by: Lance Yang Reported-by: Eero Tamminen Closes: https://lore.kernel.org/lkml/CAMuHMdW7Ab13DdGs2acMQcix5ObJK0O2dG_Fxzr8_g58Rc1_0g@mail.gmail.com Reviewed-by: Masami Hiramatsu (Google) Cc: John Paul Adrian Glaubitz Cc: Anna Schumaker Cc: Boqun Feng Cc: Finn Thain Cc: Geert Uytterhoeven Cc: Ingo Molnar Cc: Joel Granados Cc: John Stultz Cc: Kent Overstreet Cc: Lance Yang Cc: Mingzhe Yang Cc: Peter Zijlstra Cc: Sergey Senozhatsky Cc: Steven Rostedt Cc: Tomasz Figa Cc: Waiman Long Cc: Will Deacon Cc: Yongliang Gao Cc: Signed-off-by: Andrew Morton --- include/linux/hung_task.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/hung_task.h b/include/linux/hung_task.h index 34e615c76ca5..c4403eeb7144 100644 --- a/include/linux/hung_task.h +++ b/include/linux/hung_task.h @@ -20,6 +20,10 @@ * always zero. So we can use these bits to encode the specific blocking * type. * + * Note that on architectures where this is not guaranteed, or for any + * unaligned lock, this tracking mechanism is silently skipped for that + * lock. + * * Type encoding: * 00 - Blocked on mutex (BLOCKER_TYPE_MUTEX) * 01 - Blocked on semaphore (BLOCKER_TYPE_SEM) @@ -45,7 +49,7 @@ static inline void hung_task_set_blocker(void *lock, unsigned long type) * If the lock pointer matches the BLOCKER_TYPE_MASK, return * without writing anything. */ - if (WARN_ON_ONCE(lock_ptr & BLOCKER_TYPE_MASK)) + if (lock_ptr & BLOCKER_TYPE_MASK) return; WRITE_ONCE(current->blocker, lock_ptr | type); @@ -53,8 +57,6 @@ static inline void hung_task_set_blocker(void *lock, unsigned long type) static inline void hung_task_clear_blocker(void) { - WARN_ON_ONCE(!READ_ONCE(current->blocker)); - WRITE_ONCE(current->blocker, 0UL); } -- cgit v1.2.3 From 0fbbcab7f9082cdc233da5e5e353f69830f11956 Mon Sep 17 00:00:00 2001 From: Randy Dunlap Date: Fri, 17 Oct 2025 00:07:42 -0700 Subject: cgroup/misc: fix misc_res_type kernel-doc warning Format the kernel-doc for SCALE_HW_CALIB_INVALID correctly to avoid a kernel-doc warning: Warning: include/linux/misc_cgroup.h:26 Enum value 'MISC_CG_RES_TDX' not described in enum 'misc_res_type' Fixes: 7c035bea9407 ("KVM: TDX: Register TDX host key IDs to cgroup misc controller") Signed-off-by: Randy Dunlap Signed-off-by: Tejun Heo --- include/linux/misc_cgroup.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/misc_cgroup.h b/include/linux/misc_cgroup.h index 71cf5bfc6349..0cb36a3ffc47 100644 --- a/include/linux/misc_cgroup.h +++ b/include/linux/misc_cgroup.h @@ -19,7 +19,7 @@ enum misc_res_type { MISC_CG_RES_SEV_ES, #endif #ifdef CONFIG_INTEL_TDX_HOST - /* Intel TDX HKIDs resource */ + /** @MISC_CG_RES_TDX: Intel TDX HKIDs resource */ MISC_CG_RES_TDX, #endif /** @MISC_CG_RES_TYPES: count of enum misc_res_type constants */ -- cgit v1.2.3 From ce831bffcef3d8f9691b5537d74ffa1b1256c017 Mon Sep 17 00:00:00 2001 From: Thomas Hellström Date: Wed, 15 Oct 2025 19:07:26 +0200 Subject: drm/xe/uapi: Hide the madvise autoreset behind a VM_BIND flag The madvise implementation currently resets the SVM madvise if the underlying CPU map is unmapped. This is in an attempt to mimic the CPU madvise behaviour. However, it's not clear that this is a desired behaviour since if the end app user relies on it for malloc()ed objects or stack objects, it may not work as intended. Instead of having the autoreset functionality being a direct application-facing implicit UAPI, make the UMD explicitly choose this behaviour if it wants to expose it by introducing DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET, and add a semantics description. v2: - Kerneldoc fixes. Fix a commit log message. Fixes: a2eb8aec3ebe ("drm/xe: Reset VMA attributes to default in SVM garbage collector") Cc: Matthew Brost Cc: Himal Prasad Ghimiray Cc: "Falkowski, John" Cc: "Mrozek, Michal" Signed-off-by: Thomas Hellström Reviewed-by: Himal Prasad Ghimiray Link: https://lore.kernel.org/r/20251015170726.178685-2-thomas.hellstrom@linux.intel.com (cherry picked from commit 59a2d3f38ab23cce4cd9f0c4a5e08fdfe9e67ae7) Signed-off-by: Lucas De Marchi --- drivers/gpu/drm/xe/xe_svm.c | 5 +++++ drivers/gpu/drm/xe/xe_vm.c | 12 +++++++++--- drivers/gpu/drm/xe/xe_vm_types.h | 1 + include/uapi/drm/xe_drm.h | 15 +++++++++++++++ 4 files changed, 30 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/drivers/gpu/drm/xe/xe_svm.c b/drivers/gpu/drm/xe/xe_svm.c index da2a412f80c0..129e7818565c 100644 --- a/drivers/gpu/drm/xe/xe_svm.c +++ b/drivers/gpu/drm/xe/xe_svm.c @@ -302,6 +302,11 @@ static int xe_svm_range_set_default_attr(struct xe_vm *vm, u64 range_start, u64 if (!vma) return -EINVAL; + if (!(vma->gpuva.flags & XE_VMA_MADV_AUTORESET)) { + drm_dbg(&vm->xe->drm, "Skipping madvise reset for vma.\n"); + return 0; + } + if (xe_vma_has_default_mem_attrs(vma)) return 0; diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c index f7a0931eb66c..63c65e3d207b 100644 --- a/drivers/gpu/drm/xe/xe_vm.c +++ b/drivers/gpu/drm/xe/xe_vm.c @@ -620,7 +620,8 @@ static void xe_vma_ops_incr_pt_update_ops(struct xe_vma_ops *vops, u8 tile_mask, XE_VMA_READ_ONLY | \ XE_VMA_DUMPABLE | \ XE_VMA_SYSTEM_ALLOCATOR | \ - DRM_GPUVA_SPARSE) + DRM_GPUVA_SPARSE | \ + XE_VMA_MADV_AUTORESET) static void xe_vm_populate_rebind(struct xe_vma_op *op, struct xe_vma *vma, u8 tile_mask) @@ -2270,6 +2271,8 @@ vm_bind_ioctl_ops_create(struct xe_vm *vm, struct xe_vma_ops *vops, op->map.vma_flags |= XE_VMA_SYSTEM_ALLOCATOR; if (flags & DRM_XE_VM_BIND_FLAG_DUMPABLE) op->map.vma_flags |= XE_VMA_DUMPABLE; + if (flags & DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET) + op->map.vma_flags |= XE_VMA_MADV_AUTORESET; op->map.pat_index = pat_index; op->map.invalidate_on_bind = __xe_vm_needs_clear_scratch_pages(vm, flags); @@ -3253,7 +3256,8 @@ ALLOW_ERROR_INJECTION(vm_bind_ioctl_ops_execute, ERRNO); DRM_XE_VM_BIND_FLAG_NULL | \ DRM_XE_VM_BIND_FLAG_DUMPABLE | \ DRM_XE_VM_BIND_FLAG_CHECK_PXP | \ - DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR) + DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR | \ + DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET) #ifdef TEST_VM_OPS_ERROR #define SUPPORTED_FLAGS (SUPPORTED_FLAGS_STUB | FORCE_OP_ERROR) @@ -3368,7 +3372,9 @@ static int vm_bind_ioctl_check_args(struct xe_device *xe, struct xe_vm *vm, XE_IOCTL_DBG(xe, (prefetch_region != DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC && !(BIT(prefetch_region) & xe->info.mem_region_mask))) || XE_IOCTL_DBG(xe, obj && - op == DRM_XE_VM_BIND_OP_UNMAP)) { + op == DRM_XE_VM_BIND_OP_UNMAP) || + XE_IOCTL_DBG(xe, (flags & DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET) && + (!is_cpu_addr_mirror || op != DRM_XE_VM_BIND_OP_MAP))) { err = -EINVAL; goto free_bind_ops; } diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h index a3b422b27ae8..d6e2a0fdd4b3 100644 --- a/drivers/gpu/drm/xe/xe_vm_types.h +++ b/drivers/gpu/drm/xe/xe_vm_types.h @@ -46,6 +46,7 @@ struct xe_vm_pgtable_update_op; #define XE_VMA_PTE_COMPACT (DRM_GPUVA_USERBITS << 7) #define XE_VMA_DUMPABLE (DRM_GPUVA_USERBITS << 8) #define XE_VMA_SYSTEM_ALLOCATOR (DRM_GPUVA_USERBITS << 9) +#define XE_VMA_MADV_AUTORESET (DRM_GPUVA_USERBITS << 10) /** * struct xe_vma_mem_attr - memory attributes associated with vma diff --git a/include/uapi/drm/xe_drm.h b/include/uapi/drm/xe_drm.h index 40ff19f52a8d..517489a7ec60 100644 --- a/include/uapi/drm/xe_drm.h +++ b/include/uapi/drm/xe_drm.h @@ -1013,6 +1013,20 @@ struct drm_xe_vm_destroy { * valid on VMs with DRM_XE_VM_CREATE_FLAG_FAULT_MODE set. The CPU address * mirror flag are only valid for DRM_XE_VM_BIND_OP_MAP operations, the BO * handle MBZ, and the BO offset MBZ. + * - %DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET - Can be used in combination with + * %DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR to reset madvises when the underlying + * CPU address space range is unmapped (typically with munmap(2) or brk(2)). + * The madvise values set with &DRM_IOCTL_XE_MADVISE are reset to the values + * that were present immediately after the &DRM_IOCTL_XE_VM_BIND. + * The reset GPU virtual address range is the intersection of the range bound + * using &DRM_IOCTL_XE_VM_BIND and the virtual CPU address space range + * unmapped. + * This functionality is present to mimic the behaviour of CPU address space + * madvises set using madvise(2), which are typically reset on unmap. + * Note: free(3) may or may not call munmap(2) and/or brk(2), and may thus + * not invoke autoreset. Neither will stack variables going out of scope. + * Therefore it's recommended to always explicitly reset the madvises when + * freeing the memory backing a region used in a &DRM_IOCTL_XE_MADVISE call. * * The @prefetch_mem_region_instance for %DRM_XE_VM_BIND_OP_PREFETCH can also be: * - %DRM_XE_CONSULT_MEM_ADVISE_PREF_LOC, which ensures prefetching occurs in @@ -1119,6 +1133,7 @@ struct drm_xe_vm_bind_op { #define DRM_XE_VM_BIND_FLAG_DUMPABLE (1 << 3) #define DRM_XE_VM_BIND_FLAG_CHECK_PXP (1 << 4) #define DRM_XE_VM_BIND_FLAG_CPU_ADDR_MIRROR (1 << 5) +#define DRM_XE_VM_BIND_FLAG_MADVISE_AUTORESET (1 << 6) /** @flags: Bind flags */ __u32 flags; -- cgit v1.2.3 From d54c676d4fe0543d1642ab7a68ffdd31e8639a5d Mon Sep 17 00:00:00 2001 From: Bart Van Assche Date: Tue, 14 Oct 2025 15:02:43 -0700 Subject: scsi: core: Fix the unit attention counter implementation scsi_decide_disposition() may call scsi_check_sense(). scsi_decide_disposition() calls are not serialized. Hence, counter updates by scsi_check_sense() must be serialized. Hence this patch that makes the counters updated by scsi_check_sense() atomic. Cc: Kai Mäkisara Fixes: a5d518cd4e3e ("scsi: core: Add counters for New Media and Power On/Reset UNIT ATTENTIONs") Signed-off-by: Bart Van Assche Reviewed-by: Ewan D. Milne Link: https://patch.msgid.link/20251014220244.3689508-1-bvanassche@acm.org Signed-off-by: Martin K. Petersen --- drivers/scsi/scsi_error.c | 4 ++-- include/scsi/scsi_device.h | 10 ++++------ 2 files changed, 6 insertions(+), 8 deletions(-) (limited to 'include') diff --git a/drivers/scsi/scsi_error.c b/drivers/scsi/scsi_error.c index 746ff6a1f309..1c13812a3f03 100644 --- a/drivers/scsi/scsi_error.c +++ b/drivers/scsi/scsi_error.c @@ -554,9 +554,9 @@ enum scsi_disposition scsi_check_sense(struct scsi_cmnd *scmd) * happened, even if someone else gets the sense data. */ if (sshdr.asc == 0x28) - scmd->device->ua_new_media_ctr++; + atomic_inc(&sdev->ua_new_media_ctr); else if (sshdr.asc == 0x29) - scmd->device->ua_por_ctr++; + atomic_inc(&sdev->ua_por_ctr); } if (scsi_sense_is_deferred(&sshdr)) diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h index 6d6500148c4b..993008cdea65 100644 --- a/include/scsi/scsi_device.h +++ b/include/scsi/scsi_device.h @@ -252,8 +252,8 @@ struct scsi_device { unsigned int queue_stopped; /* request queue is quiesced */ bool offline_already; /* Device offline message logged */ - unsigned int ua_new_media_ctr; /* Counter for New Media UNIT ATTENTIONs */ - unsigned int ua_por_ctr; /* Counter for Power On / Reset UAs */ + atomic_t ua_new_media_ctr; /* Counter for New Media UNIT ATTENTIONs */ + atomic_t ua_por_ctr; /* Counter for Power On / Reset UAs */ atomic_t disk_events_disable_depth; /* disable depth for disk events */ @@ -693,10 +693,8 @@ static inline int scsi_device_busy(struct scsi_device *sdev) } /* Macros to access the UNIT ATTENTION counters */ -#define scsi_get_ua_new_media_ctr(sdev) \ - ((const unsigned int)(sdev->ua_new_media_ctr)) -#define scsi_get_ua_por_ctr(sdev) \ - ((const unsigned int)(sdev->ua_por_ctr)) +#define scsi_get_ua_new_media_ctr(sdev) atomic_read(&sdev->ua_new_media_ctr) +#define scsi_get_ua_por_ctr(sdev) atomic_read(&sdev->ua_por_ctr) #define MODULE_ALIAS_SCSI_DEVICE(type) \ MODULE_ALIAS("scsi:t-" __stringify(type) "*") -- cgit v1.2.3 From 00aaae60faf554c27c95e93d47f200a93ff266ef Mon Sep 17 00:00:00 2001 From: Ioana Ciornei Date: Tue, 14 Oct 2025 18:53:53 +0300 Subject: gpio: regmap: add the .fixed_direction_output configuration parameter There are GPIO controllers such as the one present in the LX2160ARDB QIXIS FPGA which have fixed-direction input and output GPIO lines mixed together in a single register. This cannot be modeled using the gpio-regmap as-is since there is no way to present the true direction of a GPIO line. In order to make this use case possible, add a new configuration parameter - fixed_direction_output - into the gpio_regmap_config structure. This will enable user drivers to provide a bitmap that represents the fixed direction of the GPIO lines. Signed-off-by: Ioana Ciornei Acked-by: Bartosz Golaszewski Reviewed-by: Michael Walle Signed-off-by: Bartosz Golaszewski --- drivers/gpio/gpio-regmap.c | 26 ++++++++++++++++++++++++-- include/linux/gpio/regmap.h | 5 +++++ 2 files changed, 29 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/drivers/gpio/gpio-regmap.c b/drivers/gpio/gpio-regmap.c index ab9e4077fa60..f4267af00027 100644 --- a/drivers/gpio/gpio-regmap.c +++ b/drivers/gpio/gpio-regmap.c @@ -31,6 +31,7 @@ struct gpio_regmap { unsigned int reg_clr_base; unsigned int reg_dir_in_base; unsigned int reg_dir_out_base; + unsigned long *fixed_direction_output; #ifdef CONFIG_REGMAP_IRQ int regmap_irq_line; @@ -134,6 +135,13 @@ static int gpio_regmap_get_direction(struct gpio_chip *chip, unsigned int base, val, reg, mask; int invert, ret; + if (gpio->fixed_direction_output) { + if (test_bit(offset, gpio->fixed_direction_output)) + return GPIO_LINE_DIRECTION_OUT; + else + return GPIO_LINE_DIRECTION_IN; + } + if (gpio->reg_dat_base && !gpio->reg_set_base) return GPIO_LINE_DIRECTION_IN; if (gpio->reg_set_base && !gpio->reg_dat_base) @@ -284,6 +292,17 @@ struct gpio_regmap *gpio_regmap_register(const struct gpio_regmap_config *config goto err_free_gpio; } + if (config->fixed_direction_output) { + gpio->fixed_direction_output = bitmap_alloc(chip->ngpio, + GFP_KERNEL); + if (!gpio->fixed_direction_output) { + ret = -ENOMEM; + goto err_free_gpio; + } + bitmap_copy(gpio->fixed_direction_output, + config->fixed_direction_output, chip->ngpio); + } + /* if not set, assume there is only one register */ gpio->ngpio_per_reg = config->ngpio_per_reg; if (!gpio->ngpio_per_reg) @@ -300,7 +319,7 @@ struct gpio_regmap *gpio_regmap_register(const struct gpio_regmap_config *config ret = gpiochip_add_data(chip, gpio); if (ret < 0) - goto err_free_gpio; + goto err_free_bitmap; #ifdef CONFIG_REGMAP_IRQ if (config->regmap_irq_chip) { @@ -309,7 +328,7 @@ struct gpio_regmap *gpio_regmap_register(const struct gpio_regmap_config *config config->regmap_irq_line, config->regmap_irq_flags, 0, config->regmap_irq_chip, &gpio->irq_chip_data); if (ret) - goto err_free_gpio; + goto err_free_bitmap; irq_domain = regmap_irq_get_domain(gpio->irq_chip_data); } else @@ -326,6 +345,8 @@ struct gpio_regmap *gpio_regmap_register(const struct gpio_regmap_config *config err_remove_gpiochip: gpiochip_remove(chip); +err_free_bitmap: + bitmap_free(gpio->fixed_direction_output); err_free_gpio: kfree(gpio); return ERR_PTR(ret); @@ -344,6 +365,7 @@ void gpio_regmap_unregister(struct gpio_regmap *gpio) #endif gpiochip_remove(&gpio->gpio_chip); + bitmap_free(gpio->fixed_direction_output); kfree(gpio); } EXPORT_SYMBOL_GPL(gpio_regmap_unregister); diff --git a/include/linux/gpio/regmap.h b/include/linux/gpio/regmap.h index 622a2939ebe0..87983a5f3681 100644 --- a/include/linux/gpio/regmap.h +++ b/include/linux/gpio/regmap.h @@ -38,6 +38,10 @@ struct regmap; * offset to a register/bitmask pair. If not * given the default gpio_regmap_simple_xlate() * is used. + * @fixed_direction_output: + * (Optional) Bitmap representing the fixed direction of + * the GPIO lines. Useful when there are GPIO lines with a + * fixed direction mixed together in the same register. * @drvdata: (Optional) Pointer to driver specific data which is * not used by gpio-remap but is provided "as is" to the * driver callback(s). @@ -85,6 +89,7 @@ struct gpio_regmap_config { int reg_stride; int ngpio_per_reg; struct irq_domain *irq_domain; + unsigned long *fixed_direction_output; #ifdef CONFIG_REGMAP_IRQ struct regmap_irq_chip *regmap_irq_chip; -- cgit v1.2.3 From db82b8dbf5f06d7b1abec4e1326ed8c02fa16897 Mon Sep 17 00:00:00 2001 From: Rafael J. Wysocki Date: Mon, 20 Oct 2025 17:03:28 +0200 Subject: PM: runtime: Fix conditional guard definitions Since pm_runtime_get_active() returns 0 on success, all of the DEFINE_GUARD_COND() macros in pm_runtime.h need the "_RET == 0" condition at the end of the argument list or they would not work correctly. Fixes: 9a0abc39450a ("PM: runtime: Add auto-cleanup macros for "resume and get" operations") Reported-by: kernel test robot Link: https://lore.kernel.org/linux-pm/202510191529.BCyjKlLQ-lkp@intel.com/ Signed-off-by: Rafael J. Wysocki Reviewed-by: Jonathan Cameron Reviewed-by: Dan Williams Tested-by: Farhan Ali Link: https://patch.msgid.link/5943878.DvuYhMxLoT@rafael.j.wysocki --- include/linux/pm_runtime.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/pm_runtime.h b/include/linux/pm_runtime.h index a3f44f6c2da1..0b436e15f4cd 100644 --- a/include/linux/pm_runtime.h +++ b/include/linux/pm_runtime.h @@ -629,13 +629,13 @@ DEFINE_GUARD(pm_runtime_active_auto, struct device *, * device. */ DEFINE_GUARD_COND(pm_runtime_active, _try, - pm_runtime_get_active(_T, RPM_TRANSPARENT)) + pm_runtime_get_active(_T, RPM_TRANSPARENT), _RET == 0) DEFINE_GUARD_COND(pm_runtime_active, _try_enabled, - pm_runtime_resume_and_get(_T)) + pm_runtime_resume_and_get(_T), _RET == 0) DEFINE_GUARD_COND(pm_runtime_active_auto, _try, - pm_runtime_get_active(_T, RPM_TRANSPARENT)) + pm_runtime_get_active(_T, RPM_TRANSPARENT), _RET == 0) DEFINE_GUARD_COND(pm_runtime_active_auto, _try_enabled, - pm_runtime_resume_and_get(_T)) + pm_runtime_resume_and_get(_T), _RET == 0) /** * pm_runtime_put_sync - Drop device usage counter and run "idle check" if 0. -- cgit v1.2.3 From f6ceec6434b5efff62cecbaa2ff74fc29b96c0c6 Mon Sep 17 00:00:00 2001 From: Ralf Lici Date: Tue, 21 Oct 2025 12:09:40 +0200 Subject: net: datagram: introduce datagram_poll_queue for custom receive queues Some protocols using TCP encapsulation (e.g., espintcp, openvpn) deliver userspace-bound packets through a custom skb queue rather than the standard sk_receive_queue. Introduce datagram_poll_queue that accepts an explicit receive queue, and convert datagram_poll into a wrapper around datagram_poll_queue. This allows protocols with custom skb queues to reuse the core polling logic without relying on sk_receive_queue. Cc: Sabrina Dubroca Cc: Antonio Quartulli Signed-off-by: Ralf Lici Reviewed-by: Sabrina Dubroca Reviewed-by: Antonio Quartulli Link: https://patch.msgid.link/20251021100942.195010-2-ralf@mandelbit.com Signed-off-by: Paolo Abeni --- include/linux/skbuff.h | 3 +++ net/core/datagram.c | 44 ++++++++++++++++++++++++++++++++++---------- 2 files changed, 37 insertions(+), 10 deletions(-) (limited to 'include') diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index fb3fec9affaa..a7cc3d1f4fd1 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -4204,6 +4204,9 @@ struct sk_buff *__skb_recv_datagram(struct sock *sk, struct sk_buff_head *sk_queue, unsigned int flags, int *off, int *err); struct sk_buff *skb_recv_datagram(struct sock *sk, unsigned int flags, int *err); +__poll_t datagram_poll_queue(struct file *file, struct socket *sock, + struct poll_table_struct *wait, + struct sk_buff_head *rcv_queue); __poll_t datagram_poll(struct file *file, struct socket *sock, struct poll_table_struct *wait); int skb_copy_datagram_iter(const struct sk_buff *from, int offset, diff --git a/net/core/datagram.c b/net/core/datagram.c index cb4b9ef2e4e3..c285c6465923 100644 --- a/net/core/datagram.c +++ b/net/core/datagram.c @@ -920,21 +920,22 @@ fault: EXPORT_SYMBOL(skb_copy_and_csum_datagram_msg); /** - * datagram_poll - generic datagram poll + * datagram_poll_queue - same as datagram_poll, but on a specific receive + * queue * @file: file struct * @sock: socket * @wait: poll table + * @rcv_queue: receive queue to poll * - * Datagram poll: Again totally generic. This also handles - * sequenced packet sockets providing the socket receive queue - * is only ever holding data ready to receive. + * Performs polling on the given receive queue, handling shutdown, error, + * and connection state. This is useful for protocols that deliver + * userspace-bound packets through a custom queue instead of + * sk->sk_receive_queue. * - * Note: when you *don't* use this routine for this protocol, - * and you use a different write policy from sock_writeable() - * then please supply your own write_space callback. + * Return: poll bitmask indicating the socket's current state */ -__poll_t datagram_poll(struct file *file, struct socket *sock, - poll_table *wait) +__poll_t datagram_poll_queue(struct file *file, struct socket *sock, + poll_table *wait, struct sk_buff_head *rcv_queue) { struct sock *sk = sock->sk; __poll_t mask; @@ -956,7 +957,7 @@ __poll_t datagram_poll(struct file *file, struct socket *sock, mask |= EPOLLHUP; /* readable? */ - if (!skb_queue_empty_lockless(&sk->sk_receive_queue)) + if (!skb_queue_empty_lockless(rcv_queue)) mask |= EPOLLIN | EPOLLRDNORM; /* Connection-based need to check for termination and startup */ @@ -978,4 +979,27 @@ __poll_t datagram_poll(struct file *file, struct socket *sock, return mask; } +EXPORT_SYMBOL(datagram_poll_queue); + +/** + * datagram_poll - generic datagram poll + * @file: file struct + * @sock: socket + * @wait: poll table + * + * Datagram poll: Again totally generic. This also handles + * sequenced packet sockets providing the socket receive queue + * is only ever holding data ready to receive. + * + * Note: when you *don't* use this routine for this protocol, + * and you use a different write policy from sock_writeable() + * then please supply your own write_space callback. + * + * Return: poll bitmask indicating the socket's current state + */ +__poll_t datagram_poll(struct file *file, struct socket *sock, poll_table *wait) +{ + return datagram_poll_queue(file, sock, wait, + &sock->sk->sk_receive_queue); +} EXPORT_SYMBOL(datagram_poll); -- cgit v1.2.3 From b2284768c6b32aa224ca7d0ef0741beb434f03aa Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Wed, 22 Oct 2025 11:44:21 +0800 Subject: virtio-net: zero unused hash fields When GSO tunnel is negotiated virtio_net_hdr_tnl_from_skb() tries to initialize the tunnel metadata but forget to zero unused rxhash fields. This may leak information to another side. Fixing this by zeroing the unused hash fields. Acked-by: Michael S. Tsirkin Fixes: a2fb4bc4e2a6a ("net: implement virtio helpers to handle UDP GSO tunneling") Cc: Signed-off-by: Jason Wang Reviewed-by: Xuan Zhuo Link: https://patch.msgid.link/20251022034421.70244-1-jasowang@redhat.com Signed-off-by: Jakub Kicinski --- include/linux/virtio_net.h | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'include') diff --git a/include/linux/virtio_net.h b/include/linux/virtio_net.h index 20e0584db1dd..4d1780848d0e 100644 --- a/include/linux/virtio_net.h +++ b/include/linux/virtio_net.h @@ -401,6 +401,10 @@ virtio_net_hdr_tnl_from_skb(const struct sk_buff *skb, if (!tnl_hdr_negotiated) return -EINVAL; + vhdr->hash_hdr.hash_value = 0; + vhdr->hash_hdr.hash_report = 0; + vhdr->hash_hdr.padding = 0; + /* Let the basic parsing deal with plain GSO features. */ skb_shinfo(skb)->gso_type &= ~tnl_gso_type; ret = virtio_net_hdr_from_skb(skb, hdr, true, false, vlan_hlen); -- cgit v1.2.3 From bb65e0c141f879cdf54db11ae446ee3605fb54d5 Mon Sep 17 00:00:00 2001 From: Alexei Lazar Date: Wed, 22 Oct 2025 15:29:39 +0300 Subject: net/mlx5: Add PPHCR to PCAM supported registers mask Add the PPHCR bit to the port_access_reg_cap_mask field of PCAM register to indicate that the device supports the PPHCR register and the RS-FEC histogram feature. Signed-off-by: Alexei Lazar Reviewed-by: Yael Chemla Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761136182-918470-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- include/linux/mlx5/mlx5_ifc.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h index 07614cd95bed..1b0b36aa2a76 100644 --- a/include/linux/mlx5/mlx5_ifc.h +++ b/include/linux/mlx5/mlx5_ifc.h @@ -10833,7 +10833,9 @@ struct mlx5_ifc_pcam_regs_5000_to_507f_bits { u8 port_access_reg_cap_mask_127_to_96[0x20]; u8 port_access_reg_cap_mask_95_to_64[0x20]; - u8 port_access_reg_cap_mask_63_to_36[0x1c]; + u8 port_access_reg_cap_mask_63[0x1]; + u8 pphcr[0x1]; + u8 port_access_reg_cap_mask_61_to_36[0x1a]; u8 pplm[0x1]; u8 port_access_reg_cap_mask_34_to_32[0x3]; -- cgit v1.2.3 From 0d92808024b4e9868cef68d16f121d509843e80e Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 1 Oct 2025 10:55:58 -0400 Subject: Bluetooth: HCI: Fix tracking of advertisement set/instance 0x00 This fixes the state tracking of advertisement set/instance 0x00 which is considered a legacy instance and is not tracked individually by adv_instances list, previously it was assumed that hci_dev itself would track it via HCI_LE_ADV but that is a global state not specifc to instance 0x00, so to fix it a new flag is introduced that only tracks the state of instance 0x00. Fixes: 1488af7b8b5f ("Bluetooth: hci_sync: Fix hci_resume_advertising_sync") Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci.h | 1 + net/bluetooth/hci_event.c | 4 ++++ net/bluetooth/hci_sync.c | 5 ++--- 3 files changed, 7 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 9ecc70baaca9..8d0e703bc929 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -434,6 +434,7 @@ enum { HCI_USER_CHANNEL, HCI_EXT_CONFIGURED, HCI_LE_ADV, + HCI_LE_ADV_0, HCI_LE_PER_ADV, HCI_LE_SCAN, HCI_SSP_ENABLED, diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index d790b0d4eb9a..1dabf5a7ae18 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1609,6 +1609,8 @@ static u8 hci_cc_le_set_ext_adv_enable(struct hci_dev *hdev, void *data, if (adv && !adv->periodic) adv->enabled = true; + else if (!set->handle) + hci_dev_set_flag(hdev, HCI_LE_ADV_0); conn = hci_lookup_le_connect(hdev); if (conn) @@ -1619,6 +1621,8 @@ static u8 hci_cc_le_set_ext_adv_enable(struct hci_dev *hdev, void *data, if (cp->num_of_sets) { if (adv) adv->enabled = false; + else if (!set->handle) + hci_dev_clear_flag(hdev, HCI_LE_ADV_0); /* If just one instance was disabled check if there are * any other instance enabled before clearing HCI_LE_ADV diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index d160e5e1fe8a..28ad08cd7d70 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -2606,9 +2606,8 @@ static int hci_resume_advertising_sync(struct hci_dev *hdev) /* If current advertising instance is set to instance 0x00 * then we need to re-enable it. */ - if (!hdev->cur_adv_instance) - err = hci_enable_ext_advertising_sync(hdev, - hdev->cur_adv_instance); + if (hci_dev_test_and_clear_flag(hdev, HCI_LE_ADV_0)) + err = hci_enable_ext_advertising_sync(hdev, 0x00); } else { /* Schedule for most recent instance to be restarted and begin * the software rotation loop -- cgit v1.2.3 From e8785404de06a69d89dcdd1e9a0b6ea42dc6d327 Mon Sep 17 00:00:00 2001 From: Pauli Virtanen Date: Fri, 3 Oct 2025 22:07:32 +0300 Subject: Bluetooth: MGMT: fix crash in set_mesh_sync and set_mesh_complete There is a BUG: KASAN: stack-out-of-bounds in set_mesh_sync due to memcpy from badly declared on-stack flexible array. Another crash is in set_mesh_complete() due to double list_del via mgmt_pending_valid + mgmt_pending_remove. Use DEFINE_FLEX to declare the flexible array right, and don't memcpy outside bounds. As mgmt_pending_valid removes the cmd from list, use mgmt_pending_free, and also report status on error. Fixes: 302a1f674c00d ("Bluetooth: MGMT: Fix possible UAFs") Signed-off-by: Pauli Virtanen Reviewed-by: Paul Menzel Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/mgmt.h | 2 +- net/bluetooth/mgmt.c | 26 +++++++++++++++----------- 2 files changed, 16 insertions(+), 12 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/mgmt.h b/include/net/bluetooth/mgmt.h index 74edea06985b..bca0333f1e99 100644 --- a/include/net/bluetooth/mgmt.h +++ b/include/net/bluetooth/mgmt.h @@ -853,7 +853,7 @@ struct mgmt_cp_set_mesh { __le16 window; __le16 period; __u8 num_ad_types; - __u8 ad_types[]; + __u8 ad_types[] __counted_by(num_ad_types); } __packed; #define MGMT_SET_MESH_RECEIVER_SIZE 6 diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index a3d16eece0d2..24e335e3a727 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -2175,19 +2175,24 @@ static void set_mesh_complete(struct hci_dev *hdev, void *data, int err) sk = cmd->sk; if (status) { + mgmt_cmd_status(cmd->sk, hdev->id, MGMT_OP_SET_MESH_RECEIVER, + status); mgmt_pending_foreach(MGMT_OP_SET_MESH_RECEIVER, hdev, true, cmd_status_rsp, &status); - return; + goto done; } - mgmt_pending_remove(cmd); mgmt_cmd_complete(sk, hdev->id, MGMT_OP_SET_MESH_RECEIVER, 0, NULL, 0); + +done: + mgmt_pending_free(cmd); } static int set_mesh_sync(struct hci_dev *hdev, void *data) { struct mgmt_pending_cmd *cmd = data; - struct mgmt_cp_set_mesh cp; + DEFINE_FLEX(struct mgmt_cp_set_mesh, cp, ad_types, num_ad_types, + sizeof(hdev->mesh_ad_types)); size_t len; mutex_lock(&hdev->mgmt_pending_lock); @@ -2197,27 +2202,26 @@ static int set_mesh_sync(struct hci_dev *hdev, void *data) return -ECANCELED; } - memcpy(&cp, cmd->param, sizeof(cp)); + len = cmd->param_len; + memcpy(cp, cmd->param, min(__struct_size(cp), len)); mutex_unlock(&hdev->mgmt_pending_lock); - len = cmd->param_len; - memset(hdev->mesh_ad_types, 0, sizeof(hdev->mesh_ad_types)); - if (cp.enable) + if (cp->enable) hci_dev_set_flag(hdev, HCI_MESH); else hci_dev_clear_flag(hdev, HCI_MESH); - hdev->le_scan_interval = __le16_to_cpu(cp.period); - hdev->le_scan_window = __le16_to_cpu(cp.window); + hdev->le_scan_interval = __le16_to_cpu(cp->period); + hdev->le_scan_window = __le16_to_cpu(cp->window); - len -= sizeof(cp); + len -= sizeof(struct mgmt_cp_set_mesh); /* If filters don't fit, forward all adv pkts */ if (len <= sizeof(hdev->mesh_ad_types)) - memcpy(hdev->mesh_ad_types, cp.ad_types, len); + memcpy(hdev->mesh_ad_types, cp->ad_types, len); hci_update_passive_scan_sync(hdev); return 0; -- cgit v1.2.3 From 76e20da0bd00c556ed0a1e7250bdb6ac3e808ea8 Mon Sep 17 00:00:00 2001 From: Frédéric Danis Date: Mon, 6 Oct 2025 10:35:44 +0200 Subject: Revert "Bluetooth: L2CAP: convert timeouts to secs_to_jiffies()" This reverts commit c9d84da18d1e0d28a7e16ca6df8e6d47570501d4. It replaces in L2CAP calls to msecs_to_jiffies() to secs_to_jiffies() and updates the constants accordingly. But the constants are also used in LCAP Configure Request and L2CAP Configure Response which expect values in milliseconds. This may prevent correct usage of L2CAP channel. To fix it, keep those constants in milliseconds and so revert this change. Fixes: c9d84da18d1e ("Bluetooth: L2CAP: convert timeouts to secs_to_jiffies()") Signed-off-by: Frédéric Danis Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/l2cap.h | 4 ++-- net/bluetooth/l2cap_core.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/l2cap.h b/include/net/bluetooth/l2cap.h index 4bb0eaedda18..00e182a22720 100644 --- a/include/net/bluetooth/l2cap.h +++ b/include/net/bluetooth/l2cap.h @@ -38,8 +38,8 @@ #define L2CAP_DEFAULT_TX_WINDOW 63 #define L2CAP_DEFAULT_EXT_WINDOW 0x3FFF #define L2CAP_DEFAULT_MAX_TX 3 -#define L2CAP_DEFAULT_RETRANS_TO 2 /* seconds */ -#define L2CAP_DEFAULT_MONITOR_TO 12 /* seconds */ +#define L2CAP_DEFAULT_RETRANS_TO 2000 /* 2 seconds */ +#define L2CAP_DEFAULT_MONITOR_TO 12000 /* 12 seconds */ #define L2CAP_DEFAULT_MAX_PDU_SIZE 1492 /* Sized for AMP packet */ #define L2CAP_DEFAULT_ACK_TO 200 #define L2CAP_DEFAULT_MAX_SDU_SIZE 0xFFFF diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 805c752ac0a9..d08320380ad6 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -282,7 +282,7 @@ static void __set_retrans_timer(struct l2cap_chan *chan) if (!delayed_work_pending(&chan->monitor_timer) && chan->retrans_timeout) { l2cap_set_timer(chan, &chan->retrans_timer, - secs_to_jiffies(chan->retrans_timeout)); + msecs_to_jiffies(chan->retrans_timeout)); } } @@ -291,7 +291,7 @@ static void __set_monitor_timer(struct l2cap_chan *chan) __clear_retrans_timer(chan); if (chan->monitor_timeout) { l2cap_set_timer(chan, &chan->monitor_timer, - secs_to_jiffies(chan->monitor_timeout)); + msecs_to_jiffies(chan->monitor_timeout)); } } -- cgit v1.2.3 From 751463ceefc3397566d03c8b64ef4a77f5fd88ac Mon Sep 17 00:00:00 2001 From: Luiz Augusto von Dentz Date: Wed, 22 Oct 2025 16:03:19 -0400 Subject: Bluetooth: hci_core: Fix tracking of periodic advertisement Periodic advertising enabled flag cannot be tracked by the enabled flag since advertising and periodic advertising each can be enabled/disabled separately from one another causing the states to be inconsistent when for example an advertising set is disabled its enabled flag is set to false which is then used for periodic which has not being disabled. Fixes: eca0ae4aea66 ("Bluetooth: Add initial implementation of BIS connections") Signed-off-by: Luiz Augusto von Dentz --- include/net/bluetooth/hci_core.h | 1 + net/bluetooth/hci_event.c | 7 +++++-- net/bluetooth/hci_sync.c | 4 ++-- 3 files changed, 8 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 2924c2bf2a98..b8100dbfe5d7 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -244,6 +244,7 @@ struct adv_info { bool enabled; bool pending; bool periodic; + bool periodic_enabled; __u8 mesh; __u8 instance; __u8 handle; diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 1dabf5a7ae18..d37db364acf7 100644 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1607,7 +1607,7 @@ static u8 hci_cc_le_set_ext_adv_enable(struct hci_dev *hdev, void *data, hci_dev_set_flag(hdev, HCI_LE_ADV); - if (adv && !adv->periodic) + if (adv) adv->enabled = true; else if (!set->handle) hci_dev_set_flag(hdev, HCI_LE_ADV_0); @@ -3963,8 +3963,11 @@ static u8 hci_cc_le_set_per_adv_enable(struct hci_dev *hdev, void *data, hci_dev_set_flag(hdev, HCI_LE_PER_ADV); if (adv) - adv->enabled = true; + adv->periodic_enabled = true; } else { + if (adv) + adv->periodic_enabled = false; + /* If just one instance was disabled check if there are * any other instance enabled before clearing HCI_LE_PER_ADV. * The current periodic adv instance will be marked as diff --git a/net/bluetooth/hci_sync.c b/net/bluetooth/hci_sync.c index 28ad08cd7d70..73fc41b68b68 100644 --- a/net/bluetooth/hci_sync.c +++ b/net/bluetooth/hci_sync.c @@ -1607,7 +1607,7 @@ int hci_disable_per_advertising_sync(struct hci_dev *hdev, u8 instance) /* If periodic advertising already disabled there is nothing to do. */ adv = hci_find_adv_instance(hdev, instance); - if (!adv || !adv->periodic || !adv->enabled) + if (!adv || !adv->periodic_enabled) return 0; memset(&cp, 0, sizeof(cp)); @@ -1672,7 +1672,7 @@ static int hci_enable_per_advertising_sync(struct hci_dev *hdev, u8 instance) /* If periodic advertising already enabled there is nothing to do. */ adv = hci_find_adv_instance(hdev, instance); - if (adv && adv->periodic && adv->enabled) + if (adv && adv->periodic_enabled) return 0; memset(&cp, 0, sizeof(cp)); -- cgit v1.2.3 From d50f21091358b2b29dc06c2061106cdb0f030d03 Mon Sep 17 00:00:00 2001 From: Dimitri John Ledkov Date: Sun, 26 Oct 2025 20:21:00 +0000 Subject: kbuild: align modinfo section for Secureboot Authenticode EDK2 compat Previously linker scripts would always generate vmlinuz that has sections aligned. And thus padded (correct Authenticode calculation) and unpadded calculation would be same. As in https://github.com/rhboot/pesign userspace tool would produce the same authenticode digest for both of the following commands: pesign --padding --hash --in ./arch/x86_64/boot/bzImage pesign --nopadding --hash --in ./arch/x86_64/boot/bzImage The commit 3e86e4d74c04 ("kbuild: keep .modinfo section in vmlinux.unstripped") added .modinfo section of variable length. Depending on kernel configuration it may or may not be aligned. All userspace signing tooling correctly pads such section to calculation spec compliant authenticode digest. However, if bzImage is not further processed and is attempted to be loaded directly by EDK2 firmware, it calculates unpadded Authenticode digest and fails to correct accept/reject such kernel builds even when propoer Authenticode values are enrolled in db/dbx. One can say EDK2 requires aligned/padded kernels in Secureboot. Thus add ALIGN(8) to the .modinfo section, to esure kernels irrespective of modinfo contents can be loaded by all existing EDK2 firmware builds. Fixes: 3e86e4d74c04 ("kbuild: keep .modinfo section in vmlinux.unstripped") Cc: stable@vger.kernel.org Signed-off-by: Dimitri John Ledkov Link: https://patch.msgid.link/20251026202100.679989-1-dimitri.ledkov@surgut.co.uk Signed-off-by: Nathan Chancellor --- include/asm-generic/vmlinux.lds.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index 8a9a2e732a65..e04d56a5332e 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -832,7 +832,7 @@ defined(CONFIG_AUTOFDO_CLANG) || defined(CONFIG_PROPELLER_CLANG) /* Required sections not related to debugging. */ #define ELF_DETAILS \ - .modinfo : { *(.modinfo) } \ + .modinfo : { *(.modinfo) . = ALIGN(8); } \ .comment 0 : { *(.comment) } \ .symtab 0 : { *(.symtab) } \ .strtab 0 : { *(.strtab) } \ -- cgit v1.2.3 From 12a1c9353c47c0fb3464eba2d78cdf649dee1cf7 Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Mon, 27 Oct 2025 09:27:32 +0900 Subject: block: fix op_is_zone_mgmt() to handle REQ_OP_ZONE_RESET_ALL REQ_OP_ZONE_RESET_ALL is a zone management request. Fix op_is_zone_mgmt() to return true for that operation, like it already does for REQ_OP_ZONE_RESET. While no problems were reported without this fix, this change allows strengthening checks in various block device drivers (scsi sd, virtioblk, DM) where op_is_zone_mgmt() is used to verify that a zone management command is not being issued to a regular block device. Fixes: 6c1b1da58f8c ("block: add zone open, close and finish operations") Cc: stable@vger.kernel.org Signed-off-by: Damien Le Moal Reviewed-by: Chaitanya Kulkarni Reviewed-by: Christoph Hellwig Reviewed-by: Johannes Thumshirn Signed-off-by: Jens Axboe --- include/linux/blk_types.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 8e8d1cc8b06c..d8ba743a89b7 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -478,6 +478,7 @@ static inline bool op_is_zone_mgmt(enum req_op op) { switch (op & REQ_OP_MASK) { case REQ_OP_ZONE_RESET: + case REQ_OP_ZONE_RESET_ALL: case REQ_OP_ZONE_OPEN: case REQ_OP_ZONE_CLOSE: case REQ_OP_ZONE_FINISH: -- cgit v1.2.3 From 19de03b312d69a7e9bacb51c806c6e3f4207376c Mon Sep 17 00:00:00 2001 From: Damien Le Moal Date: Mon, 27 Oct 2025 09:27:33 +0900 Subject: block: make REQ_OP_ZONE_OPEN a write operation A REQ_OP_OPEN_ZONE request changes the condition of a sequential zone of a zoned block device to the explicitly open condition (BLK_ZONE_COND_EXP_OPEN). As such, it should be considered a write operation. Change this operation code to be an odd number to reflect this. The following operation numbers are changed to keep the numbering compact. No problems were reported without this change as this operation has no data. However, this unifies the zone operation to reflect that they modify the device state and also allows strengthening checks in the block layer, e.g. checking if this operation is not issued against a read-only device. Fixes: 6c1b1da58f8c ("block: add zone open, close and finish operations") Cc: stable@vger.kernel.org Signed-off-by: Damien Le Moal Reviewed-by: Chaitanya Kulkarni Reviewed-by: Christoph Hellwig Reviewed-by: Johannes Thumshirn Signed-off-by: Jens Axboe --- include/linux/blk_types.h | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'include') diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index d8ba743a89b7..44c30183ecc3 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -341,15 +341,15 @@ enum req_op { /* write the zero filled sector many times */ REQ_OP_WRITE_ZEROES = (__force blk_opf_t)9, /* Open a zone */ - REQ_OP_ZONE_OPEN = (__force blk_opf_t)10, + REQ_OP_ZONE_OPEN = (__force blk_opf_t)11, /* Close a zone */ - REQ_OP_ZONE_CLOSE = (__force blk_opf_t)11, + REQ_OP_ZONE_CLOSE = (__force blk_opf_t)13, /* Transition a zone to full */ - REQ_OP_ZONE_FINISH = (__force blk_opf_t)13, + REQ_OP_ZONE_FINISH = (__force blk_opf_t)15, /* reset a zone write pointer */ - REQ_OP_ZONE_RESET = (__force blk_opf_t)15, + REQ_OP_ZONE_RESET = (__force blk_opf_t)17, /* reset all the zone present on the device */ - REQ_OP_ZONE_RESET_ALL = (__force blk_opf_t)17, + REQ_OP_ZONE_RESET_ALL = (__force blk_opf_t)19, /* Driver private requests */ REQ_OP_DRV_IN = (__force blk_opf_t)34, -- cgit v1.2.3 From 48cbf50531d8eca15b8a811717afdebb8677de9b Mon Sep 17 00:00:00 2001 From: Shawn Guo Date: Fri, 24 Oct 2025 16:23:44 +0800 Subject: regmap: irq: Correct documentation of wake_invert flag Per commit 9442490a0286 ("regmap: irq: Support wake IRQ mask inversion") the wake_invert flag is to support enable register, so cleared bits are wake disabled. Fixes: 68622bdfefb9 ("regmap: irq: document mask/wake_invert flags") Cc: stable@vger.kernel.org Signed-off-by: Shawn Guo Link: https://patch.msgid.link/20251024082344.2188895-1-shawnguo2@yeah.net Signed-off-by: Mark Brown --- include/linux/regmap.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/regmap.h b/include/linux/regmap.h index 4e1ac1fbcec4..55343795644b 100644 --- a/include/linux/regmap.h +++ b/include/linux/regmap.h @@ -1643,7 +1643,7 @@ struct regmap_irq_chip_data; * @status_invert: Inverted status register: cleared bits are active interrupts. * @status_is_level: Status register is actuall signal level: Xor status * register with previous value to get active interrupts. - * @wake_invert: Inverted wake register: cleared bits are wake enabled. + * @wake_invert: Inverted wake register: cleared bits are wake disabled. * @type_in_mask: Use the mask registers for controlling irq type. Use this if * the hardware provides separate bits for rising/falling edge * or low/high level interrupts and they should be combined into -- cgit v1.2.3 From 18cd0a9c7aaf880502e4aff3ea30022f97d6c103 Mon Sep 17 00:00:00 2001 From: PIYUSH CHOUDHARY Date: Mon, 20 Oct 2025 00:05:08 +0530 Subject: video: fb: Fix typo in comment in fb.h Fix typo: "verical" -> "vertical" in macro description Signed-off-by: PIYUSH CHOUDHARY Signed-off-by: Helge Deller Cc: stable@vger.kernel.org --- include/uapi/linux/fb.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/uapi/linux/fb.h b/include/uapi/linux/fb.h index cde8f173f566..22acaaec7b1c 100644 --- a/include/uapi/linux/fb.h +++ b/include/uapi/linux/fb.h @@ -319,7 +319,7 @@ enum { #define FB_VBLANK_HAVE_VCOUNT 0x020 /* the vcount field is valid */ #define FB_VBLANK_HAVE_HCOUNT 0x040 /* the hcount field is valid */ #define FB_VBLANK_VSYNCING 0x080 /* currently in a vsync */ -#define FB_VBLANK_HAVE_VSYNC 0x100 /* verical syncs can be detected */ +#define FB_VBLANK_HAVE_VSYNC 0x100 /* vertical syncs can be detected */ struct fb_vblank { __u32 flags; /* FB_VBLANK flags */ -- cgit v1.2.3 From a1f3058930745d2b938b6b4f5bd9630dc74b26b7 Mon Sep 17 00:00:00 2001 From: Quanmin Yan Date: Fri, 10 Oct 2025 16:16:59 +0800 Subject: fbcon: Set fb_display[i]->mode to NULL when the mode is released Recently, we discovered the following issue through syzkaller: BUG: KASAN: slab-use-after-free in fb_mode_is_equal+0x285/0x2f0 Read of size 4 at addr ff11000001b3c69c by task syz.xxx ... Call Trace: dump_stack_lvl+0xab/0xe0 print_address_description.constprop.0+0x2c/0x390 print_report+0xb9/0x280 kasan_report+0xb8/0xf0 fb_mode_is_equal+0x285/0x2f0 fbcon_mode_deleted+0x129/0x180 fb_set_var+0xe7f/0x11d0 do_fb_ioctl+0x6a0/0x750 fb_ioctl+0xe0/0x140 __x64_sys_ioctl+0x193/0x210 do_syscall_64+0x5f/0x9c0 entry_SYSCALL_64_after_hwframe+0x76/0x7e Based on experimentation and analysis, during framebuffer unregistration, only the memory of fb_info->modelist is freed, without setting the corresponding fb_display[i]->mode to NULL for the freed modes. This leads to UAF issues during subsequent accesses. Here's an example of reproduction steps: 1. With /dev/fb0 already registered in the system, load a kernel module to register a new device /dev/fb1; 2. Set fb1's mode to the global fb_display[] array (via FBIOPUT_CON2FBMAP); 3. Switch console from fb to VGA (to allow normal rmmod of the ko); 4. Unload the kernel module, at this point fb1's modelist is freed, leaving a wild pointer in fb_display[]; 5. Trigger the bug via system calls through fb0 attempting to delete a mode from fb0. Add a check in do_unregister_framebuffer(): if the mode to be freed exists in fb_display[], set the corresponding mode pointer to NULL. Signed-off-by: Quanmin Yan Reviewed-by: Thomas Zimmermann Signed-off-by: Helge Deller Cc: stable@vger.kernel.org --- drivers/video/fbdev/core/fbcon.c | 19 +++++++++++++++++++ drivers/video/fbdev/core/fbmem.c | 1 + include/linux/fbcon.h | 2 ++ 3 files changed, 22 insertions(+) (limited to 'include') diff --git a/drivers/video/fbdev/core/fbcon.c b/drivers/video/fbdev/core/fbcon.c index 96cc9b389246..9bd3c3814b5c 100644 --- a/drivers/video/fbdev/core/fbcon.c +++ b/drivers/video/fbdev/core/fbcon.c @@ -2810,6 +2810,25 @@ int fbcon_mode_deleted(struct fb_info *info, return found; } +static void fbcon_delete_mode(struct fb_videomode *m) +{ + struct fbcon_display *p; + + for (int i = first_fb_vc; i <= last_fb_vc; i++) { + p = &fb_display[i]; + if (p->mode == m) + p->mode = NULL; + } +} + +void fbcon_delete_modelist(struct list_head *head) +{ + struct fb_modelist *modelist; + + list_for_each_entry(modelist, head, list) + fbcon_delete_mode(&modelist->mode); +} + #ifdef CONFIG_VT_HW_CONSOLE_BINDING static void fbcon_unbind(void) { diff --git a/drivers/video/fbdev/core/fbmem.c b/drivers/video/fbdev/core/fbmem.c index 53f1719b1ae1..eff757ebbed1 100644 --- a/drivers/video/fbdev/core/fbmem.c +++ b/drivers/video/fbdev/core/fbmem.c @@ -544,6 +544,7 @@ static void do_unregister_framebuffer(struct fb_info *fb_info) fb_info->pixmap.addr = NULL; } + fbcon_delete_modelist(&fb_info->modelist); fb_destroy_modelist(&fb_info->modelist); registered_fb[fb_info->node] = NULL; num_registered_fb--; diff --git a/include/linux/fbcon.h b/include/linux/fbcon.h index 81f0e698acbf..f206370060e1 100644 --- a/include/linux/fbcon.h +++ b/include/linux/fbcon.h @@ -18,6 +18,7 @@ void fbcon_suspended(struct fb_info *info); void fbcon_resumed(struct fb_info *info); int fbcon_mode_deleted(struct fb_info *info, struct fb_videomode *mode); +void fbcon_delete_modelist(struct list_head *head); void fbcon_new_modelist(struct fb_info *info); void fbcon_get_requirement(struct fb_info *info, struct fb_blit_caps *caps); @@ -38,6 +39,7 @@ static inline void fbcon_suspended(struct fb_info *info) {} static inline void fbcon_resumed(struct fb_info *info) {} static inline int fbcon_mode_deleted(struct fb_info *info, struct fb_videomode *mode) { return 0; } +static inline void fbcon_delete_modelist(struct list_head *head) {} static inline void fbcon_new_modelist(struct fb_info *info) {} static inline void fbcon_get_requirement(struct fb_info *info, struct fb_blit_caps *caps) {} -- cgit v1.2.3 From 24990d89c23de4dbef6b0b3d58383cafefdd6983 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Oct 2025 12:58:00 +0100 Subject: trace: tcp: add three metrics to trace_tcp_rcvbuf_grow() While chasing yet another receive autotuning bug, I found useful to add rcv_ssthresh, window_clamp and rcv_wnd. tcp_stream 40597 [068] 2172.978198: tcp:tcp_rcvbuf_grow: time=50307 rtt_us=50179 copied=77824 inq=0 space=40960 ooo=0 scaling_ratio=219 rcvbuf=131072 rcv_ssthresh=107474 window_clamp=112128 rcv_wnd=110592 tcp_stream 40597 [068] 2173.028528: tcp:tcp_rcvbuf_grow: time=50336 rtt_us=50206 copied=110592 inq=0 space=77824 ooo=0 scaling_ratio=219 rcvbuf=509444 rcv_ssthresh=328658 window_clamp=435813 rcv_wnd=331776 tcp_stream 40597 [068] 2173.078830: tcp:tcp_rcvbuf_grow: time=50305 rtt_us=50070 copied=270336 inq=0 space=110592 ooo=0 scaling_ratio=219 rcvbuf=509444 rcv_ssthresh=431159 window_clamp=435813 rcv_wnd=434176 tcp_stream 40597 [068] 2173.129137: tcp:tcp_rcvbuf_grow: time=50313 rtt_us=50118 copied=434176 inq=0 space=270336 ooo=0 scaling_ratio=219 rcvbuf=2457847 rcv_ssthresh=1299511 window_clamp=2102611 rcv_wnd=1302528 tcp_stream 40597 [068] 2173.179451: tcp:tcp_rcvbuf_grow: time=50318 rtt_us=50041 copied=1019904 inq=0 space=434176 ooo=0 scaling_ratio=219 rcvbuf=2457847 rcv_ssthresh=2087445 window_clamp=2102611 rcv_wnd=2088960 Signed-off-by: Eric Dumazet Signed-off-by: Matthieu Baerts (NGI0) Reviewed-by: Neal Cardwell Link: https://patch.msgid.link/20251028-net-tcp-recv-autotune-v3-2-74b43ba4c84c@kernel.org Signed-off-by: Jakub Kicinski --- include/trace/events/tcp.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include') diff --git a/include/trace/events/tcp.h b/include/trace/events/tcp.h index 9d2c36c6a0ed..6757233bd064 100644 --- a/include/trace/events/tcp.h +++ b/include/trace/events/tcp.h @@ -218,6 +218,9 @@ TRACE_EVENT(tcp_rcvbuf_grow, __field(__u32, space) __field(__u32, ooo_space) __field(__u32, rcvbuf) + __field(__u32, rcv_ssthresh) + __field(__u32, window_clamp) + __field(__u32, rcv_wnd) __field(__u8, scaling_ratio) __field(__u16, sport) __field(__u16, dport) @@ -245,6 +248,9 @@ TRACE_EVENT(tcp_rcvbuf_grow, tp->rcv_nxt; __entry->rcvbuf = sk->sk_rcvbuf; + __entry->rcv_ssthresh = tp->rcv_ssthresh; + __entry->window_clamp = tp->window_clamp; + __entry->rcv_wnd = tp->rcv_wnd; __entry->scaling_ratio = tp->scaling_ratio; __entry->sport = ntohs(inet->inet_sport); __entry->dport = ntohs(inet->inet_dport); @@ -264,11 +270,14 @@ TRACE_EVENT(tcp_rcvbuf_grow, ), TP_printk("time=%u rtt_us=%u copied=%u inq=%u space=%u ooo=%u scaling_ratio=%u rcvbuf=%u " + "rcv_ssthresh=%u window_clamp=%u rcv_wnd=%u " "family=%s sport=%hu dport=%hu saddr=%pI4 daddr=%pI4 " "saddrv6=%pI6c daddrv6=%pI6c skaddr=%p sock_cookie=%llx", __entry->time, __entry->rtt_us, __entry->copied, __entry->inq, __entry->space, __entry->ooo_space, __entry->scaling_ratio, __entry->rcvbuf, + __entry->rcv_ssthresh, __entry->window_clamp, + __entry->rcv_wnd, show_family_name(__entry->family), __entry->sport, __entry->dport, __entry->saddr, __entry->daddr, -- cgit v1.2.3 From b1e014a1f3275a6f3d0f2b30b8117447fc3915f5 Mon Sep 17 00:00:00 2001 From: Eric Dumazet Date: Tue, 28 Oct 2025 12:58:01 +0100 Subject: tcp: add newval parameter to tcp_rcvbuf_grow() This patch has no functional change, and prepares the following one. tcp_rcvbuf_grow() will need to have access to tp->rcvq_space.space old and new values. Change mptcp_rcvbuf_grow() in a similar way. Signed-off-by: Eric Dumazet [ Moved 'oldval' declaration to the next patch to avoid warnings at build time. ] Signed-off-by: Matthieu Baerts (NGI0) Reviewed-by: Neal Cardwell Link: https://patch.msgid.link/20251028-net-tcp-recv-autotune-v3-3-74b43ba4c84c@kernel.org Signed-off-by: Jakub Kicinski --- include/net/tcp.h | 2 +- net/ipv4/tcp_input.c | 14 +++++++------- net/mptcp/protocol.c | 20 ++++++++------------ 3 files changed, 16 insertions(+), 20 deletions(-) (limited to 'include') diff --git a/include/net/tcp.h b/include/net/tcp.h index 5ca230ed526a..ab20f549b8f9 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -370,7 +370,7 @@ void tcp_delack_timer_handler(struct sock *sk); int tcp_ioctl(struct sock *sk, int cmd, int *karg); enum skb_drop_reason tcp_rcv_state_process(struct sock *sk, struct sk_buff *skb); void tcp_rcv_established(struct sock *sk, struct sk_buff *skb); -void tcp_rcvbuf_grow(struct sock *sk); +void tcp_rcvbuf_grow(struct sock *sk, u32 newval); void tcp_rcv_space_adjust(struct sock *sk); int tcp_twsk_unique(struct sock *sk, struct sock *sktw, void *twp); void tcp_twsk_destructor(struct sock *sk); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 31ea5af49f2d..cb4e07f84ae2 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -891,18 +891,20 @@ static inline void tcp_rcv_rtt_measure_ts(struct sock *sk, } } -void tcp_rcvbuf_grow(struct sock *sk) +void tcp_rcvbuf_grow(struct sock *sk, u32 newval) { const struct net *net = sock_net(sk); struct tcp_sock *tp = tcp_sk(sk); - int rcvwin, rcvbuf, cap; + u32 rcvwin, rcvbuf, cap; + + tp->rcvq_space.space = newval; if (!READ_ONCE(net->ipv4.sysctl_tcp_moderate_rcvbuf) || (sk->sk_userlocks & SOCK_RCVBUF_LOCK)) return; /* slow start: allow the sender to double its rate. */ - rcvwin = tp->rcvq_space.space << 1; + rcvwin = newval << 1; if (!RB_EMPTY_ROOT(&tp->out_of_order_queue)) rcvwin += TCP_SKB_CB(tp->ooo_last_skb)->end_seq - tp->rcv_nxt; @@ -943,9 +945,7 @@ void tcp_rcv_space_adjust(struct sock *sk) trace_tcp_rcvbuf_grow(sk, time); - tp->rcvq_space.space = copied; - - tcp_rcvbuf_grow(sk); + tcp_rcvbuf_grow(sk, copied); new_measure: tp->rcvq_space.seq = tp->copied_seq; @@ -5270,7 +5270,7 @@ end: } /* do not grow rcvbuf for not-yet-accepted or orphaned sockets. */ if (sk->sk_socket) - tcp_rcvbuf_grow(sk); + tcp_rcvbuf_grow(sk, tp->rcvq_space.space); } static int __must_check tcp_queue_rcv(struct sock *sk, struct sk_buff *skb, diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index a8a3bdf95543..052a0c62023f 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -194,17 +194,18 @@ static bool mptcp_ooo_try_coalesce(struct mptcp_sock *msk, struct sk_buff *to, * - mptcp does not maintain a msk-level window clamp * - returns true when the receive buffer is actually updated */ -static bool mptcp_rcvbuf_grow(struct sock *sk) +static bool mptcp_rcvbuf_grow(struct sock *sk, u32 newval) { struct mptcp_sock *msk = mptcp_sk(sk); const struct net *net = sock_net(sk); - int rcvwin, rcvbuf, cap; + u32 rcvwin, rcvbuf, cap; + msk->rcvq_space.space = newval; if (!READ_ONCE(net->ipv4.sysctl_tcp_moderate_rcvbuf) || (sk->sk_userlocks & SOCK_RCVBUF_LOCK)) return false; - rcvwin = msk->rcvq_space.space << 1; + rcvwin = newval << 1; if (!RB_EMPTY_ROOT(&msk->out_of_order_queue)) rcvwin += MPTCP_SKB_CB(msk->ooo_last_skb)->end_seq - msk->ack_seq; @@ -334,7 +335,7 @@ end: skb_set_owner_r(skb, sk); /* do not grow rcvbuf for not-yet-accepted or orphaned sockets. */ if (sk->sk_socket) - mptcp_rcvbuf_grow(sk); + mptcp_rcvbuf_grow(sk, msk->rcvq_space.space); } static void mptcp_init_skb(struct sock *ssk, struct sk_buff *skb, int offset, @@ -2049,10 +2050,7 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied) if (msk->rcvq_space.copied <= msk->rcvq_space.space) goto new_measure; - msk->rcvq_space.space = msk->rcvq_space.copied; - if (mptcp_rcvbuf_grow(sk)) { - int copied = msk->rcvq_space.copied; - + if (mptcp_rcvbuf_grow(sk, msk->rcvq_space.copied)) { /* Make subflows follow along. If we do not do this, we * get drops at subflow level if skbs can't be moved to * the mptcp rx queue fast enough (announced rcv_win can @@ -2065,10 +2063,8 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied) ssk = mptcp_subflow_tcp_sock(subflow); slow = lock_sock_fast(ssk); /* subflows can be added before tcp_init_transfer() */ - if (tcp_sk(ssk)->rcvq_space.space) { - tcp_sk(ssk)->rcvq_space.space = copied; - tcp_rcvbuf_grow(ssk); - } + if (tcp_sk(ssk)->rcvq_space.space) + tcp_rcvbuf_grow(ssk, msk->rcvq_space.copied); unlock_sock_fast(ssk, slow); } } -- cgit v1.2.3 From 34892cfec0c2d96787c4be7bda0d5f18d7dacf85 Mon Sep 17 00:00:00 2001 From: Shahar Shitrit Date: Sun, 26 Oct 2025 22:03:01 +0200 Subject: net: tls: Change async resync helpers argument Update tls_offload_rx_resync_async_request_start() and tls_offload_rx_resync_async_request_end() to get a struct tls_offload_resync_async parameter directly, rather than extracting it from struct sock. This change aligns the function signatures with the upcoming tls_offload_rx_resync_async_request_cancel() helper, which will be introduced in a subsequent patch. Signed-off-by: Shahar Shitrit Reviewed-by: Sabrina Dubroca Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761508983-937977-2-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- .../ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c | 9 +++++++-- include/net/tls.h | 21 ++++++++------------- 2 files changed, 15 insertions(+), 15 deletions(-) (limited to 'include') diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c index d7a11ff9bbdb..5fbc92269585 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_rx.c @@ -425,12 +425,14 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi, { struct mlx5e_ktls_rx_resync_buf *buf = wi->tls_get_params.buf; struct mlx5e_ktls_offload_context_rx *priv_rx; + struct tls_offload_context_rx *rx_ctx; u8 tracker_state, auth_state, *ctx; struct device *dev; u32 hw_seq; priv_rx = buf->priv_rx; dev = mlx5_core_dma_dev(sq->channel->mdev); + rx_ctx = tls_offload_ctx_rx(tls_get_ctx(priv_rx->sk)); if (unlikely(test_bit(MLX5E_PRIV_RX_FLAG_DELETING, priv_rx->flags))) goto out; @@ -447,7 +449,8 @@ void mlx5e_ktls_handle_get_psv_completion(struct mlx5e_icosq_wqe_info *wi, } hw_seq = MLX5_GET(tls_progress_params, ctx, hw_resync_tcp_sn); - tls_offload_rx_resync_async_request_end(priv_rx->sk, cpu_to_be32(hw_seq)); + tls_offload_rx_resync_async_request_end(rx_ctx->resync_async, + cpu_to_be32(hw_seq)); priv_rx->rq_stats->tls_resync_req_end++; out: mlx5e_ktls_priv_rx_put(priv_rx); @@ -482,6 +485,7 @@ static bool resync_queue_get_psv(struct sock *sk) static void resync_update_sn(struct mlx5e_rq *rq, struct sk_buff *skb) { struct ethhdr *eth = (struct ethhdr *)(skb->data); + struct tls_offload_resync_async *resync_async; struct net_device *netdev = rq->netdev; struct net *net = dev_net(netdev); struct sock *sk = NULL; @@ -527,7 +531,8 @@ static void resync_update_sn(struct mlx5e_rq *rq, struct sk_buff *skb) seq = th->seq; datalen = skb->len - depth; - tls_offload_rx_resync_async_request_start(sk, seq, datalen); + resync_async = tls_offload_ctx_rx(tls_get_ctx(sk))->resync_async; + tls_offload_rx_resync_async_request_start(resync_async, seq, datalen); rq->stats->tls_resync_req_start++; unref: diff --git a/include/net/tls.h b/include/net/tls.h index 857340338b69..b90f3b675c3c 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -451,25 +451,20 @@ static inline void tls_offload_rx_resync_request(struct sock *sk, __be32 seq) /* Log all TLS record header TCP sequences in [seq, seq+len] */ static inline void -tls_offload_rx_resync_async_request_start(struct sock *sk, __be32 seq, u16 len) +tls_offload_rx_resync_async_request_start(struct tls_offload_resync_async *resync_async, + __be32 seq, u16 len) { - struct tls_context *tls_ctx = tls_get_ctx(sk); - struct tls_offload_context_rx *rx_ctx = tls_offload_ctx_rx(tls_ctx); - - atomic64_set(&rx_ctx->resync_async->req, ((u64)ntohl(seq) << 32) | + atomic64_set(&resync_async->req, ((u64)ntohl(seq) << 32) | ((u64)len << 16) | RESYNC_REQ | RESYNC_REQ_ASYNC); - rx_ctx->resync_async->loglen = 0; - rx_ctx->resync_async->rcd_delta = 0; + resync_async->loglen = 0; + resync_async->rcd_delta = 0; } static inline void -tls_offload_rx_resync_async_request_end(struct sock *sk, __be32 seq) +tls_offload_rx_resync_async_request_end(struct tls_offload_resync_async *resync_async, + __be32 seq) { - struct tls_context *tls_ctx = tls_get_ctx(sk); - struct tls_offload_context_rx *rx_ctx = tls_offload_ctx_rx(tls_ctx); - - atomic64_set(&rx_ctx->resync_async->req, - ((u64)ntohl(seq) << 32) | RESYNC_REQ); + atomic64_set(&resync_async->req, ((u64)ntohl(seq) << 32) | RESYNC_REQ); } static inline void -- cgit v1.2.3 From c15d5c62ab313c19121f10e25d4fec852bd1c40c Mon Sep 17 00:00:00 2001 From: Shahar Shitrit Date: Sun, 26 Oct 2025 22:03:02 +0200 Subject: net: tls: Cancel RX async resync request on rcd_delta overflow When a netdev issues a RX async resync request for a TLS connection, the TLS module handles it by logging record headers and attempting to match them to the tcp_sn provided by the device. If a match is found, the TLS module approves the tcp_sn for resynchronization. While waiting for a device response, the TLS module also increments rcd_delta each time a new TLS record is received, tracking the distance from the original resync request. However, if the device response is delayed or fails (e.g due to unstable connection and device getting out of tracking, hardware errors, resource exhaustion etc.), the TLS module keeps logging and incrementing, which can lead to a WARN() when rcd_delta exceeds the threshold. To address this, introduce tls_offload_rx_resync_async_request_cancel() to explicitly cancel resync requests when a device response failure is detected. Call this helper also as a final safeguard when rcd_delta crosses its threshold, as reaching this point implies that earlier cancellation did not occur. Signed-off-by: Shahar Shitrit Reviewed-by: Sabrina Dubroca Signed-off-by: Tariq Toukan Link: https://patch.msgid.link/1761508983-937977-3-git-send-email-tariqt@nvidia.com Signed-off-by: Jakub Kicinski --- include/net/tls.h | 6 ++++++ net/tls/tls_device.c | 4 +++- 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'include') diff --git a/include/net/tls.h b/include/net/tls.h index b90f3b675c3c..c7bcdb3afad7 100644 --- a/include/net/tls.h +++ b/include/net/tls.h @@ -467,6 +467,12 @@ tls_offload_rx_resync_async_request_end(struct tls_offload_resync_async *resync_ atomic64_set(&resync_async->req, ((u64)ntohl(seq) << 32) | RESYNC_REQ); } +static inline void +tls_offload_rx_resync_async_request_cancel(struct tls_offload_resync_async *resync_async) +{ + atomic64_set(&resync_async->req, 0); +} + static inline void tls_offload_rx_resync_set_type(struct sock *sk, enum tls_offload_sync_type type) { diff --git a/net/tls/tls_device.c b/net/tls/tls_device.c index a64ae15b1a60..71734411ff4c 100644 --- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -723,8 +723,10 @@ tls_device_rx_resync_async(struct tls_offload_resync_async *resync_async, /* shouldn't get to wraparound: * too long in async stage, something bad happened */ - if (WARN_ON_ONCE(resync_async->rcd_delta == USHRT_MAX)) + if (WARN_ON_ONCE(resync_async->rcd_delta == USHRT_MAX)) { + tls_offload_rx_resync_async_request_cancel(resync_async); return false; + } /* asynchronous stage: log all headers seq such that * req_seq <= seq <= end_seq, and wait for real resync request -- cgit v1.2.3 From 14a7f2392f42bbb71c1a5ea68930006221fcd80a Mon Sep 17 00:00:00 2001 From: Yonghong Song Date: Wed, 29 Oct 2025 11:36:46 -0700 Subject: bpf: Make migrate_disable always inline to avoid partial inlining The build fails with llvm 21/22: $ make LLVM=1 -j ... LD vmlinux.o GEN .vmlinux.objs ... BTF .tmp_vmlinux1.btf.o ... AS .tmp_vmlinux2.kallsyms.o LD vmlinux.unstripped BTFIDS vmlinux.unstripped WARN: resolve_btfids: unresolved symbol migrate_enable WARN: resolve_btfids: unresolved symbol migrate_disable make[2]: *** [vmlinux.unstripped] Error 255 make[2]: *** Deleting file 'vmlinux.unstripped' make[1]: *** [Makefile:1242: vmlinux] Error 2 make: *** [Makefile:248: __sub-make] Error 2 Two functions with identical names but different addresses are considered ambiguous and removed by "pahole" from vmlinux BTF. Later resolve_btfids warns since it cannot find them. Commit 378b7708194f ("sched: Make migrate_{en,dis}able() inline") made them inlineable in most places, but in vmlinux built with llvm 21 and 22 there are four symbols for migrate_{enable,disable}: three static functions and one global function. Fix the issue by marking migrate_{enable,disable} as always inline. The alternative is to mark them as notrace/nokprobe which is more drastic. Only bpf programs are prevented from attaching to these functions. The rest of the tracing shouldn't be affected. [note: Peter ok-ed the patch, Alexei rewrote commit log] Fixes: 378b7708194f ("sched: Make migrate_{en,dis}able() inline") Signed-off-by: Yonghong Song Acked-by: Menglong Dong Link: https://lore.kernel.org/r/20251029183646.3811774-1-yonghong.song@linux.dev Signed-off-by: Alexei Starovoitov --- include/linux/sched.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index cbb7340c5866..b469878de25c 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2407,12 +2407,12 @@ static inline void __migrate_enable(void) { } * be defined in kernel/sched/core.c. */ #ifndef INSTANTIATE_EXPORTED_MIGRATE_DISABLE -static inline void migrate_disable(void) +static __always_inline void migrate_disable(void) { __migrate_disable(); } -static inline void migrate_enable(void) +static __always_inline void migrate_enable(void) { __migrate_enable(); } -- cgit v1.2.3