In the Linux kernel, the following vulnerability has been resolved: eventfs: Initialize ei->children and ei->list in init_ei() eventfs_create_dir() allocates the eventfs_inode and initializes it with init_ei(). But this does not initialize the eventfs_inode list_heads. If the eventfs_create_dir() fails due to memory pressure, it will call free_ei() before it initialized the lists, and that checks to make sure the eventfs_inode has no children. But because the list wasn't initialized, it will give a false warning. Fix it by moving the list initialization into init_ei(). [ Rewrote change log ]
In the Linux kernel, the following vulnerability has been resolved: fsnotify: Fix stale object mask after concurrent mark updates When a mark gets a new event bit, fanotify and inotify may avoid recalculating the object mask if the cached aggregate already contains that bit. This is racy with a recalculation triggered by a concurrent update to another mark on the same connector. The concurrent scan can read the mark before the new bit is added, while the updater reads the old aggregate before that scan publishes its result. The updater then skips recalculation and the scan publishes a mask without the bit, leaving the object mask stale after both updates complete. This can be reproduced with two fanotify groups watching the same inode: one thread removes FAN_MODIFY from one existing mark while another thread adds FAN_MODIFY to the other mark. After both fanotify_mark() calls return, writes can fail to produce FAN_MODIFY for the group whose mark now contains the bit. This was reproduced on an unmodified v6.12.95 kernel. The equivalent inotify interleaving loses IN_MODIFY events. For normal fanotify additions, recalculate whenever the raw mark mask changes. The normal mask is not cleared asynchronously, so an unchanged addition cannot introduce missing interest. Always recalculate ignore-mask updates because FS_MODIFY handling may clear the ignore mask without taking mark->lock, making snapshot comparisons unreliable. Always recalculate after updating an existing inotify watch. Its replace path temporarily sets mark->mask to zero, so a concurrent scan can observe zero even when the old and final masks are equal. Assigning the replacement mask directly would avoid the transient zero, but existing-watch updates are infrequent, so unconditional recalculation is simpler.
In the Linux kernel, the following vulnerability has been resolved: accel/rocket: fix NULL dereference and integer overflow in rocket_job_push() rocket_job_push() allocates a temporary array to hold all input and output GEM object pointers: bos = kvmalloc_array(job->in_bo_count + job->out_bo_count, sizeof(void *), GFP_KERNEL); memcpy(bos, job->in_bos, job->in_bo_count * sizeof(void *)); memcpy(&bos[job->in_bo_count], job->out_bos, ...); Two bugs exist: 1. Missing NULL check: if kvmalloc_array() fails, bos is NULL and the subsequent memcpy() dereferences it, causing a kernel NULL pointer dereference. 2. Integer overflow: in_bo_count and out_bo_count are both u32, set directly from userspace-supplied in_bo_handle_count and out_bo_handle_count with no prior validation. Their sum is computed in u32 arithmetic and can wrap to a smaller value, causing the allocation count passed to kvmalloc_array() to be smaller than intended. Subsequent uses still operate on the original counts when copying and locking objects, which may lead to out-of-bounds accesses on the temporary array. Fix by using check_add_overflow() to detect count overflow before the allocation, and adding a NULL check on the allocation result.
In the Linux kernel, the following vulnerability has been resolved: accel/rocket: initialize job domain before cleanup paths rocket_ioctl_submit_job() releases rjob through rocket_job_put() on allocation error paths. rocket_job_cleanup() unconditionally calls rocket_iommu_domain_put(job->domain), but job->domain is assigned only after task copying and BO lookups. A failure before that assignment can therefore clean up a job with a NULL domain pointer. Take the per-file domain reference before the first error path can release rjob. Also clear rjob->tasks after freeing it in rocket_copy_tasks(), so the common cleanup path cannot free the task array again after a task-copy error.
In the Linux kernel, the following vulnerability has been resolved: accel/rocket: Fix error path handling in rocket_job_run() In rocket_job_run(), after taking an extra fence reference for job->done_fence via dma_fence_get(), the error paths have three bugs: - The dma_fence reference held by job->done_fence is never released, causing a reference leak. - pm_runtime_get_sync() increments the usage counter even on failure, but the error path does not decrement it, leaking the runtime PM reference and preventing the NPU from suspending. - A valid but unsignaled fence is returned to the DRM scheduler, which triggers WARN("Fence ... released with pending signals!") when the scheduler drops its reference. Fix by replacing pm_runtime_get_sync() with pm_runtime_resume_and_get() which auto-balances the usage counter on failure, releasing both fence references on error, and returning ERR_PTR(ret) instead of the unsignaled fence. [tomeu: Refactored error paths to use consolidated goto labels]
In the Linux kernel, the following vulnerability has been resolved: acpi/apei/ghes: Use raw_spinlock_t for CXL CPER work locks The CXL CPER work registration and unregistration helpers acquire cxl_cper_work_lock and cxl_cper_prot_err_work_lock with a spinlock guard(), which leaves local interrupts enabled. The corresponding post paths (cxl_cper_post_event(), cxl_cper_post_prot_err()) execute in hard IRQ context (they are called from the GHES error notification path) and acquire the same locks with an irqsave guard(). If a CPU is holding one of these locks via a spinlock guard() when a GHES interrupt arrives on the same CPU, the IRQ handler spins on the held lock waiting for it to release, while the lock holder is preempted by the IRQ. The result is a deadlock. Convert both locks from spinlock_t to raw_spinlock_t and use guard() at all call sites. On PREEMPT_RT kernels spinlock_t is backed by rt_mutex and sleeping from hard IRQ context is not permitted; raw_spinlock_t is safe in both contexts. Add WARN_ONCE to both register functions to surface double-registration bugs at runtime. Restructure both unregister functions to clear the global work pointer under the lock before calling cancel_work_sync(), closing the window where a CPER interrupt could schedule work on a pointer about to be freed. Add kfifo_reset() after cancel_work_sync() so stale entries are not replayed on next module load. Both kfifos are single-consumer: only one work_struct is registered at a time, enforced by the WARN_ONCE guard in the register functions. kfifo_reset() is safe outside the lock because cancel_work_sync() has already quiesced the consumer, and no new consumer can register until the current module exit completes and a fresh module init runs. Remove the redundant cancel_work_sync() call from cxl_ras_exit() and cxl_pci_driver_exit(). The CPER unregister functions now quiesce the work internally.
In the Linux kernel, the following vulnerability has been resolved: dm-io: clone the source bio instead of copying its biovec For DM_IO_BIO requests, do_region() built each destination bio by walking the source bio's biovec and re-adding the pages one at a time, tracking the remaining transfer in sectors. The vector lengths are byte granular and need not be sector aligned (e.g. a misaligned O_DIRECT buffer split across pages), so the sector-based accounting could lose a sub-sector fragment: to_sector() truncated the remainder and the outer loop spun forever submitting empty bios, hanging the I/O. There is no need to rebuild the biovec at all. The destination reads into (or writes from) exactly the same pages as the source bio, so the bio can simply clone the source's biovec with bio_alloc_clone() and remap it to the target device. The clone inherits the source's iterator and alignment, and the block layer splits it to the target's limits on submission, so the whole region maps to a single cloned bio with no manual page copying or sector accounting. This removes the per-page copy path (and its open-coded bvec dpages helpers) for bio-backed I/O and fixes the hang on misaligned direct I/O to a dm-mirror device. Page-list, vma and kmem sources keep the existing copy path.
In the Linux kernel, the following vulnerability has been resolved: dm-io: report non-retryable errors separatedly The error codes BLK_STS_NOTSUPP and BLK_STS_INVAL should not cause leg failure on dm-raid1. This patch changes the interface to dm-io, so that it reports two error bitmaps - error_bits and unsup_bits. The unsup_bit bitmap tracks BLK_STS_NOTSUPP or BLK_STS_INVAL errors, the error_bits bitmap tracks all the other errors. dm-raid1 is changed so that it won't fail a leg if it receives an error in the unsup_bits bitmap. This patch (with 62dc37a819a5) fixes misbehavior if the user uses unaligned bio vectors on dm-raid1.
In the Linux kernel, the following vulnerability has been resolved: dm-era: fix shadowed superblock leak on take-snap failure metadata_take_snap() bumps the live superblock refcount and then dm_tm_shadow_block() allocates a new block for the metadata snapshot. If the subsequent dm_sm_inc_block() of writeset_tree_root or era_array_root fails, the function only unlocks the clone and returns. The newly allocated shadow block is never returned to the metadata space map, so each failed take-snap permanently leaks one metadata block. Free the clone with dm_sm_dec_block() on those error paths, matching the final step of metadata_drop_snap().
In the Linux kernel, the following vulnerability has been resolved: dm raid1: reserve space for NUL-terminator in build_constructor_string() Reserve space for the termination NUL after the maximum 20 decimal digits of a long long value to avoid buffer overflow in sprintf().
In the Linux kernel, the following vulnerability has been resolved: cpufreq: apple-soc: Fix OPP table cleanup apple_soc_cpufreq_init() adds OPP tables from firmware, but some failure paths do not remove them. The driver also uses dev_pm_opp_remove_all_dynamic(), which is not the right cleanup helper for OPP tables loaded from firmware. Use the cpumask OPP helper after the policy CPU mask has been populated. Pair it with the matching cpumask remove helper on failure paths and in apple_soc_cpufreq_exit(). This also removes the separate dev_pm_opp_set_sharing_cpus() call, as the cpumask helper loads the DT OPP tables for all CPUs in the policy.
In the Linux kernel, the following vulnerability has been resolved: kho: fix size calculation in kho_preserved_memory_reserve() kho_preserved_memory_reserve() calculates the size of a preservation by doing 1 << (order + PAGE_SHIFT). Since the '1' is a 32-bit integer, it can only be shifted by 31. That is, it will only work for preservations up to 2 GiB. Larger preservations will trigger undefined behaviour. While preservations larger than 2 GiB can't be obtained via folios currently, they can be obtained via kho_preserve_pages(). For example, memblock reserve_mem uses kho_preserve_pages(). Reservations larger than 2 GiB are valid and will trigger this bug if properly aligned. Fix it by using 1UL for shifting.
In the Linux kernel, the following vulnerability has been resolved: jbd2: check need_resched() when skipping busy checkpoint buffers journal_shrink_one_cp_list() skips busy checkpoint buffers when called with JBD2_SHRINK_BUSY_SKIP. The continue statement on this path also skips the need_resched() check at the end of the loop body. Consequently, when a checkpoint list contains mostly busy buffers, the shrinker can walk the entire list while holding journal->j_list_lock, even when a reschedule has been requested. Large checkpoint lists under memory pressure can therefore cause long lock hold times and leave other CPUs spinning on j_list_lock, resulting in soft lockups or RCU stalls. Route the busy-buffer path through the need_resched() check so that the shrinker can release j_list_lock and reschedule promptly, restoring parity with the clean-buffer path, which already checks need_resched(). This does not change which checkpoint buffers are eligible for removal.
In the Linux kernel, the following vulnerability has been resolved: module: validate string table section types In elf_validity_cache_sechdrs, section sizes and offsets are validated, unless the section type is SHT_NULL or SHT_NOBITS. Later, elf_validity_cache_secstrings and elf_validity_cache_index_str access the section name table (.shstrtab) and symbol string table (.strtab) headers without first ensuring that their types are SHT_STRTAB. If a section type is SHT_NULL or SHT_NOBITS, sh_offset has not been validated and may reference out-of-bounds memory when dereferenced in elf_validity_cache_secstrings or elf_validity_cache_strtab. Validate that both string section headers are of type SHT_STRTAB before caching them.
In the Linux kernel, the following vulnerability has been resolved: params: fix charp corruption on allocation failure param_set_charp() stores charp parameters in allocated memory after slab is available, and releases the previous value when the parameter is updated. The previous value is released before the replacement allocation succeeds. If kmalloc_parameter() fails, the setter returns -ENOMEM with the parameter left as NULL. Failing zswap's compressor update before zswap is initialized can later trigger: BUG: kernel NULL pointer dereference, address: 0000000000000000 RIP: 0010:strcmp+0x10/0x30 Call Trace: zswap_setup+0x3b1/0x490 zswap_enabled_param_set+0x5b/0xa0 param_attr_store+0x93/0xe0 module_attr_store+0x1c/0x30 kernfs_fop_write_iter+0x116/0x1f0 Allocate and copy the replacement first, then replace the parameter value only after allocation succeeds.
In the Linux kernel, the following vulnerability has been resolved: sunrpc: fix use-after-free in __rpc_clnt_handle_event and __rpc_clnt_remove_pipedir Normal client creation goes through rpc_setup_pipedir(), which records clnt->pipefs_sb, but the mount-event path in __rpc_clnt_handle_event() calls rpc_setup_pipedir_sb() directly and never refreshes that field. The umount path also removes the directory without clearing clnt->pipefs_sb. After a late pipefs mount or any remount, rpc_clnt_remove_pipedir() compares the current superblock against a stale pipefs_sb pointer and skips cleanup, leaving pipefs dentries whose inode private data still points at a freed rpc_clnt, leading to a potential use-after-free during subsequent rpc_info_open() or rpc_show_info() calls. Fix this by properly updating clnt->pipefs_sb upon mount events and clearing it during unmount or failure paths.
In the Linux kernel, the following vulnerability has been resolved: svcrdma: Reject oversized Read segments at decode time The RPC/RDMA Read list decoder stores wire-supplied segment lengths without validation. xdr_count_read_segments() checks 4-byte alignment for non-zero position values but does not cap the segment length. An oversized rs_length reaches svc_rdma_build_read_segment(), which derives nr_bvec from it and can drive a large dynamic bvec allocation before verifying that enough rq_pages remain. If the post-allocation page-overrun guard fires, the freshly acquired rw context is not returned, leaking the resource. Reject any segment whose length exceeds the receive context's page budget during Read list decoding, consistent with how xdr_check_write_chunk() bounds Write segment counts against rc_maxpages. Also return the rw context on the existing post-allocation overrun path in svc_rdma_build_read_segment(), keeping that defensive guard balanced.
In the Linux kernel, the following vulnerability has been resolved: svcrdma: Use svc_xprt_put to free listener on create failure svc_rdma_create() calls kfree(cma_xprt) when svc_rdma_create_listen_id() fails. svc_xprt_init() has already acquired a net namespace reference via get_net_track(); kfree bypasses svc_xprt_free() which releases it. Replace the kfree() with svc_xprt_put() so the kref_init birth reference drops to zero and svc_xprt_free() dispatches svc_rdma_free() to clean up properly. sc_cm_id is still NULL at that point; the preceding patch added the necessary NULL guard in svc_rdma_free(). svc_xprt_free() also drops the module reference via module_put(), but the caller _svc_xprt_create() does the same on xpo_create failure, double-putting the single try_module_get() it acquired. Take a compensating __module_get() before the svc_xprt_put() to keep the count balanced, matching the convention in svc_rdma_accept()'s error path.
In the Linux kernel, the following vulnerability has been resolved: udf: reject VAT indexes equal to the entry count UDF 1.50 virtual partition mapping uses the VAT as an array of physical block mappings. s_num_entries stores the number of entries in that array, not the highest valid index. The valid VAT indexes are therefore below s_num_entries. udf_get_pblock_virt15() currently rejects only indexes greater than s_num_entries. A crafted image can request index s_num_entries, pass the bounds check, and make the kernel read one entry past the allocated VAT table. Change the check to reject block >= s_num_entries, so the count is handled as an exclusive upper bound. A crafted UDF image reproduced this on origin/master commit 0e35b9b6ec0ffcc5e23cbdec09f5c622ad532b53 with a KASAN slab-out-of-bounds report in udf_get_pblock_virt15(). Trail of Bits has a reproducer that triggers kernel panic demonstrating the bug, and can share it if needed.
In the Linux kernel, the following vulnerability has been resolved: sched_ext: Replace SCX_RQ_BAL_KEEP with a dispatch verdict return SCX_RQ_BAL_KEEP tells the pick to keep running the previous task, a leftover from when balancing and picking were separate operations. An rq-level flag only works while dispatches and picks pair up one to one, which core scheduling breaks: selections interleave through dispatch's lock drops and a pick can consume a stale flag, keeping a task that has since been dequeued. Fixing core scheduling support requires the decision to travel with the dispatch that made it. Make scx_dispatch_sched() and balance_one() return an explicit verdict instead and drop the flag's plumbing from the tools autogen enum headers. Also factor the pick-side invocation, its follow-up queueing and the post-dispatch checks out of do_pick_task_scx() into dispatch_pick(). No functional changes intended. v2: Drop the SCX_RQ_BAL_KEEP plumbing from the tools autogen enum headers as well (Andrea).
In the Linux kernel, the following vulnerability has been resolved: sched_ext: Fix this_rq() assumptions in dispatch kfuncs Under core scheduling, dispatch runs from within the core-wide pick and can target a sibling rq, so ops.dispatch() may execute on a CPU different from the dispatched rq's. Several kfunc paths assumed the two always coincide: - scx_dsq_move() decided whether an rq lock is held by testing this_rq()'s rq flags and lock-danced accordingly. A dispatch for a sibling took the unlocked-context branch and acquired the source rq lock on top of the already held dispatched rq lock which could deadlock. - scx_bpf_sub_dispatch() dispatched this_rq() with its stashed sub_dispatch_prev, which is NULL when dispatching for a sibling. - finish_dispatch(), scx_bpf_dsq_reenq() and scx_bpf_dsq_nr_queued() resolved SCX_DSQ_LOCAL to this CPU's local DSQ rather than the dispatched rq's. The latter two are callable from other rq-locked operations too, where SCX_DSQ_LOCAL now likewise resolves to the op's rq. This changes behavior also without core scheduling, e.g. for ops.enqueue() running a remote wakeup on the waking CPU, and is intended: which CPU happens to execute an operation is incidental, the op's rq is what it is operating on, and the resolution now matches the insert side where SCX_DSQ_LOCAL dispatches land on the task's rq. Use the rq tracked by scx_locked_rq(), which is set to the dispatched rq around ops invocations and NULL in unlocked contexts.
In the Linux kernel, the following vulnerability has been resolved: sched_ext: Fix rq->core_pick corruption under core scheduling Core scheduling's pick_next_task() picks what to run on every SMT sibling of the core in a single pass under the shared core-wide rq lock. The selection state is consistent only while the lock is held continuously, so ->pick_task() originally could not release it. However, since 4c95380701f5 ("sched/ext: Fold balance_scx() into pick_task_scx()"), sched_ext runs dispatch from inside the pick and dispatching can drop the rq lock. To support this, pick_next_task() has been updated to restart the whole selection when a pick returns RETRY_TASK after releasing the lock. When selections on the same core interleave through the dropped lock, they corrupt each other's state: one clears the other's rq->core_pick leading to a NULL deref, or invalidates its keep-the-previous-task decision leaving a dequeued task running, which deadlocks the next wakeup and matches the reported hard hangs. A cookied ping-pong load on an SMT machine makes the interleavings frequent and kills the kernel within seconds. Fix it by making the pick return RETRY_TASK whenever dispatch released the rq lock, so that a selection only ever commits picks made under a continuously held lock. The previous patch's rq->scx.lock_drop_seq counts the releases. A dispatch that touched nothing never releases the lock and its verdict, including "nothing to run", stands: retries are bounded, each following a dispatch that actually did something, and an idle CPU does not loop. If another dispatch is already in flight on the rq, skip dispatching and pick from what is already queued locally - the in-flight dispatch has released the lock, so its own selection will retry and re-pick this rq, while returning RETRY_TASK here would only spin on the lock that dispatch needs to finish. Balance callbacks must run in the context that queued them, so they can only be queued on the CPU's own rq. When dispatching for another rq, run the deferred work directly instead - that rq may consume all its picks through the core-sched fast path and never queue the callback itself. The put_prev_task_scx() warning about a runnable task being left behind assumed that dispatch ran as part of the very pick that is switching away. That now only holds on the non-core path, so gate it and drop the cookie-match test, which is always true without core scheduling, from its condition.
In the Linux kernel, the following vulnerability has been resolved: sched_ext: Don't BUG_ON a destroyed DSQ in process_deferred_reenq_users scx_bpf_dsq_reenq() queues a deferred reenq (dru) that runs from run_deferred(), not ops.dispatch(). If the DSQ is destroyed before the dru runs, process_deferred_reenq_users() sees dsq->id == SCX_DSQ_INVALID and hits the BUG_ON. destroy_dsq() doesn't flush pending drus, so just skip. tj: Read dsq->id once with READ_ONCE(). Reading it separately in the INVALID check and the BUG_ON would leave a window where destroy_dsq() can invalidate the id between the two reads and still trigger the BUG_ON.
In the Linux kernel, the following vulnerability has been resolved: scsi: core: Fill in DMA padding bytes in scsi_alloc_sgtables() During fuzz testing, the following issue was discovered: BUG: KMSAN: uninit-value in __dma_map_sg_attrs+0x217/0x310 __dma_map_sg_attrs+0x217/0x310 dma_map_sg_attrs+0x4a/0x70 ata_qc_issue+0x9f8/0x1420 __ata_scsi_queuecmd+0x1657/0x1740 ata_scsi_queuecmd+0x79a/0x920 scsi_queue_rq+0x4472/0x4f40 blk_mq_dispatch_rq_list+0x1cca/0x3ee0 __blk_mq_sched_dispatch_requests+0x458/0x630 blk_mq_sched_dispatch_requests+0x15b/0x340 __blk_mq_run_hw_queue+0xe5/0x250 __blk_mq_delay_run_hw_queue+0x138/0x780 blk_mq_run_hw_queue+0x4bb/0x7e0 blk_mq_sched_insert_request+0x2a7/0x4c0 blk_execute_rq+0x497/0x8a0 sg_io+0xbe0/0xe20 scsi_ioctl+0x2b36/0x3c60 sr_block_ioctl+0x319/0x440 blkdev_ioctl+0x80f/0xd70 __se_sys_ioctl+0x219/0x420 __x64_sys_ioctl+0x93/0xe0 x64_sys_call+0x1d6c/0x3ad0 do_syscall_64+0x4c/0xa0 entry_SYSCALL_64_after_hwframe+0x6e/0xd8 Uninit was created at: __alloc_pages+0x5c0/0xc80 alloc_pages+0xe0e/0x1050 blk_rq_map_user_iov+0x2b77/0x6100 blk_rq_map_user_io+0x2fa/0x4d0 sg_io+0xad6/0xe20 scsi_ioctl+0x2b36/0x3c60 sr_block_ioctl+0x319/0x440 blkdev_ioctl+0x80f/0xd70 __se_sys_ioctl+0x219/0x420 __x64_sys_ioctl+0x93/0xe0 x64_sys_call+0x1d6c/0x3ad0 do_syscall_64+0x4c/0xa0 entry_SYSCALL_64_after_hwframe+0x6e/0xd8 Bytes 14-15 of 16 are uninitialized Memory access of size 16 starts at ffff88800cbdb000 When processing the last unaligned element of the scatterlist, it is supplemented with missing bytes in the amount of pad_len. These bytes remain uninitialized, which leads to a problem. Extend last_sg->length by pad_len first, then use sg_zero_buffer() to zero those pad_len bytes. sg_zero_buffer() uses sg_miter internally, which correctly handles sg entries spanning multiple pages and padding that crosses a page boundary. Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
In the Linux kernel, the following vulnerability has been resolved: scsi: fnic: Use GFP_ATOMIC for VLAN alloc under spinlock fnic_fcoe_process_vlan_resp() allocates a VLAN descriptor with kzalloc_obj() (default GFP_KERNEL) while holding vlans_lock via spin_lock_irqsave(). GFP_KERNEL may sleep, which is not allowed in this atomic context and can trigger a sleeping-from-invalid-context warning or deadlock. Pass GFP_ATOMIC so the allocation is safe under the IRQ-safe spinlock.
In the Linux kernel, the following vulnerability has been resolved: remoteproc: scp: Fix device reference leak on failed lookup Make sure to drop the reference taken to the SCP device when attempting to look up its driver data before the driver has been bound. Note that holding a reference to a device does not prevent its driver data from going away.
In the Linux kernel, the following vulnerability has been resolved: RDMA/ionic: Embed counter driver data in rdma_counter allocation Commit 7e53b31acc7f ("RDMA/core: Create and destroy rdma_counter using rdma_zalloc_drv_obj()") requires drivers implementing counter ops to embed struct rdma_counter in a driver-specific struct, register its size via INIT_RDMA_OBJ_SIZE, and provide a counter_init callback. The ionic driver was merged without this adaptation, causing a NULL pointer dereference in alloc_and_bind() since rdma_zalloc_drv_obj() allocates zero bytes when size_rdma_counter is unset. Consolidate struct ionic_counter into a new struct ionic_rdma_counter that embeds struct rdma_counter, replace the xarray with a lightweight ida for ID allocation, and add the required counter_init and INIT_RDMA_OBJ_SIZE declarations.
In the Linux kernel, the following vulnerability has been resolved: RDMA/uverbs: Add UVERBS_ATTR_UHW to UVERBS_METHOD_REG_MR The original commit missed that three drivers (mthca, irdma, siw) have UHW data associated with reg_mr that cannot be passed through the ioctl. They also assume that the udata cannot be NULL, so failing to pass a valid udata can trigger a NULL udata crash in those drivers. This never happens in real systems since in rdma-core ibv_cmd_reg_mr_ex() does not accept a udata and those three drivers don't use it, however a malicious userspace could trigger it.
In the Linux kernel, the following vulnerability has been resolved: RDMA/uverbs: Guard legacy bundles without method_elm The legacy write() path dispatches through a uverbs_api_write_method, but the uverbs_attr_bundle passed to provider code does not have an ioctl method element. If malformed provider input causes the common uverbs validation code to emit an error message, uverbs_get_handler_fn() dereferences the uninitialized method_elm pointer. Initialize method_elm explicitly for legacy bundles and make uverbs_get_handler_fn() return NULL when no ioctl method is present. The legacy dispatcher continues to use its local write method, while the ioctl path continues to use the registered ioctl handler.
In the Linux kernel, the following vulnerability has been resolved: ring-buffer: Free cpu_buffer::free_page with subbuf_order When sub-buffers use an order greater than 0, cpu_buffer->free_page is allocated with subbuf_order. Use the correct order for cpu_buffer->free_page.
In the Linux kernel, the following vulnerability has been resolved: orangefs: fix double-free of trailer_buf on readdir copy failure On a readdir downcall, orangefs_devreq_write_iter() frees op->downcall.trailer_buf with vfree() when copy_from_iter_full() fails, but does not clear the pointer before goto Efault. The waiter in do_readdir() is then woken with a negative status and frees the same pointer again on its r < 0 path, causing a deterministic double-free. A client holding /dev/pvfs2-req triggers it by sending a readdir downcall whose declared trailer_size exceeds the bytes it supplies. Clear the pointer after freeing so the readdir-side vfree() becomes a no-op.
In the Linux kernel, the following vulnerability has been resolved: ocfs2: always run deallocs on copy-on-write completion Local fuzzing of 6.12.94 has found the following memory leak caused by doing 'copy_file_range()' within the same filesystem: unreferenced object 0xffff88812192c980 (size 32): comm "syz.0.49", pid 12095, jiffies 4294964143 hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 08 00 00 00 00 00 00 00 ................ c0 c5 92 21 81 88 ff ff 00 02 00 00 00 06 00 00 ...!............ backtrace (crc 7068d63f): kmemleak_alloc_recursive include/linux/kmemleak.h:42 [inline] slab_post_alloc_hook mm/slub.c:4152 [inline] slab_alloc_node mm/slub.c:4197 [inline] __kmalloc_cache_noprof+0x168/0x2c0 mm/slub.c:4358 kmalloc_noprof include/linux/slab.h:878 [inline] ocfs2_find_per_slot_free_list fs/ocfs2/alloc.c:6618 [inline] ocfs2_cache_block_dealloc+0x155/0x4b0 fs/ocfs2/alloc.c:6786 ocfs2_cache_extent_block_free fs/ocfs2/alloc.c:6819 [inline] ocfs2_unlink_path+0x286/0x450 fs/ocfs2/alloc.c:2613 ocfs2_rotate_subtree_left fs/ocfs2/alloc.c:2779 [inline] __ocfs2_rotate_tree_left+0x1f6f/0x2da0 fs/ocfs2/alloc.c:2985 ocfs2_rotate_tree_left+0x283/0xe00 fs/ocfs2/alloc.c:3237 ocfs2_try_to_merge_extent+0xf56/0x1a20 fs/ocfs2/alloc.c:3825 ocfs2_split_extent+0x15f4/0x2940 fs/ocfs2/alloc.c:5138 ocfs2_clear_ext_refcount+0x2f6/0x550 fs/ocfs2/refcounttree.c:3098 ocfs2_replace_clusters fs/ocfs2/refcounttree.c:3131 [inline] ocfs2_make_clusters_writable fs/ocfs2/refcounttree.c:3255 [inline] ocfs2_replace_cow+0x991/0x1660 fs/ocfs2/refcounttree.c:3349 ocfs2_refcount_cow_hunk fs/ocfs2/refcounttree.c:3427 [inline] ocfs2_refcount_cow+0x5e1/0x9f0 fs/ocfs2/refcounttree.c:3470 ocfs2_prepare_inode_for_write fs/ocfs2/file.c:2340 [inline] ocfs2_file_write_iter+0xbda/0x1880 fs/ocfs2/file.c:2451 iter_file_splice_write+0x890/0xf60 fs/splice.c:743 do_splice_from fs/splice.c:944 [inline] direct_splice_actor+0x232/0x480 fs/splice.c:1167 splice_direct_to_actor+0x4b4/0xb60 fs/splice.c:1111 do_splice_direct_actor fs/splice.c:1210 [inline] do_splice_direct+0x10f/0x1c0 fs/splice.c:1236 do_sendfile+0x430/0xbf0 fs/read_write.c:1388 unreferenced object 0xffff88812192c5c0 (size 32): comm "syz.0.49", pid 12095, jiffies 4294964143 hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ 29 70 00 00 00 00 00 00 19 00 00 00 00 00 00 00 )p.............. backtrace (crc afec850f): kmemleak_alloc_recursive include/linux/kmemleak.h:42 [inline] slab_post_alloc_hook mm/slub.c:4152 [inline] slab_alloc_node mm/slub.c:4197 [inline] __kmalloc_cache_noprof+0x168/0x2c0 mm/slub.c:4358 kmalloc_noprof include/linux/slab.h:878 [inline] kzalloc_noprof include/linux/slab.h:1014 [inline] ocfs2_cache_block_dealloc+0x25c/0x4b0 fs/ocfs2/alloc.c:6793 ocfs2_cache_extent_block_free fs/ocfs2/alloc.c:6819 [inline] ocfs2_unlink_path+0x286/0x450 fs/ocfs2/alloc.c:2613 ocfs2_rotate_subtree_left fs/ocfs2/alloc.c:2779 [inline] __ocfs2_rotate_tree_left+0x1f6f/0x2da0 fs/ocfs2/alloc.c:2985 ocfs2_rotate_tree_left+0x283/0xe00 fs/ocfs2/alloc.c:3237 ocfs2_try_to_merge_extent+0xf56/0x1a20 fs/ocfs2/alloc.c:3825 ocfs2_split_extent+0x15f4/0x2940 fs/ocfs2/alloc.c:5138 ocfs2_clear_ext_refcount+0x2f6/0x550 fs/ocfs2/refcounttree.c:3098 ocfs2_replace_clusters fs/ocfs2/refcounttree.c:3131 [inline] ocfs2_make_clusters_writable fs/ocfs2/refcounttree.c:3255 [inline] ocfs2_replace_cow+0x991/0x1660 fs/ocfs2/refcounttree.c:3349 ocfs2_refcount_cow_hunk fs/ocfs2/refcounttree.c:3427 [inline] ocfs2_refcount_cow+0x5e1/0x9f0 fs/ocfs2/refcounttree.c:3470 ocfs2_prepare_inode_for_write fs/ocfs2/file.c:2340 [inline] ocfs2_file_write_iter+0xbda/0x1880 fs/ocfs2/file.c:2451 iter_file_splice_write+0x890/0xf60 fs/splice.c:743 do_splice_from fs/splice.c:9 ---truncated---
In the Linux kernel, the following vulnerability has been resolved: ocfs2: cluster: don't sleep while holding o2hb_live_lock in o2hb_region_pin() Patch series "ocfs2: cluster: o2hb_region_pin() fixes", v2. This series fixes three related issues in o2hb_region_pin(), all are from the original implementation in commit: 58a3158a5d17 ("ocfs2/cluster: Pin/unpin o2hb regions"): 1) It is called with o2hb_live_lock (a spinlock) held, but the underlying configfs_depend_item() sleeps (takes inode rwsem and pins the filesystem). This triggers BUG under CONFIG_DEBUG_ATOMIC_SLEEP. 2) When called from the configfs drop_item callback, it creates a lock order inversion: parent inode_lock -> configfs root inode_lock, which can deadlock against subsystem unregistration paths taking root -> parent. 3) If pinning fails partway through o2hb_region_inc_user(), the o2hb_dependent_users counter is leaked and partially-pinned regions are never released, leaving heartbeat regions unprotected on subsequent mounts. Patch 1 reworks o2hb_region_pin() to drop o2hb_live_lock across each sleeping configfs_depend_item() call, using a config_item reference to keep the region alive while unlocked. Patch 2 adds a from_callback parameter to select configfs_depend_item_unlocked() when called from configfs context, avoiding the inode_lock nesting. Patch 3 fixes the error path in o2hb_region_inc_user() to unpin and decrement the counter on failure. This patch (of 3): o2hb_region_pin() is always called with the o2hb_live_lock spinlock held (from o2hb_region_inc_user() and o2hb_heartbeat_group_drop_item()), but it calls o2nm_depend_item() -> configfs_depend_item(), which sleeps: it pins the configfs filesystem and takes the configfs root inode rwsem. Under CONFIG_DEBUG_ATOMIC_SLEEP this triggers: BUG: sleeping function called from invalid context at kernel/locking/rwsem.c in_atomic(): 1, ... name: mount.ocfs2 down_write configfs_depend_item o2hb_region_pin o2hb_region_inc_user o2hb_register_callback dlm_register_domain_handlers ... ocfs2_dlm_init ocfs2_mount_volume ocfs2_fill_super Rework o2hb_region_pin() to pin one region at a time with the lock dropped across the sleeping call: under o2hb_live_lock find the next eligible region and take a config_item reference to keep it alive, drop the lock, call o2nm_depend_item(), then retake the lock and record the pin. The config_item_put() is done with the lock released as well, since o2hb_region_release() also acquires o2hb_live_lock and can sleep. The region list may change while unlocked, so the scan restarts from the top after each pin. Local heartbeat still pins only the matching region; global heartbeat pins all eligible regions. The unpin path is unaffected: configfs_undepend_item() only takes a spinlock and does not sleep.
In the Linux kernel, the following vulnerability has been resolved: ocfs2: fix readdir position truncation on 32-bit kernels In ocfs2_dir_foreach_blk_el(), the directory cookie position is rebuilt with ctx->pos = (ctx->pos & ~(sb->s_blocksize - 1)) | offset; `ctx->pos` is loff_t (signed 64-bit), while `sb->s_blocksize` is unsigned long. On 32-bit kernels unsigned long is 32-bit, so the mask ~(sb->s_blocksize - 1) is computed as a 32-bit unsigned value (e.g. 0xfffff000 for a 4 KiB block size). In the AND expression with the 64-bit `ctx->pos`, that unsigned operand is zero-extended to 64 bits per the usual arithmetic conversions, yielding 0x00000000fffff000. The high 32 bits of `ctx->pos` are silently cleared, even though directory size is allowed to exceed 4 GiB. When readdir() crosses the 4 GiB boundary on a 32-bit kernel the position is reset back into the first 4 GiB block, making the re-validation path re-enumerate already-returned dirents indefinitely. This is ocfs2_dir_foreach_blk_el(), the extent-list readdir path taken for all non-inline directories, so a directory large enough to cross 4 GiB reaches it. This is the same class of bug that commit 3dce5bb82c97 ("exfat: Fix bitwise operation having different size") fixed in exfat, and the fix mirrors the equivalent ext4 fix in this series. Cast the operand to loff_t so the mask is 64-bit before the AND: ctx->pos = (ctx->pos & ~((loff_t)sb->s_blocksize - 1)) | offset; 64-bit kernels are unaffected.
In the Linux kernel, the following vulnerability has been resolved: lockd: fix NULL dereference on lockowner allocation failure nlmclnt_locks_init_private() installs NLM file lock operations even when nlmclnt_find_lockowner() fails to allocate a lockowner. nlmclnt_proc() then returns -ENOMEM, but the VFS still tears down the partially initialized file_lock and calls locks_release_private(). That invokes nlmclnt_locks_release_private(), which dereferences fl->fl_u.nfs_fl.owner and crashes because the owner was never installed. Clear fl_ops before attempting to initialize the NLM private state, and install the NLM lock operations only after a lockowner has been allocated successfully.
In the Linux kernel, the following vulnerability has been resolved: power: supply: bq24257: fix use-after-free on remove The STAT-pin interrupt is devm-managed, so it stays armed until the devm cleanup that runs after remove() returns. remove() cancels bq->iilimit_setup_work while the threaded handler can still fire; that handler reschedules the work and dereferences bq, so the work runs against freed memory once devm frees bq. Make the delayed work device-managed with devm_delayed_work_autocancel(), registered before the interrupt request. The devm cleanup then releases the interrupt first, so the handler can no longer reschedule the work, and cancels the work before bq is freed. The explicit cancel_delayed_work_sync() in remove() is no longer needed and is dropped. Found by static analysis.
In the Linux kernel, the following vulnerability has been resolved: power: supply: bq256xx: drain usb_work before freeing the charger The USB-PHY notifier queues usb_work, whose handler calls power_supply_changed(bq->charger). The reset devm action only unregisters the notifier and was registered before the power supplies, so devm frees bq->charger on unwind before the action runs; a usb_work still queued can then dereference it. Register the reset action after the power supplies, so it unregisters the notifiers and drains usb_work before the supplies are released. Initialize usb_work and obtain the PHY references before registering the notifiers, so the worker cannot run before the supplies exist. Found by static analysis.
In the Linux kernel, the following vulnerability has been resolved: power: supply: lp8788-charger: fix use-after-free on remove lp8788_charger_remove() flushes charger_work before unregistering the IRQs. An IRQ thread can queue charger_work after flush_work() has returned. The work can then run after devres frees pchg and dereference it in lp8788_charger_event(). Unregister the IRQs first. free_irq() waits for any running threaded handler, so no handler can queue more work afterwards. Then use cancel_work_sync() to cancel pending work or wait for running work to finish. This issue was found by an in-house static analysis tool.
In the Linux kernel, the following vulnerability has been resolved: power: supply: qcom_battmgr: fix use-after-free qcom_battmgr_pdr_notify() queues enable_work when the PMIC GLINK service comes up, and the worker recovers battmgr through container_of() to issue firmware requests. The PMIC GLINK client stays on the client list until its devres release action runs, so a PDR notification can keep queueing the work, and a pending or running worker can access battmgr after devres frees it. Make enable_work device-managed with devm_work_autocancel(), registered before the PMIC GLINK client is allocated. The devres cleanup then releases the client first, so no further notification can queue the work, and cancels the work before battmgr is freed. This issue was found by an in-house static analysis tool.
In the Linux kernel, the following vulnerability has been resolved: power: supply: twl4030_charger: cancel workers via devm bci is devm-allocated. Two workers (bci->work and bci->current_worker) dereference it. twl4030_bci_remove() disables charging and masks interrupts. It cancels neither worker. A worker pending at remove() can run after devm frees bci. The USB transceiver comes from devm_usb_get_phy_by_node(). devm unregisters its notifier only after remove() returns. A cancel_work_sync() in remove() can then race a notifier reschedule. devm_work_autocancel() and devm_delayed_work_autocancel() avoid that. They cancel the workers during devm release, before bci is freed. The current_worker is registered first, since devm will cancel in reverse order and bci->work can reschedule current_worker. [Move comment about order into the commit message]
In the Linux kernel, the following vulnerability has been resolved: power: supply: ucs1002: fix use-after-free on remove ucs1002 has no remove callback, so unbind runs entirely through devm. The alert IRQ handler queues the health_poll delayed work, and the work reschedules itself while the chip reports a bad-health condition. devm frees the alert IRQ, which only synchronizes the handler; it does not cancel the delayed work, which can then run after devm frees the driver data and dereference it. Register health_poll with devm_delayed_work_autocancel() before the alert IRQ is requested. devm then frees the IRQ before cancelling the work, so the handler can no longer queue it and the work is cancelled before the driver data is freed. This issue was found by an in-house static analysis tool.
In the Linux kernel, the following vulnerability has been resolved: s390/cpum_cf: Handle CPU hotplug via prepare/dead callbacks The command 'perf stat -e cycles -- <command>' crashes the kernel when CPUs are hotplug added during that run. Root cause is the allocation of struct cpu_cf_events at first event initialization. The allocation is dynamic and the first event that has task context creates such a structure for each online CPU. This is not sufficient. CPUs may be offline during event creation and can be set online during the perf run time. For example commands # echo 0 > /sys/devices/system/cpu/cpu1/online # perf stat -e cycles -i -- stress-ng -t10s --matrix X # sleep 1 # echo 1 > /sys/devices/system/cpu/cpu1/online create an event for CPUs 0,2-X. Since the events are created with task-context, the scheduler will eventually schedule the program on CPU1. This CPU has not created and initialized any per CPU event infrastructure as that CPU was not online at the time of the perf invocation. Thus when the scheduler runs stress-ng on CPU1, the function cpumf_pmu_add() refers to a NULL pointer: struct cpu_cf_events *cpuhw = this_cpu_cfhw(); This function call is invoked after the task stress-ng has been made runnable on CPU1. And this_cpu_cfhw() returns NULL. The result is a panic: Unable to handle kernel pointer dereference in virtual kernel address space Failing address: 0000000000000000 TEID: 0000000000000483 .... Krnl PSW : 0404d00180000000 000003ef8291fd0c (cpumf_pmu_add+0x3c/0x80) .... Call Trace: [<000003ef8291fd0c>] cpumf_pmu_add+0x3c/0x80 [<000003ef82bb5e3e>] event_sched_in+0xae/0x190 [<000003ef82bb60d6>] merge_sched_in+0x1b6/0x390 [<000003ef82bb65b8>] visit_groups_merge.constprop.0.isra.0+0x308/0x5b0 [<000003ef82bb689a>] pmu_groups_sched_in+0x3a/0x50 [<000003ef82bb6a30>] ctx_sched_in+0x180/0x260 [<000003ef82bb780c>] perf_event_context_sched_in+0x11c/0x2d0 [<000003ef82bb79ee>] __perf_event_task_sched_in+0x2e/0xc0 [<000003ef82994834>] finish_task_switch.isra.0+0x1a4/0x250 .... Last Breaking-Event-Address: [<000003ef8291f1d8>] this_cpu_cfhw+0x38/0x40 The issue arises only in per-task context when the CPUMF facility is used and the scheduler picks a random CPU for such a process to run on. The scheduler enables the CPUMF infrastructure via PMU callback functions pmu::add() and pmu::del(). Introduce a CPU hotplug prepare/dead callback pair which creates and removes the per CPU counter data while the CPU is offline. Count the users which track every CPU (cpu == -1), that is perf_event_open() events with task context and /dev/hwctr device sessions, in the new counter cpu_cf_root::tskcnt, protected by pmc_reserve_mutex. This ensures the infrastructure is available when new CPU is selected to run the per-task context process. In cpum_cf_free_root() and cpum_cf_free_cpu() ensure the reference pointer to data structures is set to NULL before the data is freed to prevent interrupt handlers to access stale data. [gor@linux.ibm.com: change commit message]
In the Linux kernel, the following vulnerability has been resolved: s390/dasd: Do not complete a failed ESE read as successful dasd_int_handler() completes an NRF read of an unallocated ESE track by calling ese_read() and unconditionally marking the request DASD_CQR_SUCCESS. dasd_eckd_ese_read() can return an error before it has zeroed the destination buffer: a failed sense-data parse or a current track outside the requested range both return early, leaving the destination pages untouched. The request is still completed successfully, so the block layer is handed stale / uninitialized memory instead of zeros. Check the ese_read() return value and fail the request through the normal error path instead of forcing DASD_CQR_SUCCESS.
In the Linux kernel, the following vulnerability has been resolved: s390/dasd: Guard sysfs discipline callbacks against unallocated private data Several sysfs show/store handlers call a discipline callback that dereferences device->private, either directly or through the DASD_DEFINE_ATTR() macro. During dasd_generic_set_online() the discipline is assigned before check_device() allocates device->private, so an unprivileged read of one of these world-readable attributes in that window dereferences a NULL pointer and panics. Guard the dereference inside each callback that actually touches device->private.
In the Linux kernel, the following vulnerability has been resolved: PCI: plda: Fix use-after-free of event IRQs during teardown plda_pcie_irq_domain_deinit() removes pcie->event_domain via irq_domain_remove(), but the per-event IRQs mapped from that domain are requested with devm_request_irq() in plda_init_interrupts(). The actual free_irq() for a devm-managed IRQ is deferred by devres until after the calling probe()/remove() function returns. This means irq_domain_remove() can free the domain's internal data before the deferred free_irq() for IRQs still mapped into it has run. When devres later processes that deferred cleanup, it can end up dereferencing the already-freed domain. Free each event IRQ explicitly with devm_free_irq() before removing the domain. This triggers the free immediately and removes the IRQ from the devres tracking list, so devres will not attempt to free it a second time later. Also dispose of the event, INTx, and MSI IRQ mappings with irq_dispose_mapping() before their owning domains are removed. Finally, guard the calls to irq_set_chained_handler_and_data() for pcie->irq, pcie->msi_irq, and pcie->intx_irq so they only run when those fields hold a valid (>0) IRQ number. This is a pre-existing issue, flagged by automated review during work on an earlier, unrelated patch to this driver. Build-tested and boot-tested on StarFive VisionFive v1.2A board
In the Linux kernel, the following vulnerability has been resolved: PCI: plda: Fix IRQ domain leaks in the error paths of plda_init_interrupts() plda_init_interrupts() initializes IRQ domains and creates IRQ mapping but does not unwind them when later step fails. If platform_get_irq() or either irq_create_mapping() fails in plda_init_interrupts(), the domains are never deinitialized. If irq_create_mapping() fails, port->intx_irq stays initialized. Hence, remove the IRQ domains in the error path by calling plda_pcie_irq_domain_deinit(). Since plda_pcie_irq_domain_deinit() now disposes of the intx_irq and msi_irq mappings itself before removing their domains, the msi_irq mapping failure path can go directly to err_irq_domain_deinit instead of disposing of port->intx_irq separately first. This issue was found by automated review of sashiko-bot [mani: commit log]
In the Linux kernel, the following vulnerability has been resolved: iommu/amd: Put PCI device after handling PPR faults iommu_call_iopf_notifier() looks up the requester with pci_get_domain_bus_and_slot(), which returns a PCI device with its reference count incremented. Neither the successful iommu_report_device_fault() path nor the abort path drops that reference, so every handled PPR request leaks a PCI device reference. This is the same ownership rule that was fixed for the old iommu_v2 ppr_notifier() path by commit 6cf0981c2233 ("iommu/amd: Fix pci device refcount leak in ppr_notifier()"), but iommu_call_iopf_notifier() was added later as a separate PPR/IOPF notifier path. Drop the PCI device reference after handling the PPR entry.
In the Linux kernel, the following vulnerability has been resolved: iommu/sva: Set handle->dev before the SVA handle is visible iommu_attach_device_pasid() installs the new SVA attach handle in the group PASID lookup before iommu_sva_bind_device() returns. A concurrent bind can therefore find and reuse the same handle after iommu_sva_lock is dropped. handle->dev was initialized after dropping iommu_sva_lock. This leaves a window where a racing bind can return a handle whose dev pointer is still NULL. A subsequent iommu_sva_unbind_device() can then dereference it via handle->dev->iommu_group. Initialize handle->dev before releasing iommu_sva_lock so any visible SVA handle is fully initialized.
In the Linux kernel, the following vulnerability has been resolved: iommu: Fix dev_iommu memory leak when device_add fails in iommu_mock_device_add iommu_mock_device_add() first calls iommu_fwspec_init(), which on success allocates both dev->iommu (via dev_iommu_get()) and dev->iommu->fwspec. If the subsequent device_add(dev) call fails, the error path only calls iommu_fwspec_free(dev), which frees fwspec but leaves dev->iommu still allocated. This triggers the following kmemleak report when fuzzing with Syzkaller: BUG: memory leak unreferenced object 0xffff888011e0a200 (size 192): comm "syz.1.1695", pid 24885, jiffies 4295222527 hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 00 00 00 00 ad 4e ad de .............N.. ff ff ff ff 00 00 00 00 ff ff ff ff ff ff ff ff ................ backtrace (crc 25df5bb3): kmemleak_alloc_recursive include/linux/kmemleak.h:44 [inline] slab_post_alloc_hook mm/slub.c:4575 [inline] slab_alloc_node mm/slub.c:4899 [inline] __kmalloc_cache_noprof+0x47a/0x710 mm/slub.c:5415 kmalloc_noprof include/linux/slab.h:950 [inline] kzalloc_noprof include/linux/slab.h:1188 [inline] dev_iommu_get+0x10c/0x1a0 drivers/iommu/iommu.c:408 iommu_fwspec_init+0x288/0x4d0 drivers/iommu/iommu.c:3087 iommu_mock_device_add+0x46/0xb0 drivers/iommu/iommu.c:385 mock_dev_create drivers/iommu/iommufd/selftest.c:1025 [inline] iommufd_test_mock_domain drivers/iommu/iommufd/selftest.c:1066 [inline] iommufd_test+0x2f8a/0x6190 drivers/iommu/iommufd/selftest.c:2072 iommufd_fops_ioctl+0x367/0x540 drivers/iommu/iommufd/main.c:533 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:597 [inline] __se_sys_ioctl fs/ioctl.c:583 [inline] __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:583 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0x116/0x800 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f Fix this by calling dev_iommu_free(dev) instead of iommu_fwspec_free(dev) in the device_add() failure path. dev_iommu_free() frees both fwspec and the outer dev_iommu struct and clears dev->iommu.
In the Linux kernel, the following vulnerability has been resolved: iommufd: Avoid locking internal accesses during unmap iommufd_access_notify_unmap() skips internal accesses because they do not have an external unmap callback to invoke. However, the current test calls iommufd_lock_obj() before checking whether the access is internal. If iommufd_lock_obj() succeeds, the loop then sees the internal access and continues, bypassing the matching iommufd_put_object() used by the normal unmap path. This leaks the object reference taken by iommufd_lock_obj(). Check for internal accesses first so skipped entries are never locked.