Export limit exceeded: 376903 CVEs match your query. Please refine your search to export 10,000 CVEs or fewer.
Export limit exceeded: 21578 CVEs match your query. Please refine your search to export 10,000 CVEs or fewer.
Search
Search Results (21578 CVEs found)
| CVE | Vendors | Products | Updated | CVSS v3.1 |
|---|---|---|---|---|
| CVE-2026-68266 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 7.0 High |
| In the Linux kernel, the following vulnerability has been resolved: drm/xe: Hold a dma-buf reference for imported BOs An imported dma-buf BO is created as a ttm_bo_type_sg BO whose reservation object is the exporter's dma_buf->resv. The importer, however, only takes a dma-buf reference after a successful dma_buf_dynamic_attach(). Until then nothing keeps the exporter alive, so if the exporter is freed while the BO still references its resv, a later access to that resv is a use-after-free: Oops: general protection fault, probably for non-canonical address 0x6b6b6b6b6b6b6b9c Workqueue: ttm ttm_bo_delayed_delete [ttm] RIP: 0010:mutex_can_spin_on_owner+0x3f/0xc0 This can be reached on two paths: - dma_buf_dynamic_attach() fails, or - ttm_bo_init_reserved() fails during BO creation. In both cases the BO already has bo->base.resv pointing at the exporter resv, and sg BOs are always torn down via ttm_bo_delayed_delete(), which locks bo->base.resv asynchronously - potentially after the exporter has been freed. Take the dma-buf reference in xe_bo_init_locked(), before ttm_bo_init_reserved(), so it also covers a creation failure there, and release it in xe_ttm_bo_destroy(). The reference is held for the whole BO lifetime, keeping the shared resv alive on every path. v2: - Reworked the fix to avoid creating the imported sg BO before dma_buf_dynamic_attach() succeeds. - Attach with importer_priv == NULL and make invalidate_mappings ignore incomplete imports. v3: - Dropped the xe-side reordering approach since importer_priv must be valid when dma_buf_dynamic_attach() publishes the attachment. - Per Christian's suggestion on the v1 thread, keyed the check on import_attach rather than removing the sg guard entirely. - Fixes both xe and amdgpu in a single TTM patch. v4: - Moved import_attach check to after dma_resv_copy_fences() so fences are copied before returning for successful imports (Thomas). - Removed exporter-alive claim from commit message (Thomas). v5: - Add drm/xe patch to keep imported sg BOs off the LRU before attach succeeds; the TTM fix alone is not sufficient for xe if the BO is already LRU-visible. (Thomas) v4 patch: https://patchwork.freedesktop.org/patch/736663/?series=169129&rev=2 - Patch 1 (drm/ttm) carries Christian's Reviewed-by from v4. v6: - Reworked the fix based on Thomas' suggestion. Instead of the TTM resv individualization (v1-v5) plus the xe off-LRU/placement handling (v5), just hold a dma-buf reference for the imported BO lifetime so the shared resv can never be freed while the BO still references it. Single xe patch, no TTM change. (Thomas) - Take the reference in xe_bo_init_locked() before ttm_bo_init_reserved() so a TTM creation failure is covered too (Thomas). - Dropped the v5 series (drm/ttm + drm/xe off-LRU); the off-LRU approach also regressed in CI BAT via ttm_bo_pipeline_gutting() creating a ghost BO that outlived the exporter. Link to v5: https://patchwork.freedesktop.org/series/169984/ v7: - Move changelog above --- so it stays in the commit message. - Reorder changelog entries oldest-to-newest. (Thomas) (cherry picked from commit 3516f3fae6be35642f8f06f8a218da6425c0306a) | ||||
| CVE-2026-68105 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 4.1 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: Fix kernel panic during driver load failure Avoid kernel panic if MES init fails during driver load. The KIQ ring is falsely marked as ready as ASICs that use MES, KIQ is owned by MES. BUG: kernel NULL pointer dereference, address: 0000000000000000 RIP: 0010:gfx_v12_1_wait_reg_mem+0x5a/0x1f0 [amdgpu] Call Trace: gfx_v12_1_ring_emit_reg_write_reg_wait+0x1f/0x30 [amdgpu] amdgpu_gmc_fw_reg_write_reg_wait+0xb2/0x190 [amdgpu] amdgpu_gmc_flush_gpu_tlb+0x1cc/0x230 [amdgpu] amdgpu_gart_invalidate_tlb+0x81/0xa0 [amdgpu] amdgpu_gart_unbind+0x72/0x90 [amdgpu] amdgpu_ttm_backend_unbind+0xa4/0xb0 [amdgpu] amdgpu_ttm_tt_unpopulate+0x13/0xd0 [amdgpu] amdttm_tt_unpopulate+0x29/0x70 [amdttm] ttm_bo_put+0x1eb/0x360 [amdttm] amdgpu_bo_free_kernel+0xf9/0x1f0 [amdgpu] amdgpu_ih_ring_fini+0x5a/0x90 [amdgpu] amdgpu_irq_fini_hw+0x58/0x80 [amdgpu] amdgpu_device_fini_hw+0x4e0/0x5b0 [amdgpu] amdgpu_driver_load_kms+0x60/0xa0 [amdgpu] amdgpu_pci_probe+0x28e/0x6d0 [amdgpu] pci_device_probe+0x19f/0x220 really_probe+0x1ed/0x340 driver_probe_device+0x1e/0x80 __driver_attach+0xd3/0x1a0 bus_for_each_dev+0x68/0xa0 bus_add_driver+0x19f/0x270 driver_register+0x5d/0xf0 do_one_initcall+0xac/0x200 do_init_module+0x1ec/0x280 __se_sys_finit_module+0x2de/0x310 do_syscall_64+0x6a/0x250 entry_SYSCALL_64_after_hwframe+0x4b/0x53 (cherry picked from commit 4623b958dd6da0f4c3026afdf330626a09ecb0f0) | ||||
| CVE-2026-68115 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 4.7 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu/gfx10: replace BUG_ON() with WARN_ON() There's no need to crash the kernel for these cases. (cherry picked from commit ac6f00beb658239bced4aaed9efbb04a35348d48) | ||||
| CVE-2026-68116 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: vxlan: mdb: Fix source list corruption on a failed replace When replacing the source list of an MDB remote entry, all existing sources are first marked for deletion and vxlan_mdb_remote_srcs_add() is then called to add the new source list. Sources present in the new list have their deletion mark cleared, and any sources left marked afterwards are removed. If vxlan_mdb_remote_srcs_add() fails partway through, its error path deletes all entries on the remote's source list. That rollback is only correct for its other caller, vxlan_mdb_remote_add(), where the remote was just allocated and the list contains solely entries added during the call. On the replace path the list also holds pre-existing sources, so a failed replace tears them down together with their (S, G) forwarding entries instead of leaving the entry unchanged. This is reachable from an existing (*, G) remote. An EXCLUDE filter that loses sources starts forwarding traffic that should be blocked, while an INCLUDE filter that loses sources drops traffic that should be forwarded. Mark entries created during the current pass with a new VXLAN_SGRP_F_NEW flag. On failure, delete only those entries and clear the deletion mark on the pre-existing ones, so a failed replace leaves the source list untouched. Retain the flag until the whole operation succeeds and then clear it. Also stop vxlan_mdb_remote_src_add() from deleting a pre-existing entry it only looked up when adding that entry's forwarding entry fails. | ||||
| CVE-2026-68122 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.9 Medium |
| In the Linux kernel, the following vulnerability has been resolved: ovpn: fix peer refcount leak in TCP error paths When either the TCP RX or TX error path calls ovpn_peer_hold() followed by schedule_work(&peer->tcp.defer_del_work), and the work item is already pending from the other path, schedule_work() returns false and the work runs only once. Since ovpn_tcp_peer_del_work() calls ovpn_peer_put() exactly once, the extra reference taken by the losing path is never dropped, leaking the peer object. The race window: CPU0 (strparser/RX error): CPU1 (tcp_tx_work/TX error): ovpn_peer_hold() <- refcnt+1 ovpn_peer_hold() <- refcnt+2 schedule_work() <- queued schedule_work() <- NO-OP (work already pending) ovpn_tcp_peer_del_work runs: ovpn_peer_del() ovpn_peer_put() <- refcnt+1 <- peer never freed Fix by checking the return value of schedule_work() in both paths and calling ovpn_peer_put() to drop the extra reference if the work was already pending. ovpn_peer_hold() is kept unconditional in the TX path as it cannot fail at that point. | ||||
| CVE-2026-68124 | 1 Linux | 1 Linux Kernel | 2026-08-10 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: mctp: serial: handle zero-length frames to prevent rx buffer overflow The MCTP serial receive state machine reads a frame length byte in mctp_serial_push_header() case 2 and validates it upper-bound-only: if (c > MCTP_SERIAL_FRAME_MTU) { dev->rxstate = STATE_ERR; } else { dev->rxlen = c; dev->rxpos = 0; dev->rxstate = STATE_DATA; ... } A length of zero passes this check, so rxlen is set to 0 and the state machine advances to STATE_DATA. In mctp_serial_push() STATE_DATA, the incoming byte is stored and rxpos incremented before the terminator is dev->rxbuf[dev->rxpos] = c; dev->rxpos++; dev->rxstate = STATE_DATA; if (dev->rxpos == dev->rxlen) { dev->rxpos = 0; dev->rxstate = STATE_TRAILER; } With rxlen == 0 the "rxpos == rxlen" terminator can never fire (rxpos is already 1 on the first data byte), so subsequent bytes are written past the end of the fixed 74-byte rxbuf, which is the last member of the netdev private area. Every following data byte is an attacker-controlled 1-byte out-of-bounds heap write, and the overflow continues until a frame (0x7e) or escape byte resets the parser -- effectively unbounded. Reaching this requires CAP_NET_ADMIN to attach the N_MCTP line discipline and bring the resulting mctpserialN netdev up, after which the bytes arrive via the tty receive path. Route a zero-length frame straight to STATE_TRAILER instead of STATE_DATA. The trailer/framing bytes are still consumed, and the frame resolves to a zero-length skb that the MCTP core rejects; the parser never enters STATE_DATA with rxlen == 0, so the out-of-bounds write can no longer occur. KASAN, on a frame of 0x7e 0x01 0x00 followed by data bytes (before this change): UBSAN: array-index-out-of-bounds in drivers/net/mctp/mctp-serial.c:370 index 74 is out of range for type 'u8 [74]' BUG: KASAN: slab-out-of-bounds in mctp_serial_tty_receive_buf Write of size 1 at addr ... by task kworker/u16:0 mctp_serial_tty_receive_buf tty_ldisc_receive_buf flush_to_ldisc Allocated by task 152: alloc_netdev_mqs mctp_serial_open v2: route zero-length frames to STATE_TRAILER instead of STATE_ERR so the trailer/framing bytes are still consumed (Jeremy Kerr). Found by 0sec automated security-research tooling (https://0sec.ai). | ||||
| CVE-2026-68125 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 6.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: mac802154: llsec: reject frames shorter than the authentication tag llsec_do_decrypt_auth() computes the associated-data length for the AEAD request as assoclen += datalen - authlen; where datalen is the number of bytes after the MAC header and authlen (4, 8 or 16) is the length of the authentication tag. Nothing verifies that the frame actually carries at least authlen payload bytes. A secured frame whose payload is shorter than the tag makes datalen - authlen negative; assoclen is then passed to aead_request_set_ad() as an unsigned value close to 4 GiB, so crypto_aead_decrypt() walks far off the end of the scatterlist that only spans the real frame. The frame is fully attacker-controlled and reaches this path from any IEEE 802.15.4 peer in radio range. Reject frames whose payload is shorter than the authentication tag before the subtraction. Dynamically reproduced on a KASAN kernel as a general-protection-fault in the AEAD scatterwalk, and the fix confirmed. | ||||
| CVE-2026-68232 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/gpusvm: Fix MM reference leak in drm_gpusvm_range_evict If kvmalloc_array() fails in drm_gpusvm_range_evict(), the MM reference acquired earlier is not released, resulting in a reference leak. Fix this by dropping the MM reference on the kvmalloc_array() failure path. | ||||
| CVE-2026-68239 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/ttm: Account for NULL and handle pages in ttm_pool_backup Pages in ttm_pool_backup can be NULL or backup handles (ttm_backup_page_ptr_is_handle()), neither of which can be passed to set_pages_array_wb() or freed. Add a dedicated WB pass before the dma/purge loop that walks allocations using the same i += num_pages stride, skipping NULL and handle entries, and calls set_pages_array_wb() once per contiguous run of real pages. Apply the same NULL/handle guard to the dma/purge loop. Fixes the following oops: Oops: general protection fault, kernel NULL pointer dereference 0x0: 0000 [#1] SMP NOPTI RIP: 0010:__cpa_process_fault+0xf8/0x770 RSP: 0018:ffffc90000a87718 EFLAGS: 00010287 RAX: 0000000000000000 RBX: ffffc90000a87868 RCX: 0000000000000000 RDX: 0000000000001000 RSI: 0005088000000000 RDI: ffffffff827c5f34 RBP: 0005088000000000 R08: ffffc90000a877cb R09: ffffc90000a877d0 R10: 0000000000000000 R11: 000000000000001b R12: 000ffffffffff000 R13: ffffc90000a87868 R14: ffffc90000a87868 R15: ffff88815b882ae0 FS: 0000000000000000(0000) GS:ffff8884ec840000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f930b844000 CR3: 000000000262e003 CR4: 0000000008f70ef0 PKRU: 55555554 Call Trace: <TASK> __change_page_attr_set_clr+0x989/0xe90 ? __purge_vmap_area_lazy+0x6c/0x3a0 ? _vm_unmap_aliases+0x250/0x2a0 set_pages_array_wb+0x7f/0x120 ttm_pool_backup+0x4c9/0x5b0 [ttm] ? dma_resv_wait_timeout+0x3b/0xf0 ttm_tt_backup+0x32/0x60 [ttm] ttm_bo_shrink+0x66/0x110 [ttm] xe_bo_shrink_purge+0x12b/0x1b0 [xe] xe_bo_shrink+0xbb/0x270 [xe] __xe_shrinker_walk+0xf7/0x160 [xe] xe_shrinker_walk+0x9d/0xc0 [xe] xe_shrinker_scan+0x11f/0x210 [xe] do_shrink_slab+0x13b/0x270 shrink_slab+0xf1/0x400 shrink_node+0x352/0x8a0 balance_pgdat+0x32c/0x700 kswapd+0x205/0x2f0 ? __pfx_autoremove_wake_function+0x10/0x10 ? __pfx_kswapd+0x10/0x10 kthread+0xd1/0x110 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x1b1/0x200 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30 </TASK> | ||||
| CVE-2026-68257 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 7.0 High |
| In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: fix 32-bit overflow in CWSR total size calculation total_cwsr_size was computed in 32-bit before being used as a BO/SVM allocation size. With large ctx_save_restore_area_size and debug_memory_size multiplied by the XCC count, the product can wrap, yielding an undersized CWSR save area that firmware later overruns. Promote total_cwsr_size to u64 and use check_add_overflow()/ check_mul_overflow() in both kfd_queue_acquire_buffers() and kfd_queue_release_buffers(). (cherry picked from commit 319f7e13423ae3f486b9aea82f9ad2d6af0ee608) | ||||
| CVE-2026-68259 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: Check bounds in allocate_event_notification_slot The valid event ids go from 0 to KFD_SIGNAL_EVENT_LIMIT allocate_event_notification_slot has an option to specify an event id to allocate at, used by CRIU. We weren't checking the bounds on that value. Check them. v2: Lower bounds check is unecessary because of idr_alloc already rejecting negative numbers. Upper bounds check should be KFD_SIGNAL_EVENT_LIMIT since the signal mode mappings might not yet exist (cherry picked from commit 6853f1f6cbbeb3f53ebbbd7286536aeb2c5d5f50) | ||||
| CVE-2026-68157 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.3 Medium |
| In the Linux kernel, the following vulnerability has been resolved: libceph: guard missing CRUSH type name lookup Localized read selection can walk a parent bucket whose name exists in the CRUSH map while its type has no matching entry in type_names. get_immediate_parent() then dereferences a NULL type_cn and passes an invalid pointer into strcmp(), causing a null-ptr-deref. Skip such malformed parent buckets unless both the bucket name and type name metadata are present. This keeps malformed hierarchy data from crashing locality lookup and safely falls back to "not local". [ idryomov: add WARN_ON_ONCE ] | ||||
| CVE-2026-68161 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 4.1 Medium |
| In the Linux kernel, the following vulnerability has been resolved: sctp: close UDP tunnel sockets during netns teardown proc_sctp_do_udp_port() starts per-net SCTP UDP tunneling sockets when net.sctp.udp_port is set, and stops/restarts them when the sysctl value changes. The netns exit path does not stop these sockets, so a namespace can be torn down while its SCTP UDP tunnel sockets are still installed. Close the UDP tunnel sockets from sctp_ctrlsock_exit() after unregistering the per-net sysctl table. This prevents new sysctl writes from racing in while the sockets are being released, and closes the sockets before the control socket is destroyed. | ||||
| CVE-2026-68182 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 4.4 Medium |
| In the Linux kernel, the following vulnerability has been resolved: comedi: comedi_parport: deal with premature interrupt Syzbot reported a general protection fault in `comedi_get_is_subdevice_running()`, which was called from the interrupt handler `parport_interrupt()` in the "comedi_parport" driver, but it does not currently have a C reproducer for the problem. It's probably due to a premature interrupt for one of two reasons: 1. The driver sets up the interrupt handler before the comedi subdevices used by the interrupt handler have been allocated, but does not disable the interrupt in the parallel port's CTRL register first. 2. The driver uses a user-supplied I/O port base address which Syzbot would have supplied, but it might not be backed by real parallel port hardware. Change the initialization order in the driver's comedi "attach" handler (`parport_attach()`) so that the hardware registers are initialized before the interrupt handler is requested. This should prevent premature interrupts occurring for real hardware. Also add a test to the interrupt handler to ensure the comedi device is fully attached and return early if it isn't. | ||||
| CVE-2026-68192 | 1 Linux | 1 Linux Kernel | 2026-08-10 | N/A |
| In the Linux kernel, the following vulnerability has been resolved: wifi: brcmfmac: make release_scratchbuffers idempotent brcmf_pcie_release_scratchbuffers() frees the shared.scratch and shared.ringupd DMA buffers with dma_free_coherent() but does not clear the pointers afterwards, unlike the sibling release_ringbuffers() which NULLs commonrings/flowrings/idxbuf on release. Both the bus_reset .reset callback (brcmf_pcie_reset) and brcmf_pcie_remove() call release_scratchbuffers. When reset teardown has run before removal, remove's own teardown would call dma_free_coherent() a second time on the already-freed DMA allocation. NULL the pointers after free, matching release_ringbuffers(), so a later release observes that the allocation has already been released. This patch makes repeated sequential release safe; the reset-work lifetime is handled separately by the following patch. This issue was found by an in-house static analysis tool. | ||||
| CVE-2026-68248 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/i915: Return NULL on error in active_instance Avoid returning &node->base when node is NULL due to OOM during GFP_ATOMIC allocation. Discovered using AI-assisted static analysis confirmed by Intel Product Security. (cherry picked from commit 6029bc064f0b1bac184203a50fbaaf070fa18832) | ||||
| CVE-2026-68255 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 6.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: drm/virtio: bound EDID block reads to the response buffer virtio_get_edid_block() validates the read offset only against the device-supplied resp->size field, never against the fixed-size resp->edid array. The EDID block index is driven by the device-supplied extension count, so a malicious virtio-gpu backend can advertise a large size together with a high block count and read far past the array into adjacent kernel memory, which is then surfaced in the parsed EDID (an out-of-bounds read / info leak). Also reject any read whose end exceeds the size of the edid array. Conforming EDID responses stay within the array and are unaffected. | ||||
| CVE-2026-68130 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 6.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: ksmbd: defer destroy_previous_session() until after NTLM authentication In ntlm_authenticate(), destroy_previous_session() is called using a user pointer resolved from the client-supplied NTLM blob username field before the NTLMv2 response is validated. An authenticated attacker can set the NTLM blob username to match a victim account and set PreviousSessionId to the victim's session ID; destroy_previous_session() destroys the victim's session while ksmbd_decode_ntlmssp_auth_blob() subsequently rejects the request with -EPERM. Move destroy_previous_session() and the prev_id assignment to after ksmbd_decode_ntlmssp_auth_blob() returns success and use sess->user rather than the pre-authentication lookup result. This matches the ordering already used by krb5_authenticate(), where destroy_previous_session() is called only after ksmbd_krb5_authenticate() returns success. | ||||
| CVE-2026-68131 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.3 Medium |
| In the Linux kernel, the following vulnerability has been resolved: rbd: Reset positive result codes to zero in object map update path In a reply message to an RBD request, a positive result code indicates a data payload, which is not allowed for writes. While rbd_osd_req_callback() already resets a positive result code for writes to zero, rbd_object_map_callback() does not. This allows a corrupted reply to an object map update to trigger the rbd_assert(*result < 0) in __rbd_obj_handle_request(). This happens, because rbd_object_map_callback() calls rbd_obj_handle_request() -> __rbd_obj_handle_request() and passes this positive result code. From __rbd_obj_handle_request(), rbd_obj_advance_write() is called, which leaves the positive result code unchanged and returns true. Therefore, the if(done && *result) branch is executed in __rbd_obj_handle_request() and the assertion triggers. This patch fixes the issue by adjusting the logic in the rbd_object_map_callback() path. A positive result code for an object map update is now reset to zero (similar to rbd_osd_req_callback()), and the message is subsequently handled the same way as if the result code was zero from the beginning. Additionally, a WARN_ON_ONCE() is added for this case. | ||||
| CVE-2026-68132 | 1 Linux | 1 Linux Kernel | 2026-08-10 | 5.5 Medium |
| In the Linux kernel, the following vulnerability has been resolved: super: fix emergency thaw deadlock on frozen block devices do_thaw_all_callback() calls bdev_thaw() while holding sb->s_umount exclusively. If the block device was frozen via bdev_freeze() dropping the last block layer freeze reference calls fs_bdev_thaw() which reacquires s_umount: do_thaw_all_callback(sb) super_lock_excl(sb) # holds sb->s_umount bdev_thaw(sb->s_bdev) mutex_lock(&bdev->bd_fsfreeze_mutex) # bd_fsfreeze_count drops 1 -> 0 bd_holder_ops->thaw == fs_bdev_thaw get_bdev_super(bdev) bdev_super_lock(bdev, true) super_lock(sb, true) down_write(&sb->s_umount) # same task: deadlock The emergency thaw worker deadlocks against itself holding both s_umount and bd_fsfreeze_mutex. That fscks any subsequent unmount, freeze, or thaw of that filesystem and block device. [ 81.878470] sysrq: Show Blocked State [ 81.880140] task:kworker/0:1 state:D stack:0 pid:11 tgid:11 ppid:2 task_flags:0x4208060 flags:0x00080000 [ 81.884876] Workqueue: events do_thaw_all [ 81.886656] Call Trace: [ 81.887759] <TASK> [ 81.888763] __schedule+0x579/0x1420 [ 81.890372] schedule+0x3a/0x100 [ 81.891794] schedule_preempt_disabled+0x15/0x30 [ 81.893848] rwsem_down_write_slowpath+0x1ea/0x900 [ 81.895191] ? __pfx_do_thaw_all_callback+0x10/0x10 [ 81.896528] down_write+0xbd/0xc0 [ 81.897505] super_lock+0x91/0x180 [ 81.898457] ? __mutex_lock+0xa99/0x1140 [ 81.900748] ? __mutex_unlock_slowpath+0x1f/0x400 [ 81.902069] bdev_super_lock+0x5b/0x150 [ 81.903132] get_bdev_super+0x10/0x60 [ 81.904042] fs_bdev_thaw+0x23/0xf0 [ 81.904755] bdev_thaw+0x82/0x100 [ 81.905484] do_thaw_all_callback+0x2c/0x50 [ 81.906298] __iterate_supers+0x5d/0x130 [ 81.907067] do_thaw_all+0x20/0x40 [ 81.907739] process_one_work+0x206/0x5e0 [ 81.908545] worker_thread+0x1e2/0x3c0 [ 81.909339] ? __pfx_worker_thread+0x10/0x10 [ 81.910171] kthread+0xf4/0x130 [ 81.910799] ? __pfx_kthread+0x10/0x10 [ 81.911528] ret_from_fork+0x2e2/0x3b0 [ 81.912259] ? __pfx_kthread+0x10/0x10 [ 81.913010] ret_from_fork_asm+0x1a/0x30 [ 81.913806] </TASK> bdev_super_lock() even documents the violated requirement with lockdep_assert_not_held(&sb->s_umount). Acquiring bd_fsfreeze_mutex under s_umount also inverts the bd_fsfreeze_mutex vs. s_umount ordering established by bdev_{freeze,thaw}() and can thus ABBA against a concurrent block-layer freeze even when the recursive path isn't hit. Fix this by not holding s_umount around the bdev_thaw() loop at all. Pin the superblock with an active reference instead as filesystems_freeze_callback() does. The active reference keeps the superblock from being shut down and so ->s_bdev stays valid without holding s_umount. The block-layer-held freeze is dropped by fs_bdev_thaw() with FREEZE_MAY_NEST | FREEZE_HOLDER_USERSPACE exactly as a regular unfreeze would and thaw_super_locked() handles filesystem-level freezes as before. The emergency thaw path has deadlocked like this in one form or another for a long long time but the current exclusively-held shape dates back to commit [1] where thaw_bdev() already ended in thaw_super() with s_umount held by do_thaw_all_callback(). | ||||