CVE Database

In the Linux kernel, the following vulnerability has been resolved: tpm: tpm_tis_spi: Use wait_woken() in wait_for_tmp_stat() wait_event_interruptible_timeout() evaluates its condition after setting the current task state to TASK_INTERRUPTIBLE. With CONFIG_DEBUG_ATOMIC_SLEEP this triggers a warning when the IRQ wait path is used: tpm_tis_status() tpm_tis_spi_read_bytes() tpm_tis_spi_transfer_full() spi_bus_lock() mutex_lock() Address this with the following measures: 1. Call wait_tpm_stat_cond() only while tasking is running. 2. Use wait_woken() to wait for changes.

In the Linux kernel, the following vulnerability has been resolved: tpm: tpm2-sessions: wait for async KPP completion in tpm_buf_append_salt tpm_buf_append_salt() in drivers/char/tpm/tpm2-sessions.c calls crypto_kpp_generate_public_key() and crypto_kpp_compute_shared_secret() without installing a completion callback, discards both return values, and immediately frees the kpp_request via kpp_request_free(). When the resolved ecdh-nist-p256 KPP backend is asynchronous (atmel-ecc, HPRE, keembay-ocs), either operation returns -EINPROGRESS and the deferred completion worker dereferences the freed request. The path fires automatically from the hwrng_fillfn kernel thread via tpm_get_random -> tpm2_get_random -> tpm2_start_auth_session -> tpm_buf_append_salt on every entropy poll, without any userland action. Install crypto_req_done as the completion callback, wrap both KPP operations in crypto_wait_req(), and propagate errors to the caller. The wait is a no-op for synchronous backends.

In the Linux kernel, the following vulnerability has been resolved: sunrpc: fix uninitialized xprt_create_args structure The xprt_create_args structure is allocated on the stack without initialization in rpc_sysfs_xprt_switch_add_xprt_store(). While some fields are manually populated, critical fields like srcaddr, bc_xps, and flags contain uninitialized stack garbage. This can lead to: 1. Kernel panic when xs_setup_xprt() dereferences garbage srcaddr 2. Information leak if srcaddr points to sensitive stack data 3. Unpredictable behavior if flags has random bits set The fix is to zero-initialize the structure to ensure all unused fields are NULL/0, preventing the transport setup code from acting on garbage data.

In the Linux kernel, the following vulnerability has been resolved: dmaengine: tegra: Fix burst size calculation Currently, the Tegra GPC DMA hardware requires the transfer length to be a multiple of the max burst size configured for the channel. When a client requests a transfer where the length is not evenly divisible by the configured max burst size, the DMA hangs with partial burst at the end. Fix this by reducing the burst size to the largest power-of-2 value that evenly divides the transfer length. For example, a 40-byte transfer with a 16-byte max burst will now use an 8-byte burst (40 / 8 = 5 complete bursts) instead of causing a hang. This issue was observed with the PL011 UART driver where TX DMA transfers of arbitrary lengths were stuck.

In the Linux kernel, the following vulnerability has been resolved: dmaengine: dw-edma: Add spinlock to protect DONE_INT_MASK and ABORT_INT_MASK The DONE_INT_MASK and ABORT_INT_MASK registers are shared by all DMA channels, and modifying them requires a read-modify-write sequence. Because this operation is not atomic, concurrent calls to dw_edma_v0_core_start() can introduce race conditions if two channels update these registers simultaneously. Add a spinlock to serialize access to these registers and prevent race conditions. [den: update dw_edma.lock comment]

In the Linux kernel, the following vulnerability has been resolved: dmaengine: dw-edma-pcie: Reject devices without driver data dw_edma_pcie_probe() treats the PCI device ID driver_data as the template for the controller layout and copies it unconditionally. A device bound dynamically via sysfs can match the driver without that data, which leads to a NULL pointer dereference. Reject such matches before enabling the device.

In the Linux kernel, the following vulnerability has been resolved: dmaengine: sh: rz-dmac: Move interrupt request after everything is set up Once the interrupt is requested, the interrupt handler may run immediately. Since the IRQ handler can access channel->ch_base, which is initialized only after requesting the IRQ, this may lead to invalid memory access. Likewise, the IRQ thread may access uninitialized data (the ld_free, ld_queue, and ld_active lists), which may also lead to issues. Request the interrupts only after everything is set up. To keep the error path simpler, use dmam_alloc_coherent() instead of dma_alloc_coherent().

In the Linux kernel, the following vulnerability has been resolved: platform/x86/intel/tpmi: use cleanup helpers in mem_write() In mem_write(), the temporary array returned by parse_int_array_user() must be released on all exit paths. Convert the array variable to use cleanup.h scope-based cleanup so it is freed automatically on return. This also moves the array declaration next to parse_int_array_user() as required by cleanup.h usage guidelines.

In the Linux kernel, the following vulnerability has been resolved: platform/x86: dell-laptop: fix missing cleanups in init error path dell_init() initializes several resources after dell_setup_rfkill(), including the optional touchpad LED, keyboard backlight LED, battery hook, debugfs directory and dell-laptop notifier. If a later LED or backlight registration fails, the error path only tears down the battery hook and rfkill resources. This leaves the notifier, debugfs directory, keyboard backlight LED and optional touchpad LED registered after dell_init() returns an error. Add the missing cleanup calls before tearing down rfkill.

In the Linux kernel, the following vulnerability has been resolved: platform/x86: ISST: Restore SST-PP control to all domains The SST-PP control offset is only restored to power domain 0 after resume. During suspend, control values are read and stored for all power domains. Use pd_info->sst_base instead of power_domain_info->sst_base, which only points to power domain 0 base address.

In the Linux kernel, the following vulnerability has been resolved: i2c: imx: fix locked bus on SMBus block-read of 0 (atomic) SMBus 3.1 6.5.7 allows a Block Read byte count of 0, but the atomic (polling) path rejects it as -EPROTO. Worse, it returns without a NACK+STOP: the next receive cycle has already started, so the target keeps holding SDA and the bus stays stuck until a power cycle for this i2c controller. Reading I2DR to obtain the count likewise arms the next byte on the count > I2C_SMBUS_BLOCK_MAX path, which also returned -EPROTO directly and left the bus held. Handle both: NACK the in-flight dummy byte (TXAK) and extend msgs->len so the existing last-byte handling emits STOP; the dummy byte is discarded. A count of 0 is a valid empty block read; a count above I2C_SMBUS_BLOCK_MAX is still reported as -EPROTO, but only after the bus has been released. The interrupt-driven path has the same flaw from a later commit and is fixed separately, as it carries a different Fixes: tag and stable range.

In the Linux kernel, the following vulnerability has been resolved: i2c: imx: fix locked bus on SMBus block-read of 0 (IRQ) SMBus 3.1 6.5.7 allows a Block Read byte count of 0, but the interrupt-driven block-read state machine rejects it as -EPROTO. Worse, it returns without a NACK+STOP: the next receive cycle has already started, so the target keeps holding SDA and the bus stays stuck until a power cycle of this i2c controller. Accept count=0: NACK the in-flight dummy byte (TXAK) and set msg->len to 2 so i2c_imx_isr_read_continue() emits STOP via its normal last-byte path. The dummy byte is discarded; block-read callers only consume buf[0..count-1]. Reading I2DR has likewise already armed the next byte on the count > I2C_SMBUS_BLOCK_MAX error path, so NACK it (TXAK) before aborting with -EPROTO; otherwise the failing transfer's STOP cannot complete and the bus stays held. The atomic path regressed earlier (v3.16) and is fixed separately; this patch covers only the v6.13 state-machine rework.

In the Linux kernel, the following vulnerability has been resolved: i2c: mlxbf: Fix use-after-free in mlxbf_i2c_init_resource() If devm_platform_get_and_ioremap_resource() returns an error, mlxbf_i2c_init_resource() frees tmp_res before reading tmp_res->io to get the error code. This results in a use-after-free. Save the error code before freeing tmp_res.

In the Linux kernel, the following vulnerability has been resolved: tcp: defer md5sig_info kfree past RCU grace period in tcp_connect The md5+ao reconciliation in tcp_connect() (net/ipv4/tcp_output.c) has two symmetric branches: if (needs_md5) { tcp_ao_destroy_sock(sk, false); } else if (needs_ao) { tcp_clear_md5_list(sk); kfree(rcu_replace_pointer(tp->md5sig_info, NULL, ...)); } Both branches free a per-socket auth-info object while the socket is in TCP_SYN_SENT and is already on the inet ehash (inserted by inet_hash_connect() in tcp_v4_connect()). Both branches are reachable by softirq RX-path readers that load the corresponding info pointer via implicit RCU before bh_lock_sock_nested() is taken. The needs_md5 branch is fixed in the prior patch by re-introducing the call_rcu() free in tcp_ao_destroy_sock(): the equivalent per-key loop runs inside tcp_ao_info_free_rcu(), the RCU callback, so by the time it frees each tcp_ao_key all softirq readers that captured the container have already completed rcu_read_unlock(). The needs_ao branch is not symmetric in the same way. The container free can be deferred via kfree_rcu(md5sig, rcu) -- struct tcp_md5sig_info already has the required rcu member (include/net/tcp.h:1999-2002), and the rest of the tree already does this in the tcp_md5sig_info_add() rollback paths (net/ipv4/tcp_ipv4.c:1410, 1436). But the per-key teardown is done by tcp_clear_md5_list() in process context BEFORE the container's RCU grace period: it walks &md5sig->head and frees each tcp_md5sig_key with bare hlist_del + kfree. A concurrent softirq reader in __tcp_md5_do_lookup() / __tcp_md5_do_lookup_exact() (tcp_ipv4.c:1253, 1298) walks the same list via hlist_for_each_entry_rcu() and races with that bare kfree on the keys themselves -- a per-key slab use-after-free of the same class as the TCP-AO bug, on the same race window. Fix this in two halves: 1. Convert the bare kfree() in tcp_connect() to kfree_rcu() so the md5sig_info container joins the rest of the md5sig lifecycle. The local-variable lift is mechanical and required because kfree_rcu() is a macro that expects an lvalue. 2. Make tcp_clear_md5_list() RCU-safe by replacing hlist_del + kfree(key) with hlist_del_rcu + kfree_rcu(key, rcu). struct tcp_md5sig_key already carries the rcu member (include/net/tcp.h:1995) and tcp_md5_do_del() (net/ipv4/tcp_ipv4.c:1456) already uses kfree_rcu, so this restores the lifecycle invariant the rest of the file follows rather than introducing a one-off. The other caller of tcp_clear_md5_list() is tcp_md5_destruct_sock() (net/ipv4/tcp.c:412), which runs from the sock destructor when the socket is already unhashed and unreachable; the extra grace period there is unnecessary but harmless. Making the helper unconditionally RCU-safe is the cleaner contract. The needs_ao branch is not reachable by the userns reproducer used to demonstrate the AO-side splat (the repro installs both keys but ends up in the needs_md5 branch because the connect peer matches the MD5 key, not the AO key); however the symmetric race exists and a maintainer touching this code should not have to think about which branch escapes RCU and which one does not. [also credits to Qihang, who found that this races with tcp-diag]

In the Linux kernel, the following vulnerability has been resolved: xen/gntdev: fix error handling in ioctl When gntdev_ioctl_map_grant_ref() fails to copy the operation result back to userspace after successfully adding the mapping to the list, the error path returns -EFAULT without releasing the reference acquired by gntdev_alloc_map(). The mapping remains in priv->maps with a refcount of 1, causing a memory leak and a dangling list entry. Additionally, gntdev_add_map() may modify map->index to avoid overlap with existing mappings. Therefore, the index returned to userspace must be obtained after gntdev_add_map() completes. Fix this by holding the mutex across gntdev_add_map(), retrieving the correct index, and copy_to_user(). If copy_to_user() fails, remove the mapping from the list and release the reference while still holding the lock. Fix these issues by properly handling all error cases.

In the Linux kernel, the following vulnerability has been resolved: xfrm: nat_keepalive: avoid double free on send error nat_keepalive_send() frees the keepalive skb whenever the IPv4 or IPv6 send helper reports an error. That cleanup is only correct before the skb is handed to the output path. Once ip_build_and_send_pkt() or ip6_xmit() takes ownership, the networking stack may already have consumed the skb before returning an error, so freeing it again is unsafe. Handle the pre-handoff failure cases inside nat_keepalive_send_ipv4() and nat_keepalive_send_ipv6(), where the caller still owns the skb, and keep nat_keepalive_send() responsible only for family dispatch and the unsupported-family cleanup path.

In the Linux kernel, the following vulnerability has been resolved: xfrm: xfrm_interface: require CAP_NET_ADMIN in the device netns for changelink xfrmi_changelink() operates on at most two netns, dev_net(dev) and the interface link netns xi->net. They differ once the device is created in or moved to a netns other than the one the request runs in. The rtnl changelink path checks CAP_NET_ADMIN only against dev_net(dev), so a caller privileged there but not in xi->net can rewrite an interface that lives in xi->net. Gate xfrmi_changelink() on rtnl_dev_link_net_capable() at its top, before any attribute is parsed.

In the Linux kernel, the following vulnerability has been resolved: tpm: Make the TPM character devices non-seekable The TPM character devices expose a sequential command/response interface, but their open handlers leave FMODE_PREAD and FMODE_PWRITE enabled. After a command leaves a response pending, pread(fd, buf, 16, 0x1400) passes 0x1400 as *off to tpm_common_read(). The transfer length is bounded by response_length, but the offset is used unchecked when forming data_buffer + *off. A sufficiently large offset therefore causes an out-of-bounds heap read through copy_to_user() and, if the copy succeeds, an out-of-bounds zero-write through the following memset(). Positional I/O does not provide coherent semantics for this interface. An arbitrary pread offset cannot represent how much of a response has been consumed sequentially. The write callback always stores a command at the start of data_buffer, while pwrite() does not update file->f_pos and can leave the sequential read cursor stale. Call nonseekable_open() from both open handlers. This removes FMODE_PREAD and FMODE_PWRITE, causing positional reads and writes to fail with -ESPIPE before reaching the TPM callbacks, and explicitly marks the files non-seekable. Normal read() and write() continue to use the existing sequential f_pos cursor, leaving the response state machine unchanged. Tested on Linux 6.12 with KASAN and a swtpm TPM2 device: - sequential partial reads returned the complete response - pread() and preadv() with offset 0x1400 returned -ESPIPE - pwrite() and pwritev() with offset zero returned -ESPIPE - the pending response remained intact after the rejected operations - a subsequent normal command/response cycle completed normally - no KASAN report was produced.

In the Linux kernel, the following vulnerability has been resolved: spi: imx: reconfigure for PIO when DMA cannot be started When spi_imx_can_dma() selects DMA, the ECSPI is configured for DMA: spi_imx_setupxfer() sets CTRL.SMC and clears dynamic_burst, and spi_imx_dma_transfer() programs the dynamic-burst BURST_LENGTH and the SDMA watermarks. If the DMA descriptor cannot be prepared (dmaengine_prep_slave_single() returns NULL), the transfer is failed with SPI_TRANS_FAIL_NO_START and falls back to PIO. The dynamic-burst DMA path uses its own bounce buffers instead of the SPI core's mapping, so xfer->{tx,rx}_sg_mapped are not set and the core's DMA->PIO retry is skipped; the driver falls back to PIO internally. But none of the DMA-mode configuration is undone, so the PIO transfer runs with CTRL.SMC set, the wrong burst length and dynamic_burst cleared, and the transferred data is corrupted. This is easily hit on i.MX8MP boards that describe ECSPI DMA in the device tree but run SDMA on ROM firmware (no external sdma-imx7d.bin): every ECSPI DMA prepare fails. An Infineon SLB9670 TPM on ECSPI1 then returns shifted TPM2_GetCapability data, is flagged "field failure mode", /dev/tpmrm0 is never created. Set controller->fallback before re-running spi_imx_setupxfer() so the ECSPI is reconfigured exactly like a normal PIO transfer. With controller->fallback set, spi_imx_setupxfer() sees spi_imx_can_dma() return false, so it clears spi_imx->usedma and reprograms the controller (clears CTRL.SMC, restores dynamic_burst and the PIO burst length). No explicit spi_imx->usedma = false is needed: setupxfer() already updates it from the can_dma() result.

In the Linux kernel, the following vulnerability has been resolved: spi: uniphier: Fix completion initialization order before devm_request_irq() The driver calls devm_request_irq() before initializing the completion used by the interrupt handler. Because the interrupt may occur immediately after devm_request_irq(), the handler may execute before init_completion(). This may result in calling complete() on an uninitialized completion, causing undefined behavior. This has been observed with KASAN. Fix this by initializing the completion before registering the IRQ.

In the Linux kernel, the following vulnerability has been resolved: NFS: Charge unstable writes by request size, not folio size nfs_folio_mark_unstable() and nfs_folio_clear_commit() charge and uncharge NR_WRITEBACK/WB_WRITEBACK by folio_nr_pages(folio) once per *request* added to or removed from a commit list. This is correct only when a folio has a single associated request. When pg_test splits a folio into N sub-folio requests (e.g. pNFS flexfiles striping with a stripe unit smaller than the folio size, or plain wsize-limited splitting), each of the N requests independently charges the whole folio's page count, inflating the accounting by a factor of N per folio. With large folios and small stripe units this reaches multiple orders of magnitude: a 2 MiB folio split into 512 4 KiB requests can charge up to 512x its real size, pushing global dirty+writeback accounting past the system's dirty threshold and forcing every buffered writer on the host into the hard-throttle path, including unrelated in-kernel NFS server threads sharing the box. Charge each request only for the pages it actually covers.

In the Linux kernel, the following vulnerability has been resolved: nvme-apple: Prevent shared tags across queues on Apple A11 On Apple A11, tags of pending commands must be unique across the admin and IO queues, else the firmware crashes with "duplicate tag error for tag N", with N being the tag. Apply the existing workaround for M1 of reserving two tags for the admin queue to A11.

In the Linux kernel, the following vulnerability has been resolved: nvmet-auth: reject short AUTH_RECEIVE buffers nvmet_execute_auth_receive() trusts the AUTH_RECEIVE allocation length after checking only that it is nonzero and matches the transfer length. In the SUCCESS1 and FAILURE1/default states, that lets a remote NVMe-oF initiator reach the fixed-size DH-HMAC-CHAP response builders with a kmalloc() buffer shorter than the response, so nvmet_auth_success1() and nvmet_auth_failure1() write past the allocation; both only WARN_ON the short length and then format the message anyway. Impact: A remote NVMe-oF initiator with access to an auth-enabled target can trigger a 16-byte heap out-of-bounds write via a one-byte AUTH_RECEIVE allocation length. Compute the minimum response length for the current DH-HMAC-CHAP step in nvmet_auth_receive_data_len() and report a zero data length when the host-supplied allocation length is shorter, so the existing zero-length check in nvmet_execute_auth_receive() rejects the command before any builder runs. The SUCCESS1 minimum is sizeof(struct nvmf_auth_dhchap_success1_data) plus the HMAC hash length, because the response hash is written into the rval[] flexible-array tail, so the minimum is state dependent rather than a flat sizeof. CHALLENGE keeps its existing variable-length guard in nvmet_auth_challenge(). This is reachable only when in-band DH-HMAC-CHAP authentication is configured on the target.

In the Linux kernel, the following vulnerability has been resolved: nvmet-rdma: handle inline data with a nonzero offset nvmet_rdma_use_inline_sg() maps the host-controlled inline data offset into the per-command inline scatterlist. The bounds check admits any offset with off + len <= inline_data_size, but the mapping still assumes the data begins in the first inline page: sg->offset = off; sg->length = min_t(int, len, PAGE_SIZE - off); When a port is configured with inline_data_size > PAGE_SIZE (settable up to max(SZ_16K, PAGE_SIZE)), an offset in (PAGE_SIZE, inline_data_size] makes "PAGE_SIZE - off" underflow, so sg->length is set to ~4 GiB and the block backend reads far past the first inline page. num_pages(len) also ignores the offset, so an in-bounds offset whose [off, off+len) span crosses a page boundary under-counts the scatterlist. Map the offset properly: split it into a page index and an in-page offset, start the scatterlist at that page, and size the page count from page_off + len. Because the request scatterlist may now start at inline_sg[page_idx] rather than inline_sg[0], generalize the inline-SGL identity test in nvmet_rdma_release_rsp() to a range test; otherwise the persistent inline scatterlist is mistaken for an allocated one and nvmet_req_free_sgls() frees an inline page (and warns in free_large_kmalloc()).

In the Linux kernel, the following vulnerability has been resolved: nvmet: fix refcount leak in nvmet_sq_create() In nvmet_sq_create(), a reference on the ctrl is taken via kref_get_unless_zero() before calling nvmet_check_sqid(). If nvmet_check_sqid() fails, the function returns the error directly without releasing the reference, leading to a leak. Fix this by jumping to the "ctrl_put" label, which already performs the necessary nvmet_ctrl_put(ctrl). This ensures the reference is properly released on this error path.

In the Linux kernel, the following vulnerability has been resolved: netdev-genl: report NAPI thread PID in the caller's pid namespace netdev_nl_napi_fill_one() reports the NAPI kthread PID in NETDEV_A_NAPI_PID using task_pid_nr(), which returns the PID in the initial pid namespace. NETDEV_CMD_NAPI_GET does not have GENL_ADMIN_PERM and the netdev genl family is netnsok, so a caller in a child pid namespace can issue it. That caller then sees the kthread's global PID, even though the kthread is not visible in its pid namespace, where the value should be 0. Translate the PID through the caller's pid namespace, the same way commit 3799c2570982 ("io_uring/fdinfo: translate SqThread PID through caller's pid_ns") did for the io_uring SQPOLL thread. The doit and dumpit paths both run synchronously in the caller's context, so task_active_pid_ns(current) is the caller's pid namespace.

In the Linux kernel, the following vulnerability has been resolved: can: isotp: use unconditional synchronize_rcu() in isotp_release() isotp_notify() unregisters the (RCU) CAN filters via can_rx_unregister() and clears so->bound without waiting for a grace period. isotp_release() uses so->bound to decide whether it needs to call synchronize_rcu() before cancelling so->rxtimer, so when NETDEV_UNREGISTER runs first it skips that synchronize_rcu() and can cancel the timer while an in-flight isotp_rcv() is still executing and about to re-arm it via isotp_send_fc(), leading to a use-after-free timer callback on the freed socket. sakisho-bot remarked a problem with rtnl_lock held in isotp_notify(), therefore make isotp_release() always call synchronize_rcu() before cancelling the timers, regardless of so->bound. This still closes the original race (isotp_notify() clearing so->bound without waiting for in-flight isotp_rcv() callers before isotp_release() cancels the RX timer) without adding any RCU wait to the netdevice notifier path.

In the Linux kernel, the following vulnerability has been resolved: can: isotp: fix use-after-free race with concurrent NETDEV_UNREGISTER isotp_release() looked up the bound network device via dev_get_by_index() using the stored ifindex. During device unregistration the device is unlisted from the ifindex hash before the NETDEV_UNREGISTER notifier chain runs, so a concurrent isotp_release() could find no device, skip can_rx_unregister() entirely, and still proceed to free the socket. Since isotp_release() had already removed itself from the isotp notifier list at that point, isotp_notify() would never get a chance to clean up either, leaving a stale CAN filter that keeps pointing at the freed socket. Fix this the same way raw.c already does: hold a tracked reference to the bound net_device in the socket (so->dev/so->dev_tracker) from bind() onward instead of re-resolving it from the ifindex, and serialize bind()/release() with rtnl_lock() so that so->dev is always consistent with what the NETDEV_UNREGISTER notifier sees. so->dev stays valid regardless of ifindex-hash unlisting, and is only ever cleared by whichever of isotp_release()/isotp_notify() gets there first, so the filter is always removed exactly once. isotp_bind() now rejects a (re)bind with -EAGAIN while so->[tx|rx].state isn't ISOTP_IDLE yet, so a timer left running by a prior NETDEV_UNREGISTER can't act on a newly bound so->ifindex. Both checks share the same lock_sock() section, so there is no window in which a concurrent isotp_notify() clearing so->bound could be missed.

In the Linux kernel, the following vulnerability has been resolved: can: isotp: serialize TX state transitions under so->rx_lock The TX state machine (so->tx.state) is driven from three contexts: sendmsg() claiming and progressing a transfer, the RX path consuming Flow Control/echo frames, and two hrtimers timing out a stalled transfer. Mixing a lock-free cmpxchg() claim in sendmsg() with hrtimer_cancel() calls made under so->rx_lock elsewhere left windows where a frame or timer callback could act on a state that had already moved on, corrupting an unrelated transfer. so->rx_lock now covers the full lifecycle of a TX claim: sendmsg() takes it to check so->tx.state is ISOTP_IDLE, switch it to ISOTP_SENDING, bump so->tx_gen and drain the previous transfer's timers - all as one critical section. isotp_rcv_fc()/isotp_rcv_cf() already run under this lock via isotp_rcv(), and isotp_rcv_echo() now takes it itself, so none of them can ever observe a transfer mid-claim. This also means a transfer can no longer be handed to sendmsg()'s cleanup paths (signal or send error) while another thread is concurrently claiming or finishing it, so those paths can cancel timers and reset the state unconditionally. isotp_release() claims the socket the same way, so a racing sendmsg() sees a consistent ISOTP_SHUTDOWN and skips arming its timer or sending. Only the hrtimer callbacks stay outside so->rx_lock, since they run under so->rx_lock's cancellation elsewhere and taking it themselves would deadlock. so->tx_gen lets them recognize whether the transfer they timed out is still the one currently active, so they don't report an error against a transfer that has since completed or been superseded.

In the Linux kernel, the following vulnerability has been resolved: can: bcm: defer rx_op deallocation to workqueue to fix thrtimer UAF Commit f1b4e32aca08 ("can: bcm: use call_rcu() instead of costly synchronize_rcu()") replaced synchronize_rcu() in bcm_delete_rx_op() with call_rcu() and introduced the RX_NO_AUTOTIMER flag. However, this flag check was omitted for thrtimer in the packet rx fast-path. During BCM RX operation teardown, a concurrent RCU reader (bcm_rx_handler) can race and re-arm thrtimer via bcm_rx_update_and_send() after call_rcu() has been scheduled. Once the RCU grace period elapses, bcm_op is freed. The subsequently firing thrtimer then dereferences the deallocated op, causing a UAF. Adding flag checks to the rx fast-path (bcm_rx_update_and_send) does not fully close the TOCTOU race and introduces latency for every CAN frame. Conversely, calling hrtimer_cancel() directly inside the RCU callback (softirq context) is fatal as hrtimer_cancel() can sleep, triggering a "scheduling while atomic" panic. Resolve this by deferring the timer cancellation and memory free to a dedicated unbound workqueue (bcm_wq). The RCU callback now queues a work item to bcm_wq, which safely cancels both timers and deallocates memory in sleepable process context. A dedicated workqueue is used to prevent system-wide WQ saturation and is cleanly flushed/destroyed on module unload to avoid rmmod page faults. Since the deferred work can now outlive the calling context by an unbounded amount, also take a reference on op->sk when it is assigned and drop it only once the deferred work has cancelled both timers, so a socket can no longer be freed out from under a still-armed timer whose callback (bcm_send_to_user()) dereferences op->sk.