Total
18636 CVE
| CVE | Vendors | Products | Updated | CVSS v2 | CVSS v3 |
|---|---|---|---|---|---|
| CVE-2026-53176 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: IB/isert: Reject login PDUs shorter than ISER_HEADERS_LEN In drivers/infiniband/ulp/isert/ib_isert.c, isert_login_recv_done() computes the login request payload length as wc->byte_len minus ISER_HEADERS_LEN with no lower bound, and login_req_len is a signed int. A remote iSER initiator can post a login Send work request carrying fewer than ISER_HEADERS_LEN (76) bytes, so the subtraction underflows and login_req_len becomes negative. isert_rx_login_req() then reads that negative length back into a signed int, takes size = min(rx_buflen, MAX_KEY_VALUE_PAIRS), and because the min() is signed it keeps the negative value; the value is then passed as the memcpy() length and sign-extended to a multi-gigabyte size_t. The copy into the 8192-byte login->req_buf runs far out of bounds and faults, crashing the target node. The login phase precedes iSCSI authentication, so no credentials are required to reach this path. Reject any login PDU shorter than ISER_HEADERS_LEN before the subtraction, mirroring the existing early return on a failed work completion, so login_req_len can never go negative. The upper bound was already safe: a posted login buffer cannot deliver more than ISER_RX_PAYLOAD_SIZE, so the difference stays at or below MAX_KEY_VALUE_PAIRS and the existing min() clamps it; only the missing lower bound needs to be added. | |||||
| CVE-2026-53175 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: inet: frags: fix use-after-free caused by the fqdir_pre_exit() flush On netns teardown, fqdir_pre_exit() walks the fqdir rhashtable and flushes every fragment queue that is not yet complete using inet_frag_queue_flush(). That helper frees all the skbs queued on the fragment queue but does not set INET_FRAG_COMPLETE, and leaves q->fragments_tail and q->last_run_head pointing at the freed skbs. The queue itself stays in the rhashtable. fqdir_pre_exit() first lowers high_thresh to 0 to stop new queue lookups, but it cannot stop a fragment that already obtained the queue through inet_frag_find() earlier and stalled just before taking the queue lock. Once that fragment resumes after the flush and takes the queue lock, it passes the INET_FRAG_COMPLETE check and then dereferences the freed fragments_tail. inet_frag_queue_insert() reads FRAG_CB() and ->len of that pointer and, on the append path, writes ->next_frag, causing a slab use-after-free. IPv6, nf_conntrack_reasm6 and 6lowpan reassembly share the same flush path and are affected as well. Reset rb_fragments, fragments_tail and last_run_head in inet_frag_queue_flush() so a flushed queue no longer points at the freed skbs. A fragment that resumes after the flush and takes the queue lock then finds an empty queue and starts a new run instead of dereferencing the freed fragments_tail. ip_frag_reinit() already performed this reset after its own flush, so drop the now duplicate code there. | |||||
| CVE-2026-53153 | 2 Linux, Redhat | 2 Linux Kernel, Enterprise Linux | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: mm/list_lru: drain before clearing xarray entry on reparent memcg_reparent_list_lrus() clears the dying memcg's xarray entry with xas_store(&xas, NULL) before reparenting its per-node lists into the parent. This opens a window where a concurrent list_lru_del() arriving for the dying memcg sees xa_load() == NULL, walks to the parent in lock_list_lru_of_memcg(), takes the parent's per-node lock, and calls list_del_init() on an item still physically linked on the dying memcg's list. If another in-flight thread holds the dying memcg's per-node lock at the same moment (another list_lru_del, or a list_lru_walk_one running an isolate callback), both threads modify ->next/->prev pointers on the same physical list under different locks. Adjacent items can corrupt each other's links. Fix it by reversing the order: reparent each per-node list and mark the child's list lru dead and then clear the xarray entry. Any concurrent list_lru op that finds the still-set xarray entry either takes the dying memcg's per-node lock (synchronizing with the drain) or sees LONG_MIN and walks to the parent, where the items now live. | |||||
| CVE-2026-53148 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: thunderbolt: Clamp XDomain response data copy to allocation size tb_xdp_properties_request() derives the per-packet copy length from the response header without checking that it fits in the previously allocated data buffer. A malicious peer can set its length field larger than the declared data_length, causing memcpy to write past the kcalloc allocation. Clamp the per-packet copy length so that the cumulative offset never exceeds data_len. | |||||
| CVE-2026-53145 | 2 Linux, Redhat | 2 Linux Kernel, Enterprise Linux | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: drm/gem: Try to fix change_handle ioctl, attempt 4 [airlied: just added some comments on how to reenable] On-list because the cat is out of the bag and we're clearly not good enough to figure this out in private. The story thus far: 5e28b7b94408 ("drm: Set old handle to NULL before prime swap in change_handle") tried to fix a race condition between the gem_close and gem_change_handle ioctls, but got a few things wrong: - There's a confusion with the local variable handle, which is actually the new handle, and so the two-stage trick was actually applied to the wrong idr slot. 7164d78559b0 ("drm/gem: fix race between change_handle and handle_delete") tried to fix that by adding yet another code block, but forgot to add the error handling. Which meant we now have two paths, both kinda wrong. - dc366607c41c ("drm: Replace old pointer to new idr") tried to apply another fix, but inconsistently, again because of the handle confusion - this would be the right fix (kinda, somewhat, it's a mess) if we'd do the two-stage approach for the new handle. Except that wasn't the intent of the original fix. We also didn't have an igt merged for the original ioctl, which is a big no-go. This was attempted to address off-list in the original bugfix, and amd QA people claimed the bug was fixed now. Very clearly that's not the case. Here's my attempt to sort this out: - Rename the local variable to new_handle, the old aliasing with args->handle is just too dangerously confusing. - Merge the gem obj lookup with the two-stage idr_replace so that we avoid getting ourselves confused there. - This means we don't have a surplus temporary reference anymore, only an inherited from the idr. A concurrent gem_close on the new_handle could steal that. Fix that with the same two-stage approach create_tail uses. This is a bit overkill as documented in the comment, but I also don't trust my ability to understand this all correctly, so go with the established pattern we have from other ioctls instead for maximum paranoia. - Adjust error paths. I've tried to make the error and success paths common, because they are identical except for which handle is removed and on which we call idr_replace to (re)install the object again. But that made things messier to read, so I've left it at the more verbose version, which unfortunately hides the symmetry in the entire code flow a bit. - While at it, also replace the 7 space indent with 1 tab. And finally, because I flat out don't trust my abilities here at all anymore: - Disable the ioctl until we have the igt situation and everything else sorted out on-list and with full consensus. v2: Sashiko noticed that I didn't handle the error path for idr_replace correctly, it must be checked with IS_ERR_OR_NULL like in gem_handle_delete. So yeah, definitely should just the existing paths 1:1 because this is endless amounts of tricky. Also add the Fixes: line for the original ioctl, I forgot that too. | |||||
| CVE-2026-53143 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: drm/amdkfd: Fix buffer overflow in SDMA queue checkpoint/restore on GFX11 The v11 MQD manager incorrectly assigned the CP-compute variants of checkpoint_mqd/restore_mqd for KFD_MQD_TYPE_SDMA queues. These functions use sizeof(struct v11_compute_mqd) (2048 bytes) instead of sizeof(struct v11_sdma_mqd) (512 bytes), causing a 1536-byte overflow. During CRIU checkpoint of an SDMA queue on Navi3x: - checkpoint_mqd() reads 2048 bytes from a 512-byte SDMA MQD buffer, leaking 1536 bytes of adjacent GTT memory to userspace During CRIU restore: - restore_mqd() writes 2048 bytes into a 512-byte SDMA MQD buffer, corrupting 1536 bytes of adjacent GTT memory (often the ring buffer or neighboring MQDs) This is a copy-paste regression unique to v11. All other ASIC backends (cik, vi, v9, v10, v12) correctly use the SDMA-specific variants. Add checkpoint_mqd_sdma() and restore_mqd_sdma() functions that properly handle the smaller v11_sdma_mqd structure, matching the pattern used in other MQD managers. (cherry picked from commit 6fa41db7ffdec97d62433adf03b7b9b759af8c2c) | |||||
| CVE-2026-53006 | 2 Linux, Redhat | 2 Linux Kernel, Enterprise Linux | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: ipv6: fix possible UAF in icmpv6_rcv() Caching saddr and daddr before pskb_pull() is problematic since skb->head can change. Remove these temporary variables: - We only access &ipv6_hdr(skb)->saddr and &ipv6_hdr(skb)->daddr when net_dbg_ratelimited() is called in the slow path. - Avoid potential future misuse after pskb_pull() call. | |||||
| CVE-2026-53002 | 2 Linux, Redhat | 2 Linux Kernel, Enterprise Linux | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: netfilter: conntrack: remove sprintf usage Replace it with scnprintf, the buffer sizes are expected to be large enough to hold the result, no need for snprintf+overflow check. Increase buffer size in mangle_content_len() while at it. BUG: KASAN: stack-out-of-bounds in vsnprintf+0xea5/0x1270 Write of size 1 at addr [..] vsnprintf+0xea5/0x1270 sprintf+0xb1/0xe0 mangle_content_len+0x1ac/0x280 nf_nat_sdp_session+0x1cc/0x240 process_sdp+0x8f8/0xb80 process_invite_request+0x108/0x2b0 process_sip_msg+0x5da/0xf50 sip_help_tcp+0x45e/0x780 nf_confirm+0x34d/0x990 [..] | |||||
| CVE-2026-53000 | 2 Linux, Redhat | 2 Linux Kernel, Enterprise Linux | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: netfilter: nat: use kfree_rcu to release ops Florian Westphal says: "Historically this is not an issue, even for normal base hooks: the data path doesn't use the original nf_hook_ops that are used to register the callbacks. However, in v5.14 I added the ability to dump the active netfilter hooks from userspace. This code will peek back into the nf_hook_ops that are available at the tail of the pointer-array blob used by the datapath. The nat hooks are special, because they are called indirectly from the central nat dispatcher hook. They are currently invisible to the nfnl hook dump subsystem though. But once that changes the nat ops structures have to be deferred too." Update nf_nat_register_fn() to deal with partial exposition of the hooks from error path which can be also an issue for nfnetlink_hook. | |||||
| CVE-2026-52993 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: tipc: fix double-free in tipc_buf_append() tipc_msg_validate() can potentially reallocate the skb it is validating, freeing the old one. In tipc_buf_append(), it was being called with a pointer to a local variable which was a copy of the caller's skb pointer. If the skb was reallocated and validation subsequently failed, the error handling path would free the original skb pointer, which had already been freed, leading to double-free. Fix this by checking if head now points to a newly allocated reassembled skb. If it does, reassign *headbuf for later freeing operations. | |||||
| CVE-2026-52991 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: sched/psi: fix race between file release and pressure write A potential race condition exists between pressure write and cgroup file release regarding the priv member of struct kernfs_open_file, which triggers the uaf reported in [1]. Consider the following scenario involving execution on two separate CPUs: CPU0 CPU1 ==== ==== vfs_rmdir() kernfs_iop_rmdir() cgroup_rmdir() cgroup_kn_lock_live() cgroup_destroy_locked() cgroup_addrm_files() cgroup_rm_file() kernfs_remove_by_name() kernfs_remove_by_name_ns() vfs_write() __kernfs_remove() new_sync_write() kernfs_drain() kernfs_fop_write_iter() kernfs_drain_open_files() cgroup_file_write() kernfs_release_file() pressure_write() cgroup_file_release() ctx = of->priv; kfree(ctx); of->priv = NULL; cgroup_kn_unlock() cgroup_kn_lock_live() cgroup_get(cgrp) cgroup_kn_unlock() if (ctx->psi.trigger) // here, trigger uaf for ctx, that is of->priv The cgroup_rmdir() is protected by the cgroup_mutex, it also safeguards the memory deallocation of of->priv performed within cgroup_file_release(). However, the operations involving of->priv executed within pressure_write() are not entirely covered by the protection of cgroup_mutex. Consequently, if the code in pressure_write(), specifically the section handling the ctx variable executes after cgroup_file_release() has completed, a uaf vulnerability involving of->priv is triggered. Therefore, the issue can be resolved by extending the scope of the cgroup_mutex lock within pressure_write() to encompass all code paths involving of->priv, thereby properly synchronizing the race condition occurring between cgroup_file_release() and pressure_write(). And, if an live kn lock can be successfully acquired while executing the pressure write operation, it indicates that the cgroup deletion process has not yet reached its final stage; consequently, the priv pointer within open_file cannot be NULL. Therefore, the operation to retrieve the ctx value must be moved to a point *after* the live kn lock has been successfully acquired. In another situation, specifically after entering cgroup_kn_lock_live() but before acquiring cgroup_mutex, there exists a different class of race condition: CPU0: write memory.pressure CPU1: write cgroup.pressure=0 =========================== ============================= kernfs_fop_write_iter() kernfs_get_active_of(of) pressure_write() cgroup_kn_lock_live(memory.pressure) cgroup_tryget(cgrp) kernfs_break_active_protection(kn) ... blocks on cgroup_mutex cgroup_pressure_write() cgroup_kn_lock_live(cgroup.pressure) cgroup_file_show(memory.pressure, false) kernfs_show(false) kernfs_drain_open_files() cgroup_file_release(of) kfree(ctx) of->priv = NULL cgroup_kn_unlock() ... acquires cgroup_mutex ctx = of->priv; // may now be NULL if (ctx->psi.trigger) // NULL dereference Consequently, there is a possibility that of->priv is NULL, the pressure write needs to check for this. Now that the scope of the cgroup_mutex has been expanded, the original explicit cgroup_get/put operations are no longer necessary, this is because acquiring/releasing the live kn lock inherently executes a cgroup get/put operation. [1] BUG: KASAN: slab-use-after-free in pressure_write+0xa4/0x210 kernel/cgroup/cgroup.c:4011 Call Trace: pressure_write+0xa4/0x210 kernel/cgroup/cgroup.c:4011 cgroup_file_write+0x36f/0x790 kernel/cgroup/cgroup.c:43 ---truncated--- | |||||
| CVE-2026-52989 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: nvmet-tcp: propagate nvmet_tcp_build_pdu_iovec() errors to its callers Currently, when nvmet_tcp_build_pdu_iovec() detects an out-of-bounds PDU length or offset, it triggers nvmet_tcp_fatal_error(cmd->queue) and returns early. However, because the function returns void, the callers are entirely unaware that a fatal error has occurred and that the cmd->recv_msg.msg_iter was left uninitialized. Callers such as nvmet_tcp_handle_h2c_data_pdu() proceed to blindly overwrite the queue state with queue->rcv_state = NVMET_TCP_RECV_DATA Consequently, the socket receiving loop may attempt to read incoming network data into the uninitialized iterator. Fix this by shifting the error handling responsibility to the callers. | |||||
| CVE-2026-52987 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: drm/amdgpu: avoid double drm_exec_fini() in userq validate When new_addition is true, amdgpu_userq_vm_validate() calls drm_exec_fini(&exec) before iterating over the collected HMM ranges and calling amdgpu_ttm_tt_get_user_pages(). If amdgpu_ttm_tt_get_user_pages() fails in that path, the code jumps to unlock_all and calls drm_exec_fini(&exec) a second time on the same exec object. drm_exec_fini() is not idempotent: it frees exec->objects and may also drop exec->contended and finalize the ww acquire context. Route that error path directly to the range cleanup once exec has already been finalized. Issue found using a prototype static analysis tool and confirmed by code review. (cherry picked from commit 2802952e4a07306da6ebe813ff1acacc5691851a) | |||||
| CVE-2026-52973 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: futex: Drop CLONE_THREAD requirement for private default hash alloc Currently need_futex_hash_allocate_default() depends on strict pthread semantics, abusing CLONE_THREAD. This breaks the non-concurrency assumptions when doing the mm->futex_ref pcpu allocations, leading to bugs[0] when sharing the mm in other ways; ie: BUG: KASAN: slab-use-after-free in futex_hash_put ... where the +1 bias can end up on a percpu counter that mm->futex_ref no longer points at. Loosen the check to cover any CLONE_VM clone, except vfork(). Excluding vfork keeps the existing paths untouched (no overhead), and we can't race in the first place: either the parent is suspended and the child runs alone, or mm->futex_ref is already allocated from an earlier CLONE_VM. | |||||
| CVE-2026-52972 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 5.5 MEDIUM |
| In the Linux kernel, the following vulnerability has been resolved: crypto: af_alg - Cap AEAD AD length to 0x80000000 In order to prevent arithmetic overflows when checking the TX buffer size, cap the associated data length to 0x80000000. | |||||
| CVE-2026-52969 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: KVM: Reject wrapped offset in kvm_reset_dirty_gfn() kvm_reset_dirty_gfn() guards the gfn range with if (!memslot || (offset + __fls(mask)) >= memslot->npages) return; but offset is u64 and the addition is unchecked. The check can be silently bypassed by a u64 wrap. The dirty ring backing those entries is MAP_SHARED at KVM_DIRTY_LOG_PAGE_OFFSET of the vcpu fd, so the VMM can rewrite the slot and offset fields of any entry between when the kernel pushes them and when KVM_RESET_DIRTY_RINGS consumes them. On reset, kvm_dirty_ring_reset() re-reads the values via READ_ONCE() and feeds them straight back into this check; only the flags handshake is treated as the handover, the slot/offset payload is taken on trust. Crafting two entries entry[i].offset = 0xffffffffffffffc1 entry[i+1].offset = 0 makes the coalescing loop in kvm_dirty_ring_reset() compute delta = (s64)(0 - 0xffffffffffffffc1) = 63 which falls in [0, BITS_PER_LONG), so it folds entry[i+1] into the existing mask by setting bit 63. The trailing kvm_reset_dirty_gfn() call then sees offset = 0xffffffffffffffc1 and __fls(mask) = 63; the sum is 0 in u64 and the bounds check passes. That offset propagates into kvm_arch_mmu_enable_log_dirty_pt_masked() unchanged. On the legacy MMU path -- kvm_memslots_have_rmaps() == true, i.e. shadow paging, any VM that has allocated shadow roots, or a write-tracked slot -- it reaches gfn_to_rmap(), which indexes slot->arch.rmap[0][] with a near-U64_MAX gfn. That is an out-of-bounds load of a kvm_rmap_head, followed by a conditional clear of PT_WRITABLE_MASK in whatever the loaded pointer points at. The path is reachable from any process holding /dev/kvm. Range-check offset on its own first, so the addition cannot wrap. memslot->npages is bounded well below U64_MAX, so once offset < npages holds, offset + __fls(mask) (with __fls(mask) < BITS_PER_LONG) stays in range. | |||||
| CVE-2026-52955 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 9.8 CRITICAL |
| In the Linux kernel, the following vulnerability has been resolved: libceph: Fix potential out-of-bounds access in crush_decode() A message of type CEPH_MSG_OSD_MAP containing a crush map with at least one bucket has two fields holding the bucket algorithm. If the values in these two fields differ, an out-of-bounds access can occur. This is the case because the first algorithm field (alg) is used to allocate the correct amount of memory for a bucket of this type, while the second algorithm field inside the bucket (b->alg) is used in the subsequent processing. This patch fixes the issue by adding a check that compares alg and b->alg and aborts the processing in case they differ. Furthermore, b->alg is set to 0 in this case, because the destruction of the crush map also uses this field to determine the bucket type, which can again result in an out-of-bounds access when trying to free the memory pointed to by the fields of the bucket. To correctly free the memory allocated for the bucket in such a case, the corresponding call to kfree is moved from the algorithm-specific crush_destroy_bucket functions to the generic crush_destroy_bucket(). | |||||
| CVE-2026-52952 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 8.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: iommu: Fix WARN_ON in __iommu_group_set_domain_nofail() due to reset In __iommu_group_set_domain_internal(), concurrent domain attachments are rejected when any device in the group is recovering. This is necessary to fence concurrent attachments to a multi-device group where devices might share the same RID due to PCI DMA alias quirks, but triggers the WARN_ON in __iommu_group_set_domain_nofail(). Other IOMMU_SET_DOMAIN_MUST_SUCCEED callers in detach/teardown paths, such as __iommu_group_set_core_domain and __iommu_release_dma_ownership, should not be rejected, as the domain would be freed anyway in these nofail paths while group->domain is still pointing to it. So pci_dev_reset_iommu_done() could trigger a UAF when re-attaching group->domain. Honor the IOMMU_SET_DOMAIN_MUST_SUCCEED flag, allowing the callers through the group->recovery_cnt fence, so as to update the group->domain pointer. Instead add a gdev->blocked check in the device iteration loop, to prevent any concurrent per-device detachment. | |||||
| CVE-2026-52951 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: drm/xe/dma-buf: handle empty bo and UAF races There look to be some nasty races here when triggering the invalidate_mappings hook: 1) We do xe_bo_alloc() followed by the attach, before the actual full bo init step in xe_dma_buf_init_obj(). However the bo is visible on the attachments list after the attach. This is bad since exporter driver, say amdgpu, can at any time call back into our invalidate_mappings hook, with an empty/bogus bo, leading to potential bugs/crashes. 2) Similar to 1) but here we get a UAF, when the invalidate_mappings hook is triggered. For example, we get as far as xe_bo_init_locked() but this fails in some way. But here the bo will be freed on error, but we still have it attached from dma-buf pov, so if the invalidate_mappings is now triggered then the bo we access is gone and we trigger UAF and more bugs/crashes. To fix this, move the attach step until after we actually have a fully set up buffer object. Note that the bo is not published to userspace until later, so not sure what the comment "Don't publish the bo until we have a valid attachment", is referring to. We have at least two different customers reporting hitting a NULL ptr deref in evict_flags when importing something from amdgpu, followed by triggering the evict flow. Hit rate is also pretty low, which would hint at some kind of race, so something like 1) or 2) might explain this. v2: - Shuffle the order of the ops slightly (no functional change) - Improve the comment to better explain the ordering (Matt B) (cherry picked from commit af1f2ad0c59fe4e2f924c526f66e968289d77971) | |||||
| CVE-2026-52943 | 1 Linux | 1 Linux Kernel | 2026-07-15 | N/A | 7.8 HIGH |
| In the Linux kernel, the following vulnerability has been resolved: net: skbuff: fix missing zerocopy reference in pskb_carve helpers pskb_carve_inside_header() and pskb_carve_inside_nonlinear() both copy the old skb_shared_info header into a new buffer via memcpy(), which includes the destructor_arg pointer (uarg) for MSG_ZEROCOPY skbs. Neither function calls net_zcopy_get() for the new shinfo, creating an unaccounted holder: every skb_shared_info with destructor_arg set will call skb_zcopy_clear() once when freed, but the corresponding net_zcopy_get() was never called for the new copy. Repeated calls drive uarg->refcnt to zero prematurely, freeing ubuf_info_msgzc while TX skbs still hold live destructor_arg pointers. KASAN reports use-after-free on a freed ubuf_info_msgzc: BUG: KASAN: slab-use-after-free in skb_release_data+0x77b/0x810 Read of size 8 at addr ffff88801574d3e8 by task poc/220 Call Trace: skb_release_data+0x77b/0x810 kfree_skb_list_reason+0x13e/0x610 skb_release_data+0x4cd/0x810 sk_skb_reason_drop+0xf3/0x340 skb_queue_purge_reason+0x282/0x440 rds_tcp_inc_free+0x1e/0x30 rds_recvmsg+0x354/0x1780 __sys_recvmsg+0xdf/0x180 Allocated by task 219: msg_zerocopy_realloc+0x157/0x7b0 tcp_sendmsg_locked+0x2892/0x3ba0 Freed by task 219: ip_recv_error+0x74a/0xb10 tcp_recvmsg+0x475/0x530 The skb consuming the late access still referenced the same uarg via shinfo->destructor_arg copied by pskb_carve_inside_nonlinear() without a refcount bump. This has been verified to be reliably exploitable: a working proof-of-concept achieves full root privilege escalation from an unprivileged local user on a default kernel configuration. The fix follows the pattern of pskb_expand_head() which has the same memcpy/cloned structure. For pskb_carve_inside_header(), net_zcopy_get() is placed after skb_orphan_frags() succeeds, so the orphan error path needs no cleanup. For pskb_carve_inside_nonlinear(), net_zcopy_get() is placed after all failure points and just before skb_release_data(), so no error path needs cleanup at all -- matching pskb_expand_head() more closely and avoiding the need for a balancing net_zcopy_put(). | |||||