{"dataType":"CVE_RECORD","dataVersion":"5.2","cveMetadata":{"cveId":"CVE-2026-72344","assignerOrgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","state":"PUBLISHED","assignerShortName":"Linux","dateReserved":"2026-08-09T03:40:39.920Z","datePublished":"2026-08-15T05:55:49.751Z","dateUpdated":"2026-08-17T05:43:03.143Z"},"containers":{"cna":{"providerMetadata":{"orgId":"416baaa9-dc9f-4396-8d5f-8c081fb06d67","shortName":"Linux","dateUpdated":"2026-08-17T05:43:03.143Z"},"descriptions":[{"lang":"en","value":"In the Linux kernel, the following vulnerability has been resolved:\n\nnet/mlx5e: TC, skip peer flow cleanup when LAG seq is unavailable\n\nmlx5_lag_get_dev_seq() will return error when the peer isn't in the LAG\nor when no device is marked as master. Result bad memory access and kernel\ncrash[1].\n\nHence, skip the peer when lookup fails.\n\nNote: In case there are peer flows, they are cleaned before LAG cleared\nthe master mark.\n\n[1]\nRIP: 0010:mlx5e_tc_del_fdb_peers_flow+0x3d/0x350 [mlx5_core]\nCall Trace:\n <TASK>\n mlx5e_tc_clean_fdb_peer_flows+0xc1/0x130 [mlx5_core]\n mlx5_esw_offloads_unpair+0x3a/0x400 [mlx5_core]\n mlx5_esw_offloads_devcom_event+0xee/0x360 [mlx5_core]\n mlx5_devcom_send_event+0x7a/0x140 [mlx5_core]\n mlx5_esw_offloads_devcom_cleanup+0x2f/0x90 [mlx5_core]\n mlx5e_tc_esw_cleanup+0x28/0xf0 [mlx5_core]\n mlx5e_rep_tc_cleanup+0x19/0x30 [mlx5_core]\n mlx5e_cleanup_uplink_rep_tx+0x36/0x40 [mlx5_core]\n mlx5e_cleanup_rep_tx+0x55/0x60 [mlx5_core]\n mlx5e_detach_netdev+0x96/0xf0 [mlx5_core]\n mlx5e_netdev_change_profile+0x5b/0x120 [mlx5_core]\n mlx5e_netdev_attach_nic_profile+0x1b/0x30 [mlx5_core]\n mlx5e_vport_rep_unload+0xdd/0x110 [mlx5_core]\n __esw_offloads_unload_rep+0x81/0xb0 [mlx5_core]\n mlx5_eswitch_unregister_vport_reps+0x1d7/0x220 [mlx5_core]\n mlx5e_rep_remove+0x22/0x30 [mlx5_core]\n device_release_driver_internal+0x194/0x1f0\n bus_remove_device+0xe8/0x1b0\n device_del+0x159/0x3c0\n mlx5_rescan_drivers_locked+0xbc/0x2d0 [mlx5_core]\n mlx5_unregister_device+0x54/0x80 [mlx5_core]\n mlx5_uninit_one+0x73/0x130 [mlx5_core]\n remove_one+0x78/0xe0 [mlx5_core]\n pci_device_remove+0x39/0xa0"}],"metrics":[{"cvssV3_1":{"version":"3.1","vectorString":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H","baseScore":7.8,"baseSeverity":"HIGH"},"scenarios":[{"lang":"en","value":"AV:L - The bug is in mlx5e TC eswitch peer-flow teardown (mlx5e_tc_clean_fdb_peer_flows/mlx5e_tc_del_fdb_peers_flow), reached via local TC netlink, netdev/LAG reconfiguration, or driver unload—not from processing remote network packets.\nAC:L - On mlx5 LAG/merged-eswitch systems, a caller with CAP_NET_ADMIN can install TC flower rules that create peer flows, then trigger LAG unpair/teardown so mlx5_lag_get_dev_seq() fails and the unchecked negative index is used reliably.\nPR:L - Exploitation requires CAP_NET_ADMIN to install/delete mlx5e TC cls_flower offload rules and manipulate LAG/bond state; CAP_NET_ADMIN is obtainable in user/network namespaces and is not init-namespace root.\nUI:N - No victim interaction is required; the crash occurs during kernel-initiated peer-flow cleanup when LAG sequence lookup fails during eswitch unpair or mlx5e netdev/driver teardown.\nS:U - Impact is kernel memory corruption and crash on the host running mlx5; it does not cross a VM/hypervisor, IOMMU, or other security boundary into a separate authority.\nC:H - mlx5_lag_get_dev_seq() errors (e.g., -ENOENT) are used unchecked as peer_flows[]/flow->peer[] indices, causing out-of-bounds list operations that read and corrupt adjacent kernel memory beyond the flow structures.\nI:H - The faulting list_del(&flow->peer[peer_index]) and peer_flows[i] walks with a negative index perform out-of-bounds linked-list writes, corrupting kernel heap/metadata and enabling further memory control primitives.\nA:H - The fix commit documents a reproducible kernel crash (oops at mlx5e_tc_del_fdb_peers_flow) from this bad memory access during mlx5 eswitch/TC cleanup, causing complete loss of system availability."}]}],"affected":[{"product":"Linux","vendor":"Linux","defaultStatus":"unaffected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["drivers/net/ethernet/mellanox/mlx5/core/en_tc.c"],"versions":[{"version":"971b28accc09436fe6a6d5afd667dcbfb3ed7e03","lessThan":"5a95aa0198af75047d98fb72950642c89dab770f","status":"affected","versionType":"git"},{"version":"971b28accc09436fe6a6d5afd667dcbfb3ed7e03","lessThan":"7bed4af0ced82948d660205efecd551ef8bc3912","status":"affected","versionType":"git"}]},{"product":"Linux","vendor":"Linux","defaultStatus":"affected","repo":"https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git","programFiles":["drivers/net/ethernet/mellanox/mlx5/core/en_tc.c"],"versions":[{"version":"7.1","status":"affected"},{"version":"0","lessThan":"7.1","status":"unaffected","versionType":"semver"},{"version":"7.1.5","lessThanOrEqual":"7.1.*","status":"unaffected","versionType":"semver"},{"version":"7.2","lessThanOrEqual":"*","status":"unaffected","versionType":"original_commit_for_fix"}]}],"cpeApplicability":[{"nodes":[{"operator":"OR","negate":false,"cpeMatch":[{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"7.1","versionEndExcluding":"7.1.5"},{"vulnerable":true,"criteria":"cpe:2.3:o:linux:linux_kernel:*:*:*:*:*:*:*:*","versionStartIncluding":"7.1","versionEndExcluding":"7.2"}]}]}],"references":[{"url":"https://git.kernel.org/stable/c/5a95aa0198af75047d98fb72950642c89dab770f"},{"url":"https://git.kernel.org/stable/c/7bed4af0ced82948d660205efecd551ef8bc3912"}],"title":"net/mlx5e: TC, skip peer flow cleanup when LAG seq is unavailable","x_generator":{"engine":"bippy-1.2.0"}}}}