)]}' { "commit": "4a8a5b51fdf34070ff4e32703516ae6e2dbb3a1e", "tree": "b6a40c9c4716c7d0acb4a47c20dd4264a0ae0f42", "parents": [ "058b6f371e46de0bcab8e0125107c79dcd2ba61e" ], "author": { "name": "Davide Caratti", "email": "dcaratti@redhat.com", "time": "Wed Apr 08 17:24:36 2026 +0200" }, "committer": { "name": "Derek Taylor", "email": "ddtaylor@google.com", "time": "Fri Jun 26 11:21:30 2026 -0700" }, "message": "net/sched: cls_fw: fix NULL dereference of \"old\" filters before change()\n\n[ Upstream commit 65782b2db7321d5f97c16718c4c7f6c7205a56be ]\n\nLike pointed out by Sashiko [1], since commit ed76f5edccc9 (\"net: sched:\nprotect filter_chain list with filter_chain_lock mutex\") TC filters are\nadded to a shared block and published to datapath before their -\u003echange()\nfunction is called. This is a problem for cls_fw: an invalid filter\ncreated with the \"old\" method can still classify some packets before it\nis destroyed by the validation logic added by Xiang.\nTherefore, insisting with repeated runs of the following script:\n\n # ip link add dev crash0 type dummy\n # ip link set dev crash0 up\n # mausezahn crash0 -c 100000 -P 10 \\\n \u003e -A 4.3.2.1 -B 1.2.3.4 -t udp \"dp\u003d1234\" -q \u0026\n # sleep 1\n # tc qdisc add dev crash0 egress_block 1 clsact\n # tc filter add block 1 protocol ip prio 1 matchall \\\n \u003e action skbedit mark 65536 continue\n # tc filter add block 1 protocol ip prio 2 fw\n # ip link del dev crash0\n\ncan still make fw_classify() hit the WARN_ON() in [2]:\n\n WARNING: ./include/net/pkt_cls.h:88 at fw_classify+0x244/0x250 [cls_fw], CPU#18: mausezahn/1399\n Modules linked in: cls_fw(E) act_skbedit(E)\n CPU: 18 UID: 0 PID: 1399 Comm: mausezahn Tainted: G E 7.0.0-rc6-virtme #17 PREEMPT(full)\n Tainted: [E]\u003dUNSIGNED_MODULE\n Hardware name: Red Hat KVM, BIOS 1.16.3-2.el9 04/01/2014\n RIP: 0010:fw_classify+0x244/0x250 [cls_fw]\n Code: 5c 49 c7 45 00 00 00 00 00 41 5d 41 5e 41 5f 5d c3 cc cc cc cc 5b b8 ff ff ff ff 41 5c 41 5d 41 5e 41 5f 5d c3 cc cc cc cc 90 \u003c0f\u003e 0b 90 eb a0 0f 1f 80 00 00 00 00 90 90 90 90 90 90 90 90 90 90\n RSP: 0018:ffffd1b7026bf8a8 EFLAGS: 00010202\n RAX: ffff8c5ac9c60800 RBX: ffff8c5ac99322c0 RCX: 0000000000000004\n RDX: 0000000000000001 RSI: ffff8c5b74d7a000 RDI: ffff8c5ac8284f40\n RBP: ffffd1b7026bf8d0 R08: 0000000000000000 R09: ffffd1b7026bf9b0\n R10: 00000000ffffffff R11: 0000000000000000 R12: 0000000000010000\n R13: ffffd1b7026bf930 R14: ffff8c5ac8284f40 R15: 0000000000000000\n FS: 00007fca40c37740(0000) GS:ffff8c5b74d7a000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 00007fca40e822a0 CR3: 0000000005ca0001 CR4: 0000000000172ef0\n Call Trace:\n \u003cTASK\u003e\n tcf_classify+0x17d/0x5c0\n tc_run+0x9d/0x150\n __dev_queue_xmit+0x2ab/0x14d0\n ip_finish_output2+0x340/0x8f0\n ip_output+0xa4/0x250\n raw_sendmsg+0x147d/0x14b0\n __sys_sendto+0x1cc/0x1f0\n __x64_sys_sendto+0x24/0x30\n do_syscall_64+0x126/0xf80\n entry_SYSCALL_64_after_hwframe+0x77/0x7f\n RIP: 0033:0x7fca40e822ba\n Code: d8 64 89 02 48 c7 c0 ff ff ff ff eb b8 0f 1f 00 f3 0f 1e fa 41 89 ca 64 8b 04 25 18 00 00 00 85 c0 75 15 b8 2c 00 00 00 0f 05 \u003c48\u003e 3d 00 f0 ff ff 77 7e c3 0f 1f 44 00 00 41 54 48 83 ec 30 44 89\n RSP: 002b:00007ffc248a42c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c\n RAX: ffffffffffffffda RBX: 000055ef233289d0 RCX: 00007fca40e822ba\n RDX: 000000000000001e RSI: 000055ef23328c30 RDI: 0000000000000003\n RBP: 000055ef233289d0 R08: 00007ffc248a42d0 R09: 0000000000000010\n R10: 0000000000000000 R11: 0000000000000246 R12: 000000000000001e\n R13: 00000000000186a0 R14: 0000000000000000 R15: 00007fca41043000\n \u003c/TASK\u003e\n irq event stamp: 1045778\n hardirqs last enabled at (1045784): [\u003cffffffff864ec042\u003e] __up_console_sem+0x52/0x60\n hardirqs last disabled at (1045789): [\u003cffffffff864ec027\u003e] __up_console_sem+0x37/0x60\n softirqs last enabled at (1045426): [\u003cffffffff874d48c7\u003e] __alloc_skb+0x207/0x260\n softirqs last disabled at (1045434): [\u003cffffffff874fe8f8\u003e] __dev_queue_xmit+0x78/0x14d0\n\nThen, because of the value in the packet\u0027s mark, dereference on \u0027q-\u003ehandle\u0027\nwith NULL \u0027q\u0027 occurs:\n\n BUG: kernel NULL pointer dereference, address: 0000000000000038\n [...]\n RIP: 0010:fw_classify+0x1fe/0x250 [cls_fw]\n [...]\n\nSkip \"old-style\" classification on shared blocks, so that the NULL\ndereference is fixed and WARN_ON() is not hit anymore in the short\nlifetime of invalid cls_fw \"old-style\" filters.\n\n[1] https://sashiko.dev/#/patchset/20260331050217.504278-1-xmei5%40asu.edu\n[2] https://elixir.bootlin.com/linux/v7.0-rc6/source/include/net/pkt_cls.h#L86\n\nBUG\u003db/527722441\nTEST\u003dpresubmit\nRELEASE_NOTE\u003dFixed CVE-2026-53080 in the Linux kernel.\n\ncos-patch: security-moderate\nFixes: faeea8bbf6e9 (\"net/sched: cls_fw: fix NULL pointer dereference on shared blocks\")\nFixes: ed76f5edccc9 (\"net: sched: protect filter_chain list with filter_chain_lock mutex\")\nAcked-by: Jamal Hadi Salim \u003cjhs@mojatatu.com\u003e\nChange-Id: I3ae7f1c1bef6873c5bc834cc5ab4c30c0742394b\nSigned-off-by: Davide Caratti \u003cdcaratti@redhat.com\u003e\nLink: https://patch.msgid.link/e39cbd3103a337f1e515d186fe697b4459d24757.1775661704.git.dcaratti@redhat.com\nSigned-off-by: Jakub Kicinski \u003ckuba@kernel.org\u003e\nSigned-off-by: Sasha Levin \u003csashal@kernel.org\u003e\nSigned-off-by: Kernel CVE Triage Automation \u003ccloud-image-kernel-cve-triage-automation@prod.google.com\u003e\nReviewed-on: https://cos-review.googlesource.com/c/third_party/kernel/+/164210\nTested-by: Cusky Presubmit Bot \u003cpresubmit@cos-infra-prod.iam.gserviceaccount.com\u003e\nReviewed-by: Derek Taylor \u003cddtaylor@google.com\u003e\nReviewed-by: Chenglong Tang \u003cchenglongtang@google.com\u003e\n", "tree_diff": [ { "type": "modify", "old_id": "2e5e8df5ca55c45eb7a55b0b14813c61d1dbd398", "old_mode": 33188, "old_path": "net/sched/cls_fw.c", "new_id": "bc007bb47432cd8eb89768c72f09d7dd93700c2c", "new_mode": 33188, "new_path": "net/sched/cls_fw.c" } ] }