Fixed
Created: Jun 18, 2025
Updated: Oct 14, 2025
Resolved Date: Oct 13, 2025
Found In Version: 10.22.33.1
Fix Version: 10.22.33.2
Severity: Standard
Applicable for: Wind River Linux LTS 22
Component/s: Kernel
In the Linux kernel, the following vulnerability has been resolved:EOL][EOL]bpf, cgroup: Fix kernel BUG in purge_effective_progs[EOL][EOL]Syzkaller reported a triggered kernel BUG as follows:[EOL][EOL] ------------[ cut here ]------------[EOL] kernel BUG at kernel/bpf/cgroup.c:925![EOL] invalid opcode: 0000 [#1] PREEMPT SMP NOPTI[EOL] CPU: 1 PID: 194 Comm: detach Not tainted 5.19.0-14184-g69dac8e431af #8[EOL] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS[EOL] rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014[EOL] RIP: 0010:__cgroup_bpf_detach+0x1f2/0x2a0[EOL] Code: 00 e8 92 60 30 00 84 c0 75 d8 4c 89 e0 31 f6 85 f6 74 19 42 f6 84[EOL] 28 48 05 00 00 02 75 0e 48 8b 80 c0 00 00 00 48 85 c0 75 e5 <0f> 0b 48[EOL] 8b 0c5[EOL] RSP: 0018:ffffc9000055bdb0 EFLAGS: 00000246[EOL] RAX: 0000000000000000 RBX: ffff888100ec0800 RCX: ffffc900000f1000[EOL] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff888100ec4578[EOL] RBP: 0000000000000000 R08: ffff888100ec0800 R09: 0000000000000040[EOL] R10: 0000000000000000 R11: 0000000000000000 R12: ffff888100ec4000[EOL] R13: 000000000000000d R14: ffffc90000199000 R15: ffff888100effb00[EOL] FS: 00007f68213d2b80(0000) GS:ffff88813bc80000(0000)[EOL] knlGS:0000000000000000[EOL] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033[EOL] CR2: 000055f74a0e5850 CR3: 0000000102836000 CR4: 00000000000006e0[EOL] Call Trace:[EOL] <TASK>[EOL] cgroup_bpf_prog_detach+0xcc/0x100[EOL] __sys_bpf+0x2273/0x2a00[EOL] __x64_sys_bpf+0x17/0x20[EOL] do_syscall_64+0x3b/0x90[EOL] entry_SYSCALL_64_after_hwframe+0x63/0xcd[EOL] RIP: 0033:0x7f68214dbcb9[EOL] Code: 08 44 89 e0 5b 41 5c c3 66 0f 1f 84 00 00 00 00 00 48 89 f8 48 89[EOL] f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01[EOL] f0 ff8[EOL] RSP: 002b:00007ffeb487db68 EFLAGS: 00000246 ORIG_RAX: 0000000000000141[EOL] RAX: ffffffffffffffda RBX: 000000000000000b RCX: 00007f68214dbcb9[EOL] RDX: 0000000000000090 RSI: 00007ffeb487db70 RDI: 0000000000000009[EOL] RBP: 0000000000000003 R08: 0000000000000012 R09: 0000000b00000003[EOL] R10: 00007ffeb487db70 R11: 0000000000000246 R12: 00007ffeb487dc20[EOL] R13: 0000000000000004 R14: 0000000000000001 R15: 000055f74a1011b0[EOL] </TASK>[EOL] Modules linked in:[EOL] ---[ end trace 0000000000000000 ]---[EOL][EOL]Repetition steps:[EOL][EOL]For the following cgroup tree,[EOL][EOL] root[EOL] ([EOL) cg1EOL] ([EOL) cg2[EOL][EOL] 1. attach prog2 to cg2, and then attach prog1 to cg1, both bpf progs[EOL] attach type is NONE or OVERRIDE.[EOL] 2. write 1 to /proc/thread-self/fail-nth for failslab.[EOL] 3. detach prog1 for cg1, and then kernel BUG occur.[EOL][EOL]Failslab injection will cause kmalloc fail and fall back to[EOL]purge_effective_progs. The problem is that cg2 have attached another prog,[EOL]so when go through cg2 layer, iteration will add pos to 1, and subsequent[EOL]operations will be skipped by the following condition, and cg will meet[EOL]NULL in the end.[EOL][EOL] `if (pos && !(cg->bpf.flags[atype] & BPF_F_ALLOW_MULTI))`[EOL][EOL]The NULL cg means no link or prog match, this is as expected, and it's not[EOL]a bug. So here just skip the no match situation.