linux-kernel - WARNING in __perf_event

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJg=8jzsPbjKkKmZudMvFJDnmBHhRrZqPLACs4NibLNdga02MQ@mail.gmail.com>
Date: Tue, 11 Jun 2024 12:59:35 -0700
From: Marius Fleischer <fleischermarius@...il.com>
To: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>, 
	Arnaldo Carvalho de Melo <acme@...nel.org>, Alexei Starovoitov <ast@...nel.org>, Daniel Borkmann <daniel@...earbox.net>, 
	Andrii Nakryiko <andrii@...nel.org>, linux-perf-users@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Cc: syzkaller@...glegroups.com, harrisonmichaelgreen@...il.com
Subject: WARNING in __perf_event_overflow

Hi,

We would like to report the following bug which has been found by our
modified version of syzkaller.

======================================================
description: WARNING in __perf_event_overflow
affected file: kernel/events/core.c
kernel version: 5.15.159
kernel commit: 83655231580bc07485a4ac2a6c971c3a175dd27d
git tree: upstream
kernel config: attached
crash reproducer: attached
======================================================
Crash log:
WARNING: CPU: 0 PID: 157335 at kernel/events/core.c:9435
__perf_event_overflow+0x4f4/0x5b0 kernel/events/core.c:9435
Modules linked in:
CPU: 0 PID: 157335 Comm: syz-executor.5 Not tainted 5.15.159 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
RIP: 0010:__perf_event_overflow+0x4f4/0x5b0 kernel/events/core.c:9435
Code: b6 44 24 0f 84 c0 0f 84 68 fd ff ff e8 65 66 e1 ff 44 89 fe 44
89 ef e8 4a 5e e1 ff 45 39 fd 0f 84 4f fd ff ff e8 4c 66 e1 ff <0f> 0b
e9 43 fd ff ff c7 04 24 00 00 00 00 e9 6c fe ff ff 4c 89 ef
RSP: 0000:fffffe0000011ab8 EFLAGS: 00010046
RAX: 0000000080110000 RBX: ffff88807e178bd0 RCX: ffffffff81966076
RDX: ffff888032fa0000 RSI: ffffffff81966084 RDI: 0000000000000004
RBP: fffffe0000011bc0 R08: 0000000000000000 R09: ffff88807e178dcf
R10: 000000001787c2f9 R11: 0000000000000001 R12: fffffe0000011ef8
R13: 000000001787c2f9 R14: ffff88807e178ff4 R15: 0000000076bff1cf
FS:  00007f5415deb640(0000) GS:ffff8880b9c00000(0000) knlGS:ffff8880b9c00000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f5415deaf88 CR3: 0000000123ea5000 CR4: 0000000000750ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400
PKRU: 55555554
Call Trace:
 <NMI>
 handle_pmi_common+0x5c9/0xa20 arch/x86/events/intel/core.c:2899
 intel_pmu_handle_irq+0x265/0xf30 arch/x86/events/intel/core.c:2960
 perf_event_nmi_handler arch/x86/events/core.c:1745 [inline]
 perf_event_nmi_handler+0x48/0x70 arch/x86/events/core.c:1731
 nmi_handle+0x13d/0x3a0 arch/x86/kernel/nmi.c:140
 default_do_nmi+0x6b/0x170 arch/x86/kernel/nmi.c:334
 exc_nmi+0xf0/0x120 arch/x86/kernel/nmi.c:510
 end_repeat_nmi+0x16/0x31
RIP: 0010:asm_sysvec_irq_work+0x0/0x20 arch/x86/include/asm/idtentry.h:660
Code: e9 a5 03 00 00 0f 1f 44 00 00 0f 01 ca fc 6a ff e8 a5 02 00 00
48 89 c4 48 89 e7 e8 fa 1c f3 ff e9 85 03 00 00 0f 1f 44 00 00 <0f> 01
ca fc 6a ff e8 85 02 00 00 48 89 c4 48 89 e7 e8 ea 21 f3 ff
RSP: 0000:fffffe0000002fd8 EFLAGS: 00000046
RAX: 0000000000000003 RBX: 00007f54179b8f80 RCX: 00007f541787bdad
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000020000080
RBP: 00007f54178dd4a6 R08: 0000000000000008 R09: 0000000000000000
R10: ffffffffffffffff R11: 0000000000000246 R12: 0000000000000000
R13: 000000000000000b R14: 00007f54179b8f80 R15: 00007f5415dcb000
 </NMI>
 <ENTRY_TRAMPOLINE>
RIP: 0033:0x7f541787bdad
RSP: 002b:00007f5415deb028 EFLAGS: 00000246
 </ENTRY_TRAMPOLINE>
----------------
Code disassembly (best guess):
   0:    e9 a5 03 00 00           jmp    0x3aa
   5:    0f 1f 44 00 00           nopl   0x0(%rax,%rax,1)
   a:    0f 01 ca                 clac
   d:    fc                       cld
   e:    6a ff                    push   $0xffffffffffffffff
  10:    e8 a5 02 00 00           call   0x2ba
  15:    48 89 c4                 mov    %rax,%rsp
  18:    48 89 e7                 mov    %rsp,%rdi
  1b:    e8 fa 1c f3 ff           call   0xfff31d1a
  20:    e9 85 03 00 00           jmp    0x3aa
  25:    0f 1f 44 00 00           nopl   0x0(%rax,%rax,1)
* 2a:    0f 01 ca                 clac <-- trapping instruction
  2d:    fc                       cld
  2e:    6a ff                    push   $0xffffffffffffffff
  30:    e8 85 02 00 00           call   0x2ba
  35:    48 89 c4                 mov    %rax,%rsp
  38:    48 89 e7                 mov    %rsp,%rdi
  3b:    e8 ea 21 f3 ff           call   0xfff3222a
======================================================

Please note that there is a potentially related crash found by syzbot
(https://syzkaller.appspot.com/bug?extid=589d998651a580e6135d) and
patched (https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bb88f9695460bec25aa30ba9072595025cf6c8af).
This patch seems to be present in the kernel we analyzed.

Please let us know if we should provide any additional information!

Wishing you a nice day!

Best,
Marius

Download attachment "repro.syz" of type "application/octet-stream" (393 bytes)

Download attachment "config-5.15.159" of type "application/octet-stream" (227076 bytes)

View attachment "repro.c" of type "text/x-csrc" (6451 bytes)