| 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
| ||
|
Message-ID: <f143792b-c107-4048-8c78-7fa06643ce40@gmail.com>
Date: Mon, 25 Nov 2024 15:07:50 +0100
From: Eric Dumazet <eric.dumazet@...il.com>
To: Takashi Iwai <tiwai@...e.de>
Cc: syzbot <syzbot+351f8764833934c68836@...kaller.appspotmail.com>,
linux-kernel@...r.kernel.org, linux-sound@...r.kernel.org, perex@...ex.cz,
syzkaller-bugs@...glegroups.com, tiwai@...e.com
Subject: Re: [syzbot] [sound?] BUG: sleeping function called from invalid
context in vfree
On 11/25/24 2:34 PM, Takashi Iwai wrote:
> On Mon, 25 Nov 2024 13:58:09 +0100,
> Eric Dumazet wrote:
>>
>> On 11/25/24 1:54 PM, syzbot wrote:
>>> Hello,
>>>
>>> syzbot found the following issue on:
>>>
>>> HEAD commit: 9f16d5e6f220 Merge tag 'for-linus' of git://git.kernel.org..
>>> git tree: upstream
>>> console output: https://syzkaller.appspot.com/x/log.txt?x=154bc778580000
>>> kernel config: https://syzkaller.appspot.com/x/.config?x=47cc5fc1922531f
>>> dashboard link: https://syzkaller.appspot.com/bug?extid=351f8764833934c68836
>>> compiler: gcc (Debian 12.2.0-14) 12.2.0, GNU ld (GNU Binutils for Debian) 2.40
>>> userspace arch: i386
>>>
>>> Unfortunately, I don't have any reproducer for this issue yet.
>>>
>>> Downloadable assets:
>>> disk image (non-bootable): https://storage.googleapis.com/syzbot-assets/7feb34a89c2a/non_bootable_disk-9f16d5e6.raw.xz
>>> vmlinux: https://storage.googleapis.com/syzbot-assets/dddd5525a287/vmlinux-9f16d5e6.xz
>>> kernel image: https://storage.googleapis.com/syzbot-assets/7c9e082ba50b/bzImage-9f16d5e6.xz
>>>
>>> IMPORTANT: if you fix the issue, please add the following tag to the commit:
>>> Reported-by: syzbot+351f8764833934c68836@...kaller.appspotmail.com
>>>
>>> BUG: sleeping function called from invalid context at mm/vmalloc.c:3359
>>> in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 6938, name: syz.1.230
>>> preempt_count: 1, expected: 0
>>> RCU nest depth: 0, expected: 0
>>> 2 locks held by syz.1.230/6938:
>>> #0: ffff88804a2d2968 (&rmidi->open_mutex){+.+.}-{4:4}, at: class_mutex_constructor include/linux/mutex.h:201 [inline]
>>> #0: ffff88804a2d2968 (&rmidi->open_mutex){+.+.}-{4:4}, at: snd_rawmidi_output_params+0x51/0x1c0 sound/core/rawmidi.c:748
>>> #1: ffff888045e5ed48 (&substream->lock){....}-{3:3}, at: spin_lock_irq include/linux/spinlock.h:376 [inline]
>>> #1: ffff888045e5ed48 (&substream->lock){....}-{3:3}, at: class_spinlock_irq_constructor include/linux/spinlock.h:565 [inline]
>>> #1: ffff888045e5ed48 (&substream->lock){....}-{3:3}, at: resize_runtime_buffer+0x243/0x530 sound/core/rawmidi.c:727
>>> irq event stamp: 1218
>>> hardirqs last enabled at (1217): [<ffffffff8b189e32>] __raw_spin_unlock_irqrestore include/linux/spinlock_api_smp.h:151 [inline]
>>> hardirqs last enabled at (1217): [<ffffffff8b189e32>] _raw_spin_unlock_irqrestore+0x52/0x80 kernel/locking/spinlock.c:194
>>> hardirqs last disabled at (1218): [<ffffffff8b189ad5>] __raw_spin_lock_irq include/linux/spinlock_api_smp.h:117 [inline]
>>> hardirqs last disabled at (1218): [<ffffffff8b189ad5>] _raw_spin_lock_irq+0x45/0x50 kernel/locking/spinlock.c:170
>>> softirqs last enabled at (0): [<ffffffff814e57ac>] copy_process+0x1e9c/0x6f20 kernel/fork.c:2321
>>> softirqs last disabled at (0): [<0000000000000000>] 0x0
>>> Preemption disabled at:
>>> [<0000000000000000>] 0x0
>>> CPU: 1 UID: 0 PID: 6938 Comm: syz.1.230 Not tainted 6.12.0-syzkaller-09073-g9f16d5e6f220 #0
>>> Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
>>> Call Trace:
>>> <TASK>
>>> __dump_stack lib/dump_stack.c:94 [inline]
>>> dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120
>>> __might_resched+0x3c0/0x5e0 kernel/sched/core.c:8758
>>> vfree+0x75/0x890 mm/vmalloc.c:3359
>>> kvfree+0x33/0x50 mm/util.c:699
>>> resize_runtime_buffer+0x3a7/0x530 sound/core/rawmidi.c:736
>>> snd_rawmidi_output_params+0xeb/0x1c0 sound/core/rawmidi.c:751
>>> snd_rawmidi_ioctl_params_compat sound/core/rawmidi_compat.c:37 [inline]
>>> snd_rawmidi_ioctl_compat+0x3ac/0x3e0 sound/core/rawmidi_compat.c:120
>>> __do_compat_sys_ioctl+0x1cb/0x2c0 fs/ioctl.c:1004
>>> do_syscall_32_irqs_on arch/x86/entry/common.c:165 [inline]
>>> __do_fast_syscall_32+0x73/0x120 arch/x86/entry/common.c:386
>>> do_fast_syscall_32+0x32/0x80 arch/x86/entry/common.c:411
>>> entry_SYSENTER_compat_after_hwframe+0x84/0x8e
>>> RIP: 0023:0xf7f01579
>>> Code: b8 01 10 06 03 74 b4 01 10 07 03 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d b4 26 00 00 00 00 8d b4 26 00 00 00 00
>>> RSP: 002b:00000000f506555c EFLAGS: 00000296 ORIG_RAX: 0000000000000036
>>> RAX: ffffffffffffffda RBX: 0000000000000007 RCX: 00000000c0205710
>>> RDX: 0000000020000100 RSI: 0000000000000000 RDI: 0000000000000000
>>> RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
>>> R10: 0000000000000000 R11: 0000000000000296 R12: 0000000000000000
>>> R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
>>> </TASK>
>>> ----------------
>>> Code disassembly (best guess), 2 bytes skipped:
>>> 0: 10 06 adc %al,(%rsi)
>>> 2: 03 74 b4 01 add 0x1(%rsp,%rsi,4),%esi
>>> 6: 10 07 adc %al,(%rdi)
>>> 8: 03 74 b0 01 add 0x1(%rax,%rsi,4),%esi
>>> c: 10 08 adc %cl,(%rax)
>>> e: 03 74 d8 01 add 0x1(%rax,%rbx,8),%esi
>>> 1e: 00 51 52 add %dl,0x52(%rcx)
>>> 21: 55 push %rbp
>>> 22: 89 e5 mov %esp,%ebp
>>> 24: 0f 34 sysenter
>>> 26: cd 80 int $0x80
>>> * 28: 5d pop %rbp <-- trapping instruction
>>> 29: 5a pop %rdx
>>> 2a: 59 pop %rcx
>>> 2b: c3 ret
>>> 2c: 90 nop
>>> 2d: 90 nop
>>> 2e: 90 nop
>>> 2f: 90 nop
>>> 30: 8d b4 26 00 00 00 00 lea 0x0(%rsi,%riz,1),%esi
>>> 37: 8d b4 26 00 00 00 00 lea 0x0(%rsi,%riz,1),%esi
>>>
>>>
>>> ---
>>> This report is generated by a bot. It may contain errors.
>>> See https://goo.gl/tpsmEJ for more information about syzbot.
>>> syzbot engineers can be reached at syzkaller@...glegroups.com.
>>>
>>> syzbot will keep track of this issue. See:
>>> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
>>>
>>> If the report is already addressed, let syzbot know by replying with:
>>> #syz fix: exact-commit-title
>>>
>>> If you want to overwrite report's subsystems, reply with:
>>> #syz set subsystems: new-subsystem
>>> (See the list of subsystem names on the web dashboard)
>>>
>>> If the report is a duplicate of another one, reply with:
>>> #syz dup: exact-subject-of-another-report
>>>
>>> If you want to undo deduplication, reply with:
>>> #syz undup
>>
>> Bug added in
>>
>> commit 84bb065b316e8367e14a8824a8f4d21056b10c53
>> Author: Takashi Iwai <tiwai@...e.de>
>> Date: Tue Feb 27 09:52:51 2024 +0100
>>
>> ALSA: rawmidi: Use guard() for locking
>>
>> We can simplify the code gracefully with new guard() macro and co for
>> automatic cleanup of locks.
>>
>> There are a few remaining explicit mutex and spinlock calls, and those
>> are the places where the temporary unlock/relocking happens -- which
>> guard() doens't cover well yet.
>>
>> Only the code refactoring, and no functional changes.
>>
>> Signed-off-by: Takashi Iwai <tiwai@...e.de>
>> Link:https://lore.kernel.org/r/20240227085306.9764-10-tiwai@suse.de
>> <https://lore.kernel.org/r/20240227085306.9764-10-tiwai@suse.de>
> Hmm, through a quick glance, I don't see anything breaking there.
>
> Could you tell me which code path introduced the spinlock held during
> kvzalloc() call in resize_runtime_buffer() in sound/core/rawmidi.c by
> this commit...?
@@ -744,9 +724,8 @@ static int resize_runtime_buffer(struct
snd_rawmidi_substream *substream,
newbuf = kvzalloc(params->buffer_size, GFP_KERNEL);
if (!newbuf)
return -ENOMEM;
- spin_lock_irq(&substream->lock);
+ guard(spinlock_irq)(&substream->lock);
if (runtime->buffer_ref) {
- spin_unlock_irq(&substream->lock);
kvfree(newbuf);
return -EBUSY;
}
@@ -754,7 +733,6 @@ static int resize_runtime_buffer(struct
snd_rawmidi_substream *substream,
runtime->buffer = newbuf;
runtime->buffer_size = params->buffer_size;
__reset_runtime_ptrs(runtime, is_input);
- spin_unlock_irq(&substream->lock);
kvfree(oldbuf);
}
runtime->avail_min = params->avail_min;
You can see that after your patch, kvfree() is called while
guard(spinlock_irq)(&substream->lock); is still holding the lock.
Powered by blists - more mailing lists