Re: KMSAN: uninit-value in snd_midi_event_encode_byte

From: Dmitry Vyukov
Date: Mon Sep 03 2018 - 10:54:48 EST


On Mon, Sep 3, 2018 at 4:30 PM, Takashi Iwai <tiwai@xxxxxxx> wrote:
> On Mon, 03 Sep 2018 16:19:23 +0200,
> Dmitry Vyukov wrote:
>>
>> On Mon, Sep 3, 2018 at 4:04 PM, Takashi Iwai <tiwai@xxxxxxx> wrote:
>> > On Mon, 03 Sep 2018 15:41:36 +0200,
>> > Dmitry Vyukov wrote:
>> >>
>> >> On Mon, Sep 3, 2018 at 3:37 PM, Takashi Iwai <tiwai@xxxxxxx> wrote:
>> >> > On Mon, 03 Sep 2018 15:33:12 +0200,
>> >> > Dmitry Vyukov wrote:
>> >> >>
>> >> >> On Mon, Sep 3, 2018 at 3:29 PM, Takashi Iwai <tiwai@xxxxxxx> wrote:
>> >> >> > On Mon, 03 Sep 2018 15:22:42 +0200,
>> >> >> > syzbot wrote:
>> >> >> >>
>> >> >> >> > On Mon, 03 Sep 2018 03:53:03 +0200,
>> >> >> >> > syzbot wrote:
>> >> >> >>
>> >> >> >> >> syzbot has found a reproducer for the following crash on:
>> >> >> >>
>> >> >> >> >> HEAD commit: 28f0ca98eadf kmsan: don't instrument
>> >> >> >> >> do_syscall_64() and _..
>> >> >> >> >> git tree: https://github.com/google/kmsan.git/master
>> >> >> >> >> console output: https://syzkaller.appspot.com/x/log.txt?x=10556c92400000
>> >> >> >> >> kernel config:
>> >> >> >> >> https://syzkaller.appspot.com/x/.config?x=3431f03869413153
>> >> >> >> >> dashboard link:
>> >> >> >> >> https://syzkaller.appspot.com/bug?extid=194dffdb8b22fc5d207a
>> >> >> >> >> compiler: clang version 8.0.0 (trunk 339414)
>> >> >> >> >> syz repro:
>> >> >> >> >> https://syzkaller.appspot.com/x/repro.syz?x=123a520a400000
>> >> >> >> >> C reproducer: https://syzkaller.appspot.com/x/repro.c?x=16068dc1400000
>> >> >> >>
>> >> >> >> >> IMPORTANT: if you fix the bug, please add the following tag to the
>> >> >> >> >> commit:
>> >> >> >> >> Reported-by: syzbot+194dffdb8b22fc5d207a@xxxxxxxxxxxxxxxxxxxxxxxxx
>> >> >> >>
>> >> >> >> >> ==================================================================
>> >> >> >> >> BUG: KMSAN: uninit-value in snd_midi_event_encode_byte+0x569/0xff0
>> >> >> >> >> sound/core/seq/seq_midi_event.c:195
>> >> >> >> >> CPU: 1 PID: 1659 Comm: kworker/1:1H Not tainted 4.19.0-rc1+ #40
>> >> >> >> >> Hardware name: Google Google Compute Engine/Google Compute Engine,
>> >> >> >> >> BIOS Google 01/01/2011
>> >> >> >> >> Workqueue: events_highpri snd_vmidi_output_work
>> >> >> >> >> Call Trace:
>> >> >> >> >> __dump_stack lib/dump_stack.c:77 [inline]
>> >> >> >> >> dump_stack+0x14b/0x190 lib/dump_stack.c:113
>> >> >> >> >> kmsan_report+0x183/0x2b0 mm/kmsan/kmsan.c:956
>> >> >> >> >> __msan_warning+0x70/0xc0 mm/kmsan/kmsan_instr.c:645
>> >> >> >> >> snd_midi_event_encode_byte+0x569/0xff0
>> >> >> >> >> sound/core/seq/seq_midi_event.c:195
>> >> >> >> >> snd_vmidi_output_work+0x34e/0x5b0 sound/core/seq/seq_virmidi.c:161
>> >> >> >> >> process_one_work+0x1605/0x1f40 kernel/workqueue.c:2153
>> >> >> >> >> worker_thread+0x11a2/0x2590 kernel/workqueue.c:2296
>> >> >> >> >> kthread+0x465/0x4a0 kernel/kthread.c:247
>> >> >> >> >> ret_from_fork+0x35/0x40 arch/x86/entry/entry_64.S:416
>> >> >> >>
>> >> >> >> >> Uninit was stored to memory at:
>> >> >> >> >> kmsan_save_stack_with_flags mm/kmsan/kmsan.c:256 [inline]
>> >> >> >> >> kmsan_save_stack mm/kmsan/kmsan.c:271 [inline]
>> >> >> >> >> kmsan_internal_chain_origin+0x128/0x210 mm/kmsan/kmsan.c:573
>> >> >> >> >> __msan_chain_origin+0x69/0xc0 mm/kmsan/kmsan_instr.c:482
>> >> >> >> >> __snd_rawmidi_transmit_peek sound/core/rawmidi.c:1103 [inline]
>> >> >> >> >> snd_rawmidi_transmit+0xa75/0xbf0 sound/core/rawmidi.c:1228
>> >> >> >> >> snd_vmidi_output_work+0x2ac/0x5b0 sound/core/seq/seq_virmidi.c:159
>> >> >> >> >> process_one_work+0x1605/0x1f40 kernel/workqueue.c:2153
>> >> >> >> >> worker_thread+0x11a2/0x2590 kernel/workqueue.c:2296
>> >> >> >> >> kthread+0x465/0x4a0 kernel/kthread.c:247
>> >> >> >> >> ret_from_fork+0x35/0x40 arch/x86/entry/entry_64.S:416
>> >> >> >>
>> >> >> >> >> Uninit was created at:
>> >> >> >> >> kmsan_save_stack_with_flags mm/kmsan/kmsan.c:256 [inline]
>> >> >> >> >> kmsan_internal_poison_shadow+0xb8/0x1b0 mm/kmsan/kmsan.c:181
>> >> >> >> >> kmsan_kmalloc+0x98/0x100 mm/kmsan/kmsan_hooks.c:91
>> >> >> >> >> __kmalloc_node+0x7bf/0x11c0 mm/slub.c:3828
>> >> >> >> >> kmalloc_node include/linux/slab.h:555 [inline]
>> >> >> >> >> kvmalloc_node+0x19d/0x3e0 mm/util.c:423
>> >> >> >> >> kvmalloc include/linux/mm.h:577 [inline]
>> >> >> >> >> snd_rawmidi_runtime_create sound/core/rawmidi.c:132 [inline]
>> >> >> >> >> open_substream+0x3c8/0xaa0 sound/core/rawmidi.c:276
>> >> >> >> >> rawmidi_open_priv+0x347/0x1000 sound/core/rawmidi.c:327
>> >> >> >> >> snd_rawmidi_open+0x7d4/0x1120 sound/core/rawmidi.c:424
>> >> >> >> >> soundcore_open+0x9be/0xa60 sound/sound_core.c:597
>> >> >> >> >> chrdev_open+0xc26/0xdb0 fs/char_dev.c:417
>> >> >> >> >> do_dentry_open+0xce6/0x1740 fs/open.c:771
>> >> >> >> >> vfs_open+0xaf/0xe0 fs/open.c:880
>> >> >> >> >> do_last fs/namei.c:3418 [inline]
>> >> >> >> >> path_openat+0x1799/0x6870 fs/namei.c:3534
>> >> >> >> >> do_filp_open+0x259/0x610 fs/namei.c:3564
>> >> >> >> >> do_sys_open+0x630/0x940 fs/open.c:1063
>> >> >> >> >> __do_sys_open fs/open.c:1081 [inline]
>> >> >> >> >> __se_sys_open+0xad/0xc0 fs/open.c:1076
>> >> >> >> >> __x64_sys_open+0x4a/0x70 fs/open.c:1076
>> >> >> >> >> do_syscall_64+0xb8/0x100 arch/x86/entry/common.c:291
>> >> >> >> >> entry_SYSCALL_64_after_hwframe+0x63/0xe7
>> >> >> >> >> ==================================================================
>> >> >> >>
>> >> >> >> > This looks like some small race at virmidi reading and the buffer
>> >> >> >> > handling, which should be almost harmless by itself.
>> >> >> >> > Nevertheless, the fix should be easy, just replacing kvmalloc() with
>> >> >> >> > kvzalloc(), as below.
>> >> >> >>
>> >> >> >> > Let's check whether this works.
>> >> >> >>
>> >> >> >> > #syz test:
>> >> >> >> > git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git
>> >> >> >> > topic/rawmidi-fixes
>> >> >> >>
>> >> >> >> KMSAN bugs can only be tested on https://github.com/google/kmsan.git tree
>> >> >> >> because KMSAN tool is not upstreamed yet.
>> >> >> >> See
>> >> >> >> https://github.com/google/syzkaller/blob/master/docs/syzbot.md#kmsan-bugs
>> >> >> >> for details.
>> >> >> >
>> >> >> > So how is the procedure to test a patch for bugs spotted by KMSAN?
>> >> >> >
>> >> >> > In this case, the fix is trivial, so I can put it to my tree for the
>> >> >> > next pull request. Of course, it'd be better if it can be confirmed
>> >> >> > to work beforehand...
>> >> >>
>> >> >> https://github.com/google/syzkaller/blob/master/docs/syzbot.md#kmsan-bugs
>> >> >>
>> >> >> Which means you need to attach a patch because it's not in that tree.
>> >> >
>> >> > OK, it would have been helpful if the description there was more clear
>> >> > :)
>> >> >
>> >> > And no automated way for that?
>> >>
>> >> What exactly do you mean? Please elaborate.
>> >
>> > Well, it's not clear which action is required to let syzbot testing
>> > some patch. The description says:
>> >
>> > "KMSAN is not upstream yet, though, we want to upstream it later. For
>> > now, it lives in github.com/google/kmsan and is based on a
>> > reasonably fresh upstream tree. As the result, any patch testing
>> > requests for KMSAN bugs need to go to KMSAN tree
>> > (https://github.com/google/kmsan.git repo, master branch)."
>> >
>> > Here, "need to go to KMSAN tree" part doesn't explain what exactly to
>> > do. Remember that many kernel devs (including the man in the mirror)
>> > don't use github, at least, not dealing github PRs for kernel
>> > development itself, so it's a pretty unfamiliar process.
>>
>> It's not patches that need to go into that tree (i.e. no PRs). It's
>> patch testing requests.
>>
>> Tried to clarify doc:
>>
>> https://github.com/google/syzkaller/commit/bfe906cbb9c73c326f24327e1a89bc0cd0c314c7#diff-5b3b5ff5f03b01e1d31ec93aafd2f3d5
>>
>> Does this look better?
>
> Erm, sorry, still unclear to me. I'd need to explicitl attch/inline
> the patch for testing *to where*? Attach the test patch to the e-mail
> replying to the bug?
>
> Also, now you've put an example referring to
> https://github.com/google/kmsan.git master. What does it mean?
> I should have put that line in my reply?
>
>> >> > I mean, I already attached patch in
>> >> > the previous reply (together with syz-magic command to point the git
>> >> > branch URL). Do I need anything else?
>> >>
>> >> Just issue the command against "https://github.com/google/kmsan.git
>> >> master" and attach the patch.
>> >
>> > But it's basically a fix for the upstream tree, not about KMSAN
>> > itself. A PR to that tree doesn't sound right to me...
>>
>> You don't need to send a PR. You attach a patch (or inline in email)
>> and request to apply and test it in KMSAN tree.
>
> Then I must have understood "just issue the command against..."
> Does it mean the "syz test: https//github.com/google/kmsan.git
> master"?
>
> I need to learn the basic for understanding the black magic...

https://github.com/google/syzkaller/blob/master/docs/syzbot.md#kmsan-bugs
is meant to be a complement to the previous generic descriptions of
patch testing process, which is:
https://github.com/google/syzkaller/blob/master/docs/syzbot.md#testing-patches

So, you always reply to the syzbot+HASH email address so that syzbot
understands which bug we are talking about.
Then you do either (omitting # so that syzbot won't consider that as
actual test requests):

syz test: git://repo/address.git branch

or:

syz test: git://repo/address.git commit-hash

And then you can either attach a patch that needs to be applied on
top, or not attach it (if it's already in the tree, or you just want
to get another crash report).

For KMSAN you need to issue test request against
"https//github.com/google/kmsan.git master" (that's the only tree that
has KMSAN tool in it) and you need to attach/inline the patch (because
your patch is obviously not there yet).

Does this make things more clear?