possible deadlock in vfs_fallocate

[syzbot]

Hello,

syzbot hit the following crash on upstream commit
cdface5209349930ae1b51338763c8e029971b97 (Sun Apr 29 03:07:21 2018 +0000)
Merge tag 'for_linus_stable' of
git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
syzbot dashboard link:
https://syzkaller.appspot.com/bug?extid=148c2885d71194f18d28

C reproducer: https://syzkaller.appspot.com/x/repro.c?id=5054004375584768
syzkaller reproducer:
https://syzkaller.appspot.com/x/repro.syz?id=6438048191479808
Raw console output:
https://syzkaller.appspot.com/x/log.txt?id=5404215203594240
Kernel config:
https://syzkaller.appspot.com/x/.config?id=7043958930931867332
compiler: gcc (GCC) 8.0.1 20180413 (experimental)

IMPORTANT: if you fix the bug, please add the following tag to the commit:
Reported-by: syzbot+148c28...@syzkaller.appspotmail.com
It will help syzbot understand when the bug is fixed. See footer for
details.
If you forward the report, please keep this part and the footer.

random: sshd: uninitialized urandom read (32 bytes read)
random: sshd: uninitialized urandom read (32 bytes read)
random: sshd: uninitialized urandom read (32 bytes read)

======================================================
WARNING: possible circular locking dependency detected
4.17.0-rc2+ #23 Not tainted
------------------------------------------------------
syz-executor715/4492 is trying to acquire lock:
(ptrval) (sb_writers#6){.+.+}, at: file_start_write
include/linux/fs.h:2718 [inline]
(ptrval) (sb_writers#6){.+.+}, at: vfs_fallocate+0x5be/0x8d0
fs/open.c:318

but task is already holding lock:
(ptrval) (ashmem_mutex){+.+.}, at: ashmem_shrink_scan+0xac/0x560
drivers/staging/android/ashmem.c:440

which lock already depends on the new lock.


the existing dependency chain (in reverse order) is:

-> #3 (ashmem_mutex){+.+.}:
__mutex_lock_common kernel/locking/mutex.c:756 [inline]
__mutex_lock+0x16d/0x17f0 kernel/locking/mutex.c:893
mutex_lock_nested+0x16/0x20 kernel/locking/mutex.c:908
ashmem_mmap+0x53/0x460 drivers/staging/android/ashmem.c:361
call_mmap include/linux/fs.h:1789 [inline]
mmap_region+0xd13/0x1820 mm/mmap.c:1723
do_mmap+0xc79/0x11d0 mm/mmap.c:1494
do_mmap_pgoff include/linux/mm.h:2237 [inline]
vm_mmap_pgoff+0x1fb/0x2a0 mm/util.c:357
ksys_mmap_pgoff+0x4c9/0x640 mm/mmap.c:1544
__do_sys_mmap arch/x86/kernel/sys_x86_64.c:100 [inline]
__se_sys_mmap arch/x86/kernel/sys_x86_64.c:91 [inline]
__x64_sys_mmap+0xe9/0x1b0 arch/x86/kernel/sys_x86_64.c:91
do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #2 (&mm->mmap_sem){++++}:
__might_fault+0x155/0x1e0 mm/memory.c:4555
_copy_to_user+0x30/0x110 lib/usercopy.c:25
copy_to_user include/linux/uaccess.h:155 [inline]
filldir+0x1ea/0x3a0 fs/readdir.c:196
dir_emit_dot include/linux/fs.h:3378 [inline]
dir_emit_dots include/linux/fs.h:3389 [inline]
dcache_readdir+0x13a/0x620 fs/libfs.c:192
iterate_dir+0x4b0/0x5d0 fs/readdir.c:51
__do_sys_getdents fs/readdir.c:231 [inline]
__se_sys_getdents fs/readdir.c:212 [inline]
__x64_sys_getdents+0x293/0x4e0 fs/readdir.c:212
do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #1 (&sb->s_type->i_mutex_key#11){++++}:
down_write+0x87/0x120 kernel/locking/rwsem.c:70
inode_lock include/linux/fs.h:713 [inline]
do_last fs/namei.c:3274 [inline]
path_openat+0x123b/0x4e20 fs/namei.c:3501
do_filp_open+0x249/0x350 fs/namei.c:3535
do_sys_open+0x56f/0x740 fs/open.c:1093
__do_sys_open fs/open.c:1111 [inline]
__se_sys_open fs/open.c:1106 [inline]
__x64_sys_open+0x7e/0xc0 fs/open.c:1106
do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x49/0xbe

-> #0 (sb_writers#6){.+.+}:
lock_acquire+0x1dc/0x520 kernel/locking/lockdep.c:3920
percpu_down_read_preempt_disable include/linux/percpu-rwsem.h:36
[inline]
percpu_down_read include/linux/percpu-rwsem.h:59 [inline]
__sb_start_write+0x1e9/0x300 fs/super.c:1385
file_start_write include/linux/fs.h:2718 [inline]
vfs_fallocate+0x5be/0x8d0 fs/open.c:318
ashmem_shrink_scan+0x1f1/0x560 drivers/staging/android/ashmem.c:447
ashmem_ioctl+0x3bf/0x13a0 drivers/staging/android/ashmem.c:789
vfs_ioctl fs/ioctl.c:46 [inline]
file_ioctl fs/ioctl.c:500 [inline]
do_vfs_ioctl+0x1cf/0x16a0 fs/ioctl.c:684
ksys_ioctl+0xa9/0xd0 fs/ioctl.c:701
__do_sys_ioctl fs/ioctl.c:708 [inline]
__se_sys_ioctl fs/ioctl.c:706 [inline]
__x64_sys_ioctl+0x73/0xb0 fs/ioctl.c:706
do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x49/0xbe

other info that might help us debug this:

Chain exists of:
sb_writers#6 --> &mm->mmap_sem --> ashmem_mutex

Possible unsafe locking scenario:

CPU0 CPU1
---- ----
lock(ashmem_mutex);
lock(&mm->mmap_sem);
lock(ashmem_mutex);
lock(sb_writers#6);

*** DEADLOCK ***

1 lock held by syz-executor715/4492:
#0: (ptrval) (ashmem_mutex){+.+.}, at:
ashmem_shrink_scan+0xac/0x560 drivers/staging/android/ashmem.c:440

stack backtrace:
CPU: 1 PID: 4492 Comm: syz-executor715 Not tainted 4.17.0-rc2+ #23
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS
Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x1b9/0x294 lib/dump_stack.c:113
print_circular_bug.isra.36.cold.54+0x1bd/0x27d
kernel/locking/lockdep.c:1223
check_prev_add kernel/locking/lockdep.c:1863 [inline]
check_prevs_add kernel/locking/lockdep.c:1976 [inline]
validate_chain kernel/locking/lockdep.c:2417 [inline]
__lock_acquire+0x343e/0x5140 kernel/locking/lockdep.c:3431
lock_acquire+0x1dc/0x520 kernel/locking/lockdep.c:3920
percpu_down_read_preempt_disable include/linux/percpu-rwsem.h:36 [inline]
percpu_down_read include/linux/percpu-rwsem.h:59 [inline]
__sb_start_write+0x1e9/0x300 fs/super.c:1385
file_start_write include/linux/fs.h:2718 [inline]
vfs_fallocate+0x5be/0x8d0 fs/open.c:318
ashmem_shrink_scan+0x1f1/0x560 drivers/staging/android/ashmem.c:447
ashmem_ioctl+0x3bf/0x13a0 drivers/staging/android/ashmem.c:789
vfs_ioctl fs/ioctl.c:46 [inline]
file_ioctl fs/ioctl.c:500 [inline]
do_vfs_ioctl+0x1cf/0x16a0 fs/ioctl.c:684
ksys_ioctl+0xa9/0xd0 fs/ioctl.c:701
__do_sys_ioctl fs/ioctl.c:708 [inline]
__se_sys_ioctl fs/ioctl.c:706 [inline]
__x64_sys_ioctl+0x73/0xb0 fs/ioctl.c:706
do_syscall_64+0x1b1/0x800 arch/x86/entry/common.c:287
entry_SYSCALL_64_after_hwframe+0x49/0xbe
RIP: 0033:0x440179
RSP: 002b:00007ffc165d4a28 EFLAGS: 00000217 ORIG_RAX: 0000000000000010
RAX: ffffffffffffffda RBX: 6873612f7665642f RCX: 0000000000440179
RDX: 0000000000000000 RSI: 000000000000770a RDI: 0000000000000004
RBP: 00000000006ca018 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000004 R11: 0000000000000217 R12: 00000000004016a0
R13: 0000000000401730 R14: 0000000000000000 R15: 0000000000000000
random: crng init done


---
This bug is generated by a dumb bot. It may contain errors.
See https://goo.gl/tpsmEJ for details.
Direct all questions to syzk...@googlegroups.com.

syzbot will keep track of this bug report.
If you forgot to add the Reported-by tag, once the fix for this bug is
merged
into any tree, please reply to this email with:
#syz fix: exact-commit-title
If you want to test a patch for this bug, please reply with:
#syz test: git://repo/address.git branch
and provide the patch inline or as an attachment.
To mark this as a duplicate of another syzbot report, please reply with:
#syz dup: exact-subject-of-another-report
If it's a one-off invalid bug report, please reply with:
#syz invalid
Note: if the crash happens again, it will cause creation of a new bug
report.
Note: all commands must start from beginning of the line in the email body.

[Eric Biggers]

[+ashmem maintainers]

Looks like yet another locking bug in ashmem. ashmem_mutex normally ranks below
mmap_sem, but ashmem_shrink_scan() inverts that order, via the
file_start_write() in vfs_fallocate().

- Eric

[Todd Kjos]

+Joel Fernandes 

[Dmitry Vyukov]

On Wed, May 9, 2018 at 6:57 PM 'Todd Kjos' via syzkaller-bugs
<syzkall...@googlegroups.com> wrote:
>
> +Joel Fernandes
>
> On Wed, May 9, 2018 at 12:55 AM Eric Biggers <ebig...@gmail.com> wrote:
>>
>> [+ashmem maintainers]
>>
>> On Sun, Apr 29, 2018 at 10:00:03AM -0700, syzbot wrote:
>> > Hello,
>> >
>> > syzbot hit the following crash on upstream commit
>> > cdface5209349930ae1b51338763c8e029971b97 (Sun Apr 29 03:07:21 2018 +0000)
>> > Merge tag 'for_linus_stable' of
>> > git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4
>> > syzbot dashboard link:
>> > https://syzkaller.appspot.com/bug?extid=148c2885d71194f18d28
>> >
>> > C reproducer: https://syzkaller.appspot.com/x/repro.c?id=5054004375584768
>> > syzkaller reproducer:
>> > https://syzkaller.appspot.com/x/repro.syz?id=6438048191479808
>> > Raw console output:
>> > https://syzkaller.appspot.com/x/log.txt?id=5404215203594240
>> > Kernel config:
>> > https://syzkaller.appspot.com/x/.config?id=7043958930931867332
>> > compiler: gcc (GCC) 8.0.1 20180413 (experimental)

Let's test Tetsuo's patch

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
master

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger
crash:

Reported-and-tested-by:
syzbot+148c28...@syzkaller.appspotmail.com

Tested on:

commit: 48b161983ae5 Merge tag 'xarray-5.0-rc3' of git://git.infra..
git tree: upstream
kernel config: https://syzkaller.appspot.com/x/.config?x=dd3a020d055fa2e8
compiler: gcc (GCC) 9.0.0 20181231 (experimental)
patch: https://syzkaller.appspot.com/x/patch.diff?x=16e6d908c00000

Note: testing is done by a robot and is best-effort only.

[Tetsuo Handa]

#syz test: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master

---
drivers/staging/android/ashmem.c | 23 ++++++++++++++++++-----
1 file changed, 18 insertions(+), 5 deletions(-)

diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c
index 90a8a9f..d40c1d2 100644
--- a/drivers/staging/android/ashmem.c
+++ b/drivers/staging/android/ashmem.c
@@ -75,6 +75,9 @@ struct ashmem_range {
/* LRU list of unpinned pages, protected by ashmem_mutex */
static LIST_HEAD(ashmem_lru_list);

+static atomic_t ashmem_shrink_inflight = ATOMIC_INIT(0);
+static DECLARE_WAIT_QUEUE_HEAD(ashmem_shrink_wait);
+
/*
* long lru_count - The count of pages on our LRU list.
*
@@ -438,7 +441,6 @@ static int ashmem_mmap(struct file *file, struct vm_area_struct *vma)
static unsigned long
ashmem_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
{
- struct ashmem_range *range, *next;
unsigned long freed = 0;

/* We might recurse into filesystem code, so bail out if necessary */
@@ -448,17 +450,27 @@ static int ashmem_mmap(struct file *file, struct vm_area_struct *vma)
if (!mutex_trylock(&ashmem_mutex))
return -1;

- list_for_each_entry_safe(range, next, &ashmem_lru_list, lru) {
+ while (!list_empty(&ashmem_lru_list)) {
+ struct ashmem_range *range =
+ list_first_entry(&ashmem_lru_list, typeof(*range), lru);
loff_t start = range->pgstart * PAGE_SIZE;
loff_t end = (range->pgend + 1) * PAGE_SIZE;
+ struct file *f = range->asma->file;

- range->asma->file->f_op->fallocate(range->asma->file,
- FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
- start, end - start);
+ get_file(f);
+ atomic_inc(&ashmem_shrink_inflight);
range->purged = ASHMEM_WAS_PURGED;
lru_del(range);

freed += range_size(range);
+ mutex_unlock(&ashmem_mutex);
+ f->f_op->fallocate(f,
+ FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
+ start, end - start);
+ fput(f);
+ if (atomic_dec_and_test(&ashmem_shrink_inflight))
+ wake_up_all(&ashmem_shrink_wait);
+ mutex_lock(&ashmem_mutex);
if (--sc->nr_to_scan <= 0)
break;
}
@@ -713,6 +725,7 @@ static int ashmem_pin_unpin(struct ashmem_area *asma, unsigned long cmd,
return -EFAULT;

mutex_lock(&ashmem_mutex);
+ wait_event(ashmem_shrink_wait, !atomic_read(&ashmem_shrink_inflight));

if (!asma->file)
goto out_unlock;
--
1.8.3.1

[syzbot]

Hello,

syzbot has tested the proposed patch and the reproducer did not trigger
crash:

Reported-and-tested-by:
syzbot+148c28...@syzkaller.appspotmail.com

Tested on:

commit: d73aba1115cf Merge tag 'drm-fixes-2019-01-25-1' of git://a..

git tree: upstream
kernel config: https://syzkaller.appspot.com/x/.config?x=dd3a020d055fa2e8
compiler: gcc (GCC) 9.0.0 20181231 (experimental)

patch: https://syzkaller.appspot.com/x/patch.diff?x=162b5d5f400000