* [syzbot] possible deadlock in split_huge_page_to_list
@ 2021-12-15 1:03 syzbot
2021-12-15 4:19 ` Matthew Wilcox
0 siblings, 1 reply; 2+ messages in thread
From: syzbot @ 2021-12-15 1:03 UTC (permalink / raw)
To: akpm, dhowells, hughd, linux-kernel, linux-mm, syzkaller-bugs, willy
Hello,
syzbot found the following issue on:
HEAD commit: ea922272cbe5 Add linux-next specific files for 20211210
git tree: linux-next
console output: https://syzkaller.appspot.com/x/log.txt?x=14620bb9b00000
kernel config: https://syzkaller.appspot.com/x/.config?x=c1359a19d2230002
dashboard link: https://syzkaller.appspot.com/bug?extid=c915885f05d8e432e7b4
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2
syz repro: https://syzkaller.appspot.com/x/repro.syz?x=16e100bab00000
C reproducer: https://syzkaller.appspot.com/x/repro.c?x=1123964db00000
The issue was bisected to:
commit 3ebffc96befbaf9de9297b00d67091bb702fad8e
Author: Matthew Wilcox (Oracle) <willy@infradead.org>
Date: Sun Jun 28 02:19:08 2020 +0000
mm: Use multi-index entries in the page cache
bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=1276e4bab00000
final oops: https://syzkaller.appspot.com/x/report.txt?x=1176e4bab00000
console output: https://syzkaller.appspot.com/x/log.txt?x=1676e4bab00000
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+c915885f05d8e432e7b4@syzkaller.appspotmail.com
Fixes: 3ebffc96befb ("mm: Use multi-index entries in the page cache")
======================================================
WARNING: possible circular locking dependency detected
5.16.0-rc4-next-20211210-syzkaller #0 Not tainted
------------------------------------------------------
syz-executor832/3602 is trying to acquire lock:
ffffffff8bca9b40 (fs_reclaim){+.+.}-{0:0}, at: might_alloc include/linux/sched/mm.h:253 [inline]
ffffffff8bca9b40 (fs_reclaim){+.+.}-{0:0}, at: slab_pre_alloc_hook mm/slab.h:739 [inline]
ffffffff8bca9b40 (fs_reclaim){+.+.}-{0:0}, at: slab_alloc_node mm/slub.c:3145 [inline]
ffffffff8bca9b40 (fs_reclaim){+.+.}-{0:0}, at: slab_alloc mm/slub.c:3239 [inline]
ffffffff8bca9b40 (fs_reclaim){+.+.}-{0:0}, at: kmem_cache_alloc+0x3e/0x3a0 mm/slub.c:3244
but task is already holding lock:
ffff888011b43cc8 (&mapping->i_mmap_rwsem){++++}-{3:3}, at: i_mmap_lock_read include/linux/fs.h:507 [inline]
ffff888011b43cc8 (&mapping->i_mmap_rwsem){++++}-{3:3}, at: split_huge_page_to_list+0x533/0x5440 mm/huge_memory.c:2657
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (&mapping->i_mmap_rwsem){++++}-{3:3}:
down_write+0x90/0x150 kernel/locking/rwsem.c:1514
i_mmap_lock_write include/linux/fs.h:492 [inline]
dma_resv_lockdep+0x362/0x55a drivers/dma-buf/dma-resv.c:713
do_one_initcall+0x103/0x650 init/main.c:1303
do_initcall_level init/main.c:1378 [inline]
do_initcalls init/main.c:1394 [inline]
do_basic_setup init/main.c:1413 [inline]
kernel_init_freeable+0x6b1/0x73a init/main.c:1618
kernel_init+0x1a/0x1d0 init/main.c:1507
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
-> #0 (fs_reclaim){+.+.}-{0:0}:
check_prev_add kernel/locking/lockdep.c:3063 [inline]
check_prevs_add kernel/locking/lockdep.c:3186 [inline]
validate_chain kernel/locking/lockdep.c:3801 [inline]
__lock_acquire+0x2a07/0x54a0 kernel/locking/lockdep.c:5027
lock_acquire kernel/locking/lockdep.c:5639 [inline]
lock_acquire+0x1ab/0x510 kernel/locking/lockdep.c:5604
__fs_reclaim_acquire mm/page_alloc.c:4550 [inline]
fs_reclaim_acquire+0x115/0x160 mm/page_alloc.c:4564
might_alloc include/linux/sched/mm.h:253 [inline]
slab_pre_alloc_hook mm/slab.h:739 [inline]
slab_alloc_node mm/slub.c:3145 [inline]
slab_alloc mm/slub.c:3239 [inline]
kmem_cache_alloc+0x3e/0x3a0 mm/slub.c:3244
xas_split_alloc+0x142/0x480 lib/xarray.c:1017
split_huge_page_to_list+0xd35/0x5440 mm/huge_memory.c:2683
split_huge_page include/linux/huge_mm.h:192 [inline]
truncate_inode_partial_folio+0xaed/0xe00 mm/truncate.c:275
shmem_undo_range+0xc43/0x19b0 mm/shmem.c:959
shmem_truncate_range mm/shmem.c:1037 [inline]
shmem_fallocate+0xb11/0xe80 mm/shmem.c:2675
vfs_fallocate+0x48d/0xe10 fs/open.c:308
ksys_fallocate fs/open.c:331 [inline]
__do_sys_fallocate fs/open.c:339 [inline]
__se_sys_fallocate fs/open.c:337 [inline]
__x64_sys_fallocate+0xcf/0x140 fs/open.c:337
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&mapping->i_mmap_rwsem);
lock(fs_reclaim);
lock(&mapping->i_mmap_rwsem);
lock(fs_reclaim);
*** DEADLOCK ***
3 locks held by syz-executor832/3602:
#0: ffff888073d6a460 (sb_writers#3){.+.+}-{0:0}, at: ksys_fallocate fs/open.c:331 [inline]
#0: ffff888073d6a460 (sb_writers#3){.+.+}-{0:0}, at: __do_sys_fallocate fs/open.c:339 [inline]
#0: ffff888073d6a460 (sb_writers#3){.+.+}-{0:0}, at: __se_sys_fallocate fs/open.c:337 [inline]
#0: ffff888073d6a460 (sb_writers#3){.+.+}-{0:0}, at: __x64_sys_fallocate+0xcf/0x140 fs/open.c:337
#1: ffff888011b43a70 (&sb->s_type->i_mutex_key#8){+.+.}-{3:3}, at: inode_lock include/linux/fs.h:777 [inline]
#1: ffff888011b43a70 (&sb->s_type->i_mutex_key#8){+.+.}-{3:3}, at: shmem_fallocate+0x154/0xe80 mm/shmem.c:2651
#2: ffff888011b43cc8 (&mapping->i_mmap_rwsem){++++}-{3:3}, at: i_mmap_lock_read include/linux/fs.h:507 [inline]
#2: ffff888011b43cc8 (&mapping->i_mmap_rwsem){++++}-{3:3}, at: split_huge_page_to_list+0x533/0x5440 mm/huge_memory.c:2657
stack backtrace:
CPU: 0 PID: 3602 Comm: syz-executor832 Not tainted 5.16.0-rc4-next-20211210-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
<TASK>
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0xcd/0x134 lib/dump_stack.c:106
check_noncircular+0x25f/0x2e0 kernel/locking/lockdep.c:2143
check_prev_add kernel/locking/lockdep.c:3063 [inline]
check_prevs_add kernel/locking/lockdep.c:3186 [inline]
validate_chain kernel/locking/lockdep.c:3801 [inline]
__lock_acquire+0x2a07/0x54a0 kernel/locking/lockdep.c:5027
lock_acquire kernel/locking/lockdep.c:5639 [inline]
lock_acquire+0x1ab/0x510 kernel/locking/lockdep.c:5604
__fs_reclaim_acquire mm/page_alloc.c:4550 [inline]
fs_reclaim_acquire+0x115/0x160 mm/page_alloc.c:4564
might_alloc include/linux/sched/mm.h:253 [inline]
slab_pre_alloc_hook mm/slab.h:739 [inline]
slab_alloc_node mm/slub.c:3145 [inline]
slab_alloc mm/slub.c:3239 [inline]
kmem_cache_alloc+0x3e/0x3a0 mm/slub.c:3244
xas_split_alloc+0x142/0x480 lib/xarray.c:1017
split_huge_page_to_list+0xd35/0x5440 mm/huge_memory.c:2683
split_huge_page include/linux/huge_mm.h:192 [inline]
truncate_inode_partial_folio+0xaed/0xe00 mm/truncate.c:275
shmem_undo_range+0xc43/0x19b0 mm/shmem.c:959
shmem_truncate_range mm/shmem.c:1037 [inline]
shmem_fallocate+0xb11/0xe80 mm/shmem.c:2675
vfs_fallocate+0x48d/0xe10 fs/open.c:308
ksys_fallocate fs/open.c:331 [inline]
__do_sys_fallocate fs/open.c:339 [inline]
__se_sys_fallocate fs/open.c:337 [inline]
__x64_sys_fallocate+0xcf/0x140 fs/open.c:337
do_syscall_x64 arch/x86/entry/common.c:50 [inline]
do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f6a5c9e8819
Code: 28 c3 e8 5a 14 00 00 66 2e 0f 1f 84 00 00 00 00 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 c0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007ffcf59ddff8 EFLAGS: 00000246 ORIG_RAX: 000000000000011d
RAX: ffffffffffffffda RBX: 0000000000000000 RCX: 00007f6a5c9e8819
RDX: 0000000000000004 RSI: 0000000000000003 RDI: 0000000000000004
RBP: 00007f6a5c9a8390 R08: 0000000000000000 R09: 000000000
---
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@googlegroups.com.
syzbot will keep track of this issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
For information about bisection process see: https://goo.gl/tpsmEJ#bisection
syzbot can test patches for this issue, for details see:
https://goo.gl/tpsmEJ#testing-patches
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [syzbot] possible deadlock in split_huge_page_to_list
2021-12-15 1:03 [syzbot] possible deadlock in split_huge_page_to_list syzbot
@ 2021-12-15 4:19 ` Matthew Wilcox
0 siblings, 0 replies; 2+ messages in thread
From: Matthew Wilcox @ 2021-12-15 4:19 UTC (permalink / raw)
To: syzbot; +Cc: akpm, dhowells, hughd, linux-kernel, linux-mm, syzkaller-bugs
On Tue, Dec 14, 2021 at 05:03:26PM -0800, syzbot wrote:
> commit 3ebffc96befbaf9de9297b00d67091bb702fad8e
> Author: Matthew Wilcox (Oracle) <willy@infradead.org>
> Date: Sun Jun 28 02:19:08 2020 +0000
>
> mm: Use multi-index entries in the page cache
>
> bisection log: https://syzkaller.appspot.com/x/bisect.txt?x=1276e4bab00000
> final oops: https://syzkaller.appspot.com/x/report.txt?x=1176e4bab00000
> console output: https://syzkaller.appspot.com/x/log.txt?x=1676e4bab00000
Well, this is all entirely plausible:
+ xas_split_alloc(&xas, head, compound_order(head),
+ mapping_gfp_mask(mapping) & GFP_RECLAIM_MASK);
It looks like I can fix this by moving the memory allocation before
the acquisition of the i_mmap_lock. Any objections to this:
+++ b/mm/huge_memory.c
@@ -2653,6 +2653,13 @@ int split_huge_page_to_list(struct page *page, struct lis
t_head *list)
goto out;
}
+ xas_split_alloc(&xas, head, compound_order(head),
+ mapping_gfp_mask(mapping) & GFP_RECLAIM_MASK);
+ if (xas_error(&xas)) {
+ ret = xas_error(&xas);
+ goto out;
+ }
+
anon_vma = NULL;
i_mmap_lock_read(mapping);
@@ -2679,15 +2686,6 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
unmap_page(head);
- if (mapping) {
- xas_split_alloc(&xas, head, compound_order(head),
- mapping_gfp_mask(mapping) & GFP_RECLAIM_MASK);
- if (xas_error(&xas)) {
- ret = xas_error(&xas);
- goto out_unlock;
- }
- }
-
/* block interrupt reentry in xa_lock and spinlock */
local_irq_disable();
if (mapping) {
(relative to the above patch)
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2021-12-15 4:19 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-15 1:03 [syzbot] possible deadlock in split_huge_page_to_list syzbot
2021-12-15 4:19 ` Matthew Wilcox
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.