Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

[syzbot]

syzbot has found a reproducer for the following issue on:

HEAD commit:    c0b832517f62 Add linux-next specific files for 20240402
git tree:       linux-next
console+strace: https://syzkaller.appspot.com/x/log.txt?x=14af7dd9180000
kernel config:  https://syzkaller.appspot.com/x/.config?x=afcaf46d374cec8c
dashboard link: https://syzkaller.appspot.com/bug?extid=7b219b86935220db6dd8
compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1729f003180000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17fa4341180000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/0d36ec76edc7/disk-c0b83251.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/6f9bb4e37dd0/vmlinux-c0b83251.xz
kernel image: https://storage.googleapis.com/syzbot-assets/2349287b14b7/bzImage-c0b83251.xz
mounted in repro: https://storage.googleapis.com/syzbot-assets/9760c52a227c/mount_0.gz

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com

==================================================================
BUG: KASAN: slab-out-of-bounds in __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
Read of size 8 at addr ffff888020485fa8 by task kworker/u8:2/35

CPU: 0 PID: 35 Comm: kworker/u8:2 Not tainted 6.9.0-rc2-next-20240402-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
Workqueue: writeback wb_workfn (flush-7:1)
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114
 print_address_description mm/kasan/report.c:377 [inline]
 print_report+0x169/0x550 mm/kasan/report.c:488
 kasan_report+0x143/0x180 mm/kasan/report.c:601
 __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
 __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
 _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
 spin_lock include/linux/spinlock.h:351 [inline]
 wb_writeback+0x66f/0xd30 fs/fs-writeback.c:2160
 wb_check_old_data_flush fs/fs-writeback.c:2233 [inline]
 wb_do_writeback fs/fs-writeback.c:2286 [inline]
 wb_workfn+0xba1/0x1090 fs/fs-writeback.c:2314
 process_one_work kernel/workqueue.c:3218 [inline]
 process_scheduled_works+0xa2c/0x1830 kernel/workqueue.c:3299
 worker_thread+0x86d/0xd70 kernel/workqueue.c:3380
 kthread+0x2f0/0x390 kernel/kthread.c:388
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:243
 </TASK>

Allocated by task 5052:
 kasan_save_stack mm/kasan/common.c:47 [inline]
 kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
 poison_kmalloc_redzone mm/kasan/common.c:370 [inline]
 __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:387
 kasan_kmalloc include/linux/kasan.h:211 [inline]
 __do_kmalloc_node mm/slub.c:4048 [inline]
 __kmalloc_noprof+0x200/0x410 mm/slub.c:4061
 kmalloc_noprof include/linux/slab.h:664 [inline]
 tomoyo_realpath_from_path+0xcf/0x5e0 security/tomoyo/realpath.c:251
 tomoyo_get_realpath security/tomoyo/file.c:151 [inline]
 tomoyo_path_perm+0x2b7/0x740 security/tomoyo/file.c:822
 security_inode_getattr+0xd8/0x130 security/security.c:2269
 vfs_getattr+0x45/0x430 fs/stat.c:173
 vfs_fstat fs/stat.c:198 [inline]
 vfs_fstatat+0xd6/0x190 fs/stat.c:300
 __do_sys_newfstatat fs/stat.c:468 [inline]
 __se_sys_newfstatat fs/stat.c:462 [inline]
 __x64_sys_newfstatat+0x125/0x1b0 fs/stat.c:462
 do_syscall_64+0xfb/0x240
 entry_SYSCALL_64_after_hwframe+0x72/0x7a

Freed by task 5052:
 kasan_save_stack mm/kasan/common.c:47 [inline]
 kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
 kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:579
 poison_slab_object+0xe0/0x150 mm/kasan/common.c:240
 __kasan_slab_free+0x37/0x60 mm/kasan/common.c:256
 kasan_slab_free include/linux/kasan.h:184 [inline]
 slab_free_hook mm/slub.c:2180 [inline]
 slab_free mm/slub.c:4363 [inline]
 kfree+0x149/0x350 mm/slub.c:4484
 tomoyo_realpath_from_path+0x5a9/0x5e0 security/tomoyo/realpath.c:286
 tomoyo_get_realpath security/tomoyo/file.c:151 [inline]
 tomoyo_path_perm+0x2b7/0x740 security/tomoyo/file.c:822
 security_inode_getattr+0xd8/0x130 security/security.c:2269
 vfs_getattr+0x45/0x430 fs/stat.c:173
 vfs_fstat fs/stat.c:198 [inline]
 vfs_fstatat+0xd6/0x190 fs/stat.c:300
 __do_sys_newfstatat fs/stat.c:468 [inline]
 __se_sys_newfstatat fs/stat.c:462 [inline]
 __x64_sys_newfstatat+0x125/0x1b0 fs/stat.c:462
 do_syscall_64+0xfb/0x240
 entry_SYSCALL_64_after_hwframe+0x72/0x7a

The buggy address belongs to the object at ffff888020484000
 which belongs to the cache kmalloc-4k of size 4096
The buggy address is located 4008 bytes to the right of
 allocated 4096-byte region [ffff888020484000, ffff888020485000)

The buggy address belongs to the physical page:
page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x20480
head: order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0
flags: 0xfff80000000040(head|node=0|zone=1|lastcpupid=0xfff)
page_type: 0xffffefff(slab)
raw: 00fff80000000040 ffff888015042140 dead000000000100 dead000000000122
raw: 0000000000000000 0000000000040004 00000001ffffefff 0000000000000000
head: 00fff80000000040 ffff888015042140 dead000000000100 dead000000000122
head: 0000000000000000 0000000000040004 00000001ffffefff 0000000000000000
head: 00fff80000000003 ffffea0000812001 ffffea0000812048 00000000ffffffff
head: 0000000800000000 0000000000000000 00000000ffffffff 0000000000000000
page dumped because: kasan: bad access detected
page_owner tracks the page as allocated
page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 1, tgid -957297381 (swapper/0), ts 1, free_ts 0
 set_page_owner include/linux/page_owner.h:32 [inline]
 post_alloc_hook+0x1f3/0x230 mm/page_alloc.c:1490
 prep_new_page mm/page_alloc.c:1498 [inline]
 get_page_from_freelist+0x2e7e/0x2f40 mm/page_alloc.c:3454
 __alloc_pages_noprof+0x256/0x6c0 mm/page_alloc.c:4712
 __alloc_pages_node_noprof include/linux/gfp.h:244 [inline]
 alloc_pages_node_noprof include/linux/gfp.h:271 [inline]
 alloc_slab_page+0x5f/0x120 mm/slub.c:2249
 allocate_slab+0x5a/0x2e0 mm/slub.c:2412
 new_slab mm/slub.c:2465 [inline]
 ___slab_alloc+0xea8/0x1430 mm/slub.c:3599
 __slab_alloc+0x58/0xa0 mm/slub.c:3684
 __slab_alloc_node mm/slub.c:3737 [inline]
 slab_alloc_node mm/slub.c:3915 [inline]
 kmalloc_node_trace_noprof+0x20c/0x300 mm/slub.c:4087
 kmalloc_node_noprof include/linux/slab.h:677 [inline]
 bdi_alloc+0x4f/0x140 mm/backing-dev.c:894
 __alloc_disk_node+0xb8/0x590 block/genhd.c:1347
 __blk_mq_alloc_disk+0x17d/0x260 block/blk-mq.c:4166
 loop_add+0x448/0xba0 drivers/block/loop.c:2032
 loop_init+0x17a/0x230 drivers/block/loop.c:2275
 do_one_initcall+0x248/0x880 init/main.c:1258
 do_initcall_level+0x157/0x210 init/main.c:1320
 do_initcalls+0x3f/0x80 init/main.c:1336
page_owner free stack trace missing

Memory state around the buggy address:
 ffff888020485e80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
 ffff888020485f00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff888020485f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
                                  ^
 ffff888020486000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 ffff888020486080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
==================================================================


---
If you want syzbot to run the reproducer, reply with:
#syz test: git://repo/address.git branch-or-commit-hash
If you attach or paste a git patch, syzbot will apply it before testing.

Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

[Jan Kara]

On Tue 02-04-24 07:38:25, syzbot wrote:
> syzbot has found a reproducer for the following issue on:
> 
> HEAD commit:    c0b832517f62 Add linux-next specific files for 20240402
> git tree:       linux-next
> console+strace: https://syzkaller.appspot.com/x/log.txt?x=14af7dd9180000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=afcaf46d374cec8c
> dashboard link: https://syzkaller.appspot.com/bug?extid=7b219b86935220db6dd8
> compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1729f003180000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17fa4341180000
> 
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/0d36ec76edc7/disk-c0b83251.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/6f9bb4e37dd0/vmlinux-c0b83251.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/2349287b14b7/bzImage-c0b83251.xz
> mounted in repro: https://storage.googleapis.com/syzbot-assets/9760c52a227c/mount_0.gz
> 
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com
> 
> ==================================================================
> BUG: KASAN: slab-out-of-bounds in __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
> Read of size 8 at addr ffff888020485fa8 by task kworker/u8:2/35

Looks like the writeback cleanups are causing some use-after-free issues.
The code KASAN is complaining about is:

		/*
		 * Nothing written. Wait for some inode to
		 * become available for writeback. Otherwise
		 * we'll just busyloop.
		 */
		trace_writeback_wait(wb, work);
		inode = wb_inode(wb->b_more_io.prev);
>>>>>		spin_lock(&inode->i_lock); <<<<<<
		spin_unlock(&wb->list_lock);
		/* This function drops i_lock... */
		inode_sleep_on_writeback(inode);

in wb_writeback(). Now looking at the changes indeed the commit
167d6693deb ("fs/writeback: bail out if there is no more inodes for IO and
queued once") is buggy because it will result in trying to fetch 'inode'
from empty b_more_io list and thus we'll corrupt memory. I think instead of
modifying the condition:

		if (list_empty(&wb->b_more_io)) {

we should do:

-		if (progress) {
+		if (progress || !queued) {
                        spin_unlock(&wb->list_lock);
                        continue;
                }

Kemeng?

								Honza

> CPU: 0 PID: 35 Comm: kworker/u8:2 Not tainted 6.9.0-rc2-next-20240402-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
> Workqueue: writeback wb_workfn (flush-7:1)
> Call Trace:
>  <TASK>
>  __dump_stack lib/dump_stack.c:88 [inline]
>  dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114
>  print_address_description mm/kasan/report.c:377 [inline]
>  print_report+0x169/0x550 mm/kasan/report.c:488
>  kasan_report+0x143/0x180 mm/kasan/report.c:601
>  __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
>  lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
>  __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
>  _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
>  spin_lock include/linux/spinlock.h:351 [inline]
>  wb_writeback+0x66f/0xd30 fs/fs-writeback.c:2160
>  wb_check_old_data_flush fs/fs-writeback.c:2233 [inline]
>  wb_do_writeback fs/fs-writeback.c:2286 [inline]
>  wb_workfn+0xba1/0x1090 fs/fs-writeback.c:2314
>  process_one_work kernel/workqueue.c:3218 [inline]
>  process_scheduled_works+0xa2c/0x1830 kernel/workqueue.c:3299
>  worker_thread+0x86d/0xd70 kernel/workqueue.c:3380
>  kthread+0x2f0/0x390 kernel/kthread.c:388
>  ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
>  ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:243
>  </TASK>
> 
> Allocated by task 5052:
>  kasan_save_stack mm/kasan/common.c:47 [inline]
>  kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
>  poison_kmalloc_redzone mm/kasan/common.c:370 [inline]
>  __kasan_kmalloc+0x98/0xb0 mm/kasan/common.c:387
>  kasan_kmalloc include/linux/kasan.h:211 [inline]
>  __do_kmalloc_node mm/slub.c:4048 [inline]
>  __kmalloc_noprof+0x200/0x410 mm/slub.c:4061
>  kmalloc_noprof include/linux/slab.h:664 [inline]
>  tomoyo_realpath_from_path+0xcf/0x5e0 security/tomoyo/realpath.c:251
>  tomoyo_get_realpath security/tomoyo/file.c:151 [inline]
>  tomoyo_path_perm+0x2b7/0x740 security/tomoyo/file.c:822
>  security_inode_getattr+0xd8/0x130 security/security.c:2269
>  vfs_getattr+0x45/0x430 fs/stat.c:173
>  vfs_fstat fs/stat.c:198 [inline]
>  vfs_fstatat+0xd6/0x190 fs/stat.c:300
>  __do_sys_newfstatat fs/stat.c:468 [inline]
>  __se_sys_newfstatat fs/stat.c:462 [inline]
>  __x64_sys_newfstatat+0x125/0x1b0 fs/stat.c:462
>  do_syscall_64+0xfb/0x240
>  entry_SYSCALL_64_after_hwframe+0x72/0x7a
> 
> Freed by task 5052:
>  kasan_save_stack mm/kasan/common.c:47 [inline]
>  kasan_save_track+0x3f/0x80 mm/kasan/common.c:68
>  kasan_save_free_info+0x40/0x50 mm/kasan/generic.c:579
>  poison_slab_object+0xe0/0x150 mm/kasan/common.c:240
>  __kasan_slab_free+0x37/0x60 mm/kasan/common.c:256
>  kasan_slab_free include/linux/kasan.h:184 [inline]
>  slab_free_hook mm/slub.c:2180 [inline]
>  slab_free mm/slub.c:4363 [inline]
>  kfree+0x149/0x350 mm/slub.c:4484
>  tomoyo_realpath_from_path+0x5a9/0x5e0 security/tomoyo/realpath.c:286
>  tomoyo_get_realpath security/tomoyo/file.c:151 [inline]
>  tomoyo_path_perm+0x2b7/0x740 security/tomoyo/file.c:822
>  security_inode_getattr+0xd8/0x130 security/security.c:2269
>  vfs_getattr+0x45/0x430 fs/stat.c:173
>  vfs_fstat fs/stat.c:198 [inline]
>  vfs_fstatat+0xd6/0x190 fs/stat.c:300
>  __do_sys_newfstatat fs/stat.c:468 [inline]
>  __se_sys_newfstatat fs/stat.c:462 [inline]
>  __x64_sys_newfstatat+0x125/0x1b0 fs/stat.c:462
>  do_syscall_64+0xfb/0x240
>  entry_SYSCALL_64_after_hwframe+0x72/0x7a
> 
> The buggy address belongs to the object at ffff888020484000
>  which belongs to the cache kmalloc-4k of size 4096
> The buggy address is located 4008 bytes to the right of
>  allocated 4096-byte region [ffff888020484000, ffff888020485000)
> 
> The buggy address belongs to the physical page:
> page: refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x20480
> head: order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0
> flags: 0xfff80000000040(head|node=0|zone=1|lastcpupid=0xfff)
> page_type: 0xffffefff(slab)
> raw: 00fff80000000040 ffff888015042140 dead000000000100 dead000000000122
> raw: 0000000000000000 0000000000040004 00000001ffffefff 0000000000000000
> head: 00fff80000000040 ffff888015042140 dead000000000100 dead000000000122
> head: 0000000000000000 0000000000040004 00000001ffffefff 0000000000000000
> head: 00fff80000000003 ffffea0000812001 ffffea0000812048 00000000ffffffff
> head: 0000000800000000 0000000000000000 00000000ffffffff 0000000000000000
> page dumped because: kasan: bad access detected
> page_owner tracks the page as allocated
> page last allocated via order 3, migratetype Unmovable, gfp_mask 0xd20c0(__GFP_IO|__GFP_FS|__GFP_NOWARN|__GFP_NORETRY|__GFP_COMP|__GFP_NOMEMALLOC), pid 1, tgid -957297381 (swapper/0), ts 1, free_ts 0
>  set_page_owner include/linux/page_owner.h:32 [inline]
>  post_alloc_hook+0x1f3/0x230 mm/page_alloc.c:1490
>  prep_new_page mm/page_alloc.c:1498 [inline]
>  get_page_from_freelist+0x2e7e/0x2f40 mm/page_alloc.c:3454
>  __alloc_pages_noprof+0x256/0x6c0 mm/page_alloc.c:4712
>  __alloc_pages_node_noprof include/linux/gfp.h:244 [inline]
>  alloc_pages_node_noprof include/linux/gfp.h:271 [inline]
>  alloc_slab_page+0x5f/0x120 mm/slub.c:2249
>  allocate_slab+0x5a/0x2e0 mm/slub.c:2412
>  new_slab mm/slub.c:2465 [inline]
>  ___slab_alloc+0xea8/0x1430 mm/slub.c:3599
>  __slab_alloc+0x58/0xa0 mm/slub.c:3684
>  __slab_alloc_node mm/slub.c:3737 [inline]
>  slab_alloc_node mm/slub.c:3915 [inline]
>  kmalloc_node_trace_noprof+0x20c/0x300 mm/slub.c:4087
>  kmalloc_node_noprof include/linux/slab.h:677 [inline]
>  bdi_alloc+0x4f/0x140 mm/backing-dev.c:894
>  __alloc_disk_node+0xb8/0x590 block/genhd.c:1347
>  __blk_mq_alloc_disk+0x17d/0x260 block/blk-mq.c:4166
>  loop_add+0x448/0xba0 drivers/block/loop.c:2032
>  loop_init+0x17a/0x230 drivers/block/loop.c:2275
>  do_one_initcall+0x248/0x880 init/main.c:1258
>  do_initcall_level+0x157/0x210 init/main.c:1320
>  do_initcalls+0x3f/0x80 init/main.c:1336
> page_owner free stack trace missing
> 
> Memory state around the buggy address:
>  ffff888020485e80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>  ffff888020485f00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
> >ffff888020485f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>                                   ^
>  ffff888020486000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>  ffff888020486080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
> ==================================================================
> 
> 
> ---
> If you want syzbot to run the reproducer, reply with:
> #syz test: git://repo/address.git branch-or-commit-hash
> If you attach or paste a git patch, syzbot will apply it before testing.
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

[Christian Brauner]

On Wed, Apr 03, 2024 at 11:47:17AM +0200, Jan Kara wrote:
> On Tue 02-04-24 07:38:25, syzbot wrote:
> > syzbot has found a reproducer for the following issue on:
> > 
> > HEAD commit:    c0b832517f62 Add linux-next specific files for 20240402
> > git tree:       linux-next
> > console+strace: https://syzkaller.appspot.com/x/log.txt?x=14af7dd9180000
> > kernel config:  https://syzkaller.appspot.com/x/.config?x=afcaf46d374cec8c
> > dashboard link: https://syzkaller.appspot.com/bug?extid=7b219b86935220db6dd8
> > compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
> > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1729f003180000
> > C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17fa4341180000
> > 
> > Downloadable assets:
> > disk image: https://storage.googleapis.com/syzbot-assets/0d36ec76edc7/disk-c0b83251.raw.xz
> > vmlinux: https://storage.googleapis.com/syzbot-assets/6f9bb4e37dd0/vmlinux-c0b83251.xz
> > kernel image: https://storage.googleapis.com/syzbot-assets/2349287b14b7/bzImage-c0b83251.xz
> > mounted in repro: https://storage.googleapis.com/syzbot-assets/9760c52a227c/mount_0.gz
> > 
> > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com
> > 
> > ==================================================================
> > BUG: KASAN: slab-out-of-bounds in __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
> > Read of size 8 at addr ffff888020485fa8 by task kworker/u8:2/35
> 
> Looks like the writeback cleanups are causing some use-after-free issues.
> The code KASAN is complaining about is:
> 
> 		/*
> 		 * Nothing written. Wait for some inode to
> 		 * become available for writeback. Otherwise
> 		 * we'll just busyloop.
> 		 */
> 		trace_writeback_wait(wb, work);
> 		inode = wb_inode(wb->b_more_io.prev);
> >>>>>		spin_lock(&inode->i_lock); <<<<<<
> 		spin_unlock(&wb->list_lock);
> 		/* This function drops i_lock... */
> 		inode_sleep_on_writeback(inode);
> 
> in wb_writeback(). Now looking at the changes indeed the commit
> 167d6693deb ("fs/writeback: bail out if there is no more inodes for IO and
> queued once") is buggy because it will result in trying to fetch 'inode'
> from empty b_more_io list and thus we'll corrupt memory. I think instead of
> modifying the condition:
> 
> 		if (list_empty(&wb->b_more_io)) {
> 
> we should do:
> 
> -		if (progress) {
> +		if (progress || !queued) {
>                         spin_unlock(&wb->list_lock);
>                         continue;
>                 }
> 
> Kemeng?

Fwiw, I observed this on xfstest too the last few days and tracked it
down to this series. Here's the splat I got in case it helps:

Apr 05 00:33:06 localhost kernel: ==================================================================
Apr 05 00:33:06 localhost kernel: BUG: KASAN: slab-out-of-bounds in __lock_acquire.isra.0+0x1075/0x1280
Apr 05 00:33:06 localhost kernel: Read of size 8 at addr ffff88810ed40f48 by task kworker/u128:2/305560
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: CPU: 5 PID: 305560 Comm: kworker/u128:2 Not tainted 99.9.0-rc2-gdebeafad51e2 #262
Apr 05 00:33:06 localhost kernel: Hardware name: QEMU Standard PC (Q35 + ICH9, 2009)/Incus, BIOS unknown 2/2/2022
Apr 05 00:33:06 localhost kernel: Workqueue: writeback wb_workfn (flush-259:0)
Apr 05 00:33:06 localhost kernel: Call Trace:
Apr 05 00:33:06 localhost kernel:  <TASK>
Apr 05 00:33:06 localhost kernel:  dump_stack_lvl+0x5a/0x90
Apr 05 00:33:06 localhost kernel:  print_report+0xce/0x650
Apr 05 00:33:06 localhost kernel:  ? __virt_addr_valid+0x217/0x320
Apr 05 00:33:06 localhost kernel:  kasan_report+0xd7/0x110
Apr 05 00:33:06 localhost kernel:  ? __lock_acquire.isra.0+0x1075/0x1280
Apr 05 00:33:06 localhost kernel:  ? __lock_acquire.isra.0+0x1075/0x1280
Apr 05 00:33:06 localhost kernel:  __lock_acquire.isra.0+0x1075/0x1280
Apr 05 00:33:06 localhost kernel:  lock_acquire+0x136/0x330
Apr 05 00:33:06 localhost kernel:  ? wb_writeback+0x255/0x870
Apr 05 00:33:06 localhost kernel:  _raw_spin_lock+0x33/0x40
Apr 05 00:33:06 localhost kernel:  ? wb_writeback+0x255/0x870
Apr 05 00:33:06 localhost kernel:  wb_writeback+0x255/0x870
Apr 05 00:33:06 localhost kernel:  ? __pfx_wb_writeback+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ? __pfx_lock_release+0x10/0x10
Apr 05 00:33:06 localhost kernel:  wb_workfn+0x221/0xc80
Apr 05 00:33:06 localhost kernel:  ? __pfx_wb_workfn+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ? lock_acquire+0x136/0x330
Apr 05 00:33:06 localhost kernel:  process_one_work+0x82d/0x1790
Apr 05 00:33:06 localhost kernel:  ? __pfx_process_one_work+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ? assign_work+0x16c/0x240
Apr 05 00:33:06 localhost kernel:  worker_thread+0x724/0x1300
Apr 05 00:33:06 localhost kernel:  ? __kthread_parkme+0xba/0x1f0
Apr 05 00:33:06 localhost kernel:  ? __pfx_worker_thread+0x10/0x10
Apr 05 00:33:06 localhost kernel:  kthread+0x2ed/0x3d0
Apr 05 00:33:06 localhost kernel:  ? __pfx_kthread+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ret_from_fork+0x31/0x70
Apr 05 00:33:06 localhost kernel:  ? __pfx_kthread+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ret_from_fork_asm+0x1a/0x30
Apr 05 00:33:06 localhost kernel:  </TASK>
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: Allocated by task 1:
Apr 05 00:33:06 localhost kernel:  kasan_save_stack+0x33/0x60
Apr 05 00:33:06 localhost kernel:  kasan_save_track+0x14/0x30
Apr 05 00:33:06 localhost kernel:  __kasan_kmalloc+0xaa/0xb0
Apr 05 00:33:06 localhost kernel:  psi_cgroup_alloc+0x57/0x2b0
Apr 05 00:33:06 localhost kernel:  cgroup_mkdir+0x4f8/0xfb0
Apr 05 00:33:06 localhost kernel:  kernfs_iop_mkdir+0x133/0x1c0
Apr 05 00:33:06 localhost kernel:  vfs_mkdir+0x3b9/0x610
Apr 05 00:33:06 localhost kernel:  do_mkdirat+0x27e/0x300
Apr 05 00:33:06 localhost kernel:  __x64_sys_mkdir+0x65/0x80
Apr 05 00:33:06 localhost kernel:  do_syscall_64+0x64/0x190
Apr 05 00:33:06 localhost kernel:  entry_SYSCALL_64_after_hwframe+0x71/0x79
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: Last potentially related work creation:
Apr 05 00:33:06 localhost kernel:  kasan_save_stack+0x33/0x60
Apr 05 00:33:06 localhost kernel:  __kasan_record_aux_stack+0xad/0xc0
Apr 05 00:33:06 localhost kernel:  insert_work+0x32/0x1f0
Apr 05 00:33:06 localhost kernel:  __queue_work+0x5cb/0xcb0
Apr 05 00:33:06 localhost kernel:  call_timer_fn+0x16d/0x490
Apr 05 00:33:06 localhost kernel:  __run_timers+0x488/0x980
Apr 05 00:33:06 localhost kernel:  run_timer_base+0xfb/0x170
Apr 05 00:33:06 localhost kernel:  run_timer_softirq+0x1a/0x30
Apr 05 00:33:06 localhost kernel:  __do_softirq+0x26a/0x7d2
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: Second to last potentially related work creation:
Apr 05 00:33:06 localhost kernel:  kasan_save_stack+0x33/0x60
Apr 05 00:33:06 localhost kernel:  __kasan_record_aux_stack+0xad/0xc0
Apr 05 00:33:06 localhost kernel:  insert_work+0x32/0x1f0
Apr 05 00:33:06 localhost kernel:  __queue_work+0x5cb/0xcb0
Apr 05 00:33:06 localhost kernel:  call_timer_fn+0x16d/0x490
Apr 05 00:33:06 localhost kernel:  __run_timers+0x488/0x980
Apr 05 00:33:06 localhost kernel:  timer_expire_remote+0xe6/0x150
Apr 05 00:33:06 localhost kernel:  tmigr_handle_remote+0x6e2/0xe00
Apr 05 00:33:06 localhost kernel:  __do_softirq+0x26a/0x7d2
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: The buggy address belongs to the object at ffff88810ed40000
                                   which belongs to the cache kmalloc-2k of size 2048
Apr 05 00:33:06 localhost kernel: The buggy address is located 2792 bytes to the right of
                                   allocated 1120-byte region [ffff88810ed40000, ffff88810ed40460)
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: The buggy address belongs to the physical page:
Apr 05 00:33:06 localhost kernel: page: refcount:1 mapcount:0 mapping:0000000000000000 index:0xffff88810ed45000 pfn:0x10ed40
Apr 05 00:33:06 localhost kernel: head: order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0
Apr 05 00:33:06 localhost kernel: flags: 0x200000000000840(slab|head|node=0|zone=2)
Apr 05 00:33:06 localhost kernel: page_type: 0xffffffff()
Apr 05 00:33:06 localhost kernel: raw: 0200000000000840 ffff888100042f00 ffffea00040f1a00 dead000000000002
Apr 05 00:33:06 localhost kernel: raw: ffff88810ed45000 0000000080080006 00000001ffffffff 0000000000000000
Apr 05 00:33:06 localhost kernel: head: 0200000000000840 ffff888100042f00 ffffea00040f1a00 dead000000000002
Apr 05 00:33:06 localhost kernel: head: ffff88810ed45000 0000000080080006 00000001ffffffff 0000000000000000
Apr 05 00:33:06 localhost kernel: head: 0200000000000003 ffffea00043b5001 dead000000000122 00000000ffffffff
Apr 05 00:33:06 localhost kernel: head: 0000000800000000 0000000000000000 00000000ffffffff 0000000000000000
Apr 05 00:33:06 localhost kernel: page dumped because: kasan: bad access detected
Apr 05 00:33:06 localhost kernel:
Apr 05 00:33:06 localhost kernel: Memory state around the buggy address:
Apr 05 00:33:06 localhost kernel:  ffff88810ed40e00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
Apr 05 00:33:06 localhost kernel:  ffff88810ed40e80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
Apr 05 00:33:06 localhost kernel: >ffff88810ed40f00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
Apr 05 00:33:06 localhost kernel:                                               ^
Apr 05 00:33:06 localhost kernel:  ffff88810ed40f80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
Apr 05 00:33:06 localhost kernel:  ffff88810ed41000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
Apr 05 00:33:06 localhost kernel: ==================================================================
Apr 05 00:33:06 localhost kernel: Disabling lock debugging due to kernel taint
Apr 05 00:33:06 localhost kernel: INFO: trying to register non-static key.
Apr 05 00:33:06 localhost kernel: The code is fine but needs lockdep annotation, or maybe
Apr 05 00:33:06 localhost kernel: you didn't initialize this object before use?
Apr 05 00:33:06 localhost kernel: turning off the locking correctness validator.
Apr 05 00:33:06 localhost kernel: CPU: 5 PID: 305560 Comm: kworker/u128:2 Tainted: G    B              99.9.0-rc2-gdebeafad51e2 #262
Apr 05 00:33:06 localhost kernel: Hardware name: QEMU Standard PC (Q35 + ICH9, 2009)/Incus, BIOS unknown 2/2/2022
Apr 05 00:33:06 localhost kernel: Workqueue: writeback wb_workfn (flush-259:0)
Apr 05 00:33:06 localhost kernel: Call Trace:
Apr 05 00:33:06 localhost kernel:  <TASK>
Apr 05 00:33:06 localhost kernel:  dump_stack_lvl+0x5a/0x90
Apr 05 00:33:06 localhost kernel:  register_lock_class+0x11dd/0x1860
Apr 05 00:33:06 localhost kernel:  ? add_taint+0x2a/0x90
Apr 05 00:33:06 localhost kernel:  ? end_report+0x85/0x180
Apr 05 00:33:06 localhost kernel:  ? __pfx_register_lock_class+0x10/0x10
Apr 05 00:33:06 localhost kernel:  __lock_acquire.isra.0+0x7f/0x1280
Apr 05 00:33:06 localhost kernel:  lock_acquire+0x136/0x330
Apr 05 00:33:06 localhost kernel:  ? wb_writeback+0x255/0x870
Apr 05 00:33:06 localhost kernel:  _raw_spin_lock+0x33/0x40
Apr 05 00:33:06 localhost kernel:  ? wb_writeback+0x255/0x870
Apr 05 00:33:06 localhost kernel:  wb_writeback+0x255/0x870
Apr 05 00:33:06 localhost kernel:  ? __pfx_wb_writeback+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ? __pfx_lock_release+0x10/0x10
Apr 05 00:33:06 localhost kernel:  wb_workfn+0x221/0xc80
Apr 05 00:33:06 localhost kernel:  ? __pfx_wb_workfn+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ? lock_acquire+0x136/0x330
Apr 05 00:33:06 localhost kernel:  process_one_work+0x82d/0x1790
Apr 05 00:33:06 localhost kernel:  ? __pfx_process_one_work+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ? assign_work+0x16c/0x240
Apr 05 00:33:06 localhost kernel:  worker_thread+0x724/0x1300
Apr 05 00:33:06 localhost kernel:  ? __kthread_parkme+0xba/0x1f0
Apr 05 00:33:06 localhost kernel:  ? __pfx_worker_thread+0x10/0x10
Apr 05 00:33:06 localhost kernel:  kthread+0x2ed/0x3d0
Apr 05 00:33:06 localhost kernel:  ? __pfx_kthread+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ret_from_fork+0x31/0x70
Apr 05 00:33:06 localhost kernel:  ? __pfx_kthread+0x10/0x10
Apr 05 00:33:06 localhost kernel:  ret_from_fork_asm+0x1a/0x30
Apr 05 00:33:06 localhost kernel:  </TASK>

Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

[Jan Kara]

[-- Attachment #1: Type: text/plain, Size: 3241 bytes --]

On Fri 05-04-24 13:05:59, Christian Brauner wrote:
> On Wed, Apr 03, 2024 at 11:47:17AM +0200, Jan Kara wrote:
> > On Tue 02-04-24 07:38:25, syzbot wrote:
> > > syzbot has found a reproducer for the following issue on:
> > > 
> > > HEAD commit:    c0b832517f62 Add linux-next specific files for 20240402
> > > git tree:       linux-next
> > > console+strace: https://syzkaller.appspot.com/x/log.txt?x=14af7dd9180000
> > > kernel config:  https://syzkaller.appspot.com/x/.config?x=afcaf46d374cec8c
> > > dashboard link: https://syzkaller.appspot.com/bug?extid=7b219b86935220db6dd8
> > > compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
> > > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1729f003180000
> > > C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17fa4341180000
> > > 
> > > Downloadable assets:
> > > disk image: https://storage.googleapis.com/syzbot-assets/0d36ec76edc7/disk-c0b83251.raw.xz
> > > vmlinux: https://storage.googleapis.com/syzbot-assets/6f9bb4e37dd0/vmlinux-c0b83251.xz
> > > kernel image: https://storage.googleapis.com/syzbot-assets/2349287b14b7/bzImage-c0b83251.xz
> > > mounted in repro: https://storage.googleapis.com/syzbot-assets/9760c52a227c/mount_0.gz
> > > 
> > > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > > Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com
> > > 
> > > ==================================================================
> > > BUG: KASAN: slab-out-of-bounds in __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
> > > Read of size 8 at addr ffff888020485fa8 by task kworker/u8:2/35
> > 
> > Looks like the writeback cleanups are causing some use-after-free issues.
> > The code KASAN is complaining about is:
> > 
> > 		/*
> > 		 * Nothing written. Wait for some inode to
> > 		 * become available for writeback. Otherwise
> > 		 * we'll just busyloop.
> > 		 */
> > 		trace_writeback_wait(wb, work);
> > 		inode = wb_inode(wb->b_more_io.prev);
> > >>>>>		spin_lock(&inode->i_lock); <<<<<<
> > 		spin_unlock(&wb->list_lock);
> > 		/* This function drops i_lock... */
> > 		inode_sleep_on_writeback(inode);
> > 
> > in wb_writeback(). Now looking at the changes indeed the commit
> > 167d6693deb ("fs/writeback: bail out if there is no more inodes for IO and
> > queued once") is buggy because it will result in trying to fetch 'inode'
> > from empty b_more_io list and thus we'll corrupt memory. I think instead of
> > modifying the condition:
> > 
> > 		if (list_empty(&wb->b_more_io)) {
> > 
> > we should do:
> > 
> > -		if (progress) {
> > +		if (progress || !queued) {
> >                         spin_unlock(&wb->list_lock);
> >                         continue;
> >                 }
> > 
> > Kemeng?
> 
> Fwiw, I observed this on xfstest too the last few days and tracked it
> down to this series. Here's the splat I got in case it helps:

OK, since this is apparently causing more issues and Kemeng didn't reply
yet, here's a fix in the form of the patch. It has passed some basic
testing. Feel free to fold it into Kemeng's patch so that we don't keep
linux-next broken longer than necessary. Thanks!

								Honza
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

[-- Attachment #2: 0001-writeback-Fix-memory-corruption-in-writeback-code.patch --]
[-- Type: text/x-patch, Size: 1848 bytes --]

From cede4bc05f7a9a38f21b5943c11592fdb098b4f4 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Fri, 5 Apr 2024 13:57:28 +0200
Subject: [PATCH] writeback: Fix memory corruption in writeback code

Commit 167d6693deb3 ("fs/writeback: bail out if there is no more inodes
for IO and queued once") made the loop in wb_writeback() continue, even
if we didn't have any inodes in b_more_io list when we didn't queue any
inodes into b_io list yet. Conceptually this is fine however the loop in
this case takes the first inode from b_more_io list and waits for
writeback on it to complete. When b_more_io list is empty, this results
in accesses beyond the wb->b_more_io list head corrupting struct
wb_writeback and memory beyond it. Fix the problem by directly
restarting the loop in this case instead of going through waiting on
inode in b_more_io list.

Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com
Fixes: 167d6693deb3 ("fs/writeback: bail out if there is no more inodes for IO and queued once")
Signed-off-by: Jan Kara <jack@suse.cz>
---
 fs/fs-writeback.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index f7ed4192d0f8..92a5b8283528 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -2137,7 +2137,7 @@ static long wb_writeback(struct bdi_writeback *wb,
 		 * mean the overall work is done. So we keep looping as long
 		 * as made some progress on cleaning pages or inodes.
 		 */
-		if (progress) {
+		if (progress || !queued) {
 			spin_unlock(&wb->list_lock);
 			continue;
 		}
@@ -2145,7 +2145,7 @@ static long wb_writeback(struct bdi_writeback *wb,
 		/*
 		 * No more inodes for IO, bail
 		 */
-		if (list_empty(&wb->b_more_io) && queued) {
+		if (list_empty(&wb->b_more_io)) {
 			spin_unlock(&wb->list_lock);
 			break;
 		}
-- 
2.35.3

Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

[Christian Brauner]

On Fri, Apr 05, 2024 at 03:23:46PM +0200, Jan Kara wrote:
> On Fri 05-04-24 13:05:59, Christian Brauner wrote:
> > On Wed, Apr 03, 2024 at 11:47:17AM +0200, Jan Kara wrote:
> > > On Tue 02-04-24 07:38:25, syzbot wrote:
> > > > syzbot has found a reproducer for the following issue on:
> > > > 
> > > > HEAD commit:    c0b832517f62 Add linux-next specific files for 20240402
> > > > git tree:       linux-next
> > > > console+strace: https://syzkaller.appspot.com/x/log.txt?x=14af7dd9180000
> > > > kernel config:  https://syzkaller.appspot.com/x/.config?x=afcaf46d374cec8c
> > > > dashboard link: https://syzkaller.appspot.com/bug?extid=7b219b86935220db6dd8
> > > > compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
> > > > syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1729f003180000
> > > > C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17fa4341180000
> > > > 
> > > > Downloadable assets:
> > > > disk image: https://storage.googleapis.com/syzbot-assets/0d36ec76edc7/disk-c0b83251.raw.xz
> > > > vmlinux: https://storage.googleapis.com/syzbot-assets/6f9bb4e37dd0/vmlinux-c0b83251.xz
> > > > kernel image: https://storage.googleapis.com/syzbot-assets/2349287b14b7/bzImage-c0b83251.xz
> > > > mounted in repro: https://storage.googleapis.com/syzbot-assets/9760c52a227c/mount_0.gz
> > > > 
> > > > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > > > Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com
> > > > 
> > > > ==================================================================
> > > > BUG: KASAN: slab-out-of-bounds in __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
> > > > Read of size 8 at addr ffff888020485fa8 by task kworker/u8:2/35
> > > 
> > > Looks like the writeback cleanups are causing some use-after-free issues.
> > > The code KASAN is complaining about is:
> > > 
> > > 		/*
> > > 		 * Nothing written. Wait for some inode to
> > > 		 * become available for writeback. Otherwise
> > > 		 * we'll just busyloop.
> > > 		 */
> > > 		trace_writeback_wait(wb, work);
> > > 		inode = wb_inode(wb->b_more_io.prev);
> > > >>>>>		spin_lock(&inode->i_lock); <<<<<<
> > > 		spin_unlock(&wb->list_lock);
> > > 		/* This function drops i_lock... */
> > > 		inode_sleep_on_writeback(inode);
> > > 
> > > in wb_writeback(). Now looking at the changes indeed the commit
> > > 167d6693deb ("fs/writeback: bail out if there is no more inodes for IO and
> > > queued once") is buggy because it will result in trying to fetch 'inode'
> > > from empty b_more_io list and thus we'll corrupt memory. I think instead of
> > > modifying the condition:
> > > 
> > > 		if (list_empty(&wb->b_more_io)) {
> > > 
> > > we should do:
> > > 
> > > -		if (progress) {
> > > +		if (progress || !queued) {
> > >                         spin_unlock(&wb->list_lock);
> > >                         continue;
> > >                 }
> > > 
> > > Kemeng?
> > 
> > Fwiw, I observed this on xfstest too the last few days and tracked it
> > down to this series. Here's the splat I got in case it helps:
> 
> OK, since this is apparently causing more issues and Kemeng didn't reply
> yet, here's a fix in the form of the patch. It has passed some basic
> testing. Feel free to fold it into Kemeng's patch so that we don't keep
> linux-next broken longer than necessary. Thanks!

Thanks! Folded and I mentioned that I folded a fix from you into the
commit with a link to this patch.

Re: [syzbot] [nilfs?] KASAN: slab-out-of-bounds Read in wb_writeback

[Kemeng Shi]

on 4/5/2024 9:23 PM, Jan Kara wrote:
> On Fri 05-04-24 13:05:59, Christian Brauner wrote:
>> On Wed, Apr 03, 2024 at 11:47:17AM +0200, Jan Kara wrote:
>>> On Tue 02-04-24 07:38:25, syzbot wrote:
>>>> syzbot has found a reproducer for the following issue on:
>>>>
>>>> HEAD commit:    c0b832517f62 Add linux-next specific files for 20240402
>>>> git tree:       linux-next
>>>> console+strace: https://syzkaller.appspot.com/x/log.txt?x=14af7dd9180000
>>>> kernel config:  https://syzkaller.appspot.com/x/.config?x=afcaf46d374cec8c
>>>> dashboard link: https://syzkaller.appspot.com/bug?extid=7b219b86935220db6dd8
>>>> compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
>>>> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1729f003180000
>>>> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=17fa4341180000
>>>>
>>>> Downloadable assets:
>>>> disk image: https://storage.googleapis.com/syzbot-assets/0d36ec76edc7/disk-c0b83251.raw.xz
>>>> vmlinux: https://storage.googleapis.com/syzbot-assets/6f9bb4e37dd0/vmlinux-c0b83251.xz
>>>> kernel image: https://storage.googleapis.com/syzbot-assets/2349287b14b7/bzImage-c0b83251.xz
>>>> mounted in repro: https://storage.googleapis.com/syzbot-assets/9760c52a227c/mount_0.gz
>>>>
>>>> IMPORTANT: if you fix the issue, please add the following tag to the commit:
>>>> Reported-by: syzbot+7b219b86935220db6dd8@syzkaller.appspotmail.com
>>>>
>>>> ==================================================================
>>>> BUG: KASAN: slab-out-of-bounds in __lock_acquire+0x78/0x1fd0 kernel/locking/lockdep.c:5005
>>>> Read of size 8 at addr ffff888020485fa8 by task kworker/u8:2/35
>>>
>>> Looks like the writeback cleanups are causing some use-after-free issues.
>>> The code KASAN is complaining about is:
>>>
>>> 		/*
>>> 		 * Nothing written. Wait for some inode to
>>> 		 * become available for writeback. Otherwise
>>> 		 * we'll just busyloop.
>>> 		 */
>>> 		trace_writeback_wait(wb, work);
>>> 		inode = wb_inode(wb->b_more_io.prev);
>>>>>>>> 		spin_lock(&inode->i_lock); <<<<<<
>>> 		spin_unlock(&wb->list_lock);
>>> 		/* This function drops i_lock... */
>>> 		inode_sleep_on_writeback(inode);
>>>
>>> in wb_writeback(). Now looking at the changes indeed the commit
>>> 167d6693deb ("fs/writeback: bail out if there is no more inodes for IO and
>>> queued once") is buggy because it will result in trying to fetch 'inode'
>>> from empty b_more_io list and thus we'll corrupt memory. I think instead of
>>> modifying the condition:
>>>
>>> 		if (list_empty(&wb->b_more_io)) {
>>>
>>> we should do:
>>>
>>> -		if (progress) {
>>> +		if (progress || !queued) {
>>>                         spin_unlock(&wb->list_lock);
>>>                         continue;
>>>                 }
>>>
>>> Kemeng?
>>
>> Fwiw, I observed this on xfstest too the last few days and tracked it
>> down to this series. Here's the splat I got in case it helps:
> 
> OK, since this is apparently causing more issues and Kemeng didn't reply
> yet, here's a fix in the form of the patch. It has passed some basic
> testing. Feel free to fold it into Kemeng's patch so that we don't keep
> linux-next broken longer than necessary. Thanks!
Sorry for the late reply as I was on vacation these days. Also sorry
for the bug introduced. The change looks good to me. Thanks a lot
for helping to fix this in time.

Kemeng
> 
> 								Honza
>