[syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot]

Hello,

syzbot found the following issue on:

HEAD commit:    d2bafcf224f3 Merge tag 'cgroup-for-6.11-rc4-fixes' of git:..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=12d5fa7b980000
kernel config:  https://syzkaller.appspot.com/x/.config?x=4fc2afd52fd008bb
dashboard link: https://syzkaller.appspot.com/bug?extid=ba3c0273042a898c230e
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=137c040b980000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=16ddb015980000

Downloadable assets:
disk image: https://storage.googleapis.com/syzbot-assets/7569f02310fb/disk-d2bafcf2.raw.xz
vmlinux: https://storage.googleapis.com/syzbot-assets/e30fee7b6c1d/vmlinux-d2bafcf2.xz
kernel image: https://storage.googleapis.com/syzbot-assets/2ffddebac153/bzImage-d2bafcf2.xz
mounted in repro #1: https://storage.googleapis.com/syzbot-assets/cd08557ae343/mount_0.gz
mounted in repro #2: https://storage.googleapis.com/syzbot-assets/90ff8392fe84/mount_2.gz
mounted in repro #3: https://storage.googleapis.com/syzbot-assets/9c5b91850930/mount_7.gz

The issue was bisected to:

commit f22b5dcbd71edea087946511554956591557de9a
Author: Christoph Hellwig <hch@lst.de>
Date:   Wed May 31 06:04:59 2023 +0000

    btrfs: remove non-standard extent handling in __extent_writepage_io

bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=10974043980000
final oops:     https://syzkaller.appspot.com/x/report.txt?x=12974043980000
console output: https://syzkaller.appspot.com/x/log.txt?x=14974043980000

IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+ba3c0273042a898c230e@syzkaller.appspotmail.com
Fixes: f22b5dcbd71e ("btrfs: remove non-standard extent handling in __extent_writepage_io")

assertion failed: block_start != EXTENT_MAP_HOLE, in fs/btrfs/extent_io.c:1488
------------[ cut here ]------------
kernel BUG at fs/btrfs/extent_io.c:1488!
Oops: invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI
CPU: 1 UID: 0 PID: 5722 Comm: syz-executor399 Not tainted 6.11.0-rc4-syzkaller-00255-gd2bafcf224f3 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/06/2024
RIP: 0010:__extent_writepage_io+0x1224/0x1400 fs/btrfs/extent_io.c:1488
Code: f7 07 90 0f 0b e8 dc 68 df fd 48 c7 c7 e0 92 2c 8c 48 c7 c6 c0 a0 2c 8c 48 c7 c2 80 92 2c 8c b9 d0 05 00 00 e8 9d 15 f7 07 90 <0f> 0b e8 b5 68 df fd 48 8b 3c 24 e8 bc 9d ff ff 48 89 c7 48 c7 c6
RSP: 0018:ffffc900097d6ee8 EFLAGS: 00010246
RAX: 000000000000004e RBX: 0000000000000000 RCX: 2f9fd79be1350f00
RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000
RBP: dffffc0000000000 R08: ffffffff817402ac R09: 1ffff920012fad7c
R10: dffffc0000000000 R11: fffff520012fad7d R12: ffffea0001ccbdc8
R13: 1ffffd40003997b9 R14: fffffffffffffffd R15: 000000000007b000
FS:  00007f11a84976c0(0000) GS:ffff8880b9300000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f11a0fff000 CR3: 0000000020e4c000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 __extent_writepage fs/btrfs/extent_io.c:1578 [inline]
 extent_write_cache_pages fs/btrfs/extent_io.c:2251 [inline]
 btrfs_writepages+0x12a2/0x2760 fs/btrfs/extent_io.c:2373
 do_writepages+0x35d/0x870 mm/page-writeback.c:2683
 filemap_fdatawrite_wbc+0x125/0x180 mm/filemap.c:397
 __filemap_fdatawrite_range mm/filemap.c:430 [inline]
 filemap_fdatawrite_range+0x120/0x180 mm/filemap.c:448
 btrfs_fdatawrite_range+0x53/0xe0 fs/btrfs/file.c:3794
 btrfs_direct_write+0x558/0xb40 fs/btrfs/direct-io.c:952
 btrfs_do_write_iter+0x2a1/0x760 fs/btrfs/file.c:1505
 do_iter_readv_writev+0x60a/0x890
 vfs_writev+0x37c/0xbb0 fs/read_write.c:971
 do_pwritev fs/read_write.c:1072 [inline]
 __do_sys_pwritev2 fs/read_write.c:1131 [inline]
 __se_sys_pwritev2+0x1ca/0x2d0 fs/read_write.c:1122
 do_syscall_x64 arch/x86/entry/common.c:52 [inline]
 do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
 entry_SYSCALL_64_after_hwframe+0x77/0x7f
RIP: 0033:0x7f11a850bd79
Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 41 1d 00 00 90 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 b0 ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f11a8497158 EFLAGS: 00000212 ORIG_RAX: 0000000000000148
RAX: ffffffffffffffda RBX: 00007f11a85916d8 RCX: 00007f11a850bd79
RDX: 0000000000000001 RSI: 0000000020000240 RDI: 0000000000000004
RBP: 00007f11a85916d0 R08: 0000000000000000 R09: 0000000000000003
R10: 0000000000002000 R11: 0000000000000212 R12: 00007f11a85916dc
R13: 000000000000006e R14: 00007ffd9dcad860 R15: 00007ffd9dcad948
 </TASK>
Modules linked in:
---[ end trace 0000000000000000 ]---
RIP: 0010:__extent_writepage_io+0x1224/0x1400 fs/btrfs/extent_io.c:1488
Code: f7 07 90 0f 0b e8 dc 68 df fd 48 c7 c7 e0 92 2c 8c 48 c7 c6 c0 a0 2c 8c 48 c7 c2 80 92 2c 8c b9 d0 05 00 00 e8 9d 15 f7 07 90 <0f> 0b e8 b5 68 df fd 48 8b 3c 24 e8 bc 9d ff ff 48 89 c7 48 c7 c6
RSP: 0018:ffffc900097d6ee8 EFLAGS: 00010246
RAX: 000000000000004e RBX: 0000000000000000 RCX: 2f9fd79be1350f00
RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000
RBP: dffffc0000000000 R08: ffffffff817402ac R09: 1ffff920012fad7c
R10: dffffc0000000000 R11: fffff520012fad7d R12: ffffea0001ccbdc8
R13: 1ffffd40003997b9 R14: fffffffffffffffd R15: 000000000007b000
FS:  00007f11a84976c0(0000) GS:ffff8880b9200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005654e2821068 CR3: 0000000020e4c000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400


---
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

If the report is already addressed, let syzbot know by replying with:
#syz fix: exact-commit-title

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.

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

Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[Qu Wenruo]

在 2024/8/29 08:04, syzbot 写道:
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit:    d2bafcf224f3 Merge tag 'cgroup-for-6.11-rc4-fixes' of git:..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=12d5fa7b980000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=4fc2afd52fd008bb
> dashboard link: https://syzkaller.appspot.com/bug?extid=ba3c0273042a898c230e
> 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=137c040b980000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=16ddb015980000
>
> Downloadable assets:
> disk image: https://storage.googleapis.com/syzbot-assets/7569f02310fb/disk-d2bafcf2.raw.xz
> vmlinux: https://storage.googleapis.com/syzbot-assets/e30fee7b6c1d/vmlinux-d2bafcf2.xz
> kernel image: https://storage.googleapis.com/syzbot-assets/2ffddebac153/bzImage-d2bafcf2.xz
> mounted in repro #1: https://storage.googleapis.com/syzbot-assets/cd08557ae343/mount_0.gz
> mounted in repro #2: https://storage.googleapis.com/syzbot-assets/90ff8392fe84/mount_2.gz
> mounted in repro #3: https://storage.googleapis.com/syzbot-assets/9c5b91850930/mount_7.gz
>
> The issue was bisected to:
>
> commit f22b5dcbd71edea087946511554956591557de9a
> Author: Christoph Hellwig <hch@lst.de>
> Date:   Wed May 31 06:04:59 2023 +0000
>
>      btrfs: remove non-standard extent handling in __extent_writepage_io
>
> bisection log:  https://syzkaller.appspot.com/x/bisect.txt?x=10974043980000
> final oops:     https://syzkaller.appspot.com/x/report.txt?x=12974043980000
> console output: https://syzkaller.appspot.com/x/log.txt?x=14974043980000
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+ba3c0273042a898c230e@syzkaller.appspotmail.com
> Fixes: f22b5dcbd71e ("btrfs: remove non-standard extent handling in __extent_writepage_io")
>
> assertion failed: block_start != EXTENT_MAP_HOLE, in fs/btrfs/extent_io.c:1488

This means there is a dirty page range for us to writeback, but no
delalloc range for it.

That patch is not the cause, but only exposed it.

I guess there is some page fault involved or interrupted direct IO?

Anyway the bisection doesn't make much sense.

Thanks,
Qu

> ------------[ cut here ]------------
> kernel BUG at fs/btrfs/extent_io.c:1488!
> Oops: invalid opcode: 0000 [#1] PREEMPT SMP KASAN PTI
> CPU: 1 UID: 0 PID: 5722 Comm: syz-executor399 Not tainted 6.11.0-rc4-syzkaller-00255-gd2bafcf224f3 #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 08/06/2024
> RIP: 0010:__extent_writepage_io+0x1224/0x1400 fs/btrfs/extent_io.c:1488
> Code: f7 07 90 0f 0b e8 dc 68 df fd 48 c7 c7 e0 92 2c 8c 48 c7 c6 c0 a0 2c 8c 48 c7 c2 80 92 2c 8c b9 d0 05 00 00 e8 9d 15 f7 07 90 <0f> 0b e8 b5 68 df fd 48 8b 3c 24 e8 bc 9d ff ff 48 89 c7 48 c7 c6
> RSP: 0018:ffffc900097d6ee8 EFLAGS: 00010246
> RAX: 000000000000004e RBX: 0000000000000000 RCX: 2f9fd79be1350f00
> RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000
> RBP: dffffc0000000000 R08: ffffffff817402ac R09: 1ffff920012fad7c
> R10: dffffc0000000000 R11: fffff520012fad7d R12: ffffea0001ccbdc8
> R13: 1ffffd40003997b9 R14: fffffffffffffffd R15: 000000000007b000
> FS:  00007f11a84976c0(0000) GS:ffff8880b9300000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00007f11a0fff000 CR3: 0000000020e4c000 CR4: 00000000003506f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
>   <TASK>
>   __extent_writepage fs/btrfs/extent_io.c:1578 [inline]
>   extent_write_cache_pages fs/btrfs/extent_io.c:2251 [inline]
>   btrfs_writepages+0x12a2/0x2760 fs/btrfs/extent_io.c:2373
>   do_writepages+0x35d/0x870 mm/page-writeback.c:2683
>   filemap_fdatawrite_wbc+0x125/0x180 mm/filemap.c:397
>   __filemap_fdatawrite_range mm/filemap.c:430 [inline]
>   filemap_fdatawrite_range+0x120/0x180 mm/filemap.c:448
>   btrfs_fdatawrite_range+0x53/0xe0 fs/btrfs/file.c:3794
>   btrfs_direct_write+0x558/0xb40 fs/btrfs/direct-io.c:952
>   btrfs_do_write_iter+0x2a1/0x760 fs/btrfs/file.c:1505
>   do_iter_readv_writev+0x60a/0x890
>   vfs_writev+0x37c/0xbb0 fs/read_write.c:971
>   do_pwritev fs/read_write.c:1072 [inline]
>   __do_sys_pwritev2 fs/read_write.c:1131 [inline]
>   __se_sys_pwritev2+0x1ca/0x2d0 fs/read_write.c:1122
>   do_syscall_x64 arch/x86/entry/common.c:52 [inline]
>   do_syscall_64+0xf3/0x230 arch/x86/entry/common.c:83
>   entry_SYSCALL_64_after_hwframe+0x77/0x7f
> RIP: 0033:0x7f11a850bd79
> Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 41 1d 00 00 90 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 b0 ff ff ff f7 d8 64 89 01 48
> RSP: 002b:00007f11a8497158 EFLAGS: 00000212 ORIG_RAX: 0000000000000148
> RAX: ffffffffffffffda RBX: 00007f11a85916d8 RCX: 00007f11a850bd79
> RDX: 0000000000000001 RSI: 0000000020000240 RDI: 0000000000000004
> RBP: 00007f11a85916d0 R08: 0000000000000000 R09: 0000000000000003
> R10: 0000000000002000 R11: 0000000000000212 R12: 00007f11a85916dc
> R13: 000000000000006e R14: 00007ffd9dcad860 R15: 00007ffd9dcad948
>   </TASK>
> Modules linked in:
> ---[ end trace 0000000000000000 ]---
> RIP: 0010:__extent_writepage_io+0x1224/0x1400 fs/btrfs/extent_io.c:1488
> Code: f7 07 90 0f 0b e8 dc 68 df fd 48 c7 c7 e0 92 2c 8c 48 c7 c6 c0 a0 2c 8c 48 c7 c2 80 92 2c 8c b9 d0 05 00 00 e8 9d 15 f7 07 90 <0f> 0b e8 b5 68 df fd 48 8b 3c 24 e8 bc 9d ff ff 48 89 c7 48 c7 c6
> RSP: 0018:ffffc900097d6ee8 EFLAGS: 00010246
> RAX: 000000000000004e RBX: 0000000000000000 RCX: 2f9fd79be1350f00
> RDX: 0000000000000000 RSI: 0000000080000000 RDI: 0000000000000000
> RBP: dffffc0000000000 R08: ffffffff817402ac R09: 1ffff920012fad7c
> R10: dffffc0000000000 R11: fffff520012fad7d R12: ffffea0001ccbdc8
> R13: 1ffffd40003997b9 R14: fffffffffffffffd R15: 000000000007b000
> FS:  00007f11a84976c0(0000) GS:ffff8880b9200000(0000) knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: 00005654e2821068 CR3: 0000000020e4c000 CR4: 00000000003506f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>
>
> ---
> 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
>
> If the report is already addressed, let syzbot know by replying with:
> #syz fix: exact-commit-title
>
> 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.
>
> 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
>

Re: [syzbot] Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.

***

Subject: Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)
Author: lizhi.xu@windriver.com

#syz test

diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 744e8952abb0..c81b23d0d73e 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -262,6 +262,7 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em)
 			em->len += merge->len;
 			em->block_len += merge->block_len;
 			em->block_start = merge->block_start;
+			printk("em: %p, blockstart: %llu, mblockstart: %llu, %s\n", em, em->block_start, merge->block_start, __func__);
 			em->generation = max(em->generation, merge->generation);
 			em->flags |= EXTENT_FLAG_MERGED;
 
@@ -393,7 +394,11 @@ static int add_extent_mapping(struct btrfs_inode *inode,
 	if (ret)
 		return ret;
 
+	printk("em: %p, blockstart: %llu, em refs: %d, %s\n", 
+		em, em->block_start, refcount_read(&em->refs), __func__);
 	setup_extent_mapping(inode, em, modified);
+	printk("setuped, em: %p, blockstart: %llu, em refs: %d, %s\n", 
+		em, em->block_start, refcount_read(&em->refs), __func__);
 
 	if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(root)))
 		percpu_counter_inc(&fs_info->evictable_extent_maps);
@@ -643,6 +648,8 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode,
 			}
 			free_extent_map(existing);
 		}
+	} else if (!ret) {
+		ASSERT(em->block_start != EXTENT_MAP_HOLE);
 	}
 
 	ASSERT(ret == 0 || ret == -EEXIST);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 3a2b902b2d1f..b588381458df 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6972,7 +6972,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	}
 
 	write_lock(&em_tree->lock);
+	printk("em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", em, em->block_start, start, len, __func__);
 	ret = btrfs_add_extent_mapping(inode, &em, start, len);
+	printk("ret: %d, em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", ret, em, em->block_start, start, len, __func__);
 	write_unlock(&em_tree->lock);
 out:
 	btrfs_free_path(path);

Re: [syzbot] Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.

***

Subject: Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)
Author: lizhi.xu@windriver.com

#syz test

diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 10ac5f657e38..3a6d2b758587 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -384,6 +384,7 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em)
 		free_extent_map(merge);
 		dec_evictable_extent_maps(inode);
 	}
+	printk("em: %p, blockstart: %llu, mblockstart: %llu, %s\n", em, em->block_start, merge->block_start, __func__);
 }
 
 /*
@@ -493,7 +494,11 @@ static int add_extent_mapping(struct btrfs_inode *inode,
 	if (ret)
 		return ret;
 
+	printk("em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, em->block_start, refcount_read(&em->refs), __func__);
 	setup_extent_mapping(inode, em, modified);
+	printk("setuped, em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, em->block_start, refcount_read(&em->refs), __func__);
 
 	if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(root)))
 		percpu_counter_inc(&fs_info->evictable_extent_maps);
@@ -743,6 +748,8 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode,
 			}
 			free_extent_map(existing);
 		}
+	} else if (!ret) {
+		ASSERT(em->block_start != EXTENT_MAP_HOLE);
 	}
 
 	ASSERT(ret == 0 || ret == -EEXIST);
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b1b6564ab68f..d54c8d2ff1c1 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6958,7 +6958,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	}
 
 	write_lock(&em_tree->lock);
+	printk("em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", em, em->block_start, start, len, __func__);
 	ret = btrfs_add_extent_mapping(inode, &em, start, len);
+	printk("ret: %d, em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", ret, em, em->block_start, start, len, __func__);
 	write_unlock(&em_tree->lock);
 out:
 	btrfs_free_path(path);

Re: [syzbot] Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.

***

Subject: Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)
Author: lizhi.xu@windriver.com

#syz test

diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 10ac5f657e38..397eee4c4217 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -384,6 +384,7 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em)
 		free_extent_map(merge);
 		dec_evictable_extent_maps(inode);
 	}
+	printk("em: %p, blockstart: %llu, mblockstart: %llu, %s\n", em, extent_map_block_start(em), extent_map_block_start(merge), __func__);
 }
 
 /*
@@ -493,7 +494,11 @@ static int add_extent_mapping(struct btrfs_inode *inode,
 	if (ret)
 		return ret;
 
+	printk("em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, extent_map_block_start(em), refcount_read(&em->refs), __func__);
 	setup_extent_mapping(inode, em, modified);
+	printk("setuped, em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, extent_map_block_start(em), refcount_read(&em->refs), __func__);
 
 	if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(root)))
 		percpu_counter_inc(&fs_info->evictable_extent_maps);
@@ -743,6 +748,8 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode,
 			}
 			free_extent_map(existing);
 		}
+	} else if (!ret) {
+		ASSERT(extent_map_block_start(em) != EXTENT_MAP_HOLE);
 	}
 
 	ASSERT(ret == 0 || ret == -EEXIST);
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index 5154a8f1d26c..a84ef0c7d601 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -154,6 +154,7 @@ static inline u64 extent_map_block_start(const struct extent_map *em)
 	if (em->disk_bytenr < EXTENT_MAP_LAST_BYTE) {
 		if (extent_map_is_compressed(em))
 			return em->disk_bytenr;
+		printk("em: %p, disk byte nr: %llu, offset: %llu, %s\n", em, em->disk_bytenr, em->offset, __func__);
 		return em->disk_bytenr + em->offset;
 	}
 	return em->disk_bytenr;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b1b6564ab68f..45f3b31aacc3 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6958,7 +6958,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	}
 
 	write_lock(&em_tree->lock);
+	printk("em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", em, extent_map_block_start(em), start, len, __func__);
 	ret = btrfs_add_extent_mapping(inode, &em, start, len);
+	printk("ret: %d, em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", ret, em, extent_map_block_start(em), start, len, __func__);
 	write_unlock(&em_tree->lock);
 out:
 	btrfs_free_path(path);

Re: [syzbot] Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot]

For archival purposes, forwarding an incoming command email to
linux-kernel@vger.kernel.org.

***

Subject: Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)
Author: lizhi.xu@windriver.com

#syz test

diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 10ac5f657e38..fc6e83c6e69b 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -384,6 +384,8 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em)
 		free_extent_map(merge);
 		dec_evictable_extent_maps(inode);
 	}
+	printk("em: %p, blockstart: %llu, mblockstart: %llu, merged: %d, %s\n", em,
+		extent_map_block_start(em), extent_map_block_start(merge), em->flags & EXTENT_FLAG_MERGED, __func__);
 }
 
 /*
@@ -493,7 +495,11 @@ static int add_extent_mapping(struct btrfs_inode *inode,
 	if (ret)
 		return ret;
 
+	printk("em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, extent_map_block_start(em), refcount_read(&em->refs), __func__);
 	setup_extent_mapping(inode, em, modified);
+	printk("setuped, em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, extent_map_block_start(em), refcount_read(&em->refs), __func__);
 
 	if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(root)))
 		percpu_counter_inc(&fs_info->evictable_extent_maps);
@@ -743,6 +749,11 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode,
 			}
 			free_extent_map(existing);
 		}
+	} else if (!ret) {
+		if (em->flags & EXTENT_FLAG_MERGED)
+			ASSERT(extent_map_block_start(em) != EXTENT_MAP_HOLE);
+		else
+			return -EINVAL;
 	}
 
 	ASSERT(ret == 0 || ret == -EEXIST);
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index 5154a8f1d26c..a84ef0c7d601 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -154,6 +154,7 @@ static inline u64 extent_map_block_start(const struct extent_map *em)
 	if (em->disk_bytenr < EXTENT_MAP_LAST_BYTE) {
 		if (extent_map_is_compressed(em))
 			return em->disk_bytenr;
+		printk("em: %p, disk byte nr: %llu, offset: %llu, %s\n", em, em->disk_bytenr, em->offset, __func__);
 		return em->disk_bytenr + em->offset;
 	}
 	return em->disk_bytenr;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b1b6564ab68f..45f3b31aacc3 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6958,7 +6958,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	}
 
 	write_lock(&em_tree->lock);
+	printk("em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", em, extent_map_block_start(em), start, len, __func__);
 	ret = btrfs_add_extent_mapping(inode, &em, start, len);
+	printk("ret: %d, em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", ret, em, extent_map_block_start(em), start, len, __func__);
 	write_unlock(&em_tree->lock);
 out:
 	btrfs_free_path(path);

Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[Edward Adam Davis]

#syz test

diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c
index 10ac5f657e38..4e2c30acca67 100644
--- a/fs/btrfs/extent_map.c
+++ b/fs/btrfs/extent_map.c
@@ -384,6 +384,8 @@ static void try_merge_map(struct btrfs_inode *inode, struct extent_map *em)
 		free_extent_map(merge);
 		dec_evictable_extent_maps(inode);
 	}
+	printk("em: %p, blockstart: %llu, mblockstart: %llu, %s\n", em,
+		extent_map_block_start(em), em->flags & EXTENT_FLAG_MERGED, __func__);
 }
 
 /*
@@ -493,7 +495,11 @@ static int add_extent_mapping(struct btrfs_inode *inode,
 	if (ret)
 		return ret;
 
+	printk("em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, extent_map_block_start(em), refcount_read(&em->refs), __func__);
 	setup_extent_mapping(inode, em, modified);
+	printk("setuped, em: %p, blockstart: %llu, em refs: %d, %s\n",
+		em, extent_map_block_start(em), refcount_read(&em->refs), __func__);
 
 	if (!btrfs_is_testing(fs_info) && is_fstree(btrfs_root_id(root)))
 		percpu_counter_inc(&fs_info->evictable_extent_maps);
@@ -743,6 +749,11 @@ int btrfs_add_extent_mapping(struct btrfs_inode *inode,
 			}
 			free_extent_map(existing);
 		}
+	} else if (!ret) {
+		if (em->flags & EXTENT_FLAG_MERGED)
+			ASSERT(extent_map_block_start(em) != EXTENT_MAP_HOLE);
+		else
+			return -EINVAL;
 	}
 
 	ASSERT(ret == 0 || ret == -EEXIST);
diff --git a/fs/btrfs/extent_map.h b/fs/btrfs/extent_map.h
index 5154a8f1d26c..a84ef0c7d601 100644
--- a/fs/btrfs/extent_map.h
+++ b/fs/btrfs/extent_map.h
@@ -154,6 +154,7 @@ static inline u64 extent_map_block_start(const struct extent_map *em)
 	if (em->disk_bytenr < EXTENT_MAP_LAST_BYTE) {
 		if (extent_map_is_compressed(em))
 			return em->disk_bytenr;
+		printk("em: %p, disk byte nr: %llu, offset: %llu, %s\n", em, em->disk_bytenr, em->offset, __func__);
 		return em->disk_bytenr + em->offset;
 	}
 	return em->disk_bytenr;
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index b1b6564ab68f..45f3b31aacc3 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -6958,7 +6958,9 @@ struct extent_map *btrfs_get_extent(struct btrfs_inode *inode,
 	}
 
 	write_lock(&em_tree->lock);
+	printk("em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", em, extent_map_block_start(em), start, len, __func__);
 	ret = btrfs_add_extent_mapping(inode, &em, start, len);
+	printk("ret: %d, em: %p, blockstart: %llu, start: %llu, len: %llu, %s\n", ret, em, extent_map_block_start(em), start, len, __func__);
 	write_unlock(&em_tree->lock);
 out:
 	btrfs_free_path(path);

Re: [syzbot] [btrfs?] kernel BUG in __extent_writepage_io (2)

[syzbot]

Hello,

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

Reported-by: syzbot+ba3c0273042a898c230e@syzkaller.appspotmail.com
Tested-by: syzbot+ba3c0273042a898c230e@syzkaller.appspotmail.com

Tested on:

commit:         d5d547aa Merge tag 'random-6.11-rc6-for-linus' of git:..
git tree:       upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=167ccf8d980000
kernel config:  https://syzkaller.appspot.com/x/.config?x=a0455552d0b27491
dashboard link: https://syzkaller.appspot.com/bug?extid=ba3c0273042a898c230e
compiler:       Debian clang version 15.0.6, GNU ld (GNU Binutils for Debian) 2.40
patch:          https://syzkaller.appspot.com/x/patch.diff?x=10dcde33980000

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