BUG_ON() will be triggered when writing files concurrently,
because the same page is writtenback multiple times.
1597 void folio_end_writeback(struct folio *folio)
1598 {
......
1618 if (!__folio_end_writeback(folio))
1619 BUG();
......
1625 }
kernel BUG at mm/filemap.c:1619!
Call Trace:
<TASK>
f2fs_write_end_io+0x1a0/0x370
blk_update_request+0x6c/0x410
blk_mq_end_request+0x15/0x130
blk_complete_reqs+0x3c/0x50
__do_softirq+0xb8/0x29b
? sort_range+0x20/0x20
run_ksoftirqd+0x19/0x20
smpboot_thread_fn+0x10b/0x1d0
kthread+0xde/0x110
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x22/0x30
</TASK>
Below is the concurrency scenario:
[Process A] [Process B] [Process C]
f2fs_write_raw_pages()
- redirty_page_for_writepage()
- unlock page()
f2fs_do_write_data_page()
- lock_page()
- clear_page_dirty_for_io()
- set_page_writeback() [1st writeback]
.....
- unlock page()
generic_perform_write()
- f2fs_write_begin()
- wait_for_stable_page()
- f2fs_write_end()
- set_page_dirty()
- lock_page()
- f2fs_do_write_data_page()
- set_page_writeback() [2st writeback]
This problem was introduced by the previous commit 7377e853967b ("f2fs:
compress: fix potential deadlock of compress file"). All pagelocks were
released in f2fs_write_raw_pages(), but whether the page was
in the writeback state was ignored in the subsequent writing process.
Let's fix it by waiting for the page to writeback before writing.
Cc: Christoph Hellwig <[email protected]>
Fixes: 4c8ff7095bef ("f2fs: support data compression")
Fixes: 7377e853967b ("f2fs: compress: fix potential deadlock of compress file")
Signed-off-by: Qi Han <[email protected]>
Signed-off-by: Yangtao Li <[email protected]>
---
v2:
-avoid the else
fs/f2fs/compress.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c
index 93fec1d37899..9b7149534a58 100644
--- a/fs/f2fs/compress.c
+++ b/fs/f2fs/compress.c
@@ -1456,6 +1456,12 @@ static int f2fs_write_raw_pages(struct compress_ctx *cc,
if (!PageDirty(cc->rpages[i]))
goto continue_unlock;
+ if (PageWriteback(cc->rpages[i])) {
+ if (wbc->sync_mode == WB_SYNC_NONE)
+ goto continue_unlock;
+ f2fs_wait_on_page_writeback(cc->rpages[i], DATA, true, true);
+ }
+
if (!clear_page_dirty_for_io(cc->rpages[i]))
goto continue_unlock;
--
2.35.1
On 2023/3/21 1:22, Yangtao Li wrote:
> BUG_ON() will be triggered when writing files concurrently,
> because the same page is writtenback multiple times.
>
> 1597 void folio_end_writeback(struct folio *folio)
> 1598 {
> ......
> 1618 if (!__folio_end_writeback(folio))
> 1619 BUG();
> ......
> 1625 }
>
> kernel BUG at mm/filemap.c:1619!
> Call Trace:
> <TASK>
> f2fs_write_end_io+0x1a0/0x370
> blk_update_request+0x6c/0x410
> blk_mq_end_request+0x15/0x130
> blk_complete_reqs+0x3c/0x50
> __do_softirq+0xb8/0x29b
> ? sort_range+0x20/0x20
> run_ksoftirqd+0x19/0x20
> smpboot_thread_fn+0x10b/0x1d0
> kthread+0xde/0x110
> ? kthread_complete_and_exit+0x20/0x20
> ret_from_fork+0x22/0x30
> </TASK>
>
> Below is the concurrency scenario:
>
> [Process A] [Process B] [Process C]
> f2fs_write_raw_pages()
> - redirty_page_for_writepage()
> - unlock page()
> f2fs_do_write_data_page()
> - lock_page()
> - clear_page_dirty_for_io()
> - set_page_writeback() [1st writeback]
> .....
> - unlock page()
>
> generic_perform_write()
> - f2fs_write_begin()
> - wait_for_stable_page()
>
> - f2fs_write_end()
> - set_page_dirty()
>
> - lock_page()
> - f2fs_do_write_data_page()
> - set_page_writeback() [2st writeback]
>
> This problem was introduced by the previous commit 7377e853967b ("f2fs:
> compress: fix potential deadlock of compress file"). All pagelocks were
> released in f2fs_write_raw_pages(), but whether the page was
> in the writeback state was ignored in the subsequent writing process.
> Let's fix it by waiting for the page to writeback before writing.
>
> Cc: Christoph Hellwig <[email protected]>
> Fixes: 4c8ff7095bef ("f2fs: support data compression")
> Fixes: 7377e853967b ("f2fs: compress: fix potential deadlock of compress file")
> Signed-off-by: Qi Han <[email protected]>
> Signed-off-by: Yangtao Li <[email protected]>
Reviewed-by: Chao Yu <[email protected]>
Thanks,
Hello:
This patch was applied to jaegeuk/f2fs.git (dev)
by Jaegeuk Kim <[email protected]>:
On Tue, 21 Mar 2023 01:22:18 +0800 you wrote:
> BUG_ON() will be triggered when writing files concurrently,
> because the same page is writtenback multiple times.
>
> 1597 void folio_end_writeback(struct folio *folio)
> 1598 {
> ......
> 1618 if (!__folio_end_writeback(folio))
> 1619 BUG();
> ......
> 1625 }
>
> [...]
Here is the summary with links:
- [f2fs-dev,v2] f2fs: compress: fix to call f2fs_wait_on_page_writeback() in f2fs_write_raw_pages()
https://git.kernel.org/jaegeuk/f2fs/c/08a8457cfd78
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html