linux-f2fs-devel.lists.sourceforge.net archive mirror
 help / color / mirror / Atom feed
From: Chao Yu <chao@kernel.org>
To: jaegeuk@kernel.org
Cc: linux-kernel@vger.kernel.org, linux-f2fs-devel@lists.sourceforge.net
Subject: [f2fs-dev] [PATCH 5/5] f2fs: fix to avoid race condition of atomic write
Date: Mon,  9 Jan 2023 11:44:53 +0800	[thread overview]
Message-ID: <20230109034453.490176-5-chao@kernel.org> (raw)
In-Reply-To: <20230109034453.490176-1-chao@kernel.org>

Thread A				Kworker
- application crashs
 - do_exit
  - close_files
   - filp_close
    - flush (f2fs_file_flush)
					- writepages
					 - f2fs_write_cache_pages
					  - f2fs_write_single_data_page
					   - f2fs_do_write_data_page
					    - check f2fs_is_atomic_file
     - f2fs_abort_atomic_write
      - check f2fs_is_atomic_file
      - iput(cow_inode)
      - cow_inode = NULL
					    - set_new_dnode(cow_inode)

Fix this issue by covering f2fs_do_write_data_page() with i_atomic_sem.

Fixes: 3db1de0e582c ("f2fs: change the current atomic write way")
Signed-off-by: Chao Yu <chao@kernel.org>
---
 fs/f2fs/data.c | 30 +++++++++++++++++++++++++-----
 1 file changed, 25 insertions(+), 5 deletions(-)

diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index c940da1c540f..1645b8a1b904 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -2637,13 +2637,24 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio)
 	struct dnode_of_data dn;
 	struct node_info ni;
 	bool ipu_force = false;
+	bool atomic_locked = false;
 	int err = 0;
 
 	/* Use COW inode to make dnode_of_data for atomic write */
-	if (f2fs_is_atomic_file(inode))
+	if (f2fs_is_atomic_file(inode)) {
+		f2fs_down_write(&F2FS_I(inode)->i_atomic_sem);
+		atomic_locked = true;
+
+		if (!f2fs_is_atomic_file(inode)) {
+			/* atomic write is aborted */
+			err = -ENOENT;
+			goto out_err;
+		}
+
 		set_new_dnode(&dn, F2FS_I(inode)->cow_inode, NULL, NULL, 0);
-	else
+	} else {
 		set_new_dnode(&dn, inode, NULL, NULL, 0);
+	}
 
 	if (need_inplace_update(fio) &&
 	    f2fs_lookup_read_extent_cache_block(inode, page->index,
@@ -2652,7 +2663,8 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio)
 						DATA_GENERIC_ENHANCE)) {
 			f2fs_handle_error(fio->sbi,
 						ERROR_INVALID_BLKADDR);
-			return -EFSCORRUPTED;
+			err = -EFSCORRUPTED;
+			goto out_err;
 		}
 
 		ipu_force = true;
@@ -2661,8 +2673,10 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio)
 	}
 
 	/* Deadlock due to between page->lock and f2fs_lock_op */
-	if (fio->need_lock == LOCK_REQ && !f2fs_trylock_op(fio->sbi))
-		return -EAGAIN;
+	if (fio->need_lock == LOCK_REQ && !f2fs_trylock_op(fio->sbi)) {
+		err = -EAGAIN;
+		goto out_err;
+	}
 
 	err = f2fs_get_dnode_of_data(&dn, page->index, LOOKUP_NODE);
 	if (err)
@@ -2710,6 +2724,9 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio)
 			set_inode_flag(inode, FI_UPDATE_WRITE);
 		}
 		trace_f2fs_do_write_data_page(fio->page, IPU);
+
+		if (atomic_locked)
+			f2fs_up_write(&F2FS_I(inode)->i_atomic_sem);
 		return err;
 	}
 
@@ -2747,6 +2764,9 @@ int f2fs_do_write_data_page(struct f2fs_io_info *fio)
 out:
 	if (fio->need_lock == LOCK_REQ)
 		f2fs_unlock_op(fio->sbi);
+out_err:
+	if (atomic_locked)
+		f2fs_up_write(&F2FS_I(inode)->i_atomic_sem);
 	return err;
 }
 
-- 
2.25.1



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

  parent reply	other threads:[~2023-01-09  3:45 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-01-09  3:44 [f2fs-dev] [PATCH 1/5] f2fs: introduce trace_f2fs_replace_atomic_write_block Chao Yu
2023-01-09  3:44 ` [f2fs-dev] [PATCH 2/5] f2fs: clear atomic_write_task in f2fs_abort_atomic_write() Chao Yu
2023-01-09  3:44 ` [f2fs-dev] [PATCH 3/5] f2fs: fix to abort atomic write only during do_exist() Chao Yu
2023-01-09  3:44 ` [f2fs-dev] [PATCH 4/5] f2fs: fix to avoid race condition of f2fs_abort_atomic_write() Chao Yu
2023-01-09  3:44 ` Chao Yu [this message]
2023-01-30 23:00 ` [f2fs-dev] [PATCH 1/5] f2fs: introduce trace_f2fs_replace_atomic_write_block patchwork-bot+f2fs
2023-01-31 19:10 ` patchwork-bot+f2fs

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20230109034453.490176-5-chao@kernel.org \
    --to=chao@kernel.org \
    --cc=jaegeuk@kernel.org \
    --cc=linux-f2fs-devel@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).