FAILED: patch "[PATCH] ocfs2: fix data corruption after failed write" failed to apply to 5.10-stable tree

Commit Message

David Sterba via Ocfs2-devel March 20, 2023, 11:57 a.m. UTC

The patch below does not apply to the 5.10-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <[e-mail]>.

To reproduce the conflict and resubmit, you may use the following commands:

git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y
git checkout FETCH_HEAD
git cherry-pick -x 90410bcf873cf05f54a32183afff0161f44f9715
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[e-mail]>' --in-reply-to '[e-mail]' --subject-prefix 'PATCH 5.10.y' HEAD^..

Possible dependencies:

90410bcf873c ("ocfs2: fix data corruption after failed write")

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 90410bcf873cf05f54a32183afff0161f44f9715 Mon Sep 17 00:00:00 2001
From: Jan Kara via Ocfs2-devel <[e-mail]>
Date: Thu, 2 Mar 2023 16:38:43 +0100
Subject: [PATCH] ocfs2: fix data corruption after failed write

When buffered write fails to copy data into underlying page cache page,
ocfs2_write_end_nolock() just zeroes out and dirties the page.  This can
leave dirty page beyond EOF and if page writeback tries to write this page
before write succeeds and expands i_size, page gets into inconsistent
state where page dirty bit is clear but buffer dirty bits stay set
resulting in page data never getting written and so data copied to the
page is lost.  Fix the problem by invalidating page beyond EOF after
failed write.

Link: https://lkml.kernel.org/r/[e-mail]
Fixes: 6dbf7bb55598 ("fs: Don't invalidate page buffers in block_write_full_page()")
Signed-off-by: Jan Kara <[e-mail]>
Reviewed-by: Joseph Qi <[e-mail]>
Cc: Mark Fasheh <[e-mail]>
Cc: Joel Becker <[e-mail]>
Cc: Junxiao Bi <[e-mail]>
Cc: Changwei Ge <[e-mail]>
Cc: Gang He <[e-mail]>
Cc: Jun Piao <[e-mail]>
Cc: <[e-mail]>
Signed-off-by: Andrew Morton <[e-mail]>

Comments

Joseph Qi March 21, 2023, 2:06 a.m. UTC | #1

Hi Greg,
It can be cleanly applied for linux-5.10.y and linux-4.19.y in my desktop.
I'm not sure how it happens.

Thanks,
Joseph

On 3/20/23 7:57 PM, gregkh--- via Ocfs2-devel wrote:
> 
> The patch below does not apply to the 5.10-stable tree.
> If someone wants it applied there, or to any other stable or longterm
> tree, then please email the backport, including the original git commit
> id to <[e-mail]>.
> 
> To reproduce the conflict and resubmit, you may use the following commands:
> 
> git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-5.10.y
> git checkout FETCH_HEAD
> git cherry-pick -x 90410bcf873cf05f54a32183afff0161f44f9715
> # <resolve conflicts, build, test, etc.>
> git commit -s
> git send-email --to '<[e-mail]>' --in-reply-to '[e-mail]' --subject-prefix 'PATCH 5.10.y' HEAD^..
> 
> Possible dependencies:
> 
> 90410bcf873c ("ocfs2: fix data corruption after failed write")
> 
> thanks,
> 
> greg k-h
> 
> ------------------ original commit in Linus's tree ------------------
> 
> From 90410bcf873cf05f54a32183afff0161f44f9715 Mon Sep 17 00:00:00 2001
> From: Jan Kara via Ocfs2-devel <[e-mail]>
> Date: Thu, 2 Mar 2023 16:38:43 +0100
> Subject: [PATCH] ocfs2: fix data corruption after failed write
> 
> When buffered write fails to copy data into underlying page cache page,
> ocfs2_write_end_nolock() just zeroes out and dirties the page.  This can
> leave dirty page beyond EOF and if page writeback tries to write this page
> before write succeeds and expands i_size, page gets into inconsistent
> state where page dirty bit is clear but buffer dirty bits stay set
> resulting in page data never getting written and so data copied to the
> page is lost.  Fix the problem by invalidating page beyond EOF after
> failed write.
> 
> Link: https://lkml.kernel.org/r/[e-mail]
> Fixes: 6dbf7bb55598 ("fs: Don't invalidate page buffers in block_write_full_page()")
> Signed-off-by: Jan Kara <[e-mail]>
> Reviewed-by: Joseph Qi <[e-mail]>
> Cc: Mark Fasheh <[e-mail]>
> Cc: Joel Becker <[e-mail]>
> Cc: Junxiao Bi <[e-mail]>
> Cc: Changwei Ge <[e-mail]>
> Cc: Gang He <[e-mail]>
> Cc: Jun Piao <[e-mail]>
> Cc: <[e-mail]>
> Signed-off-by: Andrew Morton <[e-mail]>
> 
> diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
> index 1d65f6ef00ca..0394505fdce3 100644
> --- a/fs/ocfs2/aops.c
> +++ b/fs/ocfs2/aops.c
> @@ -1977,11 +1977,26 @@ int ocfs2_write_end_nolock(struct address_space *mapping,
>  	}
>  
>  	if (unlikely(copied < len) && wc->w_target_page) {
> +		loff_t new_isize;
> +
>  		if (!PageUptodate(wc->w_target_page))
>  			copied = 0;
>  
> -		ocfs2_zero_new_buffers(wc->w_target_page, start+copied,
> -				       start+len);
> +		new_isize = max_t(loff_t, i_size_read(inode), pos + copied);
> +		if (new_isize > page_offset(wc->w_target_page))
> +			ocfs2_zero_new_buffers(wc->w_target_page, start+copied,
> +					       start+len);
> +		else {
> +			/*
> +			 * When page is fully beyond new isize (data copy
> +			 * failed), do not bother zeroing the page. Invalidate
> +			 * it instead so that writeback does not get confused
> +			 * put page & buffer dirty bits into inconsistent
> +			 * state.
> +			 */
> +			block_invalidate_folio(page_folio(wc->w_target_page),
> +						0, PAGE_SIZE);
> +		}
>  	}
>  	if (wc->w_target_page)
>  		flush_dcache_page(wc->w_target_page);
> 
> 
> _______________________________________________
> Ocfs2-devel mailing list
> [e-mail]
> https://oss.oracle.com/mailman/listinfo/ocfs2-devel

Bagas Sanjaya March 21, 2023, 2:13 a.m. UTC | #2

On Tue, Mar 21, 2023 at 10:06:30AM +0800, Joseph Qi wrote:
> Hi Greg,
> It can be cleanly applied for linux-5.10.y and linux-4.19.y in my desktop.
> I'm not sure how it happens.
> 

If you can apply the backport, why don't you post it here?

Joseph Qi March 21, 2023, 2:21 a.m. UTC | #3

On 3/21/23 10:13 AM, Bagas Sanjaya wrote:
> On Tue, Mar 21, 2023 at 10:06:30AM +0800, Joseph Qi wrote:
>> Hi Greg,
>> It can be cleanly applied for linux-5.10.y and linux-4.19.y in my desktop.
>> I'm not sure how it happens.
>>
> 
> If you can apply the backport, why don't you post it here?
> 
My mistake, it requires feature folio.
Will adapt and post it later.

Thanks,
Joseph

Patch

diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 1d65f6ef00ca..0394505fdce3 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -1977,11 +1977,26 @@  int ocfs2_write_end_nolock(struct address_space *mapping,
 	}
 
 	if (unlikely(copied < len) && wc->w_target_page) {
+		loff_t new_isize;
+
 		if (!PageUptodate(wc->w_target_page))
 			copied = 0;
 
-		ocfs2_zero_new_buffers(wc->w_target_page, start+copied,
-				       start+len);
+		new_isize = max_t(loff_t, i_size_read(inode), pos + copied);
+		if (new_isize > page_offset(wc->w_target_page))
+			ocfs2_zero_new_buffers(wc->w_target_page, start+copied,
+					       start+len);
+		else {
+			/*
+			 * When page is fully beyond new isize (data copy
+			 * failed), do not bother zeroing the page. Invalidate
+			 * it instead so that writeback does not get confused
+			 * put page & buffer dirty bits into inconsistent
+			 * state.
+			 */
+			block_invalidate_folio(page_folio(wc->w_target_page),
+						0, PAGE_SIZE);
+		}
 	}
 	if (wc->w_target_page)
 		flush_dcache_page(wc->w_target_page);