On Thu, Feb 06, 2025 at 03:31:02PM GMT, Greg Kroah-Hartman wrote:
On Thu, Feb 06, 2025 at 08:41:33PM +0900, Koichiro Den wrote:
On Mon, Dec 30, 2024 at 04:43:39PM GMT, Greg Kroah-Hartman wrote:
6.12-stable review patch. If anyone has any objections, please let me know.
From: Filipe Manana fdmanana@suse.com
commit 2c8507c63f5498d4ee4af404a8e44ceae4345056 upstream.
During swap activation we iterate over the extents of a file and we can have many thousands of them, so we can end up in a busy loop monopolizing a core. Avoid this by doing a voluntary reschedule after processing each extent.
CC: stable@vger.kernel.org # 5.4+ Reviewed-by: Qu Wenruo wqu@suse.com Signed-off-by: Filipe Manana fdmanana@suse.com Signed-off-by: David Sterba dsterba@suse.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org
fs/btrfs/inode.c | 2 ++ 1 file changed, 2 insertions(+)
--- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7117,6 +7117,8 @@ noinline int can_nocow_extent(struct ino ret = -EAGAIN; goto out; }
cond_resched();if (file_extent)
Hi, please let me confirm; is this backport really ok? I mean, should the cond_resched() be added to btrfs_swap_activate() loop? I was able to reproduce the same situation:
$ git rev-parse HEAD 319addc2ad901dac4d6cc931d77ef35073e0942f $ b4 mbox --single-message c37ea7a8de12e996091ba295b2f201fbe680c96c.1733929328.git.fdmanana@suse.com 1 messages in the thread Saved ./c37ea7a8de12e996091ba295b2f201fbe680c96c.1733929328.git.fdmanana@suse.com.mbx $ patch -p1 < ./c37ea7a8de12e996091ba295b2f201fbe680c96c.1733929328.git.fdmanana@suse.com.mbx patching file fs/btrfs/inode.c Hunk #1 succeeded at 7117 with fuzz 1 (offset -2961 lines). $ git diff diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 58ffe78132d9..6fe2ac620464 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7117,6 +7117,8 @@ noinline int can_nocow_extent(struct inode *inode, u64 offset, u64 *len, ret = -EAGAIN; goto out; } + + cond_resched(); }if (file_extent)
The same goes for all the other stable branches applied. Sorry if I'm missing something.
Hm, looks like patch messed this up :(
Can you send a revert for the branches that this was wrong on, and then the correct fix, and I'll be glad to queue them all up.
Sure, I'll send them later. Thanks for the quick response!
-Koichiro
thanks,
greg k-h