On Wed, Oct 19, 2016 at 12:57:19PM +0800, Wang Xiaoguang wrote:
> hi,
>
> On 10/18/2016 06:32 PM, Holger Hoffstätte wrote:
> > On Tue, 18 Oct 2016 15:56:13 +0800, Wang Xiaoguang wrote:
> >
> >> In btrfs_truncate_inode_items()->btrfs_async_run_delayed_refs(), we
> >> swap the arg2 and arg3 wrongly, fix this.
> >>
> >> Signed-off-by: Wang Xiaoguang <wangxg.fnst@xxxxxxxxxxxxxx>
> >> ---
> >> fs/btrfs/inode.c | 4 ++--
> >> 1 file changed, 2 insertions(+), 2 deletions(-)
> >>
> >> diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
> >> index 2b790bd..2f1372b 100644
> >> --- a/fs/btrfs/inode.c
> >> +++ b/fs/btrfs/inode.c
> >> @@ -4605,8 +4605,8 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans,
> >> BUG_ON(ret);
> >> if (btrfs_should_throttle_delayed_refs(trans, root))
> >> btrfs_async_run_delayed_refs(root,
> >> - trans->transid,
> >> - trans->delayed_ref_updates * 2, 0);
> >> + trans->delayed_ref_updates * 2,
> >> + trans->transid, 0);
> >> if (be_nice) {
> >> if (truncate_space_check(trans, root,
> >> extent_num_bytes)) {
> > Reviewed-by: Holger Hoffstätte <holger@xxxxxxxxxxxxxxxxxxxxxx>
> >
> > Passing the wrong transid..why did this ever work?
> Indeed this bug just impacts asynchronous delayed refs handle when we
> truncate
> inodes. In delayed_ref_async_start(), there is such codes:
>
> trans = btrfs_join_transaction(async->root);
> if (trans->transid > async->transid)
> goto end;
> ret = btrfs_run_delayed_refs(trans, async->root, async->count);
>
> From this codes, we can see that this just influence whether can we handle
> delayed refs or the number of delayed refs to handle, this may impact
> performance, but will not result in missing delayed refs, all delayed
> refs will
> be handled in btrfs_commit_transaction().
You should put this kind of information to the changelog itself, now
added.
--
To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
