Re: [PATCH v2 24/34] gfs2: Remove use of b_end_io in gfs2_meta_read_endio()

Jan Kara <[email protected]> Sat, 30 May 2026 14:09:41 +0200
Newsgroups dev.linux.lists.gfs2,org.kernel.vger.linux-fsdevel
Message-ID <bowj5ekdjodcabs54zgceka5u5eyvgxermwyerttxyj4whutxg@iea6aqy5hfwj>
On Thu 28-05-26 18:31:37, Matthew Wilcox (Oracle) wrote:
> All buffer heads submitted by gfs2_submit_bhs() use
> end_buffer_read_sync() so we can call it directly.
> 
> Signed-off-by: Matthew Wilcox (Oracle) <[email protected]>
> Cc: [email protected]

Looks good. Feel free to add:

Reviewed-by: Jan Kara <[email protected]>

								Honza

> ---
>  fs/gfs2/meta_io.c | 4 +---
>  1 file changed, 1 insertion(+), 3 deletions(-)
> 
> diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c
> index d407dd476e72..df5f05054f6e 100644
> --- a/fs/gfs2/meta_io.c
> +++ b/fs/gfs2/meta_io.c
> @@ -212,7 +212,7 @@ static void gfs2_meta_read_endio(struct bio *bio)
>  		do {
>  			struct buffer_head *next = bh->b_this_page;
>  			len -= bh->b_size;
> -			bh->b_end_io(bh, !bio->bi_status);
> +			end_buffer_read_sync(bh, bio->bi_status == BLK_STS_OK);
>  			bh = next;
>  		} while (bh && len);
>  	}
> @@ -275,7 +275,6 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags,
>  		unlock_buffer(bh);
>  		flags &= ~DIO_WAIT;
>  	} else {
> -		bh->b_end_io = end_buffer_read_sync;
>  		get_bh(bh);
>  		bhs[num++] = bh;
>  	}
> @@ -288,7 +287,6 @@ int gfs2_meta_read(struct gfs2_glock *gl, u64 blkno, int flags,
>  			unlock_buffer(bh);
>  			brelse(bh);
>  		} else {
> -			bh->b_end_io = end_buffer_read_sync;
>  			bhs[num++] = bh;
>  		}
>  	}
> -- 
> 2.47.3
> 
-- 
Jan Kara <[email protected]>
SUSE Labs, CR