[Cluster-devel] [GFS2 PATCH] GFS2: Use single-block reservations for directories

[Cluster-devel] [GFS2 PATCH] GFS2: Use single-block reservations for directories

[Bob Peterson]

Hi,

This patch changes the multi-block allocation code, such that
directory inodes only get a single block reserved in the bitmap.
That way, the bitmaps are more tightly packed together, and there
are fewer spans of free blocks for in-use block reservations.
This means it takes less time to find a free span of blocks in the
bitmap, which speeds things up. This increases the performance of
some workloads by almost 2X. In Nate's mockup.py script (which does
(1) create dir, (2) create dir in dir, (3) create file in that dir)
the test executes in 23 steps rather than 43 steps, a 47%
performance improvement.

Regards,

Bob Peterson
Red Hat File Systems

Signed-off-by: Bob Peterson <rpeterso@redhat.com> 
---
diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
index 0c5a575..5232525 100644
--- a/fs/gfs2/rgrp.c
+++ b/fs/gfs2/rgrp.c
@@ -1401,9 +1401,14 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip,
 	u32 extlen;
 	u32 free_blocks = rgd->rd_free_clone - rgd->rd_reserved;
 	int ret;
+	struct inode *inode = &ip->i_inode;
 
-	extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested);
-	extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks);
+	if (S_ISDIR(inode->i_mode))
+		extlen = 1;
+	else {
+		extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested);
+		extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks);
+	}
 	if ((rgd->rd_free_clone < rgd->rd_reserved) || (free_blocks < extlen))
 		return;
 

[Cluster-devel] [GFS2 PATCH] GFS2: Use single-block reservations for directories

[Steven Whitehouse]

Hi,

Now in the -nmw tree. Thanks,

Steve.

On Tue, 2013-05-14 at 13:04 -0400, Bob Peterson wrote:
> Hi,
> 
> This patch changes the multi-block allocation code, such that
> directory inodes only get a single block reserved in the bitmap.
> That way, the bitmaps are more tightly packed together, and there
> are fewer spans of free blocks for in-use block reservations.
> This means it takes less time to find a free span of blocks in the
> bitmap, which speeds things up. This increases the performance of
> some workloads by almost 2X. In Nate's mockup.py script (which does
> (1) create dir, (2) create dir in dir, (3) create file in that dir)
> the test executes in 23 steps rather than 43 steps, a 47%
> performance improvement.
> 
> Regards,
> 
> Bob Peterson
> Red Hat File Systems
> 
> Signed-off-by: Bob Peterson <rpeterso@redhat.com> 
> ---
> diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c
> index 0c5a575..5232525 100644
> --- a/fs/gfs2/rgrp.c
> +++ b/fs/gfs2/rgrp.c
> @@ -1401,9 +1401,14 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip,
>  	u32 extlen;
>  	u32 free_blocks = rgd->rd_free_clone - rgd->rd_reserved;
>  	int ret;
> +	struct inode *inode = &ip->i_inode;
>  
> -	extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested);
> -	extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks);
> +	if (S_ISDIR(inode->i_mode))
> +		extlen = 1;
> +	else {
> +		extlen = max_t(u32, atomic_read(&rs->rs_sizehint), requested);
> +		extlen = clamp(extlen, RGRP_RSRV_MINBLKS, free_blocks);
> +	}
>  	if ((rgd->rd_free_clone < rgd->rd_reserved) || (free_blocks < extlen))
>  		return;
>  
>