linux-ext4.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2] ext2: don't set *count in the case of failure in ext2_try_to_allocate()
@ 2019-10-26  9:07 Chengguang Xu
  2019-10-30 10:15 ` Jan Kara
  0 siblings, 1 reply; 2+ messages in thread
From: Chengguang Xu @ 2019-10-26  9:07 UTC (permalink / raw)
  To: jack; +Cc: linux-ext4, Chengguang Xu

Currently we set *count to num(value 0) in the failure
of block allocation in ext2_try_to_allocate(). Without
reservation, we reuse *count(value 0) to retry block
allocation and wrong *count will cause only allocating
maximum 1 block even though having sufficent free blocks
in that block group. Finally, it probably cause significant
fragmentation.

Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>
---
v1->v2:
- Add detail explanation of effect to changelog.

 fs/ext2/balloc.c | 1 -
 1 file changed, 1 deletion(-)

diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c
index e0cc55164505..29fc3a5054f8 100644
--- a/fs/ext2/balloc.c
+++ b/fs/ext2/balloc.c
@@ -736,7 +736,6 @@ ext2_try_to_allocate(struct super_block *sb, int group,
 	*count = num;
 	return grp_goal - num;
 fail_access:
-	*count = num;
 	return -1;
 }
 
-- 
2.21.0




^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH v2] ext2: don't set *count in the case of failure in ext2_try_to_allocate()
  2019-10-26  9:07 [PATCH v2] ext2: don't set *count in the case of failure in ext2_try_to_allocate() Chengguang Xu
@ 2019-10-30 10:15 ` Jan Kara
  0 siblings, 0 replies; 2+ messages in thread
From: Jan Kara @ 2019-10-30 10:15 UTC (permalink / raw)
  To: Chengguang Xu; +Cc: jack, linux-ext4

On Sat 26-10-19 17:07:21, Chengguang Xu wrote:
> Currently we set *count to num(value 0) in the failure
> of block allocation in ext2_try_to_allocate(). Without
> reservation, we reuse *count(value 0) to retry block
> allocation and wrong *count will cause only allocating
> maximum 1 block even though having sufficent free blocks
> in that block group. Finally, it probably cause significant
> fragmentation.
> 
> Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>

Thanks! Patch applied.

								Honza

> ---
> v1->v2:
> - Add detail explanation of effect to changelog.
> 
>  fs/ext2/balloc.c | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/fs/ext2/balloc.c b/fs/ext2/balloc.c
> index e0cc55164505..29fc3a5054f8 100644
> --- a/fs/ext2/balloc.c
> +++ b/fs/ext2/balloc.c
> @@ -736,7 +736,6 @@ ext2_try_to_allocate(struct super_block *sb, int group,
>  	*count = num;
>  	return grp_goal - num;
>  fail_access:
> -	*count = num;
>  	return -1;
>  }
>  
> -- 
> 2.21.0
> 
> 
> 
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2019-10-30 10:15 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-10-26  9:07 [PATCH v2] ext2: don't set *count in the case of failure in ext2_try_to_allocate() Chengguang Xu
2019-10-30 10:15 ` Jan Kara

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).