All of lore.kernel.org
 help / color / mirror / Atom feed
* [f2fs-dev] [PATCH -next] f2fs: fix min_seq_blocks can not make sense in some scenes.
@ 2021-07-31  3:26 Laibin Qiu
  2021-08-01  7:08 ` Chao Yu
  0 siblings, 1 reply; 2+ messages in thread
From: Laibin Qiu @ 2021-07-31  3:26 UTC (permalink / raw)
  To: jaegeuk, chao; +Cc: linux-f2fs-devel

F2FS have dirty page count control for batched sequential
write in writepages, and get the value of min_seq_blocks by
blocks_per_seg * segs_per_sec(segs_per_sec defaults to 1).
But in some scenes we set a lager section size, Min_seq_blocks
will become too large to achieve the expected effect(eg. 4thread
sequential write, the number of merge requests will be reduced).

Signed-off-by: Laibin Qiu <qiulaibin@huawei.com>
---
 fs/f2fs/segment.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index f9b7fb785e1d..0f976cefe425 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -5159,7 +5159,7 @@ int f2fs_build_segment_manager(struct f2fs_sb_info *sbi)
 		sm_info->ipu_policy = 1 << F2FS_IPU_FSYNC;
 	sm_info->min_ipu_util = DEF_MIN_IPU_UTIL;
 	sm_info->min_fsync_blocks = DEF_MIN_FSYNC_BLOCKS;
-	sm_info->min_seq_blocks = sbi->blocks_per_seg * sbi->segs_per_sec;
+	sm_info->min_seq_blocks = sbi->blocks_per_seg;
 	sm_info->min_hot_blocks = DEF_MIN_HOT_BLOCKS;
 	sm_info->min_ssr_sections = reserved_sections(sbi);
 
-- 
2.25.1



_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

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

* Re: [f2fs-dev] [PATCH -next] f2fs: fix min_seq_blocks can not make sense in some scenes.
  2021-07-31  3:26 [f2fs-dev] [PATCH -next] f2fs: fix min_seq_blocks can not make sense in some scenes Laibin Qiu
@ 2021-08-01  7:08 ` Chao Yu
  0 siblings, 0 replies; 2+ messages in thread
From: Chao Yu @ 2021-08-01  7:08 UTC (permalink / raw)
  To: Laibin Qiu, jaegeuk; +Cc: linux-f2fs-devel

On 2021/7/31 11:26, Laibin Qiu wrote:
> F2FS have dirty page count control for batched sequential
> write in writepages, and get the value of min_seq_blocks by
> blocks_per_seg * segs_per_sec(segs_per_sec defaults to 1).
> But in some scenes we set a lager section size, Min_seq_blocks
> will become too large to achieve the expected effect(eg. 4thread
> sequential write, the number of merge requests will be reduced).

I guess we can fix this problem by configuring
/sys/fs/f2fs/<dev>/min_seq_blocks for your case.

But, IMO, it doesn't make sense that default min_seq_blocks configuration
should be related to section size.

> 
> Signed-off-by: Laibin Qiu <qiulaibin@huawei.com>

Reviewed-by: Chao Yu <chao@kernel.org>

Thanks,


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

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

end of thread, other threads:[~2021-08-01  7:08 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-31  3:26 [f2fs-dev] [PATCH -next] f2fs: fix min_seq_blocks can not make sense in some scenes Laibin Qiu
2021-08-01  7:08 ` Chao Yu

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.