From: Wu Bo via Linux-f2fs-devel <linux-f2fs-devel@lists.sourceforge.net>
To: Jaegeuk Kim <jaegeuk@kernel.org>, Chao Yu <chao@kernel.org>
Cc: Wu Bo <wubo.oduw@gmail.com>, Wu Bo <bo.wu@vivo.com>,
linux-f2fs-devel@lists.sourceforge.net
Subject: [f2fs-dev] [PATCH v2 3/3] f2fs-tools: cache free segments count to improve perfmance
Date: Thu, 28 Sep 2023 03:20:51 -0600 [thread overview]
Message-ID: <188b05d1764d1ba48f9c704ff28df853f3094f67.1695890686.git.bo.wu@vivo.com> (raw)
In-Reply-To: <cover.1695890686.git.bo.wu@vivo.com>
'get_free_segments()' is implemented by traversing all segments to
calculate the total free segments. It cosume much time.
Every time when call 'find_next_free_block()' this calculation will
do it again. So if we cache the free segments count, it will greatly
improve performance of dfrag & resize & sload.
---
fsck/f2fs.h | 1 +
fsck/mount.c | 9 +++++++--
fsck/segment.c | 2 ++
3 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/fsck/f2fs.h b/fsck/f2fs.h
index e65644e..a7cdc4c 100644
--- a/fsck/f2fs.h
+++ b/fsck/f2fs.h
@@ -197,6 +197,7 @@ struct f2fs_sm_info {
unsigned int main_segments;
unsigned int reserved_segments;
unsigned int ovp_segments;
+ unsigned int free_segments;
};
struct f2fs_dentry_ptr {
diff --git a/fsck/mount.c b/fsck/mount.c
index 098e73d..0a37bc4 100644
--- a/fsck/mount.c
+++ b/fsck/mount.c
@@ -123,7 +123,7 @@ void update_free_segments(struct f2fs_sb_info *sbi)
if (c.dbg_lv)
return;
- MSG(0, "\r [ %c ] Free segments: 0x%x", progress[i % 5], get_free_segments(sbi));
+ MSG(0, "\r [ %c ] Free segments: 0x%x", progress[i % 5], SM_I(sbi)->free_segments);
fflush(stdout);
i++;
}
@@ -2430,6 +2430,10 @@ static int build_sit_entries(struct f2fs_sb_info *sbi)
check_block_count(sbi, segno, &sit);
seg_info_from_raw_sit(sbi, se, &sit);
+ if (se->valid_blocks == 0x0 &&
+ is_usable_seg(sbi, segno) &&
+ !IS_CUR_SEGNO(sbi, segno))
+ SM_I(sbi)->free_segments++;
}
start_blk += readed;
} while (start_blk < sit_blk_cnt);
@@ -2485,6 +2489,7 @@ static int early_build_segment_manager(struct f2fs_sb_info *sbi)
sm_info->ovp_segments = get_cp(overprov_segment_count);
sm_info->main_segments = get_sb(segment_count_main);
sm_info->ssa_blkaddr = get_sb(ssa_blkaddr);
+ sm_info->free_segments = 0;
if (build_sit_info(sbi) || build_curseg(sbi)) {
free(sm_info);
@@ -2806,7 +2811,7 @@ int find_next_free_block(struct f2fs_sb_info *sbi, u64 *to, int left,
if (*to > 0)
*to -= left;
- if (get_free_segments(sbi) <= SM_I(sbi)->reserved_segments + 1)
+ if (SM_I(sbi)->free_segments <= SM_I(sbi)->reserved_segments + 1)
not_enough = 1;
while (*to >= SM_I(sbi)->main_blkaddr && *to < end_blkaddr) {
diff --git a/fsck/segment.c b/fsck/segment.c
index 0307bdd..1cb7d02 100644
--- a/fsck/segment.c
+++ b/fsck/segment.c
@@ -77,6 +77,8 @@ int reserve_new_block(struct f2fs_sb_info *sbi, block_t *to,
se = get_seg_entry(sbi, GET_SEGNO(sbi, blkaddr));
offset = OFFSET_IN_SEG(sbi, blkaddr);
se->type = type;
+ if (se->valid_blocks == 0)
+ SM_I(sbi)->free_segments--;
se->valid_blocks++;
f2fs_set_bit(offset, (char *)se->cur_valid_map);
if (need_fsync_data_record(sbi)) {
--
2.25.1
_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
prev parent reply other threads:[~2023-09-28 9:13 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-28 9:20 [f2fs-dev] [PATCH v2 0/3] f2fs-tools: cache free segments count to improve perfmance Wu Bo via Linux-f2fs-devel
2023-09-28 9:20 ` [f2fs-dev] [PATCH v2 1/3] f2fs-tools: use 'IS_CUR_SEGNO()' to check if it is current segment Wu Bo via Linux-f2fs-devel
2023-10-07 1:39 ` Chao Yu
2023-09-28 9:20 ` [f2fs-dev] [PATCH v2 2/3] f2fs-tools: skip not matched segment when finding free block Wu Bo via Linux-f2fs-devel
2023-09-28 9:20 ` Wu Bo via Linux-f2fs-devel [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=188b05d1764d1ba48f9c704ff28df853f3094f67.1695890686.git.bo.wu@vivo.com \
--to=linux-f2fs-devel@lists.sourceforge.net \
--cc=bo.wu@vivo.com \
--cc=chao@kernel.org \
--cc=jaegeuk@kernel.org \
--cc=wubo.oduw@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).