From: Jaegeuk Kim <jaegeuk@kernel.org>
To: linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org,
linux-f2fs-devel@lists.sourceforge.net
Cc: Jaegeuk Kim <jaegeuk@kernel.org>
Subject: [PATCH 2/5] f2fs: write small sized IO to hot log
Date: Sat, 25 Mar 2017 00:59:30 -0700 [thread overview]
Message-ID: <20170325075933.21072-2-jaegeuk@kernel.org> (raw)
In-Reply-To: <20170325075933.21072-1-jaegeuk@kernel.org>
It would better split small and large IOs separately in order to get more
consecutive big writes.
The default threshold is set to 64KB, but configurable by sysfs/min_hot_blocks.
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
---
fs/f2fs/data.c | 5 +++++
fs/f2fs/f2fs.h | 2 ++
fs/f2fs/segment.c | 13 ++++++-------
fs/f2fs/segment.h | 1 +
fs/f2fs/super.c | 2 ++
5 files changed, 16 insertions(+), 7 deletions(-)
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 090413236b27..7275697bc940 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -1511,6 +1511,11 @@ static int f2fs_write_cache_pages(struct address_space *mapping,
pagevec_init(&pvec, 0);
+ if (get_dirty_pages(mapping->host) <= SM_I(F2FS_M_SB(mapping))->min_hot_blocks)
+ set_inode_flag(mapping->host, FI_HOT_DATA);
+ else
+ clear_inode_flag(mapping->host, FI_HOT_DATA);
+
if (wbc->range_cyclic) {
writeback_index = mapping->writeback_index; /* prev offset */
index = writeback_index;
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 6fbdcac01d9a..19e28127a725 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -674,6 +674,7 @@ struct f2fs_sm_info {
unsigned int ipu_policy; /* in-place-update policy */
unsigned int min_ipu_util; /* in-place-update threshold */
unsigned int min_fsync_blocks; /* threshold for fsync */
+ unsigned int min_hot_blocks; /* threshold for hot block allocation */
/* for flush command control */
struct flush_cmd_control *fcc_info;
@@ -1713,6 +1714,7 @@ enum {
FI_DO_DEFRAG, /* indicate defragment is running */
FI_DIRTY_FILE, /* indicate regular/symlink has dirty pages */
FI_NO_PREALLOC, /* indicate skipped preallocated blocks */
+ FI_HOT_DATA, /* indicate file is hot */
};
static inline void __mark_inode_dirty_flag(struct inode *inode,
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index c5a5258f71c5..0cba28f95bb8 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -1797,18 +1797,16 @@ static int __get_segment_type_6(struct page *page, enum page_type p_type)
if (p_type == DATA) {
struct inode *inode = page->mapping->host;
- if (S_ISDIR(inode->i_mode))
- return CURSEG_HOT_DATA;
- else if (is_cold_data(page) || file_is_cold(inode))
+ if (is_cold_data(page) || file_is_cold(inode))
return CURSEG_COLD_DATA;
- else
- return CURSEG_WARM_DATA;
+ if (is_inode_flag_set(inode, FI_HOT_DATA))
+ return CURSEG_HOT_DATA;
+ return CURSEG_WARM_DATA;
} else {
if (IS_DNODE(page))
return is_cold_node(page) ? CURSEG_WARM_NODE :
CURSEG_HOT_NODE;
- else
- return CURSEG_COLD_NODE;
+ return CURSEG_COLD_NODE;
}
}
@@ -2915,6 +2913,7 @@ int 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_hot_blocks = DEF_MIN_HOT_BLOCKS;
sm_info->trim_sections = DEF_BATCHED_TRIM_SECTIONS;
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index 5e8ad4280a50..4d0dd9f7f4ed 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -540,6 +540,7 @@ static inline int utilization(struct f2fs_sb_info *sbi)
*/
#define DEF_MIN_IPU_UTIL 70
#define DEF_MIN_FSYNC_BLOCKS 8
+#define DEF_MIN_HOT_BLOCKS 16
enum {
F2FS_IPU_FORCE,
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 49434f951ace..c4fa1ace8e55 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -294,6 +294,7 @@ F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, ipu_policy, ipu_policy);
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_ipu_util, min_ipu_util);
F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_hot_blocks, min_hot_blocks);
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, dirty_nats_ratio, dirty_nats_ratio);
@@ -319,6 +320,7 @@ static struct attribute *f2fs_attrs[] = {
ATTR_LIST(ipu_policy),
ATTR_LIST(min_ipu_util),
ATTR_LIST(min_fsync_blocks),
+ ATTR_LIST(min_hot_blocks),
ATTR_LIST(max_victim_search),
ATTR_LIST(dir_level),
ATTR_LIST(ram_thresh),
--
2.11.0
next prev parent reply other threads:[~2017-03-25 7:59 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-25 7:59 [PATCH 1/5] f2fs: relax node version check for victim data in gc Jaegeuk Kim
2017-03-25 7:59 ` Jaegeuk Kim [this message]
2017-03-25 7:59 ` [PATCH 3/5] f2fs: allocate node and hot data in the beginning of partition Jaegeuk Kim
2017-03-25 7:59 ` [PATCH 4/5] f2fs: start SSR much eariler to avoid FG_GC Jaegeuk Kim
2017-03-25 7:59 ` [PATCH 5/5] f2fs: fix wrong max cost initialization Jaegeuk Kim
2017-03-25 9:05 ` [f2fs-dev] [PATCH 1/5] f2fs: relax node version check for victim data in gc Chao Yu
2017-03-25 21:27 ` Jaegeuk Kim
2017-03-27 8:18 ` Chao Yu
2017-03-29 7:04 ` heyunlei
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=20170325075933.21072-2-jaegeuk@kernel.org \
--to=jaegeuk@kernel.org \
--cc=linux-f2fs-devel@lists.sourceforge.net \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/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).