| Message ID | 549B7C11.8030704@huawei.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Thu, 25 Dec 2014 10:53:05 +0800 alex chen <alex.chen@huawei.com> wrote: > Add a mount option to support JBD2 feature: > JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT. When this feature is opened, > journal commit block can be written to disk without waiting for > descriptor blocks, which can improve journal commit performance. This > option will enable 'journal_checksum' internally. > > Using the fs_mark benchmark, using journal_async_commit shows a 50% > improvement, the files per second go up from 215.2 to 317.5. > > test script: > fs_mark -d /mnt/ocfs2/ -s 10240 -n 1000 > > default: > FSUse% Count Size Files/sec App Overhead > 0 1000 10240 215.2 17878 > > with journal_async_commit option: > FSUse% Count Size Files/sec App Overhead > 0 1000 10240 317.5 17881 For some reason this patch is a bit mangled and I had to apply the first hunk by hand. > fs/ocfs2/ocfs2.h | 2 ++ > fs/ocfs2/super.c | 17 +++++++++++++++++ Documentation/filesystems/ocfs2.txt needs an update.
On 2015/1/7 7:58, Andrew Morton wrote: > On Thu, 25 Dec 2014 10:53:05 +0800 alex chen <alex.chen@huawei.com> wrote: > >> Add a mount option to support JBD2 feature: >> JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT. When this feature is opened, >> journal commit block can be written to disk without waiting for >> descriptor blocks, which can improve journal commit performance. This >> option will enable 'journal_checksum' internally. >> >> Using the fs_mark benchmark, using journal_async_commit shows a 50% >> improvement, the files per second go up from 215.2 to 317.5. >> >> test script: >> fs_mark -d /mnt/ocfs2/ -s 10240 -n 1000 >> >> default: >> FSUse% Count Size Files/sec App Overhead >> 0 1000 10240 215.2 17878 >> >> with journal_async_commit option: >> FSUse% Count Size Files/sec App Overhead >> 0 1000 10240 317.5 17881 > > For some reason this patch is a bit mangled and I had to apply the > first hunk by hand. > Sorry for my mistake? I will send the new patch based on linux-next kernel. Thanks. Alex >> fs/ocfs2/ocfs2.h | 2 ++ >> fs/ocfs2/super.c | 17 +++++++++++++++++ > > Documentation/filesystems/ocfs2.txt needs an update. OK > > > > . >
diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index 3e6dc76..0d53bc4 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -278,6 +278,8 @@ enum ocfs2_mount_options writes */ OCFS2_MOUNT_HB_NONE = 1 << 13, /* No heartbeat */ OCFS2_MOUNT_HB_GLOBAL = 1 << 14, /* Global heartbeat */ + OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT = 1 << 15, /* Journal async commit */ }; #define OCFS2_OSB_SOFT_RO 0x0001 diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index da33a14..ef5d11a 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -185,6 +185,7 @@ enum { Opt_coherency_full, Opt_resv_level, Opt_dir_resv_level, + Opt_journal_async_commit, Opt_err, }; @@ -216,6 +217,7 @@ static const match_table_t tokens = { {Opt_coherency_full, "coherency=full"}, {Opt_resv_level, "resv_level=%u"}, {Opt_dir_resv_level, "dir_resv_level=%u"}, + {Opt_journal_async_commit, "journal_async_commit"}, {Opt_err, NULL} }; @@ -1493,6 +1495,9 @@ static int ocfs2_parse_options(struct super_block *sb, option < OCFS2_MAX_RESV_LEVEL) mopt->dir_resv_level = option; break; + case Opt_journal_async_commit: + mopt->mount_opt |= OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT; + break; default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -1599,6 +1604,9 @@ static int ocfs2_show_options(struct seq_file *s, struct dentry *root) if (osb->osb_dir_resv_level != osb->osb_resv_level) seq_printf(s, ",dir_resv_level=%d", osb->osb_resv_level); + if (opts & OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT) + seq_printf(s, ",journal_async_commit"); + return 0; } @@ -2468,6 +2476,15 @@ static int ocfs2_check_volume(struct ocfs2_super *osb) goto finally; } + if (osb->s_mount_opt & OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT) + jbd2_journal_set_features(osb->journal->j_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + else + jbd2_journal_clear_features(osb->journal->j_journal, + JBD2_FEATURE_COMPAT_CHECKSUM, 0, + JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT); + if (dirty) { /* recover my local alloc if we didn't unmount cleanly. */ status = ocfs2_begin_local_alloc_recovery(osb,
Add a mount option to support JBD2 feature: JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT. When this feature is opened, journal commit block can be written to disk without waiting for descriptor blocks, which can improve journal commit performance. This option will enable 'journal_checksum' internally. Using the fs_mark benchmark, using journal_async_commit shows a 50% improvement, the files per second go up from 215.2 to 317.5. test script: fs_mark -d /mnt/ocfs2/ -s 10240 -n 1000 default: FSUse% Count Size Files/sec App Overhead 0 1000 10240 215.2 17878 with journal_async_commit option: FSUse% Count Size Files/sec App Overhead 0 1000 10240 317.5 17881 Signed-off-by: Alex Chen <alex.chen@huawei.com> Signed-off-by: Weiwei Wang <wangww631@huawei.comm> Reviewed-by: Joseph Qi <joseph.qi@huawei.com> --- fs/ocfs2/ocfs2.h | 2 ++ fs/ocfs2/super.c | 17 +++++++++++++++++ 2 files changed, 19 insertions(+)