From: Mark Fasheh <mfasheh@suse.com>
To: linux-kernel@vger.kernel.org
Cc: ocfs2-devel@oss.oracle.com, Joel Becker <joel.becker@oracle.com>,
Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>,
Mark Fasheh <mfasheh@suse.com>
Subject: [PATCH 31/32] ocfs2: Pagecache usage optimization on ocfs2
Date: Tue, 17 Mar 2009 13:01:21 -0700 [thread overview]
Message-ID: <1237320082-12921-32-git-send-email-mfasheh@suse.com> (raw)
In-Reply-To: <1237320082-12921-1-git-send-email-mfasheh@suse.com>
From: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
A page can have multiple buffers and even if a page is not uptodate, some buffers
can be uptodate on pagesize != blocksize environment.
This aops checks that all buffers which correspond to a part of a file
that we want to read are uptodate. If so, we do not have to issue actual
read IO to HDD even if a page is not uptodate because the portion we
want to read are uptodate.
"block_is_partially_uptodate" function is already used by ext2/3/4.
With the following patch random read/write mixed workloads or random read after
random write workloads can be optimized and we can get performance improvement.
Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
---
fs/ocfs2/aops.c | 23 ++++++++++++-----------
1 files changed, 12 insertions(+), 11 deletions(-)
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 8e1709a..b2c52b3 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -1956,15 +1956,16 @@ static int ocfs2_write_end(struct file *file, struct address_space *mapping,
}
const struct address_space_operations ocfs2_aops = {
- .readpage = ocfs2_readpage,
- .readpages = ocfs2_readpages,
- .writepage = ocfs2_writepage,
- .write_begin = ocfs2_write_begin,
- .write_end = ocfs2_write_end,
- .bmap = ocfs2_bmap,
- .sync_page = block_sync_page,
- .direct_IO = ocfs2_direct_IO,
- .invalidatepage = ocfs2_invalidatepage,
- .releasepage = ocfs2_releasepage,
- .migratepage = buffer_migrate_page,
+ .readpage = ocfs2_readpage,
+ .readpages = ocfs2_readpages,
+ .writepage = ocfs2_writepage,
+ .write_begin = ocfs2_write_begin,
+ .write_end = ocfs2_write_end,
+ .bmap = ocfs2_bmap,
+ .sync_page = block_sync_page,
+ .direct_IO = ocfs2_direct_IO,
+ .invalidatepage = ocfs2_invalidatepage,
+ .releasepage = ocfs2_releasepage,
+ .migratepage = buffer_migrate_page,
+ .is_partially_uptodate = block_is_partially_uptodate,
};
--
1.5.6
next prev parent reply other threads:[~2009-03-17 20:46 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-03-17 20:00 [git patches] Ocfs2 updates for 2.6.30 Mark Fasheh
2009-03-17 20:00 ` [PATCH 01/32] ocfs2/hb: Expose the list of heartbeating nodes via debugfs Mark Fasheh
2009-03-17 20:00 ` [PATCH 02/32] ocfs2: Move struct recovery_map to a header file Mark Fasheh
2009-03-17 20:00 ` [PATCH 03/32] ocfs2: Expose the file system state via debugfs Mark Fasheh
2009-03-17 20:00 ` [PATCH 04/32] ocfs2: Remove debugfs file local_alloc_stats Mark Fasheh
2009-03-17 20:00 ` [PATCH 05/32] ocfs2: Introduce dir lookup helper struct Mark Fasheh
2009-03-17 20:00 ` [PATCH 06/32] ocfs2: Add a name indexed b-tree to directory inodes Mark Fasheh
2009-03-17 20:00 ` [PATCH 07/32] ocfs2: Store dir index records inline Mark Fasheh
2009-03-17 20:00 ` [PATCH 08/32] ocfs2: Introduce dir free space list Mark Fasheh
2009-03-17 20:00 ` [PATCH 09/32] ocfs2: Increase max links count Mark Fasheh
2009-03-17 20:01 ` [PATCH 10/32] ocfs2: Add total entry count to dx_root_block Mark Fasheh
2009-03-17 20:01 ` [PATCH 11/32] ocfs2: Enable indexed directories Mark Fasheh
2009-03-17 20:01 ` [PATCH 12/32] ocfs2: re-order ocfs2_empty_dir checks Mark Fasheh
2009-03-17 20:01 ` [PATCH 13/32] ocfs2: fix leaf start calculation in ocfs2_dx_dir_rebalance() Mark Fasheh
2009-03-17 20:01 ` [PATCH 14/32] ocfs2: Optimize inode allocation by remembering last group Mark Fasheh
2009-03-17 20:01 ` [PATCH 15/32] ocfs2: Allocate inode groups from global_bitmap Mark Fasheh
2009-03-17 20:01 ` [PATCH 16/32] ocfs2: Optimize inode group allocation by recording last used group Mark Fasheh
2009-03-17 20:01 ` [PATCH 17/32] ocfs2/dlm: Encapsulate adding and removing of mle from dlm->master_list Mark Fasheh
2009-03-17 20:01 ` [PATCH 18/32] ocfs2/dlm: Clean up struct dlm_lock_name Mark Fasheh
2009-03-17 20:01 ` [PATCH 19/32] ocfs2/dlm: Refactor dlm_clean_master_list() Mark Fasheh
2009-03-17 20:01 ` [PATCH 20/32] ocfs2/dlm: Create and destroy the dlm->master_hash Mark Fasheh
2009-03-17 20:01 ` [PATCH 21/32] ocfs2/dlm: Activate dlm->master_hash for master list entries Mark Fasheh
2009-03-17 20:01 ` [PATCH 22/32] ocfs2/dlm: Indent dlm_cleanup_master_list() Mark Fasheh
2009-03-17 20:01 ` [PATCH 23/32] ocfs2/dlm: Track number of mles Mark Fasheh
2009-03-17 20:01 ` [PATCH 24/32] ocfs2/dlm: Improve lockres counts Mark Fasheh
2009-03-17 20:01 ` [PATCH 25/32] ocfs2/dlm: dlm_set_lockres_owner() and dlm_change_lockres_owner() inlined Mark Fasheh
2009-03-17 20:01 ` [PATCH 26/32] ocfs2/dlm: Show the number of lockres/mles in dlm_state Mark Fasheh
2009-03-17 20:01 ` [PATCH 27/32] ocfs2/dlm: Remove struct dlm_lock_name in struct dlm_master_list_entry Mark Fasheh
2009-03-17 20:01 ` [PATCH 28/32] ocfs2/dlm: Do not purge lockres that is being migrated dlm_purge_lockres() Mark Fasheh
2009-03-17 20:01 ` [PATCH 29/32] ocfs2/dlm: Tweak mle_state output Mark Fasheh
2009-03-17 20:01 ` [PATCH 30/32] ocfs2: fix rare stale inode errors when exporting via nfs Mark Fasheh
2009-03-17 20:01 ` Mark Fasheh [this message]
2009-03-17 20:01 ` [PATCH 32/32] ocfs2: recover orphans in offline slots during recovery and mount Mark Fasheh
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=1237320082-12921-32-git-send-email-mfasheh@suse.com \
--to=mfasheh@suse.com \
--cc=hifumi.hisashi@oss.ntt.co.jp \
--cc=joel.becker@oracle.com \
--cc=linux-kernel@vger.kernel.org \
--cc=ocfs2-devel@oss.oracle.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
all inboxes | Powered by JetHome®