From: Jan Kara <jack@suse.cz>
To: Dmitry Monakhov <dmonakhov@openvz.org>
Cc: linux-fsdevel@vger.kernel.org, jack@suse.cz, hch@infradead.org,
Dmitry Monakhov <dmonakhov@gmail.com>
Subject: Re: [PATCH 08/11] quota: make per-sb dq_data_lock
Date: Wed, 6 Oct 2010 13:01:22 +0200 [thread overview]
Message-ID: <20101006110122.GI3676@quack.suse.cz> (raw)
In-Reply-To: <1286302827-31043-9-git-send-email-dmonakhov@gmail.com>
On Tue 05-10-10 22:20:24, Dmitry Monakhov wrote:
> Currently dq_data_lock is global, which is bad for scalability.
> In fact different super_blocks has no shared quota's data.
> So we may simply convert global lock to per-sb lock.
>
> Signed-off-by: Dmitry Monakhov <dmonakhov@gmail.com>
...
> @@ -657,9 +657,9 @@ static int ocfs2_write_dquot(struct dquot *dquot)
> mlog_errno(status);
> goto out;
> }
> - mutex_lock(&dq_opt(dqopt)->dqio_mutex);
> + mutex_lock(&dq_opt(dquot)->dqio_mutex);
> status = ocfs2_local_write_dquot(dquot);
> - mutex_unlock(&dq_opt(dqopt)->dqio_mutex);
> + mutex_unlock(&dq_opt(dquot)->dqio_mutex);
> ocfs2_commit_trans(osb, handle);
I guess this hunk belongs to the first patch.
> diff --git a/fs/ocfs2/quota_local.c b/fs/ocfs2/quota_local.c
> index dc78764..c078799 100644
> --- a/fs/ocfs2/quota_local.c
> +++ b/fs/ocfs2/quota_local.c
> @@ -288,14 +288,15 @@ static void olq_update_info(struct buffer_head *bh, void *private)
> struct mem_dqinfo *info = private;
> struct ocfs2_mem_dqinfo *oinfo = info->dqi_priv;
> struct ocfs2_local_disk_dqinfo *ldinfo;
> + struct super_block *sb = oinfo->dqi_gqinode->i_sb;
You can directly store struct quota_info *, here.
>
> ldinfo = (struct ocfs2_local_disk_dqinfo *)(bh->b_data +
> OCFS2_LOCAL_INFO_OFF);
> - spin_lock(&dq_data_lock);
> + spin_lock(&sb_dqopt(sb)->dq_data_lock);
> ldinfo->dqi_flags = cpu_to_le32(info->dqi_flags & DQF_MASK);
> ldinfo->dqi_chunks = cpu_to_le32(oinfo->dqi_chunks);
> ldinfo->dqi_blocks = cpu_to_le32(oinfo->dqi_blocks);
> - spin_unlock(&dq_data_lock);
> + spin_unlock(&sb_dqopt(sb)->dq_data_lock);
> }
Honza
--
Jan Kara <jack@suse.cz>
SUSE Labs, CR
next prev parent reply other threads:[~2010-10-06 11:02 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-10-05 18:20 (unknown), Dmitry Monakhov
2010-10-05 18:20 ` [PATCH 01/11] quota: add wrapper function Dmitry Monakhov
2010-10-06 8:56 ` Christoph Hellwig
2010-10-06 10:01 ` Jan Kara
2010-10-05 18:20 ` [PATCH 02/11] quota: Convert dq_state_lock to per-sb dq_state_lock Dmitry Monakhov
2010-10-06 10:04 ` Jan Kara
2010-10-05 18:20 ` [PATCH 03/11] quota: add quota format lock Dmitry Monakhov
2010-10-06 10:05 ` Jan Kara
2010-10-05 18:20 ` [PATCH 04/11] quota: make dquot lists per-sb Dmitry Monakhov
2010-10-06 8:57 ` Christoph Hellwig
2010-10-06 9:39 ` Dmitry
2010-10-06 10:22 ` Jan Kara
2010-10-06 10:40 ` Dmitry
2010-10-06 10:54 ` Jan Kara
2010-10-05 18:20 ` [PATCH 05/11] quota: make per-sb hash array Dmitry Monakhov
2010-10-06 10:38 ` Jan Kara
2010-10-05 18:20 ` [PATCH 06/11] quota: remove global dq_list_lock Dmitry Monakhov
2010-10-05 18:20 ` [PATCH 07/11] quota: rename dq_lock Dmitry Monakhov
2010-10-05 18:20 ` [PATCH 08/11] quota: make per-sb dq_data_lock Dmitry Monakhov
2010-10-06 11:01 ` Jan Kara [this message]
2010-10-05 18:20 ` [PATCH 09/11] quota: protect dquot mem info with objects's lock Dmitry Monakhov
2010-10-06 12:37 ` Jan Kara
2010-10-06 13:17 ` Dmitry
2010-10-06 13:41 ` Jan Kara
2010-10-06 14:19 ` Dmitry
2010-10-06 13:30 ` Jan Kara
2010-10-06 13:41 ` Dmitry
2010-10-05 18:20 ` [PATCH 10/11] quota: drop dq_data_lock where possible Dmitry Monakhov
2010-10-05 18:20 ` [PATCH 11/11] quota: relax dq_data_lock dq_lock locking consistency Dmitry Monakhov
2010-10-06 11:56 ` Jan Kara
2010-10-06 7:08 ` [PATCH 0/11] RFC quota scalability V1 Dmitry
2010-10-06 9:44 ` Jan Kara
2010-10-06 10:15 ` Dmitry
2010-10-06 10:47 ` Jan Kara
2010-10-10 3:50 ` Brad Boyer
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=20101006110122.GI3676@quack.suse.cz \
--to=jack@suse.cz \
--cc=dmonakhov@gmail.com \
--cc=dmonakhov@openvz.org \
--cc=hch@infradead.org \
--cc=linux-fsdevel@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).