From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Andrew Morton <akpm@linux-foundation.org>
Cc: Jens Axboe <jens.axboe@oracle.com>,
linux-kernel@vger.kernel.org, knikanth@novell.com
Subject: Re: [PATCH 4/6] block: cfq: make the io contect sharing lockless
Date: Thu, 24 Jan 2008 08:42:52 -0800 [thread overview]
Message-ID: <20080124164252.GB14706@linux.vnet.ibm.com> (raw)
In-Reply-To: <20080123140819.d53bc976.akpm@linux-foundation.org>
On Wed, Jan 23, 2008 at 02:08:19PM -0800, Andrew Morton wrote:
> > On Tue, 22 Jan 2008 10:49:19 +0100 Jens Axboe <jens.axboe@oracle.com> wrote:
> > The io context sharing introduced a per-ioc spinlock, that would protect
> > the cfq io context lookup. That is a regression from the original, since
> > we never needed any locking there because the ioc/cic were process private.
> >
> > The cic lookup is changed from an rbtree construct to a radix tree, which
> > we can then use RCU to make the reader side lockless. That is the performance
> > critical path, modifying the radix tree is only done on process creation
> > (when that process first does IO, actually) and on process exit (if that
> > process has done IO).
>
> Perhaps Paul would review the rcu usage here sometime?
I will take a look -- long plane trip about 36 hours from now. ;-)
Thanx, Paul
> > +/*
> > + * Add cic into ioc, using cfqd as the search key. This enables us to lookup
> > + * the process specific cfq io context when entered from the block layer.
> > + * Also adds the cic to a per-cfqd list, used when this queue is removed.
> > + */
> > +static inline int
>
> There's a lot of pointless inlining in there.
>
> > +++ b/block/ll_rw_blk.c
> > @@ -3831,6 +3831,16 @@ int __init blk_dev_init(void)
> > return 0;
> > }
> >
> > +static void cfq_dtor(struct io_context *ioc)
> > +{
> > + struct cfq_io_context *cic[1];
> > + int r;
> > +
> > + r = radix_tree_gang_lookup(&ioc->radix_root, (void **) cic, 0, 1);
> > + if (r > 0)
> > + cic[0]->dtor(ioc);
> > +}
>
> Some comments here might help others who are wondering why we can't just
> use radix_tree_lookup().
>
> > +static void cfq_exit(struct io_context *ioc)
> > +{
> > + struct cfq_io_context *cic[1];
> > + int r;
> > +
> > + rcu_read_lock();
> > + r = radix_tree_gang_lookup(&ioc->radix_root, (void **) cic, 0, 1);
> > + rcu_read_unlock();
> > +
> > + if (r > 0)
> > + cic[0]->exit(ioc);
> > +}
>
> ditto.
>
> > /* Called by the exitting task */
> > void exit_io_context(void)
> > {
> > struct io_context *ioc;
> > - struct cfq_io_context *cic;
> >
> > task_lock(current);
> > ioc = current->io_context;
> > @@ -3876,11 +3891,7 @@ void exit_io_context(void)
> > if (atomic_dec_and_test(&ioc->nr_tasks)) {
> > if (ioc->aic && ioc->aic->exit)
> > ioc->aic->exit(ioc->aic);
> > - if (ioc->cic_root.rb_node != NULL) {
> > - cic = rb_entry(rb_first(&ioc->cic_root),
> > - struct cfq_io_context, rb_node);
> > - cic->exit(ioc);
> > - }
> > + cfq_exit(ioc);
> >
> > put_io_context(ioc);
> > }
> > @@ -3900,7 +3911,7 @@ struct io_context *alloc_io_context(gfp_t gfp_flags, int node)
> > ret->last_waited = jiffies; /* doesn't matter... */
> > ret->nr_batch_requests = 0; /* because this is 0 */
> > ret->aic = NULL;
> > - ret->cic_root.rb_node = NULL;
> > + INIT_RADIX_TREE(&ret->radix_root, GFP_ATOMIC | __GFP_HIGH);
>
> Did this need to be atomic?
>
>
next prev parent reply other threads:[~2008-01-24 16:43 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-22 9:49 [PATCH 0/6] IO context sharing Jens Axboe
2008-01-22 9:49 ` [PATCH 1/6] ioprio: move io priority from task_struct to io_context Jens Axboe
2008-01-23 22:07 ` Andrew Morton
2008-01-24 7:30 ` Jens Axboe
2008-01-22 9:49 ` [PATCH 2/6] io context sharing: preliminary support Jens Axboe
2008-01-23 22:08 ` Andrew Morton
2008-01-24 7:36 ` Jens Axboe
2008-01-22 9:49 ` [PATCH 3/6] io_context sharing - cfq changes Jens Axboe
2008-01-22 9:49 ` [PATCH 4/6] block: cfq: make the io contect sharing lockless Jens Axboe
2008-01-23 22:08 ` Andrew Morton
2008-01-24 7:36 ` Jens Axboe
2008-01-24 16:42 ` Paul E. McKenney [this message]
2008-01-22 9:49 ` [PATCH 5/6] io_context sharing - anticipatory changes Jens Axboe
2008-01-22 9:49 ` [PATCH 6/6] kernel: add CLONE_IO to specifically request sharing of IO contexts Jens Axboe
2008-01-22 14:49 ` [PATCH 0/6] IO context sharing Peter Zijlstra
2008-01-22 18:21 ` Jens Axboe
2008-01-23 3:50 ` David Chinner
2008-01-23 8:44 ` Andi Kleen
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=20080124164252.GB14706@linux.vnet.ibm.com \
--to=paulmck@linux.vnet.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=jens.axboe@oracle.com \
--cc=knikanth@novell.com \
--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).