From: NeilBrown <neilb@suse.de>
To: Shaohua Li <shli@kernel.org>
Cc: linux-raid@vger.kernel.org, axboe@kernel.dk,
dan.j.williams@intel.com, shli@fusionio.com
Subject: Re: [patch 06/10 v3] raid5: reduce chance release_stripe() taking device_lock
Date: Mon, 2 Jul 2012 10:57:31 +1000 [thread overview]
Message-ID: <20120702105731.2be963e6@notabene.brown> (raw)
In-Reply-To: <20120625072641.504876596@kernel.org>
[-- Attachment #1: Type: text/plain, Size: 4513 bytes --]
On Mon, 25 Jun 2012 15:24:53 +0800 Shaohua Li <shli@kernel.org> wrote:
> release_stripe() is a place conf->device_lock is heavily contended. We take the
> lock even stripe count isn't 1, which isn't required.
>
> Signed-off-by: Shaohua Li <shli@fusionio.com>
> ---
> drivers/md/raid5.c | 73 +++++++++++++++++++++++++++++------------------------
> 1 file changed, 40 insertions(+), 33 deletions(-)
>
> Index: linux/drivers/md/raid5.c
> ===================================================================
> --- linux.orig/drivers/md/raid5.c 2012-06-25 14:37:21.000000000 +0800
> +++ linux/drivers/md/raid5.c 2012-06-25 14:38:13.899130571 +0800
> @@ -196,49 +196,56 @@ static int stripe_operations_active(stru
> test_bit(STRIPE_COMPUTE_RUN, &sh->state);
> }
>
> -static void __release_stripe(struct r5conf *conf, struct stripe_head *sh)
> +static void handle_release_stripe(struct r5conf *conf, struct stripe_head *sh)
> {
> - if (atomic_dec_and_test(&sh->count)) {
> - BUG_ON(!list_empty(&sh->lru));
> - BUG_ON(atomic_read(&conf->active_stripes)==0);
> - if (test_bit(STRIPE_HANDLE, &sh->state)) {
> - if (test_bit(STRIPE_DELAYED, &sh->state) &&
> - !test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
> - list_add_tail(&sh->lru, &conf->delayed_list);
> - else if (test_bit(STRIPE_BIT_DELAY, &sh->state) &&
> - sh->bm_seq - conf->seq_write > 0)
> - list_add_tail(&sh->lru, &conf->bitmap_list);
> - else {
> - clear_bit(STRIPE_DELAYED, &sh->state);
> - clear_bit(STRIPE_BIT_DELAY, &sh->state);
> - list_add_tail(&sh->lru, &conf->handle_list);
> - }
> - md_wakeup_thread(conf->mddev->thread);
> - } else {
> - BUG_ON(stripe_operations_active(sh));
> - if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
> - if (atomic_dec_return(&conf->preread_active_stripes)
> - < IO_THRESHOLD)
> - md_wakeup_thread(conf->mddev->thread);
> - atomic_dec(&conf->active_stripes);
> - if (!test_bit(STRIPE_EXPANDING, &sh->state)) {
> - list_add_tail(&sh->lru, &conf->inactive_list);
> - wake_up(&conf->wait_for_stripe);
> - if (conf->retry_read_aligned)
> - md_wakeup_thread(conf->mddev->thread);
> - }
> + BUG_ON(!list_empty(&sh->lru));
> + BUG_ON(atomic_read(&conf->active_stripes)==0);
> + if (test_bit(STRIPE_HANDLE, &sh->state)) {
> + if (test_bit(STRIPE_DELAYED, &sh->state) &&
> + !test_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
> + list_add_tail(&sh->lru, &conf->delayed_list);
> + else if (test_bit(STRIPE_BIT_DELAY, &sh->state) &&
> + sh->bm_seq - conf->seq_write > 0)
> + list_add_tail(&sh->lru, &conf->bitmap_list);
> + else {
> + clear_bit(STRIPE_DELAYED, &sh->state);
> + clear_bit(STRIPE_BIT_DELAY, &sh->state);
> + list_add_tail(&sh->lru, &conf->handle_list);
> + }
> + md_wakeup_thread(conf->mddev->thread);
> + } else {
> + BUG_ON(stripe_operations_active(sh));
> + if (test_and_clear_bit(STRIPE_PREREAD_ACTIVE, &sh->state))
> + if (atomic_dec_return(&conf->preread_active_stripes)
> + < IO_THRESHOLD)
> + md_wakeup_thread(conf->mddev->thread);
> + atomic_dec(&conf->active_stripes);
> + if (!test_bit(STRIPE_EXPANDING, &sh->state)) {
> + list_add_tail(&sh->lru, &conf->inactive_list);
> + wake_up(&conf->wait_for_stripe);
> + if (conf->retry_read_aligned)
> + md_wakeup_thread(conf->mddev->thread);
> }
> }
> }
>
> +static void __release_stripe(struct r5conf *conf, struct stripe_head *sh)
> +{
> + if (atomic_dec_and_test(&sh->count))
> + handle_release_stripe(conf, sh);
> +}
> +
> static void release_stripe(struct stripe_head *sh)
> {
> struct r5conf *conf = sh->raid_conf;
> unsigned long flags;
>
> - spin_lock_irqsave(&conf->device_lock, flags);
> - __release_stripe(conf, sh);
> - spin_unlock_irqrestore(&conf->device_lock, flags);
> + local_irq_save(flags);
> + if (atomic_dec_and_lock(&sh->count, &conf->device_lock)) {
> + handle_release_stripe(conf, sh);
> + spin_unlock(&conf->device_lock);
> + }
> + local_irq_restore(flags);
> }
>
> static inline void remove_hash(struct stripe_head *sh)
Thanks. I've applied this patch and it should appear in my -next shortly.
I renamed "handle_release_stripe" to "do_release_stripe", partly because I
think that is more consistent with practice in Linux, but mostly because
"handle" means something else inside raid5.c and I don't want to encourage
confusion.
Thanks,
NeilBrown
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 828 bytes --]
next prev parent reply other threads:[~2012-07-02 0:57 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-25 7:24 [patch 00/10 v3] raid5: improve write performance for fast storage Shaohua Li
2012-06-25 7:24 ` [patch 01/10 v3] raid5: use wake_up_all for overlap waking Shaohua Li
2012-06-28 7:26 ` NeilBrown
2012-06-28 8:53 ` Shaohua Li
2012-06-25 7:24 ` [patch 02/10 v3] raid5: delayed stripe fix Shaohua Li
2012-07-02 0:46 ` NeilBrown
2012-07-02 0:49 ` Shaohua Li
2012-07-02 0:55 ` NeilBrown
2012-06-25 7:24 ` [patch 03/10 v3] raid5: add a per-stripe lock Shaohua Li
2012-07-02 0:50 ` NeilBrown
2012-07-02 3:16 ` Shaohua Li
2012-07-02 7:39 ` NeilBrown
2012-07-03 1:27 ` Shaohua Li
2012-07-03 12:16 ` majianpeng
2012-07-03 23:56 ` NeilBrown
2012-07-04 1:09 ` majianpeng
2012-06-25 7:24 ` [patch 04/10 v3] raid5: lockless access raid5 overrided bi_phys_segments Shaohua Li
2012-06-25 7:24 ` [patch 05/10 v3] raid5: remove some device_lock locking places Shaohua Li
2012-06-25 7:24 ` [patch 06/10 v3] raid5: reduce chance release_stripe() taking device_lock Shaohua Li
2012-07-02 0:57 ` NeilBrown [this message]
2012-06-25 7:24 ` [patch 07/10 v3] md: personality can provide unplug private data Shaohua Li
2012-07-02 1:06 ` NeilBrown
2012-06-25 7:24 ` [patch 08/10 v3] raid5: make_request use batch stripe release Shaohua Li
2012-07-02 2:31 ` NeilBrown
2012-07-02 2:59 ` Shaohua Li
2012-07-02 5:07 ` NeilBrown
2012-06-25 7:24 ` [patch 09/10 v3] raid5: raid5d handle stripe in batch way Shaohua Li
2012-07-02 2:32 ` NeilBrown
2012-06-25 7:24 ` [patch 10/10 v3] raid5: create multiple threads to handle stripes Shaohua Li
2012-07-02 2:39 ` NeilBrown
2012-07-02 20:03 ` Dan Williams
2012-07-03 8:04 ` Shaohua Li
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=20120702105731.2be963e6@notabene.brown \
--to=neilb@suse.de \
--cc=axboe@kernel.dk \
--cc=dan.j.williams@intel.com \
--cc=linux-raid@vger.kernel.org \
--cc=shli@fusionio.com \
--cc=shli@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).