From: Michael Wang <wangyun@linux.vnet.ibm.com>
To: Peter Zijlstra <a.p.zijlstra@chello.nl>, Ingo Molnar <mingo@elte.hu>
Cc: LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v4] sched: Avoid unnecessary work in reweight_entity
Date: Thu, 16 Feb 2012 22:32:46 +0800 [thread overview]
Message-ID: <4F3D138E.9010904@linux.vnet.ibm.com> (raw)
In-Reply-To: <4F3D0F41.1010205@linux.vnet.ibm.com>
Hi, All
I have correct the issue about invoke update_curr too many times, it caused by
the context in dequeu_entity, the se will be cfs_rq->curr and se->on_rq is 0.
I suppose this patch can avoid some work like:
1. reduce times to check "se->on" from 2 to 1.
2. reduce times to check "parent_entity(se)" from 2 to 1.
3. reduce times to check "entity_is_task(se)" from 2 to 1.
4. reduce times to set "cfs_rq->load.inv_weight" from 4 to 2.
5. reduce times to set "rq_of(cfs_rq)->load.inv_weight" from 2 to 1.
6. no need to use "list_add(&se->group_node, &cfs_rq->tasks);" any more.
7. no need to use "list_del_init(&se->group_node);" any more.
8. no need to do "cfs_rq->nr_running++" and "cfs_rq->nr_running--" any more.
Please tell me if you found in some conditions it will not work or even reduce
the performance, any comments are appreciated :)
Best Regards.
Michael Wang
On 02/16/2012 10:14 PM, Michael Wang wrote:
> From: Michael Wang <wangyun@linux.vnet.ibm.com>
>
> In original code, using account_entity_dequeue and account_entity_enqueue
> as a pair will do some work unnecessary, such like remove node from list
> and add it back again immediately.
>
> And because there are no really enqueue and dequeue happen here, it is not
> suitable to use account_entity_dequeue and account_entity_enqueue here.
>
> This patch combine the work of them in order to save time.
>
> v2:
> Add more description and revise wrong code.
>
> v3:
> Mark add_cfs_task_weight and sub_add_cfs_task_weight as inline.
>
> v4:
> Now we invoke update_curr when really necessary.
>
> Signed-off-by: Michael Wang <wangyun@linux.vnet.ibm.com>
> ---
> kernel/sched/fair.c | 27 +++++++++++++++++++++------
> kernel/sched/sched.h | 8 ++++++++
> 2 files changed, 29 insertions(+), 6 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 7c6414f..fda63ec 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -777,16 +777,28 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se)
> */
>
> #if defined CONFIG_SMP && defined CONFIG_FAIR_GROUP_SCHED
> -static void
> +static inline void
> add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight)
> {
> cfs_rq->task_weight += weight;
> }
> +
> +static inline void
> +sub_add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long sub, unsigned long add)
> +{
> + cfs_rq->task_weight -= sub;
> + cfs_rq->task_weight += add;
> +}
> #else
> static inline void
> add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long weight)
> {
> }
> +
> +static inline void
> +sub_add_cfs_task_weight(struct cfs_rq *cfs_rq, unsigned long sub, unsigned long add)
> +{
> +}
> #endif
>
> static void
> @@ -938,6 +950,7 @@ static inline void update_entity_shares_tick(struct cfs_rq *cfs_rq)
> {
> }
> # endif /* CONFIG_SMP */
> +
> static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
> unsigned long weight)
> {
> @@ -945,13 +958,15 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
> /* commit outstanding execution time */
> if (cfs_rq->curr == se)
> update_curr(cfs_rq);
> - account_entity_dequeue(cfs_rq, se);
> + update_load_sub_add(&cfs_rq->load, se->load.weight, weight);
> + if (!parent_entity(se)) {
> + update_load_sub_add(&rq_of(cfs_rq)->load, se->load.weight, weight);
> + }
> + if (entity_is_task(se)) {
> + sub_add_cfs_task_weight(cfs_rq, se->load.weight, weight);
> + }
> }
> -
> update_load_set(&se->load, weight);
> -
> - if (se->on_rq)
> - account_entity_enqueue(cfs_rq, se);
> }
>
> static void update_cfs_shares(struct cfs_rq *cfs_rq)
> diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
> index 98c0c26..ec4430f 100644
> --- a/kernel/sched/sched.h
> +++ b/kernel/sched/sched.h
> @@ -779,6 +779,14 @@ static inline void update_load_sub(struct load_weight *lw, unsigned long dec)
> lw->inv_weight = 0;
> }
>
> +static inline void update_load_sub_add(struct load_weight *lw, unsigned long dec,
> + unsigned long inc)
> +{
> + lw->weight -= dec;
> + lw->weight += inc;
> + lw->inv_weight = 0;
> +}
> +
> static inline void update_load_set(struct load_weight *lw, unsigned long w)
> {
> lw->weight = w;
prev parent reply other threads:[~2012-02-16 14:33 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-02-06 9:37 [PATCH] sched: Avoid unnecessary work in reweight_entity Michael Wang
2012-02-06 10:31 ` Michael Wang
2012-02-07 2:18 ` Michael Wang
2012-02-07 2:36 ` [PATCH v2] " Michael Wang
2012-02-08 2:10 ` [PATCH v3] sched: " Michael Wang
2012-02-16 14:14 ` [PATCH v4] " Michael Wang
2012-02-16 14:29 ` Peter Zijlstra
2012-02-17 2:28 ` Michael Wang
2012-02-17 6:03 ` [PATCH v5] " Michael Wang
2012-02-18 1:43 ` Michael Wang
2012-02-20 13:08 ` Peter Zijlstra
2012-02-23 10:40 ` Michael Wang
2012-02-24 2:08 ` Michael Wang
2012-02-25 13:56 ` Michael Wang
2012-02-27 4:12 ` Srivatsa Vaddagiri
2012-02-27 5:07 ` Michael Wang
2012-02-27 5:10 ` Srivatsa Vaddagiri
2012-02-27 6:21 ` Michael Wang
2012-02-27 6:44 ` Srivatsa Vaddagiri
2012-02-16 14:32 ` Michael Wang [this message]
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=4F3D138E.9010904@linux.vnet.ibm.com \
--to=wangyun@linux.vnet.ibm.com \
--cc=a.p.zijlstra@chello.nl \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
/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).