From: Wanpeng Li <kernellwp@gmail.com>
To: "pang.xunlei" <pang.xunlei@linaro.org>, linux-kernel@vger.kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>,
Steven Rostedt <rostedt@goodmis.org>,
Juri Lelli <juri.lelli@gmail.com>,
Wanpeng Li <wanpeng.li@linux.intel.com>
Subject: Re: [PATCH v2 1/4] sched/deadline: Modify cpudl.free_cpus to reflect rd->span
Date: Thu, 20 Nov 2014 17:00:20 +0800 [thread overview]
Message-ID: <546DADA4.4050604@gmail.com> (raw)
In-Reply-To: <1416411982-1204-1-git-send-email-pang.xunlei@linaro.org>
Hi,
On 11/19/14, 11:46 PM, pang.xunlei wrote:
> Currently, cpudl.free_cpus contains all cpus during init(see cpudl_init()),
> so when calling cpudl_find() we have to add rd->span cpumask(cpus_allowed is
> undependable when performing clustered scheduling using the cpuset) to avoid
> selecting the cpu outside current root domain, see find_later_rq().
>
> This patch adds cpudl_set_freecpu() to initialize cpudl.free_cpus when doing
> rq_attach_root(), so we can avoid the extra rd->span operation when calling
> cpudl_find().
>
> Signed-off-by: pang.xunlei <pang.xunlei@linaro.org>
Reviewed-by: Wanpeng Li <wanpeng.li@linux.intel.com>
I think this patch solve one problem I meet in my progress to handle dl
task migration during cpu hotplug.
https://www.mail-archive.com/linux-kernel@vger.kernel.org/msg770579.html
Regards,
Wanpeng Li
> ---
> kernel/sched/core.c | 2 ++
> kernel/sched/cpudeadline.c | 18 ++++++++++++++----
> kernel/sched/cpudeadline.h | 1 +
> kernel/sched/deadline.c | 3 ---
> 4 files changed, 17 insertions(+), 7 deletions(-)
>
> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 240157c..1b417de 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -5584,6 +5584,8 @@ static void rq_attach_root(struct rq *rq, struct root_domain *rd)
> rq->rd = rd;
>
> cpumask_set_cpu(rq->cpu, rd->span);
> + cpudl_set_freecpu(&rd->cpudl, rq->cpu);
> +
> if (cpumask_test_cpu(rq->cpu, cpu_active_mask))
> set_rq_online(rq);
>
> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
> index 539ca3c..c79f0d7 100644
> --- a/kernel/sched/cpudeadline.c
> +++ b/kernel/sched/cpudeadline.c
> @@ -107,7 +107,9 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
> int best_cpu = -1;
> const struct sched_dl_entity *dl_se = &p->dl;
>
> - if (later_mask && cpumask_and(later_mask, later_mask, cp->free_cpus)) {
> + if (later_mask &&
> + cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed) &&
> + cpumask_and(later_mask, later_mask, cpu_active_mask)) {
> best_cpu = cpumask_any(later_mask);
> goto out;
> } else if (cpumask_test_cpu(cpudl_maximum(cp), &p->cpus_allowed) &&
> @@ -186,6 +188,16 @@ out:
> }
>
> /*
> + * cpudl_set_freecpu - Set the cpudl.free_cpus
> + * @cp: the cpudl max-heap context
> + * @cpu: rd attached cpu
> + */
> +void cpudl_set_freecpu(struct cpudl *cp, int cpu)
> +{
> + cpumask_set_cpu(cpu, cp->free_cpus);
> +}
> +
> +/*
> * cpudl_init - initialize the cpudl structure
> * @cp: the cpudl max-heap context
> */
> @@ -203,7 +215,7 @@ int cpudl_init(struct cpudl *cp)
> if (!cp->elements)
> return -ENOMEM;
>
> - if (!alloc_cpumask_var(&cp->free_cpus, GFP_KERNEL)) {
> + if (!zalloc_cpumask_var(&cp->free_cpus, GFP_KERNEL)) {
> kfree(cp->elements);
> return -ENOMEM;
> }
> @@ -211,8 +223,6 @@ int cpudl_init(struct cpudl *cp)
> for_each_possible_cpu(i)
> cp->elements[i].idx = IDX_INVALID;
>
> - cpumask_setall(cp->free_cpus);
> -
> return 0;
> }
>
> diff --git a/kernel/sched/cpudeadline.h b/kernel/sched/cpudeadline.h
> index 020039b..4a10a65 100644
> --- a/kernel/sched/cpudeadline.h
> +++ b/kernel/sched/cpudeadline.h
> @@ -24,6 +24,7 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
> struct cpumask *later_mask);
> void cpudl_set(struct cpudl *cp, int cpu, u64 dl, int is_valid);
> int cpudl_init(struct cpudl *cp);
> +void cpudl_set_freecpu(struct cpudl *cp, int cpu);
> void cpudl_cleanup(struct cpudl *cp);
> #endif /* CONFIG_SMP */
>
> diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
> index 5285332..bd83272 100644
> --- a/kernel/sched/deadline.c
> +++ b/kernel/sched/deadline.c
> @@ -1187,9 +1187,6 @@ static int find_later_rq(struct task_struct *task)
> * We have to consider system topology and task affinity
> * first, then we can look for a suitable cpu.
> */
> - cpumask_copy(later_mask, task_rq(task)->rd->span);
> - cpumask_and(later_mask, later_mask, cpu_active_mask);
> - cpumask_and(later_mask, later_mask, &task->cpus_allowed);
> best_cpu = cpudl_find(&task_rq(task)->rd->cpudl,
> task, later_mask);
> if (best_cpu == -1)
prev parent reply other threads:[~2014-11-20 9:00 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-19 15:46 [PATCH v2 1/4] sched/deadline: Modify cpudl.free_cpus to reflect rd->span pang.xunlei
2014-11-19 15:46 ` [PATCH v2 2/4] sched/deadline: Fix wrong cpudl_find() in check_preempt_equal_dl() pang.xunlei
2014-11-19 15:46 ` [PATCH v2 3/4] sched/deadline: add the "set_flag" argument to cpudl_find() pang.xunlei
2014-11-19 16:24 ` Steven Rostedt
2014-11-20 14:58 ` pang.xunlei
2014-11-20 15:06 ` pang.xunlei
2014-11-19 15:46 ` [PATCH v2 4/4] sched/deadline: change cpudl_find() to return bool instead of best_cpu pang.xunlei
2014-11-19 16:30 ` Steven Rostedt
2014-11-20 9:00 ` Wanpeng Li [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=546DADA4.4050604@gmail.com \
--to=kernellwp@gmail.com \
--cc=juri.lelli@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=pang.xunlei@linaro.org \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=wanpeng.li@linux.intel.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;
as well as URLs for NNTP newsgroup(s).