From: Yicong Yang <yangyicong@huawei.com>
To: Tim Chen <tim.c.chen@linux.intel.com>,
Yicong Yang <yangyicong@hisilicon.com>, <peterz@infradead.org>,
<mingo@redhat.com>, <juri.lelli@redhat.com>,
<vincent.guittot@linaro.org>, <gautham.shenoy@amd.com>,
<linux-kernel@vger.kernel.org>,
<linux-arm-kernel@lists.infradead.org>
Cc: <dietmar.eggemann@arm.com>, <rostedt@goodmis.org>,
<bsegall@google.com>, <bristot@redhat.com>,
<prime.zeng@huawei.com>, <jonathan.cameron@huawei.com>,
<ego@linux.vnet.ibm.com>, <srikar@linux.vnet.ibm.com>,
<linuxarm@huawei.com>, <21cnbao@gmail.com>,
<song.bao.hua@hisilicon.com>, <guodong.xu@linaro.org>
Subject: Re: [PATCH v2 2/2] sched/fair: Scan cluster before scanning LLC in wake-up path
Date: Thu, 27 Jan 2022 11:05:52 +0800 [thread overview]
Message-ID: <4d4099d8-e890-9e56-c395-5f521d98081a@huawei.com> (raw)
In-Reply-To: <62bbdd77f70f7b46a044685668e33fb031812c38.camel@linux.intel.com>
On 2022/1/27 10:36, Tim Chen wrote:
> On Wed, 2022-01-26 at 18:30 -0800, Tim Chen wrote:
>> On Thu, 2022-01-27 at 10:02 +0800, Yicong Yang wrote:
>>> On 2022/1/27 9:14, Tim Chen wrote:
>>>> On Wed, 2022-01-26 at 16:09 +0800, Yicong Yang wrote:
>>>>> From: Barry Song <song.bao.hua@hisilicon.com>
>>>>>
>>>>> For platforms having clusters like Kunpeng920, CPUs within the
>>>>> same
>>>>> cluster have lower latency when synchronizing and accessing
>>>>> shared
>>>>> resources like cache. Thus, this patch tries to find an idle
>>>>> cpu
>>>>> within the cluster of the target CPU before scanning the whole
>>>>> LLC
>>>>> to gain lower latency.
>>>>>
>>>>> Note neither Kunpeng920 nor x86 Jacobsville supports SMT, so
>>>>> this
>>>>> patch doesn't consider SMT for this moment.
>>>>>
>>>>> Testing has been done on Kunpeng920 by pinning tasks to one
>>>>> numa
>>>>> and two numa. On Kunpeng920, Each numa has 8 clusters and each
>>>>> cluster has 4 CPUs.
>>>>>
>>>>> With this patch, We noticed enhancement on tbench within one
>>>>> numa or cross two numa.
>>>>>
>>>>> On numa 0:
>>>>> 5.17-rc1 patched
>>>>> Hmean 1 324.73 ( 0.00%) 378.01 * 16.41%*
>>>>> Hmean 2 645.36 ( 0.00%) 754.63 * 16.93%*
>>>>> Hmean 4 1302.09 ( 0.00%) 1507.54 * 15.78%*
>>>>> Hmean 8 2612.03 ( 0.00%) 2982.57 * 14.19%*
>>>>> Hmean 16 5307.12 ( 0.00%) 5886.66 * 10.92%*
>>>>> Hmean 32 9354.22 ( 0.00%) 9908.13 * 5.92%*
>>>>> Hmean 64 7240.35 ( 0.00%) 7278.78 * 0.53%*
>>>>> Hmean 128 6186.40 ( 0.00%) 6187.85 ( 0.02%)
>>>>>
>>>>> On numa 0-1:
>>>>> 5.17-rc1 patched
>>>>> Hmean 1 320.01 ( 0.00%) 378.44 * 18.26%*
>>>>> Hmean 2 643.85 ( 0.00%) 752.52 * 16.88%*
>>>>> Hmean 4 1287.36 ( 0.00%) 1505.62 * 16.95%*
>>>>> Hmean 8 2564.60 ( 0.00%) 2955.29 * 15.23%*
>>>>> Hmean 16 5195.69 ( 0.00%) 5814.74 * 11.91%*
>>>>> Hmean 32 9769.16 ( 0.00%) 10872.63 * 11.30%*
>>>>> Hmean 64 15952.50 ( 0.00%) 17281.98 * 8.33%*
>>>>> Hmean 128 13113.77 ( 0.00%) 13895.20 * 5.96%*
>>>>> Hmean 256 10997.59 ( 0.00%) 11244.69 * 2.25%*
>>>>> Hmean 512 14623.60 ( 0.00%) 15526.25 * 6.17%*
>>>>>
>>>>> This will also help to improve the MySQL. With MySQL server
>>>>> running on numa 0 and client running on numa 1, both QPS and
>>>>> latency is imporved on read-write case:
>>>>> 5.17-rc1 patched
>>>>> QPS-16threads 143333.2633 145077.4033(+1.22%)
>>>>> QPS-24threads 195085.9367 202719.6133(+3.91%)
>>>>> QPS-32threads 241165.6867 249020.74(+3.26%)
>>>>> QPS-64threads 244586.8433 253387.7567(+3.60%)
>>>>> avg-lat-16threads 2.23 2.19(+1.19%)
>>>>> avg-lat-24threads 2.46 2.36(+3.79%)
>>>>> avg-lat-36threads 2.66 2.57(+3.26%)
>>>>> avg-lat-64threads 5.23 5.05(+3.44%)
>>>>>
>>>>> Tested-by: Yicong Yang <yangyicong@hisilicon.com>
>>>>> Signed-off-by: Barry Song <song.bao.hua@hisilicon.com>
>>>>> Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
>>>>> ---
>>>>> kernel/sched/fair.c | 46
>>>>> +++++++++++++++++++++++++++++++++++++++++
>>>>> ----
>>>>> 1 file changed, 42 insertions(+), 4 deletions(-)
>>>>>
>>>>> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
>>>>> index 5146163bfabb..2f84a933aedd 100644
>>>>> --- a/kernel/sched/fair.c
>>>>> +++ b/kernel/sched/fair.c
>>>>> @@ -6262,12 +6262,46 @@ static inline int
>>>>> select_idle_smt(struct
>>>>> task_struct *p, struct sched_domain *sd
>>>>>
>>>>> #endif /* CONFIG_SCHED_SMT */
>>>>>
>>>>> +#ifdef CONFIG_SCHED_CLUSTER
>>>>> +/*
>>>>> + * Scan the cluster domain for idle CPUs and clear cluster
>>>>> cpumask
>>>>> after scanning
>>>>> + */
>>>>> +static inline int scan_cluster(struct task_struct *p, int
>>>>> prev_cpu,
>>>>> int target)
>>>>> +{
>>>>> + struct cpumask *cpus =
>>>>> this_cpu_cpumask_var_ptr(select_idle_mask);
>>>>> + struct sched_domain *sd =
>>>>> rcu_dereference(per_cpu(sd_cluster,
>>>>> target));
>>>>> + int cpu, idle_cpu;
>>>>> +
>>>>> + /* TODO: Support SMT case while a machine with both
>>>>> cluster and
>>>>> SMT born */
>>>>
>>>> This is probably a clearer comment
>>>>
>>>> /* TODO: Support SMT system with cluster topology */
>>>>
>>>>> + if (!sched_smt_active() && sd) {
>>>>> + for_each_cpu_and(cpu, cpus,
>>>>> sched_domain_span(sd)) {
>>>>> + idle_cpu = __select_idle_cpu(cpu, p);
>>>>> */
>>>>> -static int select_idle_cpu(struct task_struct *p, struct
>>>>> sched_domain *sd, bool has_idle_core, int target)
>>>>> +static int select_idle_cpu(struct task_struct *p, struct
>>>>> sched_domain *sd, bool has_idle_core, int prev_cpu, int target)
>>>>> {
>>>>> struct cpumask *cpus =
>>>>> this_cpu_cpumask_var_ptr(select_idle_mask);
>>>>> int i, cpu, idle_cpu = -1, nr = INT_MAX;
>>>>> @@ -6282,6 +6316,10 @@ static int select_idle_cpu(struct
>>>>> task_struct
>>>>> *p, struct sched_domain *sd, bool
>>>>>
>>>>> cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
>>>>>
>>>>> + idle_cpu = scan_cluster(p, prev_cpu, target);
>>>>
>>>> Shouldn't "cpus" from
>>>>
>>>> cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
>>>>
>>>> be passed to scan_cluster, to make sure that the cpu returned is
>>>> in the affinity mask of the task? I don't see p->cpus_ptr
>>>> being checked in scan_cluster to make sure the cpu found is in
>>>> the
>>>> affinity mask.
>>>>
>>>
>>> The cpus scanned in scan_cluster() is the intersection of
>>> select_idle_mask and sched_domain_span(cluster_sd), and
>>> we limited the select_idle_mask in the tasks' affinity mask
>>> before we enter scan_cluster() here.
>>
>> Ah, I missed the fact that cpus point to the select_idle_mask.
>>
>
> I think it will be easier to read the code if you pass "cpus" directly
> to scan cluster, rather than making this implicit, and having this
> assignment
>
> *cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
>
> again in scan_cluster.
sure. It does look more readable and I think we can change to that. :)
Thanks.
next prev parent reply other threads:[~2022-01-27 3:05 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-26 8:09 [PATCH v2 0/2] sched/fair: Wake task within the cluster when possible Yicong Yang
2022-01-26 8:09 ` [PATCH v2 1/2] sched: Add per_cpu cluster domain info and cpus_share_resources API Yicong Yang
2022-01-27 15:26 ` Gautham R. Shenoy
2022-01-26 8:09 ` [PATCH v2 2/2] sched/fair: Scan cluster before scanning LLC in wake-up path Yicong Yang
2022-01-27 1:14 ` Tim Chen
2022-01-27 2:02 ` Yicong Yang
2022-01-27 2:30 ` Tim Chen
2022-01-27 2:36 ` Tim Chen
2022-01-27 3:05 ` Yicong Yang [this message]
2022-01-27 15:41 ` Gautham R. Shenoy
2022-01-27 20:21 ` Barry Song
2022-01-28 7:13 ` Srikar Dronamraju
2022-01-27 18:40 ` Barry Song
2022-02-01 9:38 ` Srikar Dronamraju
2022-02-01 20:20 ` Barry Song
2022-02-04 7:33 ` Srikar Dronamraju
2022-02-04 10:28 ` Barry Song
2022-02-04 10:49 ` Barry Song
2022-02-04 17:41 ` Tim Chen
2022-02-05 17:16 ` Chen Yu
2022-02-06 0:26 ` Barry Song
2022-02-07 15:14 ` Gautham R. Shenoy
2022-02-08 5:42 ` Barry Song
2022-02-16 9:12 ` Barry Song
2022-02-16 9:19 ` Song Bao Hua (Barry Song)
2022-02-16 10:00 ` Yicong Yang
2022-02-17 18:00 ` Tim Chen
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=4d4099d8-e890-9e56-c395-5f521d98081a@huawei.com \
--to=yangyicong@huawei.com \
--cc=21cnbao@gmail.com \
--cc=bristot@redhat.com \
--cc=bsegall@google.com \
--cc=dietmar.eggemann@arm.com \
--cc=ego@linux.vnet.ibm.com \
--cc=gautham.shenoy@amd.com \
--cc=guodong.xu@linaro.org \
--cc=jonathan.cameron@huawei.com \
--cc=juri.lelli@redhat.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=prime.zeng@huawei.com \
--cc=rostedt@goodmis.org \
--cc=song.bao.hua@hisilicon.com \
--cc=srikar@linux.vnet.ibm.com \
--cc=tim.c.chen@linux.intel.com \
--cc=vincent.guittot@linaro.org \
--cc=yangyicong@hisilicon.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).