| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <3d612766-592f-760f-ca43-4ea2257a8bb2@huawei.com>
Date: Thu, 14 Jul 2022 17:41:04 +0800
From: Yicong Yang <yangyicong@...wei.com>
To: <peterz@...radead.org>, <mingo@...hat.com>,
<juri.lelli@...hat.com>, <vincent.guittot@...aro.org>,
<tim.c.chen@...ux.intel.com>, <gautham.shenoy@....com>,
<linux-kernel@...r.kernel.org>,
<linux-arm-kernel@...ts.infradead.org>
CC: <dietmar.eggemann@....com>, <rostedt@...dmis.org>,
<bsegall@...gle.com>, <bristot@...hat.com>,
<prime.zeng@...wei.com>, <jonathan.cameron@...wei.com>,
<ego@...ux.vnet.ibm.com>, <srikar@...ux.vnet.ibm.com>,
<linuxarm@...wei.com>, <21cnbao@...il.com>,
<guodong.xu@...aro.org>, <hesham.almatary@...wei.com>,
<john.garry@...wei.com>, <shenyang39@...wei.com>,
<kprateek.nayak@....com>, <yu.c.chen@...el.com>,
<wuyun.abel@...edance.com>, Yicong Yang <yangyicong@...ilicon.com>
Subject: Re: [PATCH v5 2/2] sched/fair: Scan cluster before scanning LLC in
wake-up path
Hi,
a friendly ping..
On 2022/6/30 14:55, Yicong Yang wrote:
> From: Barry Song <song.bao.hua@...ilicon.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:
> tip/core patched
> Hmean 1 345.89 ( 0.00%) 393.96 * 13.90%*
> Hmean 2 697.77 ( 0.00%) 786.04 * 12.65%*
> Hmean 4 1392.51 ( 0.00%) 1570.26 * 12.76%*
> Hmean 8 2800.61 ( 0.00%) 3083.98 * 10.12%*
> Hmean 16 5514.27 ( 0.00%) 6116.00 * 10.91%*
> Hmean 32 10869.81 ( 0.00%) 10782.98 * -0.80%*
> Hmean 64 8315.22 ( 0.00%) 8519.84 * 2.46%*
> Hmean 128 6324.47 ( 0.00%) 7159.35 * 13.20%*
>
> On numa 0-1:
> tip/core patched
> Hmean 1 348.68 ( 0.00%) 387.91 * 11.25%*
> Hmean 2 693.57 ( 0.00%) 774.91 * 11.73%*
> Hmean 4 1369.26 ( 0.00%) 1475.48 * 7.76%*
> Hmean 8 2772.99 ( 0.00%) 2984.61 * 7.63%*
> Hmean 16 4825.83 ( 0.00%) 5873.13 * 21.70%*
> Hmean 32 10250.32 ( 0.00%) 11688.06 * 14.03%*
> Hmean 64 16309.51 ( 0.00%) 19889.48 * 21.95%*
> Hmean 128 13022.32 ( 0.00%) 16005.64 * 22.91%*
> Hmean 256 11335.79 ( 0.00%) 13821.74 * 21.93%*
>
> Tested-by: Yicong Yang <yangyicong@...ilicon.com>
> Signed-off-by: Barry Song <song.bao.hua@...ilicon.com>
> Signed-off-by: Yicong Yang <yangyicong@...ilicon.com>
> Reviewed-by: Tim Chen <tim.c.chen@...ux.intel.com>
> ---
> kernel/sched/fair.c | 44 +++++++++++++++++++++++++++++++++++++++++---
> 1 file changed, 41 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index f80ae86bb404..dff5dec0d792 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -6323,6 +6323,40 @@ 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, struct cpumask *cpus,
> + int target, int *nr)
> +{
> + struct sched_domain *sd = rcu_dereference(per_cpu(sd_cluster, target));
> + int cpu, idle_cpu;
> +
> + /* TODO: Support SMT system with cluster topology */
> + if (!sched_smt_active() && sd) {
> + for_each_cpu_and(cpu, cpus, sched_domain_span(sd)) {
> + if (!--*nr)
> + return -1;
> +
> + idle_cpu = __select_idle_cpu(cpu, p);
> + if ((unsigned int)idle_cpu < nr_cpumask_bits)
> + return idle_cpu;
> + }
> +
> + cpumask_andnot(cpus, cpus, sched_domain_span(sd));
> + }
> +
> + return -1;
> +}
> +#else
> +static inline int scan_cluster(struct task_struct *p, struct cpumask *cpus,
> + int target, int *nr)
> +{
> + return -1;
> +}
> +#endif
> +
> /*
> * Scan the LLC domain for idle CPUs; this is dynamically regulated by
> * comparing the average scan cost (tracked in sd->avg_scan_cost) against the
> @@ -6383,6 +6417,10 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool
> }
> }
>
> + idle_cpu = scan_cluster(p, cpus, target, &nr);
> + if ((unsigned int)idle_cpu < nr_cpumask_bits)
> + return idle_cpu;
> +
> for_each_cpu_wrap(cpu, cpus, target + 1) {
> if (has_idle_core) {
> i = select_idle_core(p, cpu, cpus, &idle_cpu);
> @@ -6390,7 +6428,7 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool
> return i;
>
> } else {
> - if (!--nr)
> + if (--nr <= 0)
> return -1;
> idle_cpu = __select_idle_cpu(cpu, p);
> if ((unsigned int)idle_cpu < nr_cpumask_bits)
> @@ -6489,7 +6527,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
> /*
> * If the previous CPU is cache affine and idle, don't be stupid:
> */
> - if (prev != target && cpus_share_cache(prev, target) &&
> + if (prev != target && cpus_share_lowest_cache(prev, target) &&
> (available_idle_cpu(prev) || sched_idle_cpu(prev)) &&
> asym_fits_capacity(task_util, prev))
> return prev;
> @@ -6515,7 +6553,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
> p->recent_used_cpu = prev;
> if (recent_used_cpu != prev &&
> recent_used_cpu != target &&
> - cpus_share_cache(recent_used_cpu, target) &&
> + cpus_share_lowest_cache(recent_used_cpu, target) &&
> (available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) &&
> cpumask_test_cpu(p->recent_used_cpu, p->cpus_ptr) &&
> asym_fits_capacity(task_util, recent_used_cpu)) {
>
Powered by blists - more mailing lists