Messages in this thread Patch in this message | | | Date | Tue, 31 Jul 2018 12:59:16 +0200 | From | Peter Zijlstra <> | Subject | Re: [PATCHv4 01/12] sched: Add static_key for asymmetric cpu capacity optimizations |
| |
Combined with that SD_ASYM.. rework I ended up with the below.
Holler if you want it changed :-)
---
Subject: sched: Add static_key for asymmetric cpu capacity optimizations From: Morten Rasmussen <morten.rasmussen@arm.com> Date: Wed, 4 Jul 2018 11:17:39 +0100
The existing asymmetric cpu capacity code should cause minimal overhead for others. Putting it behind a static_key, it has been done for SMT optimizations, would make it easier to extend and improve without causing harm to others moving forward.
cc: Ingo Molnar <mingo@redhat.com> cc: Peter Zijlstra <peterz@infradead.org>
Cc: valentin.schneider@arm.com Cc: mingo@redhat.com Cc: vincent.guittot@linaro.org Cc: dietmar.eggemann@arm.com Cc: gaku.inami.xh@renesas.com Signed-off-by: Morten Rasmussen <morten.rasmussen@arm.com> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: http://lkml.kernel.org/r/1530699470-29808-2-git-send-email-morten.rasmussen@arm.com --- kernel/sched/fair.c | 3 +++ kernel/sched/sched.h | 1 + kernel/sched/topology.c | 9 ++++++++- 3 files changed, 12 insertions(+), 1 deletion(-)
--- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6186,6 +6186,9 @@ static int wake_cap(struct task_struct * { long min_cap, max_cap; + if (!static_branch_unlikely(&sched_asym_cpucapacity)) + return 0; + min_cap = min(capacity_orig_of(prev_cpu), capacity_orig_of(cpu)); max_cap = cpu_rq(cpu)->rd->max_cpu_capacity; --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1185,6 +1185,7 @@ DECLARE_PER_CPU(int, sd_llc_id); DECLARE_PER_CPU(struct sched_domain_shared *, sd_llc_shared); DECLARE_PER_CPU(struct sched_domain *, sd_numa); DECLARE_PER_CPU(struct sched_domain *, sd_asym); +extern struct static_key_false sched_asym_cpucapacity; struct sched_group_capacity { atomic_t ref; --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -398,6 +398,7 @@ DEFINE_PER_CPU(int, sd_llc_id); DEFINE_PER_CPU(struct sched_domain_shared *, sd_llc_shared); DEFINE_PER_CPU(struct sched_domain *, sd_numa); DEFINE_PER_CPU(struct sched_domain *, sd_asym); +DEFINE_STATIC_KEY_FALSE(sched_asym_cpucapacity); static void update_top_cache_domain(int cpu) { @@ -1708,6 +1709,7 @@ build_sched_domains(const struct cpumask struct rq *rq = NULL; int i, ret = -ENOMEM; struct sched_domain_topology_level *tl_asym; + bool has_asym = false; alloc_state = __visit_domain_allocation_hell(&d, cpu_map); if (alloc_state != sa_rootdomain) @@ -1723,8 +1725,10 @@ build_sched_domains(const struct cpumask for_each_sd_topology(tl) { int dflags = 0; - if (tl == tl_asym) + if (tl == tl_asym) { dflags |= SD_ASYM_CPUCAPACITY; + has_asym = true; + } sd = build_sched_domain(tl, cpu_map, attr, sd, dflags, i); @@ -1776,6 +1780,9 @@ build_sched_domains(const struct cpumask } rcu_read_unlock(); + if (has_asym) + static_branch_enable_cpuslocked(&sched_asym_cpucapacity); + if (rq && sched_debug_enabled) { pr_info("root domain span: %*pbl (max cpu_capacity = %lu)\n", cpumask_pr_args(cpu_map), rq->rd->max_cpu_capacity);
| |