Re: [Patch v3 1/3] lib: Restrict cpumask_local_spread to houskeeping CPUs
From: Nitesh Narayan Lal <hidden>
Date: 2020-06-24 20:38:59
Also in:
linux-pci, lkml
On 6/24/20 3:26 PM, Andrew Morton wrote:
On Tue, 23 Jun 2020 15:23:29 -0400 Nitesh Narayan Lal [off-list ref] wrote:quoted
From: Alex Belits <redacted> The current implementation of cpumask_local_spread() does not respect the isolated CPUs, i.e., even if a CPU has been isolated for Real-Time task, it will return it to the caller for pinning of its IRQ threads. Having these unwanted IRQ threads on an isolated CPU adds up to a latency overhead. Restrict the CPUs that are returned for spreading IRQs only to the available housekeeping CPUs. ...--- a/lib/cpumask.c +++ b/lib/cpumask.c@@ -6,6 +6,7 @@ #include <linux/export.h> #include <linux/memblock.h> #include <linux/numa.h> +#include <linux/sched/isolation.h> /** * cpumask_next - get the next cpu in a cpumask@@ -205,22 +206,27 @@ void __init free_bootmem_cpumask_var(cpumask_var_t mask) */ unsigned int cpumask_local_spread(unsigned int i, int node) { - int cpu; + int cpu, hk_flags; + const struct cpumask *mask; + hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; + mask = housekeeping_cpumask(hk_flags); /* Wrap: we always want a cpu. */ - i %= num_online_cpus(); + i %= cpumask_weight(mask); if (node == NUMA_NO_NODE) { - for_each_cpu(cpu, cpu_online_mask) + for_each_cpu(cpu, mask) { if (i-- == 0) return cpu; + } } else { /* NUMA first. */ - for_each_cpu_and(cpu, cpumask_of_node(node), cpu_online_mask) + for_each_cpu_and(cpu, cpumask_of_node(node), mask) { if (i-- == 0) return cpu; + } - for_each_cpu(cpu, cpu_online_mask) { + for_each_cpu(cpu, mask) { /* Skip NUMA nodes, done above. */ if (cpumask_test_cpu(cpu, cpumask_of_node(node))) continue;Are you aware of these changes to cpu_local_spread()? https://lore.kernel.org/lkml/1582768688-2314-1-git-send-email-zhangshaokun@hisilicon.com/ (local) I don't see a lot of overlap but it would be nice for you folks to check each other's homework ;)
Sure, I will take a look. Thanks
-- Nitesh
Attachments
- signature.asc [application/pgp-signature] 833 bytes