Message ID | 20200622234510.240834-2-nitesh@redhat.com |
---|---|
State | New |
Headers | show |
Series | Preventing job distribution to isolated CPUs | expand |
On Mon, Jun 22, 2020 at 07:45:08PM -0400, Nitesh Narayan Lal wrote: > From: Alex Belits <abelits@marvell.com> > > 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. > > Signed-off-by: Alex Belits <abelits@marvell.com> > Signed-off-by: Nitesh Narayan Lal <nitesh@redhat.com> > --- > lib/cpumask.c | 43 +++++++++++++++++++++++++------------------ > 1 file changed, 25 insertions(+), 18 deletions(-) > > diff --git a/lib/cpumask.c b/lib/cpumask.c > index fb22fb266f93..cc4311a8c079 100644 > --- 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 > @@ -205,28 +206,34 @@ void __init free_bootmem_cpumask_var(cpumask_var_t mask) > */ > unsigned int cpumask_local_spread(unsigned int i, int node) > { > - int cpu; > + int cpu, m, n, hk_flags; > + const struct cpumask *mask; > > + hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; > + mask = housekeeping_cpumask(hk_flags); > + m = cpumask_weight(mask); > /* Wrap: we always want a cpu. */ > - i %= num_online_cpus(); > + n = i % m; > + while (m-- > 0) { I are confuzled. What do we need this outer loop for? Why isn't something like: i %= cpumask_weight(mask); good enough? That voids having to touch the test. Still when you're there, at the very least you can fix the horrible style: > + if (node == NUMA_NO_NODE) { > + for_each_cpu(cpu, mask) > + if (n-- == 0) > + return cpu; { } > + } else { > + /* NUMA first. */ > + for_each_cpu_and(cpu, cpumask_of_node(node), mask) > + if (n-- == 0) > + return cpu; { } > > + for_each_cpu(cpu, mask) { > + /* Skip NUMA nodes, done above. */ > + if (cpumask_test_cpu(cpu, > + cpumask_of_node(node))) > + continue; No linebreak please. > > + if (n-- == 0) > + return cpu; > + } > } > } > BUG(); > -- > 2.18.4 >
On 6/23/20 5:21 AM, Peter Zijlstra wrote: > On Mon, Jun 22, 2020 at 07:45:08PM -0400, Nitesh Narayan Lal wrote: >> From: Alex Belits <abelits@marvell.com> >> >> 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. >> >> Signed-off-by: Alex Belits <abelits@marvell.com> >> Signed-off-by: Nitesh Narayan Lal <nitesh@redhat.com> >> --- >> lib/cpumask.c | 43 +++++++++++++++++++++++++------------------ >> 1 file changed, 25 insertions(+), 18 deletions(-) >> >> diff --git a/lib/cpumask.c b/lib/cpumask.c >> index fb22fb266f93..cc4311a8c079 100644 >> --- 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 >> @@ -205,28 +206,34 @@ void __init free_bootmem_cpumask_var(cpumask_var_t mask) >> */ >> unsigned int cpumask_local_spread(unsigned int i, int node) >> { >> - int cpu; >> + int cpu, m, n, hk_flags; >> + const struct cpumask *mask; >> >> + hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; >> + mask = housekeeping_cpumask(hk_flags); >> + m = cpumask_weight(mask); >> /* Wrap: we always want a cpu. */ >> - i %= num_online_cpus(); >> + n = i % m; >> + while (m-- > 0) { > I are confuzled. What do we need this outer loop for? > > Why isn't something like: > > i %= cpumask_weight(mask); > > good enough? That voids having to touch the test. Makes sense. Thanks > Still when you're there, at the very least you can fix the horrible > style: Sure. > > >> + if (node == NUMA_NO_NODE) { >> + for_each_cpu(cpu, mask) >> + if (n-- == 0) >> + return cpu; > { } > >> + } else { >> + /* NUMA first. */ >> + for_each_cpu_and(cpu, cpumask_of_node(node), mask) >> + if (n-- == 0) >> + return cpu; > { } > >> >> + for_each_cpu(cpu, mask) { >> + /* Skip NUMA nodes, done above. */ >> + if (cpumask_test_cpu(cpu, >> + cpumask_of_node(node))) >> + continue; > No linebreak please. > >> >> + if (n-- == 0) >> + return cpu; >> + } >> } >> } >> BUG(); >> -- >> 2.18.4 >>
diff --git a/lib/cpumask.c b/lib/cpumask.c index fb22fb266f93..cc4311a8c079 100644 --- 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 @@ -205,28 +206,34 @@ void __init free_bootmem_cpumask_var(cpumask_var_t mask) */ unsigned int cpumask_local_spread(unsigned int i, int node) { - int cpu; + int cpu, m, n, hk_flags; + const struct cpumask *mask; + hk_flags = HK_FLAG_DOMAIN | HK_FLAG_WQ; + mask = housekeeping_cpumask(hk_flags); + m = cpumask_weight(mask); /* Wrap: we always want a cpu. */ - i %= num_online_cpus(); + n = i % m; + while (m-- > 0) { + if (node == NUMA_NO_NODE) { + for_each_cpu(cpu, mask) + if (n-- == 0) + return cpu; + } else { + /* NUMA first. */ + for_each_cpu_and(cpu, cpumask_of_node(node), mask) + if (n-- == 0) + return cpu; - if (node == NUMA_NO_NODE) { - for_each_cpu(cpu, cpu_online_mask) - if (i-- == 0) - return cpu; - } else { - /* NUMA first. */ - for_each_cpu_and(cpu, cpumask_of_node(node), cpu_online_mask) - if (i-- == 0) - return cpu; + for_each_cpu(cpu, mask) { + /* Skip NUMA nodes, done above. */ + if (cpumask_test_cpu(cpu, + cpumask_of_node(node))) + continue; - for_each_cpu(cpu, cpu_online_mask) { - /* Skip NUMA nodes, done above. */ - if (cpumask_test_cpu(cpu, cpumask_of_node(node))) - continue; - - if (i-- == 0) - return cpu; + if (n-- == 0) + return cpu; + } } } BUG();