Re: [PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-25 Thread H. Peter Anvin
On 03/23/18 18:01, Eric Dumazet wrote: > > Indeed, assuming a daemon can have threads running on all cpus :/ > > Some environments like to partition cpus for different jobs/containers. > Then it doesn't need to run it on all cpus... just the one that that particular daemon is "responsible"

Re: [PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-25 Thread H. Peter Anvin
On 03/23/18 18:01, Eric Dumazet wrote: > > Indeed, assuming a daemon can have threads running on all cpus :/ > > Some environments like to partition cpus for different jobs/containers. > Then it doesn't need to run it on all cpus... just the one that that particular daemon is "responsible"

Re: [PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-23 Thread Eric Dumazet
On 03/23/2018 03:17 PM, H. Peter Anvin wrote: > On 03/23/18 14:58, Eric Dumazet wrote: >> I noticed high latencies caused by a daemon periodically reading various >> MSR and cpuid on all cpus. KASAN kernels would see ~10ms latencies >> simply reading one cpuid. Even without KASAN, sending IPI to

Re: [PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-23 Thread Eric Dumazet
On 03/23/2018 03:17 PM, H. Peter Anvin wrote: > On 03/23/18 14:58, Eric Dumazet wrote: >> I noticed high latencies caused by a daemon periodically reading various >> MSR and cpuid on all cpus. KASAN kernels would see ~10ms latencies >> simply reading one cpuid. Even without KASAN, sending IPI to

Re: [PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-23 Thread H. Peter Anvin
On 03/23/18 14:58, Eric Dumazet wrote: > I noticed high latencies caused by a daemon periodically reading various > MSR and cpuid on all cpus. KASAN kernels would see ~10ms latencies > simply reading one cpuid. Even without KASAN, sending IPI to CPU > in deep sleep state or blocking hard IRQ in a

Re: [PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-23 Thread H. Peter Anvin
On 03/23/18 14:58, Eric Dumazet wrote: > I noticed high latencies caused by a daemon periodically reading various > MSR and cpuid on all cpus. KASAN kernels would see ~10ms latencies > simply reading one cpuid. Even without KASAN, sending IPI to CPU > in deep sleep state or blocking hard IRQ in a

[PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-23 Thread Eric Dumazet
I noticed high latencies caused by a daemon periodically reading various MSR and cpuid on all cpus. KASAN kernels would see ~10ms latencies simply reading one cpuid. Even without KASAN, sending IPI to CPU in deep sleep state or blocking hard IRQ in a a long section, then waiting for the answer can

[PATCH v3 2/2] x86, cpuid: allow cpuid_read() to schedule

2018-03-23 Thread Eric Dumazet
I noticed high latencies caused by a daemon periodically reading various MSR and cpuid on all cpus. KASAN kernels would see ~10ms latencies simply reading one cpuid. Even without KASAN, sending IPI to CPU in deep sleep state or blocking hard IRQ in a a long section, then waiting for the answer can