Re: [PATCH v2] lib/smp_processor_id: Don't use cpumask_equal()

2019-10-07 Thread Peter Zijlstra
On Thu, Oct 03, 2019 at 04:36:08PM -0400, Waiman Long wrote: > The check_preemption_disabled() function uses cpumask_equal() to see > if the task is bounded to the current CPU only. cpumask_equal() calls > memcmp() to do the comparison. As x86 doesn't have __HAVE_ARCH_MEMCMP, > the slow memcmp()

Re: [PATCH v2] lib/smp_processor_id: Don't use cpumask_equal()

2019-10-04 Thread Waiman Long
On 10/4/19 5:20 AM, Sebastian Andrzej Siewior wrote: > On 2019-10-03 16:36:08 [-0400], Waiman Long wrote: >> The check_preemption_disabled() function uses cpumask_equal() to see >> if the task is bounded to the current CPU only. cpumask_equal() calls >> memcmp() to do the comparison. As x86

Re: [PATCH v2] lib/smp_processor_id: Don't use cpumask_equal()

2019-10-04 Thread Juri Lelli
Hi, On 03/10/19 16:36, Waiman Long wrote: > The check_preemption_disabled() function uses cpumask_equal() to see > if the task is bounded to the current CPU only. cpumask_equal() calls > memcmp() to do the comparison. As x86 doesn't have __HAVE_ARCH_MEMCMP, > the slow memcmp() function in

Re: [PATCH v2] lib/smp_processor_id: Don't use cpumask_equal()

2019-10-04 Thread Sebastian Andrzej Siewior
On 2019-10-03 16:36:08 [-0400], Waiman Long wrote: > The check_preemption_disabled() function uses cpumask_equal() to see > if the task is bounded to the current CPU only. cpumask_equal() calls > memcmp() to do the comparison. As x86 doesn't have __HAVE_ARCH_MEMCMP, > the slow memcmp() function in

[PATCH v2] lib/smp_processor_id: Don't use cpumask_equal()

2019-10-03 Thread Waiman Long
The check_preemption_disabled() function uses cpumask_equal() to see if the task is bounded to the current CPU only. cpumask_equal() calls memcmp() to do the comparison. As x86 doesn't have __HAVE_ARCH_MEMCMP, the slow memcmp() function in lib/string.c is used. On a RT kernel that call