lkml.org 
[lkml]   [2012]   [Jan]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [PATCH] Avoid mask based num_possible_cpus and num_online_cpus
(1/18/12 1:52 PM), Venki Pallipadi wrote:
> On Tue, Jan 17, 2012 at 9:55 PM, KOSAKI Motohiro
> <kosaki.motohiro@gmail.com> wrote:
>> (1/17/12 9:07 PM), Venkatesh Pallipadi wrote:
>>> Kernel's notion of possible cpus (from include/linux/cpumask.h)
>>> * cpu_possible_mask- has bit 'cpu' set iff cpu is populatable
>>>
>>> * The cpu_possible_mask is fixed at boot time, as the set of CPU id's
>>> * that it is possible might ever be plugged in at anytime during the
>>> * life of that system boot.
>>>
>>> #define num_possible_cpus() cpumask_weight(cpu_possible_mask)
>>>
>>> and on x86 cpumask_weight() calls hweight64 and hweight64 (on older kernels
>>> and systems with !X86_FEATURE_POPCNT) or a popcnt based alternative.
>>>
>>> i.e, We needlessly go through this mask based calculation everytime
>>> num_possible_cpus() is called.
>>>
>>> The problem is there with cpu_online_mask() as well, which is fixed value at
>>> boot time in !CONFIG_HOTPLUG_CPU case and should not change that often even
>>> in HOTPLUG case.
>>>
>>> Though most of the callers of these two routines are init time (with few
>>> exceptions of runtime calls), it is cleaner to use variables
>>> and not go through this repeated mask based calculation.
>>>
>>> Signed-off-by: Venkatesh Pallipadi<venki@google.com>
>>> ---
>>> include/linux/cpumask.h | 8 ++++++--
>>> kernel/cpu.c | 9 +++++++++
>>> 2 files changed, 15 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h
>>> index 4f7a632..2eb04dd 100644
>>> --- a/include/linux/cpumask.h
>>> +++ b/include/linux/cpumask.h
>>> @@ -80,9 +80,13 @@ extern const struct cpumask *const cpu_online_mask;
>>> extern const struct cpumask *const cpu_present_mask;
>>> extern const struct cpumask *const cpu_active_mask;
>>>
>>> +extern int nr_online_cpus;
>>> +
>>> #if NR_CPUS> 1
>>> -#define num_online_cpus() cpumask_weight(cpu_online_mask)
>>> -#define num_possible_cpus() cpumask_weight(cpu_possible_mask)
>>> +
>>> +#define num_online_cpus() (nr_online_cpus)
>>> +#define num_possible_cpus() (nr_cpu_ids)
>>
>> nr_cpu_ids mean maximum cpu id of cpus. if cpu id are sparse, maximum id
>> doesn't match number of cpus.
>>
>
> Yes. But will it be sparse in any arch? I saw some of the users of
> num_possible_cpus() doing things like allocating a buffer for that
> size and then indexing it using get_cpu(). So, I thought it would be
> better to use the existing nr_cpu_ids instead of inventing another
> variable. If indeed any arch is depending on this being sparse, we can
> have a new variable similar to num_possible_cpus and also audit all
> users of num_possible_cpus to see whether they should be using
> nr_cpu_ids instead.

If my remember is correct, sparc does.




\
 
 \ /
  Last update: 2012-01-18 20:23    [W:0.063 / U:15.260 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site