[lkml]   [2013]   [Oct]   [9]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH] sched_clock: fix postinit no sched_clock function check
On Wednesday 09 October 2013 07:59 PM, John Stultz wrote:
> On 10/02/2013 11:07 AM, Santosh Shilimkar wrote:
>> On Wednesday 02 October 2013 01:48 PM, Will Deacon wrote:
>>> On Wed, Oct 02, 2013 at 06:42:40PM +0100, Stephen Boyd wrote:
>>>> On 10/02/13 10:27, Santosh Shilimkar wrote:
>>>>> Really... I have not created patch out of fun.
>>>>> Its broken on my keystone machine at least where the sched_clock is
>>>>> falling back on jiffy based sched_clock even in presence of arch_timer
>>>>> sched_clock.
>>>> How is that possible? sched_clock_func is only assigned by
>>>> arch/arm/kernel/arch_timer.c when the architected timer is detected and
>>>> sched_clock() in kernel/time/sched_clock.c calls that function pointer
>>>> unconditionally. The only way I see this happening is if the architected
>>>> timer rate is zero.
>>> ^^^^^^^^^^^^^^^^^^
>>> *cough* CNTFRQ *cough*
>> :) CNTFRQ as such is fine. I think the below print mis-lead me mostly.
>> sched_clock: ARM arch timer >56 bits at 6144kHz, resolution 162ns
>> sched_clock: 32 bits at 100 Hz, resolution 10000000ns, wraps every 4294967286ms
>> So yes, now the subject patch actually just avoids the jiffy sched_clock()
>> registration and nothing else. Even without the patch arch_timer sched_clock
>> will be in use.
> Just wanted to follow up here, as I've not been paying close attention.
> Is this issue then resolved, or is something still needed to be queued
> for 3.12/3.13?
There is no regression as I initially thought. Patch fixes the
miss-leading sched_clock print and also prevents timer to handle
wrapping which is not needed.

So no big deal and I don't mind if we don't apply it.


 \ /
  Last update: 2013-10-10 02:41    [W:0.072 / U:49.904 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site