Re: Make yield_task_fair more efficient

[Balbir Singh <balbir@linux.vnet.ibm.com>]

Balbir Singh wrote:
> Ingo Molnar wrote:
>> * Balbir Singh <balbir@linux.vnet.ibm.com> wrote:
>>
>>> If you insist that sched_yield() is bad, I might agree, but how does 
>>> my patch make things worse. [...]
>> it puts new instructions into the hotpath.
>>
>>> [...] In my benchmarks, it has helped the sched_yield case, why is 
>>> that bad? [...]
>> I had the same cache for the rightmost task in earlier CFS (it's a 
>> really obvious thing) but removed it. It wasnt a bad idea, but it hurt 
>> the fastpath hence i removed it. Algorithms and implementations are a 
>> constant balancing act.
> 
> This is more convincing, was the code ever in git? How did you measure the
> overhead? What are your plans for reports with regressions where
> kernel.compat_sched_yield is set to 1?
> 

Ingo,

I was looking through nptl/sysdeps/unix/sysv/linux/pthread_yield.c in the glibc
sources and it looks like pthread_yield() calls sched_yield(). Applications
using compat_sched_yield and pthread_yield() are also going to be impacted.

I searched for pthread_yield and sched_yield using google codesearch to look at
the applications that use these routines, the search list is too big for me to
mark applications as candidates for potential improvement.

-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL