"Qingqing Zhou" <zhouqq@cs.toronto.edu> writes:
> I agree on this. But before changing it, we need to inspect those spinlocks
> one by one to making sure two things (1) if there is out-of-line-call, make
> sure no CHECK_FOR_INTERRUPTS(); (2) ImmediateInterruptsOK is false (99% sure
> now).
I'm sure of those things already ... but feel free to look for yourself.
>> I think it would be better to substitute our own qsort only on those
>> platforms where we have specifically proved it's a win.
>>
> Our tests indicates that BSD version is better ... but it is just a
> home-brew test.
Better than what is the operative question.
regards, tom lane
