[PATCH 4 of 8 v4] ui: log time spent blocked on stdio

[Simon Farnsworth]

On 13/02/2017 18:00, Bryan O'Sullivan wrote:
>
> On Mon, Feb 13, 2017 at 9:29 AM, Simon Farnsworth <simonfar at fb.com
> <mailto:simonfar at fb.com>> wrote:
>
>     The perfwrite microbenchmark shifts to:
>     Linux:
>     ! wall 3.316087 comb 0.900000 user 0.810000 sys 0.090000 (best of 3)
>     Mac:
>     ! wall 0.939282 comb 0.580000 user 0.470000 sys 0.110000 (best of 8)
>
>     If I open-code timings in ui.write instead of using the context
>     manager, I see:
>     Linux:
>     ! wall 3.478070 comb 0.500000 user 0.420000 sys 0.080000 (best of 3)
>     Mac:
>     ! wall 0.218112 comb 0.220000 user 0.150000 sys 0.070000 (best of 15)
>
>
> Thanks for adding this data. (For future reference, it's good to include
> an analysis of the difference instead of just the plain numbers, as I
> have to flip back and forth between two emails and memorize the numbers
> to do the comparison in my head.)
>
> These numbers suggest to me that open-coding the performance measurement
> is the right thing to do for ui.write and ui.flush.
>
I'll make that change locally, but I'm not going to resubmit for a few 
days, to give other interested parties time to weigh in.
-- 
Simon Farnsworth