This is the mail archive of the
mailing list for the glibc project.
Re: [PATCH] Add math benchmark latency test
On Wednesday 16 August 2017 07:13 PM, Szabolcs Nagy wrote:
> yes in principle the throughput is number of calls/unit time,
> but the inverse of that is just as useful, it's the rate at
> which calls can be issued and then it can be easily compared
> to the latency number, but i guess printing it as calls/sec
> would work too.
I don't dispute its utility, I'd just like it to match what it is
called, so calls/unit time for throughput and time/call for latency.
BTW, this will only work for functions that have the same first argument
and return types. This is OK for now, but it might be something to fix
in future. A comment to that effect in the code would be very useful
because I'm pretty sure I'll forget about it in the next 15 minutes or
so and Wilco will too, maybe later than that :)