Okay, I'm sorry to admit that I don't immediately remember the answer to this question ...
(but I also don't feel that bad, since I think this information should be more apparent in the presentation itself)
... but: Is perf.rlo measuring the performance of the generated object binary? Or is it measuring the compilation time?
I'm looking in particular at stats like https://perf.rust-lang.org/compare.html?start=eab3eb38df8dca99110b6149b3a15deeb4ef0413&end=4cb14446465d8d3cabfb3706f424a2938628b9f6&stat=wall-time
It's all compilation time
of a bootstrapped compiler?
i.e.when we use perf to measure performance of a change like #59546
It would not surprise me if compilation took longer
since e.g. the compiler may be working harder to try to optimize the code in question
the more important thing for me to know about a case like that is "how did the object code performance change?"
(am I not looking at this problem the right way?)
Then https://github.com/anp/lolbench is for you
okay, and we should be striving to integrate that with
@rust-timer in some way?
(lolbench is pretty separate though so that might be long-term)