Stream: t-compiler/wg-parallel-rustc

Topic: custom perf script


simulacrum (Dec 03 2019 at 21:30, on Zulip):

@bjorn3 actually -- one recommendation is to do perf script --no-inline which is a massive speedup usually

simulacrum (Dec 03 2019 at 21:40, on Zulip):

I think that might get you most of the gains

bjorn3 (Dec 04 2019 at 10:22, on Zulip):

I will try that @simulacrum

bjorn3 (Dec 16 2019 at 20:27, on Zulip):

@simulacrum using --no-inline worked perfectly. Now it only takes a few seconds. I wish there was a way to not use it, but instead limit the output to samples containing a certain frame (only my code instead of all of rustc) though.

Last update: Jan 28 2020 at 00:35UTC