[llvm-dev] RFC: EfficiencySanitizer

Renato Golin via llvm-dev llvm-dev at lists.llvm.org
Thu Apr 21 09:32:30 PDT 2016

On 21 April 2016 at 00:50, Derek Bruening <bruening at google.com> wrote:
> I'm not sure I understand the question: are you asking whether not gathering
> data on time spent in the kernel is an issue?  Or you're asking how to
> measure aspects of performance without using sampling or hardware
> performance counters?

The latter. I mean, counting hits with mcount is easy, but other
information like cache miss needs kernel support, no?

> There is no cache miss information here: we are not using hardware perf
> counters nor running a software cache simulation.  We are measuring
> particular aspects of application behavior that tend to affect performance,
> often abstracted away from the precise microarchitecture you're running on.

Ah, right, I think that explains my confusion.


More information about the llvm-dev mailing list