[PATCH] D101844: [MicroBenchmarks] Add initial loop vectorization benchmarks.
Roman Lebedev via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Thu May 13 09:47:46 PDT 2021
lebedev.ri added a comment.
In D101844#2757052 <https://reviews.llvm.org/D101844#2757052>, @Meinersbur wrote:
> As said, we already had discussions about the point of limiting execution time. I was arguing that benchmarks for L3 memory effect have to be sufficiently large. This isn't even the case here. Why not just decrease `N`?
That isn't going to affect the total runtime of this executable.
> Whether each single math function counts as its own benchmark doesn't make that much of a difference here.
>
>> I can disable the benchmarking part if [...] `TEST_SUITE_BENCHMARKING_ONLY`
>
> This was what I am suggesting.
Repository:
rT test-suite
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D101844/new/
https://reviews.llvm.org/D101844
More information about the llvm-commits
mailing list