-
-
Notifications
You must be signed in to change notification settings - Fork 1.9k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
perf: add new when-then-otherwise kernels #15089
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #15089 +/- ##
==========================================
+ Coverage 81.08% 81.17% +0.09%
==========================================
Files 1342 1346 +4
Lines 174150 174777 +627
Branches 2459 2459
==========================================
+ Hits 141210 141877 +667
+ Misses 32473 32434 -39
+ Partials 467 466 -1 ☔ View full report in Codecov by Sentry. |
b2adc5c
to
20f48e4
Compare
Fixes #15068.
This PR does not include the SIMD implementation. I'll do that in a follow-up PR, but we won't use the SIMD implementation for ARM anyway due to bad codegen for
SimdMask::select
on ARM, so I can already include benchmarks for ARM.On Apple M1 I got the following speedups (
mask%
is the fraction of elements inthen
vsotherwise
):