[PATCH] D87169: SVML support for log10, sqrt
Alexander Yermolovich via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed Sep 23 14:38:04 PDT 2020
ayermolo added inline comments.
================
Comment at: llvm/include/llvm/Analysis/VecFuncs.def:312
+
+TLI_DEFINE_VECFUNC("llvm.sqrt.f64", "__svml_sqrt2", 2)
+TLI_DEFINE_VECFUNC("llvm.sqrt.f64", "__svml_sqrt4", 4)
----------------
hoy wrote:
> craig.topper wrote:
> > I don't think we want to use SVML for llvm.sqrt. We have a hardware instruction for that.
> Good point. We have the SQRTPS/SQRTPD AVX instructions for that.
You are right: https://software.intel.com/sites/landingpage/IntrinsicsGuide/#text=_mm_sqrt_p&expand=5824,5823,5356,5823,5825,5365,5356
Was going to submit a fix up patch later this week.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D87169/new/
https://reviews.llvm.org/D87169
More information about the llvm-commits
mailing list