[libclc] [libclc] Improve performance and precision of reciprocal functions (PR #203805)
Wenju He via cfe-commits
cfe-commits at lists.llvm.org
Sun Jun 14 17:50:36 PDT 2026
================
@@ -231,4 +231,5 @@ libclc_set_source_options(-fapprox-func
libclc_set_source_options(-cl-fp32-correctly-rounded-divide-sqrt
math/clc_div_cr.cl
math/clc_sqrt_cr.cl
+ math/clc_tanpi.cl
----------------
wenju-he wrote:
This change is not obvious from the commit message. Does __clc_tanpi use __clc_recip_fast?
https://github.com/llvm/llvm-project/pull/203805
More information about the cfe-commits
mailing list