[llvm] [AMDGPU] Gate fold_pow(x, ±0.5)->sqrt/rsqrt on nsz+ninf fast-math flags (PR #205592)

Jay Foad via llvm-commits llvm-commits at lists.llvm.org
Thu Jun 25 01:09:40 PDT 2026

Previous message: [llvm] [AMDGPU] Gate fold_pow(x, ±0.5)->sqrt/rsqrt on nsz+ninf fast-math flags (PR #205592)
Next message: [llvm] [AMDGPU] Gate fold_pow(x, ±0.5)->sqrt/rsqrt on nsz+ninf fast-math flags (PR #205592)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

jayfoad wrote:

> pow(-Inf, 0.5) == +Inf

TIL

https://github.com/llvm/llvm-project/pull/205592

Previous message: [llvm] [AMDGPU] Gate fold_pow(x, ±0.5)->sqrt/rsqrt on nsz+ninf fast-math flags (PR #205592)
Next message: [llvm] [AMDGPU] Gate fold_pow(x, ±0.5)->sqrt/rsqrt on nsz+ninf fast-math flags (PR #205592)
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

More information about the llvm-commits mailing list