diff options
author | Meredith Julian <35236176+mjulian31@users.noreply.github.com> | 2025-07-24 14:32:59 -0700 |
---|---|---|
committer | GitHub <noreply@github.com> | 2025-07-24 14:32:59 -0700 |
commit | be58069515e38d25a8e6ae5f1ef3b7b2e5eddbd1 (patch) | |
tree | eda93ea4797e070df2847351dcb17343cd475979 /llvm/unittests/Support/ProgramStackTest.cpp | |
parent | 581ba1cbf70bc5f89a095807c16f668a9b00ded9 (diff) | |
download | llvm-be58069515e38d25a8e6ae5f1ef3b7b2e5eddbd1.zip llvm-be58069515e38d25a8e6ae5f1ef3b7b2e5eddbd1.tar.gz llvm-be58069515e38d25a8e6ae5f1ef3b7b2e5eddbd1.tar.bz2 |
[LLVM][NVPTX] Upstream tanh intrinsic for libdevice (#149596)
Currently __nv_fast_tanhf() in libdevice maps to an nvvm intrinsic that
has not been upstreamed, which is causing issues when using the NVPTX
backend from upstream. Instead of upstreaming the intrinsic, we can
instead use the existing Intrinsic::tanh with the afn flag. This change
adds NVPTX backend support for ISD::TANH, adds auto-upgrade for the old
tanh_approx intrinsic to @llvm.tanh.f32 with afn flag so that libdevice
works properly upstream, and adds a basic codegen test and a case to the
auto-upgrade test.
Diffstat (limited to 'llvm/unittests/Support/ProgramStackTest.cpp')
0 files changed, 0 insertions, 0 deletions