diff options
| author | choikwa <5455710+choikwa@users.noreply.github.com> | 2025-05-19 14:11:31 -0400 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-05-19 14:11:31 -0400 |
| commit | 77de8a0c0abc9d245a7c6278670554b47ae183ea (patch) | |
| tree | b9bd5035e3e2e3b2b0d6c3870c2885d4e465f0cf /clang/lib/Serialization/ModuleFile.cpp | |
| parent | 7a242387c950c7060143da6da0e6fb91f36bb458 (diff) | |
| download | llvm-77de8a0c0abc9d245a7c6278670554b47ae183ea.zip llvm-77de8a0c0abc9d245a7c6278670554b47ae183ea.tar.gz llvm-77de8a0c0abc9d245a7c6278670554b47ae183ea.tar.bz2 | |
[AMDGPU][clang] provide device implementation for __builtin_logb and … (#129347)
…__builtin_scalbn
Clang generates library calls for __builtin_* functions which can be a
problem for GPUs that cannot handle them. This patch generates call to
device implementation for __builtin_logb and ldexp intrinsic for
__builtin_scalbn.
Diffstat (limited to 'clang/lib/Serialization/ModuleFile.cpp')
0 files changed, 0 insertions, 0 deletions