This is an archive of the discontinued LLVM Phabricator instance.

[CUDA] Added --[no-]cuda-include-ptx=sm_XX|all option.
ClosedPublic

Authored by tra on Apr 10 2018, 10:55 AM.

Details

Summary

Currently we always include PTX into the fatbin along
with the GPU code. It about doubles the size of the GPU binary
we need to carry in the executable. These options allow control
inclusion of PTX into GPU binary.

This patch does not change the defaults, though we may consider
making no-PTX the default in the future.

Diff Detail

Repository
rC Clang

Event Timeline

tra created this revision.Apr 10 2018, 10:55 AM
jlebar accepted this revision.Apr 10 2018, 11:06 AM

Where do we document the default values for these flags? Like, how is a user supposed to figure out whether the default is to include ptx or not?

This revision is now accepted and ready to land.Apr 10 2018, 11:06 AM
tra updated this revision to Diff 141882.Apr 10 2018, 11:16 AM

Documented new options in ClangCommandLineReference.rst

tra added a comment.Apr 10 2018, 11:18 AM

Where do we document the default values for these flags? Like, how is a user supposed to figure out whether the default is to include ptx or not?

I've updated docs/ClangCommandLineReference.rst.

This revision was automatically updated to reflect the committed changes.