This is an archive of the discontinued LLVM Phabricator instance.

Implement get_local_size for AMDGPU (allows us to use get_global_id)
ClosedPublic

Authored by kzhuravl-AMD on Mar 18 2016, 2:53 PM.

Diff Detail

Repository
rL LLVM

Event Timeline

kzhuravl-AMD retitled this revision from to Implement get_local_size for AMDGPU (allows us to use get_global_id).
kzhuravl-AMD updated this object.
arsenm edited edge metadata.Mar 18 2016, 4:25 PM

We need to add a separate HSA ABI triple to the libclc build for this

kzhuravl-AMD edited edge metadata.

Separated into separate HSA ABI triple

arsenm added inline comments.Mar 29 2016, 12:20 AM
configure.py
106 ↗(On Diff #51826)

The SI devices should be removed here

Removed SI devices

kzhuravl-AMD marked an inline comment as done.Mar 29 2016, 10:29 AM
tstellarAMD accepted this revision.Apr 1 2016, 5:09 PM
tstellarAMD edited edge metadata.

LGTM.

This revision is now accepted and ready to land.Apr 1 2016, 5:09 PM
This revision was automatically updated to reflect the committed changes.