This is an archive of the discontinued LLVM Phabricator instance.

[AMDGPU] Assembler: Fix parsing error with floating-point literals passed to integer instructions
ClosedPublic

Authored by SamWot on Jul 4 2016, 6:43 AM.

Details

Summary

Previous changes for sext() modifier introduced regression: when you pass fp literal to integer instruction assembler was returning error. Now it just converts that float to binary representation.

Diff Detail

Event Timeline

SamWot updated this revision to Diff 62676.Jul 4 2016, 6:43 AM
SamWot retitled this revision from to [AMDGPU] Assembler: Fix parsing error with floating-point literals passed to integer instructions.
SamWot updated this object.
SamWot added a reviewer: vpykhtin.
SamWot updated this revision to Diff 62677.Jul 4 2016, 6:49 AM
SamWot edited edge metadata.

Removed strange X86 changes from previous diff.

SamWot added a project: Restricted Project.Jul 4 2016, 6:50 AM
vpykhtin accepted this revision.Jul 5 2016, 6:44 AM
vpykhtin edited edge metadata.

LGTM.

This revision is now accepted and ready to land.Jul 5 2016, 6:44 AM
This revision was automatically updated to reflect the committed changes.