This is an archive of the discontinued LLVM Phabricator instance.

[AMDGPU][MC][GFX11] Correct encoding of neg modifier for v_dot2_f32_bf16
ClosedPublic

Authored by dp on Dec 21 2022, 3:14 AM.

Details

Summary

Fix a bug with neg_lo:[0,1,0] and neg_hi:[0,1,0] modifiers - they are accepted but not encoded.

Diff Detail

Event Timeline

dp created this revision.Dec 21 2022, 3:14 AM
Herald added a project: Restricted Project. · View Herald TranscriptDec 21 2022, 3:14 AM
dp requested review of this revision.Dec 21 2022, 3:14 AM
Herald added a project: Restricted Project. · View Herald TranscriptDec 21 2022, 3:14 AM
This revision is now accepted and ready to land.Dec 29 2022, 11:49 AM