Page MenuHomePhabricator

[ARM] Make v2f62 scalar_to_vector legal

Authored by dmgreen on Feb 5 2021, 9:32 AM.



Because we mark all operations as expand for v2f64, scalar_to_vector would end up lowering through a stack store/reload. But it is pretty simple to implement, only inserting a D reg into an undef vector. This helps clear up some inefficient codegen from soft calling conventions.

Diff Detail

Event Timeline

dmgreen created this revision.Feb 5 2021, 9:32 AM
dmgreen requested review of this revision.Feb 5 2021, 9:32 AM
Herald added a project: Restricted Project. · View Herald TranscriptFeb 5 2021, 9:32 AM
simon_tatham accepted this revision.Feb 8 2021, 1:15 AM

You've got a typo in the subject line :-) v2f64, not v2f62.

This revision is now accepted and ready to land.Feb 8 2021, 1:15 AM

Ah, yep. Thanks

This revision was automatically updated to reflect the committed changes.