| commit | de5be4657f8311b78bcea068c282c61351efcdfb | [log] [tgz] |
|---|---|---|
| author | Bradley Smith <bradley.smith@arm.com> | Mon Mar 23 15:59:54 2015 +0000 |
| committer | Bradley Smith <bradley.smith@arm.com> | Mon Mar 23 15:59:54 2015 +0000 |
| tree | 5580c835c93bee7d005a79243adcf6c689ab782b | |
| parent | cab2a096fdf27f1c3cf32ebc777e3560ac0dd0f5 [diff] |
[ARM] Add more pattern matching for f16 <-> f64 conversions Specifically when the conversion is done in two steps, f16 -> f32 -> f64. For example: %1 = tail call float @llvm.convert.from.fp16.f32(i16 %0) %conv = fpext float %1 to double to: vcvtb.f64.f16 git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232954 91177308-0d34-0410-b5e6-96231b3b80d8