| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Revert "[ARM] Add more pattern matching for f16 <-> f64 conversions" | Bradley Smith | 2015-03-23 | 1 | -31/+0 |
| | | | | | | | | | | | This change is incorrect since it converts double rounding into single rounding, which can produce different results. Instead this optimization will be done by modifying Clang's codegen to not produce double rounding in the first place. This reverts commit r232954. llvm-svn: 232962 | ||||
| * | [ARM] Add more pattern matching for f16 <-> f64 conversions | Bradley Smith | 2015-03-23 | 1 | -0/+31 |
| 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 llvm-svn: 232954 | |||||
 |
index : bcm5719-llvm | |
| Project Ortega BCM5719 LLVM | Raptor Computing Systems |
| summaryrefslogtreecommitdiffstats |
|