Scott McMurray 9aa9a846b6 Allow transmutes to produce OperandValues instead of always using allocas ...

LLVM can usually optimize these away, but especially for things like transmutes of newtypes it's silly to generate the `alloc`+`store`+`load` at all when it's actually a nop at LLVM level.

2023-04-04 18:44:29 -07:00

..

simd-intrinsic-float-abs.rs

…

simd-intrinsic-float-ceil.rs

…

simd-intrinsic-float-cos.rs

…

simd-intrinsic-float-exp2.rs

…

simd-intrinsic-float-exp.rs

…

simd-intrinsic-float-floor.rs

…

simd-intrinsic-float-fma.rs

…

simd-intrinsic-float-fsqrt.rs

…

simd-intrinsic-float-log2.rs

…

simd-intrinsic-float-log10.rs

…

simd-intrinsic-float-log.rs

…

simd-intrinsic-float-minmax.rs

…

simd-intrinsic-float-pow.rs

…

simd-intrinsic-float-powi.rs

…

simd-intrinsic-float-sin.rs

…

simd-intrinsic-generic-arithmetic-saturating.rs

Fix codegen tests

2023-02-14 19:21:58 -05:00

simd-intrinsic-generic-bitmask.rs

Try to fix codegen tests for ??? LLVM 14 ???

2023-02-14 19:49:49 -05:00

simd-intrinsic-generic-extract-insert.rs

Try to fix codegen tests for ??? LLVM 14 ???

2023-02-14 19:49:49 -05:00

simd-intrinsic-generic-gather.rs

…

simd-intrinsic-generic-scatter.rs

…

simd-intrinsic-generic-select.rs

…

simd-intrinsic-transmute-array.rs

Allow transmutes to produce OperandValues instead of always using allocas

2023-04-04 18:44:29 -07:00