llvm-project/llvm/test/Transforms/CodeGenPrepare/AArch64
Florian Hahn 786c687810
[AArch64] Add support for FMA intrinsics to shouldSinkOperands.
If the fma operates on a legal vector type, the indexed variants can be
used, if the second operand is a splat of a valid index.

Reviewed By: dmgreen

Differential Revision: https://reviews.llvm.org/D126234
2022-05-27 10:37:03 +01:00
..
combine-address-mode.ll [CGP] Fix the crash for combining address mode when having cyclic dependency 2021-08-26 22:52:42 +08:00
free-zext.ll
gather-scatter-opt-inseltpoison.ll
gather-scatter-opt.ll
large-offset-gep.ll [AArch64] Async unwind - function epilogues 2022-04-12 16:50:50 +01:00
lit.local.cfg
overflow-intrinsics.ll
sink-free-instructions-inseltpoison.ll [CodeGenPrepare] The instruction to be sunk should be inserted before its user in a block 2021-08-17 18:58:15 +08:00
sink-free-instructions.ll [AArch64] Add support for FMA intrinsics to shouldSinkOperands. 2022-05-27 10:37:03 +01:00
trunc-weird-user.ll
widen_switch.ll
zext-to-shuffle.ll [AArch64] Add test cases where zext can be lowered to series of tbl. 2022-02-25 15:36:32 +00:00