diff options
| author | David Sherwood <david.sherwood@arm.com> | 2021-04-06 11:06:58 +0100 |
|---|---|---|
| committer | David Sherwood <david.sherwood@arm.com> | 2021-04-21 09:39:45 +0100 |
| commit | 57ca65e21e9a48040e929a506a259ee2e4b42088 (patch) | |
| tree | b0561fa86726456243f4cd3db5620a5e5f19a401 /flang/lib/Frontend/CompilerInvocation.cpp | |
| parent | cf61cf07248a0c395b63453876f4a12d21148c61 (diff) | |
| download | llvm-57ca65e21e9a48040e929a506a259ee2e4b42088.zip llvm-57ca65e21e9a48040e929a506a259ee2e4b42088.tar.gz llvm-57ca65e21e9a48040e929a506a259ee2e4b42088.tar.bz2 | |
[AArch64] Add instruction costs for FP_TO_UINT and FP_TO_SINT with half types
We were missing some instruction costs when converting vectors of
floating point half types into integers, so I've added those here.
I also manually generated assembly code for each FP->int case and
looked at the number of instructions generated, which meant
adjusting some of the existing costs too.
I've updated an existing test to reflect the new costs:
Analysis/CostModel/AArch64/sve-fptoi.ll
Differential Revision: https://reviews.llvm.org/D99935
Diffstat (limited to 'flang/lib/Frontend/CompilerInvocation.cpp')
0 files changed, 0 insertions, 0 deletions