diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-10-12 13:21:50 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-10-12 13:21:50 +0000 |
| commit | 1b59a16c0b8bd89445b97eee0fde2a152001dbaf (patch) | |
| tree | 4a8db0f7b553bc9e38ed955915a44b71eb0df974 /llvm/test/Transforms/SLPVectorizer | |
| parent | ba229557dd208a39086c5e2bd8db619dca7d9b45 (diff) | |
| download | bcm5719-llvm-1b59a16c0b8bd89445b97eee0fde2a152001dbaf.tar.gz bcm5719-llvm-1b59a16c0b8bd89445b97eee0fde2a152001dbaf.zip | |
[CostModel][X86] Improve sum reduction costs.
I can't see any notable differences in costs between SSE2 and SSE42 arches for FADD/ADD reduction, so I've lowered the target to just SSE2.
I've also added vXi8 sum reduction costs in line with the PSADBW codegen and discussions on PR42674.
llvm-svn: 374655
Diffstat (limited to 'llvm/test/Transforms/SLPVectorizer')
| -rw-r--r-- | llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll b/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll index 4a69a35420b..34ceacac3f2 100644 --- a/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll +++ b/llvm/test/Transforms/SLPVectorizer/X86/remark_horcost.ll @@ -120,7 +120,7 @@ for.body: ; preds = %for.body, %entry ; YAML-NEXT: Function: foo ; YAML-NEXT: Args: ; YAML-NEXT: - String: 'Vectorized horizontal reduction with cost ' - ; YAML-NEXT: - Cost: '-2' + ; YAML-NEXT: - Cost: '-4' ; YAML-NEXT: - String: ' and with tree size ' ; YAML-NEXT: - TreeSize: '1' |
