[PATCH] D103144: [X86][Costmodel] Load/store v2i16 VF=2 interleaving costs
Roman Lebedev via Phabricator via llvm-commits
llvm-commits at lists.llvm.org
Wed May 26 12:54:39 PDT 2021
lebedev.ri added a comment.
And one more example: store VF16 stride 6:
- haswell `48` https://godbolt.org/z/jfb1rss8E
- broadwell `48` https://godbolt.org/z/v73sTPrax
- skylake `48` `https://godbolt.org/z/fn8Y9334x`
- zen1 `20` https://godbolt.org/z/rnTvzYnTr (we seem to round 0.5 down?)
- zen2 `20` https://godbolt.org/z/effcqPfaP (we seem to round 0.5 down?)
- zen3 `20` https://godbolt.org/z/1fTjhf3To
Thus, `48`.
Sidenote, i'm still unsure whether we should specify that Zen 3 has `FeatureFastVariableShuffle`.
Byte-wise variable shuffle is fast, but some are ucoded.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D103144/new/
https://reviews.llvm.org/D103144
More information about the llvm-commits
mailing list