[PATCH] D112873: [X86] `X86TTIImpl::getInterleavedMemoryOpCostAVX512()`: fallback to scalarization cost computation for mask

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri created this revision.
lebedev.ri added reviewers: RKSimon, pengfei, dorit, Ayal, hsaito, fhahn.
lebedev.ri added a project: LLVM.
Herald added a subscriber: hiraditya.
lebedev.ri requested review of this revision.

I don't really buy that masked interleaved memory loads/stores are supported on X86.
There is zero costmodel test coverage, no actual cost modelling for the generation
of the mask repetition, and basically only two LV tests.
Additionally, i'm not very interested in AVX512.

I don't know if this really helps "soft" block over at
https://reviews.llvm.org/D111460#inline-1075467,
but i think it can't make things worse at least.

When we are being told that there is a masking, instead of
completely giving up and falling back to
fully scalarizing `BasicTTIImplBase::getInterleavedMemoryOpCost()`,
let's correctly query the cost of masked memory ops,
keep all the pretty shuffle cost modelling,
but scalarize the cost computation for the mask replication.

I think, not scalarizing the shuffles themselves
may adjust the computed costs a bit,
and maybe hopefully just enough to hide the "regressions"
at https://reviews.llvm.org/D111460#inline-1075467
I do mean hide, because the test coverage is non-existent.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D112873

Files:
  llvm/lib/Target/X86/X86TargetTransformInfo.cpp

-------------- next part --------------
A non-text attachment was scrubbed...
Name: D112873.383609.patch
Type: text/x-patch
Size: 5572 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20211030/734e46d9/attachment.bin>