[X86] Recognise VPMADD52L/H patterns on AVX512IFMA/AVXIFMA targets

[RKSimon]

AVX512IFMA/AVXIFMA targets should be able to fold vXi64 mul-add patterns assuming the multiplicands are known to be u52 integers

X86ISD::VPMADD52L patterns can be matched for (X * Y) + Z if the upper 12 bits of X, Y AND (X * Y) are all known zero

X86ISD::VPMADD52H patterns are trickier as we extract the upper 52 bits of the (X * Y) 104-bit mulu extended result, but at least basic matching might be achievable

[llvmbot]

@llvm/issue-subscribers-backend-x86

Author: Simon Pilgrim (RKSimon)

AVX512IFMA/AVXIFMA targets should be able to fold vXi64 mul-add patterns assuming the multiplicands are known to be u52 integers

X86ISD::VPMADD52L patterns can be matched for (X * Y) + Z if the upper 12 bits of X, Y AND (X * Y) are all known zero

X86ISD::VPMADD52H patterns are trickier as we extract the upper 52 bits of the (X * Y) 104-bit mulu extended result, but at least basic matching might be achievable

[llvmbot]

Hi!

This issue may be a good introductory issue for people new to working on LLVM. If you would like to work on this issue, your first steps are:

1. Check that no other contributor is working on this issue. If someone is assigned to the issue or claimed to be working on it, ping the person. After one week without a response, the assignee may be changed.
2. Leave a comment indicating that you are working on the issue, or just create a pull request after following the steps below. Mention this issue in the description of the pull request.
3. Fix the issue locally.
4. Run the test suite locally. Remember that the subdirectories under test/ create fine-grained testing targets, so you can e.g. use make check-clang-ast to only run Clang's AST tests.
5. Create a Git commit.
6. Run git clang-format HEAD~1 to format your changes.
7. Open a pull request to the upstream repository on GitHub. Detailed instructions can be found in GitHub's documentation. Mention this issue in the description of the pull request.

If you have any further questions about this issue, don't hesitate to ask via a comment in the thread below.

[llvmbot]

@llvm/issue-subscribers-good-first-issue

Author: Simon Pilgrim (RKSimon)

AVX512IFMA/AVXIFMA targets should be able to fold vXi64 mul-add patterns assuming the multiplicands are known to be u52 integers

X86ISD::VPMADD52L patterns can be matched for (X * Y) + Z if the upper 12 bits of X, Y AND (X * Y) are all known zero

X86ISD::VPMADD52H patterns are trickier as we extract the upper 52 bits of the (X * Y) 104-bit mulu extended result, but at least basic matching might be achievable

[RKSimon]

CC @XChy

[RKSimon]

It might be worth using VPMADD52L for vXi64 multiplies if value tracking says we can - on Intel targets it is MUCH faster than VPMULLQ

[Arghnews]

@RKSimon I'd be interested in having a go at this, but I'm unclear considering the linked commits how much work - if any - is left to do?

[RKSimon]

Actual matching hasn't been attempted yet - everything has purely been about value tracking / demanded bits so far.

A basic matcher for VPMADD52L in combineAdd should be relatively quick and easy.