[Clang] VectorExprEvaluator::VisitCallExpr / InterpretBuiltin - allow SSE/AVX/AVX512 square root intrinsics to be used in constexpr

[RKSimon]

_mm_sqrt_pd / _mm_mask_sqrt_pd / _mm_maskz_sqrt_pd
_mm_sqrt_ps / _mm_mask_sqrt_ps / _mm_maskz_sqrt_ps
_mm256_sqrt_pd /_mm256_mask_sqrt_pd / _mm256_maskz_sqrt_pd
_mm256_sqrt_ps /_mm256_mask_sqrt_ps / _mm256_maskz_sqrt_ps
_mm512_sqrt_pd / _mm512_mask_sqrt_pd / _mm512_maskz_sqrt_pd
_mm512_sqrt_ps / _mm512_mask_sqrt_ps / _mm512_maskz_sqrt_ps

These should probably be limited to non-constrained fp cases

NOTE: We are NOT intending to handle the related rcp and rsqrt estimate intrinsics

[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-backend-x86

Author: Simon Pilgrim (RKSimon)

``` _mm_sqrt_pd / _mm_mask_sqrt_pd / _mm_maskz_sqrt_pd _mm_sqrt_ps / _mm_mask_sqrt_ps / _mm_maskz_sqrt_ps _mm256_sqrt_pd /_mm256_mask_sqrt_pd / _mm256_maskz_sqrt_pd _mm256_sqrt_ps /_mm256_mask_sqrt_ps / _mm256_maskz_sqrt_ps _mm512_sqrt_pd / _mm512_mask_sqrt_pd / _mm512_maskz_sqrt_pd _mm512_sqrt_ps / _mm512_mask_sqrt_ps / _mm512_maskz_sqrt_ps ``` These should probably be limited to non-constrained fp cases

NOTE: We are NOT intending to handle the related rcp and rsqrt estimate intrinsics

[llvmbot]

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

Author: Simon Pilgrim (RKSimon)

``` _mm_sqrt_pd / _mm_mask_sqrt_pd / _mm_maskz_sqrt_pd _mm_sqrt_ps / _mm_mask_sqrt_ps / _mm_maskz_sqrt_ps _mm256_sqrt_pd /_mm256_mask_sqrt_pd / _mm256_maskz_sqrt_pd _mm256_sqrt_ps /_mm256_mask_sqrt_ps / _mm256_maskz_sqrt_ps _mm512_sqrt_pd / _mm512_mask_sqrt_pd / _mm512_maskz_sqrt_pd _mm512_sqrt_ps / _mm512_mask_sqrt_ps / _mm512_maskz_sqrt_ps ``` These should probably be limited to non-constrained fp cases

NOTE: We are NOT intending to handle the related rcp and rsqrt estimate intrinsics

[bojle]

Hi, @RKSimon I would like to work on this one

[bojle]

I am almost done with this #161547

However, I am stumbling on this issue: The intrinsics manual mentions _mm512_sqrt_pd to have one argument.

However, BuiltinsX86.td describes sqrtpd512 to have two args, the vector and an int. Are there any references to this? It breaks my code that expects total arguments to all sqrt variants to be 1.

@RKSimon

[RKSimon]

AVX512 _mm512_sqrt_pd shares the same builtin with _mm512_sqrt_round_pd (etc.) - you will have to check for the additional argument and ensure that it is set to 4 (== _MM_FROUND_CUR_DIRECTION) - potentially we could add support for AVX512 rounding modes in the future, but tbh I see it as a stretch goal at best.