[Clang] VectorExprEvaluator::VisitCallExpr / InterpretBuiltin - add MMX/SSE/AVX/AVX512 VPSADBW / VMPSADBW / VDBPSADBW intrinsics to be used in constexpr

[RKSimon]

The horizontal sum of difference intrinsics can all be used in constexpr with suitable handling of the builtins inside VectorExprEvaluator::VisitCallExpr and InterBuiltin.cpp, and testing in the relevent *-builtins.c files.

_mm_sad_epu8 _mm256_sad_epu8 _mm512_sad_epu8 _mm_sad_pu8

_mm_mpsadbw_epu8 _mm256_mpsadbw_epu8

_mm_dbsad_epu8 _mm_mask_dbsad_epu8 _mm_maskz_dbsad_epu8
_mm256_dbsad_epu8 _mm256_mask_dbsad_epu8 _mm256_maskz_dbsad_epu8
_mm512_dbsad_epu8 _mm512_mask_dbsad_epu8 _mm512_maskz_dbsad_epu8

These intrinsics can be tricky to correctly evaluate, so I'd recommend previous knowledge of SSE/AVX instructions is required: VPSADBW is pretty straightfoward, but VMPSADBW / VDBPSADBW is messy.

Care needs to be taken with the signed/unsigned extension of the arguments - we might need to adjust the builtin signatures.

[llvmbot]

@llvm/issue-subscribers-backend-x86

Author: Simon Pilgrim (RKSimon)

The horizontal sum of difference intrinsics can all be used in constexpr with suitable handling of the builtins inside VectorExprEvaluator::VisitCallExpr and InterBuiltin.cpp, and testing in the relevent *-builtins.c files. ``` _mm_sad_epu8 _mm256_sad_epu8 _mm512_sad_epu8 _mm_sad_pu8

_mm_mpsadbw_epu8 _mm256_mpsadbw_epu8

_mm_dbsad_epu8 _mm_mask_dbsad_epu8 _mm_maskz_dbsad_epu8
_mm256_dbsad_epu8 _mm256_mask_dbsad_epu8 _mm256_maskz_dbsad_epu8
_mm512_dbsad_epu8 _mm512_mask_dbsad_epu8 _mm512_maskz_dbsad_epu8

These intrinsics can be tricky to correctly evaluate, so I'd recommend previous knowledge of SSE/AVX instructions is required: VPSADBW is pretty straightfoward, but VMPSADBW / VDBPSADBW is messy.

Care needs to be taken with the signed/unsigned extension of the arguments - we might need to adjust the builtin signatures.
</details>

[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-clang-frontend

Author: Simon Pilgrim (RKSimon)

The horizontal sum of difference intrinsics can all be used in constexpr with suitable handling of the builtins inside VectorExprEvaluator::VisitCallExpr and InterBuiltin.cpp, and testing in the relevent *-builtins.c files. ``` _mm_sad_epu8 _mm256_sad_epu8 _mm512_sad_epu8 _mm_sad_pu8

_mm_mpsadbw_epu8 _mm256_mpsadbw_epu8

_mm_dbsad_epu8 _mm_mask_dbsad_epu8 _mm_maskz_dbsad_epu8
_mm256_dbsad_epu8 _mm256_mask_dbsad_epu8 _mm256_maskz_dbsad_epu8
_mm512_dbsad_epu8 _mm512_mask_dbsad_epu8 _mm512_maskz_dbsad_epu8

These intrinsics can be tricky to correctly evaluate, so I'd recommend previous knowledge of SSE/AVX instructions is required: VPSADBW is pretty straightfoward, but VMPSADBW / VDBPSADBW is messy.

Care needs to be taken with the signed/unsigned extension of the arguments - we might need to adjust the builtin signatures.
</details>

[llvmbot]

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

Author: Simon Pilgrim (RKSimon)

The horizontal sum of difference intrinsics can all be used in constexpr with suitable handling of the builtins inside VectorExprEvaluator::VisitCallExpr and InterBuiltin.cpp, and testing in the relevent *-builtins.c files. ``` _mm_sad_epu8 _mm256_sad_epu8 _mm512_sad_epu8 _mm_sad_pu8

_mm_mpsadbw_epu8 _mm256_mpsadbw_epu8

_mm_dbsad_epu8 _mm_mask_dbsad_epu8 _mm_maskz_dbsad_epu8
_mm256_dbsad_epu8 _mm256_mask_dbsad_epu8 _mm256_maskz_dbsad_epu8
_mm512_dbsad_epu8 _mm512_mask_dbsad_epu8 _mm512_maskz_dbsad_epu8

These intrinsics can be tricky to correctly evaluate, so I'd recommend previous knowledge of SSE/AVX instructions is required: VPSADBW is pretty straightfoward, but VMPSADBW / VDBPSADBW is messy.

Care needs to be taken with the signed/unsigned extension of the arguments - we might need to adjust the builtin signatures.
</details>

[Mohxen]

Hey Simon @RKSimon, I would like to work on this.

[RKSimon]

ping @Mohxen ?

[Mohxen]

Hi @RKSimon,

Sorry for the delay — I missed this thread.

I’ve started implementing constexpr support for VPSADBW (__builtin_ia32_psadbw128) and wanted to check whether my approach is on the right track. If there are similar threads or prior discussions that would give me a better understanding, please let me know.
What I’ve done so far

Added EvaluatePSADBW128 in ExprConstant.cpp to compute the two SAD blocks.
Added the corresponding builtin case in VectorExprEvaluator::VisitCallExpr.
Added a test under clang/test/AST/ByteCode/x86-psadbw-psadbw128.cpp.

Thanks for any hints or guidance!