Skip to content

[Clang] VectorExprEvaluator::VisitCallExpr - add MMX/SSE/AVX/AVX512 PMULHRSW intrinsics to be used in constexpr #155805

@RKSimon

Description

@RKSimon

Similar to #152524 - add handling for __builtin_ia32_pmulhrs to VectorExprEvaluator::VisitCallExpr and InterBuiltin.cpp

_mm_mulhrs_pi16
mm_mulhrs_epi16 mm256_mulhrs_epi16 mm512_mulhrs_epi16
mm_mask_mulhrs_epi16 mm256_mask_mulhrs_epi16 mm512_mask_mulhrs_epi16
mm_maskz_mulhrs_epi16 mm256_maskz_mulhrs_epi16 mm512_maskz_mulhrs_epi16

Further info on how to constant fold PMULHRSW can be found in simplifyX86pmulh in X86InstCombineIntrinsic.cpp

Metadata

Metadata

Assignees

Labels

backend:X86clang:frontendLanguage frontend issues, e.g. anything involving "Sema"clang:headersHeaders provided by Clang, e.g. for intrinsicsconstexprAnything related to constant evaluationgood first issuehttps://github.com/llvm/llvm-project/contribute

Type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions