Searched refs:FMLA (Results 1 – 6 of 6) sorted by relevance
| /llvm-project-15.0.7/llvm/test/CodeGen/AArch64/ |
| H A D | machine-combiner-instr-fmf.mir | 203 # Can create FMLA, because both the fmul and fadd have all fast-math flags. 242 # Can create FMLA, because both the fmul and fadd have the contract fast-math flag. 281 # Do not create FMLA, because we don't have the contract flag on the FADD. 321 # Do create FMLA, because we have the contract flag on the FADD. 361 # Do not create FMLA, as nsz flag does not allow it.
|
| H A D | sve-intrinsics-fp-arith.ll | 522 ; FMLA 562 ; FMLA (Indexed)
|
| /llvm-project-15.0.7/llvm/lib/Target/AArch64/ |
| H A D | AArch64TargetTransformInfo.cpp | 1001 auto FMLA = Builder.CreateIntrinsic(Intrinsic::aarch64_sve_fmla, in instCombineSVEVectorFMLA() local 1003 FMLA->setFastMathFlags(FAddFlags); in instCombineSVEVectorFMLA() 1004 return IC.replaceInstUsesWith(II, FMLA); in instCombineSVEVectorFMLA() 1100 if (auto FMLA = instCombineSVEVectorFMLA(IC, II)) in instCombineSVEVectorFAdd() local 1101 return FMLA; in instCombineSVEVectorFAdd()
|
| H A D | AArch64InstrInfo.td | 4590 // NOTE: The operands of the PatFrag are reordered on FMLA/FMLS because the 4592 defm FMLA : SIMDThreeSameVectorFPTied<0, 0, 0b001, "fmla", 6366 defm FMLA : SIMDFPIndexedTied<0, 0b0001, "fmla">; 6370 // NOTE: Operands are reordered in the FMLA/FMLS PatFrags because the 6375 defm : SIMDFPIndexedTiedPatterns<"FMLA", 6377 defm : SIMDFPIndexedTiedPatterns<"FMLA",
|
| /llvm-project-15.0.7/llvm/test/tools/llvm-mca/AArch64/Cortex/ |
| H A D | forwarding-A57.s | 284 # FMUL, FMULX, FMLA, FMLS are impacted
|
| /llvm-project-15.0.7/clang/include/clang/Basic/ |
| H A D | arm_neon.td | 782 def FMLA : SInst<"vfma", "....", "dQd">;
|