; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=riscv32 -mattr=+d -verify-machineinstrs < %s \
; RUN: -target-abi=ilp32d | FileCheck -check-prefixes=CHECKIFD,RV32IFD %s
; RUN: llc -mtriple=riscv64 -mattr=+d -verify-machineinstrs < %s \
; RUN: -target-abi=lp64d | FileCheck -check-prefixes=CHECKIFD,RV64IFD %s
; RUN: llc -mtriple=riscv32 -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefix=RV32I %s
; RUN: llc -mtriple=riscv64 -verify-machineinstrs < %s \
; RUN: | FileCheck -check-prefix=RV64I %s
; These tests are each targeted at a particular RISC-V FPU instruction.
; Compares and conversions can be found in double-fcmp.ll and double-convert.ll
; respectively. Some other double-*.ll files in this folder exercise LLVM IR
; instructions that don't directly match a RISC-V instruction.
define double @fadd_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fadd_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fadd.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fadd_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fadd_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fadd double %a, %b
ret double %1
}
define double @fsub_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fsub_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fsub.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fsub_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call __subdf3@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fsub_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call __subdf3@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fsub double %a, %b
ret double %1
}
define double @fmul_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fmul_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fmul.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fmul_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call __muldf3@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmul_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call __muldf3@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fmul double %a, %b
ret double %1
}
define double @fdiv_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fdiv_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fdiv.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fdiv_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call __divdf3@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fdiv_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call __divdf3@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fdiv double %a, %b
ret double %1
}
declare double @llvm.sqrt.f64(double)
define double @fsqrt_d(double %a) nounwind {
; CHECKIFD-LABEL: fsqrt_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fsqrt.d fa0, fa0
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fsqrt_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call sqrt@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fsqrt_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call sqrt@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = call double @llvm.sqrt.f64(double %a)
ret double %1
}
declare double @llvm.copysign.f64(double, double)
define double @fsgnj_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fsgnj_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fsgnj.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fsgnj_d:
; RV32I: # %bb.0:
; RV32I-NEXT: lui a2, 524288
; RV32I-NEXT: and a2, a3, a2
; RV32I-NEXT: slli a1, a1, 1
; RV32I-NEXT: srli a1, a1, 1
; RV32I-NEXT: or a1, a1, a2
; RV32I-NEXT: ret
;
; RV64I-LABEL: fsgnj_d:
; RV64I: # %bb.0:
; RV64I-NEXT: srli a1, a1, 63
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: slli a0, a0, 1
; RV64I-NEXT: srli a0, a0, 1
; RV64I-NEXT: or a0, a0, a1
; RV64I-NEXT: ret
%1 = call double @llvm.copysign.f64(double %a, double %b)
ret double %1
}
; This function performs extra work to ensure that
; DAGCombiner::visitBITCAST doesn't replace the fneg with an xor.
define i32 @fneg_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fneg_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fadd.d ft0, fa0, fa0
; CHECKIFD-NEXT: fneg.d ft1, ft0
; CHECKIFD-NEXT: feq.d a0, ft0, ft1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fneg_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv a2, a0
; RV32I-NEXT: mv a3, a1
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: lui a2, 524288
; RV32I-NEXT: xor a3, a1, a2
; RV32I-NEXT: mv a2, a0
; RV32I-NEXT: call __eqdf2@plt
; RV32I-NEXT: seqz a0, a0
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fneg_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv a1, a0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a1, a0, a1
; RV64I-NEXT: call __eqdf2@plt
; RV64I-NEXT: seqz a0, a0
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fadd double %a, %a
%2 = fneg double %1
%3 = fcmp oeq double %1, %2
%4 = zext i1 %3 to i32
ret i32 %4
}
define double @fsgnjn_d(double %a, double %b) nounwind {
; TODO: fsgnjn.s isn't selected on RV64 because DAGCombiner::visitBITCAST will
; convert (bitconvert (fneg x)) to a xor.
;
; CHECKIFD-LABEL: fsgnjn_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fsgnjn.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fsgnjn_d:
; RV32I: # %bb.0:
; RV32I-NEXT: not a2, a3
; RV32I-NEXT: lui a3, 524288
; RV32I-NEXT: and a2, a2, a3
; RV32I-NEXT: slli a1, a1, 1
; RV32I-NEXT: srli a1, a1, 1
; RV32I-NEXT: or a1, a1, a2
; RV32I-NEXT: ret
;
; RV64I-LABEL: fsgnjn_d:
; RV64I: # %bb.0:
; RV64I-NEXT: not a1, a1
; RV64I-NEXT: slli a0, a0, 1
; RV64I-NEXT: srli a0, a0, 1
; RV64I-NEXT: srli a1, a1, 63
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: or a0, a0, a1
; RV64I-NEXT: ret
%1 = fsub double -0.0, %b
%2 = call double @llvm.copysign.f64(double %a, double %1)
ret double %2
}
declare double @llvm.fabs.f64(double)
; This function performs extra work to ensure that
; DAGCombiner::visitBITCAST doesn't replace the fabs with an and.
define double @fabs_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fabs_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fadd.d ft0, fa0, fa1
; CHECKIFD-NEXT: fabs.d ft1, ft0
; CHECKIFD-NEXT: fadd.d fa0, ft1, ft0
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fabs_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv a3, a1
; RV32I-NEXT: slli a1, a1, 1
; RV32I-NEXT: srli a1, a1, 1
; RV32I-NEXT: mv a2, a0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fabs_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv a1, a0
; RV64I-NEXT: slli a0, a0, 1
; RV64I-NEXT: srli a0, a0, 1
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fadd double %a, %b
%2 = call double @llvm.fabs.f64(double %1)
%3 = fadd double %2, %1
ret double %3
}
declare double @llvm.minnum.f64(double, double)
define double @fmin_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fmin_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fmin.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fmin_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call fmin@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmin_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call fmin@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = call double @llvm.minnum.f64(double %a, double %b)
ret double %1
}
declare double @llvm.maxnum.f64(double, double)
define double @fmax_d(double %a, double %b) nounwind {
; CHECKIFD-LABEL: fmax_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fmax.d fa0, fa0, fa1
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fmax_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call fmax@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmax_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call fmax@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = call double @llvm.maxnum.f64(double %a, double %b)
ret double %1
}
declare double @llvm.fma.f64(double, double, double)
define double @fmadd_d(double %a, double %b, double %c) nounwind {
; CHECKIFD-LABEL: fmadd_d:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fmadd.d fa0, fa0, fa1, fa2
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fmadd_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmadd_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = call double @llvm.fma.f64(double %a, double %b, double %c)
ret double %1
}
define double @fmsub_d(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fmsub_d:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft0, fa2, ft0
; RV32IFD-NEXT: fmsub.d fa0, fa0, fa1, ft0
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fmsub_d:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft0, fa2, ft0
; RV64IFD-NEXT: fmsub.d fa0, fa0, fa1, ft0
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fmsub_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a3
; RV32I-NEXT: mv s1, a2
; RV32I-NEXT: mv s2, a1
; RV32I-NEXT: mv s3, a0
; RV32I-NEXT: mv a0, a4
; RV32I-NEXT: mv a1, a5
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv a4, a0
; RV32I-NEXT: lui a0, 524288
; RV32I-NEXT: xor a5, a1, a0
; RV32I-NEXT: mv a0, s3
; RV32I-NEXT: mv a1, s2
; RV32I-NEXT: mv a2, s1
; RV32I-NEXT: mv a3, s0
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmsub_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a1
; RV64I-NEXT: mv s1, a0
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a2, a0, a1
; RV64I-NEXT: mv a0, s1
; RV64I-NEXT: mv a1, s0
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%c_ = fadd double 0.0, %c ; avoid negation using xor
%negc = fsub double -0.0, %c_
%1 = call double @llvm.fma.f64(double %a, double %b, double %negc)
ret double %1
}
define double @fnmadd_d(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fnmadd_d:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft1, fa0, ft0
; RV32IFD-NEXT: fadd.d ft0, fa2, ft0
; RV32IFD-NEXT: fnmadd.d fa0, ft1, fa1, ft0
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fnmadd_d:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft1, fa0, ft0
; RV64IFD-NEXT: fadd.d ft0, fa2, ft0
; RV64IFD-NEXT: fnmadd.d fa0, ft1, fa1, ft0
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fnmadd_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s4, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s5, 4(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: mv s2, a3
; RV32I-NEXT: mv s3, a2
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s4, a0
; RV32I-NEXT: mv s5, a1
; RV32I-NEXT: mv a0, s1
; RV32I-NEXT: mv a1, s0
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv a4, a0
; RV32I-NEXT: lui a0, 524288
; RV32I-NEXT: xor a2, s5, a0
; RV32I-NEXT: xor a5, a1, a0
; RV32I-NEXT: mv a0, s4
; RV32I-NEXT: mv a1, a2
; RV32I-NEXT: mv a2, s3
; RV32I-NEXT: mv a3, s2
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s4, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s5, 4(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmadd_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s2, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: mv s1, a1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s2, a0
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a2, a1, 63
; RV64I-NEXT: xor a1, s2, a2
; RV64I-NEXT: xor a2, a0, a2
; RV64I-NEXT: mv a0, a1
; RV64I-NEXT: mv a1, s1
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s2, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%a_ = fadd double 0.0, %a
%c_ = fadd double 0.0, %c
%nega = fsub double -0.0, %a_
%negc = fsub double -0.0, %c_
%1 = call double @llvm.fma.f64(double %nega, double %b, double %negc)
ret double %1
}
define double @fnmadd_d_2(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fnmadd_d_2:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft1, fa1, ft0
; RV32IFD-NEXT: fadd.d ft0, fa2, ft0
; RV32IFD-NEXT: fnmadd.d fa0, ft1, fa0, ft0
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fnmadd_d_2:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft1, fa1, ft0
; RV64IFD-NEXT: fadd.d ft0, fa2, ft0
; RV64IFD-NEXT: fnmadd.d fa0, ft1, fa0, ft0
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fnmadd_d_2:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s4, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s5, 4(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: mv s2, a1
; RV32I-NEXT: mv s3, a0
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: mv a1, a3
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s4, a0
; RV32I-NEXT: mv s5, a1
; RV32I-NEXT: mv a0, s1
; RV32I-NEXT: mv a1, s0
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv a4, a0
; RV32I-NEXT: lui a0, 524288
; RV32I-NEXT: xor a3, s5, a0
; RV32I-NEXT: xor a5, a1, a0
; RV32I-NEXT: mv a0, s3
; RV32I-NEXT: mv a1, s2
; RV32I-NEXT: mv a2, s4
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s4, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s5, 4(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmadd_d_2:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s2, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: mv s1, a0
; RV64I-NEXT: mv a0, a1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s2, a0
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a2, a1, 63
; RV64I-NEXT: xor a1, s2, a2
; RV64I-NEXT: xor a2, a0, a2
; RV64I-NEXT: mv a0, s1
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s2, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%b_ = fadd double 0.0, %b
%c_ = fadd double 0.0, %c
%negb = fsub double -0.0, %b_
%negc = fsub double -0.0, %c_
%1 = call double @llvm.fma.f64(double %a, double %negb, double %negc)
ret double %1
}
define double @fnmadd_d_3(double %a, double %b, double %c) nounwind {
; CHECKIFD-LABEL: fnmadd_d_3:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fmadd.d ft0, fa0, fa1, fa2
; CHECKIFD-NEXT: fneg.d fa0, ft0
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fnmadd_d_3:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lui a2, 524288
; RV32I-NEXT: xor a1, a1, a2
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmadd_d_3:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a0, a0, a1
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = call double @llvm.fma.f64(double %a, double %b, double %c)
%neg = fneg double %1
ret double %neg
}
define double @fnmadd_nsz(double %a, double %b, double %c) nounwind {
; CHECKIFD-LABEL: fnmadd_nsz:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fnmadd.d fa0, fa0, fa1, fa2
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fnmadd_nsz:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lui a2, 524288
; RV32I-NEXT: xor a1, a1, a2
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmadd_nsz:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a0, a0, a1
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = call nsz double @llvm.fma.f64(double %a, double %b, double %c)
%neg = fneg nsz double %1
ret double %neg
}
define double @fnmsub_d(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fnmsub_d:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft0, fa0, ft0
; RV32IFD-NEXT: fnmsub.d fa0, ft0, fa1, fa2
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fnmsub_d:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft0, fa0, ft0
; RV64IFD-NEXT: fnmsub.d fa0, ft0, fa1, fa2
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fnmsub_d:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: mv s2, a3
; RV32I-NEXT: mv s3, a2
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: lui a2, 524288
; RV32I-NEXT: xor a1, a1, a2
; RV32I-NEXT: mv a2, s3
; RV32I-NEXT: mv a3, s2
; RV32I-NEXT: mv a4, s1
; RV32I-NEXT: mv a5, s0
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmsub_d:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: mv s1, a1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a0, a0, a1
; RV64I-NEXT: mv a1, s1
; RV64I-NEXT: mv a2, s0
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%a_ = fadd double 0.0, %a
%nega = fsub double -0.0, %a_
%1 = call double @llvm.fma.f64(double %nega, double %b, double %c)
ret double %1
}
define double @fnmsub_d_2(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fnmsub_d_2:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft0, fa1, ft0
; RV32IFD-NEXT: fnmsub.d fa0, ft0, fa0, fa2
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fnmsub_d_2:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft0, fa1, ft0
; RV64IFD-NEXT: fnmsub.d fa0, ft0, fa0, fa2
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fnmsub_d_2:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: mv s2, a1
; RV32I-NEXT: mv s3, a0
; RV32I-NEXT: mv a0, a2
; RV32I-NEXT: mv a1, a3
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv a2, a0
; RV32I-NEXT: lui a0, 524288
; RV32I-NEXT: xor a3, a1, a0
; RV32I-NEXT: mv a0, s3
; RV32I-NEXT: mv a1, s2
; RV32I-NEXT: mv a4, s1
; RV32I-NEXT: mv a5, s0
; RV32I-NEXT: call fma@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmsub_d_2:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: mv s1, a0
; RV64I-NEXT: mv a0, a1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a1, a0, a1
; RV64I-NEXT: mv a0, s1
; RV64I-NEXT: mv a2, s0
; RV64I-NEXT: call fma@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%b_ = fadd double 0.0, %b
%negb = fsub double -0.0, %b_
%1 = call double @llvm.fma.f64(double %a, double %negb, double %c)
ret double %1
}
define double @fmadd_d_contract(double %a, double %b, double %c) nounwind {
; CHECKIFD-LABEL: fmadd_d_contract:
; CHECKIFD: # %bb.0:
; CHECKIFD-NEXT: fmadd.d fa0, fa0, fa1, fa2
; CHECKIFD-NEXT: ret
;
; RV32I-LABEL: fmadd_d_contract:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -16
; RV32I-NEXT: sw ra, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 4(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: call __muldf3@plt
; RV32I-NEXT: mv a2, s1
; RV32I-NEXT: mv a3, s0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: lw ra, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 4(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 16
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmadd_d_contract:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -16
; RV64I-NEXT: sd ra, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: call __muldf3@plt
; RV64I-NEXT: mv a1, s0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: ld ra, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 16
; RV64I-NEXT: ret
%1 = fmul contract double %a, %b
%2 = fadd contract double %1, %c
ret double %2
}
define double @fmsub_d_contract(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fmsub_d_contract:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft0, fa2, ft0
; RV32IFD-NEXT: fmsub.d fa0, fa0, fa1, ft0
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fmsub_d_contract:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft0, fa2, ft0
; RV64IFD-NEXT: fmsub.d fa0, fa0, fa1, ft0
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fmsub_d_contract:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s4, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s5, 4(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a3
; RV32I-NEXT: mv s1, a2
; RV32I-NEXT: mv s2, a1
; RV32I-NEXT: mv s3, a0
; RV32I-NEXT: mv a0, a4
; RV32I-NEXT: mv a1, a5
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s4, a0
; RV32I-NEXT: mv s5, a1
; RV32I-NEXT: mv a0, s3
; RV32I-NEXT: mv a1, s2
; RV32I-NEXT: mv a2, s1
; RV32I-NEXT: mv a3, s0
; RV32I-NEXT: call __muldf3@plt
; RV32I-NEXT: mv a2, s4
; RV32I-NEXT: mv a3, s5
; RV32I-NEXT: call __subdf3@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s4, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s5, 4(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fmsub_d_contract:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s2, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a1
; RV64I-NEXT: mv s1, a0
; RV64I-NEXT: mv a0, a2
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s2, a0
; RV64I-NEXT: mv a0, s1
; RV64I-NEXT: mv a1, s0
; RV64I-NEXT: call __muldf3@plt
; RV64I-NEXT: mv a1, s2
; RV64I-NEXT: call __subdf3@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s2, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%c_ = fadd double 0.0, %c ; avoid negation using xor
%1 = fmul contract double %a, %b
%2 = fsub contract double %1, %c_
ret double %2
}
define double @fnmadd_d_contract(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fnmadd_d_contract:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft1, fa0, ft0
; RV32IFD-NEXT: fadd.d ft2, fa1, ft0
; RV32IFD-NEXT: fadd.d ft0, fa2, ft0
; RV32IFD-NEXT: fnmadd.d fa0, ft1, ft2, ft0
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fnmadd_d_contract:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft1, fa0, ft0
; RV64IFD-NEXT: fadd.d ft2, fa1, ft0
; RV64IFD-NEXT: fadd.d ft0, fa2, ft0
; RV64IFD-NEXT: fnmadd.d fa0, ft1, ft2, ft0
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fnmadd_d_contract:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s4, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s5, 4(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: mv s2, a3
; RV32I-NEXT: mv s3, a2
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s4, a0
; RV32I-NEXT: mv s5, a1
; RV32I-NEXT: mv a0, s3
; RV32I-NEXT: mv a1, s2
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s2, a0
; RV32I-NEXT: mv s3, a1
; RV32I-NEXT: mv a0, s1
; RV32I-NEXT: mv a1, s0
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s0, a0
; RV32I-NEXT: mv s1, a1
; RV32I-NEXT: mv a0, s4
; RV32I-NEXT: mv a1, s5
; RV32I-NEXT: mv a2, s2
; RV32I-NEXT: mv a3, s3
; RV32I-NEXT: call __muldf3@plt
; RV32I-NEXT: lui a2, 524288
; RV32I-NEXT: xor a1, a1, a2
; RV32I-NEXT: mv a2, s0
; RV32I-NEXT: mv a3, s1
; RV32I-NEXT: call __subdf3@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s4, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s5, 4(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmadd_d_contract:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s2, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: mv s1, a1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s2, a0
; RV64I-NEXT: mv a0, s1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s1, a0
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s0, a0
; RV64I-NEXT: mv a0, s2
; RV64I-NEXT: mv a1, s1
; RV64I-NEXT: call __muldf3@plt
; RV64I-NEXT: li a1, -1
; RV64I-NEXT: slli a1, a1, 63
; RV64I-NEXT: xor a0, a0, a1
; RV64I-NEXT: mv a1, s0
; RV64I-NEXT: call __subdf3@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s2, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%a_ = fadd double 0.0, %a ; avoid negation using xor
%b_ = fadd double 0.0, %b ; avoid negation using xor
%c_ = fadd double 0.0, %c ; avoid negation using xor
%1 = fmul contract double %a_, %b_
%2 = fneg double %1
%3 = fsub contract double %2, %c_
ret double %3
}
define double @fnmsub_d_contract(double %a, double %b, double %c) nounwind {
; RV32IFD-LABEL: fnmsub_d_contract:
; RV32IFD: # %bb.0:
; RV32IFD-NEXT: fcvt.d.w ft0, zero
; RV32IFD-NEXT: fadd.d ft1, fa0, ft0
; RV32IFD-NEXT: fadd.d ft0, fa1, ft0
; RV32IFD-NEXT: fnmsub.d fa0, ft1, ft0, fa2
; RV32IFD-NEXT: ret
;
; RV64IFD-LABEL: fnmsub_d_contract:
; RV64IFD: # %bb.0:
; RV64IFD-NEXT: fmv.d.x ft0, zero
; RV64IFD-NEXT: fadd.d ft1, fa0, ft0
; RV64IFD-NEXT: fadd.d ft0, fa1, ft0
; RV64IFD-NEXT: fnmsub.d fa0, ft1, ft0, fa2
; RV64IFD-NEXT: ret
;
; RV32I-LABEL: fnmsub_d_contract:
; RV32I: # %bb.0:
; RV32I-NEXT: addi sp, sp, -32
; RV32I-NEXT: sw ra, 28(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s0, 24(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s1, 20(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s2, 16(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s3, 12(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s4, 8(sp) # 4-byte Folded Spill
; RV32I-NEXT: sw s5, 4(sp) # 4-byte Folded Spill
; RV32I-NEXT: mv s0, a5
; RV32I-NEXT: mv s1, a4
; RV32I-NEXT: mv s2, a3
; RV32I-NEXT: mv s3, a2
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv s4, a0
; RV32I-NEXT: mv s5, a1
; RV32I-NEXT: mv a0, s3
; RV32I-NEXT: mv a1, s2
; RV32I-NEXT: li a2, 0
; RV32I-NEXT: li a3, 0
; RV32I-NEXT: call __adddf3@plt
; RV32I-NEXT: mv a2, a0
; RV32I-NEXT: mv a3, a1
; RV32I-NEXT: mv a0, s4
; RV32I-NEXT: mv a1, s5
; RV32I-NEXT: call __muldf3@plt
; RV32I-NEXT: mv a2, a0
; RV32I-NEXT: mv a3, a1
; RV32I-NEXT: mv a0, s1
; RV32I-NEXT: mv a1, s0
; RV32I-NEXT: call __subdf3@plt
; RV32I-NEXT: lw ra, 28(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s0, 24(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s1, 20(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s2, 16(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s3, 12(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s4, 8(sp) # 4-byte Folded Reload
; RV32I-NEXT: lw s5, 4(sp) # 4-byte Folded Reload
; RV32I-NEXT: addi sp, sp, 32
; RV32I-NEXT: ret
;
; RV64I-LABEL: fnmsub_d_contract:
; RV64I: # %bb.0:
; RV64I-NEXT: addi sp, sp, -32
; RV64I-NEXT: sd ra, 24(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s0, 16(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s1, 8(sp) # 8-byte Folded Spill
; RV64I-NEXT: sd s2, 0(sp) # 8-byte Folded Spill
; RV64I-NEXT: mv s0, a2
; RV64I-NEXT: mv s1, a1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv s2, a0
; RV64I-NEXT: mv a0, s1
; RV64I-NEXT: li a1, 0
; RV64I-NEXT: call __adddf3@plt
; RV64I-NEXT: mv a1, a0
; RV64I-NEXT: mv a0, s2
; RV64I-NEXT: call __muldf3@plt
; RV64I-NEXT: mv a1, a0
; RV64I-NEXT: mv a0, s0
; RV64I-NEXT: call __subdf3@plt
; RV64I-NEXT: ld ra, 24(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s0, 16(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s1, 8(sp) # 8-byte Folded Reload
; RV64I-NEXT: ld s2, 0(sp) # 8-byte Folded Reload
; RV64I-NEXT: addi sp, sp, 32
; RV64I-NEXT: ret
%a_ = fadd double 0.0, %a ; avoid negation using xor
%b_ = fadd double 0.0, %b ; avoid negation using xor
%1 = fmul contract double %a_, %b_
%2 = fsub contract double %c, %1
ret double %2
}