[TLI][AArch64] Add SLEEF mappings to scalable vector functions for fmod and fmodf

[jolanta.jensen]

Summary

This patch adds SLEEF mappings to scalable vector functions for fmod and fmodf.

llvm/include/llvm/Analysis/VecFuncs.def

 TLI_DEFINE_VECFUNC("exp2", "_ZGVsMxv_exp2",  SCALABLE(2), MASKED)
 TLI_DEFINE_VECFUNC("exp2f", "_ZGVsMxv_exp2f", SCALABLE(4), MASKED)
 TLI_DEFINE_VECFUNC("llvm.exp2.f64", "_ZGVsMxv_exp2", SCALABLE(2), MASKED)
 TLI_DEFINE_VECFUNC("llvm.exp2.f32", "_ZGVsMxv_exp2f", SCALABLE(4), MASKED)
 
 TLI_DEFINE_VECFUNC("exp10", "_ZGVsMxv_exp10",  SCALABLE(2), MASKED)
 TLI_DEFINE_VECFUNC("exp10f", "_ZGVsMxv_exp10f", SCALABLE(4), MASKED)
 
+TLI_DEFINE_VECFUNC("fmod", "_ZGVsMxvv_fmod", SCALABLE(2), MASKED)
+TLI_DEFINE_VECFUNC("fmodf", "_ZGVsMxvv_fmodf", SCALABLE(4), MASKED)
+
 TLI_DEFINE_VECFUNC("lgamma", "_ZGVsMxv_lgamma",  SCALABLE(2), MASKED)
 TLI_DEFINE_VECFUNC("lgammaf", "_ZGVsMxv_lgammaf", SCALABLE(4), MASKED)
 
 TLI_DEFINE_VECFUNC("log", "_ZGVsMxv_log",  SCALABLE(2), MASKED)
 TLI_DEFINE_VECFUNC("logf", "_ZGVsMxv_logf", SCALABLE(4), MASKED)
 TLI_DEFINE_VECFUNC("llvm.log.f64", "_ZGVsMxv_log", SCALABLE(2), MASKED)
 TLI_DEFINE_VECFUNC("llvm.log.f32", "_ZGVsMxv_logf", SCALABLE(4), MASKED)
 

llvm/test/Transforms/LoopVectorize/AArch64/sleef-calls-aarch64.ll

define void @exp10_f32(float* nocapture %varray) {
   %iv.next = add nuw nsw i64 %iv, 1
   %exitcond = icmp eq i64 %iv.next, 1000
   br i1 %exitcond, label %for.end, label %for.body
 
   for.end:
   ret void
 }
 
+declare double @fmod(double, double) #0
+declare float @fmodf(float, float) #0
+
+define void @fmod_f64(double* nocapture %varray) {
+  ; CHECK-LABEL: @fmod_f64(
+  ; SVE:      [[TMP5:%.*]] = call <vscale x 2 x double> @_ZGVsMxvv_fmod(<vscale x 2 x double> [[TMP4:%.*]], <vscale x 2 x double> [[TMP4:%.*]], <vscale x 2 x i1> {{.*}})
+  ; CHECK:    ret void
+  ;
+  entry:
+  br label %for.body
+
+  for.body:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %tmp = trunc i64 %iv to i32
+  %conv = sitofp i32 %tmp to double
+  %call = tail call double @fmod(double %conv, double %conv)
+  %arrayidx = getelementptr inbounds double, double* %varray, i64 %iv
+  store double %call, double* %arrayidx, align 8
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond = icmp eq i64 %iv.next, 1000
+  br i1 %exitcond, label %for.end, label %for.body
+
+  for.end:
+  ret void
+}
+
+define void @fmod_f32(float* nocapture %varray) {
+  ; CHECK-LABEL: @fmod_f32(
+  ; SVE:      [[TMP5:%.*]] = call <vscale x 4 x float> @_ZGVsMxvv_fmodf(<vscale x 4 x float> [[TMP4:%.*]], <vscale x 4 x float> [[TMP4:%.*]], <vscale x 4 x i1> {{.*}})
+  ; CHECK:    ret void
+  ;
+  entry:
+  br label %for.body
+
+  for.body:
+  %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
+  %tmp = trunc i64 %iv to i32
+  %conv = sitofp i32 %tmp to float
+  %call = tail call float @fmodf(float %conv, float %conv)
+  %arrayidx = getelementptr inbounds float, float* %varray, i64 %iv
+  store float %call, float* %arrayidx, align 4
+  %iv.next = add nuw nsw i64 %iv, 1
+  %exitcond = icmp eq i64 %iv.next, 1000
+  br i1 %exitcond, label %for.end, label %for.body
+
+  for.end:
+  ret void
+}
+
 declare double @lgamma(double) #0
 declare float @lgammaf(float) #0
 declare double @llvm.lgamma.f64(double) #0
 declare float @llvm.lgamma.f32(float) #0
 
 define void @lgamma_f64(double* nocapture %varray) {
   ; CHECK-LABEL: @lgamma_f64(
   ; NEON:     [[TMP5:%.*]] = call <2 x double> @_ZGVnN2v_lgamma(<2 x double> [[TMP4:%.*]])