Differential D36852 Diff 112165 test/GPGPU/libdevice-functions-copied-into-kernel.ll

Changeset View

Standalone View

test/GPGPU/libdevice-functions-copied-into-kernel.ll

	Show All 14 Lines

	; Check that we model the kernel as a scop.			; Check that we model the kernel as a scop.
	; SCOP: Function: f			; SCOP: Function: f
	; SCOP-NEXT: Region: %entry.split---%for.end			; SCOP-NEXT: Region: %entry.split---%for.end

	; Check that the intrinsic call is present in the kernel IR.			; Check that the intrinsic call is present in the kernel IR.
	; KERNEL-IR: %p_expf = tail call float @__nv_expf(float %A.arr.i.val_p_scalar_)			; KERNEL-IR: %p_expf = tail call float @__nv_expf(float %A.arr.i.val_p_scalar_)
	; KERNEL-IR: %p_cosf = tail call float @__nv_cosf(float %p_expf)			; KERNEL-IR: %p_cosf = tail call float @__nv_cosf(float %p_expf)
				; KERNEL-IR: %p_logf = tail call float @__nv_logf(float %p_cosf)

	; Check that kernel launch is generated in host IR.			; Check that kernel launch is generated in host IR.
	; the declare would not be generated unless a call to a kernel exists.			; the declare would not be generated unless a call to a kernel exists.
	; HOST-IR: declare void @polly_launchKernel(i8, i32, i32, i32, i32, i32, i8)			; HOST-IR: declare void @polly_launchKernel(i8, i32, i32, i32, i32, i32, i8)


	; void f(float A, float B, int N) {			; void f(float A, float B, int N) {
	; for(int i = 0; i < N; i++) {			; for(int i = 0; i < N; i++) {
	; float tmp0 = A[i];			; float tmp0 = A[i];
	; float tmp1 = expf(tmp1);			; float expf = expf(tmp1);
	; tmp1 = cosf(tmp1);			; cosf = cosf(expf);
	; B[i] = tmp1;			; logf = logf(cosf);
				; B[i] = logf;
	; }			; }
	; }			; }

	target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"			target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"

	define void @f(float* %A, float* %B, i32 %N) {			define void @f(float* %A, float* %B, i32 %N) {
	entry:			entry:
	br label %entry.split			br label %entry.split

	entry.split: ; preds = %entry			entry.split: ; preds = %entry
	%cmp1 = icmp sgt i32 %N, 0			%cmp1 = icmp sgt i32 %N, 0
	br i1 %cmp1, label %for.body.lr.ph, label %for.end			br i1 %cmp1, label %for.body.lr.ph, label %for.end

	for.body.lr.ph: ; preds = %entry.split			for.body.lr.ph: ; preds = %entry.split
	br label %for.body			br label %for.body

	for.body: ; preds = %for.body.lr.ph, %for.body			for.body: ; preds = %for.body.lr.ph, %for.body
	%indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]			%indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]
	%A.arr.i = getelementptr inbounds float, float* %A, i64 %indvars.iv			%A.arr.i = getelementptr inbounds float, float* %A, i64 %indvars.iv
	%A.arr.i.val = load float, float* %A.arr.i, align 4			%A.arr.i.val = load float, float* %A.arr.i, align 4
	; Call to intrinsics that should be part of the kernel.			; Call to intrinsics that should be part of the kernel.
	%expf = tail call float @expf(float %A.arr.i.val)			%expf = tail call float @expf(float %A.arr.i.val)
	%cosf = tail call float @cosf(float %expf)			%cosf = tail call float @cosf(float %expf)
				%logf = tail call float @logf(float %cosf)
	%B.arr.i = getelementptr inbounds float, float* %B, i64 %indvars.iv			%B.arr.i = getelementptr inbounds float, float* %B, i64 %indvars.iv
	store float %expf, float* %B.arr.i, align 4			store float %logf, float* %B.arr.i, align 4

	%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1			%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
	%wide.trip.count = zext i32 %N to i64			%wide.trip.count = zext i32 %N to i64
	%exitcond = icmp ne i64 %indvars.iv.next, %wide.trip.count			%exitcond = icmp ne i64 %indvars.iv.next, %wide.trip.count
	br i1 %exitcond, label %for.body, label %for.cond.for.end_crit_edge			br i1 %exitcond, label %for.body, label %for.cond.for.end_crit_edge

	for.cond.for.end_crit_edge: ; preds = %for.body			for.cond.for.end_crit_edge: ; preds = %for.body
	br label %for.end			br label %for.end

	for.end: ; preds = %for.cond.for.end_crit_edge, %entry.split			for.end: ; preds = %for.cond.for.end_crit_edge, %entry.split
	ret void			ret void
	}			}

	; Function Attrs: nounwind readnone			; Function Attrs: nounwind readnone
	declare float @expf(float) #0			declare float @expf(float) #0
	declare float @cosf(float) #0			declare float @cosf(float) #0
				declare float @logf(float) #0

	attributes #0 = { nounwind readnone }			attributes #0 = { nounwind readnone }