llvm/test/Transforms/InstCombine/gep-vector.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -instcombine %s -S | FileCheck %s

@block = global [64 x [8192 x i8]] zeroinitializer, align 1

define <2 x i8*> @vectorindex1() {
; CHECK-LABEL: @vectorindex1(
; CHECK-NEXT:    ret <2 x i8*> getelementptr inbounds ([64 x [8192 x i8]], [64 x [8192 x i8]]* @block, <2 x i64> zeroinitializer, <2 x i64> <i64 1, i64 2>, <2 x i64> zeroinitializer)
;
  %1 = getelementptr inbounds [64 x [8192 x i8]], [64 x [8192 x i8]]* @block, i64 0, <2 x i64> <i64 0, i64 1>, i64 8192
  ret <2 x i8*> %1
}

define <2 x i8*> @vectorindex2() {
; CHECK-LABEL: @vectorindex2(
; CHECK-NEXT:    ret <2 x i8*> getelementptr inbounds ([64 x [8192 x i8]], [64 x [8192 x i8]]* @block, <2 x i64> zeroinitializer, <2 x i64> <i64 1, i64 2>, <2 x i64> <i64 8191, i64 1>)
;
  %1 = getelementptr inbounds [64 x [8192 x i8]], [64 x [8192 x i8]]* @block, i64 0, i64 1, <2 x i64> <i64 8191, i64 8193>
  ret <2 x i8*> %1
}

define <2 x i8*> @vectorindex3() {
; CHECK-LABEL: @vectorindex3(
; CHECK-NEXT:    ret <2 x i8*> getelementptr inbounds ([64 x [8192 x i8]], [64 x [8192 x i8]]* @block, <2 x i64> zeroinitializer, <2 x i64> <i64 0, i64 2>, <2 x i64> <i64 8191, i64 1>)
;
  %1 = getelementptr inbounds [64 x [8192 x i8]], [64 x [8192 x i8]]* @block, i64 0, <2 x i64> <i64 0, i64 1>, <2 x i64> <i64 8191, i64 8193>
  ret <2 x i8*> %1
}

; Negative test - datalayout's alloc size for the 2 types must match.

define i32* @bitcast_vec_to_array_gep(<7 x i32>* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @bitcast_vec_to_array_gep(
; CHECK-NEXT:    [[ARR_PTR:%.*]] = bitcast <7 x i32>* [[X:%.*]] to [7 x i32]*
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr [7 x i32], [7 x i32]* [[ARR_PTR]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    ret i32* [[GEP]]
;
  %arr_ptr = bitcast <7 x i32>* %x to [7 x i32]*
  %gep = getelementptr [7 x i32], [7 x i32]* %arr_ptr, i64 %y, i64 %z
  ret i32* %gep
}

; Negative test - datalayout's alloc size for the 2 types must match.

define i32* @bitcast_array_to_vec_gep([3 x i32]* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @bitcast_array_to_vec_gep(
; CHECK-NEXT:    [[VEC_PTR:%.*]] = bitcast [3 x i32]* [[X:%.*]] to <3 x i32>*
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds <3 x i32>, <3 x i32>* [[VEC_PTR]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    ret i32* [[GEP]]
;
  %vec_ptr = bitcast [3 x i32]* %x to <3 x i32>*
  %gep = getelementptr inbounds <3 x i32>, <3 x i32>* %vec_ptr, i64 %y, i64 %z
  ret i32* %gep
}

; Sizes and types match - safe to remove bitcast.

define i32* @bitcast_vec_to_array_gep_matching_alloc_size(<4 x i32>* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @bitcast_vec_to_array_gep_matching_alloc_size(
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr <4 x i32>, <4 x i32>* [[X:%.*]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    ret i32* [[GEP]]
;
  %arr_ptr = bitcast <4 x i32>* %x to [4 x i32]*
  %gep = getelementptr [4 x i32], [4 x i32]* %arr_ptr, i64 %y, i64 %z
  ret i32* %gep
}

; Sizes and types match - safe to remove bitcast.

define i32* @bitcast_array_to_vec_gep_matching_alloc_size([4 x i32]* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @bitcast_array_to_vec_gep_matching_alloc_size(
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds [4 x i32], [4 x i32]* [[X:%.*]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    ret i32* [[GEP]]
;
  %vec_ptr = bitcast [4 x i32]* %x to <4 x i32>*
  %gep = getelementptr inbounds <4 x i32>, <4 x i32>* %vec_ptr, i64 %y, i64 %z
  ret i32* %gep
}

; Negative test - datalayout's alloc size for the 2 types must match.

define i32 addrspace(3)* @bitcast_vec_to_array_addrspace(<7 x i32>* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @bitcast_vec_to_array_addrspace(
; CHECK-NEXT:    [[ARR_PTR:%.*]] = bitcast <7 x i32>* [[X:%.*]] to [7 x i32]*
; CHECK-NEXT:    [[ASC:%.*]] = addrspacecast [7 x i32]* [[ARR_PTR]] to [7 x i32] addrspace(3)*
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr [7 x i32], [7 x i32] addrspace(3)* [[ASC]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    ret i32 addrspace(3)* [[GEP]]
;
  %arr_ptr = bitcast <7 x i32>* %x to [7 x i32]*
  %asc = addrspacecast [7 x i32]* %arr_ptr to [7 x i32] addrspace(3)*
  %gep = getelementptr [7 x i32], [7 x i32] addrspace(3)* %asc, i64 %y, i64 %z
  ret i32 addrspace(3)* %gep
}

; Negative test - datalayout's alloc size for the 2 types must match.

define i32 addrspace(3)* @inbounds_bitcast_vec_to_array_addrspace(<7 x i32>* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @inbounds_bitcast_vec_to_array_addrspace(
; CHECK-NEXT:    [[ARR_PTR:%.*]] = bitcast <7 x i32>* [[X:%.*]] to [7 x i32]*
; CHECK-NEXT:    [[ASC:%.*]] = addrspacecast [7 x i32]* [[ARR_PTR]] to [7 x i32] addrspace(3)*
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds [7 x i32], [7 x i32] addrspace(3)* [[ASC]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    ret i32 addrspace(3)* [[GEP]]
;
  %arr_ptr = bitcast <7 x i32>* %x to [7 x i32]*
  %asc = addrspacecast [7 x i32]* %arr_ptr to [7 x i32] addrspace(3)*
  %gep = getelementptr inbounds [7 x i32], [7 x i32] addrspace(3)* %asc, i64 %y, i64 %z
  ret i32 addrspace(3)* %gep
}

; Sizes and types match - safe to remove bitcast.

define i32 addrspace(3)* @bitcast_vec_to_array_addrspace_matching_alloc_size(<4 x i32>* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @bitcast_vec_to_array_addrspace_matching_alloc_size(
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr <4 x i32>, <4 x i32>* [[X:%.*]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    [[TMP1:%.*]] = addrspacecast i32* [[GEP]] to i32 addrspace(3)*
; CHECK-NEXT:    ret i32 addrspace(3)* [[TMP1]]
;
  %arr_ptr = bitcast <4 x i32>* %x to [4 x i32]*
  %asc = addrspacecast [4 x i32]* %arr_ptr to [4 x i32] addrspace(3)*
  %gep = getelementptr [4 x i32], [4 x i32] addrspace(3)* %asc, i64 %y, i64 %z
  ret i32 addrspace(3)* %gep
}

; Sizes and types match - safe to remove bitcast.

define i32 addrspace(3)* @inbounds_bitcast_vec_to_array_addrspace_matching_alloc_size(<4 x i32>* %x, i64 %y, i64 %z) {
; CHECK-LABEL: @inbounds_bitcast_vec_to_array_addrspace_matching_alloc_size(
; CHECK-NEXT:    [[GEP:%.*]] = getelementptr inbounds <4 x i32>, <4 x i32>* [[X:%.*]], i64 [[Y:%.*]], i64 [[Z:%.*]]
; CHECK-NEXT:    [[TMP1:%.*]] = addrspacecast i32* [[GEP]] to i32 addrspace(3)*
; CHECK-NEXT:    ret i32 addrspace(3)* [[TMP1]]
;
  %arr_ptr = bitcast <4 x i32>* %x to [4 x i32]*
  %asc = addrspacecast [4 x i32]* %arr_ptr to [4 x i32] addrspace(3)*
  %gep = getelementptr inbounds [4 x i32], [4 x i32] addrspace(3)* %asc, i64 %y, i64 %z
  ret i32 addrspace(3)* %gep
}

; Negative test - avoid doing bitcast on i8*, because '16' should be scaled by 'vscale'.

define i8* @test_accumulate_constant_offset_vscale_nonzero(<vscale x 16 x i1> %pg, i8* %base) {
; CHECK-LABEL: @test_accumulate_constant_offset_vscale_nonzero
; CHECK-NEXT:   %bc = bitcast i8* %base to <vscale x 16 x i8>*
; CHECK-NEXT:   %gep = getelementptr <vscale x 16 x i8>, <vscale x 16 x i8>* %bc, i64 1, i64 4
; CHECK-NEXT:   ret i8* %gep
  %bc = bitcast i8* %base to <vscale x 16 x i8>*
  %gep = getelementptr <vscale x 16 x i8>, <vscale x 16 x i8>* %bc, i64 1, i64 4
  ret i8* %gep
}

define i8* @test_accumulate_constant_offset_vscale_zero(<vscale x 16 x i1> %pg, i8* %base) {
; CHECK-LABEL: @test_accumulate_constant_offset_vscale_zero
; CHECK-NEXT:   %[[RES:.*]] = getelementptr i8, i8* %base, i64 4
; CHECK-NEXT:   ret i8* %[[RES]]
  %bc = bitcast i8* %base to <vscale x 16 x i8>*
  %gep = getelementptr <vscale x 16 x i8>, <vscale x 16 x i8>* %bc, i64 0, i64 4
  ret i8* %gep
}