; RUN: llc -mtriple=r600 -mcpu=redwood -disable-promote-alloca-to-vector < %s | FileCheck %s -check-prefix=R600 -check-prefix=FUNC
; RUN: llc -mtriple=r600 -mcpu=redwood < %s | FileCheck %s -check-prefix=R600-VECT -check-prefix=FUNC
; RUN: opt -S -mtriple=r600-unknown-unknown -mcpu=redwood -passes=amdgpu-promote-alloca -disable-promote-alloca-to-vector < %s | FileCheck -check-prefix=OPT %s
target datalayout = "A5"
declare i32 @llvm.r600.read.tidig.x() nounwind readnone
; FUNC-LABEL: {{^}}mova_same_clause:
; R600: LDS_WRITE
; R600: LDS_WRITE
; R600: LDS_READ
; R600: LDS_READ
; OPT: call range(i32 0, 257) i32 @llvm.r600.read.local.size.y()
; OPT: call range(i32 0, 257) i32 @llvm.r600.read.local.size.z()
; OPT: call range(i32 0, 256) i32 @llvm.r600.read.tidig.x()
; OPT: call range(i32 0, 256) i32 @llvm.r600.read.tidig.y()
; OPT: call range(i32 0, 256) i32 @llvm.r600.read.tidig.z()
define amdgpu_kernel void @mova_same_clause(ptr addrspace(1) nocapture %out, ptr addrspace(1) nocapture %in) #0 {
entry:
%stack = alloca [5 x i32], align 4, addrspace(5)
%0 = load i32, ptr addrspace(1) %in, align 4
%arrayidx1 = getelementptr inbounds [5 x i32], ptr addrspace(5) %stack, i32 0, i32 %0
store i32 4, ptr addrspace(5) %arrayidx1, align 4
%arrayidx2 = getelementptr inbounds i32, ptr addrspace(1) %in, i32 1
%1 = load i32, ptr addrspace(1) %arrayidx2, align 4
%arrayidx3 = getelementptr inbounds [5 x i32], ptr addrspace(5) %stack, i32 0, i32 %1
store i32 5, ptr addrspace(5) %arrayidx3, align 4
%2 = load i32, ptr addrspace(5) %stack, align 4
store i32 %2, ptr addrspace(1) %out, align 4
%arrayidx12 = getelementptr inbounds [5 x i32], ptr addrspace(5) %stack, i32 0, i32 1
%3 = load i32, ptr addrspace(5) %arrayidx12
%arrayidx13 = getelementptr inbounds i32, ptr addrspace(1) %out, i32 1
store i32 %3, ptr addrspace(1) %arrayidx13
ret void
}
; This test checks that the stack offset is calculated correctly for structs.
; All register loads/stores should be optimized away, so there shouldn't be
; any MOVA instructions.
;
; XXX: This generated code has unnecessary MOVs, we should be able to optimize
; this.
; FUNC-LABEL: {{^}}multiple_structs:
; R600-NOT: MOVA_INT
%struct.point = type { i32, i32 }
define amdgpu_kernel void @multiple_structs(ptr addrspace(1) %out) #0 {
entry:
%a = alloca %struct.point, addrspace(5)
%b = alloca %struct.point, addrspace(5)
%a.y.ptr = getelementptr inbounds %struct.point, ptr addrspace(5) %a, i32 0, i32 1
%b.y.ptr = getelementptr inbounds %struct.point, ptr addrspace(5) %b, i32 0, i32 1
store i32 0, ptr addrspace(5) %a
store i32 1, ptr addrspace(5) %a.y.ptr
store i32 2, ptr addrspace(5) %b
store i32 3, ptr addrspace(5) %b.y.ptr
%a.indirect = load i32, ptr addrspace(5) %a
%b.indirect = load i32, ptr addrspace(5) %b
%0 = add i32 %a.indirect, %b.indirect
store i32 %0, ptr addrspace(1) %out
ret void
}
; Test direct access of a private array inside a loop. The private array
; loads and stores should be lowered to copies, so there shouldn't be any
; MOVA instructions.
; FUNC-LABEL: {{^}}direct_loop:
; R600-NOT: MOVA_INT
define amdgpu_kernel void @direct_loop(ptr addrspace(1) %out, ptr addrspace(1) %in) #0 {
entry:
%prv_array_const = alloca [2 x i32], addrspace(5)
%prv_array = alloca [2 x i32], addrspace(5)
%a = load i32, ptr addrspace(1) %in
%b_src_ptr = getelementptr inbounds i32, ptr addrspace(1) %in, i32 1
%b = load i32, ptr addrspace(1) %b_src_ptr
store i32 %a, ptr addrspace(5) %prv_array_const
%b_dst_ptr = getelementptr inbounds [2 x i32], ptr addrspace(5) %prv_array_const, i32 0, i32 1
store i32 %b, ptr addrspace(5) %b_dst_ptr
br label %for.body
for.body:
%inc = phi i32 [0, %entry], [%count, %for.body]
%x = load i32, ptr addrspace(5) %prv_array_const
%y = load i32, ptr addrspace(5) %prv_array
%xy = add i32 %x, %y
store i32 %xy, ptr addrspace(5) %prv_array
%count = add i32 %inc, 1
%done = icmp eq i32 %count, 4095
br i1 %done, label %for.end, label %for.body
for.end:
%value = load i32, ptr addrspace(5) %prv_array
store i32 %value, ptr addrspace(1) %out
ret void
}
; FUNC-LABEL: {{^}}short_array:
; R600-VECT: MOVA_INT
define amdgpu_kernel void @short_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%0 = alloca [2 x i16], addrspace(5)
%1 = getelementptr inbounds [2 x i16], ptr addrspace(5) %0, i32 0, i32 1
store i16 0, ptr addrspace(5) %0
store i16 1, ptr addrspace(5) %1
%2 = getelementptr inbounds [2 x i16], ptr addrspace(5) %0, i32 0, i32 %index
%3 = load i16, ptr addrspace(5) %2
%4 = sext i16 %3 to i32
store i32 %4, ptr addrspace(1) %out
ret void
}
; FUNC-LABEL: {{^}}char_array:
; R600-VECT: MOVA_INT
define amdgpu_kernel void @char_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%0 = alloca [2 x i8], addrspace(5)
%1 = getelementptr inbounds [2 x i8], ptr addrspace(5) %0, i32 0, i32 1
store i8 0, ptr addrspace(5) %0
store i8 1, ptr addrspace(5) %1
%2 = getelementptr inbounds [2 x i8], ptr addrspace(5) %0, i32 0, i32 %index
%3 = load i8, ptr addrspace(5) %2
%4 = sext i8 %3 to i32
store i32 %4, ptr addrspace(1) %out
ret void
}
; Make sure we don't overwrite workitem information with private memory
; FUNC-LABEL: {{^}}work_item_info:
; R600-NOT: MOV T0.X
; Additional check in case the move ends up in the last slot
; R600-NOT: MOV * TO.X
define amdgpu_kernel void @work_item_info(ptr addrspace(1) %out, i32 %in) #0 {
entry:
%0 = alloca [2 x i32], addrspace(5)
%1 = getelementptr inbounds [2 x i32], ptr addrspace(5) %0, i32 0, i32 1
store i32 0, ptr addrspace(5) %0
store i32 1, ptr addrspace(5) %1
%2 = getelementptr inbounds [2 x i32], ptr addrspace(5) %0, i32 0, i32 %in
%3 = load i32, ptr addrspace(5) %2
%4 = call i32 @llvm.r600.read.tidig.x()
%5 = add i32 %3, %4
store i32 %5, ptr addrspace(1) %out
ret void
}
; Test that two stack objects are not stored in the same register
; The second stack object should be in T3.X
; FUNC-LABEL: {{^}}no_overlap:
define amdgpu_kernel void @no_overlap(ptr addrspace(1) %out, i32 %in) #0 {
entry:
%0 = alloca [3 x i8], align 1, addrspace(5)
%1 = alloca [2 x i8], align 1, addrspace(5)
%2 = getelementptr inbounds [3 x i8], ptr addrspace(5) %0, i32 0, i32 1
%3 = getelementptr inbounds [3 x i8], ptr addrspace(5) %0, i32 0, i32 2
%4 = getelementptr inbounds [2 x i8], ptr addrspace(5) %1, i32 0, i32 1
store i8 0, ptr addrspace(5) %0
store i8 1, ptr addrspace(5) %2
store i8 2, ptr addrspace(5) %3
store i8 1, ptr addrspace(5) %1
store i8 0, ptr addrspace(5) %4
%5 = getelementptr inbounds [3 x i8], ptr addrspace(5) %0, i32 0, i32 %in
%6 = getelementptr inbounds [2 x i8], ptr addrspace(5) %1, i32 0, i32 %in
%7 = load i8, ptr addrspace(5) %5
%8 = load i8, ptr addrspace(5) %6
%9 = add i8 %7, %8
%10 = sext i8 %9 to i32
store i32 %10, ptr addrspace(1) %out
ret void
}
define amdgpu_kernel void @char_array_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%alloca = alloca [2 x [2 x i8]], addrspace(5)
%gep1 = getelementptr inbounds [2 x [2 x i8]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 1
store i8 0, ptr addrspace(5) %alloca
store i8 1, ptr addrspace(5) %gep1
%gep2 = getelementptr inbounds [2 x [2 x i8]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 %index
%load = load i8, ptr addrspace(5) %gep2
%sext = sext i8 %load to i32
store i32 %sext, ptr addrspace(1) %out
ret void
}
define amdgpu_kernel void @i32_array_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%alloca = alloca [2 x [2 x i32]], addrspace(5)
%gep1 = getelementptr inbounds [2 x [2 x i32]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 1
store i32 0, ptr addrspace(5) %alloca
store i32 1, ptr addrspace(5) %gep1
%gep2 = getelementptr inbounds [2 x [2 x i32]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 %index
%load = load i32, ptr addrspace(5) %gep2
store i32 %load, ptr addrspace(1) %out
ret void
}
define amdgpu_kernel void @i64_array_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%alloca = alloca [2 x [2 x i64]], addrspace(5)
%gep1 = getelementptr inbounds [2 x [2 x i64]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 1
store i64 0, ptr addrspace(5) %alloca
store i64 1, ptr addrspace(5) %gep1
%gep2 = getelementptr inbounds [2 x [2 x i64]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 %index
%load = load i64, ptr addrspace(5) %gep2
store i64 %load, ptr addrspace(1) %out
ret void
}
%struct.pair32 = type { i32, i32 }
define amdgpu_kernel void @struct_array_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%alloca = alloca [2 x [2 x %struct.pair32]], addrspace(5)
%gep0 = getelementptr inbounds [2 x [2 x %struct.pair32]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 0, i32 1
%gep1 = getelementptr inbounds [2 x [2 x %struct.pair32]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 1, i32 1
store i32 0, ptr addrspace(5) %gep0
store i32 1, ptr addrspace(5) %gep1
%gep2 = getelementptr inbounds [2 x [2 x %struct.pair32]], ptr addrspace(5) %alloca, i32 0, i32 0, i32 %index, i32 0
%load = load i32, ptr addrspace(5) %gep2
store i32 %load, ptr addrspace(1) %out
ret void
}
define amdgpu_kernel void @struct_pair32_array(ptr addrspace(1) %out, i32 %index) #0 {
entry:
%alloca = alloca [2 x %struct.pair32], addrspace(5)
%gep0 = getelementptr inbounds [2 x %struct.pair32], ptr addrspace(5) %alloca, i32 0, i32 0, i32 1
%gep1 = getelementptr inbounds [2 x %struct.pair32], ptr addrspace(5) %alloca, i32 0, i32 1, i32 0
store i32 0, ptr addrspace(5) %gep0
store i32 1, ptr addrspace(5) %gep1
%gep2 = getelementptr inbounds [2 x %struct.pair32], ptr addrspace(5) %alloca, i32 0, i32 %index, i32 0
%load = load i32, ptr addrspace(5) %gep2
store i32 %load, ptr addrspace(1) %out
ret void
}
define amdgpu_kernel void @select_private(ptr addrspace(1) %out, i32 %in) nounwind {
entry:
%tmp = alloca [2 x i32], addrspace(5)
%tmp2 = getelementptr inbounds [2 x i32], ptr addrspace(5) %tmp, i32 0, i32 1
store i32 0, ptr addrspace(5) %tmp
store i32 1, ptr addrspace(5) %tmp2
%cmp = icmp eq i32 %in, 0
%sel = select i1 %cmp, ptr addrspace(5) %tmp, ptr addrspace(5) %tmp2
%load = load i32, ptr addrspace(5) %sel
store i32 %load, ptr addrspace(1) %out
ret void
}
; AMDGPUPromoteAlloca does not know how to handle ptrtoint. When it
; finds one, it should stop trying to promote.
; FUNC-LABEL: ptrtoint:
; SI-NOT: ds_write
; SI: buffer_store_dword v{{[0-9]+}}, v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], s{{[0-9]+}} offen
; SI: buffer_load_dword v{{[0-9]+}}, v{{[0-9]+}}, s[{{[0-9]+:[0-9]+}}], s{{[0-9]+}} offen ;
define amdgpu_kernel void @ptrtoint(ptr addrspace(1) %out, i32 %a, i32 %b) #0 {
%alloca = alloca [16 x i32], addrspace(5)
%tmp0 = getelementptr inbounds [16 x i32], ptr addrspace(5) %alloca, i32 0, i32 %a
store i32 5, ptr addrspace(5) %tmp0
%tmp1 = ptrtoint ptr addrspace(5) %alloca to i32
%tmp2 = add i32 %tmp1, 5
%tmp3 = inttoptr i32 %tmp2 to ptr addrspace(5)
%tmp4 = getelementptr inbounds i32, ptr addrspace(5) %tmp3, i32 %b
%tmp5 = load i32, ptr addrspace(5) %tmp4
store i32 %tmp5, ptr addrspace(1) %out
ret void
}
attributes #0 = { nounwind "amdgpu-waves-per-eu"="1,2" "amdgpu-flat-work-group-size"="1,256" }
Codebase Browser
llvm