mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-25 14:32:53 +00:00
0ff4287fe2
IR for CUDA should use "nvptx[64]-nvidia-cuda", and IR for NV OpenCL should use "nvptx[64]-nvidia-nvcl" git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@184579 91177308-0d34-0410-b5e6-96231b3b80d8
27 lines
811 B
LLVM
27 lines
811 B
LLVM
; RUN: llc < %s -march=nvptx -mcpu=sm_20 | FileCheck %s
|
|
|
|
target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v16:16:16-v32:32:32-v64:64:64-v128:128:128-n16:32:64"
|
|
target triple = "nvptx-nvidia-cuda"
|
|
|
|
; Ensure global variables in address space 0 are promoted to address space 1
|
|
|
|
; CHECK: .global .align 4 .u32 myglobal = 42;
|
|
@myglobal = internal global i32 42, align 4
|
|
; CHECK: .global .align 4 .u32 myconst = 42;
|
|
@myconst = internal constant i32 42, align 4
|
|
|
|
|
|
define void @foo(i32* %a, i32* %b) {
|
|
; CHECK: cvta.global.u32
|
|
%ld1 = load i32* @myglobal
|
|
; CHECK: cvta.global.u32
|
|
%ld2 = load i32* @myconst
|
|
store i32 %ld1, i32* %a
|
|
store i32 %ld2, i32* %b
|
|
ret void
|
|
}
|
|
|
|
|
|
!nvvm.annotations = !{!0}
|
|
!0 = metadata !{void (i32*, i32*)* @foo, metadata !"kernel", i32 1}
|