File: v1024.ll

package info (click to toggle)
llvm-toolchain-20 1%3A20.1.6-1~exp1
  • links: PTS, VCS
  • area: main
  • in suites: experimental
  • size: 2,111,304 kB
  • sloc: cpp: 7,438,677; ansic: 1,393,822; asm: 1,012,926; python: 241,650; f90: 86,635; objc: 75,479; lisp: 42,144; pascal: 17,286; sh: 10,027; ml: 5,082; perl: 4,730; awk: 3,523; makefile: 3,349; javascript: 2,251; xml: 892; fortran: 672
file content (30 lines) | stat: -rw-r--r-- 1,376 bytes parent folder | download | duplicates (6)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
; RUN: llc -mtriple=amdgcn -mcpu=gfx908 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s

; Check that we do not use AGPRs for v32i32 type

; GCN-LABEL: {{^}}test_v1024:
; GCN-NOT: v_accvgpr
; GCN-COUNT-8: global_store_dwordx4
; GCN-NOT: v_accvgpr
define amdgpu_kernel void @test_v1024(i1 %c0) {
entry:
  %alloca = alloca <32 x i32>, align 16, addrspace(5)
  call void @llvm.memset.p5.i32(ptr addrspace(5) %alloca, i8 0, i32 128, i1 false)
  br i1 %c0, label %if.then.i.i, label %if.else.i

if.then.i.i:                                      ; preds = %entry
  call void @llvm.memcpy.p5.p5.i64(ptr addrspace(5) align 16 %alloca, ptr addrspace(5) align 4 undef, i64 128, i1 false)
  br label %if.then.i62.i

if.else.i:                                        ; preds = %entry
  br label %if.then.i62.i

if.then.i62.i:                                    ; preds = %if.else.i, %if.then.i.i
  call void @llvm.memcpy.p1.p5.i64(ptr addrspace(1) align 4 undef, ptr addrspace(5) align 16 %alloca, i64 128, i1 false)
  ret void
}

declare void @llvm.memset.p5.i32(ptr addrspace(5) nocapture readonly, i8, i32, i1 immarg)
declare void @llvm.memcpy.p5.p5.i64(ptr addrspace(5) nocapture writeonly, ptr addrspace(5) nocapture readonly, i64, i1 immarg)

declare void @llvm.memcpy.p1.p5.i64(ptr addrspace(1) nocapture writeonly, ptr addrspace(5) nocapture readonly, i64, i1 immarg)