File: sched-setprio.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 (20 lines) | stat: -rw-r--r-- 877 bytes parent folder | download | duplicates (9) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
 
; RUN: llc -mtriple=amdgcn -mcpu=gfx908 -verify-machineinstrs < %s | FileCheck --check-prefix=GCN %s

declare void @llvm.amdgcn.s.setprio(i16)
declare <4 x float> @llvm.amdgcn.mfma.f32.4x4x1f32(float, float, <4 x float>, i32, i32, i32)

; GCN-LABEL: {{^}}test_mfma_f32_4x4x1f32:
; GCN: s_setprio 1
; GCN: v_mfma
; GCN: v_mfma
; GCN: s_setprio 0
define amdgpu_kernel void @test_mfma_f32_4x4x1f32(ptr addrspace(1) %arg) #0 {
bb:
  %in.1 = load <4 x float>, ptr addrspace(1) %arg
  call void @llvm.amdgcn.s.setprio(i16 1)
  %mai.1 = tail call <4 x float> @llvm.amdgcn.mfma.f32.4x4x1f32(float 1.0, float 2.0, <4 x float> %in.1, i32 0, i32 0, i32 0)
  %mai.2 = tail call <4 x float> @llvm.amdgcn.mfma.f32.4x4x1f32(float 3.0, float 4.0, <4 x float> %mai.1, i32 0, i32 0, i32 0)
  call void @llvm.amdgcn.s.setprio(i16 0)
  store <4 x float> %mai.2, ptr addrspace(1) %arg
  ret void
}