File: adjust-writemask-vectorized.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 (14 lines) | stat: -rw-r--r-- 735 bytes parent folder | download | duplicates (9) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
 
; RUN: llc -mtriple=amdgcn -mcpu=gfx90a -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s

; Check that write mask is 0xf.

; GCN-LABEL: {{^}}sample_2d_vectorized_use:
; GCN: image_sample v[{{[0-9:]+}}], v[{{[0-9:]+}}], s[{{[0-9:]+}}], s[{{[0-9:]+}}] dmask:0xf
define amdgpu_ps <4 x float> @sample_2d_vectorized_use(<8 x i32> inreg %rsrc, <4 x i32> inreg %samp, float %s, float %t, <4 x float> %a) {
main_body:
  %v = call <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32 15, float %s, float %t, <8 x i32> %rsrc, <4 x i32> %samp, i1 0, i32 0, i32 0)
  %r = fadd <4 x float> %v, %a
  ret <4 x float> %r
}

declare <4 x float> @llvm.amdgcn.image.sample.2d.v4f32.f32(i32, float, float, <8 x i32>, <4 x i32>, i1, i32, i32)