File: vectorize_i24.ll

package info (click to toggle)
llvm-toolchain-17 1%3A17.0.6-22
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid, trixie
  • size: 1,799,624 kB
  • sloc: cpp: 6,428,607; ansic: 1,383,196; asm: 793,408; python: 223,504; objc: 75,364; f90: 60,502; lisp: 33,869; pascal: 15,282; sh: 9,684; perl: 7,453; ml: 4,937; awk: 3,523; makefile: 2,889; javascript: 2,149; xml: 888; fortran: 619; cs: 573
file content (21 lines) | stat: -rw-r--r-- 568 bytes parent folder | download | duplicates (10) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
 
; RUN: opt -mtriple=nvptx64-nvidia-cuda -passes=load-store-vectorizer -S -o - %s | FileCheck %s

; We don't need to vectorize this.  Just make sure it doesn't crash.

; CHECK-LABEL: @int24x2
; CHECK: load i24
; CHECK: load i24
; CHECK: store i24
; CHECK: store i24
define void @int24x2(ptr nocapture align 4 %ptr) {
  %ptr0 = getelementptr i24, ptr %ptr, i64 0
  %ptr1 = getelementptr i24, ptr %ptr, i64 1

  %l0 = load i24, ptr %ptr0, align 4
  %l1 = load i24, ptr %ptr1, align 1

  store i24 %l1, ptr %ptr0, align 4
  store i24 %l0, ptr %ptr1, align 1

  ret void
}