File: extload-align.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 (23 lines) | stat: -rw-r--r-- 1,062 bytes parent folder | download | duplicates (4) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
 
; RUN: llc -debug-only=machine-scheduler -march=amdgcn -mtriple=amdgcn-- -verify-machineinstrs %s -o - 2>&1| FileCheck -check-prefix=DEBUG %s
target datalayout = "A5"
; REQUIRES: asserts

; Verify that the extload generated from %eval has the default
; alignment size (2) corresponding to the underlying memory size (i16)
; size and not 4 corresponding to the sign-extended size (i32).

; DEBUG: {{^}}# Machine code for function extload_align:
; DEBUG: (volatile load (s16) from %ir.a, addrspace 5)
; DEBUG: {{^}}# End machine code for function extload_align.

define amdgpu_kernel void @extload_align(ptr addrspace(5) %out, i32 %index) #0 {
  %v0 = alloca [4 x i16], addrspace(5)
  %a2 = getelementptr inbounds [4 x i16], ptr addrspace(5) %v0, i32 0, i32 1
  store volatile i16 0, ptr addrspace(5) %v0
  store volatile i16 1, ptr addrspace(5) %a2
  %a = getelementptr inbounds [4 x i16], ptr addrspace(5) %v0, i32 0, i32 %index
  %val = load volatile i16, ptr addrspace(5) %a
  %eval = sext i16 %val to i32
  store i32 %eval, ptr addrspace(5) %out
  ret void
}