File: machinelicm-convergent.mir

package info (click to toggle)
llvm-toolchain-21 1%3A21.1.6-3
  • links: PTS, VCS
  • area: main
  • in suites: forky, sid
  • size: 2,245,028 kB
  • sloc: cpp: 7,619,726; ansic: 1,434,018; asm: 1,058,748; python: 252,740; f90: 94,671; objc: 70,685; lisp: 42,813; pascal: 18,401; sh: 8,601; ml: 5,111; perl: 4,720; makefile: 3,675; awk: 3,523; javascript: 2,409; xml: 892; fortran: 770
file content (33 lines) | stat: -rw-r--r-- 777 bytes parent folder | download | duplicates (5) 
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
31
32
33
 
# RUN: llc -mtriple=amdgcn  -run-pass=early-machinelicm -o - %s | FileCheck %s
# RUN: llc -mtriple=amdgcn  -passes=early-machinelicm -o - %s | FileCheck %s

# Test to check machine LICM does not hoist convergent instructions,
# DS_PERMUTE_B32 in this example.

---
# CHECK-LABEL: name: _amdgpu_cs_main
# CHECK: bb.1:
# CHECK: DS_PERMUTE_B32

name: _amdgpu_cs_main
body: |
  bb.0:
    liveins: $vgpr0, $vgpr1
    successors: %bb.1

    %0:vgpr_32 = COPY $vgpr0
    %1:vgpr_32 = COPY $vgpr1

  bb.1:
    successors: %bb.1, %bb.2

    %2:vgpr_32 = DS_PERMUTE_B32 %0, %1, 0, implicit $exec
    %3:vgpr_32 = V_ADD_CO_U32_e32 %0, %2, implicit-def $vcc, implicit $exec
    S_CBRANCH_SCC1 %bb.1, implicit undef $scc
    S_BRANCH %bb.2

  bb.2:
    $vgpr0 = COPY %3
    S_ENDPGM 0

...