[llvm] [AMDGPU] Introduce conditional barrier pseudo instruction (PR #171604)

[via llvm-commits]

github-actions[bot] wrote:

<!--PREMERGE ADVISOR COMMENT: Linux-->
# :penguin: Linux x64 Test Results

* 166977 tests passed
* 2927 tests skipped
* 1 test failed

## Failed Tests
(click on a test name to see its output)

### LLVM
<details>
<summary>LLVM.CodeGen/AMDGPU/llc-pipeline.ll</summary>

```
Exit Code: 1

Command Output (stdout):
--
# RUN: at line 2
/home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/llc -O0 -mtriple=amdgcn--amdhsa -disable-verify -debug-pass=Structure < /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll 2>&1    | /home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/FileCheck -match-full-lines -strict-whitespace -check-prefix=GCN-O0 /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
# executed command: /home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/llc -O0 -mtriple=amdgcn--amdhsa -disable-verify -debug-pass=Structure
# note: command had no output on stdout or stderr
# executed command: /home/gha/actions-runner/_work/llvm-project/llvm-project/build/bin/FileCheck -match-full-lines -strict-whitespace -check-prefix=GCN-O0 /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
# .---command stderr------------
# | /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll:133:15: error: GCN-O0-NEXT: is not on the line after the previous match
# | ; GCN-O0-NEXT:        Remove Redundant DEBUG_VALUE analysis
# |               ^
# | <stdin>:120:1: note: 'next' match was here
# |         Remove Redundant DEBUG_VALUE analysis
# | ^
# | <stdin>:118:27: note: previous match ended here
# |         SI Fix VGPR copies
# |                           ^
# | <stdin>:119:1: note: non-matching line after previous match is here
# |         AMDGPU Expand Conditional Barriers
# | ^
# | 
# | Input file: <stdin>
# | Check file: /home/gha/actions-runner/_work/llvm-project/llvm-project/llvm/test/CodeGen/AMDGPU/llc-pipeline.ll
# | 
# | -dump-input=help explains the following input dump.
# | 
# | Input was:
# | <<<<<<
# |           .
# |           .
# |           .
# |         115:         SI Lower WWM Copies 
# |         116:         AMDGPU Reserve WWM Registers 
# |         117:         Fast Register Allocator 
# |         118:         SI Fix VGPR copies 
# |         119:         AMDGPU Expand Conditional Barriers 
# |         120:         Remove Redundant DEBUG_VALUE analysis 
# | next:133     !~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~  error: match on wrong line
# |         121:         Fixup Statepoint Caller Saved 
# |         122:         Lazy Machine Block Frequency Analysis 
# |         123:         Machine Optimization Remark Emitter 
# |         124:         Prologue/Epilogue Insertion & Frame Finalization 
# |         125:         Post-RA pseudo instruction expansion pass 
# |           .
# |           .
# |           .
# | >>>>>>
# `-----------------------------
# error: command failed with exit status: 1

--

```
</details>

If these failures are unrelated to your changes (for example tests are broken or flaky at HEAD), please open an issue at https://github.com/llvm/llvm-project/issues and add the `infrastructure` label.

https://github.com/llvm/llvm-project/pull/171604