[llvm] [AMDGPU] Implement vop3p complex pattern optmization for gisel (PR #130234)

[LLVM Continuous Integration via llvm-commits]

llvm-ci wrote:

LLVM Buildbot has detected a new failure on builder `amdgpu-offload-rhel-9-cmake-build-only` running on `rocm-docker-rhel-9` while building `llvm` at step 4 "annotate".

Full details are available at: https://lab.llvm.org/buildbot/#/builders/205/builds/6808

<details>
<summary>Here is the relevant piece of the build log for the reference</summary>

```
Step 4 (annotate) failure: '../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py --jobs=32' (failure)
...
[7590/7780] Linking CXX executable bin/llvm-jitlink
[7591/7780] Linking CXX executable bin/llvm-objdump
[7592/7780] Generating ../../bin/llvm-otool
[7593/7780] Linking CXX executable bin/llvm-profgen
[7594/7780] Linking CXX executable bin/llvm-rtdyld
[7595/7780] Linking CXX executable bin/sancov
[7596/7780] Building CXX object tools/flang/lib/FrontendTool/CMakeFiles/flangFrontendTool.dir/ExecuteCompilerInvocation.cpp.o
[7597/7780] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSubtarget.cpp.o
[7598/7780] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNSubtarget.cpp.o
[7599/7780] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o
FAILED: lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o 
ccache /usr/bin/c++ -DGTEST_HAS_RTTI=0 -DLLVM_EXPORTS -D_DEBUG -D_GLIBCXX_ASSERTIONS -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/lib/Target/AMDGPU -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/include -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/include -fPIC -fno-semantic-interposition -fvisibility-inlines-hidden -Werror=date-time -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wno-missing-field-initializers -pedantic -Wno-long-long -Wimplicit-fallthrough -Wno-uninitialized -Wno-nonnull -Wno-class-memaccess -Wno-redundant-move -Wno-pessimizing-move -Wno-noexcept-type -Wdelete-non-virtual-dtor -Wsuggest-override -Wno-comment -Wno-misleading-indentation -Wctad-maybe-unsupported -fdiagnostics-color -ffunction-sections -fdata-sections -O3 -DNDEBUG -std=c++17 -fPIC  -fno-exceptions -funwind-tables -fno-rtti -UNDEBUG -MD -MT lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o -MF lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o.d -o lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o -c /home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp: In member function ‘bool llvm::AMDGPUInstructionSelector::selectG_TRUNC(llvm::MachineInstr&) const’:
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:2531: warning: enumerated and non-enumerated type in conditional expression [-Wextra]
 2531 |         DstSize < 32 ? AMDGPU::sub0 : TRI.getSubRegFromChannel(0, DstSize / 32);
      | 
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp: In function ‘llvm::SmallVector<std::pair<const llvm::MachineOperand*, SrcStatus> > getSrcStats(const llvm::MachineOperand*, const llvm::MachineRegisterInfo&, searchOptions, int)’:
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:4669: error: could not convert ‘Statlist’ from ‘SmallVector<[...],4>’ to ‘SmallVector<[...],3>’
 4669 |   return Statlist;
      | 
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp: In function ‘SrcStatus getNegStatus(const llvm::MachineOperand*, SrcStatus, const llvm::MachineRegisterInfo&)’:
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:4554: warning: control reaches end of non-void function [-Wreturn-type]
 4554 | }
      | 
ninja: build stopped: subcommand failed.
['ninja'] exited with return code 1.
The build step threw an exception...
Traceback (most recent call last):
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py", line 50, in step
    yield
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py", line 41, in main
    run_command(["ninja"])
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py", line 63, in run_command
    util.report_run_cmd(cmd, cwd=directory)
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-zorg/zorg/buildbot/builders/annotated/util.py", line 49, in report_run_cmd
    subprocess.check_call(cmd, shell=shell, *args, **kwargs)
  File "/usr/lib64/python3.9/subprocess.py", line 373, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['ninja']' returned non-zero exit status 1.
@@@STEP_FAILURE@@@
Step 7 (build cmake config) failure: build cmake config (failure)
...
[7590/7780] Linking CXX executable bin/llvm-jitlink
[7591/7780] Linking CXX executable bin/llvm-objdump
[7592/7780] Generating ../../bin/llvm-otool
[7593/7780] Linking CXX executable bin/llvm-profgen
[7594/7780] Linking CXX executable bin/llvm-rtdyld
[7595/7780] Linking CXX executable bin/sancov
[7596/7780] Building CXX object tools/flang/lib/FrontendTool/CMakeFiles/flangFrontendTool.dir/ExecuteCompilerInvocation.cpp.o
[7597/7780] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUSubtarget.cpp.o
[7598/7780] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/GCNSubtarget.cpp.o
[7599/7780] Building CXX object lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o
FAILED: lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o 
ccache /usr/bin/c++ -DGTEST_HAS_RTTI=0 -DLLVM_EXPORTS -D_DEBUG -D_GLIBCXX_ASSERTIONS -D_GNU_SOURCE -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/lib/Target/AMDGPU -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/include -I/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/include -fPIC -fno-semantic-interposition -fvisibility-inlines-hidden -Werror=date-time -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wno-missing-field-initializers -pedantic -Wno-long-long -Wimplicit-fallthrough -Wno-uninitialized -Wno-nonnull -Wno-class-memaccess -Wno-redundant-move -Wno-pessimizing-move -Wno-noexcept-type -Wdelete-non-virtual-dtor -Wsuggest-override -Wno-comment -Wno-misleading-indentation -Wctad-maybe-unsupported -fdiagnostics-color -ffunction-sections -fdata-sections -O3 -DNDEBUG -std=c++17 -fPIC  -fno-exceptions -funwind-tables -fno-rtti -UNDEBUG -MD -MT lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o -MF lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o.d -o lib/Target/AMDGPU/CMakeFiles/LLVMAMDGPUCodeGen.dir/AMDGPUInstructionSelector.cpp.o -c /home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp: In member function ‘bool llvm::AMDGPUInstructionSelector::selectG_TRUNC(llvm::MachineInstr&) const’:
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:2531: warning: enumerated and non-enumerated type in conditional expression [-Wextra]
 2531 |         DstSize < 32 ? AMDGPU::sub0 : TRI.getSubRegFromChannel(0, DstSize / 32);
      | 
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp: In function ‘llvm::SmallVector<std::pair<const llvm::MachineOperand*, SrcStatus> > getSrcStats(const llvm::MachineOperand*, const llvm::MachineRegisterInfo&, searchOptions, int)’:
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:4669: error: could not convert ‘Statlist’ from ‘SmallVector<[...],4>’ to ‘SmallVector<[...],3>’
 4669 |   return Statlist;
      | 
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp: In function ‘SrcStatus getNegStatus(const llvm::MachineOperand*, SrcStatus, const llvm::MachineRegisterInfo&)’:
/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-project/llvm/lib/Target/AMDGPU/AMDGPUInstructionSelector.cpp:4554: warning: control reaches end of non-void function [-Wreturn-type]
 4554 | }
      | 
ninja: build stopped: subcommand failed.
['ninja'] exited with return code 1.
The build step threw an exception...
Traceback (most recent call last):
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py", line 50, in step
    yield
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py", line 41, in main
    run_command(["ninja"])
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/build/../llvm-zorg/zorg/buildbot/builders/annotated/amdgpu-offload-cmake.py", line 63, in run_command
    util.report_run_cmd(cmd, cwd=directory)
  File "/home/botworker/bbot/amdgpu-offload-rhel-9-cmake-build-only/llvm-zorg/zorg/buildbot/builders/annotated/util.py", line 49, in report_run_cmd
    subprocess.check_call(cmd, shell=shell, *args, **kwargs)
  File "/usr/lib64/python3.9/subprocess.py", line 373, in check_call
    raise CalledProcessError(retcode, cmd)
subprocess.CalledProcessError: Command '['ninja']' returned non-zero exit status 1.
program finished with exit code 0
elapsedTime=69.057063

```

</details>

https://github.com/llvm/llvm-project/pull/130234