[mlgo][aot] requrie the model is autogenerated for test determinism

[mtrofin]

Summary

The tests that exercise the 'release' mode, where the model is AOT-ed,
check the output has certain properties, to validate that, indeed, a
different policy from the default one was exercised. For determinism, we
can't reliably check that output for an arbitrary learned policy, since
it could be that policy happens to mimic the default one in that
particular case.

This patch adds a requirement that those tests run only when the model
is autogenerated (e.g. on build bots).

[phosek]

llvm/test/lit.site.cfg.py.in

57

llvm_canonicalize_cmake_booleans replaces every variable with 0 or 1 (depending on whether the variable evaluates to true or not in CMake) so this condition will always be false.

[phosek]

I'd probably use a new variable, for example LLVM_INLINER_MODEL_AUTOGENERATED and set it like this:

set(LLVM_INLINER_MODEL_AUTOGENERATED 0)
if(LLVM_INLINER_MODEL_PATH)
  string(TOUPPER ${LLVM_INLINER_MODEL_PATH} uppercase_LLVM_INLINER_MODEL_PATH)
  if(uppercase_LLVM_INLINER_MODEL_PATH STREQUAL "AUTOGENERATED")
    set(LLVM_INLINER_MODEL_AUTOGENERATED 1)
  endif()
endif()

Then I'd use @LLVM_INLINER_MODEL_AUTOGENERATED@ inside lit.site.cfg.py.in.

[mtrofin]

feedback

[phosek]

LGTM

llvm/CMakeLists.txt

set(TENSORFLOW_AOT_COMPILER
     "${TENSORFLOW_AOT_PATH}/../../../../bin/saved_model_cli"
     CACHE PATH "Path to the Tensorflow AOT compiler")
   include_directories(${TENSORFLOW_AOT_PATH}/include)
   add_subdirectory(${TENSORFLOW_AOT_PATH}/xla_aot_runtime_src
     ${CMAKE_ARCHIVE_OUTPUT_DIRECTORY}/tf_runtime)
   install(TARGETS tf_xla_runtime EXPORT LLVMExports
     ARCHIVE DESTINATION lib${LLVM_LIBDIR_SUFFIX} COMPONENT tf_xla_runtime)
   set_property(GLOBAL APPEND PROPERTY LLVM_EXPORTS tf_xla_runtime)
+  # Once we add more modules, we should handle this more automatically.
+  if (NOT DEFINED LLVM_INLINER_MODEL_PATH
+      OR "${LLVM_INLINER_MODEL_PATH}" STREQUAL ""
+      OR "${LLVM_INLINER_MODEL_PATH}" STREQUAL "autogenerate")
+    set(LLVM_INLINER_MODEL_PATH "autogenerate")
+    set(LLVM_INLINER_MODEL_AUTOGENERATED 1)
+  endif()
 endif()
 
 # Configure the three LLVM configuration header files.
 configure_file(
   ${LLVM_MAIN_INCLUDE_DIR}/llvm/Config/config.h.cmake
   ${LLVM_INCLUDE_DIR}/llvm/Config/config.h)
 configure_file(
   ${LLVM_MAIN_INCLUDE_DIR}/llvm/Config/llvm-config.h.cmake

llvm/lib/Analysis/CMakeLists.txt

 if (DEFINED LLVM_HAVE_TF_AOT OR DEFINED LLVM_HAVE_TF_API)
   include(TensorFlowCompile)
   set(LLVM_INLINER_MODEL_PATH_DEFAULT "models/inliner-Oz")
 
   # This url points to the most recent most which is known to be compatible with
   # LLVM. When better models are published, this url should be updated to aid
   # discoverability.
   set(LLVM_INLINER_MODEL_CURRENT_URL "https://github.com/google/ml-compiler-opt/releases/download/inlining-Oz-v1.1/inlining-Oz-99f0063-v1.1.tar.gz")
 
   if (DEFINED LLVM_HAVE_TF_AOT)
-    # If the path is empty, autogenerate the model
-    if (NOT DEFINED LLVM_INLINER_MODEL_PATH OR "${LLVM_INLINER_MODEL_PATH}" STREQUAL "")
-      set(LLVM_INLINER_MODEL_PATH "autogenerate")
-      message(WARNING "LLVM_INLINER_MODEL_PATH was not set: autogenerating a model to finish the build.")
-    endif()
-
     tf_find_and_compile(
       ${LLVM_INLINER_MODEL_PATH}
       ${LLVM_INLINER_MODEL_CURRENT_URL}
       ${LLVM_INLINER_MODEL_PATH_DEFAULT}
       "models/inlining/config.py"
       serve
       action
       InlinerSizeModel

llvm/test/CMakeLists.txt

llvm_canonicalize_cmake_booleans(
   LLVM_LINK_LLVM_DYLIB
   LLVM_TOOL_LTO_BUILD
   LLVM_USE_INTEL_JITEVENTS
   LLVM_BUILD_EXAMPLES
   LLVM_ENABLE_PLUGINS
   LLVM_BYE_LINK_INTO_TOOLS
   LLVM_HAVE_TF_AOT
   LLVM_HAVE_TF_API
+  LLVM_INLINER_MODEL_AUTOGENERATED
   LLVM_ENABLE_EXPENSIVE_CHECKS
   )
 
 configure_lit_site_cfg(
   ${CMAKE_CURRENT_SOURCE_DIR}/lit.site.cfg.py.in
   ${CMAKE_CURRENT_BINARY_DIR}/lit.site.cfg.py
   MAIN_CONFIG
   ${CMAKE_CURRENT_SOURCE_DIR}/lit.cfg.py

llvm/test/Transforms/Inline/ML/bounds-checks.ll

 ; Test behavior when inlining policy grows size out of control.
 ; In all cases, the end result is the same: mandatory inlinings must happen.
-; However, when we discover we 'trip' over the artificially-low size increase 
+; However, when we discover we 'trip' over the artificially-low size increase
 ; factor, we don't inline anymore.
 ; REQUIRES: have_tf_aot
+; REQUIRES: llvm_inliner_model_autogenerated
 ; RUN: opt -passes=scc-oz-module-inliner -enable-ml-inliner=release -ml-advisor-size-increase-threshold=10.0 -S < %s 2>&1 | FileCheck %s --check-prefix=CHECK --check-prefix=NOBOUNDS
 ; RUN: opt -passes=scc-oz-module-inliner -enable-ml-inliner=release -ml-advisor-size-increase-threshold=1.0 -S < %s 2>&1 | FileCheck %s --check-prefix=CHECK --check-prefix=BOUNDS
 
 target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-grtev4-linux-gnu"
 
 declare i64 @f1()
 

 attributes #0 = { alwaysinline }
 
 ; CHECK-LABEL: @top
 ; f2 must always be inlined, so we won't find a call to it in @top()
 ; CHECK-NOT: call i64 @f2
 ; @some-function isn't mandatory, and when we set the increase threshold too low,
 ; it won't be inlined.
 ; NOBOUNDS-NOT: @some_function
-; BOUNDS: call i64 @some_function
+; BOUNDS: call i64 @some_function
 

llvm/test/Transforms/Inline/ML/ml-test-release-mode.ll

 ; The default inliner doesn't elide @adder, it believes it's too costly to inline 
 ; adder into switcher. The ML inliner carries out that inlining, resulting in
 ; a smaller result (part of it is that adder gets elided).
 ;
 ; This test uses Inputs/test-module.ll, as it will share it with a similar test
 ; for the 'development' mode.
 ;
 ; REQUIRES: have_tf_aot
+; REQUIRES: llvm_inliner_model_autogenerated
 ; RUN: opt -passes=scc-oz-module-inliner -enable-ml-inliner=release -S < %S/Inputs/test-module.ll 2>&1 | FileCheck %S/Inputs/test-module.ll --check-prefix=CHECK
 ; RUN: opt -passes=scc-oz-module-inliner -enable-ml-inliner=default -S < %S/Inputs/test-module.ll 2>&1 | FileCheck %S/Inputs/test-module.ll --check-prefix=DEFAULT

llvm/test/lit.cfg.py

config.substitutions.append(
                                   config.llvm_shlib_ext)))
 
 if config.have_tf_aot:
     config.available_features.add("have_tf_aot")
 
 if config.have_tf_api:
     config.available_features.add("have_tf_api")
 
+if config.llvm_inliner_model_autogenerated:
+    config.available_features.add("llvm_inliner_model_autogenerated")
+
 def have_cxx_shared_library():
     readobj_exe = lit.util.which('llvm-readobj', config.llvm_tools_dir)
     if not readobj_exe:
         print('llvm-readobj not found')
         return False
 
     try:
         readobj_cmd = subprocess.Popen(

llvm/test/lit.site.cfg.py.in

 config.llvm_host_triple = '@LLVM_HOST_TRIPLE@'
 config.host_arch = "@HOST_ARCH@"
 config.have_opt_viewer_modules = @LLVM_HAVE_OPT_VIEWER_MODULES@
 config.libcxx_used = @LLVM_LIBCXX_USED@
 config.has_plugins = @LLVM_ENABLE_PLUGINS@
 config.linked_bye_extension = @LLVM_BYE_LINK_INTO_TOOLS@
 config.have_tf_aot = @LLVM_HAVE_TF_AOT@
 config.have_tf_api = @LLVM_HAVE_TF_API@
+config.llvm_inliner_model_autogenerated = @LLVM_INLINER_MODEL_AUTOGENERATED@
 config.expensive_checks = @LLVM_ENABLE_EXPENSIVE_CHECKS@
 
 # Support substitution of the tools_dir with user parameters. This is
 # used when we can't determine the tool dir at configuration time.
 try:
     config.llvm_tools_dir = config.llvm_tools_dir % lit_config.params
     config.llvm_lib_dir = config.llvm_lib_dir % lit_config.params
     config.llvm_shlib_dir = config.llvm_shlib_dir % lit_config.params