[CI] Add Python Script for Computing Projects/Runtimes to Test

This patch adds a python script, compute_projects, and associated unit
tests for computing the projects and runtimes that need to be tested in
premerge. Rewriting in Python opens up a couple new
improvements/opportunities:
1. I personally find python to be much easier to work with than shell
   scripts for tasks like this. Particularly it becomes a lot easier to
   work with paths with proper array support.
2. Unit testing becomes easier which makes it a lot easier to reason
   about behavior changes, especially in review.
3. Most of the configuration is now setup in some dictionaries, which
   makes changes much easier to apply for most of the common changes.

This preserves the behavior of the existing premerge scripts as much as
possible.

Reviewers: ldionne, lnihlen, Endilll, tstellar, Keenuts

Reviewed By: Keenuts

Pull Request: #132634

Author: boomanaiden154

.ci/compute_projects.py

@@ -0,0 +1,222 @@
+# Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+# See https://llvm.org/LICENSE.txt for license information.
+# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+"""Computes the list of projects that need to be tested from a diff.
+
+Does some things, spits out a list of projects.
+"""
+
+from collections.abc import Set
+import pathlib
+import platform
+import sys
+
+# This mapping lists out the dependencies for each project. These should be
+# direct dependencies. The code will handle transitive dependencies. Some
+# projects might have optional dependencies depending upon how they are built.
+# The dependencies listed here should be the dependencies required for the
+# configuration built/tested in the premerge CI.
+PROJECT_DEPENDENCIES = {
+    "llvm": set(),
+    "clang": {"llvm"},
+    "bolt": {"clang", "lld", "llvm"},
+    "clang-tools-extra": {"clang", "llvm"},
+    "compiler-rt": {"clang", "lld"},
+    "libc": {"clang", "lld"},
+    "openmp": {"clang", "lld"},
+    "flang": {"llvm", "clang"},
+    "lldb": {"llvm", "clang"},
+    "libclc": {"llvm", "clang"},
+    "lld": {"llvm"},
+    "mlir": {"llvm"},
+    "polly": {"llvm"},
+}
+
+# This mapping describes the additional projects that should be tested when a
+# specific project is touched. We enumerate them specifically rather than
+# just invert the dependencies list to give more control over what exactly is
+# tested.
+DEPENDENTS_TO_TEST = {
+    "llvm": {
+        "bolt",
+        "clang",
+        "clang-tools-extra",
+        "lld",
+        "lldb",
+        "mlir",
+        "polly",
+        "flang",
+    },
+    "lld": {"bolt", "cross-project-tests"},
+    # TODO(issues/132795): LLDB should be enabled on clang changes.
+    "clang": {"clang-tools-extra", "compiler-rt", "cross-project-tests"},
+    "clang-tools-extra": {"libc"},
+    "mlir": {"flang"},
+}
+
+DEPENDENT_RUNTIMES_TO_TEST = {"clang": {"libcxx", "libcxxabi", "libunwind"}}
+
+EXCLUDE_LINUX = {
+    "cross-project-tests",  # TODO(issues/132796): Tests are failing.
+    "openmp",  # https://github.com/google/llvm-premerge-checks/issues/410
+}
+
+EXCLUDE_WINDOWS = {
+    "cross-project-tests",  # TODO(issues/132797): Tests are failing.
+    "compiler-rt",  # TODO(issues/132798): Tests take excessive time.
+    "openmp",  # TODO(issues/132799): Does not detect perl installation.
+    "libc",  # No Windows Support.
+    "lldb",  # TODO(issues/132800): Needs environment setup.
+    "bolt",  # No Windows Support.
+}
+
+# These are projects that we should test if the project itself is changed but
+# where testing is not yet stable enough for it to be enabled on changes to
+# dependencies.
+EXCLUDE_DEPENDENTS_WINDOWS = {
+    "flang",  # TODO(issues/132803): Flang is not stable.
+}
+
+EXCLUDE_MAC = {
+    "bolt",
+    "compiler-rt",
+    "cross-project-tests",
+    "flang",
+    "libc",
+    "libcxx",
+    "libcxxabi",
+    "libunwind",
+    "lldb",
+    "openmp",
+    "polly",
+}
+
+PROJECT_CHECK_TARGETS = {
+    "clang-tools-extra": "check-clang-tools",
+    "compiler-rt": "check-compiler-rt",
+    "cross-project-tests": "check-cross-project",
+    "libcxx": "check-cxx",
+    "libcxxabi": "check-cxxabi",
+    "libunwind": "check-unwind",
+    "lldb": "check-lldb",
+    "llvm": "check-llvm",
+    "clang": "check-clang",
+    "bolt": "check-bolt",
+    "lld": "check-lld",
+    "flang": "check-flang",
+    "libc": "check-libc",
+    "lld": "check-lld",
+    "lldb": "check-lldb",
+    "mlir": "check-mlir",
+    "openmp": "check-openmp",
+    "polly": "check-polly",
+}
+
+
+def _add_dependencies(projects: Set[str]) -> Set[str]:
+    projects_with_dependents = set(projects)
+    current_projects_count = 0
+    while current_projects_count != len(projects_with_dependents):
+        current_projects_count = len(projects_with_dependents)
+        for project in list(projects_with_dependents):
+            if project not in PROJECT_DEPENDENCIES:
+                continue
+            projects_with_dependents.update(PROJECT_DEPENDENCIES[project])
+    return projects_with_dependents
+
+
+def _compute_projects_to_test(modified_projects: Set[str], platform: str) -> Set[str]:
+    projects_to_test = set()
+    for modified_project in modified_projects:
+        # Skip all projects where we cannot run tests.
+        if modified_project not in PROJECT_CHECK_TARGETS:
+            continue
+        projects_to_test.add(modified_project)
+        if modified_project not in DEPENDENTS_TO_TEST:
+            continue
+        for dependent_project in DEPENDENTS_TO_TEST[modified_project]:
+            if (
+                platform == "Windows"
+                and dependent_project in EXCLUDE_DEPENDENTS_WINDOWS
+            ):
+                continue
+            projects_to_test.add(dependent_project)
+    if platform == "Linux":
+        for to_exclude in EXCLUDE_LINUX:
+            if to_exclude in projects_to_test:
+                projects_to_test.remove(to_exclude)
+    elif platform == "Windows":
+        for to_exclude in EXCLUDE_WINDOWS:
+            if to_exclude in projects_to_test:
+                projects_to_test.remove(to_exclude)
+    elif platform == "Darwin":
+        for to_exclude in EXCLUDE_MAC:
+            if to_exclude in projects_to_test:
+                projects_to_test.remove(to_exclude)
+    else:
+        raise ValueError("Unexpected platform.")
+    return projects_to_test
+
+
+def _compute_projects_to_build(projects_to_test: Set[str]) -> Set[str]:
+    return _add_dependencies(projects_to_test)
+
+
+def _compute_project_check_targets(projects_to_test: Set[str]) -> Set[str]:
+    check_targets = set()
+    for project_to_test in projects_to_test:
+        if project_to_test not in PROJECT_CHECK_TARGETS:
+            continue
+        check_targets.add(PROJECT_CHECK_TARGETS[project_to_test])
+    return check_targets
+
+
+def _compute_runtimes_to_test(projects_to_test: Set[str]) -> Set[str]:
+    runtimes_to_test = set()
+    for project_to_test in projects_to_test:
+        if project_to_test not in DEPENDENT_RUNTIMES_TO_TEST:
+            continue
+        runtimes_to_test.update(DEPENDENT_RUNTIMES_TO_TEST[project_to_test])
+    return runtimes_to_test
+
+
+def _compute_runtime_check_targets(runtimes_to_test: Set[str]) -> Set[str]:
+    check_targets = set()
+    for runtime_to_test in runtimes_to_test:
+        check_targets.add(PROJECT_CHECK_TARGETS[runtime_to_test])
+    return check_targets
+
+
+def _get_modified_projects(modified_files: list[str]) -> Set[str]:
+    modified_projects = set()
+    for modified_file in modified_files:
+        modified_projects.add(pathlib.Path(modified_file).parts[0])
+    return modified_projects
+
+
+def get_env_variables(modified_files: list[str], platform: str) -> Set[str]:
+    modified_projects = _get_modified_projects(modified_files)
+    projects_to_test = _compute_projects_to_test(modified_projects, platform)
+    projects_to_build = _compute_projects_to_build(projects_to_test)
+    projects_check_targets = _compute_project_check_targets(projects_to_test)
+    runtimes_to_test = _compute_runtimes_to_test(projects_to_test)
+    runtimes_check_targets = _compute_runtime_check_targets(runtimes_to_test)
+    # We use a semicolon to separate the projects/runtimes as they get passed
+    # to the CMake invocation and thus we need to use the CMake list separator
+    # (;). We use spaces to separate the check targets as they end up getting
+    # passed to ninja.
+    return {
+        "projects_to_build": ";".join(sorted(projects_to_build)),
+        "project_check_targets": " ".join(sorted(projects_check_targets)),
+        "runtimes_to_build": ";".join(sorted(runtimes_to_test)),
+        "runtimes_check_targets": " ".join(sorted(runtimes_check_targets)),
+    }
+
+
+if __name__ == "__main__":
+    current_platform = platform.system()
+    if len(sys.argv) == 2:
+        current_platform = sys.argv[1]
+    env_variables = get_env_variables(sys.stdin.readlines(), current_platform)
+    for env_variable in env_variables:
+        print(f"{env_variable}='{env_variables[env_variable]}'")

.ci/compute_projects_test.py

@@ -0,0 +1,166 @@
+# Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+# See https://llvm.org/LICENSE.txt for license information.
+# SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+"""Does some stuff."""
+
+import unittest
+
+import compute_projects
+
+
+class TestComputeProjects(unittest.TestCase):
+    def test_llvm(self):
+        env_variables = compute_projects.get_env_variables(
+            ["llvm/CMakeLists.txt"], "Linux"
+        )
+        self.assertEqual(
+            env_variables["projects_to_build"],
+            "bolt;clang;clang-tools-extra;flang;lld;lldb;llvm;mlir;polly",
+        )
+        self.assertEqual(
+            env_variables["project_check_targets"],
+            "check-bolt check-clang check-clang-tools check-flang check-lld check-lldb check-llvm check-mlir check-polly",
+        )
+        self.assertEqual(
+            env_variables["runtimes_to_build"], "libcxx;libcxxabi;libunwind"
+        )
+        self.assertEqual(
+            env_variables["runtimes_check_targets"],
+            "check-cxx check-cxxabi check-unwind",
+        )
+
+    def test_llvm_windows(self):
+        env_variables = compute_projects.get_env_variables(
+            ["llvm/CMakeLists.txt"], "Windows"
+        )
+        self.assertEqual(
+            env_variables["projects_to_build"],
+            "clang;clang-tools-extra;lld;llvm;mlir;polly",
+        )
+        self.assertEqual(
+            env_variables["project_check_targets"],
+            "check-clang check-clang-tools check-lld check-llvm check-mlir check-polly",
+        )
+        self.assertEqual(
+            env_variables["runtimes_to_build"], "libcxx;libcxxabi;libunwind"
+        )
+        self.assertEqual(
+            env_variables["runtimes_check_targets"],
+            "check-cxx check-cxxabi check-unwind",
+        )
+
+    def test_llvm_mac(self):
+        env_variables = compute_projects.get_env_variables(
+            ["llvm/CMakeLists.txt"], "Darwin"
+        )
+        self.assertEqual(
+            env_variables["projects_to_build"],
+            "clang;clang-tools-extra;lld;llvm;mlir",
+        )
+        self.assertEqual(
+            env_variables["project_check_targets"],
+            "check-clang check-clang-tools check-lld check-llvm check-mlir",
+        )
+        self.assertEqual(
+            env_variables["runtimes_to_build"], "libcxx;libcxxabi;libunwind"
+        )
+        self.assertEqual(
+            env_variables["runtimes_check_targets"],
+            "check-cxx check-cxxabi check-unwind",
+        )
+
+    def test_clang(self):
+        env_variables = compute_projects.get_env_variables(
+            ["clang/CMakeLists.txt"], "Linux"
+        )
+        self.assertEqual(
+            env_variables["projects_to_build"],
+            "clang;clang-tools-extra;compiler-rt;lld;llvm",
+        )
+        self.assertEqual(
+            env_variables["project_check_targets"],
+            "check-clang check-clang-tools check-compiler-rt",
+        )
+        self.assertEqual(
+            env_variables["runtimes_to_build"], "libcxx;libcxxabi;libunwind"
+        )
+        self.assertEqual(
+            env_variables["runtimes_check_targets"],
+            "check-cxx check-cxxabi check-unwind",
+        )
+
+    def test_clang_windows(self):
+        env_variables = compute_projects.get_env_variables(
+            ["clang/CMakeLists.txt"], "Windows"
+        )
+        self.assertEqual(
+            env_variables["projects_to_build"], "clang;clang-tools-extra;llvm"
+        )
+        self.assertEqual(
+            env_variables["project_check_targets"], "check-clang check-clang-tools"
+        )
+        self.assertEqual(
+            env_variables["runtimes_to_build"], "libcxx;libcxxabi;libunwind"
+        )
+        self.assertEqual(
+            env_variables["runtimes_check_targets"],
+            "check-cxx check-cxxabi check-unwind",
+        )
+
+    def test_bolt(self):
+        env_variables = compute_projects.get_env_variables(
+            ["bolt/CMakeLists.txt"], "Linux"
+        )
+        self.assertEqual(env_variables["projects_to_build"], "bolt;clang;lld;llvm")
+        self.assertEqual(env_variables["project_check_targets"], "check-bolt")
+        self.assertEqual(env_variables["runtimes_to_build"], "")
+        self.assertEqual(env_variables["runtimes_check_targets"], "")
+
+    def test_lldb(self):
+        env_variables = compute_projects.get_env_variables(
+            ["lldb/CMakeLists.txt"], "Linux"
+        )
+        self.assertEqual(env_variables["projects_to_build"], "clang;lldb;llvm")
+        self.assertEqual(env_variables["project_check_targets"], "check-lldb")
+        self.assertEqual(env_variables["runtimes_to_build"], "")
+        self.assertEqual(env_variables["runtimes_check_targets"], "")
+
+    def test_mlir(self):
+        env_variables = compute_projects.get_env_variables(
+            ["mlir/CMakeLists.txt"], "Linux"
+        )
+        self.assertEqual(env_variables["projects_to_build"], "clang;flang;llvm;mlir")
+        self.assertEqual(
+            env_variables["project_check_targets"], "check-flang check-mlir"
+        )
+        self.assertEqual(env_variables["runtimes_to_build"], "")
+        self.assertEqual(env_variables["runtimes_check_targets"], "")
+
+    def test_flang(self):
+        env_variables = compute_projects.get_env_variables(
+            ["flang/CMakeLists.txt"], "Linux"
+        )
+        self.assertEqual(env_variables["projects_to_build"], "clang;flang;llvm")
+        self.assertEqual(env_variables["project_check_targets"], "check-flang")
+        self.assertEqual(env_variables["runtimes_to_build"], "")
+        self.assertEqual(env_variables["runtimes_check_targets"], "")
+
+    def test_invalid_subproject(self):
+        env_variables = compute_projects.get_env_variables(
+            [".ci/compute_projects.py"], "Linux"
+        )
+        self.assertEqual(env_variables["projects_to_build"], "")
+        self.assertEqual(env_variables["project_check_targets"], "")
+        self.assertEqual(env_variables["runtimes_to_build"], "")
+        self.assertEqual(env_variables["runtimes_check_targets"], "")
+
+    def test_top_level_file(self):
+        env_variables = compute_projects.get_env_variables(["README.md"], "Linux")
+        self.assertEqual(env_variables["projects_to_build"], "")
+        self.assertEqual(env_variables["project_check_targets"], "")
+        self.assertEqual(env_variables["runtimes_to_build"], "")
+        self.assertEqual(env_variables["runtimes_check_targets"], "")
+
+
+if __name__ == "__main__":
+    unittest.main()