Merge #55687

55687: *: bazel-ify cockroachdb r=irfansharif a=irfansharif

This commit introduces a new build system to cockroachdb: Bazel.
The hope here is to eventually replace our current Makefile. It sets the
foundation for introducing Bazel in our CI systems for parallelized
builds/test runs, artifact caching, remote builds, and many other such
profits (literally, it could reduce infra spend).

We originally prototyped a pseudo bazel-ified cockroachdb in #52824,
which resulted in a week long effort (with otan@, jamesl@, and myself) to
wrap it up in earnest in #55258. This is the polished version of the
resulting work, the write up for which can be found at
http://go.crdb.dev/bazel-hackweek.

---

Most of the BUILD files here were auto-generated through Gazelle (see
doc). Some of these auto-generated files required further manual
tweaking to get just right (see doc). These were mostly around the packages that
link into c-dependencies, but also the few packages that rely on
auto-generated code (again, see doc). These specific files are the ones
intended for review (they're also the ones responsible for
auto-generation), among a few others. I've listed them out here:

```
	new file:   .bazelrc
	modified:   .gitattributes
	modified:   .gitignore
	new file:   BUILD.bazel
	new file:   DEPS.bzl
	modified:   Makefile
	new file:   WORKSPACE
	new file:   pkg/ccl/gssapiccl/BUILD.bazel
	new file:   pkg/ccl/storageccl/engineccl/BUILD.bazel
	new file:   pkg/cli/BUILD.bazel
	new file:   pkg/cmd/cockroach-oss/BUILD.bazel
	new file:   pkg/cmd/cockroach-short/BUILD.bazel
	new file:   pkg/cmd/cockroach/BUILD.bazel
	new file:   pkg/col/coldata/BUILD.bazel
	new file:   pkg/geo/geoproj/BUILD.bazel
	modified:   pkg/geo/geoproj/geoproj.go
	modified:   pkg/geo/geoproj/proj.cc
	new file:   pkg/geo/geos/BUILD.bazel
	new file:   pkg/sql/colconv/BUILD.bazel
	new file:   pkg/sql/colexec/BUILD.bazel
	new file:   pkg/sql/colexec/COLEXEC.bzl
	modified:   pkg/sql/colexec/and_or_projection_tmpl.go
	modified:   pkg/sql/colexec/any_not_null_agg_tmpl.go
	modified:   pkg/sql/colexec/avg_agg_tmpl.go
	modified:   pkg/sql/colexec/bool_and_or_agg_tmpl.go
	modified:   pkg/sql/colexec/cast_tmpl.go
	new file:   pkg/sql/colexec/colbuilder/BUILD.bazel
	new file:   pkg/sql/colexec/colexec.go
	modified:   pkg/sql/colexec/concat_agg_tmpl.go
	modified:   pkg/sql/colexec/const_tmpl.go
	modified:   pkg/sql/colexec/count_agg_tmpl.go
	modified:   pkg/sql/colexec/default_agg_tmpl.go
	modified:   pkg/sql/colexec/default_cmp_expr_tmpl.go
	modified:   pkg/sql/colexec/default_cmp_proj_ops_tmpl.go
	modified:   pkg/sql/colexec/default_cmp_sel_ops_tmpl.go
	modified:   pkg/sql/colexec/distinct_tmpl.go
	modified:   pkg/sql/colexec/hash_aggregator_tmpl.go
	modified:   pkg/sql/colexec/hash_utils_tmpl.go
	modified:   pkg/sql/colexec/hashjoiner_tmpl.go
	modified:   pkg/sql/colexec/hashtable_tmpl.go
	modified:   pkg/sql/colexec/is_null_ops_tmpl.go
	modified:   pkg/sql/colexec/mergejoinbase_tmpl.go
	modified:   pkg/sql/colexec/mergejoiner_tmpl.go
	modified:   pkg/sql/colexec/min_max_agg_tmpl.go
	modified:   pkg/sql/colexec/ordered_synchronizer_tmpl.go
	modified:   pkg/sql/colexec/proj_const_ops_tmpl.go
	modified:   pkg/sql/colexec/proj_non_const_ops_tmpl.go
	modified:   pkg/sql/colexec/quicksort_tmpl.go
	modified:   pkg/sql/colexec/rank_tmpl.go
	modified:   pkg/sql/colexec/relative_rank_tmpl.go
	modified:   pkg/sql/colexec/row_number_tmpl.go
	modified:   pkg/sql/colexec/rowstovec_tmpl.go
	modified:   pkg/sql/colexec/select_in_tmpl.go
	modified:   pkg/sql/colexec/selection_ops_tmpl.go
	modified:   pkg/sql/colexec/sort_tmpl.go
	modified:   pkg/sql/colexec/substring_tmpl.go
	modified:   pkg/sql/colexec/sum_agg_tmpl.go
	modified:   pkg/sql/colexec/utils_tmpl.go
	modified:   pkg/sql/colexec/values_differ_tmpl.go
	modified:   pkg/sql/colexec/vec_comparators_tmpl.go
	modified:   pkg/sql/colexec/window_peer_grouper_tmpl.go
	new file:   pkg/sql/colflow/BUILD.bazel
	new file:   pkg/sql/opt/BUILD.bazel
	new file:   pkg/sql/opt/exec/BUILD.bazel
	new file:   pkg/sql/opt/exec/execbuilder/BUILD.bazel
	new file:   pkg/sql/opt/exec/explain/BUILD.bazel
	new file:   pkg/sql/opt/memo/BUILD.bazel
	new file:   pkg/sql/opt/norm/BUILD.bazel
	new file:   pkg/sql/opt/optgen/lang/BUILD.bazel
	new file:   pkg/sql/opt/xform/BUILD.bazel
	new file:   pkg/sql/parser/BUILD.bazel
	new file:   pkg/sql/parser/sql-gen.sh
	modified:   pkg/sql/parser/sql.y
	new file:   pkg/storage/BUILD.bazel
	modified:   vendor
```

The changes to all the _tmpl.go files were needed to include necessary
import analysis/compilation for bazel when generating the .eg.go files
(also summarized in doc up above), and should be uncontroversial.

I think it's easiest to pull the branch down locally and to click around
to see what works, and what doesn't. I would start at the top-level
BUILD.bazel and WORSPACE file. I've added comments to explain what most
of added machinery does.

Release note: None


Co-authored-by: irfan sharif <[email protected]>

.bazelrc

@@ -0,0 +1,9 @@
+# Bazel's usual sandboxing does not try to protect against accidental
+# dependencies on # files from the host filesystem. See [1]. This is a problem
+# for us, for when users have `proj` installed locally. They can run into
+# issues where bazel tries to use `/usr/local/include/proj_api.h` instead of
+# `bazel-build/.../c-deps/proj/lib`. The work around here, suggested by [1], is
+# to ensure the sandbox does not read /usr/local/include.
+#
+# [1]: https://github.com/bazelbuild/bazel/issues/8053.
+build --sandbox_block_path=/usr/local/include

.gitattributes

@@ -1,2 +1,3 @@
 *.pb.* -diff
 *.eg.go -diff
+DEPS.bzl -diff

BUILD.bazel

@@ -0,0 +1,215 @@
+load("@bazel_gazelle//:def.bzl", "gazelle")
+
+# The following directives inform gazelle how to auto-generate BUILD.bazel
+# files throughout the repo. By including them here, we can run gazelle using
+# `bazel run //:gazelle` instead of invoking gazelle directly.
+#
+# NB: Be sure to re-generate BUILD.bazel files if any of the directives below
+# are changed.
+
+# Define the gazelle prefix and what the autogenerated BUILD files should be
+# named as.
+#
+# gazelle:prefix github.com/cockroachdb/cockroach
+# gazelle:build_file_name BUILD.bazel
+
+# We disable protobuf generation for our dependencies
+#
+# gazelle:proto disable_global
+
+# Gazelle is unable to resolve this specific package.
+#
+# TODO(irfansharif): I'm not sure why this is. Is it because it's a proto only
+# package?
+#
+# gazelle:resolve go github.com/grpc-ecosystem/grpc-gateway/internal //vendor/github.com/grpc-ecosystem/grpc-gateway/internal
+
+# We exclude a few things from gazelle consideration:
+# - The protobuf C dependency, it's are already bazel-ified, lest we overwrite
+#   those build files.
+# - The artifacts directory
+# - All checked-in autogenerated files (they'll get autogenerated on the fly
+#   through bazel).
+# - A testdata "repo" that looks like a go package, but isn't
+#   (testdata/src/example.com)
+#
+# gazelle:exclude c-deps/protobuf
+# gazelle:exclude artifacts
+# gazelle:exclude **/zcgo_flags.go
+# gazelle:exclude **/zcgo_flags_*.go
+# gazelle:exclude **/*.og.go
+# gazelle:exclude **/*.eg.go
+# gazelle:exclude pkg/sql/parser/sql.go
+# gazelle:exclude pkg/sql/opt/rule_name_string.go
+# gazelle:exclude pkg/sql/opt/rule_name_string.go
+# gazelle:exclude pkg/cmd/prereqs/testdata
+
+# TODO(irfansharif): Today we let bazel take over the vendor directory and
+# define each vendored dependency as a bazel target. We should be able to have
+# bazel ignore all of vendor/ and create those dependencies on the fly.
+# Deleting the vendor directory and running `bazel run //:gazelle` shows what
+# that would look like (though that doesn't quite work yet).
+
+# TODO(irfansharif): Document a few usage patterns for bazel and how to
+# understand all the autogen stuff. Probably as a tech note. Here are a few
+# short hands I've used so far:
+#
+#   bazel test //pkg/kv/kvserver/concurrency/...
+#   bazel build //pkg/sql/opt
+#   bazel build //pkg/sql/opt:all
+#   bazel build //pkg/sql/colexec:gen-exec
+#   bazel build //pkg/cmd/cockroach-short
+#   bazel build //:libjemalloc
+#   bazel query //pkg/sql/colexec:all
+#   bazel run //:gazelle
+#   bazel run //pkg/cmd/cockroach-short -- demo
+#   bazel run //pkg/sql/opt/optgen/cmd/langgen -- -h
+#   bazel run //:gazelle -- update-repos -from_file=go.mod -to_macro=DEPS.bzl%go_deps
+#
+# The //<stuff> names can also be fully qualified using @cockroach, and that
+# appears in certain parts of the codebase/elsewhere. Specifically it'll look
+# like:
+#
+#   bazel build @cockroach//pkg/cmd/cockroach-short
+#
+# TODO(irfansharif): We should probably define shorthands for the more common
+# ones from up above.
+
+# TODO(irfansharif): The way we currently generate code through bazel, that
+# code is only available within the bazel sandbox. Bazel ignores all
+# the pre-generated code that is already checked into the codebase (through
+# `make generate`/etc.)through the exclude directives above. It's generating
+# everything on the fly.
+#
+# As we move towards bazel, we'll want to introduce a mechanism that implants
+# the generated code within the sandbox placing them "back into" the
+# appropriate packages. This is to ensure we don't break existing IDEs and code
+# editors which rely on files existing in the same package itself (as opposed
+# to the sandbox). This way we could continue checking in auto-generated code.
+# We should provide a bazel alternative for `make generate` that does this very
+# same thing. See [1], this is a long standing issue for folks using using Go,
+# bazel, and autogenerated code.
+#
+# [1]: https://github.com/bazelbuild/rules_go/issues/512.
+
+# TODO(irfansharif): We'll need to pin toolchains somewhere to make sure
+# everything below works as expected.
+
+gazelle(
+    name = "gazelle",
+    prefix = "github.com/cockroachdb/cockroach",
+)
+
+# Load the components that lets us use cmake/make in third party deps.
+load("@rules_foreign_cc//tools/build_defs:cmake.bzl", "cmake_external")
+load("@rules_foreign_cc//tools/build_defs:configure.bzl", "configure_make")
+
+# TODO(irfansharif): All the cmake cache entries below were cargo-culted from
+# the Makefile. We may need to programatically set them depending on the
+# machine bazel is being run on. For now the only tested variant are Macbooks.
+
+# Define the build target for libjemalloc.
+configure_make(
+    name = "libjemalloc",
+    autoconf = True,
+    configure_env_vars = {
+        "AR": "",
+    },
+    configure_in_place = True,
+    configure_options = [
+        "--enable-prof",
+    ],
+    lib_source = "@jemalloc//:all",
+    make_commands = [
+        "make build_lib_static",
+        "mkdir -p libjemalloc/lib",
+        "cp lib/libjemalloc.a libjemalloc/lib",
+        "cp -r include libjemalloc",
+    ],
+    static_libraries = ["libjemalloc.a"],
+    visibility = ["//visibility:public"],
+)
+
+# Define the build targets for libprotobuf and protoc.
+cmake_external(
+    name = "libprotobuf",
+    binaries = ["protoc"],
+    cache_entries = {
+        "CMAKE_BUILD_TYPE": "Release",
+        "CMAKE_TARGET_MESSAGES": "OFF",
+        "protobuf_BUILD_TESTS": "OFF",
+    },
+    lib_source = "@protobuf//:all",
+    static_libraries = ["libprotobuf.a"],
+    visibility = ["//visibility:public"],
+    working_directory = "cmake",
+)
+
+# Define the build target for libproj.
+cmake_external(
+    name = "libproj",
+    cache_entries = {
+        "CMAKE_BUILD_TYPE": "Release",
+        "BUILD_LIBPROJ_SHARED": "OFF",
+    },
+    lib_source = "@proj//:all",
+    static_libraries = ["libproj.a"],
+    visibility = ["//visibility:public"],
+)
+
+# TODO(irfansharif): libgeos has not been worked out yet. We'll need to
+# similarly ensure the lib/libgeos.so and lib/libegeos_c.so are in the right
+# place.
+
+# Define the build target for libroach.
+#
+# Bazel also expects the library archive and the include headers to be placed
+# in a certain path, so we fix it all up accordingly within make_commands.
+cmake_external(
+    name = "libroach",
+    cache_entries = {
+        "CMAKE_TARGET_MESSAGES": "OFF",
+        "CMAKE_BUILD_TYPE": "Release",
+    },
+    lib_source = "@libroach//:all",
+    make_commands = [
+        "make roach",
+        "mkdir -p libroach/lib",
+        "cp libroach.a libroach/lib/libroach.a",
+        "cp -r $EXT_BUILD_ROOT/external/libroach/include libroach",
+    ],
+    static_libraries = ["libroach.a"],
+    tools_deps = [
+        "@libroach//:all",
+    ],
+    visibility = ["//visibility:public"],
+    deps = [],
+)
+
+# Define the build target for kerberos.
+#
+# TODO(irfansharif): Kerboros is not used for anything other than Linux.
+# The following has not been tested as yet.
+configure_make(
+    name = "libkrb5",
+    autoreconf = True,
+    configure_command = "src/configure",
+    configure_env_vars = {
+        "CPFLAGS": "",
+        "CXXFLAGS": "",
+    },
+    configure_options = [
+        "--enable-static",
+        "--disable-shared",
+    ] + select({
+        "//conditions:default": ["AR=/usr/bin/ar"],
+    }),
+    lib_source = "@krb5//:all",
+    make_commands = [
+        "make",
+        "mkdir -p libkrb5/lib",
+        "cp libkrb5/libgssapi_krb5.a libkrb5/lib",
+    ],
+    static_libraries = ["libgssapi_krb5.a"],
+    visibility = ["//visibility:public"],
+)