bazel: remove rules_cc

[keith]

The plans for bazel to move to rules_cc have been postponed without any
communication. There's no value to us in using this right now, but it
will be trivial to re-adopt in the future if needed. But it has the
downside of using a fork of bazel's crosstool, that has to be updated
independently of bazel, which doesn't always happen as improvements are
made.

More details: bazelbuild/rules_cc#86 bazelbuild/bazel#14150

This also required a --host_action_env addition to mirror the variables we
pass through to actions in general. This is required because C++ toolchain
setup which discovers linkers in the cc configuration which uses PATH directly,
but then host actions didn't use PATH because of
--incompatible_strict_action_env, so gcc couldn't discover the path of lld even
though -fuse-ld=lld was passed.

Fixes #16608

Signed-off-by: Keith Smiley keithbsmiley@gmail.com

[repokitteh-read-only]

CC @envoyproxy/dependency-shepherds: Your approval is needed for changes made to (bazel/.*repos.*\.bzl)|(bazel/dependency_imports\.bzl)|(api/bazel/.*\.bzl)|(.*/requirements\.txt)|(.*\.patch).
envoyproxy/dependency-shepherds assignee is @moderation

🐱

Caused by: #19760 was opened by keith.

see: more, trace.

[keith]

Hoping we're good here now that we're on 5.x, because we were waiting for some upstream change there

[phlax]

this is breaking due to gcc linking issues

ERROR: /build/tmp/_bazel_envoybuild/b570b5ccd0454dc9af9f65ab1833764d/external/com_google_protobuf/BUILD:513:10: Linking external/com_google_protobuf/protoc failed: (Exit 1): gcc failed: error executing command 

not sure of the exact cause, but as discussed offline i think removing rules_cc is breaking the gcc toolchain as we were seeing a similar symptom during the bazel 5 upgrade

[keith]

I think I got to the bottom of that and updated the description accordingly, we'll see how CI feels

[keith]

Depends on #20814

[keith]

/retest

[repokitteh-read-only]

Retrying Azure Pipelines:
Check envoy-presubmit isn't fully completed, but will still attempt retrying.
Retried failed jobs in: envoy-presubmit

🐱

Caused by: a #19760 (comment) was created by @keith.

see: more, trace.

[keith]

/retest

[repokitteh-read-only]

Retrying Azure Pipelines:
Retried failed jobs in: envoy-presubmit

🐱

Caused by: a #19760 (comment) was created by @keith.

see: more, trace.

[keith]

Ok this is ready for review! The key here was to update buildifier to no longer try to enforce using rules_cc (they used to but flipped the default since things have stalled), and to pass through the host_action_env to make sure the setup uses the right tools

[moderation]

Always like deps removal 👍

/lgtm deps

[keith]

bump

[htuch]

LGTM, thanks!

[adisuissa]

Following this PR, a few of us have started seeing differences between running the check_format.py tool with docker and without.

When running check_format using docker (ci/run_envoy_docker.sh './ci/do_ci.sh check_format'), it runs well and doesn't complain.
When running without docker (tools/code_format/check_format.py check), it will complain about BUILD files which are missing some definitions. For example, it will complain that the file "source/common/protobuf/BUILD" is missing cc_proto_library as it is being used.

It seems that after this PR the cc_proto_library definition is missing from the file. Should this be reverted?
@phlax can you think why this PR could show different results when running with and without docker?

cc @jmarantz @sergiitk

[keith]

Can you update buildifier locally? There was a default flip that we're relying on that I update CI for. I can probably also bake that into the script to avoid this issue for older versions. Ideally we would start vendoring buildifier to avoid this.

[adisuissa]

Upgrading buildifier fixed this, thanks!