[ABLD-395] Framework to create toolchains for OS provided tools

[aiuto]

What does this PR do?

• Creates a tiny framework for finding an OS provided tool and telling Bazel it is a toolchain.
• Use it for macos pkgbuild and codesign.

The important feature of the way this is implemented is that the toolchain always safely resolves on all platforms, so you can bazel cquery across platforms, even though you will fail if you try to build. Combined with exec_compatible_with you can make targets that only build if the required tools are available. This is important when there are different tools available as OSes evolve (or are simply different by vendor, like debian and redhat) and we need to create paths that can gracefully use different tools depending on what is available.

There is also a capability to use a bazel target as the tool implementation instead of looking it up in $PATH. That is useful for building alternate implementations (let's say a windows tool on linux). That will come into use in full remote execution mode, when the target and exec hosts can be different architectures.

Examples:

• I'm building a tarball, see if I have the xz toolchain and use that if available, otherwise fall back to a slower compresser
• The OS is evolving a tool, if we have version 1, it might have one name. In version 2, there might be a better tool. We want to make rules that can do the right thing regardless of what OS we have. This is a huge win for external users who may be using a different base linux to build.

Motivation

This is an important part of our supply chain security posture. We want to account for every binary we use to build the product. Ideally that would include even basic low level things, like bash and cat. In practice we can make the tradeoff to stop at declaring the things that might vary across OSes or from OS release to release.

Eventually, we will expand this the framework to include package metadata (https://github.com/bazel-contrib/supply-chain/tree/main/metadata) that can be easily gathered into a workspace BOM.

The more immediate need is to create toolchains for pkgbuild and codesign to create rules to make macos packages. #54115

Describe how you validated your changes

With the BUILD file

genrule(
    name = "test",
    cmd = select({
        "@macos_codesign//:have_codesign": "echo GOT IT",
        "//conditions:default": "echo wump, wump, wump",
    }),
    outs = ["test.out"],
)

macos:

$ bazel cquery --output=build  //ztony:test
genrule(
  name = "test",
  outs = ["//ztony:test.out"],
  cmd = "echo GOT IT",
)
$ bazel cquery @macos_codesign//:all
INFO: Analyzed 7 targets (0 packages loaded, 6 targets configured).
INFO: Found 7 targets...
@macos_codesign//:codesign_auto (eab6d8f)
@macos_codesign//:codesign_toolchain (eab6d8f)
@macos_codesign//:codesign_toolchain_type (eab6d8f)
@macos_codesign//:have_codesign (eab6d8f)
@macos_codesign//:is_codesign_available (eab6d8f)
@macos_codesign//:no_codesign (eab6d8f)
@macos_codesign//:zzz_codesign_missing_toolchain (eab6d8f)

linux:

$ bazel cquery --output=build  //ztony:test
genrule(
  name = "test",
  outs = ["//ztony:test.out"],
  cmd = "echo wump, wump, wump",
)
$ bazel cquery @macos_codesign//:all
@macos_codesign//:codesign_auto (a938602)
@macos_codesign//:codesign_toolchain (a938602)
@macos_codesign//:codesign_toolchain_type (a938602)
@macos_codesign//:have_codesign (a938602)
@macos_codesign//:is_codesign_available (a938602)
@macos_codesign//:no_codesign (a938602)
@macos_codesign//:zzz_codesign_missing_toolchain (a938602)

Additional

This is an alternate version of #54154. I think I like it better.
[ABLD-386]: https://datadoghq.atlassian.net/browse/ABLD-386

[dd-octo-sts]

Files inventory check summary

File checks results against ancestor 9569d8dc:

Results for datadog-agent_7.83.0~devel.git.536.d9d76a4.pipeline.128686930-1_amd64.deb:

No change detected

[dd-octo-sts]

Static quality checks

✅ Please find below the results from static quality gates
Comparison made with ancestor 9569d8d
📊 Static Quality Gates Dashboard
🔗 SQG Job

33 successful checks with minimal change (< 2 KiB)

	Quality gate	Current Size
✅	agent_deb_amd64	753.928 MiB
✅	agent_deb_amd64_fips	707.604 MiB
✅	agent_heroku_amd64	308.200 MiB
✅	agent_msi	639.265 MiB
✅	agent_rpm_amd64	753.912 MiB
✅	agent_rpm_amd64_fips	707.588 MiB
✅	agent_rpm_arm64	730.352 MiB
✅	agent_rpm_arm64_fips	687.280 MiB
✅	agent_suse_amd64	753.912 MiB
✅	agent_suse_amd64_fips	707.588 MiB
✅	agent_suse_arm64	730.352 MiB
✅	agent_suse_arm64_fips	687.280 MiB
✅	docker_agent_amd64	812.582 MiB
✅	docker_agent_arm64	814.086 MiB
✅	docker_agent_jmx_amd64	1003.480 MiB
✅	docker_agent_jmx_arm64	993.636 MiB
✅	docker_cluster_agent_amd64	209.998 MiB
✅	docker_cluster_agent_arm64	223.077 MiB
✅	docker_cws_instrumentation_amd64	7.439 MiB
✅	docker_cws_instrumentation_arm64	6.877 MiB
✅	docker_dogstatsd_amd64	39.304 MiB
✅	docker_dogstatsd_arm64	37.433 MiB
✅	docker_host_profiler_amd64	305.307 MiB
✅	docker_host_profiler_arm64	316.621 MiB
✅	dogstatsd_deb_amd64	30.045 MiB
✅	dogstatsd_deb_arm64	28.085 MiB
✅	dogstatsd_rpm_amd64	30.045 MiB
✅	dogstatsd_suse_amd64	30.045 MiB
✅	iot_agent_deb_amd64	46.274 MiB
✅	iot_agent_deb_arm64	42.957 MiB
✅	iot_agent_deb_armhf	43.722 MiB
✅	iot_agent_rpm_amd64	46.274 MiB
✅	iot_agent_suse_amd64	46.273 MiB

[cit-pr-commenter-54b7da]

Regression Detector

Regression Detector Results

Metrics dashboard
Target profiles
Run ID: c996791b-7301-426a-8631-256a959aeb77

Baseline: 29a80d4
Comparison: c2cf4c2
Diff

Optimization Goals: ✅ No significant changes detected

Fine details of change detection per experiment

perf	experiment	goal	Δ mean %	Δ mean % CI	trials	links
➖	quality_gate_logs	% cpu utilization	+1.04	[+0.17, +1.91]	1	Logs bounds checks dashboard
➖	quality_gate_metrics_logs	memory utilization	+0.73	[+0.48, +0.97]	1	Logs bounds checks dashboard
➖	quality_gate_idle	memory utilization	+0.14	[+0.09, +0.18]	1	Logs bounds checks dashboard
➖	quality_gate_security_idle	memory utilization	+0.09	[+0.04, +0.14]	1	Logs bounds checks dashboard
➖	quality_gate_security_no_fs_load	memory utilization	+0.05	[-0.04, +0.14]	1	Logs bounds checks dashboard
➖	quality_gate_security_mean_fs_load	memory utilization	-0.09	[-0.12, -0.05]	1	Logs bounds checks dashboard
➖	quality_gate_idle_all_features	memory utilization	-0.17	[-0.21, -0.13]	1	Logs bounds checks dashboard
➖	quality_gate_private_action_runner	memory utilization	-0.29	[-0.41, -0.17]	1	Logs bounds checks dashboard

Bounds Checks: ✅ Passed

perf	experiment	bounds_check_name	replicates_passed	observed_value	links
✅	quality_gate_idle	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle	memory_usage	10/10	148.31MiB ≤ 154MiB	bounds checks dashboard
✅	quality_gate_idle	total_bytes_received	10/10	731.79KiB ≤ 819.20KiB	bounds checks dashboard
✅	quality_gate_idle_all_features	intake_connections	10/10	3 ≤ 4	bounds checks dashboard
✅	quality_gate_idle_all_features	memory_usage	10/10	492.20MiB ≤ 512MiB	bounds checks dashboard
✅	quality_gate_idle_all_features	total_bytes_received	10/10	1.12MiB ≤ 1.25MiB	bounds checks dashboard
✅	quality_gate_logs	intake_connections	10/10	3 ≤ 6	bounds checks dashboard
✅	quality_gate_logs	memory_usage	10/10	186.67MiB ≤ 195MiB	bounds checks dashboard
✅	quality_gate_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_logs	total_bytes_received	10/10	264.34MiB ≤ 292MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	cpu_usage	10/10	363.65 ≤ 2000	bounds checks dashboard
✅	quality_gate_metrics_logs	intake_connections	10/10	3 ≤ 6	bounds checks dashboard
✅	quality_gate_metrics_logs	memory_usage	10/10	414.19MiB ≤ 430MiB	bounds checks dashboard
✅	quality_gate_metrics_logs	missed_bytes	10/10	0B = 0B	bounds checks dashboard
✅	quality_gate_metrics_logs	total_bytes_received	10/10	0.93GiB ≤ 1.04GiB	bounds checks dashboard
✅	quality_gate_private_action_runner	memory_usage	10/10	71.14MiB ≤ 75MiB	bounds checks dashboard
✅	quality_gate_security_idle	cpu_usage	10/10	27.62 ≤ 100	bounds checks dashboard
✅	quality_gate_security_idle	memory_usage	10/10	301.85MiB ≤ 330MiB	bounds checks dashboard
✅	quality_gate_security_mean_fs_load	cpu_usage	10/10	60.29 ≤ 200	bounds checks dashboard
✅	quality_gate_security_mean_fs_load	memory_usage	10/10	280.39MiB ≤ 310MiB	bounds checks dashboard
✅	quality_gate_security_no_fs_load	cpu_usage	10/10	20.42 ≤ 100	bounds checks dashboard
✅	quality_gate_security_no_fs_load	memory_usage	10/10	283.60MiB ≤ 320MiB	bounds checks dashboard

Explanation

Confidence level: 90.00%
Effect size tolerance: |Δ mean %| ≥ 5.00%

Performance changes are noted in the perf column of each table:

• ✅ = significantly better comparison variant performance
• ❌ = significantly worse comparison variant performance
• ➖ = no significant change in performance

A regression test is an A/B test of target performance in a repeatable rig, where "performance" is measured as "comparison variant minus baseline variant" for an optimization goal (e.g., ingress throughput). Due to intrinsic variability in measuring that goal, we can only estimate its mean value for each experiment; we report uncertainty in that value as a 90.00% confidence interval denoted "Δ mean % CI".

For each experiment, we decide whether a change in performance is a "regression" -- a change worth investigating further -- if all of the following criteria are true:

1. Its estimated |Δ mean %| ≥ 5.00%, indicating the change is big enough to merit a closer look.

2. Its 90.00% confidence interval "Δ mean % CI" does not contain zero, indicating that if our statistical model is accurate, there is at least a 90.00% chance there is a difference in performance between baseline and comparison variants.

3. Its configuration does not mark it "erratic".

Replicate Execution Details

We run multiple replicates for each experiment/variant. However, we allow replicates to be automatically retried if there are any failures, up to 8 times, at which point the replicate is marked dead and we are unable to run analysis for the entire experiment. We call each of these attempts at running replicates a replicate execution. This section lists all replicate executions that failed due to the target crashing or being oom killed.

Note: In the below tables we bucket failures by experiment, variant, and failure type. For each of these buckets we list out the replicate indexes that failed with an annotation signifying how many times said replicate failed with the given failure mode. In the below example the baseline variant of the experiment named experiment_with_failures had two replicates that failed by oom kills. Replicate 0, which failed 8 executions, and replicate 1 which failed 6 executions, all with the same failure mode.

Experiment	Variant	Replicates	Failure	Logs	Debug Dashboard
experiment_with_failures	baseline	0 (x8) 1 (x6)	Oom killed		Debug Dashboard

The debug dashboard links will take you to a debugging dashboard specifically designed to investigate replicate execution failures.

❌ Retried Profiling Replicate Execution Failures (ddprof)

Note: Profiling replicas may still be executing. See the debug dashboard for up to date status.

Experiment	Variant	Replicates	Failure	Debug Dashboard
quality_gate_idle_all_features	baseline	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	baseline	10	Oom killed	Debug Dashboard
quality_gate_metrics_logs	comparison	10	Oom killed	Debug Dashboard

CI Pass/Fail Decision

✅ Passed. All Quality Gates passed.

• quality_gate_metrics_logs, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_metrics_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle_all_features, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_private_action_runner, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_mean_fs_load, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_mean_fs_load, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_no_fs_load, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_no_fs_load, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check total_bytes_received: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check memory_usage: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check missed_bytes: 10/10 replicas passed. Gate passed.
• quality_gate_logs, bounds check intake_connections: 10/10 replicas passed. Gate passed.
• quality_gate_security_idle, bounds check cpu_usage: 10/10 replicas passed. Gate passed.
• quality_gate_security_idle, bounds check memory_usage: 10/10 replicas passed. Gate passed.

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: c6e763d290

ℹ️ About Codex in GitHub

Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".

[alopezz]

This is probably mostly fine, but I need a bit more time to make sure I understand enough of how the toolchain missing case is being handled. In the meantime I do have a suggestion of two.

[alopezz]

This name seems to describe the function poorly. Probably something like local_toolchain_repository would be more apt.

[aiuto]

It's not that either, because it doesn't make the repository. It returns a rule, that you have to call to make the thing you want. So it's really a builder pattern. Think of the expansion in MODULE.bazel. It would be

# This next line is the effect of load(..., "find_macos_pkgbuild")
find_macos_pkgbuild = make_repo_builder(name = NAME, tool_name = "pkgbuild")
# This calls the function to make the repository.
find_macos_pkgbuild(
    name = "macos_pkgbuild",
    exec_compatible_with = ["@platforms//os:macos"],
)

So, maybe we rename find_macos_pkgbuild. Maybe "os_provided_macos_pkgbuild".
That's wordy, but more accurate.

[alopezz]

That's fair, but the name is still a bit too generic, it could be something like make_local_toolchain_repository_rule or make_os_provided_tool_repo_rule or similar, at least something that points at this being intended to wrap local tools would be desirable in my opinion.

[aiuto]

Changed to make_local_toolchain_repository_rule

[alopezz]

A couple more questions.

[alopezz]

And another question that I have is whether we actually need to be templated with the TOOL_NAME baked in or whether a single common rule can be used to define all these toolchains instead.

[aiuto]

For just the two I have done, yes. But that gets in the way of future expansion, where I want different things to customize the template. pkgbuild and codesign, do not have version commands, but for tools that do, I want the repo rule to extract the version (which has to be tool specific) and put that in as metadata so we can have the SBOM say things like otool: cctools-1030.6.3

I have that ready in another PR, but it's not needed yet. Essentially the repo_builder call will use a private impl function instead of the default, and your private impl will call do the lookup and then call write_toolchain_repo to emit the repo.

[chouquette]

LGTM but codex seems to have a good point

[alopezz]

One final question.

[alopezz]

Did you check whether it can actually be used in exec_compatible_with clauses?

According to https://bazel.build/reference/be/platforms-and-toolchains#toolchain_args, only constraint_setting's can be passed here, and the attribute is nonconfigurable.

[aiuto]

You're right it can't be used there. I got ahead of myself. We were going to unify constraint_value and config_setting but never got around to it. A rule using it would have to examine the valid field.
I fixed the comment.