Lberki 7.4.0 analysis cache discard #23652
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…Unix cc_toolchain (#21798) This was missed with a9e5a32. Without this change, `gcc_quoting_for_param_files` does nothing, because `FeatureConfiguration.isEnabled()` will always return false. Without `gcc_quoting_for_param_files`, linking can fail for files with spaces in them. Closes #21699. PiperOrigin-RevId: 617092387 Change-Id: Ifcd4378d2120e2e75e2e72a4b0059b41ba5ce833 Commit bd2da6e Co-authored-by: Brentley Jones <github@brentleyjones.com> Co-authored-by: Ian (Hee) Cha <heec@google.com>
…m operations (#21861) This requires making the implementation of `renameTo` in `JavaIoFileSystem` and `WindowsFileSystem` atomic. Fixes #21823 Closes #21779. PiperOrigin-RevId: 620719510 Change-Id: Id3ff4baa56c6ef693196e8268614b483713542a5 Commit 753dc97 Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
Original commits: * 239b89a * 707d065 * b532a46 * e3bd725 * 5dc6170 * 982b9ad * 40824e1 * 594bd12 * 849da31 * 11498b5 * 55ec5ea * ad8ea5a * 6f47024 * 8024a75 * 3798b55 * 6783339 * 59cf396 --------- Signed-off-by: iancha1992 <heec@google.com> Co-authored-by: Googler <tanzhengwei@google.com> Co-authored-by: Zheng Wei Tan <tanzhengwei143@gmail.com> Co-authored-by: Googler <jingwen@google.com> Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im> Co-authored-by: Googler <chiwang@google.com> Co-authored-by: Yun Peng <pcloudy@google.com> Co-authored-by: Googler <cushon@google.com> Co-authored-by: iancha1992 <heec@google.com> Co-authored-by: hvd <hvadehra@gmail.com> Co-authored-by: Sren Gjesse <sgjesse@google.com>
#21892) …lTest to prevent breakage when running on JDK 21." Amazingly, this fixes the build for release-7.2.0. For some reason, without this PR, BazelAndroidLocalTestTest fails with some sort of "artifact does not have generating action" error: https://storage.googleapis.com/bazel-untrusted-buildkite-artifacts/018ea0c4-55df-44fe-8f65-cc40d8f29482/src/test/java/com/google/devtools/build/lib/bazel/rules/android/AndroidTests/test.log But that error goes away as soon as I delete the `if` block at line 73. This reverts commit c70a4b7.
This is in preparation for tracking the hashes of remote (non-`file:` URL) files in the lockfile. If the tests use local registries, they wouldn't be representative of the default situation. Work towards #20369 Closes #21906. PiperOrigin-RevId: 622290283 Change-Id: Ibe825d2bede84c1b0672dbb699aaf3ee5168a813 Commit 3f9d80c Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
If `--experimental_remote_output_service` is set, Bazel issues a StartBuild RPC to the output service. `StartBuildResponse.initial_output_path_contents` still needs to be handled in a follow-up change. Highlights: - Add a new class `BazelOutputService` which is exclusive to `RemoteOutputService`. Only `startBuild` is implemented by this PR. Other methods will be implemented in follow-up PRs. - Add `--experimental_remote_output_service_output_path_prefix` flag which can be used to set the value of `StartBuildRequest.output_path_prefix` (defaults to empty string). - Reuse `RemoteModule#createChannel` method to create `ReferenceCountedChannel` for `BazelOutputService`. Since there is no server capabilities associated with this channel, the parameter is changed to nullable. - In `RemoteExecutionService#downloadOutputs`, the code skip downloading outputs if `BazelOutputService` is available, instead it calls `stageArtifacts` (although, it currently does nothing). Tests will be added once a MVP is done. Working towards #21630. Closes #21674. PiperOrigin-RevId: 617224074 Change-Id: I872ed06387e90182de2c7c69175989e9a0f663ce Closes #21738.
If `--experimental_remote_output_service` is set, Bazel issue a FinalizeBuild RPC to the output service after the build is done. Working towards #21630. Closes #21739. PiperOrigin-RevId: 617487580 Change-Id: Ib07e9c97f74dfa2fe6f78ec46f7674f59e39ae12 Commit c8187ac Co-authored-by: Chi Wang <chiwang@google.com>
If `--experimental_remote_output_service` is set, Bazel issues a Clean RPC to the output service when running `bazel clean`. Working towards #21630. Closes #21743. PiperOrigin-RevId: 617501624 Change-Id: Idf83b45913f5b35acfb1a29231f2fa7e5aedc850 Commit 3b6aac0 Co-authored-by: Chi Wang <chiwang@google.com>
If `--experimental_remote_output_service` is set, Bazel issue StageArtifacts RPC to the output service after hitting the remote cache or executing action remotely. Working towards #21630. Closes #21746. PiperOrigin-RevId: 617834441 Change-Id: I85fb544a69d0ca779a793038ff0e6a4b755dd637 Commit 4484f86 Co-authored-by: Chi Wang <chiwang@google.com>
If `--experimental_remote_output_service` is set, Bazel issues a FinalizeArtifacts RPC to the output service after an action has been executed. Working towards #21630. Closes #21759. PiperOrigin-RevId: 618190373 Change-Id: I8ed981fcea7ebe146de0b1c48b1933392c86253f Commit 586bd13 Co-authored-by: Chi Wang <chiwang@google.com>
…#21946) If `--experimental_remote_output_service` is set, a `XattrProvider` which uses `BatchStat` RPC to get the digest of a file will be used for Bazel to get digests for outputs. Working towards #21630. Closes #21812. PiperOrigin-RevId: 619539349 Change-Id: I71364e78c45c37e834c71e120a109c8de602db77 Commit 30aa5d0 Co-authored-by: Chi Wang <chiwang@google.com>
This is required by the language spec, but was not implemented in Bazel. See https://github.com/bazelbuild/starlark/blob/master/spec.md#max Fixes #15022 Also take the opportunity to adjust sorted's signature for `key` to match. RELNOTES: Starlark `min` and `max` buitins now allow a `key` callback, similarly to `sorted`. PiperOrigin-RevId: 623547043 Change-Id: I71d44aa715793f9f2260f9b20b876694154ff352 Commit cf66672 Co-authored-by: Googler <arostovtsev@google.com>
If `--experimental_remote_output_service` is set, `BatchStat` RPC will be used by Bazel to check stat of many output files at once during the start of incremental builds. Working towards #21630. Closes #21870. PiperOrigin-RevId: 621500767 Change-Id: I5473d6ac0992215b1fe320b3b76c6510575476cd Commit 9e37f4e Co-authored-by: Chi Wang <chiwang@google.com>
Context: https://bazelbuild.slack.com/archives/C014RARENH0/p1709547684815279?thread_ts=1709324251.252319&cid=C014RARENH0 Closes #21938. PiperOrigin-RevId: 623421163 Change-Id: I450b1b1e3b3c0544b391e571911f62f0709aa62c (cherry picked from commit c8fcfd4)
The argument was already available on 'print' and is necessary for full control over the formatting of error messages that rely on StarlarkValue's `debugPrint`. Work towards #20486 Closes #21961. PiperOrigin-RevId: 625022818 Change-Id: I895b5844d8543f936fc31d367b12bc6291a10bf8 Commit 6b2e0db Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
This makes the build event stream and starlark debugging proto libraries public. This is for consumption by the vscode-bazel plugin, which already uses these but in a very hacky way. Fixes #16335 Fixes #3684 Closes #21196. PiperOrigin-RevId: 623740758 Change-Id: Ic47d6dd05fcf2af8beec4d9212d4c7fd65bcda44 Commit adfe016 Co-authored-by: Cameron Martin <cameronmartin123@gmail.com>
) Fixes #21621 (comment) PiperOrigin-RevId: 625286656 Change-Id: Ibc5d153af96912de39b146e50f081030d1da7257 Commit 2d54af1 Co-authored-by: Googler <pcloudy@google.com>
Ensures that yanked version information is only downloaded once per module name, which reduces the number of files fetched during module resolution. On my machine, this reduces `bazel mod deps` runtime on Bazel itself by ~15%. This also prepares for storing yanked version information in the lockfile. Work towards #20369 Closes #21909. PiperOrigin-RevId: 625617556 Change-Id: Ie4184def3b868470f93d584bbbbd7f704e1e9a82 Commit 3052eff Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
Whenever the sandbox base changed via --sandbox_base, the trash base used by the AsynchronousTreeDeleter wasn't being updated accordingly. If the new sandbox base was on a different filesystem (like tmpfs), new async deletions would result in the error (Invalid cross-device link) at the time of moving. This would then trigger a precondition crash later when we checked whether everything had been deleted correctly. PiperOrigin-RevId: 625430544 Change-Id: I1547b16bdb803eb5dcee649c9988bad73442a140 Commit dfd1763 Co-authored-by: Pedro Liberal Fernandez <plf@google.com>
Found to fix flaky local behavior of remote tests when invoking `bazel shutdown`: ``` Load240418 11:57:38.114:XT 85 [com.google.devtools.build.remote.worker.ExecutionServer.lambda$waitExecution$1] Work failed: 9bcaa872-d921-4889-a762-66c242993c59 io.grpc.StatusRuntimeException: CANCELLED: call already cancelled. Use ServerCallStreamObserver.setOnCancelHandler() to disable this exception at io.grpc.Status.asRuntimeException(Status.java:526) at io.grpc.stub.ServerCalls$ServerCallStreamObserverImpl.onNext(ServerCalls.java:366) at com.google.devtools.build.remote.worker.ExecutionServer.lambda$waitExecution$1(ExecutionServer.java:171) at com.google.common.util.concurrent.DirectExecutor.execute(DirectExecutor.java:31) at com.google.common.util.concurrent.AbstractFuture.executeListener(AbstractFuture.java:1298) at com.google.common.util.concurrent.AbstractFuture.complete(AbstractFuture.java:1059) at com.google.common.util.concurrent.AbstractFuture.set(AbstractFuture.java:784) at com.google.common.util.concurrent.TrustedListenableFutureTask$TrustedFutureInterruptibleTask.afterRanInterruptiblySuccess(TrustedListenableFutureTask.java:136) at com.google.common.util.concurrent.InterruptibleTask.run(InterruptibleTask.java:89) at com.google.common.util.concurrent.TrustedListenableFutureTask.run(TrustedListenableFutureTask.java:82) at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1144) at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:642) at java.base/java.lang.Thread.run(Thread.java:1583) ``` Closes #22049. PiperOrigin-RevId: 626953580 Change-Id: Id914dfd77496a50427f59638bd81794f021b9425 Commit ba8ccaf Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
When a command line item is subject to special default stringification behavior (currently, `Label` and `File` instances are), it does not suffice to fingerprint the information that computes the custom stringification; it is also necessary to track whether each individual item is subject to this special formatting. Otherwise, as demonstrated by numerous new test cases, command line items have identical fingerprints to their naive stringification, but may result in distinct command lines. This is fixed by tracking element types via additional UUID markers, either individually or for entire `NestedSet`s based on the `Depset` type. Closes #22025 Original commit: d8613c2 PiperOrigin-RevId: 627040284 Change-Id: Id0725e7d7a58b439fde4f6644de09ebedb56a14c
Fixes bazelbuild/buildtools#1263 Closes #22045. PiperOrigin-RevId: 626307672 Change-Id: Ib088e73835ea3d7076976a9de4ab4207ff491b27 Commit 7979dd6 --------- Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
…nce. This didn't work at HEAD because the check there compared the flags in the new configuration with the flags of the old configuration, but overwrote the old configuration with the new one, so running a command that discarded the analysis cache was allowed the second time. The fix is not as trivial as moving checking the value before setting the new configuration because creating the new configuration entails a Skyframe evaluation, which is a waste if we end up erroring out. So the check is now done by checking the build options of the old configuration and the new build options, which doesn't require creating the new configuration first. Fixes #23491. RELNOTES: None. PiperOrigin-RevId: 675555324 Change-Id: Ib402b699ed9ba9f04542896a40ff1180351471db
lberki
requested review from
a team,
brandjon,
tetromino,
oquenchil,
gregestren,
fweikert,
Wyverald,
meteorcloudy,
ahumesky,
ted-xie and
kevin1e100
as code owners
github-actions
bot
added
team-Android
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
No description provided.