* Remove BzlmodRepoRuleHelper

The helper logic is only used in the BzlmodRepoRuleFunction so no need to have it in a separate place.
- Remove BzlmodRepoRuleHelper Interface & its implementations
- Refactor BzlmodRepoRuleFunction to use the helper logic
- Update tests and Build files accordingly

PiperOrigin-RevId: 486712547
Change-Id: I9a274a6a0afcc77be56bbe20e0c4d17c41c31c58

* Separate selection from bazel dependency graph

* Created new K/F/V for Bazel dependency graph named "BazelDepGraph"
* Refactored "BazelModuleResolution" to only run resolution (discovery, selection and checks) to create pruned and unpruned graph and not create any dependency value
* BazelDepGraphResolution calls BazelModuleResolution and extracts the dependency graph from it
* Updated tests

PiperOrigin-RevId: 499026445
Change-Id: Id1237f9d09015ffe8987d933431fccfcfa0c0963

* Add native.module_{name,version}

Extension authors often want to write some macro and change its behavior depending on which module is using it. For example, for rules_go, they want to look at the `go_deps` tags in a certain module and only allow access to repos generated from those tags.

We do this by introducing two new methods on `native`, callable only during the loading phase. They return the name and version of the module associated with the current repo. If the repo is from WORKSPACE, they return `None`. If the repo is generated by an extension, they return info about the module hosting the extension.

The implementation works by storing the "associated module" information in `RepositoryMappingValue`. I had attempted to store them in `BzlmodRepoRuleValue` or even `RepositoryDirectoryValue`, but those are not the right places since they might happen before we even evaluate the MODULE.bazel file (i.e. for non-registry overrides).

Fixes https://github.com/bazelbuild/bazel/issues/17652

.

RELNOTES: Added `native.module_name()` and `native.module_version()` to allow BUILD macro authors to acquire information about which Bazel module the current repo is associated with.
PiperOrigin-RevId: 518849334
Change-Id: I06b4bc95b5a57de2412ee02544240b054c708165

* fix BUILD

* remove test case that was accidentally cherry-picked

---------

Co-authored-by: salma-samy <salmasamy@google.com>
Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

Closes #17888.

PiperOrigin-RevId: 519729739
Change-Id: I0ae8b18acd1a88de4da40b6883cf24c5ce6586bb

Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>

Pick up the latest JDK 17 bugfixes for remotejdk users.

Closes #16816.

PiperOrigin-RevId: 498117497
Change-Id: Ib4e818b09b8c4306d96b328c8c3829da5ee88fa1

Co-authored-by: Benjamin Peterson <benjamin@engflow.com>

Fixes #17754.

What we have seen prior to this change was that sometimes for quick tests the output was swallowed. After a lot of poking it became clear that the culprit is the use of subshell and `tee`, e.g. if you remove `tee` completely from the picture the behavior never shows up.

The issue is that with a fast test, `tee` seems to be killed (or its parent subshell) before the printing the output to stdout.

With this change, we reduce the number of subshells and processes to set up and reduce the chance of the race condition but not remove it.

However, for practical purposes, the race condition is gone.

With the reproduction steps in #17754, and this command
```
for i in {1..10000}; do /tmp/bazel run :foo &> /tmp/log ; grep -q "useful echo" /tmp/log ; if [ $? -eq 0 ]; then echo -n '+'; else  echo -n '-'; fi; done
```
a bazel from head fails ~3900 out of 10000 times.

After this commit, it never failed.

Closes #17846.

PiperOrigin-RevId: 518794237
Change-Id: I8c1862d3a274799b864f0f5f42b85d6df5af78c7

Co-authored-by: Tobias Werth <twerth@google.com>

Resolves #15605

Closes #17645.

PiperOrigin-RevId: 514491743
Change-Id: I17ea9fb57682b668bff02bc64fefd75edb2cf2ee

Co-authored-by: Adam Lavin <adam@lavoaster.co.uk>

Closes #17847.

PiperOrigin-RevId: 518664847
Change-Id: I99d09be093e785c7d1ad437d40866423638be118

Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>

There are several points where ByteStreamUploader may discover that the server already has the blob fully uploaded. These points tried to effect an early return from the upload code, generally by "lying" to higher layers that the upload fully finished. That could lead to bugs. For example, consider the added test case: A compressed upload completes its writing but receives an error instead of a server ACK. On retry, QueryWriteStatus reveals the blob exists and returns its uncompressed size. This confused the checkCommittedSize logic, which expected the final committed size of a compressed upload to be the total compressed data size or -1. The code added by https://github.com/bazelbuild/bazel/commit/daa3dbe22adb03338c75b53ea97954c9434099b4

 also looks broken in the case of compressed uploads.

Rework the uploader code, so that early returns throw a AlreadyExists exception. The exception control flow naturally reflects the desire to escape quickly to the top level.

Closes #17791.

PiperOrigin-RevId: 517389227
Change-Id: I23a2ae92fd4ad27dad750418c128c0d0b245e573

Co-authored-by: Benjamin Peterson <benjamin@engflow.com>

Empty chunkers are strange in that they emit one empty chunk and then end. This change makes them properly resetable. Previously, reset() on a empty chunker could result in a NullPointerException.

Also, it's important to call close() on the underlying data stream even if it's empty.

Closes #17797.

PiperOrigin-RevId: 517119206
Change-Id: Iff7908d6cd0633aa2a355ea89f8e647a9fefffcd

Co-authored-by: Benjamin Peterson <benjamin@engflow.com>

Fixes https://github.com/bazelbuild/bazel/issues/12924.

RELNOTES[NEW]: The aquery and cquery commands now respect the --query_file flag just like the query command.

PiperOrigin-RevId: 487689456
Change-Id: Ia2c9d85e88fdf769a823eaf7b6585a77d654ae70

* Add support for alias targets to cquery's `providers`

When applied to an `alias` target, the `providers` function of `cquery`'s `--output=starlark` mode now returns the providers of the aliased target rather than `None`.

This is achieved by moving `getProvidersDict` from `AbstractConfiguredTarget` up to `ConfiguredTarget`.

Fixes #17749

Closes #17753.

PiperOrigin-RevId: 516707744
Change-Id: I840588d605e3a64b968a019cf4bf43b56d18f4f5

* Remove extra code

---------

Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>
Co-authored-by: keertk <keerthanakumar@google.com>

Currently, we evaluate WORKSPACE (`WorkspaceFileFunction`) and WORKSPACE-loaded .bzl files (`BzlLoadFunction` with `BzlLoadValue.KeyForWorkspace`) with the repo mapping purely computed from previous WORKSPACE chunks. This is unlike BUILD-loaded .bzl files from WORKSPACE-defined repos (`BzlLoadFunction` with `BzlLoadValue.KeyForBuild`, which is the same as `RepositoryMappingFunction`), which take the mappings of the Bzlmod root module into account. This CL fixes that discrepancy by doing the same "repo mapping composition" everywhere.

Fixes https://github.com/bazelbuild/bazel/issues/17655



RELNOTES: Fixed an issue where WORKSPACE and WORKSPACE-loaded .bzl files couldn't see the Bzlmod root module's mappings when Bzlmod is enabled.
PiperOrigin-RevId: 515318590
Change-Id: I4babc922f6cdb932d17ce18d9a9d9d427dbed2eb

Co-authored-by: Googler <wyv@google.com>

[6.2.0] Include build-tools/X.Y.Z/{lib,lib64}/** in the minimal set of SDK files used by the Android integration tests. (#17826)

This is necessary to reenable remote caching/execution for these tests (and make CI runs faster!). In the absence of this change, the tests will fail because the aapt/appt2/aidl/zipalign tools can't be dynamically linked in the remote environment. See issue #8235.

We need to wait until this change makes it into a Bazel release before reenabling the tests, as they inherit the SDK files from the Bazel that runs them.

Related to #8235.

PiperOrigin-RevId: 517180365
Change-Id: I2baa66af15af85349187e75e3acd4e3cb5d84a49

With TTL based discarding and upcoming lease extension, remote cache eviction error won't happen if remote cache can guarantee the TTL. However, if it happens, it usually means the remote cache is under high load and it could possibly evict more blobs that Bazel wouldn't aware of. Following builds could still fail for the same error (caused by different blobs).

This PR changes to remove all remote metadata when the remove cache eviction error happens (which should be rare with the help from TTL based discarding and lease extension) to make sure next incremental build can success.

Part of #16660.

Closes #17747.

PiperOrigin-RevId: 516519657
Change-Id: Ia99770b9d314ca62801b73dc96d09ed8ac2233f6

Co-authored-by: Chi Wang <chiwang@google.com>

Previously, after two window resize events we would stop waiting for the child process.

`wait` is interrupted by any trapped signal (and we trap all signals), so we need to call it in a loop until the child really exited.

Fixes https://github.com/bazelbuild/bazel/issues/17215



Closes #17301.

PiperOrigin-RevId: 514395755
Change-Id: Ifb763a55d53f230cde8b1ca77761d94ea6e43e1f

Co-authored-by: Tobias Werth <twerth@google.com>

This is an amendment to #17374

We have a C++ toolchain config that's being developed to support standalone wasm. We discovered that #17374 wasn't complete. Our example was poorly written and didn't actually create a wasm dynamic library. These changes allow us to successfully create a standalone wasm dynamic library. Sorry for the botched attempt previously.

I would like to add tests, but I'm unsure how to approach such a tests considering the automatic toolchain doesn't support wasm AFAICT.

Closes #17698.

PiperOrigin-RevId: 516204125
Change-Id: Iced5cc80a3151ffde7116b6264c89eaf40466ff5

Co-authored-by: Ezekiel Warren <zekewarren@gmail.com>

* Propagate runfiles for singlejar.

Since the singlejar toolchain attribute is currently defined as a single file and carried around as an Artifact instead of a FilesToRunProvider, it's not possible to implement it as a wrapper script that dispatches to an actual implementation somewhere in its runfiles. This would be useful to experiment with cross-platform action sharing.

PiperOrigin-RevId: 492487124
Change-Id: Ib0f80314eae09bd865b3f31a4180bf068833cdf4

* Add test coverage support to android_local_test

Adding test coverage support to `android_local_test`.

https://github.com/bazelbuild/bazel/issues/15827



Closes #15840.

RELNOTES: Adds coverage metric support to android_local_test
PiperOrigin-RevId: 508549884
Change-Id: I6977efa51ca1c7a6df1f776fe1a326d07989a185

---------

Co-authored-by: Googler <tjgq@google.com>
Co-authored-by: Benjamin Lee <ben@ben.cm>
Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

PR #15590 accidentally introduced a bug where zip-based binaries weren't
cleaning up the temporary runfiles directory they extracted their zip into when
they were run outside of a test invocation.

To fix, explicitly keep track of when the module space needs to be deleted or
not, and pass that long to the code that decides how to execute the program and
how to clean it up afterwards.

Fixes #17342

PiperOrigin-RevId: 513256904
Change-Id: I8c3d322248f734a6290a8a7ee5c8725fae5203dc

Co-authored-by: Richard Levasseur <rlevasseur@google.com>

* Update java_tools v11.12

Release tracking issue: https://github.com/bazelbuild/java_tools/issues/64



Closes #17642.

PiperOrigin-RevId: 513763909
Change-Id: I197d05c9b72a79bdcb62716027ee5d87335fba88

* Make EmptyTopLevelDeclaration a warning

PiperOrigin-RevId: 501864491
Change-Id: I06db1964964a2a817f6255153d7ee0121397720d

---------

Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>
Co-authored-by: Googler <cushon@google.com>

Individual file prefetches within a single prefetchFiles() can race against each other, so they must synchronize when writing to the DirectoryContext.

Closes #17678.

PiperOrigin-RevId: 515024483
Change-Id: Ic8097979d06ab143b4d63f5e90f871f8cbf83959

Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

If the service returns an `ALREADY_EXISTS` error, then we assume that the proper file is present remotely.

Prior art: https://github.com/bazelbuild/bazel/pull/12112

Fixes https://github.com/bazelbuild/bazel/issues/12111



Closes #17692.

PiperOrigin-RevId: 515339566
Change-Id: Iafdd288148e47197cc047d39c9a5e5b6c95acee1

Co-authored-by: Yannic Bonenberger <yannic@engflow.com>

This fixes a race condition where writing to stdin of the subprocess fails because the subprocess terminated quickly (e.g., because it did not need the request to provide credentials, or some other error occurred in the helper before it reads from stdin).

```
Starting local Bazel server and connecting to it...
INFO: Writing tracer profile to '/private/var/tmp/_bazel_yannic/bb6c3b02ccc98aaca966df445b5024eb/command.profile.gz'
INFO: Invocation ID: 86e227bc-187e-4d28-8f04-40b1ce9a36a4
ERROR: java.io.IOException: Stream closed
	at java.base/java.lang.ProcessBuilder$NullOutputStream.write(Unknown Source)
	at java.base/java.io.OutputStream.write(Unknown Source)
	at java.base/java.io.BufferedOutputStream.flushBuffer(Unknown Source)
	at java.base/java.io.BufferedOutputStream.flush(Unknown Source)
	at java.base/java.io.FilterOutputStream.close(Unknown Source)
	at java.base/sun.nio.cs.StreamEncoder.implClose(Unknown Source)
	at java.base/sun.nio.cs.StreamEncoder.close(Unknown Source)
	at java.base/java.io.OutputStreamWriter.close(Unknown Source)
	at com.google.devtools.build.lib.authandtls.credentialhelper.CredentialHelper.getCredentials(CredentialHelper.java:82)
	at com.google.devtools.build.lib.authandtls.credentialhelper.CredentialHelperCredentials.getCredentialsFromHelper(CredentialHelperCredentials.java:108)
	at com.github.benmanes.caffeine.cache.BoundedLocalCache.lambda$doComputeIfAbsent$14(BoundedLocalCache.java:2413)
	at java.base/java.util.concurrent.ConcurrentHashMap.compute(Unknown Source)
	at com.github.benmanes.caffeine.cache.BoundedLocalCache.doComputeIfAbsent(BoundedLocalCache.java:2411)
	at com.github.benmanes.caffeine.cache.BoundedLocalCache.computeIfAbsent(BoundedLocalCache.java:2394)
	at com.github.benmanes.caffeine.cache.LocalCache.computeIfAbsent(LocalCache.java:108)
	at com.github.benmanes.caffeine.cache.LocalManualCache.get(LocalManualCache.java:62)
	at com.google.devtools.build.lib.authandtls.credentialhelper.CredentialHelperCredentials.getRequestMetadata(CredentialHelperCredentials.java:80)
	at com.google.auth.Credentials.blockingGetToCallback(Credentials.java:112)
	at com.google.auth.Credentials$1.run(Credentials.java:98)
	at io.grpc.stub.ClientCalls$ThreadlessExecutor.runQuietly(ClientCalls.java:771)
	at io.grpc.stub.ClientCalls$ThreadlessExecutor.waitAndDrain(ClientCalls.java:754)
	at io.grpc.stub.ClientCalls.blockingUnaryCall(ClientCalls.java:157)
	at build.bazel.remote.execution.v2.CapabilitiesGrpc$CapabilitiesBlockingStub.getCapabilities(CapabilitiesGrpc.java:218)
	at com.google.devtools.build.lib.remote.RemoteServerCapabilities.lambda$get$0(RemoteServerCapabilities.java:84)
	at com.google.devtools.build.lib.remote.ReferenceCountedChannel.lambda$withChannelBlocking$2(ReferenceCountedChannel.java:85)
	at com.google.devtools.build.lib.remote.ReferenceCountedChannel.lambda$withChannel$4(ReferenceCountedChannel.java:108)
	at io.reactivex.rxjava3.internal.operators.single.SingleUsing.subscribeActual(SingleUsing.java:59)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap$SingleFlatMapCallback.onSuccess(SingleFlatMap.java:85)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap$SingleFlatMapCallback$FlatMapSingleObserver.onSuccess(SingleFlatMap.java:112)
	at io.reactivex.rxjava3.internal.operators.single.SingleMap$MapSingleObserver.onSuccess(SingleMap.java:65)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnDispose$DoOnDisposeObserver.onSuccess(SingleDoOnDispose.java:84)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnError$DoOnError.onSuccess(SingleDoOnError.java:52)
	at io.reactivex.rxjava3.internal.operators.observable.ObservableSingleSingle$SingleElementObserver.onComplete(ObservableSingleSingle.java:110)
	at io.reactivex.rxjava3.internal.observers.DeferredScalarDisposable.complete(DeferredScalarDisposable.java:85)
	at io.reactivex.rxjava3.subjects.AsyncSubject.subscribeActual(AsyncSubject.java:233)
	at io.reactivex.rxjava3.core.Observable.subscribe(Observable.java:13176)
	at io.reactivex.rxjava3.internal.operators.observable.ObservableSingleSingle.subscribeActual(ObservableSingleSingle.java:36)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnError.subscribeActual(SingleDoOnError.java:35)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnDispose.subscribeActual(SingleDoOnDispose.java:38)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleMap.subscribeActual(SingleMap.java:35)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap$SingleFlatMapCallback.onSuccess(SingleFlatMap.java:85)
	at io.reactivex.rxjava3.internal.operators.single.SingleCreate$Emitter.onSuccess(SingleCreate.java:68)
	at com.google.devtools.build.lib.remote.grpc.TokenBucket$1.onNext(TokenBucket.java:79)
	at io.reactivex.rxjava3.internal.util.NotificationLite.accept(NotificationLite.java:247)
	at io.reactivex.rxjava3.subjects.BehaviorSubject$BehaviorDisposable.test(BehaviorSubject.java:507)
	at io.reactivex.rxjava3.subjects.BehaviorSubject$BehaviorDisposable.emitFirst(BehaviorSubject.java:468)
	at io.reactivex.rxjava3.subjects.BehaviorSubject.subscribeActual(BehaviorSubject.java:224)
	at io.reactivex.rxjava3.core.Observable.subscribe(Observable.java:13176)
	at com.google.devtools.build.lib.remote.grpc.TokenBucket.lambda$acquireToken$0(TokenBucket.java:64)
	at io.reactivex.rxjava3.internal.operators.single.SingleCreate.subscribeActual(SingleCreate.java:40)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap.subscribeActual(SingleFlatMap.java:37)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleDefer.subscribeActual(SingleDefer.java:43)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap.subscribeActual(SingleFlatMap.java:37)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.core.Single.blockingGet(Single.java:3644)
	at com.google.devtools.build.lib.remote.ReferenceCountedChannel.withChannelBlocking(ReferenceCountedChannel.java:85)
	at com.google.devtools.build.lib.remote.RemoteServerCapabilities.lambda$get$1(RemoteServerCapabilities.java:83)
	at com.google.devtools.build.lib.remote.Retrier.execute(Retrier.java:244)
	at com.google.devtools.build.lib.remote.RemoteRetrier.execute(RemoteRetrier.java:125)
	at com.google.devtools.build.lib.remote.RemoteRetrier.execute(RemoteRetrier.java:114)
	at com.google.devtools.build.lib.remote.RemoteServerCapabilities.get(RemoteServerCapabilities.java:81)
	at com.google.devtools.build.lib.remote.RemoteModule.getAndVerifyServerCapabilities(RemoteModule.java:209)
	at com.google.devtools.build.lib.remote.RemoteModule.beforeCommand(RemoteModule.java:542)
	at com.google.devtools.build.lib.runtime.BlazeCommandDispatcher.execExclusively(BlazeCommandDispatcher.java:379)
	at com.google.devtools.build.lib.runtime.BlazeCommandDispatcher.exec(BlazeCommandDispatcher.java:233)
	at com.google.devtools.build.lib.server.GrpcServerImpl.executeCommand(GrpcServerImpl.java:550)
	at com.google.devtools.build.lib.server.GrpcServerImpl.lambda$run$1(GrpcServerImpl.java:614)
	at io.grpc.Context$1.run(Context.java:566)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
	at java.base/java.lang.Thread.run(Unknown Source)

ERROR: Failed to query remote execution capabilities: Stream closed
java.io.IOException: Stream closed
	at java.base/java.lang.ProcessBuilder$NullOutputStream.write(Unknown Source)
	at java.base/java.io.OutputStream.write(Unknown Source)
	at java.base/java.io.BufferedOutputStream.flushBuffer(Unknown Source)
	at java.base/java.io.BufferedOutputStream.flush(Unknown Source)
	at java.base/java.io.FilterOutputStream.close(Unknown Source)
	at java.base/sun.nio.cs.StreamEncoder.implClose(Unknown Source)
	at java.base/sun.nio.cs.StreamEncoder.close(Unknown Source)
	at java.base/java.io.OutputStreamWriter.close(Unknown Source)
	at com.google.devtools.build.lib.authandtls.credentialhelper.CredentialHelper.getCredentials(CredentialHelper.java:82)
	at com.google.devtools.build.lib.authandtls.credentialhelper.CredentialHelperCredentials.getCredentialsFromHelper(CredentialHelperCredentials.java:108)
	at com.github.benmanes.caffeine.cache.BoundedLocalCache.lambda$doComputeIfAbsent$14(BoundedLocalCache.java:2413)
	at java.base/java.util.concurrent.ConcurrentHashMap.compute(Unknown Source)
	at com.github.benmanes.caffeine.cache.BoundedLocalCache.doComputeIfAbsent(BoundedLocalCache.java:2411)
	at com.github.benmanes.caffeine.cache.BoundedLocalCache.computeIfAbsent(BoundedLocalCache.java:2394)
	at com.github.benmanes.caffeine.cache.LocalCache.computeIfAbsent(LocalCache.java:108)
	at com.github.benmanes.caffeine.cache.LocalManualCache.get(LocalManualCache.java:62)
	at com.google.devtools.build.lib.authandtls.credentialhelper.CredentialHelperCredentials.getRequestMetadata(CredentialHelperCredentials.java:80)
	at com.google.auth.Credentials.blockingGetToCallback(Credentials.java:112)
	at com.google.auth.Credentials$1.run(Credentials.java:98)
	at io.grpc.stub.ClientCalls$ThreadlessExecutor.runQuietly(ClientCalls.java:771)
	at io.grpc.stub.ClientCalls$ThreadlessExecutor.waitAndDrain(ClientCalls.java:754)
	at io.grpc.stub.ClientCalls.blockingUnaryCall(ClientCalls.java:157)
	at build.bazel.remote.execution.v2.CapabilitiesGrpc$CapabilitiesBlockingStub.getCapabilities(CapabilitiesGrpc.java:218)
	at com.google.devtools.build.lib.remote.RemoteServerCapabilities.lambda$get$0(RemoteServerCapabilities.java:84)
	at com.google.devtools.build.lib.remote.ReferenceCountedChannel.lambda$withChannelBlocking$2(ReferenceCountedChannel.java:85)
	at com.google.devtools.build.lib.remote.ReferenceCountedChannel.lambda$withChannel$4(ReferenceCountedChannel.java:108)
	at io.reactivex.rxjava3.internal.operators.single.SingleUsing.subscribeActual(SingleUsing.java:59)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap$SingleFlatMapCallback.onSuccess(SingleFlatMap.java:85)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap$SingleFlatMapCallback$FlatMapSingleObserver.onSuccess(SingleFlatMap.java:112)
	at io.reactivex.rxjava3.internal.operators.single.SingleMap$MapSingleObserver.onSuccess(SingleMap.java:65)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnDispose$DoOnDisposeObserver.onSuccess(SingleDoOnDispose.java:84)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnError$DoOnError.onSuccess(SingleDoOnError.java:52)
	at io.reactivex.rxjava3.internal.operators.observable.ObservableSingleSingle$SingleElementObserver.onComplete(ObservableSingleSingle.java:110)
	at io.reactivex.rxjava3.internal.observers.DeferredScalarDisposable.complete(DeferredScalarDisposable.java:85)
	at io.reactivex.rxjava3.subjects.AsyncSubject.subscribeActual(AsyncSubject.java:233)
	at io.reactivex.rxjava3.core.Observable.subscribe(Observable.java:13176)
	at io.reactivex.rxjava3.internal.operators.observable.ObservableSingleSingle.subscribeActual(ObservableSingleSingle.java:36)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnError.subscribeActual(SingleDoOnError.java:35)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleDoOnDispose.subscribeActual(SingleDoOnDispose.java:38)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleMap.subscribeActual(SingleMap.java:35)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap$SingleFlatMapCallback.onSuccess(SingleFlatMap.java:85)
	at io.reactivex.rxjava3.internal.operators.single.SingleCreate$Emitter.onSuccess(SingleCreate.java:68)
	at com.google.devtools.build.lib.remote.grpc.TokenBucket$1.onNext(TokenBucket.java:79)
	at io.reactivex.rxjava3.internal.util.NotificationLite.accept(NotificationLite.java:247)
	at io.reactivex.rxjava3.subjects.BehaviorSubject$BehaviorDisposable.test(BehaviorSubject.java:507)
	at io.reactivex.rxjava3.subjects.BehaviorSubject$BehaviorDisposable.emitFirst(BehaviorSubject.java:468)
	at io.reactivex.rxjava3.subjects.BehaviorSubject.subscribeActual(BehaviorSubject.java:224)
	at io.reactivex.rxjava3.core.Observable.subscribe(Observable.java:13176)
	at com.google.devtools.build.lib.remote.grpc.TokenBucket.lambda$acquireToken$0(TokenBucket.java:64)
	at io.reactivex.rxjava3.internal.operators.single.SingleCreate.subscribeActual(SingleCreate.java:40)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap.subscribeActual(SingleFlatMap.java:37)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleDefer.subscribeActual(SingleDefer.java:43)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.internal.operators.single.SingleFlatMap.subscribeActual(SingleFlatMap.java:37)
	at io.reactivex.rxjava3.core.Single.subscribe(Single.java:4855)
	at io.reactivex.rxjava3.core.Single.blockingGet(Single.java:3644)
	at com.google.devtools.build.lib.remote.ReferenceCountedChannel.withChannelBlocking(ReferenceCountedChannel.java:85)
	at com.google.devtools.build.lib.remote.RemoteServerCapabilities.lambda$get$1(RemoteServerCapabilities.java:83)
	at com.google.devtools.build.lib.remote.Retrier.execute(Retrier.java:244)
	at com.google.devtools.build.lib.remote.RemoteRetrier.execute(RemoteRetrier.java:125)
	at com.google.devtools.build.lib.remote.RemoteRetrier.execute(RemoteRetrier.java:114)
	at com.google.devtools.build.lib.remote.RemoteServerCapabilities.get(RemoteServerCapabilities.java:81)
	at com.google.devtools.build.lib.remote.RemoteModule.getAndVerifyServerCapabilities(RemoteModule.java:209)
	at com.google.devtools.build.lib.remote.RemoteModule.beforeCommand(RemoteModule.java:542)
	at com.google.devtools.build.lib.runtime.BlazeCommandDispatcher.execExclusively(BlazeCommandDispatcher.java:379)
	at com.google.devtools.build.lib.runtime.BlazeCommandDispatcher.exec(BlazeCommandDispatcher.java:233)
	at com.google.devtools.build.lib.server.GrpcServerImpl.executeCommand(GrpcServerImpl.java:550)
	at com.google.devtools.build.lib.server.GrpcServerImpl.lambda$run$1(GrpcServerImpl.java:614)
	at io.grpc.Context$1.run(Context.java:566)
	at java.base/java.util.concurrent.ThreadPoolExecutor.runWorker(Unknown Source)
	at java.base/java.util.concurrent.ThreadPoolExecutor$Worker.run(Unknown Source)
	at java.base/java.lang.Thread.run(Unknown Source)

INFO: Streaming build results to: https://opal.cluster.engflow.com/invocation/86e227bc-187e-4d28-8f04-40b1ce9a36a4


```

Closes #17666.

PiperOrigin-RevId: 514377908
Change-Id: I772a5e462146e3e9699b45090120634a057a5532

Co-authored-by: Yannic Bonenberger <yannic@engflow.com>

This ensures consistent behavior of string operations even if the individual operations do not set a locale.

Without this change, Bazel can't operate in e.g. a Turkish locale, where it fails with error messages such as:

In rule 'test', size 'medium' is not a valid size.

This is because Turkish case mapping rules make it so that a capital ASCII 'I' lowercases to a non-ASCII variant of 'i'.

Fixes #17541

Closes #17702.

PiperOrigin-RevId: 515339563
Change-Id: I8417d0befd76ba6d140588be5f7e50529af3f6c7

Co-authored-by: Fabian Meumertzheim <fabian@meumertzhe.im>

Since 17cfa015, the java_stub_template.txt file calls into the locale tool to check for the presence of cetain locales.  These calls correctly send error messages to stderr.

Unfortunately, *bash* itself is doing locale lookups as soon as it sees that the LANG or LC_* variables are touched, and this results in warnings that pollute the output.  In our builds, the output is polluted with hundreds of warnings like these:

```
bazel-out/aarch64-opt-exec-CF41B97/bin/external/bfg/lang/java/src/main/java/com/google/devtools/build/bfg/JavaSourceFileParserCli: line 100: warning: setlocale: LC_CTYPE: cannot change locale (): No such file or directory
bazel-out/aarch64-opt-exec-CF41B97/bin/external/bfg/lang/java/src/main/java/com/google/devtools/build/bfg/JavaSourceFileParserCli: line 102: warning: setlocale: LC_CTYPE: cannot change locale (): No such file or directory
```

There is nothing we can do to avoid these warnings.  Trying to set the LANG or LC_* variables via --action_env and --host_action_env has no effect (other than trigger more/different warnings).

The problem stems from https://bugzilla.redhat.com/show_bug.cgi?id=1361965

, which has been fixed in various distros but hasn't made its way to the one we are currently using.

Given that it's bash that is complaining, we can easily side-step this issue by patching the template and wrapping the problematic invocations behind "env".  In this way, bash never gets to see the locale changes. This change does precisely this.

Closes #17670.

PiperOrigin-RevId: 515350095
Change-Id: I1887003802aa329e7fb96f644a2fba906daf5165

Co-authored-by: Julio Merino <julio.merino+oss@snowflake.com>

Due to the existence of templated tree artifacts (tree artifacts where each file is produced by a separate action) and nested artifacts (artifacts whose output path is a descendant of another artifact's), the prefetcher must not assume every tree artifact is fetched as a whole.

In addition, concurrent actions calling the prefetcher might write to the same directory in the output tree, so they must synchronize when making the directory temporarily writable.

Fixes #16333.

PiperOrigin-RevId: 513205572
Change-Id: I827c4643643f63c9425e63bdf9177805c5f0f409

actionId was added by https://github.com/bazelbuild/bazel/pull/11236

 to track download requests for input prefetches.

However, according to the REAPI spec, action_id is

> An identifier that ties multiple requests to the same action. For example, multiple requests to the CAS, Action Cache, and Execution API are used in order to compile foo.cc.

The input prefetches are requested by local actions that depend on outputs of remote executed action (which is identified by the action_id). These requests are not part of generating action hence shouldn't be bound with the action_id.

We could include ActionExecutionMetadata of the local action for the requests for the tracking purpose, but that desires another CL.

PiperOrigin-RevId: 510430853
Change-Id: I8d64a4f7033320a394e02a2b11498f09a3d15310

Co-authored-by: Googler <chiwang@google.com>

While non-empty tree artifacts in the inputs to an action are expanded into the files they contain and omitted from the input mapping, empty tree artifacts are still present, as they signal the need to create the directory. Thus, the check added in 763f966a

 is incorrect.

I'm explicitly skipping the empty tree artifacts in prefetchFiles() as otherwise they get skipped as a result of FileArtifactValue#isRemote() returning false for the FileArtifactValue associated with an empty tree artifact (even if it was produced remotely!), which is extremely subtle.

Closes #17183.

PiperOrigin-RevId: 501207095
Change-Id: Ib52727d6fdc6b7a291a61fba33914e57531fb1f4

Co-authored-by: Tiago Quelhas <tjgq@google.com>

This reverts commit 0f9912ba to fix the regression reported in #17686

Baseline: d60ce2c7

Cherry picks:

   + 490f8bad:
     kshyanashree is the release manager of 6.0.0
   + 4e66d934:
     Automated rollback of commit
     2a289091.
   + 48a8d01b:
     Allow C/C++ coverage collection for external targets
   + ec7be346:
     Temporarily set parent directory of the input to writable if it
     is not.
   + b0984342:
     Infer CPUs for tvOS and watchOS when on Apple Silicon
   + 5cea7dda:
     Improve the documentation for ctx.actions.symlink.
   + a82d26f1:
     Add quotes to "Tip"
   + 75b0f3aa:
     Write a repo mapping manifest in the runfiles directory (#16555)
   + 07c5c1aa:
     Ensure repository names don't start with `~`
   + 30f6c823:
     Escape tilde more gracefully
   + cf3f48ca:
     Relax `Label` repo visibility validation
   + 4477823e:
     Use "apparent name" instead of "local name" in Bzlmod docs
   + 55f4f483:
     Bazel DevSite: Add required css file.
   + a706994a:
     Remove logic that increases delay between progress updates over
     time
   + 1d514ab0:
     Remove references to io_bazel repository
   + b0357bd3:
     Relnotes python script
   + 8f956511:
     Allow Java coverage collection for external targets
   + bddb191d:
     Fix race condition in package-bazel.sh
   + a7578527:
     Decanonicalize labels emitted by {a,c,}query if possible
   + 9d250edb:
     Add uniquify parameter to TemplateDict.add_joined
   + f02bcf8d:
     Fix identical gcov json file name problem
   + a1593309:
     Add `$(rlocationpath(s) ...)` expansion
   + 56f54daf:
     Rollup of SBOM correctness fixes (#16655)
   + 72e6e948:
     [cc] Add cc toolchain to starlark cc_proto_library (#16661)
   + 8f285138:
     Make C++ runfiles library repo mapping aware
   + 95f9adc5:
     Always collect FileProvider's filesToBuild as data runfiles
   + 4959d34e:
     fix: export constraints.bzl file from @local_config_platform so
     it can be used in downstream bzl_library targets
   + 4919d4a6:
     Add --host_per_file_copt (#16695)
   + 0a23d469:
     Moving some tests for `RemoteActionFileSystem` of BwoB to a base
     class.
   + 9296068b:
     [remote/downloader] Don't include headers in `FetchBlobRequest`
   + 128d833f:
     [remote] Respect whether the server supports action cache updates
   + 38c50191:
     [remote/downloader] Migrate `Downloader` to take `Credentials`
     (#16732)
   + 5929cb72:
     Stage repository mapping manifest as a root symlink (#16733)
   + 455454a5:
     Expose current repository name to Java with @AutoBazelRepository
   + 97dea593:
     Implement getDirectoryEntries and readdir for
     RemoteActionFileSystem.
   + aa45f5f3:
     Move integration tests for BwoB to a base class and add more
     tests there.
   + 1b073ac0:
     Make Java runfiles library repo mapping aware
   + 148bbb1c:
     Use `_repo_mapping` in C++ runfiles library (#16752)
   + 25558ada:
     Add --experimental_action_cache_store_output_metadata to the
     expansio…
   + 6e945e8a:
     Treat `DEBUG` events as progress-like.
   + 1940c5d6:
     redact 'token' strings from logging
   + 0b645254:
     Make Bash runfiles library repo mapping aware (#16753)
   + 4caae75b:
     Automated rollback of commit
     8f956511.
   + 13ff6d9e:
     Fix Bash rlocation failure with stricter Bash options
   + bd88c7e4:
     Make bazel Android remote deps compatible with bzlmod (#16772)
   + 6f6d4cca:
     Flip `--incompatible_always_include_files_in_data`
   + 94355b1b:
     Add required `--add-opens` server JVM args also with
     non-embedded JDK
   + 8349c95f:
     Also collect clang resource directory with
     `-no-canonical-prefixes`
   + dce6ed77:
     Make `bazel run` works with minimal mode
   + ca8674cc:
     Include source files with `cquery --output=files` (#16826)
   + 9cb5e0a3:
     Fsync before rename after copy in DiskCacheClient
   + 23ffce53:
     Update GetActionResult for disk cache to check referenced files
     when …
   + 0bc0ff52:
     Make Java runfiles library compilable with JDK 8
   + fe169654:
     [6.0.0] Default --incompatible_strict_conflict_checks to true.
     (#16864)
   + 99ca86f7:
     Revert "[6.0.0] Default --incompatible_strict_conflict_checks to
     true. (#16864)" (#16872)
   + 312238f2:
     Make C++ runfiles library build with `-Werror=shadow`
   + 2baa5a42:
     Keep credentials cached across build commands.
   + 8818a57f:
     [6.0.0] Only inject metadata for outputs that cannot be
     reconstructed by skyf… (#16879)
   + 31e4bf40:
     Update java_tools v11.9
   + cd406660:
     replace _get_canonical_form(label) with str(label)
   + e2bc2374:
     Avoid exceptions from hermetic sandbox for unsupported artifact
     subclasses
   + b564d145:
     Remove warnings about ignored starlark options
   + 7ccc6610:
     [6.0.0] Emit Tree objects in topological order (#16904)
   + 86dee6d2:
     Correctly match regex with tree artifact
   + 14925b57:
     Always use target's attributes to set Python version (#16959)
   + a247502d:
     Add 'toolchain' parameter to actions.{run,run_shell}
   + 497b7e40:
     Fix Bash `runfiles_current_repository` with RBE
   + 0a2c4edb:
     Fix corner cases in Bash runfiles library
   + 33b514b2:
     Fix runfiles creation with MANIFEST when building without the
     bytes
   + 3a13af41:
     Remove LCOV merger dependency of `cc_test` without coverage
     (#17004)
   + 25ba76c1:
     Include full tree artifact in inputs when prefetcher doesn't
     support partial tree artifacts.
   + 06deebfb:
     Release 6.0.0 (2022-12-19)
   + c2bfb2e9:
     Enable git_submodules tests and merge git_repository and
     starlark_git_repository (#17247)
   + e8182ce3:
     have 'bazel test' non-test targets depend on
     --remote_download_output
   + c3245cd0:
     Add `SpellChecker` suggestions for common Bzlmod errors
   + 8ec87434:
     Use xcrun to invoke install_name_tool
   + 2b2bea4e:
     Extra resources
   + 5b4de12b:
     Do not clear `--platforms` on no-op change to `--cpu`
   + db3c0130:
     Report `AspectCreationException` to the user
   + 53e9fea3:
     Use long executable path instead of `argv[0]` in all launchers
   + abae5ca3:
     Add sanitizer features to unix_cc_toolchain_config
   + db684196:
     Allow `map_each` to return `None` in `TemplateDict#add_joined`
   + 66b58ee9:
     Always emit trailing zero in human-readable download progress
   + 17b8e448:
     Upload all logs in BEP even with minimal upload
   + 28d2daae:
     Set `--experimental_action_listeners` to default in `exec` config
   + 16427c90:
     Do not count tests as failed that have not started
   + 094a0aab:
     Accept tildes in --override_module
   + 5f2866f8:
     Do the AC integrity check for disk part of the combined cache.
   + cd10d508:
     Fix `--nozip_undeclared_test_outputs` on Windows
   + 073f54b9:
     Allow pyd in extensions of dynamic libraries
   + ac504cbb:
     Register JavaCompileActionContext in Bazel.
   + f090433f:
     Rollback #14510 because it causes remote test execution to fail
   + 942dddaa:
     Prepare backwards compatible usage of optional C++ toolchain
   + 88b51f57:
     Emit LLVM coverage for source file paths with a `tmp` segment
   + bb0b8862:
     Enable archive_param_file feature by default for Windows
   + baf97c00:
     Allow `TemplateDict#map_each` callback to return a list of
     strings (#17306)
   + fcb00774:
     Only try to create groups of test actions in the ui.
   + 3c47b471:
     Find `libtool` when using `BAZEL_USE_CPP_ONLY_TOOLCHAIN`.
   + 1da8a825:
     Add -dead_strip in default opt link flags for darwin
   + b0c5eb30:
     Dont query remote cache but always use bytestream protocol
   + 00e9af19:
     Allow Java coverage collection for external targets.
   + dd24a002:
     Test and fix root symlink edge case in runfiles library
   + af97d983:
     [6.1.0] Fix broken CI (#17422)
   + b8595718:
     Add `native.package_relative_label` function
   + 82168d44:
     Make Bazel more responsive and use less memory when --jobs is
     high (#17398)
   + 84c1ed43:
     Multiplex worker fixes and tests for Android busybox tools
     (#17371)
   + 07590818:
     Improve error message for concurrent modifications (#17396)
   + 27bc896f:
     [6.1.0] Make bazel coverage work with minimal mode (#17397)
   + 544b8164:
     [6.1.0] Fix cc_binary bug related to cc_shared_library on
     Windows and prepare for future removal of
     --experimental_cc_shared_library flag (#17445)
   + 9a7aa21c:
     Fix Bash `runfiles_current_repository` for tools
   + 911eedc0:
     Fix label unambiguous canonical form to correctly report
     non-visible repo names
   + e132653e:
     Remove O1 from sanitizer feature flag defaults
   + ba9e2f8f:
     Remove usage of gRPC Context cancellation in the remote
     execution client.
   + de03a236:
     Allow -v to libtool
   + 4e35c02c:
     Prettify labels in action progress messages with Bzlmod (#17278)
   + 1be0ac3e:
     Expand tree outputs before eagerly prefetching them for local
     actions. (#17494)
   + c2666510:
     Fix Java coverage collection with Java 8 runtime
   + 1a6ffe6d:
     Add a flag to disable execution log sorting.
   + fbec8e2a:
     Reduce flakiness on Windows for BwoB tests
   + 420659a9:
     Make AutoBazelRepositoryProcessor compatible with Java 8
   + dde6d20b:
     Do not recommend `shallow_since` for `git_repository`
   + f76fc616:
     Allow the timeout value for execute calls on a mac to be set via
     an environment variable
   + 773d232f:
     Fix RPATHs for cc toolchain solib when sibling layout is used
   + 5932b3b8:
     [6.1.0] Add --host_features (#17528)
   + a0fa77cc:
     Exit with code 39 if remote cache evicted blobs that Bazel need
     during an invocation (#17496)
   + 1a438b41:
     Only fetch @remote_coverage_tools when collecting coverage
     (#17512)
   + ee1daaf9:
     [6.1.0]Only fetch @remote_coverage_tools when collecting
     coverage (#17287)
   + 350e3295:
     [6.1.0]Fix symlink file creation overhead (#17488)
   + 5406c953:
     [6.1.0] Cleanup stale state when remote cache evicted (#17538)
   + 2d1b570e:
     Bump version number in runfiles.bash init code
   + 3ab8a0a5:
     [6.1.0]Let `aquery` print effective environment for all
     `CommandAction`s (#17274)
   + 51e6c38a:
     Properly report repo fetch progress during main repo mapping
     computation
   + 744108e8:
     [docs] Update migrating to Xcode tutorial (#17563)
   + 9dc7a7e6:
     Update //third_party:jsr305 (#17557)
   + 43dadb27:
     Bump minimum supported macOS versions to 10.13
   + 7d9d23c1:
     Correctly set default subprocess factory when loading class
     `SubprocessBuilder`.
   + 77f82f4f:
     Add an --incompatible_strict_conflict_checks alias for
     --experimental_strict_conflict_checks.
   + e05345de:
     Add support for wrapping system streams in WorkRequestHandler
   + 68e1924c:
     Also handle remote cache eviction for tree artifacts. (#17601)
   + 4a6d056a:
     Add host transition to lcov_merger in Java version of py_test
   + 546e9e29:
     Fix Bazel 6.0 crash regression (#17613)
   + 7e328bb7:
     Include Bazel version information in profile metadata. (#17616)
   + aafe1235:
     [6.1.0] Handle remote cache eviction when uploading inputs for
     remote actions. (#17605)
   + 938e3482:
     [6.1.0] Rerun the artifact conflict check when
     --incompatible_strict_conflict_checks changes. (#17592)
   + 034a2815:
     Report background download for BwoB (#17619)
   + 5afbce52:
     [6.1.0] Flag for writable outputs (experimental) (#17617)
   + d4105e65:
     [6.1.0] Allow .wasm cc executable and dynamic library extension
     (#17440)
   + a306f4f7:
     make desugar dependencies deterministic
   + 37953c56:
     Apply exec transition to lcov_merger in sh_test and cc_test
   + 1d73d72a:
     [6.1.0]Fix --host_features with multiple transitions (#17641)

Incompatible changes:

  - Bazel no longer increases the delay between progress updates when
    there is no cursor control.
  - `--incompatible_always_include_files_in_data` is flipped
    to true. See https://github.com/bazelbuild/bazel/issues/16654 for
    details.
  - `cquery --output=files` also outputs source files.

    Closes #16602.
  - `--incompatible_strict_conflict_checks` is flipped to true. See
    https://github.com/bazelbuild/bazel/issues/16729 for details.
  - This changes the behavior of Python version in exec/host
    configuration. Mitigation is to set Python version on the targets.
  - `--features` only applies to targets built in the target
    configuration, and `--host_features` is used for the host / exec
    configuration (gated behind `--incompatible_use_host_features`)

    Fixes https://github.com/bazelbuild/bazel/issues/13839

    Closes #16626.

Important changes:

  - The new path variable `$(rlocationpath ...)` and its plural form
    `$(rlocationpaths ...)` can be used to expand labels to the paths
    accepted by the `Rlocation` function of runfiles libraries. This
    is the preferred way to access data dependencies at runtime and
    works on all platforms, even when runfiles are not enabled (e.g.,
    on Windows by default).

    Work towards #16124
    Fixes #10923

    Closes #16428.
  - Starlark `print()` statements are now emitted iff the line of
    code is executed. They are no longer replayed on subsequent
    invocations unless the Starlark code is re-executed.
    Additionally, multiple identical `print()` statements (same
    string from the same line of code, e.g. from a loop) are all
    emitted and no longer deduplicated.
  - Fixes a bug where some compilation flags would not be applied to
    a cc_test
  - Added a `native.package_relative_label()` function, which
    converts a label string to a Label object in the context of the
    calling package, in contrast to `Label()`, which does so in the
    context of the current .bzl file. Both functions now also accept
    relative labels such as `:foo`, and are idempotent.

This release contains contributions from many people at Google, as well as Brentley Jones, Ezekiel Warren, Fabian Meumertzheim, keertk, Keith Smiley, Krzysztof Naglik, kshyanashree, lunch-glide-pepper, oquenchil, Ted Kaplan, Ted Kaplan, UebelAndre, Xùdōng Yáng, Yannic, yuzhy8701.

* Fix --host_features with multiple transitions

Fixes https://github.com/bazelbuild/bazel/issues/13839#issuecomment-1448208320



Closes #17625.

PiperOrigin-RevId: 513249316
Change-Id: Ied4ab6958febaf320ae9e0f906c8071db0077e43

* Update exec to host

---------

Co-authored-by: Keith Smiley <keithbsmiley@gmail.com>
Co-authored-by: keertk <110264242+keertk@users.noreply.github.com>

* Apply exec transition to lcov_merge in sh_test

* Apply exec transition to :lcov_merger in cc_test

Desugar dependencies are added to the result jar file as metadata, this desugar dependency object contains a few lists that we found out the order of this list can be different in different builds with the same inputs, therefore the final result will have a different hash and it cause cache miss in the builds solution is to make the lists in this object sorted. So for the same input, we always get the same output

Closes #16859.

PiperOrigin-RevId: 500842557
Change-Id: I051e74e7147d590fcfe5cda9731b8e012396bb65

Co-authored-by: mohamadk <mohamad.khaleghy@gmail.com>
Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

* Add .wasm cc artifact executable extension

This allows creating a custom cc_toolchain that compiles to WebAssembly that outputs a wasm file with the standard extension, without needing to have a genrule that renames the file

Closes #16091.

PiperOrigin-RevId: 503921110
Change-Id: I3ed9e50a945d2e415e2ab34bd99ceea481c987f0

* Add .wasm cc dynamic library extension

Similar to #16091 except also allow the `.wasm` extension for dynamic libraries.

Closes #17374.

PiperOrigin-RevId: 507668339
Change-Id: Id85f98bb804e573e5f93a922b67baf75a22baf16

---------

Co-authored-by: Scott Bennett <scottbennett@google.com>
Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

This feature is tied to an experimental flag `--experimental_writable_outputs`. When enabled, Bazel will set the permissions of all output files to 0755 instead of 0555.

RELNOTES: None.
PiperOrigin-RevId: 500786227
Change-Id: I59e15f3fec09c40a052a60b00da209547f10d7fc

Co-authored-by: Googler <cparsons@google.com>
Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>
Co-authored-by: keertk <110264242+keertk@users.noreply.github.com>

When building without the bytes, downloads can happen when Bazel need to fetch inputs for local actions, download outputs of toplevel targets, and etc. The downloads happen silently at the background.

This PR makes output downloads of toplevel targets visible to the UI.

For downloads during the build, the UI looks like:

```
[2 / 5] 2 actions, 1 running
    [Prepa] Executing genrule //:foo
    Executing genrule //:bar; 2s
    Fetching bazel-out/darwin-fastbuild/bin/baz; 50.8 MiB / 1.0 GiB; 4s
```

For downloads after build is completed, the UI looks like:

```
INFO: 2 processes: 1 remote cache hit, 1 internal.
INFO: Build completed successfully, 2 total actions
    Fetching bazel-out/darwin-fastbuild/bin/baz; 48.8 MiB / 1.0 GiB
```

Fixes https://github.com/bazelbuild/bazel/issues/17417.

Closes #17604.

PiperOrigin-RevId: 512934756
Change-Id: I502489420ab9b84efb74ceabbe3dd4cb4a7a62e2

[6.1.0] Rerun the artifact conflict check when --incompatible_strict_conflict_checks changes. (#17592)

Related to #16729.

PiperOrigin-RevId: 511432374
Change-Id: I00b0bff5731a3468bf0a56c4a44e95590da7b463

Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

* Extract code for cleaning stale state when remote CAS evicted blobs into another class.

PiperOrigin-RevId: 512049304
Change-Id: I87e9e6bcd4d4c4d8be27be13cfb8ba70b199b6e6

* Handle remote cache eviction when uploading inputs for remote actions.

Previously, we handle remote cache eviction when downloading inputs for local actions. However, it's possible that when Bazel need to re-execute an remote action, it detected that some inputs are missing from remote CAS. In this case, Bazel will try to upload inputs by reading from local filesystem. Since the inputs were generated remotely, not downloaded and evicted remotely, the upload will fail with FileNotFoundException.

This CL changes the code to correctly handles above case by reading through ActionFS when uploading inputs and propagate CacheNotFoundException.

Related to #16660.

PiperOrigin-RevId: 512568547
Change-Id: I3a28cadbb6285fa3727e1603f37abf8843c093c9

* Fix tests

---------

Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

This is basically the same as
https://github.com/bazelbuild/bazel/pull/17562/files before we did the JSON
trace profile refactoring.

---

This is helpful when building tools on top of the JSON profile which might to need to distinguish with what Bazel version the profile was written to be able to parse or interpret it correctly.

Closes bazelbuild#17562.

PiperOrigin-RevId: 511761823
Change-Id: I9c31494bfc07be898250d75821dd080e5b68b6db

Stop crashes when incompatible target skipping mixes with ---auto_cpu_environment_group.

Fixes https://github.com/bazelbuild/bazel/issues/17561

.

PiperOrigin-RevId: 512125121
Change-Id: If5960a6abb08f8fe4f2643af6249c8528b7a2c51

Closes #17590.

Change-Id: If5960a6abb08f8fe4f2643af6249c8528b7a2c51
PiperOrigin-RevId: 512820070

Co-authored-by: Googler <gregce@google.com>

Co-authored-by: kshyanashree <109167932+kshyanashree@users.noreply.github.com>

Related to #17366.

PiperOrigin-RevId: 511849585
Change-Id: I1c4acd469280628b15441a8e7a97ff572c42c1ff

Co-authored-by: Googler <tjgq@google.com>