| Commit message (Collapse) | Author | Age |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In this change:
(1) add a sanity check to verify that GNU bintools
are on the PATH
(2) correct the PATH on Windows if (1) failed
(3) use "<user>/My Documents/Temp" as the default
temp directory instead of "%windir%/Temp", because
the latter is non-writable
Motivated by:
https://github.com/bazelbuild/bazel/issues/2431
https://github.com/bazelbuild/bazel/issues/2449
--
PiperOrigin-RevId: 146006796
MOS_MIGRATED_REVID=146006796
|
|
|
|
|
|
|
|
|
|
|
|
| |
This is necessary since it's now passed in --javabase= even if no bootstrap build is requested.
Fixes #2384.
--
Change-Id: Icc15d9ec174ce9ed82b703c13449d88ec1d936a1
Reviewed-on: https://cr.bazel.build/8408
PiperOrigin-RevId: 145268264
MOS_MIGRATED_REVID=145268264
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Turns out, we couldn't run jarjar because the Java launcher script looks for the .jars in the runfiles and build-runfiles is stubbed out during bootstrapping.
The only reason why this worked at all is that sandboxing *also* doesn't work during bootstrapping but it causes the creation of symlinks that happened to be just in the right place for the Java launcher to find the .jars .
The fix is:
- Explicitly disable sandboxing during bootstrapping so that coincidences like this don't happen again
- Pass a --javabase and --host_javabase option during the bootstrap build so that we don't need any symlinks to access to JVM
- Invoke jarjar using its deploy jar instead of the launcher script.
That was fun.
--
PiperOrigin-RevId: 145083357
MOS_MIGRATED_REVID=145083357
|
|
|
|
|
|
|
|
| |
Bootstrap is broken at HEAD but that was hidden by another change.
--
PiperOrigin-RevId: 145075041
MOS_MIGRATED_REVID=145075041
|
|
|
|
|
|
|
|
| |
Hard-coding the version is causing difficulties to update error prone, so let's not do it.
--
PiperOrigin-RevId: 145061484
MOS_MIGRATED_REVID=145061484
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 144296949
MOS_MIGRATED_REVID=144296949
|
|
|
|
|
|
|
|
|
| |
Closes #2139.
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/2139
PiperOrigin-RevId: 142570236
MOS_MIGRATED_REVID=142570236
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also build the JNI library while bootstraping.
This was once submitted in commit 4a249b6962d32ed4cfd4165dfdae4a555b00bb69 but got
rolled back due to some test breakage that's long
since fixed. In this change I'm slightly modifying
the original code in compile.sh.
Using JNI methods however is necessary because we
can't implement WindowsFileOperations.GetLongPath
in native Java, and having that code is a
prerequisite for the fix of https://github.com/bazelbuild/bazel/issues/2145
See also https://github.com/bazelbuild/bazel/issues/2238
--
PiperOrigin-RevId: 142535019
MOS_MIGRATED_REVID=142535019
|
|
|
|
|
|
|
|
|
|
| |
Use $PATHSEP to create a Java classpath.
Fixes https://github.com/bazelbuild/bazel/issues/2262
--
PiperOrigin-RevId: 142445784
MOS_MIGRATED_REVID=142445784
|
|
|
|
|
|
| |
--
PiperOrigin-RevId: 141396618
MOS_MIGRATED_REVID=141396618
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If a user happens to have spaces in the path name,
`scripts/bootstrap/buildenv.sh` will fail with the error `usage: dirname
path`. Quoting the argument to `dirname` when setting `WORKSPACE_DIR`
avoids this issue.
Closes #2030.
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/2030
PiperOrigin-RevId: 141067912
MOS_MIGRATED_REVID=141067912
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In the error message about not being able to ./compile.sh from
a plain checkout, add full instructions on how to build bazel.
This will hopefully reduce the need for questions about this
error message.
--
Change-Id: I79c280c4c8522cf6f38ec731947365178aef9d58
Reviewed-on: https://cr.bazel.build/7594
PiperOrigin-RevId: 140846434
MOS_MIGRATED_REVID=140846434
|
|
|
|
|
|
|
|
|
|
|
|
| |
...when neither run on a distribution artefact nor
PROTOC is set. In this case, it might be that the user
was trying to build a development version of bazel,
which should be done with 'bazel build //src:bazel'.
--
Change-Id: Ib76601f625b36515a94af7508dfd73a3a352710f
Reviewed-on: https://cr.bazel.build/7536
MOS_MIGRATED_REVID=140350570
|
|
|
|
|
|
|
|
|
|
| |
the order of the error_prone_core jar and the guava jar. The default
alphabetical-directory ordering is bad for the same reason that care is needed
in the order of those jars in the rules in third_party/BUILD (see the comments
there for more info).
--
MOS_MIGRATED_REVID=140051569
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change ./compile.sh to look for pregenerated protoc output and use it,
if present. If not, insist on PROTOC and GRPC_JAVA_PLUGIN to be specified
via the environment. In this way compile.sh does not depend on committed
binaries any more.
--
Change-Id: I0d19a16535b606d2a6ab26ca349a7f9db8e224b7
Reviewed-on: https://bazel-review.googlesource.com/#/c/6731
MOS_MIGRATED_REVID=140037537
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
//third_party/protobuf/3.0.0/.
The issue is that the proto-compiler in third_party/protobuf/ (not 3.0.0/) is actually 3.0.0-beta4, which lacks support for proto maps.
This CL will be followed by the removal of the binaries in third_party/protobuf/.
This is a temporary workaround to allow me to submit code that uses proto maps.
--
MOS_MIGRATED_REVID=138815162
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In our bazel-srcs tar ball, files are packed with permissions 550.
Copying into a temporary directory may preserve permissions. So,
when overwriting one of those files in the temporary directory,
grant the user write permissions before writing to that file. Otherwise,
bootstrapping from //:bazel-srcs via ./compile.sh as normal user
will fail on some systems.
--
Change-Id: I70b4645c9bc3c25b9fd356bacdb959205c64b931
Reviewed-on: https://bazel-review.googlesource.com/#/c/6695
MOS_MIGRATED_REVID=136360945
|
|
|
|
|
|
|
|
|
|
|
| |
We have added support for Linux s390x platform in Bazel.
Closes #1891.
--
Reviewed-on: https://github.com/bazelbuild/bazel/pull/1891
MOS_MIGRATED_REVID=135777047
|
|
|
|
|
|
|
|
|
|
|
| |
Bazel in the will provide a machine-readable stream of important build
events. These interfaces set up the framework and expectations about
the produced events and the entities distributing those events.
--
Change-Id: If2c3b2e11c31b0136b57eadeef2d2f8f8fe5e2e7
Reviewed-on: https://bazel-review.googlesource.com/#/c/6272
MOS_MIGRATED_REVID=134522369
|
|
|
|
|
|
|
|
|
|
| |
Since commit 94d905836c167a21d2321d26c646fde07f5fc522 we no longer need JNI for WindowsFileSystem.isJunction.
This is a partial rollback of commit 4a249b6962d32ed4cfd4165dfdae4a555b00bb69.
Affects https://github.com/bazelbuild/bazel/issues/1680
--
MOS_MIGRATED_REVID=134506403
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When `compile.sh` builds bazel using bazel, it
copies the resulting binary to `output/bazel`.
However sometimes the convenience symlink
`bazel-bin` is not created, probably because an
old one is still around and cannot be deleted.
That is clearly a bug, but to work around it, the
bootstrap builder shouldn't attempt to rely on the
creation of these symlinks in the first place.
This change updates compile.sh to use `bazel info`
to locate the `bazel-bin` directory's real path,
and attempt to copy the bazel binary from there.
This works around
https://github.com/bazelbuild/bazel/issues/1827
--
MOS_MIGRATED_REVID=134398451
|
|
|
|
|
|
|
|
|
| |
Pass "-c opt" when building bazel.
Fixes https://github.com/bazelbuild/bazel/issues/1830
--
MOS_MIGRATED_REVID=134307054
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The bootstrap build is special in that it does not
use the Bazel client hence we cannot pass a
--bazelrc flag as there'd be nothing to parse it.
While Bazel is being bootstrapped it's nice to
have colors on a smart terminal, but when the
bootstrapping is done by Jenkins as part of a CI
test, we need the output to be readable, i.e. be
free of control characters, so we should pass
--colors=no for example.
By exporting BAZEL_BOOTSTRAP_STARTUP_OPTIONS the
user can control what startup flags will the
bootstrap build use, so we can pass these
dumbing-down flags during CI builds.
--
MOS_MIGRATED_REVID=134273969
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
It now always uses mixed path style (drive letter + forward slashes) and
expands abbreviated paths (when c:\\looongpath\\foo is abbreviated to
c:\\looong~1\foo).
The result will always look like c:/looongpath/foo/.
Fixes https://github.com/bazelbuild/bazel/issues/1714
--
MOS_MIGRATED_REVID=132218638
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Still broken. :(
*** Original change description ***
Windows bootstrapping: Fix compile.sh when $TMP is unset.
Fixes https://github.com/bazelbuild/bazel/issues/1714
--
MOS_MIGRATED_REVID=132217266
|
|
|
|
|
|
|
| |
Fixes https://github.com/bazelbuild/bazel/issues/1714
--
MOS_MIGRATED_REVID=132060993
|
|
|
|
|
|
|
| |
Fixes https://github.com/bazelbuild/bazel/issues/1680
--
MOS_MIGRATED_REVID=132051176
|
|
|
|
|
|
|
|
|
|
|
| |
of xcode to select between, relaxing the requirement that users specify --xcode_version and --ios_sdk_version even if they have only one version installed.
RELNOTES: On mac devices, detect locally installed versions of xcode to:
1. Use a sensible default if xcode is required but --xcode_version is unspecified.
2. Use sensible default iOS SDK version for the targeted version of xcode if ios_sdk_version is unspecified.
--
MOS_MIGRATED_REVID=131841587
|
|
|
|
|
|
|
|
|
| |
text around the version string. Fixes #1678.
--
Change-Id: I765699c43119e28d5f0ef8245b91e263d6af6dc0
Reviewed-on: https://bazel-review.googlesource.com/c/5570
MOS_MIGRATED_REVID=131080854
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=130406085
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
The underlying issue (wrong operator '==' instead of '=' for test (a.k.a '[')) was fixed by commit 31b059fc07831f86be951609128b7cd51613df98 which removed the offending code completely.
*** Original change description ***
Revert build-runfiles to use /bin/bash
If /bin/sh is dash (default on Ubuntu), this output a lot of garbage error, which make thing that the dummy build-runfiles is incorrect.
See http://ci.bazel.io/view/Bazel%20bootstrap%20and%20maintenance/job/Bazel/JAVA_VERSION=1.8,PLATFORM_NAME=ubuntu_15.10-x86_64/731/console
--
MOS_MIGRATED_REVID=129432905
|
|
|
|
|
|
|
|
|
| |
If /bin/sh is dash (default on Ubuntu), this output a lot of garbage error, which make thing that the dummy build-runfiles is incorrect.
See http://ci.bazel.io/view/Bazel%20bootstrap%20and%20maintenance/job/Bazel/JAVA_VERSION=1.8,PLATFORM_NAME=ubuntu_15.10-x86_64/731/console
--
MOS_MIGRATED_REVID=129430198
|
|
|
|
|
|
|
|
|
| |
bootstrap.
Previously, we deleted the temp dir and after that, we tried to print errfile in it, and were surprised that it didn't exist.
--
MOS_MIGRATED_REVID=129320375
|
|
|
|
|
|
|
| |
on Windows by default)
--
MOS_MIGRATED_REVID=129319018
|
|
|
|
|
|
|
|
|
| |
bootstrapping.
It apparently happened not to make a difference, but still.
--
MOS_MIGRATED_REVID=129314926
|
|
|
|
|
|
|
| |
--
Change-Id: I2da9049019b3965975fab9b7f606d099d6eab2ff
Reviewed-on: https://bazel-review.googlesource.com/#/c/4040/
MOS_MIGRATED_REVID=128208129
|
|
|
|
|
|
|
| |
reference by future repository rules. Removes the xcode-locator binary file under tools/objc. Originally, the precompiled binary was going to be referenced, but it's easier to build from source in the repository rule.
--
MOS_MIGRATED_REVID=128063694
|
|
|
|
|
|
|
|
|
| |
-XX:TieredStopAtLevel=1 instead of -client, as the latter is ignored on 64-bit JVMs since at least JDK 6.
The new flags have been shown to cut down CPU time by 50% due to reduced time spent in JIT and improving overall run time for short running actions, some times also cutting them in half.
--
MOS_MIGRATED_REVID=127536578
|
|
|
|
|
|
|
| |
This helps a lot when trying to debug Bazel from an IDE.
--
MOS_MIGRATED_REVID=127529499
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=127198779
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change the bootstrapping process so that setting the BAZEL_WRKDIR
environment variable will tell the bootstrap process to mostly write
to that directory---apart from adding the symlinks next to the WORKSPACE
file. So setting this variable will avoid the usual writes to random
places in the file system (like /tmp and the user's home directory).
--
Change-Id: I9d1af747e75cc2a7bb1af08308acc9dfd927e920
Reviewed-on: https://bazel-review.googlesource.com/#/c/3963
MOS_MIGRATED_REVID=127075535
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Apparently we now try to open output files for the process twice: once when we are constructing the output streams, and the second time when we tell the process to redirect its outputs. This causes the outputs to be empty on Windows
*** Original change description ***
Do redirection of stdout / stderr in Java instead of reimplementing it in every process wrapper again.
--
MOS_MIGRATED_REVID=126801016
|
|
|
|
|
|
| |
This makes the regeneration instruction in much simpler (since now they do publish 64-bit Linux binaries), and more importantly, fixes a race condition between a client request and the gRPC server timeout thread.
Change-Id: I9c8b628530fa0192804f8e67b285963ac26110c4
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
...instead of cleaning up the last one once for each tmpdir.
Commit 63e8d6321 changed the command registered atexit from
an explict "rm -rf '${DIR}'" to a call to a cleanup_tempdir
function, with the function containing the remove command.
As, however, the function name is not unique (in fact, it is
a constant), that function will get overridden at each invocation
of the tempdir function. The fact that the function is registered
several times with atexit doesn't help, as it will always remove
the last created temporary directory. Fix this, by creating function
names that encode the directory to be removed in the name. Fixes #1466.
--
Change-Id: I833aef8ee5423412f058e74c8c9e2f4bb53a0cba
Reviewed-on: https://bazel-review.googlesource.com/#/c/3955
MOS_MIGRATED_REVID=126406196
|
|
|
|
|
|
|
|
|
| |
We have the source so build from sources instead. This was
causing strange behavior were the input file was conflicting
with the output file of the corresponding java_library but
was matched by the glob.
Change-Id: Ie0b70215d1862627747bd7b83f075743c904c967
|
|
|
|
|
|
|
|
|
| |
that's what it does.
In particular, it affects Windows.
--
MOS_MIGRATED_REVID=126293339
|
|
|
|
|
|
|
| |
every process wrapper again.
--
MOS_MIGRATED_REVID=126279021
|
|
|
|
|
|
|
|
|
| |
Fixes 1444.
--
Change-Id: If9afa2bebcd35b0c0882685588d8e1f3b18f5344
Reviewed-on: https://bazel-review.googlesource.com/#/c/3890/
MOS_MIGRATED_REVID=125771832
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When comparing two strings with test(1) we need to
make sure that the strings are passed each as precisely
one argument to test (a.k.a. [). Shell variables, like $1,
however may expand to zero or more arguments; hence quote
appropriately.
--
Change-Id: I7fd1cee970ed24e6c332b0da361bc916aafcb006
Reviewed-on: https://bazel-review.googlesource.com/#/c/3881
MOS_MIGRATED_REVID=125671405
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=125589653
|