| Commit message (Collapse) | Author | Age |
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105876178
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105875176
|
|
|
|
|
|
|
|
| |
In checkAttrValNonEmpty, also check dict values for emptiness
when the attribute was declared non-empty.
--
MOS_MIGRATED_REVID=105871620
|
|
|
|
|
|
|
|
|
| |
Fixes #518.
RELNOTES: Add support for Windows-created zip files with non-posix permissions.
--
MOS_MIGRATED_REVID=105865785
|
|
|
|
|
|
|
|
| |
Use SkylarkList everywhere rather than either List or GlobList.
Keep a GlobList underneath a MutableList, where applicable.
--
MOS_MIGRATED_REVID=105864035
|
|
|
|
|
|
|
| |
--
Change-Id: Ib82d20394b85433d7b0aae4a71f07ec1af068fa0
Reviewed-on: https://bazel-review.googlesource.com/#/c/2140/
MOS_MIGRATED_REVID=105856861
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of two metasyntactic variables and one random bit of magic which
changes relative paths to be children of crosstool, we now have three
consistent pieces of metasyntax; except that since we can't actually
remove the old syntax, we have either way for the time being.
RELNOTES[NEW]: accept %crosstool_top% in cxx_builtin_include_directory
--
MOS_MIGRATED_REVID=105854779
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105854596
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105851371
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105844221
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105841866
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105841398
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105840775
|
|
|
|
|
|
|
| |
Preparation for running multiple commands in parallel.
--
MOS_MIGRATED_REVID=105840167
|
|
|
|
|
|
|
|
|
| |
hte uniqueifer and size them smaller when we're under the memoization threshold.
A heap dump (with retained garbage) showed many of these sets sitting around, most at under 60% fill ratio.
--
MOS_MIGRATED_REVID=105818028
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Uses tests that don't run on Bazel
*** Original change description ***
Implement aspect(...) Skylark function.
--
MOS_MIGRATED_REVID=105808850
|
|
|
|
|
|
|
|
|
| |
and --nodistinct_host_configuration explicit. This particularly
helps triaging the Bazel tests when --experimental_dynamic_configs
is on.
--
MOS_MIGRATED_REVID=105801590
|
|
|
|
|
|
|
|
|
|
| |
cxx_builtin_include directories which start with "%workspace%/" are taken
to be relative to the build workspace, and not relative to crosstool-top.
RELNOTES[NEW]: cpxx_builtin_include_directory specifications allow more flexibility.
--
MOS_MIGRATED_REVID=105789270
|
|
|
|
|
|
|
|
|
|
|
| |
android_library rule when the android_library is being used as a forwarding rule
(i.e., has no sources).
RELNOTES: When used as a forwarding rule (i.e., has no sources), android_library
will also forward any exported_plugins in its dependencies.
--
MOS_MIGRATED_REVID=105787789
|
|
|
|
|
|
|
| |
RELNOTES: None
--
MOS_MIGRATED_REVID=105787681
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
While a normal set is theoretically sufficient, it can cause hard-to-reproduce
problems. In particular, the iteration order of the expansion result depends
on the iteration order of the requested targets. If there are multiple
requests for the same set of targets, but with different orders, the results
would depend on which request was made first. If a downstream function also
inadvertantly depends on the iteration order, it can be hard to debug why it
ended up with a different order than it requested.
Alternatively, we could sort the result before returning it. We generally
expect the key set to be smaller than the result set, so this should be
ever so slightly more efficient.
--
MOS_MIGRATED_REVID=105765514
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Breaks many android targets in the depot.
*** Original change description ***
Trying again with checking for the presence of the "resources" attribute.
Change the resource dependency handling to separate between the transitive and direct resources from libraries.
This slightly increases the complexity of resource propagation. The initial algorithm was to simply merge all transitive ResourceContainers together with any new ResourceContainer and propagate them via the AndroidResourcesProvider. The new algorithm is encapsulated inside a new ResourceDependencies class which...
***
ROLLBACK_OF=105631635
--
MOS_MIGRATED_REVID=105752015
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
release. Note the following peculiarities of the current situation:
- Sandboxed execution still silently falls back to non-sandboxed execution due to LocalGccStrategy explicitly requesting local execution.
- However, builds are still correct due to local execution using hdrs_check.
- hdrs_check, even though it was intended to default to "strict" in Bazel from the start, is still set to "loose" by default, so you might accidentally be using non-declared header files in your cc_* targets. In this case, your builds will break when the default becomes hdrs_check=strict and sandboxing becomes available. It is recommended that you check whether your builds are affected, by specifying --hdrs_check=strict manually and fixing your BUILD files to explicitly list all used includes in the srcs / hdrs attributes of your cc_* targets.
These remaining issues will be fixed in an upcoming change.
This change also adds regression tests for the issue.
--
MOS_MIGRATED_REVID=105747212
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105742091
|
|
|
|
|
|
|
|
| |
This is still not hooked up, but it's now passing all the relevant tests that
apply directly to the LoadingPhaseRunner; there are still a couple of TODOs.
--
MOS_MIGRATED_REVID=105739507
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105734972
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105648425
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Change the resource dependency handling to separate between the transitive and direct resources from libraries.
This slightly increases the complexity of resource propagation. The initial algorithm was to simply merge all transitive ResourceContainers together with any new ResourceContainer and propagate them via the AndroidResourcesProvider. The new algorithm is encapsulated inside a new ResourceDependencies class which works as follows:
1. Collect resources from the deps into transitive and direct NestedSets
2. If a rule provides a ResourceContainer, merge the transitive and direct into a new transitive set
3. Export the provider
This results having a rule without resources "forward" the merged sets of transitive and direct resources to the next rule.
--
MOS_MIGRATED_REVID=105631635
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This fixes android_resources being dropped on the ground because the
source jar isn't included.
Previously, the java compilation args were used. This includes a bunch
of implicit and generated source, like idl sources. The source jars
include output from resource processing the rule itself, so it wasn't
possible to start including that.
What we want are all explicit sources only. This CL does that.
--
MOS_MIGRATED_REVID=105624870
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105624527
|
|
|
|
|
|
|
| |
Minimize indirection wrt singletons, turns out we had one that was completely unused.
--
MOS_MIGRATED_REVID=105621494
|
|
|
|
|
|
|
| |
For debugging. Simply serializes the proto to a text file.
--
MOS_MIGRATED_REVID=105621279
|
|
|
|
|
|
|
|
|
|
|
|
| |
(https://github.com/bazelbuild/bazel/issues/510)
This compiles with the assumption that all we were using SIOCGIFINDEX for was to verify
that the name was valid. I think the code as written before was actually wrong as the
index value, and the flags value for a ifreq structure actually shared memory, so we were
potentially setting flags that we didn't want to set based on the actual index returned.
--
MOS_MIGRATED_REVID=105619856
|
|
|
|
|
|
|
| |
A previous commit caused install_base_key to be a large .zip file instead of 32 alphanumeric characters. Thus, when Bazel tried to extract itself, it tried to create a directory of the name "$INSTALL_BASE/<40M path segment>/<long path>", and it allocated said 40M each time it walked up a path segment trying to create parent directories.
--
MOS_MIGRATED_REVID=105600489
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
*** Reason for rollback ***
Breaks [].
*** Original change description ***
Change the resource dependency handling to separate between the transitive and direct resources from libraries.
This slightly increases the complexity of resource propagation. The initial algorithm was to simply merge all transitive ResourceContainers together with any new ResourceContainer and propagate them via the AndroidResourcesProvider. The new algorithm is encapsulated inside a new ResourceDependencies class which works as follows:
1. Collect resources from the deps into transitive and...
***
--
MOS_MIGRATED_REVID=105598448
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105596479
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Most (all?) messages for this exception are complete description of the failure (e.g. "failed to do X", "could not do X"). With the previous wording the user got something like:
ERROR: [...] Writing file for rule '[...]' failedfailed to create file '[...]'
Now:
ERROR: [...] Writing file for rule '[...]' failed (failed to create file '[...]')
--
MOS_MIGRATED_REVID=105591532
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105591091
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105585492
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105552023
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105539869
|
|
|
|
|
|
|
|
|
| |
since root causes are associated with top-level targets and labels, and the filter is for all top-level targets and labels.
I noticed this when a huge --noanalyze build spent most of its time in filtering here. The passed-in "collection" was a list, which meant that we could have sped it up by using a set, but why not just get rid of it all.
--
MOS_MIGRATED_REVID=105536485
|
|
|
|
|
|
|
|
|
| |
former uses the latter). This way we parse BUILD files exactly once.
This is part of a series of changes with the net result being that we open, read, and parse each BUILD file exactly once.
--
MOS_MIGRATED_REVID=105528075
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
and direct resources from libraries.
This slightly increases the complexity of resource propagation. The initial algorithm was to simply merge all transitive ResourceContainers together with any new ResourceContainer and propagate them via the AndroidResourcesProvider. The new algorithm is encapsulated inside a new ResourceDependencies class which works as follows:
1. Collect resources from the deps into transitive and direct NestedSets
2. If a rule provides a ResourceContainer, merge the transitive and direct into a new transitive set
3. Export the provider
This results having a rule without resources "forward" the merged sets of transitive and direct resources to the next rule.
--
MOS_MIGRATED_REVID=105515074
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105512492
|
|
|
|
|
|
|
| |
because it breaks some internal tests.
--
MOS_MIGRATED_REVID=105511798
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105511114
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105504785
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105499267
|
|
|
|
|
| |
--
MOS_MIGRATED_REVID=105498175
|