Then plumb them to LabelAttribute.
This refactoring is required because the previous implementation did not
handle properties in shards other than the first one (e.g.
version_script) well. In addition, it also makes the code paths between
bp2build and analysis more similar.
Bug: 186650430
Test: Presubmits.
Change-Id: Ic4393e8ae47f4e88816bf45c89399efd61494d22
Use a concurrency-safe writer in runWithTimeout to avoid data races
on the bytes.Buffer passed in during tests.
Bug: 181095653
Fixes: 187149270
Test: Test_runWithTimeout
Test: go test -race ./cmd/run_with_timeout
Change-Id: I57a889765cb9ee7b42983f0906313e0c2d1e414e
Maintain compatibility with previous behavior by ignoring dex_import
during hidden API processing.
Bug: 179354495
Test: m nothing
Change-Id: I976b02129bf981b7b61dce233567d6f89e04f92d
This CL adds the ability to filter cc_library modules to only generate
their static variant of their shared variant isn't ready.
For example, the libstdc++ library is buildable as a static library,
which libc depends on. However, the shared variant of libstdc++ depends
on libc, which has to come later.
This CL introduces that abstraction to break up bp2build conversion into
more atomic steps to help with conversion.
Test: TH (bazel build //bionic/... incl. libstdc++'s static variant)
Bug: 186489250
Bug: 186822597
Change-Id: I3e2fe748e4e3d3b656760da4807f342d67c8f45f
But add it to the mixed builds deny list instead of the bp2build deny list.
Test: Waiting for another postsubmit to run.
Change-Id: Ief51356758304b3f13c95df401e310d4f5e236f7
(Some info has been relocated to blockers in Buganizer)
Also remove libjemalloc5 and fmtlib_ndk from the deny list as they don't seem to need to be there at the moment.
Test: bazel build //bionic/...
Test: build/bazel/scripts/run_presubmits.sh
Change-Id: I87e9dd3d4a83291efbcc15d6b99187b57d45ad18
The latter doesn't exist anymore, so there's no more need for the
bp2build file manifest.
Bug: 185817076
Test: TH
Change-Id: I6638b84eb371ced98474e3ca5026e3b33b824767
The behaviour is semantically identical, however,
the tool additionally sorts the symbols by address,
compresses frame unwind information more efficiently,
and improves random-accessibility for lazy decompression.
Overall, the changes balance and the output size is same,
however, libunwindstack can access the data much faster
while using less memory (due to the lazy decompression).
It will also enable further improvements in the future.
Bug: 110133331
Test: ART unwinding tests, run prefetto on the device.
Change-Id: Id48f9fe67fb67fcf2b90cc3b217b71bb8f5147ca
R8 processes are sometimes hanging on the build servers. Wrap R8 with
run_with_timeout to dump the stacks with jstack and kill the process
after 30 minutes. Switch from running with the r8-compat-proguard
shell script to running the jar directly so that jstack gets the pid
of the java process.
Bug: 181095653
Test: m checkbuild
Test: m NetworkStackNextIntegrationTests
Test: m USE_RBE=true RBE_R8=true RBE_R8_EXEC_STRATEGY=remote NetworkStackNextIntegrationTests
Change-Id: If6996bd8eb39c7a8453d79e825004339c009ade2
Add a command that can be used to wrap actions with a timeout, and
optionally run an extra debugging command on timeout.
Bug: 181095653
Test: run_with_timeout_test.go
Change-Id: I91df5c3fb5277968717815a4ad4612113766dab1
This combines properties among "multilib" and "arch" stanzas in selects
generated by bp2build.
With this fix, libc_gdtoa may be removed from the denylist.
This change also refactors a portion of arch.go, specifically bp2build's
arch mutator, adding a number of comments along the way, to hopefully
make this code clearer for future readers.
Test: mixed_libc.sh
Change-Id: If2beea672957cfb1af6760406ba507181ec38f77
Doesn't work when depends on arch/target/etc., but good enough for
libdl_android.
Bug: 186650430
Test: Presubmits.
Change-Id: Ib0facb41a89454717c74663e5e078aedd33d1b9c
The previous attempt, which simply skipped the hidden API processing
altogether when unbundled builds were enabled failed when attempting to
build module snapshots as while they enabled an unbundled build they
actually need the hidden API processing to be performed.
This change just checks whether missing dependencies are allowed and if
so it fakes up any missing files so that the build will only fail if
they are not present AND they are used.
Bug: 186695448
Bug: 185828824
Test: tapas Calendar
m -j60
Change-Id: Ie13fed05af0aba51f45f6791fce944d0e4285037
The main difference between the dexpreopt_bootjars singleton and the
platform_bootclasspath singleton module is the way they find the
modules to use. The former searches all modules, the latter adds
dependencies on the modules that they need. This change separates the
finding of the modules from the copying of the boot jars for those
modules to make it easier to move the remaining functionality to
platform_bootclasspath.
This temporarily creates a singleton specific copy of the hidden API
function isModuleInConfiguredList() to select the modules in place of
the logic in the getBootJar() method. There is a slight loss of context
information from the error messages but as these methods will be
removed once the boot image creation has moved this is not an issue.
While switching the isModuleInConfiguredListForSingleton() to use the
SingletonContext the error message was fixed to include the name of
the module with the issue.
Bug: 177892522
Test: m nothing
Change-Id: Iaea906da95d9da5301fb964fc593890f2216d336
The getBootImageJar function will be removed once the boot image
creation has been moved to the platform_bootclasspath and
bootclasspath_fragment module types. However, the consistency checks
that it performs are still useful so this change moves them out
first.
The ART boot image related checks are now performed in the
bootclasspath_fragment module type. A previous change accidentally
disabled the checks when the contents property was not empty which has
been fixed. Also, the error messages have been tweaked to make it clear
that the art-bootclasspath-fragment is now the source of truth as to
its contents not the configuration.
The framework boot image related checks are now performed in the
platform_bootclasspath module type.
Initially, this change included an extra check to make sure that
UpdatableBootJars comes from updatable APEXes but that broke because
framework-wifi and framework-tethering are not currently marked as
updatable in AOSP.
Bug: 177892522
Test: m nothing
Change-Id: I80fb600fa2c7cec4566b3461c6a33c4c6f0743f4
This CL updates the error messages and reorganizes the bottom half of
the bp2build denylist.
Test: TH
Change-Id: Ib15d019ba9cd006a78474477af34c66b5e418348
The modules in a flattened apex have suffixes that ensure they don't
conflict with the platform versions, so they can ignore HideFromMake.
This ensures that Make can install symbols for all libraries in
apexes.
Fixes: 180622230
Test: banchan com.android.adbd && m && ls $OUT/symbols/apex/com.android.adbd/lib
Change-Id: I76d1aba2eed6ae27d9a954bf88f40f0d00fc6a98
This annotation is required for linting, but is not useful in stubs.
Bug: 185579441
Test: cherry-pick http://ag/14098354 and regenerate stubs
Change-Id: I7e4da99a8843b9c709f3b59f125f43cbd66bda4e
Bug: 179354495
Test: rm out/soong/make_vars*
m nothing
grep INTERNAL_PLATFORM_HIDDENAPI_FLAGS out/soong/make_vars*
- make sure it is still out/soong/hiddenapi/hiddenapi-flags.csv
Change-Id: I86e56512a9a2091f446bad25294d41ea1f4341ee
A module will be installed to /debug_ramdisk if debug_ramdisk is set to
true.
This is a reland of f84e9c05e2, with a fix
that removes /first_stage_ramdisk.
Bug: 184004542
Test: soong test
Change-Id: I739de63cfec6b0fec5a90f7c4741fc4d884d209c
Brian Egizi