Context
- droidstubs module are either generated from the java_sdk_library
module or defined in the bp files.
- Since droidstubs module contains API text file property,
java_api_contribution module can by dynamically created from
droidstubs.
Implementation
- Add `api_surface` property in droidstubs module. This property is
either inherited from the java_sdk_library or written in the module
definition in the bp file.
- Add defaultable hook in droidstubs module to generate the child
java_api_contribution module.
Test: m
Change-Id: Ica43d65614723c623cd0c155266f9844e69e5d5e
Module Lib API surface takes modules that are not part of the API
surface as classpath during compilation of the JAR file. Thus, add libs
property to add deps in bp module definition.
Test: m
Change-Id: I8967d53430fa4cc8e4aa431770ba4e97893f8c61
The -libraryjars argument to r8 was missing transitive dependencies, and
so complained when there were classes used in the program JAR which were
not provided via libraries. This CL propagates transitive dependencies
to the r8 command to reduce the warnings that are generated for missing classes.
Bug: 242088131
Change-Id: Ifad7bc7c5d406e3de8d98ea963c97e88c84b45a1
Now that libs conversion is happening for the needed module_types,
removing if statement with old hacky libs conversion
Test: Presubmit
Bug: 244210934
Change-Id: I92bb2080eeb7dcb85d087a3ff6054bdf8675ec54
RuleBuilder creates a new rule in the ninja file every time
it's used, but in this case the rules were all exactly the same
except that the input/output file paths were different.
This can be converted to a single static rule instead.
Bug: 262629589
Test: m nothing
Change-Id: Iaa887c66a757da13293a3614c000d3be02a2a1b0
Since this class is now part of ART's public API, we do not need to add
it separately to the relevant system modules
Bug: 261244752
Test: TH
Change-Id: Ib318ef006b868d482b2cb8ff4820afe3f352483d
Context
- Some Java API surfaces have subset relationships (i.e. public API
surface is a subset of system API surface). Thus, simply listing
all API files that contributes to hierarchical API surfaces when
defining java_api_library modules can lead to repetition of code in
Android.bp files.
- Utilizing java_defaults can resolve this problem by
grouping java_api_contribution modules by API surfaces, and reduce
code repetition and enhance readability.
Implementation
- Modify java_api_library module to utilize java_defaults modules.
Test: m
Change-Id: I6adead208ca36c21de72ee39c4f532b84af8a153
Enforce target sdk version flag could only be used by `android_app`,
moving the check into a common function `generateAndroidBuildActions`.
This would ensure that the `enforce_target_sdk_version` flag
can also be set by `android_test` and `android_test_helper_app`.
Bug: b/227460469
Test: m nothing
Change-Id: I86e0bf684a5083221dae53907d9f548a0390b673
adds support for *-> android_library and android_library->* edges
Change-Id: I41d4e1d1b8106a17e67951d47e67b59ef3170d17
Test: manually inspected build files for libWallpaperPicker & android-suppor-v4
Lint raises an issue ObsoleteLintCustomCheck if it detects
incompatibilities between a custom lint check jar and the lint api
itself.
This is a warning by default, but in AOSP it should be fatal.
Any custom checks are expected to run. If they don't, there is risk
of introducing bugs.
ObsoleteLintCustomCheck will now cause build breakage so that it will
be investigated immediately.
Bug: N/A
Test: TH
Change-Id: I3a46c93d5b53ee0e1add7c287af791bd1b41f1c1
This reverts commit daa220ae8c.
Reason for revert: Suspected to be a cause for build failure (b/264446640). Will use go/abtd to verify
Change-Id: Ibfdf893f2426b1119923467027e0acedd08808c1
`min_sdk_version` is used to represent the api_level of the device and
its type will eventually become android.ApiLevel. OTOH,
`sdk_version` property represents the API surface a module builds against
_and_ the version of that API surface. For
R8/D8, the additional `--android-platform-build` should be determined
using the sdk_version of the soong module and not min_sdk_version, since
min_sdk_version will not contain information about the api surface used
for compilation.
The unit test for `core_platform_app` in TestR8 were passing since
min_sdk_version was not set, and therefore it implicitly defaulted to
sdk_version.
Also created a custom struct to propagate params to the helper dex
functions
Test: In build/soong, go test ./java
Test: TH
Bug: 208456999
Change-Id: I08ac6f496444d603557e498c8a1794af665abc7a
Some tests need to add custom tradefed options, but still want to
keep most of the soong autogenerated tradefed xml file.
Expose a test_options: { tradefed_options: [...] } property that
will allow tests to add more options to the autogenerated xml file.
Fixes: 184895128
Test: go test, and verified that the ninja files did not change for aosp_arm64
Change-Id: I50d4ad139322e9e207202f1e1a50f5bbb424aa6f
This creates a new API surface for the APIs provided by ART to compile
other API domains (e.g. this contains LambdaMetaFactory). The scope of
this API surface is restricted to compile-time. Unlike the other API
surfaces (e.g. public) whose APIs will exist on device in the form of
implementation libraries, these classes are not guaranteed to exist on
device.
This also updates the naming convention map used in Multi-tree API
export
Bug: 261244752
Test: TH
Change-Id: I164714d68618a3c289bac1a695958e36fdbbda5d
- The converter runs for api_bp2build, not bp2build workspace
- Since droidstubs is an internal module created by java_sdk_library,
the conversion encompasses the latter as well
- Since droidstubs do not have an api_surface attribute, this conversion
uses naming convention to infer the api_surface represented by the api
file e.g. *stubs.source -> publicapi, *stubs.source.system -> systemapi)
- Also adds an SdkIntraCore enum to represent the API surface provided
by one core module to another
There is also ongoing work to check in java_api_contribution modules in
Soong. Once we have that, we can update this converter to operate on
that module type instead
Test: go test ./bp2build
Change-Id: Ia85828e04c738d9ffcc524856d7c3034ee29bbf9
Currently, stubs are compiled with Java version 1.8.
To align with this, set the java version passed to javac invocation as
1.8 for java_api_library module. Helper function is used to sync the
java version of droiddoc module and java_api_library module.
Test: m
Change-Id: I103b94aed9747a2f2dd65d13de0ae9c0f33db5c0
android_library modules are only dexpreopted when built into an app,
and the app will already get the dexpreopt compat dependencies itself.
Adding the compat dependencies to android_library modules causes
circular dependencies when the android_library is used to build
part of the framework, which is then used to build the compat libraries,
which is added as a dependency to the andorid_library.
Test: TestUsesLibraries
Change-Id: I3aca780858c7e5f19d270bcbbe18e6cff8616b43
Check for ALLOW_MISSING_DEPENDENCIES before reporting an error
with a prebuilt JNI library missing the required architecture.
Bug: 263138197
Test: lunch aosp_riscv64-userdebug && m
Change-Id: Ifa02c3c690cca1bf7b2b62cf9888f80d96073d71
This reverts commit 8ec823cba1.
Reason for revert: DroidMonitor: Potential culprit for Bug b/262965953 - verifying through ABTD before revert submission. This is part of the standard investigation process, and does not mean your CL will be reverted.
Change-Id: I236cc36981d8b30527ca286632727f8ca267e969
Ensure proguard flag files specified by java_library targets propagate
to downstream targets. This matches bazel behavior, as well as
behavior for transitive android_library deps.
Bug: 212623640
Test: m + presubmit
Change-Id: I61abaf1fd43d6f06bb610383751bb26e9a1bafd6
Add AndroidMkEmitAssignList to emit a line to assign the items from
the given list of string arrays.
Test: treehugger
Change-Id: Id5acbef38ea4e91349bd2461f226db352d4b8123
Add a test that verifies that that missing certificate path inserted
when AllowMissingDependencies is set uses a well formed extension
that won't trip up sign_target_files_apks.
Fixes: 259861670
Test: TestCertificates
Change-Id: I00656809698aed5f9d3aaadaecda9848d2882223
That method was provided to allow a java_sdk_library_import that was
part of a versioned sdk snapshot to create the name of its component
modules consistent with the way in which modules were named within the
sdk snapshot. As versioned sdk snapshots have all been removed this is
no longer necessary.
Bug: 260237150
Test: m nothing
Change-Id: Iee35abd6d210d9a33dfd8ef2f9c77a53329fda24
Previously, as all versioned sdk snapshots have been removed from
Android.bp files this method would always return false. This change
effectively replaces all calls to it with false, and then optimizes
away any unused code.
Bug: 260237150
Test: m nothing
Change-Id: I1b717ee8345e807bd888451f6e7e3c3a0d391ee2
Previously, various places duplicated the logic to check whether a
module was part of a versioned snapshot. This change replaces that code
with a call to the existing IsModuleInVersionedSdk(Module) func.
This makes the code consistent in preparation for a change that will
remove all calls to IsModuleInVersionedSdk(Module). It is separated out
from that change as it is easier to see that the replaced code is
equivalent to the IsModuleInVersionedSdk(Module) func when it is not
mixed in with other changes. It also makes the following change easier
to review.
Bug: 260237150
Test: m nothing
Change-Id: I44d665bf02d2fc97632a03b5a3d2225ea3c2827d
AndroidGlobalLintChecker.jar provides a set of lint checks
that should be run across the entire tree.
Bug: 236558918
Test: manually tested, treehugger
Change-Id: I2a868f1d78c969eefa2c29477fc8ecab1043df39
Some tests need to add custom tradefed options, but still want to
keep most of the soong autogenerated tradefed xml file.
Expose a test_options: { tradefed_options: [...] } property that
will allow tests to add more options to the autogenerated xml file.
Fixes: 184895128
Test: go test, and verified that the ninja files did not change for aosp_arm64
Change-Id: I75f7eb002c8325ce7cdc76e12e76e16195320620
These will be built using the toolchains of the respective inner tree.
This will eventually allow us to drop them from the module SDK (albeit
we still need to make them work for S/T builds)
Test: TH
Bug: 261476442
Change-Id: I43726a2a171d25ace90dc3ddc73e228ffadfb37c
java_api_contribution is useful to java_api_library when the api files
are not colocated.
If they are colocated (e.g. in Multi-tree assembled api_surfaces
directory), it is useful to refer to them directly without nedding to
create a java_api_contribution module.
Test: In build/soong, go test ./java
Change-Id: I5b4e557068a1e5c71a80c76452030e72ec83a696
Mark Robolectric tests as belonging to the `robolectric-tests` suite
since tools expect test modules to belong to at least one suite.
Note that tests not marked with a suite implicitly get annotated with
the `null-suite`.
This is also required for Atest Bazel mode which only includes files of
test modules with a non-empty suite attribute in the Tradefed test
directory.
See aosp/2181295 for more details.
Test: m out/target/product/vsoc_x86_64/module-info.json
Bug: 222437352
Bug: 228799625
Change-Id: I9dedaf1661a351c32d2affbaf1f9b89d672acc3a
The first line of each section will be
```
include $(CLEAR_VARS) # <module type>
...
```
Bug: 257037252
Test: treehugger
Change-Id: Iba586155b682fe4e1e5817d8397eda8f9d9c8789
Context
- Create phony rule for java_api_library module with its artifact
(*.jar) as dependency
- Fix source directory base mislinkage issue for input text files passed
from java_api_contributions modules
Test: m
Change-Id: I055332fd5bc7ef0d03e7b311b0efa93f4454102f
This leads to an arm64 apex being used on a device that is mixed
x86_64 with nativebridge=arm64. A device like that doesn't appear
to work with arm64 binaries. For example, the boringssl-self-check
binary crashes on boot.
Bug: 260115309
Test: unit test
Test: boot emulator with this combination
Change-Id: Ic4a91974290a05b1799f755fcf52ef226d68f4c2
This change stops the build system from dexpreopting system server jars
for the secondary arch. System server jars are only loaded by the system
server, so the artifacts for the the secondary arch are not used.
Bug: 260577273
Test: Build a system image and no longer see artifacts for the
secondary arch.
Change-Id: I0b131a02e996ed50e7aec3bb0a221f8ceed9ce57
The min_sdk_version was added to the java_import in Tiramisu. This
change will propagate the min_sdk_version set on a java_library to the
java_import snapshot. If possible the min_sdk_version will be resolved
into a numerical version to ensure consistent behavior across build
releases.
Bug: 260560424
Test: m nothing
BUILD_NUMBER=fixed packages/modules/common/build/mainline_modules_sdks.sh
# Ran above before and after this change and made sure that only
# Tiramisu, UpsideDownCake and latest were changed and those changes
# were the addition of a min_sdk_version property.
#
# Ran the following command in tm-mainline-prod:
# lunch cf_x86_64_phone-userdebug
# m ART_MODULE_BUILD_FROM_SOURCE=false nothing`.
#
# It failed with:
# module "prebuilt_okhttp-norepackage" variant "android_common": should support min_sdk_version(33) for "AdServicesApk": min_sdk_version is not specified.
#
# Unpacked the Tiramisu art snapshot generated with this change into
# tm-mainline-prod and reran the previous command and it succeeded.
Change-Id: I9d9d730845554fc175d17f38c038e4e3c7d39e07
JavaDoc's implementation of `aidlFlags` omits `min_sdk_version`, and
therefore aidl compiler would default it to a default version.
To fix this, pass the min_sdk_version explicitly
Bug: 253122520
Test: TH
Test: go test ./java
Change-Id: Ia8f639174f8361136596d0e7b3286606f84705cd
In some cases the java_sdk_library needs to compile against hidden
platform APIs, in order to do so it needs to add impl_only_libs
dependency on the framework library, and set the sdk_version property to
the "core_*" one (e.g. core_platform).
However, if this java_sdk_library depends on the Android SDK in it's
stable API surface, then the setup above will break the stubs srcs
generation, as it uses the sdk_version of the java_sdk_library (e.g.
core_platform), which doesn't have symbols from the Android SDK.
This chage solves the problem by propagating the stub_only_libs to the
invocation that generates the stub sources. It allows the
java_sdk_libraries to add a stub_only_libs dependency on the
android_module_lib_stubs_current stubs library to fix their stubs build.
For more context see another change in this topic:
https://android-review.git.corp.google.com/c/platform/packages/modules/Virtualization/+/2310569
Bug: 243512044
Test: builds
Test: m nothing
Change-Id: Id745cfb2789901672561050c1c3f8bb163922379
sign_target_files_apks and check_target_files_signatures require that
certificates have a .x509.pem extension and reject missing.pem. Using
missing.x509.pem instead.
Bug: 259861670
Change-Id: I0b2367213a8cbe390c231a8609bb04df2b125eda
Context
- Update java_api_library module's prop name from api_providers to
api_contributions to adjust to module rename
- Update variables name correspondingly to enhance readability
Test: m
Change-Id: I93b941a572e04bed6084109d151ba83a82715651
Before this change, ANDROID_LINT_EXTRA_CHECK_MODULES was only read if
ANDROID_LINT_CHECK was also set. This change allows an entire module of
checks to be added to lint without having to specify all of the
individuals with ANDROID_LINT_CHECK.
This is part of a workaround to run some lint checks globally, but is
useful to provide more flexibility generally.
Bug: 236558918
Test: manually tested with AndroidFrameworkLintChecker
Change-Id: I2d3ec8a997d902c9d00a577a56f4b152b10e078a
Context
- fuzz_test.go is not listed in the soong/java test sources and is not
running on CI.
Implementation
- Fix fuzz_test.go and add to test sources
Test: m
Change-Id: If2685c646dec4885fc42d7760992309bfa9da382
Context
- Droidstubs module is currently responsible not only for java api stubs
generation, but also for checking api equality and compatibility.
- Generating stubs with incomplete api text file that does not list
entire api surface is done through metalava implicitly adding unlisted
methods from java source files to the stubs in droidstubs module.
- These factors make java stubs generation harder to debug, thus
introduce `java_api_contribution` and `java_api_library` modules to make java
api stubs and jar generation more explicit in Android.bp level and
eventually easier to debug.
Implementation
- `java_api_contribution` module is included in api domains' directory and
lists api text file directory to be added to the api surface
- `java_api_library` collects all api text file that forms the api surface
and creates stubs invoking metalava. Generated java stub files are
converted into `.srcjar`, and eventually `.jar` file which is the complete
api surface.
Test: m
Change-Id: I86f097cc8592334a5eaa900cec12764c5fcc09e7
This change is a no-op change. It doesn't enable userfaultfd GC by
default. ENABLE_UFFD_GC=true can be passed to the build system to enable
userfaultfd GC for testing purposes.
Bug: 242553398
Test: -
1. lunch aosp_oriole-userdebug
2. ENABLE_UFFD_GC=true m
3. See "--runtime-arg -Xgc:CMC" in .invocation files.
Change-Id: I789b49a71b9604fd41bf1ef77d0ac5bac4cbdf25
java_library modules with .kt srcs or common_srcs will be converted into
module type kt_jvm_library
Test: m bp2build, manually inspected build files for allowlisted modules
Bug: 258688914
Change-Id: I8293a11c8247b4b76358d0991f82c6b61b58adc3
The sdk snapshot must not be including implementation code for boot
libraries, the implementation is provided by dex jars within the
corresponding APEX. However, the snapshot does need a module for each
boot library so that the build can seamlessly access the dex files from
the APEX.
A java_library boot library (like core-oj) is represented in the
snapshot by a java_import module which requires a jar file to be
provided, otherwise it is disabled. However, that is provided purely
to keep Soong happy and should never be used.
Previously, the snapshot would contain an empty file for the jar. As
an empty file is an invalid jar any tool (like compiler) that tried
to consume it would fail which was the correct behavior. Unfortunately,
the error message that was produced was not very helpful, it was just
some variant on `invalid file` which lead to a lot of bugs being
raised.
This change replaces that empty file with a reference to the output
from a genrule which runs a script which produces a more useful error
message, with information on how to fix the issue, and fails the build.
It also adds a Name() method to the SdkMemberProperties type as that is
needed in AddInternalModule() to construct the name of the additional
module.
Tested as follows:
In AOSP/master make the following changes:
1. Temporarily set visibility on core-oj and core-libart to
//visibility:public.
2. Run packages/modules/common/build/mainline_modules_sdks.py to create
the snapshots.
For each of the S, T and latest snapshots I did the following in the
s-aml-prebuilt-test, t-aml-prebuilt-test and aosp/master branches:
1. Created an Android.bp file containing the following:
java_library {
name: "broken",
static_libs: [
"prebuilt_core-libart",
"prebuilt_core-oj",
],
}
2. Fix the visibility issues and run `m broken` where it fails with an
invalid file.
3. Delete the contents of the prebuilts/module_sdk/art/current/sdk
directory.
4. Unpack the relevant version of the art-module-sdk snapshot into the
directory.
5. Run `m broken` where it fails with the helpful message.
6. Test the instructions on how to use the ninja -t path tool to
identify the cause of the problem and fix it.
Bug: 257969510
Test: See above.
Change-Id: I125bde2d7202afff84c97daebcef37e21c548a3a
Remove all modules from legacyCorePlatformApiModules that exist
in internal master and build without the legacy stubs in an
aosp_sunfish-userdebug build.
Bug: 161973015
Test: builds
Change-Id: I2e6248ec0cde0d751f66f32bc7e787b7adbfb75b
Remove all modules from legacyCorePlatformApiModules that exist
in AOSP and build without the legacy stubs. The remaining modules
that exist in AOSP in all products and still need the legacy stubs are:
framework-jobscheduler
framework-minus-apex
framework-minus-apex-intdefs
FrameworksCoreTests
services
services.core.unboosted
Settings-core
SettingsLib
SettingsRoboTests
telephony-common
TeleService
The remainder of the modules still in the list either only exist in
internal master or are product-specific.
Bug: 161973015
Test: builds
Change-Id: I8995670aaaa8dce8e4a870ec1700d1dd7be7c95e
Remove modules that don't exist in an internal master
aosp_sunfish-userdebug build and also have no references in
codesearch.
Bug: 161973015
Test: builds
Change-Id: I4372f140b3fc7d6a9cda01a243f200fe084ef099
Sort the modules in legacyCorePlatformApiModules to simplify removing
unnecessary entries. Also remove a comment that doesn't add any value
and makes it harder to run scripts against the list.
Bug: 161973015
Test: builds
Change-Id: I0886baa42209600cfd1e16e5d4919a784efe29ab
SystemUI seems to build fine against the stable core platform API.
Bug: 161973015
Test: m SystemUI-core SystemUISharedLib SystemUI-tests
Change-Id: I584f98f991e0b9c8b1a5dba1fcc1bef0745dbdc0
There are no riscv64 prebuilts in apex sets, use arm64 ones for
now.
Test: lunch aosp_riscv64-userdebug && m droid dist
Change-Id: Iee2d669e01d93504642223939857634b7fd1c1ba
Allow genrules to access the turbine header jar outputs
from java_library modules using the .hjar output file
tag.
Bug: 251871740
Test: m checkbuild
Change-Id: Ib1ec9734323a51583057fb458f791e1e0bd0d767
Since Bazel's java_import requires a jars attribute to be specified,
the generated neverlink-duplicated module is of type java_library
Change-Id: I14a866dfc583507a9462add50d95060cbfe540c5
Bug: 244210934
Test: m bp2build, go test ./bp2build, manual inspection of generated Build and jar files
Any device java_binary that doesn't have a specific wrapper must
have a main_class property, which is used to generate its default
wrapper. Otherwise its build should fail.
Bug: 250851599
Test: TestDeviceBinaryWrapperGeneration in java_test.go
Change-Id: Ice4c580bcfc1b92f95e217b39e984c55d25a3a02
Rather than individually denylisting filegroups until we prioritize a
solution for mixed builds that will correctly integrate into uses such
as proto, aidl, gensrcs, etc.
Test: mixed_droid
Change-Id: Iddbd391af7dd7cabc892b2b26dbc68e3aa506471
Allow new robolectric to break soong's exepected machine type restrictions
Test: mma in /external/robolectric
Bug: 244627502
Change-Id: If58a36b2d84804d586d9c8a773e2e739867fa987
Soong supports string properties, but they are overloaded, and can mean
one of three things:
* path reference
* module reference
* string literal
Bazel has different types: label and string attributes. Thus there needs
to be a way to categorize them correctly in bp2build.
This CL introduces a new function to be used on properties like
apex_key.private_key / apex_key.public_key, as well as
android_app.certificate / apex.certificate.
It is important to disambiguate the prop betenn a string literal
attribute or file/rule target label attribute, so this functions does
just that. The new attributes are then further handled by their
respective macros (apex_key, android_binary, apex).
Bug: 253557437
Fixes: 253557437
Test: presubmits, new tests
Change-Id: Id8111cdd60d3aabcae7d17fe9da84d0ee3966023
android_test_import is to import prebuilt test apks into the build rule.
It doesn't support properties like test_config, test_config_template,
and etc, but those properties were anyhow added to the module type and
setting those properties didn't trigger any error.
Fixing this surprising behavior by adding only the properties that are
actualy being used.
Bug: N/A
Test: N/A
Change-Id: I93e821e108861fa1249bc5e0882d706849bcf43f
ag/19500070 causes errorprone to fail with out of memory errors.
Also in aosp/2215048 the regular javac heap sized was increased
to 4096MB, so the errorprone heap size wasn't any larger anymore.
Bug: 240473481
Test: m RUN_ERROR_PRONE=true SystemUIGoogleScreenshotTestsLib on internal master
Change-Id: Ie6bdce9f2e7620c076213f4c3313960fd537967b
Follow-up to aosp/2065271. "private" is a Soong internal `SdkKind`
definition and is probably more meaningul for devs who work on
the build system. Changing it to "private platform APIs" should
hopefully make it more meaningful to a wider audience.
(Context: https://yaqs.corp.google.com/eng/q/4927173287831666688)
Test: None
Change-Id: Ied7198bb529eeaa85bc741177b414e06a7262365
suggested-fixes.zip is useful even when lint is exiting with an
error. Save the exit code from the lint executable, create the
zip file, then re-exit with the exit code.
Bug: 216456886
Test: Introduce lint error, verify suggested-fixes.zip is created.
Change-Id: I0ba6190e3de0744e53b2a59ba3016861f2f115e2
Currently, Robolectric test written in Kotlin may not run, especially
when shards are in used, for example SettingsRoboTests has 10 shards.
This is because Robolectric test currently only recognize java files,
adding kt files to fix.
Rename current uniqueSrcFiles to uniqueJavaFiles, and compiledJavaSrcs
to uniqueSrcFiles. uniqueSrcFiles will contains both Java and Kotlin
files.
Note: android.FirstUniquePaths cannot be used, seems the behavior is
different and cause build error.
Bug: 252355400
Test: cd build/soong && mm
Test: m RunSettingsRoboTests with Kotlin tests
Change-Id: Id530ae4dcabffe01a06f44fe4234ffc67b73a601
Previously, in a build containing source and prebuilt art
bootclasspath_fragments, the bootImageVariant.licenseMetadataFile was
set twice with the source always being set after the prebuilt and
so winning.
This change only sets bootImageVariant.licenseMetadataFile for the
active module so it will use the prebuilt's license file if that is
preferred.
Bug: 245956352
Test: m nothing
Change-Id: I948c7e5123169452f67c85ad98c4bbdb90a5d2de
Most of the fields in the bootImageConfig/Variant structs are assigned
inside a Once func so are guaranteed to be only set once. However, some
are assigned outside. This change adds comprehensive tests for those
structs and verifies that the constant fields are preserved and the
mutated fields have the correct value.
The check for the constant fields is added in a new TestBootImageConfig
test.
The check for the mutated fields is added into
TestSnapshotWithBootclasspathFragment_ImageName as that test checks an
art bootclasspath_fragment in the following configurations:
* source on its own
* prebuilt on its own
* source and prebuilt with source preferred
* source and prebuilt with prebuilt
It reveals a couple of interesting facts:
* All the *installs fields are set to the same value irrespective of
whether the source or prebuilt is preferred. The information is
constructed solely from information already within the
bootImageConfig/Variant and so can be moved within Once.
* The licenseMetadataFile is incorrect when prebuilt is preferred.
That is due to both the source and prebuilt modules setting it and
the source module always wins as the source module depends on the
prebuilt so always runs its GenerateAndroidBuildActions after it.
Those issues will be cleaned up in following changes.
Bug: 245956352
Test: m nothing
Change-Id: If917cfbcb3b1c842a8682d51cc1ee1fed1c51add
Sometimes it is necessary for some functionality to be disabled while
running in unit tests, e.g. functionality that requires external
information such as error prone tools and configuration. Sometimes it
is necessary for some functionality to be enabled while running in
unit tests, e.g. functionality that makes state available for
testing but which is not necessary at runtime.
Previously, that was done by checking to see if TestProductVariables
was nil. This change adds a method to abstract that.
Bug: 245956352
Test: m nothing
Change-Id: I7845b79328e7180623161a9bf897568089da4e4f
* changes:
Remove deviceInstalls from bootImageVariant
Remove profilePathOnHost from bootImageConfig
Document bootImageConfig and bootImageVariant structs
* changes:
Move fuzzer's CollectAllSharedDependencies into GenerateAndroidBuildActions
Support AllowMissingDependencies in prebuilt_apex modules
Support AllowMissingDependencies for apex dependencies
Add AllowMissingDependencies support for prebuilt_etc module with no src property
Make OutputFileForModule work for AllowMissingDependencies
Fix panics when target arch is riscv64
Previously, a java module with sdk_version: "system_server_current",
would only be able to access the system server or public API of a
java_sdk_library. This change allows it to access the system server,
module lib, system and public APIs in that order.
The apiScope structs define the characteristics of each of the
different API scopes used as required by the java_sdk_library. They are
organized into a hierarchy which is used for two different purposes.
The first purpose is to define an extension hierachy. If scope X
extends Y then X provides a superset of all API elements (classes,
fields, methods, etc) provided by Y. That is reflected in the fact that
the .txt file for X would be a delta on the .txt file for Y. So, system
extends public and so system_current.txt only contains additional API
elements to add to current.txt.
The second purpose is when a java_sdk_library/import is asked to
provide a specific API scope. e.g. a library that has:
sdk_version: "module_current"
will ask each of the SDK libraries it depends upon for a module-lib
API. However, not all of them will provide an API for that scope. In
that case it will find the closest suitable API scope.
Previously, it did that by traversing up the API extension until it
found an API scope that it did provide and return that. As
system_server_current extended the public API that meant that a library
which has:
sdk_version: "system_server_current"
would provide a system server API if available, and if not fall
straight back to public. That meant that the library could not access
system or module-lib APIs even though it is running in the system
server which should be able to access all APIs.
One way to fix this would have been to just have system server API
scope extend module-lib but that would have had a number of nasty
side effects:
* It would have created a significant overhead as every module that
provides a system server API would also have to provide a module-lib
and system API, along with their corresponding .txt files.
* Each existing java_sdk_library that provided a system server API
would need those .txt files created.
* Generating sdk snapshots for older releases would have been more
complicated.
* It would have confused developers.
All of that would be unnecessary because the system server API scope is
intended to be provided by libraries that are used solely by the system
server so there is no point in them providing anything other than a
system server API.
So, instead a separate access hierarchy was added which is the same as
the extension hierarchy for all existing scopes except for the
system server scope, which instead of just being able to access the
public API will be able to access the module-lib scope, which can in
turn access system and it can in turn access public.
That achieves what we want which is a library that is loaded into the
system server to be able to access all API scopes.
Bug: 204176972
Test: m nothing
Change-Id: I854df63fcaeba32afbc1eb0d1a501238022673d0
The use of this field to return information from buildBootImageVariant
up the call stack to one of the callers resulted in data races being
detected. This change simply passes the deviceInstalls up the call
stack.
Bug: 245956352
Test: m nothing
go test -race ./sdk/... -run TestSnapshotWithBootclasspathFragment_ImageName -test.count 100
# Run the previous command without this change and sometimes it
# shows the data race around deviceInstalls. When run with this
# change it reports no data races.
Change-Id: I3c73920dcb17a6c89a63c6a9c3a0bb049a98a690
The use of this field to return information from bootImageProfileRule
up the call stack to one of the users resulted in data races being
detected. This change simply returns the profile path back up the call
stack.
Bug: 245956352
Test: m nothing
go test -race ./sdk/... -run TestSnapshotWithBootclasspathFragment_ImageName -test.count 100
# Run the previous command without this change and sometimes it
# shows the data race around profilePathOnHost. With this change
# that data race is not reported. Although there is still another
# data race.
Change-Id: I03b09e514cc94f2a6c9d5117d3b2f130cc2e4f5b
Add warnings to the structs to explain how they are supposed to be used
and deprecate fields that are used incorrectly.
Bug: 245956352
Test: m nothing
Change-Id: I090698287b96fd37102b88beb5d7252977bddc54
Fix panics in api_level.go and apex.go when using riscv64 as the target
arch.
Bug: 250918230
Test: lunch aosp_riscv64-userdebug && m ALLOW_MISSING_DEPENDENCIES=true nothing
Change-Id: I85c7685f3d14fa2dc7ffbcdea7f490feca304ef7
Context
- Android-<target>.mk currently does not contain information about partition for its dependent unembedded jni libraries, but only lists the name of the unembedded jni libraries.
- If an android_app module depends on an unembedded jni library that is located in a different partition, make cannot find the library.
Implementation
- Create a string field partition in jniLib struct.
- Add variable "LOCAL_SOONG_JNI_LIBS_PARTITION_<target>", an array of mappings of the name of the jni library to its partition.
Bug: 154162945
Test: m
Change-Id: I6b8e1272ff59dc70e3dd6ce8c6c8e4686dad76df
The certificate module is handled the same in Soong between android apps
and apexes, so share the bp2build code as well.
There are a few changes in this CL:
- If override_apex.certificate is unset, the generated apex also unsets
it. This prevents the generated apex from using the base apex's
certificate, which is most likely incorrect (e.g. google variant using
the cert for the aosp variant). Instead, rely on the default
certificate handling in the macro.
- If the certificate prop is a string, then it gets generated into
certificate_name in order to disambiguate. This behavior is identical
to android_app.
Test: added various unit tests.
Bug: 249089160
Fixes: 249089160
Change-Id: I99e18964ff546429a985d0f64dc21e2c69d35d9d
Revert submission 2164342-use-jdk17
Reason for revert: DroidMonitor-triggered revert due to breakage b/250426490
Reverted Changes:
Ib1a24e7ea:Update path for libjli.so
I3375bd4e4:Use the toolchain from JDK 17
I9183e5511:Use the toolchain from JDK 17
Change-Id: I28fc32290dd337b7d44bb98bc52956ee84f01aaa
Add a simple test and add necessary mock dependencies to allow future
resoure shrinking use with platform targets.
Test: m
Bug: 202959019
Change-Id: Id2dd44d52ce5ea62c06784caab0af6276248cb3f
Previously, only the prefer property was copied from the
java_sdk_library_import module to its child modules which meant that if
the use_source_config_var property was used to control the prefer
property that the child modules would never be used. That can cause
build breakages when building against prebuilts as some parts of the
build will use prebuilt files from java_sdk_library_import and some
will use source files from the corresponding java_sdk_library.
This change copies the use_source_config_var property too.
It also adds tests to verify that dependencies on child modules of a
java_sdk_library use the prebuilt child modules of the corresponding
java_sdk_library_import. That revealed a bug with the handling of stub
sources where the prefer property was set after creating the module
which has also been fixed.
Bug: 249192297
Test: m nothing
# Cherry pick into branch broken by previous behavior and make
# sure that it fixes it.
Change-Id: I5719c257f8457bcb2238bc7965215512a20f1095
Lint was updated, and adds this new error-level check that we
have preexisting occurances of.
Bug: 247885568
Bug: 249337906
Bug: 249335143
Test: m lint-check on internal master
Change-Id: I88c337dd1e12b2722a97e878b5277201df0c163d
The API finalization process for platform releases copies various files
from the apistubs/ dist directory into prebuilts/sdk/<n>/... Having
them use the same name in both places makes that process simpler. For
most modules the name of the file is derived from the name of the
module by appending a suffix but unfortunately, for some modules that
does not work. e.g. the conscrypt.txt file is produced by the
conscrypt.module.public.api module. The dist_stem property was added to
java_sdk_library to allow the stem name of the file to differ from the
module name.
The API finalization process for extension APIs does something similar
as it extracts various files from the snapshots and copies them into
the appropriate extension API specific directory in
prebuilts/sdk/extensions/<n>. Instead of copying files from the
apistubs/ dist directory (which is not built for trains) it copies them
from sdk snapshots that are built as part of the train.
Previously, the sdk snapshot used to derive the name of the files
within the snapshot from the name of the module, it ignored the
dist_stem property. This change causes it to use the dist_stem property
to make it consistent with the apistubs/ dist directory naming.
The file name is created in
sdkLibrarySdkMemberProperties.AddPropertyToSet() which does not have
access to the dist_stem property. So, it has to be supplied in the
sdkLibrarySdkMemberProperties instance in the new Stem property.
Bug: 248258460
Test: m nothing
BUILD_NUMBER=fixed packages/modules/common/build/mainline_modules_sdks.sh --build-release=latest
# Ran the previous command before and after and make sure that the
# conscrypt, art and icu sdk snapshots use the dist_stem value but
# none of the other snapshots are affected.
Change-Id: Ied52003de63dcdb86a252a39bb8781f85d51a6ff
This adds a new flag to the optimize section to allow invoking the resource shrinker
Bug: 246217952
Test: m
Change-Id: I2e7851af1189db2a6adf6f9f9f444a1d7f3a8d60
Due to GMS target SDK requirements (https://docs.partner.android.com/gms/policies/preview/mba#min-target-sdk)
we need to ensure all the mainline packages to target latest API
level. Currently, the team chases each module to bump their target
API level.
Updating the build code to make sure that mainline modules
(i.e updatable modules) target the latest sdk version by default.
It would be by default set to 10000 before SDK finalization and updated to the new API level after finalization.
Effectively it would mean:
1. '10000' in aosp and internal master
2. Finalized number in development branches like "33" in "tm-dev"
3. As sdk hasn't been finalised in "udc-dev", it would be "10000"
which would be automatically updated to finalized version after sdk finalization.
For local development if the target sdk version is required to be set,
users would need to set Updatable flag to false.
Enforce_default_target_sdk_version flag needs to be used in bp file,
if this feature needs to be tested locally when updatable: false.
Bug: b/227460469
Test: m nothing
Change-Id: I05e0ae74ae44fd73df75e91b926bcc924446253f
Provide the `--android-platform-build` flag to R8/D8 for targets that do
not specify a stable min SDK. This includes all targets that do not
specificy an SDK version (e.g., services.jar, SystemUI.apk) and all
targets that specify core_platform (e.g., framework.jar).
Follow-up work will expand the set of cases that qualify for this flag,
but for now this captures some of the highest priority targets.
Bug: 232073181
Test: m + presubmit + validate verbose.log
Change-Id: Idad32c80490a101c049b7f937807a615564ca8ef
Always use the latest extension as the latest API, regardless of
the current base extension version. This makes it so that we can
bump the base in master without changing which txt is considered
latest.
Also extend the error check to apply more widely (which caught an
error in the extension 3 finalization).
Bug: 228017107
Test: m checkapi (with api break & bumped base)
Test: prebuilt_apis_test.go
Change-Id: Ia10eb41ff500c566736f0d29a29984a99386bb6f
I046d75dbbd4f21f4a2b6851f558e430e9879fcff fixed android_app modules
with missing certificate dependencies when AllowMissingDependencies
was set, but the same problem can occur in android_app_import and
android_rro modules. Move the AllowMissingDependencies handling
into processMainCert so that it applies to all of them.
Bug: 246649647
Test: TestAppImportMissingCertificateAllowMissingDependencies
Change-Id: Ic7dd3e61e0e3af15c53b583cf680b1e52394a018
This reverts commit ff9b6faba2.
Reason for revert: No longer needed
Bug: 194063708
Test: m nothing
Change-Id: Ic2f6761c9a0a00a612724cb429f002a16522e53f
When an sdk snapshot is targeted at release X then it cannot include
bootclasspath fragment libraries which are not present in that build as
otherwise it causes build failures. It should also not include any
unsupported libraries, i.e. libraries that cannot work on that release.
This change causes sdk snapshot to exclude libraries that have a
min_sdk_version > target build release
It also ensures that hidden API flags do not include any information
from excluded libraries.
Bug: 240406019
Test: BUILD_NUMBER=fixed packages/modules/common/build/mainline_modules_sdks.sh
# Ran the previous command with and without this change to make
# sure that this change excludes framework-connectivity-t library from the
# tethering sdk snapshot for S, including from the hidden API flag files.
Change-Id: I57969b85a12e9e5a3fc76c055b260cec5d5f7d7f
Previously, the sdk snapshot and platform_bootclasspath both used the
HiddenAPIInfo/Provider to retrieve the hidden API related information
they needed. This change separates them by adding a new
HiddenAPIInfoForSdk/Provider that is used to provide the information
needed by the sdk snapshot. This allows a follow up change to generate
different information for the sdk and for the platform depending on the
target build release of the sdk snapshot.
Bug: 240406019
Test: packages/modules/common/build/mainline_modules_sdks.sh
# Ran the previous command with and without this change to make
# sure that this change does not change the sdk snapshot
# contents.
Change-Id: I30cea6cf8dae1be467b7063430d6197e5baa684b
Previously, the rules for encoding hidden API flags into dex files were
generated in the hiddenAPIRulesForBootclasspathFragment alongside the
rules for creating the hidden API flag files.
This change separates the encoding from the generation of the flags.
This will allow a follow up change to generate API flags for the sdk
snapshot separately without affecting the encoded dex files.
Bug: 240406019
Test: m nothing
Change-Id: I3ab3ff5780e7b2e322665d3ea15f1438f2e403bd
On coverage builds, R8 will fail to properly optimize and fail the build
if ignore_warnings: false, because jacoco injects dependencies on
jacocoagent classes, but the jacocoagent library is not part of the
classpath libraries passed in to R8 in its arguments.
Instead we can add jacocoagent as a libs dependency for these modules so
that it will get pulled into the r8 flags.
Bug: 243903417
Test: m
Change-Id: Icc24cc260b896fc800125a0318308d823ccf7a83
Teach soong to pass metalava the two new command line arguments
--sdk-extensions-root <dir> and --sdk-extensions-info <file> when
generating the API levels file (--generate-api-levels api-versions.xml).
The directory hierarcy in a droiddoc_exported_dir module has special
meaning, e.g. extensions/1/public/*-stubs.jar are the mainline module
stubs containing the public API for SdkExt version 1. Update the logic
where the directories are scanned for android.jar files to locate the
SDK extension jars (extensions/<int>/public/*-stubs.jar).
Also introduce a new field on droidstubs properties
(Extensions_info_file) to set the value of --sdk-extensions-info.
Note: if Extensions_info_file is not set, neither
--sdk-extensions-root or --sdk-extensions-info will be passed to
metalava.
Bug: 228828986
Test: go test -run TestDroidstubsWithSdkExtensions ./java
Test: m sdk dist && grep -e from= -e module= out/dist/data/api-versions.xml # needs APIs to be listed in the info file
Change-Id: I682e34d328fc93d3eded8565ffee40961307901a
This will enable AIDL tags in perfetto traces for the following SDKs:
1. SdkSystemServer
2. SdkCore
3. SdkCorePlatform
SdkModule requires exposing some Trace methods as
@SystemApi(MODULE_LIBRARIES). This will come as a follow up
Test: Manual
Bug: 161393989
Change-Id: I01d520afa8e26eabb97318192151b707b002c28b
On coverage builds, R8 will fail to properly optimize and fail the build
if ignore_warnings: false, because jacoco injects dependencies on
jacocoagent classes, but the jacocoagent library is not part of the
classpath libraries passed in to R8 in its arguments.
Instead we can add jacocoagent as a libs dependency for these modules so
that it will get pulled into the r8 flags.
Bug: 243903417
Test: add optimize.ignore_warnings: false to NetworkStackAppDefaults &&
make SKIP_ABI_CHECKS=true PRODUCT=cf_x86_phone-userdebug \
EMMA_INSTRUMENT=true EMMA_INSTRUMENT_FRAMEWORK=true \
CLANG_COVERAGE=true COVERAGE_PATHS="" \
TARGET_USERDATAIMAGE_PARTITION_SIZE=17179869184 \
SKIP_BOOT_JARS_CHECK=true \
NetworkStack
Change-Id: I71313b1fd35437ec52890f5c30b3fd1381c7c3d2
* changes:
add bp2build unit tests for aidl_interface
AIDL source generation accounts for Bazel paths
convert .aidl srcs for java_library
add Partition method to LabelListAttribute
The AIDL source generation rule sets include flags based on the relative
path of .aidl sources. For .aidl sources provided by Bazel targets, e.g.
in a filegroup, the same directory could be added to the include path
twice. Instead we need to ensure that if a Bazel source provides the
include path, that we don't add it again from a Soong source.
Bug: 229251008
Test: USE_BAZEL_ANALYSIS=1 m api-stubs-docs-non-updatable
Change-Id: I4997039003242b43e0e52ccf41729acb4ad11324
Test: go test ./bp2build
Test: b build //frameworks/base/services/tests/servicestests/aidl:servicestests-aidl
Test: enable //packages/modules/NetworkStack/common/networkstackclient
&& disable restriction on Android SDK in javaLibraryBp2Build
&& b build //packages/modules/NetworkStack/common/networkstackclient:ipmemorystore-aidl-interfaces-V10-java
Change-Id: Ifb817daf09a3983ea1c84948ed9f02a79f95784b
Some protobuf modules like perfetto_trace-full take a long time
to analyze with errorprone, in that case it took over 10 minutes.
Skip errorprone on protobuf or aidl-only modules because having
warnings on generated source code isn't useful anyways.
Bug: 242630963
Test: m RUN_ERROR_PRONE=true javac-check
Change-Id: I5e73accd0b56a0a71e75927778c5c04a0d358e67
Previously, hidden API properties were only allowed on
bootclasspath_fragment and platform_bootclasspath module types. This
change allows them to be specified on java_sdk_library modules too. It
involves the following changes:
1. Add the properties to the java.Module.
2. Populate and provide a HiddenAPIPropertyInfo struct from
java_sdk_library modules.
3. Modify bootclasspath_fragment to merge information gathered from its
content libraries as if it was specified on the fragment itself.
Bug: 240406019
Test: m nothing
packages/modules/common/build/mainline_modules_sdks.sh
# Ran the previous command with and without this change to make
# sure that this change does not change the sdk snapshot
# contents.
Change-Id: I64eb71c2039ddc14cf380689d0cec7ec221f5b88
All usages of HiddenAPIFlagFileProperties use it as the type of a
Hidden_api field. This change pushes the field inside the struct and
just embeds the struct inside the structs that need it making it
simpler and more consistent with how the HiddenApiPackageProperties
struct is used.
This is extracted as a separate change as while the change is simple
it does affect a lot of lines. Keeping it separate makes it easier to
review the changes.
Bug: 240406019
Test: packages/modules/common/build/mainline_modules_sdks.sh
# Ran the previous command with and without this change to make
# sure that this change does not change the sdk snapshot
# contents.
Change-Id: I664453029574ef68dc5712f7bf137a1a6e78e864
In preparation for adding hidden API properties to individual
java_sdk_library modules this change refactors the handling of that
information.
1. Renames HiddenApiPackageProperties to HiddenAPIPackageProperties
to be consistent with other similar structs.
2. Creates a new HiddenAPIPropertyInfo struct to be used to provide
property derived information from one module to another. It includes
both flag files as well as the information about which packages are
used.
3. The HiddenAPIPropertyInfo is embedded within the existing
HiddenAPIFlagInput in place of the existing flag files field.
4. The fields in the HiddenAPIPropertyInfo struct are populated from
the HiddenAPI*Properties structs.
5. Access to package information is retrieved from the info struct
instead of directly from the properties. That means that no changes
will be needed to this code when info from other modules is merged
into the struct in a follow up change.
Bug: 240406019
Test: packages/modules/common/build/mainline_modules_sdks.sh
# Ran the previous command with and without this change to make
# sure that this change does not change the sdk snapshot
# contents.
Change-Id: I773279a4b621bcc3c40e3bfe193f2c7b0caeccd6
In mixed builds currently, filegroup doesn't use path prop when creating the paths to the srcs. It defaults to ModuleDir.
Hence, when java.genAidlIncludeFlags [1] calls srcFile.Rel() to eventually create the AIDL include dir for AIDL flags, srcFile.Rel() returns the filepath relative to the module directory instead. This CL appends path prop to module dir when creating relativeRoot.
This fixes the bridge between converted filegroup that set path prop (e.g. libbinder_aidl) to unconverted module (for example, droidstubs). The fix is needed for the child CL aosp/2186599 to convert libbinder_aidl to Bazel. Without this fix, module-lib-api-stubs-docs-non-updatable (unconverted module that depends on libbinder_aidl) can't be built in mixed builds.
[1]: https://cs.android.com/android/platform/superproject/+/master:build/soong/java/gen.go;l=123?q=java%2Fgen.go
Test: go test
Bug: 243010121
Change-Id: Ic2dd2ab9199c62010303a5b8c611d722f4a4118d
Previously, there were two separate fields for each of the API and
removed API files because in some cases it needed a WritablePath and in
other cases it could only be a Path.
This change refactors the code so that the WritablePath is no longer
needed, the duplicate fields deleted and the remaining field is changed
to Path. It also removes a few unused fields.
Bug: 234113632
Test: m nothing
Change-Id: If81f5fe2aa05d9d7b2bc23869fdb2397649d67b3
The majority of libcore NewApi warnings come from references to new
methods, constants, fields added in a file and then referenced in the
same source file. Maintaining suppressions is not scalable: they need
to work across multiple API versions in different branches.
Bug: 208656169
Test: TH
Change-Id: If5e746f25f1bd34931ee8c89b363aa92c81e23b3
