Revert submission 3060229-ron-tmc-symlinks
Reason for revert: Droidmonitor created revert due to b/344045516
Reverted changes: /q/submissionid:3060229-ron-tmc-symlinks
Change-Id: If5045366677163560cdae95c0ab74256b4b49b9a
Now the derived modules will have symlink's to base's testcase dir
rather than copyfile files from base's intermediates dir.
I also removed storing the "base" module as dependency and only use
data from the provider in GenerateAndroidBuildActions and
AndroidMkEntries.
I did have to add two fields to the provider for this.
To ensure the symlinks always resolve to a physical file, I also added
goals such that building `derived-target` will also build `build-target`
Create new Make variable: LOCAL_SOONG_INSTALLED_COMPATIBILITY_SUPPORT_FILES
for tasks/general-tests.mk,device-tests.mk to read.
This allows "support" files that are installed with soong rules rather
than make rules to still end up in the zips.
Sample dir structure:
% tree -ls testcases aosp_cf_x86_64_phone[4:31:54]/0
[ 4096] testcases
├── [ 4096] FrameworksServicesTests
│ ├── [ 4096] data
│ │ └── [ 851] broken_shortcut.xml
│ ├── [ 2800] FrameworksServicesTests.config
│ ├── [ 12695] JobTestApp.apk
│ ├── [ 8599] MediaButtonReceiverHolderTestHelperApp.apk
│ ├── [ 16791] SimpleServiceTestApp1.apk
│ ├── [ 16791] SimpleServiceTestApp2.apk
│ ├── [ 16791] SimpleServiceTestApp3.apk
│ ├── [ 1017763] SuspendTestApp.apk
│ └── [ 4096] x86_64
│ └── [ 79827767] FrameworksServicesTests.apk
└── [ 4096] FrameworksServicesTests_contentprotection
├── [ 4096] data
│ └── [ 54] broken_shortcut.xml -> ../../FrameworksServicesTests/data/broken_shortcut.xml
├── [ 3005] FrameworksServicesTests_contentprotection.config
├── [ 41] JobTestApp.apk -> ../FrameworksServicesTests/JobTestApp.apk
├── [ 69] MediaButtonReceiverHolderTestHelperApp.apk -> ../FrameworksServicesTests/MediaButtonReceiverHolderTestHelperApp.apk
├── [ 52] SimpleServiceTestApp1.apk -> ../FrameworksServicesTests/SimpleServiceTestApp1.apk
├── [ 52] SimpleServiceTestApp2.apk -> ../FrameworksServicesTests/SimpleServiceTestApp2.apk
├── [ 52] SimpleServiceTestApp3.apk -> ../FrameworksServicesTests/SimpleServiceTestApp3.apk
├── [ 45] SuspendTestApp.apk -> ../FrameworksServicesTests/SuspendTestApp.apk
├── [ 36] test_module_config.manifest
└── [ 4096] x86_64
├── [ 64] FrameworksServicesTests.apk -> ../../FrameworksServicesTests/x86_64/FrameworksServicesTests.apk
└── [ 36] UNUSED-FrameworksServicesTests.apk
Test: m clean && atest FrameworksServicesTests_contentprotection
Test: m clean && atest CtsDevicePolicyManagerTestCases_ParentProfileApiDisabled
Bug: b/332320956
Change-Id: I8466f253fa559bc74cef4533edf263650e96bbfb
* changes:
Revert^2 "Refactor how jni_libs dependencies are added"
Revert^2 "Add make java.dependencyTag installable or not"
Revert "Revert "APK-in-APEX should set use_embedded_native_libs:..."
Revert^2 "Install jni symlinks in Soong"
Revert submission 3094658-revert-3088867-QEHDREAABY
Reason for revert: not a regression
Reverted changes: /q/submissionid:3094658-revert-3088867-QEHDREAABY
Change-Id: Ie1e4a196f181643a3e64e573b132d52d7455461b
b7646e4d4f
This is a relanding of I0930cb1ebb8ca8a6efd64b1ce2cdfd1c47fe19ef plus
some forward fix described below:
Export non-embedded JNI lib names via LOCAL_REQUIRED_MODULES
The non-embedded JNI libs are installed as the dependencies of the APK.
However, that dependency is not revealed to the Make world and as a
result, the JNI libs are dropped from the file_list.txt file which Make
uses to filter files to include in the image file.
Adding the lib names to LOCAL_REQUIRED_MODULES fixes it.
Bug: 341335305
Bug: 330276359
Test: m out/target/product/vsoc_x86_64_only/obj/PACKAGING/system_intermediates/file_list.txt
and check if libcarservicejni.so is there
Test: go test ./... under soong/java
Change-Id: If915a05909129c92fab7a6cbbd0c4c55f5ced598
An android app is automatically creating a corresponding "product
characteristics" RRO. The app's resources are filtered by the aapt
product config, but the rro resources are not. This cl makes the RRO
resources also filtered.
When building the CN version, it was expected that only the resources
corresponding to the CN version would be retained, but the actual result
was that other unexpected areas appeared on the region setting interface
in boot interface. After investigation, it was found that the
framework-res__nosdcard__auto_generated_characteristics_rro.apk apk
integrated resources other than the CN version that caused this problem,
and the integration of this apk was not filtered by aaptconfig.
The purpose of this modification is to add aaptconfig filtering, and it
has been verified that this problem can be fixed.
Change-Id: I85d5d07397bd2b0759d35b9f3a07f015346af4e9
Signed-off-by: wybxyx <wybxyx@gmail.com>
This replaces the global override of min_sdk_version (via
`APEX_GLOBAL_MIN_SDK_VERSION_OVERRIDE`) with an min_sdk_version override
that can be set by each individual soong override_app.
The use case for this are go apps which are only installed in T and
above, even though the base AOSP apexes might be installable on < T
devices.
Test: added a unit test
Bug: 295311875
Change-Id: Ie2e738a6786bb24417c675617f7c78358017c96c
Revert submission 3088867
Reason for revert: On cf_auto build, we are seeing a symlink of jni lib pointing to lib64/libcarservicejni.so, which does not exist on the device.
Reverted changes: /q/submissionid:3088867
Bug: b/341335305
Change-Id: Ie4ff26fe47e1a0ab5e97b185c61cb94473cc5fb8
Revert submission 3088867
Reason for revert: On cf_auto build, we are seeing a symlink of jni lib pointing to lib64/libcarservicejni.so, which does not exist on the device.
Reverted changes: /q/submissionid:3088867
Bug: b/341335305
Change-Id: I3064caef0cb55d1594951835a1173c667f58e607
Revert submission 3088867
Reason for revert: On cf_auto build, we are seeing a symlink of jni lib pointing to lib64/libcarservicejni.so, which does not exist on the device.
Reverted changes: /q/submissionid:3088867
Bug: b/341335305
Change-Id: Ibda68b268e74ffca1840faae01a2f3fe2e00be30
This CL brings three changes in how jni_libs are depended on.
1. SDK variants of the jni_libs are depended on only when they can be
embedded. This is because SDK variants are not installable.
Previously, app could depend on SDK variants without embedding them, but
this didn't cause a problem because the installation of the jni libs was
done in Make. However, as it's done in Soong, we need to depend on a
correct variant.
2. Non-SDK variants of the jni_libs are now tagged with jniInstallTag.
This automatically installs the libraries along with the app. The
installation of the jni libs is no longer done in app.
3. checking of the sdk version of the jni libs is done only when they
are embedded. Doing the check even when the libs are not embedded
triggers a lot of false alarms. Note that with #1, many platform apps
have started depending on non-SDK variants.
Bug: 330276359
Test: m
Change-Id: I1bc9ceb8f79b102caeb23476c3fb03989e184a91
So far, we automatically have embedded JNI libs for APKs in APEX, even
if use_embedded_native_libs was not set or set to false.
With this change, such APKs should explicitly set the property.
This is in preparation for an upcoming change for using different
dependency tags to JNI deps depending on whether they are embedded (thus
no install is needed), or not (thus installation to the partition is
needed).
shouldEmbedJni function now doesn't make use of the ApexInfoProvider
which is not available during deps mutator.
Bug: N/A
Test: m
Change-Id: I836171aacba19dbfa0e09f16a773ae498c56d60b
The installation of the symlink
(<partition>/app/MyApp/lib/<arch>/libfoo.so) and its target
(/system/lib64/libfoo.so) are now done int Soong.
I gave up the idea of always embedding jni libs to apps, due to a
hard-to-fix regression in storage usage. Specifically, consider this
case.
app --(jni_lib)--> libfoo
--(jni_lib)--> libbar
libfoo --(shared_lib)--> libbar
Ideally, with the embedding idea, both libfoo and libbar should be
embedded into the app and there should be no libfoo or libbar outside of
the apk, unless there's no dependency to any of them from outside of the
apk.
However, the previous implementation installed libbar to /system/lib64,
leading two copies of the lib; one in /system/lib64, the other in the
apk.
This was happening because libbar was also considered as a transitive
dep of libfoo, and therefore a dependency to its installation path is
added to the apk. The problem here is that the app doesn't know that
libfoo depends on libbar. We in theory can write a very delicate
dependency traverse routine which filters libbar out of the transitive
deps, but that looked too complicated.
Bug: 339923078
Bug: 330276359
Test: Build and watch any bloatbuster warning
Test: m aosp_cf_system_x86_64
The following three files are found
* system/lib64/libnfc_nci_jni.so
* system/etc/libnfc-nci.conf
* system/priv-app/NfcNci/lib/arm64/libnfc_nci_jni.so
Change-Id: I0930cb1ebb8ca8a6efd64b1ce2cdfd1c47fe19ef
* changes:
Revert "Revert^2 "Always embed jni libs and store uncompressed""
Revert "Revert "Revert "Collect transitve deps of jni libs only for bund...""
Revert "Add SkipToTransitiveDepsTag interface for dependency tags"
Revert "Install transitive deps of jni libs, but not the jni libs themselves"
Since we will not shrink the resources anyway this is a waste of time
Bug: 338429777
Test: m SystemUI
Change-Id: I388f045294131a304920756607792091627921bc
We do this because otherwise we will be embedding platform libs in the
bundled apps bloating their size.
Bug: 330276359
Test: Build Galery2 and check if libc++.so is not included there.
Change-Id: Ie866b192e9816bcb71b7ca3d74331d14d85592a4
Previously, unlike unbundled apps, if a platform app depends on a native
library, it is not embedded in the app, but installed to <partition>/lib
and the app is instead provided with a symlink to the lib. This actualy
is a legacy from the original Android where native libraries couldn't be
executed while embedded in the apk. To be executed, the native libs had
to be extracted to a mutable storage, but that had a risk of breaking
the verified boot, so libs couldn't be embedded.
Since API level 23, execute-in-place of native libs has become
possible. But platform apps had to opt-in (use_embedded_native_libs:
true) for the feature for a reason that is unclear today. Perhaps, it
was to save disk space in case when multiple apps share the same native
library, but such cases are found to be very rare, or non-existing.
With this CL, two changes are introduced:
1. jni libs are always embededd for all apps (bundled or unbundled)
2. use_embedded_native_libs is deprecated unless the module type is
android_test or android_test_helper_app. It is now a
no-op property. It's left just to not break existing bp files.
This will make unbundled apps bigger, but given that unbundled apps
built using android platform build system are tests, the size increase
should be acceptible.
Bug: 330276359
Test: m
Change-Id: I7df993ea85bce1c0a7222000c403a974818c3362
Bug: 331528424
Test: m --no-skip-soong-tests
Ignore-AOSP-First: Depends on internal changes. Will cherry-pick once merged.
Merged-In: Ied2821f11b6a5056ecf577e1e25765bc6dd212c0
Change-Id: Ied2821f11b6a5056ecf577e1e25765bc6dd212c0
This is a no-op change for a majority of cases.
Before this change, the contruction of the manifest check inputs is
confusing. It mutates uses_libs properties in place just for the
manifest check, by replacing module names with library names for
direct dependencies and merging library names from CLC for both direct
denpendencies and transitive denpendencies, and then constructs manifest
check inputs from those mutated uses_libs properties. This is
error-prone and leads to insistency: the goal is to check that the CLC
matches the manifest, but the inputs to the check don't reflect the CLC.
After this change, we no longer mutate uses_libs properties in place.
Instead, we maintain a separate list of missing denpendencies, and then
construct manifest check inputs directly from the CLC for all existing
libraries, no matter they are direct or transtive, and from the separate
list of missing libraries. This change makes the logic more
consistent and straightforward, and it also allows us to easily do the
next change, which is to propagate transtive missing denpendencies.
In fact, this change revealed several bugs around library optionality
and order in CLC construction, and fixed them.
Bug: 331528424
Test: m --no-skip-soong-tests
Ignore-AOSP-First: Depends on internal changes. Will cherry-pick once merged.
Merged-In: I0de82e76c47995b54aba9efd41538d950256a95f
Change-Id: I0de82e76c47995b54aba9efd41538d950256a95f
As part of aosp/3022586 where we added the idea of "test-only" modules
and top_level_test_targets, this CL implements that for java modules.
We let users set "test-only" on java_library, but not on other modules
where the module kind is implicitly test-only, like java_test.
The implementation, not the user decides it is test-only.
We also exclude it from java_defaults.
% gqui from "flatten(~/aosp-main-with-phones/out/soong/ownership/all_teams.pb, teams)" proto team.proto:AllTeams 'select teams.kind, count(*) where teams.test_only = true and teams.kind not like "%cc_%" group by teams.kind'
+--------------------------+----------+
| teams.kind | count(*) |
+--------------------------+----------+
| android_test | 1382 |
| android_test_helper_app | 1680 |
| java_fuzz | 5 |
| java_test | 774 |
| java_test_helper_library | 29 |
+--------------------------+----------+
% gqui from "flatten(~/aosp-main-with-phones/out/soong/ownership/all_teams.pb, teams)" proto team.proto:AllTeams 'select teams.kind, count(*) where teams.top_level_target = true and teams.kind not like "%cc_%" group by teams.kind'
+--------------+----------+
| teams.kind | count(*) |
+--------------+----------+
| android_test | 1382 |
| java_fuzz | 5 |
| java_test | 774 |
+--------------+----------+
Test: m nothing --no-skip-soong-tests
Test: go test ./java
Test: m all_teams
Bug: b/327280661
Change-Id: I9c3ad947dc3d68d6427abada27449526d69daa6b
This pairs with `test_module_config` but also works on the base is a
`java_test_host` module.
e.g.
test_module_config_host {
name: "CtsOsHostTestCases_DERIVED_2566",
base: "CtsOsHostTestCases",
test_suites: ["general-tests"],
include_filters: [ "android.os.cts.StaticSharedLibsHostTests" ],
exclude_annotations: [ "androidx.test.filters.FlakyTest","org.junit.Ignore" ],
}
The new module is composed of the previous and shares much of the same
code.
With respect to build size,
Without this change, if you build CtsAppSecurityHostTestCases, there
will be several copies of the jar (and related apks) :
*) 1 in framework
out/host/linux-x86/framework/CtsAppSecurityHostTestCases.jar
*) 1 in testcases for the test
out/host/linux-x86/testcases/CtsAppSecurityHostTestCases/CtsAppSecurityHostTestCases.jar
*) 1 per compatibility suite testcases
out/host/linux-x86/mts-documentsui/android-mts-documentsui/testcases/CtsAppSecurityHostTestCases/CtsAppSecurityHostTestCases.jar
out/host/linux-x86/mts-mediaprovider/android-mts-mediaprovider/testcases/CtsAppSecurityHostTestCases/CtsAppSecurityHostTestCases.jar
out/host/linux-x86/mts/android-mts/testcases/CtsAppSecurityHostTestCases/CtsAppSecurityHostTestCases.jar
out/host/linux-x86/cts/android-cts/testcases/CtsAppSecurityHostTestCases/CtsAppSecurityHostTestCases.jar
out/host/linux-x86/mts-mainline-infra/android-mts-mainline-infra/testcases/CtsAppSecurityHostTestCases/CtsAppSecurityHostTestCases.jar
A dervived test using CtsAppSecurityHostTestCases as base adds one more
to its testcases dir:
*) derived testcase.
out/host/linux-x86/testcases/CtsAppSecurityHostTestCases_presubmit_ExternalStorageApp/CtsAppSecurityHostTestCases.jar
Fixes: b/327280990
Test: m clean && m CtsOsHostTestCases_DERIVED_2566# as above
Test: atest CtsOsHostTestCases_DERIVED_2566 --collect-tests-only
Test: migrated the 71 TestMapping instances to Android.bp and build them.
Ran tests on some of them.
Ran some original `test_module_config` tests derived from `android_test` as well
TODO: Add actions to validate the given filters are valid for the given
test apks/jars.
Change-Id: I115eedb6ff6ba8e72bb49e71867daf49d25ca0f1
If the flag is set we will:
- pass --optimized-resource-shrinking to the r8 invocation
- use non final fields for the generated R classes
- not pass the aapt2 generated proguard rules, R8 will do the tracing of xml files
Bug: 325905703
Test: This is simply passing the flag through to R8, this is off by default
Change-Id: Ib2043f3201578c3bcd39c1de9a524fd78f6d795c
This change adds the support that was added to android_app in
https://r.android.com/2854663 for android_library modules.
Implementation details:
- Move `Flags_packages` to aaptProperties, so that it can be utilized
for both android_app and android_library.
- Wrap `VisitDirectDeps` of aconfig_declarations to a function that
takes a ModuleContext as an input, so that it can be utilized in the
`GenerateAndroidBuildActions` of both android_app and android_library.
Test: m nothing --no-skip-soong-tests
Bug: 330222981
Change-Id: I8a755f5ca615c8a1651afcd2ec441fc9fbd82c61
Bug: 329762127
Test: cd build/soong/java ; go test -run TestJavaLibHostWithStem
Test: cd build/soong/java ; go test -run TestAppStem
Test: cd build/soong/java ; go test -run TestAndroidAppImport_ArchVariants
Test: cd build/soong/java ; go test -run TestOverrideAndroidAppStem
Test: cd build/soong/java ; go test -run TestOverrideAndroidApp
Test: cd build/soong/java ; go test -run TestPackageNameOverride
Change-Id: I8fbdc82116f9a0c68121fbbeb1ef4f78f2b6fea3
This is a followup to aosp/2999198 and adds information about apps.
Each app will have an entry in this file with the following properties
- Name, mandatory
- Is_prebuilt, mandatory
- Prebuilt_info_file_path, optional
Implementation details
- Move prebuiltInfoProvider out of build/soong/apex to
build/soong/android. This allows build/soong/java to use it.
- Introduce a new `prebuilt_info` prop to `android_app_set` and
`android_app_import`
- All app module types will set a prebuiltInfoProvider in
GenerateAndroidBuildActions, including the source app module types
Test: m nothing --no-skip-soong-tests
Test: m out/soong/prebuilt_info.json
Test: ls -l out/soong/prebuilt_info.json --human-readable
-rw------- 1 spandandas primarygroup 317K Mar 11 23:46 out/soong/prebuilt_info.json
Test: #modified trunk_staging.locally to select prebuilts of some
mainline apps. Spot-checked that `is_prebuilt` and
`prebuilt_info_file_path` get populated appropriately
Bug: 327480225
Change-Id: I5078e0ec26c9568194550909962b90111a5223f7
New soong module type `TestModuleConfig`
This module type allows to write:
test_module_config {
name: "DerivedFrameworks1",
base: "FrameworksServicesTests",
include_filters: ["com.android.server.accessibility.FingerprintGestureDispatcherTest"],
}
This goal is to put the test options (include_fitler in this case) in
Android.bp rather than TEST_MAPPING to be able to name a set of options
and to allow the CI to more precise about which part of a TestModule
might be failing.
In general, it is recommended to split up tests rather than use this
approach, but that isn't always feasible.
More details on project here: https://docs.google.com/document/d/1MpA4BS6mTr-1D-K9MVmzdHkSufAvIkkM4gPEUglNodk/edit?tab=t.0
Migration plan here: https://docs.google.com/document/d/1x3w-BvKO3E-GbY8YOSBuzJC6-pkLkuhfjSyZ1hJQBYM/edit?tab=t.0
Some details of final output:
* All the [data] files from Derived are the same as base.
* The apk names is changed to be the Module name
* The test options are added to .config file
* test-file-name in the .config file (for base.apk) is changed to be
new apk.
* module-info.json is hand inspected between base and derived.
* I added a .manifest file so future tools know what base is, if they
need it.
We are working a mechanism to use a CAS so the zip of the testcases
won't explode. (however the trident disk image will get bigger)
% tree -ls out/target/product/shiba/testcases/DerivedFrameworks1 out/target/product/shiba/testcases/FrameworksServicesTests
[ 4096] out/target/product/shiba/testcases/DerivedFrameworks1
├── [ 4096] arm64
│ └── [ 72346281] DerivedFrameworks1.apk
├── [ 20887] BstatsTestApp.apk
├── [ 4096] data
│ └── [ 851] broken_shortcut.xml
├── [ 3157] DerivedFrameworks1.config
├── [ 12695] JobTestApp.apk
├── [ 8599] MediaButtonReceiverHolderTestHelperApp.apk
├── [ 16791] SimpleServiceTestApp1.apk
├── [ 16791] SimpleServiceTestApp2.apk
├── [ 16791] SimpleServiceTestApp3.apk
├── [ 1017540] SuspendTestApp.apk
└── [ 36] test_module_config.manifest
[ 4096] out/target/product/shiba/testcases/FrameworksServicesTests
├── [ 4096] arm64
│ └── [ 72346281] FrameworksServicesTests.apk
├── [ 20887] BstatsTestApp.apk
├── [ 4096] data
│ └── [ 851] broken_shortcut.xml
├── [ 2866] FrameworksServicesTests.config
├── [ 12695] JobTestApp.apk
├── [ 8599] MediaButtonReceiverHolderTestHelperApp.apk
├── [ 16791] SimpleServiceTestApp1.apk
├── [ 16791] SimpleServiceTestApp2.apk
├── [ 16791] SimpleServiceTestApp3.apk
└── [ 1017540] SuspendTestApp.apk
Fixes: 314148134
Test: atest DerivedFrameworks1 DerivedFrameworks2
Test: m blueprint_tests
Change-Id: Ib73a3404557e0bd583b065f0cf2fd55fba9ccdbe
At ToT, an optional_uses_libs is not added to the build time CLC if it
does not exist in the tree. One edge case here is
java_sdk_library_import, which might exist in the tree, but without an
implementation. This cause issues during analysis when we try to verify
the correctness of the build time CLC.
This CL disables dexpreopt if a dependency does not have an
implementation. To limit inadvertent side effects, this is restricted to
java_sdk_library(_import) module types. (more precisely, it is restricted
to java_sdk_library_import, since the source module type will always
have an impl)
Bug: 315802285
Test: Added a unit test
Test: m nothing
Test: printf debugging in internal main, verified that this CL does not
disable dexpreopt on any android app inadvertently
Change-Id: I173fc2f3ff654fe4091e9a43322164afd3222ee7
Overridden packages can have different package names. This change reads
output APKs to parse package name, instead of manifest xml.
Bug: 325011777
Test: boot and idmap2 dump
Change-Id: I748b9db8bd6282d7f3a1478cf94b21580c7683e7
Because
1) it has no effect.
2) it will simplify build commands, which will minimize differences
among targets using generate_product_characteristics_rro
Bug: 322932641
Test: TH
Change-Id: I16afae597038da7489989483f8c926916d207df5
