- Adds all required details for bootclasspath_fragment to implement
classpath_fragment.
- Keeps the actual boot jars in platform-bootclasspath to begin with.
- Makes sure to put the file in apex/etc/classpath on device. Note that
for platform versions of classpath fragment AndroidMkEntries perform
the installation, while for APEXes it must be plumbed via apex.go.
Bug: 180105615
Test: m && launch_cvd; atest CtsClasspathsTestCases
Change-Id: I6101ebdf5b8bcbe95c0b7ce21f3f67a2685aef50
Previously, deapexer modules were created by the LoadHook which meant
that the deapexer module could not use any information derived from the
dependencies of the prebuilt_apex/apex_set modules. This change moves
the creation into a separate mutator that runs after
ComponentDepsMutator and before DepsMutator. That means that a follow
up change can use information from dependencies added by the former
mutator in order to create the deapexer module and the deapexer module
can itself add dependencies onto other modules.
This change also dedups the logic to determine whether a deapexer
module is needed by pushing it down into the createDeapexerModule
method which was renamed to createDeapexerModuleIfNeeded to reflect its
conditional nature.
Bug: 187266082
Test: m nothing
m SOONG_CONFIG_art_module_source_build=false nothing
Change-Id: I65316473ff1e4b2827ff48ab5a870a8ce5c0475a
Should manually cut dependencies instead. Previously, this was used to
take advantage of the limited dependencies of variant libs in some APEXes.
Fixes: 175318864
Test: m nothing
Change-Id: Id559fd1bea5c8b8696cf7ab8acc711a7c3844d14
The dex boot jar for the apex must have had hidden API flags encoded
into it. Currently, the hidden API processing is done within the java
modules themselves so the apex gets the dex boot jar from them.
However, as part of the hidden API modularization work the hidden API
encoding will be performed by the bootclasspath_fragment so this change
prepares for that by delegating the retrieval of the dex boot jars to
the bootclasspath_fragment, via BootclasspathFragmentApexContentInfo.
For the moment that simply delegates straight back to the java module
so this change does not change the build. It will however make it
easier to switch hidden API encoding to the bootclasspath_fragment in
future.
Bug: 179354495
Test: m com.android.art
- verify that this change does not change its contents
Change-Id: I12eba333749be976bcc72661bb9d6be6cc3c56e3
The apexBundle.GenerateAndroidBuildActions is 400+ lines long which
makes it difficult to understand what is happening.
Bug: 177892522
Test: m com.android.art
- verify that this change does not change its contents
Change-Id: I07cb31d246377869771bdb18fdc6291365adf399
Currently, it only contains ART boot image related information, i.e.
.art, .oat and .vdex files. However, follow up changes will extend that
to include other information from bootclasspath_fragment.
Bug: 177892522
Test: m nothing
Change-Id: I2b226131c0eccff0c739a18f265f90caa10a91d9
Also renames files, tests, module types in a similar fashion.
There are still some references to image and boot image. They are kept
for the following reasons:
* image_name - this is the name of an ART boot image, i.e. the
collection of .art/.oat/.vdex files.
* BootImageInfo - again this is related to the ART boot image.
* .../art_boot_images/... paths - ditto.
Bug: 177892522
Test: m nothing
Change-Id: Ie1f4738061d131fee75de48bc26a7601481bad4d
This change ensures that bootclasspath_fragment only depends on source
modules and prebuilt_bootclasspath_fragment only depends on prebuilt
modules.
It does that in two ways:
1. It adds the dependencies in ComponentDepsMutator method which is
called before any renaming of prebuilts is done which makes it very
easy to add a dependency directly onto either the source or prebuilt
as required.
2. It uses a tag which prevents dependencies on a source module from
being replaced with a dependency on a prebuilt module which ensures
that a dependency on the source modules is not replaced with a
dependency on a prebuilt module.
Bug: 177892522
Test: m nothing
Change-Id: Ibcfae39083afbc07fcf729ead3ed5f5d020845bf
ApexInfo is not part of the properties struct. It can handle structs
having private fields.
Bug: 1663140
Test: m
Change-Id: Ib07d4410f0ce187c9de347da34b84b814b2eb537
Previously, SdkSpec was constructed only from the user string. It didn't
make use of the Config struct where information about the latest stable
SDK version, etc. is recorded. As a result, the build system couldn't
check if the sdk version "current" is referring to the in-development
(i.e. not-yet-frozen) SDK version or the latest stable version.
"current" was always assumed to be in-development (IsPreview() returns
true) even when Platform_sdk_final == true.
As the first step for fixing that, this change requires
android.EarlyModuleContext to be passed when constructing SdkSpec from
the user string.
In the following changes, "current" will be mapped to either
FutureApiLevel (10000) or one of the FinalApiLevels() depending on
whether the platform SDK was finalized or not.
Bug: 175678607
Test: m
Change-Id: Ifea12ebf147ecccf12e7266dd382819806571543
This change fixes a bug that shared lib dependencies of an rlib is not
installed to the APEX even when the rlib is part of the APEX.
Bug: N/A
Test: m
Change-Id: I88fe461584499839d8018d6b4292374592e7562b
Previously, SdkSpec was constructed only from the user string. It didn't
make use of the Config struct where information about the latest stable
SDK version, etc. is recorded. As a result, the build system couldn't
check if the sdk version "current" is referring to the in-development
(i.e. not-yet-frozen) SDK version or the latest stable version.
"current" was always assumed to be in-development (IsPreview() returns
true) even when Platform_sdk_final == true.
As the first step for fixing that, this change requires
android.EarlyModuleContext to be passed when constructing SdkSpec from
the user string.
In the following changes, "current" will be mapped to either
FutureApiLevel (10000) or one of the FinalApiLevels() depending on
whether the platform SDK was finalized or not.
Bug: 175678607
Test: m
Change-Id: Ifea12ebf147ecccf12e7266dd382819806571543
When `test_for` dependencies are added from libraries to APEX modules,
they can be created from the platform variants of the libraries, since
those are used for building tests. Hence we need an alias from the
platform variant of the APEX module to have a target for those
dependencies.
This is only necessary for libraries that are split by the APEX
mutator, i.e. is a member of some APEX. Normally that's not the case
for test libraries, but there may be exceptions (read
com.android.art.testing).
This relands https://r.android.com/1654679 after decoupling it from the
topic that caused b/184239856.
Test: m nothing
Bug: 183882457
Change-Id: If643c75ce9bc25fa01ad9d1e3ba8e1d060d03bb2
Remove Android-targeting gcc toolchain libraries -- libgcc,
libgcc_stripped, libatomic, and libgcov. Also remove libunwind_llvm,
which is replaced with a libunwind toolchain prebuilt.
The __atomic_* library functions are now part of the compiler-rt
builtins library:
https://android-review.googlesource.com/c/toolchain/llvm_android/+/1625025
Bug: http://b/153025717
Test: treehugger
Change-Id: I971d0a4a49f1aaeb3546e80b6d94208277a171ac
Revert submission 1658000
Reason for revert: Breaks full-eng build: b/184239856
Reverted Changes:
I4f8ead785:Avoid internal APEX stubs for libsigchain and clea...
I68affdf69:Allow dependencies from platform variants to APEX ...
I54b33784e:Rename libdexfile_external_static to libdexfile_st...
Id68ae9438:libdexfile_external is being replaced by libdexfil...
I12ac84eb4:libdexfile_external is replaced by libdexfile.
If05dbffc8:Rename libdexfile_external_static to libdexfile_st...
Ia011fa3a8:Merge libdexfile_external into libdexfile.
Change-Id: If494dc5385042a4620a76a9eadc1613ae0eb1655
When `test_for` dependencies are added from libraries to APEX modules,
they can be created from the platform variants of the libraries, since
those are used for building tests. Hence we need an alias from the
platform variant of the APEX module to have a target for those
dependencies.
This is only necessary for libraries that are split by the APEX
mutator, i.e. is a member of some APEX. Normally that's not the case
for test libraries, but there may be exceptions (read
com.android.art.testing).
Test: m nothing
Bug: 183882457
Change-Id: I68affdf69d7ec05c0ee8730e8ec04d7cb9e0e44a
This avoids having to specify boot libraries in both the boot_image
and separately as java_libs on the apex. Simply add them to the
boot_image (happens automatically ATM when using image_name: "art")
and add the boot_image to the apex.
Bug: 177892522
Test: m nothing
Change-Id: I7e0c41665604b73780cdf0dc555067497b1e6ef0
This is part of the work to rename boot_image to bootclasspath_fragment
which is being done for two reasons:
1. To avoid clashing with the bootimg module type.
2. To better reflect what this represents.
While a bootclasspath_fragment can create what ART calls a boot image
(which is different to what the bootimg module type represents) it does
not have to do so.
Bug: 177892522
Test: m nothing
Change-Id: Ib45604be7adc790ded9e27a2ac812dd7522ca8db
Creates a new deptag type for it so that it can implement the marker
interfaces that will exclude it from being added to the APEX and from
visibility enforcement. The latter is probably not an issue ATM because
the dependencies are added after visibility checks are enforced but
this code is undergoing lots of refactoring so that may change.
Bug: 177892522
Test: m nothing
Change-Id: Ibd167d557adec761a2e3eed78f4d334c40a04fb9
The ExcludeFromApexContentsTag marker interface was added to avoid
every implementation of DepIsInSameApex() from having to deal with the
special tags, like PrebuiltDepTag. Unfortunately, when adding that
not all calls to DepIsInSameApex() were protected which meant that the
BootImageModule, which panics if it doesn't recognize a tag, was
causing failures. This change documents the need and improves the
consistency.
A follow up change will add a test for this.
Bug: 182992071
Test: m nothing
Change-Id: If0bf9a7447ebf7a0bb0c88e91951a7220d4af45c
Adds the AlwaysRequireApexVariantTag interface to enable
ApexInfoMutator to differentiate between a tag that is excluded from
apex contents but still requires an apex variant and a tag that is
excluded from apex contents and does not require an apex variant.
That is needed to support the sdkMemberVersionedDepTag which excludes
the target from being added to the APEX but requires an APEX variant.
A more detailed explanation is in the comments.
The AlwaysRequireApexVariant() method follows the pattern used in
ReplaceSourceWithPrebuilt of having a method that returns a bool to
trigger the behavior and not say ExcludeFromApexContentsTag that simply
relies on the tag implementing an interface to trigger. That is because
the former is more flexible and allows a tag type to parameterize the
behavior if necessary.
The tags that this will exclude from creating an apex variant are:
* PrebuiltDepTag - by the time the apex variant has been created any
preferred prebuilts will have replaced the sources so there is no
need to create an APEX variant if the only dependency path from the
APEX to the prebuilt is via this tag.
* hiddenApiAnnotationsDependencyTag - the target of which is a purely
build time artifect and MUST NEVER end up in the APEX.
It will also stop calling DepIsInSameApex for any dependency created
by the sdkMemberVersionedDepTag. Which will fix the issue reported in
the bug.
Bug: 182992071
Test: m nothing
Change-Id: I9569e488d6446ca45d3ea8f32a9b74524eb865df
It just provides the metadata needed by the global singleton as the
rest is in the apex.
Bug: 182402754
Test: m nothing
Change-Id: I511df7a3a06dab13ddb9ad63392ae5310dfee9c4
An upcoming change to create a prebuilt_platform_compat_config module
will break if the apex tries to use it instead of a
platform_compat_config because the former does not provide all the
information that the apex needs. This change will allow the
compatConfigsTag to be configured to prevent the prebuilt from being
used even when it is preferred.
Bug: 182402754
Test: m nothing
Change-Id: Ib9dc06c038f7cf3fc229f3c4d2b025335a4715b4
While debugging an issue with some work I was doing on boot image
modules I noticed that markPlatformAvailability() is calling
DepIsInSameApex() even when the dependency tag indicates that it is
excluded from the apex contents.
Test: m droid
Change-Id: Iac49049546a886f7a3a0d9640ffd31ce24c61364
1) suggest a fix at the end of the message
2) add new lines around the dependency path so that they are visually
separated from rest of the error message
Bug: N/A
Test: m with an intentional break
error: bionic/apex/Android.bp:32:1: module "com.android.runtime" variant "android_common_com.android.runtime_image": "libutils_headers" requires "libsystem_headers" that doesn't list the APEX under 'apex_available'.
Dependency path:
via tag apex.dependencyTag: { name:executable payload:true}
-> crash_dump{os:android,image:,arch:arm_armv8-a,sdk:,apex:apex10000}
via tag cc.libraryDependencyTag: { Kind:staticLibraryDependency Order:normalLibraryDependency wholeStatic:false reexportFlags:false explicitlyVersioned:false dataLib:false ndk:false staticUnwinder:false makeSuffix: skipApexAllowedDependenciesCheck:false excludeInApex:false}
-> libtombstoned_client_static{os:android,image:,arch:arm_armv8-a,sdk:,link:static,apex:apex10000}
via tag cc.libraryDependencyTag: { Kind:staticLibraryDependency Order:normalLibraryDependency wholeStatic:true reexportFlags:true explicitlyVersioned:false dataLib:false ndk:false staticUnwinder:false makeSuffix: skipApexAllowedDependenciesCheck:false excludeInApex:false}
-> libcutils{os:android,image:,arch:arm_armv8-a,sdk:,link:static,asan:,apex:apex10000}
via tag cc.libraryDependencyTag: { Kind:headerLibraryDependency Order:normalLibraryDependency wholeStatic:false reexportFlags:false explicitlyVersioned:false dataLib:false ndk:false staticUnwinder:false makeSuffix: skipApexAllowedDependenciesCheck:false excludeInApex:false}
-> libutils_headers{os:android,image:,arch:arm_armv8-a,sdk:,asan:,apex:apex10000}
via tag cc.libraryDependencyTag: { Kind:headerLibraryDependency Order:normalLibraryDependency wholeStatic:false reexportFlags:true explicitlyVersioned:false dataLib:false ndk:false staticUnwinder:false makeSuffix: skipApexAllowedDependenciesCheck:false excludeInApex:false}
-> libsystem_headers{os:android,image:,arch:arm_armv8-a,sdk:,asan:,apex:apex10000}
Consider adding "com.android.runtime" to 'apex_available' property of "libsystem_headers"
Change-Id: I09f92c3086ea433780133a33ba0ad73baee6dc41
Originally, when the prebuilt_apex was first created, it selected the
source to use in its DepsMutator. It did that because that was a
convenient place for it to perform that work which had to be:
* After the arch mutator had run so MultiTargets() was available.
* Before the prebuilt_select mutator runs as that relied on the Source
property to have been set.
Change 064b70c9 then duplicated the call from the DepsMutator of the
deapexer module type that was added as part of the work to make dex
files available for hiddenapi processing.
Change 356f7d45 moved it out of the the DepsMutator methods into its
their own mutators, presumably because it interfered with the
Soong -> Bazel conversion work.
This change improves the existing PrebuiltSrcsSupplier mechanism to
support reporting errors so that the logic for selecting the source can
be done on demand rather than in separate mutators.
The main complication was that PrebuiltSrcsSupplier is called with a
BaseModuleContext for both source and prebuilt modules so it cannot use
any methods on it that are related to the current module. That
necessitated adding MultiTargets() to android.Module.
Bug: 181267622
Test: m droid
Change-Id: I106c78fd21016f051a315b82b470d8f12b1f820b
Some filesystems (like boot_image type) are arch-specific. This change
allows us to have filesystems property inside the arch.<arch> struct.
Bug: 178978059
Test: m
Change-Id: I09faa6186b4a3fa35f348fc0805d894b4dc27ffd
This currently expands all globs, still need to support converting glob
syntax.
Test: go build_conversion_test
Test: GENERATE_BAZEL_FILES=true m nothing
Test: m nothing
Bug: 165114590
Change-Id: If7b26e8e663d17566fad9614ca87a8da1f095284
Previously, the apex module had to hard code behavior specific to the
art apex module in order to include the art boot image. This change
adds support to the apex module to allow the boot images to be
specified per apex.
In combination with a change to add the "art-boot-image" to the ART
apex this allows the custom code for handling the art boot image in
apex to be removed.
That custom apex code also included the logic to ensure that the
GlobalSoongConfig was initialized for use by the dex_bootjars
singleton. That logic has been moved from the APEX to the boot_image
module. That ensures that it will be run if and only if a boot_image
module is present in the checked out repos. So, limited manifest
checkouts which do not contain the art or frameworks/base repos (which
is where the boot_image modules are defined) will not attempt to run
this logic, which would fail because dex2oat would not be present.
Bug: 177892522
Test: m droid
Change-Id: I02d25fbef6e864e31eb5e0f4eb50358c79486db0
If we don't check "current", it won't be checked even in the finalized
branch.
If we don't check "preview", it should be done during the SDK
finalization. It'd be better done before the SDK finalization regarding
that setting min_sdk_version is to get approval from deps library owners.
Bug: 177833148
Test: m (soong tests)
Change-Id: I712b61cfe5a134fbb69c73956d26fb3a1e5c011e
This reverts commit 556b2ad77e.
Original change-id: Ib1b2f0dd2f9ae85b1545c6cc5bb4c5bbdfac1c15
Reason for revert: Add excutable permission to the gen_ndk_backedby_apex.sh.
Test: TARGET_BUILD_APPS=com.android.adbd m dist apps_only
Change-Id: Ib587ba200cd7f2f61d478452b43c329d72b2de06
Add gen_ndk_usedby_apex.sh script to generate NDK API list used by Mainlain modules when modules get build.
Test: TARGET_BUILD_APPS=com.android.adbd m dist apps_only
Change-Id: Ib1b2f0dd2f9ae85b1545c6cc5bb4c5bbdfac1c15
Bug: 176171716
Test: build_mainline_modules.sh and ensure that the ART boot image is
built: `find out -type f -name '*.art'` is nonempty for every arch.
Change-Id: Ibc45581eef2b205c750a30709780cf659ba7cfa1
Currently, when a module is included in an APEX, the dependencies of the
modules are listed in LOCAL_REQUIRED_MODULES of the APEX. There are two
purposes for this:
1) for native dependencies, they are installed to
$(TARGET_OUT)/apex/<apexname> directories which isn't packaged as an
*.img. However, as a side effect of the installation, their symbol files
are placed under $(TARGET_OUT)/symbols directory to aid debugging.
2) to implement the symlink optimization. When the APEX is not
updatable, the dependencies are not included inside the APEX, but
installed directly to /system partition because the same files might be
used outside of the APEX. The files in the APEX are replaced with
symlinks to the system copy.
So far, the module name like "libfoo" was directly used in
LOCAL_REQUIRED_MODULES. This becomes problematic when only a single arch
variant of the module is used by the APEX. The build system will install
both arch variants to the system partition.
This change fixes the problem by appending ":32" or ":64" suffix
when composing LOCAL_REQUIRED_MODULES.
Bug: N/A
Test: m
Test: Cherry-pick I285c5d1bb9b27265c8589f2588d95eafa324d412 and its
dependencies from internal master. `m nothing` doesn't show the artifact
path requirement error.
Change-Id: I78feae1d5b18f93b0f984d3b1558812fd1689a96
With the addition of the compile_dex property to the
java_import module it becomes possible to include a
java_import module in an apex. This change allows the
dependency and adds a test.
Bug: 177228901
Test: m nothing
Test: new TestApexWithJavaImport
Change-Id: I9336dade1857109e2fd21f7d57e1dc4abc4a402c
Dexpreopt and boot jars package check all require access to dex
implementation jars created for java_library and java_sdk_library. They
were available when building from source but not when building from
prebuilts, even though they are embedded within the .apex files that
are referenced from prebuilt_apex.
This changes adds support to prebuilt_apex to export the dex
implementation jars and updates java_import to use those exported dex
implementation jars.
In a source build dexpreopt/boot jars package check access the apex (or
platform) specific variant of a java_library, e.g. core-oj, from which
it retrieves the dex implementation jar path.
After this change in a prebuilt build dexpreopt/boot jars package check
behave in the same way except in this case they retrieve the dex
implementation jar path from the apex (or platform) specific variant of
the java_import, e.g. core-oj.
The work to export files from a `.apex` file for use by other modules
is performed by a new `deapexer` module type. It is not used directly
in an `Android.bp` file but instead is created implicitly by
`prebuilt_apex`,
In order to do that this contains the following changes:
* Adds a new `dexapexer` module type to handle the exporting of files
from the `.apex` file.
* Adds an exported_java_libs property to prebuilt_apex to specify the
set of libraries whose dex implementation jars need exporting.
* Creates apex specific variants of the libraries listed in the
exported_java_libs property.
* Adds the set of exported files to the ApexInfo to make them available
to the apex specific variants.
* Prevents the prebuilt_apex variants from being merged together as
they will not be compatible.
* Modifies java_import to use the exported file for variants of a
prebuilt_apex.
* Adds a ninja rule to unpack (using deapexer) the contents of the
prebuilt_apex's apex file, verify that the required files are present
and make them available as outputs for other rules to use.
* Some minor refactorings to support these changes.
* Adds tests to cover prebuilt only, prebuilt with source preferred,
and prebuilt preferred with source.
Test: m nothing
Bug: 171061220
Change-Id: Ic9bed81fb65b92f0d59f64c0bce168a9ed44cfac
For the platform libc++ STL, remove the
-Wl,--exclude-libs,libunwind_llvm.a argument, which is redundant with
the same argument in deviceGlobalLdflags.
Bug: http://b/153025717
Test: device boots
Change-Id: Idd7791d52f74aab2d5f59419fb75f841fc29a2eb
android_filesystem modules can be included in APEX via the new
`filesystems` property. The filesystem images are placed at
./etc/fs/<modulename>.img.
Bug: 172413888
Test: m nothing
Change-Id: I215ca7a32ff1988a0de4e1f71397684e189839ea
The APEX symlink optimization is a build-time trick to save the
storage/RAM usage of APEX by replacing some files in APEX with symlinks
to the files in the system partition. The optimization however is
automatically turned off for 'updatable: true' APEXes because doing the
optimization for them will hide the sys-health implication until when
the APEXes are built unbundled (i.e. prebuilt) and thus the
optimization is impossible.
TARGET_FORCE_APEX_SYMLINK_OPTIMIZATION forcibly disables the safety net.
When it is set to true, the symlink optimization is done regardless of
the 'updatable' property. This is useful for some of the devices like Go
where most APEXes (even the 'updatable: true' ones) should be
effectively non-updatable.
Bug: 175630508
Test: TARGET_FORCE_APEX_SYMLINK_OPTIMIZATION=true m and check that
updatable APEXes have symlinks to system libs
Change-Id: I26f72e5d5ebccc2d1e09c2a2f743db14937eb39a
The NDK build in downstream branches with prebuilt apexes break
with multiple rules to copy the apex after
I8038ed5c6f05c989ac21ec06c4552fb3136b9a7a renamed SkipInstall to
HideFromMake, which is shadowed by an existing HideFromMake method.
Call through from the existing HideFromMake as a quick-fix for the
build.
Fixes: 175911330
Test: NDK build
Change-Id: Id475671d5dad888913bfb1af2e4be1f2454c3211
This fixes a regression in https://r.android.com/1515154 when building
the host ART APEX (com.android.art.host).
Test: art/build/apex/runtests.sh
Bug: 166468760
Change-Id: I39a00019b7bfa77a865e7287c66add3566b34d87
Refactor the apexInfoMutator to delegate the work to an implementation
of ApexInfoMutator and then move the existing functionality into the
apexBundle.ApexInfoMutator(). This will allow a follow up change to
customize the behavior of this mutator for prebuilt_apex modules.
Test: m nothing
Bug: 171061220
Change-Id: I7ccd55818e02a606c5494e215f4370b635ddd0a5
This change fixes a bug that headers are not re-exported from stub
variants of a library, if the headers are not from header libs, but from
shared or static libs. This is because only header lib dependencies
are respected for stubs variants.
The fix is as follows. 1) dependencies to the shared/static libs are
added even for stubs variants. 2) instead, in depsToPaths, they are
treated like header libs (i.e. don't contribute to linkFile) for the
stubs variants.
Bug: 174558745
Test: m
Change-Id: Iab6c77e7817055d0f2d09cb114186b30164fc231
Generally ...DepsMutator methods add dependencies between modules but
the apexDepsMutator does not which can be confusing. This renames
apexDepsMutator to apexInfoMutator and adds some extra documentation to
clarify its function. It also renames the registered name for the
mutator and its associated providers from apex_deps to apex_info.
Test: m nothing
Bug: 171061220
Change-Id: Ic074a281215b23e982448ccff7ac075236123bee
This CL fixes a bug that rust_ffi_shared libraries can't be included in
an APEX either directly (via native_shared_libs property) or indirectly.
Bug: 175105284
Test: m
Change-Id: I2e2d3e11b3901f3412b9725e47a8ebf31f93847b
1. Soong can now detect PRODUCT_COMPRESSED_APEX flag
We don't want APEX to be compressed on all devices. Only those that
have explicitely set PRODUCT_COMPRESSED_APEX flag.
2. Handle "compressible" field in soong build rule
On devices that supports APEX compression, all APEX will be
compressed by default. If any apex does not want to be compressed,
they will need to state that by setting "compressible" field to
false
3. Can use apex_compression_tool to compress APEX
Note we compress the APEX after it has been signed. That way, when
we decompress we will get a signed APEX.
4. Place the compressed APEX in system with .capex extension
This makes it easy to identify. We still preserve the original
extension so that when we decompress, we can just rename by cuttif
off the .capex extension.
Note: with this change, we can create a system image with compressed
APEX, but we cannot boot with it since platform doesn't know how to
handle .capex files. Platform support will be added on follow up CLs.
Bug: 172911362
Test: OVERRIDE_PRODUCT_COMPRESSED_APEX=true m (apex_test.go)
Test: observed $OUT/system/apex has .capex files
Change-Id: I20ac4c4ceb521924c751a6017f979b2d808fdded
The property is used to exclude some shared and static libs when the
module is built for an APEX.
Bug: 166468760
Test: m
Change-Id: I0dcaa4ae94c01aa00dc5539c60d3054c57fd8824
The distinction between foo and prebuilt_foo doesn't exist in make, so
this could create invalid dependencies when both source and prebuilt
modules exist and the prebuilts are preferred.
Test: `m` with ART module prebuilts with prefer:true in the tree
Test: m nothing
Bug: 172480615
Change-Id: I90b76a8f38493882b3330d1b6789159852e59d55
Add gen_ndk_usedby_apex.sh script to generate NDK API list used by Mainlain modules when modules get build.
Test: TARGET_BUILD_APPS=com.android.adbd m dist apps_only
Forrest run: android-build.googleplex.com/builds/forrest/run/L97800000742565481
Change-Id: I39e5aa7c74eff46aba06808642b2bd67ae45ba1e
We will have some APEXes having rust binaries and libraries. So, adding
the support for the types of modules.
rust.Module now inherits from android.ApexModuleBase and implements
the android.ApexModule interface.
Bug: 172414324
Test: m
Exempt-From-Owner-Approval: rebased after +2 from the owner
Change-Id: I356ef4c45f782a6460f001e83af96d1710642d80
A previous change handles dist properties automatically for all module
types and as a result has made the apexBundle setting of DistFiles
redundant so this change removes that.
Test: m nothing
m dist sdk - before and after this change, compare result to
make sure that there are no significant differences.
Test the effect on the apex by following instructions in
http://b/172951145
Bug: 174226317
Change-Id: If6f741b5b73b3335f4c478a6528f5f311cd633e4
Change https://r.android.com/1335521 added tag property to the Dist
struct so that it could be used to select one of a number of different
output files to copy to the dist instead of the single file that the
module type made available for dist. The output files were selected
by passing the tag to OutputFiles(tag).
Module types that wanted to support this new approach had to explicitly
set AndroidMkEntries.DistFiles = GenerateTaggedDistFiles(module).
Unfortunately, doing that had a side effect of changing the behavior of
dist entries without a tag.
That was because the change treated a tag that was not specified, as
being the same as "". So, prior to the change no tag meant use the
default dist file but after it meant use the paths returned by
OutputFiles(""). That changed the behavior of the java.Library type
which affected the behavior of the android_app module type.
Prior to the change the java_library would make the
Library.outputFile available for dist when no tag was specified. After
that change it would make Library.outputFile plus
Library.extraOutputFiles. The latter is usually empty except for
android_app which adds some extra files into there which will now be
copied to the dist. That change may have been intentional but there
was no mention of it in the change or the bug. Even if it wasn't
intentional it may still be beneficial.
Any module type that wants to add support for tags in dist runs the
risk of introducing similar changes in behavior. This change
differentiates between the tag not being set and the tag being set to
"" to avoid that possibility and to make the default behavior
explicit for those module types that have switched.
It does so as follows:
* Adds a DefaultDistTag constant that is used when the tag is not set.
It is a string that is unlikely to be used as an actual tag as it
does not start with a . and uses some special characters.
* The DefaultDistTag is used in MakeDefaultDistFiles(paths) to indicate
that the supplied paths are the default ones and and also in
GenerateTaggedDistFiles() for Dist structures that have no tag
property set.
* The DefaultDistTag is passed to OutputFiles(tag) just in case the
module type has explicitly defined the paths to associate with that
tag in there. If it has then it overrides the legacy behavior. If it
has not then it is just ignored and falls back to using the previous
behavior.
* The java.Library.OutputFiles(tag) method explicitly handles the
DefaultDistTag and returns Library.outputFile for it which restores
the behavior from before the change that added dist.tag support.
* Similar change was made to apexBundle.OutputFiles(tag) in order to
preserve its previous behaviour.
* The customModule used by TestGetDistContributions has been modified
to also preserve its previous behavior after this change.
Test: m nothing
m dist sdk - before and after this change, compare result to
make sure that there are no significant differences.
Test the effect on the apex by following instructions in
http://b/172951145
Bug: 174226317
Change-Id: Ib8f0d9307751cc2ed34e3d9a5538d3c144666f6d
Mostly documentation changes, but includes a few refactorings like
changing the variable names, reordering functions, reordering statements
in logical order, etc.
Bug: 173472337
Test: m
Change-Id: Iefdc8e463bcda60187b98e8c90661e220e8cdf40
Mostly documentation changes, but includes a few refactorings like
changing the variable names, reording functions, reordering statements
in logical order, etc.
Bug: 173472337
Test: m
Change-Id: Ie1799c0972d63da823ad375f008018de782529d1
Mostly documentation changes, but includes a few refactorings like
changing the variable names, reording functions, reordering statements
in logical order, etc.
Bug: 173472337
Test: m
Change-Id: I000c76e818722ed06bac03d9de87588b23552b08
The source code is reorganized following the order of the execution for
better readability. The order is as follows:
1) init
2) properties
3) module struct
4) deps mutator
5) post-deps mutator
6) generate build actions
7) module initialization for different sub types
8) misc (e.g. apex_available check, etc.)
Behavior is not touched.
Bug: 173472337
Test: m
Merged-In: I919fca468fa376273d498029e6ac2fa37998a4a6
Change-Id: I919fca468fa376273d498029e6ac2fa37998a4a6
which was added to support built-time common apex. But the common apex
was abandoned and it is used nowhere.
This is a manual revert of 5c998b9ff8.
Bug: 173329465
Test: m
Change-Id: I684e6b8bfe2ef5b53943329f4aaa5a5ccba27738
This error tracks `apex_available` closure, but the message doesn't
indicate this variable specifically. Calling it out to try to make
the problem more searchable/discoverable/intuitable.
Bug: N/A
Test: apex_test.go
Change-Id: Ib254ace1dbd4e77d073ed4f98ee181dd86adfcfc
Prepare for making the relationship between an llndk_library stubs
module and the cc_library implementation module explicit by
adding an llndk_stubs property. Each cc_library will be updated
to point to its llndk_library, and the llndk_library name will
be changed to make the .llndk suffix explicit. Then the implicit
connection and suffix can be removed.
Bug: 170784825
Test: m checkbuild
Change-Id: I6b0482a3f286ec29b2e928551aa4317749f2b499
A global variant was used to store the global mapping between
modules and APEXes. Replace it with storing pointers to APEX
contents inside each module so that they can query the contents
of any APEXes they belong to.
Bug: 146393795
Test: all Soong tests
Test: single line change to build.ninja host install dependency ordering
Test: no Android-${TARGET_PRODUCT}.mk, make_vars-${TARGET_PRODUCT}.mk or late-${TARGET_PRODUCT}.mk
Change-Id: Id2d7b73ea27f8c3b41d30820bdd86b65c539bfa4
When building an APEX, the build system included artifacts from all
targets that are returned from ctx.MultiTargets(). This however has
became a problem as we add new host targets like linux_bionic_arm64.
When there are multiple host targets having different OSes,
ctx.MultiTargets() returns one target per a host OS. For example, when
linux_bionic_arm64 is enabled, ctx.MultiTargets() returns the following
two targets because linux_bionic != linux_glibc.
* linux_glibc_86_64
* linux_bionic_arm64
Some dependencies (like soong_zip, etc.) are not enabled for the new
host target (because they don't have to) and therefore the build fails.
Since the purpose of host APEX is to package some host tools for
testing, etc., we actually don't need the second target for APEX.
Fixing the problem by not skipping "cross host" targets which can't be
natively executed on the host machine.
Bug: 169454252
Test: HOST_CROSS_OS=linux_bionic HOST_CROSS_ARCH=arm64 m
Test: DIST_DIR=$(pwd)/out/dist \
./art/tools/dist_linux_bionic.sh com.android.support.apexer
Change-Id: Idaded56e4b4fc0195c8bbc760e4b4728016bd5b4
Flattened APEXes are installed in /<partition>/apex/<name>, so its
file_contexts file should be amended accordingly.
For example, file_contexts of an apex "com.android.foo" with should be
modified by prepending actual install path to the path regex.
/bin/foo <sepolicy label>
=> /system/apex/com\.android\.foo/bin/foo <sepolicy label>
File_contexts file is emitted to "make" via LOCAL_FILE_CONTEXTS variable
for apex_manifest.pb module.
Make collects these files and use them to build file_contexts.bin.
Also removed global state which is used to keep the list of all
flattened apexes.
Bug: 166518492
Test: m file_contexts.bin
check intermediate file_contexts.modules.tmp file
in $OUT/obj/ETC/file_contexts.bin_intermediates/
Change-Id: I780370b6e576964a5c09f9477f3361f341dd576b
Keeping the int constant around for now as FutureApiLevelInt because
it's still useful in places that haven't adopted ApiLevel yet for
testing if their non-ApiLevel API level is current or not.
Test: treehugger
Bug: http://b/154667674
Change-Id: I47a7012703f41fdeb56f91edf9c83afa93042deb
AddFarVariationDependencies was broken, which allowed apex to add
dependencies on tests by requesting an empty test_per_src variation
even though some test dependencies did not have a test_per_src
variation. Add an alias from the pre-test_per_src variation
variant to the empty test_per_src variation, and drop the
test_per_src variation when requesting the depencency.
Test: TestApexWithTests
Change-Id: I2ed7bdd761027956141e25d4d81f9074afe628b6
This change fixes a bug that notice files for some module type (e.g.
prebuilt_etc) were not included in APEX. This happened because we relied
on WalkPayloadDeps which actually doesn't traverse module types that
don't implement ApexModule interface. prebuilt_etc is one such module
type. Fixing the problem by also iterating the filesInfo array which
has info about all modules that are included in the APEX.
Bug: 166575301
Test: m com.android.tzdata and inspect the built artifact.
NOTICE.html.gz is there.
Change-Id: Iceb055b60184aef2a3e65e44680304853eb79a53
bpf program is put to an APEX via 'bpfs' property. It is
placed under etc/bpf directory in it.
Fix: 167530625
Test: m
Change-Id: Ia36b486f0cffb619ecc4f7a318cde881abc5baf4
prebuilt_firmware module is one of many prebuilt_etc-like modules. When
it is soc-specific, it is installed in /vendor/firmware. Similarly, when
prebuilt_firmware is embeded in a vendor apex, installing it in
<apex>/firmware instead of <apex>/etc.
Bug: 162701747
Test: lunch sunfish-userdebug
m && device boots && vibrator works
Change-Id: I00d28cde42259aaf8221e3897df77efc42b0c1ca
The affected modules weren't actually built in the affected build,
delay the platform availability check in Make instead to allow
the build to proceed.
Bug: 154888298
Test: m checkbuild
Change-Id: I21778e38e04635545b352dee2043a33ec9da246b
Add the new property payload_fs_type in apex module type. Either 'f2fs'
or 'ext4'. Default 'ext4'.
Test: m
Bug: 158453869
Change-Id: I36f373251bd597e11acb62af75437a87e2a531ec
Merged-In: I36f373251bd597e11acb62af75437a87e2a531ec
AddFarVariationDependencies was broken, which allowed sdk modules to
request dependencies using image and version variations, even for
host modules that do not have image or version variations. Make
the image and version variations conditional on device sdk modules.
Test: go test ./sdk
Change-Id: I59b7a32a3782254fd5feb828a5258ee13d4db812
APEX variants that share the same SDK version and updatability
almost always use identical command line arguments to build but
with different intermediates directories. This causes unnecessary
build time and disk space for duplicated work.
Deduplicate APEX variants that would build identically. Create
aliases from the per-APEX variations to the new shared variations
so that the APEX modules can continue to depend on them via the
APEX name as the variation.
This has one significant change in behavior. Before this change,
if an APEX had two libraries in its direct dependencies and one
of those libraries depended on the other, and the second library
had stubs, then the first library would depend on the implementation
of the second library and not the stubs. After this change, if
the first library is also present in a second APEX but the second
library is not, then the common variant shared between the two
APEXes would use the stubs, not the implementation.
In a correctly configured set of build rules this change will
be irrelevant, because if the compilation worked for the second
APEX using stubs then it will work for the common variant using
stubs. However, if an incorrect change to the build rules is
made this could lead to confusing errors, as a previously-working
common variant could suddenly stop building when a module is added
to a new APEX without its dependencies that require implementation
APIs to compile.
This change reduces the number of modules in an AOSP arm64-userdebug
build by 3% (52242 to 50586), reduces the number of variants of the
libcutils module from 74 to 53, and reduces the number of variants
of the massive libart[d] modules from 44 to 32.
This relands I0529837476a253c32b3dfb98dcccf107427c742c with a fix
to always mark permissions XML files of java_sdk_library modules as
unique per apex since they contain the APEX filename, and a fix
to UpdateUniqueApexVariationsForDeps to check ApexInfo.InApexes
instead of DepIsInSameApex to check if two modules are in the same
apex to account for a module that depends on another in a way that
doesn't normally include the dependency in the APEX (e.g. a libs
property), but the dependency is directly included in the APEX.
Bug: 164216768
Test: go test ./build/soong/apex/...
Change-Id: I2ae170601f764e5b88d0be2e0e6adc84e3a4d9cc
APEX variants that share the same SDK version and updatability
almost always use identical command line arguments to build but
with different intermediates directories. This causes unnecessary
build time and disk space for duplicated work.
Deduplicate APEX variants that would build identically. Create
aliases from the per-APEX variations to the new shared variations
so that the APEX modules can continue to depend on them via the
APEX name as the variation.
This has one significant change in behavior. Before this change,
if an APEX had two libraries in its direct dependencies and one
of those libraries depended on the other, and the second library
had stubs, then the first library would depend on the implementation
of the second library and not the stubs. After this change, if
the first library is also present in a second APEX but the second
library is not, then the common variant shared between the two
APEXes would use the stubs, not the implementation.
In a correctly configured set of build rules this change will
be irrelevant, because if the compilation worked for the second
APEX using stubs then it will work for the common variant using
stubs. However, if an incorrect change to the build rules is
made this could lead to confusing errors, as a previously-working
common variant could suddenly stop building when a module is added
to a new APEX without its dependencies that require implementation
APIs to compile.
This change reduces the number of modules in an AOSP arm64-userdebug
build by 3% (52242 to 50586), reduces the number of variants of the
libcutils module from 74 to 53, and reduces the number of variants
of the massive libart[d] modules from 44 to 32.
Bug: 164216768
Test: go test ./build/soong/apex/...
Change-Id: I0529837476a253c32b3dfb98dcccf107427c742c
In preparation for reusing the same variation for multiple apexes,
rename ApexName to ApexVariationName.
Bug: 164216768
Test: all soong tests
Change-Id: I88f2c5b192ffa27acd38e01952d0cefd413222a0
Remove ipmemorystore-aidl-interfaces-java in allowlist for
com.android.tethering
Bug: 146436251
Test: m nothing
Change-Id: I8b7f97d59c2fd191f467ad50a8fd66d9a5d37704
When a vendor APEX sets use_vndk_as_stable: true, then the apex doesn't
bundle VNDK libraries but instead it uses them from the "vndk" namespace
at runtime.
":vndk" is a pseudo-name for VNDK libraries, which is processed by
linkerconfig so that the generated linker configuration allows the
namespace for the apex to access VNDK libs.
Bug: 159576928
Test: test with a test-only vendor apex
it runs okay with VNDK libs. (lsof shows that it opens libs from
VNDK APEX)
Change-Id: I924ce0fea696162575f59589b8f858deddf436b3
Vendor APEXes(in general, APEXes for non-system partitions) which is
supposed to be tied to a specific VNDK version can set this new property
so that it excludes VNDK libs and use them from VNDK APEX (provided by
system parition).
For these APEXes to use VNDK libs from VNDK APEX, linkerconfig should
link "vndk" linker namespace to the namespaces of these APEXes.
Bug: 159195575
Test: m (soong test added)
Change-Id: If90650973239ef7aab0ff084488bda57d9b0364e
This change fixes this error:
```
TARGET module com.android.adbd.flattened requires non-existent TARGET
module: prebuilt_libclang_rt.ubsan_standalone-arm-android
```
apexFile.moduleName is used as Make dependency name, so it should use
m.BaseModuleName() instead of m.Name(), because soong may prepend
"prebuilt_" to or mutate the output of m.Name() in other ways.
android/androidmk.go emits Android.mk modules with
`LOCAL_MODULE := module.BaseModuleName() + <SubName>`, so replace
apexFile.moduleName with BaseModuleName() + <SubName> as much as
possible.
Bug: 7456955
Test: Add unit test in apex/apex_test.go
Test: lunch blueline_hwasan && SANITIZE_TARGET='hwaddress fuzzer' m nothing
Test: Verify out/soong/Android-blueline_hwasan.mk
Change-Id: If8537fc1bedbe6c3405de3662a5df210a073c43f
Previously, an APEX variant was created for a module that was a member
of an SDK just in case it had to be replaced with an APEX requested
snapshotted version of that member. That was necessary because that was
the only way to have APEX specific replacements.
Since then a new method called ReplaceDependenciesIf() has been added
which provides fine grained control over which dependencies are
replaced. This change uses that new method to handle the replacements
which allows the APEX variants to be removed.
Bug: 161928524
Test: m nothing
Change-Id: If3869dd6753dc182b099af566b20fbc9c9c6eff7
apexDepsMutator marks all dependencies of apex modules. Previously, it
was converted from WalkDeps() to Top-down mutator to avoid the pitfall
of WalkDeps() bug. (It did't handle multiple visits via different
paths.)
Because WalkDeps() problem solved in aosp/1277516, apexDepsMutator can
be reverted to use WalkDeps().
Even though there's no observable difference between them, I revert this
for the up-coming change, which requires different pruning strategies
per apexes.
Bug: 159195575
Test: m
Change-Id: Ib09cbc7a3dfd143dd37b660b1aea6c71392ce2e3
