The DepIsInSameApex() and RequiredSdks() methods were defined in a few
places to avoid having to depend on the whole ApexModule/SdkAware
interfaces directly. However, that has a couple of issues:
1) It duplicates functionality making it difficult to change, changes
to the definitions outside the main interfaces do not cause compile
time failures, instead they result in a runtime change in behavior
which can be difficult to debug.
2) IDE navigation (specifically in Intellij) does not detect that the
duplicate definitions can resolve to the definitions in the main
interface.
This change extracts the methods into their own interfaces and reuses
those interfaces instead of duplicating the methods to fix both of
these issues.
Bug: 152878661
Test: m nothing
Change-Id: I0cfdf342a14eb0bfb82b1bd17e0633d81c7facfb
Apex can use codenames like "Q", "R" for its min_sdk_version property.
Also, cc_library can use codenames for its stubs.versions.
Bug: 152655956
Test: vendor/google/build/build_mainline_modules.sh
Change-Id: I077ad7b2ac5d90b4c8708921e43846206f05ba70
Delegate the responsibility for determining whether the
walkPayloadDeps() should visit a child dependency to its do function.
This is needed to allow the visitor in checkApexAvailability() to avoid
checking the apex_available setting after crossing the APEX boundary.
Bug: 152878661
Test: m droid
Change-Id: If46a2f74b6eca670befc3aeae430650e85542346
The apex available check can traverse quite a long path (5+ steps) to
get from the apex to a module that is missing the apex from its
apex_available property. Understanding where that dependency came from
can often require examining the dependency path which can be difficult.
This change adds the path to the error to simplify that process.
Test: m nothing
Bug: 152762638
Change-Id: Ic4eb169dc2026cd8339d49e23b25d6d1c3879750
Coverage build adds extra dependencies for the runtime libs. Marking
them and their transitive dependenciese with apex_available doesn't make
any value; it instead adds noise.
Bug: 150999716
Test: m
Change-Id: I08e51cec1de6ded624ef9d4c41c1e5b2e611ff38
The marked library(ies) were available to the adbd APEX via the
hand-written whitelist in build/soong/apex/apex.go. Trying to remove the
whitelist by adding apex_available property to the Android.bp of the
libraries.
Bug: 150999716
Test: m
Change-Id: I6afbe905dda11056559521b937803dcd85b32fe5
The marked library(ies) were available to the APEXes via the hand-written
whitelist in build/soong/apex/apex.go. Trying to remove the whitelist
by adding apex_available property to the Android.bp of the libraries.
Bug: 150999716
Test: m
Change-Id: I91d1b1076733a949ca2a959ba640ce34d0233492
The NDK prebuilts are implicitly used when building with sdk_version set.
Make the module types be available to any apex so that we don't need to
manually add apex_available property to the module definitions manually.
Bug: 150999716
Test: m
Change-Id: I0870afa4c74b4a06ab1273dff84615778561ecc9
Revert submission 1254009-vdex-symlinks
Reason for revert: This causing some devices to fail to boot. See b/151836042 for details.
Reverted Changes:
Iced89071b:Expect vdex files in a target-independent director...
Ifbceb8457:Share vdex files in the ART apex between architect...
Change-Id: I5bd88f8e61d442eed921d840c90777a2750ddb16
Test: aosp_walleye-userdebug boots.
Test: Check symlinks to *.vdex files in the ART apex:
$ adb shell 'find /apex/com.android.art -name '*.vdex' | xargs ls -l'
lrw-r--r-- 1 system system 23 1970-01-01 01:00 /apex/com.android.art/javalib/arm/boot-apache-xml.vdex -> ../boot-apache-xml.vdex
lrw-r--r-- 1 system system 25 1970-01-01 01:00 /apex/com.android.art/javalib/arm/boot-bouncycastle.vdex -> ../boot-bouncycastle.vdex
lrw-r--r-- 1 system system 23 1970-01-01 01:00 /apex/com.android.art/javalib/arm/boot-core-icu4j.vdex -> ../boot-core-icu4j.vdex
lrw-r--r-- 1 system system 24 1970-01-01 01:00 /apex/com.android.art/javalib/arm/boot-core-libart.vdex -> ../boot-core-libart.vdex
lrw-r--r-- 1 system system 19 1970-01-01 01:00 /apex/com.android.art/javalib/arm/boot-okhttp.vdex -> ../boot-okhttp.vdex
lrw-r--r-- 1 system system 12 1970-01-01 01:00 /apex/com.android.art/javalib/arm/boot.vdex -> ../boot.vdex
lrw-r--r-- 1 system system 23 1970-01-01 01:00 /apex/com.android.art/javalib/arm64/boot-apache-xml.vdex -> ../boot-apache-xml.vdex
lrw-r--r-- 1 system system 25 1970-01-01 01:00 /apex/com.android.art/javalib/arm64/boot-bouncycastle.vdex -> ../boot-bouncycastle.vdex
lrw-r--r-- 1 system system 23 1970-01-01 01:00 /apex/com.android.art/javalib/arm64/boot-core-icu4j.vdex -> ../boot-core-icu4j.vdex
lrw-r--r-- 1 system system 24 1970-01-01 01:00 /apex/com.android.art/javalib/arm64/boot-core-libart.vdex -> ../boot-core-libart.vdex
lrw-r--r-- 1 system system 19 1970-01-01 01:00 /apex/com.android.art/javalib/arm64/boot-okhttp.vdex -> ../boot-okhttp.vdex
lrw-r--r-- 1 system system 12 1970-01-01 01:00 /apex/com.android.art/javalib/arm64/boot.vdex -> ../boot.vdex
-rw-r--r-- 1 system system 1229 1970-01-01 01:00 /apex/com.android.art/javalib/boot-apache-xml.vdex
-rw-r--r-- 1 system system 2043 1970-01-01 01:00 /apex/com.android.art/javalib/boot-bouncycastle.vdex
-rw-r--r-- 1 system system 2883 1970-01-01 01:00 /apex/com.android.art/javalib/boot-core-icu4j.vdex
-rw-r--r-- 1 system system 865 1970-01-01 01:00 /apex/com.android.art/javalib/boot-core-libart.vdex
-rw-r--r-- 1 system system 395 1970-01-01 01:00 /apex/com.android.art/javalib/boot-okhttp.vdex
-rw-r--r-- 1 system system 7125 1970-01-01 01:00 /apex/com.android.art/javalib/boot.vdex
Bug: 150934453
Change-Id: Ifbceb845749f4c218693f4118e8b35b59ff26de1
libselinux no longer is included in any APEX. Only the platform variant
of it (/system/lib/libselinux.so) exists and APEXes link to it.
Removing the lib name from the whitelist to make it clear that the
library is not available to any APEX.
Bug: 151053366
Bug: 150999716
Test: m
Exempt-From-Owner-Approval: cherry-pick from aosp
Merged-In: Id4fb933141ad32ff5217a58f1c7d689cc657e9ea
(cherry picked from commit 1731f6ae3f)
Change-Id: Id4fb933141ad32ff5217a58f1c7d689cc657e9ea
min_sdk_version = 29 implies that the module should support Android10.
Bug: 150431944
Test: m
Merged-In: Iad90a239898f59456900ae7816b90379b1b43406
Change-Id: Iad90a239898f59456900ae7816b90379b1b43406
(cherry picked from commit 5417f775e5)
Exempt-From-Owner-Approval: cp from aosp
Native modules within APEX should be linked with proper stub version
according to its min_sdk_version.
For example, when min_sdk_version is set to "29", libfoo in the apex
would be linked to libbar of version 29 from platform, even if it has
a newer version like 30.
Bug: 145796956
Test: m nothing (soong tests)
Merged-In: I4a0b2002587bc24b7deeb5d59b6eeba5e1db5b1f
Change-Id: I4a0b2002587bc24b7deeb5d59b6eeba5e1db5b1f
(cherry picked from commit 03b5185b88)
Exempt-From-Owner-Approval: got ORV already.
libselinux no longer is included in any APEX. Only the platform variant
of it (/system/lib/libselinux.so) exists and APEXes link to it.
Removing the lib name from the whitelist to make it clear that the
library is not available to any APEX.
Bug: 151053366
Bug: 150999716
Test: m
Change-Id: Id4fb933141ad32ff5217a58f1c7d689cc657e9ea
Makes sure that the module snapshots do not rely on the white list
of apex available settings so that when those lists are removed it is
not necessary to update any snapshots.
Bug: 142935992
Test: m nothing
Change-Id: Iedcff7dfc2646a4da77258d16e06657dd2f411f9
Test apexes no longer check to see whether their contents are available
so the special handling is no longer necessary.
Bug: 142935992
Test: m nothing
Change-Id: Iecae7dcbb87908d19c672f74d3c1ed8810d4485b
The whitelistedApexAvailable used to map references to
com.android.art.debug/release to com.android.art before looking it up
in the white list. This change removed that mapping and simply added
both to the white list.
Bug: 142935992
Test: m nothing
Change-Id: Ibad76fb73988688eb303e056197986ee9a6119ae
Merged-In: Ibad76fb73988688eb303e056197986ee9a6119ae
NativeBridgeRelativePath should be appended after /bin or /lib and
before subdir.
Bug: N/A
Test: m (add a soong test)
Change-Id: Id0c44c66b4900caa291e816cf3361fdec8ff9421
Checking apex_available was missing some corner cases.
For example, the deps of share deps of cc_library modules are missed
while those from cc_library_shared are correctly tracked.
This was due to..
* calling DepIsInSameApex in WalkDeps: both work fine separately, but
when they are used together, it fails to work. It's due to how WalkDeps
works. (We might fix this bug too risky since it is used very widely)
* incorrect receiver for DepIsInSameApex in apex_deps mutator: receiver
is supposed to be parent, but child was used before. Interestingly lots
of deps are within the same group of module types(cc to cc, java to
java), it has worked. (note that receiver's DepIsInSameApex
implementation can be different).
This change fixes them by..
* walkPayloadDeps is now relying on ApexVariation, which is calculated
correctly by TopDown apex_deps mutator.
* use correct receiver for DepIsInSameApex in apex_deps mutator, which
requires for java.SdkLibrary to override the method and for
java.Library/Import to use passed dep instead of receiver to check its
membership of sdk.
Exempt-From-Owner-Approval: cherry-pick from aosp/master
Bug: 151071238
Test: build/boot
Merged-In: I0569ef4bb8e79635e4d97a89f421a8d8b7d26456
(cherry picked from commit 5e9013be22)
Change-Id: I0569ef4bb8e79635e4d97a89f421a8d8b7d26456
Checking apex_available was missing some corner cases.
For example, the deps of share deps of cc_library modules are missed
while those from cc_library_shared are correctly tracked.
This was due to..
* calling DepIsInSameApex in WalkDeps: both work fine separately, but
when they are used together, it fails to work. It's due to how WalkDeps
works. (We might fix this bug too risky since it is used very widely)
* incorrect receiver for DepIsInSameApex in apex_deps mutator: receiver
is supposed to be parent, but child was used before. Interestingly lots
of deps are within the same group of module types(cc to cc, java to
java), it has worked. (note that receiver's DepIsInSameApex
implementation can be different).
This change fixes them by..
* walkPayloadDeps is now relying on ApexVariation, which is calculated
correctly by TopDown apex_deps mutator.
* use correct receiver for DepIsInSameApex in apex_deps mutator, which
requires for java.SdkLibrary to override the method and for
java.Library/Import to use passed dep instead of receiver to check its
membership of sdk.
Bug: 151071238
Test: build/boot
Change-Id: I0569ef4bb8e79635e4d97a89f421a8d8b7d26456
When the check for apex_available has failed, the build system now shows
the module that brought the unavailable module into the APEX.
Bug: 151051671
Test: m
Merged-In: Id1a3fda67fe56fdc2dc90ec800d10689415de4d6
(cherry picked from commit 7bd9444b0f)
Change-Id: Id1a3fda67fe56fdc2dc90ec800d10689415de4d6
When the check for apex_available has failed, the build system now shows
the module that brought the unavailable module into the APEX.
Bug: 151051671
Test: m
Change-Id: Id1a3fda67fe56fdc2dc90ec800d10689415de4d6
Native modules within APEX should be linked with proper stub version
according to its min_sdk_version.
For example, when min_sdk_version is set to "29", libfoo in the apex
would be linked to libbar of version 29 from platform, even if it has
a newer version like 30.
Bug: 145796956
Test: m nothing (soong tests)
Change-Id: I4a0b2002587bc24b7deeb5d59b6eeba5e1db5b1f
Add a min_sdk_version property apexes. Currently a noop, but will
be used to enforce that dependencies are compatible with the
specified version.
Test: m checkbuild
Bug: 149591522
Merged-In: I923773c90fe15becbffae3986791aa9edde8f8f6
Change-Id: I923773c90fe15becbffae3986791aa9edde8f8f6
(cherry picked from commit 50317874ff)
If an APEX contains APKs and the manifest package name of the APKs are
overridden (either via override_android_app
orPRODUCT_MANIFEST_PACKAGE_NAME_OVERRIDES), that the path to the APK
(relative in the APEX) and the overridden manifest package name is
recorded in the bundle config file.
Exempt-From-Owner-Approval: cherry-pick from master
Bug: 148002117
Test: m
Merged-In: Ibb90bcefb77fa6b2dad77cb2facc6079de9ab154
(cherry picked from commit cfaa1643e8)
Change-Id: Ibb90bcefb77fa6b2dad77cb2facc6079de9ab154
If an APEX contains APKs and the manifest package name of the APKs are
overridden (either via override_android_app
orPRODUCT_MANIFEST_PACKAGE_NAME_OVERRIDES), that the path to the APK
(relative in the APEX) and the overridden manifest package name is
recorded in the bundle config file.
Bug: 148002117
Test: m
Change-Id: Ibb90bcefb77fa6b2dad77cb2facc6079de9ab154
Which is the list of JNI libraries that are embeded inside the apex.
jni_libs is handled just like native_shared_libs except that it is
stored in apex_manifest.
When linkerconfig finds an apex with JNI libs, it exposes the namespace
for the apex as visible so that libnativeloader can link the namespace
to the corresponding classloader-namespace.
Bug: 149363889
Test: m nothing(runs soong test)
Change-Id: I52ebe38b44545e6e8853e34a3404a235c858112a
Embed common properties (apexNativeDependencies) so that the logic for
adding dependencies for native modules can be reused.
Bug: N/A
Test: m
Change-Id: I789b526c09eea14213ab1544590ed2238ed8c625
Symlinking doesn't make sense for host APEXes.
Bug: 150255435
Test: m com.android.art.host and inspect the built APEX; there is
no symlink.
Merged-In: I28492dfaaef471117a430be05255fbef76e557b0
(cherry picked from commit 9b96418dfe)
Change-Id: I28492dfaaef471117a430be05255fbef76e557b0
Symlinking doesn't make sense for host APEXes.
Bug: 150255435
Test: m com.android.art.host and inspect the built APEX; there is
no symlink.
Change-Id: I28492dfaaef471117a430be05255fbef76e557b0
Because APK-in-APEX embeds its jni_libs in it. We don't have to follow
deps of jni_libs.
Bug: 146992436
Test: m com.android.tethering
deapexer extract com.android.tethering.apex apex
ls apex # there should be no /lib dir
Merged-In: Ifa1a6430a420ae7376b155cd59b8ece462cced7e
Change-Id: Ifa1a6430a420ae7376b155cd59b8ece462cced7e
(cherry picked from commit b7bebe2616)
Because APK-in-APEX embeds its jni_libs in it. We don't have to follow
deps of jni_libs.
Bug: 146992436
Test: m com.android.tethering
deapexer extract com.android.tethering.apex apex
ls apex # there should be no /lib dir
Change-Id: Ifa1a6430a420ae7376b155cd59b8ece462cced7e
Add a min_sdk_version property apexes. Currently a noop, but will
be used to enforce that dependencies are compatible with the
specified version.
Test: m checkbuild
Bug: 149591522
Change-Id: I923773c90fe15becbffae3986791aa9edde8f8f6
The optimization is confusing syshealth because the modules will use
more memory when they are switched to the prebuilt (unbundled) or
updated via Play.
Let's have a per-module switch to control the behavior and turn it on
only for non-updatable modules like ART.
Bug: 149805758
Test: m
Change-Id: Ieb842c47f31f3b06e403b1e1f9e463c3e5524107
This change fixes a regression that was introduced with
I597bccbb177b6b6320c3a3edeff467243230d384. With the change, the content
of the permissions XML file for a java_sdk_library was determined before
the java_sdk_library is mutated for an APEX. As a result, the file path
to the implementation jar library was always set to
/system/framework/*.jar regardless of whether the java_sdk_library is
part of an APEX or not.
This change fixes the problem, by creating the permissions XML file via
a new module type. The content of the xml file is determined after the
xml file is mutated for APEXes.
Bug: 149600642
Test: m
Change-Id: Id21f8d8285df49b0b3be1daf0f101f2bc978eeb0
Bug: 149075752
Test: tapas com.android.conscrypt com.android.tethering arm64
Test: m out/target/product/generic_arm64/{,symbols/}apex/com.android.{tethering,conscrypt}/lib64/libc++.so
Test: Verified that unwinder was dynamically linked to tethering's
Test: libc++ and statically linked to conscrypt's.
Test: lunch flame-userdebug && m
Test: Verified that unwinder was dynamically linked to /system/lib64/libc++.so
Change-Id: I98eed7cb4316962b19b5c12e150c224c25d0e91d
apex { name: ["myapex"], native_shared_libs: ["libX", "libY"] }
cc_library { name: "libX", shared_libs: ["libY"] }
cc_library { name: "libY", shared_libs: ["libZ"], stubs: {...} }
apexDepsMutator was a bottom up mutator and it uses WalkDeps to traverse
the dependency tree rooted at myapex in a depth-first order. While
traversing the tree, if calls BuildForApex for a module that will be
part of the APEX.
libY is visited twice. Once via libX and once via myapex. If the visit
from libX was before the visit from myapex (since this is a depth-first
traversing), BuildForApex is not called for libY and its dependency
libZ, because libY provides a stub. And then when libY is again visited
via myapex, BuildForApex is correctly called for the module, but not for
its dependencies libZ because the paths from libY to libZ was already
visited.
As a result, the apex variant of libY has a dependency to the non-apex
variant of libZ.
Fixing the problem by changing the mutator a top-down one.
Bug: 148645937
Test: m
Change-Id: Ib2cb28852087c63a568b3fd036504e9261cf0782
The apex dependency map wasn't updated for test or non-installable
APEXes to work around the problem that a module being in such APEX
prevented the module from being installed in the system partition.
Since that problem is not happening any more, removing the unnecessary
work-around.
Bug: 123892969
Test: m
Change-Id: I43e07a9611a3e08ff39b9a64454b1c67949d35bc
This library is empty, and its functionality has moved
into libbinder/libhwbinder.
Bug: 148692216
Test: N/A
Change-Id: I5874efda9ab43fc00cf90395a1aabde45cf49579
The APEX dependency is more correctly tracked. Previously, the
dependency was tracked while we gather modules that will be installed to
an APEX. This actually was incorrect because we skipped many dependency
types that we don't need to follow to gather the modules list, such as
the headers dependency.
Now, the dependency is tracked directly when a module is mutated for an
APEX. In other words, if a module is mutated for an apex X, then the
module will appear in the X-deps-into.txt file.
This change also changes the format of the txt file. It now clearly
shows why a module is included in the APEX by showing the list of
modules that depend on the module.
Bug: 146323213
Test: m
Change-Id: I0a70cf9cce56e36565f9d55683fdaace8748a081
The function visits dependencies of an APEX that contribute to the
payload. checkApexAvailability is rewritten using the generic function.
There is no change in behavior.
Bug: N/A
Test: m
Change-Id: I1a8b4eb0a60a432f667a61b4f6f457c3b8f1cd3d
Previously, a java_sdk_library called "SDKLIB" would create a
prebuilt_etc module called "SDKLIB.xml" which installs the generated
XML permission file to /etc/permissions/SDKLIB.xml. That module
depended on the java_sdk_library "SDKLIB" to generate the XML file
as one of its outputs by specifying srcs: [":SDKLIB{.xml}"].
If the java_sdk_library is replaced by a prebuilt then the SDKLIB.xml
module expects the prebuilt to provide the XML permissions file which
it doesn't because that is an implementation detail and so the build
breaks.
A couple of alternative approaches were looked at to fix this. One was
to have the logic that replaced the source module with the prebuilt to
inform the source module that it was being replaced so it could disable
its created module. That lead to a dependency cycle where
SDKLIB -> SDKLIB.xml -> SDKLIB{.xml}
Another solution was to mark dependency tags in such a way that the
prebuilt could automatically identify and disable the SDKLIB.xml
module. Similar to how the visibility code will ignore dependencies
that are tagged with ExcludeFromVisibilityEnforcementTag. That became
very convoluted.
Instead the java_sdk_library was changed so that it was not responsible
for creating the XML permissions file. Instead it created a genrule
called "gen-SDKLIB.xml" to create it and then "SDKLIB.xml" depended on
that. The java_sdk_library also depended on the genrule to make the XML
permissions file available for APEX and testing.
Some refactoring of the APEX code and tests was necessary because they
had knowledge of the internal implementation of java_sdk_library. The
refactoring insulates them a little better from those details.
Bug: 148080325
Test: m droid && TARGET_BUILD_APPS=Camera2 m
Change-Id: I597bccbb177b6b6320c3a3edeff467243230d384
Currently, com.android.runtime provides libc/libm/libdl to system. But
they are supposed to be used from symlinks under /system/lib not
directly from runtime apex.
This helps linkerconfig to generate ld.config.txt automatically for
apexes.
Bug: 144664390
Test: m com.android.runtime
deapexer info com.android.runtime.apex
Change-Id: I1620e88e489fba88a06cc3bd6eb5b86a9b581e4f
This change fixes a bug that apex_available is not enforced for static
dependencies. For example, a module with 'apex_available:
["//apex_available:platform"]' was able to be statically linked to any
APEX. This was happening because the check was done on the modules that
are actually installed to an APEX. Static dependencies of the modules
were not counted as they are not installed to the APEX as files.
Fixing this bug by doing the check by traversing the tree in the method
checkApexAvailability.
This change includes a few number of related changes:
1) DepIsInSameApex implementation for cc.Module was changed as well.
Previuosly, it returned false only when the dependency is actually a
stub variant of a lib. Now, it returns false when the dependency has one
or more stub variants. To understand why, we need to recall that when
there is a dependency to a lib having stubs, we actually create two
dependencies: to the non-stub variant and to the stub variant during the
DepsMutator phase. And later in the build action generation phase, we
choose one of them depending on the context. Also recall that an APEX
variant is created only when DepIsInSameApex returns true. Given these,
with the previous implementatin of DepIsInSameApex, we did create apex
variants of the non-stub variant of the dependency, while not creating
the apex variant for the stub variant. This is not right; we needlessly
created the apex variant. The extra apex variant has caused no harm so
far, but since the apex_available check became more correct, it actually
breaks the build. To fix the issue, we stop creating the APEX variant
both for non-stub and stub variants.
2) platform variant is created regardless of the apex_available value.
This is required for the case when a library X that provides stub is in
an APEX A and is configured to be available only for A. In that case,
libs in other APEX can't use the stub library since the stub library is
mutated only for apex A. By creating the platform variant for the stub
library, it can be used from outside as the default dependency variation
is set to the platform variant when creating the APEX variations.
3) The ApexAvailableWhitelist is added with the dependencies that were
revealed with this change.
Exempt-From-Owner-Approval: cherry-pick from internal
Bug: 147671264
Test: m
Merged-In: Iaedc05494085ff4e8af227a6392bdd0c338b8e6e
(cherry picked from commit fa89944c79)
Change-Id: Iaedc05494085ff4e8af227a6392bdd0c338b8e6e
This reverts commit a952495643.
Reason for revert: PermissionController may need to be built outside of APEX
Fixes: 148957736
Change-Id: I80a2cd666ea5028741250dd2cebf2bd7d7b331ad
This change fixes a bug that apex_available is not enforced for static
dependencies. For example, a module with 'apex_available:
["//apex_available:platform"]' was able to be statically linked to any
APEX. This was happening because the check was done on the modules that
are actually installed to an APEX. Static dependencies of the modules
were not counted as they are not installed to the APEX as files.
Fixing this bug by doing the check by traversing the tree in the method
checkApexAvailability.
This change includes a few number of related changes:
1) DepIsInSameApex implementation for cc.Module was changed as well.
Previuosly, it returned false only when the dependency is actually a
stub variant of a lib. Now, it returns false when the dependency has one
or more stub variants. To understand why, we need to recall that when
there is a dependency to a lib having stubs, we actually create two
dependencies: to the non-stub variant and to the stub variant during the
DepsMutator phase. And later in the build action generation phase, we
choose one of them depending on the context. Also recall that an APEX
variant is created only when DepIsInSameApex returns true. Given these,
with the previous implementatin of DepIsInSameApex, we did create apex
variants of the non-stub variant of the dependency, while not creating
the apex variant for the stub variant. This is not right; we needlessly
created the apex variant. The extra apex variant has caused no harm so
far, but since the apex_available check became more correct, it actually
breaks the build. To fix the issue, we stop creating the APEX variant
both for non-stub and stub variants.
2) platform variant is created regardless of the apex_available value.
This is required for the case when a library X that provides stub is in
an APEX A and is configured to be available only for A. In that case,
libs in other APEX can't use the stub library since the stub library is
mutated only for apex A. By creating the platform variant for the stub
library, it can be used from outside as the default dependency variation
is set to the platform variant when creating the APEX variations.
3) The ApexAvailableWhitelist is added with the dependencies that were
revealed with this change.
Bug: 147671264
Test: m
Change-Id: Iaedc05494085ff4e8af227a6392bdd0c338b8e6e
apex module accepts PlatformCompatConfigIntf as prebuilt,
and places it in the etc folder of the apex.
Test: m
Test: flash device with dummy config in mediaprovider APEX -
the config is present
Change-Id: Ifc62cd262f6c6571c1bf6c2943879aa20877ecad
Leaving out LOCAL_CERTIFICATE for flattened apex APKs causes
the apkcerts.txt to have empty keys for those APKs, which
confuses the signing tools. Set LOCAL_CERTIFICATE for them.
Also refactor the Certificate support to avoid introducing
duplicated handling for presigned certificates.
Bug: 147765187
Test: m apkcerts-list
Change-Id: Ife07661761cd5a89c9f009b8ce041db4dff9ec54
This change fixes a bug that license info for non-flattened APEXes are
not captured in /system/etc/NOTICE.xml.gz file. For non-flatted APEXes,
we have been creating NOTICE.html.gz file by concatenating all the
license infos of the modules that contributes to the APEX and embedding
the file into the asset directory of the APEX. Then at runtime, the info
is shown through the "Google Play System Update Licenses" UI. However,
this was problematic because the UI only shows license info for the
Google-signed APEXes, leaving OEM-signed APEXes (a.k.a. optional
modules).
The problem is now fixed by associating a merged license file with each
APEX and exporting them to Make, so that the merged license files are
included in the partition level /system/etc/NOTICE.xml.gz file
regardless of whether the APEX is a Google-signed one or not.
This also fixes a bug that license info entries are created for the
runtime paths /apex/<apex_name>/<path_to_a_file>, which is not necessary
as they are already included in the license info of the containing APEX.
Bug: N/A
Test: Go to Settings->About Phone->Legal information and check
that a) /system/apex/*.apex files are shown and b) /apex/<apex_name>/*
files are not shown
Change-Id: I2c25c803b6a4c39b24bb3f724502699382fab50c
This reverts commit 5df3b11f78.
Reason for revert: re-land with a fix
Fix a broken soong test
Add implicit dependency (libprofile-clang-extra) to make a test pass.
Bug: n/a
Test: m
Change-Id: I0b179199bc032501354f8e24782837453781bd8c
VNDK APEX is supposed to contain "vendor" variants of VNDK libraries.
This is different from normal APEXes which have "apex" variants.
Bug: 146758869
Test: build / flash / boot
Change-Id: I5e035678c337334092616b58d2e0e404788a6639
Exempt-From-Owner-Approval: Got ORV, but rebased with resolving merge conflicts.
This reverts commit 31c65d4fe4.
Bug: 144533348
Test: checkout master-art-host and run
ALLOW_MISSING_DEPENDENCIES=true DIST_DIR=out/dist /art/tools/dist_linux_bionic.sh -j80 com.android.art.host
the result is successful
Change-Id: Ica11eec9b64867088b16720a41c6d83905976ec5
This means everything that goes into apexes need to be
explicitly labeled to be available for apex.
Whitelist the current offenders. This list should be
trimmed down.
Bug: 147364041
Test: m
Test: multiproduct_kati -only-soong
Exempt-From-Owner-Approval: cherry-pick from internal
Change-Id: I837299c6a15d46f8a5ba544b613776b1cc27d7b8
Merged-In: I837299c6a15d46f8a5ba544b613776b1cc27d7b8
(cherry picked from commit 93488cbb10d4882845abb732f8e53714f0982031)
Used in coverage builds.
Test: NATIVE_COVERAGE=true \
EMMA_INSTRUMENT_FRAMEWORK=true \
EMMA_INSTRUMENT=true \
m nothing
Change-Id: I6309d8149fce016eb9689326eeae221914e4c3bb
This means everything that goes into apexes need to be
explicitly labeled to be available for apex.
Whitelist the current offenders. This list should be
trimmed down.
Bug: 147364041
Test: m
Test: multiproduct_kati -only-soong
Exempt-From-Owner-Approval: Just added more whitelist entries
Change-Id: I837299c6a15d46f8a5ba544b613776b1cc27d7b8
This reverts commit 7cb4d378e7.
Test: m
Test: ALLOW_MISSING_DEPENDENCIES=true DIST_DIR=out/dist ./art/tools/dist_linux_bionic.sh -j80 com.android.art.host
(in the master-art-host branch)
Change-Id: I9beca73aafdf42f03bfa19cf1634b2641dac417b
This reverts commit 956305c61c.
Reason for revert: broke master-art-host branch
Exempt-From-Owner-Approval: reverting a bad change
Change-Id: Id7faed4ee85328c7c65847a3543ea9e67a3d50b3
For each APEX, <apexname>-installed-files.txt is dist'ed to show the
list of files and their sizes that are included in the APEX.
Bug: 147605944
Test: m dist and examine the txt files
Change-Id: I565479523e51280fc88d5fbf8ea3f48ac0ae9fee
This change fixes a bug that jacoco-report-classes-all.jar does not
include info for APK-in-APEX such as the MediaProvider apk in
com.android.mediaprovider APEX.
Firstly, LOCAL_SOONG_JACOCO_REPORT_CLASSES_JAR is correctly set also for
the APKs included in APEXes. Secondly, the Make modules for the embedded
APKs are now built with soong_app_prebuilt.mk to correctly import the
jacoco file into the Make world.
Bug: 147296855
Test: execute the following command in internal master.
$ choosecombo cf_x86_phone userdebug
$ NINJA_ARGS="-t path out/target/product/vsoc_x86/jacoco-report-classes-all.jar out/target/common/obj/ETC/MediaProvider.com.android.mediaprovider_intermediates/jacoco-report-classes.jar" EMMA_INSTRUMENT=true EMMA_INSTRUMENT_FRAMEWORK=true m
The result shows that there is a path as follows:
out/target/product/vsoc_x86/jacoco-report-classes-all.jar
out/target/product/vsoc_x86/apex/com.android.mediaprovider/priv-app/MediaProvider/MediaProvider.apk
out/target/product/vsoc_x86/obj/ETC/MediaProvider.com.android.mediaprovider_intermediates/package.apk
out/target/common/obj/ETC/MediaProvider.com.android.mediaprovider_intermediates/jacoco-report-classes.jar
Change-Id: I52d11534a34eb35219bfafca4453e75a1b701c0e
When prebuilt_apex overrides/replaces other apex, the same symlinks need
to be created as well.
Bug: 143192278
Test: 1. add prebuilt_apex with vndk apex
with overrides set as ["com.android.vndk.current"]
2. m <prebuilt apex>
3. check if vndk symlinks are created
Change-Id: I8ee9c981ea9c7202ccf5143b3f43e6848773cd63
When a module is included in an APEX and the module is not directly
installed to the system partition, the 'required' property of the module
is lost. The APEX containing the module now carries the property from
its dependencies.
Bug: 146549048
Test: m
Change-Id: Ie21cc7b8420c6126c855e66c2363644de4a7ef42
We need to have a way to see the list of modules that directly or
indirectly contribute to an APEX. People find it difficult to determine
whether a module is included in which APEXes because APEX tracks
indirect dependencies as well as direct dependencies. Therefore, just
looking at Android.bp for the APEX itself doesn't give the answer.
This change adds a new make target <apex_name>-deps-info, which
generates out/soong/<apex_name>-deps-info.txt file that shows the
internal and external dependencies of the said APEX.
Here, internal means the dependencies are actually part of the
APEX, while external means the dependencies are still external to the
APEX.
Bug: 146323213
Test: m (apex_test amended)
Change-Id: I33d1ccf5d1ca335d71cd6ced0f5f66b8c3886d13
Create an APEX variation of a module only when it is either directly
included in an APEX or the dependency to the module demands that the
module should be included. For example, a non-static dependency to a
java library no longer creates an APEX variant of the library.
Bug: 146907857
Test: m (apex_test amended)
Test: cat out/soong/build.ninja | grep "Module:" | wc -l
reduced from 47132 to 45881
Change-Id: Icd79fb4b60668d852b9f018343b934b5e42a8eda
Permission XML file for java_sdk_library is generated by
java_sdk_library itself now.
And, build rule is switched to android.WriteFile since "echo -e" is
not supported from build server. (-e is printed to output)
Bug: 145474221
Test: m com.android.cronet and check its permissions xml file
also, m org.chromium.net.cronet.xml (created dynamically)
Change-Id: Iffb119151c49bc4fe6c4386fa267cca193f37dbc
java_import doesn't support dex jar which is necessary to be packaged in
an APEX. Supporting this would require non-trivial work. By the way
java_import is not used and there is a workaround.(defining java_library
and static-linking with it)
We drop the support for `java_import` in APEX.
Bug: 139175488
Test: m (soong tests amended)
Change-Id: I924386571079090c701276d87f665ce7fbb6f074
When a java_sdk_library module is added, both impl jar and permission
xml files are packaged together.
For example, when a java_sdk_library "foo" is listed, following two
entries will be in an APEX package.
/javalibs/foo.jar
/etc/permissions/foo.xml
Bug: 145474221
Test: m com.android.cronet
deapexer list com.android.cronet.apex
Change-Id: If5883c02255e9309f20810b1532d3fbe73bf4e95
In case that an apex module depends on a module with stubs directly
*and* indirectly, the build system should follow the deps graph further.
Note that WalkDeps() visits deps in DFS and it won't visit again visited
modules.
Bug: n/a
Test: m (apex_test.go amended)
Change-Id: I1a6f135dbda6d1eb641575a3fbbc9bbee0622076
This change fixes a bug that LOCAL_PATH for modules included in an APEX
is set to the path of the APEX bundle, not to the path of the embedded
module. For example, LOCAL_PATH of libconscrypt included in
com.android.adbd was set to /system/core/adb instead of
/external/boringssl. This caused a problem that NOTICE file in
/external/boringssl is not tagged to libconscrypt, but the NOTICE file
for adbd is.
Fixing the problem by recording the module directories of the included
modules and emitting it in LOCAL_PATH.
Bug: 145347092
Test: Settings -> About Phone -> Legal Information -> Third-party
license. The license for /apex/com.android.adbd/lib64/libconscrypt.so is
OpenSSL.
Change-Id: I76f1830d5a10af63fa74dcc2a42730ffabb8c4ed
APEXes with "legacy_android10_support" will have apex_manifest.json for
compatibility as well as apex_manifest.pb.
Bug: 143951586
Test: m (soong tests)
Change-Id: I019252aee5a9423f4b180ba1026e6e99c9961437
GSI targets are supposed to have both 'flattened' and 'unflattened'
APEXes. By adding 'flattened' APEX as REQUIRED moduled for 'unflattened'
APEX, both will be installed togetther.
This is done by a new variable PRODUCT_INSTALL_EXTRA_FLATTENED_APEXES,
which is set only for GSI targets.
Bug: 137802149
Test: lunch aosp_arm64-userdebug && m
resulting apex images under /system/apex
and flattened apexes under /system/system_ext/apex
Change-Id: I336e2674e427b358542e0045b2a49dfa3d84095b
Previously, it was defaultable only for cc_library. Now, it is
defaultable for all module types.
Bug: 128708192
Test: m
Change-Id: If248f593da6f4b724bf889ccd7f2e077d48069b5
This change is to make it easier to add new fields to the struct.
transitiveDep field is added to distinguish apexFiles coming from
transitive dependencies of the APEX. We will later use the info to
reduce the size of bundled APEXes by replacing the transitive deps with
symlinks to the corresponding files in the system partition outside of
the APEX.
Bug: 144533348
Test: m
Change-Id: I283859f2f2f1b5cfb3025569f168ba8569b22bb9
This build rule is specific to platform APEXes.
For non-platform APEXes, MakeAsSystemExt() is not applied.
This fixes the cases of "soc_specific: true" apexes which fails to
build.
Bug: 139053989
Test: m nothing (soong tests)
Change-Id: I98d0257499647ab41cdaa62a3671d89addbdf833
Exempt-From-Owner-Approval: got +1 before rebasing
When an override_apex named Foo overrides an apex module named Bar, the
Make modules from Foo have Foo as their suffix. Previously the suffix
was Bar for both of the overriding and the overridden APEXes, causing
name conflicts in the Make side.
Bug: 144338929
Test: apex_test.go
Change-Id: I1396910ab294ba5f5e0585af6d37f1eab9460250
For platform APEXes, file_contexts should point a file under
/system/sepolicy.
Bug: 144732805
Test: m
Change-Id: Ib2d5db715bbebc80a6178d1c42e387b268cc4a0d
prebult_etc module type does not respect prefer32, i.e. its primary arch
is 64-bit on 64/32-bit device even when built with TARGET_PREFER_32_BIT
is true. However, the apex module type respects prefer32 and therefore
when TARGET_PREFER_32_BIT is true its primary arch becomes 32-bit. Then
the problem is that the apex tries to depend on 32-bit variant of the
prebuilt_etc modules which don't exist.
Fixing the problem by force using the first arch of the device when
adding dependencies to prebuilt_etc modules.
Bug: 144532908
Test: choosecombo 1 aosp_arm64 userdebug; TARGET_PREFER_32_BIT=true m;
Change-Id: I7642c57b05a837495587bbe4d3589d8549607862
Prepare for making the image mutator available to all modules and
moving it between the os and arch mutators by moving it into the
android package and using an interface implemented by the module
types to control it.
Bug: 142286466
Test: No unexpected changes to out/soong/build.ninja
Change-Id: I0dcc9c7b5ec80edffade340c367f6ae4da34151b
override_apex module type is used to override existing apex module with
certain properties overridden. Currently, only the 'apps' property is
overridable.
Bug: 144338929
Test: m
Change-Id: Ic050b062093cda29ce78126cc92dd6097647f7db
apex.go is too big. Separate the build rule and android.mk generation
logic into builder.go and androidmk.go, respectively. prebuilt_apex is
moved to prebuilt.go as well.
No refactoring has been made other than the splitting.
Test: m
Change-Id: I839ab0a1ba2b70ce82d98ac1fa8e3534808b5fd3
APEXes are ETC type, which should be overridden via
LOCAL_OVERRIDES_MODULES. This change fixes a bug that
LOCAL_OVERRIDES_PACKAGES was used for APEXes.
Bug: 140792287
Test: m
Test: add com.google.android.tzdata to PRODUCT_PACKAGES and build.
/system/apex/com.google.android.tzdata.apex exists, but
/system/apex/com.android.tzdata.apex doesn't.
Change-Id: Id65743b36e0b706d6ffd8cae0597cc0a42a83fb7
If some apex module can be installed for both flattened and unflattened,
flattened apex should be installed to system_ext partition.
So add MakeAsSystemExt func to change the partition from system to
system_ext.
Bug: 139053989
Test: m -j
Change-Id: I3e3430413a9045d96130af99e249567b1a26ed7e
If a JNI lib is depended on by an APK that is included in an APEX, the
lib is embedded inside the APK.
This change also fixes a bug that APKs are not mutated for APEXes.
Bug: 144135069
Test: m (apex_test.go amended)
Change-Id: I21ac24412b30c05afc03385655c6b196130dffe3
If a JNI lib is depended on by an APK that is included in an APEX, the
lib is embedded inside the APK.
This change also fixes a bug that APKs are not mutated for APEXes.
Bug: 144135069
Test: m (apex_test.go amended)
Change-Id: Icf490d2701a7ede8bcad7e671fc72be9c8d7c01e
This reverts commit bf0e47648a.
Reason for revert: coverage build with EMMA_INSTRUMENT_FRAMEWORK=true
is fixed by inspecting the environment variable and not generating
boot image in case it is set.
Dexpreopt artifacts for the libcore part of the boot class path are
now packaged in the ART apex. The system image still contains
dexpreopt artifacts for the full set of boot class path libraries
(both libcore and framework); the libcore part will be removed and
boot image extension will be used in a follow-up CL.
Since this is specific to the ART apex and makes no sense for other
apexes, the implementation adds a boolean flag "is ART apex" rather
than a new apex module property.
Build rules for the new set of dexpreopt artifacts are created using
a new variant of the global boot image config. Previously we had two
variants: "default" (for the system image) and "apex" (for the
JIT-zygote experiment). This patch adds a third "art" variant.
Test: m
Test: m art/build/apex/runtests.sh
Bug: 144091989
Change-Id: I113c0d39222d6d697cb62cd09d5010607872fc2b
For Q compatibility, .json files are also bundled.
Three different apex_manifest files are built from input.
- apex_manifest.json: Q-readable .json file
- apex_manifest_full.json: input + dependency
- apex_manifest.pb: apex: converted from apex_manifest_full.json
apexer will handle these files.
Bug: 143654022
Test: m
Change-Id: I9697094057f0c0543282b1b46b7535cf21431176
These files were highly coupled with vndk version, so having them inside
the corresponding VNDK APEX is reasonable.
These files are used by linkerconfig and libnativeloader. In the future,
they reference these files from and VNDK APEX.
Bug: 141450808
Test: m com.android.vndk.current
Change-Id: I055a979d2636ddd8844a0afff81f6ba441f7965e
When use_vendor is used, native modules are built with
__ANDROID_VNDK__ and __ANDROID_APEX__, which may cause
compatibility issues. (e.g. libbinder)
Even though libbinder restricts its availability via
'apex_available' property and relies on yet another macro
__ANDROID_APEX_<NAME>__, we restrict usage of "use_vendor:" from
other APEX modules to avoid similar problems.
Bug: 142684427
Test: m -j
Change-Id: Ibc781de2efcd20cb6688a183b08e908a8a6e2593
Create variant of image, zip or flattened according to
TARGET_FLATTEN_APEX and payload type.
If payload type is zip, only zip variant is created because flattened
apex is not supported. And if payload type is image, image and flattened
variants are created.
Bug: 139053989
Test: m -j
Change-Id: Ibde18490d23ec602c4cca97cf97db90a562e014e
This reverts commit d5df949385.
Bug: 143594594
Bug: 143593500
Reason for revert: Some builds are failing.
Change-Id: I69986b472bce39266095e526fcd7ef5f48ece85e
Exempt-From-Owner-Approval: Going back to green.
Since apex_vndk deals with its own dependencies, it should not follow
dependencies while packaging.
Bug: 139772411
Test: TARGET_VNDK_USE_CORE_VARIANT=true m com.android.vndk.current
see if there are unexpected libs in the apex
for example, android.hardware.audio.common@2.0.so should not be
included in the apex since it is not listed as "must use vendor
variant".
Change-Id: Ibada600b3099fb19630d1e327a2e09cd26b2deb0
While modifying the CTS shim apex packages, it was observed that the
output of the image content file is not necessarily sorted, which in
turn can cause failure when checking the file against a whitelist.
Bug: 138429615
Test: Succesfully built a modified version of CTS Shim v3 that was
previously failing because of the issue this CL fixes.
Change-Id: I901859ae08feb6012f34b851e125977e1c0100d9
This patch adds dexpreopt files for the libore part of the
bootclasspath to the ART apex.
Since this is specific to the ART apex and makes not sense for other
apexes, the patch does not add a new module property, but only a
boolean flag denoting that this is an ART apex.
Dexpreopt artifacts packaged into the ART apex differ from those that
are packaged in the system image: it inludes only the libcore part of
bootclasspath jars, but not the framework part. When the boot image
extension is implementd, dexpreopt artifacts for the libcore jars will
be removed from the system image (but for now they are both in the
apex and in the system image).
Build rules for the new set of dexpreopt artifacts are created using
a new variant of the global boot image config. Previously we had two
variants: "default" (for the system image) and "apex" (for the
JIT-zygote experiment). This patch adds the third "libcore" variant.
Test: m
Test: m com.android.art deapexer \
&& find $ANDROID_BUILD_TOP -type f -name 'com.android.art.*.apex \
| xargs deapexer | grep boot \
Expect to find dexpreopt/$ARCH/boot-art*.{art,oat,vdex} files.
Test: m art/build/apex/runtests.sh
Change-Id: I353ef90304bc5e18c3055ea379b3b223e5c38948
With VNDK APEX, the location of VNDK lib is changed.
But vndk libs of older VNDK versions still depend on old location.
For current VNDK version, those hard-coded references will be fixed.
TODO(b/142911355): [VNDK APEX] Fix hard-coded references to /system/lib/vndk
Bug: 143192278
Bug: 142912195
Test: m com.android.vndk.current and check if /system/lib/vndk
OVERRIDE_TARGET_FLATTEN_APEX m com.android.vndk.current
Change-Id: I2ebacde7fcd1c7621e4509e08a76765e1dfeb059
When an APEX is built with uses_sdks, any depedndency from the APEX to
the outside of the APEX should be from the SDKs that the APEX is built
against.
Bug: 138182343
Test: m
Change-Id: I1c2ffe8d28ccf648d928ea59652c2d0070bf10eb
The arch variants are hardcoded in every module type. Refactor
them out into a Target.Variations() method in preparation for
splitting the arch mutator into two, which will require using
different variations.
Test: m checkbuild
Change-Id: I28ef7cd5168095ac888fe77f04e27f9ad81978c0
Older VNDK libraries are provided as vndk_prebuilt_shared modules. Those
are added to corresponding VNDK APEX as dependencies.
With VNDK APEX installed, VNDK libs are unnecessary. By the way, since
there can be vendor modules which depend on VNDK libs, Make targets are
still emitted with UNINSTALLABLE=true.
Android.mk has additional modules for vndk libraries which are named
with apex name as suffices. For example, if libfoo is a vndk library,
then libfoo.vendor is its vendor variant and it would be in
/system/lib/vndk. But with vndk apex, it has additional
libfoo.com.android.vndk.current variant.
Bug: 141451661
Bug: 139772411
Test: m (soong tests)
Test: boot with aosp_arm64 system image on Q vendor device
Change-Id: I269c28a4d4c4e2f1518bd51df558438fe5316774
APK in a flattened APEX is installed as 'ETC' class module, instead of
the 'APP' class. This is to prevent Make from doing app-specific
amendments (e.g. such as adding module name after my_module_path)
to the paths and filenames which are all correctly set in the Soong
side.
Test: add `installable:true` to development/samples/Snake/Android.mk
and add 'Snake' to the `apps` property of an APEX. The build is
successful.
Test: build the APEX on a device with TARGET_FLATTEN_APEX=true
The APK is at system/apex/<apexname>/app/Snake/Snake.apk
Bug: 142537672
Change-Id: I314bf1ab7abd4d4a4e9fa210442c004f54d8ccca
This fixes the bug: flattened apex is installed under
/system/apex/<soong module name> not /system/apex/<apex name>.
Flattened apexes can be seen as pre-activated under /system/apex, which
is bind-mount to /apex. Therefore, install dir for flattend apex should
be the same with the activation point(or mountpoint) of non-flattened apex.
When apex_name: is set, it should be used as its name instead of name:.
This change also emits makevar APEX_FILE_CONTEXTS_INFOS which is list of
pairs(<apex_name>:<file_contexts>) and is used by sepolicy to modify
file_contexts files for flattened apexes properly.
Bug: 123314817
Bug: 142300241
Test: add a test apex module(foo) with apex_name(foobar)
Test: OVERRIDE_TARGET_FLATTEN_APEX=true m foo
Test: see if the apex is installed correctly(/system/apex/foobar)
Test: see if the file_contexts for foo is correctly modified
Change-Id: I97a07de5cc772cd1d91e95ae059d282038028916
The check doesn't make sense for host.
Test: ALLOW_MISSING_DEPENDENCIES=true DIST_DIR=out/dist
./art/tools/dist_linux_bionic.sh -j110 com.android.art.host
is successful
Change-Id: Icdb4bb557a83b3c8044e6c73721dab47c56888a1
This allows us to change `logpersist.start` to `sh_binary` and unblocks
breakpad symbol uploading
Test: convert logpersist.start to sh_binary and verify symlinks are correct
Change-Id: I1b86c512df73a336205ca35216445a08e43bb879
Add a ToMakePath() method that returns a new path that points out
out/ instead of out/soong/, and replace the
"$(OUT_DIR)/" + path.RelPathString()
pattern with
path.ToMakePath().String()
Bug: 141877526
Test: m checkbuild
Change-Id: I391b9f2ed78c83a58d905d48355ce9b01d610d16
Create a new type InstallPath that is similar to OutputPath to
differentiate intermediates output paths from installed output
paths.
RelPathString is a poorly defined, undocumented function that is
primarily used to get an install path relative to out/soong to
generate an equivalent install path for Make relative to $(OUT_DIR).
Move it to InstallPath for now, and fix the one remaining user on
OutputPath.
Add a method to create an NDK install path so that ndk_sysroot.go
doesn't have to do it manually with PathForOutput.
Bug: 141877526
Test: m checkbuild
Change-Id: I83c5a0bd1fd6c3dba8d3b6d20d039f64f353ddd5
This change reverts following three changes to remove the no_apex
property. no_apex: true is equivalent to apex_available:
["//apex_available:platform"].
Revert "fix: "no_apex" can be put in defaults"
This reverts commit cc372c5b1d.
Revert "Add no_apex check for static library"
This reverts commit 2db7f46d0c.
Revert "Add no_apex property"
This reverts commit 4f7dd9b4db.
Bug: 139870423
Bug: 128708192
Test: m
Change-Id: Ia4b094e371e9f8adff94ae6dc3ebb8e081381d4e
apex_available property controls the availability of a module to APEXes.
For example, `apex_available: ["myapex", "otherapex"]` makes the module
available only to the two APEXes: myapex and otherapex, and nothing
else, even to the platform.
If the module is intended to be available to any APEX, then a pseudo
name "//apex_available:anyapex" can be used.
If the module is intended to be available to the platform, then another
pseudo name "//apex_available:platform" is used.
For now, if unspecified, this property defaults to ["//apex_available:platform",
"//apex_available:anyapex"], which means the module is available to everybody.
This will be reduced to ["//apex_available:platform"], when marking for
apex_available for existing modules are finished.
Bug: 139870423
Bug: 128708192
Test: m
Change-Id: Id4b233c3056c7858f984cbf9427cfac4118b2682
VNDK APEX should be named after the version of VNDK libs.
For example, if vndk_version is 29, then the apex name should be
com.android.vndk.v29. If vndk_version is not set or is 'current', then
the platform vndk version is implied.
This is done with setting "apex_name" with proper name.
Bug: 139774701
Test: m (soong test added)
Change-Id: I918252f12ccd351886030fe9139a020d6cf1ff32
All APEXes need to ship w/o libbinder since it does not offer a stable
wire protocol. Removing this since nothing adds a libbinder dep here.
Bug: 139016109
Test: build
Merged-In: Ic6a3fcb68054d8c7a5f2f64795a0e2889ce1abe8
(cherry picked from commit 1c3108d629)
Change-Id: Ic6a3fcb68054d8c7a5f2f64795a0e2889ce1abe8
For now, apex_name is used only for flattened apex.
Even if apex_name is set, the activation point of unflattened is
determined by 'name' in apex_manifest.json.
This change make apex_name as priority.
If apex_name is set, then use this for apex name
- update name in apex_manifest.json to apex_name
- do not check if key filename matches (use apex_name as key name)
This can be useful if soong wants to rename apex module. Simply setting
apex_name has the same effect of renaming "activation point" of apex.
But: 139774701
Test: m (soong test)
Change-Id: I8ea3645e4aa8f317997bc1443ec308ed0595b1c2
The return value is true if TARGET_FLATTEN_APEX is true,
TARGET_BUILD_APPS is false and the module is flatten variant.
Only the module name of flattened variant is appended with
".flattened" as suffix.
Test: m -j
Bug: 139716748
Change-Id: I966a178b43c5d08bd69d19e97e59456aa66d9d28
This change introduces a new module type named 'sdk'. It is a logical
group of prebuilt modules that together provide a context (e.g. APIs)
in which Mainline modules (such as APEXes) are built.
A prebuilt module (e.g. java_import) can join an sdk by adding it to the
sdk module as shown below:
sdk {
name: "mysdk#20",
java_libs: ["myjavalib_mysdk_20"],
}
java_import {
name: "myjavalib_mysdk_20",
srcs: ["myjavalib-v20.jar"],
sdk_member_name: "myjavalib",
}
sdk {
name: "mysdk#21",
java_libs: ["myjavalib_mysdk_21"],
}
java_import {
name: "myjavalib_mysdk_21",
srcs: ["myjavalib-v21.jar"],
sdk_member_name: "myjavalib",
}
java_library {
name: "myjavalib",
srcs: ["**/*/*.java"],
}
An APEX can specify the SDK(s) that it wants to build with via the new
'uses_sdks' property.
apex {
name: "myapex",
java_libs: ["libX", "libY"],
uses_sdks: ["mysdk#20"],
}
With this, libX, libY, and their transitive dependencies are all built
with the version 20 of myjavalib (the first java_import module) instead
of the other one (which is for version 21) and java_library having the
same name (which is for ToT).
Bug: 138182343
Test: m (sdk_test.go added)
Change-Id: I7e14c524a7d6a0d9f575fb20822080f39818c01e
Translated second architectures now go in NativeBridgeArch instead
of DeviceSecondaryArch.
This reapplies I568046330abc002d4eed582cb999b62a5eaba790 with
ctx.Config().HasMulitlibConflict() added to fix the NDK build,
which has arm64, arm, x86_64, and x86 architectures enabled.
Test: m checkbuild
Test: OUT_DIR=out_ndk build/soong/scripts/build-ndk-prebuilts.sh
Test: no change to build.ninja or Android-aosp_cf_x86_phone.mk
Change-Id: Iadcafbd64bfb9579ae7c86914927c43a062b0c8e
Remove the mention to the (APEX) manifest (`apex_manifest.json`) from
this field's documentation, as `Apex_name` is neither read from nor
written to that file.
Test: n/a
Change-Id: Ia0dfac4f35f1ea697f379bcb19dd11af8705a0a0
Vendor variant is now divided into several vendor.{version} variants,
depending on their intended usages:
vendor.{BOARD_VNDK_VERSION}: vendor and vendor_available modules
vendor.{PLATFORM_VNDK_VERSION}: VNDK modules in the source tree
vendor.{snapshot_ver}: VNDK snapshot modules
This also affects exported module names from Soong to Make. But to
maintain backward compatibility, ".{BOARD_VNDK_VERSION}" suffix will not
be emitted for modules having version BOARD_VNDK_VERSION, so that vendor
modules still can be referred as-is.
Bug: 65377115
Bug: 68123344
Test: clean build and boot blueline
Change-Id: Ib9016a0f1fe06b97e9423fd95142653a89a343fa
If payload type of APEX is zip, flattened variant and non-flattened
variant create the same MK rules (it's build error). So only
non-flattened variant case, MK rule is created.
And if a.flattenedConfigValue is true, soong don't mutate variants. So
when a.flattenedConfigValue is true, MK rule should be created.
This CL is a partial revert of 4c04713871
Bug: 139716748
Test: m -j com.android.art.host
Change-Id: I37b2fa17e203fb668a0255cda37f10fcfadb3f85
The NOTICE files are missing from prebuilt apexes, and it turns out they
were excluded when building bundles.
Bug: 140317706
Test: Ran build_mainline_modules.sh and checked bundle base modules.
Change-Id: I92c4231f2007e1d8cd9c2bd044201458803c0fd7
Merged-In: I92c4231f2007e1d8cd9c2bd044201458803c0fd7
Ensure that it ends up in the bionic/ subdir in the Runtime APEX and is
symlinked from /system/lib(64). That makes it available in the default
(platform) linker namespace and no longer requires it to be accessible
through the runtime namespace. All this makes it consistent with how the
other Bionic libs are handled, and avoids the need for various special
cases in ld.config.txt files (to be cleaned up later).
With this there might no longer be a need for a linker namespace for the
Runtime APEX, and we could consider removing the kludge with the
/apex/com.android.runtime/${LIB}/bionic subdirectories and the special case
in getCopyManifestForNativeLibrary in apex.go. However, keeping it calls out
the special treatment of those libraries, and allows a namespace to be added
again if necessary.
Test: Build and boot on taimen_hwasan-userdebug
Test: adb shell /apex/com.android.art/bin/dexdump on taimen_hwasan-userdebug
Test: atest CtsCompilationTestCases on taimen_hwasan-userdebug
Bug: 140734238
Bug: 140790209
Change-Id: Ieb506bfa5d5c159db391273c7eba41d7909de286
The apexBundle module is mutated flattened variant and unflattened
variant. So we can install flattened apex and unflattened apex at the
same time with adding {Apex module name}.flattened to PRODUCT_PACKAGES.
Bug: 139716748
Test: m -j && Add com.android.art.debug.flattened to PRODUCT PACAGES and
check system/apex/com.android.art.debug directory
Change-Id: I0d8c9a155e47c5d0ff13e2f55d4080d91ef5093d
This commit switches the build rule for APEX nativeTests from
`$(BUILD_PREBUILT)` to `soong_cc_prebuilt.mk`. Since these are not real
prebuilts, `soong_cc_prebuilt.mk` should be sufficient.
Using `soong_cc_prebuilt.mk` also fixes a problem reported by prebuilt
ELF checker because `LOCAL_SHARED_LIBRARIES` are not generated for
nativeTests.
Bug: 139525855
Test: lunch aosp_walleye-userdebug && make check-elf-files
Change-Id: I7bfc29f9c2708896dad4e7cfc214480e1205a51e
"apex_vndk" is a variant of "apex" module.
apex_vndk {
name: "com.android.vndk",
..
}
This rule is used to produce a VNDK APEX per vndk version.
It supports automatic inclusion of vndk libs.
If "vndk_version" property is set, the prebuilt vndk libs of
the version will be included in the apex bundle.
apex_vndk {
name: "com.android.vndk.v29"
vndk_version: "29",
...
}
Otherwise, platform's vndk version is used.
This will replace /system/{lib}/vndk-{ver} and vndk-sp-{ver}.
Bug: 134357236
Bug: 139772411
Test: m com.android.vndk
Change-Id: Ib5c86e625839389670d13c683a7427198ef6852f
New property "apps" is added to APEX soong module. This property
configure a list of APKs to package inside APEX payload. For each
android_app module package it's APK into /app/$(android_app module Name)
directory.
Bug: 139906763
Test: m -j
Change-Id: Ic6655f211951af08097c76e157a07ebc52425718
So far a very basic version.
In case there is a diff, soong will emit an error message with a command
to resolve the diff. Example:
New unexpected files were added to com.android.apex.cts.shim.v3. To fix the build run following command:
system/apex/tools/update_whitelist.sh system/apex/shim/build/default_shim_whitelist.txt out/soong/.intermediates/system/apex/shim/build/com.android.apex.cts.shim.v3/android_common_com.android.apex.cts.shim.v3/com.android.apex.cts.shim.v3-content.txt
Test: m checkbuild
Bug: 139125405
Change-Id: I57e694f394c56105dc7363a684983605c7bf5e7d
This includes a few changes that make AndroidMkEntries more resemble
AndroidMkData, especially in terms of how extra entries are added.
Most importantly it can now have multiple custom functions.
Test: Soong tests
Change-Id: Ibf9102624d16d0c1c9894a2794fc7c797bb34c9a
Builtin APEXes (/system, ...) don't need hashtree.
So if we remove hashtree from builtin APEXes we can save more space in
/system.
However, this change only affects to the apex bundles which are built
from source, not from prebuilts.
Removing hashtree from prebuilts will be handled in a follow CL.
Bug: 139957269
Test: m
Test: see if /system/apex/*.apex have hashtree
Change-Id: Id755382b2153bf6bbaf2c480f3bbb91b6f62a564
Allow modules to depend on the flattened flavor of an APEX package
using `module.{flattened}` (returning the output install path of the
flattened APEX).
Also add this flattened output install path as an implicit output
dependency of a flattened APEX phony module.
Test: m checkbuild
Bug: 139277987
Change-Id: I8c55d5ff2922a36fcc8ec0819fbe3e93a772ef8a
Merged-In: I8c55d5ff2922a36fcc8ec0819fbe3e93a772ef8a
If no_apex module is static library, this module is not included to
filesInfo. So add a check tat the static library is no_apex among
indirect dependencies.
Bug: 139016109
Test: m -j
Change-Id: I46ddf099715aea0e088027a1141e282969cef0e1
This change adds 'no_apex' property which, when set to true, prevents
the module from being installed to any APEX. If the module is included
either directly or transitively in an APEX, but build fails.
Bug: 139016109
Test: m
Change-Id: If1478aa9660a3442f7dd1ffe45e4ca5611a6acbe
"use_vendor: true" APEX modules may bundle vendor variants of cc
modules, which can depend on llndk stubs.
In that case, those llndk libraries should be also counted as required
libraries. (stored in 'requireNativeLibs' key in apex_manifest.json)
Bug: 138695532
Test: m nothing (runs soong tests)
Change-Id: If7ad4dec0e723c8d0c73ca60453b555063e14694
java_import can be included in apex via 'java_libs' property.
Bug: 139175488
Test: m (apex_test.go updated)
Change-Id: I3680a47cdac93b0cb2d41da8df3f8defa2bbe670
When the native coverage is enabled, APEXes (and files there) are built
for native coverage as well.
Bug: 138952487
Test: make -j NATIVE_COVERAGE=true COVERAGE_PATHS='*' com.android.resolv
find out -name "*.gcno" | grep DnsResolver shows files
Test: libnetd_resolv.zip is found under
$(TARGET_OUT)/apex/com.android.resolv/lib directory
Change-Id: I97bcee9bf8ffc0dc71453abbdb613ed56ea2cdb4
To generate ld.config.txt dynamically(b/123722631), each APEX should
provide some dependency information:
a) list of libraries which other APEXes(or system) can use from this apex
b) list of libraries which this apex uses from other APEXes(or system)
This change puts dependency information in apex_manifest.json at
build-time with two additional keys:
a) provideNativeLibs
b) requireNativeLibs
Bug: 138695532
Test: m (runs soong tests)
Test: find $OUT/apex -name apex_manifest.json -exec cat {} \;
(shows contents of apex_manifest.json files)
Change-Id: Iaad12c8c35454222ad177ce923cce76ef12a8a5a
For some APEX packages (i.e. the Runtime Testing APEX), the set of
files to copy can be so large that the copy commands (which are part
of the Ninja shell command executed for an APEX package) may exceed
the maximum length of argument to the exec() functions (ARG_MAX). To
work around this limitation, record these copy commands in a Ninja
response file (rspfile) and `source` this file in the Ninja command to
execute them.
Test: m nothing (`apex/apex_test.go` amended)
Test: m com.android.runtime.testing (with CL https://android-review.googlesource.com/c/platform/art/+/1008034/ cherry-picked)
Bug: 129534335
Change-Id: I09ff2d9cf66bfd4cbc12cb724a45d455d08da0b2
To build the platform for ASAN, we do
`m && SANITIZE_TARGET='addresss' m`
However, at the end of the second build, the system partition could have
conflicting APEXes; prebuilt APEXes from the first build and
source-built APEXes from the second build. Since the file names for the
prebuilt and the source-built are different (e.g.
com.google.android.media.apex v.s. com.android.media.apex), we end up
having two files for the same APEX. This is confusing apexd at runtime
and the device fails to boot.
To fix this, when building a non-prebuilt APEX, the prebuilt APEX might
have been installed by the previous build is deleted.
Bug: 138146044
Test: lunch aosp_cf_x86_pasan; m && SANITIZE_TARGET='address' m
check that out/target/product/vsoc_x86/system/apex has
com.android.*.apex only.
Change-Id: Ib5a021a297cf0173ea5a3b50e9398b1cf295c558
In `apex.apexBundle.GenerateAndroidBuildActions`, we used to pass the
"all tests" ("") module as `module` for all `apexFile` objects created
from a test module using `test_per_src: true`. An immediate issue of
this situation was that the "" module is hidden from Make, which made
all the generated `apexFile` objects hidden from Make too. This would
break the construction of flattened APEXes, as they rely on Make logic
to install their files.
Instead of collecting `test_per_src` test variations' output files in
`cc.Module.GenerateAndroidBuildActions` and using them in
`apex.apexBundle.GenerateAndroidBuildActions` as part of handling the
"" variation as a direct dependency of an `apexBundle`, process them
as indirect dependencies (and do nothing for the "" variation direct
dependency).
Adjust the indirect dependency logic in
`apex.apexBundle.GenerateAndroidBuildActions` to allow not only
shared/runtime native libraries as indirect dependencies of an
`apexBundle`, but also `test_per_src` tests.
Test: m (`apex/apex_test.go` amended)
Bug: 129534335
Change-Id: I845e0f0dd3a98d61d0b7118c5eaf61f3e5335724
This change fixes a bug that a symlink in the bin/* directory of an APEX
is not added to the fs_config file.
Bug: 137918291
Test: choosecombo aosp_x86_arm
TARGET_FLATTEN_APEX=false m com.android.runtime.debug is successful
Change-Id: Ib10b635f08e03b18e8ad5407d8de6bac2c85326d
If a test module with a `test_per_src` property set to `true` is
included in an APEX module, add all the variants for mutator
`test_per_src` as dependencies of the APEX module (not just the
first one).
This is done by adding variation "" of mutator `test_per_src` when
adding a test dependency to an APEX module, which creates an indirect
dependency of the APEX module on all the `test_per_src` variants of
the test module. When generating outputs for the APEX bundle, fetch
and include the set of test outputs from the "" variant.
Test: m (`apex/apex_test.go` amended)
Bug: 129534335
Change-Id: I1c99855971a8a9b2fc5b964a420e882b6791d4e6
This change fixes a bug that APEX prebuilts are auto-disabled in
platform builds (i.e. non-unbundled builds).
Bug: 137282010
Test: m com.android.conscrypt
check that the apex is from prebuilt directory
Merged-In: I935ef3896e80864bdcc1ca5f6fc12b63c9588c0d
(cherry picked from commit 895e224d8f)
Change-Id: I935ef3896e80864bdcc1ca5f6fc12b63c9588c0d
This change introduces NativeBridgeRelativePath and
NativeBridgeSecondaryRelativePath product variables to
make relative path to native_bridge binaries configurable.
It also removes moves information about native bridge
host architecture to Target and sets it during decodeTargetProductVariables
Test: make PRODUCT-cf_x86_phone-userdebug dist
Change-Id: Ie736e81eae507e1775566ce9f29135011b12af27
This change fixes two problems:
1) the prebuilt apexes are force disabled for the unbundled builds
because we need to build the modules from the source then
2) the dependencies from an sdk_library module to
*.stubs.{public|system|tests} are not added for the unbundled build
because the stubs modules are disabled.
Bug: 137282010
Test: unbundled mainline builds are successful
Test: build com.android.media and inspect the jar file to see if
hiddenapi flags are there
$ cd out/dist/mainline_modules_arm
$ unzip com.android.media.apex apex_payload.img
$ mkdir -p mnt
$ sudo mount -o ro,loop apex_payload.img mnt
$ dexdump2 mnt/javalib/updatable-media.jar | grep hiddenapi
shows results
Change-Id: I2c00af07aac4a15770d3acab011a36e2e4803bfc
Some prebuilt APEXes are enabled with scudo, which causes crash on
devices with asan.
Bug: 137216042
Test: build walleye_hwasan and check if apexes are not from prebuilts
Merged-In: Ic436ad06e724af952d99fb8e66358a595450eb84
(cherry picked from commit 3b98a509f1)
Change-Id: Ic436ad06e724af952d99fb8e66358a595450eb84
When the device does not support non-flattened APEX (i.e.
TARGET_FLATTEN_APEX is set to true), then ignore the prebuilt_apex
modules even when `prefer: true`.
Bug: 136251130
Bug: 136662772
Test: build marlin, and check if /system/apex has flattened APEXes
Merged-In: I9f3dfefc3bd357d6750fbb1e418937d095720b04
(cherry picked from commit 0a573d798f)
Change-Id: I9f3dfefc3bd357d6750fbb1e418937d095720b04
Bug: 137015854
Test: m
Test: build cf_x86_phone and ensure there is no dangling symlink
/system/lib/arm/libm.so
Change-Id: I0f5efdb9e1f5e4e213ca4ebaa28ace0a130b36a6
Instead of circumventing the limitation of Prebuilt implementation by
picking a source path itself, it now uses the same mechanism as
archMutator and replaces the source path in advance so that Prebuilt
always sees the corrent source path.
Because this requires the Apk field to be a string pointer, the single
source prebuilt implementation is being updated to be reflection-based.
Test: Soong unit tests, m soong_docs, TreeHugger
Change-Id: I2304f15e32d632f74f95f0d9e9bf1f75ff3e2225
For APEXes to share C++ native libraries, we need a new kind of depedency
between APEXes: "providing" APEXes and "using" APEXes. To reflect this
dependency two new properties are added.
provide_cpp_shared_libs: bool
this indicates that the current APEX module provides the native C++ shared
libs to other APEXes.
uses: []string
this indicates that the current APEX module uses the native C++ shared
libraries from APEXes listed.
With these two, "using" APEXes can omit shared libraries in its APEX
bundle and use them from the "providing" APEXes.
Note that without corresponding changes in ld.config.txt, this won't
work.(The linker namespaces should be configured so that user APEX can
access provided libs.)
Bug: 136975105
Test: m nothing (this will trigger soong's test)
Change-Id: Iec6f9f67bcbde01145acc383f862ba21c8197536
Instead of outputting an aggregated NOTICE file as an intermediate build
resource to allow Make to include it in the final system-wide NOTICE,
process and embed it as an asset in the final APEX. This allows us to
update the NOTICE contents automatically when an APEX is updated.
Fixes: 135218846
Test: Built mainline modules, apex_test.go
Change-Id: Ic851b330fe93be1f602907d44ecc7886c3b0171b
If embed_notices or ALWAYS_EMBED_NOTICES is set, collect NOTICE files
from all dependencies of the android_app, merge them with the app's own
one (if exists), transform it to HTML, gzip it, and put it as an asset
in the final APK output.
Bug: 135460391
Test: app_test.go
Change-Id: I52d92e2fd19b3f5f396100424665c5cc344190d8
apexMutator only modifies the currently visited module, it can
visit modules in parallel.
Test: builds
Change-Id: I7d0ad142f5161742bd25dc57bac64c0f4a733ecf
I found these while trying to build an aosp_arm system image with RBE,
which only makes the sources that you depend on available to every rule.
The hardcoded prebuilts/sdk path is a bit unfortunate, but that's
currently hardcoded as a default in the script as well.
Bug: 130111713
Test: treehugger
Test: build a system image with RBE
Change-Id: I4415563017e053749788b0a537a48d61a2161935
Add an empty GenerateAndroidBuildActiosn to DefaultsModuleBase
so that every defaults module doesn't need to provide one. This
will also allow adding an implementation in the next patch.
Test: m checkbuild
Change-Id: I13554bdb3a287c2f18e1efab74d4f08a1ba8620c
blueprint.BaseModuleContext is the set of methods available to all
module-specific calls (GenerateBuildActions or mutators). The
android package split the same functionality across baseContext (nee
androidBaseContext), BaseModuleContext, and BaseContext.
Consolidate all of them into android.BaseModuleContext.
Test: m checkbuild
Change-Id: I2d7f5c56fd4424032cb93edff6dc730ff33e4f1e
There are cases where a module needs to refer to an intermediate
output of another module instead of its final output. For example,
a module may want to use the .jar containing .class files from
another module whose final output is a .jar containing classes.dex
files. Support a new ":module{.tag}" format in any property that
is annotated with `android:"path"`, which will query the target
module for its ".tag" output(s).
Test: path_properties_test.go, paths_test.go
Test: no unexpected changes in build.ninja
Change-Id: Icd3c9b0d83ff125771767c04046fcffb9fc3f65a
When HWASAN is enabled, the runtime is conceptually part of Bionic (and
mutually depends on it), so it needs to be treated in the same way as the
Bionic libs.
Now there are only two copies of the runtime: the one in
/system/lib64/bootstrap (which won't be used by ordinary processes) and the
one in the runtime APEX.
This reduces the size of the HWASAN system image and fixes an issue where
multiple copies of the HWASAN runtime were being loaded into 64-bit binaries in
APEXes because the linker namespace for the binary is different from the one
for its dependent libraries outside of APEXes. HWASAN only supports loading
one copy of the runtime per process, so this was causing such binaries to
crash on startup.
Change-Id: I228896e193a035e6dfba9f6e28d0b2e12fc163ea
apexkeys.txt now correctly lists prebuilt APEXes with keys specified as
PRESIGNED.
This change also fixes a bug that non-installable APEXes are listed in
the file.
Bug: 131130235
Test: m out/soong/apexkeys.txt and check that
com.android.apex.cts.shim.apex is listed there with PRESIGNED keys.
Merged-In: Ib6d391a82864714743a1cc59cd655bea917b5073
Change-Id: Ib6d391a82864714743a1cc59cd655bea917b5073
(cherry picked from commit a41f12a6fa)
This commit allows a module to opt in for ABI checks even when it is
not an LLNDK/VNDK module.
Bug: 131421213
Test: Add `header_abi_checker { enabled: true, }` to some module
Change-Id: Ie09d262e651cbb44d7d0eba652f55dc1e1e52962
Target SDK version is used for targeting an APEX to a specific set of
platform builds. Usually, the targeting is unrestricted (in case the
APEX can run on all platforms), or based on platform SDK version (e.g.
28 for P). However, when the platform is under development and SDK is
not finalized, the targeting should be much more fine-grained; the
APEX should be targeted to a very specific build that supports the same
set of APIs that the APEX was built against.
To support that, target sdk version is automatically set by the build
system. When the platform is released or SDK is finalized, the target sdk
version set to the SDK version number. If not, it is set to
<version_code>.<fingerprint> (e.g., Q.123456).
Note that the target sdk version set by the build system is used only
when the target sdk version is not explicitly set in
AndroidManifest.xml.
Bug: 130541924
Test: UNBUNDLED_BUILD_SDKS_FROM_SOURCE=true \
UNBUNDLED_BUILD_TARGET_SDK_WITH_API_FINGERPRINT=true \
TARGET_BUILD_APPS=com.android.tzdata m
build.ninja has --target_sdk_version Q.$$(cat out/soong/api_fingerprint.txt)
Test: aapt dump badging out/dist/com.android.tzdata.apex | grep \
targetSdkVersion shows:
targetSdkVersion:'Q.6ee443d9ad5f0cca7a43cfa97b7fc62a'
Change-Id: I086230d787f01075c28fc3f0163550300fa00212
* Makes it more inline with prebuilt_etc;
* For shim apexes, prebuilt_apex modules have pattern of
com.android.apex.cts.shim.v1_prebuilt, but I would prefer
pre-installed shim to be: /system/apex/com.android.apex.cts.shim.apex
Bug: 128677967
Bug: 127789981
Test: m
Change-Id: I34e3e078733420b5cf777fd6e3ce4d8c5796b19b
In case of shim apexes, we prebuilt all of them, but only need to
install v1 to a system partition.
Bug: 128677967
Test: manually checked that non-installable prebuilts don't end in /system
Change-Id: I112432abfd8f03cc7d7379ea3cab3f5491ace49c
android/soong/common was renamed to android/soong/android long
ago, but the pctx package path was still "android/soong/common".
This required all users of rules defined in android/soong/android
to import "android/soong/android" and then
pctx.Import("android/soong/common").
Test: m checkbuild
Change-Id: I20d096522760538f7cfc2bec3d4bfeba99b275d4
The public key associated with an APEX is always included in the APEX.
Obviously, the public keys are no longer installed to
/system/etc/security/apex
Bug: 128344735
Test: m
Change-Id: I1e1aef1d32597a447b57d49ab80bbfb921fa8178
Arch-specific source can be specified for prebuilt_apex as follows.
arch: {
arm64: {
src: "myapex-arm64.apex",
},
},
A note on the implementation. The Src property was not tagged as
`android:"arch_variant"` as usual. Instead, multiple Src properties are
explicitly declared like
struct Arch {
struct Arm {
struct Src *string
}
...
}
Corresponding Src property is manually selected according to the
MultiTargets()[0].
This is because prebuilt_apex is mutated only for android_common, in
order to have the same arch variant with the apex module type.
Therefore, we can't rely on the arch_variant tag.
Bug: 127789981
Test: m (apex_test amended)
Change-Id: I77dbe626171d8975f549bdb4af3c487232cf05f7
Notice file for an APEX is created by merging notice files for the
modules included in it (plus the notice file for the APEX itself if
specified).
Notice files having the same content are not duplicated; it is emitted
only once.
Bug: 128701495
Test: m (apex_test is amended)
Test: m and inspect $(PRODUCT_OUT)/obj/NOTICE.txt to check there are
license entries for /system/apex/*.apex files
Change-Id: I169d91038291a6c71615de97cf5b03174afab5d4
Move the logic from ctx.ExpandSources into android.PathsForModuleSrc
and ctx.ExpandSource into android.PathForModuleSrc, and deprecate
them. When combined with the pathDepsMutator this will let all
properties that take source paths also take filegroups or genrule
outputs, as long as they are tagged with `android:"path"`.
Test: All soong tests
Change-Id: I01625e76b5da19240e9649bf26a014eeeafcab8f
This change fixes the problem that symbol files for APEXes are installed to
incorrect path when TARGET_FLATTEN_APEX is set to true or the canonical
name of an APEX is different fro the module name of the APEX.
For the case when TARGET_FLATTEN_APEX is true, LOCAL_SOONG_SYMBOL_PATH
is set to point to the runtime path of a file (e.g. /apex/<name>/*).
For the case of the different canonical and module names, apex_name
property is added to explicitly specify the canonical name
Bug: 120846816
Test: m and inspect that symbol files exist under
$(PRODUCT_OUT)/symbols/apex/com.android.runtime/
Change-Id: Idfec88d6a30a18c225b0d87b868b9f1e0a617e38
Add `android:"path"` to all properties that take paths to source
files, and remove the calls to ExtractSource(s)Deps, the
pathsDepsMutator will add the necessary SourceDepTag dependency.
Test: All soong tests
Change-Id: I488ba1a5d680aaa50b04fc38acf693e23c6d4d6d
These additional binary types are useful for some apexs. Add the
ability to include them. Due to the nature of the resulting artifacts
only py binaries with embedded launchers and host go binaries are
supported.
Test: m com.android.support.apexer
Bug: 119332365
Bug: 119332362
Change-Id: I27c253d3647cf7bbe15896610d7a74a5f93e8bec
This reverts commit aa65e17016.
Reason for revert: Not compatible with PRODUCT_PACKAGES, and so has very limited use.
Change-Id: Ib141d3984a6f12bb50989e66037494c466b066f1
This new module type replaces the inherit-package function in make by
allowing developers to override the name, the certificate, and the
manifest package name of an android_app module.
Bug: 122957760
Fixes: 123640028
Test: app_test.go + BrowserGoogle
Change-Id: Iefe447e7078b25039233221361ef95c83a29973a
Flattened apexs were not including symlinks. This could cause expected
files not to be present.
Test: lunch aosp_marlin-userdebug
m droid
stat $OUT/system/apex/com.android.runtime.debug/bin/dalvikvm
Bug: 124924906
Change-Id: I04e696602b776b383bf7198eb19220353df74192
Host apexs don't have a system-image associated with them so we need
to include all libraries, including those with stubs.
Test: ./art/test/testrunner/run_build_test_target.py -j50 art-linux-bionic-x64-zipapex
Bug: 125417922
Change-Id: Ie76e6a34e8be1057b29e2e005597c3e4b5fb1f9c
By default, if a lib is included in an APEX, all its direct and indirect
dependencies are also included in the same APEX. However, when one of
the dependencies have stable API (i.e. has stubs: {...}) then the lib
having stable API and its dependencies are not included in the APEX.
However, the problem here is that the lib having stable API might not be
installed on the system, thus causing error at runtime. This can happen
if there is no other module in the platform that depends on the lib.
This change fixes the problem by adding such libraries as external
dependencies so that they are also installed on the device along with
the APEXes using them.
Bug: 124831003
Test: m installclean; m com.android.resolv
libbinder_ndk, libvndksupport are found under system/lib
Change-Id: I457e03ff3fce37e0890c64d911e6e0ea6d0c6dd6
apexkeys.txt is a text file having APEX-to-keys mappings. The file is
included in the target-files package where it is used when re-sign the
APEXes with release keys.
Each line of the file consists of 5 fields:
1) name: file name of the APEX
2) public_key: the public key for the apex_payload.img of the APEX
3) private_key: the private key used to sign the apex_payload.img
4) container_certificate: the certificate used to sign the APEX zip
container
5) container_private_key: the private key used to sign the APEX zip
container
Bug: 124406181
Test: m out/soong/apexkeys.txt and inspect the content
Test: TARGET_BUILD_APPS=com.android.tzdata m dist and make sure
out/dist/apexkeys.txt exists
Change-Id: I1daa13ec50956323b97e15e8df7f1fbe5ea21d63
The manifest and androidManifest properties in the apex module type now
supports ":name" syntax.
Bug: 123857186
Test: m (apex_test amended)
Change-Id: Ic4e5a73cf73260d156ec61d07932ad07b2561413
The properties are no longer required as an APEX module is mutated with
required sanitizers which are gathered by scanning the sanitizers that
are enabled for its direct dependencies.
Bug: 124128094
Test: m on marlin
The extractor libs are found under /system/apex/com.android.media
Merged-In: I55961d400dcbac067a5c0dcecb90e399d4991a70
Change-Id: I55961d400dcbac067a5c0dcecb90e399d4991a70
(cherry picked from commit abda0eb76b)
Zip apex files are not really installed so should not ever be
flattened. Unfortunately we didn't send this information to make which
would still unzip everything.
Test: ALLOW_MISSING_DEPENDENCIES=true
./art/tools/build_linux_bionic_tests.sh com.android.runtime.host
Bug: 124333446
Change-Id: I3eda579f8c40f768f1ef2be867967f436396cd4c
AndroidManifest.xml file can be specified via the 'androidManifest'
property in the apex module type. It can be used to have a custom
AndroidManifest.xml that have additional tags (such as <uses-sdk> or
<uses-feature>) for precise targeting.
The property is optional; if unspecified, then one is auto-generated as
before.
Bug: 123857186
Test: m apex_test_build_features
Change-Id: Id7ee0471661887cfa11124cbaa3beea8cffcdda4
Zipapex's cannot be flattened so do not prevent them from being built
even if TARGET_FLATTEN_APEX is set.
Test: lunch aosp_arm-eng;
ALLOW_MISSING_DEPENDENCIES=true \
./art/tools/build_linux_bionic.sh com.android.runtime.host
Bug: 124333446
Change-Id: I4bef65cce100136a8f7852692e841b62ed9e399f
It is used to configure an apex to prefere specific sanitizer variants
if available. For example, if a lib is with sanitize: {cfi: true} then
an APEX with prefer_sanitize: {cfi: true} will use the sanitized variant
of the lib.
Bug: 124128094
Test: m on marlin
extractor libraries are found under /system/apex/com.android.media
Change-Id: I858778eef78c5791cdeb497c7c11688cb128b5fe
Some test and other miscallaneous use-cases need apexs that are
installable but will not affect the installation of other artifacts
onto the device. For this purpose we added a test_apex type. These
test_apex targets create apex files and have all the normal attributes
but are not considered the source of any of their contents for the
device by the rest of the build-system.
Test: build and boot
Test: ALLOW_MISSING_DEPENDENCIES=true \
DIST_DIR=$ANDROID_BUILD_TOP/out/dist \
./art/tools/dist_linux_bionic.sh -j72 showcommands com.android.runtime.host
Bug: 123591866
Bug: 123892969
Change-Id: I7f50be0ac0425cd87868145e18bcee6962d472ef
We were marking any libraries included in any apex (including
non-installable apex's like com.android.runtime.host) as
uninstallable. This could cause phones to become unbootable if these
apex's are modified.
Bug: 123892969
Test: m droid && boot device.
Change-Id: Ief9004bbe7b106ee8f52715ce5bd7bb5accec290
94427265d1 broke the flattened APEX
by not updating moduleNames
Test: TARGET_FLATTEN_APEX=true m
Files are under /system/apex
Change-Id: I14b1a6f8b2244d35e1accdf9888cfef65f4b0a03
When an APEX is non-installable, the make rules for the APEX files in
the APEX are not emitted as they will never get installed.
androidMkForType() is refactored so that make rules for the APEX files
are created in a separate function androidMkForFiles().
Test: m checkbuild tests
Bug: 123290268
Change-Id: Ibe8817d1e9c6312fb5c6f986dced8aa3e823664a
Build rules for both flattened and non-flattend APEXes are created
regardless of TARGET_FLATTEN_APEX. The selection is made in AndroidMk.
This allows other module to reference an APEX via :module syntax
irrespective of TARGET_FLATTEN_APEX.
Bug: 123780484
Test: TARGET_FLATTEN_APEX=true m out/soong/.intermediates/art/build/apex/art-check-debug-apex-gen/gen/art-check-debug-apex-gen.dummy
with aosp/891696 applied
Change-Id: Ia49415ec3d18cfc5081461be76900c73ea803dca
relative_install_path for cc_library is respected by APEX.
relative_install_path for cc_binary is not yet respected because doing
it will break the path to the dynamic linker in the runtime APEX.
That change should be done along with changes in init, bionic, etc.
Bug: 123721777
Test: m (apex_test.go amended)
Change-Id: I855f8eda0d4255d563861ac96d0d3e2c669e9a2a
apex will normally place libc and some other system libraries into a
separate directory. This is to ensure that early startup works
correctly. Some apex targets want to have these libraries at the
normal places however.
Test: ./art/tools/build_linux_bionic.sh com.android.runtime.host
Bug: 123591866
Bug: 120266448
Change-Id: Ib5a67a43fe3eea6188b4df9215e743f1634045f3
Some apex targets need to be able to change their contents based on if
they are host or target or what libc they are using. This adds support
for doing this using the standard 'target: {...}' idiom.
Test: m com.android.runtime.host
Test: ./art/tools/build_linux_bionic.sh com.android.runtime.host
Bug: 123591866
Change-Id: If73bee650cdeb277c0e603763aa0b0108656bfdd
Some modules rely on symlink_preferred_arch to have expected files
present. This change makes apexs include these symlinks.
Test: m com.android.runtime.debug
pushd $(mktemp -d)
mkdir mnt
unzip $OUT/apex/system/com.android.runtime.debug.apex
sudo mount -o loop,ro apex_payload.img mnt
Ensure that mnt/bin/dalvikvm and mnt/bin/dex2oatd both exist and
are symlinks to mnt/bin/dalvikvm64 and mnt/bin/dex2oatd32
respectively.
Bug: 119942078
Bug: 122373634
Bug: 123079311
Change-Id: I47868fbedc5bdd3141a836c488f79e91e0a6ddfe
This fixes a problem that APEX is unconditionally mutated for all
sanitizer types. This can make an APEX to include sanitized version
of a lib even when SANITIZE_TARGET is not set. It can happen when lib X
is directly depended on by an APEX (e.g. via native_shared_libs) and X
has a sanitized variant due to the dependency from another library Y
which is force sanitized (via 'sanitize' property). In that case,
regardless of lib Y is in the APEX or not, the APEX chooses the
sanitized variant of lib X since the dependency from APEX to the lib is
created with AddFarVariationDependency.
Fixing this problem by mutating the APEX for a sanitizer type only when
the device is requested to be sanitized.
Bug: 122717287
Test: add libnetd_client to com.android.runtime APEX.
Inspect build.ninja to verify that libnetd_client.so in the runtime APEX
is not a sanitized one.
Change-Id: I918bc8407137d74c5456142b3a29de13df68c0b3
If native-bridge/ndk translation is supported for a config the build
system will know about 2 copies of binaries, one for each ABI. We only
want to actually install the 'Native' (non-translated) ABI into the
apex in these situations.
Test: lunch aosp_cf_x86_phone-userdebug
m showcommands com.android.runtime.debug
examine the command output to make sure only a single copy of
dex2oat is being installed.
Change-Id: I9480e52855fd9f5564a579099309fb38f1c50367
Master-art configurations do not have frameworks/base and therefore do
not have hidden API flags. Pass --no-force-assign-all to `hiddenapi`
when frameworks/base does not exist to disable the corresponding
assertion. This enables us to enforce the assertion on non-master-art
builds and also get rid of logspam about missing flags on ART buildbots.
Test: art/tools/buildbot-build.sh on master-art
Bug: 123143676
Change-Id: I074d9554fb11dab3eef904016375730520107ec2
soong_java_prebuilt.mk sets LOCAL_MODULE_SUFFIX := .jar. Therefore, we
need to remove .jar suffix from LOCAL_MODULE_STEM, otherwise, we
will have .jar.jar suffix.
Bug: 123092860
Test: TARGET_FLATTEN_APEX=true apex.test
Check /system/apex/apex.test/javalib/core-oj.jar is there.
Change-Id: Ie19b20523753ec32ab16266f63d26c4df685cc34
This change first introduces the interface "Sanitizeable" that
module types other than cc.Module can use to be handled by the sanitizer
mutator. APEX module, by implementing the interface, gets sanitizer
variants.
In doing so, sanitizer.go is refactored so that modules have explicit
dependencies to the runtime sanitizer libraries. This allows the runtime
library to be packaged into the APEX when required. This also completes
the dependency graph; updating the prebuilt sanitizer runtime will
trigger rebuilding of modules using the runtime.
Bug: 120894259
Bug: 121038155
Test: SANITIZE_TARGET=hwaddress m apex.test
Test: TARGET_FLATTEN_APEX=true SANITIZE_TARGET=address m
Change-Id: Ia91576ff48cda3c996350308b75bf83fcf7c23d7
This reverts commit 769e50b3c9.
Reason for revert: Likely causing build failure on aosp_x86_64-eng in aosp_master:
"build/make/core/base_rules.mk:271: error: art/build/apex: MODULE.TARGET.SHARED_LIBRARIES.com.android.runtime.host.libart-compiler already defined by art/build/apex."
Change-Id: I83b7caa04b2648e4e4914aae2fa5878516634eed
The unstripped file for libraries in APEXes are available via
out/target/product/<device>/symbols/apex/<apex_name>/path_to_lib.
This change make the symbol files available by installing the individual
files in APEXes to the directory where the APEXes will be mounted at
(i.e. runtime directory which is /apex/<apex_name>). Note that the files
are not actually packaged to a filesystem image; they are installed just
to create the symbol files under the out directory where developers can
use them for debugging.
Bug: 120846816
Test: m com.android.runtime.debug
There are unstripped files under
out/target/product/walleye/symbols/apex/com.android.runtime.debug
Change-Id: Ib182e2bf8787b7669ccba13814491db35370f468
When PRODUCT_DEFAULT_DEV_CERTIFICATE is set to /vendor/foo/devkeys/test,
then the public/private key pairs for an apex_key is searched at
/vendor/foo/devkeys directory.
To be specific,
/system/timezone/Android.bp:
apex_key {
name: "timezone.key",
public_key: "com.android.tzdata.avbpubkey",
private_key: "com.android.tzdata.pem",
}
When PRODUCT_DEFAULT_DEV_CERTIFICATE isn't set, the keys are searched at
/system/timezone, which is the path where Android.bp is located.
With PRODUCT_DEFAULT_DEV_CERTIFICATE set to /vendor/foo/devkeys/test,
the keys are searched at /vendor/foo/devkeys.
Bug: 121224311
Test: m (apex_test updated)
Test: m with crosshatch (PRODUCT_DEFAULT_DEV_CERTIFICATE is set to
/vendor/google/...)
Test: m with cheets (PRODUCT_DEFAULT_DEV_CERTIFICATE is set, but there
is no apex key there. The product is with TARGET_FLATTEN_APEX := true)
Change-Id: I213bbb96c433d851f9cc982871459fd7fb4fe47d
This change fixes a bug that when built with TARGET_FLATTEN_APEX=true,
apex manifests are installed without being renamed to
apex_manifest.json.
Test: TARGET_FLATTEN_APEX=true
Test: /system/apex/*/apex_mnifest.json exist
Test: device boots to the UI
Change-Id: Ib8baeb475babbf4faf6cc073e266bb3038ac1b2d
If a lib is directly included in an APEX (via native_shared_libs
property) and the lib has stubs (via stubs.versions property), then the
ordinary non-stubs variant of the library is renamed to
<libname>.bootstrap in the makefile. At the same time, the stubs variant
of the lib becomes visible and it's name is <libname>.
This ensures that modules in Android.mk build against the stubs
variant thus preventing them from using private APIs in the lib.
The non-stubs variant, however, is used if the module explicitly has
set the new 'bootstrap' property to true. This is useful for building
some early binaries (such as init and vold) which need to run before
APEXes are activated. Since they can't use the bionic libs from the
runtime APEX, they should use the bionic libs left in the system
partition which is called the boostrap bionic.
Bug: 120266448
Test: m
Test: m with https://android-review.googlesource.com/c/platform/bionic/+/849044
Change-Id: I882b8aeb5b29460f07b4424e4f8eb844d6c9a9b0
It is a list of <module_name>:<manifest_name> pairs. When the module
name of an APK or an APEX matches with <module_name>, then its app
manifest name is overridden to <manifest_name>.
<module_name> and <manifest_name> can be patterns as in
com.android.%:com.mycompany.android.%.release
Note that, in case of APEXes, the manifest name refers to the name of
the zip container. The apex manifest name (which is specified in
apex_manifest.json) is not overridden.
Test: m with PRODUCT_MANIFEST_PACKAGE_NAME_OVERRIDES for
1) an APK in Android.mk
2) an APK in ANdroid.bp
3) an APEX
and check that manifest names are modified as specified
Change-Id: Ie58882d90884695e893944c43d9c8803b283e93d
Executables under /system/bin are configured as (uid:gid)=(root:shell)
by fs_config.cpp. Therefore, an executable that is moved/copied from
/system/bin to an APEX should be configured as such.
Test: adb shell ls -al /system/bin/linker
/apex/com.android.runtime/bin/linker shows
-rwxr-xr-x 1 root shell 1133528 1970-01-01 09:00 /apex/com.android.runtime/bin/linker
-rwxr-xr-x 1 root shell 1133528 1970-01-01 09:00 /system/bin/linker
Change-Id: Ibb698aab237362fed312da2af809cceead8a1092
When an apex key is marked as 'installable: false' and the build is
debuggable, the pubic key file for the apex key is bundled with the APEX
that is signed with the key.
This eliminates the need to install the public keys for the
testing-purpose APEX in the system partition.
Bug: 122047804
Test: m
Change-Id: Ifa5914891463dbf4c21484ea440836521b2f90b1
Paul Duffin