Building all java to dex is very slow. Make javac checkbuilds act
like jack checkbuilds and only build to class files for unused
modules, and add a java-dex target that can be used to build
everything to dex.
Test: m -j checkbuild
Change-Id: Idcf9c9e0223b9dd88aaa53beb0cb50a230034b0f
Allow working around Jack bugs by using javac to compile *.java
to class files, and then pass through Jack for desugaring,
proguarding, and dexing.
Also make ANDROID_COMPILE_WITH_JACK=false override LOCAL_JACK_ENABLED.
Bug: 37483961
Test: m -j checkbuild
Test: Add LOCAL_JACK_ENABLED := javac_frontend to a target
Change-Id: I2189055aac9deccc36b8eab8e949796dddadf7a1
And by default 'speed' compile those apps at build time.
bug:33799337
Test: m -j32
(cherry picked from commit 9af483488a)
Change-Id: I66f1c675369fccc03b6529c08cb030264d15ba1c
Ensure aar/res exists, otherwise aapt2 may fail with an error.
Bug: 33381544
Test: Attempt to build aar prebuilt
Change-Id: Ifbeed546fe1a3a48c3dae33f95b04acc559f5b30
This path needs to be set if we're going to build aar prebuilts with
aapt2.
Bug: 33381544
Test: Attempt to build an aar prebuilt with LOCAL_USE_AAPT2 := true, it
doesn't fail writing to /
Change-Id: I28c85736986221607a1d7a295ff1522b993a2049
Test: make with corresponding configurations in BoardConfig.mk
- dtbo.img is copied to the OUT directory
- avbtool info_image --image dtbo.img shows avb hashes
- avbtool info_image --image vbmeta.img shows dtbo partition info
Bug: 37730955
Change-Id: If3edaab860926c743f18fd031c911b076eb09fab
Soong already uses it, just not as a prebuilt module.
Test: Compare soong_to_convert.txt before/after
Change-Id: I6ffa0d967635d52b7b4f8b2b0465cd50254f22b9
Updates the test module build rules to also copy over any
extra test configs that match the naming convention of
LOCAL_MODULE_*.config
Bug: 37315907
Test: forrest test HelloWorldTests_HalloWelt
Change-Id: Ie476285fe780cb19264f1fb0294008b21893a4ed
JNI libs used by bundled APKs have been installed outside of the APK
(e.g. /system/lib and /vendor/lib) to save space because multiple APKs
may share the same set of JNI libs.
However, this causes problem for treble-ized devices where access to
/vendor/lib is prohibited for system processes including Java apps.
So, the JNI libs are embedded into the APK when the APK is installed to
/vendor partition and the device is fully treble-ized.
Bug: 37491282
Test: Move an APK and its JNI libraries to vendor partition by setting
LOCAL_PROPRIETARY_MODULE := true and build:
The jni lib does not exist under /system or /vendor. unzipping
the apk reveals the jni lib under ./lib/<abi> directory.
Change-Id: I8389761dfbd28d181869d79f0bf988a1e4df2fca
This is what intermediates-dir-for does (though it also auto-detects
some common classes, which we don't)
Test: build-aosp_arm64.ninja is identical
Change-Id: I4a3efa74e060d82658b856e68318cd92403a6149
64-bit only apps should only be enabled if the target supports 64-bit.
Test: lunch aosp_x86_arm-userdebug; m -j PMTest_Java64
Change-Id: Ibdc863730e97b1a89efb44840a1c633e9951a1d1
This CL disables CFI if both CFI and ASAN flags are enabled. This
allows ASAN to take precedence where needed, preventing build errors
that would otherwise arise.
Bug: 30227045
Test: SANITIZE_TARGET="address" m -j40
Change-Id: I9073ace0a10eb554d14e418a9b23cc8a8277607d
libprotobuf-cpp-* libs are VNDK, However, since the '.vendor' prefix is
added before libprotobuf-cpp-* libs are added to my-*-libraries by
LOCAL_PROTOC_OPTIMIZE_TYPE, the clients are still linked against to the
platform version of the protobuf libs. As a result, link type error
occurs.
In order to avoid this, '.vendor' prefix is appended after
my-*-*libraries are all updated.
Test: normal build successfull
Test: add vendor_provided:true to libprotobuf-cpp-* and build successful
with BOARD_VNDK_VERSION=true
Change-Id: I1ed7978f03e0db56b4cbee86e286c0787f7a3afc
All the new features are turned off for now, since multiple branches and
products need to be verified before they can be turned on. So everything
should behave the same as today, except for no partition-based
warnings.
Instead of the current link type checks that happen during the build,
run as many as possible immediately after loading all the Android.mk
files. If we're allowing missing dependencies ('mm',
ALLOW_MISSING_DEPENDENCIES, tapas, etc), we'll defer the link type
checks to during the build. If we're not allowing missing dependencies,
we'll produce a better error message to the user about the missing
dependencies.
See core/main.mk for a description of the storage format.
This also remove the partition-based type checking. It hasn't worked all
that well, particularly with ASAN builds. The new VNDK checks will
handle the most pressing cases.
Test: Verify all link_type files and dependencies are the same:
grep link_type: out/build-aosp_arm64.ninja | sed -E "s/ rule[0-9]+//" | sort
Change-Id: Id643658b9d9e84f99f5db0d526aad88c1f5d3417
For A/B OTAs, ota_from_target_files.py calls brillo_update_payload
script to generate the payload entry in an A/B OTA package.
Bug: 37443093
Test: lunch aosp_marlin-userdebug && rm
out/host/linux-x86/bin/brillo_update_payload && m otapackage
Change-Id: I3845786b6c934c709f32e12a63cd6ea1f303d380
Update the fuzz_test makefile to link against the new libFuzzer
library and update the coverage flags to the new version.
Test: built a fuzzer and verified correct linking.
Change-Id: Ie6e6da9a202a258abf805dbdc8cdd3490fcb2406
Add soong_javac_wrapper to all javac commands. soong_javac_wrapper
colorizes javac output, and hides noisy messages.
Previously attempted in I7e8fcd2e4e1ed3ff530a8ccfe931ceb7e411e0ad
and I449a61ed2c95b3e1afd51a31a11baab15d97e2f2.
Test: m -j ANDROID_COMPILE_WITH_JACK=false
Change-Id: I9d7a20d766fadab323a4cf95e5734741b3f578bb
The build system currently uses -fPIC for position independent
libraries, and -fpie for position independent executables. However,
these levels are incompatible and conflict when an executable includes
a static lib with the -fPIC flag - such as when building with
LTO. This CL changes the flag for executables to the compatible -fPIE
instead, which is equivalent for all architectures except PowerPC.
Bug: 30227045
Test: ENABLE_CFI=true m -j40 checkbuild # builds and boots
Change-Id: Ieb20bc2528300d88ff05e8787c5840ab6f03fb38
Uses avbtool to sign vendor.img if BOARD_AVB_ENABLE is set.
It also allows appending additional arguments to avbtool via
BOARD_AVB_VENDOR_ADD_HASHTREE_FOOTER_ARGS.
e.g.,
BOARD_AVB_ENABLE := true
BOARD_AVB_VENDOR_ADD_HASHTREE_FOOTER_ARGS := --generate_fec
Bug: 35415839
Test: "make" with the above variables and use avbtool to check vbmeta is
appended to vendor.img
Test: "make dist" with the above variables
Change-Id: I8ada38dff3def6d34613e77c67944def8a49f464
If PRIVATE_PREFIX is unset transform-prebuilt-to-target references
$(DISPLAY) instead of $(HOST_DISPLAY) or $(TARGET_DISPLAY), which
causes unnecessary kati regenerations when the DISPLAY environment
variable changes.
Bug: 36901093
Bug: 37333696
Test: grep for $DISPLAY in out/build-${TARGET_PRODUCT}.ninja
Change-Id: Ia9802fc2b07da9b74e8f80dfdcef121408cd9817
Prebuilts may already be stripped. Instead of requiring every stripped
prebuilt to be annotated with LOCAL_STRIP_MODULE := false, just make
the default (mini-debug-info) pass through already-stripped modules
without failing.
Bug: 36793128
Test: m -j checkbuild
Change-Id: Iafc4614ceb91bbbb2e5ff82caefd18753f8961b1
The codebase is growing more uses of annotation processors,
through a mixture of LOCAL_ADDITIONAL_DEPENDENCIES and
LOCAL_JACK_FLAGS. Add LOCAL_ANNOTATION_PROCESSORS and handle
the rest in the build system.
For now we also need LOCAL_ANNOTATION_PROCESSOR_CLASSES to
work around bad behavior in grok (b/25860419).
Test: m -j java
Change-Id: I465b021e65d25ed3ea517333a82ec4ebd63fc038
Soong modules may be split into both /system and /vendor variants.
Ensure that Make libraries link against the correct version based on
whether LOCAL_USE_VNDK is set.
Bug: 36426473
Bug: 36079834
Test: build-aosp_arm64.ninja files are the same before/after
Test: attempt building with BOARD_VNDK_VERSION:=current
Change-Id: I229bc290373743406275f3ca16081eae04c27987
This will allow individual modules to enable error prone checks
as needed so that the change to treat an error prone check as a
compiler error can be done atomically with the changes to fix
the issues that check reveals.
Tested with the following steps, cleaning up the targets in
between to ensure that it was rebuilt everytime.
Set LOCAL_ERROR_PRONE_FLAGS := -Xep:ClassCanBeStatic:WARN
compiler error in junit-params-test target. Built
javac-check-junit-params-test with RUN_ERROR_PRONE=true to make
sure that compilation failed.
Built the javac-check-junit-params-test without RUN_ERROR_PRONE
set to make sure that the flag is not included unless explicitly
requested.
Built the junit-params-test as normal to make sure it still
worked properly.
Bug: 31507496
Test: See above
Change-Id: I88b5800dacd0e30dd9dcfa5f42e89d1428957365
Merged-In: I88b5800dacd0e30dd9dcfa5f42e89d1428957365
(cherry picked from commit 3ca9298d99)
Instead of using the NDK headers and libraries, add LL-NDK specific
headers and library stubs for VNDK users. This allows us to provide an
expanded liblog interface.
Test: aosp_arm; m -j
Test: Enable BOARD_VNDK_VERSION on aosp_arm; m -j
Change-Id: I0197f44c91218c73b9567a05320c91a2baaae39b
I6730e2d3ec38004874265b2a690442dec57b33f4 introduced duplicate rules
to build classes.jar for static java library prebuilts, and left rules
to copy $(built_dex_intermediates) to $(built_dex) for modules that
don't build dex files.
Move prebuilts LOCAL_BUILT_MODULE back to javalib.jar to match
non-prebuilt modules, and remove the manual building of javalib.jar.
Add a check around $(built_dex_intermediates) for static java libraries.
Bug: 36901093
Test: m -j ANDROID_COMPILE_WITH_JACK=false
Change-Id: I291510b56162f0d35553e3c7ccd59b1382224d62
Jack can silently handle using a java library as a static java
library by merging in the java library dex file. This causes
problems when switching to javac, because dex doesn't support
dex merging with multidex enabled?
Make the output files consistent between java libraries and static
java libraries. Java libraries will now produce:
classes-pre-proguard.jar: the classes before proguard processing
classes.jar: the final jar file containing classes before dexing
javalib.jar: a jar containing classes.dex
Static java libraries will eventually only produce
classes-pre-proguard.jar and classes.jar. All inter-library
linking is done with classes.jar, so a java library can be
used as a static java library.
There are too many dependencies outside the build system that
expect javalib.jar to exist for static and host java libraries,
so for now continue to build a javalib.jar that is a copy of
classes.jar.
Test: m -j ANDROID_COMPILE_WITH_JACK=false java
Test: m -j java
Bug: 36901093
Change-Id: I6730e2d3ec38004874265b2a690442dec57b33f4
Downstream users of a library always link against javalib.jar, which
is either the dex jar for java libraries or a copy of
proguard.classes.jar for static java libraries. In preparation for
making java libraries a superset of static java libraries, make
classes.jar always be the final class-containing jar for use
downstream, and always create a classes-pre-proguard.jar for
users that need a jar without shrinking or obfuscation.
Also rename the intermediates to be consitent: classes-<tool>.jar
Test: m -j ANDROID_COMPILE_WITH_JACK=false java
Change-Id: I3df8b9a4edcd5db996f1fedc54c8a782d4f36a92
Merged-In: I3df8b9a4edcd5db996f1fedc54c8a782d4f36a92
(cherry picked from commit 6fabefa88e)
This CL addresses the problem that native tests build for multiple
architectures and also build for host.
First it restructures each testcase folder to the following layout:
<testcase>/<testcase>.config
<testcase>/<arch 1>/testcase
<testcase>/<arch 2>/testcase
Secondly it now uses the new soong-zip prefix capability to
separate out host and target testcases in the device-tests &
general-tests zip files.
Bug:36692141
Test: `make device-tests general-tests cts -j` &
ensured all built successfully.
Change-Id: I4e8b084ee56fa8940914911649ff35507d87eb27
This lets Soong pass -isystem or -I as necessary (or potentially, even
other cflags in the future).
This is not available for Android.mk use, nor exposed directly to
Android.bp users.
Test: m -j
Merged-In: Id37d4692d5fbddce467bd777903b20169f44dd6e
Change-Id: Id37d4692d5fbddce467bd777903b20169f44dd6e
There are many references to ndk stub libraries in the third party
notices:
out/soong/ndk/platforms/android-10/arch-arm/usr/lib/libc.so
out/soong/ndk/platforms/android-10/arch-arm/usr/lib/liblog.so
...
These are just stub libraries, and aren't distributed on the device. The
real versions of these libraries will be distributed on the device, and
already have the license information.
So let Soong set LOCAL_NO_NOTICE_FILE to prevent these from appearing.
Bug: 36867708
Test: Diff aosp_arm's system/etc/NOTICE.html.gz file before/after
Change-Id: Ifa3693bde7d3b6b9dc7a83122f9cfa24997fac50
Board-based rule can cause unexpected regression because build-time overlays
are different among products. So each product should be tested with its own
product-based rule considering its build-time overlays before applying
enforcing RRO.
Additionally RRO conversion is mandatory only for overlays of which target is
included in AOSP system image and is critical for CTS/VTS tests with AOSP
system image. So inclusive rule is more suitable instead of exclusive rule
to avoid unexpected regression due to unnecessary RRO conversion.
Note that we still support conversion for all the overlays by specifying
PRODUCT_ENFORCE_RRO_TARGETS as "*".
Test: building succeeded and tested with auto-generated RROs.
Bug: 36231603
Change-Id: I8e1d701d4f78b818c89ef3e7638110105370c5bc
(cherry picked from commit 2a209997e8)
This patch updates the logic to use the PRODUCT_IOT variable instead of
BRILLO.
Bug: 36702887
Test: `make`; Image doesn't have the dev key.
Change-Id: I1751e97d8cdfeba83c7e4720a017a5f4dcfd49da
Test: Temporarily added additional codenames for OPD1 -> O and
OPM1 -> O-MR1. Lunched aosp_sailfish-eng, built build.prop and
checked that ro.build.version.all_codenames was "O". Then
lunched aosp_sailfish-eng-PPR1 and checked that it was
"O,O-MR1,P".
Bug: None
Change-Id: I51189f5b40e3e7f99e2fc08bb5616c51a23e3b19
This CL moves the location of ASAN-ified libraries on disk in the
following manner:
/data/lib* --> /data/asan/system/lib*
/data/vendor/* --> /data/asan/vendor/*
There are a couple of advantages to this, including better isolation
from other components, and more transparent linker renaming and
SELinux policies.
Bug: 36574794
Bug: 36674745
Test: m -j40 && SANITIZE_TARGET="address" m -j40 and the device
boots. All sanitized libraries are correctly located in /data/asan/*.
Change-Id: Ic6ba8e43e31df2ea92b85fd60f572823b6883ba2
For tar-ed ASAN artifacts, enforce ownership by the system uid.
Bug: 36458146
Test: m && m SANITIZE_TARGET=true SANITIZE_TARGET_SYSTEM=true
Change-Id: I190cba3d160f15a89ef74f26e7aaa853a449929f
Don't copy jar files for disabled stages. Instead, set the name of
the output to the name of the input so the next stage will directly
pick up the output of the previous stage.
Test: m -j ANDROID_COMPILE_WITH_JACK=false java
Change-Id: Ib7268cbff7ea7ff2ad2caf994aa145e1b7a12b83
This CL is to generate every static RRO package for its target package
automatically at build-time.
BOARD_ENFORCE_RRO build variable is added to specify whether enforcing
RRO is required or not.
BOARD_ENFORCE_RRO_EXEMPT_SOURCES build variable is added to specify
the module list of which item should be exempt from enforcing RRO.
Test: tested on bullhead and sailfish
Bug: 34097942
Change-Id: I455b2ce34e66c57a540c299b5e378b7c4e78d5b8
(cherry picked from commit 3070610b72)
Add build system support for asan.tar.bz in the system image. This
is triggered by SANITIZE_TARGET_SYSTEM=true.
Bug: 36458146
Test: m && m SANITIZE_TARGET=true SANITIZE_TARGET_SYSTEM=true
Change-Id: I725f99f16a8c9182b1c7ad26580c73d3708f8554
This reverts commit 440079b84b.
Adding the pipe interferes with error codes, which results
in an empty classes-full-debug.jar output file being left
after the failure. The next build sees the output file with
an up-to-date timestamp and doesn't rerun the failing rule.
Bug: 36666657
Change-Id: I6658edb766d8ba3120f88e3d8d6eda6ea6c691da
Add MIN_PLATFORM_VERSION and MAX_PLATFORM_VERSION to track
the range of releases that are expected to be released from
the current branch.
Also simplify version_defaults.mk by moving most of the code
to envsetup.mk.
Test: build/make/tests/envsetup_tests.sh
Change-Id: I4f19c31c267e202f8f5ba1384a8b4385d725f9d7
End-of-line comments cause the variable to be set to a space, which
confuses using ifdef on it.
Test: builds
Change-Id: I7b1eb3b9febff1a5a5fca1e6623b2b36d538dacc
Merged-In: I7b1eb3b9febff1a5a5fca1e6623b2b36d538dacc
This change sets Jack default arguments to enable some of lambda
optimizations, namely: lambda grouping in scope of a single package
and using singletin lambda objects for stateless lambdas.
Test: tested manually
Change-Id: Id2d0c37e3462f1e3137cfef2684cf6e74df6089b
BOARD_VNDK_VERSION controls which version of the VNDK is used for the
build. We only support compiling against the current VNDK, and we don't
support installing old prebuilt VNDK libraries yet, so ensure this
variable is set to "current".
Once we support installing old prebuilt VNDK libraries, we'll also need
to disable building modules that use the VNDK.
Test: build-aosp_arm.ninja is the same before/after
Test: Ensure there are no boards that set BOARD_VNDK_VERSION
Test: Set BOARD_VNDK_VERSION := invalid
Change-Id: Ic26c9f44e356a0734f522b4538cb866d5b901d43
If a module has multiple dependencies, the current regexp for making
the dependencies absolute paths only modifies the first dependency.
Test: build sailfish with module configuration including multiple
dependencies
Change-Id: I8f2b40c8c5ec228aa8f831086bcdd561c8714910
fileslist.go is still disabled by default. To enable, use:
USE_FILESLIST_GO=true m -j
Bug: 36274890
Test: Manual, with m -j
Change-Id: I722e17fa7fb6ba91e2b2c44cd7524d974cbe4a56
This got lost from the previous backport, since systemotherimage wasn't
in AOSP yet.
(The Merged-In is a random Change-Id only on master so that this only
merges up to master)
Test: lunch aosp_marlin-eng; m -j systemotherimage
Change-Id: I92c05f27c92d877f3b83e5c5ae2e76728eb162bd
Merged-In: Iafed66e19bc114552b0261dd299379dbb37873c6
LOCAL_OVERRIDES_MODULES is similar to LOCAL_OVERRIDES_PACKAGES, but is
for executables.
Bug: 36491275
Test: LOCAL_OVERRIDES_MODULES := foo should prevent a target executable
foo from being installed. If foo is not a target executable, error
should be reported.
Change-Id: I558f3d94105f9e319f8413c1290799f1c70ee131
add_img_to_target_files may use these to create the recovery patch, so
add a dependency. They were previously a transitive dependency through
the system image to the standalone recovery patch creation.
Bug: 36575896
Test: rm out/host/linux-x86/bin/imgdiff; m -j target-files-package
Change-Id: I2a43220e94c09393e88e1d9950032f5665a0d2ce
This used to happen in the system image creation, but the target-files
didn't really depend on the system image, so I removed that dependency.
But then we weren't creating the symlink all the time.
This should be safe to run in parallel with the same link step in the
system creation.
Bug: 36540023
Test: m -j target-files-package, ensure vendor symlink is created
without system.img.
Change-Id: I37059fa1ce759a0870dc4c65c5515050992245e2
lunch can now take combos in the form:
$TARGET_PRODUCT
$TARGET_PRODUCT-$TARGET_BUILD_VARIANT
$TARGET_PRODUCT-$TARGET_BUILD_VARIANT-$TARGET_PLATFORM_VERSION
If all 3 are not specified the unspecified ones will take the
default values provided by the build system (eng, and currently
OPR1).
In addition, error handling for invalid products, variants and
versions is moved to the build system.
Bug: 34972208
Test: build/make/tests/envsetup_tests.sh
Change-Id: Ib0aaa98633448ba9bd8df911704c9cb3a8ebbe85
Modules compiled with javac 1.7 may have LOCAL_STATIC_JAVA_LIBRARIES
pointing to jars that contain code compiled with javac 1.8, and may
need desugaring.
Test: m -j ANDROID_COMPILE_WITH_JACK=false checkbuild
Change-Id: I470fc0a2fba51179a3ef95986cb3030194cd8f1c
Replace the current naming scheme for TARGET_PLATFORM_VERSION
(O, ODR1, OMR1, P) with one that matches the user-visible
branch version (OPR1, OPD1, OPM1, PPR1).
Temporarily map P -> PPR1 to decouple this change from updating
the build servers.
Bug: 34972208
Test: m -j TARGET_PLATFORM_VERSION=PPR1
Test: m -j TARGET_PLATFORM_VERSION=P
Change-Id: Ib00b13abe79a744e344b88bb1d800524ab09d585
(cherry picked from commit d8bc190a63)
Replace the current naming scheme for TARGET_PLATFORM_VERSION
(O, ODR1, OMR1, P) with one that matches the user-visible
branch version (OPR1, OPD1, OPM1, PPR1).
Temporarily map P -> PPR1 to decouple this change from updating
the build servers.
Bug: 34972208
Test: m -j TARGET_PLATFORM_VERSION=PPR1
Test: m -j TARGET_PLATFORM_VERSION=P
Change-Id: Ib00b13abe79a744e344b88bb1d800524ab09d585
Merged-In: Ib00b13abe79a744e344b88bb1d800524ab09d585
We don't actually need to build the system image before
target-files-package, we just need to make sure everything is present in
$(PRODUCT_OUT)/system. So switch the dependencies around.
This can't be done as easily for the vendor image, since it may be a
prebuilt, or we may need to install some kernel modules into
$(PRODUCT_OUT)/vendor when building vendor.img. It tends to be smaller
anyway, so the time savings isn't as big.
Test: Make sure the build server outputs the same files
Test: Ensure target-files-package is identical before/after.
Change-Id: I84e4f878491ab2b71582f4f49418dbfbba75590f
And move simple cases of $(ACP) to cp. That does change OTA/bin/updater
from 0640 to 0700, but that's more consistent anyways.
Test: m -j target-files-package; ensure it's the same before/after.
Change-Id: I37b65eb9ddccd2f360aa007f929ece35c9e44f9b
We've just created the target files package and still have the extracted
directory, so pass that to ota_from_target_files so that it doesn't have
to re-extract what we just compressed.
This saves a little bit of time -- for bullhead-userdebug on internal
master, this brings the time to build an OTA from ~340s -> ~310s. Much
of the time is still spent generating and signing the OTA.
Test: lunch bullhead-userdebug; m -j otapackage
Test: bullhead-ota-*.zip is identical before/after
Change-Id: Ib51cf6b02123e9c49280e9957c6590645470a52c
When building target-files-package, dir: META has to be copied first
to help early validation of the .zip file.
In order to prevent $(zip_root)/META from getting copied twice to $@.list,
use the method (find -path -prune -o -print) instead of (grep pattern -v).
Test: m target-files-package
Bug: b/31676493
Change-Id: Id6b4d77fb62aa136253de88573cc51a575b90be2
Static java libraires are designed to be included in another
java library. If desugar runs on the static java library,
the resulting jar will contain the desugared lambda classes.
When the static jar is included into another java library
and desugar runs again, it desugars the static java library
classes again, producing the same desugared lambda classes
and colliding in the final jar.
Test: m -j ANDROID_COMPILE_WITH_JACK=false
Change-Id: I0508b2cc9c56dbbd4ac0aba44fb99ea2a27715e0
DEFAULT_APP_TARGET_SDK is PLATFORM_VERSION_CODENAME (for example 'P')
for non-release builds, and PLATFORM_SDK_VERSION (for example 25) for
release builds. Use it for dx and desugar, and translate
PLATFORM_VERSION_CODENAME to 10000 to enable features in the current
development version.
Bug: 36087246
Bug: 36118520
Test: m -j ANDROID_COMPILE_WITH_JACK=false
Change-Id: I3d6bbc36a8869e4ced6bb850bc32f37b4cba3f85
The vendor image will have more than just proprietary modules in it
under Treble, so let's stop marking open source code as proprietary just
to move it to vendor.
Bug: 36452052
Test: compare build-aosp_arm.ninja before/after, no changes.
Test: Try defining LOCAL_VENDOR_MODULE / LOCAL_PROPRIETARY_MODULE in the
same and different ways, ensure it works.
Change-Id: I0ab046bcb7563a20d44ed00667e18497ef0873a8
This means that we won't have to rebuild the odex files after an
installclean.
Test: profile.prof is identical before/after
Test: m -j; m installclean; m -j
Change-Id: I53c630ce0c5a985d856e26bd7bb3d8c76958aaab
This reverts commit 842a985f90. It's
causing test failures, warnings and complaints, so backing it out and
we'll resolve those before putting it back in.
This also bundles in the revert of a minor warning/error fixup for this
code, and does the correct thing with the CleanSpec.mk (which can't just
be reverted).
Bug: 33681361
Test: m -j
Change-Id: Ic889ce6a4737a497ad6bf633424ce1860713f7d0
Warning messages like, "libart_fake: framework library must be installed
to system/lib64 but requested to be installed at system/fake-libs64.
Please fix.", are temporarily muted since some of those warnings might
be false alarms.
Let's just silence the warning for everybody and investigate the cases
internally by turning on the build flag SHOW_MODULE_PATH_WARNINGS.
Test: the warning messages should not be shown unless built with
SHOW_MODULE_PATH_WARNINGS=true.
Change-Id: I5eb3ab3385c92a5192723ff1008277c321162191
Compute PLATFORM_VERSION and PLATFORM_VERSION_CODENAME from
TARGET_PLATFORM_VERSION. Add explicit values for O, even though
they are the same as the fall backs, as placeholders for future
values.
Bug: 35445510
Test: manual
Merged-In: I30a7142f732c85b637d2ce072373bc867cedcd42
Change-Id: I30a7142f732c85b637d2ce072373bc867cedcd42
Using ; to join commands in a rule causes failures to be ignored by
make. Use && instead, and add true at the end as the second operand
to the last && operator.
Also inline copy-test-in-batch, batching is no longer necessary as
kati will fall back to a shell script if the recipe is too long to
fit in a command line.
Test: builds
Change-Id: I4a2528bf2a15106cfabaae0336662c4a0464271d
/file_contexts.bin is needed only by devices which do not have
PRODUCT_FULL_TREBLE set to true. Adding this file to devices which
have PRODUCT_FULL_TREBLE set to true causes confusion and wastes
about 800 kB of space.
/file_contexts.bin was being added unconditionally to all devices due
to the build system needing the combined file_contexts file for
host-side labelling of filesystems in images and for inclusion into
target files ZIP (again, to be used for host-side labelling later).
This change switches these targets to depend on the same file, but
from intermediates directory. As a result, everything continues to
work just fine, but without pulling in /file_contexts.bin. On devices
which need /file_contexts.bin, it is pulled in by
system/core/init/Android.mk.
Test: /file_contexts.bin still there on bullhead, but not there on
sailfish with PRODUCT_FULL_TREBLE set to true.
Test: Clean build, device boots up with, no new denials.
Reboot to recovery, recovery boots up fine too, no denials.
This was tested on sailfish-eng and marlin-eng.
Bug: 36002414
Change-Id: I9bbbb08bbf7d874bba0bafcc19bcbf9eec564326
This makes recoveryimage/bootimage targets explicitly depend on SELinux
policy artifacts needed by recovery. Prior to this change, the
dependency was contrived: these targets would rely on copying these
files from "root" directory, and then overwriting sepolicy file with
sepolicy.recovery.
Test: Clean build, flash, device boots up just fine. No new denials.
Reboot into recovery, recovery boot up just fine, no denials.
This was tested on bullhead (non A/B device) and sailfish (A/B
device).
Bug: 33642277
Change-Id: Id2369253d0c7e02e079ae01ac7140b41880fd5ad
Shared libraries are now installed to different directories depending on
their types.
* NDK libraries: /system/lib/ndk (with symlink from /system/lib)
* VNDK libraries: /system/lib/vndk
* VNDK-ext libraries: /system/lib/vndk-ext
* Framework-only libraries: /system/lib
* Vendor-only libraries: /vendor/lib
* Same-process HALs: /vendor/lib/sameprocess
However, if LOCAL_MODULE_PATH is explicitly set, then it is respected,
with a warning message. Module owners are highly encouraged to
investigate the warnings and use alternatives to LOCAL_MODULE_PATH;
combination of LOCAL_[PROPRIETARY|OEM|ODM]_MODULE, LOCAL_MODULE_CLASS
and LOCAL_RELATIVE_PATH will cover most of the cases.
Furthermore, for each shared libraries whose path is changed, a symolic
link from the original path to the new path is *temporarily* generated.
e.g. /system/lib/libbase.so -> vndk/libbase.so. This is
to prevent sudden breakage of the code expecting the lib from the old
path. This symbolic links will eventually be removed before O launch
(b/34917183).
Finally, BOARD_SAME_PROCESS_HAL_DEPS is added. It contains the list of
shared libraries implementing the same-process HALs and its internal sub
libraries. This is designed to be defined in BoardConfig.mk
Bug: 33681361
Test: build & run. Libraries must be in the correct directories.
Symlinks from the old path to the new path must exist.
Change-Id: I46130aac09ae65400fd4d0abbc2e12dddd154fb1
This also makes the resulting zip files more repeatable -- the file list
is sorted and the entries have static timestamps.
On my machine, this saves ~30 seconds for android-cts.zip, it now takes
less than a second.
Test: m -j cts; compare output from without this change
Change-Id: Ia71e35878ff98ba9775115860530e87eee47739f
BBOTA v1 and v2 (introduced in L and L MR1 respectively) don't support
resumable OTA. We shouldn't generate packages using v1/v2 at the risk of
bricking devices.
BBOTA v3 (since M) and v4 (since N) both support resumable OTAs. BBOTA
v4 additionally supports using FEC to possibly recover a corrupted
image.
Bug: 33694730
Test: Generate full and incremental OTAs w/ and w/o the CL. They should
give identical packages (in v4).
Change-Id: Ib89d9cd63ba08e8e9aa4131bed18876b89d244c0
Remove the following functions that are needed for file-based OTAs only:
- SetPermissions()
- SetPermissionsRecursive()
- MakeSymlinks()
- DeleteFiles()
- DeleteFilesIfNotMatching()
- RenameFiles()
- SkipNextActionIfTargetExists()
Bug: 35853185
Test: Verified there's no reference to these functions.
Change-Id: Iff24a9d705476211effaef28eed2a147fa5fcbce
For aosp_marlin (on AOSP), this speeds up target files creation from
~57 seconds to ~15 seconds.
For marlin-userdebug (on internal master), this speeds up target files
creation from ~300 seconds to ~95 seconds.
This does make some minor changes to the resulting target_files zip:
* All of the META files are actually at the beginning of the zip now,
previously the ones created by add_img_to_target_files would be at
the end.
* The images are slightly larger. Go's zip implementation in soong_zip
isn't quite as good as the `zip` tool.
* vendor_filesystem_config.txt isn't present if we're not building a
vendor image.
Test: aosp_marlin target_files zip changes look reasonable
Test: marlin target_files zip changes look reasonable
Test: bullhead target_files zip changes look reasonable
Change-Id: Ia5ce6870b85559d65107c3b94332391b4b43a0ea
Otherwise it doesn't work for incremental builds.
Bug: 34929159
Bug: 35997300
Test: repo sync && m
Change-Id: I78ead57cf89a9d0f26dbf4f3a225a9357bdca1b3
Support BOARD_VENDOR_KERNEL_MODULES which accepts a list of the
kernel module binaries. Binaries listed there are copied to
$(TARGET_OUT_VENDOR)/lib/modules and 'depmod' is executed to generate
the dependencies. BOARD_RECOVERY_KERNEL_MODULES is similarly
implemented to provide a way to specify kernel modules to be copied
to the recovery image.
Bug: 33305644
Test: android master build on marlin and load dummy modules
Merged-In: I0a58a270dbb0d2d38fd089f4abff66e62e14fac1
Change-Id: Ia5f429508ca3d7c77448358e42af6efa98fa8624
Colin Cross