This CL gets us closer to having incremental builds for ASAN by
separating the generation of sanitized intermediates from their
non-sanitized counterparts.
Bug: 38145756
Test: time (m -j40 && SANITIZE_TARGET="address" m -j40) # reduces from
17 mins to ~9 mins with these changes.
Change-Id: I7e3e7d88d276d834192b2e2427643f70e89d731e
- explicitly set the path where desugar will dump temporary class files.
This ensures that the system property is set before it is read during
InnerClassLambdaMetafactory.<clinit>. Before this CL, the system
property was set by Desugar.createAndRegisterLambdaDumpDirectory(),
which may run too late.
- explicitly specify -source 1.8 for droiddoc's javadoc run.
Previously, the command used the language version of the build
toolchain, which might fail due the backward-incompatible restrictions
imposed by the proposed module system.
Some Android build targets use LOCAL_JAVA_LANGUAGE_VERSION 1.7,
but droiddoc combines sources from multiple build targets and there
where no backwards incompatible changes in 1.8, so it should be fine
to use 1.8 for code from either language level.
Bug: 38318052
Bug: 38225656
Bug: 38177295
Test: make clean && make checkbuild docs
(using OpenJDK 8 toolchain)
Change-Id: Ie5fe0502dfe2f99fc58b1a723b7c18d5896a2646
This CL changes the build system to always look for shared JNI
libraries at their unsanitized install locations.
Bug: 38309771
Test: m -j40 && SANITIZE_TARGET="address" m -j40
Change-Id: Icb9d4f5365def6ea7a780553f455f41d2cb8b8bf
We no longer need to use $PRODUCT_OUT/obj/lib as a linker path, we can
directly specify the shared libraries to the linker.
So install both into the standard built module path, and declare a copy
rule to the old OVERRIDE_BUILT_MODULE_PATH location. This way we can
incrementally move users from the old location to the standard path.
Test: m -j
Change-Id: I3c6140ac26e5e5ca486337ce0192e15e530da32c
Merged-In: I3c6140ac26e5e5ca486337ce0192e15e530da32c
Remove timestamps from jar files in host java libraries to get
consistent results between builds to help with build artifact
caching.
Bug: 38215808
Test: sha1sum desugar.jar && touch Desugar.java && m -j desugar && sha1sum desugar.jar
Change-Id: I1ecac9b2b80f673937086686540ab0736259df1b
* changes:
Move auto installclean to soong_ui
Move version checking to soong_ui
Move clean/clobber to soong_ui
Allow disabling of CleanSpec functionality
Set CUSTOM_IMAGE_AVB_ENABLE := true to enable avb, add_hashtree_footer
args can be added in CUSTOM_IMAGE_AVB_ADD_HASHTREE_FOOTER_ARGS.
Bug: 38319818
Test: m custom_images
Change-Id: Ia452dc5ce8b55bcbd3abba9e965b72e78fd8c104
This way kati doesn't need to keep state for which build got run last,
and we have to run kati less often. This was forcing another kati run
for an empty out directory, and a kati run (or two) every time you
switched products that shared a device.
Bug: 35970961
Test: m clean; m -j blueprint_tools; m -j blueprint_tools; m -j blueprint_tools
Test: lunch aosp_arm-eng; m -j blueprint_tools; lunch full-eng; m -j blueprint_tools; <repeat>
Change-Id: I825a0868fb7059016a940c76244527432e3e7cff
This removes versions_checked.mk that caused kati to regenerate the
ninja file 3 times for a clean out directory. Kati needs to regenerate
every time that we write a file that we also read. soong_ui doesn't have
this problem.
Bug: 35970961
Test: m clean; m -j blueprint_tools; m -j blueprint_tools; m -j blueprint_tools
Test: $OUT_DIR/versions_checked.mk no longer exists
Change-Id: I14b67dc275ea3daa77a7315c2985dc73d77fd07c
So that we don't have to load up all the makefile state just to remove
the output directory.
Starting from a completely empty out directory:
kati: 16s
soong_ui: 2.0s
From a minimal out directory (m -j blueprint_tools):
kati: 3.8s
soong_ui: 0.4s
Test: m -j clean
Test: m -j clobber
Change-Id: Iec1ce032c7cc7ed102430080e857fc421b66309c
This is only intended for testing purposes -- so that we can run build
system tests and verify that kati only gets run once for each
configuration. CleanSpecs necessarily write state that require us to
re-run kati twice.
Test: export NO_ANDROID_CLEANSPEC=true; rm -rf out; m -j; m -j
Change-Id: I635a8d8b6754dff82fbc7f20c500d06d7cf6c4f2
Previous change that moved com.android.internal.util.Predicate
to the legacy-android-test JAR caused a proguard build breakage
due to duplicate classes in the application and library. This
fixes it in the same way as previous issues with junit and
android.test classes were resolved by excluding those classes
from the application JAR.
Bug: 30188076
Test: make checkbuild and make -j ANDROID_FORCE_JACK_ENABLED=disabled checkbuild
Change-Id: Ica0089607187f970251ddba339a1f71d4cad80cc
This CL ensures that the _asan suffix is added to the build flavor
only when it doesn't already contain _asan (or _asan_coverage). This
correctly prevents it from appending an extra _asan to ASAN lunch
configs that already include it in the flavor, while allowing
sanitized targets for generic configs to behave as they used to.
Bug: 38145756
Test: _asan is correctly appended only for lunch configs that don't
already specify it in the build flavor.
Change-Id: Ia7d9356f717b97d1c64e3237ca31cc507f27734f
This is useful for devices with low disk space with different
build variants.
Bug: 37469715
Test: Regular image builds successfully, errors occur when
the headroom size is greater than available partition space.
Change-Id: I526cdd0f84981bbd16e3afcfe1cd7fc43dce98ef
With LOCAL_USE_AAPT2, resource directories like a/b/../res cause
problems, since ninja will canonicalize the path before creating the
intermediate resource directory, so it creates <intermediates>/a/res
while we give AAPT2 <intermediates>/a/b/../res, which fails to open.
Bug: 37716307
Test: Switch LOCAL_USE_AAPT2:=true for TelecommUnitTests, mma
Test: lunch aosp_marlin-userdebug; m -j
Change-Id: Id0d167e68185a119390e7b7e3c344895e77ca0e3
We shouldn't give non-clean paths to tools -- if a/b/../file was
specified, we can simplify that path to a/file, and not need to create
a/b just to make the path name work.
The testcases come from golang's filepath.Clean tests, this should be
compatible with that implementation.
Bug: 37716307
Test: TEST_MAKE_clean_path=true m -j blueprint_tools
Change-Id: I290a02b0a1e4a7c2b9255bca3c881589b521c402
We have code that acts slightly differently when the static analyzer is
running, so that it can produce more accurate diagnostics (e.g. less
false positives). It uses __clang_analyzer__ to detect the static
analyzer.
When the static analyzer is run via clang-tidy, __clang_analyzer__
doesn't get defined.
Bug: None
Test: WITH_TIDY=1 m. clang-tidy now acts as expected in code made for
the static analyzer
Change-Id: Ib2a815c0bd67553af465b64207bb480fb52cfaf8
Setting TARGET_PLATFORM_VERSION to DEFAULT_PLATFORM_VERSION during
"lunch sailfish-userdebug" causes unnecessary pain when
DEFAULT_PLATFORM_VERSION becomes invalid after branching. Only
set TARGET_PLATFORM_VERSION if it was explicitly requested with
lunch sailfish-userdebug-OPR1.
Test: build/make/tests/envsetup_tests.sh
Bug: 34972208
Bug: 37208937
Change-Id: I40ba4617e73803c5bec1a8d317382f70fb3ec3a2
Merged-In: I40ba4617e73803c5bec1a8d317382f70fb3ec3a2
CL I2189055aac9deccc36b8eab8e949796dddadf7a1 moved the
ANDROID_COMPILE_WITH_JACK=false check to configure_local_jack.mk,
which broke some modules that use ifndef LOCAL_JACK_ENABLED to
change their behavior we jack is disabled. Clear
DEFAULT_JACK_ENABLED when ANDROID_COMPILE_WITH_JACK=false so
that LOCAL_JACK_ENABLED is cleared.
Bug: 37483961
Test: m -j checkbuild
Change-Id: I0d08946c57182689624c57f4952d7cecaa0b82b3
Sort the inputs to remove duplicates, which fixes:
found two file paths to be copied into dest path: "target/testcases/minikin_perftests/minikin_perftests", both ["target/testcases/minikin_perftests/minikin_perftests"]"out/target/product/bullhead/testcases/minikin_perftests/minikin_perftests" and ["target/testcases/minikin_perftests/minikin_perftests"]"out/target/product/bullhead/testcases/minikin_perftests/minikin_perftests"!
Test: m -j device-tests
Change-Id: I8d7e83cd077c62461c9a08e7f1b49c321fbcb0e3
TARGET_USES_HWC2 is going away, but propagate it to soong for now
to unblock converting dependency chains that end in libhwui.
Test: soong tests
Change-Id: Ie436548a60a08a487b1d85ff55706b824b297833
Bug: 37626838
Since Lollipop, RS generated bitcode is embedded as strings in
generated Java files, and no longer needed in the resources.
This CL makes packaging resources conditional on RS target API, and
not do so unless it is below 21.
Test: mm in frameworks/rs/tests/java_api/Refocus, ImageProcessing2, and
ImageProcessing_jb
Change-Id: I79a90ed4b96cb78b22a64a35b539d9d67351f4c4
(cherry picked from commit e2c0901918)
