Commit I66b57de91042bfd56ba54a3659843d8cf7873955 added the support
for different boot partition sizes but didn't consider the case of
building boot-test-harness.img. This CL fixes this by allowing
'boot-test-harness.img' as a boot image variant.
Bug: 181618515
Test: `m bootimage_test_harness` pass
Change-Id: I190d3bf26725ac54999758efbca66a94c5cd5c86
For cases with a my_register_name, move into $(call license-metadata-rule).
Bug: 68860345
Bug: 151177513
Bug: 151953481
Test: m nothing
Change-Id: Ib6c610960347db84503831b1e1519622696d4500
Merged-in: Ib6c610960347db84503831b1e1519622696d4500
Remove extra dot from depreceation and obsoletion warnings to be consistent.
Changes
x: warning: BOARD_PLAT_PUBLIC_SEPOLICY_DIR has been deprecated. Use SYSTEM_EXT_PUBLIC_SEPOLICY_DIRS instead..
to
x: warning: BOARD_PLAT_PUBLIC_SEPOLICY_DIR has been deprecated. Use SYSTEM_EXT_PUBLIC_SEPOLICY_DIRS instead.
Test: m
Change-Id: I60a4258e549570e261252cffee69e5640f62295e
Signed-off-by: Timi <timi.rautamaki@gmail.com>
* changes:
Keep the first and last snapshot of variables.
Generate GenericConfig objects from MakeConfig objects.
Emit and parse the product config variables from kati/make
Add class to fork and exec kati, based on the commandline option given.
Add a CSV parser to parse the output from kati.
Take the unwanted JNI lib removal action out of JNI uncompression
function so that they work independently when building prebuilt apks.
Fixes: 180475879
Test: Manually with a test apk.
Change-Id: I7392d0bf4ca7db76f27e883df08f7cb256162440
This to allow operating on the complete ART_APEX_JARS list from make in
a follow-up CL.
Test: m nothing
Test: m nothing EMMA_INSTRUMENT=true EMMA_INSTRUMENT_FRAMEWORK=true
Bug: 180325915
Change-Id: Ic08148edf25738f6f4769e5359a573237a38b0e9
* changes:
Set BUILD_BROKEN_VENDOR_PROPERTY_NAMESPACE for goldfish
Add BUILD_BROKEN_VENDOR_PROPERTY_NAMESPACE to BoardConfig
Add PRODUCT_SHIPPING_API_LEVEL to productVariables
TARGET_BUILD_APPS.
Test: Run build_mainline_modules.sh and check that
SOONG_CONFIG_art_module_source_build is true even if the fallback
default is false.
Bug: 172480615
Change-Id: Ie0e9566d4d09b1e54af00c06021aa355adcf0a1a
* Default value will be empty and not work in -header-filter.
Bug: 179530304
Test: make with WITH_TIDY=1 and various setting of DEFAULT_TIDY_HEADER_DIRS
Change-Id: I3f784f545acd12285a5bf6dcae1bab42c9185fe7
Library order is important because it is used to construct class loader
context, which is then written into OAT/ODEX files and chacked against
class loader context constructed by PackageManager on the device. If the
orders are different, dexpreopted code is rejected.
This CL fixes a few problems that caused reordering:
- 'filter' function arguments are swapped so that patterns list comes
first, and the library list second
- JSON representation of class loader context is changed to avoid
unmarshaling it to Go maps, which may reorder keys
- library list is no longer sorted (it's unclear why it was sorted)
Bug: 132357300
Test: lunch cf_x86_64_phone-userdebug && m && launch_cvd \
adb wait-for-device && adb root && adb logcat \
| grep -E 'ClassLoaderContext [a-z ]+ mismatch'
# empty grep output, no errors
Change-Id: Ie76996d497e60da0948f1879d6db589ff3e968a2
This adds the configuration necessary to allow the cts_root
test suite to be built.
Bug: 162933947
Test: m cts_root
Merged-In: I4732bf82d3cedc77e30d7fa5eff285d382992853
Change-Id: I4732bf82d3cedc77e30d7fa5eff285d382992853
Jarjar doesn't exit with a nonzero return code when there is a syntax
error in a rules file and doesn't write the output file. Without a
nonzero return code ninja prints a warning but continues, which leads
to stale results on incremental builds where the output file already
exists, or delayed errors on clean builds whre the output file didn't
exist.
Delete the output file before running jarjar, and then check if it
exists after running jarjar.
Fixes: 119516143
Test: m out/target/common/obj/JAVA_LIBRARIES/noto-emoji-compat-java_intermediates/classes-jarjar.jar
Change-Id: Ib99167403948ee11f039dd4d2d7e63052bb92136
Use_lmkd_stats_log used to control LMKD_LOG_STATS when compiling lmkd.
However, for newer Android releases, we mandate that lmkd informs statsd of process kills (change id I8cb123b9488fbc6e88863c2f0e75f1422bcd282e)
After aosp/1555318 we compile lmkd with statsd by default, so this setting is a noop.
Change-Id: I91bd08198fe2275fb8aca77ae5ea637df360d416
Test: build
Bug: 177985094
repack_bootimg is useful to add files into a ramdisk, then repack
a boot.img. Note that mkbootimg, unpack_bootimg, lz4, minigzip, etc.,
that repack_bootimg depends are already in otatools.zip.
Bug: 174443192
Test: `m otatools-package`, unzip $OUT/otatools.zip, export $PATH
with ./otatools/bin and checks unpack_bootimg can run
Change-Id: Id4fbf2076b50522b95ab46dfeb1e2b7115d6606d
BOARD_<partition>_KERNEL_MODULES_BLOCKLIST_FILE<kernel_version> checks
the syntax of and installs a prebuilt modules.blocklist file.
Example usage:
```
BOARD_VENDOR_KERNEL_MODULES_BLOCKLIST_FILE := \
$(wildcard <path to prebuilt modules.blocklist>)
```
Then the modules blocklist would be install to
(vendor|vendor_dlkm)/lib/modules/modules.blocklist.
Bug: 175165656
Test: Create a blocklist with incorrect syntax, verify build fails
Test: Verify the new macro indeed installs the blocklist file
Change-Id: Icb75aae1bdb45546792f28f4c4e8d06d9439391a
target_files.zip passes a list of files to soong_zip in a file list
file, which soong_zip interprets as a glob. Adding a file called "["
causes soong_zip to error with "syntax error in pattern". Use an RSP
file instead, which after I1de7829c668538204fc4a2ac793a4f88dbc886c9
does not interpret entries as globs.
Bug: 179297188
Test: m target-files-package
soong_zip: don't glob RSP file entries
File arguments are considered globs, but RSP file entries should be
considered filenames without globbing. Escape RSP file entries so
they don't get treated as globs later.
Bug: 179297188
Test: zip_test.go
Change-Id: I0736f2d1802114217cba03ee32b67479c9d98da8
builds.
This CL is a nop since the default is still true, but it gives us a way
to toggle it without affecting coverage builds.
Test: env NATIVE_COVERAGE=true build/soong/soong_ui.bash \
--dumpvar-mode SOONG_CONFIG_art_module_source_build
env CLANG_COVERAGE=true build/soong/soong_ui.bash \
--dumpvar-mode SOONG_CONFIG_art_module_source_build
env NATIVE_COVERAGE=true CLANG_COVERAGE=true \
build/soong/soong_ui.bash \
--dumpvar-mode SOONG_CONFIG_art_module_source_build
check that all these print "true" when the default in the "else"
clause is "false".
Bug: 172480615
Change-Id: I648ed911773c59a5b2a117dbf3ddcb2ef338830d
Filter LOCAL_OPTIONAL_USES_LIBRARIES by PRODUCT_PACKAGES. This has the
drawback that some present libraries may be filtered out as well, as the
full list of product packages is unknown until all Android.mk files have
been read.
This CL unblocks <uses-library> fixes for individual apps, which will be
necessary to dexpreopt them in the nearest future when the &-classpath
hack is removed in ART. Apps that are not fixed by then will loose
dexpreopt completely. This fix, although not perfect, lowers the risk of
loosing dexpreopt for many apps. In the future it may be replaced with a
better solution.
Bug: 132357300
Test: lunch aosp_cf_x86_64_phone-userdebug && m && launch_cvd \
adb wait-for-device && adb root && adb logcat \
| grep -E 'ClassLoaderContext [a-z ]+ mismatch'
# empty grep output, no errors
Change-Id: I188cf240259f7a448623450e6a949e0784a21f87
Bowgo Tsai