On devices with prebuilt generic boot image:
TARGET_NO_KERNEL := true
BOARD_PREBUILT_BOOTIMAGE is set
In this case, the process for extracting the kernel
config and version must also try extracting from the
prebuilt boot image.
Test: build on device with prebuilt boot image
Fixes: 193588301
Change-Id: I43dd8e206bcd1c9f3f5f51869f126791c50af01a
Bug: 188179858
Test: compare dexpreopt_config.zip files from
1. TARGET_BUILD_UNBUNDLED_IMAGE=true m dexpreopt_config_zip
2. m dexpreopt_config_zip
(note that m clean should run between steps)
Change-Id: I92623b59e4fcac397c7fed3d7ab1070fe5281027
This is a workaround for the problem that we cannot control from make
which APEX gets picked for deapexing.
Cherry-picked from http://ag/15107826.
Test: env TARGET_PRODUCT=aosp_cf_x86_64_phone \
SOONG_CONFIG_art_module_source_build=false \
build/soong/soong_ui.bash \
--dumpvar-mode SOONG_CONFIG_art_module_source_build
This shows true
Test: env TARGET_PRODUCT=cf_x86_64_phone \
SOONG_CONFIG_art_module_source_build=false \
build/soong/soong_ui.bash \
--dumpvar-mode SOONG_CONFIG_art_module_source_build
This shows false in internal
Bug: 192006406
Bug: 172480615
Change-Id: I8990b726affeb08eabf130d20be45129a7c1c58d
Merged-In: I8990b726affeb08eabf130d20be45129a7c1c58d
(1) Incorrect path to librsjni.so because of typo in earlier change;
fixed by fixing typo.
(2) clang did not specify the target architecture (I don't know what
changed so that this no longer works); fixed by specifying target
architecture and changing from gcc linker to lld (gcc linker didn't
understand the options it was being passed by clang).
Test: blueline-userdebug
Test: make -j FORCE_BUILD_RS_COMPAT=true ENABLE_RSTESTS=true \
RSTestForward_21_1_2 RSTestForward_22_0_2 \
RSTestForward_23_0_3 RSTestForward_24_0_3 RSTestForward_25_0_2 \
RSTestBackward RSTestBackward19 RSTest RSTest_Compat RSTest_Compat19
Test: unzip -t ${PATH_TO_APK}
# confirm that .so files are present in RSTest_Compat[19] but not
for the other APKs
# confirm that .bc files are present in RSTestBackward19 but not
for the other APKs
Test: adb shell am instrument [--abi armeabi-v7a] \
-w ${TEST_PACKAGE_NAME}/androidx.test.runner.AndroidJUnitRunner
# Run four times:
# "--abi" argument present or absent; always omit for
# RSTest_Compat[19], which only support platform default ABI
# "adb shell setprop debug.rs.forcecompat 0" or "1"
# Note that RSTestBackward19 only supports 32-bit ABI, so both
# runs (with and without "--abi" argument) are actually armeabi-v7a
Test: After run, check logcat for RenderScript_jni
# (missing): not built against support library; expected for all
tests except RSTest_Compat[19]
# "RenderScript_jni: RS native mode"; expected for RSTest_Compat[19]
when setprop debug.rs.forcecompat is 0
# "RenderScript_jni: RS compat mode"; expected for RSTest_Compat[19]
when setprop debug.rs.forcecompat is 1
Change-Id: Icfaa226f2e1bfa909cb3bcbbe960f2804c355d2a
At least until the device builds are largely disabled and host builds
become opt-in rather than opt-out. Otherwise every host test that
depends on a linux-only library breaks the mac build, even if it won't
be used.
Bug: 187222815
Change-Id: Ic598f2b66a949b90e5b3fbebf463f28278e4834f
But only for a TARGET_BUILD_UNBUNDLED_IMAGE build. Since
the unbundled image build attempts to remove as many
framework sources as possible, we'd like to remove the
sources and dependencies of some blueprint_tools as
well.
Test: m nothing dist
Test: Verify graph is unchanged
Test: m nothing dist TARGET_BUILD_UNBUNDLED_IMAGE=true
Test: Verify that graph no longer includes blueprint_tools
Bug: 192629969
Change-Id: Ibfb52b841379e20855d0a935a13ad047af2fe58d
When set, module builds will prefer source where applicable.
Bug: 191978129
Test: m; MODULE_BUILD_FROM_SOURCE=true m
Change-Id: Ie7f928674a511a98f17653fa814a42194ee0e9f9
Bug: 158843648
Test: check if dexpreopt config for the module defined in mk has
DexPreoptImageLocationsOnDevice field.
Change-Id: Ie8c7b1c8c5a6797f71920d9ce671dde0e1f489a3
This reverts commit 2e43e34d53.
The kernel bootconfig feature has been updated to handle mixed subkeys
and values, so androidboot.hardware is ok to use now.
Test: build and boot cuttlefish with 'hardware=cutf_vm'
Test: build and boot cuttlefish with 'androidboot.hardware=cutf_vm'
Bug: 191502832
Change-Id: I7c3f330a73abe6cf1fb7d14aaaa6cbc3e874a4aa
* changes:
Add BUILD_USES_RECOVERY_AS_BOOT to soong config
Directly create ramdisk dirs in ramdisk image rule
Add BOARD_BUILD_SYSTEM_ROOT_IMAGE to config vars
The droidcore-unbundled target replaces the droidcore
target when TARGET_BUILD_UNBUNDLED_IMAGE=true. This
target is similar to droidcore in terms of dependencies,
but it has a smaller set of dist files. The intention is
to use TARGET_BUILD_UNBUNDLED_IMAGE=true with a
VSDK-using vendor build to supply Java dependencies from
prebuilts to avoid building (or needing) Java framework
sources.
Test: build with and without TARGET_BUILD_UNBUNDLED_IMAGE=true
Test: m nothing dist (before patch)
Test: m nothing dist (after patch)
Test: m nothing dist TARGET_BUILD_UNBUNDLED_IMAGE=true (after patch)
Test: examine ninja build graph for previous 3 tests
Bug: 188176942
Bug: 187064252
Merged-In: Ia7f8200c1bb1aaaa57c177d1698a114efe26bec7
Change-Id: Ia7f8200c1bb1aaaa57c177d1698a114efe26bec7
These directories have been created with post install cmds of
init_first_stage. To migrate init_first_stage to Soong, the directory
rules are now written directly in ramdisk image rule.
Bug: 187196593
Test: "m installclean; m" and see ramdisk output
Change-Id: Ic76c325ce102347f20b282572e3edbb5b4359aaf
