When creating ext4 images with mke2fs, it may skip allocating some
blocks if they contain all zeros. As a result, there could be less
blocks listed in the block map than the actual file length.
For example, for a file with a length of 112200-byte (27+ blocks),
the listed blocks in block.map could be '43665-43688' (24 blocks).
Because some all-zero blocks are not taking actual space.
The generated ext4 images are perfectly valid - kernel will figure out
that data block is not allocated and writes all zeros into user buffer.
However, we can't fully reconstruct a file from its block list in our
Python script. Ideally this can be avoided by mounting or parsing an
ext4 image directly, which is yet to be supported in our script.
This CL skips checking for such files to avoid failing
validate_target_files.py.
Bug: 65213616
Test: validate_target_files.py passes on targets with mke2fs generated
images (e.g. marlin).
Change-Id: Id9cc59e345b9283844044ef94ceb5702f0ca0526
(cherry picked from commit b418c30e3a)
PRODUCT_ENFORCE_RRO_EXCLUDED_OVERLAYS will be used to specify overlays
to be excluded from enforcing RRO.
The excluded overlays will be applied into the original package.
Bug: 63600240
Bug: 65001751
Test: succeeded building with
PRODUCT_ENFORCE_RRO_EXCLUDED_OVERLAYS := \
vendor/google/google_overlay/static_only \
vendor/google/nexus_overlay/common/static_only
and confirmed that config_webview_packages existed in framework-res.apk
with the overlayed value, but not in framework-res__auto_generated_rro.apk.
Merged-In: I0bfb44fc7726710bb78d9100404bc6dd29d06a73
Change-Id: I0bfb44fc7726710bb78d9100404bc6dd29d06a73
(cherry picked from commit ccee95e6ec)
GSI inherited from aosp_base_telephony.mk, but it included some settings for
emulator. So this patch changes to inherit from more pure makefiles to
includes AOSP and telephony modules.
Bug: 65332895
Test: Test on some Pixel devices with GSI, boot to home screen
Change-Id: I98523e864ccf06ddba4422bc8380bc6ff0095d5a
To support test automation, disable adb authorization in GSI.
Applied the patch, and device won't ask for adb authorization after
overriding system image with GSI.
Bug: 64489353
Test: boot up with GSI, and check by 'adb devices'
Change-Id: Ibaa3aaa90a89ae03b8b23abf37f53ede64d15789
Update location to framework/base/config instead of frameworks/base
Fixed dex2oat dependency to my_out_boot_image_profile_location
instead of my_out_profile_location.
Bug: 37966211
Test: make and flash
Change-Id: I446f11216cf6342861bbbf56ac5c4da2f36fb755
Use c_m.<level>.xml for matrices which correspond
to released API levels.
Test: manual
Test: build system complains when files are missing
Bug: 63603463
Change-Id: I16bcab9a7548876ec2bedf9c5d70a126ac50b8d5
System Server starts "TimeZoneRulesManagerService" if
"config_enableUpdateableTimeZoneRules" is true in config.xml.
Some projects would turn on "config_enableUpdateableTimeZoneRules"
by resource overlaying in vendor partition.
So, GSI needs to include all required modules for
TimeZoneRulesManagerService.
Bug: 65001751
Test: Build pass and boot with GSI into home screen
Change-Id: Ief929b23411c4fb85633696f47aabd0f065aab91
Renamed WITH_DEXPREOPT_BOOT_IMG_ONLY to
WITH_DEXPREOPT_BOOT_IMG_AND_SYSTEM_SERVER_ONLY and changed the
behavior accordingly.
Preopt system server jars since selinux prevents system server from
loading anything from /data. If we don't do this they will need to
be extracted which is not favorable for RAM usage or performance.
Test: make and flash and look at system server maps
Bug: 65122284
Bug: 62356545
Change-Id: I49f2a23337fdd2bc2e628b683d26a0c135b2aabe
2ND CPU variant is set to generic as it already is for its non-AB
counterpart.
Bug: 65042524
Test: No regression on CTS with arm64_ab GSI
Change-Id: I513444178e663b154d6ccfae763c59a1091b163e
Configstore was up-revisioned from 1.0 to 1.1. However no new
APIs were added to 1.1. Therefore configstore in mr1 will be down
revisioned to 1.0.
Bug: 64999074
Test: build & run
Change-Id: Ife60328110c6d6cb3e8024fb4a5caabc4c7c034f
When BOARD_PROPERTY_OVERRIDES_SPLIT_ENABLED is enabled, the build system
may attempt to create a symlink in TARGET_ROOT_OUT before TARGET_ROOT_OUT
is created. Also make the symbolic link depend on
INSTALLED_DEFAULT_PROP_TARGET so that the symbolic link is properly created
when only bootimage is made.
Bug: 63645797
Test: on non-A/B device, make installclean && make bootimag and then verify
that the symbolic link exists at $(TARGET_ROOT_OUT)/default.prop
Change-Id: Ied7b32abd7a5fd5456c4cb5ff68a56debe6254b7
(cherry picked from 58c1b0426f)
with cas/camera/drm hals
BUG: 64726466
Test: build aosp_x86-eng, boot emulator
without cas/camera/drm hals killed by
selinux avc denials; camera can take
pictures
Change-Id: I8a154fdb5d7aa7763aab2b912976eb9e1d329cdb
We were previously using LOCAL_BUILT_MODULE, which wouldn't work in the
case that "nostripping" wasn't specified (i.e when we stripped).
The issue is that the built module has its classes.dex entry stripped,
which means that profman can't do anything meaningful with it.
In this change, we use the right set of files for JAVA_LIBRARIES as well
as prebuilt PACKAGES. Fixing apps built from source is a larger task and
has been left for a future change. Depending on ART changes, we might
need to disable stripping for all apps and/or align userdebug and user
builds with each other. This will be tackled separately.
Test: make
Bug: 64896089
Change-Id: I4fd256d187b66763f354cc6001953469c2cd8fbb
TreeHugger Robot