Tiramisu has been finalized in tm-dev, remove the codename from
branches downstream of tm-dev.
Bug: 225745567
Test: treehugger
Change-Id: I12d5a0d018cd8a7dc53136dc67cd23b78b56d55a
The value of BOARD_AVB_VBMETA_SYSTEM_IMAGE_ARGS and
BOARD_AVB_VBMETA_VENDOR_IMAGE_ARGS are usually the same, e.g.,
--padding_size 4096 --rollback_index 1656979200.
However, in a split build, the value of
BOARD_AVB_VBMETA_SYSTEM_IMAGE_ARGS might be empty.
Using BOARD_AVB_VBMETA_VENDOR_IMAGE_ARGS to correctly
assign the value for vbmeta_vendor.img.
Bug: 232360330
Test: TreeHugger
Change-Id: Ia26816bd309084c3a4c0df6b9f2f3fc5a28bc31a
Before aosp/1851043, it was triggered by checkbuild because the
timestamp is attached to $LOCAL_BUILT_MODULE. With aosp/1851043, the
timestamp is not triggered by checkbuild and when a new VNDK lib is
added the check might get skipped in AOSP tree while it can break in the
downstream trees when actually used.
This change adds the same_vndk_variants.timestamp files to checkbuild
again so that CI catches the errors.
Bug: 233616647
Test: prebuilts/build-tools/linux-x86/bin/ninja -f \
out/combined-aosp_cf_x86_64_phone.ninja -t query checkbuild \
| grep same_vndk_variants.timestamp
Change-Id: I26043ec001ca2473fccfc19368d035374c663a21
The SDK no longer includes system images, so no longer needs to depend
on targets that are only relevant for the system images.
Remove all the explicit dependencies that end up depending on
"everything" but are not included in the sdk zip.
Two positive side-effects:
- fewer unused things get built (now ~32k targets, down from ~108k)
- apexes no longer get built, eliminating the risk of mismatch between
module sdk prebuilts and source-built AOSP apexes.
There are some potential risks with this CL:
- because we no longer forcefully rebuild everything, if there are bugs
in dependencies of exising files in the SDK, incremental builds may
not rebuild them.
- some files that get built into the SDK may be being built as a
side-effect of other files included in the SDK, so removing one file
may cause another to not get built. The solution to that is to add
the file not getting built to ALL_SDK_FILES.
Bug: 205008975
Test: m sdk sdk_repo dist && diff sdk.zip before/after (identical)
Merged-In: I97284864f66aa88556fbe16864f45b04be97634e
Change-Id: I97284864f66aa88556fbe16864f45b04be97634e
The system image itself is being removed from the sdk zip,
so remove the NOTICE file for it at the same time.
Bug: 205008975
Test: m sdk sdk_repo dist
Merged-In: Ib75b3e5919cd203e58e03efc1cb0a378283b0c34
Change-Id: Ib75b3e5919cd203e58e03efc1cb0a378283b0c34
The makefile implementation of dexpreopt is lagging behind the Soong
implementation in the way it handles transitive dependencies: Soong
propagates class loader context through transitive dependencies, such
as static libraries, but makefiles only propagate that information
through uses-library dependencies.
This discrepancy is caused by the difficulty of handling class loader
contexts in makefies: since there is no topological order in handling
module and its dependencies, the information is propagated via file
level dependencies on dexpreopt.configs and merged into the module's
dexpreopt.config with a python script.
To handle transitive dependencies correctly, it is necessary to
generate dexpreopt.config files for all libraries, not just those that
enable dexpreopt.
This change only prepares dex_preopt_odex_install.mk for that and
updates if for corresponding changes in manifest_check.py. It does not
enable generation of dexpreopt.config files for non uses-library
dependencies (e.g. static libraries). This will be done as a follow-up.
Changes:
- renamed option for dependency dexpreopt.config files
- added option for self dexpreopt.config
- moved variable definitions so that they can be used both by
uses-library check and dexpreopt command
- add 'Optional' field in makefile-generated dexpreopt.config to
align it on par with Soong-generated configs
- modify dex_preopt_config_merger.py to add uses-libraries propagated
via dexpreopt.config files
Bug: 214255490
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'
# a few unrelated errors caused by missing optional uses-libraries
# android.net.ipsec.ike
# androidx.window.extensions
# androidx.window.sidecar
# that were present before this patch
Change-Id: I6522319a8415f22f90fc96059a34675fb830e5cc
Except for santizer and framework Java coverage tests as they require
specialized prebuilts built from a matching configuration and we do not
provide them.
And products that use dex2oat(d)s as they are not available as a
prebuilt.
And products that cross compile to arm64 and linux_bionic on the host
as prebuilts for that are not available.
And atv/tv products as they break when building against prebuilts.
Bug: 230478310
Bug: 223592962
Test: presubmit
Merged-In: I3ed091156c4b39dad6eaf578144ac3d070025988
Merged-In: I9731400acfade95b3eab8f2d99fddb4cf856aa91
Merged-In: I43fa4ae8db51012eaf78d70fe65e9e0a3f7eca16
Change-Id: Ib01ba9e2e3c2225c88d52b6811edf04aff02c81e
This reverts commit de9f0b12f1.
Reason for revert: breaks build_test and build_test2 on git_tm-dev.
Reverting the topic together does not work due to limitations with TH. However, this is the change that caused the error, the other changes were just fixing issues unrelated to this specific problem.
Bug: 233412962
Bug: 230478310
Bug: 223592962
Change-Id: I4d4de4f4c63c9c34fb083edbe76b09852af357e4
Test: presubmit
