Passing variables via a makefile instead of
rblf_cli / rblf_env allows us to give them correct
types while converting the makefile to starlark,
as opposed to the variables always being strings
when given via rblf_cli / rblf_env.
This also allows us to remove some hand-converted
starlark code.
Bug: 201700692
Test: ./out/soong/rbcrun ./build/make/tests/run.rbc
Change-Id: I58c4f20b29171c14e5ae759beb26a849426f6961
dict(**h[0]) was only copying the top level dictionary
object, but not any other dictionaries/lists inside of
it. This was causing us to not find all the changes to
variables the board configuration was making, and so
some variables were not included in the output.
Bug: 201700692
Test: ./build/bazel/ci/rbc_regression_test.sh mainline_system_x86_64-userdebug
Test: ./out/soong/rbcrun ./build/make/tests/run.rbc
Change-Id: I4467eead7e597f6f49119a8c4832126f39d646e7
It's not guaranteed that the requested image size to mkfs is precisely
respected, due to metadata alignment and such. For accurate care maps
use the real image size rather than requested.
Bug: 205541521
Test: smartsync to 7892270, check that care_map.pb has the right block
count
Change-Id: I60fe64f720db13d3c3c4f1d8968341d7293217c9
These variables' values show up in the command line
of certain build commands, so they need to be stable
to have stable ninja files.
The starlark board configuration strips these variables,
causing a discrepency between the starlark and make
versions of board configuration.
Bug: 201700692
Test: ./build/bazel/ci/rbc_regression_test.sh beagle_x15-userdebug
Change-Id: Id053435409821a3fe5997c07610ef835e0c83112
Some module types include BUILD_NOTICE_FILE a second time after
the include from base_rules.mk. This won't work after a future
change, and doesn't seem to be necessary as none of the files
leave any modified LOCAL* variables in place for notice_files.mk
to read.
Bug: 207445310
Test: m checkbuild
Change-Id: I8c68888d6bf26926f5c2216abe147ccf28ddc2b8
external/zucchini/version_info.h contains version info for zucchini
diffing algorithm. OTA tools will use these version info to check for
compatibility and determine if zucchini can be used safely. Owners of
zucchini agreed to keep the format of version_info.h stable so it's
parser friendly.
Bug: 194237829
Test: m dist
Change-Id: If8e37533933b346d3f03adab9263679910ab328c
Certain board configurations reference $(PRODUCT_OUT)
through deferred expansion, which will no longer work
after conversion to starlark.
Bug: 201700692
Test: build/bazel/ci/rbc_regression_test.sh -b yukawa-userdebug
Change-Id: I02055f5c4a05e540c1752d5964d4db4306292c3b
This is the list of jars that system_server loads dynamically using
separate classloaders. We will rely on this variable to decide which
jars to preopt in the build system and on the device for system_server.
The list is supposed to be in sync with the code in SystemServer.java.
There will be a follow-up CL to add a comment in SystemServer.java to
remind developers to keep them in sync.
Bug: 203198541
Test: m nothing
Change-Id: I305a73218ef2d2c61ac3795d21026b2afe7007fd
Treble doesn't support T system + O vendor, so removing 26.0 (N) and
27.0 (O) prebuilts and compat files.
Bug: 207815515
Test: build
Change-Id: Ie12737cc5ca8725caf834d9bf1bdfe840ac516c0
This is a follow-up CL for https://r.android.com/1900565.
This is needed because product makefiles may need to add their own
APEX boot jars, and they cannot insert them in the correct alphabetical
order on the list. So it needs to be sorted automatically. Do this
after processing all product makefiles and populating the variable,
but before passing it to Soong.
Bug: 207474880
Test: lunch aosp_cf_x86_64_phone-userdebug && launch_cvd
# cuttlefish device boots successfully
Change-Id: I86b58ef8aeac595e401d872f59a2936a67bedeec
dexopt defaults to dex2oat32 unless instructed otherwise. Since
dex2oat32 doesn't exist on a 64-bit-only build, make sure dex2oat64
is selected instead.
Test: boot aosp_cf_x86_64_only_phone and check that installd doesn't
try to use dex2oat32
Change-Id: Ia67e746894684a52a4e5b765bfde0f6dd0efbf6e
This is needed because product makefiles may need to add their own
system server jars to this variable, and they cannot insert them in
the correct alphabetical order on the list. So it needs to be sorted
automatically. Do this after processing all product makefiles and
populating the variable, but before passing it to Soong.
Bug: 207474880
Test: atest art_standalone_dexpreopt_tests:art_standalone_dexpreopt_tests.DexpreoptTest#ForSystemServer
Change-Id: I80dbf555c395c92539c506d3d8b1f2f101a58bce
Test: lunch armv8
- Check that PATH has no "::".
- Check that an executable script in CWD doesn't get executed.
Change-Id: I3a97eb661c614ff2e09723e9ccca0bdeb63dbdc5
TARGET_BUILD_APPS is supposed to be a list of apps to
build, so having it be a bool was incorrect. Change it
to properly reflect what it is, because the starlark
product configuration relies on soong's interpretation
of variables types.
Bug: 201700692
Test: Presubmits
Change-Id: I00109b57abbcda86077f0465f982f54b4a77a7f0
mk2rbc has a few predefined variables that can't be
set in the product/board config makefiles, unless they're
set to their predefined values. Exclude these variables
from the board config input variables so that they don't
conflict.
Bug: 201700692
Test: m RBC_BOARD_CONFIG=1 mainline_system_x86_64-userdebug (ninja files differ still but this fixes a compilation error)
Change-Id: Idc11b2c5029d28116236b289ad1f09afaaf83cc3
The board config can actually change a few product config
variables that are not made read-only.
Bug: 201700692
Test: ./build/bazel/ci/rbc_regression_test.sh -b aosp_cf_x86_64_phone
Change-Id: Ib2ac04ac2c5a56d32bca51fc8189e44e180075ee
location of ota_from_target_files changed from
out/host/linux-x86/bin to
out/soong/host/linux-x86/bin . This changes relative position of
signapj.jar. To fix, use ANDROID_HOST_OUT as search path
Change-Id: I5397171566e9d7598b5ef16ae26641f0c183d748
builds (reland).
- Use art/build/boot/boot-image-profile.txt for the primary boot image
in the ART module, both when it's built from source in platform and
as an unbundled module.
- Use frameworks/base/boot/boot-image-profile.txt for the framework
extension image in platform, but not in unbundled builds.
This should obsolete the combined profile
frameworks/base/config/boot-image-profile.txt.
This relands https://r.android.com/1881863 with a fix to allow multiple
values on PRODUCT_DEX_PREOPT_BOOT_IMAGE_PROFILE_LOCATION.
Test: build/soong/soong_ui.bash --dumpvar-mode \
PRODUCT_DEX_PREOPT_BOOT_IMAGE_PROFILE_LOCATION
and check that it prints both art/build/boot/boot-image-profile.txt
and frameworks/base/boot/boot-image-profile.txt in a platform build
on master.
Test: build/soong/soong_ui.bash --dumpvar-mode \
PRODUCT_DEX_PREOPT_BOOT_IMAGE_PROFILE_LOCATION
and check that it prints only art/build/boot/boot-image-profile.txt
in an unbundled build on master-art.
Test: banchan com.android.art && m
on master-art and check that
out/soong/.intermediates/art/build/apex/com.android.art/android_common_com.android.art_image/image.apex/javalib/x86_64/boot.oat
shrinks from 14 MB to 4.7.
Test: m droid
on master together with https://r.android.com/1895131 and check that
out/soong/.intermediates/art/build/apex/com.android.art/android_common_com.android.art_image/image.apex/javalib/x86_64/boot.oat
and out/target/product/vsoc_x86_64/system/framework/x86_64/boot-framework.oat
are identical.
Test: lunch armv8-eng && art/tools/buildbot-build.sh
on master-art
Bug: 174746397
Change-Id: I9114271bc69cf0888150b2c778a086bc50b73045
