So far, we have *_system.mk and *_vendor.mk.
This patch split all product settings from *_vendor.mk to new
*_product.mk
This patch also create a new telephony.mk to include all
telephony_*.mk
Bug: 119747870
Test: lunch mainline_arm64-user; make -j
Test: compare the content with and without the patch
Test: the output files in out folder are the same
Change-Id: I79aefdd1971b3b5724a3ec858de4109b645c765a
Merged-In: I79aefdd1971b3b5724a3ec858de4109b645c765a
This product variable is used to generate the loader config (to use
either /system/product or /product), and the vendor one is likely
used in similar ways. Having these partitions is required for mainline
devices, so set it in the shared BoardConfig.
Bug: 80410283
Test: make
Change-Id: I6f2d94ff43572a28df54ea8c60f2d838a3b9216c
The app will move to product soon, but temporarily
whitelist that mainline_arm64 adds it to /system.
Test: make
Change-Id: I7b0ce5623a160bcb0f8b89f3053a007019ff0b62
It is present in media_vendor.mk (and soon media_product.mk),
which is the correct place.
Test: make
Change-Id: I7ffda211cf4b01a22ec01c661959defcf59bbc4b
The Android Runtime APEX is always included on all products.
Test: make systemimage on marlin-userdebug
Test: make systemimage on sailfish-userdebug
Bug: 113373927
Change-Id: I47b02ee44562b8da17bd960825a3489572018d93
Rename ANS to ONS
Bug: 118347827
Test: compile
Merged-In: Id01f0fc595e736f081d97309f62426cc54ac9e94
Change-Id: Id01f0fc595e736f081d97309f62426cc54ac9e94
Bionic libs are moved to the runtime APEX and thus are available via the
path /apex/com.android.runtime/lib[64]. However, for processes that are
started before the APEX is ready, a copy of the bionic libs should be left
under /system/lib[64]. This change ensures that the bootstrap bionic
libs are always installed to the system partition.
Note that, at runtime, the bionic libs in the runtime APEX are
bind-mounted to /system/lib/<libname>.so which essentially hides the
bootstrap bionic libs to the processes after the activation of the
runtime APEX.
Bug: 120266448
Test: m and check that libc, libdl, libm are under
/system/lib[64]
Change-Id: I9d5e8e718aa020c012f312a70ed54652130048dd
It's 2019 and the CDD has required NEON for many years now.
I've left unbundled apps alone for now.
I think the recent fix here for https://issuetracker.google.com/116070933
was incorrect. (I've commented on the bug.)
Test: treehugger
Change-Id: Ia02305cfdff9fd18a85359bc44d8eb9cab1413bd
Individual products can add their own calculator package if desired.
Bug: 120905067
Test: lunch aosp_arm64 && make -j8
Merged-In: I60490b6e2a42c4304768075e8fef9f07cb72d268
Change-Id: I60490b6e2a42c4304768075e8fef9f07cb72d268
Legacy GSI had been changed to include BoardConfigGsiCommon.mk
instead of treble_common*.mk. However, some default settings are
different than the original. This patch fixes
BOARD_BUILD_SYSTEM_ROOT_IMAGE in Legacy GSI makefiles.
The patch also removes redundant settings and fix some comments.
Bug: 122282239
Test: make
Change-Id: Ie4eb8cf410bfd2edc74648c0e5ee81028242a1b7
It's 2019 and the CDD has required NEON for many years now.
I've left unbundled apps alone for now.
Test: treehugger
Change-Id: Iee22d4605a81e2840063a203b76db7df5b4f3c37
This flag turns on 64-bit mediadrmserver for mainline. Keep the default
value for GSI for now.
Bug: 80410283
Test: make
Change-Id: Ie09b258f140b56121e4b531775b2a052b4dc6c15
Generate and store minimum amount of information necessary to
create backtraces in native tools (for crashes and profilers).
The data is compressed and takes <5% relative to JIT code size
(which is less than recent stack map savings; so win overall).
Averages for non-compiled run of maps which filled JIT code cache to ~4MB:
Q: code:1.55kb/method data:0.738kb/method debug:0.064kb/method(*this CL*)
P: code:1.62kb/method data:0.956kb/method debug:0.835kb/method(if enabled)
I also measured the P to Q on-disk stack maps savings in CL/762841 as ~28%
(not all of that is applicable to JITed code, so that is an overestimate).
Bug: 111350693
Test: Device boots, mini-debug-info is generated.
Change-Id: I95883bd4f2d5c2308149ab0ba641b0a703c7f81b
All these modules are to be moved to /product, as they are not
part of the core system image. Move them out of the _system mks
as a start, and temporarily whitelist them so that they can be
moved iteratively.
Bug: 122353713
Test: make
Change-Id: If658e6b4565924cd915a183400c5b2b32984eca4
Merged-In: If658e6b4565924cd915a183400c5b2b32984eca4
The patch modifies the Legacy GSI to use BoardConfigGsiCommon.mk
instead of board config treble-common*.mk, and delete these
treble-common*.mk files.
After the patch, the GSI common board configs are in single file
BoardConfigGsiCommon.mk
The patch also includes:
- Moving `TARGET_COPY_OUT_VENDOR := vendor` to
BoardConfigGsiCommon.mk, because GSI always contains no vendor
contents.
- Configure the binder interface bitness explicitly.
- Unify the structure of each BoardConfig.mk.
Bug: 122282239
Test: compare the out folder between with and without the patch
Test: output files are all the same
Change-Id: Ie5901446920dda317dea848e6d5a7bbcf10a32cf
Contains mainline_system and the relevant AOSP packages for the
other partitions. Intended to be inherited by the aosp_X variants
of mainline devices.
Bug: 118826166
Test: build mainline_arm64
Change-Id: I7d1953bf5ccda4d9652dca4d32efb75b31015342
Similar to I1720644d89ec5289fbe99f95ebcdfbb3f3b20e67,
but for mainline_system. This reduces the diff between
full_base.mk and mainline_system, a step on the way to
get aosp_blueline inheriting mainline_system.
Bug: 118826166
Test: make
Change-Id: I52bfe5dfb3e1a22db2d46fe9d1f372ee25533fdd
The binder interface bitness is always 64bits from P.
To keep 32bits binder interface for Legacy GSI, we need to explict
configure it.
Bug: 122282239
Test: lunch aosp_arm_ab-userdebug; make -j
Change-Id: I90b45adb52c398d1c0cf3f877199f9dd1e0b3f86
soong_ui is passing the Android.mk file list to Kati, and it ignores
Android.mk files found via symlinks, so we don't need the workarounds
for build/make/target/board/Android.mk and build/target/board/Android.mk
both existing.
We've got dist support in Soong now, so we can remove the signapk
Android.mk
Bug: 122332221
Test: m dist out/dist/signapk.jar out/dist/libconscrypt_openjdk_jni.so
Change-Id: I2ac9016b04146340c5ac80f7090d00e186023574
dmctl can list all device mapper targets and their tables which is
extremely useful information for a bugreport. Especially to compare
the device mapper setup with the partition metadata.
Bug: 120916687
Test: 'dmctl list device -v'
Change-Id: I9e3afcf146b2814b0304f69cdf6029303beb57c7
Merged-In: I9e3afcf146b2814b0304f69cdf6029303beb57c7
Signed-off-by: Sandeep Patil <sspatil@google.com>
Home is overriden by Launcher3QuickStep, so listing these two
in the same mk is a noop.
Test: make
Change-Id: I2865a5346c63efd78e434185041f320ca344e854
This is the same list of languages that was in mainline_system.mk
before, but sorted alphabetically.
Make languages_full inherit this list and add the only locale present
that was previously unique (the en_XC pseudo-locale).
Test: make
Change-Id: I09f0c7521740d4c779d5fd2a50baa92b7cdfe8d4
This doesn't appear to be a real locale. The interpretation would be
khmer spoken in the Marshall Islands, which doesn't appear to be a thing.
The locale doesn't appear in https://lh.2xlibre.net/locales/ or
frameworks/base/core/res/res/values/locale_config.xml.
It might have been a typo when it was introduced in internal change
I78d91916fdbb04502466c26bf7d160fe679aaa31.
Test: make
Change-Id: Id1deebdd8ad2a5601c15b84e010be766b3a951e9
Needed for NFC. Has a proprietary override, which is the
only reason I had not added it earlier.
Test: make
Bug: 80410283
Change-Id: I1fad3d2bab0ff2fa8bbd32388049f0f8c86b68a5
GSI includes a default apns-conf.xml. This patch align the
apns-conf.xml in Legacy GSI with the one in non-Legacy GSI.
Bug: 122238063
Test: lunch aosp_arm64_ab-userdebug; make -j and check output
Change-Id: Ia9e2e411d9f62896f75e190c72260688bed5369d
Some non-phone form factors don't want cameraserver to save storage and RAM.
Pushing it down to handheld and other build files allows selective inclusion to
happen there.
Intended to be a no-op refactoring, will disable for ATV in a follow-up CL.
Bug: 118369095
Test: built system image for random device as smoke test
Change-Id: I260fa5c9e512341d6a6824afebd5e8273fa5e66c
Merged-In: I260fa5c9e512341d6a6824afebd5e8273fa5e66c
Uses skip_mount.cfg for `init` to skip mounting product and
product_service paritions in GSI. The patch also removes
the un-mount in `init.gsi.rc`.
With this patch, a device with flashing GSI will not use the
content in product/prodcut_service partitions on the device
and will always use the content under /system/product in GSI.
The patch also rename the target/product/vndk to
target/product/gsi.
Bug: 120208657
Bug: 117246558
Test: Boot blueline master/P with flashing aosp_arm64-userdebug
Test: /product and /product_service should be symbolic links
Change-Id: Idc9c60e5a05b3d90fc4ab64b0ac000571849b19b
libunwind has no dependency on vendor modules.
Remove it from VNDK list.
Bug: 120948701
Test: build and boot
Change-Id: Ic4c0080fea7da021334aaa1a81b3b7377b06f62f
The device currently targeted does not have a product partition.
Adding product-specific apps to it therefore trips up the artifact path
check for /system.
Workaround that by whitelisting /system/product/% for now. This should
be removed when the device has the right partitions.
Bug: 110072687
Test: build mainline_arm64
Merged-In: I54175bc51df5dfe82c64f66f255a2cea6ff9b9ab
Change-Id: I54175bc51df5dfe82c64f66f255a2cea6ff9b9ab
For go devices, the network stack should not run in a separate process:
NetworkStackLib is used in-process instead.
Test: built & booted a go and non-go config
Bug: b/112869080
Change-Id: I18b0446a0c0d19519ce0ef344f185b54ad7d4ac6
Needed to get some space on marlin-eng :(
Also avoids static linking libart into perfprofd. Roland will follow-up by reverting the no-perfprofd changes.
bug: 73175642
Test: lunch marlin-eng && m -j32
Change-Id: Ic1c79c6dd1052981801b92604f9891a8a63b4ea0
Partial rollback of Id5d114bde8b93800fe52c791ccc47cdffd86f382.
It is likely mainline also wants some of the BOARD_AVB_SYSTEM.. flags,
but I'm not entirely sure which ones.
Test: run build_mixed with this patch
Change-Id: I6ef3c01f6c74e6d93cf798b26b8c7a4785bbefb4
Make the mainline board config only import a subset of the GSI
options. The current set of attributes is the minimal set to
keep mainline_system_arm64 building, as well as not producing
any diffs in the current set of files in /system. It's possible
we'll need to move a few more over, but we'll do that on a
case-by-case basis.
Make the GSI config inherit the mainline one. The only diff for GSI
should be the addition of BOARD_USES_SYSTEM_OTHER_ODEX.
Bug: 119911662
Test: build mainline_system_arm64
Change-Id: Id5d114bde8b93800fe52c791ccc47cdffd86f382
- LMKD_LOG_STATS being false is a disk-saving flag not suitable
for mainline
- USE_HWC2 should likely be cleaned up, but toggles some timing
behavior in libhwui
Bug: 80410283
Test: build mainline_system_arm64
Change-Id: I8a1be50bd2cc48405656cd1daad8e8a1ca555871
This includes the dns resolv APEX in the system image
Bug: 119527674
Test: build / boot crosshatch
Change-Id: Ie92db9979fb63bfabcd29590d0ea236fd8717970
To better test CBRS, we want IRadio 1.1 to be Android P plus CBRS
HAL interfaces, while 1.2 will be 1.1 plus all other Android Q
interfaces. So we are creating V1_2 folder and moving everything
currently defined in android.hardware.radio.config.V1_1 there.
Bug: 117805040
Test: build and telephony unittest
Change-Id: Icf208036385d1d9994d07fe7d839b290f2234622
Merged-In: Icf208036385d1d9994d07fe7d839b290f2234622
To better test CBRS, we want IRadio 1.3 to be Android P plus CBRS
HAL interfaces, while 1.4 will be 1.3 plus all other Android Q
interfaces. So we are creating V1_4 folder and moving everything
currently defined in android.hardware.radio.V1_3 there.
Bug: 117805040
Test: build and telephony unittest
Change-Id: I0a0c651b25caa4d0bd88bf51c02a9f7160d0caff
Merged-In: I0a0c651b25caa4d0bd88bf51c02a9f7160d0caff
This is a types-only HAL used by any HAL as a convenient primitive for
effectively using safe_union, so it should be able to be
double-loadable.
Bug: 119790516
Test: build and boot walleye
Change-Id: I8f48ec6e3a0f47bfd5bb90d8c496848fbed85687
These are libraries commonly depended on by things on the product
partition. Add them explicitly for now.
Bug: 80410283
Test: build mainline_system_arm64
Change-Id: I6696990243397aa7109f88951ff120acf2a0f7e8
rss_hwm_reset is binary that reset RSS high-water mark counters for all
currently running processes. It runs in a separate process because it
needs dac_override capability.
The binary is not a persistent service. It is invoked from a system
service by flipping sys.rss_hwm_reset.on property.
Bug: 119603799
Test: adb shell ls /system/bin
Change-Id: Iee0b54ec6486d8f23048819493f6d710db3feff6
This ensures all -userdebug builds have the sanitizer-status binary,
which is used to check which build flags were used for a given image.
It is important to distinguish between what the build fingerprint
claims, and what is actually available on the phone.
Change-Id: Iac1eabb8c5ed885a057f7a302b1887b7e9c4ccbc
Merged-In: Iac1eabb8c5ed885a057f7a302b1887b7e9c4ccbc
(cherry-picked from 3a26f80cd4)
Likely not the final setting we'll settle with for this device,
but brings mainline_system builds closer to the device builds
it's targeting.
Bug: 80410283
Test: build mainline_arm64
Change-Id: Ib672d1a060d0ad6a4e1faa210461bbbbaf6f2bc6
This is a workaround solution. It let GSI to unmount product
partition when booting. The patch avoid the problem that
- share UID between between system and product partition
- the product partition override the content under /system/product
Bug: 119391482
Test: Boot aosp_arm64-userdebug on a blueline device with Android P
Change-Id: I48808fcc1ca527a23eb1446b97176b08268b500e
This includes the time zone data APEX in the system image
which should flush out any obvious issues.
Bug: 119026403
Test: build / boot taimen
Change-Id: I3d9a8289d8af405a373c94891b482540e59a3571
Unless specified otherwise (using `DONT_INCLUDE_RUNTIME_APEX`).
This reverts commit 4c39bd2c6e.
Test: make systemimage on taimen-userdebug
Test: make systemimage on marlin-eng
Test: make systemimage on marlin-userdebug
Test: make systemimage on sailfish-userdebug
Bug: 113373927
Change-Id: I1f1f0ffcc7bb6bf26d4df72613a806d2d73c5e82
This reverts commit cb6e935db8.
Reason for revert: Increases the system image size beyond the
limit for some targets (e.g. marlin).
Bug: 113373927
Change-Id: Ia609523a60179c32d377d0735049c944464ea0cc
Required by certain ringtones, apparently (though I couldn't find
what actually uses this library).
Bug: 80410283
Test: lunch mainline_arm64; m
Change-Id: I36b5b07156b41c442903d7d07e6b9d9f4abe2345
This is disabled by default, but can be enabled via a
config overlay.
Bug: 116191025
Bug: 80410283
Test: make
Change-Id: Ibfe5d85580ccccc774de166750d91efd74458a8d
Remove apex_debug_key, in favor of apex.test.key which is under
/system/apex and is built with soong.
Bug: 118213152
Test: /system/etc/security/apex/testkey exists
Change-Id: Iaa0facea2d26cadd48783778e8915bc3c560de10
Some libs, such as libandroid_net.so and some other Renderscript related
ones, can't be built in PDK due to missing frameworks/base.
Bug: 118634643
Test: build PDK; build doesn't break due to missing VNDK libs
Change-Id: Ic2bf657d796c2044c3b8f49b2963ca84760d54d1
Locales not in this list will get stripped.
Bug: 80410283
Test: lunch mainline_arm64; m
Change-Id: I90a63e234e2f2ee32da8f9f257248f5ee6354151
Merged-In: I90a63e234e2f2ee32da8f9f257248f5ee6354151
(cherry picked from commit d994b0dc98)
Prevents additions of non-aosp artifacts to these products.
Bug: 80410283
Test: lunch mainline_arm64; m nothing
Test: lunch mainline_system_arm64; m nothing
Change-Id: Id12a77bd7f4cc9917aecbca76e8a2842d7c77613
The existing 'generic' targets are emulator devices, and targeting
it has the side-effect of including emulator-specific artifacts.
Create a mainline device instead, which is really generic. This also
allows specifying tweaks unsuitable for the emulator.
Include BoardConfigGsiCommon for now, even though it includes a few
things we likely won't want in the future.
Bug: 80410283
Bug: 111538404
Test: lunch mainline_arm64; m
Change-Id: I15e7cb41fcbd3c0592a091ba7a8575ddf10bda71
Let this be defined in the leaves instead. As is it frequently causes
problems where the product inheriting this also inherit another value
of PRODUCT_SHIPPING_API_LEVEL, leading to invalid values unless
overridden again.
Bug: 80410283
Test: lunch mainline_arm64; m nothing
Test: lunch mainline_system_arm64; m nothing
Change-Id: Id08b6cc0cd8815404e548c521c9e40706ba8ca16
libvixl-arm and libvixl-arm64 are not used by any vendor libraries, but
only by ART. Removing them from VNDK.
Test: m
Change-Id: I5b83be13cd70fb12581fe561798484a8907022c6
To make the generic_x86_arm bootable.
Third party translabor binary can be added on the resulting
system.img and be tested at least on emulator.
BUG: 117673798
Change-Id: Iffbcf9217bbedc12ba83a4851e84ecbf299bf8e7
Merged-In: Iffbcf9217bbedc12ba83a4851e84ecbf299bf8e7
Missed adding this app in I5afabc7dcc4ca320f06fdc4a2669fb1b81e53b0b.
It is usually added in conjunction with RcsService.
Bug: 80410283
Test: lunch mainline_arm64; m nothing
Change-Id: Ic721d1dcef479a03f4b603bd77a91801ea9dd9ec
With a workaround for preopt files that get installed into an unknown
path. The BoardConfig defines whether the files go in /system_other,
so allow that.
This reverts commit 46bf4b510c.
Test: lunch mainline_arm64-userdebug; m nothing
Change-Id: Ia75670eb6e3730ad0ff17f5b81954b69f6cab22b
With system-as-root, both first and second stage init belong to
system, but without system-as-root, first stage belongs to vendor and
second stage belongs to system. This change references two phony
packages that include the right parts of init for systen and vendor.
Bug: 117606184
Test: mainline builds work properly
Change-Id: I997309655fef28cb1c3d4b3be989c035b5c6488d
SzuWei Lin