When I removed the actions that ran the python link type checker, I
forgot that the Make version hadn't been enabled yet. Enable the Make
version.
Bug: 143748700
Test: treehugger (particularly, build_test on the downstream branches)
Change-Id: I293b3be5ca7828c9d7148e13d88b0f0bba54d964
Merged-In: I293b3be5ca7828c9d7148e13d88b0f0bba54d964
Merged-In: I08090db44d47f809fbaa0c4ad373e46e6a3fbe2e
If true, incorrect order of system server jars does not cause a build
error. The order is non-optimal if some jar X precedes its dependency Y.
In that case dexpreopt will be unable to resolve any rerefences from X
to Y.
Test: lunch cf_x86_phone-userdebug && m nothing
Test: lunch aosp_car_arm && m nothing
Bug: 140451054
Change-Id: I70efd365fa9fbd9e40792d646f8471d009164924
Merged-In: I70efd365fa9fbd9e40792d646f8471d009164924
(cherry picked from commit 17715b48f0)
Refactor the build support for kernel modules. This refactoring is
leveraged to add support for directories of kernel modules in each
partition. To add a directory of kernel modules, define
BOARD_KERNEL_MODULE_DIRS to include the directory name, for example:
BOARD_KERNEL_MODULE_DIRS := 5.4
The build will then check all the usual kernel module related macros
with an extension of _5.4, for example you can define
BOARD_VENDOR_KERNEL_MODULES_5.4 := <modules to copy>
BOARD_VENDOR_KERNEL_MODULES_ARCHIVE_5.4 = <path to archive file>
BOARD_VENDOR_KERNEL_MODULES_LOAD_5.4 := <modules for load list>
Bug: 157645635
Change-Id: I1687d0ec85c1dcf21350350623b4689aba3f6717
Moving this makes each of the partition's kernel module Makefile
generation more similar, enabling a forthcoming cleanup/refactoring.
If recovery-is-boot and there are kernel modules defined in
BOARD_GENERIC_RAMDISK_KERNEL_MODULES, those kernel modules will
now be included in the default assignment to
BOARD_RECOVERY_KERNEL_MODULES_LOAD. If this is not desired then
BOARD_RECOVERY_KERNEL_MODULES_LOAD may simply be set to
BOARD_RECOVERY_KERNEL_MODULES (or a custom list of recovery kernel
modules to load).
Bug: 157645635
Change-Id: Ia1392b7d20d07cee75420edd95dcee1dc825516e
Currently the two special-case calls to create just a module load list
(as opposed to copying a set of modules and creating a module load list)
are inside checks that the related board macro for the module set is not
empty.
While it would be incorrect to create a non-empty module load list with
an empty set of modules, this error would quickly surface at runtime.
More importantly, moving these out will permit refactoring this code to
clean it up.
Bug: 157645635
Change-Id: I199788d49072d4c59737b3ef2c75546852cf3e36
This check is meant to initialize BOARD_ODM_KERNEL_MODULES_LOAD to
BOARD_ODM_KERNEL_MODULES if BOARD_ODM_KERNEL_MODULES_LOAD is empty.
Bug: 157645635
Change-Id: I9c8ac0e64bb002e926d15b446a63801d7cfa9911
It may be needed in class loader context if the target API version
is less than 30.
Test: lunch aosp_cf_x86_phone-userdebug && m
Change-Id: Ifa5ab18d1dc322184100207ad6eee4bdcf74deb0
The module-load-list-copy-paths fn takes five parameters.
Merged-In: I7d1736da1599e265793173a741053d8a09f1622f
Change-Id: I9286c9e4ca81caf5392bc804dead718bd028b193
* changes:
Drop the dependency from vendor/build.prop to system/build.prop
Remove ro.bootimage.* sysprops
ro.build.expect.* sysprops are created in a separate rule
Make use of ADDITIONAL_ODM_PROPERTIES
Refactor post_process_props.py
remove code for default.prop
Make more use of ADDITIONAL_VENDOR_PROPERTIES
ADDITIONAL_*_PROPERTIES are props determined by the build system
First, make it safer for incremental builds. We used to just be
depending on the bootable/recovery/tools/recovery_l10n/res dir as a
dependency, but that would only trigger a rebuild if a direct child
file/directory was added or removed (so essentially, when a translation
was addded or removed). If a translation was updated, it wouldn't
re-trigger recovery_text_res.
So instead of depending on the directory, depend on the results from
running `find` to list all of the contents in that directory.
Next, move it out of PACKAGING. `m installclean`, which runs in between
incremental builds on our build servers, deletes the obj/PACKAGING
directory. So when recovery_text_res used that directory, we were having
to rebuild it on every incremental build, which isn't optimal when it
takes nearly 90s to build.
So with:
$ lunch aosp_coral-eng
$ m bootimage
$ m installclean
$ m bootimage
The second `m bootimage` went from executing 666 actions in 93s to
executing 658 actions in 6s.
Finally, remove the last uses of $(call include-path-for,recovery),
since they were all in this file (and they weren't using it for include
paths...). We'd like to remove all of these and switch them to header
libraries or other use cases, but in this case, it's just shorter to use
the real path.
Test: treehugger
Test: m bootimage;
touch bootable/recovery/tools/recovery_l10n/res/values/strings.xml
m bootimage
Test: m bootimage; m installclean; m bootimage
Change-Id: I005592e49443aab45ed039a2f0c63f7a69035565
Ramdisks are currently compressed using gzip, which is generally
smaller, but slower to decompress than lz4. Provide an option for an
lz4 scheme to reduce the time taken to unpack initramfs during boot.
Bug: 156129966
Signed-off-by: J. Avila <elavila@google.com>
Change-Id: Iac9538e6ee6ec51e6b487de2101f53bb5d9c54c8
aosp_arm64 has no emulators after the patch:
https://android-review.googlesource.com/c/platform/build/+/1286895
The patch creates all other emulator_* devices by cloning from
generic_*, and modify all sdk_phone_* to use them.
Bug: 111538404
Test: build sdk_phone_armv7-userdebug
Test: build sdk_phone_x86_64-userdebug
Test: build sdk_phone_x86-userdebug
Change-Id: I2b82e999ea919f005729f322cd948e6b72ba6843
The former doesn't use anything from the latter. The dependency is not
needed.
Bug: 117892318
Test: m
Change-Id: I61aeea9feb1ac4467af4f185f3d6930b03145afd
ro.bootimage.build.date.* are not used anywhere.
ro.bootimage.build.fingerprint was used in the recovery mode, which has
now been switched to ro.build.fingerprint.
Bug: 117892318
Test: m
Change-Id: I3776c4154f86ddb908b24c6899037c6a7df037df
This change is required to unify the build rules for */build.prop files;
The rules shouldn't have partition-specific actions.
Bug: 117892318
Test: m
Manually inspect ro.build.expect.[bootloader|baseband] props in
vendor/build.prop
Change-Id: I3b29b2728f779a9d7e1d71a5aa69b02ea940b3c6
Properties that were defined in the build rule for odm/build.prop are
now defined via ADDITIONAL_ODM_PROPERTIES.
Bug: 117892318
Test: m
Change-Id: I6220e217d0724b6e27127786e7a908e1e662d1a0
The script is refactored to not parse lines everytime a prop is looked
up or updated.
Also it is now built using python_binary_host.
Bug: 117892318
Test: m
Change-Id: I1677ca5aa919c54b7b0740fdc52e1399009f9a12
default.prop files are gone. Remove the code in post_process_prop.py
that handles the file.
Bug: 1313599
Test: m
Change-Id: I3b4ff4adac57a406b70300d34ece7028cb52eb2d
Properties that were defined in the build rule for vendor/build.prop are
now defined via ADDITIONAL_VENDOR_PROPERTIES.
Bug: 117892318
Test: m
Change-Id: Ib3dc9d001a8dcd0aa38f1acae384a96b67b5af8a
This change defines ADDITIONAL_<partition>_PROPERTIES to store properties
determined by the build system itself.
ADDITIONAL_SYSTEM_PROPERTIES was originally ADDITIONAL_BUILD_PROPERTIES.
ADDITIONAL_VENDOR_PROPERTIES was originally FINAL_VENDOR_DEFAULT_PROPERTIES.
ADDITIONAL_PRODUCT_PROPERTIES has been there and thus not changed.
ADDITIONAL_ODM_PROPERTIES is newly defined, although it is empty now.
In addition, ADDITIONAL_VENDOR_PROPERTIES are now defined in
core/main.mk along with other ADDITIONAL_*_PROPERTIES.
Bug: 117892318
Test: m
Change-Id: Ic41bba2e5695355436ac8df353866cfc044e1e0b
We've converted all known users of mainline.mk to use
mainline_system.mk. Don't include mainline.mk directly in the future,
instead include mainline_system.mk and other partition specific
makefiles. Remove mainline.mk to prevent people from accidentally
including it again.
Bug: 141443017
Test: TH
Change-Id: Id161f29951982a100482a6201130adeebdb49846
The size of the ramdisk has greatly increased, which can be a source of
boot time regressions. Strip the modules of debug symbols for the vendor
ramdisk and vendor partitions; vendor stripping can be skipped via BOARD
config.
Bug: 156395749
Signed-off-by: J. Avila <elavila@google.com>
Merged-In: I248d4e2097992f00975b344b5da41aa5d8016812
Change-Id: I248d4e2097992f00975b344b5da41aa5d8016812
Sasha Smundak