This CL adds the support for specifying lists of directories in
build/soong/android/bazel.go, which are then written into
out/soong/bp2build/MANIFEST. Using this configuration,
modules/directories can either default to bp2build_available: true or
false, while still retaining the ability to opt-in or out at the module level.
It also ensures that ConvertWithBp2Build returns true iff the module
type has a registered bp2build converter.
Test: go tests
Test: demo.sh full
Test: TreeHugger presubmits for bp2build and mixed builds.
Change-Id: I0e0f6f4b1b2ec045f2f1c338f7084defc5d23a55
PlatformSepolicyVersion and BoardSepolicyVers haven't assigned correctly
so far. Below is the reason why it hasn't been discovered yet.
DeviceConfig.PlatformSepolicyVersion() was added to support mixed
sepolicy build (setting BOARD_SEPOLICY_VERS and building vendor sepolicy
with old plat policy files). Soong compares PlatformSepolicyVersion()
and BoardSepolicyVers(), and used old vendor sources if both are
different. Back then, the only place where such logic played a role was
selinux contexts. Test codes were running as intended: after setting
BOARD_SEPOLICY_VERS the context files were built against old policies.
But there were two mistakes:
1) PlatformSepolicyVersion() was not added to soong_config.mk, so it was
always empty.
2) BOARD_SEPOLICY_VERS was set to default in system/sepolicy/Android.mk,
which was processed after soong_config. So if BOARD_SEPOLICY_VERS was
not set in BoardConfig.mk, BoardSepolicyVers() was empty, not
PLATFORM_SEPOLICY_VERSION.
And there were no issues as Soong only checked equality.
To fix the issue correctly, this commit adds the variable
PlatformSepolicyVersion, and then handles BoardSepolicyVers correctly by
returning PlatformSepolicyVersion if it's empty.
Test: set BOARD_SEPOLICY_VERS and see vendor_file_contexts changes
Change-Id: Ie55827f47d2903e9739eccacc3874d354d5c9f2d
Remove the references to the android package in remotexec so that
the android package can reference the remoteexec package. This
will allow RuleBuilder to integrate directly with remoteexec.
Bug: 182612695
Test: m checkbuild
Change-Id: I15be5ef126d8aacbd605518638f341daf6f31bb3
This requires linking Blueprint into soong_ui. It lets us avoid the
complicated dance of Ninja files and shell scripts: now the information
as to how soong_build is built is passed directly to Blueprint using a
struct that contains all the information the command line arguments used
to contain.
The ability to run Blueprint from the command line is kept (for now).
Some variables in bootstrap/command.go needed public accessor functions
because soong_build reads them. This will be disentangled by moving the
flag parsing to soong_build.
The presence of the flag definitions in Blueprint means that soong_ui
now also accepts them. This is not a problem in practice because they
are ignored and because soong_ui itself is hidden behind a few layers of
shell scripts.
Test: Presubmits + the new bootstrap_test.sh .
Change-Id: I6dca478f356f56a8aee1e457d71439272351390b
Calling ExistentPathForSource from a PackageVarContext has to fall
back to a less optimal glob dependency implementation that adds the
file to out/soong/build.ninja.d. Checking for "frameworks/base"
adds the directory to build.ninja.d, resulting in any writes to that
directory potentially causing a regeneration. Check for
"frameworks/base/Android.bp", which will already cause a regeneration
if its timestamp changes and therefore not increase regenerations.
Bug: 158342713
Test: m nothing && touch frameworks/base/foo && m nothing
Change-Id: Iac7ac0445dbaa365c90e9a012f1f32020afc6c08
The end goal of this exercise is to remove all mutable global state so
that multiple Blueprint instances can be run in the same process.
Test: Presubmit.
Change-Id: Idb34b0920f2c7d92efb9328ce8a78b7306f89571
This lets us not leak $PATH to soong_build. It was only needed for
Darwin tests anyway.
Test: Will ask jingwen@ who has a Mac.
Change-Id: I4647e41275b323fe6283580f8f92718c6229f23e
The old error message was wrong because this message is emitted
specifically when a non-hermetic tool is not found.
Test: Manual ("m nothing")
Change-Id: I857c1c2bb059aff894e8bd3dd0c09b030c82035d
aosp/1609473 broke this by making originalEnv empty unless
InitEnvironment() is called, which doesn't happen in tests.
Bug: 181835672
Test: "m nothing" on Linux and Darwin.
Change-Id: I33f9cce65a9151e620e1dddafb62d4c599ed2cac
Adds the test fixture support and converts a few tests to exercise the
code and show how it works.
Bug: 181070625
Test: m nothing
Change-Id: I0a2b40fff93b6041f9aa8c4ef0aba91da1bc8bf3
KYTHE_JAVA_SOURCE_BATCH_SIZE environment variable controls this setting.
The limit is 1000 if this variable is not set.
Fixes: 179932118
Test: run prebuilts/build-tools/build-prebuilts.sh, use it to build kzips, check
Change-Id: I9ad57dfd1d2c2dce5cff755b1bd61cf933420bd3
Previously, ConfiguredJarList would accept an empty apex name,
e.g. ":jar" which makes no sense as every apex has to have a non-empty
name. This change makes an empty apex invalid.
In order to improve the test coverage of the TestConfiguredJarList test
this change also changes the implementation of
CreateTestConfiguredJarList([]string) to marshal the supplied strings
into a json list and then unmarshal into a ConfiguredJarList which more
closely matches how it is used at runtime.
Bug: 178361284
Test: m nothing
Change-Id: I7dfec6b4cc1923aa99746e976da0393922ef0791
By enabling these hiddenapi CSV files to be prebuilt, it
becomes possible to create a split build that supports
the hiddenapi encode dex step, but doesn't contain all
of the java sources needed to generate the CSV files.
Bug: 175048716
Test: m nothing
Test: new TestHiddenAPISingletonWithPrebuiltCsvFile
Test: local build without prebuilt hiddenapi
Test: local build with prebuilt hiddenapi
Change-Id: I805ea3ec05860d2df211a3985ec025bf36f0d775
Vendors can now generate only needed modules by setting the following
Makefile variables:
- DIRECTED_VENDOR_SNAPSHOT: set to true
- VENDOR_SNAPSHOT_MODULES: list of snapshot candidates
e.g.
DIRECTED_VENDOR_SNAPSHOT := true
VENDOR_SNAPSHOT_MODULES := toybox_vendor sh_vendor libbase libcutils ...
Bug: 157967325
Test: m dist vendor-snapshot after setting those in BoardConfig.mk
Change-Id: I6515a43d9358d94483d7c7fa2b066f9dd457f6aa
Now newer system policy and older vendor policy can be built together by
setting following variables:
- BOARD_SEPOLICY_VERS
- BOARD_REQD_MASK_POLICY (copy of older system/sepolicy/reqd_mask)
- BOARD_PLAT_VENDOR_POLICY (copy of older system/sepolicy/vendor)
- BOARD_(SYSTEM_EXT|PRODUCT)_(PUBLIC|PRIVATE)_PREBUILT_DIRS (copy of
older system_ext and product policies)
Bug: 168159977
Test: try normal build and mixed build
Test: boot and check selinux denials
Change-Id: I20e1986cc1c877f5e3a7965e03bd8ade84fd7230
The APEX symlink optimization is a build-time trick to save the
storage/RAM usage of APEX by replacing some files in APEX with symlinks
to the files in the system partition. The optimization however is
automatically turned off for 'updatable: true' APEXes because doing the
optimization for them will hide the sys-health implication until when
the APEXes are built unbundled (i.e. prebuilt) and thus the
optimization is impossible.
TARGET_FORCE_APEX_SYMLINK_OPTIMIZATION forcibly disables the safety net.
When it is set to true, the symlink optimization is done regardless of
the 'updatable' property. This is useful for some of the devices like Go
where most APEXes (even the 'updatable: true' ones) should be
effectively non-updatable.
Bug: 175630508
Test: TARGET_FORCE_APEX_SYMLINK_OPTIMIZATION=true m and check that
updatable APEXes have symlinks to system libs
Change-Id: I26f72e5d5ebccc2d1e09c2a2f743db14937eb39a
1. Soong can now detect PRODUCT_COMPRESSED_APEX flag
We don't want APEX to be compressed on all devices. Only those that
have explicitely set PRODUCT_COMPRESSED_APEX flag.
2. Handle "compressible" field in soong build rule
On devices that supports APEX compression, all APEX will be
compressed by default. If any apex does not want to be compressed,
they will need to state that by setting "compressible" field to
false
3. Can use apex_compression_tool to compress APEX
Note we compress the APEX after it has been signed. That way, when
we decompress we will get a signed APEX.
4. Place the compressed APEX in system with .capex extension
This makes it easy to identify. We still preserve the original
extension so that when we decompress, we can just rename by cuttif
off the .capex extension.
Note: with this change, we can create a system image with compressed
APEX, but we cannot boot with it since platform doesn't know how to
handle .capex files. Platform support will be added on follow up CLs.
Bug: 172911362
Test: OVERRIDE_PRODUCT_COMPRESSED_APEX=true m (apex_test.go)
Test: observed $OUT/system/apex has .capex files
Change-Id: I20ac4c4ceb521924c751a6017f979b2d808fdded
No current use cases for soong.config and user-editable local
configuration options for Soong, so let's remove it to reduce the API
surface.
The MegaDevice configuration is used for building every module, for
every cpu variant, for every architecture, but it apparently isn't
used for a while now.
Test: m nothing
Test: TH
Fixes: 174188200
Change-Id: I2550e70ff6f9c8b57e9a7cc517d6a119a032a27a
Jingwen Chen