This reverts commit e7c71c344d.
The intermediates created by the separated rust compile steps are very
large, on the order of 60GB. This is more than CI can support for now,
revert the split into compile and link phases. This comes at the expense
of incremental build time, especially when modifying C/C++ sources that
are dependencies of rust modules.
Bug: 293349612
Test: builds
Change-Id: I81169e7d0727330c8de5e7688dcdc87fe7b8d3b5
Also add script to remove the Bazel output base.
This will assist with supporting movable checkouts alongside
mixed builds.
Bug: 259191764
Test: m && (move topic and prepare_moved_top.sh) && m
Test: m && prepare_moved_top.sh && m
Test: build/soong/tests/relative_symlinks_test.sh
Change-Id: I0f53da8d99f752fad496cf3ac61b01f001b7296d
`conv_linker_config validate` command is used to validate the linker
configuration embedded in APEX to detect common mistakes.
For example, when used in APEX, linker configuration can't set
provideLibs/requireLibs. For APEX, there are
provideSharedLibs/requireSharedLibs in APEX manifest for that purpose.
One might make mistake by setting provideLibs in linker config.
Now, when these unsupported properties are set, there'll be build-time
error like:
// set provideLibs key in com.android.art's linker config.
$ m com.android.art
...image.apex/etc/linker.config.pb: provideLibs is set. Use provideSharedLibs in apex_manifest
Bug: 264341796
Test: m com.android.art (see above)
Change-Id: Ibaf7322616ad333569e6d721680f3d72243402a2
So that we don't erroneously add skip_preprocessed_apk_checks
everywhere.
Bug: 185811447
Test: Presubmits
Change-Id: Icb1c6163d170ca4181c5c6a814b51fda4777746f
When targetSdk is >= 30, the system verifies that you use a valid
signature V2+ certificate. Uncompressing ndk/dex files or aligning
the zip file will break a signature V2, so these apks should really
just set preprocessed: true.
Fixes: 185811447
Test: Presubmits
Change-Id: Id89c42bcd5b5daa6eda1716bff4023423298036b
- Don't pass `--as-needed` to the linker on Mac OS which is unsupported
there
- Use `--force_load` rather than `--Wl,--whole-archive` on Mac OS
- Scan `rustc`'s linker arguments for `-dylib` and `-dynamiclib`, which
it can use instead of `-shared` on Mac OS:
7bd81ee190/compiler/rustc_codegen_ssa/src/back/linker.rs (L319)
Test: m libhalf serde_derive
Bug: 291164566
Change-Id: Iecd6c2532fa31c9476834f49b109de98cbd2dccf
Regarding aosp/2624311, `blkid_path` was removed from the command, so the number of parameters should be reduced by 1.
Test: MODULE_BUILD_FROM_SOURCE=false m (in internal-main branch)
Bug: 279858383
Change-Id: Iad494b757e2a4152a6867098c3eb7419c3bba71b
deapexer doesn't need it.
Bug: 279858383
Test: presubmit
Test: m MODULE_BUILD_FROM_SOURCE=false droid
Change-Id: Ie8311c74682f366cf6aeebb669500d01675cd0bf
Before this change, dexpreopt was often broken with optional libraries.
This was because the CLC construction was done in Soong at an early
stage, where we don't have sufficient information to determine whether
an optional library is installed or not.
For example, the "Settings" package uses an optional library called
"androidx.window.extensions". On some devices, the library is installed,
but on some other devices, it's not. Soong always adds the library to
the CLC, meaning the CLC is wrong for devices which don't have the
library. This change fixes the problem. See the tests below.
After this change, the CLC construction is done by a Python script
invoked at a very late stage. It uses product_packages.txt, which is
generated by Make, to determine whether an optional library is
installed or not, and filter out libraries that are not installed.
Note that optional libraries are still added as dependencies by Soong.
This is because dependencies have to be added at an early stage. This
means what dex2oat eventually uses will be a subset of the dependencies,
which is fine.
Bug: 282877248
Test: m
Test: atest construct_context_test
Test: -
1. lunch aosp_cf_x86_64_phone-userdebug && m
2. Check the .invocation file of the "Settings" package (defined in
.bp file)
3. See androidx.window.extensions
Test: -
1. lunch aosp_redfin-userdebug && m
2. Check the .invocation file of the "Settings" package (defined in
.bp file)
3. Don't see androidx.window.extensions
Test: Check the .invocation file of the "Dialer" package (defined in
.mk file)
Test: -
1. Build a Pixel 5 system image and flash it to a Pixel 5 device.
2. adb shell pm art dump
3. See "reason=prebuilt" instead of "reason=vdex".
(https://diff.googleplex.com/#key=fB6Ls9q2QGSN, before: left,
after: right)
Change-Id: Ia112bd7c2328373e68db6bffb74bf34030f683d8
It was inlined into the usage in the other cl in this topic.
Bug: 280685526
Test: ./build/bazel/ci/rbc_dashboard.py --quick aosp_arm64-userdebug
Change-Id: I68af9186e17a996baf4676d97cdc9895abcb7e06
This script will be used to change the package that is being allowed to
use privileged permissions.
Test: python scripts/modify_permissions_allowlist.py
Bug: 242509786
Change-Id: I1d7cb3c906ad39aef42ad4394953251db40d1bf1
Teach rustc to generate object files instead of a linked executable
for binary and shared library crates. This lets us skip the lengthy
Rust compilation phase when only the C++ dependencies of these crates
have changed.
This works using a Python script that replaces the linker invoked by
rustc and converts the linker command line into a rspfile that can be
used during the link phase. The script makes certain assumptions about
how rustc invokes the linker, but since we control the version of Rust
we use, we can update the script if those assumptions ever break. I've
asked the Rust developers about adding an official way to do this so
hopefully we won't need to rely on this hack in the future.
The rustc rule for Kythe has been altered to pass linker=true, somewhat
consistent with the main rustc rule, which now doesn't invoke the actual
linker either. `m xref_rust` still succeeds, but hopefully we can find
someone who knows more about the rest of the Kythe pipeline who can let
us know whether it should still work. Otherwise, let's use this CL as
an opportunity to find out whether the rest of the pipeline is happy
with linker=true, since otherwise we will find it hard to maintain the
Kythe rule in the future.
Change-Id: Ifb9c8689c1bd78c0afcf6fdfd111140d20377536
```
ERROR: /buildbot/src/android/master/out/soong/workspace/external/libcxx/BUILD.bazel:30:18: While resolving toolchains for target //external/libcxx:libc++: com.google.devtools.build.lib.packages.NoSuchTargetException: no such target '@soong_injection//product_config_platforms/products/ndk-eng:ndk-eng_android_riscv64': target 'ndk-eng_android_riscv64' not declared in package 'product_config_platforms/products/ndk-eng' defined by /buildbot/src/android/master/out/bazel/output/external/soong_injection/product_config_platforms/products/ndk-eng/BUILD (Tip: use `query "@soong_injection//product_config_platforms/products/ndk-eng:*"` to see all the targets in that package)
```
Bug: https://github.com/google/android-riscv64/issues/10
Bug: http://b/274509603
Bug: http://b/262192655
Test: build/soong/scripts/build-ndk-prebuilts.sh
Change-Id: I4b6633ab5aab3f01953b628d78ac1d79f2a4ee3c
These will only be provisional, but there are enough different groups
that want *something* to make a start on porting that it's probably
time...
Ignore missing dependencies in build-ndk-prebuilts.sh for now, because
various pieces (that the NDK doesn't care about) are still missing.
Bug: http://b/273792258
Test: build/soong/scripts/build-ndk-prebuilts.sh
Change-Id: I39202a953a73dc449a10a887d497d09079c43402
This will enable its bp2build conversion, and is needed for java_system_modules support in Bazel
Change-Id: I4f3ff5e36c8cd7f78efbb42f641efb2f76a8b71d
Bug: 215230098
to cover the recent features.
- conv_linker_config proto with empty input
- conv_linker_config proto with existing output
- conv_linker_config proto with --append
- conv_linker_config proto with --force
Bug: n/a
Test: conv_linker_config_test
Change-Id: I0de79b6e05c2608e0e2f30dfbf04d8289672f362
`conv_linker_config proto -s` should work with multiple json input
files, but ParseDict() overwrites list fields (e.g. provideLibs), not
appending elements.
Also added a test.
Bug: 264330513
Test: conv_linker_config_test
Change-Id: Idc482f941201f15e5fc276c0ffc0dfeaa09d0cc2
`proto` sub command now fails when the output file exists.
This is to avoid accidental overwrite.
To handle the case properly, it support --force and --append.
This is to support amending /{partition}/etc/linker.config.pb in the
build process.
Bug: 244531518
Test: manual testing
Change-Id: I0af8c83015e485f2c7533221cae8caf6143403c8
PRODUCT_VENDOR_LINKER_CONFIG_FRAGMENTS lists json files for vendor
linker config. It's annoying to handle the case of empty list.
`proto` subcommand now works for empty input. This is useful to generate
the empty linker config.
Bug: 244531518
Test: conv_linker_config proto --source --output output.pb
Change-Id: Iec6de67a979814a818730c393d9a4a7ca5d2eebe
Tests that run against mainline modules generally need to be built
twice: once for aosp (unsigned) and once for internal (signed).
The override_android_test rule is really useful for generating a
signed version of the unsigned test.
However, mainline tests need to use a MainlineTestModuleController
to only run tests on targets with the expected module. Since unsigned
and signed mainline modules have different package names, this means
the test config needs to be updated when overriding a test.
Add a new "mainline_package_name" argument so that android test rules
can update the targeted mainline package. Without this, the only way
to create a mainline package test is to instead use android_test, which
results in a lot of copy-pasta.
Test: added tests to soong
Test: manual, using a locally-modified RkpdAppUnitTest
Test: test_config_fixer_test.py
Change-Id: Idaffd63f225719a2bfda41018fda630b17db0080
This is part of OpenJDK17 update. sun.misc.JarIndex was moved to
jdk.internal.util.jar.
Bug: 260718199
Test: m
Change-Id: I7107f980d6afe500acff88164b1e6cb04cded896
Previously, the signature_trie python library would reject classes
which started with a lower case character which in turn caused the
verify_overlaps tool to fail. That meant that it was impossible to
add obfuscated classes (which commonly used lower case characters
for their name) from being used on the bootclasspath.
This change removes that restriction and the accompanying test.
Bug: 265833521
Test: TH and partner testing
Change-Id: I70710484e427f64d79fb30301f3413f3b67b27e7
conv_linker_config proto -s input1.json:input2.json generates a merged
proto output file.
Bug: 264330513
Test: conv_linker_config proto -s input1.json:input2.json
Change-Id: I4806fc606d115c42a1c745fba67960ce6db8c8d7
Test: `m nothing && build/soong/scripts/run-ckati.sh FOO BAR`, see FOO
and BAR assignments in out/ckati.trace
Change-Id: Ife8bb6d716bfde5ee9f75b85b1e5738ab2238dce
A previous change to the deapexer script to check that blkid_path and
fsckerofs_path were set broke the scripts/unpack-prebuilt-apex.sh
scripts use of deapexer. This change fixes it by passing the paths
through to the script which then passes it on to deapexer.
Bug: 255963179
Bug: 240288941
Bug: 259958590
Test: m MODULE_BUILD_FROM_SOURCE=false droid
Change-Id: Ief7f36219b7fe1cf7104c08163e10d8e39f181d0
This script is responsible for generating <module>_using.txt, which is a
file containing all NDK symbols used by the shared libraries in an APEX.
This script doesn't work as-is in the bazel sandbox; namely it tries to
create tmp directories in read only input dirs, for example:
"mkdir: cannot create directory 'bazel-out/android_target-fastbuild-ST-7fe6d3cd3905/bin/packages/modules/adb/apex/com.android.adbd_staging_dir/tmpUnzipped': Permission denied"
The staging dir inputs are also symlinks, so the `find` call needs a new
-L flag to follow symlinks.
This CL fixes that by using a tmpdir, and also refactors the script to
be less fragile.
I'm not a fan of the camelCase, but I'll save that for another change.
Test: presubmits
Bug: 239081455
Bug: 257226023
Fixes: 257226023
Change-Id: I1d0fe93c47c92d046e72c0ab32014a5dc8bbd597
This allows go sources under subdirectories of prebuilts/ to be
referencable by soong_ui and soong_build. Without this change,
microfactory does not correctly include go libraries under prebuilts.
Test: Conjunction with aosp/2260126
Change-Id: I4cba6fb04d1e274af759fa92f9ae4b2e3ea18447
The supported scripts for doing that now live in
packages/modules/common/build.
Test: Check in CodeSearch that the script isn't actively used nor
mentioned in docs.
Bug: 180394948
Change-Id: I787cbe469bec2121f89de7b74695484310b00b5e
Even though the soong build system generates a bundle module for an
apex, the bundle module itself should be further processed to be an
AppBundle (.aab) file which can be uploaded to Play.
This script fills the gap by invoking bundletool to create an
AppBundle (.aab) file out of soong-built bundle module for an apex.
(Note: uploading APEX bundle (.aab) to Play is not supported yet.)
You can create an .aab file by:
- TARGET_BUILD_APPS={apex name} m dist
- m build-apex-bundle
- build-apex-bundle --output out.aab out/dist/{apex name}-base.zip
For now it creates a single-ABI APEX bundle. In the future it can be
extended to support multiple-ABI APEX bundles.
Bug: 236673372
Test: m build-apex-bundle
Test: TARGET_BUILD_APPS=com.google.cf.bt m dist
Test: build-apex-bundle --output bt.aab out/dist/com.google.cf.bt-base.zip
Change-Id: Id321efcd42c0fe60294a8348047c9ebbf7acf391
