Rewriting LLNDK dependencies with .llndk suffix requries referencing
a global data structure to determine if a given library is an LLNDK
library and therefore needs the .llndk suffix. References to
global data structures from mutators must be removed to support
incremental Soong analysis. Instead, move the LLNDK stubs rules
into the vendor variant of the implementing cc_library so that
the original name can be used.
As an incremental step, the llndk_library modules are left in
place, and the properties are copied into the cc_library via
the dependency specified by the llndk_stub property. A followup
will move the LLNDK properties directly into the cc_library and
delete the llndk_library modules.
The global list of LLNDK libraries is kept for now as it is used
to generate the vndk.libraries.txt file.
Bug: 170784825
Test: m checkbuild
Test: compare Soong outputs
Test: all Soong tests
Change-Id: I2a942b21c162541a49e27b2e5833c9aebccff1d0
Use WriteFileRule instead of a rule that calls printf for generating the
module source file for rust_protobuf.
Bug: 171361369
Test: rust_protobuf modules compile
Change-Id: Ie83a30e19d7d09875f3d209662526783862d8ea0
We don't have Rust VNDK support yet, but static linkage can be
supported in the interim. This adds support for making rust_ffi_static
libraries available to CC vendor modules.
Since rust_ffi_static modules will link against rlibs, we allow rlib
linkage into vendor as well, but only for the variants which use the
rlib libstd.
Bug: 172525289
Test: New Soong tests pass
Test: Example vendor cc_binary links against rust_ffi_static module.
Change-Id: Idf3aeb51e32293866f1ad965e329aa6b9e0bf2ef
Refactors parts of CC to prepare for preliminary support for using Rust
static libraries in vendor images. Some previously private functions are
made public, and additional functions are added to LinkableInterface so
GetMakeLinkType can be passed a LinkableInterface.
Bug: 172525289
Test: m
Change-Id: I5fda48e79532fe9ceab255e18d910af58048a123
Add the min_sdk_version property to Rust modules so they can declare
a minimum SDK version they support for use with APEX modules.
Test: New Soong test passes.
Bug: 174862583
Change-Id: I2829053a320f50c218783dee5adbeff9cef81e8e
Previously, the dep tag used by PackagingBase was fixed, which prevented
some of its clients (e.g. cvd-host-package) from opting in to
android.InstallAlwaysNeededDependencyTag. Now, PackagingBase.AddDeps
accepts the dependency tag to use.
Also, dependencies toward rust dylib, rlib, and proc_macro are
configured to return true on InstallDepNeeded. This is required to
install shared_lib dependencies of the rust modules when they are
depended on by a rust binary.
Exempt-From-Owner-Approval: a trivial change after +2 from the owner.
This has to land ASAP as many users are affected by acloud being
unavailable.
Bug: N/A
Test: m
Test: acloud create --local-instance --local-image
Change-Id: If22aee7c6f314fcb03b9d4fe6901a2557f1e899c
We will have some APEXes having rust binaries and libraries. So, adding
the support for the types of modules.
rust.Module now inherits from android.ApexModuleBase and implements
the android.ApexModule interface.
Bug: 172414324
Test: m
Exempt-From-Owner-Approval: rebased after +2 from the owner
Change-Id: I356ef4c45f782a6460f001e83af96d1710642d80
The usage of the well known type Empty requires a hack in the module
above the grpc implementation, this is now the generated stem_mod.rs
This also adds additional implicit dependencies that are required by
the grpc protobuf generated code. This includes the addition of a
'header_libs' property for library dependencies which export include
paths required by protos.
We also now include both the protos and the grpcio in the library
variant via the mod_stem.rs.
Bug: 172592789
Bug: 171504899
Test: m nothing
Test: Example rust_grpcio module build command includes dependencies,
include paths.
Change-Id: I187a13cd5cdea991828a1020314de16727e4f74e
The stubs methods are very specific to cc for now, move them out
of LinkableInterface so they are not shared with rust. Instead,
create a cc.Module.library field that contains the libraryInterface
to simplify calling libraryInterface methods on cc modules.
Test: all Soong tests
Test: no change to Soong outputs
Change-Id: I0289d866ce1f7a765631fe3101a62b1b4988ba1c
Add vendor_ramdisk_available and vendor_ramdisk attribute to
various rules. When a vendor_ramdisk variant of a module is
generated, it is installed to $OUT/vendor-ramdisk.
It is similar to a ramdisk image.
Test: m nothing -j
Change-Id: Ib2d16459f3094dbe21c3bdb7c016cb4b2bf62765
Implement stubsVersions on *llndkStubDecorator and *stubDecorator to
handle the special cases in versionSelectorMutator.
Test: m checkbuild
Change-Id: Idc985c52f91450df42c0275b2b2acef3f2ed8868
Export information about static libraries, shared libraries and
exported flags through Providers instead of accessing the module
directly. Much more is left to be converted, but this significantly
simplifies the dependencies on libraries with stubs by making it easy
for a module to masquerade as another by simply exporting the
providers from the other module. Instead of depending on all the
versions of a library and then picking which one to use later, it
can depend only on the implementation variant and then select the
right SharedLibraryInfo from the variant.
Test: m checkbuild
Test: only expected changes to build.ninja
Change-Id: I1fd9eb4d251cf96ed8398d586efc3e0817663c76
* Upgrade to new protobuf-codegen; use its standard
default output without local change.
* Allow a sourceProvider to have multiple output files.
For a stem.proto file, output stem.rs and mod_stem.rs.
* New protobuf-codegen option gen_mod_rs always generates
output file named "mod.rs". To generate multiple .proto
files into the same output directory, we need to rename
mod.rs to mod_<stem>.rs.
* Instead of using the gen_mod_rs option and renaming mod.rs
to mod_<stem>.rs, we generate the same mod_<stem>.rs
directly with a simple printf command.
Bug: 170256643
Test: atest -c --host --include-subdirs external/crosvm
Change-Id: Ia09e41029099a6de4d35c96dbabd9ba5514c9019
* changes:
Don't create version variations of sdk modules
Create fewer empty version variants
Don't make SplitPerApiLevel imply UseSdk
Remove vendor crt special case
Fix apex_test.go and add it to Android.bp
Don't create empty version variants for binaries, objects, rust
rlibs or rust dylibs.
Test: no change to build.ninja
Change-Id: I62d4d43da476eafdb258a08b5ada758bb2971a1a
UseSdk was returning true when SplitPerApiLevel returned true,
which was causing the platform variant of SplitPerApiLevel
module to compile against the SDK. Check SplitPerApiLevel
separately in the sdkMutator instead.
Test: m checkbuild
Change-Id: I0ae667d48a3b7b96709a6cad8e8ea9701659fc2a
Instead of returning a boolean, return an enum value to improve
readability and provide greater flexibility for future modifications.
Bug: 168729404
Test: Soong tests pass
Change-Id: Iddcdae8c34be09e476404382e43d1ea5935bae65
There is no need to export the linkdirs from the SourceProvider
variant. Remove them to reduce differences in build.ninja from
a later patch that moves the flag exporting into compile(), which
isn't called by the SourceProvider variant.
Test: m checkbuild
Change-Id: I9c4d3a336a07cb9074376303bfa277c05d893b98
rust_bindgen modules can't inherit properties in cc_defaults that would
be useful for generating bindings (such as cflags). This CL moves these
common properties out into a new struct in cc and adds that struct to
cc_default.
Additionally, Cppflags is added to rust_bindgen to make sure that these
get picked up as well from cc_defaults.
Bug: 163598610
Test: rust_bindgen module uses cflags in cc_defaults.
Test: New Soong test passes
Change-Id: I702442a355244dc01954083f98a2eebbcea12e47
Until now, source provider modules duplicated the rule to generate the
source for each variant. Add a inter-variant dependency between the
source and the other variants (e.g. rlib, dylib) to avoid this
duplication. Add documentation on this behaviour.
Bug: 162588681
Test: m
Change-Id: I41c9e2220f8875245415e17374852e540dfd47ec
The current state of linkage is that device targets always link
libstd dynamically except for rust_ffi_static which requires a static
libstd linkage. However this prevents producing rust_ffi_static
modules which depend on other Rust libraries as those dependencies
will link libstd dynamically and cause a collision. We also want our
rust_test modules to statically link in libstd as well.
This adds a linkage mutator for rlibs that creates a variant for each
libstd linkage. Dependent modules can then select the variant that
matches their linkage of libstd.
Also fixes an issue where installation paths were being generated for
rlibs and static libs even though they weren't being installed. This broke
when adding the linkage mutator as Make would complain about multiple
targets producing the same output.
Bug: 168729404
Test: rust_ffi_static module with other rustlib dependency can be built.
Change-Id: I955b484bf5809e8fc5517750c7f8df82d3ca8895
A per-context variable is used to store the list of modules that
contain stubs and their available versions. Stores the list of the
stubs versions on the implementation module, and then use the new
return values from AddVariationDependencies to expand dependencies
on implementation libraries to also depend on the stubs libraries.
Adds a new mutator pass to propagate list of stub versions to llndk
libraries.
Also creates an alias version variation called "latest" to allow
depending on the latest version without having to know what it is.
Test: all Soong tests
Test: no change to build.ninja, Android-${TARGET_PRODUCT}.mk, make_vars-${TARGET_PRODUCT}.mk or late-${TARGET_PRODUCT}.mk
Change-Id: If19659e2e5828c860fd4d679ef79a414b7ea2efc
With If4300583edfb6d75bd4d984b38f73b6a406b4447, linux_bionic supports
arm64 and rust toolchain was configured to support the new target.
However, rust module types still have been disabled for linux_bionic
targets which forces many rust modules to be annotated with
target.linux_bionic_arm64.enabled: true, which will be very cumbersome.
Therefore, stop disabling the rust modules for linux_bionic.
Bug: 159685774
Test: HOST_CROSS_OS=linux_bionic m nothing
Change-Id: I68b97a43e8252fb82da6e87b3481fa237e1f2691
rust_bindgen implicits currently include dependency-exported include
directories. This is incorrect, and causes issues when the include
directory is a generated directory.
Instead, remove the dependency-exported include directories from the
implicits list, and also add dependency-exported headers to the
implicits to handle these correctly.
Bug: 166779501
Test: Rust projects continue to build, test-case no longer failing.
Change-Id: Ia2da25de9e712b7306f19603dc9d14de6bac3fb3
Reuses the cc.Stripper logic. Abstracts Stripper to avoid the spreading
of references to the cc package.
rustc requires unstripped libraries (precisely, with the `.rustc`
section) when building dependent targets. Contrary to cc, the output of
a compiler module will remain unstripped and only an extra build rule
will be added. This rule will be referenced at install time (in
baseCompiler.install or androidmk).
This change drastically reduces the size of the installed libraries:
(unstripped, from out/target/product/crosshatch/system)
$ find . -name \*.dylib.so -print0 | du -c --files0-from=-
149996 total
(stripped, with this change)
$ find . -name \*.dylib.so -print0 | du -c --files0-from=-
42380 total
Bug: 153430439
Test: cd external/rust; mma
Change-Id: I94fd8bbcec97e0610aa325d3db4460be84d01734
The cc toolchain information is required when using cc.Stripper.
Move to a Module method to avoid importing the cc package everywhere.
Test: m nothing
Bug: 153430439
Change-Id: I497dcdff4e767ecb8b8688c73cf88850c302683d
In order to support cc dependencies which do not start with the 'lib'
prefix, we can't pass them through the -l flag. Instead, we can pass
them directly to linker flags.
Bug: 166151658
Test: cd external/rust/; mma
Test: cd external/crosvm/; mma
Test: Test linking to a cc dep that does not begin with 'lib'
Change-Id: I5acbf3d3405e66446f3eae600b35683c4eb3d8a5
Device binaries currently are linked statically by default. Instead we
should be linking these dynamic by default. To avoid conflicts when
manually specifying rlib dependencies on modules, we always link libstd
dynamically for all device modules except static libraries.
This removes the "prefer_dynamic" property entirely to avoid confusion.
Bug: 165161105
Test: m profcollectd is built dynamically.
Test: cd external/rust/; mma
Test: cd external/crosvm/; mma
Change-Id: I25ac897040acbcc2d97c791a33e8e01610632272
Move the linting properties to an enum with 4 possible options:
"default", "android", "vendor" or "none". The previous logic for
default, based on the module's location, is kept. It is now possible to
force the upgrade to a certain lint level for some modules (e.g.
external/[...]/android). Update the unit tests and documentation.
Bug: 163400111
Test: m
Change-Id: I8e464b04401158ed2d3c518a9b72f145a9835c99
We don't need the prebuilt versions. The NDK CRT objects are (now)
built from the platform sources and the only difference is that the
NDK CRT objects also include an ELF note that identifies the NDK
version, which isn't helpful for anything built by the platform.
Add a `crt` property to cc_object that allows CRT objects to identify
themselves. CRT objects, unlike other modules, will have a variant
built per-API level they support, rather than just an SDK variant and
a platform variant. This is needed because new CRT objects will rely
on APIs not available in old libcs and old CRT objects will not
support all the features of a modern one.
Test: treehugger
Bug: http://b/159925977
Change-Id: I6595485fa1bfe0ad4945193d344b863f64eec654
Colin Cross