This commit fixes `shared_libs` for `cc_prebuilt_binary`,
`cc_prebuilt_library_static`, and `cc_prebuilt_library_shared`. Before
this commit, all shared libraries in `shared_libs` are dropped from the
generated `LOCAL_SHARED_LIBRARIES`. This commit fixes the problem by
delegating `linkerDeps()` to `libraryDecorator.linkerDeps()`.
This commit also fixes the dependencies to NDK shared libraries. Before
this change, those dependencies are mapped to a `ndkStubDepTag` and then
ignored by the computation of `AndroidMkSharedLibs`. This commit adds
it back.
Bug: 123053270
Bug: 119084334
Test: libclang_rt.scudo-*.so can be checked with 2 more hacks.
Change-Id: I59e37e1a3fe0c329e8cb7032e5671f117f7832a0
libclang_rt.*.so depends on libc++, which is not a part of NDK. Thus,
this workaround must be added in order not to break the build when the
prebuilt dependencies are checked.
Bug: 121358700
Test: make checkbuild
Change-Id: Icaeb7adf96fb72829053e198b659ebcb19a035fc
This commit removes a break statement so that apex stub libraries are
added to LOCAL_SHARED_LIBRARIES for prebuilt ABI checker.
This CL doesn't change I09b78e38df285033ef6e9c85f7ea4b0274e85070 [1].
The libraries provided by apex modules are not installed to
`/system/lib[64]` because their stub libraries have
`LOCAL_UNINSTALLABLE_MODULE := true`.
[1] https://android-review.googlesource.com/844555
Bug: 120266448
Bug: 119084334
Test: Add `stubs: { symbol_file: "libnetd_resolv.map.txt", versions: ["1"]}`
to libnetd_resolv, create a program that depends on
`libnetd_resolve`, build the program, and do not find
`/system/lib/libnetd_resolv.so`.
Change-Id: Iffa2c1a0eac9c4940ec1fa05fbacb9806272c31e
This commit skips the ABI checks on LL-NDK-Private because
LL-NDK-Private libs are only used by other VNDK-core or VNDK-SP libs on
the system partition, are NOT used by vendor modules, and do not
constitute a system-vendor interface.
Bug: 122938657
Test: development/vndk/tools/header-checker/utils/create_reference_dumps.py
Change-Id: Ia2af4250ef1443f8ea3ed5ab111668462f120979
This change first introduces the interface "Sanitizeable" that
module types other than cc.Module can use to be handled by the sanitizer
mutator. APEX module, by implementing the interface, gets sanitizer
variants.
In doing so, sanitizer.go is refactored so that modules have explicit
dependencies to the runtime sanitizer libraries. This allows the runtime
library to be packaged into the APEX when required. This also completes
the dependency graph; updating the prebuilt sanitizer runtime will
trigger rebuilding of modules using the runtime.
Bug: 120894259
Bug: 121038155
Test: SANITIZE_TARGET=hwaddress m apex.test
Test: TARGET_FLATTEN_APEX=true SANITIZE_TARGET=address m
Change-Id: Ia91576ff48cda3c996350308b75bf83fcf7c23d7
This reverts commit 769e50b3c9.
Reason for revert: Likely causing build failure on aosp_x86_64-eng in aosp_master:
"build/make/core/base_rules.mk:271: error: art/build/apex: MODULE.TARGET.SHARED_LIBRARIES.com.android.runtime.host.libart-compiler already defined by art/build/apex."
Change-Id: I83b7caa04b2648e4e4914aae2fa5878516634eed
The unstripped file for libraries in APEXes are available via
out/target/product/<device>/symbols/apex/<apex_name>/path_to_lib.
This change make the symbol files available by installing the individual
files in APEXes to the directory where the APEXes will be mounted at
(i.e. runtime directory which is /apex/<apex_name>). Note that the files
are not actually packaged to a filesystem image; they are installed just
to create the symbol files under the out directory where developers can
use them for debugging.
Bug: 120846816
Test: m com.android.runtime.debug
There are unstripped files under
out/target/product/walleye/symbols/apex/com.android.runtime.debug
Change-Id: Ib182e2bf8787b7669ccba13814491db35370f468
If a lib is directly included in an APEX (via native_shared_libs
property) and the lib has stubs (via stubs.versions property), then the
ordinary non-stubs variant of the library is renamed to
<libname>.bootstrap in the makefile. At the same time, the stubs variant
of the lib becomes visible and it's name is <libname>.
This ensures that modules in Android.mk build against the stubs
variant thus preventing them from using private APIs in the lib.
The non-stubs variant, however, is used if the module explicitly has
set the new 'bootstrap' property to true. This is useful for building
some early binaries (such as init and vold) which need to run before
APEXes are activated. Since they can't use the bionic libs from the
runtime APEX, they should use the bionic libs left in the system
partition which is called the boostrap bionic.
Bug: 120266448
Test: m
Test: m with https://android-review.googlesource.com/c/platform/bionic/+/849044
Change-Id: I882b8aeb5b29460f07b4424e4f8eb844d6c9a9b0
This commit stops building lsdump files for APEX variants since APEX
variants are local to APEX modules themselves.
Bug: 121986692
Test: make findlsdumps # compare $ANDROID_PRODUCT_OUT/lsdump_paths.txt
Change-Id: I37fcd152d0d84d235a354ea53e53e808dd71464a
This change fixes following problem:
1) a native lib having stubs is defined.
2) the lib is included in an APEX.
3) a static binary is linking the lib from outside of the APEX.
4) then, the dependency from the binary to the lib is vanishing.
This is happening because cc.depsToPaths() mistakely does not
distinguish static lib deps from shared lib deps. For shared lib deps,
it creates two dependencies (one for stubs variant and the other for
non-stubs variant) and choose the stubs variant when the lib and the
current module is not in the same APEX (i.e. dependency to the non-stubs
variant is discarded). However, since we don't have stubs variant for
static library, it ends up having no dependency to the library if the
link is static.
Fixing the issue by skipping the variant selection routine when the link
is static.
Test: m (apex_test added)
Test: build with https://android-review.googlesource.com/c/platform/bionic/+/849044
Change-Id: I21102a31cc5c0b105da2affdd035bd5cc571a6ab
use_vendor, when set to true, brings vendor variant of the native
libraries and binaries to the APEX.
Bug: 115707625
Test: m (apex_test updated)
Change-Id: Ib4e996f8652f4ce4645a9c22f6914e2ab35edda6
The runtime APEX is built for host as well as for target. Therefore
stubs libs should be available also for host.
Bug: 120670568
Test: follow the repro step shown in b/120670568#comment4
Change-Id: I350fe490848ae9ceb55aade0521bdfaf48ed083f
This change fixes a problem that APEX variations are created for the
modules that actually shouldn't built for any APEX. For example,
consider this case.
apex { name: "myapex", native_shared_libs: ["mylib"],}
cc_library { name: "mylib", shared_libs: ["libfoo#10"],}
cc_library { name: "libfoo",
shared_libs: ["libbar"],
stubs: { versions: ["10"], }, }
cc_library { name: "libbar", ...}
Before this change, both the stubs and non-stubs variations of libfoo
were mutated with apexMuator, which is incorrect for the non-stubs
varia; there is no dependency chain from the apex "myapex" to the
non-stubs variation, but to the stubs variation due to the #10 syntax.
This was happening becauses we used the name of the module to determine
whether it should be built for APEX or not. Both stubs and non-stubs
variations have the same module name "libfoo".
Fixing this issue by recording the list of APEX variations required
directly on the module. So, the stubs variation of libfoo has myapex in
its apex variations list, but the non-stubs variation doesn't, and thus
apexMutator does not pick up the non-stubs variation.
Test: m (apex_test updated and passing)
Test: cherry-pick ag/5747464 and m
Change-Id: I31e618626809a828a55fff513ef5f81f79637afa
When the stubs variant of a library is dependend by a platform component
and the library is included in one or more APEX, the library is not
installed to the platform, because it is provided by APEX.
Bug: 120266448
Test: m
Test: add stubs: { versions: ["1"], }, to libnetd_resolv
then build netd. libnetd_resolv.so does not exist under /system.
Change-Id: I09b78e38df285033ef6e9c85f7ea4b0274e85070
When a native module is built for an APEX and is depending on a native
library having stubs (i.e. stubs.versions property is set), the stubs
variant is used unless the dependent lib is directly included in the
same APEX with the depending module.
Example:
apex {
name: "myapex",
native_shared_libs: ["libX", "libY"],
}
cc_library {
name: "libX",
shared_libs: ["libY", "libZ"],
}
cc_library {
name: "libY",
stubs: { versions: ["1", "2"], },
}
cc_library {
name: "libZ",
stubs: { versions: ["1", "2"], },
}
In this case, libX is linking to the impl variant of libY (that provides
private APIs) while libY is linking to the version 2 stubs of libZ. This is
because libY is directly included in the same apex via
native_shared_libs property, but libZ isn't.
Bug: 112672359
Test: apex_test added
Change-Id: If9871b70dc74a06bd828dd4cd1aeebd2e68b837c
VNDK libs are system defined libraries. They must not be product
specific. Adding sanity check and a test to prevent setting
`product_specific: true` on VNDK libs.
Bug: 119575107
Test: building a vndk lib with 'product_specific: true,'
must return error.
Change-Id: Ie0326540a692573f076ee08baf5d2e2f09d1007e
AArch64 execute-only memory is only supported when using lld as the
linker. There's still a few modules which don't use lld, so in those
cases we need to disable this option.
Bug: 77958880
Test: Module with use_clang_lld false builds without XOM
Test: Module without use_clang_lld defined builds with XOM
Change-Id: I4ab961c4d7342c54c6b40b9facfe18a45ed883bd
Before it just mentioned 'double_loadable', but in this context,
LL-NDK and VNDK-SP libraries also work.
Bug: 119790516
Test: N/A
Change-Id: Ie09a959fd5e05b9cb73db30d3cc2853694577dfe
Adds build system support for generating AArch64 binaries with
execute-only memory layouts via a new xom module property. Also adds
support for an ENABLE_XOM build flag for global builds.
Bug: 77958880
Test: make -j ENABLE_XOM=true
Change-Id: Ia2ea981498dd12941aaf5ca807648ae37527e3ee
Shadow call stack (SCS) is a security mitigation that uses a
separate stack (the SCS) for return addresses.
The effect of setting sanitize.scs on a shared library is to build
the library and all of its static library dependencies with SCS. This
is similar to CFI and the other sanitizers.
Bug: 112907825
Bug: 119557795
Change-Id: I82fb2b38b10eac911c4d2d120b74fea4af0622ad
This reverts commit 555c114283.
Reason for revert: The namespace issue in the pi-dev-plus-aosp-without-vendor branch is now fixed.
Change-Id: I26ed591447797a8ee505f43bdd209162418b6c5e
Arch property struct types are created at runtime. Go has a limit
of 2**16 bytes for the name of a type, and the type of a struct
created at runtime is a string containing all the names and types
of its fields. To avoid going over the limit, split the runtime
created structs into multiple shards.
Also undo MoreBaseLinkerProperties now that it is no longer
required.
Bug: 80437643
Test: m checkbuild
Test: no change to out/soong/build.ninja
Change-Id: I035b20332ec63f3d4b1696855c5b0b0a810597b7
llvm-ar takes a --plugin argument but it is ignored for compatibility,
so passing it has no effect.
Change-Id: I4fc51d226d66cf2a43462d3d4ccc12e6e5ebb226
Export static libraries through LOCAL_STATIC_LIBRARIES and
LOCAL_WHOLE_STATIC_LIBRARIES. This enables dependency-based NOTICE file
generation. Also, add a notice property in the libwinpthread module.
Bug: 36073965
Test: cc_test.go
Change-Id: If1ca1f9159e80cf8fbe71df7a13ca5d6a1f63b40
Export static libraries through LOCAL_STATIC_LIBRARIES and
LOCAL_WHOLE_STATIC_LIBRARIES. This enables dependency-based NOTICE file
generation. Also, add a notice property in the libwinpthread module.
Bug: 36073965
Test: cc_test.go
Change-Id: Ic63ca523b40acac82bbe876f7aa40ecd495907c5
A cc_library or cc_library_shared can be configured to have stubs
variants of the lib.
cc_library_shared {
name: "libfoo",
srcs: ["foo.cpp"],
stubs: {
symbol_file: "foo.map.txt",
versions: ["1", "2", "3"],
},
}
then, stubs variants of libfoo for version 1, 2, and 3 are created
from foo.map.txt. Each version has the symbols from the map file where
each symbol is annotated with the version that the symbol was introduced
via the 'introduced=<ver>' syntax. The versions don't need to be in sync
with the platform versions (e.g., P for 28). The versions are local to
the library.
For another library or executable to use the versioned stubs lib, use
the new 'name#ver' syntax to specify the version:
cc_binary {
name: "test",
....
shared_libs: ["libFoo#2"],
}
Internally, a new mutator 'version' is applied to all cc.Module objects.
By default, a variant named 'impl' is created for the non-stub version.
If the versions property is set, additional variations are created per a
version with the mutable property BuildStubs set as true, which lets the
compiler and the linker to build a stubs lib from the symbol file
instead from the source files.
This feature will be used to enforce stable interfaces among APEXs. When
a lib foo in an APEX is depending on a lib bar in another APEX, then bar
should have stable interface (in C lang) and foo should be depending on
one of the stubs libs of bar. Only libraries in the same APEX as foo can
link against non-stub version of it.
Bug: 112672359
Test: m (cc_test added)
Change-Id: I2488be0b9d7b7b8d7761234dc1c9c0e3add8601c
In order to simplify the wrapper function, and stop using a linker
script, generate a set of flags to pass to LLD. Then run
host_bionic_inject on the linked binary in order to verify the
embedding, and give the wrapper function the address of the original
entry point (_start).
Bug: 31559095
Test: build host bionic with prebuilts/build-tools/build-prebuilts.sh
Change-Id: I53e326050e0f9caa562c6cf6f76c4d0337bb6faf
ApexModule is the interface for APEX-aware modules. The module type apex
uses the interface to get APEX-specific information from other modules,
such as the list of APEXs that a module should be built for.
A module that is included in an APEX will be built specificaly for the
APEX. This is especially required for shared libraries; we shouldn't
just copy the artifacts built for platform, because they may be linking
against private (=unstable) symbols that are not available to APEXs
which are basically unbundled.
This CL, as a first step, makes cc.Module an APEX-aware module type.
Bug: 112672359
Test: m apex.test; the built apex has all the direct and transitive
shared lib dependencies of the libs and executables listed in Android.bp
Change-Id: I21f6a586654779984f0f5154b2a08b2adbf2168b
Clang is always used now, so we can remove all the GCC checks. Removing
GCC-specific configuration will happen in the next CL.
Test: m
Change-Id: I4835ecf6062159315d0dfb07b098e60bff033a8a
JNI testing will need to create basic native shared library
modules, export the minimum mutators and module types for
the required dependencies of a native shared library.
Bug: 80095087
Test: cc_test.go
Change-Id: Ibe7bc88b69cb0851291cb09a4c0c6cdb421b8651
prefer32 needs to be set differently for app and native modules.
Make it use lambda provided by the module type instead of trying
to make archMutator figure it out.
Test: m checkbuild
Change-Id: Ibf8af35fdd3e1721725539d1f5452f4439d2125c
APEX needs to know the location of the output file of cc.Module to copy
it to an intermediate directory and package there.
Bug: 112672359
Test: m apex.test
Change-Id: Iaa19b29c2859df96f50716001dbd57df3f596050
Export Soong cc modules to Make using a new soong_cc_prebuilt.mk that
bypasses all of prebuilt_internal.mk, dynamic_binary.mk and binary.mk.
This also means that stripping is handled in Soong instead of Make.
Relands If9008c50920779048480f5eeeb0084f26006c998 with fixes for
mac builds.
Bug: 113936524
Test: m checkbuild
Change-Id: I9710ff57f0793f36eb889eabd08bd60a365a88dd
Export Soong cc modules to Make using a new soong_cc_prebuilt.mk that
bypasses all of prebuilt_internal.mk, dynamic_binary.mk and binary.mk.
This also means that stripping is handled in Soong instead of Make.
Bug: 113936524
Test: m checkbuild
Change-Id: If9008c50920779048480f5eeeb0084f26006c998
We need to keep support in Soong until toolchain_library no longer uses
GCC.
Test: out/build-aosp_arm64.ninja is the same before/after
Test: build_test on downstream branches
Change-Id: Ib78bcd6d5544afc3a3164f65fa72a3f08b2acf29
This commit extends ABI stability checks to all NDK shared libs
(including the framework-related ones such as libandroid.so).
Bug: 112404572
Bug: 79576032
Test: make findlsdumps # (and check the output)
Change-Id: I0147c60ce0c90d187f85b996911d98326a0c37ae
Dependencies of disabled modules may not have the right variations
for calling AddVariationDependencies, skip DepsMutator completely
on disabled modules.
Bug: 112707915
Test: m checkbuild
Test: prebuilts/build-tools/build-prebuilts.sh
Change-Id: I702e591437e5e6eb9f91f3b7eb32bacc4bd5d249
ctx.AddDependency will succeed if the named dependency only has a
single variant, even if that variant is the wrong architecture.
Use ctx.AddVariationDependency(nil, ...) instead, which requires
that all variations of the calling module match the dependency.
Bug: 112707915
Test: no change to out/soong/build.ninja
Test: using a device dependency in a host java module is an error
Change-Id: I70b661a57d4412eb63b8c9841febfb756e9e025d
This CL fixes a bug that when a module is configured as 'vendor: true' &&
'recovery_available: true', the link type of the recovery variant of the
module is incorrectly set to 'native:vendor'. This was because,
androidmk.go emits 'LOCAL_PROPRIETARY_MODULE := true' whenever
Proprietary property is set to true, regardless of whether it is a
recovery variant or not. This in turn makes LOCAL_USE_VNDK := true for
the module which in turn causes the link type to be 'native:vendor'.
Fixing the bug by resetting the properties like Proprietary, Vendor,
Soc_specific, etc. for the recovery variants.
Bug: 113277544
Test: m -j (test added)
Change-Id: I5d6ae76e46ef8fcd9204d386d0809862a7b0ff7e
This is a new sanitizer similar to ASan, but with a few differences
from the build system perspective:
* Only runs on AArch64.
* Supports static binaries.
* Bionic itself will be built with HWASan.
* Does not have any "if a library is sanitized than the binary must
be, too" requirements unlike ASan. Even better, individual static
libraries can be freely sanitized or not. We propagate "nosanitize"
from binaries to static libraries anyway, because otherwise there
is no good way to disable hwasan on a binary as a whole.
Same a CFI, we export a list of sanitized static libraries to make.
In fact, we export separate lists for vendor and regular libraries,
because it is possible for one to be sanitized without the other
(i.e. there can be mylib.hwasan.vendor w/o mylib.hwasan or vice
versa).
Bug: 112438058, 112709969
Test: manual, part of a bigger patch set
Change-Id: Ie4fdeb522ac03cf9684526882e84dfee3807b6a7
Most everything in Soong is using clang (and we're getting very close
for Make as well), so remove the ability to select GCC to build a
module.
We still need to keep the internal functionality around to find libgcc.a
and other toolchain_library modules with GCC.
Test: m blueprint_tools
Change-Id: I5ef3e3836b9ad3e160669ac32aee39698c9b72c3
This commit cleans up `createVndkSourceAbiDump()` and renames it to
`shouldcreateVndkSourceAbiDump()`.
Test: lunch aosp_walleye-userdebug && make
Change-Id: Iff4379e2812c4b5c5baff288b938eed5d92e024f
