recovery variants are not built with NDK (useSdk() is false for
recovery). So LOCAL_SDK_VERSION shouldn't be emitted for the recovery
variants as well.
Test: m -j liblzma.recovery does not show any link type check error
Change-Id: Icb6c4fc766060e3f1555c405d5785ef5a015c629
Add a module type for compiling .c to .o using clang --target=bpf.
Bug: 72981744
Test: m
Change-Id: Ie5712bfabe34901a84ed33737053306ca56ebeb4
Merged-In: Ie5712bfabe34901a84ed33737053306ca56ebeb4
(cherry picked from commit 3ba736a7e1)
Add core-lambda-stubs to the bootclasspath for modules with no
sdk_version or with sdk_version: "current", and to the classpath
for modules that specify a specific sdk_version number.
Fixes compiling modules with lambdas against the SDK.
Bug: 80428539
Test: java_test.go
Change-Id: I5e700f2dd86f1a6b84b7a55dd9bfe21a448d3fb6
When pack_relocations is false and clang lld is used,
pass --pack-dyn-relocs=none to lld.
Bug: 80093890
Test: build and boot with USE_CLANG_LLD=true
Change-Id: I0ffe77a111d7fbab5afaa1395d09734a8a390e09
The aidl files are allowed when creating the droiddoc module in
java_sdk_library. But The properties of AIDL, local_include_dir
and include_dir, are not passed to the droiddoc module.
These properties are add because they are needed to generate the
java files.
Bug: 77575606
Test: build && java_test.go
Change-Id: I1006c9ce7a5c7e424416d39fb921c27a3fa86afa
This reverts commit c59a92cb1a.
Reason for revert: tests are broken with long OUT_DIRs
They're directly calling net.Listen, and not using the fallback
for long socket names.
Change-Id: Id14cbd499fd9b36c6926b7552d3554340cb0916c
This reverts commit 96c957ae20.
Fixes issues on some machines where the socket in TMPDIR ended up with a
unix domain socket pathname over 107 characters long, which Go will
reject due to underlying limitations in the system calls. If this
happens, we'll fall back to opening the directory, then using
/proc/self/fd/#/<file>, or manually creating a similar symlink in /tmp.
Also fixes some issues on Mac where os.Executable returns the symlink
instead of the underlying file, sending a message over a unix domain
socket will block if the reader isn't reading, and sandboxing was
preventing us from running `ps`.
Test: m blueprint_tools
Test: m blueprint_tools on mac
Change-Id: Ib19ccfe10cb0a79f1476fb1d5cd20ed0495be367
There was a bug that recovery.cflags is ignored. Respect the property
when building for recovery.
Bug: 79146551
Test: m -j
Change-Id: I24c1c1b593c83eb1992721571c6d5c42483a41a0
I5d439353d46ba6659ec8d16967693d2b9e62ab5a will need the manifests
from android libraries to merge into modules that reference them
with LOCAL_STATIC_ANDROID_LIBRARIES.
Bug: 3434784
Test: m checkbuild
Change-Id: I47b9def5b9c2ecc423550eb8a988cce96038a81e
The current rule runs a specific preprocessor over the source files,
and I'm adding support for generic preprocessed headers in a follow
up patch.
Test: make checkbuild
Bug: None
Change-Id: I1f3193cd595f151309e6321e1b41d8d16085379d
android_test module are APKs that can be run as tests, either
as standalone unit tests or as instrumentation tests for another
APK.
Test: m checkbuild
Change-Id: I16661701637e4048fd99442029c3e195ebf373a4
Libraries that are direct or indirect dependencies of modules installed
to recovery partition (e.g. adbd) are marked as recovery_available:
true. This allows a recovery variant of the lib is created when it is
depended by other recovery or recovery_available modules.
Bug: 79146551
Test: m -j
Change-Id: I1181fa0646d840c0194e4af2043ec228ec9eff16
recovery_available property is required in cc_genrule. Specifically,
we will mark libminijail as recovery_available:true as part of building
adbd and other stuffs for recovery. Some source code of libminijail is
created via cc_genrule, so we need recovery_available in the module
type.
Bug: 79146551
Test: m -j
Change-Id: I0cf0d9b1004dda055373573e5c5a7debd112685f
This reverts commit bfb0aed2dd.
Reason for revert: <LOCAL_REQUIRED_MODULES overrides the "required:" property in Android.bp>
Test: m -j sepolicy_tests
Bug: b/80140289
Change-Id: I4e51007a7d4dda69cbf4f31253e9a56d3f5e35c7
droiddoc now accepts aidl files as inputs. This in turn allows us to
feed aidl files to java_sdk_library modules. This is required as some
java_sdk_library internally uses AIDL files and thus we need to specify
*.aidl files in the srcs property. Since the srcs property is internally
given to the droiddoc module as well as the runtime library, droiddoc
should be able to handle aidl files.
Bug: 77575606
Test: java_test.go
Change-Id: If7a8559a2a1d8ac1056b061d24e3a5ee5253453f
As I mentioned in b/79751992 #4, when launcher loads __main__ as entry point,
the module dict got set using absolute pkg path which
was runfiles.google.protobuf which caused the type checking failed
(expected: google.protobuf for type()). After then, starting with non entrypoint
python files, the importing module dict would be set with from...import...
So remove dir:runfiles/ from zip section so that during entrypoint Python file
importing the proto files using
from...import..., the type checking can be passed.
Bug: b/79751992
Test: m clean && m out/host/linux-x86/bin/perf_proto_stack;
And execute it:
usage: perf_proto_stack [-h] [--syms SYMS] [--json-out JSON_OUT]
[--print-samples] [--skip-kernel-syms]
[--print-pid-histogram] [--print-sym-histogram]
[--print-dso-histogram] [--parallel PARALLEL]
file [file ...]
perf_proto_stack: error: too few arguments
Change-Id: I95a4a9ee95ce7f90d893ea0529c384d7b944abe4
Add a platform_apis property. The logic to verify it isn't
hooked up yet, but this will allow translating modules to
Soong without losing the annotation.
Test: m checkbuild
Change-Id: I75ff2d7b94247873f5bb1547a9b347a8e73c3191
LOCAL_PROGUARD_FLAG_FILES should convert to
optimize.proguard_flags_files, not optimize.proguard_flag_files.
Test: m checkbuild
Change-Id: I1471afc006a2a3892af37d2cb73c51636f05d7f7
After https://android-review.googlesource.com/688488
BOARD_PLAT_PUBLIC[PRIVATE]_SEPOLICY_DIR can now specify multiple
directories.
Bug: n/a
Test: build sepolicy
Change-Id: I838c76736a3f1bd8759e24739cf82fea33a531b4
Bug: 78133793
Test: Built target with cortex-a53 and verified that the flag is still
Test: there.
Test: Build target with cortex-a55/cortex-a75 and verified that the flag
Test: is not there.
Test: Built and booted on a taimen, ran bionic unit tests.
Change-Id: Ibd7f497ecf46e5781cd0265463b0782babdf062a
Merged-In: Ibd7f497ecf46e5781cd0265463b0782babdf062a
(cherry picked from commit a98d721485)
This will allow us to track (and eventually limit) the commands that the
build references via $PATH. These are mostly implicit dependencies on
the host system -- for Linux, we assume something similar to Ubuntu
14.04 with a few extra packages, but this will let us better define
that.
This will not catch uses of tools with absolute paths (/bin/bash, etc),
but most uses shouldn't be relying on absolute path names anyways.
Adds ~400ms on the first startup, ~140ms on subsequent runs, and
overhead of a few ms for every forwarded execution.
Test: m
Test: build/soong/build_test.bash
Test: Add `gcc --version`, TEMPORARY_DISABLE_PATH_RESTRICTIONS=true m
Change-Id: Id68cbb1c8ceef65bbbb10751e83722c7662d2351
-check-all-apis currently, produces noise for some libraries
(eg:libstagefright_soft_vpxenc), so remove this till a good solution
arrives.
Bug: 79928919
Test: create abi reference for libstagefright_soft_vpxenc; m -j
libstagefright_soft_vpxenc.vendor; no warnings about struct LAYER_CONTEXT
layout changing.
Change-Id: I8f30700c1b619c72504ca9aef69075f6e334ba5f
Jiyong Park