Bug: http://b/74395273
Update profile-search logic to look for profile files named
<profile_file>.<arbitrary-version> in the profileProject and use that if
available. This works around an issue where ccache serves stale cache
entries when the profile file has changed.
Test: touch toolchain/pgo-profiles/art/dex2oat.profdata.foo and check it
gets used.
Change-Id: I3de095c263b952cede25af7d0b63dc7280914d76
Merged-In: I3de095c263b952cede25af7d0b63dc7280914d76
(cherry picked from commit 8aed42c798)
Vendor binaries cannot directly link with vndk libraries which are not vendor
available. We don't need them to be abi stable.
Bug: 66301104
Test: cd external/cblas; touch src/cblas_cher.c; mm -j64 does not
produce an lsdump for libblas, since it isn't vendor available,
even though it is vndk_enabled: true.
Change-Id: Ib2e1eaa06c4dc2e05623bd7b9aa3a83010f76bd1
Merged-In: Ib2e1eaa06c4dc2e05623bd7b9aa3a83010f76bd1
(cherry picked from commit 22963cda0e)
Bug: http://b/73127367
Bug: http://b/68659946
These new prebuilts have fixes for the bugs mentioned above.
Test: Build and verify that issues in both bugs above are fixed.
Change-Id: I410a43b70920d509118f8e6fdb012ef6b078d657
Merged-In: I410a43b70920d509118f8e6fdb012ef6b078d657
(cherry picked from commit 4a825ee14a)
Allow native modules to specify use_version_lib, which will make
an android::build::GetBuildNumber() function available. For host
builds, the function will return the build number at the time that
the module was linked. For device modules it will return the
value of the ro.build.version.incremental property.
Bug: 71719137
Test: build_version_test
Test: m build_version_test && touch build/make/core/Makefile build/soong/cc/libbuildversion/tests/build_version_test.cpp && m build_version_test shows different build numbers for binary and library tests.
Change-Id: I6f7d40b7574bb8206866c4e39bad9c710c796e32
Parse the machine-specific symbol tables into a custom symbol table
object and use a single function to find the offset and size of the
symbol in the custom table. Make the functions to convert a
machine-specific symbol table into the custom table also take mocks,
and provide functions to dump mocks from real files. Add tests
that take the mocks and verify they produce the right offset and
size.
Test: cmd/symbol_inject/*_test.go
Change-Id: I25654032b5017dd13a9a1fe29f8b8826ce5bc82b
asan variants can sometimes have extra exported symbols (this is a vndk
extension violation). Since asan variants are only used for testing,
don't do abi diffs for them.
Bug: 66301104
Test: Without the change, libc++.so.lsdump exists for the asan variant;
with the change, it does not.
Test: make -j64.
Change-Id: I61a4c7a3e9aa0028a54ad0ca8715e8c77aebad94
This is needed so that we can embed the paths of lsdump files in a file,
in order to make lookup faster, while creating reference dumps.
Test: mm -j64 in external/libjpeg-turbo for aosp_arm64_ab;
out/soong/make_vars-aosp_arm64_ab.mk contains
SOONG_LSDUMP_PATHS := <list of lsdump paths>
Change-Id: I04608429e1add307cc1ee79d2f0c348fb041613c
binder_size_t has a different size for builds with different primary
arches.
Also maintain seperate reference dumps for different arch variants,
since different cflags may be legally specified for them (similar to
what GSI does)
Test: create reference dump for libjpeg at
prebuilts/abi-dumps/vndk/current/arm64/arm64_armv8-a/source-based/libjpeg.so.lsdump.gz
mm -j64; header-abi-diff gets invoked.
Change-Id: I55eae4d4811c9754fe8dbd1009c7929fea119eeb
GlobFiles had allowed results to be anywhere in the source tree,
restrict it to results within the current module directory.
Then use it for ExpandSources and other places where we only want files.
This fixes using '*' in cc_test's `data` property, which can only
support files.
The only thing this changes today is that java_resource_dirs and
java_resources no longer pass directories to soong_zip's -f argument.
core-libart previously added some icu directories, now it only passes
files.
Bug: 71906438
Test: only expected changes in out/soong/build.ninja
Test: add data: ["**/*"] to a cc_test, build successfully
Change-Id: Iff1bd8c005a48e431c740706d7e23f4f957d8b1d
We prefer not to use absolute paths in modules, but to reference modules
that have associated paths. This a few benefits:
* it's easier to move a module than to update all the references
* if the module doesn't exist, we treat it as a normal missing
dependency, not having to deal with the missing dependency in path.go
* implementing visibility(etc) in the future would be simpler if there
was a module attached to the reference, so we don't have to do various
path-based lookups to try and match things up.
So define a `droiddoc_template` module, which takes a path, and will run
the glob once in that module. All of the `droiddoc` modules can then
specify it through the `custom_template` property, which will pull the
necessary data.
Also fix that htmldirs should be references from the local path, the
htmldir2 argument never being specified, and complain if more than two
htmldirs are specified, or if the custom template isn't specified.
Test: m core-docs
Test: out/soong/build.ninja is nearly identical
- line numbers in comments
- adds directories to droiddoc template dependency lists, which
is more correct, since we need to rerun on added or removed
files too.
Change-Id: Iff630bddb3818b8eeed439de7e41fc7fbe7cdcb0
Some exe files have a .data symbol at the same address as the
soong_build_number symbol. If the .data symbol is after
soong_build_number in the symbol list, symbol_inject would think
the end address was the same as the start address, and use
uint32(-1) as the size.
Use sort.Search to find the first symbol whose section number is
the same as the target symbol, but whose address is higher than
the target symbol.
Test: manual
Change-Id: I51d6e53c6b906222ba68c5cf93be944843e23550
