Also filter out gcc-only flags -Wno-clobbered and -fno-devirtualize
when compiled with clang/llvm.
BUG: 19872411
Change-Id: I6de57583be04da607f569df65e93531787dbb789
(cherry picked from AOSP commit 90036610dd)
When an app's LOCAL_SDK_VERSION is lower than the support library's
LOCAL_SDK_VERSION, we artifically raises the "SDK version" "linked"
by ProGuard, to
- suppress ProGuard warnings of referencing symbols unknown to the
lower SDK version.
- prevent ProGuard stripping subclass in the support library that
extends class added in the higher SDK version.
This allows us to remove the dangerous ProGuard flag
"-dontwarn android.support.**".
Notes:
- We don't raise the app's LOCAL_SDK_VERSION, so that the app's own
code is still unable to use the higher SDK version.
- For platform build, we can't just raise to the "current" SDK,
because would break apps that use APIs removed from the current SDK.
We raise it to framework.jar instead, which contains the most complete
symbol set.
Bug: 20658265
Change-Id: I90099073457a65cb8031fbaec6b396d68ce614a7
(cherry picked from commit 8e8d1e7aba)
USE_MINGW=1 mm didn't work in directories that contained target modules
because the build system would use the Windows locations and extensions
when trying to find the host GCC prebuilts. Windows is the target OS,
not the OS we're building from.
Change-Id: Ic994fed15388d0c7d393f71ba28fe7afdc659f5c
We need PT_LOAD segments to match for the gdb sake.
If we pack module after stripping symbolic version
PT_LOAD differ from actual ones; this confuses gdb.
Bug: http://b/20687795
Change-Id: If7b1ffcda918d0cc47051a30ca1202007ed62403
(cherry picked from commit 258b29cf76)
We need PT_LOAD segments to match for the gdb sake.
If we pack module after stripping symbolic version
PT_LOAD differ from actual ones; this confuses gdb.
Bug: http://b/20687795
Change-Id: If7b1ffcda918d0cc47051a30ca1202007ed62403
- We don't need LOCAL_PAGE_ALIGN_JNI_SHARED_LIBRARIES now, for we always
page-align jni shared libraries and store them umcompressed.
- For prebuilt apks, we don't extract jni any more; Instead we always run
uncompress-shared-libs on them.
- For apks built from source, we still install the jni separately, because
that way multiple apks can share the same jni and it saves space.
With this change, for most prebuilt apks, we don't need to specify
LOCAL_PREBUILT_JNI_LIBS ("@lib/<abi>/foo.so") any more, for the build
system automatically replaces the embedded jni with uncompressed files;
But if a prebuilt is a fat apk (i.e. containing jni not needed by the
current product architecture), you still need LOCAL_PREBUILT_JNI_LIBS to
specify what jni to keep. Otherwise all embedded jni will be replaced with
uncompressed files, that wastes space.
Bug: 8076853
Change-Id: Icf07e0998ac3602e6e05e80fed836fbafca33e01
The relocation packer is causing the kernel to load the executable
overlapping ASAN's shadow space.
Bug: http://b/20665974
Change-Id: Ifc5914f4fbed5f4f00ed1c795d01cf2fcb849cfe
Always link libm with asan. Hasn't been a problem before because ASAN
was only ever used when libc++ was as well, which already links libm.
Pass --no-as-needed for host modules.
These aren't needed for the target builds because the target uses the
shared RTL.
Change-Id: I5d6a3c2dd557b5231be2b7711df6438693753a88
ASan runtime library (when using dynamic linking) must be the first
dependency of the main executable to achieve correct symbol
interposition. This matches how the clang driver works.
In multilib setup, ASan-RT name depends on the target arch:
/system/lib/libclang_rt.asan-arm-android.so
/system/lib64/libclang_rt.asan-arm64-android.so
We also set RPATH to /system/lib/asan or /system/lib64/asan
to have a place for ASan-only versions of system libraries.
Change-Id: I5c0cdb89e5e08a1950eb276e406da9f31a6e52dd
This is needed for projects that have known ubsan issues that we can't
fix right away (perhaps because it's an upstream project that we're
diligent about keeping in sync with upsteam).
Also make the normal ubsan use -fno-sanitize-recover=all by default.
Change-Id: I1b0f3309792f32dbd08c18816d7306e76c8d7c30
Dan Albert