OSDN Git Service
Elliott Hughes [Fri, 4 Dec 2015 23:57:51 +0000 (15:57 -0800)]
Make all of ndk_cruft.cpp one big extern "C".
Change-Id: Iddd9b5f36e661e443fa13d12326e72df8c0d209c
Elliott Hughes [Mon, 7 Dec 2015 19:10:32 +0000 (19:10 +0000)]
Merge "Fix stdio static initializers to make both clang and GCC happy."
Elliott Hughes [Mon, 7 Dec 2015 19:07:15 +0000 (11:07 -0800)]
Fix stdio static initializers to make both clang and GCC happy.
Previously only clang was happy. GCC said:
error: missing initializer for field 'wcio_mbstate_in' of 'struct wchar_io_data'
Change-Id: I25a11b64f4dfa22a5dd5daded152191fe2cfacaf
Elliott Hughes [Mon, 7 Dec 2015 18:19:07 +0000 (18:19 +0000)]
Merge "Revert "Revert "Remove __sinit and __sdidinit."""
Elliott Hughes [Sat, 5 Dec 2015 02:03:12 +0000 (18:03 -0800)]
Revert "Revert "Remove __sinit and __sdidinit.""
This reverts commit
c8bae05f3ff9f1c736f7be70fa17d02795d748bb.
We were breaking init (ueventd) because we initialize system properties
before we initialize stdio. The new system property implementation uses
stdio to read from /property_contexts, so we end up touching stdio data
structures before they've been initialized.
This second attempt takes things further by removing the stdio initialization
function altogether. The data structures for stdin/stdout/stderr can be
statically initialized as data, and -- since we already had to give the
atexit implementation a backdoor for stdio -- we can just admit that we
need to clean up stdio, and that we always do so last.
This patch also removes the 17 statically pre-allocated file structures,
so the first fopen will now allocate a block of 10 (the usual overflow
behavior). I did this just to make my life simpler, but it's not actually
necessary to remove it if we want it back.
Change-Id: I936b2eb5e88e4ebaf5516121872b71fc88e5609c
Dimitry Ivanov [Sat, 5 Dec 2015 05:25:57 +0000 (05:25 +0000)]
Merge "Allow dlopening public libs using absolute path"
Dimitry Ivanov [Sat, 5 Dec 2015 02:28:49 +0000 (18:28 -0800)]
Allow dlopening public libs using absolute path
dlopen on isolated namespaces should be able to open
public libraries using absolute path not only soname.
Bug: http://b/
25853820
Change-Id: If574a67853dc51226f0f376e9e2d108316002f84
Elliott Hughes [Sat, 5 Dec 2015 01:53:21 +0000 (01:53 +0000)]
Merge "Revert "Remove __sinit and __sdidinit.""
Elliott Hughes [Sat, 5 Dec 2015 01:47:20 +0000 (17:47 -0800)]
Revert "Remove __sinit and __sdidinit."
This reverts commit
4371961e00ad83fca033992c8a19c7d262fe6f84.
This broke booting; ueventd crashes with a null pointer dereference
somewhere in __sfp (but the kernel doesn't unwind, so I don't know
what was calling __sfp).
Change-Id: I65375fdfdf1d339a06558b4057b580cacd6324e2
Tom Cherry [Sat, 5 Dec 2015 00:21:46 +0000 (00:21 +0000)]
Merge "Do not create prop files for ctl.* properties"
Tom Cherry [Fri, 4 Dec 2015 23:53:25 +0000 (15:53 -0800)]
Do not create prop files for ctl.* properties
Change-Id: Ia6660c68c9e0cb89938751dbc0747ee038394778
Elliott Hughes [Fri, 4 Dec 2015 23:54:15 +0000 (23:54 +0000)]
Merge "Fix GCC build."
Elliott Hughes [Fri, 4 Dec 2015 23:53:25 +0000 (15:53 -0800)]
Fix GCC build.
error: '__sdidinit' initialized and declared 'extern' [-Werror]
Change-Id: I97d9c174da6a99ca61b72572dfb4694813372a60
Elliott Hughes [Fri, 4 Dec 2015 23:30:09 +0000 (23:30 +0000)]
Merge "Track rename from base/ to android-base/."
Elliott Hughes [Fri, 4 Dec 2015 23:27:46 +0000 (15:27 -0800)]
Track rename from base/ to android-base/.
Change-Id: I0a4592945400b1fa6892bf7c1fa8659fd711efa3
Elliott Hughes [Fri, 4 Dec 2015 23:27:07 +0000 (23:27 +0000)]
Merge "Remove __sinit and __sdidinit."
Tom Cherry [Fri, 4 Dec 2015 23:06:31 +0000 (23:06 +0000)]
Merge "add checks for initialization for system properties"
Tom Cherry [Fri, 4 Dec 2015 19:34:42 +0000 (11:34 -0800)]
add checks for initialization for system properties
If a __system_property* function is called before
__system_properties_init() then the app will will abort. This commit
returns either an error code or a safe return value instead.
Bug
26027140
Change-Id: I95ffd143e9563658ab67a397991e84fb4c46ab77
Yabin Cui [Fri, 4 Dec 2015 00:58:05 +0000 (00:58 +0000)]
Merge "Clear pthread_internal_t allocated on user provided stack."
Yabin Cui [Thu, 3 Dec 2015 21:01:42 +0000 (13:01 -0800)]
Clear pthread_internal_t allocated on user provided stack.
Several parts in pthread_internal_t should be initialized
to zero, like tls, key_data and thread_local_dtors. So
just clear the whole pthread_internal_t is more convenient.
Bug:
25990348
Change-Id: Ibb6d1200ea5e6e1afbc77971f179197e8239f6ea
Yabin Cui [Fri, 4 Dec 2015 00:29:47 +0000 (00:29 +0000)]
Merge "Don't use StringPrintf() in gtest runner."
Yabin Cui [Fri, 4 Dec 2015 00:28:03 +0000 (16:28 -0800)]
Don't use StringPrintf() in gtest runner.
Change-Id: I0cd0b3cbb952c65b1c449e88ce12964b93029538
Tom Cherry [Thu, 3 Dec 2015 23:48:54 +0000 (23:48 +0000)]
Merge "Remove c++14'isms from system_properties"
Dimitry Ivanov [Thu, 3 Dec 2015 23:44:59 +0000 (23:44 +0000)]
Merge "libm: stop exporting __muldc3"
Tom Cherry [Thu, 3 Dec 2015 23:38:52 +0000 (15:38 -0800)]
Remove c++14'isms from system_properties
Change-Id: If78e7d2770e8f8321f0d1824c3c52f93820dd325
Dimitry Ivanov [Thu, 3 Dec 2015 23:25:37 +0000 (15:25 -0800)]
libm: stop exporting __muldc3
Bug: http://b/
26004493
Change-Id: I991268d5e7b59bca646da25d0ad90a1cb4a9d4b9
Tom Cherry [Thu, 3 Dec 2015 22:55:32 +0000 (22:55 +0000)]
Merge "Separate properties by selabel"
Yabin Cui [Thu, 3 Dec 2015 22:29:29 +0000 (22:29 +0000)]
Merge "Improve unit test runner."
Elliott Hughes [Thu, 3 Dec 2015 21:23:03 +0000 (13:23 -0800)]
Remove __sinit and __sdidinit.
We're eagerly initializing stdio now, so this can all be simplified.
Change-Id: Icb288f8dd0ee08f02bea0d23670f75e78bed6b99
Yabin Cui [Tue, 17 Nov 2015 04:39:58 +0000 (20:39 -0800)]
Improve unit test runner.
1. Read unit test's output while the test is running. Previously
we only read output when the test finishes, which has trouble
when the test outputs too many stuff.
2. Report failed unit test's exit code. It is useful when the
test doesn't fail in ASSERT_xxx, but in somewhere else.
Bug:
25392375
Change-Id: Ie90823337f7c2ee25fa489a5534801d991258f95
Dan Willemsen [Thu, 3 Dec 2015 01:56:00 +0000 (01:56 +0000)]
Merge "Re-add putw for LP32"
Dan Willemsen [Thu, 3 Dec 2015 01:26:15 +0000 (17:26 -0800)]
Re-add putw for LP32
This was missed when switching to LOCAL_SRC_FILES_EXCLUDE
Change-Id: I6ea23c9eb31abe11e0ec4abfc2ee2f2a43c76ce2
Tom Cherry [Wed, 23 Sep 2015 23:09:47 +0000 (16:09 -0700)]
Separate properties by selabel
The purpose of this change is to add read access control to the property
space.
In the current design, a process either has access to the single
/dev/__properties__ file and therefore all properties that it contains
or it has access to no properties. This change separates properties
into multiple property files based on their selabel, which allows
creation of sepolicies that allow read access of only specific sets of
properties to specific domains.
Bug
21852512
Change-Id: Ice265db79201ca811c6b6cf6d851703f53224f03
Dimitry Ivanov [Wed, 2 Dec 2015 18:44:20 +0000 (18:44 +0000)]
Merge "Explicitly disallow default c-tor"
Dimitry Ivanov [Wed, 2 Dec 2015 00:57:19 +0000 (16:57 -0800)]
Explicitly disallow default c-tor
Change-Id: Ia52995a459443159e80383d5b396c3edd90a08ae
Chih-hung Hsieh [Tue, 1 Dec 2015 00:23:32 +0000 (00:23 +0000)]
Merge "Add bionic-unit-tests-gcc{32,64}, compiled with gcc."
Yabin Cui [Mon, 30 Nov 2015 23:57:02 +0000 (23:57 +0000)]
Merge "Init stdio in __libc_init_common."
Yabin Cui [Mon, 30 Nov 2015 23:56:41 +0000 (23:56 +0000)]
Merge "Enable using clang to build __cxa_thread_atexit_impl."
Yabin Cui [Mon, 30 Nov 2015 22:07:58 +0000 (14:07 -0800)]
Enable using clang to build __cxa_thread_atexit_impl.
Remove previous workaround as we no longer use
__thread in __cxa_thread_atexit_impl.cpp.
Change-Id: Ic1062995db488859b341acdda0b5f6635e10d7e8
Yabin Cui [Mon, 30 Nov 2015 21:59:47 +0000 (21:59 +0000)]
Merge "Don't use __thread in __cxa_thread_finalize()."
Yabin Cui [Tue, 24 Nov 2015 23:37:06 +0000 (15:37 -0800)]
Init stdio in __libc_init_common.
Previously we call __sinit() lazily. But it is likely to cause data
races like in https://android-review.googlesource.com/#/c/183237/. So
we prefer to call __sinit() explicitly at libc initialization.
Bug:
25392375
Change-Id: I181ea7a4b2e4c7350b45f2e6c86886ea023e80b8
Yabin Cui [Mon, 30 Nov 2015 21:38:59 +0000 (21:38 +0000)]
Merge "Fix pthread_test according to tsan report."
Chih-hung Hsieh [Mon, 30 Nov 2015 18:58:59 +0000 (18:58 +0000)]
Merge "Consider $(use_clang) when building libc_thread_atexit_impl module"
Chih-Hung Hsieh [Fri, 20 Nov 2015 22:13:24 +0000 (14:13 -0800)]
Add bionic-unit-tests-gcc{32,64}, compiled with gcc.
* bionic-unit-tests{32,64} are compiled with clang/llvm.
* Skip one single test in __cxa_thread_atexit_test.cpp
when compiled with aarch64 clang/llvm.
Aarch64 clang/llvm generates relocation references to
"thread_local" symbols not supported by Android linker.
BUG:
25642296
Change-Id: Ia0497b79c4b335228afeb48a26e0592217909953
Nikola Veljkovic [Wed, 25 Nov 2015 18:03:12 +0000 (19:03 +0100)]
Consider $(use_clang) when building libc_thread_atexit_impl module
Change-Id: Ia369d94979f418198a2ee891385942d9bc5604d7
Yabin Cui [Wed, 25 Nov 2015 01:24:06 +0000 (17:24 -0800)]
Don't use __thread in __cxa_thread_finalize().
Currently we use __thread variable to store thread_local_dtors,
which makes tsan test fork_atexit.cc hang. The problem is as below:
The main thread creates a worker thread, the worker thread calls
pthread_exit() -> __cxa_thread_finalize() -> __emutls_get_address()
-> pthread_once(emutls_init) -> emutls_init().
Then the main thread calls fork(), the child process cals
exit() -> __cxa_thread_finalize() -> __emutls_get_address()
-> pthread_once(emutls_init).
So the child process is waiting for pthread_once(emutls_init)
to finish which will never occur.
It might be the test's fault because POSIX standard says if a
multi-threaded process calls fork(), the new process may only
execute async-signal-safe operations until exec functions are
called. And exit() is not async-signal-safe. But we can make
bionic more reliable by not using __thread in
__cxa_thread_finalize().
Bug:
25392375
Change-Id: Ife403dd7379dad8ddf1859c348c1c0adea07afb3
Colin Cross [Tue, 24 Nov 2015 21:33:54 +0000 (21:33 +0000)]
Merge "Remove cflags and ldflags from crt defaults"
Colin Cross [Tue, 24 Nov 2015 20:21:20 +0000 (20:21 +0000)]
Merge "move arch variant structs down a level"
Colin Cross [Mon, 23 Nov 2015 22:12:19 +0000 (14:12 -0800)]
Remove cflags and ldflags from crt defaults
After ToolchainCflags were added to soong crt no longer needs to try to
recreate the bare minimum cflags for compiling for each architecture.
Also always use GCC to match crt.mk
Change-Id: I01a833ab70d989033c84f072e3660d060189688c
Dimitry Ivanov [Tue, 24 Nov 2015 02:23:34 +0000 (02:23 +0000)]
Merge "Introduce anonymous namespace"
Dmitriy Ivanov [Mon, 23 Nov 2015 19:26:35 +0000 (11:26 -0800)]
Introduce anonymous namespace
The anonymous namespace is introduced to
handle cases when linker can not find the
caller. This usually happens when caller
code was not loaded by dynamic linker;
for example mono-generated code.
Bug: http://b/
25844435
Bug: http://b/
22548808
Change-Id: I9e5b1d23c1c75bc78548d68e79216a6a943a33cf
Colin Cross [Mon, 23 Nov 2015 22:12:15 +0000 (14:12 -0800)]
move arch variant structs down a level
Use blueprint's new anonymous embedded struct feature to move the arch
variant properties down a level, replacing arch.cortex_a9.srcs with
arch.arm.cortex_a9.srcs, while still supporting top-level properties
like arch.arm.srcs.
Change-Id: Ib41c80e3549440d5efdfd293a15cffa3f51a0fe4
Yabin Cui [Mon, 23 Nov 2015 18:57:26 +0000 (18:57 +0000)]
Merge "Change _stdio_handles_locking into _caller_handles_locking."
Dimitry Ivanov [Mon, 23 Nov 2015 07:28:32 +0000 (07:28 +0000)]
Merge "Handling invalid section headers"
Dmitriy Ivanov [Sat, 21 Nov 2015 01:28:12 +0000 (17:28 -0800)]
Handling invalid section headers
The linker crashes if native library has invalid section
headers. This change adds boundary checks on target offsets
and generates dlerror instead of crash.
Bug: http://b/
25800330
Change-Id: Ibe282029997302b9b557637c3aad064d7d0febc5
Yabin Cui [Sat, 21 Nov 2015 01:50:29 +0000 (01:50 +0000)]
Merge "Use FUTEX_WAIT_BITSET to avoid converting timeouts."
Yabin Cui [Thu, 19 Nov 2015 21:52:16 +0000 (13:52 -0800)]
Change _stdio_handles_locking into _caller_handles_locking.
It is reported by tsan that funlockfile() can unlock an unlocked mutex.
It happens when printf() is called before fopen() or other stdio stuff.
As FLOCKFILE(fp) is called before __sinit(), _stdio_handles_locking is false,
and _FLOCK(fp) will not be locked. But then cantwrite(fp) in __vfprintf()
calls__sinit(), which makes _stdio_handles_locking become true, and
FUNLOCKFILE(fp) unlocks _FLOCK(fp).
Change _stdio_handles_locking into _caller_handles_locking,
so __sinit() won't change its value. Add test due to my previous fault.
Bug:
25392375
Change-Id: I483e3c3cdb28da65e62f1fd9615bf58c5403b4dd
Dimitry Ivanov [Fri, 20 Nov 2015 21:42:18 +0000 (21:42 +0000)]
Merge "Move some utility functions to linker_utils"
Dmitriy Ivanov [Fri, 20 Nov 2015 21:34:11 +0000 (13:34 -0800)]
Move some utility functions to linker_utils
Also adds unit-tests for page_start, page_offset, and safe_add
Change-Id: Ia1325b4682d367328a01599a19848e4ffcd2c0ea
Dimitry Ivanov [Fri, 20 Nov 2015 19:16:14 +0000 (19:16 +0000)]
Merge "Improve error message for files with no sections"
Dmitriy Ivanov [Fri, 20 Nov 2015 18:42:02 +0000 (10:42 -0800)]
Improve error message for files with no sections
Bug: http://b/
25801618
Change-Id: I4f4f368e727ff48c84781279e3d17d4ac2d1b6b0
Chih-hung Hsieh [Fri, 20 Nov 2015 18:27:14 +0000 (18:27 +0000)]
Merge "Disable clang for mips/mips64 libc."
Chih-Hung Hsieh [Fri, 20 Nov 2015 17:55:35 +0000 (09:55 -0800)]
Disable clang for mips/mips64 libc.
* Many processes, including adbd, failed to start in mips/mips64
emulator when libc.so was compiled by clang.
BUG:
25291096
Change-Id: If3434ebdca4a3a6bf6102b120ee838a7ab66cd74
Elliott Hughes [Fri, 20 Nov 2015 17:19:19 +0000 (17:19 +0000)]
Merge "mmap: fix calculation of is_private_anonymous variable"
Vitaly Vul [Thu, 24 Sep 2015 12:28:59 +0000 (14:28 +0200)]
mmap: fix calculation of is_private_anonymous variable
Currently is_private_anonymous is calculated as true if _either_
MAP_PRIVATE or MAP_ANONYMOUS is set, which is a mistake.
According to Documentation/vm/ksm.txt, "KSM only merges anonymous
(private) pages, never pagecache (file) pages". MAP_PRIVATE can
still be set on file cache pages so in order to not redundantly
set MADV_MERGEABLE on pages that are not fitted for it, both
MAP_PRIVATE and MAP_ANONYMOUS should be set.
Along with this fix, add an extra check that the mapped page is
not a stack page before setting MADV_MERGEABLE for it. Stack pages
change too quickly and always end up in KSM 'page_volatile' list.
Change-Id: If4954142852f17cc61f02985ea1cb625a7f3dec6
Josh Gao [Fri, 20 Nov 2015 01:55:56 +0000 (01:55 +0000)]
Merge changes I73a39c6d,Iab2af242
* changes:
Fix test build break.
extend the PTRDIFF_MAX size check to mremap
Yabin Cui [Fri, 20 Nov 2015 00:40:00 +0000 (00:40 +0000)]
Merge "Fix tsan abort for lake of __libc_auxv initialization."
Colin Cross [Thu, 19 Nov 2015 22:48:33 +0000 (22:48 +0000)]
Merge "Update libc/Android.bp to match libc/Android.mk"
Yabin Cui [Thu, 19 Nov 2015 22:24:06 +0000 (14:24 -0800)]
Fix tsan abort for lake of __libc_auxv initialization.
If tsan is used, the following callchain can happen:
__libc_preinit() -> __libc_init_globals() ->
__libc_init_vdso() -> strcmp() -> __tsan_init()
-> sysconf(_SC_PAGE_SIZE) -> getauxval().
But __libc_auxv is initialized in __libc_init_common(),
after __libc_init_globals(). One simple way to fix
this is to initialize __libc_auxv at __libc_init_globals().
Bug:
25392375
Change-Id: I3893b1f567d5f3b7a8c881c0c1b8234b06b7751b
Colin Cross [Thu, 19 Nov 2015 21:32:49 +0000 (13:32 -0800)]
Update libc/Android.bp to match libc/Android.mk
Change-Id: I4dfadc55688213f095949c56306e5071b2ab6135
Josh Gao [Thu, 19 Nov 2015 21:44:20 +0000 (13:44 -0800)]
Fix test build break.
Change-Id: I73a39c6d4bb9baebbfb5572e9259bcd405116039
Yabin Cui [Thu, 5 Nov 2015 23:36:08 +0000 (15:36 -0800)]
Use FUTEX_WAIT_BITSET to avoid converting timeouts.
Add unittests for pthread APIs with timeout parameter.
Bug:
17569991
Change-Id: I6b3b9b2feae03680654cd64c3112ce7644632c87
Hung-ying Tyan [Thu, 19 Nov 2015 16:38:51 +0000 (16:38 +0000)]
Merge "Add /odm/lib to shared lib search path"
Daniel Micay [Sat, 7 Nov 2015 15:40:26 +0000 (10:40 -0500)]
extend the PTRDIFF_MAX size check to mremap
This removes another way to obtain objects larger than PTRDIFF_MAX. The
only known remaining hole is now jemalloc's merging of virtual memory
spans.
Technically this could be wrapped in an __LP64__ ifndef since it can't
occur on 64-bit due to the 1:1 split. It doesn't really matter either
way.
Change-Id: Iab2af242b775bc98a59421994d87aca0433215bd
Junichi Uekawa [Thu, 19 Nov 2015 12:44:19 +0000 (12:44 +0000)]
Merge "Do not depend on host bits to get the right size to write."
Junichi Uekawa [Wed, 18 Nov 2015 01:18:59 +0000 (10:18 +0900)]
Do not depend on host bits to get the right size to write.
x86_64 32-bit or 64-bit relocations do not depend on ELF bit size, they
are 32-bit or 64-bit respectively.
Known compiler that emits such code is nacl-clang which emits
R_X86_64_PC32 which should write 32 bits but ended up writing 64 bits.
Change-Id: Ibb6b484c0fea6a7e291362148e8ac749d6674529
Yabin Cui [Thu, 19 Nov 2015 01:55:37 +0000 (01:55 +0000)]
Merge "Implement pthread spin."
Yabin Cui [Thu, 19 Nov 2015 01:55:25 +0000 (01:55 +0000)]
Merge "Avoid tsan warning about pthread_mutex_destroy."
Yabin Cui [Wed, 18 Nov 2015 00:03:18 +0000 (16:03 -0800)]
Implement pthread spin.
In order to run tsan unit tests, we need to support pthread spin APIs.
Bug:
18623621
Bug:
25392375
Change-Id: Icbb4a74e72e467824b3715982a01600031868e29
Hung-ying Tyan [Thu, 12 Nov 2015 03:48:19 +0000 (11:48 +0800)]
Add /odm/lib to shared lib search path
ODM's shared libs should live in /odm/lib on the ODM partition.
BUG:
25654402
Change-Id: I3f5c26f208af87ff2817e490e51d0b4f5b8ec98e
Dimitry Ivanov [Wed, 18 Nov 2015 03:43:59 +0000 (03:43 +0000)]
Merge "Ignore target sdk version for the public namespace"
Dmitriy Ivanov [Wed, 18 Nov 2015 02:36:50 +0000 (18:36 -0800)]
Ignore target sdk version for the public namespace
This fixes the bug with using the libraries loaded
prior to android_set_target_sdk_version call.
Bug: http://b/
22548808
Change-Id: I3ca2d367b0fa930a437bbb65f780834803d2ef0a
Yabin Cui [Tue, 17 Nov 2015 04:19:31 +0000 (20:19 -0800)]
Avoid tsan warning about pthread_mutex_destroy.
If calling pthread_mutex_trylock from pthread_mutex_destroy, tsan
warns about an attempt to destroy a locked mutex.
Bug:
25392375
Change-Id: I5feee20e7a0d0915adad24da874ec1ccce241381
Yabin Cui [Tue, 17 Nov 2015 05:06:16 +0000 (21:06 -0800)]
Fix pthread_test according to tsan report.
1. Fix leak threads and data races related to spin_flag.
2. Increase stack size to run under tsan.
This doesn't pass all pthread tests, as some tests are used
to run intentionally in race situations.
Bug:
25392375
Change-Id: Icfba3e141e7170abd890809586e89b99adc8bd02
Yabin Cui [Tue, 17 Nov 2015 00:22:54 +0000 (00:22 +0000)]
Merge "Implement pthread barrier."
Dimitry Ivanov [Mon, 16 Nov 2015 22:25:50 +0000 (22:25 +0000)]
Merge "Remove the warning about empty LD_LIBRARY_PATH"
Dmitriy Ivanov [Mon, 16 Nov 2015 22:23:37 +0000 (14:23 -0800)]
Remove the warning about empty LD_LIBRARY_PATH
Change-Id: I3eca11512d8055ab94dd0a6badcd83ce6440d675
Yabin Cui [Fri, 6 Nov 2015 06:06:09 +0000 (22:06 -0800)]
Implement pthread barrier.
Bug:
24341262
Change-Id: I5472549e5d7545c1c3f0bef78235f545557b9630
Chih-hung Hsieh [Mon, 16 Nov 2015 21:31:36 +0000 (21:31 +0000)]
Merge "Enable clang to compile more tests."
Dimitry Ivanov [Mon, 16 Nov 2015 21:23:18 +0000 (21:23 +0000)]
Merge "Fix arm64 and x86 builds"
Dmitriy Ivanov [Mon, 16 Nov 2015 21:17:27 +0000 (13:17 -0800)]
Fix arm64 and x86 builds
Change-Id: Ic3746b2f200f42218b6ac857c1631e873c767c17
Dimitry Ivanov [Mon, 16 Nov 2015 20:14:37 +0000 (20:14 +0000)]
Merge "Introducing linker namespaces"
Chih-Hung Hsieh [Mon, 16 Nov 2015 19:13:04 +0000 (11:13 -0800)]
Enable clang to compile more tests.
* Disable optimization only in gtest.h of atexit_test.c for arm/aarch64
to keep VTT for std::__1::basic_stringstream<char, std::__1::char_traits<char>,
std::__1::allocator<char> > to link with g++ compiled modules.
* bionic-unit-tests source files are not affected by clang x86_64 fp128 bug
so they can be compiled with clang.
BUG:
25643775
Change-Id: I3da2a0de61edcdca07b7fcd73a16de9da4a1f7d6
Dmitriy Ivanov [Fri, 30 Oct 2015 00:01:24 +0000 (17:01 -0700)]
Introducing linker namespaces
Bug: http://b/
22548808
Change-Id: Ia3af3c0a167f1d16447a3d83bb045d143319b1e1
Elliott Hughes [Fri, 13 Nov 2015 18:24:41 +0000 (18:24 +0000)]
Merge "Improve libc time zone fallback behavior."
Elliott Hughes [Fri, 13 Nov 2015 16:38:48 +0000 (08:38 -0800)]
Improve libc time zone fallback behavior.
We should fall back to GMT if neither the environment variable nor the
system property is set. This is the case if you wipe a WiFi-only device,
because we currently only take the time zone from cell networks.
Bug: http://b/
24773112
Change-Id: I90d236d4d492b6562d75021bd312030b91c1e298
Elliott Hughes [Fri, 13 Nov 2015 02:58:30 +0000 (02:58 +0000)]
Merge "Fix strftime if tm_zone is null."
Nick Kralevich [Fri, 13 Nov 2015 01:09:44 +0000 (01:09 +0000)]
Merge "FORTIFY_SOURCE: make sure gcc unittests are compiled with gcc"
Elliott Hughes [Fri, 13 Nov 2015 00:51:31 +0000 (16:51 -0800)]
Fix strftime if tm_zone is null.
Upstream tzcode said "On platforms with tm_zone, strftime.c now assumes it
is not NULL". Which is fine for any struct tm generated by tzcode, but not
necessarily true of a struct tm constructed by arbitrary code. In particular,
Netflix on Nexus Player was failing to start because they format "%Z" with
a struct tm whose tm_zone is null (the other fields are valid, but, yeah,
that's probably not intentional).
glibc takes a null tm_zone to mean "the current time zone", so let's do that
too. (Historically Android would use the empty string, and POSIX doesn't
clarify which of this is the appropriate behavior when tm_zone is null.)
Bug: http://b/
25170306
Change-Id: Idbf68bfe90d143aca7dada8607742905188b1d33
Nick Kralevich [Thu, 12 Nov 2015 23:55:13 +0000 (15:55 -0800)]
FORTIFY_SOURCE: make sure gcc unittests are compiled with gcc
The default compiler is clang. We need to explicitly indicate
that we should use gcc.
Change-Id: I37859c2e303f2a86a2565fe72eda1fda7c557a59