...by inlining them.
Also fix a couple of harmless bugs in passing. I've added tests, but in
both cases I don't think it was actually possible to hit the bad behavior:
we'd hit another test and fail immediately after in an externally
indistinguishable way.
Bug: N/A
Test: readelf
Change-Id: I8466050b0bfe2b7b94c76b383cf10c1d9d28debd
Upstream's killpg is diverging further from glibc behavior, so let's just fork.
Bug: N/A
Test: ran tests
Change-Id: I70a3543018bc0a5c0bbf019ac527043b90568fda
If the hardware debug support is missing from the CPU, the kernel can
end up in the state where it reports that it supports N(>0) watchpoints,
but the "maximum supported watchpoint size" is zero.
This is not a really sensible API, but it is what we have. I'm going to
speak to folks upstream about whether we can have it report something
more sensible, but that will take a while (if ever).
So, for the time being, detect this situation and treat it as if no
hardware watchtpoints/breakpoints are supported.
Test: bionic-unit-tests --gtest_filter="sys_ptrace.*"
Change-Id: Id6a013c9bec8d8b450b4a36c28431f6b7fd15879
Bug: 63116544
* TTYDEF_OFLAG referred to the BSD OXTABS constant instead of the Linux XTABS,
so references to it wouldn't compile.
* CEOL was incorrect (compared to `stty -a`).
* ttydefchars wouldn't compile (removed rather than attempt to fix something
of unclear utility).
Bug: http://b/63145226
Test: new test
Change-Id: I823e5eaab789cafd2eeea03f77ea553346822f74
With this, stack frame 0 is the abort, not tgkill.
arm:
#00 pc 0001a41c /system/lib/libc.so (abort+63)
arm64:
#00 pc 000000000001d75c /system/lib64/libc.so (abort+120)
Also "include what you use" for <sys/syscall.h>.
Bug: N/A
Test: ran `crasher abort` and `crasher64 abort`
Change-Id: I6517ac67b39b4133e890d52efc115071c812958b
Building ruby actually trips over both of these:
* if the RTLD_ constants aren't #defined, it uses its own incorrect values.
* if the REG_ constants aren't #defined, it confuses x86 with x86-64.
In all other places where we have enums in our headers, we already match
existing glibc practice.
Bug: http://b/62531921
Test: builds
Change-Id: I5b3aab25a1a24611bdc58f2eda4104a78e9f841c
Move all tests into stdlib_test.cpp since that's where the definition lives
in bionic.
Add a sweep test and a various size test.
Test: Run new unit tests on glibc and angler.
Change-Id: Ief1301f402bea82ce90240500dd6a01636dbdbae
This breaks code that declares `errno` itself for whatever crazy reason:
b.c:22:12: error: conflicting types for '__errno'
extern int errno;
^
/usr/local/google/ndkports/toolchain/bin/../sysroot/usr/include/errno.h:47:20: note: expanded from macro 'errno'
#define errno (*__errno())
^
/usr/local/google/ndkports/toolchain/bin/../sysroot/usr/include/errno.h:44:15: note: previous declaration is here
volatile int* __errno(void) __attribute_const__;
^
Bug: N/A
Test: built various bits of GNU source
Change-Id: I27c03bf3bde419a001f98f1ea6c267c847f31271
r279663: enhance to handle 32-bit aligned data on 64-bit systems
r279666: qsort(3): small style(9) cleanups.
r288030: Use ANSI C prototypes. Eliminates -Wold-style-definition
warnings.
r318514: Use size_t.
r318515: Reduce the possible recursion level to log2(N) by always
recursing on the smaller partition instead.
Bug: 38450217
Test: builds, ./tests/run-on-host.sh 64
Change-Id: Ieb622e6f54f8f6d6f62e1bdb491d3b5c8e2f9965
Test fcntl#falloc_punch is wrong. It checks that fallocate() with mode
FALLOC_FL_PUNCH_HOLE fails on ext4 file system on older kernels. The
test fails to ensure that the file it creates is indeed on an ext4
partition. On an Angelfish device for example, the file is created on an
f2fs partition, which supports FALLOC_FL_PUNCH_HOLE, and thus the test
fails (wrongly).
Change-Id: I23c1ba4d0fcee81551531779e93ac3d5e19ba1d7
Fixes: 62220977
Test: run bionic-unit-tests as per bionic/README.md###Device tests
Signed-off-by: Nick Desaulniers <ndesaulniers@google.com>
Callers are supposed to #include <scsi/sg.h> but if we tell soong to add
bionic/libc/kernel/android/ to the include path, the uapi headers in there
would be (unintentionally) accessible as either <linux/name.h> or
<uapi/linux/name.h>.
Bug: N/A (hit while upgrading strace to 4.17)
Test: builds
Change-Id: I8d47dd51da688c38f747a255d401dfb2c209c805
The number of available physical pages is not constant and can
potentially change between calls to get_avphys_pages and
sysconf(_SC_AVPHYS_PAGES) calls.
Relax the assert to check that sconf(_SC_AVPHYS_PAGES) <= phys_pages.
Bug: http://b/31502852
Test: build and run bionic-unit-tests
Change-Id: Ibf8873d2151d93239391d2638dfbf055b70cde3a
Add the mallopt function, and only a single option so far.
Bug: 36401135
Test: Built and booted bullhead.
Test: Ran jemalloc unit tests.
Test: Ran bionic unit tests.
Test: Ran a test that allocated and free'd a large piece of memory,
Test: and verified that after changing the parameter, the PSS
Test: sticks around (decay timer set to 1), the PSS is purged (decay
Test: timer set to 0).
(cherry picked from commit a1c0d2fd4c)
Change-Id: Ie8125dc3a5615a859efb66dc91a6feeb0d57b478
ETC did end up causing problems, since it's not per-arch. But
SHARED_LIBRARIES isn't correct either, since the build system tries to
read shared libraries as ELF files (to create the toc files).
NATIVE_TESTS is slightly better -- they are not libraries, so we don't
attempt to generate toc files, but they are considered per-arch.
Bug: 38463793
Test: Run CtsBionicTestCases
Change-Id: I01b4b093359e0c39eb2f8639d5c5dde9e304ed20
We've had pthread_*_setpshared for long enough that there are no
__INTRODUCED_IN guards.
Found because fio's configure script was confused by this.
Bug: N/A
Test: ran tests
Change-Id: I07b4d937741e4dcd7e615f2371b17c827341917a
This is a bit bogus because it's been removed from glibc (though not
thoroughly) and is never useful on Android (because the system calls
in question are compiled out of Android kernels, and SELinux would
disallow them even if you weren't running an Android kernel). This
also means that on glibc you need to include <linux/sem.h> for this
and on bionic you need <sys/sem.h> (and for either if you #include
the other file, you won't get this union).
Bug: https://github.com/android-ndk/ndk/issues/400
Test: added new test
Change-Id: I47f721da77515531f616d6ad8479bfbc9b60ee47
It's going away, since we don't have a single library on device anymore,
and it's simpler to keep the host looking like the device. This doesn't
seem to need it anymore either -- the libraries are found relative to
the test binary.
Test: bionic/tests/run-on-host.sh glibc
Change-Id: Ie9bf6388ecd773623181ec9eadb35d6e0ba0a441
DlExtRelroSharingTest.VerifyMemorySaving verifies whether
android_dlopen_ext(ANDROID_DLEXT_USE_RELRO) really saves memory.
This patch compares Pss of the tested library rather than total
Pss of the process, in case different behavior of dlopen() and
android_dlopen_ext() introduces random Pss usage.
Test: DlExtRelroSharingTest.VerifyMemorySaving
(cherry picked from commit 81aad0058c)
Bug: https://issuetracker.google.com/38226388
Change-Id: I4248628fa951b38818fe2af363fdf78abf09a112
It's going away, so use the built versions in the per-library
intermediate directories instead.
Test: `m -j bionic-unit-tests` produces the same output before/after
Change-Id: Ia5cd874d13129b2be89cdc26e4ec598be4003c87
Add the mallopt function, and only a single option so far.
Bug: 36401135
Test: Built and booted bullhead.
Test: Ran jemalloc unit tests.
Test: Ran bionic unit tests.
Test: Ran a test that allocated and free'd a large piece of memory,
Test: and verified that after changing the parameter, the PSS
Test: sticks around (decay timer set to 1), the PSS is purged (decay
Test: timer set to 0).
Change-Id: I6927929b0c539c1023d34772d9e26bb6a8a45877
In pthread_leak.detach test, compare the memory usage after two same
scenarios - all children threads have exited. Thus, test emulated by
NativeBridge, which may reserve a memory pool, will pass if it's not
unbounded leak.
Bug: https://issuetracker.google.com/37920774
Test: CtsBionicTestCases
(cherry picked from commit cf17b487ef)
Change-Id: Ib47b26975489b397783e29d65f67fe775b060589
In pthread_leak.detach test, compare the memory usage after two same
scenarios - all children threads have exited. Thus, test emulated by
NativeBridge, which may reserve a memory pool, will pass if it's not
unbounded leak.
Bug: https://issuetracker.google.com/37920774
Test: CtsBionicTestCases
Change-Id: Ia51aa67bacb588284747652b36ea5d8e85bea832
greylist is needed only for application namepaces. Since we started using
linker-namesapces for vendors and other platform apps linker should not
enable greylist workaround by default.
Bug: http://b/37731053
Bug: https://issuetracker.google.com/38146125
Test: sailfish builds and boots
Test: bionic-unit-tests and linker-unit-tests pass
Change-Id: Iee83db6fb1ae754f5ade18491321d9bca3b5ead4
(cherry picked from commit e8ffe56a9c)
greylist is needed only for application namepaces. Since we started using
linker-namesapces for vendors and other platform apps linker should not
enable greylist workaround by default.
Bug: http://b/37731053
Bug: https://issuetracker.google.com/38146125
Test: sailfish builds and boots
Test: bionic-unit-tests and linker-unit-tests pass
Change-Id: Iee83db6fb1ae754f5ade18491321d9bca3b5ead4
There are new optimizations for these functions, so adding some extra
testing for these routines.
Also, clean up the strchr test slightly with some extra comments.
Test: Ran new tests on glibc version, and on angler.
Change-Id: I41bf4e5e2c84295cc1ce9d2226ed57c2d228d7b8
Since clang does not produce IRELATIVE relocations
this test started passing when we switched away from gcc
This is still a problem because it now leads to
situation where static ifuncs become globally visible
but this is a different problem.
Test: bionic-unit-tests --gtest_filer=dl*
Change-Id: Id27ba5093b88519c00acbc43ab8a991a671a9f51
This library is used by a number of different libraries in the system.
Make it easy for platform libraries to use this library and create
an actual exported include file.
Change the names of the functions to reflect the new name of the library.
Run clang_format on the async_safe_log.cpp file since the formatting is
all over the place.
Bug: 31919199
Test: Compiled for angler/bullhead, and booted.
Test: Ran bionic unit tests.
Test: Ran the malloc debug tests.
Change-Id: I8071bf690c17b0ea3bc8dc5749cdd5b6ad58478a
Make shared namespace inherit namespace links and
search/permitted paths from the parent namespace.
Bug: http://b/37854032
Test: bionic_unit_tests --gtest_filter=dl*:Dl*
Change-Id: I174661d4a1dd0cbe4a378179073719aa955f3592
(cherry picked from commit ec43dd6c36)
After the new rebase of Clang, we now support the alloc_size attribute,
so we can remove the warning about switching to alloc_size once Clang
has support.
Also fix an issue where 'false' is returned in a main method.
Test: make cts
Bug: 37752547
Change-Id: I32ee814aaa0338567cb8174a68d7ee1bfece2f75
Make shared namespace inherit namespace links and
search/permitted paths from the parent namespace.
Bug: http://b/37854032
Test: bionic_unit_tests --gtest_filter=dl*:Dl*
Change-Id: I174661d4a1dd0cbe4a378179073719aa955f3592
Mistakenly removed years ago, along with its genuinely duplicated companions.
This patch also redefines one of those rather than use the Linux uapi header
definition, which leads to signed/unsigned warnings in calling code.
Bug: https://github.com/android-ndk/ndk/issues/377
Test: ran tests
Change-Id: Ib25b17949f04969f12b945ca88ab87d080677cf9
8-byte watchpoints are not supported on 32-bit kernels (at least on x86,
probably arm as well). 8-byte watchpoints are not required for this
test, so just decrease the size of watchpoints on these architectures.
Test: bionic-unit-tests-static --gtest_filter=sys_ptrace.*
Bug: 37662849
Change-Id: I450a78fc1d6b8807fe1c776f5fa62d329788969e
(cherry picked from commit 778c048a8f)
DlExtRelroSharingTest.VerifyMemorySaving verifies whether
android_dlopen_ext(ANDROID_DLEXT_USE_RELRO) really saves memory.
This patch compares Pss of the tested library rather than total
Pss of the process, in case different behavior of dlopen() and
android_dlopen_ext() introduces random Pss usage.
Test: DlExtRelroSharingTest.VerifyMemorySaving
Change-Id: I696eca736018fca5d9ec9385ac1c79ced280281d
8-byte watchpoints are not supported on 32-bit kernels (at least on x86,
probably arm as well). 8-byte watchpoints are not required for this
test, so just decrease the size of watchpoints on these architectures.
Change-Id: Ie96b4bc8ce0ad8c924ac737083cf9fe0182aee4d
Test: bionic-unit-tests-static --gtest_filter=sys_ptrace.*
Bug: 37662849
This reverts commit 6d59318d78.
Bug: http://b/37245203
Bug: http://b/37590327
Test: bionic-unit-tests on angler and marlin
Test: boot an angler and make sure mediacodec does not crash
Fix the list of static functions coming from swab.h to match the
ones exported by glibc.
Force the definition of __HAVE_BUILTIN_BSWAP{16,32,64}__ so the headers
use the builtin.
Add a unit test to guarantee that kernel header updates do not break this.
Test: Built and booted angler.
Test: Built the bionic unit tests for arm, arm64, mips, x86, x86_64.
Test: Ran the new test on angler and glibc.
Change-Id: I4ce229e3f198c204186d72bf22dd97b5cdf239e4
The dl_iterate_phdr test is new, but the exidx test is largely copy & paste
from ndk_translation.
Bug: http://b/31556066
Test: ran tests
Change-Id: If754f1459ef1d789ee98a89cd2215188edca51da
This change adds two tests for dlopen from temporary files.
1. One Uses memfd_create() can be used to load libraries directly
from memory. This requires relaxing namespace accessibility check
in order to make this work in isolated namespaces.
2. Another checks that open with O_TMPFILE works.
Bug: http://b/37245203
Test: bionic-unit-tests --gtest_filter=dl*:Dl*
Change-Id: I3be1d7198ca17e7f1ba022a0d86c64d59a493506
(cherry picked from commit bb8b22a087)
This change adds two tests for dlopen from temporary files.
1. One Uses memfd_create() can be used to load libraries directly
from memory. This requires relaxing namespace accessibility check
in order to make this work in isolated namespaces.
2. Another checks that open with O_TMPFILE works.
Bug: http://b/37245203
Test: bionic-unit-tests --gtest_filter=dl*:Dl*
Change-Id: I3be1d7198ca17e7f1ba022a0d86c64d59a493506
Netflix was using this, and looking the header file, although
__system_property_find_nth has been available since the beginning of time,
__system_property_foreach only appeared in 16. So anyone who wants to run
on pre-JellyBean devices would want to use __system_property_find_nth.
It's pretty much a one-liner in terms of __system_property_foreach anyway,
so it doesn't cost us anything to keep it.
Also restore slightly better tests than we originally removed.
Bug: http://b/36566667
Test: ran tests
(cherry picked from commit 438e01940b)
Change-Id: I639f2142ad4ba049b990b13ccccd255be4b4f479
Netflix was using this, and looking the header file, although
__system_property_find_nth has been available since the beginning of time,
__system_property_foreach only appeared in 16. So anyone who wants to run
on pre-JellyBean devices would want to use __system_property_find_nth.
It's pretty much a one-liner in terms of __system_property_foreach anyway,
so it doesn't cost us anything to keep it.
Also restore slightly better tests than we originally removed.
Bug: http://b/36566667
Test: ran tests
Change-Id: Id268c2c2e848da17bb0a5a5420af234d9dcb829a
And add a thing to remind us to enable the malloc FORTIFY test when we
pull a new clang in.
Bug: 34747525
Test: vts -m BionicUnitTests on bullhead
Change-Id: If341a27756d608a0fa77ba6518e9bcc725f7632c
libcxx provides const-correct overloads for a few string.h functions.
These overloads use clang's enable_if attribute, so they're preferred
over our FORTIFY'ed equivalents.
This weakens _FORTIFY_SOURCE=2 when used with some of these functions,
since clang needs to see __pass_object_size in order to pass an accurate
result for __builtin_object_size(s, 1) at a callsite. Since those
functions don't have __pass_object_size on their params, clang can't do
that. This makes LLVM lower the __builtin_object_size calls, which means
we get the same result as __builtin_object_size(s, 0).
We have to provide all of the overloads in Bionic, since enable_if is
only used to disambiguate overloads with (otherwise) the same type. In
other words:
// overload 1
char *strchr(const char *, int s) __attribute__((enable_if(1, "")));
// overload 2
char *strchr(char *, int s);
void foo() {
char cs[1] = {};
strchr(static_cast<const char *>(cs), '\0'); // calls overload #1.
strchr(cs, '\0'); // calls overload #2.
}
Bug: 34747525
Test: m checkbuild on bullhead internal master + AOSP. vts -m
BionicUnitTests passes on both. Surprisingly, the only code that this
seems to break is contained in Bionic.
Change-Id: Ie406f42fb3d1c5bf940dc857889876fc39b57c90
Used by CFI, so broke cfi_test#early_init@x86, but I've added a specific
test for this (and a similar test for getauxval from preinit, which this
patch does not fix).
Bug: http://b/35885875
Test: ran tests
(cherry picked from commit eb04ed506f)
Change-Id: Ie9da1bf4c2dcc23be7d75c739ae8549309813f88
Used by CFI, so broke cfi_test#early_init@x86, but I've added a specific
test for this (and a similar test for getauxval from preinit, which this
patch does not fix).
Bug: http://b/35885875
Test: ran tests
Change-Id: I43885bedfb88c0a26b4474bd3c27a87dec7bbc97
Modify the test slightly to make sure that the unwinder properly unwinds
through at least 2 frames outside of the signal handler. This guarantees
that if the first frame outside of the handler happens to be garbage,
this test will fail.
Bug: 34468756
Test: Ran the unit tests on fugu, angler (both 32 bit and 64 bit).
Change-Id: I6b76ac9fc1df9ed6fd5bbcc6f5fa4bf458354dff
Prevent the compiler from being too smart and allocating a stack buffer
at the beginning of a function.
Bug: http://b/36206043
Test: 32/64-bit dynamic tests pass, static ones still don't
Change-Id: I90c575be43a9dd6c4fefc0d8b514f1ae0405b994
snprintf to a buffer of length PATH_MAX consumes about 7kB of stack.
Bug: http://b/35858739
Test: bionic-unit-tests --gtest_filter="*big_enough*"
Change-Id: I34a7f42c1fd2582ca0d0a9b7e7a5290bc1cc19b1
Apparently that "backdoor" is no longer needed - the proper way is
to reinitialize properties:
https://android-review.googlesource.com/#/c/181794/24/tests/system_properties_test.cpp
Also removes mentions of libnativehelper test (it no longer uses
__system_property_area__) and removes useless "extern" declaration
(actual use was removed long ago).
Test: refactoring CL, existsing tests still pass
BUG=21852512
BUG=34114501
Change-Id: I2223cab2fcb671ea180ad4470a7aba5c9cd20bd8
Do not load second copy of libraries that are supposed to
be provided by linked namespaces. Also do not print
error in the log if caller tries to open shared library
using absolute path for apps targeting N+.
Bug: http://b/35454141
Bug: http://b/26833548
Bug: http://b/35338922
Test: run bionic-unit-tests --gtest_filter=dl*
Change-Id: Icf3aeedff18d287d2ba0b3df3808b100f3ef5f7a
Some versions of kernel set AT_BASE to 0
if dynamic loader does not have PT_INTERP
set.
Bug: http://b/30739481
Test: run /system/bin/linker64 and /system/bin/linker
Change-Id: I1b67777166fe917d3ee1a97277045ca6f5db0084
Also, for the stuff that's also in <netinet/in.h> as real functions,
check that they're there too (and as functions rather than macros,
since that was historically not true).
Bug: http://b/28432448
Test: ran tests
Change-Id: I7e4ae926f7e02de3b6dd38d1953e5b3b43d44f74
This tests for the presence of a kernel bug that meant that the kernel
would sometimes fail to report the watchpoint hit if the hardware
reported a address which did not exactly match the address range being
watched (which it is allowed to do per ARM spec if the instruction
accesses a larger block of memory than the region being watched). This
bug was fixed in linux kernel 4.9, and has been backported to older
android kernels.
Bug: 30802222
Bug: 30919905
Test: bionic-unit-tests --gtest_filter="sys_ptrace.*"
Change-Id: I80c35b29eaf28e2dbacb9e8ee5317fdea653fc87
Somewhat unsurprisingly, very few commands are happy to be run like this,
in particular multiplexed commands like toybox. But that's no reason for
the linker to get in the way too.
Bug: http://b/33276926
Test: new test
Change-Id: I6dd71ea0183f4da83571039c2198ebb6ed38520e
So far this is the only issue we've hit in vendor code, and we've hit
it several times already. Rather than try to fix bullhead (the current
problem), let's just admit that the special case of 0 is a lot less
worrying.
Also fix the test expectations to correspond to the new abort message.
Bug: http://b/35455349 (crashes on 0)
Bug: http://b/35622944 (tests)
Test: ran tests
Change-Id: Iec57011fa699a954ebeaec151db2193e36d1ef35
