This adds a test which verifies that watchpoints work on devices
which support them. It is motivated by issues experienced by Nexus
5 devices, where watchpoints work only sporadically.
These tests spawn a process, trace it, and verify it hits a
watchpoint of various sizes while executing on all available CPUs
(to help catch a recent bug, where some debugging functionality
was not available on non-boot CPUs). All x86(_64) targets are
expected to support this functionality. In case of arm(64),
watchpoints are expected to work on devices where the relevant
syscall indicates watchpoint presence. In case of mips, this test
is skipped as I could not verify that the test would work
correctly due to lack of hardware, and this test is intrinsically
hardware-dependend.
Change-Id: I035ac1822816cd1a3f3b56b985c2fb5fd664185d
linker didn't remove link to the soinfo from shared
namespaces on soinfo_unload, because it didn't keep
record of all namespaces the library is added to.
This change adds test for this and also fixes the
problem by introducing list of secondary namespaces
to soinfo, which is used to remove soinfo in
soinfo::remove_all_links().
Bug: http://b/28115950
Change-Id: Ifbf6e54f92fa6e88f86b6a8dd6dc22d4553afd22
Even though bionic linker does not support RTLD_LAZY - it
is good to know that its behavior matches glibc.
Bug: http://b/27930475
Change-Id: I1a47263aaa3dc44f9ac61fe77deb55a21e7f881a
The XXXX now lines up to the underlying uid and has no offset.
Work with AID_OEM_RESERVED uids.
Test uses hard coded values to catch changes in the API expectations
that may occur in private/android_filesystem_config.h.
SideEffects: names change, some product dependencies.
Bug: 27999086
Change-Id: Ic2b4c36de74ae009a44e14711c75834293828207
This was fixed upstream years ago. While we're here, let's switch to the
OpenBSD copy (because that's our majority upstream BSD, not because they
found and fixed this bug first).
Bug: http://b/28035006
Change-Id: I53dd915a8122bfd7a6d58f01f9902d1586a47e23
It turns out that at least the Nexus 9 kernel is built without CONFIG_QUOTA.
If we decide we're going to mandate quota functionality, I'm happy for us to
be a part of CTS that ensures that happens, but I don't want to be first, so
there's not much to test here other than "will it compile?". The strace
output looks right though.
Bug: http://b/27948821
Bug: http://b/27952303
Change-Id: If667195eee849ed17c8fa9110f6b02907fc8fc04
Implement the legacy SysV signal handling functions sighold(),
sigignore(), sigpause(), sigrelse(), and sigset() in terms of the newer
POSIX signal APIs. As of POSIX 2013 the SysV signal APIs are deprecated
but still required.
Change-Id: I4ca40e3d706605a7d1a30dc76c78b2b24586387d
Signed-off-by: Greg Hackmann <ghackmann@google.com>
Any caller of brk needs to handle failure by falling back to mmap as
there's no guarantee that other mappings are not placed above the brk
heap. Since jemalloc starts with mmap in the first place and dlmalloc
knows how to fall back to mmap, it's already a legacy API in Android.
Forcing it to succeed rules out stronger ASLR implementations where the
brk heap might not have any room to grow. It's also valid for the brk
implementation to simply return -1 with errno set to ENOMEM in every
case and that's a good way of finding anything relying on it. It would
make sense for it to be an optional kernel feature.
The brk heap is also not guaranteed to be initially page aligned. PaX
ASLR randomizes the internal bits. It was only mentioned in a comment to
explain the design of the test though.
Bug: 24233096
Change-Id: I16e9bc8677e796c73915b830b99b0ce39c02b31d
The compiler in master doesn't leave a gap between the buffer and the stack
guard canary.
Bug: http://b/27815668
Change-Id: Iae6681683881055c2719d661a2179f5305c23135
Built for fugu this was working fine, but built for generic x86/x86-64
the compiler was (a) optimizing out all the stack writes and (b) inserting
enough padding on x86-64 for the canary to be safely out of the way.
While here, let's tighten up this test so we test that it's sufficient to
only overwrite the buffer by one byte...
(cherry-pick of 6f90c1ac59eb4e7811b7b0e408615049c935c0a9.)
Bug: http://b/27815668
Change-Id: I80a646de4b30fd5c78df20fdaa7e3eb163585caf
Previously, the implementation of setjmp on x86_64 claimed that
sigprocmask would write to two longs' worth of bytes.
Bug: http://b/27856501
Change-Id: I9f32b40ac773a0cd91a976aace5bfba6e67fb0f8
Handle no longer is a pointer to soinfo of
a corresponding library. This is done to
prevent access to linker internal fields.
Bug: http://b/25593965
Change-Id: I62bff0d0e5b2dc842e6bf0babb30fcc4c000be24
POSIX makes "the CPU-time clock of the calling thread" (i.e.,
CLOCK_THREAD_CPUTIME_ID) a special case which returns EINVAL instead of
ENOTSUP.
However, the clock_nanosleep syscall treats this clock just like any
other, and returns -EOPNOTSUPP to indicate an unimplemented nanosleep
handler. So we need to handle this ourselves in userspace.
This change fixes the LTP clock_nanosleep01 testcase.
Change-Id: If3bed940d276834bcd114d8c17f96197e9384711
Signed-off-by: Greg Hackmann <ghackmann@google.com>
Make it easier to diagnose applications mucking with the contents of
jmp_buf by checksumming its contents.
Bug: http://b/27417786
Change-Id: I473bc2871dece23a9b9d02481945246160d671c6
bionic has the Linux-specific mntent.h but is missing hasmntopt().
Change-Id: I0ab7b83626c969704add4e64b37a6fc715d4a723
Signed-off-by: Greg Hackmann <ghackmann@google.com>
{get,set}domainname aren't in POSIX but are widely-implemented
extensions.
The Linux kernel provides a setdomainname syscall but not a symmetric
getdomainname syscall, since it expects userspace to get the domain name
from uname(2).
Change-Id: I96726c242f4bb646c130b361688328b0b97269a0
Signed-off-by: Greg Hackmann <ghackmann@google.com>
"I have none of these capabilities" is a legitimate response, and
distinct from "I don't know what my capabilities are". It's the latter
I meant to test for.
Bug: http://b/27810985
Change-Id: Id165423e62390051cbf23fdda54f28d94566cbe7
Even though the size parameter to epoll_create(2) is (otherwise) unused,
passing in size <= 0 is explicitly documented as an error.
This change fixes the LTP epoll01 testcase.
Change-Id: I044a38be823c2fa956b57e77cc66571dfae8a4bb
Signed-off-by: Greg Hackmann <ghackmann@google.com>
The clone syscall accepts NULL child stacks, interpreting this to mean
the child gets a copy of the parent's stack with copy-on-write
semantics. However clone(2) is explicitly documented to treat this an
an error.
"Fortunately" every architecture's __bionic_clone implementation pushes
something onto the child stack before making the clone syscall. So we
know fixing this won't break legacy apps, because any app that tried
using a NULL child stack would have died with SIGSEGV.
This change fixes the LTP clone04 testcase.
Change-Id: I663b34f34bc8dad2aa405c46e4eed4418cccca0d
Signed-off-by: Greg Hackmann <ghackmann@google.com>
Also guard both these GNU extensions with _GNU_SOURCE.
Also improve the tests to test each case on both the current thread and
another thread, since the code paths are totally different.
Bug: http://b/27810459
Change-Id: I72b05bca5c5b6ca8ba4585b8edfb716a1c252f92
pthread_barrier_smoke test uses WaitUntilThreadSleep() to wait until
BarrierTestHelper threads sleep in pthread_barrier_wait(). But this
is flaky as there a two futex_wait places in pthread_barrier_wait.
This patch modifies this test to avoid using WaitUntilThreadSleep().
Bug: 27780937
Change-Id: I4c36b82cce9345d5088f8854b289dc5bf7a08e8c
