The test none/tests/mips32/MoveIns.c did not initialize destination register,
and that is important for movn.s and movn.d instructions. This improves the
test as we will get the same stdout with different compilers and options now.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12661
Linaro gcc 4.6.3, running on an A9. Not sure why these need to be
updated, but they results are same as native on said platform, so am
not worried.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12660
This patch fixes incorrect handling of sys_shmdt for MIPS. Linux wrappers have
been added for sys_sigprocmask, sys_timerfd_create, sys_timerfd_gettime, and
sys_timerfd_settime on MIPS.
The bug has been reported at https://bugs.kde.org/show_bug.cgi?id=270777 as
sh_mat issue, and it can be reproduced with shmat-sample.c from
https://bugs.kde.org/show_bug.cgi?id=222545.
The change also fixes sigprocmask from memcheck tests.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12658
Syscall pipe returns two descriptors in two registers, and thus a generic
Linux wrapper is not sufficient. This fixes none/tests/fdleak_pipe.
MIPS port is currently tracked as bug #270777.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12655
such that pthread_cancel() works properly. See also Ulrich Drepper, Cancellation
and C++ Exceptions, 2010 (http://udrepper.livejournal.com/21541.html).
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12651
Note that VG_(arena_memalign) is not used by core or tools for the moment.
We have one single maxima for both the V core/tools and the client.
Enhanced memcheck/tests/memalign2.c to test 4 Mb and 16 Mb alignments.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12642
ARM by simply scanning up and looking for words that look like they
might be return addresses. Last-ditch hack for when the CFI trail
goes cold.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12641
This is needed to have the outer valgrind able to do stack
traces of the inner Valgrind before the main thread runs.
Without this, the outer crashes (segv) when doing a stack trace.
In mtV.txt, an ugly kludge was described to avoid this crash.
This is the clean solution replacing the kludge.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12638
GDB can create watchpoints watching the same address.
This was causing assertion failures.
To handle this, hash table (with key watched address) is replaced
by an xarray of address/lengh/kind.
Fully identical watches are ignored (either not inserted, and
not causing a problem if already deleted).
gdbserver_tests/mcwatchpoint enhanced to test duplicated watchpoints
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12637
vg_regtest failed on MIPS as filter_stderr was not an executable.
MIPS port is currently tracked as bug #270777.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12625
on s390 has only 8 byte alignment gcc cant guarantee 16 byte alignment
for local variables. As a global variable gcc can do that.
git-svn-id: svn://svn.valgrind.org/valgrind/trunk@12620
