https://bugs.kde.org/show_bug.cgi?id=496552
Paul Floyd <pjfl...@wanadoo.fr> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |pjfl...@wanadoo.fr
--- Comment #2 from Paul Floyd <pjfl...@wanadoo.fr> ---
That's a lot of errors. Unless you have solid proof you should assume that the
errors are genuine. They may be harmless and/or impossible for you to fix, but
that is not the same thing.
Can you get debuginfo?
Errors like
==10:51:28.388424 24239== Conditional jump or move depends on uninitialised
value(s)
==10:51:28.388531 24239== at 0x5520D89:
instanceKlass::link_class_impl(instanceKlassHandle, bool, Thread*) (in
/build/share/java/jdk-1.6.10_02/jre/lib/amd64/server/libjvm.so)
will be clearer if you can see the line:file info for libjvm.so. If these
errors really are benign then you should use a suppression file.
==10:51:28.895572 24239== Invalid write of size 4
==10:51:28.895611 24239== at 0xD158900: ???
==10:51:28.895617 24239== by 0xD1473EA: ???
==10:51:28.895635 24239== by 0x554464C: JavaCalls::call_helper(JavaValue*,
methodHandle*, JavaCallArguments*, Thread*) (in
/build/share/java/jdk-1.6.10_02/jre/lib/amd64/server/libjvm.so)
[snip]
==10:51:28.895740 24239== Address 0x5f46550 is on thread 2's stack
==10:51:28.895744 24239== 4096 bytes below stack pointer
These are even harder. I assume that the ??? addresses are in the Java
generated code. I have no experience in that kind of debugging. Maybe using
valgrind+gdb would help (I don't know what extras you can use in gdb). Also
these errors are all exact multiples from 1 to 6 of4k pages.
Thirdly there are some invalid reads.
==11:30:09.646556 24239== Address 0x10 is not stack'd, malloc'd or (recently)
free'd
^^^
doesn't that cause a segfault?
Finally errors like
==11:43:12.263417 24239== Invalid read of size 8
==11:43:12.263493 24239== at 0x4AA74E3: vfprintf (in
/usr/lib64/libc-2.17.so)
==11:43:12.263514 24239== by 0x4AD6F38: vsnprintf (in
/usr/lib64/libc-2.17.so)
==11:43:12.263519 24239== by 0x4AB23C1: snprintf (in
/usr/lib64/libc-2.17.so)
==11:43:12.263545 24239== by 0x19CFD9A2:
TransactX::LogFacility::formatLogLine(char*, int, TransactX::MITime&,
Transact::LogSeverity, ch
ar const*, char const*, char const*, int, unsigned long, char const*, int, int,
char const*) (LogFacility.cpp:151)
That looks like it is in your code. You should be able to examine the source to
work out what is wrong.
--
You are receiving this mail because:
You are watching all bug changes.
