This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [RFA] Fix leak in linespec.c


>>>>> "Philippe" == Philippe Waroquiers <philippe.waroquiers@skynet.be> writes:

Philippe> 54 gdb/testsuite/v7_outputs/gdb.base/reread/gdb.log == Conditional jump or move depends on uninitialised value(s)

I think I have a fix for this one.  I plan to check it in soon, maybe
today, as it's been around for quite a while.

Philippe> 1 gdb/testsuite/v7_outputs/gdb.cp/inherit/gdb.log == Invalid read of size 8
Philippe> 1 gdb/testsuite/v7_outputs/gdb.cp/virtbase/gdb.log == Invalid read of size 2
Philippe> 5 gdb/testsuite/v7_outputs/gdb.cp/virtbase/gdb.log == Invalid read of size 8

I believe I saw these with ASAN as well.  I think what happens here is
that the pretty-printer code creates a value from a virtual base class
slice of an object, but doesn't "inflate" it to the full object,
resulting in some out-of-bounds reads.

Tom


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]