[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-bugs-dist
Subject:    [valgrind] [Bug 337794] memcheck: stack overflow segv message should show callstack
From:       Philippe Waroquiers <philippe.waroquiers () skynet ! be>
Date:       2014-07-28 21:41:05
Message-ID: bug-337794-17878-jgILcdYzPI () http ! bugs ! kde ! org/
[Download RAW message or body]

https://bugs.kde.org/show_bug.cgi?id=337794

Philippe Waroquiers <philippe.waroquiers@skynet.be> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |philippe.waroquiers@skynet.
                   |                            |be

--- Comment #1 from Philippe Waroquiers <philippe.waroquiers@skynet.be> ---
In case of (suspected) stackoverflow, the code explicitely produces
a stacktrace of only one single IP.

This logic was added in m_signals.c by the below revision.
Removing this is trivial (just do not do the check to call
VG_(record_depth_1_ExeContext) in m_signals.c for the SEGV handling.
Unclear if the below revision is just to be very cautious
or if a real problem is/was created by a full exe context

------------------------------------------------------------------------
r7304 | sewardj | 2007-12-21 02:24:59 +0100 (Fri, 21 Dec 2007) | 11 lines

Add a new method VG_(record_depth_1_ExeContext), a trivial derivative
of VG_(record_ExeContext), which just records the first stack frame
but does not attempt to unwind the (guest) stack.  This is useful in
situations where we suspect unwinding the stack might cause a
segfault.

Use this in m_signals, when getting a backtrace following a guest
segfault.

-- 
You are receiving this mail because:
You are watching all bug changes.
[prev in list] [next in list] [prev in thread] [next in thread]