On Thu, Feb 10, 2005 at 10:25:25PM +0100, Reuben Thomas wrote: > > I agree that the line your patch delete is not needed. But I > >don't understand how it can prevent a crash. > > The only thing I can think of is that if the supplied vasprintf is used > with built-in vsprintf, and vasprintf gives a shorter result, then > vsprintf could overrun a buffer.
Indeed, but the supplied vasprintf is not used on amd64. Another solution is that vasprintf failed due to a memory allocation failure. Since vasprintf return code is not tested, calling vsprintf could generate a segfault. -- Nicolas Duboc <[EMAIL PROTECTED]>
signature.asc
Description: Digital signature
