On Thu, Jul 27, 2000 at 09:17:43AM -0400, Thomas Briggs wrote:
> 
>    Backed out part of Goran Thyni's patch which caused Win32 to GPF startup
> (due to attempting to strdup a freed pointer).  Why this didn't cause
> problems on Linux I have no idea (I tested it without this fix and it didn't
> seg fault... ;shrug)

The pointer that is strdup:ed was pointing into a released
parsetree. Under happy circumtances it work OK, but it is
a time bomb.
Good catch.

regards,
G�ran.





Reply via email to