Mailing-List: contact cygwin-developers-help AT sourceware DOT cygnus DOT com; run by ezmlm List-Subscribe: List-Archive: List-Post: List-Help: , Sender: cygwin-developers-owner AT sourceware DOT cygnus DOT com Delivered-To: mailing list cygwin-developers AT sourceware DOT cygnus DOT com From: Chris Faylor Date: Sat, 20 May 2000 02:28:40 -0400 To: cygwin-developers AT sourceware DOT cygnus DOT com Subject: Win95 problems revealed? Message-ID: <20000520022840.A10995@cygnus.com> Mail-Followup-To: cgf AT cygnus DOT com, cygwin-developers AT sourceware DOT cygnus DOT com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.1.12i I may have figured out at least one of the long long long standing problems with Windows 95. The function GetThreadContext seems to sometimes return bogus information. This function is used by the signal handler code to figure out things about the thread which is supposed to receive a signal. In particular, cygwin gets information concerning the stack pointer and current PC from this function. So, if the information is wrong, the results are catastrophic. I've really ripped the signal handler apart in the last few months and this has been partly responsible for revealing the Windows 95 bug. I noticed the problem by running rxvt continuously on my Windows 95 system. It would randomly cause a blue screen. Liberal use of printfs pinpointed the problem. I've worked around the problem in the current snapshot so I would be interested in hearing if I have broken anything before I unleash this on the world. So, if you have negative experiences with the latest code on Windows 95 or NT, I'd like to hear about it. Please send email to cygwin-developers with your experiences. Thanks, cgf