[Search for users] [Overall Top Noters] [List of all Conferences] [Download this site]

Conference noted::windows95

Title:Microsoft Windows 95 ("Chicago")
Notice:Please read topics 1 to 22 before writing anything
Moderator:EEMELI::BACKSTROM
Created:Mon Nov 14 1994
Last Modified:Fri Jun 06 1997
Last Successful Update:Fri Jun 06 1997
Number of topics:2958
Total number of notes:19968

2834.0. "GPFs in WIN95" by KAOFS::LEBLANC () Fri Mar 07 1997 12:23

    I have a PLx 566 running WIN95 and I get General Protection Faults
    (GPFs) intermittantly. It can happen within any application; Screen
    Saver, KEAterm, Lotus Notes. I always have to reboot the system to
    recover.
    
    I read somewhere that duplicate .DLL files can be a problem and cause
    GPFs but I'm don't feel confident deleting these things. Is there
    any suggestions as to how I should proceed?
    
    Is there a way to PURGE files in WIN95 or DOS as there is in VMS?
    
    Thanks
    
    Al
     
    
T.RTitleUserPersonal
Name
DateLines
2834.1COOKIE::FROEHLINLet's RAID the Internet!Fri Mar 07 1997 14:5910
    GPFs are either caused by a hardware problem (bad memory, not likely)
    or software bugs (most likely). The GPF screen show you where the GPF
    happened (module name) to give an indication. Mostly it's in USER or
    KERNEL, which doesn't help. Some hardware (video and printer) comes
    with buggy support software. Does it happen whenever you print
    something? Also there's a chance that the support files on disk are
    corrupted in which case you have to re-install basically everything.
    (happens to me sometimes running my own kernel-termite-programs).
    
    Guenther
2834.2skylab.zko.dec.com::FISHERGravity: Not just a good idea. It's the law!Fri Mar 07 1997 15:4111
I've had this happen in LOTs of different computers, though often I don't get a
blue screen...I just get a halt.

The only thing in common with all the machines I have seen it on is that they all
have a Diamond graphics card with the Diamond W95 driver.  I wonder if there may be
some common code in all of the Diamond drivers which is buggy.  That's my current
hypothesis, anyway.  I suppose I could go back to the Microsoft generic S3 (or
whichever chip) driver, but it happens only often enough to be annoying...never
enough to push me to action!

Burns