Hamish Moffatt was screaming at Peter McGrath because of Win32s 1.30
?.


Sorry to butt in to a thread here, but I want to add my 2 cents
worth....

HM>  > This is actually not an OS/2 problem, but a hardware one. Running the
HM>  > above program does NOT lock up my OS/2 machine (but DOES lock up WIN95)
HM>  > Why is this so I hear you ask?. Well a geniune Intel processor (486 or
HM>  > Pentium) has a watchdog timer which regularly generates a non maskable
HM>  > interrupt. OS/2 exploits this and regains control of the computer (in
HM>  > the above situation) and you can kill the offending program.
HM> 
HM> How long should it take to realise? I didn't leave it long 
HM> before a reset, but I have seen OS/2 recover from some 
HM> crashes within about a minute previously. However, the 
HM> mouse cursor at least usually still worked. I'm using an 
HM> AMD 5x86-133.
HM> 

I use an AMD486DX4-100, and was running telix as a background task
(while downloading), and reading a mail packet in the foreground with
an os/2 mail reader. I crashed the reader (ie it stopped responding)
and pressed the enter button (to end it). nothing happened. This is not
a new thing with my pc, and I knew (suspected anyway) that the
background d/l would be still going. so I left it and watched tv for 15
mins while the dl finished. I came back to reboot it, only to find that
warp had killed the mail reader and the downlaod was complete, without
errors.

If I had been using windoze or win95 software, I doubt for one second
that the background task would have not locked up.

moral: patience when dealing with nonresponsive tasks is required. :-)

 Party On!

___
 X KWQ/2 1.2c NR X Any sufficiently advanced bug will become a feature.

--- Maximus/2 3.00