In message <9704031822.AA09766@lightning.oscs.montana.edu>Michael L. Hitch writ
es
>On Apr  3,  3:06am, maximum entropy wrote:
>> >From: Thomas Seidmann <seidmann@dcs.elf.stuba.sk>
>> 
>> >problems with the behaviour of the mouse on the 5000/25. First, there is a
>> >part of the screen (approximately the right 1/5), which can't be reached
>> >by the pointer. Second, while having more than one window on the desktop,
>> >mouse motions cause unpredictable effects, for example selection in one
>> >window, focus changes etc. which should be all fired by pressing a mouse
>> >button (which never happened). Any clue? TIA.
>> 
>> No clues here, but I thought I'd mention I'm having similar symptoms
>> with Xcfbpmax on a 5000/120 with PMAG-AA frame buffer.  What frame
>> buffers are in your machines?
>
>  The mfbPosCursor() procedure in pmax/dev/mfb.c is making the same checks
>on the cursor position as the ims332PosCursor() for the 5000/xx.  

Well, that's a bug....

>The x
>position should compare with the framebuffer width and the y position
>with the framebuffer height.
>
>Michael

Sorry, I'm confused.   which width and height? 

The screen pixel width/height (1280x1024) or where they line in memory
(2048 bytes between verticall-adjacent pixels, from the preceding patch)?

i think you're saying the patch posted to port-pmax needs to be applied,
*and* the cursor-in-range checks need to be fixed, e.g, like the sfb checks.
Is that right? 

(I have yet another sinus infection and it leaves me rather fuzzy-headed,
so I'm almost certainly missing the obvious.)