[comp.windows.x] bug? in X11R3 Sun server

mouse@LARRY.MCRCIM.MCGILL.EDU (der Mouse) (04/04/89)

There appears to be a bug in our server.  We run X11R3; this was
observed on a Sun-3/60 (with the cgfour display).

On the "monochrome" screen, Xsun will sometimes botch a display update.
In particular, when I start an xterm with

	% xterm -geometry -50+50 &

then the left edge of the window suffers from this problem.  To be more
precise, the leftmost eight pixels of the window are displayed both
where they should be and in the eight pixels immediately to the left of
where they should be.  This happens only when xterm scrolls the window,
and only to the area that gets scrolled.  

To more completely specify exactly how xterm is being started:  I have
the following lines in my .Xdefaults file:

	*Foreground:white
	*Background:black
	*borderColor:white
	
	XTerm*reverseWrap:on
	
	xclock*update:60
	xclock*hands:black
	xclock*highlight:white

(presumably the xclock settings don't have much to do with xterm :-),
the following my .Xdefaults-`hostname` file,

	*Background:#000048

and xrdb -query prints nothing.  It appears to be independent of what
other windows are present and happens both with no window manager
running and with my own (still under development).

This appears to happen at other places on the screen; I have seen
display glitches appearing sometimes under the mouse cursor which look
like the same problem, but they aren't repeatable at will.  This is.

When I use xmag to look at the corrupted area, the magnified view also
shows the corruption, but I doubt that means much, because xmag will
also magnify the crud resulting from "echo foo > /dev/console".

Are we missing a patch, or is this a real bug?  (More details, right up
to a sample screendump, are available on request.)

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu