Re: BUG REPORT: redisplay assertion failure
Martin Buchholz <martin(a)xemacs.org> writes:
Fatal error: assertion failed, file
/net/wanchai/export1/xemacs/home/martin/x/ws/dev/src/redisplay.c, line 6799, cur_elt >=
0
This has been reproduced in debug and optimized, mule and latin-1
xemacs. I cannot (yet) reproduce it in a -vanilla xemacs.
This is probably (somehow) related to the redisplay fix for the
widening crash (1998-09-11 in the ChangeLog). There were some other
reports of this happening but nobody could ever reproduce it. I still
hope that the fix is OK and we are just exposing different redisplay
bugs this way...
Are you using scroll-in-place? That seems to be pretty good at
triggering problems in redisplay (which is a Good Thing).
I might work on fixing this, but I'm afraid I won't return
from an
excursion into redisplay.c for a year or so.
Yeah, I know what you mean, but it would certainly be great for XEmacs
if *you* spend a year improving/fixing redisplay :-) I really think we
should do something about redisplay: either rewrite it or at least
generate some helpful documentation.
#8 0x1ebb10 in redisplay_window (window={...}, skip_selected=0) at
/net/wanchai/export1/xemacs/home/martin/x/ws/dev/src/redisplay.c:5166
could you print the values of:
w->start_at_line_beg
beginning_of_line_p(b,startp)
in this frame?
Currently I don't have any time at all to look into this (I got a
deadline for some papers on Friday and still need to prepare the talk
for next week :-) Maybe we can talk about it in Japan.
Gunnar
--
Gunnar Evermann
Speech, Vision & Robotics Group
Engineering Department
Cambridge University