Re: abort in isearch-backward

[Kim F. Storm]

Eric Hanchrow <address@hidden> writes:

> I cannot reliably reproduce this.
>
> Here's what I can remember that seems pertinent:
>
> I was using Gnus to read a newsgroup
> (nntp+news.gmane.org:gmane.comp.version-control.subversion.tortoisesvn.devel,
> if it matters), and had just asked for the entire list of article
> headers by putting point on that line in the *Summary* buffer (the
> number at the beginning of the line was 0, meaning there were no
> unread messages), hitting Enter, and then hitting Enter again (to
> accept the default answer to Gnu's question "How many articles").
> After a while, the screen showed lots of article headers (the default
> answer was about 9000 headers).  Now, I've got Gnus set up to "hide
> all threads" in its summary buffers; I don't know if that's relevant.
> Anyway, I did M-> to put point at the end of the buffer, and then did
> C-r k e y w RET.  That searched backwards and found the word
> "keyword", as expected.  I then hit C-r again to search for the next
> occurrence, and got the following 'abort':
>
> (gdb) bt
> #0  abort () at emacs.c:454
> #1  0x08068722 in next_element_from_buffer (it=0xbfffdc30) at xdisp.c:5630

If you still have the crashed emacs in gdb, can you pls try this:

xbacktrace
up 1
p *it

Thanks.

-- 
Kim F. Storm <address@hidden> http://www.cua.dk