emacs-pretest-bug
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Assertion failure in xdisp.c


From: Kim F. Storm
Subject: Re: Assertion failure in xdisp.c
Date: Tue, 16 Nov 2004 01:10:06 +0100
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/21.3.50 (gnu/linux)

Did we reach any conclusion on this?

I cannot reproduce it now on GNU/Linux (don't think I tried earlier)...

Could you print *row and *w when this happens.


Stefan <address@hidden> writes:

> src/emacs -Q lisp/gnus/mm-util.el -f outline-minor-mode -f reveal-mode
> C-u 1000 M-x hide-sublevels RET  [ this hides all the function bodies. ]
> C-e                              [ shows the previously hidden file header ]
> C-a M->                          [ The first line show not visible any more ]
> M-<                              [ Back on the first line, the copyright
>                                    notice should be hidden again.  If not
>                                    try M-> M-< a few more times. ]
> C-e
>
> This last C-e (which should re-show the header with the copyright notice)
> aborts.  It aborts inside `find_first_unchanged_at_end_row' at the test:
>
>         if (!row->enabled_p || !MATRIX_ROW_DISPLAYS_TEXT_P (row))
>             abort ();
>
> On my Mac OS X machine, catching this abort in the debugger leads to
> a confusing stack trace (GDB tells me I'm in try_window_id called from
> try_window_id, and doesn't mention find_first_unchanged_at_end_row).
> Replacing the above by
>
>         eassert (row->enabled_p && MATRIX_ROW_DISPLAYS_TEXT_P (row));
>
> helped a great deal.
>
> If I remove the assertion, everything works just dandy.  Is the assertion
> still correct (and thus some bug needs to be fixed) or should the assertion
> be fixed/eliminated?
>
>
>         Stefan

--
Kim F. Storm  http://www.cua.dk





reply via email to

[Prev in Thread] Current Thread [Next in Thread]