Gregor Zattler <telegr...@gmx.net> writes: > Now I think I found something specific. I do open the org-mode > file in question with #+STARTUP: showeverything because otherwise > it’s slow. Today I discovered org-goto. This presents the > buffer in overview mode. So I started the profiler, started > org-goto and typed 11 letters in order to jump to a heading. > This takes quite some time while my notebooks fan is making noise > and Emacs uses 100% cpu according to htop. > > I did this two times with the same org-mode file: > > - with Org-mode version 8.3beta (release_8.3beta-1291-gd6e601 @ > /home/grfz/src/org-mode/lisp/) on GNU Emacs 25.0.50.4 > (i686-pc-linux-gnu, GTK+ Version 3.14.5) of 2015-07-17 on boo > > - with Org-mode version 8.2.10 (release_8.2.10 @ > /usr/local/stow/emacs-snapshot/share/Emacs/25.0.50/lisp/org/) > on the same version of Emacs. > > I do not know how to read these reports, but to me it seems, that > most of the cpu cycles are used by org-context, org-end-of-item, > org-list-struct and org-list-context. The memory reports do not > seem to be of interest but I attach them for the sake of > completeness.
I removed a call to `org-context' in 985420eceb3661aae8b1caafc95e959ab12ad1cb. Does it fix your problem? For the record, `org-context' predates Org Elements and should ultimately be removed from the code base. Regards,
