Re: [O] C-c C-c in table takes 12 seconds

[Rasmus]

Hi,

Nicolas Goaziou <address@hidden> writes:

> Karl Voit <address@hidden> writes:
>
>>> Issue: Whenever I change table cells via TAB, it takes twelve
>>> seconds with 100% CPU core:
>>>
>>> org-table-next-field very slow:
>>> #+BEGIN_EXAMPLE
>>> - command-execute                           31397  99%
>>>  - call-interactively                       31397  99%
>>>   - yas-expand                              31298  98%
>>>    - yas--fallback                          31298  98%
>>>     - call-interactively                    31298  98%
>>>      - org-cycle                            31298  98%
>>>       - call-interactively                  31291  98%
>>>        - org-table-next-field               31291  98%
>>>         - org-table-align                   15676  49%
>>>          + org-activate-bracket-links          36   0%
>>> [...]
>>> #+END_EXAMPLE
>
> Could you provide an ECM? I'm unable to reproduce it.
>
> In fact I remember encountering issues using yasnippet within Org
> tables, but never related to speed.

So I don't know if this is related.  But as discussed previously I also
see major random lockdown issues with tables.  For me it seems to be a
combination of company, flyspell, tables, and maybe entities.  I have
yasnippet installed but only 'cause it's required by some packages.  I
never activate it.

When looking at the profiler, I notice avl-tree--do-delete (and other
avl functions) and org-element--cache-compare taking lots of speed.  I
think I have also had this issues in mail-buffers where I use orgstruct
modes.  Again org-element--cache showed up.

Attached is a backtrace from yesterday where I had to stop Emacs with
kill -usr2.  The file I was working on has many tables with conjugations
of words and some entities.

I can get slowdowns in emacs -q but not the 100%

Cheers,
Rasmus

-- 
Bang bang

backtrace.txt
Description: Text document