Re: scratch/igc: Implications of MPS being asynchronous

[Eli Zaretskii]

> Date: Sat, 25 Jan 2025 10:00:12 +0000
> From: Pip Cet <pipcet@protonmail.com>
> Cc: Gerd Möllmann <gerd.moellmann@gmail.com>, stefankangas@gmail.com, 
> emacs-devel@gnu.org, eller.helmut@gmail.com
> 
> "Eli Zaretskii" <eliz@gnu.org> writes:
> 
> > I now see this comment in compact_font_caches:
> >
> >   compact_font_caches (void)
> >   {
> >     // Cannot be done with MPS, which doesn't seem to be tragic
> >     // because there is an option to turn it off.
> >   # ifndef HAVE_MPS
> >
> > Why cannot font caches be compacted with MPS?
> 
> How important is this?

I don't know.  The person who added that to GC considered it extremely
important; see

  https://lists.gnu.org/archive/html/emacs-devel/2013-10/msg00740.html

But since then we found that it could also cause problems, see
bug#24634, bug#24565, bug#16093, bug#16128.  Some, if not all, of them
were since fixed.

> How often do we need it, and when we need it,
> would it be acceptable to delay for one or several "GC cycles" until
> eventually performing this operation?

I think delaying the compaction should not be a problem.

> If you have the time, could you briefly suggest how we might want to
> test this code?  I have many fonts but that's about all I know, and some
> people have thousands of them, IIUC.

The above reference to a discussion on emacs-devel has a code snippet
you could use.