Re: macOS metal rendering engine in mac port

[Aaron Jensen]

On Sun, May 23, 2021 at 8:56 PM Aaron Jensen <aaronjensen@gmail.com> wrote:
>
> Here are some interesting profile pictures. On the left is emacs -Q w/
> display-line-numbers-mode. On the right is a partial version of my
> config with some of the things I mentioned before removed. Most of the
> difference between the two comes in the amount of time spent in
> merge_faces. My config ends up spending roughly 7.5x more time in
> merge_faces (1500ms vs 200ms). What causes that to get slower?

One thing of note is that emacs -Q has 129 faces for me. My config has
anywhere from 600-800 faces depending on what's loaded it looks like.
lface_from_face_name_no_resolve does a linear time search for a face
every time it's called. It's called over 700k times when I do my
xdisp.c scroll up test. merge_faces is called over 150k times, each
time with line-number-current-line or line-number and 0's for face_id
and base_face_id. I don't really know what this code is doing at all,
but all of this seems suspect. Thoughts?