How I made Magit fast again

TLDR: here are the settings that made the biggest difference for me. YMMV.

(setq magit-commit-show-diff nil
magit-revert-buffers 1)

Magit is awesome. And it's getting better with regular releases, a more consistent interface, and much more. But since the release of 2.1, it's generally been slower for me. I'm not the only one. In particular, the status buffer would take multiple seconds to refresh after almost any action such as committing, checking out a branch, stashing/popping, deleting a file, etc.

Tarsius, Magit's maintainer, is clearly aware of the performance issues and working to fix them. It can't be easy to diagnose performance problems given the multitude of ways Magit can be configured, plus the huge variety of characteristics among all the git repos out there. Nonetheless, I'm sure performance will improve in future versions.

But I needed to do something about it in the near term. I searched online and while Magit has a page dedicated to perf settings, none of them helped me much. So I grepped the Magit source for defcustom and read all the docstrings in search of things to try. Here's what I found.

Setting 1: magit-commit-show-diff

Docs say: "Whether the relevant diff is automatically shown when committing." In practice, I found that this added a lot of time to committing. From the magit status page, I would hit c c and wait and wait for the commit message edit buffer to open. After setting this to nil, it opens quickly.

I prefer to review diffs as I stage files and hunks, not before writing the commit message, so this has the bonus of better matching my workflow.

Setting 2: ttmagit-revert-buffers

Docs say: "How file-visiting buffers in the current repository are reverted." The default value is usage which prints a message explaining what this setting is. Long ago I'd switched it to silent: "Revert the buffers synchronously and be quiet about it."

I want my buffers reverted because it's quite annoying to do work in a buffer whose underlying file has changed, so the option nil (don't revert) was no good. But I noticed a value I had never seen before:

NUMBER    An integer or float.  Revert the buffers asynchronously,
          mentioning each one as it is being reverted.  If user
          input arrives, then stop reverting.  After NUMBER
          seconds resume reverting.

This seemed promising. So I tried it with NUMBER set to 1. And it feels great. The status buffer can still take a while to finish refreshing, but I no longer have to wait for it. Emacs is responsive almost right away.

The risk, of course, is that not all buffers have been reverted. However, in practice, this has not been a problem for me. I haven't encoutered the dread "buffer has changed since visited. Save anyway?" prompt yet after several days.

The implication of this is that the time it takes to refresh the status buffer grows with the number of buffers visiting files in a given repo. And my experience also bears this out: with magit-revert-buffers set to silent, I found that the status buffer for a repo took much longer to refresh if I had 100 buffers visiting its files than if I had 20. So if you prefer to keep that setting, then you can get faster refresh times by vigilantly killing buffers when you're done working with them.

I went back to magit-revert-buffers = 1. With this setting, magit echoes the names of buffers as it reverts them. They fly by fast but a few times I've spotted the names of buffers I knew were not part of the repo I was working with, which leads to me to wonder if Magit might be unintentionally reverting buffers it doesn't need to revert -- that would be unfortunate.