Thanks.

The commit message says "are no" instead of "of not". You might consider something simpler because users can really filter out whatever they want.

Nit: The Committer's guide says the subject line after the prefix should be capitalized. https://docs.freebsd.org/en/articles/committers-guide/#_capitalize_the_first_letter_of_the_subject

Thanks @guest-patmaddox. Much better. I haven't had much time to look into this or bug 292973. Do you have any thoughts on the shortcomings I mentioned in my last comment?

• Replace the hacky tree-sitter indent helper functions with cleaner rules from Pat's configuration with some minor tweaks:
  • Use c-ts-mode--simple-indent-rules instead of c-ts-mode--indent-styles and c-ts-indent-offset instead of c-ts-mode-indent-offset.
  • Add a fix for initializer list opening braces.
• Provide, a closing footer, and an improved docstring.

Really update to the latest commit.

• Update to latest commit
• Do not include non-reproducible information such as build date, instead display RPIBOOT: FreeBSD package (version 20260527)

I believe Emacs calls ls --dired and pipes the output. I think that means we have a problem because you are using ftell(stdout) (*in three different places), which will always return -1 when the output is redirected to a pipe.

Very cool.

Base the port on the HEAD of upstream's repository, which includes some helpful fixes since the last tag. Upstream is using date-base tags, so just using lastest commit and calling it 20260527 should be fine.

Tweak COMMENT

Rename the port to rpi-usbboot

We could, and there are advantages to having the portname match the directory. For example, when you have the package installed and, based on the package name, search the ports tree for rpi-usboot, you'll get a hit.

Add an rpi- prefix for the package name

Accepting because everything looks correct to me, with the caveat that I don't often look at this code, and I haven't tested (because I don't have an Intel processor whose signature appears in an extended table).

Looks good. There is a typo in the commit message: `s/sotred/stored/'.

Looks good to me. There is a typo in the commit message: s/sepcific/specific/.

Thanks both. Nice catch to fix the inverted condition. If I understand correctly, before this fix, we would never have detected an extended signature, and thus never applied microcode for processors only listed in the extended signature.

Fix a typo (s/feee/free/)

• Use pread(2) as suggested by markj, which means we can remove extsig_off save/restore around copy_entry()
• Reference the 'Optional Extended Signature Table' in vol 3A section 12.11.2 of the March 2026 SDM

In D57054#1307326, @kib wrote:

I looked at the referenced bug. Could you please conduct an experiment where the failing machine is booting kernel with just the WRMSR write to actually update the ucode is commented out? I mean, remove only the instruction, but keep the setup for wrmsr_safe intact. Does the machine boot in this case?

In D57054#1307326, @kib wrote:

I looked at the referenced bug. Could you please conduct an experiment where the failing machine is booting kernel with just the WRMSR write to actually update the ucode is commented out? I mean, remove only the instruction, but keep the setup for wrmsr_safe intact. Does the machine boot in this case?

In D57054#1307284, @imp wrote:

What are the risks this a avoids?

Thanks for sharing. I'm going to spend more time testing these three configurations (@imp's, the classic one in freebsd.el, and the tree-sitter configuration) by opening lots of source files and doing things like C-M-\.

Another minor comment tweak.

Minor comment tweak

This was meant to be a temporary review to share ideas with @thj. Closing.

Looks good. I'll commit with some minor tweaks to the commit message to comply with the guidelines in the Committer's Guide, and to remove mention of the patches for the UUID includes. BTW, if you're not already using it, there is a commit template you can use by doing git config --add core.hooksPath .hooks. Look good to you?