# 20 October 2021 - coreboot Leadership

##Attendees:

 Felix Singer, David Hendricks, Arthur, Werner, Piotr, Lean Sheng Tan, 
Marshall, Jason Glenesk, Jay Talbott, Matt DeVillier, Patrick Georgi, Tim 
Crawford, Martin, Stefan, Felix Held

## Agenda:

* [patrick g] “Verified+1” permissions on gerrit repos that aren’t build tested 
(e.g. homepage.git): open up? For which groups?
    * For which repos?
    * Blobs - Only people appointed by Stefan for legal reasons.
    * Homepage: All submitters? Owner’s +2 or verified doesn’t count.

* [martin] Clean up the coreboot root folder?
    * /spd, /gnat.adc, /Makefile.inc, /toolchain.inc -> src
        * Move spd directory to src/lib (or src/device?)
        * Look at moving gnat.adc to /src.  Might be difficult to move
        * Move toolchain.inc to /src - only included by top level makefile
        * (keep Makefile.inc in place, otherwise it has to refer to ../util 
which is ugly)
    * /configs -> util/abuild
        * Having this here may help with visibility of platforms
        * Leave this here, remove subdirectory
    * /MAINTAINERS, /AUTHORS -> Documentation
        * Leave both of these at the top level
        * MAINTAINERS is machine readable and not really “Documentation”, 
AUTHORS is a “standard file”
    * Remove COPYING - it’s a copy of the GPL V2 license, which we have under 
/LICENSES
        * Leave this here: standard location so license checking and compliance 
tools are happier to find it there.
    * Update README.MD to point into /Documentation and each .md in other 
directories.
        * Marshall will do this.  Thanks!
    * Add a .mailmap file [to correctly map authors to an email 
address](https://git-scm.com/docs/gitmailmap).
        * Yes
    * Should we lock it to the result after moving anything so that additional 
objects can’t be created without deliberation?
        * Add a lint test to keep other files from being added without 
discussion.

* [Martin] When do we need another coreboot EFI working group meeting?
    * Next tuesday - send out invites after this meeting.
    * Lots of issues joining last time.

* [Werner] Discuss patch 
[CB:55367](https://review.coreboot.org/c/coreboot/+/55367) regarding Intel PSE 
(Programmable Service Engine)
    * Must be loaded at boot time by a piece of firmware.
    * Discussed osfc2020 - zephyr based, but not open yet.
    * Patch just adds the loading from cbfs so that FSP can load it.
    * Yet another blob without notifications.
    * Blob is unfortunately mandatory for the platform.
    * Initializes the ARM processor then stops running?
    * It may be possible to write an open version?
    * The process of open sourcing the code for this takes a long time, so this 
is (in theory) a temporary solution.
    * The processor is EC-like, handles realtime interface, has I2C interface.
    * PCIe devices can be assigned to either the device or X86.
    * Sheng and Werner will verify that this is more similar to an EC than the 
ME.
    * If this is EC-like, this shouldn’t be an issue.

* [Arthur] Logging: (optional) printing the error level in front of the message?
    * Would need to have a little logic to allow a line to continue without 
printing again.
    * Loglevel in brackets.
    * Sure, let's try this - so long as it's optional, there's not much of a 
downside.
* [Martin] Adding requirements for blobs in coreboot’s repos
    * Nico [suggested requesting/requiring some additional items for 
blobs](https://mail.coreboot.org/hyperkitty/list/[email protected]/thread/AV5LYYLMMFWYIR5R2OW6A2M7XGI2SBHX/).
    * We already have[ a list of 
requirements](https://review.coreboot.org/plugins/gitiles/blobs/+/refs/heads/master/README.md),
 so updating that doesn’t seem unreasonable.
    * How do we want to handle non-coreboot-owned blob repos - 
fsp/intel_microcode/amd_blobs/qc_blobs/
    * Let’s review the blobs directory and see what’s still needed.
    * Revisit next meeting

## Mailing list discussions of interest:

* Mailing list question about [arm64 ports of 
coreboot](https://mail.coreboot.org/hyperkitty/list/[email protected]/thread/TOWWCZL6XUKIPQO5SRX4JRHEPO3SI2XJ/)
* Mailing list: [firmware and bootloader log 
spec](https://mail.coreboot.org/hyperkitty/list/[email protected]/thread/OUGJENEVQAVL7AA2YIIHAW6BQNNSALGQ/)
    * (LPC 2021 System Boot and Security MC discussion about 
that)[https://youtu.be/JwU1_hyOzMI?t=4329]

## Ongoing items:

* [New videoconferencing system?](#bookmark=id.ehzxgz9a1ta0)
    * Big Blue Button (maintained install: senfcall.de, [coreboot 
room](https://lecture.senfcall.de/pat-yut-ayj-bc6) set up by pgeorgi)
        * At the end of the meeting, please try to join the coreboot room here, 
so we can see if it’s working for people.
        * What happens if the host doesn’t join - seems fine.
        * This was tested by a number of people after the meeting, and it seems 
better than bluejeans.  There weren't any reports of people not being able to 
get in, though some people had a little difficulty getting their audio set up.  
It seems like they were able to resolve this
    * Bluejeans. coreboot.org has an account here already, but lots of folks 
had issues joining the coreboot EFI working group meeting on Oct 12, so 
probably not?
        * Kind of weird - different versions per subscription level?
        * Definite NO.

* Documentation updates
    * Hiring a tech writer?
        * Not right now.

* coreboot 4.15 release ~Nov 1
    * First notice went out to mailing list.
    * Contribute to the same patch or check it in and add to it?
        * Martin to review.

* [Updating the 
builder](https://docs.google.com/document/d/10jQc1N67Pck2pE3pkyke7g-YFDIBqum6SkQH7UdXSpE/edit#)
    * Felix testing NixOS.
    * Need to determine builder differences - decide what builder can build 
what at runtime based on configuration.
    * Packaged coreboot toolchain for NixOS.
        * 4.14 package - making a stable release.
    * Set up build test with flashrom
        * Test both with linux & bsd
_______________________________________________
coreboot mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to