# 2022-05-04 - coreboot Leadership Next meeting 2022-05-16
## Significant Decisions * Pointing one vendor's mainboard directly at another vendor's mainboard directory to reduce duplicated code is not allowed going forward. * For ODMs with rebranded devices, it is allowed, if desired by the submitter to implement the code as a variant of the ODM's mainboard. If the vendor wishes to maintain a mainboard implementation separate from the ODM's, that is allowed as well, and should not be forced. * Add a checker for markdown files similar to checkpatch. * Add a documentation build check on jenkins. * Get rid of the doxygen builds. The doxygen comments do not need to be removed, but it's not a recommended documentation method for coreboot anymore. * Get the SMM patches reviewed and merged. Arthur will work to port them to the required branches after they're merged to the master branch. (Thank you Arthur). * Intel should be thanked for opening the PSE firmware. * coreboot needs to support generation of a software BOM.
## Attendees Arthur Heymans, David Hendricks, Felix Held, Felix Singer, Jay Talbott, Julius Werner, Kapil Porwal, Lean Sheng Tan, Martin Roth, Matt DeVillier, Patrick Georgi, Ron Minnich, Tarun Tuli, Tim Crawford, Werner Zeh, Piotr Król
## Minutes * [Martin] How do we want to handle variants between vendors? * Example: Vendor1 and Vendor2 both have a design based on a reference design, so they’re very similar in most respects. To minimize repeated code, we could have vendor2’s mainboard Kconfig point into vendor1’s mainboard implementation. This just seems… awkward. * Can we just not worry about de-duping the mainboard code between the two vendors in that case? * Code that can be moved to EC or to SoC or other generic places should be moved there. * Let’s not point to the other mainboard. * [Felix h] what If it’s actually the same mainboard? * [Jay] Editing one mainboard affecting another by an unrelated vendor would be a huge quagmire - a lot more risk. * [Ron] This burned us badly early on. * Would it make sense to build it as a variant of a device? * So the main board would be under the clevo directory as a variant. * Allow this case. All other cases would need to be discussed again first.
* [Martin] Add markdown checker? This can be added to the checkpatch build to give feedback on markdown files. * [https://review.coreboot.org/c/coreboot/+/63805%5D(https://review.coreboot.or...) * Yes, let’s do this. * Let’s also add a test for the documentation in jenkins * [https://pre-commit.com/%5D(https://pre-commit.com/)
* [Martin] What do we want to do about doxygen? Can we kill it off, or should it be fixed up? * [https://doxygen.coreboot.org/%5D(https://doxygen.coreboot.org/) * [Ron] This was a 20 year experiment that failed. It’s time to kill it.
* [Martin] Who will add the patches for the SMM security issue to the 4.14. 4.15 & 4.16 branches? * Need to look at 4.11 branch as well as the SMMv2 was ported back to the 4.11 branch. * After the patch is merged to master, Arthur will look at at porting it back. * Needs to be reviewed.
* [Werner] Intel has released EHL PSE to the public with an apache license [https://github.com/intel/pse-fw%5D(https://github.com/intel/pse-fw) * A first test confirmed that the basic functionality (mainly OS and HW-init) is up and running with a simple ‘Hello, world’ application. * Tooling issues on the build machine. * Werner will work with martin to add required tools to the coreboot build machines. * Smaller than the released binary. * Good enough to get all of the NICs running. * Thank you letter to Intel for doing this. * Werner will write this up and post it, then announce on the mailing list.
* [Jay]: Any sign of the server, Martin? * Not yet. * Final meeting will be tomorrow, and Jay will ask about it again at that point.
* [Martin]: Sent out a survey to consultants. * Will send out a reminder for companies who have not responded. * Then will update the consulting page and submit for review.
* [Arthur] Software BOM generation: Working on creating a list of contents in a firmware image. Mandatory feature to sell to the US government in the future. * Design is WIP. * coreboot is designed to be able to add things after the image is built. * Look at one approach for fmap region and one for cbfs. * Look at a verification tool. * [https://review.coreboot.org/c/coreboot/+/63639%5D(https://review.coreboot.or...)
## Notice Decisions shown here are not necessarily final, and are based on the current information available. If there are questions or comments about decisions made, or additional information to present, please put it on the leadership meeting agenda and show up if possible to discuss it.
Of course items may also be discussed on the mailing list, but as it's difficult to interpret tone over email, controversial topics frequently do not have good progress in those discussions. For particularly difficult issues, it may be best to try to schedule another meeting.
-
coreboot org