Hi Louis, On Fri, Aug 09, 2024 at 07:59:49PM +0200, Louis Chauvet wrote: > Hi everyone, > > I'm excited to share some good news! I've recently completed the addition > of a ConfigFS interface to VKMS, which allows to configure VKMS from > user-space, at runtime. You should be able to: > - Create new devices > - Create planes/crtc/encoders > - Configure rotation, color range, color encoding > - Link planes, crtc and encoders. Nice! Thanks for the hard work. During the last 2 or 3 weeks I worked on ConfigFS support as well and I have an RFC series ready to be sent to the mailing list. My version is a bit simpler than yours because it doesn't implement extra features (color formats, etc) and it can be applied on drm-misc-next. I'll send the RFC as soon as I finish typing this email :) I don't have any preference about which ConfigFS series we end up using as long as we get the feature implemented. Instead, I think that having 2 different implementations is a great opportunity for better reviews because we have a good idea about how the implementation might look like and it'll make easier to find bugs. > The entire series can be found on my GitHub repository: > https://github.com/Fomys/linux/tree/b4/new-configfs > > This series is big, consisting of over 40 commits. Although it's not > completely cleaned up, all commits compile successfully and (almost) pass > checkpatch. > > I plan to split this series into several smaller ones: > > - Adding support for additional color formats > 4a4f75873cca..d74006d7f9c4 > - Reintroducing the writeback algorithm > 9e74d259e1be..f839dcf6a7d8 > - Clarifying documentation > b3bfd0ba2283..93945f0fbfc7 > - Properly splitting headers > c70018038572..67258bd8a180 > - Switching to drmm_ helpers > 844e701e1d6d..f3050d125f64 > - Using a proper platform device driver > 4658f99dfe3e..a3258e4d7550 > - Introducing a vkms_config structure > 95ad6da46638..5b2d080b4626 > - Adding ConfigFS support > 866ad75102ae..f900ad18ab8c > > What's currently missing: > > - A deep cleanup to ensure checkpatch compliance and proper > functionality for every commit > - Updating documentation > - Reviews > > The primary area where I need assistance is reviews and testers. I'm aware > that Maìra is very busy and can't review quickly, but any other > individuals who can test and/or review this series would be greatly > appreciated. > > My next step is to add connector support, but as I will use this work and > my previous series [1], I would like to see it merged first. > > If any of the original authors would like to be credited for > their contributions, please let me know. I rewrote most of the code, but > the general idea was originally from them. > > José, I am sorry, I think I missed your mail where you told me you already > reviewed some commits: > > > I reviewed the first 9 patches and added a few comments on your > > GitHub fork. > > I am not able to find any comments, can you send me the link to the page > with them? I would like to read/apply them before submitting the first > part of the series. My comments were in the GitHub commits, but I think they were lost after rebasing/force pushing. Hopefully, I saved a copy of your patches with my suggestions applied and commit messages where missing: https://github.com/JoseExposito/linux/commits/patch-vkms-header-refactor/ I think you should be able to extract the diff easily. It might be a better idea to share the patches in the mailing list for broader discussion and to avoid loosing review comments. Thanks a lot for your work Louis! José Expósito > Thanks for your time, > Louis Chauvet > > [1]: https://lore.kernel.org/all/20240809-yuv-v10-0-1a7c764166f7@xxxxxxxxxxx/ > > -- > Louis Chauvet, Bootlin > Embedded Linux and Kernel engineering > https://bootlin.com
