On 2019-03-30, Jonas Smedegaard <d...@jones.dk> wrote: > Quoting Vagrant Cascadian (2019-03-30 19:32:10) >> On 2019-03-30, Jonas Smedegaard wrote: >> > As subject says, please have u-boot support Olimux Teres-I DIY laptop. ... >> I've glanced at the changes, and overall they look ok. A little more >> invasive than one might want this late in the freeeze, though... > > As you maybe noticed from the history of the git, I first tried older > more minimal patches, before I stumbled upon this current one which > actually works for me: Today was first time ever that I booted my > Teres-I off of an up-to-date (and then patched) u-boot.
Congrats! I know from experience how much of an adventure that can be... >> Have the patches been submitted upstream? Is the added .dts based on >> the one included in the linux kernel? > > Short version: No. That makes it harder to consider, to be honest. I've made the mistake of maintaining invasive third-party patches before... would prefer to not make that mistake again. Especially where we're at in the freeze. I would consider including them in an upload to experimental, with the understanding that they might be removed again later. I was planning on uploading a v2019.04-rc* to experimental soon, and maybe it would make sense to target that. > I am not in contact with the author of the patch, just lifted it our of > their rpm source package and applied ot to our u-boot package. > > This is as far as I know the newest conversation related to this (and > the post that clued me in on searching for work done by Torsten Duwe): > https://lkml.org/lkml/2019/2/4/1012 > > My understanding is that it is is being pushed to Linux kernel actively > by Icenowy Zheng, and for u-boot seemingly only exists as the work by > Torsten Duwe which seemingly is not pushing it upstream actively. > > My (wild) guess is that Icenowy will push it to u-boot when in Linux. Thanks for a little deeper background. It still sounds unclear when the changes might plausibly land upstream, though. Maybe nudging some people to get the patches upstream would be feasible; a u-boot merge window opens in a couple weeks. >> It would be helpful to either include the patches in the bug report, >> or submit a merge request on salsa.debian.org. Mixing the two methods >> feels a bit suboptimal to me. > > I was awaiting your response to know what you preferred. > > I am not familiar with Gitlab merge requests, but am willing to try if > guided. But easiest for me is to simply push directly to the u-boot git > if that's ok with you (I got write access already since you put it in > the debian area on salsa). If you want me to add it in a separate > branch, then what branch name do you prefer? At the moment, I'd prefer to hold off on these patches until upstream status is a little more solid, or keep the patches only in experimental till then. Do feel free to push to a "wip/teres-i" branch in the u-boot salsa; I think that would make it easier for me to review. live well, vagrant
signature.asc
Description: PGP signature
