I have a gorgeous Surface Pro 11 X1 Elite that can run just enough Linux to tease me with how beautiful it could be, but it's still unstable enough that I can't daily it.

Torture.

It's happening: https://www.phoronix.com/news/HP-EliteBook-X-G2q-Linux

When will folks learn companies only support Linux or any other FOSS to the extent their own business goals?

None of them are on the game for the well being of the community or whatever.

Profits and lower R&D costs, that is all.

I recently tried to get BSD/Linux to work on my omnibook X 14 and... it's been a journey!

Eventually I got it to work well with [1] and extracted firmware off github because I had wiped Windows and all partitions into oblivion.

I was looking for the bliss of fan-less linux with ARM. The joy! [2]

[1] https://discourse.ubuntu.com/t/ubuntu-concept-snapdragon-x-e...

[2] the fans are ON permanently

Nah they'd rather try to continue to force snapdragon on windows where no one actually cares about this and the experience is trash.

AArch64 is dead for Windows and client Linux, and the knife is in Qualcomm's hands.

Never trust Qualcomm in the linux space. It only leads to frustration. Somehow I've had to learn this lesson more than once because I'm not the smartest ARM addict.

It is sad to see that they still do not support Snapdragon products with Linux offically as a product

I was just about to complain about being locked into one kernel version, being at the mercy of a single vendor, but you said it so much better than I would have :) Thanks.

We'll see. They purposely dropped all server/console support and now only offer a Debian desktop image, which seems crazy for an SBC. Not exactly a welcome mat:

> Other variants that were previously provided AS-IS are no longer provided. Interested users need to build those by themselves.

https://github.com/radxa-build/radxa-dragon-q6a

AI / NPU use cases have been severely hampered as well:

https://gist.github.com/Foadsf/3cc2e0ed357c3ac7180589701bf83...

I've personally been wrestling with their broken I2C for a couple weeks.

Really want to love this board but lots of sharp edges at the moment. Hopefully Qualcomm keeps dedicating resources to improving things - I know it's hard work!

MNT just started offering the QCS6490 for their laptops.

https://mnt.re/media/reform_md/2026-06-30-june-update.html

Sure, it's not great, it's just better (for some tasks) then every tool it's competing with. Buildroot is simpler, but not as flexible, etc...

Eudora was the first email client I used in the 90s, I didn't know that qualcomm was responsible for it's disappearance. Looks like the source was released in its last breath[0]; i'm a bit surprised it hasn't been scooped up and maintained anywhere

[0] https://computerhistory.org/blog/the-eudora-email-client-sou...

It has now been decades since I used Eudora. I still feel I lost the best email client due to Qualcomm’s neglect

Imagine you wrote a WYSIWYG text editor, like Libre Office Writer. You have all sorts of functionality and an overall architecture which makes it sane to upkeep the project & have things work well together. Then someone else makes a custom font, but kind of does it their own way and with a different approach making it a one off from the way the rest of the fonts all work and are used in the program maybe using a custom font file format parser and different UI element even though you know it could have just used the normal, already maintained and planned out code paths.

You can of course merge anything with the right license if you so like, like that one off font code into your editor, but if it doesn't fit well into the overall project or meet the general quality standards of it then it's not practical to and can actually be worse than not including it. Upstreaming is about submitting something the maintainer can reasonably accept and maintain, not just about whether working code is available. GPL licensed code provides the latter, it's still up to someone (either the original company or some other interested person) to make it fit right first.

Upstream requires a level of quality most developers cannot meet.

Ofcourse they can. But which particular person will do it ?

The "upstream" people deal with their own drivers, subsystems or tasks which takes up their time - but if someone feels they want to take on this too, they'll do it (normally that doesn't happen - it's up to the original authors to take responsibility)

Linux tries to avoid special cases. That means that when someone shows up with a new driver that's either not something that fits into an existing category, or which sort of (but doesn't entirely) overlap with an existing driver, there's an extended set of design discussions about how to make this new driver fit into existing infrastructure in a way that's consistent with what's there and which also allows new things to exist.

That sounds great from a design perspective, but it can also lead to cases where people are attempting to design for utter unknowns - potential futures that may or may not exist, theoretical understandings of how hardware works, that kind of thing. It frequently prevents new drivers being merged without significant modification, and sometimes it results in a need to entirely rearchitect the relevant part of the kernel before the driver can even be considered (and also now you need to split that driver into three parts). Upstreaming is hard.