Have the drm.git kernel modules been abandoned?

[nanonyme]

re: binary drivers - Linux is not just an x86 OS, you know. There are other processor architectures out there, important ones, and they also run Linux and use a lot of the same types of devices. Advocating binary-only drivers is extremely short-sighted.

The original point of talking of a static ABI in this thread was not related to binary drivers. It was just that it also harms sensible development of out-of-tree opensource kernel driver development. (kernel developers don't think you're supposed to be doing that anyway but rather have your drivers integrated to the kernel tree and then everyone has to download whole kernel tree to get the drivers) Dunno who sidetracked the conversation yet again to proprietary code, that was quite offtopic.

[bridgman]

The 6xx/7xx 3D has been ported across and is running on *BSD already; not sure of the KMS status but I think rnoland is working on it.

[yoshi314]

btw. what are *BSD developers going to do now? i mean, linux is slowly getting KMS and drm is adapting to it accordingly. how is bsd going to maintain their own drivers now?

[highlandsun]

re: binary drivers - Linux is not just an x86 OS, you know. There are other processor architectures out there, important ones, and they also run Linux and use a lot of the same types of devices. Advocating binary-only drivers is extremely short-sighted.

It's not even x86 vs ARM / Sparc / PowerPC whatever that causes this problem. Even now I don't have a stable driver for the modem in my laptop, because the vendor's driver depends on an x86 blob therefore it won't run in my x86-64 kernel. This is a ridiculous state of affairs and I've been disassembling / decompiling that particular blob just so I can recompile it for x86-64. If you insist on locking me out of hardware that I own after paying in good faith for everything contained within, then I am going to continue to break the locks that you've placed on my property.

[energyman]

It's nothing tragic, just a deep personal hatred about the way drivers work in Linux :P They come as a "big, fat bunch", all inside the kernel. Reminds me of those "super duper 846-codecs pack" on Windows; I hated those too. Drivers should be separate entities. But I guess since Linux totally lacks a driver interface, we have to live with it.

sucks to be you :P

And I hate to hunt around for drivers and love the fact that everything is in one nice tarball.

[nanonyme]

I've been meaning to take a look when I have enough spare time, haven't yet had.

[Ant P.]

Yeah, I guess I should apologise, that wasn't very helpful.

Here's a better link for those complaining about Linux's development model. Let us know how it goes.

[nanonyme]

I'm off this discussion. I feel it was somewhat enlighting to me and probably someone else until fanboys came in.

If you mean my comment, it was imo perfectly justified. The post behind the link was a worst-case scenario. Quoting that page is like quoting the articles that say LHC is gonna create a black hole every time when talking of particle accelerators. Some people like playing with feelings more than reason. There are good reasons for not having a stable ABI but I consider the lwn.net link more of a flame-bait than intellectual conversation.

[mdias]

What the hell? Are you serious? What kind of article is that?

I'm off this discussion. I feel it was somewhat enlighting to me and probably someone else until fanboys came in.

Bye.

[RealNC]

Fglrx and nvidia used to crash back when I used them. The open-source ATI driver never does.

That is a very naive assumption ("it works for me = it works everywhere"). The open source driver does crash here far more often (by orders of magnitude) than fglrx.