Bettering Radeon Gallium3D Performance With PCI-E 2.0

[liam]

But I imagine looking at e.g. the power saving code could still tell a very good programmer why radeon on the "low" profile still uses more power than fglrx.

I mean, why doesn't AMD simply have somebody going over the code of catalyst, deleting everything patented or "secret" and release the nonfunctional rest? Could still be helpful for low level stuff like power saving and communicating properly with the hardware...

Do you have any idea who long that would take? I think Bridgeman has said catalyst includes something like 10s of millions of lines of code (IIRC). Going through that, line by line, and having legal agree with you, would take many years.

[AnonymousCoward]

galium3d is supposed to do the same -share code among OSes- but i haven't seen anyone jumping on it and releasing an FOSS driver for all OSes

What systems would that be? For Windows and Macos it'll presumably take a lot of work to get them working at all, and then the proprietary drivers will still be better, and all those small niche systems like Haiku and so on will probably be hard-pressed to come up with the required manpower.

[Eduard Munteanu]

Peoples will care about free software only if someone will find a way to stop piracy, when they will have to pay 600$ for a microsoft office suite plus another 600$ for a retail copy of windows just because someone sent them a fucked .doc (or whatever they call it in office 20${XY}) something will change.

But companies generally care about that sort of legal stuff. I do wonder why exactly FOSS adoption isn't all that commonplace in companies, even partial adoption (e.g. some free office suite on Windows), especially when your employees don't have specific requirements. It could be MS gives you a better TCO/risk, it could be free stuff indeed has some rough edges or it could be management isn't comfortable with saying "no, ask for documents in a non-proprietary format from your contacts/customers". In any case, I guess it's easy to conclude proprietary software is the way to go and not give it a second thought, since it obviously worked out for many others.

[89c51]

Exactly. That's the primary reason proprietary drivers exist at all for open OSes -- they let you share code with proprietary OSes and as a consequence deliver a lot more functionality and performance than the same number of developers could provide in a separate code base.

galium3d is supposed to do the same -share code among OSes- but i haven't seen anyone jumping on it and releasing an FOSS driver for all OSes

too big of a change i assume but anyway

[whizse]

So what are the plans for using pcie_gen2=1 by default? Couldn't the incompatible motherboards be blacklisted instead of turning it off for everyone?

[bridgman]

2) Catalyst is made of share code for the 90% of it and the linux team is very little (20? 30? maybe even 50 peoples?) because of that, those developers alone wouldn't be able to develop a competitive stack because the shared part is made by thousands of developers.

Exactly. That's the primary reason proprietary drivers exist at all for open OSes -- they let you share code with proprietary OSes and as a consequence deliver a lot more functionality and performance than the same number of developers could provide in a separate code base.

[darkbasic]

And they probably would if it made sense.

They don't and they will never because:

1) Peoples do not care.
2) Catalyst is made of share code for the 90% of it and the linux team is very little (20? 30? maybe even 50 peoples?) because of that, those developers alone wouldn't be able to develop a competitive stack because the shared part is made by thousands of developers.

Peoples will care about free software only if someone will find a way to stop piracy, when they will have to pay 600$ for a microsoft office suite plus another 600$ for a retail copy of windows just because someone sent them a fucked .doc (or whatever they call it in office 20${XY}) something will change.

[agd5f]

This has been addressed before, but the closed source driver has a completely different coding style and kernel/userspace interfaces. Most of the code would have to be completely rewritten before it had even a remote chance of making it upstream. There are plenty of OS-agnostic drivers out there (realtek, etc.) who release common shaded code for all OSes and it never gets merged upstream because the coding styles don't match or the driver uses it's own infrastructure for certain things rather than some common infrastructure provided by the kernel. It's not really feasible to redesign the entire closed source driver (much of which is planned years in advance as new chips and initiatives are planned) just so it could be shared more easily with open source. The market for open source drivers just isn't big enough.

And before you reply, yes, I know, chicken and egg.

[Eduard Munteanu]

And they probably would if it made sense. But since Linux' graphics stack is a moving target, what would that accomplish? Plus, they'd still have to maintain Catalyst for Windows anyway.

I'm not sure if it's legally-possible, but a shared FOSS driver stack on Windows might help AMD too. The Linux graphics stack is a moving target but, AFAIK, Gallium should do fairly well in terms of OS abstraction. They'd just need to make releases, i.e. pick, compile and bundle the various bits and pieces like distros do, while offloading some of the coding burden to the community (including the Linux people for the common parts).

The problem is the FOSS stack is behind Catalyst in terms of features and performance and I'm not sure anybody has done that work for Windows, which is quite a disincentive at the moment.

[sk8erboi101]

How do I put this command when booting the kernel?