A Game Developer's Perspective On Linux Driver Quality

[omglotsofdots]

Well, some bugs found in article - wtf ??? intel driver #1, driver #2 ??? in what galaxy ??? intel has only one opensource driver, thats it. And more fixes - intel only rules in cpu market, their "gpus" sucks, and they are terrible at software/drivers stuff.

These folks actually have so much money and their org charts are so deep and wide they can afford two entirely different driver teams! (That's right - for this vendor, on one platform you get GL driver #1, and another you get GL driver #2, and they are completely different codebases and teams.)

Hope its clear now.

[mmstick]

This myth has already been debunked. There is no IP that keeps AMD from open sourcing their drivers. It's their "OpenGL secret sacue" that they don't want others to see. AMD said it themselves.

http://www.phoronix.com/scan.php?pag...t_kernel&num=1

There was more to it than secret sauce. It's really just that the driver is so huge, with so many lines of code, full of so many bugs, and not even they know what a lot of their own code does, lacking so much documentation, that open sourcing it would be pointless -- it's better to develop the cleaner existing open source drivers. They can't just ditch Catalyst either since they need it for Windows and proprietary professional software on Windows that need OpenGL. I do think they should just ditch Catalyst for Linux though and focus more on the open source driver.

[Vim_User]

This myth has already been debunked. There is no IP that keeps AMD from open sourcing their drivers. It's their "OpenGL secret sacue" that they don't want others to see. AMD said it themselves.

http://www.phoronix.com/scan.php?pag...t_kernel&num=1

And yet they still can not release UVD on RS780/880/RV790 in the FOSS driver due to legal reasons.

[whitecat]

And yet they still can not release UVD on RS780/880/RV790 in the FOSS driver due to legal reasons.

Do you have more information about that ?

[tarceri]

They can't just ditch Catalyst either since they need it for Windows and proprietary professional software on Windows that need OpenGL. I do think they should just ditch Catalyst for Linux though and focus more on the open source driver.

Gallium is designed so that drivers can be made to run on different OS's. So they could use the opensource drivers on windows too.

[przemoli]

Gallium is designed so that drivers can be made to run on different OS's. So they could use the opensource drivers on windows too.

Designed=/=Implemented

Any existing Non Linux drivers out there using Gallium?
(*BSD was porting some, are those working?)

[tarceri]

Designed=/=Implemented

Any existing Non Linux drivers out there using Gallium?
(*BSD was porting some, are those working?)

What is =/= meant to mean?

I'm pretty sure VMware uses it for other systems (well at least Windows) and also has a Directx state tracker. They just haven't released the code.

Edit: Also from what I recall the BSD porting was about adding compatability (DRI2 etc) to the BSD kernel itself rather than adding a new winsys to Gallium.

[Vim_User]

Do you have more information about that ?

According to bridgman, the code for that is going around between developers and the legal department, but on a low priority. If the reason the Catalyst driver will not be open sourced is secret OpenGL ingredients, why don't we have the non-OpenGL parts already? The truth is, no one works on this and we will never see UVD on those chips.

[Ancurio]

Gallium is designed so that drivers can be made to run on different OS's. So they could use the opensource drivers on windows too.

Wouldn't those drivers have to be signed my Microsoft though? I wonder what their stance on freely licensed kernel drivers are. (no sarcasm)

[Leinad]

Driver #C3?

What about Intel for Mac Os X?