Red Hat Joins Khronos, The Group Behind OpenGL & Vulkan

[bridgman]

I would rather check his commits count. In fact I wouldn't even need that. Everyone who bothered himself to learn about Linux graphics stack a bit should be very well aware who is airlied (and you too, ofc) and would have no questions why RH joins Khronos. I would rather ask another question: why you haven't do it earlier, RH?

Yep, the post count comment was in direct response to someone who was surprised that RH employees were "finally allowed to post here".

When I volunteered to set up the open source graphics effort I ran across a forum post saying "if AMD was serious about open source drivers they would call Dave Airlie and ask him what they should do"... so I did.

[smitty3268]

When I volunteered to set up the open source graphics effort I ran across a forum post saying "if AMD was serious about open source drivers they would call Dave Airlie and ask him what they should do"... so I did.

If AMD was serious about open source drivers they'd send me free samples of all their new hardware.

Your move, bridgman.

[Hamish Wilson]

Yep, the post count comment was in direct response to someone who was surprised that RH employees were "finally allowed to post here".

I see my sarcasm did not carry over, apparently. I was actually referring to the people slagging Red Hat off.

I would have thought banging on about assassins and Lovecraft would have been enough to make people realize I was being facetious though.

[bridgman]

I see my sarcasm did not carry over, apparently. I was actually referring to the people slagging Red Hat off.

Sorry, my sarcasm detector has never been that good, and it's pretty much disappeared until this cold goes away.

I would have thought banging on about assassins and Lovecraft would have been enough to make people realize I was being facetious though.

On Phoronix. You're kidding, right ?

[libv]

Yep, the post count comment was in direct response to someone who was surprised that RH employees were "finally allowed to post here".

When I volunteered to set up the open source graphics effort I ran across a forum post saying "if AMD was serious about open source drivers they would call Dave Airlie and ask him what they should do"... so I did.

That seems to clash with my timeline of how things unrolled. You seem to have come in late, after Chris Schlaeger had talked to Egbert Eich (april 16th 2007), and after we at SuSE had written our proposal for an open source strategy and distributed it. This proposal was sent to the next level of management inside SuSE on May 31st and we got very favourable comments back 4d later. But AMD had shown interest in a proposal from us since late april, early may, and probably started soliciting for a counterproposal from ATI after that. Dave has publicly claimed that he talked to you somewhere in june... For any definition of "somewhere", could've even been late june, at which point the SuSE guys had already got a 5k budget from AMD for picking out commercially available hw (monday 18th of june 2007).

So please, put your money where your mouth is, and provide a url for that.

[GreatEmerald]

Wow, this is quite the thread.

[bridgman]

That seems to clash with my timeline of how things unrolled. You seem to have come in late, after Chris Schlaeger had talked to Egbert Eich (april 16th 2007), and after we at SuSE had written our proposal for an open source strategy and distributed it. This proposal was sent to the next level of management inside SuSE on May 31st and we got very favourable comments back 4d later. But AMD had shown interest in a proposal from us since late april, early may, and probably started soliciting for a counterproposal from ATI after that. Dave has publicly claimed that he talked to you somewhere in june... For any definition of "somewhere", could've even been late june, at which point the SuSE guys had already got a 5k budget from AMD for picking out commercially available hw (monday 18th of june 2007).

Both are correct AFAIK. There were two independent sets of discussions going on, and they didn't get totally aligned until early 2008. The discussions I was involved in started in May 2007 IIRC.

A couple of people were involved with both discussions and cross-pollinated some of the content, eg suggesting to me that we work with SUSE on the initial driver development leveraging the existing AMD/SUSE relationship & budget. I took a plan to Dirk in Aug 2007 and secured approval/budget -- AFAICS the approved plan was fairly close to the SUSE proposal with the exception of KMS and AtomBIOS.

[libv]

Both are correct AFAIK. There were two independent sets of discussions going on, and they didn't get totally aligned until early 2008. The discussions I was involved in started in May 2007 IIRC.

A couple of people were involved with both discussions and cross-pollinated some of the content, eg suggesting to me that we work with SUSE on the initial driver development leveraging the existing AMD/SUSE relationship & budget. I took a plan to Dirk in Aug 2007 and secured approval/budget -- AFAICS the approved plan was fairly close to the SUSE proposal with the exception of KMS and AtomBIOS.

KMS was a crappy brainfart from someone who never had been near a display driver before, and who got handheld through turning the (poor and restrictive) RandR1.2 interface into a header for inclusion in kernel code. This was march 2007. The first stab at getting a driver up on it was in May 2007 with jesse posting patches for intel on the 17th of may 2007. Given that we sent in our proposal, after a month of back and forth between me and egbert, on May 31st, any plan for a stable, well-supported, enterprise ready driver would definitely not have included KMS in may 2007 (Expecially given the fact that actual display driver developers with rather extensive and proven experience made that plan).

Atombios was doing things the ATI way, the fglrx way, and from what I could tell then and now, the main reason AMD wanted an open source driver was to get the nasty fglrx stain on the ATI image from ruining the amd chip/server market. It's still amazing how you managed to remarket things so that people booed down proper C code but cheered for that same fglrx/ati way that they were trashing before.

August 2007? Dirk? I mentioned Chris Schlaeger before, and the 16th of April 2007. I also do not see that link to that forum post you mentioned.

[bridgman]

Given that we sent in our proposal, after a month of back and forth between me and egbert, on May 31st, any plan for a stable, well-supported, enterprise ready driver would definitely not have included KMS in may 2007 (Expecially given the fact that actual display driver developers with rather extensive and proven experience made that plan).

Correct, your proposal did not include AtomBIOS or KMS, but unfortunately I didn't know about that proposal until early 2008. I had seen one of the statement-of-work documents (basically list of deliverables) but that's it. If I had known in 2007 that there were two different proposals that would certainly have helped. I explained all this the last time we met, remember ?

Atombios was doing things the ATI way, the fglrx way, and from what I could tell then and now, the main reason AMD wanted an open source driver was to get the nasty fglrx stain on the ATI image from ruining the amd chip/server market. It's still amazing how you managed to remarket things so that people booed down proper C code but cheered for that same fglrx/ati way that they were trashing before.

Um... thank you ??

Seriously, I don't think the things people disliked about fglrx had much to do with Atombios.

I also do not see that link to that forum post you mentioned.

I read it back in 2007, don't know when it was posted. It doesn't seem to exist any more -- at least casual Google searches didn't turn anything up.

Anyways, this all happened a long time ago.

[libv]

Correct, your proposal did not include AtomBIOS or KMS, but unfortunately I didn't know about that proposal until early 2008. I had seen one of the statement-of-work documents (basically list of deliverables) but that's it. If I had known in 2007 that there were two different proposals that would certainly have helped. I explained all this the last time we met, remember ?

Your supposed lack of knowledge of that proposal (which i find very very hard to believe), is still not enough a reason to have so actively supported a fork of the RadeonHD code, which produced a driver which was less technically advanced and less free than RadeonHD. The competing driver did everything the ATI way, from PLL calculation, to I2C, to HPD, to atombios, even powermanagement took well into this decade.

The fact that the SuSE account manager at AMD was moved aside in.. September 2007? And was replaced with the account manager for Red Hat, that was a marked shift in the relationship between AMD and SuSE. It also seriously hints at Redhat not being happy with not being able to catch the glory on RadeonHD, and that it threatened to damage the AMD/Red hat relationship and make things worse for AMD servers. That explains a lot more of the actions taken by a lot of people than "i claim i didn't know about your proposal".

Seriously, I don't think the things people disliked about fglrx had much to do with Atombios.

I still maintain to this day that atombios is part of a culture of abstracting badly and hiding things in black boxes, and it is that culture that led to the bad name of fglrx/catalyst, therefor atombios is a part of why fglrx was/is that hated.

I read it back in 2007, don't know when it was posted. It doesn't seem to exist any more -- at least casual Google searches didn't turn anything up.

Anyways, this all happened a long time ago.

You are the one who stated that you so selflessly volunteered, which still clashes with how I experienced things, and you seem to not be able to provide any fixed points of reference.

Plus, the methods employed by the forkers are still being employed by them today. They never got called to rights on their nasty and insidious behaviour totally unsuited for what is supposed to be an open source community, and while i doubt that they will hack repositories again, they do employ the (only slightly) less obvious tactics employed during the RadeonHD days.