RadeonSI Gallium3D Gets New State Handling

[bridgman]

Would it be possible to release the PM bits under NDA so a dev that will have no problem working under it can work on this??

Sure, as long as they understand they may never be able to release the code. Doesn't seem worth it though, since we already have working code and the challenge is getting approval to release it.

[89c51]

For everything else, please try to stick to facts. AMD has not yet opened up all of the PM bits of the GPU (but as I said before we are working on that again since it doesn't look like community devs are going to go any further with what has already been released), so it loses in the performance-per-watt when the system is idle (an oxymoron if ever there was one) but I understand the point you were trying to make.

Would it be possible to release the PM bits under NDA so a dev that will have no problem working under it can work on this??

[bridgman]

Not a schedule per se -- we basically work on documentation in the "free time" (ha ha) left over after implementing and releasing enough initial support that we know what needs to go into the documentation, and there won't be a lot of that until we are fully caught up with hardware development and can start leveraging more of the work being done by other teams.

First doc out will probably be the SI ISA guide (which is worked on by a number of teams inside AMD) then we'll backfill around that with SI graphics programming information and try to start filling some gaps in the earlier GPU generations.

[Khudsa]

Bridgman, there is some schedule for the next release of AMD specs and what specs somewhere?

Thanks!

[bridgman]

I'm sorry to go off topic, but it does relates to the point I am trying to make that Linux cannot be ignored any longer, and not being able to properly support linux in a reasonable timeframe, oss or not, may very well be the end of AMD in the long run.

Sure, but neither of those have anything to do with this thread (except for the discussion about trending towards launch time open source graphics support). That's why I'm suggesting a new one.

I'm glad AMD is not giving up the high-end (I only own AMD hardware FYI), but it is really sad to see an semi- OSS suportive company missing on so many opportunities, loosing ground to the likes of intel, nvidia, imagination, ti and qualcomm, simply because they can't see such an obvious trend. All this focus on higher performance via GPU compute and higher CPU core counts seems (for us outsiders) to come down to only one man on linux: Tom Stellar, who seems to be the whole linux GPU compute team. Is this really how AMD plans to get back in the race? Even if the guy is a superhero and manage to make clover the best OpenCl stack in the universe, AMD can't even open up the PM bits of the GPU, so it still loses in the performance-per-watt argument.

Um... no. Tom is a fairly big part of the open source OpenCL effort for GPU compute on Linux inside AMD (obviously it *is* open source and there are other devs outside AMD working on it as well). There is a larger team working on GPU compute via Catalyst, plus the whole HSA effort.

For everything else, please try to stick to facts. AMD has not yet opened up all of the PM bits of the GPU (but as I said before we are working on that again since it doesn't look like community devs are going to go any further with what has already been released), so it loses in the performance-per-watt when the system is idle (an oxymoron if ever there was one) but I understand the point you were trying to make.

[Figueiredo]

This should probably be in another thread, but I thought the message was that we planned to focus on higher performance via GPU compute and higher CPU core counts rather than continuing to push for ever-higher single-core performance, not that we weren't competing in high performance CPUs. Short statements like that can be interpreted in a lot of ways so they generally just end up reinforcing whatever you believed before... but nobody seems to want to stay awake through long statements any more

I'm sorry to go off topic, but it does relates to the point I am trying to make that Linux cannot be ignored any longer, and not being able to properly support linux in a reasonable timeframe, oss or not, may very well be the end of AMD in the long run.

I'm glad AMD is not giving up the high-end (I only own AMD hardware FYI), but it is really sad to see an semi- OSS suportive company missing on so many opportunities, loosing ground to the likes of intel, nvidia, imagination, ti and qualcomm, simply because they can't see such an obvious trend. All this focus on higher performance via GPU compute and higher CPU core counts seems (for us outsiders) to come down to only one man on linux: Tom Stellar, who seems to be the whole linux GPU compute team. Is this really how AMD plans to get back in the race? Even if the guy is a superhero and manage to make clover the best OpenCl stack in the universe, AMD can't even open up the PM bits of the GPU, so it still loses in the performance-per-watt argument.

[curaga]

This is wrong if you are 100% the copyright holder you can release the code as GPL and MIT and closed source without any problem.

True, but AMD is not the only copyright holder of drm or mesa. Any common code would be likely to be copied too.

[bridgman]

This should probably be in another thread, but I thought the message was that we planned to focus on higher performance via GPU compute and higher CPU core counts rather than continuing to push for ever-higher single-core performance, not that we weren't competing in high performance CPUs. Short statements like that can be interpreted in a lot of ways so they generally just end up reinforcing whatever you believed before... but nobody seems to want to stay awake through long statements any more

[Figueiredo]

I just find it kinda funny that AMD is deciding to stay out of so many markets. According to Rory Read they will no longer be competing for high performance CPUs, nor will they be entering the smartphone segment. If they can't provide a working linux GPU driver, they can't support android, so forget about android tablets. Ok, Windows 8 is coming, but will it be such a huge success? Are they really betting everything on it? As I said in another topic recently, Intel is coming with Valley View, they will possibly be able to support x86 versions of every linux based mobile OS available: android, chrome OS, web OS, firefox OS, ubuntu, moblin, meego, tizen, (Steam OS?), you name it, besides plain old windows, of course. Even though I agree that there may not be space for all of them it doesn't really matter for Intel. AMD on the other side, has all it's horses on windows righ now, they can't even advertise linux support on their products, so once again they are locking themselves out of potential markets. I really fear where this is going. Windows has been going strong for a long time, will this trend go on with so many new competitors?

Look at the OUYA, it could have been AMD, but even *ucking nVidia has better support for android. Only a really stupid person would deny that "winter is coming" for windows... 2014-2015 timeframe might be way too late to get on board.

AMD is giving up market share on servers, on high-end desktops and more or less holding up on notebooks. Their last stronghold seems to be consoles: Playstation and Xbox... Guess what, Steam is coming to linux, rumors of a linux based SteamBox are spreading. IF this is true, and SteamBoxes start chewing Playstation's and Xbox's marketshare, guess who will be going the way of VIA?

[russofris]

So we can expect out of the box support for "9000" series in 2014-ish? Seems like even catalyst team is behind lately, has VCE been enabled in windows?

While I don't expect OOTB support for 9000, it is good to know that the OSS team will be starting to ramp up at the same time fglrx is starting to ramp up. Unfortunately, and this is speculation, I fear that the hardware teams are so used to working with the catalyst team's processes that there will be some growing pains for the OSS team. True parity will likely never occur. That's OK though, as long as the OSS team can produce an acceptable driver in a reasonable amount of time.

While a number of us complain that OSS isn't delivered as fast as FGLRX, what we're really trying to say is that we want an acceptable driver in a reasonable amount of time. The fact that FGLRX exists annoys us, because it reminds us constantly of what-could-have-been had AMD shared our utopian software development philosophy. Really, you just want your purchased product to work, and work now, and align with your philosophy.

I'll be looking forward to AMD's 9000 offering. I hope that they can deliver a working OSS driver in an acceptable timeframe.

F