The "Bloody Mess" That Is Intel's Poulsbo Driver

AdamWill writes "Phoronix writes about the mess that is the Linux support situation for Intel's new graphics chipset, the GMA 500 — aka Poulsbo. Near the end they refer to my own post on the topic ('Okay, so after a whole day spent bashing around at this crap, I can very confidently and conclusively say, it's utterly broken'). Intel has a reputation as one of the most clued-up open source-friendly hardware companies, but if they can't sort out the mess surrounding the driver for this chipset — which is already used on the Dell Mini 12 and Sony Vaio P, and will be used on many future Intel-based systems — that reputation will take a serious hit."

Comments

Bloody Mess

[Anonymous Coward]

It must be that time of the month for intel....

Re:Bloody Mess

[timeOday]

Crap like this is what will drive companies _away_ from open source. Look at all the flak that Sun gets regarding it's handling of Open Office. Or countless other examples. The community should be grateful that these companies support FOSS at all

No, this is not OpenOffice, because drivers are only useful to customers who pay for hardware. When I shop for a laptop, I buy something with good driver support by my chosen OS, which is Linux. So if Intel wants me to consider buying something with their chips, they'd better fix the driver problem. The idea of being "grateful" to somebody making something I might want to buy is neither here nor there.

Reputation?

[clang_jangle]

I thought the intel video chipset reputation was already something like "it sucks, ATI or nvidia are much better choices".

Re:Reputation?

[lbbros]

Actually they are fine (I can even run a composited desktop on my EeePC, and that's a GMA 900), but in this case the technology isn't theirs, it was acquired from some third-party.

Re:Reputation?

[AdamWill]

That's the reputation of the power of the hardware. Yes, as far as playing your 3D games goes, you're not going to get very far with Intel.

However, up till now Intel has had a very good reputation for open source friendliness with regards to supporting the hardware, disregarding the actual power of the hardware. Intel are actively involved in maintaining the (100% open source) driver for all other Intel graphics chipsets, and they also contribute to general X.org development and the development of new technologies within X. Intel graphics hardware is generally the least powerful of the big three, but until this mess, it's been by far the best (and most openly) supported hardware in Linux.

Re:Reputation?

[timeOday]

I wouldn't recommend composting on a laptop. If you must compost, I'd place the heap in the furthest corner of the back yard.

Wonder if this is one of the reasons?

[HangingChad]

Microsoft threatening Intel unless they knock off the Linux integration. Now, all of a sudden, Intel is having all kinds of problems with their Linux drivers.

Coincidence or anti-competitive behavior in action?

Re:Wonder if this is one of the reasons?

[Pecisk]

I think it is more about the lines of "omg, there is new sublaptop market here, quick, we need solution. Damn, our video chip uses too much power. Ok, there is some niche chip which could suit us. But there is lot of NDA and proprietary stuff. Heck, let's ride with it and see if it sticks. If not, we will abandon a driver."

It is clearly a totally different video card with different chip (which have closed parts not developed by Intel). So it ends there where usually such drivers goes - to trash can.

Re:Wonder if this is one of the reasons?

[_Sprocket_]

Having spent a lot of time in various beurocracies, I can attest that malice and stupidity work very well together. While one does not guarantee the other, they are often interlinked.

You are correct when it comes to 3D performance

[Anonymous Coward]

Intel graphics chips are not for games. However, if you don't play games and you want a solid graphics card with enough 3D performance to run compiz or Quake with fully open source drivers, then Intel is what you want.

Or, it used to be. I don't know what the deal is with this new chipset.

Re:You are correct when it comes to 3D performance

[jabuzz]

The 3D is licensed from PowerVR, aka Imagination Technologies which used to be called Videologic for those with long memories.

It has nothing to do with Intel (other than that they licensed it), and historically Videologic, when they where in the PC graphics card business where tight lipped about their stuff, rather like nVidia are.

Which all sorts of sucks because the chipset does pretty good 3D for virtually no power. Which should finally mean some netbooks with decent battery lives.

Re:You are correct when it comes to 3D performance

[Anonymous Coward]

Quake is not a game. It is a murder simulator.

Re:You are correct when it comes to 3D performance

[_Sprocket_]

No. Quake is not a game. It is a network performance analyzer.

!gonvidia

[paroneayea]

I'm noticing the tag 'gonvidia', and it's true... as in terms of hardware, Nvidia does seem to be the best. But as in terms of the linux community, they pretty much create problems for everyone. And yes, I know, to the end user that's not always apparent. But the linux desktop really would be a lot farther along if it weren't for nvidia's refusal to open up to the free software community.

If Intel's new open source graphic drivers suck, then obviously yes, that's shitty. But between them and nvidia, if you're going to praise one or the other in the Linux community, it shouldn't be nvidia. Intel's graphic cards still don't support GLSL and the like, but at least you can run an open source driver and it works.

Re:!gonvidia

[dattaway]

More importantly, distributions with closed source drivers are very fragile and easily break. Having an open source driver, its easy to find what went wrong with the changes and fix that. The closed source drivers don't like change. That's my 10+ years with Linux.

Re:!gonvidia

[Racemaniac]

i don't know how it is in this case, but most of the times, the problem is that there is no information about the hardware. so even though the open source community would love to code uber awesome megadrivers, they haven't got any documents on how these chipsets work, so they can't write drivers for them

Re:!gonvidia

[Kjella]

But the linux desktop really would be a lot farther along if it weren't for nvidia's refusal to open up to the free software community.

nVidias stance is pretty simple: No open source support, period. No specifications, no features of really any kind in the open nv drivers, no help to those who ask, no nothing.

What they have delivers is addition to hardware is a great closed source driver which have simply been the best in terms of perforamnce, features and quality for anything better than integrated graphics. Catalyst (AMD/ATIs driver) has been a mess and despite improving greatly since AMD took over, they're not there yet. While AMD has opened their specifications, the open source Radeon drivers are far, far off from the closed source drivers still. AMD has still said their primary commitment is Catalyst, so who knows when if it'll ever get as good as that, which I said isn't as good as nVidia's.

nVidia has constantly been the ones pushing the boundries for what the Linux desktop can do. Just recently before Christmas they delivered the first working hardware accelerated h.264/vc-1 HD playback /VDPAU) and it's available on pretty much all mainstream nVidia cards. ATI is thinking of maybe adding UVD support to their closed source driver and any open source support is unlikely and certainly not coming soon. Poulsbo is the first I've heard from Intel that actually supports VA API and it sure isn't mainstream motherboards.

You talk as if nVidia has been keeping open source back and maybe the open source infrastructure would have been better if nVidia worked with them instead of doing their own thing. But the Linux desktop? I doubt it. It's been over a year since AMDs first release of specifications, go check out the current state of the open source drivers. When you come back, you might realize that for a long time, the best way to show a Linux desktop has been a nVidia machine with proprietary drivers, not ideologically pure but it works well. But sure, blame the guy up front plowing the road for not towing the open source community too. If the open source community could pull it off, they have the chance now as AMDs specs are in the open, that excuse is gone. Put up or STFU.

Re:!gonvidia

[darkmeridian]

Linux open source purists want special treatment from manufacturers, and it makes no sense. ATI opens specs but their Linux drivers suck. nVidia has great Linux desktop support that advances the state of the art for Linux, as you admit, but they didn't release an open-source driver so you knock them for that.

But nVidia does not release the source code for Windows, either. They are treating all the operating systems exactly the same. Why would a non-zealot go with ATI when nVidia's closed source driver is far superior?

Re:Bit of a tangent

[Anonymous Coward]

Possibly because your friend is an idiot who has no idea what he's talking about? Intel release huge amounts of documentation, freely, they were one of the first companies to write and push their own drivers into the Linux kernel tree and almost all of their hardware is well supported with OSS drivers.

Re:If it's open source, fix it.

[HeronBlademaster]

Even if the driver is open source, the chipset documentation might not be. As others have mentioned, it's hard to know how to write a good driver working with nothing more than a bad driver. You need good documentation.

Re:Bit of a tangent

[HeronBlademaster]

Intel's wireless 3945ABG Linux drivers are pretty good. The firmward microcode is released under a closed-source license, but the drivers themselves are open source (and in fact are part of the Linux kernel). That may be offensive to some OSS purists, but I'd rather have good, open-source drivers with closed firmware than non-functional open-source drivers.

It's the same with my opinion about nvidia's drivers. Sure, they're closed-source. But I'd rather nVidia give us working 3d drivers than be stuck with the crappy open-source 2d-only nv driver. It'd be nice if they were open-source, but I'm not going to refuse to use them out of some misguided idealism.

Put another way, if I'm dying of thirst, and a known thief offers me stolen water, I'm going to drink it - it's not like the water is tainted. Maybe that makes me an accessory to a crime (or, in software terms, maybe it encourages closed drivers) but it's better than dying of thirst (or, better than having no 3d drivers at all). nVidia has no real motivation to give us open-source 3d drivers in the first place, so refusing to use their closed driver won't make them change their minds.

Re:Bit of a tangent

[Creepy Crawler]

I'd agree. I had a rather nasty return on a DV6990 HP laptop. It was trash, but that's aside the point.

I went and bought a T61, all intel down to the graphics card. Better wattage drain and complete open source drivers. Ubuntu detects everything on here, with exception to the HD APS system, which I can do without (it drains batt 2w extra).

And then, I find out that Intel releases everything about their 3d system.. And because of that, Linux devs are working on a Graphical Memory Manager, called GEM. Come to find out, it only works for Intel because they're soo open. They know they sell hardware, not their drivers.

Hopefully, AMD/ATI will follow and do the same. Wonder where that leaves nVidia...

Re:If it's open source, fix it.

[AdamWill]

Intel *didn't* make an open source driver. If you read my post, you'll note that there's three different closed-source components to the driver, without which significant features break.

Aside from that, what's needed for meaningful open source development is not "here's some code, have fun". There needs to be a proper development process hosted in an accessible fashion, and proper documentation. The 'intel' driver for previous Intel chipsets satisfies all of these goals. It's 100% open-source, it's developed within X.org and so easily accessible to external contributions in a widely-understood fashion, and the hardware is properly documented.

This 'psb' driver satisfies none of the goals. It was previously hosted within Moblin (which doesn't really have much of an external development community), and even that version of the code is now not being used. It now only shows up in obscure Ubuntu Netbook Remix repositories, with no independent source that anyone can find. So there's no sane development process to which external people can sensibly contribute. It contains large closed-source chunks. And there's no public hardware documentation, which makes it very hard for anyone else to work on it in the first place.

This is what I (and anyone else stuck with one of these chips) am complaining about.

Re:Bit of a tangent

[MostAwesomeDude]

You got modded up, so I get to correct you.

GEM (Graphics Execution Manager) is only working for Intel because they have more people working on it. There's only around four or five people working on Radeon stuff, and of those, only two of us are dedicated to ATI work, and we're both students.

If you grab development snapshots, you can see Radeons working with DRI2, GEM, KMS, and all that fancy stuff.