Wanted a flying pony. Got a pony.<p>An open userspace means that I will be able to use the GPU as far into the future as I like. It means that if anything is screwed up on my ARM processor I can see it and fix it (or thank the person that beat me to it.) That's a good thing.<p>The code blob that is not open is what runs on the GPU. I'd love to have that too so that the OpenGL could possibly be extended beyond what the vendor chooses or it could be ditched as a GPU entirely and used as a big parallel DSP. BroadCom didn't give us that.<p>Some reasons I can think of:<p>• GPU litigation – If you don't have a pre-emptive patent arsenal, open sourcing your GPU code and design is like inviting your competitors' lawyers over for tea and depositions.<p>• Unbuildable – The toolchain to build the GPU code is likely a cobbled together mess of tools, many built to minimally functional in house standards.<p>• Undocumented – They probably have not spent the hundreds of thousands of dollars to properly document how the GPU works. What documentation exists is possibly not in English.<p>• Unsafe – It is possible that the GPU has HCF[1] opcodes or sequences, or at the very least hasn't been lightly proven not to. In particular, there could be thermal issues depending how you drive it.<p>• IP Ownership – Portions of their implementation could be licensed under terms that do not permit release.<p>EOM<p>[1] <a href="http://en.wikipedia.org/wiki/Halt_and_Catch_Fire" rel="nofollow">http://en.wikipedia.org/wiki/Halt_and_Catch_Fire</a>
Describing the driver as crap seems a bit harsh when the problem appears to be that the driver doesn't open-up as much gpu functionality as the author hoped. Perhaps a better title for the article would be "Raspberry Pi GPU Driver Not As Open As I Would Like".
I find this conversation, and the previous one, instructive.<p>The discussion illustrates the challenge that Linux has accommodating third parties. Two entire classes of software are categorically unable to embrace the 'rules' of open software, the graphics folks and the wireless folks (and to some extent the Printer folks). This has meant that these areas generally had a much poorer to non-existent user experience for Linux users than users of systems with support for IP protection built into them.<p>I think the 'shim' model is certainly one way to compromise here. I think the Raspberry Pi folks have done a great service to developers by helping this get done.
Every time I read an article like this it seems that the implied consensus is that companies with millions of dollars in yearly revenue are unable to hire good talent when it comes to hardware programming and all the "good" ones are open-source people who may or may not have worked for one of these companies in the past.<p>With all these "this thing is bad" articles it just seems like no hardware company has ever been able to write good drivers and are completely unable to hire the right people to do the job. This is implied in all of the history of electronics.<p>But they also almost always come across as more opinions than facts. Does the Pi GPU driver suck? I don't know because this article says so but doesn't explain why exactly. As someone else pointed out, this article is pointing out that the company simply did not create certain things the way the community wanted.<p>I'm not a hardware programmer so I don't much about such things, but what do these people actually expect out of a piece of hardware that costs around $25-50 US? What exactly are the goals these people want that require the type of access they are asking for? What goals are they failing to achieve that makes them decide the hardware and/or drivers suck? That's all I ask when one of these "this thing is bad" type of article.
It's cool that phoronix linked and quoted Dave Airlie. It's too bad they didn't quote or respond to the last line of his blog post: "(and really phoronix, you suck even more than usual at journalism)"
The nut graph from arlied's post[1] (linked to in phoronix article):<p>"So really Rasberry Pi and Broadcom - get a big FAIL for even bothering to make a press release for this, if they'd just stuck the code out there and gone on with things it would have been fine, nobody would have been any happier, but some idiot thought this crappy shim layer deserved a press release, pointless. <i>(and really phoronix, you suck even more than usual at journalism)</i>."<p>Emphasis mine.<p>[1] <a href="http://airlied.livejournal.com/76383.html" rel="nofollow">http://airlied.livejournal.com/76383.html</a>