NVIDIA's Proposal For A New API Better Than GBM Has Already Made Some Progress

[phoronix]

Phoronix: NVIDIA's Proposal For A New API Better Than GBM Has Already Made Some Progress

On the first day of the XDC2016 conference this week in Finland NVIDIA presented over their GBM vs. EGLStreams debate that's been ongoing for months with NVIDIA's lack of GBM API support by their driver being what's preventing the binary blob from working with current-generation Wayland compositors. In that session they called for a new community-driven API to suit the needs of device memory / surface allocation and could succeed the Generic Buffer Manager. By the end of XDC2016, some progress has already been made...

http://www.phoronix.com/scan.php?page=news_item&px=NVIDIA-Memory-API-Progress

[DebianLinuxero]

If they don't support GBM by themselfs, at least they could help the others who support it, nouveau.

[M@GOid]

Unbelievable. They prefer start a new API instead of supporting GMB. The whole point of people not wanting to use EGLStreams was to not throw away all the work already done. Now Nvidia want people to do even more work by implementing a new API, instead of fixing/implementing what GBM lacks.

[Azpegath]

Wow, awesome! Cooperation is beautiful, especially when talented developers get together and rise above the NIH, market and legal bullshit!

[SaucyJack]

We'll see where this goes. If it at least performs as well as gbm i don't see an issue.

[bug77]

Well, that also kind of proves there are indeed things to improve wrt GBM and Nvidia wasn't just making up excuses. At least not 100%.

[Holograph]

Wait, is EGLStreams actually an improvement over GBM? I thought it was just something they wanted to do because it would be easy for them? If it's actually better then GBM should please GTFO. Or is that just Nvidia rhetoric?

[cl333r]

Judging by the Wayland progress (by 2018 Nvidia might support it) - 2018-2008=10 years in the making, GLVND - about 3-5 years, it will take this new API about 3-4 years to be defined, agreed upon, implemented for 95+% of used hw and spread enough in distros to be useful.

[bibaheu]

I hope it doesn't become a convoluted mess due to supporting common kernel APIs shared with other Unix

[Holograph]

Judging by the Wayland progress (by 2018 Nvidia might support it) - 2018-2008=10 years in the making, GLVND - about 3-5 years, it will take this new API about 3-4 years to be defined, agreed upon, implemented for 95+% of used hw and spread enough in distros to be useful.

I think it's silly to wait to implement this, there has been plenty of opportunity to do so before anything was actually using Wayland in production environments of any sort, now we have distros finally wanting to default to Wayland and only just now are we learning that Nvidia may have been correct all along? I think that's crazy. Should have figured that out before people were trying to switch distros to use it. They've been pushing this method for a while now. They could have sooner elaborated more on why it would be better but I really think this should have been figured out already.