User Profile

[bridgman]

Didn't RX480 launch in mid-2016 ?...

[bridgman]

Right... the black art is defining/communicating the information that a scheduler would need to "make the right decisions". In the absense of...

[bridgman]

Technically you are correct - this is a scheduling problem. The challenge is that AFAIK we do not have way to pass sufficiently rich information about...

[bridgman]

Yep... 120+ commits queued up for 4.21 already:

https://cgit.freedesktop.org/~agd5f/...-next-4.21-wip...

[bridgman]

I would be surprised if you could find even a single example of that. I regard closed source firmware as a necessary evil at best. I have sometimes explained...

[bridgman]

Quick question - are you talking about POWER-specific issues with the video code or just the much older cards from before we were able to separate the...

[bridgman]

Arcturus is not post-Navi architecture. If I get more specific than that it kinda defeats the whole purpose of having code names

The only...

[bridgman]

Not even the first chip to use the architecture. It's just a chip. We did pick from star names but you have to pick from somewhere.

We do...

[bridgman]

Please go back and read my original post. Arcturus is a CHIP not an ARCHITECTURE... just the first chip where the timing worked out for us to give it...

[bridgman]

Calling it "hostile" is a bit unfair... it's just that our contractual arrangements with the semi-custom customers do not allow us to release...

[bridgman]

This is probably pretty close to what Linus first wrote for the original CoC (the Code of Conflict) then softened a bit for publication....

[bridgman]

Sorry, poor choice of words... we don't want a situation where the released kernel is broken, and that's why testing mainline is still important.

[bridgman]

Yes and no. Getting some user testing of the internal development trees is definitely helpful, but the code that goes into distros comes from the mainline...

[bridgman]

What you are talking about is the standard kernel development cycle, where new code is allowed in once per kernel cycle during a 2 week merge window,...

[bridgman]

The implementation builds on the existing dma-buf mechanism inside dma, although it extends it to support new usage scenarios. All P2P mapping is done...

[bridgman]

Our GPUs have been able to do this for a long time - this adds some standardized support for it to the drm driver framework.

Until now the...