The Updated AMD Polaris Firmware Blobs Needed For RX 480 Support Land

[vortex]

So, what is the plan for the review anyway?
I assume the 480 is going to go up against the 950/960/970/980 and 380/380X/390/390X?
Doing a review of a 480 against 1070/1080 is worthless, completely different markets.

[Michael]

So, what is the plan for the review anyway?
I assume the 480 is going to go up against the 950/960/970/980 and 380/380X/390/390X?
Doing a review of a 480 against 1070/1080 is worthless, completely different markets.

I tested all of the relevant AMD and NVIDIA GPUs I have available.

[pal666]

Proprietary firmware is a serious concern to some people and their reasoning is sound from what I can tell

the reasoning is bullshit, you just can't tell properly

[dungeon]

Sorry, I posted this in the other thread on accident, but Is there any way to get OpenCL 2.0 working with these, or with any of the newer AMD drivers beyond fglrx?

It should work with amdgpu beta i guess, but as they say cl is not fully validated there... whatever that means, maybe to do not complain if it does not work reliable/even at all/xyz... So best bet is fglrx still.

[bridgman]

I think OpenCL support in the new amdgpu stack is at 1.2 with some extra extensions... don't think you can get 2.0 functionality out of it today.

[dungeon]

I think OpenCL support in the new amdgpu stack is at 1.2 with some extra extensions... don't think you can get 2.0 functionality out of it today.

He, he, back in march Michael posted clinfo which said 2.0, so i guess there is some inconsistencies there

http://www.phoronix.com/scan.php?pag...-gpu-pro&num=1

[boltronics]

the reasoning is bullshit, you just can't tell properly

Regardless of the reason, it means you can't just install GuixSD (for example) and expect a working computer. They even disable the amdgpu DDX support out of the box last I checked (because what's the point if it won't work due to lack of microcode anyway)? And of course they disable the ability to load non-free microcode.

GuixSD is an amazing distribution, but this firmware issue (regardless of if you agree with their reasons or not) makes AMD hardware a PITA.

[bridgman]

I could make a distro that removed the X server and several moving parts; that would be a PITA as well, but it wouldn't fly because nobody could use it. That would be inconvenient.

Many years ago the FSF saw a convenient dividing line they could draw that they thought might result in getting more hardware internals opened up without totally crippling Linux graphics for everyone. Today the two vendors who support open source graphics drivers are in the "bad" camp and the last remaining vendor only supporting binary drivers is in the "good" camp. This arbitrary line is baked into a number of distros that soon are only going to be fully usable with binary drivers or with old hardware. Convenient isn't the same as good.

It's time to fix what was an arbitrary guideline at the best of times, and is now bordering on the absurd despite the best intentions of everyone involved.

[boltronics]

The main thing I want to know about the RX 480 is if it continues to be hit by the random amdgpu lock-up bug. https://bugs.freedesktop.org/show_bug.cgi?id=93652

[boltronics]

It's time to fix what was an arbitrary guideline at the best of times, and is now bordering on the absurd despite the best intentions of everyone involved.

I agree for the most part. Either the FSF should care if non-free firmware is built into flash as well and block that, or permit firmware files to be loaded on hardware that requires it. It's the biggest disagreement I have with the FSF, but I am a FSF supporter.

But probably there is still a line that should be drawn. We don't want to go down the path where proprietary firmware is needlessly complex just to avoid proprietary drivers (for example). But I don't see any easy way to avoid that without just blocking it all.

My personal hope was that AMD might be able to release an alternate free firmware for distros like GuixSD which doesn't have any of the DRM stuff that Windows requires (or maybe provide some assistance to teams that want to work on it), but I understand that would be a challenging and time-consuming task. Hopefully one day AMD sees benefit to addressing this.