[geeks] New POWER9 motherboard coming out...

Nathan Raymond nraymond at gmail.com
Fri Feb 8 13:44:12 CST 2019


On Fri, Feb 8, 2019 at 11:38 AM Jonathan Patschke <jp at celestrion.net> wrote:

> The lack of interest in explicit big-endian support for the platform seems
> largely due to IBM proclaiming that ppc64le is the way forward.  The
> saddest part of that is the implied role of AIX in IBM's vision of the
> future of big blue wearing a red hat.
>

I haven't been paying much attention to the Power architecture or IBM in
the last few years, and didn't realize that IBM had been moving the
architecture to little endian, and my first response was, "wha? but why?!"
IBM's pages explain the details of it, but not the reason behind it:

https://www.ibm.com/developerworks/library/l-power-little-endian-faq-trs/index.html

I found this info which dates from 2010 and gives a pretty plausible reason
why this happened:

https://lwn.net/Articles/408848/

"Now, the main reasons in practice are anything touching graphics.
There's quite a few IP cores out there for SoCs that don't have HW
swappers, and -tons- of more or less ugly code that can't deal with non
native pixel ordering (hell, even Xorg isn't good at it, we really only
support cards that have HW swappers today).
There's an even bigger pile of application code that deals with graphics
without any regard for endianness and is essentially unfixable."

I thought about it a while, and graphics (and all the heaps and piles of
pre-existing code cruft) does make sense (sadly) to move Power to little
endian.

Now on a positive note about Power and the Raptor implementation, I was
overjoyed to read how they handle firmware (excerpt from the first
tenfourfox.blogspot.com link I posted at the start of this discussion):

"The major selling point to the second group is that the firmware is fully
open-source and auditable even down to the FPGA level. Schematics are
included! You can download and build your own FPGA flash image, your own
BMC flash image and your own PNOR flash image. In fact, you are expected
to, though Raptor provides pre-built versions assuming you trust them and
their warrant canary. As long as you don't brick the BMC -- though this is
doable if you are incautious -- you can play around with Petitboot and
Skiboot pretty much at will and just reflash if you screw it up.
Programming the FPGA means you'll need your own SPI programmer, but there's
a JTAG port on the board and you can plug right in. Note that this scheme
isn't perfect because you still have to trust a certain amount of the other
firmware in the system, mostly in the various peripheral devices, but it's
clearly better than what you'd get from any other system and it's a strong
start towards reclaiming control of our own machines. Although I haven't
tried writing my own custom firmware yet, it's very easy to build and flash
the prefab releases, and the process is well documented. To upgrade to the
current firmware from the v1.02 my machine came with did not require
flashing the FPGA, so I could do it all from my G5 by talking to the BMC
over SSH."

That's pretty wonderful these days. I'm pretty sure the few desktop ARMv8
systems out there aren't anywhere near that open. One of the few I'm aware
of is the Gigabyte ThunderXStation, and it's aimed primarily at developers
creating apps to run on ARMv8 in the datacenter:

https://www.anandtech.com/show/12571/gigabyte-thunderxstation-cavium-thunderx2-socs

Looks like you have to call the US distributor to get any pricing, so not a
normal retail product:

https://www.phoenicselectronics.com/

The only other "desktop" ARMv8 solution I know of is really a kit for
developing an embedded solution, NVIDIA's Jetson AGX:

https://www.anandtech.com/show/13584/nvidia-xavier-agx-hands-on-carmel-and-more

The firmware in that is probably closed off or at least proprietary... I
bet for big customers NVIDIA does custom work, but of course that's not the
same thing as open. I won't be surprised if in 2 years Apple moves at least
their notebooks to their own custom ARM, but those will be wrapped in a lot
of proptietary, T2 Security Chip-esque stuff I'm sure and probably their
own GPUs.

There's RISC-V, but right now it's still just embedded or small
single-board computers. Would be nice if someone made notebooks and
desktops based on it with open-source firmware, which might happen someday,
but we're still years off from that.

- Nate


More information about the geeks mailing list