It Turns Out RISC-V Hardware So Far Isn't Entirely Open-Source

[wizard69]

What is the "analogue side" of a DRAM controller?

Transmission lines are effectively analog devices. Each line associated with a DRAM interface has to operate in an analog reality.

while i havent looked at a logic bard in a long time it is actually possible to change system timing by lengthening or shortening PCB traces. You could actually see traces lengthened to create delay lines.

[mzs.112000]

So, IMHO the best hope for an open PC platform(with real performance, not just a toy), is POWER...

[bridgman]

Core or ISA? The quote from the article suggests only the ISA is open, which I think is also true of SPARC, nowadays (and POWER, as noted).

You can even supposedly get open source implementations:

https://en.wikipedia.org/wiki/SPARC#...mplementations

I'm not sure - my understanding was same as yours... the ISA is the primary thing that is "open", but a few people have published open implementations of the core.

There might also be an open implementation of an entire SOC, but I expect that would have to be foundry/process dependent.

[e8hffff]

We need truly open hardware using the likes of RISC-V. No more backdoors to Israel or Iran, or America, or Russia.

[brrrrttttt]

What is the "analogue side" of a DRAM controller?

What is the digital side? Just a concept written on (proverbial) paper. There is really no such thing as a "digital" signal when physics gets involved.

[log0]

What is the digital side? Just a concept written on (proverbial) paper. There is really no such thing as a "digital" signal when physics gets involved.

Nope, within given context, there is such a thing as "digital".

Digital devices switch between a finite number of states, analog work with continuous ranges of values.

[L_A_G]

I knew that the final implementations would still be pretty heavily touched by proprietary tech, but I do have to admit I forgot about necessary, but also complex components with off-the-shelf IP cores like DRAM controllers and focused on synthesis tools for creating the actual hardware implementation from the VHDL/Verilog just about all silicon is designed in these days. Seems like I was vindicated in my my assertion that these things won't be all that much more "open" than most ARM-based SoCs.

[brrrrttttt]

Digital devices switch between a finite number of states, analog work with continuous ranges of values.

And what do they do between those states?

[kpedersen]

So could anyone point me towards a solution which is fully open-source?

Does one actually exist? It seems that Lemote (Stallman's last noted open-source platform) is no longer in trade of open hardware?

[-MacNuke-]

Being open is just a part of RISC-V. Other parts are "license free" and "patent free" (at least as an ISA). You still need a license if you want to use OpenPOWER.