Hacker News Reader

Comment by karma_pharmer | Hacker News Reader

Rockchip RK3399 and RK3588 SoCs

Just so people don't get confused, there is a huge world of difference between these two chips. They are not in the same category.

Rockchip RK3399 is 100% blobless. You even control the EL3 Trustzone Secure World! This is True Root.

Rockchip RK3588 still needs blobs in EL3, the highest privilege level. We've been hearing rumors for years now about "oh they'll open source it next month for sure" and it.... just. never. happens. Please stop spreading this rumor. Source or GTFO, Rockchip.

View on HN

Feb 2024, https://news.ycombinator.com/item?id=39490540 & https://review.trustedfirmware.org/c/TF-A/trusted-firmware-a...

> Rockchip have sent a few patches to the TF-A project here to support [RK3588].. From TF-A we can now build a complete working BL31 and replace the closed binary blob with an open-source binary that we can compile ourselves.. There are still some missing parts and the most important that is remaining right now is the DDR training blob, which is still closed source.

Like I said: blobs in EL3, the highest privilege level.

We've been getting these "almost there" announcements quarterly for multiple years at this point.

In decreasing order of preference:

  OSS TF-A at EL3 + OSS DDR training (RK3399)
  OSS TF-A at EL3 + closed DDR training (RK3588)
  OSS uboot + closed TF-A 
  closed uboot + closed TF-A

Those who need the features of RK3588 can compare it to competitors which are less open.

Those who don't need the features of RK3588 can use the older and fully open RK3399.

> OSS TF-A at EL3 + closed DDR training (RK3588)

That should be "blob which does stuff, including DDR training, at EL3".

That blob certainly does DDR training. Maybe it does other stuff. But all of it is done at EL3.

> In decreasing order of preference

"Warning! Diagram is not drawn to scale."

> That blob certainly does DDR training. Maybe it does other stuff. But all of it is done at EL3.

If someone wants to find out, they can load it in IDA/Ghidra?

If it was that easy we'd have a cleanroom FOSS implementation by now. It's been more than two years!

https://wiki.raptorcs.com/wiki/Project_Ortega

Cleanroom reverse engineering for the purpose of publishing new driver code, to avoid legal/IP minefields, is super-expensive. It should be a much narrower scope to determine whether a binary blob's actions are limited to memory training, since there is no requirement to publish reusable source code.

Looks like Collabora is already monitoring the blob, so it's not entirely a mystery:

  At the moment of writing this article, we have identified a few differences from the binary blob previously used, which we can highlight as following:

  Binary BL31 contains some unknown code to get HDMI-RX PHY access working.
  The cpufreq support in binary BL31 is different from TF-A. 

  There could be more issues that are unknown at the moment and users should be aware of it.

> Cleanroom reverse engineering for the purpose of publishing new driver code, to avoid legal/IP minefields, is super-expensive

Project Ortega did it to an entire gigabit ethernet controller for the princely sum of zero dollars:

https://media.ccc.de/v/37c3-11781-adventures_in_reverse_engi...

https://wiki.raptorcs.com/wiki/Project_Ortega

> Binary BL31 contains some unknown code to get HDMI-RX PHY access working.

That doesn't sound like DDR training to me. Maybe now you see the problem?

> That doesn't sound like DDR training to me. Maybe now you see the problem?

It is absolutely a problem, but it's bounded by the ability to inspect and question code/behavior outside the officially claimed rationale. We can prefer open systems and also shine a bright light on the behavior of closed systems.

Reply on news.ycombinator.com