Andrew Jeffery

I've added a table describing all the bits in each SoC's physical address space for `org.fwupd.hsi.Aspeed.iLPC2AHB.ReadWrite`. Let me know what you think. If it's helpful we can discuss how we...

> Nah, I mean if you do culvert probe when running on the SoC device it should show the user what the problems are. This should work. Is it not...

Ah, I think we're running up against two things here: 1. Naming things is hard 2. My absolute lack of any design or usage documentation Let me try to fix...

Okay, I've added the rest of the tables and covered the interesting configuration bits for all of the AST2400, AST2500 and AST2600 in the shared google doc. That should be...

Hmm, let me poke at this. I tend to order local includes before system includes so the local headers can adjust the behaviour of system headers as necessary. The fact...

Well, I think we need to decide which way we go. I'm struggling to find time to maintain it. What are your thoughts @mihais ?

I've made some comments as a first pass. I'd like to get the clean-up changes in first, then we can go over the actual graph changes.

```c cc -c -I. -O2 -g -ggdb -Wall -std=gnu11 -MMD -o ccan/tal/talloc/talloc.o ccan/tal/talloc/talloc.c ccan/tal/talloc/talloc.c: In function 'hash_ptr': ccan/tal/talloc/talloc.c:197:27: warning: cast from pointer to integer of different size [-Wpointer-to-int-cast] 197 |...

Perhaps a better direction for this is, if we haven't improved the log messages in the mean time, what could we do better to make them more informative?

Slightly trollish suggestion, but what about making InternalError assert()? I imagine that'd encourage people to sort them out, if at the cost of a bit of up-front instability.