coreboot

*More correctly, unless you revert the commit, the port does not work

I can report also having this issue.
Reverting 9554b26f9fd608 fixed it for me, and is my temporary solution.

It seems like commit 4f4322dd68575402e099e2dfda057687388f064e (which actually mentions this issue) was not enough to resolve this issue for me.

I tested it on master and don't find an issue with get_option() on an Ivy bridge system. Can you investigate to see what the problem is?

Anton Kindestam wrote:

I can report also having this issue.
Reverting 9554b26f9fd608 fixed it for me, and is my temporary solution.

It seems like commit 4f4322dd68575402e099e2dfda057687388f064e (which actually mentions this issue) was not enough to resolve this issue for me.

Hi Anton,

Can you give me some more detail? I wrote the https://review.coreboot.org/c/coreboot/+/29565 patch originally. Can you

• Run ectool -d and save it's output to a file
• Set the option through nvramcui or nvramtool
• Reboot
• Confirm that the setting applied
• Run ectool -d again, saving to a different file

And then attach the files here?

Patrick Rudolph wrote:

I tested it on master and don't find an issue with get_option() on an Ivy bridge system. Can you investigate to see what the problem is?

IIRC (because it was a little while ago when I initially debugged the patches) it's due to the changes in 9554... not allowing get_option() to work in SMM, which is where the original code sets the EC flag. I think it is mentioned in a comment in 9554, but I might be confused.

Anton Kindestam wrote:

Sorry for the late reply.

I have now tested it on coreboot-4.9-514-gdacf083c2f. It seems it only works when using the yellow USB connector, but not any of the other two. I had it working on those as well earlier (by reverting 9554b26f9fd608), unless I'm severely misremembering. Alas, reverting that doesn't seem to do the trick any longer, and I have since rebased...

Files attached. Before is before enabling usb_always_on, after is after enabling it.

I'm not sure how you got it to work with the other ports initially, as the schematics show a specialised USB charge controller (U20 in the bottom left) only on that yellow port: https://i.imgur.com/XfsHQpC.png
The two lines entering it in the bottom left appear to be what we're controlling in the EC, and as the charge controller doesn't hook up to the other ports, there must have been some other reason they were powered.

Those before/afters you posted don't appear to have the correct EC flag set - is that with my patch reverted or just the normal tree un-touched?

It is with the normal tree un-touched. Specifically at commit dacf083c2f

I think I will have to do some testing with an old coreboot build which I believe worked and with stock firmware, to see if I can have the black ports charge while suspended.