this post was submitted on 19 Feb 2024
262 points (99.6% liked)

linuxmemes

20880 readers
5 users here now

I use Arch btw


Sister communities:

Community rules

  1. Follow the site-wide rules and code of conduct
  2. Be civil
  3. Post Linux-related content
  4. No recent reposts

Please report posts and comments that break these rules!

founded 1 year ago
MODERATORS
 

Running dmesg on PinePhone reveals that idiot firmware is bored.

you are viewing a single comment's thread
view the rest of the comments
[–] [email protected] 12 points 9 months ago* (last edited 9 months ago) (1 children)

Ah, that makes sense! I didn't realise that it was the modem. It runs proprietary firmware, which developers can't really replace.

Edit: I noticed this message when trying to plug in USB devices while the keyboard addon is attached. For some reason USB devices only work without the keyboard. But I couldn't find any mention of this other than some old bug that has already been fixed.

The modem does have issues with waking up from sleep and I sometimes have to restart it with a script. WiFi has a similar problem.

[–] [email protected] 16 points 9 months ago* (last edited 9 months ago) (1 children)

I see, with that in mind the other lines in the output make more sense.

[26380.765876] anx7688 0-0028: enabling vbus_in power path
[26381.373155] anx7688 0-0028: BC 1.2 result: SDP

They show the modem's power path being enabled, the BC 1.2 result indicating the USB charging mode, and the change in the charge controller status. These lines are related to the modem's power management and charging behavior.

Finally, the output :

[26398.882595] anx7688 0-0028: cc_status changed to CC1 = SRC. Open CC2 = SRC.Open

Is related to the anx76880 USB Type-C controller on the PinePhone, indicating a change in the CC (Communication Channel) status, specifically stating that CC1 is now in the SRC (Source) state and CC2 is in the SRC.Open state.

My speculation then, doesn't seem quite right.
It seems that the modem is behaving improperly when the USB power management state is changed. When plugging in the keyboard along with the other USB devices, it triggers a USB power state change, which makes sense as it needs power from the USB port to work, but the modem firmware seems to improperly handle that change and probably hard crashes and turns off the hardware, then requires a hard reset to power back on which is detected by the "idiot firmware is bored" mechanism that sends the hard reset signal.

As for :

For some reason USB devices only work without the keyboard

That seems unrelated to this specific output. This maybe tangentially related however, pointing to a possible underlying root cause. I recommend looking into the journald logs and see if you can find other related messages. For this, Kjournald is a really good GUI that can help sort the messages.

[–] [email protected] 5 points 9 months ago* (last edited 9 months ago) (2 children)

Just to clarify, this is the keyboard addon that I was talking about. It has a built-in battery. It has a button on the side that lets you toggle the battery on and off. When I disable the battery and plug in a USB mouse, it will work, just as it does without the addon. But not with the battery enabled. So the extra battery must be interfering with the power to the USB port somehow. Surprisingly the result is the same even when plugging in a powered USB-C hub with the mouse plugged into it.

When I disable the battery, then plug in the mouse and then enable the battery again, the mouse will keep working.

I don't know much about the hardware, though. I'm even surprised that the modem has something to do with delivering power to devices.

Edit: I guess misunderstood some things. anx7688 seems to be a USB controller. The modem is separate from the SoC and is connected internally via USB. But maybe it's not a problem with the modem in this case, but with the controller.

[–] [email protected] 9 points 9 months ago* (last edited 9 months ago)

Edit: I guess misunderstood some things. anx7688 seems to be a USB controller. The modem is separate from the SoC and is connected internally via USB. But maybe it's not a problem with the modem in this case, but with the controller.

That's my bad, I should've specified that anx7688 is the USB-C bridge driver controlling all USB devices along with the modem.
In this case the modem is being effected by the USB charging state & USB-C peripherals over CC pins from anx7688.

Your other issues are very likely tangentially related to the modem error messages caused by an underlying issue within the anx7688 USB-C bridge driver's controller.

[–] [email protected] 1 points 9 months ago* (last edited 9 months ago) (1 children)

The modem is separate from the SoC and is connected internally via USB.

I guess that's a pinephone thing, to keep modem and phone separate? How is it usually connected, with all the things it can do on your phone?

[–] [email protected] 3 points 9 months ago* (last edited 9 months ago)

Yeah, it's to isolate the modem and I think it's done the same way in Librem 5. Both phones also have a killswitch that lets you power off the modem. As far as I know other phones instead have the modem built into the SoC and there is some isolation too, but I don't really know how that works and I guess it's done by the chip's manufacturer (but I'm not sure).

Like the Librem 5, the PinePhone uses separate cellular baseband and Wi-Fi/Bluetooth chips. Together with the hardware kill switches, this results in larger printed circuit boards (PCBs) and less energy efficiency compared to the mass-produced Android phone that has an integrated System on a Chip, such as the Snapdragon, Helio or Exynos. The PinePhone is thinner at 9.2 mm than the Librem 5 which is 15.5 mm thick because the PinePhone solders its wireless communication chips to the PCB whereas the Librem 5 places the cellular baseband and Wi-Fi/Bluetooth on two removable M.2 cards.[1][18][19]

https://en.wikipedia.org/wiki/PinePhone