I have a Dell U2720Q monitor and I connect the laptop to it using USB-C on both sides. So I get video over DP Alt Mode and the monitor charges the laptop via its 90W of PD.
The problem is that, occasionally, when I plug the monitor in, I get the picture, but the laptop doesn’t charge. Then I just plug it out and in again and it starts charging.
The only thing in the kernel log that appears to be related is:
Nov 30 08:18:17 kirFw kernel: ucsi_acpi USBC000:00: UCSI_GET_PDOS failed (-5)
which is reported on the first (unsuccessful) attempt, but not on the second (successful) one. I am not very familiar with the internals of USB-C, but a quick search tells me that PDO stands for Power Deliver Object, so that’s probably it.
The kernel version is 6.5.9. And I am pretty sure that I’ve never had this problem with this monitor when using my previous laptop (Dell XPS 13) with a similar (not exactly sure which one at the moment) kernel version.
Any ideas how to debug this the next time it happens?
Jumping in to say that while I am not in this particular situation, I came across this with my Dell U3419W in connection with my XPS13 (9300).
Once the left USB-C on that thing degraded the connection became unstable. The right port still works fine. I haven’t yet experienced that with the Framework.
This may not be doable (although I lack the display to compare this to).
The error doers indicate a power delivery issue for sure. Ideally, you would connect a power supply from wall to laptop using a compatible charger from wall to laptop. Charging via the display may not be something we can help with.
One thing you could try is to use another expansion bay. But it is likely you may see a similar results.
Wait, but, like, it’s just USB-PD and it’s exactly the intended use-case for USB-C, isn’t it? So it is supposed to work.
As I mentioned, the issue is flaky with my Framework (i.e. most of the time it works just fine, but sometimes, pretty rarely, I have to quickly reconnect it). And I never had this problem with Dell or when occasionally charging my phone from the monitor (don’t ask).
Is this indicative of a problem with the monitor? Or with the USB controller in the laptop? Or a bug in the kernel? I mean, there has to be some way of diagnosing it.
There are USB challenges we’re working on, to be addressed in future BIOS releases. It’s probable that this will help, but as I don’t have a similar display to test again, we may have to wait and see. What you describe fits into this USB Power related issue.