ODROID-C1/-C2 USB Detection only triggered by some devices

[Martin Blumenstingl]

Hi Anand,

On Tue, Jul 13, 2021 at 8:05 AM Anand Moon <linux.amoon at gmail.com> wrote:
[...]
> > If you're interested you can look at one of my earlier emails [0] (not
> > sure if you were Cc'ed then) where I was running the cat command as
> > suggested by Alan three times (each time plugging in different devices
> > at different times) and got three different outputs.
> >
>
> Sorry for the late reply, I have submitted small code changes for this.
> [0] https://patchwork.kernel.org/project/linux-amlogic/list/?series=514467
>
> Yes, I have observed the passing usbcore.autosuspend=-1
> with command line parameter help resolve the hotplug issue with USB
> storage device.
Thanks for reporting back.
I just tried your code changes from the series you mentioned above and
it doesn't seem to have any impact on the issue which is described in
this whole thread. Can you please elaborate which change you have seen
on the issue described here (hotplug not working) with the changes
from your series applied?

> Yes it seems that USB hub is entering into LPM
>
> dwc3 has the following quick property to handle these cases.
Minas' last statement was [1]:
"There are couple of transaction errors and some EP stalled. I want to
understand on which device/ep this happen."
I don't know whether transaction errors and EP stalls are related to
LPM, so I cannot comment on this any further.

[...]
> # usbmon
> Earlier my aim was to capture the USB PHY event via usbmon and wireshark.
> so that we could verify the handshake.
> I am attaching the Wireshark capture log using the following command.
I still don't understand how you are coming to the results which you
have attached to your email:
- the dmesg output starts more than 30 minutes after boot
- it's not clear if you ran any "lsusb -vv" command at some point
- which devices were plugged in during boot?
- ...

>From my experience at work (where I also do software development and
support) two things are important when reporting and investigating
problems:
- being as precise as possible to make sure everyone is one the same page
- at the same time focusing on relevant parts

Alan has helped us out in the beginning of this investigation so Matt
and I have been providing him with all the information he needed.
Now Minas is the dwc2 expert helping us solve this issue which is why
I am waiting for further instructions from him on how to proceed.


Best regards,
Martin


[0] https://lore.kernel.org/linux-amlogic/CAFBinCA9Y16Ej3PEBN1Rsqo=6V1AZXKOpTfc_siHP0rvVo7wWQ@mail.gmail.com/
[1] http://lists.infradead.org/pipermail/linux-amlogic/2021-July/010341.html
[2] http://lists.infradead.org/pipermail/linux-amlogic/2021-July/010343.html