On 23.03.23 13:52, Michael Laß wrote:
Hi,
Am Donnerstag, dem 23.03.2023 um 10:51 +0100 schrieb Oliver Neukum:
On 18.03.23 13:09, Michael Laß wrote:
--> no more communication
Just to be clear, the device is generating data, isn't it?
So you would have expected data from a suspended device?
Yes, the remote device is programmed to print "Hello World" every three
seconds. Nothing should have stopped it from doing that.
I am asking because the device says that it is bus powered.
That is, are we putting the device into some sleep state?
That's my impression as well. I had a look at the available USB device
quirks. Would USB_QUIRK_RESET_RESUME be the correct one in this case?
Booting with "usbcore.quirks=04b4:0008:b" fixes the issue because
autosuspend is disabled when this quirk is set. The description of that
quirk ("device can't resume correctly so reset it instead") does
however not really match the issue at hand and suggests that the device
would still be put to sleep but reset on every wakeup. Is there a
better fitting quirk available?
No. Theoretically we should make a new quirk for systems that keep
power across STR.
Regards
Oliver
