On Fri, Jan 28, 2022 at 12:12:38PM +0100, Oleksij Rempel wrote: > On Thu, Jan 27, 2022 at 02:22:49PM +0100, Greg KH wrote: > > On Thu, Jan 27, 2022 at 01:31:52PM +0100, Oleksij Rempel wrote: > > > On Thu, Jan 27, 2022 at 12:13:53PM +0100, Greg KH wrote: > > > > On Thu, Jan 27, 2022 at 12:07:42PM +0100, Oleksij Rempel wrote: > > > > > The weakest link of usbnet devices is the USB cable. > > > > > > > > The weakest link of any USB device is the cable, why is this somehow > > > > special to usbnet devices? > > > > > > > > > Currently there is > > > > > no way to automatically detect cable related issues except of analyzing > > > > > kernel log, which would differ depending on the USB host controller. > > > > > > > > > > The Ethernet packet counter could potentially show evidence of some USB > > > > > related issues, but can be Ethernet related problem as well. > > > > > > > > > > To provide generic way to detect USB issues or HW issues on different > > > > > levels we need to make use of devlink. > > > > > > > > Please make this generic to all USB devices, usbnet is not special here > > > > at all. > > > > > > Ok. I'll need some help. What is the best place to attach devlink > > > registration in the USB subsystem and the places to attach health > > > reporters? > > > > You tell us, you are the one that thinks this needs to be reported to > > userspace. What is only being reported in kernel logs that userspace > > somehow needs to see? And what will userspace do with that information? > > The user space should get an event in case there is a problem with the > USB transfers, i.e. the URB status is != 0. That's pretty brave, lots of things can have a urb status of != 0 in semi-normal operation, have you tried this? > The use space then can decide if the USB device needs to be reset, power > cycled and so on. > > What about calling a to-be-written devlink function that reports the USB > status if the URB status is not 0: > > diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c > index d0f45600b669..a90134854f32 100644 > --- a/drivers/usb/core/hcd.c > +++ b/drivers/usb/core/hcd.c > @@ -1648,6 +1648,8 @@ static void __usb_hcd_giveback_urb(struct urb *urb) > usb_unanchor_urb(urb); > if (likely(status == 0)) > usb_led_activity(USB_LED_EVENT_HOST); > + else > + devlink_report_usb_status(urb, status); Try it and do lots of transfers, device additions and removals and other things and let us know what it reports. thanks, greg k-h
