On Thu, 6 Jun 2019, Greg Kroah-Hartman wrote:

On Thu, Jun 06, 2019 at 10:24:18AM -0400, Alan Stern wrote:

quoted

On Thu, 6 Jun 2019, David Howells wrote:

quoted

Add a USB subsystem notification mechanism whereby notifications about
hardware events such as device connection, disconnection, reset and I/O
errors, can be reported to a monitoring process asynchronously.

USB I/O errors covers an awfully large and vague field.  Do we really
want to include them?  I'm doubtful.

See the other patch on the linux-usb list that wanted to start adding
KOBJ_CHANGE notifications about USB "i/o errors".

That patch wanted to add notifications only for enumeration failures
(assuming you're talking about the patch from Eugeniu Rosca), not I/O
errors in general.

So for "severe" issues, yes, we should do this, but perhaps not for all
of the "normal" things we see when a device is yanked out of the system
and the like.

Then what counts as a "severe" issue?  Anything besides enumeration 
failure?

Alan Stern