On Tue, Oct 08, 2019 at 10:09:01AM +0200, David Heinzelmann wrote: > If a device connected to an xHCI host controller disconnects from the USB bus > and then reconnects, e.g. triggered by a firmware update, then the host > controller automatically activates the connection and the port is enabled. The > implementation of hub_port_connect_change() assumes that if the port is > enabled then nothing has changed. There is no check if the USB descriptors > have changed. As a result, the kernel's internal copy of the descriptors ends > up being incorrect and the device doesn't work properly anymore. > > The solution to the problem is for hub_port_connect_change() always to > check whether the device's descriptors have changed before resuscitating > an enabled port. > > Signed-off-by: David Heinzelmann <heinzelmann.david@xxxxxxxxx> > --- > Changes in v4: > - changed commit description > Changes in v3: > - changed commit message and description > - fix code style > Changes in v2: > - fix logic error to handle return code from usb_get_device_descriptor() > properly > - fix line endings > --- > drivers/usb/core/hub.c | 196 +++++++++++++++++++++++------------------ > 1 file changed, 111 insertions(+), 85 deletions(-) What happened to Alan's ack? v5? thanks, greg k-h
