Re: [PATCH/RFC 2/2] usb: gadget: u_ether: Don't change endpoint status in eth_stop()

2012-08-09 Thread Kevin Cernekee
On Thu, Aug 9, 2012 at 1:26 AM, Felipe Balbi wrote: > I have applied this one instead: > > commit a3ab51013e6af38b3e2f9f20bf469cf8c595391b > Author: Michael Grzeschik > Date: Wed Aug 8 11:48:10 2012 +0200 > > usb: gadget: u_ether: fix kworker 100% CPU issue with still used > interfaces in

Re: [PATCH/RFC 2/2] usb: gadget: u_ether: Don't change endpoint status in eth_stop()

2012-08-09 Thread Felipe Balbi
Hi, On Sun, Jul 08, 2012 at 11:53:08AM -0700, Kevin Cernekee wrote: > This is a follow-up to an earlier thread: "gadget: Clearing ep->desc in > struct usb_ep on EP disable?" > > I took a closer look at u_ether.c and I had some concerns about the > synchronization between USB connection/disconnect

[PATCH/RFC 2/2] usb: gadget: u_ether: Don't change endpoint status in eth_stop()

2012-07-08 Thread Kevin Cernekee
This is a follow-up to an earlier thread: "gadget: Clearing ep->desc in struct usb_ep on EP disable?" I took a closer look at u_ether.c and I had some concerns about the synchronization between USB connection/disconnection vs. netdev open/close. For instance: If somebody yanks the USB cable in t