On Fri, May 04, 2012 at 03:32:13PM -0400, Alan Stern wrote: > Greg: > > There's a problem in the usb-serial core. When a serial driver > registers itself, the core automatically sets the .supports_autosuspend > flag in the usb_driver structure. The problem is that many of the > serial drivers (such as ftdi_sio) don't define their own suspend and > resume routines. Consequently when the system tries to autosuspend one > of these drivers, it ends up calling a NULL pointer. > > I gather the way this is supposed to work is that each suspend or > resume operation goes through usb_serial_suspend() or > usb_serial_resume(), which then calls the appropriate method in the > serial driver if one is defined. Is that the right approach? Yes, that is the way it is supposed to work. > Assuming it is, I've got a patch to fix the bug. I'll submit it if you > confirm that my impression is correct. Thanks, that would be great, I haven't tested suspending usb-serial devices in a long time. greg k-h -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
