On Mon, Jul 13, 2009 at 05:52:43PM +0200, javier Martin wrote: > Sorry for breaking thread but I missed the main one. > > I am testing last version of Daniel Mack's patches in a i.mx27 based > board and I am getting the error in the attached file. > For some reason (probably a mistake of mine in platform code) USB > transceiver is not being detected through ULPI > interface, however, the driver continues registering EHCI untill it crashes. > > Shouldn't be cleaner to unregister what has been registered and abort > driver execution? Well, we should probably, but there is broken boards - at least I suspect mine to have a problem there which nobody else was to reproduce with the same software stack. And these boards will still work fine even if this part of initialization fails. Hence I didn't make that a hard error. > The system should be able to start although there is an error with the > transceiver. True, and the error you're seeing does not to be related to direct ULPI communication but occurs at some later point. Could you put some printk()s in the ehci driver part and see where exactly it crashes? The logs won't currently show that. Also, posting your board-specific platform data would help. But I've never ran into a "Unhandled fault: external abort on non-linefetch (0x808) at 0xc4846184". Maybe anyone else can explain that? Daniel -- 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
