On Tue, 24 Aug 2010, Gadiyar, Anand wrote: > I quickly tried this with v2.6.36-rc1 on an OMAP4 board (similar MUSB > controller as the OMAP3 that Robert is running). > > I see that with stall=y, I have repeated messages like this on the host, > a little after enumeration: > [2091018.128077] usb 1-5: reset high speed USB device using ehci_hcd and address 71 > > Assuming Robert's problem is due to the same cause, here're a couple > of usbmon traces from my test - one with stall=n passed to g_file_storage, > and one without. > (I've added 2 blank lines in the stall=y log, just before I got the reset message). > > The corresponding dmesg logs, are at the top of the same traces. The usbmon trace indicates that the host clears the bulk-IN endpoint halt but then doesn't receive any more data. This might mean that the clear-halt code doesn't set the data toggle back to 0. Or maybe something else is going wrong; for example, perhaps the clear-halt causes the device controller to forget about data that is already queued for the bulk-IN endpoint. Robert said that adding debugging to the gadget driver fixed the problem. I don't think debugging statements would affect the data toggle. Alan Stern -- 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
