Alan, If I understand is right, at this place ohci->hcca->done_head can not be 0. Is it correct? I am now trying your patch. My modification to the ohci_irq is ok now. Thanks, Frank -----Original Message----- From: linux-usb-owner@xxxxxxxxxxxxxxx [mailto:linux-usb-owner@xxxxxxxxxxxxxxx] On Behalf Of Alan Stern Sent: Wednesday, December 30, 2009 12:35 AM To: Huang, FrankR Cc: USB list Subject: RE: A question about ohci_irq() On Tue, 29 Dec 2009, Huang, FrankR wrote: > Alan, > > I trace into it and found the ohci->hcca->done_head is 0 when > handling OHCI_INTR_WDH interrupt, that will cause the mouse hang. I do > not know why ohci->hcca->done_head will be 0 when ohci_irq() is > triggered. Can you give me some explanation for that? The only explanation I can think of is that your controller is buggy. > Because from the > 2.6.22 kernel source, I found that when entering the ohci_irq() > function, the code will justify if the ohci->hcca->done_head is 0. The > 2.6.27 kernel does not have that part code. Please try the patch below and see if it fixes the problem. Alan Stern Index: usb-2.6/drivers/usb/host/ohci-q.c =================================================================== --- usb-2.6.orig/drivers/usb/host/ohci-q.c +++ usb-2.6/drivers/usb/host/ohci-q.c @@ -861,6 +861,13 @@ static struct td *dl_reverse_done_list ( struct td *td = NULL; td_dma = hc32_to_cpup (ohci, &ohci->hcca->done_head); + + /* Some controllers don't set hcca->done_head as soon as they should */ + if (!td_dma) { + udelay(10); + td_dma = hc32_to_cpup(ohci, &ohci->hcca->done_head); + } + ohci->hcca->done_head = 0; wmb(); -- 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 -- 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
