On Thu, 3 Dec 2009, Rickard Bellini wrote: > > Hi, > > Have been playing around with the delays in hub.c and ehci-hcd.c for the remote > wakeup issue. I have found that: > > msleep(25); before usb_mark_last_busy(udev); in remote_wakeup() in hub.c > > and the change from 20ms to 30ms here: > > ehci->reset_done [i] = jiffies + msecs_to_jiffies (30); in ehci_irq() in ehci-hcd.c > > Seems to do the trick for cdc-acm, no XactErr seen during my week of testing. > At least 10000 remote wakeups was successfully handled. What about: Leave remote_wakeup() in its original form; Move the TRSMRCY 10-ms msleep in usb_port_resume() down below the SuspendCleared: label; Change the reset_done[i] delay from 20 ms to 25 ms. Does this also fix the problem? 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
