On Thu, Dec 23, 2010 at 09:43:12PM +0800, loody wrote: > hi: > > 2010/12/23 Greg KH <greg@xxxxxxxxx>: > > On Wed, Dec 22, 2010 at 10:17:59AM +0800, loody wrote: > >> Dear all: > >> I found something interesting. > >> I use usb_control_message to send 2000 vendor specific setup token to > >> device, the speed of each return is quite slow. > >> But if I just add "printk(".");" each time before I call > >> usb_control_message, the speed of each return will be faster. > > > > Really? You see this on the wire, or when running usbmon? > I try to run it with usbmon, but my kernel will panic before I can > successfully get the mon files out. Does it panic from your driver code? I suggest posting your code here for people to review. > I try to cat the output of usbmon to NFS folder, but it is still empty > after the kernel crash. > Is there any option or config can let me flush out the content of > usbmon immediately ? It should always be sent out as the data is sent across the wire. But fix your crash first, that seems like the biggest issue at the moment. good luck, 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