On Fri, Jan 09, 2009 at 10:01:21AM +0800, Kawshol Devilal SHARMA wrote: > Hello Sarah, > > Great to see the USB 3.0, I read your blog and see that you got ~125MB of > transfer speed. > > Did you do any modification to SCSI storage portion? I see the URB in Linux > if you do scatter-gather are very small in buffer size. I didn't modify the SCSI storage driver. I did modify the USB mass storage device driver and the USB core a little, in order to push the scatter gather lists down to the host controller driver, instead of having the USB core remap them. I'm sorry I can't share my scatter gather list patch yet; I'm still working around some legal issues. I didn't need to make many changes to the USB MSD driver because the USB 3.0 device was using bulk only transport (BOT), rather than the new USB Mass Storage device class, USB Attached SCSI Protocol (UASP), that is still under draft spec in the USB-IF. The new UASP class will need a new USB device driver. One of the questions that came up when looking into a UASP driver was whether the USB core API to submit scatter gather requests was adequate. My understanding was that the driver would call usb_sg_init() and then call usb_sg_wait(), thus blocking on the request. The new USB 3.0 MSD device class is looking at asynchronous command queueing over USB using bulk endpoint "stream IDs". (You can read about stream IDs in section 4.4.6.4 of the USB 3.0 spec.) The details aren't hammered out yet, but the new USB 3.0 MSD command queueing should be very similar to SCSI command queueing. The basic idea is to have the USB MSD driver queue multiple commands to the USB 3.0 device and have the device notify the host when any of the commands are done. If the driver is blocked waiting on the scatter gather transfer by calling usb_sg_wait(), it obviously can't queue any other commands. The question is, should the driver spawn a kernel thread or a work queue or something, so that each instance can call usb_sg_init() and usb_sg_wait()? Or should the USB core API be changed so that the driver can submit multiple URBs with scatter gather list pointers in them? A more important question is will the SCSI layer can support command queueing to the USB MSD driver, or will there need to be changes there also? Sarah Sharp -- 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
