On Thu, May 19, 2011 at 01:32:32PM +0800, yingchun li wrote: > On Thu, May 19, 2011 at 11:59 AM, Greg KH <greg@xxxxxxxxx> wrote: > > On Thu, May 19, 2011 at 10:08:33AM +0800, yingchun li wrote: > >> On Wed, May 18, 2011 at 9:50 PM, Greg KH <greg@xxxxxxxxx> wrote: > >> > On Wed, May 18, 2011 at 05:22:42PM +0800, yingchun li wrote: > >> >> On Tue, Apr 19, 2011 at 11:49 AM, Greg KH <greg@xxxxxxxxx> wrote: > >> >> > On Tue, Apr 19, 2011 at 11:41:29AM +0800, yingchun li wrote: > >> >> >> On Fri, Apr 15, 2011 at 4:09 AM, Greg KH <greg@xxxxxxxxx> wrote: > >> >> >> > On Thu, Apr 14, 2011 at 01:48:46PM +0800, yingchun li wrote: > >> >> >> >> Hi, > >> >> >> >> I have a mobile device, and using USB general serial driver to > >> >> >> >> put some log to PC. > >> >> >> > > >> >> >> > Ick, don't use that driver for any "real" work. What type of device is > >> >> >> > this really? We should use the real driver for it. > >> >> >> Why I cannot use this driver? does this driver have problems? > >> >> > > >> >> > It is slow, has no flow control, and you have to pass the module > >> >> > parameter to it for every time you load the module. It also only > >> >> > supports one device id (well, there are ways around that as well, but I > >> >> > don't recommend it.) > >> >> Hi, Greg > >> > > >> > <snip> > >> > > >> > For some reason you sent this only to me, which is a bit rude to > >> > everyone else on the mailing list. I'll be glad to respond if you > >> > resend it to everyone. > >> sorry, my careless, I didn't find that, thanks for note me > >> so I resend my question: > >> I do a simple test(which write to /dev/ttyGS0 repeatedly,and use the > >> minicom receive the data), and the speed is about 48KB/s(my ARM-core > >> cpu's bogomips is 276). why the speed is so slow? > >> because our PC application based on the generic serial driver, so I > >> have to use this driver, is there any way to enhance this? > >> I have try to enlarge the WRITE_BUF_SIZ from 8192 to 16K, but has > >> little effect. > > > > That is correct, the generic driver is very slow, and will stay that > > way, sorry. > > > > Please feel free to add your device id to the correct driver, which will > > go much faster, what is the chip type in this device? > Oh, thanks for your reply so quickly. and I think your western are in sleep > at this time. > In fact, it is an android device, and the chip type is ARM soc. with USB, > it has some gadget functions(mass storage, adb and usb-serial...), we use > generic usb serial to send some log to PC. > but my understanding is that the bottleneck is in my gadget usb2serial, not > PC's driver. I have test this with windows, and use hyperterminal, the result > is almost same. Wait, are you referring to the gadget side speed issues, or the host side? Please be more specific. And if you can, look at the usb bus traces to see where your speed is going, odds are you have lots of delay time on the host side. 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
