On Sat, 2010-03-20 at 22:13 +0100, Arjen de Korte wrote: > Citeren Charles Lepple <[email protected]>: > > > Several years ago on the linux-usb list, it was often pointed out > > that the kernel was supposed to honor the wMaxPacketSize field, and > > fragment the requests accordingly. While this was suggested for > > performance reasons (fewer round trips between user and kernel > > space), it should still hold true for less frequent transfers. > > This was my understanding too, but for instance the blazer_usb (and > megatec_usb) also fragment transfers in 8-byte chunks themselves. A > while back we tried to query for larger chunks, but this wasn't > reliable at all. > > It looks like we can't trust that the kernel will do this for us, so I > guess it would be better to do this in the libusb.c module. If memory > serves, it would be possible to get wMaxPacketSize somehow from the > device handle.
I was wondering whether the 8 byte limit is going to be common for all low speed devices. If I read the hid specs correctly, the reports are read from the interrupt end point and these look like they are limited to a single 8 byte packet for a low speed device. Perhaps usb_connectinfo.slow can be used to detect such slow speed devices? > Best regards, Arjen _______________________________________________ Nut-upsdev mailing list [email protected] http://lists.alioth.debian.org/mailman/listinfo/nut-upsdev
