From: Pete B. <pb...@gm...> - 2010-01-28 16:07:42
|
On 2010.01.28 15:02, Xiaofan Chen wrote: >>> The other device still fails, the error is similar to what Pete has. > > Now this is the only one which failed (Sandisk 1GB) among my 4 tested device. Interesting results. But the fact remains that we still need to figure out why these devices still fail. I tried to use usbmon on Linux to have a look at what the kernel does on a dd command for instance, but either there's some caching going on or /dev/sd# access is not picked by usbmon as the traces I get don't show the commands I'd expect (but I'm still new with usbmon). >> Yet another one now passes the test, even if it is using EP2 OUT. It also >> has EP3 IN, but I am not so sure the use of EP3 IN. The current code will use the first valid IN/OUT endpoint it sees as default, so it would use 81 and 02. > This is from Kingston (256MB). > > mcuee@ubuntu:~/Desktop/build/libusb1/windows/libusb-pbatard/examples$ > libusb:debug [bulk_transfer_cb] actual_length=0 > get_mass_storage_status: Overflow The overflow was normal, since there were still a few things that needed fixing with the read command. I have updated xsub in r121 so that overflows are averted. Also, it is now possible to use "xusb -k vid:pid", with 'vid' 'pid' your device ids in hexa, so that you don't have to recompile every time. Regards, /Pete |