Hmm, so it's still not clear to me what the bug really is, but I think CMD_GET_FW_VERSION should be retired. biot's use of the usb dscr makes much more sense.




On 27 April 2012 at 16:53 wrote:

> > Please build either libusb.git or libusb-1.0.9 from source and make
> > sure to configure it with --prefix=$prefix --enable-debug-log and
> > then send the log to the mailing list.
> I did some more testing and used a debug build of libusb as Peter
> requested. I tested 3 scenario's:
> A: original Saleae's Logic 1.1.5 32-bit ubuntu version
> B: libsigrok version 0574e4c04dac739d758ec609c3737f1a64fea7ac with Saleae
> 1.1.5 firmware
> C: latest git with known working fx2lafw from Bert
> Results
> -------
> A: works like a charm, which make me conclude that the device, bus and
> cables are ok.
> B: fails first run (with libusb error), works on second run, and fails
> every following run with "status 2 received 0 bytes". After resetting the
> device, the same behaviour is seen.
> C: fails every time with random error messages.
> The device is the lcsoft board with an additional i2c eeprom.
> Logs from B and C attached.
> -- Martin