> We always used to byte-swap just a few fields in the descriptor, to
> optimise access to those. We never bothered to put them back when we
> passed them up to userspace via usbdevfs -- we gave a structure which
> was mostly LE but had precisely four fields byteswapped to host-endian.
> The upstream version of usbutils doesn't expect this -- it expects the
> descriptor to be entirely little-endian, as it's received from the
> device. John's version of usbutils (which distro, is that, btw?)
> evidently has a hack to work around it.
I'm running Debian 'Sarge' and there are currently no bug reports for
either 'usbutils' or 'libusb'.
I would support doing things consistently, especially if the relevant
utility(s) can readily determine whether byte-swapping is necessary or
not.
-- JM
P.S. Other readers: Hey, what about the SCSI oops???
-------------------------------------------------------------------------------
-- System Information:
Debian Release: 3.1
APT prefers testing
APT policy: (500, 'testing')
Architecture: powerpc (ppc)
Kernel: Linux 2.6.10
Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968)
Versions of packages usbutils depends on:
ii libc6 2.3.2.ds1-20 GNU C Library: Shared libraries an
ii libusb-0.1-4 1:0.1.8-17 Userspace USB programming library
-- no debconf information
===============================================================================
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
