https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4664
--- Comment #8 from Chris Maynard <christopher.maynard@xxxxxxxxx> 2010-04-12 21:12:59 PDT ---
(In reply to comment #7)
> This isn't removing the pseudo-header, it's just changing the way Wiretap
> supplies it to libwireshark; instead of supplying it in the "union
> pseudo_header", it supplies it as part of the (almost-)raw packet data, by not
> stripping it off of the raw packet data, and leaving it up to the USB dissector
> to interpret the beginning part of the raw packet data as a USB pseudo-header
> rather than getting that information from the "union pseudo_header".
OK, I follow you. Does the USB dissector have enough knowledge to know the
endian-ness of the Linux USB pseudo header though? If not, I guess wiretap
could indicate it somehow by a flag? And if we do go this route, then that
endian-ness flag would also need to be made available to the code that saves
off capture files as well since the Linux USB pseudo header is apparently
supposed to be saved in host-endian format. So, if the host endian-ness
doesn't match the original endian-ness, then the raw data can't just be
written. The multi-byte fields would need to be swapped. Yes?
--
Configure bugmail: https://bugs.wireshark.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.