I forgot to mention that I would like to hear feedback by November 28th, considering it is holidays for most in the US between November 22-23rd.
Thank you for reviewing the changes. See my responses inline. Guy Harris wrote: > sagun shakya wrote: > >> I have developed a prototype with the option 1 proposed. I would like to >> hear from you if you have any comments or feedback. >> The changes made to the libpcap library can be seen at the link below: >> >> http://cr.opensolaris.org/~sagun/libpcap/ > > Unless you expect there to be systems that have libdlpi but that don't > have DLIOCRAW, there's no point in having the #if defined(DLIOCRAW) > and #ifdef DLIOCRAW tests. > > In addition, unless you expect there to be systems that have libdlpi > but that don't have the bufmod streams module, there's no point in > having the #ifdef HAVE_SYS_BUFMOD_H tests, either. > That's true. I''ll take out the #ifdef in both of the cases above. > In addition, is there a libdlpi routine for *sending* packets? If so, > that's what pcap_inject_libdlpi() should be using. > Yes, there is a dlpi_send() routine and I hadn't realize this. I'll make the changes to pcap_inject_libdlpi(). > Are there systems that support libdlpi and SunATM? If so, you should > propagate the SunATM code to pcap-lidlpi.c. Given that libdlpi so far is only on Solaris 11 and SunATM is not supported on it, I removed the SunATM code form pcap-libdlpi.c. If it turns out otherwise it can always be added. I'll update the webrev after I hear back from others as well. Thanks, Sagun
