David wrote: >>> Can you post a patch? >> >> Yes, but not before 8h from now, sorry. > > OK, no rush... Not that I don't want, but I simply can't. Technically speaking. > Philippe, I know you like the name net_stack_driver, but I realy > don't. Sorry. Also, I don't want to have a /dev/net/stack entry > as it gives the wrong impression - I mean it's NOT the stack, > it's just a means of getting at the stack... People still call the visible top of an iceberg an "iceberg", even if they'll never see it fully. Same here. The driver is the only visible part of the full stack visible from userland... I mean, if you know where to look (WTF is these /dev/* paths?) or you're libnet.so yourself (a scary feeling, btw :-) ). Not being the full stack doesn't make it a valid part of the stack. But, anyway, what is in a name? ;-) I just can't stand the "socket" feeling a /dev/net/socket entry give me. Our stack isn't made only of sockets. Asking a /dev/net/socket entry to stop sound a little weird, but whatever... > I was going to rename driver/socket.c into socket_driver.c which I think is > more meaningful. ?? At least, I agree on the last name part: *_driver.c ;-) Anyway, I've introduce some new #define so that we could easily modify the /dev/net/* entry name the driver publish. Commiting, renaming, re-building, not the hard part. But agreeing? > We need to discuss the settings, but I don't really see how usernames > and passwords relate to huge text files (etc/services and so on). They don't. I'll expand on this "network global settings" topic in a next mail... -Philippe