mbuf -> sockaddr patch for get{sock,peer}name and accept

[Tyler Retzlaff]

hi,

attached is the next patch for converting protocol
user-request methods to use sockaddr * instead of mbuf *.
This patch covers getsockname, getpeername and accept.

The only part of the patch I think needs help is what to do
with the ktrkuser(mbuftypes[MT_SONAME], addr, len) call from
copyout_sockname_sb(). Obviously mbuftypes[] makes no real
sense and addr is not in alias of an mbuf. What should be
done with the ktrace hook?

comments welcome

rtr

[Christos Zoulas]

-=-=-=-=-=-
hi,
attached is the next patch for converting protocol
user-request methods to use sockaddr * instead of mbuf *.
This patch covers getsockname, getpeername and accept.
The only part of the patch I think needs help is what to do
with the ktrkuser(mbuftypes[MT_SONAME], addr, len) call from
copyout_sockname_sb(). Obviously mbuftypes[] makes no real
sense and addr is not in alias of an mbuf. What should be
done with the ktrace hook?
comments welcome

Are you getting rid of MT_SONAME? You can just change the string
there to be "sockaddr"... It is just a string annotating what the
data you are placing in the trace are.

christos


--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de

[Tyler Retzlaff]

Post by Christos Zoulas

-=-=-=-=-=-
hi,

Are you getting rid of MT_SONAME? You can just change the string
there to be "sockaddr"... It is just a string annotating what the
data you are placing in the trace are.

MT_SONAME is still sprinkled about in other places but when the work is
complete it may be possible to eliminate it.

If just using "sockaddr" in place for now is acceptable then that that's
what I'll do.

thanks

--
Posted automagically by a mail2news gateway at muc.de e.V.
Please direct questions, flames, donations, etc. to news-***@muc.de