On Mon, 25 Jun 2001, Jaromír Dolecek wrote:

# Bill Sommerfeld wrote:
# > > Log Message:
# > > Add 'kernsa' parameter for sendit()/recvit(); if nonzero, msg->msg_name
# > > is supposed to point directly to struct mbuf or struct sockaddr in kernel
# > > space as appropriate, rather than being a pointer to memory in userland.
# > > 
# > > This is to be used by compat/* when emulation needs to wrap
# > > send{to|msg}(2)/recv{from|msg}(2) and modify the passed struct
# > > sockaddr.
# > Who reviewed this change?
# 
# This change didn't seem like one needing review.

Begging pardon, but this is a bright orange flag in my mind.  I'm not one
qualified to make kernel judgment calls due to lack of experience, but
just on general procedure, and knowing the project's charter and goals,
this attitude seems way out of line with reality.

Yes, it slows our implementation cycle down a bit, but:  ALL CHANGES SHOULD
BE REVIEWED. This is from experience -- failure to do peer review in the
workplace has gotten me in the most employment-wise fatal of trouble.

It doesn't matter how small or innocuous the change may seem, and I can't
demand it, but I would urge that the changes be peer reviewed prior to
implementation.

# E.g. MSG_COMPAT handling costs more than this.

It might seem a bit extreme, but have you done profiling on this to verify
it, or are you judging by code and feel?

				--*greywolf;
--
NetBSD: "Progress on your system is closer than it appears."