Date: Fri, 08 Jan 2010 17:56:40 -0700 (MST) From: "M. Warner Losh" <imp@bsdimp.com> To: jhb@FreeBSD.org Cc: svn-src-head@FreeBSD.org, tuexen@FreeBSD.org, svn-src-all@FreeBSD.org, src-committers@FreeBSD.org, rizzo@iet.unipi.it Subject: Re: moving sctp to a separate directory ? Message-ID: <20100108.175640.1104512900458971844.imp@bsdimp.com> In-Reply-To: <201001080812.21124.jhb@freebsd.org> References: <20100107203536.GB8230@rincewind.paeps.cx> <20100107214334.GA35184@onelab2.iet.unipi.it> <201001080812.21124.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In message: <201001080812.21124.jhb@freebsd.org>
John Baldwin <jhb@freebsd.org> writes:
: On Thursday 07 January 2010 4:43:34 pm Luigi Rizzo wrote:
: > > What do you do with udp, for instance? Compared to tcp and sctp, it's
: > > trivial in terms of code, but it's an upper layer protocol from the
: > > perspective of netinet/netinet6 - do we put it in its own directory too?
: > > Also note that this won't only cause churn for people who have patches against
: > > or (out-of-tree) branches from netinet/, but also in other kernel subsystems
: > > which rely on tcp -- nfs, for instance.
: >
: > + i find the concern about churn in external patchsets a bit weak, first of
: > all because this is bound to happen unless we stop all development,
: > and secondly because this kind of file moving or splitting happens
: > once every 10-15 years which is well beyond the lifetime of a patchset.
:
: Having the files rename is entirely different from merging changes. At least
: for svn and p4 I believe that merging a rename into a branch is not smart
: enough to merge your local changes into the new files. Instead it involves a
: big manual fixup.
:
: Also, the 10-15 years thing is completely non-relevant. What is relevant is
: if you are working on a project in a branch and someone renames files before
: you have finished your branch and merged it up to HEAD. For example, assume
: that someone else renamed the ipfw files in HEAD next week. That would
: create an utter mess for you to resolve in your current ipfw3 branch. Moving
: TCP would create similar a headache, except much more widespread since TCP is
: one of the most widely worked-on subsystems.
:
: FWIW, I do think it would be cleaner to have netinet more split up perhaps,
: but I do not think it is worth the pain that would be involved.
It is painful enough moving drivers around. I think that while well
intentioned, it will cause us nothing but grief.
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100108.175640.1104512900458971844.imp>