Date: Tue, 26 Feb 2008 10:39:58 -0800 From: Julian Elischer <julian@elischer.org> To: Andre Oppermann <andre@freebsd.org> Cc: Marko Zec <zec@freebsd.org>, Brooks Davis <brooks@freebsd.org>, Marko Zec <zec@imunes.net>, FreeBSD Current <current@freebsd.org>, Marko Zec <zec@icir.org> Subject: Re: warning of pending commit attempt. Message-ID: <47C45CFE.9030006@elischer.org> In-Reply-To: <47C458E9.5090400@freebsd.org> References: <47C39948.3080907@elischer.org> <20080226051346.GA65258@lor.one-eyed-alien.net> <47C458E9.5090400@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Andre Oppermann wrote: > Brooks Davis wrote: >> On Mon, Feb 25, 2008 at 08:44:56PM -0800, Julian Elischer wrote: >> >>> At some stage in the next few weeks I will be trying to commit >>> Marco Zec's vimage code to -current. (only 'trying' not >>> for technical reasons, but political). > ... >>> Why now? >>> The code is in a shape where teh compiled out version of hte system >>> is stable. In the compiled in version, it is functional >>> enough to provide nearly all of what people want. It needs people >>> with other interests to adapt it to their purposes and use it so that >>> it can become a solid product for future releases. >> >> The website has a snapshot with a date over a month old and many >> comments about unstable interfaces. I've seen zero reports of >> substantial testing... > > What about locking and SMP scalability? Any new choke points? not that I've seen. > > Having a patch set to glance over would be helpful as well. It's all in p4 //depot/projects/vimage/src > >>> Solaris and Linux have seen what BSD can do with jails and have upped >>> the ante. it's time for FreeBSD to tak our jails to teh next logical >>> step. >>> >>> As it will be committed it does have some missing parts to the >>> jigsaw, but it is complete enough that a system compiled in this >>> manner can >>> be fully functional and fully backwards compatible. >>> >>> Basically no userland changes need be made to get the full effect. >>> >>> I expect the usual nay-sayers no matter what is proposed, but >>> I hope we can have a decent discussion about this.. >> >> From purely procedural perspective, the "next few weeks" seems rushed and >> poorly motivated. We're still finding and fixing bugs from the last >> major round of network changes. We should at least get the first batch >> of 7.0 errata out the door before making changes that will certain make >> merging non-trivial network stack changes more difficult. We also need >> credible, qualitative reports verifying that it works and what it's >> impacts are. > > Seconded. Waiting until 7.0 has left the barn and got some exposure would > be preferrable. > >> Don't get me wrong. I think this is interesting work and that it could >> be a major asset to FreeBSD. I also recognize that it should go in >> in the next 6-9 months (12 at the outside) if it's not going to cause >> problems with 8.0. I simply don't see any valid motivation for doing it >> with undue haste. > > We also need detailed briefing on how to work with virtualization. What > have we to be careful about? How are kld's and network drivers affected? > New network related kernel structures? And so on. Make sure you read the paper(s) too.. There is some documentation required on "how to do it", especially with the new macros. How and when to use them, and how a kernel module needs to be made aware of the framwork. I think Marco told me once that he was going to do such documentation, so your reminder will probably produce some sort of "developer's notes" I hope. I'm sure that when it gets into the system a lot of people will want to play with it. I have only virtualised one module (divert sockets) and I still have a question about one thing so I understand the need for notes on how to develope for vimage. >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?47C45CFE.9030006>