Date: Thu, 21 Apr 2011 10:31:56 -0700 From: Garrett Cooper <yanegomi@gmail.com> To: Alexander Motin <mav@FreeBSD.org> Cc: FreeBSD-Current <freebsd-current@freebsd.org> Subject: Re: Switch from legacy ata(4) to CAM-based ATA Message-ID: <5998CCDE-CA96-4C37-8C03-3CC283DBB099@gmail.com> In-Reply-To: <4DAEAE1B.70207@FreeBSD.org> References: <4DAEAE1B.70207@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Apr 20, 2011, at 2:57 AM, Alexander Motin <mav@FreeBSD.org> wrote: > Hi. >=20 > With 9.0 release approaching quickly, I believe it the best time now to > manage migration from legacy ata(4) ATA to the new CAM-based one. New > ATA code present in the tree for more then a year now, used by many > people and proved it's superior functionality and reliability. The only > major issue with it now is the migration process. Sooner or later we > have to pass it, but due to major UI and API changes we can't do it > after 9.0 release. So I propose to do it the next Sunday (April 24) to > have as much time for troubleshooting as possible. >=20 > I have prepared the following patch to do it: > http://people.freebsd.org/~mav/ata_switch.patch >=20 > I haven't added geom_raid to the kernel configurations because we have > no other GEOM classes there. But tell me if you thing I should. >=20 > If somebody has any problems with new ATA stack, please repeat your > tests with latest HEAD code and contact me if problem is still there. > Next three weeks before BSDCan I am going to dedicate to fixing possibly Although this may not be a list of fixable issues, here are some observation= s (in part with the new geom raid infrastructure): 1. Channels are no longer fixed of course because ata uses cam now, and I be= lieve that device numbering is done based on probe ordering. This is fun to w= ork with when dealing with appliances or configurations that require determi= nistic probe and mount, especially when drives fail, go missing, etc, but ca= n be hacked around in device.hints. This is why it would be nice for geom la= bels to work in a sane manner. 2. When testing with USB thumb drives on 7.x with a custom kernel / sourceba= se, it was enumerating the thumbdrive as da0, the first sata drive as ada1, e= tc. This may not be true for vanilla FreeBSD or later versions of FreeBSD. 3. Directly mappable analogs between atacontrol and camcontrol don't exist i= n areas; I'll come up with a laundry list for this because I said I would ea= rlier via private email. 4. IIRC atacontrol and graid status reported different sets of raid volume s= tates, depending on the underlying controller, in particular they changed th= e textual name for "healthy" from "optimal" to "ready" somewhere around the I= CH7/8/9 era. Thanks again for all your hard work! -Garrett=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5998CCDE-CA96-4C37-8C03-3CC283DBB099>