Date: Mon, 5 Jun 2006 11:34:02 -0600 From: "Kenneth D. Merry" <ken@kdm.org> To: Matt Jacob <mjacob@freebsd.org> Cc: scsi@freebsd.org Subject: Re: changes to cam_xpt to support sequential scanning (!parallel) Message-ID: <20060605173402.GA23444@nargothrond.kdm.org> In-Reply-To: <200606040305.k5435Nm3083345@freefall.freebsd.org> References: <200606040305.k5435Nm3083345@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jun 04, 2006 at 03:05:23 +0000, Matt Jacob wrote: > > One solution that Scott suggested to some of the negotiation issues for > the MPT was to not do parallel problem, i.e., all targets in parallel. > > The patch at http://people.freebsd.org/~mjacob/CAM_PATCHES_FOR_SEQSCAN > > does just that and I've had quite good success with it so far. I add > a new flat, PIM_SEQSCAN, which cases xpt_scan_bus to *not* scan all > targets in parallel but rather to scan them sequentially. > > Comments? Sounds like a good idea to me. It'll allow us to also do the same thing for any other devices that come along that can't handle the parallel probe. I took a brief look through it, but I haven't reviewed it in detail. It looks like there are a few whitespace type changes. > The code also does some cleanups under failure conditions. > > After a bit more testing I'd like to put this in if there are no > objections so I can get the mpt U320 stuff off my back for a while. I would say go for it if things seem to work well. (I'm assuming you've also tested it with an Adaptec, Qlogic or some other controller that works with the parallel scan?) Ken -- Kenneth Merry ken@kdm.org
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060605173402.GA23444>