Date: Wed, 21 Nov 2012 11:15:45 -0700 (MST) From: Warren Block <wblock@wonkity.com> To: Arthur Chance <freebsd@qeng-ho.org> Cc: freebsd-questions@freebsd.org Subject: Re: gpt booting (Was: Re: boot problem after freebsd-update from 9.1-RC2 to 9.1-RC3) Message-ID: <alpine.BSF.2.00.1211211108520.7028@wonkity.com> In-Reply-To: <50AC9409.1030609@qeng-ho.org> References: <50ABE97B.5030408@johnea.net> <50AC041A.7050607@dreamchaser.org> <50AC3116.6010600@johnea.net> <alpine.BSF.2.00.1211202201460.42121@wonkity.com> <50AC9409.1030609@qeng-ho.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 21 Nov 2012, Arthur Chance wrote: > On 11/21/12 05:11, Warren Block wrote: >> gptboot looks for the first UFS partition. Maybe /boot/boot can be >> modified to do that also. > > It's a little more complicated than that Warren. > > AIUI gptboot first looks (in partition order) for partitions with both the > bootme and bootonce attributes set. If it doesn't find any, or if they all > failed to boot it then tries booting partitions with just the bootme > attribute. It only boots the first UFS partition if no partitions have the > bootme attribute set, and IIRC that is for compatibility with the 8.x gptboot > which didn't know the boot* attributes. > > Confusingly, there's no manual page for gptboot to document this. It's sort > of implicit in the gpart manual page, in the section on ATTRIBUTES for GPT, > but the best way to understand it is to read the code for gptfind in > > /usr/src/sys/boot/common/gpt.c Well, yes. The point is that gptboot doesn't just assume that p2, say, is where the bootable UFS partition must be. I've also noted the lack of a gptboot man page, and it's on my long list of Things That Should Be Done. There was a thread on -doc: http://lists.freebsd.org/pipermail/freebsd-doc/2012-June/020060.html Help would be greatly appreciated.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?alpine.BSF.2.00.1211211108520.7028>