Date: Fri, 4 Jan 2002 08:53:07 -0800 (PST) From: Doug Ambrisko <ambrisko@ambrisko.com> To: John Baldwin <jhb@FreeBSD.org> Cc: cvs-all@FreeBSD.org, cvs-committers@FreeBSD.org Subject: Re: cvs commit: src/sys/boot/i386/libi386 bootinfo.c Message-ID: <200201041653.g04Gr7x27489@ambrisko.com> In-Reply-To: <XFMail.020104014417.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin writes: | On 04-Jan-02 John Baldwin wrote: | > jhb 2002/01/03 23:59:27 PST | > | > Modified files: | > sys/boot/i386/libi386 bootinfo.c | > Log: | > Don't turn on RB_CDROM if the rootdev is a BIOS cd device for now as this | > breaks when cdboot is used with an MFS root. | | This isn't really right. The problem is that for an MFS root, we leave the | root dev in the loader pointing to whatever, and RB_CDROM takes precedence over | MD_ROOT. A more proper fix might be to allow the rootdev to be set to "mfs" or | "md0" or some such and have the loader allow that. Then if the user leaves cd0 | as the rootdev, RB_CDROM would give them what they asked for, but if they | change rootdev to "md0", they get what they want. However, I think that | mfs_root should override RB_CDROM just like it overrides all the other root | devices. This would allow us to not have to teach the loader about fake "md0" | devices as a kludge. Instead, the logic would be to use mfs_root if we are | given one, otherwise fall back on what the loader hints to us. | | Comments? I agree that "mfs_root should override RB_CDROM" to be consistent and then re-enable RB_CDROM like you had it before. I need to play with this some more to see if there are some other quirks. Doug A. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200201041653.g04Gr7x27489>