Date: 23 Nov 1998 01:11:12 -0600 From: Joel Ray Holveck <joelh@gnu.org> To: "Justin T. Gibbs" <gibbs@narnia.plutotech.com> Cc: current@FreeBSD.ORG Subject: Re: 3.0-RELEASE Feedback Message-ID: <864srq6fzj.fsf@detlev.UUCP> In-Reply-To: "Justin T. Gibbs"'s message of "Sun, 22 Nov 1998 16:54:07 -0700 (MST)" References: <199811222354.QAA08722@narnia.plutotech.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>> The CAM stuff is handled in async parallel. >>> Should these be completing before certain system boot-up scripts are >>> processed (rc.local, etc.)? >> If you want them to. It'll probably be semi-trivial to make this >> happen. > CAM probes for all devices prior to proceeding with the boot. What it > doesn't wait for is for all peripheral attachments to complete > initialization. Peripheral instances that have not completed initialation > will block in 'open' until the initialization is complete. With this > strategy, the system will only wait for the devices it needs to during > the boot process. This is a feature. Agreed. > The easiest way to make the system 'wait' for your cdrom drive is to > leave a disc in it and add have the system mount the drive at boot > time. If I left disks in the drive to begin with, then the probes would complete before bootup. However, I have no desire to do this. >> I'm considering making a change so that cd's are not probed for size. > Why. It's one of the ways the system ensures it can successfully talk > to the device. To keep a slightly more orderly startup. Rather than have a CD-ROM dump a "Not Ready" error code to the login prompt, and miss all mention of said CD drive in dmesg.boot, I for one would rather use a different method. If I'm reading right, we've already sent the CD-ROM a Test Unit Ready, two Inquiries, and a Mode Sense before sending a read capacity command. Are there likely to be CD-ROM failures in which these others all succeed but the Read Capacity fails with a sense code other than Medium Not Present or Logical Unit Not Ready (both of which are considered successful probes by the CD probe code)? Happy hacking, joelh -- Joel Ray Holveck - joelh@gnu.org Fourth law of programming: Anything that can go wrong wi sendmail: segmentation violation - core dumped To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?864srq6fzj.fsf>