Date: Fri, 6 Feb 1998 10:30:05 -0800 (PST) From: Tom <tom@sdf.com> To: Terry Lambert <tlambert@primenet.com> Cc: John Fieber <jfieber@indiana.edu>, calin@ibd.dbio.ro, freebsd-hackers@FreeBSD.ORG Subject: Re: Powering off the system/UPS Message-ID: <Pine.BSF.3.95q.980206102658.7371A-100000@misery.sdf.com> In-Reply-To: <199802061839.LAA12897@usr01.primenet.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 6 Feb 1998, Terry Lambert wrote: > > > I followed all your mail about how to patch the kernel. I seems that the > > > solution that would satisfy (more or less) everybody would be an extra > > > list for at_shutdown where both apm and ups drivers can insert callbacks. > > > (the way I patched my system's kernel - haha). > > > > Maybe I missed it in the thread, but why it is a Good Thing to > > put UPS support in the kernel as opposed to a purely user mode > > daemon that monitors the UPS and does the Right Thing, like > > shutdown(8), when things get to a critical state? > > Because the kernel knows about uncommitted write buffers, but the > user space daemon does not, and the user space daemon might be > fighting for a quantum with 1000 other processes, so it can't > react before its too late to save your data. Not even an issue. You want to start the shutdown well in advance anyhow, because battery life is a bit variable. The biggest problem is that a usermode program loses control as soon as it halts the system. What happens if power is resumed? The system is limbo with "Press any key to reboot" on the screen. A smarter UPS can help, but adding intelligence to the OS is easier. > Terry Lambert > terry@lambert.org > --- > Any opinions in this posting are my own and not those of my present > or previous employers. Tom
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.95q.980206102658.7371A-100000>