Date: Sat, 2 Mar 2013 01:53:18 +0100 From: Marius Strobl <marius@alchemy.franken.de> To: Chris Ross <cross+freebsd@distal.com> Cc: "freebsd-sparc64@freebsd.org" <freebsd-sparc64@freebsd.org> Subject: Re: Problems booting 9.1-STABLE on Netra X1 Message-ID: <20130302005318.GA22053@alchemy.franken.de> In-Reply-To: <5F00D05B-37BB-4342-AE56-C707B5267C17@distal.com> References: <CE371F2B-CF62-4695-A9F0-B56995BA3CC6@distal.com> <20130225101315.GA79064@alchemy.franken.de> <76C74932-5BB0-4194-86CE-F121F6D18D84@distal.com> <20130225214941.GD955@alchemy.franken.de> <3A37672E-B6F7-4D5E-8293-0ED3B203C358@distal.com> <20130226013400.GG955@alchemy.franken.de> <5F00D05B-37BB-4342-AE56-C707B5267C17@distal.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Feb 25, 2013 at 09:01:16PM -0500, Chris Ross wrote: > > On Feb 25, 2013, at 20:34 , Marius Strobl <marius@alchemy.franken.de> wrote: > >> It's clearly low-end hardware, and I could believe that it may be a "affects > >> some but not all systems". But, assuming it's more than just _one_ bogus > >> system, finding a way to deal with it at HEAD makes sense. Not necessarily > >> that the earlier code is the only or best solution. > > > > Will do, I have to think about this some more though. > > That's fine. No problem. I completely understand, and am working on my own > for my immediate needs. > Craig Butler confirmed my impression that generally there's no downside from registering a handler for the power fail interrupt on X1 (or V100 for that matter). So I did what I initially thought would be the best option if the affected machine(s) indeed run stable despite this issue and added a loader tunable in r247600. By setting hw.psycho.powerfail to 0 you can noe disable setting up the interrupt handler. Marius
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130302005318.GA22053>