Date: Mon, 10 Jan 2011 16:02:52 -0600 From: Tom Judge <tom@tomjudge.com> To: Luigi Rizzo <rizzo@iet.unipi.it> Cc: freebsd-hackers@freebsd.org, luigi@freebsd.org Subject: Re: sys/boot/boot0/boot0.S - r186598 Message-ID: <4D2B820C.3050205@tomjudge.com> In-Reply-To: <20110109224401.GA95358@onelab2.iet.unipi.it> References: <4D295820.20807@tomjudge.com> <20110109183316.GA92631@onelab2.iet.unipi.it> <4D2A0514.7040305@tomjudge.com> <20110109224401.GA95358@onelab2.iet.unipi.it>
next in thread | previous in thread | raw e-mail | index | archive | help
On 01/09/2011 04:44 PM, Luigi Rizzo wrote: > On Sun, Jan 09, 2011 at 12:57:24PM -0600, Tom Judge wrote: >> On 09/01/2011 12:33, Luigi Rizzo wrote: >>> On Sun, Jan 09, 2011 at 12:39:28AM -0600, Tom Judge wrote: >>>> Hi, >>>> >>>> Today I ran into an issue where setting the default slice with boot0cfg >>>> -s is broken. >>> a few questions inline: >>> >> Output inline, full script log attached. >> >> If you need more info let me know. > > can you take a dump of the boot sector at various stages > indicated below: > Paths to the files inline. >> <snip> > > DUMP #1: ORIGINAL BOOT SECTOR http://www.tomjudge.com/tmp/boot0/file1 > > <run boot0cfg -s 2 -v ad0 > > > DUMP #2: AFTER THE BOOT SECTOR UPDATE http://www.tomjudge.com/tmp/boot0/file2 > > <reboot without pressing an F-key> > <expect to be in s2 but actually end up in s1, as you found> > > DUMP #3: AFTER A REBOOT WITH NO KEYPRESS http://www.tomjudge.com/tmp/boot0/file3 > > <reboot, this time selecting the slice with F2> > > DUMP #4: AFTER THE SUCCESSFUL BOOT IN SLICE 2 http://www.tomjudge.com/tmp/boot0/file4 > > At least from this we can tell how #4 differs from #2/#3 > Maybe we can update the update code so that it only writes an updated active flag if the active flag has changed? This would resolv the wear issue that caused this to change behavior. Tom -- TJU13-ARIN
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4D2B820C.3050205>