Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 16 Jan 2013 02:34:29 +0100
From:      Marius Strobl <marius@alchemy.franken.de>
To:        Olivier =?iso-8859-1?Q?Cochard-Labb=E9?= <olivier@cochard.me>
Cc:        freebsd-sparc64@freebsd.org
Subject:   Re: How to retrieve the "bootpath" variable in a running FreeBSD ?
Message-ID:  <20130116013429.GA28437@alchemy.franken.de>
In-Reply-To: <CA%2Bq%2BTcoPDCZHP5xgRymEBMe3GpBPQir%2BLrOogQ6n1Ho9%2BM4oEA@mail.gmail.com>
References:  <CA%2Bq%2BTcrLq2tJd0ffLGgiswBkozs4T=pPwjby55B1FbD=pBE26A@mail.gmail.com> <20130113.075038.471841385206679419.hrs@allbsd.org> <CA%2Bq%2BTcoPDCZHP5xgRymEBMe3GpBPQir%2BLrOogQ6n1Ho9%2BM4oEA@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Jan 13, 2013 at 01:27:36AM +0100, Olivier Cochard-Labbé wrote:
> On Sat, Jan 12, 2013 at 11:50 PM, Hiroki Sato <hrs@freebsd.org> wrote:
> >
> >  Try:
> >
> >  % kenv currdev
> >
> 
> Thanks a lot's !
> 
> Now my eeprom is correctly updated, just a last bug to fix.
> After updating the eeprom from FreeBSD I need to issue a cold-reboot
> for eeprom changes being applied (poweroff+poweron or "boot" command
> in OBP): A simple "reboot" from SSH console is not enough.
> I don't know if it's my old Sun Blade 150 that have this problem or if
> it's a normal sparc behavior.
> 

Uhm, that's tricky. I haven't implemented that part but AFAICT, the
current behavior of rebooting from the previously used bootpath
rather than from what is specified via boot-device (now) is on
purpose and actually I make heavy use of it. Normally, I boot my
machines from disk set via boot-device, except for kernel development
which I typically use netbooting for. In order to do so, I interrupt
the boot process and boot via `boot net`. It's very convenient that
this sticks across reboots until one power cycles the machine or
explicitly boots from disk again without having to fiddle with
boot-device.
The behavior you are asking for can also be implemented but I can't
think of a proper logic for determining what to do when boot-device
and bootpath disagree and I don't want to get rid of the current one :)
If you need that alternate behavior the best way forward seems to
be to add a switch for it to reboot(8).

Marius




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20130116013429.GA28437>