Re: boot.cfg

["Ernesto Bascon" <ebasconp%gmail.com@localhost>]

Hi, I am testing it in a VMware Workstation (I dunno if it is a valid
test environment for Xen); with your last boot(8) the Xen kernel is
loaded but it shows:

(XEN) Dom0 has maximum 1 VCPUs
(XEN) Scrubbing Free RAM: .done.
(XEN) Xen trace buffers: disabled
(XEN) Std. Loglevel: Errors and warnings
(XEN) Guest Loglevel: Nothing (Rate-limited: Errors and warnings)
(XEN) Xen is relinquishing VGA console

After that, the NetBSD kernel starts to load quietly but the control
is never returned to the final user (actually I can ping my NetBSD
machine but I have no access it via my running sshd).



On 16/09/2008, Robert Swindells <rjs%fdy2.demon.co.uk@localhost> wrote:
>
> Manuel Bouyer wrote:
>>On Tue, Sep 16, 2008 at 05:15:10PM +0100, Robert Swindells wrote:
>>> >Are you seeing some output on the serial console ?
>>>
>>> The two laptops that I have been using to test this don't have serial
>>> ports. I can try on another system though.
>>>
>>> If I don't specify a console then I get Xen output up to the line that
>>> reads "Xen is relinquishing VGA console.", then nothing until the login
>>> prompt appears.
>
>>OK, so Xen uses both the serial and VGA console, and NetBSD thinks
>>it's a serial console.
>
>>> With "console=pc" nothing between boot(8) messages and
>>> the login prompt.
>
>>So both thinks it's a serial console. Do you pass "console=pc" to the Xen
>>kernel, or the NetBSD kernel ?
>>For a VGA console (VGA only, no output on serial console at all) you need:
>>console=vga
>>passed to xen.gz, and
>>console=pc
>>or
>>console=tty0
>>passed to the NetBSD kernel.
>
> That was it, I was missing the console=vga option to the Xen kernel.
>
> The following will now work as a boot.cfg option:
>
> menu=Boot Xen:modules enabled;load /netbsd-DOM0;multiboot xen.gz
> dom0_mem=65536 console=vga -- bootdev=wd0a ro console=pc
>
> If anyone wants to try it out there is a copy of boot(8) with the changes
> at:
>
> <ftp://ftp.netbsd.org/pub/NetBSD/misc/rjs/boot>
>
> I will generate the diffs for review tomorrow.
>
> Robert Swindells
>