Re: [PATCH qemu] spapr: Force 32bit when resetting a core

[Cédric Le Goater]

On 1/15/22 15:21, Mark Cave-Ayland wrote:
> On 14/01/2022 14:12, Cédric Le Goater wrote:
>
> > Yes, more info here :
> >
> > https://patchwork.kernel.org/project/qemu-devel/patch/1458121432-2855-1-git-send-email-lvivier@redhat.com/
> > mac99+970 only boots with a 64bit kernel. 32bit are not supported because
> > of the use of the rfi instruction which was removed in v2.01. 32bit user
> > space is supported though.
> >
> > However I was not able to build a disk with a compatible boot partition
> > for OpenBIOS. The above support only applies for kernel loaded in memory.
> > May be Mark knows how to do this ?
>
> The Mac machines generally require a HFS filesystem for booting with OpenBIOS: 
> it's a bit convoluted, but some instructions for grub can be found at 
> https://wiki.gentoo.org/wiki/GRUB_on_Open_Firmware_(PowerPC).

Initially, I installed a debian11 ppc64 on a QEMU mac99/970 machine.
Something went wrong with the bootloader at installation and I was
stuck with memory boot. I didn't manage to restore a decent boot
setup even after that.

I stole the HFS partition from a real G5 and after some manual tweaks
on the disk, I got it working.

There are a few harmless errors :
    
    >> =============================================================
    >> OpenBIOS 1.1 [Jan 15 2022 14:23]
    >> Configuration device id QEMU version 1 machine id 3
    >> CPUs: 1
    >> Memory: 1024M
    >> UUID: 00000000-0000-0000-0000-000000000000
    >> CPU type PowerPC,970
    milliseconds isn't unique.
    Welcome to OpenBIOS v1.1 built on Jan 15 2022 14:23
    Trying hd:,\\:tbxi...
    >> switching to new context:
    call-method color!: exception -21
    >> call-method color! failed with error ffffffdf
    call-method color!: exception -21
    
    ...
    call-method block-size: exception -21
    >> call-method block-size failed with error ffffffdf
    call-method block-size: exception -21
    >> call-method block-size failed with error ffffffdf
    

I guess the initial problem is in the debian installer. I will report.

> I can certainly help out if you get stuck. If this is a more obscure combination 
> then is it worth adding a tiny image for use with avocado?

No. It is really a standard setup :

    root@vm02:~# mac-fdisk -l /dev/sda
    /dev/sda
            #                    type name                length   base    ( 
size )  system
    /dev/sda1     Apple_partition_map Apple                   63 @ 1       ( 
31.5k)  Partition map
    /dev/sda2               Apple_HFS bootstrap           500001 @ 64      
(244.1M)  HFS
    /dev/sda3         Apple_UNIX_SVR2 untitled           39634766 @ 500065  ( 
18.9G)  Linux native
    /dev/sda4         Apple_UNIX_SVR2 swap               1808208 @ 40134831 
(882.9M)  Linux swap
    /dev/sda5              Apple_Free Extra                    1 @ 41943039 (  
0.5k)  Free space
    
    Block size=512, Number of Blocks=41943040
    DeviceType=0x0, DeviceId=0x0
    
    root@vm02:~# cat /proc/cpuinfo
    processor   : 0
    cpu         : PPC970, altivec supported
    clock               : 900.000000MHz
    revision    : 2.2 (pvr 0039 0202)
    
    timebase	: 100000000
    platform    : PowerMac
    model               : PowerMac3,1
    machine             : PowerMac3,1
    motherboard : PowerMac3,1 MacRISC MacRISC2 Power Macintosh
    detected as : 0 ((null))
    pmac flags  : 00000000
    pmac-generation     : NewWorld
    root@vm02:~# uname -a
    Linux vm02 5.15.0-2-powerpc64 #1 SMP Debian 5.15.5-2 (2021-12-18) ppc64 
GNU/Linux
 

Thanks,

C.