Re: [PATCH v2 0/4] apic: Fix migration breakage of >255 vcpus

[Peter Xu <peterx@redhat.com>]

On Thu, Oct 24, 2019 at 01:49:11PM -0400, John Snow wrote:
> 
> 
> On 10/23/19 4:17 AM, Kevin Wolf wrote:
> > The important difference here is legacy IDE (which works) vs. AHCI
> > (which doesn't work). If you add a -device ahci to the -M pc case, it
> > starts failing, too.
> > 
> > Not sure why AHCI fails, but I'll just CC John who is the lucky
> > maintainer of this device. :-)
> 
> Hm... It looks like SeaBIOS is identifying the drive correctly and
> perfectly well, but we're failing at boot_disk(u8 bootdrv, int
> checksig), about here:
> 
> call16_int(0x13, &br);
> 
> if (br.flags & F_CF) {
>     printf("Boot failed: could not read the boot disk\n\n");
>     return;
> }
> 
> Looking at AHCI tracing (From the QEMU side), it looks like we set up
> the drive correctly, and then never touch the port ever again -- I don't
> see an attempted read on QEMU's end.
> 
> I'll need to look through SeaBIOS source for hints, I'm not sure right
> yet. If anyone is more familiar with the SeaBIOS boot code, maybe they
> can give a pointer faster than I'll figure it out myself.

Hi, John,

I don't know seabios well, but I did have a pointer in my previous
email on where it faulted.  It seems to me that the issue is that
SeaBIOS may have got incorrect secs/cyls/heads information (and
explicitly setting secs=1,cyls=1,heads=1 on the block device fixes the
issue).

Thanks,

-- 
Peter Xu