Re: [Qemu-devel] [PATCH 2/5] exec: qemu_ram_alloc_device, qemu_ram_resize

[Paolo Bonzini <pbonzini@redhat.com>]

On 20/11/2014 15:04, Michael S. Tsirkin wrote:
> On Thu, Nov 20, 2014 at 02:35:14PM +0100, Markus Armbruster wrote:
>> What am I missing here that can justify the complexity of partially
>> overriding target configuration in the migration stream plus
>> infrastructure for resizing memory?
> 
> The justification is that sizing it properly is an unsolved problem.

Not really: sizing it properly is a tedious thing to do, but it is not a
problem at all.  We _almost_ get it right, only the DSDT can change from
one version to the next right now.  And patches exist to pad the DSDT
adequately; those patches are simpler than these ones.

I am okay with considering it "too tedious to spend more time on it",
especially because the size of the ACPI tables already changed
arbitrarily from one version to the other when it was computed in the
firmware.  On the other hand, one could also say that being able to size
tables properly is an _advantage_ of doing tables in QEMU.

I was expecting little opposition and thus thinking it was not
worthwhile to discuss the approach to take.  Apparently there _is_
opposition, so I think we should reconsider my patches.

Paolo