Re: [Qemu-devel] [PATCH v3 0/7] Let boards state maximum RAM limits in QEMUMachine struct

[Peter Maydell]

On 30 March 2011 08:48, Jes Sorensen <address@hidden> wrote:
> On 03/29/11 16:08, Peter Maydell wrote:
>> This primary aim of this patchset is to add a new 'max_ram' field to the
>> QEMUMachine structure so that a board model can specify the maximum RAM it
>> will accept.

> I am a little concerned about this approach. It should work for simple
> embedded boards, but for larger systems, it really ought to be a mask
> rather than a max address.

It's not a maximum address, it's a maximum size. For instance
the RAM isn't contiguous on some of the ARM devboards.

> Ideally I think it would be better to have a mask and then introduce a
> is_valid_memory() kinda function to check it with.

The command line option doesn't provide any means of saying
"put 64MB in this hole and another 128 over here and 32 there",
so this seems completely pointless to me. All we are trying
to do is validate what the user has asked for, so why have
a validation mechanism that can cope with impossible-to-request
arrangements?

> I fear that by introducing a simple max limit like this, we are going to
> hit problems later when we try to improve the NUMA support.

I think this is letting the best be the enemy of the good.

Even if you do want to have NUMA systems do more complex
things I think you should still have the simple "maximum
size" approach for the bulk of the supported boards which
don't need anything more complicated. So additional NUMA
features would augment, not replace this.

-- PMM