Peter Maydell <peter.mayd...@linaro.org> writes: > On 5 March 2015 at 19:24, Nikunj A Dadhania <nik...@linux.vnet.ibm.com> wrote: >> Rejection is also change of behaviour. Because till now, a VM would >> start with any memory size, even if it's less that 128MB >> (default_ram_size). With rejection, all those VMs would fail booting >> displaying the warning. Is this OK? > > No. Not all of the machines we emulate are modern machines with > gigabytes of memory -- some are very small boards which might > really only have 64K of RAM. If the user asks for 64K you should > do what they ask. > > If what you want is to reject user specified memory sizes which > are too small, this is a "minimum RAM size", which is different > from "default RAM size". It would also be nice to have a > "maximum RAM size", so we can avoid weird failures if the user > asks for 1GB on a board which only has 256MB of space for RAM > in its address map.
Yes, [min,max]_ram_size is more appropriate. At present, I have sent a v4 without changing the default behaviour when user has provided an option. > Somebody may be along shortly to complain that this doesn't account > for machines where you can only add RAM one DRAM stick at a time > and so 64MB, 128MB and 256MB might all be valid but 100MB not. Yes, and these would help memory hotplug as well. > At least, that's what happened a few years ago when I tried to > suggest something like these per-board properties... > > -- PMM Regards Nikunj
