On Tue, 16 Jul 2013 15:27:19 +0800 Hu Tao <hu...@cn.fujitsu.com> wrote:
> On Tue, Jul 16, 2013 at 08:19:48AM +0200, Paolo Bonzini wrote: > > Il 16/07/2013 03:27, Hu Tao ha scritto: > > > > I think it's the same. One "-numa mem" option = one "-device dimm" > > > > option; both define one range. Unused memory ranges may remain if you > > > > stumble upon a unusable range such as the PCI window. For example two > > > > "-numa mem,size=2G" options would allocate memory from 0 to 2G and from > > > > 4 to 6G. > > > > > > So we can drop -dimm if we agree on -numa mem? > > > > Yes, the point of the "-numa mem" proposal was to avoid the concept of a > > "partially initialized device" that you had for DIMMs. > > > > BTW, how do you specify which module you are plugging in? I.e., what if > > I have three 1G ranges at 0, 1G and 2G, and I want to plug the first and > > the third? > > I think an id is still needed to identify ranges, which can be shown to > user with `info numa' or similar command, along with the corresponding > ranges. "info numa" could get actual ranges enumerating present dimms if we would use plain -device for hotplug memory instead of -numa mem hacking. > > > > > This is especially important with hot-unplug, because then you can have > > this kind of hole in the address space. If you migrate the VM, you have > > to reproduce the situation in the destination command line. > > > > Paolo > > >
