On Wed, 2020-11-11 at 19:07 -0300, Daniel Henrique Barboza wrote: > +++ b/tests/qemuxml2argvdata/memory-hotplug-ppc64-nonuma.args > @@ -11,7 +11,7 @@ QEMU_AUDIO_DRV=none \ > -name QEMUGuest1 \ > -S \ > -machine pseries,accel=kvm,usb=off,dump-guest-core=off \ > --m size=1310720k,slots=16,maxmem=4194304k \ > +-m size=1048576k,slots=16,maxmem=4194304k \ > -realtime mlock=off \ > -smp 1,sockets=1,cores=1,threads=1 \ > -object memory-backend-ram,id=memdimm0,size=536870912 \ This doesn't look right: AFAIK the initial memory size is guest-visible, so by changing how the alignment is performed you might both change the guest ABI across guest boots (if libvirt is upgraded in between them) and break migration (if either the source or destination host is running the newer libvirt but the other side isn't). Did I miss something that makes this okay? -- Andrea Bolognani / Red Hat / Virtualization
