Am 01.07.2013 12:18, schrieb Hu Tao: > and split off memory region initialization into instance_init. > > Signed-off-by: Hu Tao <hu...@cn.fujitsu.com> > --- > hw/pci-host/q35.c | 29 +++++++++++++++-------------- > 1 file changed, 15 insertions(+), 14 deletions(-)
Thanks, rebased and applied to qom-next: https://github.com/afaerber/qemu-cpu/commits/qom-next As mentioned, pci_bus_new() in realize should be replaced with pci_bus_new_inplace() in instance_init. However investigating that, I noticed that either function registers VMState, which is supposed to be done at realize time. This being a system chipset device, I think we can get away with that (same as for PReP), but I'll do it in a follow-up so that the blame goes on me. I think the long-term solution would be extending our realize/unrealize mechanisms to BusClass. Regards, Andreas -- SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Jeff Hawn, Jennifer Guild, Felix Imendörffer; HRB 16746 AG Nürnberg
