On 15.05.2012, at 14:38, Christian Borntraeger wrote: > On 15/05/12 14:33, Avi Kivity wrote: >> On 05/15/2012 03:31 PM, Christian Borntraeger wrote: >>> On 15/05/12 14:24, Avi Kivity wrote: >>>>> Newer systems allow to write-protect the guest backing memory >>>>> and let the fault be delivered to the host, thus allowing COW. >>>>> >>>>> Use a capability bit to tell qemu if that is possible. >>>>> >>>> >>>> Asking out of ignorance: who is doing the write protection here? The >>>> guest? If so, why is qemu involved? >>> >>> It is about the host doing write protection of guest/user memory (e.g. for >>> dirty pages tracking or KSM) >> >> Ok, so why does qemu^Wuserspace need to know about it? > > By default qemu will use MAP_PRIVATE for guest pages. This will write > protect pages and thus break on s390 systems that dont support this feature. > Therefore qemu has a hack to always use MAP_SHARED for s390. But MAP_SHARED > has other problems (no dirty pages tracking, a lot more swap overhead etc.) > With this feature qemu can use the standard qemu alloc if available, otherwise > it will use the old s390 hack. I will put you on cc for the qemu patch :-) It'd be nice to have that documented somewhere :). Alex -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
