On 16/03/2015 13:18, Christian Borntraeger wrote: > Its up to a page and the content depends on the function code, selector1 > and selector2. > See arch/s390/kernel/sysinfo.c for the details or zarchitecture principles of > opration the STSI instruction. > > > Some time ago we had patches that passed the guest name > and UUID into the kernel - then Alex Graf asked, why we cant do that in > userspace. > https://lists.nongnu.org/archive/html/qemu-devel/2014-04/msg00094.html > > Now, stsi also gives the bare information as provides by upper > layers: e.g. 2.2.2 gives basic LPAR information. When doing that under > KVM or z/VM, KVM or z/VM will pass that information along to the underlying > LPAR hypervisor and give the result back to the guest. > > So I came up with the idea to let KVM do what it needs but also allow > QEMU to override (3.2.2 - this 2nd level hypervisor) If it's up to a page it definitely makes sense to do it this way. Thanks, Paolo -- 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
