Paolo, Radim, here is the first shot of nested virtualization for kvm/s390. It requires an updated qemu[1] and the nested module parameter to work. We are able to start multiple nested guests and are also able to run nested under nested. As there are certainly still some rough edges, the default of the nested paramter is off. [1] https://github.com/cohuck/qemu/tree/cpumodel-s390x This request has 1. a change in include/linux/page_ref.h (with an ack from Andrew) 2. several changes in s390 base memory managedment (patches are either from Martin or acked-by martin) 3. a potential conflict with Paolos "speed up context-tracking system calls" series. The following changes since commit 64672c95ea4c2f7096e519e826076867e8ef0938: kvm: vmx: hook preemption timer support (2016-06-16 10:07:50 +0200) are available in the git repository at: git://git.kernel.org/pub/scm/linux/kernel/git/kvms390/linux.git tags/kvm-s390-next-4.8-2 for you to fetch changes up to a411edf1320ed8fa3b4560902ac4e033c4a72bcf: KVM: s390: vsie: add module parameter "nested" (2016-06-21 09:43:47 +0200) ---------------------------------------------------------------- KVM: s390: vSIE (nested virtualization) feature for 4.8 (kvm/next) With an updated QEMU this allows to create nested KVM guests (KVM under KVM) on s390. s390 memory management changes from Martin Schwidefsky or acked by Martin. One common code memory management change (pageref) acked by Andrew Morton. The feature has to be enabled with the nested medule parameter. ---------------------------------------------------------------- David Hildenbrand (45): s390/mm: flush tlb of shadows in all situations s390/mm: shadow pages with real guest requested protection s390/mm: avoid races on region/segment/page table shadowing s390/mm: fix races on gmap_shadow creation s390/mm: take the mmap_sem in kvm_s390_shadow_fault() s390/mm: protection exceptions are corrrectly shadowed s390/mm: take ipte_lock during shadow faults s390/mm: push ste protection down to shadow pte s390/mm: prepare for EDAT1/EDAT2 support in gmap shadow s390/mm: support EDAT1 for gmap shadows s390/mm: support EDAT2 for gmap shadows s390/mm: push rte protection down to shadow pte s390/mm: support real-space for gmap shadows s390/mm: limit number of real-space gmap shadows s390/mm: remember the int code for the last gmap fault s390/mm: allow to check if a gmap shadow is valid s390/mm: don't fault everything in read-write in gmap_pte_op_fixup() KVM: s390: fast path for shadow gmaps in gmap notifier KVM: s390: backup the currently enabled gmap when scheduled out s390: introduce page_to_virt() and pfn_to_virt() mm/page_ref: introduce page_ref_inc_return KVM: s390: vsie: initial support for nested virtualization KVM: s390: vsie: optimize gmap prefix mapping KVM: s390: vsie: support setting the ibc KVM: s390: vsie: support edat1 / edat2 KVM: s390: vsie: support host-protection-interruption KVM: s390: vsie: support STFLE interpretation KVM: s390: vsie: support aes dea wrapping keys KVM: s390: vsie: support transactional execution KVM: s390: vsie: support vectory facility (SIMD) KVM: s390: vsie: support run-time-instrumentation KVM: s390: vsie: support 64-bit-SCAO KVM: s390: vsie: support shared IPTE-interlock facility KVM: s390: vsie: support guest-PER-enhancement KVM: s390: vsie: support guest-storage-limit-suppression KVM: s390: vsie: support intervention-bypass KVM: s390: vsie: support conditional-external-interception KVM: s390: vsie: support IBS interpretation KVM: s390: vsie: try to refault after a reported fault to g2 KVM: s390: vsie: speed up VCPU irq delivery when handling vsie KVM: s390: don't use CPUSTAT_WAIT to detect if a VCPU is idle KVM: s390: vsie: speed up VCPU external calls KVM: s390: vsie: correctly set and handle guest TOD KVM: s390: vsie: add indication for future features KVM: s390: vsie: add module parameter "nested" Martin Schwidefsky (6): s390/kvm: page table invalidation notifier s390/mm: use RCU for gmap notifier list and the per-mm gmap list s390/mm: extended gmap pte notifier s390/mm: add reference counter to gmap structure s390/mm: add shadow gmap support s390/mm: add kvm shadow fault function arch/s390/include/asm/gmap.h | 82 +- arch/s390/include/asm/kvm_host.h | 21 +- arch/s390/include/asm/mmu.h | 11 +- arch/s390/include/asm/mmu_context.h | 3 +- arch/s390/include/asm/page.h | 2 + arch/s390/include/asm/pgalloc.h | 2 + arch/s390/include/asm/pgtable.h | 10 +- arch/s390/include/asm/processor.h | 2 + arch/s390/include/uapi/asm/kvm.h | 12 + arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/gaccess.c | 239 ++++++ arch/s390/kvm/gaccess.h | 3 + arch/s390/kvm/interrupt.c | 5 + arch/s390/kvm/kvm-s390.c | 111 ++- arch/s390/kvm/kvm-s390.h | 10 +- arch/s390/kvm/priv.c | 1 + arch/s390/kvm/sigp.c | 10 +- arch/s390/kvm/vsie.c | 1091 ++++++++++++++++++++++++ arch/s390/mm/fault.c | 2 + arch/s390/mm/gmap.c | 1590 +++++++++++++++++++++++++++++++++-- arch/s390/mm/pgalloc.c | 39 +- arch/s390/mm/pgtable.c | 103 ++- include/linux/page_ref.h | 9 + 23 files changed, 3219 insertions(+), 141 deletions(-) create mode 100644 arch/s390/kvm/vsie.c -- 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
