[PATCH v5 0/6] KVM: X86: Implement Exit-less IPIs support

[Wanpeng Li <kernellwp@gmail.com>]

Using hypercall to send IPIs by one vmexit instead of one by one for
xAPIC/x2APIC physical mode and one vmexit per-cluster for x2APIC cluster 
mode. Intel guest can enter x2apic cluster mode when interrupt remmaping 
is enabled in qemu, however, latest AMD EPYC still just supports xapic 
mode which can get great improvement by Exit-less IPIs. This patchset 
lets a guest send multicast IPIs, with at most 128 destinations per 
hypercall in 64-bit mode and 64 vCPUs per hypercall in 32-bit mode.

Hardware: Xeon Skylake 2.5GHz, 2 sockets, 40 cores, 80 threads, the VM 
is 80 vCPUs, IPI microbenchmark(https://lkml.org/lkml/2017/12/19/141):

x2apic cluster mode, vanilla

 Dry-run:                         0,            2392199 ns
 Self-IPI:                  6907514,           15027589 ns
 Normal IPI:              223910476,          251301666 ns
 Broadcast IPI:                   0,         9282161150 ns
 Broadcast lock:                  0,         8812934104 ns

x2apic cluster mode, pv-ipi 

 Dry-run:                         0,            2449341 ns
 Self-IPI:                  6720360,           15028732 ns
 Normal IPI:              228643307,          255708477 ns
 Broadcast IPI:                   0,         7572293590 ns  => 22% performance boost 
 Broadcast lock:                  0,         8316124651 ns

x2apic physical mode, vanilla

 Dry-run:                         0,            3135933 ns
 Self-IPI:                  8572670,           17901757 ns
 Normal IPI:              226444334,          255421709 ns
 Broadcast IPI:                   0,        19845070887 ns
 Broadcast lock:                  0,        19827383656 ns

x2apic physical mode, pv-ipi

 Dry-run:                         0,            2446381 ns
 Self-IPI:                  6788217,           15021056 ns
 Normal IPI:              219454441,          249583458 ns
 Broadcast IPI:                   0,         7806540019 ns  => 154% performance boost 
 Broadcast lock:                  0,         9143618799 ns

v4 -> v5:
 * update hypercall layout description
 * fix PV IPIs send hypercall loops

v3 -> v4:
 * offset algorithm w/ __uint128_t to scale to higher APIC IDs
 * remove num_possible_cpus limit
 * pass op_64_bit to check bitmap size
 * better describe hypercall layout

v2 -> v3:
 * rename ipi_mask_done to irq_restore_exit, __send_ipi_mask return int 
   instead of bool 
 * fix build errors reported by 0day
 * split patches, nothing change 

v1 -> v2:
 * sparse apic id > 128, or any other errors, fallback to original apic hooks
 * have two bitmask arguments so that one hypercall handles 128 vCPUs 
 * fix KVM_FEATURE_PV_SEND_IPI doc
 * document hypercall
 * fix NMI selftest fails
 * fix build errors reported by 0day

Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>

Wanpeng Li (6):
  KVM: X86: Add kvm hypervisor init time platform setup callback
  KVM: X86: Implement PV IPIs in linux guest
  KVM: X86: Fallback to original apic hooks when bad happens
  KVM: X86: Implement PV IPIs send hypercall
  KVM: X86: Add NMI support to PV IPIs
  KVM: X86: Expose PV_SEND_IPI CPUID feature bit to guest

 Documentation/virtual/kvm/cpuid.txt      |   4 ++
 Documentation/virtual/kvm/hypercalls.txt |  20 ++++++
 arch/x86/include/uapi/asm/kvm_para.h     |   1 +
 arch/x86/kernel/kvm.c                    | 111 +++++++++++++++++++++++++++++++
 arch/x86/kvm/cpuid.c                     |   3 +-
 arch/x86/kvm/x86.c                       |  43 ++++++++++++
 include/uapi/linux/kvm_para.h            |   1 +
 7 files changed, 182 insertions(+), 1 deletion(-)

-- 
2.7.4