All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v5 0/8] IPI virtualization support for VM
@ 2021-12-31 14:28 Zeng Guang
  2021-12-31 14:28 ` [PATCH v5 1/8] x86/cpu: Add new VMX feature, Tertiary VM-Execution control Zeng Guang
                   ` (7 more replies)
  0 siblings, 8 replies; 47+ messages in thread
From: Zeng Guang @ 2021-12-31 14:28 UTC (permalink / raw)
  To: Paolo Bonzini, Sean Christopherson, Vitaly Kuznetsov, Wanpeng Li,
	Jim Mattson, Joerg Roedel, kvm, Dave Hansen, Tony Luck,
	Kan Liang, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	H. Peter Anvin, Kim Phillips, Jarkko Sakkinen, Jethro Beekman,
	Kai Huang
  Cc: x86, linux-kernel, Robert Hu, Gao Chao, Zeng Guang

Currently, issuing an IPI except self-ipi in guest on Intel CPU
always causes a VM-exit. It can lead to non-negligible overhead
to some workloads involving frequent IPIs when running in VMs.

IPI virtualization is a new VT-x feature, targeting to eliminate
VM-exits on source vCPUs when issuing unicast, physical-addressing
IPIs. Once it is enabled, the processor virtualizes following kinds
of operations that send IPIs without causing VM-exits:
- Memory-mapped ICR writes
- MSR-mapped ICR writes
- SENDUIPI execution

This patch series implements IPI virtualization support in KVM.

Patches 1-4 add tertiary processor-based VM-execution support
framework, which is used to enumerate IPI virtualization.

Patch 5 handles APIC-write VM exit due to writes to ICR MSR when
guest works in x2APIC mode. This is a new case introduced by
Intel VT-x.

Patch 6 implements IPI virtualization related function including
feature enabling through tertiary processor-based VM-execution in
various scenarios of VMCS configuration, PID table setup in vCPU
creation and vCPU block consideration.

Patch 7 supports guest VM to modify vCPU APIC ID in xAPIC mode in
the case of IPI virtualization enabled.

Patch 8 changes the memory allocation policy to resize the PID-pointer
table on demand. It targets to reduce overall memory footprint.

Document for IPI virtualization is now available at the latest "Intel
Architecture Instruction Set Extensions Programming Reference".

Document Link:
https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html

We did experiment to measure average time sending IPI from source vCPU
to the target vCPU completing the IPI handling by kvm unittest w/ and
w/o IPI virtualization. When IPI virtualization enabled, it will reduce
22.21% and 15.98% cycles consuming in xAPIC mode and x2APIC mode
respectively.

--------------------------------------
KVM unittest:vmexit/ipi

2 vCPU, AP was modified to run in idle loop instead of halt to ensure
no VM exit impact on target vCPU.

                Cycles of IPI
                xAPIC mode              x2APIC mode
        test    w/o IPIv  w/ IPIv       w/o IPIv  w/ IPIv
        1       6106      4816          4265      3768
        2       6244      4656          4404      3546
        3       6165      4658          4233      3474
        4       5992      4710          4363      3430
        5       6083      4741          4215      3551
        6       6238      4904          4304      3547
        7       6164      4617          4263      3709
        8       5984      4763          4518      3779
        9       5931      4712          4645      3667
        10      5955      4530          4332      3724
        11      5897      4673          4283      3569
        12      6140      4794          4178      3598
        13      6183      4728          4363      3628
        14      5991      4994          4509      3842
        15      5866      4665          4520      3739
        16      6032      4654          4229      3701
        17      6050      4653          4185      3726
        18      6004      4792          4319      3746
        19      5961      4626          4196      3392
        20      6194      4576          4433      3760

Average cycles  6059      4713.1        4337.85   3644.8
%Reduction                -22.21%                 -15.98%

--------------------------------------
IPI microbenchmark:
(https://lore.kernel.org/kvm/20171219085010.4081-1-ynorov@caviumnetworks.com)

2 vCPUs, 1:1 pin vCPU to pCPU, guest VM runs with idle=poll, x2APIC mode

Result with IPIv enabled:

Dry-run:                         0,             272798 ns
Self-IPI:                  5094123,           11114037 ns
Normal IPI:              131697087,          173321200 ns
Broadcast IPI:                   0,          155649075 ns
Broadcast lock:                  0,          161518031 ns

Result with IPIv disabled:

Dry-run:                         0,             272766 ns
Self-IPI:                  5091788,           11123699 ns
Normal IPI:              145215772,          174558920 ns
Broadcast IPI:                   0,          175785384 ns
Broadcast lock:                  0,          149076195 ns


As IPIv can benefit unicast IPI to other CPU, Normal IPI test case gain
about 9.73% time saving on average out of 15 test runs when IPIv is
enabled.

Normal IPI statistics (unit:ns):
        test    w/o IPIv        w/ IPIv
        1       153346049       140907046
        2       147218648       141660618
        3       145215772       117890672
        4       146621682       136430470
        5       144821472       136199421
        6       144704378       131676928
        7       141403224       131697087
        8       144775766       125476250
        9       140658192       137263330
        10      144768626       138593127
        11      145166679       131946752
        12      145020451       116852889
        13      148161353       131406280
        14      148378655       130174353
        15      148903652       127969674

Average time    145944306.6     131742993.1 ns
%Reduction                      -9.73%

--------------------------------------
hackbench:

8 vCPUs, guest VM free run, x2APIC mode
./hackbench -p -l 100000

                w/o IPIv        w/ IPIv
Time            91.887          74.605
%Reduction                      -18.808%

96 vCPUs, guest VM free run, x2APIC mode
./hackbench -p -l 1000000

                w/o IPIv        w/ IPIv
Time            287.504         235.185
%Reduction                      -18.198%

--------------------------------------
v4 -> v5:
1. Deal with enable_ipiv parameter following current
   vmcs configuration rule.
2. Allocate memory for PID-pointer table dynamically
3. Support guest runtime modify APIC ID in xAPIC mode
4. Helper to judge possibility to take PI block in IPIv case

v3 -> v4:
1. Refine code style of patch 2
2. Move tertiary control shadow build into patch 3
3. Make vmx_tertiary_exec_control to be static function

v2 -> v3:
1. Misc change on tertiary execution control
   definition and capability setup
2. Alternative to get tertiary execution
   control configuration

v1 -> v2:
1. Refine the IPIv enabling logic for VM.
   Remove ipiv_active definition per vCPU.

--------------------------------------

Gao Chao (1):
  KVM: VMX: enable IPI virtualization

Robert Hoo (4):
  x86/cpu: Add new VMX feature, Tertiary VM-Execution control
  KVM: VMX: Extend BUILD_CONTROLS_SHADOW macro to support 64-bit
    variation
  KVM: VMX: Detect Tertiary VM-Execution control when setup VMCS config
  KVM: VMX: dump_vmcs() reports tertiary_exec_control field as well

Zeng Guang (3):
  KVM: x86: Support interrupt dispatch in x2APIC mode with APIC-write VM
    exit
  KVM: VMX: Update PID-pointer table entry when APIC ID is changed
  KVM: VMX: Resize PID-ponter table on demand for IPI virtualization

 arch/x86/include/asm/kvm-x86-ops.h |   1 +
 arch/x86/include/asm/kvm_host.h    |   4 +
 arch/x86/include/asm/msr-index.h   |   1 +
 arch/x86/include/asm/vmx.h         |  11 ++
 arch/x86/include/asm/vmxfeatures.h |   5 +-
 arch/x86/kernel/cpu/feat_ctl.c     |   9 +-
 arch/x86/kvm/lapic.c               |  19 ++-
 arch/x86/kvm/lapic.h               |   5 +
 arch/x86/kvm/vmx/capabilities.h    |  14 +++
 arch/x86/kvm/vmx/evmcs.c           |   2 +
 arch/x86/kvm/vmx/evmcs.h           |   1 +
 arch/x86/kvm/vmx/posted_intr.c     |   9 +-
 arch/x86/kvm/vmx/vmcs.h            |   1 +
 arch/x86/kvm/vmx/vmx.c             | 184 +++++++++++++++++++++++++++--
 arch/x86/kvm/vmx/vmx.h             |  69 ++++++-----
 arch/x86/kvm/x86.c                 |   2 +
 16 files changed, 292 insertions(+), 45 deletions(-)

-- 
2.27.0


^ permalink raw reply	[flat|nested] 47+ messages in thread

end of thread, other threads:[~2022-02-23 10:26 UTC | newest]

Thread overview: 47+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-31 14:28 [PATCH v5 0/8] IPI virtualization support for VM Zeng Guang
2021-12-31 14:28 ` [PATCH v5 1/8] x86/cpu: Add new VMX feature, Tertiary VM-Execution control Zeng Guang
2021-12-31 14:28 ` [PATCH v5 2/8] KVM: VMX: Extend BUILD_CONTROLS_SHADOW macro to support 64-bit variation Zeng Guang
2021-12-31 14:28 ` [PATCH v5 3/8] KVM: VMX: Detect Tertiary VM-Execution control when setup VMCS config Zeng Guang
2021-12-31 14:28 ` [PATCH v5 4/8] KVM: VMX: dump_vmcs() reports tertiary_exec_control field as well Zeng Guang
2022-01-13 21:03   ` Sean Christopherson
2022-01-14  4:19     ` Zeng Guang
2022-01-20  1:06       ` Sean Christopherson
2022-01-20  5:34         ` Zeng Guang
2021-12-31 14:28 ` [PATCH v5 5/8] KVM: x86: Support interrupt dispatch in x2APIC mode with APIC-write VM exit Zeng Guang
2022-01-13 21:29   ` Sean Christopherson
2022-01-14  7:52     ` Zeng Guang
2022-01-14 17:34       ` Sean Christopherson
2022-01-15  2:08         ` Zeng Guang
2022-01-18  0:44           ` Yuan Yao
2022-01-18  3:06             ` Zeng Guang
2022-01-18 18:17           ` Sean Christopherson
2022-01-19  2:48             ` Zeng Guang
2021-12-31 14:28 ` [PATCH v5 6/8] KVM: VMX: enable IPI virtualization Zeng Guang
2022-01-13 21:47   ` Sean Christopherson
2022-01-14  5:36     ` Zeng Guang
2021-12-31 14:28 ` [PATCH v5 7/8] KVM: VMX: Update PID-pointer table entry when APIC ID is changed Zeng Guang
2022-01-05 19:13   ` Tom Lendacky
2022-01-06  1:44     ` Zeng Guang
2022-01-06 14:06       ` Tom Lendacky
2022-01-07  8:05         ` Zeng Guang
2022-01-07  8:31           ` Maxim Levitsky
2022-01-10  7:45             ` Chao Gao
2022-01-10 22:24               ` Maxim Levitsky
2022-01-13 22:19                 ` Sean Christopherson
2022-01-14  2:58                   ` Chao Gao
2022-01-14  8:17                     ` Maxim Levitsky
2022-01-17  3:17                       ` Chao Gao
2022-02-02 23:23                   ` Sean Christopherson
2022-02-03 20:22                     ` Sean Christopherson
2022-02-23  6:10                       ` Chao Gao
2022-02-23 10:26                         ` Maxim Levitsky
2022-01-14  0:22               ` Yuan Yao
2021-12-31 14:28 ` [PATCH v5 8/8] KVM: VMX: Resize PID-ponter table on demand for IPI virtualization Zeng Guang
2022-01-13 22:09   ` Sean Christopherson
2022-01-14 15:59     ` Zeng Guang
2022-01-14 16:18       ` Sean Christopherson
2022-01-17 15:04         ` Zeng Guang
2022-01-18 17:15           ` Sean Christopherson
2022-01-19  7:55             ` Zeng Guang
2022-01-20  1:01               ` Sean Christopherson
2022-01-24 16:40                 ` Zeng Guang

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.