[PATCH v13 00/20] KVM: ARM64: Add guest PMU support
Shannon Zhao
zhaoshenglong at huawei.com
Tue Feb 23 21:08:20 PST 2016
From: Shannon Zhao <shannon.zhao at linaro.org>
This patchset adds guest PMU support for KVM on ARM64. It takes
trap-and-emulate approach. When guest wants to monitor one event, it
will be trapped by KVM and KVM will call perf_event API to create a perf
event and call relevant perf_event APIs to get the count value of event.
Use perf to test this patchset in guest. When using "perf list", it
shows the list of the hardware events and hardware cache events perf
supports. Then use "perf stat -e EVENT" to monitor some event. For
example, use "perf stat -e cycles" to count cpu cycles and
"perf stat -e cache-misses" to count cache misses.
Below are the outputs of "perf stat -r 5 sleep 5" when running in host
and guest.
Host:
Performance counter stats for 'sleep 5' (5 runs):
0.529248 task-clock (msec) # 0.000 CPUs utilized ( +- 1.65% )
1 context-switches # 0.002 M/sec
0 cpu-migrations # 0.000 K/sec
49 page-faults # 0.092 M/sec ( +- 1.05% )
1104279 cycles # 2.087 GHz ( +- 1.65% )
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
528112 instructions # 0.48 insns per cycle ( +- 1.12% )
<not supported> branches
9579 branch-misses # 18.099 M/sec ( +- 2.40% )
5.000851904 seconds time elapsed ( +- 0.00% )
Guest:
Performance counter stats for 'sleep 5' (5 runs):
0.695412 task-clock (msec) # 0.000 CPUs utilized ( +- 1.26% )
1 context-switches # 0.001 M/sec
0 cpu-migrations # 0.000 K/sec
49 page-faults # 0.070 M/sec ( +- 1.29% )
1430471 cycles # 2.057 GHz ( +- 1.25% )
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
659173 instructions # 0.46 insns per cycle ( +- 2.64% )
<not supported> branches
10893 branch-misses # 15.664 M/sec ( +- 1.23% )
5.001277044 seconds time elapsed ( +- 0.00% )
Have a cycle counter read test like below in guest and host:
static void test(void)
{
unsigned long count, count1, count2;
count1 = read_cycles();
count++;
count2 = read_cycles();
}
Host:
count1: 3046505444
count2: 3046505575
delta: 131
Guest:
count1: 5932773531
count2: 5932773668
delta: 137
The gap between guest and host is very small. One reason for this I
think is that it doesn't count the cycles in EL2 and host since we add
exclude_hv = 1. So the cycles spent to store/restore registers which
happens at EL2 are not included.
This patchset can be fetched from [1] and the relevant QEMU version for
test can be fetched from [2].
The results of 'perf test' can be found from [3][4].
The results of perf_event_tests test suite can be found from [5][6].
Also, I have tested "perf top" in two VMs and host at the same time. It
works well.
Thanks,
Shannon
[1] https://git.linaro.org/people/shannon.zhao/linux-mainline.git KVM_ARM64_PMU_v13
[2] https://git.linaro.org/people/shannon.zhao/qemu.git PMU
[3] http://people.linaro.org/~shannon.zhao/PMU/perf-test-host.txt
[4] http://people.linaro.org/~shannon.zhao/PMU/perf-test-guest.txt
[5] http://people.linaro.org/~shannon.zhao/PMU/perf_event_tests-host.txt
[6] http://people.linaro.org/~shannon.zhao/PMU/perf_event_tests-guest.txt
Changes since v12:
* Fix bisect problem
* Move ARMV8_PMU_PMCR_LC into the patch which firstly uses it
* Inject PMU irq only when level changed to save cycles(Marc, thanks!)
Changes since v11:
* Move PMU register related defines to asm/perf_event.h and rename them
with prefix ARMV8_PMU_*
* BUG_ON when writing to PMCEID0/1 register
* Lower the PMU interrupt line when PMCR_EL0.E == 0
* Fix come coding styles
* Drop kvm_arm_pmu_irq_access
Changes since v10:
* Check if PMCR.N is zero before using GENMASK
* Fix attr.disabled
* use same way to reset counter's value in PATCH 14
Changes since v9:
* Change kvm_arm_support_pmu_v3 to a bool function [PATCH 19/21]
* Fix several typoes, change checking logic of kvm_arm_pmu_v3_init and
change irq_is_invalid to irq_is_valid [PATCH 21/21]
* Add Acks and Rb from Peter and Andrew, thanks a lot
Changes since v8:
* Fix the wrong use of r->reg in register accessors for 32bit part
* Rewrite the handle of PMUSERENR based on the new inject UND patch
* Drop the inline attribute
* Introduce SET/GET/HAS_DEVICE_ATTR for vcpu iotcl and set the PMU
overflow interrupt via this API
* Use one feature bit for PMUv3
Changes since v7:
* Rebase on kvm-arm next
* Fix the handler of PMUSERENR and add a helper to forward trap to guest
EL1
* Fix some small bugs found by Marc
Changes since v6:
* Rebase on v4.4-rc5
* Drop access_pmu_cp15_regs() so that it could use same handler for both
arch64 and arch32. And it could drop the definitions of CP15 register
offsets, also avoid same codes added twice
* Use vcpu_sys_reg() when accessing PMU registers to avoid endian things
* Add handler for PMUSERENR and some checkers for other registers
* Add kvm_arm_pmu_get_attr()
Changes since v5:
* Rebase on new linux kernel mainline
* Remove state duplications and drop PMOVSCLR, PMCNTENCLR, PMINTENCLR,
PMXEVCNTR, PMXEVTYPER
* Add a helper to check if vPMU is already initialized
* remove kvm_vcpu from kvm_pmc
Changes since v4:
* Rebase on new linux kernel mainline
* Drop the reset handler of CP15 registers
* Fix a compile failure on arch ARM due to lack of asm/pmu.h
* Refactor the interrupt injecting flow according to Marc's suggestion
* Check the value of PMSELR register
* Calculate the attr.disabled according to PMCR.E and PMCNTENSET/CLR
* Fix some coding style
* Document the vPMU irq range
Changes since v3:
* Rebase on new linux kernel mainline
* Use ARMV8_MAX_COUNTERS instead of 32
* Reset PMCR.E to zero.
* Trigger overflow for software increment.
* Optimize PMU interrupt inject logic.
* Add handler for E,C,P bits of PMCR
* Fix the overflow bug found by perf_event_tests
* Run 'perf test', 'perf top' and perf_event_tests test suite
* Add exclude_hv = 1 configuration to not count in EL2
Changes since v2:
* Directly use perf raw event type to create perf_event in KVM
* Add a helper vcpu_sysreg_write
* remove unrelated header file
Changes since v1:
* Use switch...case for registers access handler instead of adding
alone handler for each register
* Try to use the sys_regs to store the register value instead of adding
new variables in struct kvm_pmc
* Fix the handle of cp15 regs
* Create a new kvm device vPMU, then userspace could choose whether to
create PMU
* Fix the handle of PMU overflow interrupt
Shannon Zhao (20):
ARM64: Move PMU register related defines to asm/perf_event.h
KVM: ARM64: Define PMU data structure for each vcpu
KVM: ARM64: Add access handler for PMCR register
KVM: ARM64: Add access handler for PMSELR register
KVM: ARM64: Add access handler for PMCEID0 and PMCEID1 register
KVM: ARM64: Add access handler for event counter register
KVM: ARM64: Add access handler for PMCNTENSET and PMCNTENCLR register
KVM: ARM64: PMU: Add perf event map and introduce perf event creating
function
KVM: ARM64: Add access handler for event type register
KVM: ARM64: Add access handler for PMINTENSET and PMINTENCLR register
KVM: ARM64: Add access handler for PMOVSSET and PMOVSCLR register
KVM: ARM64: Add access handler for PMSWINC register
KVM: ARM64: Add helper to handle PMCR register bits
KVM: ARM64: Add access handler for PMUSERENR register
KVM: ARM64: Add PMU overflow interrupt routing
KVM: ARM64: Reset PMU state when resetting vcpu
KVM: ARM64: Free perf event of PMU when destroying vcpu
KVM: ARM64: Add a new feature bit for PMUv3
KVM: ARM: Introduce per-vcpu kvm device controls
KVM: ARM64: Add a new vcpu device control group for PMUv3
Documentation/virtual/kvm/api.txt | 12 +-
Documentation/virtual/kvm/devices/vcpu.txt | 33 ++
arch/arm/include/asm/kvm_host.h | 15 +
arch/arm/kvm/arm.c | 61 ++++
arch/arm64/include/asm/kvm_host.h | 25 +-
arch/arm64/include/asm/perf_event.h | 48 +++
arch/arm64/include/uapi/asm/kvm.h | 6 +
arch/arm64/kernel/perf_event.c | 68 +---
arch/arm64/kvm/Kconfig | 7 +
arch/arm64/kvm/Makefile | 1 +
arch/arm64/kvm/guest.c | 51 +++
arch/arm64/kvm/hyp/hyp.h | 1 +
arch/arm64/kvm/hyp/switch.c | 3 +
arch/arm64/kvm/reset.c | 7 +
arch/arm64/kvm/sys_regs.c | 569 +++++++++++++++++++++++++++--
include/kvm/arm_pmu.h | 108 ++++++
include/uapi/linux/kvm.h | 2 +
virt/kvm/arm/pmu.c | 511 ++++++++++++++++++++++++++
18 files changed, 1443 insertions(+), 85 deletions(-)
create mode 100644 Documentation/virtual/kvm/devices/vcpu.txt
create mode 100644 include/kvm/arm_pmu.h
create mode 100644 virt/kvm/arm/pmu.c
--
2.0.4
More information about the linux-arm-kernel
mailing list