On 08/12/15 13:53, Will Deacon wrote:

On Tue, Dec 08, 2015 at 01:37:14PM +0000, Marc Zyngier wrote:

quoted

On 08/12/15 12:47, Shannon Zhao wrote:

quoted

From: Shannon Zhao <redacted>

Here we plan to support virtual PMU for guest by full software
emulation, so define some basic structs and functions preparing for
futher steps. Define struct kvm_pmc for performance monitor counter and
struct kvm_pmu for performance monitor unit for each vcpu. According to
ARMv8 spec, the PMU contains at most 32(ARMV8_MAX_COUNTERS) counters.

Since this only supports ARM64 (or PMUv3), add a separate config symbol
for it.

Signed-off-by: Shannon Zhao <redacted>
---
 arch/arm64/include/asm/kvm_host.h |  2 ++
 arch/arm64/kvm/Kconfig            |  8 ++++++++
 include/kvm/arm_pmu.h             | 40 +++++++++++++++++++++++++++++++++++++++
 3 files changed, 50 insertions(+)
 create mode 100644 include/kvm/arm_pmu.h

[...]

quoted

quoted

diff --git a/arch/arm64/kvm/Kconfig b/arch/arm64/kvm/Kconfig
index a5272c0..66da9a2 100644
--- a/arch/arm64/kvm/Kconfig
+++ b/arch/arm64/kvm/Kconfig

@@ -36,6 +36,7 @@ config KVM
 	select HAVE_KVM_EVENTFD
 	select HAVE_KVM_IRQFD
 	select KVM_ARM_VGIC_V3
+	select KVM_ARM_PMU

What if HW_PERF_EVENTS is not selected? Also, selecting HW_PERF_EVENTS
is not enough, and you probably need PERF_EVENTS as well, So this should
probably read:

	select KVM_ARM_PMU if (HW_PERF_EVENTS && PERF_EVENTS)

HW_PERF_EVENTS depends on ARM_PMU which in turn depends on PERF_EVENTS.

in which case, let's make it:

	select KVM_ARM_PMU if HW_PERF_EVENTS

which should give us the minimal chain. I hate the kernel config
language! ;-)

	M.
-- 
Jazz is not dead. It just smells funny...