From: Andrew Murray <andrew.murray@arm.com>
To: Christoffer Dall <christoffer.dall@arm.com>,
Marc Zyngier <marc.zyngier@arm.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will.deacon@arm.com>,
Mark Rutland <mark.rutland@arm.com>
Cc: kvmarm@lists.cs.columbia.edu, linux-arm-kernel@lists.infradead.org
Subject: [PATCH v13 4/8] arm64: arm_pmu: Add !VHE support for exclude_host/exclude_guest attributes
Date: Tue, 9 Apr 2019 20:22:13 +0100 [thread overview]
Message-ID: <20190409192217.23459-5-andrew.murray@arm.com> (raw)
Message-ID: <20190409192213.u0QOlGiYU89TyLpV1J_gc2Ih6smtrpoDo34ZOPhsbcI@z> (raw)
In-Reply-To: <20190409192217.23459-1-andrew.murray@arm.com>
Add support for the :G and :H attributes in perf by handling the
exclude_host/exclude_guest event attributes.
We notify KVM of counters that we wish to be enabled or disabled on
guest entry/exit and thus defer from starting or stopping events based
on their event attributes.
With !VHE we switch the counters between host/guest at EL2. We are able
to eliminate counters counting host events on the boundaries of guest
entry/exit when using :G by filtering out EL2 for exclude_host. When
using !exclude_hv there is a small blackout window at the guest
entry/exit where host events are not captured.
Signed-off-by: Andrew Murray <andrew.murray@arm.com>
Acked-by: Will Deacon <will.deacon@arm.com>
---
arch/arm64/kernel/perf_event.c | 43 ++++++++++++++++++++++++++++------
1 file changed, 36 insertions(+), 7 deletions(-)
diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index cccf4fc86d67..6bb28aaf5aea 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -26,6 +26,7 @@
#include <linux/acpi.h>
#include <linux/clocksource.h>
+#include <linux/kvm_host.h>
#include <linux/of.h>
#include <linux/perf/arm_pmu.h>
#include <linux/platform_device.h>
@@ -528,11 +529,21 @@ static inline int armv8pmu_enable_counter(int idx)
static inline void armv8pmu_enable_event_counter(struct perf_event *event)
{
+ struct perf_event_attr *attr = &event->attr;
int idx = event->hw.idx;
+ u32 counter_bits = BIT(ARMV8_IDX_TO_COUNTER(idx));
- armv8pmu_enable_counter(idx);
if (armv8pmu_event_is_chained(event))
- armv8pmu_enable_counter(idx - 1);
+ counter_bits |= BIT(ARMV8_IDX_TO_COUNTER(idx - 1));
+
+ kvm_set_pmu_events(counter_bits, attr);
+
+ /* We rely on the hypervisor switch code to enable guest counters */
+ if (!kvm_pmu_counter_deferred(attr)) {
+ armv8pmu_enable_counter(idx);
+ if (armv8pmu_event_is_chained(event))
+ armv8pmu_enable_counter(idx - 1);
+ }
}
static inline int armv8pmu_disable_counter(int idx)
@@ -545,11 +556,21 @@ static inline int armv8pmu_disable_counter(int idx)
static inline void armv8pmu_disable_event_counter(struct perf_event *event)
{
struct hw_perf_event *hwc = &event->hw;
+ struct perf_event_attr *attr = &event->attr;
int idx = hwc->idx;
+ u32 counter_bits = BIT(ARMV8_IDX_TO_COUNTER(idx));
if (armv8pmu_event_is_chained(event))
- armv8pmu_disable_counter(idx - 1);
- armv8pmu_disable_counter(idx);
+ counter_bits |= BIT(ARMV8_IDX_TO_COUNTER(idx - 1));
+
+ kvm_clr_pmu_events(counter_bits);
+
+ /* We rely on the hypervisor switch code to disable guest counters */
+ if (!kvm_pmu_counter_deferred(attr)) {
+ if (armv8pmu_event_is_chained(event))
+ armv8pmu_disable_counter(idx - 1);
+ armv8pmu_disable_counter(idx);
+ }
}
static inline int armv8pmu_enable_intens(int idx)
@@ -829,11 +850,16 @@ static int armv8pmu_set_event_filter(struct hw_perf_event *event,
if (!attr->exclude_kernel)
config_base |= ARMV8_PMU_INCLUDE_EL2;
} else {
- if (attr->exclude_kernel)
- config_base |= ARMV8_PMU_EXCLUDE_EL1;
- if (!attr->exclude_hv)
+ if (!attr->exclude_hv && !attr->exclude_host)
config_base |= ARMV8_PMU_INCLUDE_EL2;
}
+
+ /*
+ * Filter out !VHE kernels and guest kernels
+ */
+ if (attr->exclude_kernel)
+ config_base |= ARMV8_PMU_EXCLUDE_EL1;
+
if (attr->exclude_user)
config_base |= ARMV8_PMU_EXCLUDE_EL0;
@@ -863,6 +889,9 @@ static void armv8pmu_reset(void *info)
armv8pmu_disable_intens(idx);
}
+ /* Clear the counters we flip at guest entry/exit */
+ kvm_clr_pmu_events(U32_MAX);
+
/*
* Initialize & Reset PMNC. Request overflow interrupt for
* 64 bit cycle counter but cheat in armv8pmu_write_counter().
--
2.21.0
_______________________________________________
kvmarm mailing list
kvmarm@lists.cs.columbia.edu
https://lists.cs.columbia.edu/mailman/listinfo/kvmarm
next prev parent reply other threads:[~2019-04-09 19:22 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-09 19:22 [PATCH v13 0/8] arm64: Support perf event modifiers :G and :H Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-09 19:22 ` [PATCH v13 1/8] arm64: arm_pmu: remove unnecessary isb instruction Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-09 19:22 ` [PATCH v13 2/8] arm64: KVM: encapsulate kvm_cpu_context in kvm_host_data Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-09 19:22 ` [PATCH v13 3/8] arm64: KVM: add accessors to track guest/host only counters Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-09 19:22 ` Andrew Murray [this message]
2019-04-09 19:22 ` [PATCH v13 4/8] arm64: arm_pmu: Add !VHE support for exclude_host/exclude_guest attributes Andrew Murray
2019-04-11 22:28 ` Suzuki K Poulose
2019-04-11 22:28 ` Suzuki K Poulose
2019-04-09 19:22 ` [PATCH v13 5/8] arm64: KVM: Enable !VHE support for :G/:H perf event modifiers Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-15 13:02 ` Suzuki K Poulose
2019-04-15 13:02 ` Suzuki K Poulose
2019-04-09 19:22 ` [PATCH v13 6/8] arm64: KVM: Enable VHE " Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-15 13:43 ` Suzuki K Poulose
2019-04-15 13:43 ` Suzuki K Poulose
2019-04-30 9:54 ` Andrew Murray
2019-04-30 9:54 ` Andrew Murray
2019-04-09 19:22 ` [PATCH v13 7/8] arm64: KVM: avoid isb's by using direct pmxevtyper sysreg Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-15 13:49 ` Suzuki K Poulose
2019-04-15 13:49 ` Suzuki K Poulose
2019-04-29 16:41 ` Andrew Murray
2019-04-29 16:41 ` Andrew Murray
2019-04-09 19:22 ` [PATCH v13 8/8] arm64: docs: document perf event attributes Andrew Murray
2019-04-09 19:22 ` Andrew Murray
2019-04-11 17:18 ` Will Deacon
2019-04-11 17:18 ` Will Deacon
2019-04-15 14:05 ` Suzuki K Poulose
2019-04-15 14:05 ` Suzuki K Poulose
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190409192217.23459-5-andrew.murray@arm.com \
--to=andrew.murray@arm.com \
--cc=catalin.marinas@arm.com \
--cc=christoffer.dall@arm.com \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=marc.zyngier@arm.com \
--cc=mark.rutland@arm.com \
--cc=will.deacon@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).