From: Marc Zyngier <marc.zyngier@arm.com>
To: Shannon Zhao <zhaoshenglong@huawei.com>,
kvmarm@lists.cs.columbia.edu, christoffer.dall@linaro.org
Cc: linux-arm-kernel@lists.infradead.org, kvm@vger.kernel.org,
will.deacon@arm.com, alex.bennee@linaro.org, wei@redhat.com,
cov@codeaurora.org, shannon.zhao@linaro.org,
peter.huangpeng@huawei.com, hangaohuai@huawei.com
Subject: Re: [PATCH v6 17/21] KVM: ARM64: Add helper to handle PMCR register bits
Date: Tue, 08 Dec 2015 17:36:23 +0000 [thread overview]
Message-ID: <56671517.3040503@arm.com> (raw)
In-Reply-To: <1449578860-15808-18-git-send-email-zhaoshenglong@huawei.com>
On 08/12/15 12:47, Shannon Zhao wrote:
> From: Shannon Zhao <shannon.zhao@linaro.org>
>
> According to ARMv8 spec, when writing 1 to PMCR.E, all counters are
> enabled by PMCNTENSET, while writing 0 to PMCR.E, all counters are
> disabled. When writing 1 to PMCR.P, reset all event counters, not
> including PMCCNTR, to zero. When writing 1 to PMCR.C, reset PMCCNTR to
> zero.
>
> Signed-off-by: Shannon Zhao <shannon.zhao@linaro.org>
> ---
> arch/arm64/kvm/sys_regs.c | 2 ++
> include/kvm/arm_pmu.h | 2 ++
> virt/kvm/arm/pmu.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 55 insertions(+)
>
> diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
> index 9baa654..110b288 100644
> --- a/arch/arm64/kvm/sys_regs.c
> +++ b/arch/arm64/kvm/sys_regs.c
> @@ -620,6 +620,7 @@ static bool access_pmu_regs(struct kvm_vcpu *vcpu,
> val &= ~ARMV8_PMCR_MASK;
> val |= p->regval & ARMV8_PMCR_MASK;
> vcpu_sys_reg(vcpu, r->reg) = val;
> + kvm_pmu_handle_pmcr(vcpu, val);
> break;
> }
> case PMCEID0_EL0:
> @@ -1218,6 +1219,7 @@ static bool access_pmu_cp15_regs(struct kvm_vcpu *vcpu,
> val &= ~ARMV8_PMCR_MASK;
> val |= p->regval & ARMV8_PMCR_MASK;
> vcpu_cp15(vcpu, r->reg) = val;
> + kvm_pmu_handle_pmcr(vcpu, val);
> break;
> }
> case c9_PMCEID0:
> diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h
> index d12450a..a131f76 100644
> --- a/include/kvm/arm_pmu.h
> +++ b/include/kvm/arm_pmu.h
> @@ -46,6 +46,7 @@ void kvm_pmu_overflow_set(struct kvm_vcpu *vcpu, u32 val);
> void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u32 val);
> void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u32 data,
> u32 select_idx);
> +void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u32 val);
> #else
> u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u32 select_idx)
> {
> @@ -58,6 +59,7 @@ void kvm_pmu_overflow_set(struct kvm_vcpu *vcpu, u32 val) {}
> void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u32 val) {}
> void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u32 data,
> u32 select_idx) {}
> +void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u32 val) {}
> #endif
>
> #endif
> diff --git a/virt/kvm/arm/pmu.c b/virt/kvm/arm/pmu.c
> index 093e211..9b9c706 100644
> --- a/virt/kvm/arm/pmu.c
> +++ b/virt/kvm/arm/pmu.c
> @@ -151,6 +151,57 @@ static u32 kvm_pmu_valid_counter_mask(struct kvm_vcpu *vcpu)
> }
>
> /**
> + * kvm_pmu_handle_pmcr - handle PMCR register
> + * @vcpu: The vcpu pointer
> + * @val: the value guest writes to PMCR register
> + */
> +void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u32 val)
> +{
> + struct kvm_pmu *pmu = &vcpu->arch.pmu;
> + struct kvm_pmc *pmc;
> + u32 enable;
> + int i;
> +
> + if (val & ARMV8_PMCR_E) {
> + if (!vcpu_mode_is_32bit(vcpu))
> + enable = vcpu_sys_reg(vcpu, PMCNTENSET_EL0);
This returns a 64bit quantity. Please explicitly select the low 32bits.
> + else
> + enable = vcpu_cp15(vcpu, c9_PMCNTENSET);
> +
> + kvm_pmu_enable_counter(vcpu, enable, true);
It really feels like there is some common stuff with the handling of
PNCTENSET,
> + } else {
> + kvm_pmu_disable_counter(vcpu, 0xffffffffUL);
> + }
> +
> + if (val & ARMV8_PMCR_C) {
> + pmc = &pmu->pmc[ARMV8_MAX_COUNTERS - 1];
Nit: it would be nice to have a #define for the cycle counter.
> + if (pmc->perf_event)
> + local64_set(&pmc->perf_event->count, 0);
> + if (!vcpu_mode_is_32bit(vcpu))
> + vcpu_sys_reg(vcpu, PMCCNTR_EL0) = 0;
> + else
> + vcpu_cp15(vcpu, c9_PMCCNTR) = 0;
> + }
> +
> + if (val & ARMV8_PMCR_P) {
> + for (i = 0; i < ARMV8_MAX_COUNTERS - 1; i++) {
> + pmc = &pmu->pmc[i];
> + if (pmc->perf_event)
> + local64_set(&pmc->perf_event->count, 0);
> + if (!vcpu_mode_is_32bit(vcpu))
> + vcpu_sys_reg(vcpu, PMEVCNTR0_EL0 + i) = 0;
> + else
> + vcpu_cp15(vcpu, c14_PMEVCNTR0 + i) = 0;
> + }
> + }
> +
> + if (val & ARMV8_PMCR_LC) {
> + pmc = &pmu->pmc[ARMV8_MAX_COUNTERS - 1];
> + pmc->bitmask = 0xffffffffffffffffUL;
> + }
> +}
> +
> +/**
> * kvm_pmu_overflow_clear - clear PMU overflow interrupt
> * @vcpu: The vcpu pointer
> * @val: the value guest writes to PMOVSCLR register
>
Thanks,
M.
--
Jazz is not dead. It just smells funny...
WARNING: multiple messages have this Message-ID (diff)
From: marc.zyngier@arm.com (Marc Zyngier)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v6 17/21] KVM: ARM64: Add helper to handle PMCR register bits
Date: Tue, 08 Dec 2015 17:36:23 +0000 [thread overview]
Message-ID: <56671517.3040503@arm.com> (raw)
In-Reply-To: <1449578860-15808-18-git-send-email-zhaoshenglong@huawei.com>
On 08/12/15 12:47, Shannon Zhao wrote:
> From: Shannon Zhao <shannon.zhao@linaro.org>
>
> According to ARMv8 spec, when writing 1 to PMCR.E, all counters are
> enabled by PMCNTENSET, while writing 0 to PMCR.E, all counters are
> disabled. When writing 1 to PMCR.P, reset all event counters, not
> including PMCCNTR, to zero. When writing 1 to PMCR.C, reset PMCCNTR to
> zero.
>
> Signed-off-by: Shannon Zhao <shannon.zhao@linaro.org>
> ---
> arch/arm64/kvm/sys_regs.c | 2 ++
> include/kvm/arm_pmu.h | 2 ++
> virt/kvm/arm/pmu.c | 51 +++++++++++++++++++++++++++++++++++++++++++++++
> 3 files changed, 55 insertions(+)
>
> diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
> index 9baa654..110b288 100644
> --- a/arch/arm64/kvm/sys_regs.c
> +++ b/arch/arm64/kvm/sys_regs.c
> @@ -620,6 +620,7 @@ static bool access_pmu_regs(struct kvm_vcpu *vcpu,
> val &= ~ARMV8_PMCR_MASK;
> val |= p->regval & ARMV8_PMCR_MASK;
> vcpu_sys_reg(vcpu, r->reg) = val;
> + kvm_pmu_handle_pmcr(vcpu, val);
> break;
> }
> case PMCEID0_EL0:
> @@ -1218,6 +1219,7 @@ static bool access_pmu_cp15_regs(struct kvm_vcpu *vcpu,
> val &= ~ARMV8_PMCR_MASK;
> val |= p->regval & ARMV8_PMCR_MASK;
> vcpu_cp15(vcpu, r->reg) = val;
> + kvm_pmu_handle_pmcr(vcpu, val);
> break;
> }
> case c9_PMCEID0:
> diff --git a/include/kvm/arm_pmu.h b/include/kvm/arm_pmu.h
> index d12450a..a131f76 100644
> --- a/include/kvm/arm_pmu.h
> +++ b/include/kvm/arm_pmu.h
> @@ -46,6 +46,7 @@ void kvm_pmu_overflow_set(struct kvm_vcpu *vcpu, u32 val);
> void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u32 val);
> void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u32 data,
> u32 select_idx);
> +void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u32 val);
> #else
> u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u32 select_idx)
> {
> @@ -58,6 +59,7 @@ void kvm_pmu_overflow_set(struct kvm_vcpu *vcpu, u32 val) {}
> void kvm_pmu_software_increment(struct kvm_vcpu *vcpu, u32 val) {}
> void kvm_pmu_set_counter_event_type(struct kvm_vcpu *vcpu, u32 data,
> u32 select_idx) {}
> +void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u32 val) {}
> #endif
>
> #endif
> diff --git a/virt/kvm/arm/pmu.c b/virt/kvm/arm/pmu.c
> index 093e211..9b9c706 100644
> --- a/virt/kvm/arm/pmu.c
> +++ b/virt/kvm/arm/pmu.c
> @@ -151,6 +151,57 @@ static u32 kvm_pmu_valid_counter_mask(struct kvm_vcpu *vcpu)
> }
>
> /**
> + * kvm_pmu_handle_pmcr - handle PMCR register
> + * @vcpu: The vcpu pointer
> + * @val: the value guest writes to PMCR register
> + */
> +void kvm_pmu_handle_pmcr(struct kvm_vcpu *vcpu, u32 val)
> +{
> + struct kvm_pmu *pmu = &vcpu->arch.pmu;
> + struct kvm_pmc *pmc;
> + u32 enable;
> + int i;
> +
> + if (val & ARMV8_PMCR_E) {
> + if (!vcpu_mode_is_32bit(vcpu))
> + enable = vcpu_sys_reg(vcpu, PMCNTENSET_EL0);
This returns a 64bit quantity. Please explicitly select the low 32bits.
> + else
> + enable = vcpu_cp15(vcpu, c9_PMCNTENSET);
> +
> + kvm_pmu_enable_counter(vcpu, enable, true);
It really feels like there is some common stuff with the handling of
PNCTENSET,
> + } else {
> + kvm_pmu_disable_counter(vcpu, 0xffffffffUL);
> + }
> +
> + if (val & ARMV8_PMCR_C) {
> + pmc = &pmu->pmc[ARMV8_MAX_COUNTERS - 1];
Nit: it would be nice to have a #define for the cycle counter.
> + if (pmc->perf_event)
> + local64_set(&pmc->perf_event->count, 0);
> + if (!vcpu_mode_is_32bit(vcpu))
> + vcpu_sys_reg(vcpu, PMCCNTR_EL0) = 0;
> + else
> + vcpu_cp15(vcpu, c9_PMCCNTR) = 0;
> + }
> +
> + if (val & ARMV8_PMCR_P) {
> + for (i = 0; i < ARMV8_MAX_COUNTERS - 1; i++) {
> + pmc = &pmu->pmc[i];
> + if (pmc->perf_event)
> + local64_set(&pmc->perf_event->count, 0);
> + if (!vcpu_mode_is_32bit(vcpu))
> + vcpu_sys_reg(vcpu, PMEVCNTR0_EL0 + i) = 0;
> + else
> + vcpu_cp15(vcpu, c14_PMEVCNTR0 + i) = 0;
> + }
> + }
> +
> + if (val & ARMV8_PMCR_LC) {
> + pmc = &pmu->pmc[ARMV8_MAX_COUNTERS - 1];
> + pmc->bitmask = 0xffffffffffffffffUL;
> + }
> +}
> +
> +/**
> * kvm_pmu_overflow_clear - clear PMU overflow interrupt
> * @vcpu: The vcpu pointer
> * @val: the value guest writes to PMOVSCLR register
>
Thanks,
M.
--
Jazz is not dead. It just smells funny...
next prev parent reply other threads:[~2015-12-08 17:36 UTC|newest]
Thread overview: 126+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-08 12:47 [PATCH v6 00/21] KVM: ARM64: Add guest PMU support Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 01/21] ARM64: Move PMU register related defines to asm/pmu.h Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 02/21] KVM: ARM64: Define PMU data structure for each vcpu Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 13:37 ` Marc Zyngier
2015-12-08 13:37 ` Marc Zyngier
2015-12-08 13:53 ` Will Deacon
2015-12-08 13:53 ` Will Deacon
2015-12-08 14:10 ` Marc Zyngier
2015-12-08 14:10 ` Marc Zyngier
2015-12-08 14:14 ` Shannon Zhao
2015-12-08 14:14 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 03/21] KVM: ARM64: Add offset defines for PMU registers Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 04/21] KVM: ARM64: Add reset and access handlers for PMCR_EL0 register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 05/21] KVM: ARM64: Add reset and access handlers for PMSELR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 06/21] KVM: ARM64: Add reset and access handlers for PMCEID0 and PMCEID1 register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 14:23 ` Marc Zyngier
2015-12-08 14:23 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 07/21] KVM: ARM64: PMU: Add perf event map and introduce perf event creating function Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 15:43 ` Marc Zyngier
2015-12-08 15:43 ` Marc Zyngier
2015-12-09 7:38 ` Shannon Zhao
2015-12-09 7:38 ` Shannon Zhao
2015-12-09 7:38 ` Shannon Zhao
2015-12-09 8:23 ` Marc Zyngier
2015-12-09 8:23 ` Marc Zyngier
2015-12-09 8:23 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 08/21] KVM: ARM64: Add access handler for PMEVTYPERn and PMCCFILTR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 16:17 ` Marc Zyngier
2015-12-08 16:17 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 09/21] KVM: ARM64: Add access handler for PMXEVTYPER register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 10/21] KVM: ARM64: Add access handler for PMEVCNTRn and PMCCNTR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 16:30 ` Marc Zyngier
2015-12-08 16:30 ` Marc Zyngier
2015-12-10 11:36 ` Shannon Zhao
2015-12-10 11:36 ` Shannon Zhao
2015-12-10 11:36 ` Shannon Zhao
2015-12-10 12:07 ` Marc Zyngier
2015-12-10 12:07 ` Marc Zyngier
2015-12-10 13:23 ` Shannon Zhao
2015-12-10 13:23 ` Shannon Zhao
2015-12-10 13:23 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 11/21] KVM: ARM64: Add access handler for PMXEVCNTR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 16:33 ` Marc Zyngier
2015-12-08 16:33 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 12/21] KVM: ARM64: Add reset and access handlers for PMCNTENSET and PMCNTENCLR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 16:42 ` Marc Zyngier
2015-12-08 16:42 ` Marc Zyngier
2015-12-09 8:35 ` Shannon Zhao
2015-12-09 8:35 ` Shannon Zhao
2015-12-09 8:35 ` Shannon Zhao
2015-12-09 8:56 ` Marc Zyngier
2015-12-09 8:56 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 13/21] KVM: ARM64: Add reset and access handlers for PMINTENSET and PMINTENCLR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 14/21] KVM: ARM64: Add reset and access handlers for PMOVSSET and PMOVSCLR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 16:59 ` Marc Zyngier
2015-12-08 16:59 ` Marc Zyngier
2015-12-09 8:47 ` Shannon Zhao
2015-12-09 8:47 ` Shannon Zhao
2015-12-09 8:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 15/21] KVM: ARM64: Add reset and access handlers for PMUSERENR register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 17:03 ` Marc Zyngier
2015-12-08 17:03 ` Marc Zyngier
2015-12-09 9:18 ` Shannon Zhao
2015-12-09 9:18 ` Shannon Zhao
2015-12-09 9:18 ` Shannon Zhao
2015-12-09 9:49 ` Marc Zyngier
2015-12-09 9:49 ` Marc Zyngier
2015-12-09 9:49 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 16/21] KVM: ARM64: Add reset and access handlers for PMSWINC register Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 17/21] KVM: ARM64: Add helper to handle PMCR register bits Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 17:36 ` Marc Zyngier [this message]
2015-12-08 17:36 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 18/21] KVM: ARM64: Add PMU overflow interrupt routing Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 17:37 ` Marc Zyngier
2015-12-08 17:37 ` Marc Zyngier
2015-12-08 12:47 ` [PATCH v6 19/21] KVM: ARM64: Reset PMU state when resetting vcpu Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 20/21] KVM: ARM64: Free perf event of PMU when destroying vcpu Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` [PATCH v6 21/21] KVM: ARM64: Add a new kvm ARM PMU device Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 12:47 ` Shannon Zhao
2015-12-08 17:43 ` Marc Zyngier
2015-12-08 17:43 ` Marc Zyngier
2015-12-08 17:56 ` [PATCH v6 00/21] KVM: ARM64: Add guest PMU support Marc Zyngier
2015-12-08 17:56 ` Marc Zyngier
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=56671517.3040503@arm.com \
--to=marc.zyngier@arm.com \
--cc=alex.bennee@linaro.org \
--cc=christoffer.dall@linaro.org \
--cc=cov@codeaurora.org \
--cc=hangaohuai@huawei.com \
--cc=kvm@vger.kernel.org \
--cc=kvmarm@lists.cs.columbia.edu \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=peter.huangpeng@huawei.com \
--cc=shannon.zhao@linaro.org \
--cc=wei@redhat.com \
--cc=will.deacon@arm.com \
--cc=zhaoshenglong@huawei.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 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.