linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: "Mi, Dapeng" <dapeng1.mi@linux.intel.com>
To: Sean Christopherson <seanjc@google.com>,
	Marc Zyngier <maz@kernel.org>,
	Oliver Upton <oliver.upton@linux.dev>,
	Tianrui Zhao <zhaotianrui@loongson.cn>,
	Bibo Mao <maobibo@loongson.cn>,
	Huacai Chen <chenhuacai@kernel.org>,
	Anup Patel <anup@brainfault.org>,
	Paul Walmsley <paul.walmsley@sifive.com>,
	Palmer Dabbelt <palmer@dabbelt.com>,
	Albert Ou <aou@eecs.berkeley.edu>, Xin Li <xin@zytor.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Andy Lutomirski <luto@kernel.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Namhyung Kim <namhyung@kernel.org>,
	Paolo Bonzini <pbonzini@redhat.com>
Cc: linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev,
	kvm@vger.kernel.org, loongarch@lists.linux.dev,
	kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org,
	linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
	Kan Liang <kan.liang@linux.intel.com>,
	Yongwei Ma <yongwei.ma@intel.com>,
	Mingwei Zhang <mizhang@google.com>,
	Xiong Zhang <xiong.y.zhang@linux.intel.com>,
	Sandipan Das <sandipan.das@amd.com>
Subject: Re: [PATCH v5 07/44] perf: Add APIs to load/put guest mediated PMU context
Date: Fri, 8 Aug 2025 15:30:40 +0800	[thread overview]
Message-ID: <f3e09bef-8bc9-48ac-983e-7a18c2ff4ad2@linux.intel.com> (raw)
In-Reply-To: <20250806195706.1650976-8-seanjc@google.com>


On 8/7/2025 3:56 AM, Sean Christopherson wrote:
> From: Kan Liang <kan.liang@linux.intel.com>
>
> Add exported APIs to load/put a guest mediated PMU context.  KVM will
> load the guest PMU shortly before VM-Enter, and put the guest PMU shortly
> after VM-Exit.
>
> On the perf side of things, schedule out all exclude_guest events when the
> guest context is loaded, and schedule them back in when the guest context
> is put.  I.e. yield the hardware PMU resources to the guest, by way of KVM.
>
> Note, perf is only responsible for managing host context.  KVM is
> responsible for loading/storing guest state to/from hardware.
>
> Suggested-by: Sean Christopherson <seanjc@google.com>
> Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
> Signed-off-by: Mingwei Zhang <mizhang@google.com>
> [sean: shuffle patches around, write changelog]
> Signed-off-by: Sean Christopherson <seanjc@google.com>
> ---
>  include/linux/perf_event.h |  2 ++
>  kernel/events/core.c       | 61 ++++++++++++++++++++++++++++++++++++++
>  2 files changed, 63 insertions(+)
>
> diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
> index 0958b6d0a61c..42d019d70b42 100644
> --- a/include/linux/perf_event.h
> +++ b/include/linux/perf_event.h
> @@ -1925,6 +1925,8 @@ extern u64 perf_event_pause(struct perf_event *event, bool reset);
>  #ifdef CONFIG_PERF_GUEST_MEDIATED_PMU
>  int perf_create_mediated_pmu(void);
>  void perf_release_mediated_pmu(void);
> +void perf_load_guest_context(unsigned long data);
> +void perf_put_guest_context(void);
>  #endif
>  
>  #else /* !CONFIG_PERF_EVENTS: */
> diff --git a/kernel/events/core.c b/kernel/events/core.c
> index 6875b56ddd6b..77398b1ad4c5 100644
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -469,10 +469,19 @@ static cpumask_var_t perf_online_pkg_mask;
>  static cpumask_var_t perf_online_sys_mask;
>  static struct kmem_cache *perf_event_cache;
>  
> +#ifdef CONFIG_PERF_GUEST_MEDIATED_PMU
> +static DEFINE_PER_CPU(bool, guest_ctx_loaded);
> +
> +static __always_inline bool is_guest_mediated_pmu_loaded(void)
> +{
> +	return __this_cpu_read(guest_ctx_loaded);
> +}
> +#else
>  static __always_inline bool is_guest_mediated_pmu_loaded(void)
>  {
>  	return false;
>  }
> +#endif
>  
>  /*
>   * perf event paranoia level:
> @@ -6379,6 +6388,58 @@ void perf_release_mediated_pmu(void)
>  	atomic_dec(&nr_mediated_pmu_vms);
>  }
>  EXPORT_SYMBOL_GPL(perf_release_mediated_pmu);
> +
> +/* When loading a guest's mediated PMU, schedule out all exclude_guest events. */
> +void perf_load_guest_context(unsigned long data)

nit: the "data" argument is not used in this patch, we may defer to
introduce it in patch 09/44.


> +{
> +	struct perf_cpu_context *cpuctx = this_cpu_ptr(&perf_cpu_context);
> +
> +	lockdep_assert_irqs_disabled();
> +
> +	guard(perf_ctx_lock)(cpuctx, cpuctx->task_ctx);
> +
> +	if (WARN_ON_ONCE(__this_cpu_read(guest_ctx_loaded)))
> +		return;
> +
> +	perf_ctx_disable(&cpuctx->ctx, EVENT_GUEST);
> +	ctx_sched_out(&cpuctx->ctx, NULL, EVENT_GUEST);
> +	if (cpuctx->task_ctx) {
> +		perf_ctx_disable(cpuctx->task_ctx, EVENT_GUEST);
> +		task_ctx_sched_out(cpuctx->task_ctx, NULL, EVENT_GUEST);
> +	}
> +
> +	perf_ctx_enable(&cpuctx->ctx, EVENT_GUEST);
> +	if (cpuctx->task_ctx)
> +		perf_ctx_enable(cpuctx->task_ctx, EVENT_GUEST);
> +
> +	__this_cpu_write(guest_ctx_loaded, true);
> +}
> +EXPORT_SYMBOL_GPL(perf_load_guest_context);
> +
> +void perf_put_guest_context(void)
> +{
> +	struct perf_cpu_context *cpuctx = this_cpu_ptr(&perf_cpu_context);
> +
> +	lockdep_assert_irqs_disabled();
> +
> +	guard(perf_ctx_lock)(cpuctx, cpuctx->task_ctx);
> +
> +	if (WARN_ON_ONCE(!__this_cpu_read(guest_ctx_loaded)))
> +		return;
> +
> +	perf_ctx_disable(&cpuctx->ctx, EVENT_GUEST);
> +	if (cpuctx->task_ctx)
> +		perf_ctx_disable(cpuctx->task_ctx, EVENT_GUEST);
> +
> +	perf_event_sched_in(cpuctx, cpuctx->task_ctx, NULL, EVENT_GUEST);
> +
> +	if (cpuctx->task_ctx)
> +		perf_ctx_enable(cpuctx->task_ctx, EVENT_GUEST);
> +	perf_ctx_enable(&cpuctx->ctx, EVENT_GUEST);
> +
> +	__this_cpu_write(guest_ctx_loaded, false);
> +}
> +EXPORT_SYMBOL_GPL(perf_put_guest_context);
>  #else
>  static int mediated_pmu_account_event(struct perf_event *event) { return 0; }
>  static void mediated_pmu_unaccount_event(struct perf_event *event) {}


  reply	other threads:[~2025-08-08  7:49 UTC|newest]

Thread overview: 65+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-08-06 19:56 [PATCH v5 00/44] KVM: x86: Add support for mediated vPMUs Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 01/44] perf: Skip pmu_ctx based on event_type Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 02/44] perf: Add generic exclude_guest support Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 03/44] perf: Move security_perf_event_free() call to __free_event() Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 04/44] perf: Add APIs to create/release mediated guest vPMUs Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 05/44] perf: Clean up perf ctx time Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 06/44] perf: Add a EVENT_GUEST flag Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 07/44] perf: Add APIs to load/put guest mediated PMU context Sean Christopherson
2025-08-08  7:30   ` Mi, Dapeng [this message]
2025-08-06 19:56 ` [PATCH v5 08/44] perf: core/x86: Register a new vector for handling mediated guest PMIs Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 09/44] perf/x86: Switch LVTPC to/from mediated PMI vector on guest load/put context Sean Christopherson
2025-08-15 11:39   ` Peter Zijlstra
2025-08-15 15:41     ` Sean Christopherson
2025-08-15 15:55       ` Sean Christopherson
2025-08-18 14:32         ` Peter Zijlstra
2025-08-18 15:25           ` Sean Christopherson
2025-08-18 16:12             ` Peter Zijlstra
2025-08-18 20:07               ` Liang, Kan
2025-08-15 13:04   ` Peter Zijlstra
2025-08-15 15:51     ` Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 10/44] perf/x86/core: Do not set bit width for unavailable counters Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 11/44] perf/x86/core: Plumb mediated PMU capability from x86_pmu to x86_pmu_cap Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 12/44] perf/x86/intel: Support PERF_PMU_CAP_MEDIATED_VPMU Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 13/44] perf/x86/amd: Support PERF_PMU_CAP_MEDIATED_VPMU for AMD host Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 14/44] KVM: VMX: Setup canonical VMCS config prior to kvm_x86_vendor_init() Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 15/44] KVM: SVM: Check pmu->version, not enable_pmu, when getting PMC MSRs Sean Christopherson
2025-08-13  9:58   ` Sandipan Das
2025-08-06 19:56 ` [PATCH v5 16/44] KVM: Add a simplified wrapper for registering perf callbacks Sean Christopherson
2025-08-22 10:32   ` Anup Patel
2025-08-06 19:56 ` [PATCH v5 17/44] KVM: x86/pmu: Snapshot host (i.e. perf's) reported PMU capabilities Sean Christopherson
2025-08-13  9:56   ` Sandipan Das
2025-08-06 19:56 ` [PATCH v5 18/44] KVM: x86/pmu: Start stubbing in mediated PMU support Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 19/44] KVM: x86/pmu: Implement Intel mediated PMU requirements and constraints Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 20/44] KVM: x86/pmu: Implement AMD mediated PMU requirements Sean Christopherson
2025-08-13  9:49   ` Sandipan Das
2025-08-06 19:56 ` [PATCH v5 21/44] KVM: x86/pmu: Register PMI handler for mediated vPMU Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 22/44] KVM: x86: Rename vmx_vmentry/vmexit_ctrl() helpers Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 23/44] KVM: x86/pmu: Move PMU_CAP_{FW_WRITES,LBR_FMT} into msr-index.h header Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 24/44] KVM: x86: Rework KVM_REQ_MSR_FILTER_CHANGED into a generic RECALC_INTERCEPTS Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 25/44] KVM: x86: Use KVM_REQ_RECALC_INTERCEPTS to react to CPUID updates Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 26/44] KVM: VMX: Add helpers to toggle/change a bit in VMCS execution controls Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 27/44] KVM: x86/pmu: Disable RDPMC interception for compatible mediated vPMU Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 28/44] KVM: x86/pmu: Load/save GLOBAL_CTRL via entry/exit fields for mediated PMU Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 29/44] KVM: x86/pmu: Use BIT_ULL() instead of open coded equivalents Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 30/44] KVM: x86/pmu: Move initialization of valid PMCs bitmask to common x86 Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 31/44] KVM: x86/pmu: Restrict GLOBAL_{CTRL,STATUS}, fixed PMCs, and PEBS to PMU v2+ Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 32/44] KVM: x86/pmu: Disable interception of select PMU MSRs for mediated vPMUs Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 33/44] KVM: x86/pmu: Bypass perf checks when emulating mediated PMU counter accesses Sean Christopherson
2025-08-13 10:01   ` Sandipan Das
2025-08-06 19:56 ` [PATCH v5 34/44] KVM: x86/pmu: Introduce eventsel_hw to prepare for pmu event filtering Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 35/44] KVM: x86/pmu: Reprogram mediated PMU event selectors on event filter updates Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 36/44] KVM: x86/pmu: Always stuff GuestOnly=1,HostOnly=0 for mediated PMCs on AMD Sean Christopherson
2025-08-06 19:56 ` [PATCH v5 37/44] KVM: x86/pmu: Load/put mediated PMU context when entering/exiting guest Sean Christopherson
2025-08-06 19:57 ` [PATCH v5 38/44] KVM: x86/pmu: Disallow emulation in the fastpath if mediated PMCs are active Sean Christopherson
2025-08-13  9:53   ` Sandipan Das
2025-08-06 19:57 ` [PATCH v5 39/44] KVM: x86/pmu: Handle emulated instruction for mediated vPMU Sean Christopherson
2025-08-06 19:57 ` [PATCH v5 40/44] KVM: nVMX: Add macros to simplify nested MSR interception setting Sean Christopherson
2025-08-06 19:57 ` [PATCH v5 41/44] KVM: nVMX: Disable PMU MSR interception as appropriate while running L2 Sean Christopherson
2025-08-06 19:57 ` [PATCH v5 42/44] KVM: nSVM: " Sean Christopherson
2025-08-06 19:57 ` [PATCH v5 43/44] KVM: x86/pmu: Expose enable_mediated_pmu parameter to user space Sean Christopherson
2025-08-06 19:57 ` [PATCH v5 44/44] KVM: x86/pmu: Elide WRMSRs when loading guest PMCs if values already match Sean Christopherson
2025-08-08  8:28 ` [PATCH v5 00/44] KVM: x86: Add support for mediated vPMUs Mi, Dapeng
2025-08-08  8:35   ` Mi, Dapeng
2025-08-13  9:45 ` Sandipan Das
2025-08-22  8:12 ` Hao, Xudong

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=f3e09bef-8bc9-48ac-983e-7a18c2ff4ad2@linux.intel.com \
    --to=dapeng1.mi@linux.intel.com \
    --cc=acme@kernel.org \
    --cc=anup@brainfault.org \
    --cc=aou@eecs.berkeley.edu \
    --cc=chenhuacai@kernel.org \
    --cc=hpa@zytor.com \
    --cc=kan.liang@linux.intel.com \
    --cc=kvm-riscv@lists.infradead.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=loongarch@lists.linux.dev \
    --cc=luto@kernel.org \
    --cc=maobibo@loongson.cn \
    --cc=maz@kernel.org \
    --cc=mingo@redhat.com \
    --cc=mizhang@google.com \
    --cc=namhyung@kernel.org \
    --cc=oliver.upton@linux.dev \
    --cc=palmer@dabbelt.com \
    --cc=paul.walmsley@sifive.com \
    --cc=pbonzini@redhat.com \
    --cc=peterz@infradead.org \
    --cc=sandipan.das@amd.com \
    --cc=seanjc@google.com \
    --cc=xin@zytor.com \
    --cc=xiong.y.zhang@linux.intel.com \
    --cc=yongwei.ma@intel.com \
    --cc=zhaotianrui@loongson.cn \
    /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).