All of lore.kernel.org
 help / color / mirror / Atom feed
From: Punit Agrawal <punit.agrawal@arm.com>
To: Christoffer Dall <christoffer.dall@linaro.org>
Cc: kvm@vger.kernel.org, Marc Zyngier <marc.zyngier@arm.com>,
	Will Deacon <will.deacon@arm.com>,
	linux-kernel@vger.kernel.org,
	Steven Rostedt <rostedt@goodmis.org>,
	Ingo Molnar <mingo@redhat.com>,
	kvmarm@lists.cs.columbia.edu,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [RFC v2 PATCH 7/7] arm64: KVM: Enable selective trapping of TLB instructions
Date: Tue, 06 Sep 2016 12:33:29 +0100	[thread overview]
Message-ID: <871t0xjm9i.fsf@e105922-lin.cambridge.arm.com> (raw)
In-Reply-To: <20160906102404.GF30513@cbox> (Christoffer Dall's message of "Tue, 6 Sep 2016 12:24:04 +0200")

Christoffer Dall <christoffer.dall@linaro.org> writes:

> On Mon, Sep 05, 2016 at 05:31:37PM +0100, Punit Agrawal wrote:
>> The TTLB bit of Hypervisor Control Register (HCR_EL2) controls the
>> trapping of guest TLB maintenance instructions. Taking the trap requires
>> a switch to the hypervisor and is an expensive operation.
>> 
>> Enable selective trapping of guest TLB instructions when the associated
>> perf trace event is enabled for a specific virtual machine.
>> 
>> Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>
>> Cc: Christoffer Dall <christoffer.dall@linaro.org>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>>  arch/arm64/kvm/perf_trace.c | 32 ++++++++++++++++++++++++++++++++
>>  1 file changed, 32 insertions(+)
>> 
>> diff --git a/arch/arm64/kvm/perf_trace.c b/arch/arm64/kvm/perf_trace.c
>> index 8bacd18..f26da1d 100644
>> --- a/arch/arm64/kvm/perf_trace.c
>> +++ b/arch/arm64/kvm/perf_trace.c
>> @@ -17,6 +17,8 @@
>>  #include <linux/kvm_host.h>
>>  #include <linux/trace_events.h>
>>  
>> +#include <asm/kvm_emulate.h>
>> +
>>  typedef int (*perf_trace_callback_fn)(struct kvm *kvm, bool enable);
>>  
>>  struct kvm_trace_hook {
>> @@ -24,7 +26,37 @@ struct kvm_trace_hook {
>>  	perf_trace_callback_fn setup_fn;
>>  };
>>  
>> +static int tlb_invalidate_trap(struct kvm *kvm, bool enable)
>> +{
>> +	int i;
>> +	struct kvm_vcpu *vcpu;
>> +
>> +	/*
>> +	 * Halt the VM to ensure atomic update across all vcpus (this
>> +	 * avoids racy behaviour against other modifications of
>> +	 * HCR_EL2 such as kvm_toggle_cache/kvm_set_way_flush).
>> +	 */
>> +	kvm_arm_halt_guest(kvm);
>> +	kvm_for_each_vcpu(i, vcpu, kvm) {
>> +		unsigned long hcr = vcpu_get_hcr(vcpu);
>> +
>> +		if (enable)
>> +			hcr |= HCR_TTLB;
>> +		else
>> +			hcr &= ~HCR_TTLB;
>> +
>> +		vcpu_set_hcr(vcpu, hcr);
>> +	}
>> +	kvm_arm_resume_guest(kvm);
>> +
>> +	return 0;
>> +}
>> +
>>  static struct kvm_trace_hook trace_hook[] = {
>> +	{
>> +		.key = "kvm_tlb_invalidate",
>
> is this key defined elsewhere?  If not, I think the name is ambiguous,
> because it's unclear if this means 'kvm the subsystem' or 'a kvm guest',
> by looking purely at the string.

The key is the name of the tracepoint that the callback is executed
against. Here it is the tracepoint defined in Patch 6/7.

I've now added a comment to the definition of "struct kvm_trace_hook" to
explain what the key is.

>
>> +		.setup_fn = tlb_invalidate_trap,
>> +	},
>>  	{ },
>>  };
>>  
>> -- 
>> 2.8.1
>> 
>
> Otherwise looks ok to me.
>
> -Christoffer
> _______________________________________________
> kvmarm mailing list
> kvmarm@lists.cs.columbia.edu
> https://lists.cs.columbia.edu/mailman/listinfo/kvmarm

WARNING: multiple messages have this Message-ID (diff)
From: punit.agrawal@arm.com (Punit Agrawal)
To: linux-arm-kernel@lists.infradead.org
Subject: [RFC v2 PATCH 7/7] arm64: KVM: Enable selective trapping of TLB instructions
Date: Tue, 06 Sep 2016 12:33:29 +0100	[thread overview]
Message-ID: <871t0xjm9i.fsf@e105922-lin.cambridge.arm.com> (raw)
In-Reply-To: <20160906102404.GF30513@cbox> (Christoffer Dall's message of "Tue, 6 Sep 2016 12:24:04 +0200")

Christoffer Dall <christoffer.dall@linaro.org> writes:

> On Mon, Sep 05, 2016 at 05:31:37PM +0100, Punit Agrawal wrote:
>> The TTLB bit of Hypervisor Control Register (HCR_EL2) controls the
>> trapping of guest TLB maintenance instructions. Taking the trap requires
>> a switch to the hypervisor and is an expensive operation.
>> 
>> Enable selective trapping of guest TLB instructions when the associated
>> perf trace event is enabled for a specific virtual machine.
>> 
>> Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>
>> Cc: Christoffer Dall <christoffer.dall@linaro.org>
>> Cc: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>>  arch/arm64/kvm/perf_trace.c | 32 ++++++++++++++++++++++++++++++++
>>  1 file changed, 32 insertions(+)
>> 
>> diff --git a/arch/arm64/kvm/perf_trace.c b/arch/arm64/kvm/perf_trace.c
>> index 8bacd18..f26da1d 100644
>> --- a/arch/arm64/kvm/perf_trace.c
>> +++ b/arch/arm64/kvm/perf_trace.c
>> @@ -17,6 +17,8 @@
>>  #include <linux/kvm_host.h>
>>  #include <linux/trace_events.h>
>>  
>> +#include <asm/kvm_emulate.h>
>> +
>>  typedef int (*perf_trace_callback_fn)(struct kvm *kvm, bool enable);
>>  
>>  struct kvm_trace_hook {
>> @@ -24,7 +26,37 @@ struct kvm_trace_hook {
>>  	perf_trace_callback_fn setup_fn;
>>  };
>>  
>> +static int tlb_invalidate_trap(struct kvm *kvm, bool enable)
>> +{
>> +	int i;
>> +	struct kvm_vcpu *vcpu;
>> +
>> +	/*
>> +	 * Halt the VM to ensure atomic update across all vcpus (this
>> +	 * avoids racy behaviour against other modifications of
>> +	 * HCR_EL2 such as kvm_toggle_cache/kvm_set_way_flush).
>> +	 */
>> +	kvm_arm_halt_guest(kvm);
>> +	kvm_for_each_vcpu(i, vcpu, kvm) {
>> +		unsigned long hcr = vcpu_get_hcr(vcpu);
>> +
>> +		if (enable)
>> +			hcr |= HCR_TTLB;
>> +		else
>> +			hcr &= ~HCR_TTLB;
>> +
>> +		vcpu_set_hcr(vcpu, hcr);
>> +	}
>> +	kvm_arm_resume_guest(kvm);
>> +
>> +	return 0;
>> +}
>> +
>>  static struct kvm_trace_hook trace_hook[] = {
>> +	{
>> +		.key = "kvm_tlb_invalidate",
>
> is this key defined elsewhere?  If not, I think the name is ambiguous,
> because it's unclear if this means 'kvm the subsystem' or 'a kvm guest',
> by looking purely at the string.

The key is the name of the tracepoint that the callback is executed
against. Here it is the tracepoint defined in Patch 6/7.

I've now added a comment to the definition of "struct kvm_trace_hook" to
explain what the key is.

>
>> +		.setup_fn = tlb_invalidate_trap,
>> +	},
>>  	{ },
>>  };
>>  
>> -- 
>> 2.8.1
>> 
>
> Otherwise looks ok to me.
>
> -Christoffer
> _______________________________________________
> kvmarm mailing list
> kvmarm at lists.cs.columbia.edu
> https://lists.cs.columbia.edu/mailman/listinfo/kvmarm

  reply	other threads:[~2016-09-06 11:34 UTC|newest]

Thread overview: 76+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-05 16:31 [RFC v2 PATCH 0/7] Add support for monitoring guest TLB operations Punit Agrawal
2016-09-05 16:31 ` Punit Agrawal
2016-09-05 16:31 ` Punit Agrawal
2016-09-05 16:31 ` [RFC v2 PATCH 1/7] perf/trace: Add notification for perf trace events Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31 ` [RFC v2 PATCH 2/7] KVM: Track the pid of the VM process Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-06  6:22   ` Christoffer Dall
2016-09-06  6:22     ` Christoffer Dall
2016-09-06  6:22     ` Christoffer Dall
2016-09-06  9:51     ` Punit Agrawal
2016-09-06  9:51       ` Punit Agrawal
2016-09-06  9:51       ` Punit Agrawal
2016-09-06 10:25       ` Christoffer Dall
2016-09-06 10:25         ` Christoffer Dall
2016-09-06 10:25         ` Christoffer Dall
2016-09-06 11:07         ` Punit Agrawal
2016-09-06 11:07           ` Punit Agrawal
2016-09-06 11:07           ` Punit Agrawal
2016-09-06 11:22           ` Christoffer Dall
2016-09-06 11:22             ` Christoffer Dall
2016-09-06 15:22             ` Punit Agrawal
2016-09-06 15:22               ` Punit Agrawal
2016-09-06 15:22               ` Punit Agrawal
2016-09-06 16:57               ` Christoffer Dall
2016-09-06 16:57                 ` Christoffer Dall
2016-09-06 17:03                 ` Punit Agrawal
2016-09-06 17:03                   ` Punit Agrawal
2016-09-06 17:03                   ` Punit Agrawal
2016-09-05 16:31 ` [RFC v2 PATCH 3/7] KVM: arm/arm64: Register perf trace event notifier Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-06  6:36   ` Christoffer Dall
2016-09-06  6:36     ` Christoffer Dall
2016-09-06  6:36     ` Christoffer Dall
2016-09-06 16:10     ` Punit Agrawal
2016-09-06 16:10       ` Punit Agrawal
2016-09-05 16:31 ` [RFC v2 PATCH 4/7] arm64: tlbflush.h: add __tlbi() macro Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-06  6:38   ` Christoffer Dall
2016-09-06  6:38     ` Christoffer Dall
2016-09-06  6:38     ` Christoffer Dall
2016-09-06 10:05     ` Punit Agrawal
2016-09-06 10:05       ` Punit Agrawal
2016-09-06 10:39       ` Christoffer Dall
2016-09-06 10:39         ` Christoffer Dall
2016-09-06 18:17   ` Will Deacon
2016-09-06 18:17     ` Will Deacon
2016-09-06 18:17     ` Will Deacon
2016-09-05 16:31 ` [RFC v2 PATCH 5/7] arm64/kvm: hyp: tlb: use __tlbi() helper Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-06  6:39   ` Christoffer Dall
2016-09-06  6:39     ` Christoffer Dall
2016-09-06  6:39     ` Christoffer Dall
2016-09-05 16:31 ` [RFC v2 PATCH 6/7] arm64: KVM: Handle trappable TLB instructions Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-06 10:21   ` Christoffer Dall
2016-09-06 10:21     ` Christoffer Dall
2016-09-06 10:21     ` Christoffer Dall
2016-09-06 15:44     ` Punit Agrawal
2016-09-06 15:44       ` Punit Agrawal
2016-09-06 16:59       ` Christoffer Dall
2016-09-06 16:59         ` Christoffer Dall
2016-09-05 16:31 ` [RFC v2 PATCH 7/7] arm64: KVM: Enable selective trapping of " Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-05 16:31   ` Punit Agrawal
2016-09-06 10:24   ` Christoffer Dall
2016-09-06 10:24     ` Christoffer Dall
2016-09-06 10:24     ` Christoffer Dall
2016-09-06 11:33     ` Punit Agrawal [this message]
2016-09-06 11:33       ` Punit Agrawal

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=871t0xjm9i.fsf@e105922-lin.cambridge.arm.com \
    --to=punit.agrawal@arm.com \
    --cc=christoffer.dall@linaro.org \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.cs.columbia.edu \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=marc.zyngier@arm.com \
    --cc=mingo@redhat.com \
    --cc=rostedt@goodmis.org \
    --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 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.