public inbox for kvm@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] KVM: Move kvm_exit tracepoint rip reading inside tracepoint
@ 2010-03-11  8:52 Avi Kivity
  2010-03-11 11:03 ` Takuya Yoshikawa
  0 siblings, 1 reply; 3+ messages in thread
From: Avi Kivity @ 2010-03-11  8:52 UTC (permalink / raw)
  To: kvm

Reading rip is expensive on vmx, so move it inside the tracepoint so we only
incur the cost if tracing is enabled.

Signed-off-by: Avi Kivity <avi@redhat.com>
---
 arch/x86/kvm/svm.c   |    2 +-
 arch/x86/kvm/trace.h |    6 +++---
 arch/x86/kvm/vmx.c   |    2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 3a2f2b9..b646e96 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -2685,7 +2685,7 @@ static int handle_exit(struct kvm_vcpu *vcpu)
 	struct kvm_run *kvm_run = vcpu->run;
 	u32 exit_code = svm->vmcb->control.exit_code;
 
-	trace_kvm_exit(exit_code, svm->vmcb->save.rip);
+	trace_kvm_exit(exit_code, vcpu);
 
 	if (unlikely(svm->nested.exit_required)) {
 		nested_svm_vmexit(svm);
diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
index b75efef..3cf9547 100644
--- a/arch/x86/kvm/trace.h
+++ b/arch/x86/kvm/trace.h
@@ -182,8 +182,8 @@ TRACE_EVENT(kvm_apic,
  * Tracepoint for kvm guest exit:
  */
 TRACE_EVENT(kvm_exit,
-	TP_PROTO(unsigned int exit_reason, unsigned long guest_rip),
-	TP_ARGS(exit_reason, guest_rip),
+	    TP_PROTO(unsigned int exit_reason, struct kvm_vcpu *vcpu),
+	TP_ARGS(exit_reason, vcpu),
 
 	TP_STRUCT__entry(
 		__field(	unsigned int,	exit_reason	)
@@ -192,7 +192,7 @@ TRACE_EVENT(kvm_exit,
 
 	TP_fast_assign(
 		__entry->exit_reason	= exit_reason;
-		__entry->guest_rip	= guest_rip;
+		__entry->guest_rip	= kvm_rip_read(vcpu);
 	),
 
 	TP_printk("reason %s rip 0x%lx",
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index ae3217d..06108f3 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -3605,7 +3605,7 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu)
 	u32 exit_reason = vmx->exit_reason;
 	u32 vectoring_info = vmx->idt_vectoring_info;
 
-	trace_kvm_exit(exit_reason, kvm_rip_read(vcpu));
+	trace_kvm_exit(exit_reason, vcpu);
 
 	/* If guest state is invalid, start emulating */
 	if (vmx->emulation_required && emulate_invalid_guest_state)
-- 
1.7.0.2


^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH] KVM: Move kvm_exit tracepoint rip reading inside tracepoint
  2010-03-11  8:52 [PATCH] KVM: Move kvm_exit tracepoint rip reading inside tracepoint Avi Kivity
@ 2010-03-11 11:03 ` Takuya Yoshikawa
  2010-03-11 11:03   ` Avi Kivity
  0 siblings, 1 reply; 3+ messages in thread
From: Takuya Yoshikawa @ 2010-03-11 11:03 UTC (permalink / raw)
  To: Avi Kivity; +Cc: kvm

Avi Kivity wrote:
> diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
> index b75efef..3cf9547 100644
> --- a/arch/x86/kvm/trace.h
> +++ b/arch/x86/kvm/trace.h
> @@ -182,8 +182,8 @@ TRACE_EVENT(kvm_apic,
>   * Tracepoint for kvm guest exit:
>   */
>  TRACE_EVENT(kvm_exit,
> -	TP_PROTO(unsigned int exit_reason, unsigned long guest_rip),
> -	TP_ARGS(exit_reason, guest_rip),
> +	    TP_PROTO(unsigned int exit_reason, struct kvm_vcpu *vcpu),

Whitespaces were inserted by accident?

> +	TP_ARGS(exit_reason, vcpu),
>  
>  	TP_STRUCT__entry(
>  		__field(	unsigned int,	exit_reason	)
> @@ -192,7 +192,7 @@ TRACE_EVENT(kvm_exit,
>  
>  	TP_fast_assign(
>  		__entry->exit_reason	= exit_reason;
> -		__entry->guest_rip	= guest_rip;
> +		__entry->guest_rip	= kvm_rip_read(vcpu);
>  	),
>  
>  	TP_printk("reason %s rip 0x%lx",
> diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
> index ae3217d..06108f3 100644
> --- a/arch/x86/kvm/vmx.c
> +++ b/arch/x86/kvm/vmx.c
> @@ -3605,7 +3605,7 @@ static int vmx_handle_exit(struct kvm_vcpu *vcpu)
>  	u32 exit_reason = vmx->exit_reason;
>  	u32 vectoring_info = vmx->idt_vectoring_info;
>  
> -	trace_kvm_exit(exit_reason, kvm_rip_read(vcpu));
> +	trace_kvm_exit(exit_reason, vcpu);
>  
>  	/* If guest state is invalid, start emulating */
>  	if (vmx->emulation_required && emulate_invalid_guest_state)


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH] KVM: Move kvm_exit tracepoint rip reading inside tracepoint
  2010-03-11 11:03 ` Takuya Yoshikawa
@ 2010-03-11 11:03   ` Avi Kivity
  0 siblings, 0 replies; 3+ messages in thread
From: Avi Kivity @ 2010-03-11 11:03 UTC (permalink / raw)
  To: Takuya Yoshikawa; +Cc: kvm

On 03/11/2010 01:03 PM, Takuya Yoshikawa wrote:
> Avi Kivity wrote:
>   
>> diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h
>> index b75efef..3cf9547 100644
>> --- a/arch/x86/kvm/trace.h
>> +++ b/arch/x86/kvm/trace.h
>> @@ -182,8 +182,8 @@ TRACE_EVENT(kvm_apic,
>>   * Tracepoint for kvm guest exit:
>>   */
>>  TRACE_EVENT(kvm_exit,
>> -	TP_PROTO(unsigned int exit_reason, unsigned long guest_rip),
>> -	TP_ARGS(exit_reason, guest_rip),
>> +	    TP_PROTO(unsigned int exit_reason, struct kvm_vcpu *vcpu),
>>     
> Whitespaces were inserted by accident?
>   

Yeah, already fixed locally.

-- 
error compiling committee.c: too many arguments to function


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2010-03-11 11:03 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-03-11  8:52 [PATCH] KVM: Move kvm_exit tracepoint rip reading inside tracepoint Avi Kivity
2010-03-11 11:03 ` Takuya Yoshikawa
2010-03-11 11:03   ` Avi Kivity

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox