Re: [PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Raghavendra K T
On 07/11/2012 02:20 AM, Ingo Molnar wrote: * Raghavendra K T wrote: --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1595,6 +1595,9 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me) continue; if (waitqueue_active(>wq))

Re: [PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Ingo Molnar
* Raghavendra K T wrote: > --- a/virt/kvm/kvm_main.c > +++ b/virt/kvm/kvm_main.c > @@ -1595,6 +1595,9 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me) > continue; > if (waitqueue_active(>wq)) > continue; > +

Re: [PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Rik van Riel
On 07/10/2012 03:31 PM, Raghavendra K T wrote: From: Raghavendra K T Currently PLE handler can repeatedly do a directed yield to same vcpu that has recently done PL exit. This can degrade the performance. Try to yield to most eligible guy instead by alternate yielding. Precisely, give chance

[PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Raghavendra K T
From: Raghavendra K T Currently PLE handler can repeatedly do a directed yield to same vcpu that has recently done PL exit. This can degrade the performance. Try to yield to most eligible guy instead by alternate yielding. Precisely, give chance to a VCPU which has: (a) Not done PLE exit at

[PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Raghavendra K T
From: Raghavendra K T raghavendra...@linux.vnet.ibm.com Currently PLE handler can repeatedly do a directed yield to same vcpu that has recently done PL exit. This can degrade the performance. Try to yield to most eligible guy instead by alternate yielding. Precisely, give chance to a VCPU which

Re: [PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Rik van Riel
On 07/10/2012 03:31 PM, Raghavendra K T wrote: From: Raghavendra K Traghavendra...@linux.vnet.ibm.com Currently PLE handler can repeatedly do a directed yield to same vcpu that has recently done PL exit. This can degrade the performance. Try to yield to most eligible guy instead by alternate

Re: [PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Ingo Molnar
* Raghavendra K T raghavendra...@linux.vnet.ibm.com wrote: --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1595,6 +1595,9 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me) continue; if (waitqueue_active(vcpu-wq))

Re: [PATCH RFC V2 2/2] kvm PLE handler: Choose better candidate for directed yield

2012-07-10 Thread Raghavendra K T
On 07/11/2012 02:20 AM, Ingo Molnar wrote: * Raghavendra K Traghavendra...@linux.vnet.ibm.com wrote: --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1595,6 +1595,9 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me) continue; if