From mboxrd@z Thu Jan  1 00:00:00 1970
From: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
Subject: Re: [PATCH RFC V4 3/3] kvm: Choose better candidate for directed
 yield
Date: Mon, 16 Jul 2012 22:37:32 +0530
Message-ID: <50044A54.2090005@linux.vnet.ibm.com>
References: <20120716082445.23477.15128.sendpatchset@codeblue.in.ibm.com> <20120716082529.23477.91096.sendpatchset@codeblue.in.ibm.com> <5003E7ED.2030701@redhat.com> <50043CF3.8020903@redhat.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Cc: "H. Peter Anvin" <hpa@zytor.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	Srikar <srikar@linux.vnet.ibm.com>,
	S390 <linux-s390@vger.kernel.org>,
	Carsten Otte <cotte@de.ibm.com>,
	Christian Borntraeger <borntraeger@de.ibm.com>,
	KVM <kvm@vger.kernel.org>, chegu vinod <chegu_vinod@hp.com>,
	"Andrew M. Theurer" <habanero@linux.vnet.ibm.com>,
	LKML <linux-kernel@vger.kernel.org>, X86 <x86@kernel.org>,
	Gleb Natapov <gleb@redhat.com>, linux390@de.ibm.com,
	Srivatsa Vaddagiri <srivatsa.vaddagiri@gmail.com>,
	Joerg Roedel <joerg.roedel@amd.com>
To: Rik van Riel <riel@redhat.com>, Avi Kivity <avi@redhat.com>
Return-path: <linux-kernel-owner@vger.kernel.org>
In-Reply-To: <50043CF3.8020903@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
List-Id: kvm.vger.kernel.org

On 07/16/2012 09:40 PM, Rik van Riel wrote:
> On 07/16/2012 06:07 AM, Avi Kivity wrote:
>
>>> +{
>>> + bool eligible;
>>> +
>>> + eligible = !vcpu->ple.cpu_relax_intercepted ||
>>> + (vcpu->ple.cpu_relax_intercepted&&
>>> + vcpu->ple.dy_eligible);
>>> +
>>> + if (vcpu->ple.cpu_relax_intercepted)
>>> + vcpu->ple.dy_eligible = !vcpu->ple.dy_eligible;
>>
>> Probably should assign 'true', since the previous value is essentially
>> random.
>
> I suspect the intended purpose of this conditional is to
> flip the eligibility of a vcpu for being selected as a
> direct yield target.
>
> In other words, that bit of the code is correct.

Yes, That is the intention.

The first intention was to make sure, not to select previously selected
pause loop exited guy.
But second intention was not to make a vcpu permanently eligible after
skipping once.

The problem is when several PL exits happen simultaneously, (most
possible in large vcpu guest), it is very much probable that same vcpu
is tried as target of directed yield slowing down yielding to a
eligible guy.

But I 'll test one more time to make sure that.

(if we can make dy_eligible true as suggested, that means probably we 
can live with only one bool variable.)