Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-03 Thread Eduardo Valentin
On Fri, Nov 03, 2017 at 11:09:53AM +0100, Paolo Bonzini wrote: > On 02/11/2017 19:43, Eduardo Valentin wrote: > > On Thu, Nov 02, 2017 at 07:24:16PM +0100, Paolo Bonzini wrote: > >> On 02/11/2017 19:08, Eduardo Valentin wrote: > >>> On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: >

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-03 Thread Eduardo Valentin
On Fri, Nov 03, 2017 at 11:09:53AM +0100, Paolo Bonzini wrote: > On 02/11/2017 19:43, Eduardo Valentin wrote: > > On Thu, Nov 02, 2017 at 07:24:16PM +0100, Paolo Bonzini wrote: > >> On 02/11/2017 19:08, Eduardo Valentin wrote: > >>> On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: >

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-03 Thread Paolo Bonzini
On 02/11/2017 19:43, Eduardo Valentin wrote: > On Thu, Nov 02, 2017 at 07:24:16PM +0100, Paolo Bonzini wrote: >> On 02/11/2017 19:08, Eduardo Valentin wrote: >>> On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: On 02/11/2017 18:45, Eduardo Valentin wrote: > Currently, the

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-03 Thread Paolo Bonzini
On 02/11/2017 19:43, Eduardo Valentin wrote: > On Thu, Nov 02, 2017 at 07:24:16PM +0100, Paolo Bonzini wrote: >> On 02/11/2017 19:08, Eduardo Valentin wrote: >>> On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: On 02/11/2017 18:45, Eduardo Valentin wrote: > Currently, the

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
On Thu, Nov 02, 2017 at 07:24:16PM +0100, Paolo Bonzini wrote: > On 02/11/2017 19:08, Eduardo Valentin wrote: > > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: > >> On 02/11/2017 18:45, Eduardo Valentin wrote: > >>> Currently, the existing qspinlock implementation will fallback to

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
On Thu, Nov 02, 2017 at 07:24:16PM +0100, Paolo Bonzini wrote: > On 02/11/2017 19:08, Eduardo Valentin wrote: > > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: > >> On 02/11/2017 18:45, Eduardo Valentin wrote: > >>> Currently, the existing qspinlock implementation will fallback to

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
Longman, On Thu, Nov 02, 2017 at 02:12:13PM -0400, Waiman Long wrote: > On 11/02/2017 02:08 PM, Eduardo Valentin wrote: > > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: > >> On 02/11/2017 18:45, Eduardo Valentin wrote: > >>> Currently, the existing qspinlock implementation will

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
Longman, On Thu, Nov 02, 2017 at 02:12:13PM -0400, Waiman Long wrote: > On 11/02/2017 02:08 PM, Eduardo Valentin wrote: > > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: > >> On 02/11/2017 18:45, Eduardo Valentin wrote: > >>> Currently, the existing qspinlock implementation will

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Paolo Bonzini
On 02/11/2017 19:08, Eduardo Valentin wrote: > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: >> On 02/11/2017 18:45, Eduardo Valentin wrote: >>> Currently, the existing qspinlock implementation will fallback to >>> test-and-set if the hypervisor has not set the PV_UNHALT flag. >>>

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Paolo Bonzini
On 02/11/2017 19:08, Eduardo Valentin wrote: > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: >> On 02/11/2017 18:45, Eduardo Valentin wrote: >>> Currently, the existing qspinlock implementation will fallback to >>> test-and-set if the hypervisor has not set the PV_UNHALT flag. >>>

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Waiman Long
On 11/02/2017 02:08 PM, Eduardo Valentin wrote: > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: >> On 02/11/2017 18:45, Eduardo Valentin wrote: >>> Currently, the existing qspinlock implementation will fallback to >>> test-and-set if the hypervisor has not set the PV_UNHALT flag.

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Waiman Long
On 11/02/2017 02:08 PM, Eduardo Valentin wrote: > On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: >> On 02/11/2017 18:45, Eduardo Valentin wrote: >>> Currently, the existing qspinlock implementation will fallback to >>> test-and-set if the hypervisor has not set the PV_UNHALT flag.

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: > On 02/11/2017 18:45, Eduardo Valentin wrote: > > Currently, the existing qspinlock implementation will fallback to > > test-and-set if the hypervisor has not set the PV_UNHALT flag. > > > > This patch gives the opportunity to guest

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
On Thu, Nov 02, 2017 at 06:56:46PM +0100, Paolo Bonzini wrote: > On 02/11/2017 18:45, Eduardo Valentin wrote: > > Currently, the existing qspinlock implementation will fallback to > > test-and-set if the hypervisor has not set the PV_UNHALT flag. > > > > This patch gives the opportunity to guest

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Paolo Bonzini
On 02/11/2017 18:45, Eduardo Valentin wrote: > Currently, the existing qspinlock implementation will fallback to > test-and-set if the hypervisor has not set the PV_UNHALT flag. > > This patch gives the opportunity to guest kernels to select > between test-and-set and the regular queueu fair lock

Re: [PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Paolo Bonzini
On 02/11/2017 18:45, Eduardo Valentin wrote: > Currently, the existing qspinlock implementation will fallback to > test-and-set if the hypervisor has not set the PV_UNHALT flag. > > This patch gives the opportunity to guest kernels to select > between test-and-set and the regular queueu fair lock

[PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
Currently, the existing qspinlock implementation will fallback to test-and-set if the hypervisor has not set the PV_UNHALT flag. This patch gives the opportunity to guest kernels to select between test-and-set and the regular queueu fair lock implementation based on the PV_DEDICATED KVM feature

[PATCHv2 1/1] locking/qspinlock/x86: Avoid test-and-set when PV_DEDICATED is set

2017-11-02 Thread Eduardo Valentin
Currently, the existing qspinlock implementation will fallback to test-and-set if the hypervisor has not set the PV_UNHALT flag. This patch gives the opportunity to guest kernels to select between test-and-set and the regular queueu fair lock implementation based on the PV_DEDICATED KVM feature