Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

On 11.04.2022 13:00, Julien Grall wrote: > On 11/04/2022 11:45, Jan Beulich wrote: >> On 11.04.2022 12:25, Julien Grall wrote: >>> On 11/04/2022 07:13, Jan Beulich wrote: --- a/xen/common/event_channel.c +++ b/xen/common/event_channel.c @@ -1559,6 +1559,16 @@ void evtchn_

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

Hi, On 11/04/2022 11:45, Jan Beulich wrote: On 11.04.2022 12:25, Julien Grall wrote: On 11/04/2022 07:13, Jan Beulich wrote: --- a/xen/common/event_channel.c +++ b/xen/common/event_channel.c @@ -1559,6 +1559,16 @@ void evtchn_move_pirqs(struct vcpu *v) unsigned int port; stru

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

On 11.04.2022 12:25, Julien Grall wrote: > On 11/04/2022 07:13, Jan Beulich wrote: >> --- a/xen/common/event_channel.c >> +++ b/xen/common/event_channel.c >> @@ -1559,6 +1559,16 @@ void evtchn_move_pirqs(struct vcpu *v) >> unsigned int port; >> struct evtchn *chn

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

Hi, On 11/04/2022 07:13, Jan Beulich wrote: --- a/xen/common/event_channel.c +++ b/xen/common/event_channel.c @@ -1559,6 +1559,16 @@ void evtchn_move_pirqs(struct vcpu *v) unsigned int port; struct evtchn *chn; +/* + * The work done below is an attempt to keep pIRQ-

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

On 08.04.2022 19:17, Julien Grall wrote: > On 08/04/2022 13:25, Jan Beulich wrote: >> On 08.04.2022 13:02, Julien Grall wrote: >>> On 08/04/2022 08:16, Jan Beulich wrote: See the code comment. The higher the rate of vCPU-s migrating across pCPU-s, the less useful this attempted optimizati

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

Hi, On 08/04/2022 13:25, Jan Beulich wrote: On 08.04.2022 13:02, Julien Grall wrote: On 08/04/2022 08:16, Jan Beulich wrote: See the code comment. The higher the rate of vCPU-s migrating across pCPU-s, the less useful this attempted optimization actually is. With credit2 the migration rate loo

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

On 08.04.2022 13:02, Julien Grall wrote: > On 08/04/2022 08:16, Jan Beulich wrote: >> See the code comment. The higher the rate of vCPU-s migrating across >> pCPU-s, the less useful this attempted optimization actually is. With >> credit2 the migration rate looks to be unduly high even on mostly id

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

Hi Jan, On 08/04/2022 08:16, Jan Beulich wrote: See the code comment. The higher the rate of vCPU-s migrating across pCPU-s, the less useful this attempted optimization actually is. With credit2 the migration rate looks to be unduly high even on mostly idle systems, and hence on large systems lo

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

Hi Luca, On 08/04/2022 11:41, Luca Fancellu wrote: On 8 Apr 2022, at 08:16, Jan Beulich wrote: See the code comment. The higher the rate of vCPU-s migrating across pCPU-s, the less useful this attempted optimization actually is. With credit2 the migration rate looks to be unduly high even o

Re: [PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

> On 8 Apr 2022, at 08:16, Jan Beulich wrote: > > See the code comment. The higher the rate of vCPU-s migrating across > pCPU-s, the less useful this attempted optimization actually is. With > credit2 the migration rate looks to be unduly high even on mostly idle > systems, and hence on large s

[PATCH RFC] evtchn: add early-out to evtchn_move_pirqs()

See the code comment. The higher the rate of vCPU-s migrating across pCPU-s, the less useful this attempted optimization actually is. With credit2 the migration rate looks to be unduly high even on mostly idle systems, and hence on large systems lock contention here isn't very difficult to observe.