Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-13 16:42 GMT+08:00 Ingo Molnar : > > * Rik van Riel wrote: > >> On Wed, 10 Aug 2016 07:39:08 +0800 >> Wanpeng Li wrote: >> >> > The regression is caused by your commit "sched,time: Count actually >> > elapsed irq & softirq time". >>

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-13 16:42 GMT+08:00 Ingo Molnar : > > * Rik van Riel wrote: > >> On Wed, 10 Aug 2016 07:39:08 +0800 >> Wanpeng Li wrote: >> >> > The regression is caused by your commit "sched,time: Count actually >> > elapsed irq & softirq time". >> >> Wanpeng, does this patch fix your issue? >> >>

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-16 22:01 GMT+08:00 Rik van Riel : > On Tue, 2016-08-16 at 14:54 +0800, Wanpeng Li wrote: >> 2016-08-16 10:11 GMT+08:00 Rik van Riel : >> > On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: >> > > 2016-08-15 23:00 GMT+08:00 Rik van Riel

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-16 22:01 GMT+08:00 Rik van Riel : > On Tue, 2016-08-16 at 14:54 +0800, Wanpeng Li wrote: >> 2016-08-16 10:11 GMT+08:00 Rik van Riel : >> > On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: >> > > 2016-08-15 23:00 GMT+08:00 Rik van Riel : >> > > > On Mon, 2016-08-15 at 16:53 +0800,

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Tue, 2016-08-16 at 14:54 +0800, Wanpeng Li wrote: > 2016-08-16 10:11 GMT+08:00 Rik van Riel : > > On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: > > > 2016-08-15 23:00 GMT+08:00 Rik van Riel : > > > > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Tue, 2016-08-16 at 14:54 +0800, Wanpeng Li wrote: > 2016-08-16 10:11 GMT+08:00 Rik van Riel : > > On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: > > > 2016-08-15 23:00 GMT+08:00 Rik van Riel : > > > > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: > > > > > 2016-08-12 23:58

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-16 10:11 GMT+08:00 Rik van Riel : > On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: >> 2016-08-15 23:00 GMT+08:00 Rik van Riel : >> > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >> > > 2016-08-12 23:58 GMT+08:00 Rik van Riel

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-16 10:11 GMT+08:00 Rik van Riel : > On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: >> 2016-08-15 23:00 GMT+08:00 Rik van Riel : >> > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >> > > 2016-08-12 23:58 GMT+08:00 Rik van Riel : >> > > [...] >> > > > Wanpeng, does the patch

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: > 2016-08-15 23:00 GMT+08:00 Rik van Riel : > > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: > > > 2016-08-12 23:58 GMT+08:00 Rik van Riel : > > > [...] > > > > Wanpeng, does the patch below work for

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Tue, 2016-08-16 at 09:31 +0800, Wanpeng Li wrote: > 2016-08-15 23:00 GMT+08:00 Rik van Riel : > > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: > > > 2016-08-12 23:58 GMT+08:00 Rik van Riel : > > > [...] > > > > Wanpeng, does the patch below work for you? > > > > > > It will break steal

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-15 23:00 GMT+08:00 Rik van Riel : > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >> 2016-08-12 23:58 GMT+08:00 Rik van Riel : >> [...] >> > Wanpeng, does the patch below work for you? >> >> It will break steal time for full dynticks guest, and

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-15 23:00 GMT+08:00 Rik van Riel : > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >> 2016-08-12 23:58 GMT+08:00 Rik van Riel : >> [...] >> > Wanpeng, does the patch below work for you? >> >> It will break steal time for full dynticks guest, and there is a >> calltrace of

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-15 23:00 GMT+08:00 Rik van Riel : > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >> 2016-08-12 23:58 GMT+08:00 Rik van Riel : >> [...] >> > Wanpeng, does the patch below work for you? >> >> It will break steal time for full dynticks guest, and

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-15 23:00 GMT+08:00 Rik van Riel : > On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: >> 2016-08-12 23:58 GMT+08:00 Rik van Riel : >> [...] >> > Wanpeng, does the patch below work for you? >> >> It will break steal time for full dynticks guest, and there is a >> calltrace of

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: > 2016-08-12 23:58 GMT+08:00 Rik van Riel : > [...] > > Wanpeng, does the patch below work for you? > > It will break steal time for full dynticks guest, and there is a > calltrace of thread_group_cputime_adjusted call stack,

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Mon, 2016-08-15 at 16:53 +0800, Wanpeng Li wrote: > 2016-08-12 23:58 GMT+08:00 Rik van Riel : > [...] > > Wanpeng, does the patch below work for you? > > It will break steal time for full dynticks guest, and there is a > calltrace of thread_group_cputime_adjusted call stack, RIP is >

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-15 16:53 GMT+08:00 Wanpeng Li : > 2016-08-12 23:58 GMT+08:00 Rik van Riel : > [...] >> Wanpeng, does the patch below work for you? > > It will break steal time for full dynticks guest, and there is a > calltrace of thread_group_cputime_adjusted call

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-15 16:53 GMT+08:00 Wanpeng Li : > 2016-08-12 23:58 GMT+08:00 Rik van Riel : > [...] >> Wanpeng, does the patch below work for you? > > It will break steal time for full dynticks guest, and there is a > calltrace of thread_group_cputime_adjusted call stack, RIP is >

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-12 23:58 GMT+08:00 Rik van Riel : [...] > Wanpeng, does the patch below work for you? It will break steal time for full dynticks guest, and there is a calltrace of thread_group_cputime_adjusted call stack, RIP is cputime_adjust+0xff/0x130. > > Everybody else, does this

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-12 23:58 GMT+08:00 Rik van Riel : [...] > Wanpeng, does the patch below work for you? It will break steal time for full dynticks guest, and there is a calltrace of thread_group_cputime_adjusted call stack, RIP is cputime_adjust+0xff/0x130. > > Everybody else, does this patch look

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Sat, 2016-08-13 at 10:42 +0200, Ingo Molnar wrote: > * Rik van Riel wrote: > > > On Wed, 10 Aug 2016 07:39:08 +0800 > > Wanpeng Li wrote: > > > > > The regression is caused by your commit "sched,time: Count > > > actually > > > elapsed irq & softirq

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Sat, 2016-08-13 at 10:42 +0200, Ingo Molnar wrote: > * Rik van Riel wrote: > > > On Wed, 10 Aug 2016 07:39:08 +0800 > > Wanpeng Li wrote: > > > > > The regression is caused by your commit "sched,time: Count > > > actually > > > elapsed irq & softirq time". > > > > Wanpeng, does this patch

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Fri, Aug 12, 2016 at 11:58:03AM -0400, Rik van Riel wrote: > On Fri, 12 Aug 2016 15:09:00 +0800 > Wanpeng Li wrote: > > 2016-08-12 10:44 GMT+08:00 Rik van Riel : > > > > If you pass ULONG_MAX as the maxtime argument to > > > steal_account_process_time(),

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Fri, Aug 12, 2016 at 11:58:03AM -0400, Rik van Riel wrote: > On Fri, 12 Aug 2016 15:09:00 +0800 > Wanpeng Li wrote: > > 2016-08-12 10:44 GMT+08:00 Rik van Riel : > > > > If you pass ULONG_MAX as the maxtime argument to > > > steal_account_process_time(), does the steal time > > > get

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

* Rik van Riel wrote: > On Wed, 10 Aug 2016 07:39:08 +0800 > Wanpeng Li wrote: > > > The regression is caused by your commit "sched,time: Count actually > > elapsed irq & softirq time". > > Wanpeng, does this patch fix your issue? > > Paolo, what is

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

* Rik van Riel wrote: > On Wed, 10 Aug 2016 07:39:08 +0800 > Wanpeng Li wrote: > > > The regression is caused by your commit "sched,time: Count actually > > elapsed irq & softirq time". > > Wanpeng, does this patch fix your issue? > > Paolo, what is your opinion on this issue? > > I can

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

> There is one copy of paravirt_steal_clock(smp_processor_id()), > but what keeps it in sync with this_rq()->prev_steal_time? > > Is it something simple like them both being zeroed out when > the structures are first allocated at boot time? Yes, more precisely both of them being equal when the

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

> There is one copy of paravirt_steal_clock(smp_processor_id()), > but what keeps it in sync with this_rq()->prev_steal_time? > > Is it something simple like them both being zeroed out when > the structures are first allocated at boot time? Yes, more precisely both of them being equal when the

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Fri, 2016-08-12 at 18:33 +0200, Paolo Bonzini wrote: > > On 10/08/2016 18:52, Rik van Riel wrote: > > Paolo, what is your opinion on this issue? > > > > I can think of all kinds of ways in which guest and host might lose > > sync with steal time, from uninitialized values at boot, to guest >

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Fri, 2016-08-12 at 18:33 +0200, Paolo Bonzini wrote: > > On 10/08/2016 18:52, Rik van Riel wrote: > > Paolo, what is your opinion on this issue? > > > > I can think of all kinds of ways in which guest and host might lose > > sync with steal time, from uninitialized values at boot, to guest >

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On 10/08/2016 18:52, Rik van Riel wrote: > Paolo, what is your opinion on this issue? > > I can think of all kinds of ways in which guest and host might lose > sync with steal time, from uninitialized values at boot, to guest > pause, followed by save to disk, and reload, to live migration,

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On 10/08/2016 18:52, Rik van Riel wrote: > Paolo, what is your opinion on this issue? > > I can think of all kinds of ways in which guest and host might lose > sync with steal time, from uninitialized values at boot, to guest > pause, followed by save to disk, and reload, to live migration,

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Fri, 12 Aug 2016 15:09:00 +0800 Wanpeng Li wrote: > 2016-08-12 10:44 GMT+08:00 Rik van Riel : > > If you pass ULONG_MAX as the maxtime argument to > > steal_account_process_time(), does the steal time > > get accounted properly at 75%? > > Yes. I

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Fri, 12 Aug 2016 15:09:00 +0800 Wanpeng Li wrote: > 2016-08-12 10:44 GMT+08:00 Rik van Riel : > > If you pass ULONG_MAX as the maxtime argument to > > steal_account_process_time(), does the steal time > > get accounted properly at 75%? > > Yes. I talked with Paolo this morning, and it

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-12 10:44 GMT+08:00 Rik van Riel : > On Thu, 2016-08-11 at 18:11 +0800, Wanpeng Li wrote: >> 2016-08-11 0:52 GMT+08:00 Rik van Riel : >> > On Wed, 10 Aug 2016 07:39:08 +0800 >> > Wanpeng Li wrote: >> > >> > > The regression is caused

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-12 10:44 GMT+08:00 Rik van Riel : > On Thu, 2016-08-11 at 18:11 +0800, Wanpeng Li wrote: >> 2016-08-11 0:52 GMT+08:00 Rik van Riel : >> > On Wed, 10 Aug 2016 07:39:08 +0800 >> > Wanpeng Li wrote: >> > >> > > The regression is caused by your commit "sched,time: Count >> > > actually >> > >

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Thu, 2016-08-11 at 18:11 +0800, Wanpeng Li wrote: > 2016-08-11 0:52 GMT+08:00 Rik van Riel : > > On Wed, 10 Aug 2016 07:39:08 +0800 > > Wanpeng Li wrote: > > > > > The regression is caused by your commit "sched,time: Count > > > actually > > > elapsed irq

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

On Thu, 2016-08-11 at 18:11 +0800, Wanpeng Li wrote: > 2016-08-11 0:52 GMT+08:00 Rik van Riel : > > On Wed, 10 Aug 2016 07:39:08 +0800 > > Wanpeng Li wrote: > > > > > The regression is caused by your commit "sched,time: Count > > > actually > > > elapsed irq & softirq time". > > > > Wanpeng,

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-11 0:52 GMT+08:00 Rik van Riel : > On Wed, 10 Aug 2016 07:39:08 +0800 > Wanpeng Li wrote: > >> The regression is caused by your commit "sched,time: Count actually >> elapsed irq & softirq time". > > Wanpeng, does this patch fix your issue? I test

Re: [PATCH] time,virt: resync steal time when guest & host lose sync

2016-08-11 0:52 GMT+08:00 Rik van Riel : > On Wed, 10 Aug 2016 07:39:08 +0800 > Wanpeng Li wrote: > >> The regression is caused by your commit "sched,time: Count actually >> elapsed irq & softirq time". > > Wanpeng, does this patch fix your issue? I test this against kvm guest (nohz_full, four

[PATCH] time,virt: resync steal time when guest & host lose sync

On Wed, 10 Aug 2016 07:39:08 +0800 Wanpeng Li wrote: > The regression is caused by your commit "sched,time: Count actually > elapsed irq & softirq time". Wanpeng, does this patch fix your issue? Paolo, what is your opinion on this issue? I can think of all kinds of ways

[PATCH] time,virt: resync steal time when guest & host lose sync

On Wed, 10 Aug 2016 07:39:08 +0800 Wanpeng Li wrote: > The regression is caused by your commit "sched,time: Count actually > elapsed irq & softirq time". Wanpeng, does this patch fix your issue? Paolo, what is your opinion on this issue? I can think of all kinds of ways in which guest and