xen: sched: fix locking of {insert, remove}_vcpu()
cover xen: sched: fix locking of {insert, remove}_vcpu()
Hi everyone,
This series originates from this other one:
http://lists.xen.org/archives/html/xen-devel/2015-09/msg03811.html
And, more specifically, from the review subthread of patch 3 (Jan's comments, mainly):
http://lists.xen.org/archives/html/xen-devel/2015-09/msg03812.html
http://lists.xen.org/archives/html/xen-devel/2015-10/msg00023.html
In fact, when reworking it, according to the review comments I got, I thought that the locking issues in insert_vcpu() and remove_vcpu() could well deserve a series on their own.
Both the paths (vCPU insertion and removel) suffer from locking related bugs, which are being fixed by patches 1 and 2 of this series. Without these two patches, sched_move_domain() ends up doing some runqueue manipulation _without_ holding the approaprite lock, for some schedulers (which include Credit).
I haven't seen the bug trigger. Actually, I haven't tried much to mess things up by means of it, TBH, but, in principle, it's possible. Patches 1 and 2 are, therefore, both candidates for being backported, I guess.
The rest of the series is basically cleanups. Patch 3 tries to make that whole 'insert vCPU' situation more clear, consistent, and easy to understand.
Patch 4 and 5 simplify both Credit2 and RTDS schedulers a (really really) small bit, by removing the (non necessary) list of vCPUs from the scheduling specific domain data structure, and, of course, the burden of having to manage it.
Patch 6 and 7 are, literally, (dump) output cleanups.
There is a branch available with this series applied at:
git://xenbits.xen.org/people/dariof/xen.git rel/sched/fix-vcpu-ins-rem
http://xenbits.xen.org/gitweb/?p=people/dariof/xen.git;a=shortlog;h=refs/heads/rel/sched/fix-vcpu-ins-rem
Regards,
Dario
---
Dario Faggioli (7):
xen: sched: fix locking of remove_vcpu() in credit1
xen: sched: fix locking for insert_vcpu() in credit1 and RTDS
xen: sched: better handle (not) inserting idle vCPUs in runqueues
xen: sched: get rid of the per domain vCPU list in RTDS
xen: sched: get rid of the per domain vCPU list in Credit2
xen: sched: fix an 'off by one \t' in credit2 debug dump
xen: sched / cpupool: dump the actual value of NOW()
xen/common/cpupool.c | 2 +- xen/common/sched_credit.c | 17 +++++++++++-- xen/common/sched_credit2.c | 57 +++++++++++++++----------------------------- xen/common/sched_rt.c | 43 +++++++++++++-------------------- xen/common/schedule.c | 27 ++++++++------------- 5 files changed, 63 insertions(+), 83 deletions(-)
--
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
PATCH 1/7 xen: sched: fix locking of remove_vcpu() in credit1
In fact, csched_vcpu_remove() (i.e., the credit1 implementation of remove_vcpu()) manipulates runqueues, so holding the runqueue lock is necessary.
Also, while there, *_lock_irq() (for the private lock) is enough, there is no need to *_lock_irqsave().
---
Cc: George Dunlap <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: Jan Beulich <[email protected]>
---
Changes from v1 (within the other series):
* split the patch (wrt the original patch, in the original series), and take care, in this one, only of remove_vcpu();
* removed pointless parentheses.
--- xen/common/sched_credit.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
xen/common/sched_credit.c +7 -3
@@ -926,7 +926,7 @@ csched_vcpu_remove(const struct scheduler *ops, struct vcpu *vc)struct csched_private *prv =CSCHED_PRIV (ops);struct csched_vcpu *const svc =CSCHED_VCPU (vc);struct csched_dom *const sdom = svc->sdom; -unsigned long flags; +spinlock_t *lock;SCHED_STAT_CRANK (vcpu_remove);
@@ -936,15 +936,19 @@ csched_vcpu_remove(const struct scheduler *ops, struct vcpu *vc)vcpu_unpause (svc->vcpu); } + lock =vcpu_schedule_lock_irq (vc); +if (__vcpu_on_runq (svc) )__runq_remove (svc); -spin_lock_irqsave (&(prv->lock), flags); +vcpu_schedule_unlock_irq (lock, vc); + +spin_lock_irq (&prv->lock);if ( !list_empty (&svc->active_vcpu_elem) )__csched_vcpu_acct_stop_locked (prv, svc); -spin_unlock_irqrestore (&(prv->lock), flags); +spin_unlock_irq (&prv->lock);BUG_ON ( sdom ==NULL );BUG_ON ( !list_empty (&svc->runq_elem) );
Re: [PATCH 1/7] xen: sched: fix locking of remove_vcpu() in credit1
Quoted 8 lines · Dario Faggioli
In fact,
csched_vcpu_remove()(i.e., the credit1 implementation ofremove_vcpu()) manipulates runqueues, so holding the runqueue lock is necessary.Also, while there,
*_lock_irq()(for the private lock) is enough, there is no need to*_lock_irqsave().Signed-off-by Dario Faggioli <[email protected]>
PATCH 2/7 xen: sched: fix locking for insert_vcpu() in credit1 and RTDS
In reply to Dario FaggioliThe insert_vcpu scheduler hook is called with an inconsistent locking strategy. In fact, it is sometimes invoked while holding the runqueue lock and sometimes when that is not the case.
For instance, in case of schedule_cpu_switch() the lock is acquired in generic code. On the other hand, in case of sched_move_domain(), locking is left as a responsibility of the schedulers implementing the hook.
This results in Credit1 and RTDS schedulers ending up (in case of sched_move_domain()) doing runqueue manipulation without holding any runqueue lock, which is a bug. (Credit2 was doing the locking by itself already.)
The right thing is to defer locking to the specific schedulers, as it's them that know what, how and when it is best to lock (as in: runqueue locks, vs. private scheduler locks, vs. both, etc.).
This patch, therefore:
- removes any locking around insert_vcpu() from generic code;
- add proper locking in the hook implementations, for both Credit1 and RTDS.
---
Cc: George Dunlap <[email protected]>
Cc: Meng Xu <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: Jan Beulich <[email protected]>
---
Changes from v1 (within the other series):
* split the patch (wrt the original patch, in the original series), and take care, in this one, only of insert_vcpu();
--- xen/common/sched_credit.c | 5 +++++ xen/common/sched_rt.c | 3 +++ xen/common/schedule.c | 6 ------ 3 files changed, 8 insertions(+), 6 deletions(-)
xen/common/sched_credit.c +5 -0
@@ -903,10 +903,15 @@ static void csched_vcpu_insert(const struct scheduler *ops, struct vcpu *vc) { struct csched_vcpu *svc = vc->sched_priv; +spinlock_t *lock; + + lock =vcpu_schedule_lock_irq (vc);if ( !__vcpu_on_runq (svc) &&vcpu_runnable (vc) && !vc->is_running )__runq_insert (vc->processor, svc); +vcpu_schedule_unlock_irq (lock, vc); +SCHED_STAT_CRANK (vcpu_insert); }
xen/common/sched_rt.c +3 -0
@@ -622,16 +622,19 @@ rt_vcpu_insert(const struct scheduler *ops, struct vcpu *vc) {struct rt_vcpu *svc =rt_vcpu (vc);s_time_t now =NOW (); +spinlock_t *lock;/* not addlocate idle vcpu to dom vcpu list */ if (is_idle_vcpu (vc) )return ; + lock =vcpu_schedule_lock_irq (vc);if ( now >= svc->cur_deadline )rt_update_deadline (now, svc);if ( !__vcpu_on_q (svc) &&vcpu_runnable (vc) && !vc->is_running )__runq_insert (ops, svc); +vcpu_schedule_unlock_irq (lock, vc);/* add rt_vcpu svc to scheduler-specific vcpu list of the dom */ list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu);
xen/common/schedule.c +0 -6
@@ -1488,9 +1488,7 @@ void __init scheduler_init(void)int schedule_cpu_switch (unsigned int cpu,struct cpupool *c) { -unsigned long flags;struct vcpu *idle; -spinlock_t *lock;void *ppriv, *ppriv_old, *vpriv, *vpriv_old;struct scheduler *old_ops =per_cpu (scheduler, cpu);struct scheduler *new_ops = (c ==NULL ) ? &ops : c->sched;
@@ -1509,8 +1507,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c)return -ENOMEM ; } - lock =pcpu_schedule_lock_irqsave (cpu, &flags); -SCHED_OP (old_ops, tick_suspend, cpu); vpriv_old = idle->sched_priv; idle->sched_priv = vpriv;
@@ -1520,8 +1516,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c)SCHED_OP (new_ops, tick_resume, cpu);SCHED_OP (new_ops, insert_vcpu, idle); -pcpu_schedule_unlock_irqrestore (lock, flags, cpu); -SCHED_OP (old_ops, free_vdata, vpriv_old);SCHED_OP (old_ops, free_pdata, ppriv_old, cpu);
Re: [PATCH 2/7] xen: sched: fix locking for insert_vcpu() in credit1 and RTDS
Quoted 20 lines · Dario Faggioli
The
insert_vcpuscheduler hook is called with an inconsistent locking strategy. In fact, it is sometimes invoked while holding the runqueue lock and sometimes when that is not the case.For instance, in case of
schedule_cpu_switch()the lock is acquired in generic code. On the other hand, in case ofsched_move_domain(), locking is left as a responsibility of the schedulers implementing the hook.This results in Credit1 and RTDS schedulers ending up (in case of
sched_move_domain()) doing runqueue manipulation without holding any runqueue lock, which is a bug. (Credit2 was doing the locking by itself already.)The right thing is to defer locking to the specific schedulers, as it's them that know what, how and when it is best to lock (as in: runqueue locks, vs. private scheduler locks, vs. both, etc.).
Quoted 7 lines
This patch, therefore:
- removes any locking aroundinsert_vcpu()from generic code;
- add proper locking in the hook implementations, for both Credit1 and RTDS.Signed-off-by Dario Faggioli <[email protected]>
Re: [PATCH 2/7] xen: sched: fix locking for insert_vcpu() in credit1 and RTDS
In reply to Dario FaggioliQuoted 110 lines · Dario Faggioli
The
insert_vcpuscheduler hook is called with an inconsistent locking strategy. In fact, it is sometimes invoked while holding the runqueue lock and sometimes when that is not the case.For instance, in case of
schedule_cpu_switch()the lock is acquired in generic code. On the other hand, in case ofsched_move_domain(), locking is left as a responsibility of the schedulers implementing the hook.This results in Credit1 and RTDS schedulers ending up (in case of
sched_move_domain()) doing runqueue manipulation without holding any runqueue lock, which is a bug. (Credit2 was doing the locking by itself already.)The right thing is to defer locking to the specific schedulers, as it's them that know what, how and when it is best to lock (as in: runqueue locks, vs. private scheduler locks, vs. both, etc.).
This patch, therefore:
- removes any locking aroundinsert_vcpu()from generic code;
- add proper locking in the hook implementations, for both Credit1 and RTDS.Signed-off-by Dario Faggioli <[email protected]>---
Cc: George Dunlap <[email protected]>
Cc: Meng Xu <[email protected]>
Cc: Andrew Cooper <[email protected]>
Cc: Jan Beulich <[email protected]>
---
Changes from v1 (within the other series):
* split the patch (wrt the original patch, in the original series), and take care, in this one, only ofinsert_vcpu();--- xen/common/sched_credit.c | 5 +++++ xen/common/sched_rt.c | 3 +++ xen/common/schedule.c | 6 ------ 3 files changed, 8 insertions(+), 6 deletions(-)xen/common/sched_credit.c +5
-0@@ -903,10 +903,15 @@ static void csched_vcpu_insert(const struct scheduler *ops, struct vcpu *vc) { struct csched_vcpu *svc = vc->sched_priv; +spinlock_t *lock; + + lock =vcpu_schedule_lock_irq (vc);if ( !__vcpu_on_runq (svc) &&vcpu_runnable (vc) && !vc->is_running )__runq_insert (vc->processor, svc); +vcpu_schedule_unlock_irq (lock, vc); +SCHED_STAT_CRANK (vcpu_insert); }xen/common/sched_rt.c +3
-0@@ -622,16 +622,19 @@ rt_vcpu_insert(const struct scheduler *ops, struct vcpu *vc) {struct rt_vcpu *svc =rt_vcpu (vc);s_time_t now =NOW (); +spinlock_t *lock;/* not addlocate idle vcpu to dom vcpu list */ if (is_idle_vcpu (vc) )return ; + lock =vcpu_schedule_lock_irq (vc);if ( now >= svc->cur_deadline )rt_update_deadline (now, svc);if ( !__vcpu_on_q (svc) &&vcpu_runnable (vc) && !vc->is_running )__runq_insert (ops, svc); +vcpu_schedule_unlock_irq (lock, vc);/* add rt_vcpu svc to scheduler-specific vcpu list of the dom */ list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu);xen/common/schedule.c +0
-4@@ -1488,9 +1488,7 @@ void __init scheduler_init(void)int schedule_cpu_switch (unsigned int cpu,struct cpupool *c) { -unsigned long flags;struct vcpu *idle; -spinlock_t *lock;void *ppriv, *ppriv_old, *vpriv, *vpriv_old;struct scheduler *old_ops =per_cpu (scheduler, cpu);struct scheduler *new_ops = (c ==NULL ) ? &ops : c->sched;@@ -1509,8 +1507,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c)return -ENOMEM ; } - lock =pcpu_schedule_lock_irqsave (cpu, &flags); -SCHED_OP (old_ops, tick_suspend, cpu); vpriv_old = idle->sched_priv; idle->sched_priv = vpriv;
xen/common/schedule.c int schedule_cpu_switch(unsigned int cpu, struct cpupool *c) · line 1523
Quoted 4 lines
xen/common/schedule.c +0
-1@@ -1520,8 +1516,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c)SCHED_OP (new_ops, tick_resume, cpu);SCHED_OP (new_ops, insert_vcpu, idle); -pcpu_schedule_unlock_irqrestore (lock, flags, cpu);
It seems to me that the locking here wasn't to protect insert_vcpu, but to prevent any scheduling events from happening on cpu until all the expected infrastructure (ticks, idle vcpu, &c) were ready. I can't immediately convince myself that removing these is safe in that regard.
Can you address this?
-George
Re: [PATCH 2/7] xen: sched: fix locking for insert_vcpu() in credit1 and RTDS
Quoted 31 lines · George Dunlap
Quoted 29 lines · Dario Faggioli
xen/common/schedule.c +0
-4@@ -1488,9 +1488,7 @@ void __init scheduler_init(void)int schedule_cpu_switch (unsigned int cpu,struct cpupool *c) { -unsigned long flags;struct vcpu *idle; -spinlock_t *lock;void *ppriv, *ppriv_old, *vpriv, *vpriv_old;struct scheduler *old_ops =per_cpu (scheduler, cpu);struct scheduler *new_ops = (c ==NULL ) ? &ops : c->sched;@@ -1509,8 +1507,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c)return -ENOMEM ; } - lock =pcpu_schedule_lock_irqsave (cpu, &flags); -SCHED_OP (old_ops, tick_suspend, cpu); vpriv_old = idle->sched_priv;idle->sched_priv = vpriv;xen/common/schedule.c +0
-0@@ -1520,8 +1516,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c)SCHED_OP (new_ops ,tick_resume ,cpu );SCHED_OP (new_ops, insert_vcpu, idle);
Quoted 10 lines
xen/common/schedule.c int schedule_cpu_switch(unsigned int cpu, struct cpupool *c) · line 1523
Quoted 1 line
xen/common/schedule.c +0
-1-pcpu_schedule_unlock_irqrestore (lock, flags, cpu);It seems to me that the locking here wasn't to protect
insert_vcpu, but
to prevent any scheduling events from happening on cpu until all the expected infrastructure (ticks, idle vcpu, &c) were ready. I can't immediately convince myself that removing these is safe in that regard.
Can you address this?
Scheduling can't happen on the cpu, until later than the end of this function, when, in cpupool_assign_cpu_locked(), we set to 1 its corresponding bit in the target cpupool's cpu_valid mask.
In fact, scheduling events happening before that, would basically mean that a cpu outside of any cpupool is somehow being considered for scheduling, which, as said, would be a bug. In fact, I sent patches back in July to cure occurrences of that behavior.
We've been discussing, basically about the same issue, with Jan in here:
https://www.choon.net/forum/read.php?22,3817262,3817489
And I'll add the promised ASSERT() and comment, when sending v2 of that patch. :-)
Make sense?
Thanks and Regards,
Dario
Signature · 4 lines
<<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
Re: [PATCH 2/7] xen: sched: fix locking for insert_vcpu() in credit1 and RTDS
Quoted 15 lines · Dario Faggioli
Quoted 9 lines · George Dunlap
It seems to me that the locking here wasn't to protect
insert_vcpu, but
to prevent any scheduling events from happening on cpu until all the
expected infrastructure (ticks, idle vcpu, &c) were ready. I can't immediately convince myself that removing these is safe in that regard.
Can you address this?Scheduling can't happen on the cpu, until later than the end of this function, when, in
cpupool_assign_cpu_locked(), we set to 1 its corresponding bit in the target cpupool'scpu_validmask.
Quoted 8 lines
We've been discussing, basically about the same issue, with Jan in here:
https://www.choon.net/forum/read.php?22,3817262,3817489And I'll add the promised
ASSERT()and comment, when sending v2 of that
patch. :-)
Or, of course, I can put them down when doing v2 of _this_ patch, if that makes it easier to understand why what I'm doing here is safe.
Yes, I think I'll do it here, I like it better that way.
Regards,
Dario
Signature · 4 lines
<<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
PATCH 3/7 xen: sched: better handle (not) inserting idle vCPUs in runqueues
In reply to Dario FaggioliIdle vCPUs should never really be explicitly inserted in any of the schedulers' runqueue. In fact, they are just put in execution immediately (either on boot, or when a pCPU is assigned to a cpupool), and it will be the first scheduling decision that --by preempting them-- will 'park' them in the queue.
In fact, avoiding inserting those vCPUs in runqueues is something that Credit2 and RTDS intercept and forbid already (when implementing insert_vcpu()). Credit1 does that implicitly, as the idle vCPU will always be already running when insert_vcpu() is called, and hence not put in the runqueue.
Let's make it *always* explicit, as that simplifies things by quite a bit. For instance, we can now introduce some BUG_ON() guards, with the twofold objective of making this assumption clear, and of catching misuse and bugs.
The check for whether we'd be inserting an idle vCPU in a queue, now, happens, once and for all schedulers, in generic code, at vCPU initialization time, while we can just avoid trying (and always failing!) doing so when doing cpupools manipulations.
---
Cc: George Dunlap <[email protected]>
Cc: Juergen Gross <[email protected]>
--- xen/common/sched_credit.c | 2 ++ xen/common/sched_credit2.c | 25 ++++++++++--------------- xen/common/sched_rt.c | 4 +--- xen/common/schedule.c | 21 +++++++++++---------- 4 files changed, 24 insertions(+), 28 deletions(-)
xen/common/sched_credit.c +2 -0
@@ -905,6 +905,8 @@ csched_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)struct csched_vcpu *svc = vc->sched_priv;spinlock_t *lock; +BUG_ON (is_idle_vcpu (vc) ); + lock =vcpu_schedule_lock_irq (vc);if ( !__vcpu_on_runq (svc) &&vcpu_runnable (vc) && !vc->is_running )
xen/common/sched_credit2.c +10 -15
@@ -870,30 +870,25 @@ csched2_vcpu_insert(const struct scheduler *ops, struct vcpu *vc) {struct csched2_vcpu *svc = vc->sched_priv;struct csched2_dom *const sdom = svc->sdom; +spinlock_t *lock;printk ("%s: Inserting %pv\n" , __func__, vc); -/* NB: On boot, idle vcpus are inserted before alloc_pdata() has -* been called for that cpu. -*/ -if ( !is_idle_vcpu (vc) ) - { -spinlock_t *lock; +BUG_ON (is_idle_vcpu (vc)); -/* FIXME: Do we need the private lock here? */ -list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); +/* FIXME: Do we need the private lock here? */ +list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); -/* Add vcpu to runqueue of initial processor */ - lock =vcpu_schedule_lock_irq (vc); +/* Add vcpu to runqueue of initial processor */ + lock =vcpu_schedule_lock_irq (vc); -runq_assign (ops, vc); +runq_assign (ops, vc); -vcpu_schedule_unlock_irq (lock, vc); +vcpu_schedule_unlock_irq (lock, vc); - sdom->nr_vcpus++; + sdom->nr_vcpus++; -SCHED_STAT_CRANK (vcpu_insert); - } +SCHED_STAT_CRANK (vcpu_insert);CSCHED2_VCPU_CHECK (vc); }
xen/common/sched_rt.c +1 -3
@@ -624,9 +624,7 @@ rt_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)s_time_t now =NOW ();spinlock_t *lock; -/* not addlocate idle vcpu to dom vcpu list */ -if (is_idle_vcpu (vc) ) -return ; +BUG_ON (is_idle_vcpu (vc) ); lock =vcpu_schedule_lock_irq (vc);if ( now >= svc->cur_deadline )
xen/common/schedule.c +11 -10
@@ -240,20 +240,22 @@ int sched_init_vcpu(struct vcpu *v, unsigned int processor)init_timer (&v->poll_timer, poll_timer_fn, v, v->processor); -/* Idle VCPUs are scheduled immediately. */ + v->sched_priv =SCHED_OP (DOM2OP (d), alloc_vdata, v, d->sched_priv); +if ( v->sched_priv ==NULL ) +return 1 ; + +TRACE_2D (TRC_SCHED_DOM_ADD , v->domain->domain_id, v->vcpu_id); + +/* Idle VCPUs are scheduled immediately, not inserting them in runqueue. */ if (is_idle_domain (d) ) {per_cpu (schedule_data, v->processor).curr = v; v->is_running =1 ; } - -TRACE_2D (TRC_SCHED_DOM_ADD , v->domain->domain_id, v->vcpu_id); - - v->sched_priv =SCHED_OP (DOM2OP (d), alloc_vdata, v, d->sched_priv); -if ( v->sched_priv ==NULL ) -return 1 ; - -SCHED_OP (DOM2OP (d), insert_vcpu, v); +else + { +SCHED_OP (DOM2OP (d), insert_vcpu, v); + }return 0 ; }
@@ -1514,7 +1516,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c) ppriv_old =per_cpu (schedule_data, cpu).sched_priv;per_cpu (schedule_data, cpu).sched_priv = ppriv;SCHED_OP (new_ops, tick_resume, cpu); -SCHED_OP (new_ops, insert_vcpu, idle);SCHED_OP (old_ops, free_vdata, vpriv_old);SCHED_OP (old_ops, free_pdata, ppriv_old, cpu);
Re: [PATCH 3/7] xen: sched: better handle (not) inserting idle vCPUs in runqueues
Quoted 20 lines · Dario Faggioli
Idle vCPUs should never really be explicitly inserted in any of the schedulers' runqueue. In fact, they are just put in execution immediately (either on boot, or when a pCPU is assigned to a cpupool), and it will be the first scheduling decision that --by preempting them-- will 'park' them in the queue.
In fact, avoiding inserting those vCPUs in runqueues is something that Credit2 and RTDS intercept and forbid already (when implementing
insert_vcpu()). Credit1 does that implicitly, as the idle vCPU will always be already running wheninsert_vcpu()is called, and hence not put in the runqueue.Let's make it *always* explicit, as that simplifies things by quite a bit. For instance, we can now introduce some
BUG_ON()guards, with the twofold objective of making this assumption clear, and of catching misuse and bugs.
Quoted 7 lines
The check for whether we'd be inserting an idle vCPU in a queue, now, happens, once and for all schedulers, in generic code, at vCPU initialization time, while we can just avoid trying (and always failing!) doing so when doing cpupools manipulations.
Signed-off-by Dario Faggioli <[email protected]>
Although it might be easier to evaluate the previous patch if this one were moved behind it in the series:
-George
Re: [PATCH 3/7] xen: sched: better handle (not) inserting idle vCPUs in runqueues
Quoted 8 lines · George Dunlap
Quoted 7 lines · Dario Faggioli
[...]
The check for whether we'd be inserting an idle vCPU in a queue, now, happens, once and for all schedulers, in generic code, at vCPU initialization time, while we can just avoid trying (and always failing!) doing so when doing cpupools manipulations.
Quoted 6 lines
Quoted 1 line
Signed-off-by Dario Faggioli <[email protected]>Although it might be easier to evaluate the previous patch if this one
were moved behind it in the series:
Indeed. I only did it like this because, in case we want to backport patch 1 and 2, we would not need to backport this one (which is more an improvement than a bugfix) too.
Quoted 2 lines
Acked-by George Dunlap <[email protected]>
Thanks. :-)
Regards,
Dario
Signature · 4 lines
<<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
Re: [PATCH 3/7] xen: sched: better handle (not) inserting idle vCPUs in runqueues
In reply to Dario FaggioliQuoted 20 lines · Dario Faggioli
Idle vCPUs should never really be explicitly inserted in any of the schedulers' runqueue. In fact, they are just put in execution immediately (either on boot, or when a pCPU is assigned to a cpupool), and it will be the first scheduling decision that --by preempting them-- will 'park' them in the queue.
In fact, avoiding inserting those vCPUs in runqueues is something that Credit2 and RTDS intercept and forbid already (when implementing
insert_vcpu()). Credit1 does that implicitly, as the idle vCPU will always be already running wheninsert_vcpu()is called, and hence not put in the runqueue.Let's make it *always* explicit, as that simplifies things by quite a bit. For instance, we can now introduce some
BUG_ON()guards, with the twofold objective of making this assumption clear, and of catching misuse and bugs.
Quoted 7 lines
The check for whether we'd be inserting an idle vCPU in a queue, now, happens, once and for all schedulers, in generic code, at vCPU initialization time, while we can just avoid trying (and always failing!) doing so when doing cpupools manipulations.
Signed-off-by Dario Faggioli <[email protected]>
Quoted 134 lines
---
Cc: George Dunlap <[email protected]>
Cc: Juergen Gross <[email protected]>--- xen/common/sched_credit.c | 2 ++ xen/common/sched_credit2.c | 25 ++++++++++--------------- xen/common/sched_rt.c | 4 +--- xen/common/schedule.c | 21 +++++++++++---------- 4 files changed, 24 insertions(+), 28 deletions(-)xen/common/sched_credit.c +2
-0@@ -905,6 +905,8 @@ csched_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)struct csched_vcpu *svc = vc->sched_priv;spinlock_t *lock; +BUG_ON (is_idle_vcpu (vc) ); + lock =vcpu_schedule_lock_irq (vc);if ( !__vcpu_on_runq (svc) &&vcpu_runnable (vc) && !vc->is_running )xen/common/sched_credit2.c +10
-15@@ -870,30 +870,25 @@ csched2_vcpu_insert(const struct scheduler *ops, struct vcpu *vc) {struct csched2_vcpu *svc = vc->sched_priv;struct csched2_dom *const sdom = svc->sdom; +spinlock_t *lock;printk ("%s: Inserting %pv\n" , __func__, vc); -/* NB: On boot, idle vcpus are inserted before alloc_pdata() has -* been called for that cpu. -*/ -if ( !is_idle_vcpu (vc) ) - { -spinlock_t *lock; +BUG_ON (is_idle_vcpu (vc)); -/* FIXME: Do we need the private lock here? */ -list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); +/* FIXME: Do we need the private lock here? */ +list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); -/* Add vcpu to runqueue of initial processor */ - lock =vcpu_schedule_lock_irq (vc); +/* Add vcpu to runqueue of initial processor */ + lock =vcpu_schedule_lock_irq (vc); -runq_assign (ops, vc); +runq_assign (ops, vc); -vcpu_schedule_unlock_irq (lock, vc); +vcpu_schedule_unlock_irq (lock, vc); - sdom->nr_vcpus++; + sdom->nr_vcpus++; -SCHED_STAT_CRANK (vcpu_insert); - } +SCHED_STAT_CRANK (vcpu_insert);CSCHED2_VCPU_CHECK (vc); }xen/common/sched_rt.c +1
-3@@ -624,9 +624,7 @@ rt_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)s_time_t now =NOW ();spinlock_t *lock; -/* not addlocate idle vcpu to dom vcpu list */ -if (is_idle_vcpu (vc) ) -return ; +BUG_ON (is_idle_vcpu (vc) ); lock =vcpu_schedule_lock_irq (vc);if ( now >= svc->cur_deadline )xen/common/schedule.c +11
-10@@ -240,20 +240,22 @@ int sched_init_vcpu(struct vcpu *v, unsigned int processor)init_timer (&v->poll_timer, poll_timer_fn, v, v->processor); -/* Idle VCPUs are scheduled immediately. */ + v->sched_priv =SCHED_OP (DOM2OP (d), alloc_vdata, v, d->sched_priv); +if ( v->sched_priv ==NULL ) +return 1 ; + +TRACE_2D (TRC_SCHED_DOM_ADD , v->domain->domain_id, v->vcpu_id); + +/* Idle VCPUs are scheduled immediately, not inserting them in runqueue. */ if (is_idle_domain (d) ) {per_cpu (schedule_data, v->processor).curr = v; v->is_running =1 ; } - -TRACE_2D (TRC_SCHED_DOM_ADD , v->domain->domain_id, v->vcpu_id); - - v->sched_priv =SCHED_OP (DOM2OP (d), alloc_vdata, v, d->sched_priv); -if ( v->sched_priv ==NULL ) -return 1 ; - -SCHED_OP (DOM2OP (d), insert_vcpu, v); +else + { +SCHED_OP (DOM2OP (d), insert_vcpu, v); + }return 0 ; }@@ -1514,7 +1516,6 @@ int schedule_cpu_switch(unsigned int cpu, struct cpupool *c) ppriv_old =per_cpu (schedule_data, cpu).sched_priv;per_cpu (schedule_data, cpu).sched_priv = ppriv;SCHED_OP (new_ops, tick_resume, cpu); -SCHED_OP (new_ops, insert_vcpu, idle);SCHED_OP (old_ops, free_vdata, vpriv_old);SCHED_OP (old_ops, free_pdata, ppriv_old, cpu);_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
PATCH 4/7 xen: sched: get rid of the per domain vCPU list in RTDS
In reply to Dario FaggioliAs, curently, there is no reason for bothering having it and keeping it updated.
In fact, it is only used for dumping and changing vCPUs parameters, but that can be achieved easily with for_each_vcpu.
---
Cc: George Dunlap <[email protected]>
Cc: Meng Xu <[email protected]>
--- xen/common/sched_rt.c | 36 +++++++++++++----------------------- 1 file changed, 13 insertions(+), 23 deletions(-)
xen/common/sched_rt.c +13 -23
@@ -160,7 +160,6 @@ struct rt_private {*/ struct rt_vcpu {struct list_head q_elem;/* on the runq/depletedq list */ -struct list_head sdom_elem;/* on the domain VCPU list */ /* Up-pointers */ struct rt_dom *sdom;
@@ -182,7 +181,6 @@ struct rt_vcpu {* Domain */ struct rt_dom { -struct list_head vcpu;/* link its VCPUs */ struct list_head sdom_elem;/* link list on rt_priv */ struct domain *dom;/* pointer to upper domain */ };
@@ -290,7 +288,7 @@ rt_dump_pcpu(const struct scheduler *ops, int cpu)static void rt_dump (const struct scheduler *ops) { -struct list_head *iter_sdom, *iter_svc, *runq, *depletedq, *iter; +struct list_head *runq, *depletedq, *iter;struct rt_private *prv =rt_priv (ops);struct rt_vcpu *svc;struct rt_dom *sdom;
@@ -319,14 +317,16 @@ rt_dump(const struct scheduler *ops) }printk ("Domain info:\n" ); -list_for_each ( iter_sdom, &prv->sdom ) +list_for_each ( iter, &prv->sdom ) { - sdom = list_entry(iter_sdom, struct rt_dom, sdom_elem); +struct vcpu *vc; + + sdom = list_entry(iter, struct rt_dom, sdom_elem);printk ("\tdomain: %d\n" , sdom->dom->domain_id); -list_for_each ( iter_svc, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc ) { - svc = list_entry(iter_svc, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (vc);rt_dump_vcpu (ops, svc); } }
@@ -527,7 +527,6 @@ rt_alloc_domdata(const struct scheduler *ops, struct domain *dom)if ( sdom ==NULL )return NULL ; -INIT_LIST_HEAD (&sdom->vcpu);INIT_LIST_HEAD (&sdom->sdom_elem); sdom->dom = dom;
@@ -587,7 +586,6 @@ rt_alloc_vdata(const struct scheduler *ops, struct vcpu *vc, void *dd)return NULL ;INIT_LIST_HEAD (&svc->q_elem); -INIT_LIST_HEAD (&svc->sdom_elem); svc->flags =0U ; svc->sdom = dd; svc->vcpu = vc;
@@ -614,8 +612,7 @@ rt_free_vdata(const struct scheduler *ops, void *priv)* This function is called in sched_move_domain() in schedule.c * When move a domain to a new cpupool. * It inserts vcpus of moving domain to the scheduler's RunQ in -* dest. cpupool; and insert rt_vcpu svc to scheduler-specific -* vcpu list of the dom +* dest. cpupool. */ static void rt_vcpu_insert (const struct scheduler *ops,struct vcpu *vc)
@@ -634,15 +631,11 @@ rt_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)__runq_insert (ops, svc);vcpu_schedule_unlock_irq (lock, vc); -/* add rt_vcpu svc to scheduler-specific vcpu list of the dom */ -list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); -SCHED_STAT_CRANK (vcpu_insert); }/* -* Remove rt_vcpu svc from the old scheduler in source cpupool; and -* Remove rt_vcpu svc from scheduler-specific vcpu list of the dom +* Remove rt_vcpu svc from the old scheduler in source cpupool. */ static void rt_vcpu_remove (const struct scheduler *ops,struct vcpu *vc)
@@ -659,9 +652,6 @@ rt_vcpu_remove(const struct scheduler *ops, struct vcpu *vc)if (__vcpu_on_q (svc) )__q_remove (svc);vcpu_schedule_unlock_irq (lock, vc); - -if ( !is_idle_vcpu (vc) ) -list_del_init (&svc->sdom_elem); }/*
@@ -1137,7 +1127,7 @@ rt_dom_cntl(struct rt_private *prv =rt_priv (ops);struct rt_dom *const sdom =rt_dom (d);struct rt_vcpu *svc; -struct list_head *iter; +struct vcpu *vc;unsigned long flags;int rc =0 ;
@@ -1145,7 +1135,7 @@ rt_dom_cntl( {case XEN_DOMCTL_SCHEDOP_getinfo:spin_lock_irqsave (&prv->lock, flags); - svc = list_entry(sdom->vcpu.next, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (sdom->dom->vcpu[0 ]); op->u.rtds.period = svc->period /MICROSECS (1 );/* transfer to us */ op->u.rtds.budget = svc->budget /MICROSECS (1 );spin_unlock_irqrestore (&prv->lock, flags);
@@ -1157,9 +1147,9 @@ rt_dom_cntl(break ; }spin_lock_irqsave (&prv->lock, flags); -list_for_each ( iter, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc ) { - struct rt_vcpu * svc = list_entry(iter, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (vc); svc->period =MICROSECS (op->u.rtds.period);/* transfer to nanosec */ svc->budget =MICROSECS (op->u.rtds.budget); }
Re: [PATCH 4/7] xen: sched: get rid of the per domain vCPU list in RTDS
Quoted 5 lines · Dario Faggioli
xen/common/sched_rt.c +0
-1@@ -319,14 +317,16 @@ rt_dump(const struct scheduler *ops) }printk ("Domain info:\n" ); -list_for_each ( iter_sdom, &prv->sdom )
xen/common/sched_rt.c rt_dump(const struct scheduler *ops) · line 327
Quoted 10 lines
xen/common/sched_rt.c +5
-2+list_for_each ( iter, &prv->sdom ) { - sdom = list_entry(iter_sdom, struct rt_dom, sdom_elem); +struct vcpu *vc; + + sdom = list_entry(iter, struct rt_dom, sdom_elem);printk ("\tdomain: %d\n" , sdom->dom->domain_id); -list_for_each ( iter_svc, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc )
Space before bracket, as you are already changing the line.
xen/common/sched_rt.c rt_dom_cntl( · line 1138
Quoted 13 lines
xen/common/sched_rt.c +2
-2{ - svc = list_entry(iter_svc, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (vc);rt_dump_vcpu (ops, svc); } }@@ -1145,7 +1135,7 @@ rt_dom_cntl( {case XEN_DOMCTL_SCHEDOP_getinfo:spin_lock_irqsave (&prv->lock, flags); - svc = list_entry(sdom->vcpu.next, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (sdom->dom->vcpu[0 ]);
This change swaps one potential bad pointer for another.
In the former case, there was no guarantee that sdom->vcpu had any entries in it, potentially making svc a wild pointer.
In the latter case, there is no guarantee that dom->vcpu has been allocated yet. You must check d->max_vcpus > 0 before dereferencing d->vcpu[].
~Andrew
Re: [PATCH 4/7] xen: sched: get rid of the per domain vCPU list in RTDS
In reply to Dario FaggioliQuoted 8 lines · Dario Faggioli
As, curently, there is no reason for bothering having it and keeping it updated.
In fact, it is only used for dumping and changing vCPUs parameters, but that can be achieved easily with
for_each_vcpu.Signed-off-by Dario Faggioli <[email protected]>
With Andrew's comments addressed:
Quoted 141 lines
---
Cc: George Dunlap <[email protected]>
Cc: Meng Xu <[email protected]>--- xen/common/sched_rt.c | 36 +++++++++++++----------------------- 1 file changed, 13 insertions(+), 23 deletions(-)xen/common/sched_rt.c +13
-23@@ -160,7 +160,6 @@ struct rt_private {*/ struct rt_vcpu {struct list_head q_elem;/* on the runq/depletedq list */ -struct list_head sdom_elem;/* on the domain VCPU list */ /* Up-pointers */ struct rt_dom *sdom;@@ -182,7 +181,6 @@ struct rt_vcpu {* Domain */ struct rt_dom { -struct list_head vcpu;/* link its VCPUs */ struct list_head sdom_elem;/* link list on rt_priv */ struct domain *dom;/* pointer to upper domain */ };@@ -290,7 +288,7 @@ rt_dump_pcpu(const struct scheduler *ops, int cpu)static void rt_dump (const struct scheduler *ops) { -struct list_head *iter_sdom, *iter_svc, *runq, *depletedq, *iter; +struct list_head *runq, *depletedq, *iter;struct rt_private *prv =rt_priv (ops);struct rt_vcpu *svc;struct rt_dom *sdom;@@ -319,14 +317,16 @@ rt_dump(const struct scheduler *ops) }printk ("Domain info:\n" ); -list_for_each ( iter_sdom, &prv->sdom ) +list_for_each ( iter, &prv->sdom ) { - sdom = list_entry(iter_sdom, struct rt_dom, sdom_elem); +struct vcpu *vc; + + sdom = list_entry(iter, struct rt_dom, sdom_elem);printk ("\tdomain: %d\n" , sdom->dom->domain_id); -list_for_each ( iter_svc, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc ) { - svc = list_entry(iter_svc, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (vc);rt_dump_vcpu (ops, svc); } }@@ -527,7 +527,6 @@ rt_alloc_domdata(const struct scheduler *ops, struct domain *dom)if ( sdom ==NULL )return NULL ; -INIT_LIST_HEAD (&sdom->vcpu);INIT_LIST_HEAD (&sdom->sdom_elem); sdom->dom = dom;@@ -587,7 +586,6 @@ rt_alloc_vdata(const struct scheduler *ops, struct vcpu *vc, void *dd)return NULL ;INIT_LIST_HEAD (&svc->q_elem); -INIT_LIST_HEAD (&svc->sdom_elem); svc->flags =0U ; svc->sdom = dd; svc->vcpu = vc;@@ -614,8 +612,7 @@ rt_free_vdata(const struct scheduler *ops, void *priv)* This function is called in sched_move_domain() in schedule.c * When move a domain to a new cpupool. * It inserts vcpus of moving domain to the scheduler's RunQ in -* dest. cpupool; and insert rt_vcpu svc to scheduler-specific -* vcpu list of the dom +* dest. cpupool. */ static void rt_vcpu_insert (const struct scheduler *ops,struct vcpu *vc)@@ -634,15 +631,11 @@ rt_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)__runq_insert (ops, svc);vcpu_schedule_unlock_irq (lock, vc); -/* add rt_vcpu svc to scheduler-specific vcpu list of the dom */ -list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); -SCHED_STAT_CRANK (vcpu_insert); }/* -* Remove rt_vcpu svc from the old scheduler in source cpupool; and -* Remove rt_vcpu svc from scheduler-specific vcpu list of the dom +* Remove rt_vcpu svc from the old scheduler in source cpupool. */ static void rt_vcpu_remove (const struct scheduler *ops,struct vcpu *vc)@@ -659,9 +652,6 @@ rt_vcpu_remove(const struct scheduler *ops, struct vcpu *vc)if (__vcpu_on_q (svc) )__q_remove (svc);vcpu_schedule_unlock_irq (lock, vc); - -if ( !is_idle_vcpu (vc) ) -list_del_init (&svc->sdom_elem); }/* @@ -1137,7 +1127,7 @@ rt_dom_cntl(struct rt_private *prv =rt_priv (ops);struct rt_dom *const sdom =rt_dom (d);struct rt_vcpu *svc; -struct list_head *iter; +struct vcpu *vc;unsigned long flags;int rc =0 ;@@ -1145,7 +1135,7 @@ rt_dom_cntl( {case XEN_DOMCTL_SCHEDOP_getinfo:spin_lock_irqsave (&prv->lock, flags); - svc = list_entry(sdom->vcpu.next, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (sdom->dom->vcpu[0 ]); op->u.rtds.period = svc->period /MICROSECS (1 );/* transfer to us */ op->u.rtds.budget = svc->budget /MICROSECS (1 );spin_unlock_irqrestore (&prv->lock, flags);@@ -1157,9 +1147,9 @@ rt_dom_cntl(break ; }spin_lock_irqsave (&prv->lock, flags); -list_for_each ( iter, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc ) { - struct rt_vcpu * svc = list_entry(iter, struct rt_vcpu, sdom_elem); + svc =rt_vcpu (vc); svc->period =MICROSECS (op->u.rtds.period);/* transfer to nanosec */ svc->budget =MICROSECS (op->u.rtds.budget); }
PATCH 5/7 xen: sched: get rid of the per domain vCPU list in Credit2
In reply to Dario FaggioliAs, curently, there is no reason for bothering having it and keeping it updated.
In fact, it is only used for dumping and changing vCPUs parameters, but that can be achieved easily with for_each_vcpu.
While there, improve alignment of comments, ad add a const qualifier to a pointer, making things more consistent with what happens everywhere else in the source file.
This also allows us to kill one of the remaining FIXMEs in the code, which is always good.
---
Cc: George Dunlap <[email protected]>
--- xen/common/sched_credit2.c | 34 +++++++++++----------------------- xen/common/sched_rt.c | 2 +- 2 files changed, 12 insertions(+), 24 deletions(-)
xen/common/sched_credit2.c +11 -23
@@ -234,9 +234,8 @@ struct csched2_private {* Virtual CPU */ struct csched2_vcpu { -struct list_head rqd_elem;/* On the runqueue data list */ -struct list_head sdom_elem;/* On the domain vcpu list */ -struct list_head runq_elem;/* On the runqueue */ +struct list_head rqd_elem;/* On the runqueue data list */ +struct list_head runq_elem;/* On the runqueue */ struct csched2_runqueue_data *rqd;/* Up-pointer to the runqueue */ /* Up-pointers */
@@ -261,7 +260,6 @@ struct csched2_vcpu {* Domain */ struct csched2_dom { -struct list_head vcpu;struct list_head sdom_elem;struct domain *dom;uint16_t weight;
@@ -772,7 +770,6 @@ csched2_alloc_vdata(const struct scheduler *ops, struct vcpu *vc, void *dd)return NULL ;INIT_LIST_HEAD (&svc->rqd_elem); -INIT_LIST_HEAD (&svc->sdom_elem);INIT_LIST_HEAD (&svc->runq_elem); svc->sdom = dd;
@@ -876,9 +873,6 @@ csched2_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)BUG_ON (is_idle_vcpu (vc)); -/* FIXME: Do we need the private lock here? */ -list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); -/* Add vcpu to runqueue of initial processor */ lock =vcpu_schedule_lock_irq (vc);
@@ -923,10 +917,6 @@ csched2_vcpu_remove(const struct scheduler *ops, struct vcpu *vc)vcpu_schedule_unlock_irq (lock, vc); -/* Remove from sdom list. Don't need a lock for this, as it's called -* syncronously when nothing else can happen. */ -list_del_init (&svc->sdom_elem); - svc->sdom->nr_vcpus--; } }
@@ -1443,7 +1433,7 @@ csched2_dom_cntl(if ( op->u.credit2.weight !=0 ) { -struct list_head *iter; +struct vcpu *vc;int old_weight; old_weight = sdom->weight;
@@ -1451,9 +1441,9 @@ csched2_dom_cntl( sdom->weight = op->u.credit2.weight;/* Update weights for vcpus, and max_weight for runqueues on which they reside */ - list_for_each ( iter, &sdom->vcpu ) + for_each_vcpu ( d, vc ) { - struct csched2_vcpu *svc = list_entry(iter, struct csched2_vcpu, sdom_elem); +struct csched2_vcpu *svc =CSCHED2_VCPU (vc);/* NB: Locking order is important here. Because we grab this lock here, we * must never lock csched2_priv.lock if we're holding a runqueue lock.
@@ -1487,7 +1477,6 @@ csched2_alloc_domdata(const struct scheduler *ops, struct domain *dom)return NULL ;/* Initialize credit and weight */ -INIT_LIST_HEAD (&sdom->vcpu);INIT_LIST_HEAD (&sdom->sdom_elem); sdom->dom = dom; sdom->weight =CSCHED2_DEFAULT_WEIGHT ;
@@ -1539,9 +1528,7 @@ csched2_free_domdata(const struct scheduler *ops, void *data)static void csched2_dom_destroy (const struct scheduler *ops,struct domain *dom) { -struct csched2_dom *sdom =CSCHED2_DOM (dom); - -BUG_ON (!list_empty (&sdom->vcpu)); +BUG_ON (CSCHED2_DOM (dom)->nr_vcpus !=0 );csched2_free_domdata (ops,CSCHED2_DOM (dom)); }
@@ -1881,7 +1868,7 @@ csched2_dump_pcpu(const struct scheduler *ops, int cpu)static void csched2_dump (const struct scheduler *ops) { -struct list_head *iter_sdom, *iter_svc; +struct list_head *iter_sdom;struct csched2_private *prv =CSCHED2_PRIV (ops);unsigned long flags;int i, loop;
@@ -1926,6 +1913,8 @@ csched2_dump(const struct scheduler *ops)list_for_each ( iter_sdom, &prv->sdom ) {struct csched2_dom *sdom; +struct vcpu *vc; + sdom = list_entry(iter_sdom, struct csched2_dom, sdom_elem); printk("\tDomain: %d w %d v %d\n\t",
@@ -1933,12 +1922,11 @@ csched2_dump(const struct scheduler *ops) sdom->weight, sdom->nr_vcpus); -list_for_each ( iter_svc, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc ) { -struct csched2_vcpu *svc; +struct csched2_vcpu *const svc =CSCHED2_VCPU (vc);spinlock_t *lock; - svc = list_entry(iter_svc, struct csched2_vcpu, sdom_elem); lock =vcpu_schedule_lock (svc->vcpu);printk ("\t%3d: " , ++loop);
xen/common/sched_rt.c +1 -1
@@ -1147,7 +1147,7 @@ rt_dom_cntl(break ; }spin_lock_irqsave (&prv->lock, flags); -for_each_vcpu ( sdom->dom, vc ) +for_each_vcpu ( d, vc ) { svc =rt_vcpu (vc); svc->period =MICROSECS (op->u.rtds.period);/* transfer to nanosec */
Re: [PATCH 5/7] xen: sched: get rid of the per domain vCPU list in Credit2
xen/common/sched_rt.c rt_dom_cntl( · line 1153
Quoted 13 lines · Dario Faggioli
xen/common/sched_rt.c +1
-1@@ -1147,7 +1147,7 @@ rt_dom_cntl(break ; }spin_lock_irqsave (&prv->lock, flags); -for_each_vcpu ( sdom->dom, vc ) +for_each_vcpu ( d, vc ) { svc =rt_vcpu (vc); svc->period =MICROSECS (op->u.rtds.period);/* transfer to nanosec */
This hunk looks like it wants to be in the previous patch.
Re: [PATCH 5/7] xen: sched: get rid of the per domain vCPU list in Credit2
Quoted 18 lines · Andrew Cooper
Quoted 13 lines · Dario Faggioli
xen/common/sched_rt.c +1
-1@@ -1147,7 +1147,7 @@ rt_dom_cntl(break ; }spin_lock_irqsave (&prv->lock, flags); -for_each_vcpu ( sdom->dom, vc ) +for_each_vcpu ( d, vc ) { svc =rt_vcpu (vc); svc->period =MICROSECS (op->u.rtds.period);/* Quoted 1 line
transfer to nanosec */
This hunk looks like it wants to be in the previous patch.
It clearly does. `stg refresh' fat-fingering, I guess.
Sorry, I'll take care when respinning! :-/
Dario
Signature · 4 lines
<<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
Re: [PATCH 5/7] xen: sched: get rid of the per domain vCPU list in Credit2
In reply to Dario Faggiolixen/common/sched_credit2.c csched2_dom_cntl( · line 1436
Quoted 6 lines · Dario Faggioli
xen/common/sched_credit2.c +1
-1@@ -1443,7 +1433,7 @@ csched2_dom_cntl(if ( op->u.credit2.weight !=0 ) { -struct list_head *iter; +struct vcpu *vc;
Any chance of starting to align on the more common practice of just v for a vcpu?
xen/common/sched_credit2.c csched2_free_domdata(const struct scheduler *ops, void *data) · line 1531
Quoted 8 lines
xen/common/sched_credit2.c +1
-3@@ -1539,9 +1528,7 @@ csched2_free_domdata(const struct scheduler *ops, void *data)static void csched2_dom_destroy (const struct scheduler *ops,struct domain *dom) { -struct csched2_dom *sdom =CSCHED2_DOM (dom); - -BUG_ON (!list_empty (&sdom->vcpu)); +BUG_ON (CSCHED2_DOM (dom)->nr_vcpus !=0 );
This is a latent bug (excuse the pun) which can be triggered by userspace. There is no guarantee or requirement that a domain registered with a scheduler has vcpus.
~Andrew
Re: [PATCH 5/7] xen: sched: get rid of the per domain vCPU list in Credit2
Quoted 11 lines · Andrew Cooper
xen/common/sched_credit2.c csched2_dom_cntl( · line 1436
Quoted 6 lines · Dario Faggioli
xen/common/sched_credit2.c +1
-1@@ -1443,7 +1433,7 @@ csched2_dom_cntl(if ( op->u.credit2.weight !=0 ) { -struct list_head *iter; +struct vcpu *vc;Any chance of starting to align on the more common practice of just v for a vcpu?
I see. I'm a bit split, though. It's v in schedule.c, but it's **always** (with only 2 exceptions) vc in sched_*.c.
I know that, if we want to align, we need to start from somewhere, but OTOH, consistency is rather helpful when reading this code.
I'll think about it...
Quoted 6 lines
Quoted 6 lines
(quoted diff) +0
-0@@ -1539,9 +1528,7 @@ csched2_free_domdata(const struct scheduler *ops, void *data) static void csched2_dom_destroy(const struct scheduler *ops, struct domain *dom) {
Quoted 9 lines
xen/common/sched_credit2.c csched2_free_domdata(const struct scheduler *ops, void *data) · line 1531
Quoted 4 lines
xen/common/sched_credit2.c +1
-3-struct csched2_dom *sdom =CSCHED2_DOM (dom); - -BUG_ON (!list_empty (&sdom->vcpu)); +BUG_ON (CSCHED2_DOM (dom)->nr_vcpus !=0 );This is a latent bug (excuse the pun) which can be triggered by userspace. There is no guarantee or requirement that a domain registered with a scheduler has vcpus.
Mmm... I think the original check wanted to catch cases where a domain (or, at least, its scheduling related bits) is being destroyed with vcpus still on.
While trying to retain the BUG_ON, even with sdom->vcpu not being there any longer, I probably went too far.
BUG_ON(CSCHED2_DOM(dom)->nr_vcpus > 0);
should be fine, I think.
Thanks and Regards,
Dario
Signature · 4 lines
<<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
Re: [PATCH 5/7] xen: sched: get rid of the per domain vCPU list in Credit2
Quoted 33 lines · Dario Faggioli
Quoted 10 lines · Andrew Cooper
xen/common/sched_credit2.c csched2_dom_cntl( · line 1436
Quoted 6 lines · Dario Faggioli
xen/common/sched_credit2.c +1
-1@@ -1443,7 +1433,7 @@ csched2_dom_cntl(if ( op->u.credit2.weight !=0 ) { -struct list_head *iter; +struct vcpu *vc;Any chance of starting to align on the more common practice of just v for a vcpu?
I see. I'm a bit split, though. It's v in
schedule.c, but it's **always** (with only 2 exceptions) vc in sched_*.c.I know that, if we want to align, we need to start from somewhere, but OTOH, consistency is rather helpful when reading this code.
I'll think about it...
Quoted 5 lines
Quoted 5 lines
(quoted diff) +0
-0@@ -1539,9 +1528,7 @@ csched2_free_domdata(const struct scheduler *ops, void *data) static void csched2_dom_destroy(const struct scheduler *ops, struct domain *dom)Quoted 9 lines
xen/common/sched_credit2.c csched2_free_domdata(const struct scheduler *ops, void *data) · line 1531
Quoted 5 lines
xen/common/sched_credit2.c +1
-3{ -struct csched2_dom *sdom =CSCHED2_DOM (dom); - -BUG_ON (!list_empty (&sdom->vcpu)); +BUG_ON (CSCHED2_DOM (dom)->nr_vcpus !=0 );This is a latent bug (excuse the pun) which can be triggered by userspace. There is no guarantee or requirement that a domain registered with a scheduler has vcpus.
Quoted 3 lines
Mmm... I think the original check wanted to catch cases where a domain (or, at least, its scheduling related bits) is being destroyed with vcpus still on.
Oops sorry - I got the condition inverted when thinking about it. Also, this is a csched2_dom rather than a struct domain.
I rescind my query.
~Andrew
Re: [PATCH 5/7] xen: sched: get rid of the per domain vCPU list in Credit2
In reply to Dario FaggioliQuoted 12 lines · Dario Faggioli
As, curently, there is no reason for bothering having it and keeping it updated.
In fact, it is only used for dumping and changing vCPUs parameters, but that can be achieved easily with
for_each_vcpu.While there, improve alignment of comments, ad add a const qualifier to a pointer, making things more consistent with what happens everywhere else in the source file.
Quoted 4 lines
This also allows us to kill one of the remaining FIXMEs in the code, which is always good.
Signed-off-by Dario Faggioli <[email protected]>
Quoted 147 lines
---
Cc: George Dunlap <[email protected]>--- xen/common/sched_credit2.c | 34 +++++++++++----------------------- xen/common/sched_rt.c | 2 +- 2 files changed, 12 insertions(+), 24 deletions(-)xen/common/sched_credit2.c +11
-23@@ -234,9 +234,8 @@ struct csched2_private {* Virtual CPU */ struct csched2_vcpu { -struct list_head rqd_elem;/* On the runqueue data list */ -struct list_head sdom_elem;/* On the domain vcpu list */ -struct list_head runq_elem;/* On the runqueue */ +struct list_head rqd_elem;/* On the runqueue data list */ +struct list_head runq_elem;/* On the runqueue */ struct csched2_runqueue_data *rqd;/* Up-pointer to the runqueue */ /* Up-pointers */ @@ -261,7 +260,6 @@ struct csched2_vcpu {* Domain */ struct csched2_dom { -struct list_head vcpu;struct list_head sdom_elem;struct domain *dom;uint16_t weight;@@ -772,7 +770,6 @@ csched2_alloc_vdata(const struct scheduler *ops, struct vcpu *vc, void *dd)return NULL ;INIT_LIST_HEAD (&svc->rqd_elem); -INIT_LIST_HEAD (&svc->sdom_elem);INIT_LIST_HEAD (&svc->runq_elem); svc->sdom = dd;@@ -876,9 +873,6 @@ csched2_vcpu_insert(const struct scheduler *ops, struct vcpu *vc)BUG_ON (is_idle_vcpu (vc)); -/* FIXME: Do we need the private lock here? */ -list_add_tail (&svc->sdom_elem, &svc->sdom->vcpu); -/* Add vcpu to runqueue of initial processor */ lock =vcpu_schedule_lock_irq (vc);@@ -923,10 +917,6 @@ csched2_vcpu_remove(const struct scheduler *ops, struct vcpu *vc)vcpu_schedule_unlock_irq (lock, vc); -/* Remove from sdom list. Don't need a lock for this, as it's called -* syncronously when nothing else can happen. */ -list_del_init (&svc->sdom_elem); - svc->sdom->nr_vcpus--; } }@@ -1443,7 +1433,7 @@ csched2_dom_cntl(if ( op->u.credit2.weight !=0 ) { -struct list_head *iter; +struct vcpu *vc;int old_weight; old_weight = sdom->weight;@@ -1451,9 +1441,9 @@ csched2_dom_cntl( sdom->weight = op->u.credit2.weight;/* Update weights for vcpus, and max_weight for runqueues on which they reside */ - list_for_each ( iter, &sdom->vcpu ) + for_each_vcpu ( d, vc ) { - struct csched2_vcpu *svc = list_entry(iter, struct csched2_vcpu, sdom_elem); +struct csched2_vcpu *svc =CSCHED2_VCPU (vc);/* NB: Locking order is important here. Because we grab this lock here, we * must never lock csched2_priv.lock if we're holding a runqueue lock. @@ -1487,7 +1477,6 @@ csched2_alloc_domdata(const struct scheduler *ops, struct domain *dom)return NULL ;/* Initialize credit and weight */ -INIT_LIST_HEAD (&sdom->vcpu);INIT_LIST_HEAD (&sdom->sdom_elem); sdom->dom = dom; sdom->weight =CSCHED2_DEFAULT_WEIGHT ;@@ -1539,9 +1528,7 @@ csched2_free_domdata(const struct scheduler *ops, void *data)static void csched2_dom_destroy (const struct scheduler *ops,struct domain *dom) { -struct csched2_dom *sdom =CSCHED2_DOM (dom); - -BUG_ON (!list_empty (&sdom->vcpu)); +BUG_ON (CSCHED2_DOM (dom)->nr_vcpus !=0 );csched2_free_domdata (ops,CSCHED2_DOM (dom)); }@@ -1881,7 +1868,7 @@ csched2_dump_pcpu(const struct scheduler *ops, int cpu)static void csched2_dump (const struct scheduler *ops) { -struct list_head *iter_sdom, *iter_svc; +struct list_head *iter_sdom;struct csched2_private *prv =CSCHED2_PRIV (ops);unsigned long flags;int i, loop;@@ -1926,6 +1913,8 @@ csched2_dump(const struct scheduler *ops)list_for_each ( iter_sdom, &prv->sdom ) {struct csched2_dom *sdom; +struct vcpu *vc; + sdom = list_entry(iter_sdom, struct csched2_dom, sdom_elem); printk("\tDomain: %d w %d v %d\n\t",@@ -1933,12 +1922,11 @@ csched2_dump(const struct scheduler *ops) sdom->weight, sdom->nr_vcpus); -list_for_each ( iter_svc, &sdom->vcpu ) +for_each_vcpu ( sdom->dom, vc ) { -struct csched2_vcpu *svc; +struct csched2_vcpu *const svc =CSCHED2_VCPU (vc);spinlock_t *lock; - svc = list_entry(iter_svc, struct csched2_vcpu, sdom_elem); lock =vcpu_schedule_lock (svc->vcpu);printk ("\t%3d: " , ++loop);xen/common/sched_rt.c +1
-1@@ -1147,7 +1147,7 @@ rt_dom_cntl(break ; }spin_lock_irqsave (&prv->lock, flags); -for_each_vcpu ( sdom->dom, vc ) +for_each_vcpu ( d, vc ) { svc =rt_vcpu (vc); svc->period =MICROSECS (op->u.rtds.period);/* transfer to nanosec */
PATCH 6/7 xen: sched: fix an 'off by one \t' in credit2 debug dump
In reply to Dario Faggioli---
Cc: George Dunlap <[email protected]>
--- xen/common/sched_credit2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
xen/common/sched_credit2.c +1 -1
@@ -1917,7 +1917,7 @@ csched2_dump(const struct scheduler *ops) sdom = list_entry(iter_sdom, struct csched2_dom, sdom_elem); -printk ("\tDomain: %d w %d v %d\n\t" , +printk ("\tDomain: %d w %d v %d\n" , sdom->dom->domain_id, sdom->weight, sdom->nr_vcpus);
Re: [PATCH 6/7] xen: sched: fix an 'off by one \t' in credit2 debug dump
Quoted 2 lines · Dario Faggioli
Signed-off-by Dario Faggioli <[email protected]>Reviewed-by Juergen Gross <[email protected]>
Oh, I didn't realize this one was going to have stuff from the previous series as well. :-)
Quoted 20 lines
---
Cc: George Dunlap <[email protected]>--- xen/common/sched_credit2.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)xen/common/sched_credit2.c +1
-1@@ -1917,7 +1917,7 @@ csched2_dump(const struct scheduler *ops) sdom = list_entry(iter_sdom, struct csched2_dom, sdom_elem); -printk ("\tDomain: %d w %d v %d\n\t" , +printk ("\tDomain: %d w %d v %d\n" , sdom->dom->domain_id, sdom->weight, sdom->nr_vcpus);
Re: [PATCH 6/7] xen: sched: fix an 'off by one \t' in credit2 debug dump
Quoted 8 lines · George Dunlap
Quoted 2 lines · Dario Faggioli
Signed-off-by Dario Faggioli <[email protected]>Reviewed-by Juergen Gross <[email protected]>Oh, I didn't realize this one was going to have stuff from the previous
series as well. :-)
Just this patch. Since I was doing a new series, this patch seemed more appropriate here than there.
I should have made it more clear in the cover, sorry.
Quoted 2 lines
Acked-by George Dunlap <[email protected]>
Thanks and Regards,
Dario
Signature · 4 lines
<<This happens because I choose it to happen!>> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)
_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel
PATCH 7/7 xen: sched / cpupool: dump the actual value of NOW()
In reply to Dario Faggiolirather than its hexadecimal representation. This makes it easier to compare the actual system time with other times being printed out (e.g., deadlines in RTDS).
---
Cc: Juergen Gross <[email protected]>
Cc: George Dunlap <[email protected]>
--- xen/common/cpupool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
xen/common/cpupool.c +1 -1
@@ -744,7 +744,7 @@ void dump_runq(unsigned char key)printk ("sched_smt_power_savings: %s\n" , sched_smt_power_savings?"enabled" :"disabled" ); -printk ("NOW=0x%08X%08X\n" , (u32)(now>>32 ), (u32 )now); +printk ("NOW=%" PRI_stime"\n" , now);print_cpumap ("Online Cpus" , &cpu_online_map);if ( !cpumask_empty (&cpupool_free_cpus) )
Re: [PATCH 7/7] xen: sched / cpupool: dump the actual value of NOW()
Quoted 5 lines · Dario Faggioli
rather than its hexadecimal representation. This makes it easier to compare the actual system time with other times being printed out (e.g., deadlines in RTDS).
Signed-off-by Dario Faggioli <[email protected]>
There are a number of other areas of code which do the same (which I noticed when doing the keyhandler refactor). I was going to find and fix them all in some free time.
Re: [PATCH 7/7] xen: sched / cpupool: dump the actual value of NOW()
Quoted 6 lines
Quoted 1 line
Quoted 1 line · On 08.10.15 at 15:12, <[email protected]> wrote:
Quoted 4 lines · Dario Faggioli
rather than its hexadecimal representation. This makes it easier to compare the actual system time with other times being printed out (e.g., deadlines in RTDS).
Quoted 7 lines
Quoted 1 line
Signed-off-by Dario Faggioli <[email protected]>Reviewed-by Andrew Cooper <[email protected]>There are a number of other areas of code which do the same (which I noticed when doing the keyhandler refactor). I was going to find and fix them all in some free time.
I think in a number of those cases printing NOW() is actually meaningless and would better be deleted than fixed.
Jan
Re: [PATCH 7/7] xen: sched / cpupool: dump the actual value of NOW()
In reply to Dario FaggioliQuoted 5 lines · Dario Faggioli
rather than its hexadecimal representation. This makes it easier to compare the actual system time with other times being printed out (e.g., deadlines in RTDS).
Signed-off-by Dario Faggioli <[email protected]>
Quoted 27 lines
---
Cc: Juergen Gross <[email protected]>
Cc: George Dunlap <[email protected]>--- xen/common/cpupool.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)xen/common/cpupool.c +1
-1@@ -744,7 +744,7 @@ void dump_runq(unsigned char key)printk ("sched_smt_power_savings: %s\n" , sched_smt_power_savings?"enabled" :"disabled" ); -printk ("NOW=0x%08X%08X\n" , (u32)(now>>32 ), (u32 )now); +printk ("NOW=%" PRI_stime"\n" , now);print_cpumap ("Online Cpus" , &cpu_online_map);if ( !cpumask_empty (&cpupool_free_cpus) )_______________________________________________
Xen-devel mailing list
[email protected]
http://lists.xen.org/xen-devel