On Wed, Jun 02, 2010 at 12:00:27PM +0300, Avi Kivity wrote:
>
> There are two separate problems: the more general problem is that
> the hypervisor can put a vcpu to sleep while holding a lock, causing
> other vcpus to spin until the end of their time slice. This can
> only be addressed with hypervisor help.
Fyi - I have a early patch ready to address this issue. Basically I am using
host-kernel memory (mmap'ed into guest as io-memory via ivshmem driver) to hint
host whenever guest is in spin-lock'ed section, which is read by host scheduler
to defer preemption.
Guest side:
static inline void spin_lock(spinlock_t *lock)
{
raw_spin_lock(&lock->rlock);
+ __get_cpu_var(gh_vcpu_ptr)->defer_preempt++;
}
static inline void spin_unlock(spinlock_t *lock)
{
+ __get_cpu_var(gh_vcpu_ptr)->defer_preempt--;
raw_spin_unlock(&lock->rlock);
}
[similar changes to other spinlock variants]
Host side:
@@ -860,6 +866,17 @@ check_preempt_tick(struct cfs_rq *cfs_rq
ideal_runtime = sched_slice(cfs_rq, curr);
delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime;
if (delta_exec > ideal_runtime) {
+ if ((sched_feat(DEFER_PREEMPT)) && (rq_of(cfs_rq)->curr->ghptr)) {
+ int defer_preempt = rq_of(cfs_rq)->curr->ghptr->defer_preempt;
+ if (((defer_preempt & 0xFFFF0000) == 0xfeed0000) && ((defer_preempt & 0x0000FFFF) != 0)) {
+ if ((rq_of(cfs_rq)->curr->grace_defer++ < sysctl_sched_preempt_defer_count)) {
+ rq_of(cfs_rq)->defer_preempt++;
+ return;
+ } else
+ rq_of(cfs_rq)->force_preempt++;
+ }
+ }
resched_task(rq_of(cfs_rq)->curr);
/*
* The current task ran long enough, ensure it doesn't get
[similar changes introduced at other preemption points in sched_fair.c]
Note that guest can only request preemption to be deferred (and not disabled via
this mechanism). I have seen good improvement (~15%) in kern compile benchmark
with sysctl_sched_preempt_defer_count set to a low value of just 2 (i.e we can
defer preemption by maximum two ticks). I intend to cleanup and post the patches
pretty soon for comments.
One pathological case where this may actually hurt is routines in guest like
flush_tlb_others_ipi() which take a spinlock and then enter a while() loop
waiting for other cpus to ack something. In this case, deferring preemption just
because guest is in critical section actually hurts! Hopefully the upper bound
for deferring preemtion and the fact that such routines may not be frequently
hit should help alleviate such situations.
- vatsa
--
To unsubscribe from this list: send the line "unsubscribe kvm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
