On Tue, Aug 30, 2016 at 11:29:51PM +0530, Bhaktipriya Shridhar wrote:
> The workqueue "irqfd_cleanup_wq" queues a single work item
> &irqfd->shutdown and hence doesn't require ordering. It is a host-wide
> workqueue for issuing deferred shutdown requests aggregated from all
> vm* instances. It is not being used on a memory reclaim path.
> Hence, it has been converted to use system_wq.
> The work item has been flushed in kvm_irqfd_release().
>
> The workqueue "wqueue" queues a single work item &timer->expired
> and hence doesn't require ordering. Also, it is not being used on
> a memory reclaim path. Hence, it has been converted to use system_wq.
>
> System workqueues have been able to handle high level of concurrency
> for a long time now and hence it's not required to have a singlethreaded
> workqueue just to gain concurrency. Unlike a dedicated per-cpu workqueue
> created with create_singlethread_workqueue(), system_wq allows multiple
> work items to overlap executions even on the same CPU; however, a
> per-cpu workqueue doesn't have any CPU locality or global ordering
> guarantee unless the target CPU is explicitly specified and thus the
> increase of local concurrency shouldn't make any difference.
>
> Signed-off-by: Bhaktipriya Shridhar <bhaktipriya96@xxxxxxxxx>
> ---
> virt/kvm/arm/arch_timer.c | 11 ++---------
> virt/kvm/eventfd.c | 22 +++-------------------
> virt/kvm/kvm_main.c | 6 ------
> 3 files changed, 5 insertions(+), 34 deletions(-)
>
> diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
> index e2d5b6f..56e0c15 100644
> --- a/virt/kvm/arm/arch_timer.c
> +++ b/virt/kvm/arm/arch_timer.c
> @@ -31,7 +31,6 @@
> #include "trace.h"
>
> static struct timecounter *timecounter;
> -static struct workqueue_struct *wqueue;
> static unsigned int host_vtimer_irq;
>
> void kvm_timer_vcpu_put(struct kvm_vcpu *vcpu)
> @@ -140,7 +139,7 @@ static enum hrtimer_restart kvm_timer_expire(struct hrtimer *hrt)
> return HRTIMER_RESTART;
> }
>
> - queue_work(wqueue, &timer->expired);
> + schedule_work(&timer->expired);
> return HRTIMER_NORESTART;
> }
>
> @@ -455,12 +454,6 @@ int kvm_timer_hyp_init(void)
> goto out_free;
> }
>
> - wqueue = create_singlethread_workqueue("kvm_arch_timer");
> - if (!wqueue) {
> - err = -ENOMEM;
> - goto out_free;
> - }
> -
> kvm_info("virtual timer IRQ%d\n", host_vtimer_irq);
> on_each_cpu(kvm_timer_init_interrupt, NULL, 1);
>
> @@ -522,7 +515,7 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
> * VCPUs have the enabled variable set, before entering the guest, if
> * the arch timers are enabled.
> */
> - if (timecounter && wqueue)
> + if (timecounter)
> timer->enabled = 1;
>
> return 0;
This was discussed when this stuff was originally added, and I think the
argument then was that it improved tracing somehow, to be able to tell
exactly which piece of work was being done.
That being said, I don't have any objections against this patch, so
unless others object, for the ARM part:
Acked-by: Christoffer Dall <christoffer.dall@xxxxxxxxxx>
Thanks,
-Christoffer
> diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c
> index e469b60..f397e9b 100644
> --- a/virt/kvm/eventfd.c
> +++ b/virt/kvm/eventfd.c
> @@ -42,7 +42,6 @@
>
> #ifdef CONFIG_HAVE_KVM_IRQFD
>
> -static struct workqueue_struct *irqfd_cleanup_wq;
>
> static void
> irqfd_inject(struct work_struct *work)
> @@ -168,7 +167,7 @@ irqfd_deactivate(struct kvm_kernel_irqfd *irqfd)
>
> list_del_init(&irqfd->list);
>
> - queue_work(irqfd_cleanup_wq, &irqfd->shutdown);
> + schedule_work(&irqfd->shutdown);
> }
>
> int __attribute__((weak)) kvm_arch_set_irq_inatomic(
> @@ -555,7 +554,7 @@ kvm_irqfd_deassign(struct kvm *kvm, struct kvm_irqfd *args)
> * so that we guarantee there will not be any more interrupts on this
> * gsi once this deassign function returns.
> */
> - flush_workqueue(irqfd_cleanup_wq);
> + flush_work(&irqfd->shutdown);
>
> return 0;
> }
> @@ -592,7 +591,7 @@ kvm_irqfd_release(struct kvm *kvm)
> * Block until we know all outstanding shutdown jobs have completed
> * since we do not take a kvm* reference.
> */
> - flush_workqueue(irqfd_cleanup_wq);
> + flush_work(&irqfd->shutdown);
>
> }
>
> @@ -622,23 +621,8 @@ void kvm_irq_routing_update(struct kvm *kvm)
> spin_unlock_irq(&kvm->irqfds.lock);
> }
>
> -/*
> - * create a host-wide workqueue for issuing deferred shutdown requests
> - * aggregated from all vm* instances. We need our own isolated single-thread
> - * queue to prevent deadlock against flushing the normal work-queue.
> - */
> -int kvm_irqfd_init(void)
> -{
> - irqfd_cleanup_wq = create_singlethread_workqueue("kvm-irqfd-cleanup");
> - if (!irqfd_cleanup_wq)
> - return -ENOMEM;
> -
> - return 0;
> -}
> -
> void kvm_irqfd_exit(void)
> {
> - destroy_workqueue(irqfd_cleanup_wq);
> }
> #endif
>
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index 02e98f3..93506d2 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -3719,12 +3719,7 @@ int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
> * kvm_arch_init makes sure there's at most one caller
> * for architectures that support multiple implementations,
> * like intel and amd on x86.
> - * kvm_arch_init must be called before kvm_irqfd_init to avoid creating
> - * conflicts in case kvm is already setup for another implementation.
> */
> - r = kvm_irqfd_init();
> - if (r)
> - goto out_irqfd;
>
> if (!zalloc_cpumask_var(&cpus_hardware_enabled, GFP_KERNEL)) {
> r = -ENOMEM;
> @@ -3805,7 +3800,6 @@ out_free_0a:
> free_cpumask_var(cpus_hardware_enabled);
> out_free_0:
> kvm_irqfd_exit();
> -out_irqfd:
> kvm_arch_exit();
> out_fail:
> return r;
> --
> 2.1.4
>
--
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
