Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ

From	Paolo Bonzini <pbonzini@redhat.com>
Newsgroups	linux.kernel
Subject	Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ
Date	2015-10-29 10:50 +0100
Message-ID	<qoS0i-6Ul-35@gated-at.bofh.it> (permalink)
References	(4 earlier) <qok8j-1XT-45@gated-at.bofh.it> <qon6a-3PX-31@gated-at.bofh.it> <qoBst-4RD-1@gated-at.bofh.it> <qoCoy-5we-31@gated-at.bofh.it> <qoLUR-3eH-3@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw


On 29/10/2015 04:11, Alex Williamson wrote:
> > The irqfd is already able to schedule a work item, because it runs with
> > interrupts disabled, so I think we can always return IRQ_HANDLED.
>
> I'm confused by this.  The problem with adding IRQF_NO_THREAD to our
> current handler is that it hits the spinlock that can sleep in
> eventfd_signal() and the waitqueue further down the stack before we get
> to the irqfd.  So if we split to a non-threaded handler vs a threaded
> handler, where the non-threaded handler either returns IRQ_HANDLED or
> IRQ_WAKE_THREAD to queue the threaded handler, there's only so much that
> the non-threaded handler can do before we start running into the same
> problem.

You're right.  I thought schedule_work used raw spinlocks (and then
everything would be done in the inject callback), but I was wrong.

Basically where irqfd_wakeup now does schedule_work, it would need to
return IRQ_WAKE_THREAD.  The threaded handler then can just do the
eventfd_signal.

Paolo

> I think that means that the non-threaded handler needs to
> return IRQ_WAKE_THREAD if we need to use the current eventfd_signal()
> path, such as if the bypass path is not available.  If we can get
> through the bypass path and the KVM irqfd side is safe for the
> non-threaded handler, inject succeeds and we return IRQ_HANDLED, right?
> Thanks,
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Thread

[RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-10-27 02:40 +0100
  Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Alex Williamson <alex.williamson@redhat.com> - 2015-10-27 04:40 +0100
    Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-10-27 07:50 +0100
      Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Paolo Bonzini <pbonzini@redhat.com> - 2015-10-27 10:30 +0100
        Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-10-27 22:40 +0100
          Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Paolo Bonzini <pbonzini@redhat.com> - 2015-10-28 01:50 +0100
            Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Alex Williamson <alex.williamson@redhat.com> - 2015-10-28 17:10 +0100
              Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Paolo Bonzini <pbonzini@redhat.com> - 2015-10-28 18:10 +0100
                Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Marcelo Tosatti <mtosatti@redhat.com> - 2015-10-29 01:00 +0100
                Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Alex Williamson <alex.williamson@redhat.com> - 2015-10-29 04:20 +0100
                Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Paolo Bonzini <pbonzini@redhat.com> - 2015-10-29 10:50 +0100
                Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-10-30 07:30 +0100
            Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-10-28 19:10 +0100
              Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Alex Williamson <alex.williamson@redhat.com> - 2015-10-28 19:20 +0100
                Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Yunhong Jiang <yunhong.jiang@linux.intel.com> - 2015-10-28 23:00 +0100
              Re: [RFC PATCH] VFIO: Add a parameter to force nonthread IRQ Paolo Bonzini <pbonzini@redhat.com> - 2015-10-28 19:30 +0100

csiph-web