Re: [Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Avi Kivity
On 02/22/2010 05:42 PM, Anthony Liguori wrote: Spurious calls to qemu_notify_event() also make it difficult to tell when it's actually necessary to call qemu_notify_event() vs. when it's just something that doesn't hurt. One improvement in this area would be to add a context parameter (which

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Anthony Liguori
On 02/22/2010 09:32 AM, Avi Kivity wrote: On 02/22/2010 05:29 PM, Anthony Liguori wrote: On 02/22/2010 09:16 AM, Marcelo Tosatti wrote: Are you concerned about spurious wakeups? Yes. Also, qemu_notify_event() is an undirected notification (wakes up all iothreads, and all devices), whereas ->h

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Avi Kivity
On 02/22/2010 05:29 PM, Anthony Liguori wrote: On 02/22/2010 09:16 AM, Marcelo Tosatti wrote: Are you concerned about spurious wakeups? Yes. Also, qemu_notify_event() is an undirected notification (wakes up all iothreads, and all devices), whereas ->handle_output() is directed (wakes up exactl

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Anthony Liguori
On 02/22/2010 09:16 AM, Marcelo Tosatti wrote: Are you concerned about spurious wakeups? Yes. Also, qemu_notify_event() is an undirected notification (wakes up all iothreads, and all devices), whereas ->handle_output() is directed (wakes up exactly what is needed). What's the underlyin

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Marcelo Tosatti
On Mon, Feb 22, 2010 at 04:51:46PM +0200, Avi Kivity wrote: > On 02/22/2010 04:29 PM, Marcelo Tosatti wrote: > >On Mon, Feb 22, 2010 at 04:20:52PM +0200, Avi Kivity wrote: > >>On 02/22/2010 03:59 PM, Marcelo Tosatti wrote: > >>>VIRTIO_PCI_QUEUE_NOTIFY is used to inform availability of new buffers,

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Avi Kivity
On 02/22/2010 04:29 PM, Marcelo Tosatti wrote: On Mon, Feb 22, 2010 at 04:20:52PM +0200, Avi Kivity wrote: On 02/22/2010 03:59 PM, Marcelo Tosatti wrote: VIRTIO_PCI_QUEUE_NOTIFY is used to inform availability of new buffers, so wakeup the iothread to process that information immediate

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Marcelo Tosatti
On Mon, Feb 22, 2010 at 04:20:52PM +0200, Avi Kivity wrote: > On 02/22/2010 03:59 PM, Marcelo Tosatti wrote: > >VIRTIO_PCI_QUEUE_NOTIFY is used to inform availability of new buffers, > >so wakeup the iothread to process that information immediately. > > > >Reported-by: Amit Shah > >Signed-off-by: M

[Qemu-devel] Re: [patch uq/master 1/2] virtio-pci: wake up iothread on VIRTIO_PCI_QUEUE_NOTIFY

2010-02-22 Thread Avi Kivity
On 02/22/2010 03:59 PM, Marcelo Tosatti wrote: VIRTIO_PCI_QUEUE_NOTIFY is used to inform availability of new buffers, so wakeup the iothread to process that information immediately. Reported-by: Amit Shah Signed-off-by: Marcelo Tosatti Index: qemu/hw/virtio-pci.c ===