Re: [PATCH] vfio: avoid SET_ACTION_TRIGGER ioctls

2020-03-09 Thread Roman Kapl
On 3/9/20 12:43 PM, Roman Kapl wrote: On 3/5/20 11:37 PM, Alex Williamson wrote: On Fri, 28 Feb 2020 13:08:00 +0100 Roman Kapl wrote: For MSI-X interrupts, remember what the last used eventfd was (KVM bypass vs QEMU) and only call vfio_set_irq_signaling if it has changed. This not only hel

Re: [PATCH] vfio: avoid SET_ACTION_TRIGGER ioctls

2020-03-09 Thread Roman Kapl
On 3/5/20 11:37 PM, Alex Williamson wrote: On Fri, 28 Feb 2020 13:08:00 +0100 Roman Kapl wrote: For MSI-X interrupts, remember what the last used eventfd was (KVM bypass vs QEMU) and only call vfio_set_irq_signaling if it has changed. This not only helps with performance, but it seems that in

Re: [PATCH] vfio: avoid SET_ACTION_TRIGGER ioctls

2020-03-05 Thread Alex Williamson
On Fri, 28 Feb 2020 13:08:00 +0100 Roman Kapl wrote: > For MSI-X interrupts, remember what the last used eventfd was (KVM > bypass vs QEMU) and only call vfio_set_irq_signaling if it has changed. > > This not only helps with performance, but it seems that interrupts can > be lost during VFIO_IRQ

[PATCH] vfio: avoid SET_ACTION_TRIGGER ioctls

2020-02-28 Thread Roman Kapl
For MSI-X interrupts, remember what the last used eventfd was (KVM bypass vs QEMU) and only call vfio_set_irq_signaling if it has changed. This not only helps with performance, but it seems that interrupts can be lost during VFIO_IRQ_SET_ACTION_TRIGGER. With the 'x-no-kvm-msix' switch and this pat