search for: gicv2m

...with virtio and network (due to the fact how our Fast Models are implemented, I mostly us block devices and 9p protocol over virtio and I get enough performance from them :-). > As arm doesn't support msi-x now, To be precise: "ARM" does "support" MSI-X :-) (google for GICv2m) The correct statement would be: "normal memory mapped devices have no interface for message signalled interrupts (like MSI-X)" > we use GSI for multiple irq. I'm not sure what GSI stands for, but looking at the code I assume it's just a "normal" peripheral interr...

As the current virtio-mmio only support single irq, so some advanced features such as vhost-net with irqfd are not supported. And the net performance is not the best without vhost-net and irqfd supporting. This patch support virtio-mmio to request multiple irqs like virtio-pci. With this patch and qemu assigning multiple irqs for virtio-mmio device, it's ok to use vhost-net with irqfd on

As the current virtio-mmio only support single irq, so some advanced features such as vhost-net with irqfd are not supported. And the net performance is not the best without vhost-net and irqfd supporting. This patch support virtio-mmio to request multiple irqs like virtio-pci. With this patch and qemu assigning multiple irqs for virtio-mmio device, it's ok to use vhost-net with irqfd on

...to the fact how our Fast Models are implemented, I mostly us block > devices and 9p protocol over virtio and I get enough performance from > them :-). > >> As arm doesn't support msi-x now, > > To be precise: "ARM" does "support" MSI-X :-) (google for GICv2m) Sorry, I mean ARM with GICv2. > > The correct statement would be: "normal memory mapped devices have no > interface for message signalled interrupts (like MSI-X)" > Yes, that's right. >> we use GSI for multiple irq. > > I'm not sure what GSI stands f...

...to the fact how our Fast Models are implemented, I mostly us block > devices and 9p protocol over virtio and I get enough performance from > them :-). > >> As arm doesn't support msi-x now, > > To be precise: "ARM" does "support" MSI-X :-) (google for GICv2m) Sorry, I mean ARM with GICv2. > > The correct statement would be: "normal memory mapped devices have no > interface for message signalled interrupts (like MSI-X)" > Yes, that's right. >> we use GSI for multiple irq. > > I'm not sure what GSI stands f...

...-pci,iommu_platform,disable-legacy=on,drive=hd0 -drive if=none,file=rootfs.bin,id=hd0 You can also try the kvmtool device [4]. For example on AMD Seattle I use the following commands to boot a guest with all devices behind a virtio-iommu, then display mapping stats. $ lkvm run -k Image --irqchip gicv2m --console virtio --vfio-pci 01:00.0 --viommu vfio --viommu virtio $ lkvm debug -a -i stats [1] [RFC] virtio-iommu version 0.6 https://www.spinics.net/lists/linux-virtualization/msg32628.html [2] [RFC PATCH v2 0/5] Add virtio-iommu driver https://www.spinics.net/lists/kvm/msg159047.html [3...

...-pci,iommu_platform,disable-legacy=on,drive=hd0 -drive if=none,file=rootfs.bin,id=hd0 You can also try the kvmtool device [4]. For example on AMD Seattle I use the following commands to boot a guest with all devices behind a virtio-iommu, then display mapping stats. $ lkvm run -k Image --irqchip gicv2m --console virtio --vfio-pci 01:00.0 --viommu vfio --viommu virtio $ lkvm debug -a -i stats [1] [RFC] virtio-iommu version 0.6 https://www.spinics.net/lists/linux-virtualization/msg32628.html [2] [RFC PATCH v2 0/5] Add virtio-iommu driver https://www.spinics.net/lists/kvm/msg159047.html [3...