First of all, interrupt mode here means per-queue Rx interrupt support. Historically, virtio PMD can only be binded to igb_uio or uio_pci_generic, and not for vfio-pci. Besides, quote from http://dpdk.org/doc/guides-16.11/rel_notes/release_2_1.html: "- Per queue RX interrupt events are only allowed in VFIO which supports multiple MSI-X vectors. - In UIO, the RX interrupt shares the same vector with other interrupts. When the RX interrupt and LSC interrupt are both enabled, only the former is available. - RX interrupt is only implemented for the linuxapp target"
As Linux starts to support vfio noiommu mode since 4.8.0, it's a good chance to enable per queue RX interrupt for virtio PMD. This patchset is an attempt to enable this. Note: this patch set is still under developping, not ready for test. Signed-off-by: Jianfeng Tan <jianfeng....@intel.com> Jianfeng Tan (5): net/virtio: add Rx descriptor check net/virtio: setup Rx fastpath interrupts net/virtio: remove Rx queue interrupts when stopping net/virtio: add Rx queue intr enable/disable functions examples/l3fwd: add parse-ptype option drivers/net/virtio/virtio_ethdev.c | 74 +++++++++++++++++++++++++++++++++++--- drivers/net/virtio/virtio_ethdev.h | 3 ++ drivers/net/virtio/virtio_rxtx.c | 9 +++++ drivers/net/virtio/virtqueue.c | 11 ------ drivers/net/virtio/virtqueue.h | 26 +++++++++++++- examples/l3fwd-power/main.c | 60 ++++++++++++++++++++++++++++++- 6 files changed, 165 insertions(+), 18 deletions(-) -- 2.7.4
