Re: [PATCH] vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM

2020-02-26 Thread Jason Wang
On 2020/2/26 下午4:12, Michael S. Tsirkin wrote: On Wed, Feb 26, 2020 at 02:20:36AM -0500, Jason Wang wrote: - Original Message - On Wed, Feb 26, 2020 at 03:06:47PM +0800, Jason Wang wrote: We turn on device IOTLB via VIRTIO_F_IOMMU_PLATFORM unconditionally on platform without IOMMU s

Re: [PATCH] vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM

2020-02-26 Thread Michael S. Tsirkin
On Wed, Feb 26, 2020 at 02:20:36AM -0500, Jason Wang wrote: > > > - Original Message - > > On Wed, Feb 26, 2020 at 03:06:47PM +0800, Jason Wang wrote: > > > We turn on device IOTLB via VIRTIO_F_IOMMU_PLATFORM unconditionally on > > > platform without IOMMU support. This can lead unnecessa

Re: [PATCH] vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM

2020-02-25 Thread Jason Wang
- Original Message - > On Wed, Feb 26, 2020 at 03:06:47PM +0800, Jason Wang wrote: > > We turn on device IOTLB via VIRTIO_F_IOMMU_PLATFORM unconditionally on > > platform without IOMMU support. This can lead unnecessary IOTLB > > transactions which will damage the performance. > > > > F

Re: [PATCH] vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM

2020-02-25 Thread Michael S. Tsirkin
On Wed, Feb 26, 2020 at 03:06:47PM +0800, Jason Wang wrote: > We turn on device IOTLB via VIRTIO_F_IOMMU_PLATFORM unconditionally on > platform without IOMMU support. This can lead unnecessary IOTLB > transactions which will damage the performance. > > Fixing this by check whether the device is ba

[PATCH] vhost: correctly turn on VIRTIO_F_IOMMU_PLATFORM

2020-02-25 Thread Jason Wang
We turn on device IOTLB via VIRTIO_F_IOMMU_PLATFORM unconditionally on platform without IOMMU support. This can lead unnecessary IOTLB transactions which will damage the performance. Fixing this by check whether the device is backed by IOMMU and disable device IOTLB. Reported-by: Halil Pasic Fix