similar to: [PATCH] vhost: relax log address alignment

virtio 1.0 only requires used address to be 4 byte aligned, vhost required 8 bytes (size of vring_used_elem). Fix up vhost to match that. Additionally, while vhost correctly requires 8 byte alignment for log, it's unconnected to used ring: it's a consequence that log has u64 entries. Tweak code to make that clearer. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

virtio 1.0 only requires used address to be 4 byte aligned, vhost required 8 bytes (size of vring_used_elem). Fix up vhost to match that. Additionally, while vhost correctly requires 8 byte alignment for log, it's unconnected to used ring: it's a consequence that log has u64 entries. Tweak code to make that clearer. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling. This is done through setup kernel address through direct mapping and co-opreate VM management with MMU notifiers. Test shows about 23% improvement on TX PPS. TCP_STREAM doesn't see obvious

Hi: This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling like SMAP. This is done through setup kernel address through direct mapping and co-opreate VM management with MMU notifiers. Test shows about 23% improvement on TX PPS. TCP_STREAM

Hi: This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling like SMAP. This is done through setup kernel address through direct mapping and co-opreate VM management with MMU notifiers. Test shows about 23% improvement on TX PPS. TCP_STREAM

virtio 1.0 only requires used address to be 4 byte aligned, vhost required 8 bytes (size of vring_used_elem). Fix up vhost to match that. Additionally, while vhost correctly requires 8 byte alignment for log, it's unconnected to used ring: it's a consequence that log has u64 entries. Tweak code to make that clearer. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

virtio 1.0 only requires used address to be 4 byte aligned, vhost required 8 bytes (size of vring_used_elem). Fix up vhost to match that. Additionally, while vhost correctly requires 8 byte alignment for log, it's unconnected to used ring: it's a consequence that log has u64 entries. Tweak code to make that clearer. Signed-off-by: Michael S. Tsirkin <mst at redhat.com> ---

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

We need to set log when updating flags of used ring, otherwise they may be missed after migration. A helper was introduced to write used_flags back to guest memory and update the log if necessary. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/vhost/vhost.c | 26 ++++++++++++++++++++++---- drivers/vhost/vhost.h | 2 ++ 2 files changed, 24 insertions(+), 4 deletions(-)

We need to set log when updating flags of used ring, otherwise they may be missed after migration. A helper was introduced to write used_flags back to guest memory and update the log if necessary. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/vhost/vhost.c | 26 ++++++++++++++++++++++---- drivers/vhost/vhost.h | 2 ++ 2 files changed, 24 insertions(+), 4 deletions(-)

On Fri, Dec 07, 2018 at 04:48:41PM +0800, Tiwei Bie wrote: > Introduce VIRTIO_RING_NO_LEGACY to support disabling legacy > macros and layout definitions. > > Suggested-by: Michael S. Tsirkin <mst at redhat.com> > Signed-off-by: Tiwei Bie <tiwei.bie at intel.com> > --- > VRING_AVAIL_ALIGN_SIZE, VRING_USED_ALIGN_SIZE and VRING_DESC_ALIGN_SIZE > are not pre-virtio

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

The ring element addresses are passed between components with different alignments assumptions. Thus, if guest/userspace selects a pointer and host then gets and dereferences it, we might need to decrease the compiler-selected alignment to prevent compiler on the host from assuming pointer is aligned. This actually triggers on ARM with -mabi=apcs-gnu - which is a deprecated configuration, but it

On Mon, Apr 06, 2020 at 09:34:00PM +0800, Jason Wang wrote: > > On 2020/4/6 ??8:50, Michael S. Tsirkin wrote: > > The ring element addresses are passed between components with different > > alignments assumptions. Thus, if guest/userspace selects a pointer and > > host then gets and dereferences it, we might need to decrease the > > compiler-selected alignment to

On Mon, Apr 06, 2020 at 09:34:00PM +0800, Jason Wang wrote: > > On 2020/4/6 ??8:50, Michael S. Tsirkin wrote: > > The ring element addresses are passed between components with different > > alignments assumptions. Thus, if guest/userspace selects a pointer and > > host then gets and dereferences it, we might need to decrease the > > compiler-selected alignment to

Details about this can be found here: https://lwn.net/Articles/750770/ What's new in this version ========================== A new VFIO device type is introduced - vfio-vhost. This addressed some comments from here: https://patchwork.ozlabs.org/cover/984763/ Below is the updated device interface: Currently, there are two regions of this device: 1) CONFIG_REGION

Details about this can be found here: https://lwn.net/Articles/750770/ What's new in this version ========================== A new VFIO device type is introduced - vfio-vhost. This addressed some comments from here: https://patchwork.ozlabs.org/cover/984763/ Below is the updated device interface: Currently, there are two regions of this device: 1) CONFIG_REGION