similar to: [PATCH net-next 0/3] basic in order support for vhost_net

On Fri, Nov 23, 2018 at 11:00:16AM +0800, Jason Wang wrote: > Device use descriptors table in order, so there's no need to read > index from available ring. This eliminate the cache contention on > avail ring completely. Well this isn't what the in order feature says in the spec. It forces the used ring to be in the same order as the available ring. So I don't think you can

Device use descriptors table in order, so there's no need to read index from available ring. This eliminate the cache contention on avail ring completely. Virito-user + vhost_kernel + XDP_DROP gives about ~10% improvement on TX from 4.8Mpps to 5.3Mpps on Intel(R) Core(TM) i7-5600U CPU @ 2.60GHz. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/vhost/vhost.c | 19

On Fri, Nov 23, 2018 at 11:00:15AM +0800, Jason Wang wrote: > This makes vhost_net to support in order feature. This is as simple as > use datacopy path when it was negotiated. An alternative is not to > advertise in order when zerocopy is enabled which tends to be > suboptimal consider zerocopy may suffer from e.g HOL issues. Well IIRC vhost_zerocopy_signal_used is used to actually

> On 6 Mar 2019, at 08:18, Jason Wang <jasowang at redhat.com> wrote: > > This is used to hide the metadata address from virtqueue helpers. This > will allow to implement a vmap based fast accessing to metadata. > > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vhost/vhost.c | 94 +++++++++++++++++++++++++++++++++++++++++---------- > 1

> On 6 Mar 2019, at 08:18, Jason Wang <jasowang at redhat.com> wrote: > > This is used to hide the metadata address from virtqueue helpers. This > will allow to implement a vmap based fast accessing to metadata. > > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vhost/vhost.c | 94 +++++++++++++++++++++++++++++++++++++++++---------- > 1

Hi: This series tries to implement basic tx batched processing. This is done by prefetching descriptor indices and update used ring in a batch. This intends to speed up used ring updating and improve the cache utilization. Test shows about ~22% improvement in tx pss. Please review. Jason Wang (5): vhost: split out ring head fetching logic vhost: introduce helper to prefetch desc index

Hi: This series tries to implement basic tx batched processing. This is done by prefetching descriptor indices and update used ring in a batch. This intends to speed up used ring updating and improve the cache utilization. Test shows about ~22% improvement in tx pss. Please review. Jason Wang (5): vhost: split out ring head fetching logic vhost: introduce helper to prefetch desc index

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. Test shows about 24% improvement on TX PPS. It should benefit other cases as well. Changes from V1: - instead of pinning pages, use MMU notifier to invalidate vmaps and remap duing

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 vmap() and resigter MMU notifier for invalidation. Test shows about 24% improvement on TX PPS. TCP_STREAM doesn't see obvious improvement.

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 vmap() and resigter MMU notifier for invalidation. Test shows about 24% improvement on TX PPS. TCP_STREAM doesn't see obvious improvement.

This makes vhost_net to support in order feature. This is as simple as use datacopy path when it was negotiated. An alternative is not to advertise in order when zerocopy is enabled which tends to be suboptimal consider zerocopy may suffer from e.g HOL issues. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/vhost/net.c | 6 ++++-- 1 file changed, 4 insertions(+), 2

On 2020/6/3 ??5:48, Michael S. Tsirkin wrote: > On Wed, Jun 03, 2020 at 03:13:56PM +0800, Jason Wang wrote: >> On 2020/6/2 ??9:05, Michael S. Tsirkin wrote: [...] >>> + >>> +static int fetch_indirect_descs(struct vhost_virtqueue *vq, >>> + struct vhost_desc *indirect, >>> + u16 head) >>> +{ >>> + struct vring_desc desc;

On 2020/6/2 ??9:05, Michael S. Tsirkin wrote: > The idea is to support multiple ring formats by converting > to a format-independent array of descriptors. > > This costs extra cycles, but we gain in ability > to fetch a batch of descriptors in one go, which > is good for code cache locality. > > When used, this causes a minor performance degradation, > it's been kept

On 2020/6/2 ??9:05, Michael S. Tsirkin wrote: > The idea is to support multiple ring formats by converting > to a format-independent array of descriptors. > > This costs extra cycles, but we gain in ability > to fetch a batch of descriptors in one go, which > is good for code cache locality. > > When used, this causes a minor performance degradation, > it's been kept

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. Test shows about 24% improvement on TX PPS. It should benefit other cases as well. Please review Jason Wang (3): vhost: generalize adding used elem vhost: fine grain userspace memory

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. Test shows about 24% improvement on TX PPS. It should benefit other cases as well. Please review Jason Wang (3): vhost: generalize adding used elem vhost: fine grain userspace memory

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 all: This RFC implement packed ring layout. The code were tested with Tiwei's RFC V3 ahttps://lkml.org/lkml/2018/4/25/34. Some fixups and tweaks were needed on top of Tiwei's code to make it run for event index. Pktgen reports about 20% improvement on PPS (event index is off). More testing is ongoing. Notes for tester: - Start from this version, vhost need qemu co-operation to work

Hi all: This RFC implement packed ring layout. The code were tested with Tiwei's RFC V3 ahttps://lkml.org/lkml/2018/4/25/34. Some fixups and tweaks were needed on top of Tiwei's code to make it run for event index. Pktgen reports about 20% improvement on PPS (event index is off). More testing is ongoing. Notes for tester: - Start from this version, vhost need qemu co-operation to work

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