similar to: [RFC] virtio: use mandatory barriers for remote processor vdevs

Getting use of virtio rings correct is tricky, and a recent patch saw an implementation of in-kernel rings (as separate from userspace). This patch attempts to abstract the business of dealing with the virtio ring layout from the access (userspace or direct); to do this, we use function pointers, which gcc inlines correctly. Signed-off-by: Rusty Russell <rusty at rustcorp.com.au> ---

Getting use of virtio rings correct is tricky, and a recent patch saw an implementation of in-kernel rings (as separate from userspace). This patch attempts to abstract the business of dealing with the virtio ring layout from the access (userspace or direct); to do this, we use function pointers, which gcc inlines correctly. Signed-off-by: Rusty Russell <rusty at rustcorp.com.au> ---

This patch-set introduces the CAIF Virtio Link layer. The purpose is to communicate with a remote processor (a modem) over shared memory. Virtio is used as the transport mechanism, and the Remoteproc framework provides configuration and management of the Virtio rings and devices. The modem and Linux host may be on the same SoC, or connected over a shared memory interface such as LLI. Zero-Copy

This patch-set introduces the CAIF Virtio Link layer. The purpose is to communicate with a remote processor (a modem) over shared memory. Virtio is used as the transport mechanism, and the Remoteproc framework provides configuration and management of the Virtio rings and devices. The modem and Linux host may be on the same SoC, or connected over a shared memory interface such as LLI. Zero-Copy

Hello everyone, This RFC implements packed ring support in virtio driver. Some functional tests have been done with Jason's packed ring implementation in vhost (RFC v5): https://lwn.net/Articles/755862/ Both of ping and netperf worked as expected. TODO: - Refinements (for code and commit log); - More tests and bug fixes if any; - Send the formal patch set; RFC v5 -> RFC v6: - Avoid

Hello everyone, This patch set implements packed ring support in virtio driver. Some functional tests have been done with Jason's packed ring implementation in vhost: https://lkml.org/lkml/2018/7/3/33 Both of ping and netperf worked as expected. v1 -> v2: - Use READ_ONCE() to read event off_wrap and flags together (Jason); - Add comments related to ccw (Jason); RFC (v6) -> v1: -

Hello everyone, This patch set implements packed ring support in virtio driver. Some functional tests have been done with Jason's packed ring implementation in vhost: https://lkml.org/lkml/2018/7/3/33 Both of ping and netperf worked as expected. v1 -> v2: - Use READ_ONCE() to read event off_wrap and flags together (Jason); - Add comments related to ccw (Jason); RFC (v6) -> v1: -

Hello everyone, This patch set implements packed ring support in virtio driver. Some functional tests have been done with Jason's packed ring implementation in vhost: https://lkml.org/lkml/2018/7/3/33 Both of ping and netperf worked as expected. RFC (v6) -> v1: - Avoid extra virtio_wmb() in virtqueue_enable_cb_delayed_packed() when event idx is off (Jason); - Fix bufs calculation in

We were cheating with our barriers; using the smp ones rather than the real device ones. That was fine, until virtio-mmio came along, which could be talking to a real device (a non-SMP CPU). Unfortunately, just putting back the real barriers (reverting d57ed95d) causes a performance regression on virtio-pci. In particular, Amos reports netbench's TCP_RR over virtio_net CPU utilization

virtio is communicating with a virtual "device" that actually runs on another host processor. Thus SMP barriers can be used to control memory access ordering. Where possible, we should use SMP barriers which are more lightweight than mandatory barriers, because mandatory barriers also control MMIO effects on accesses through relaxed memory I/O windows (which virtio does not use)

We were cheating with our barriers; using the smp ones rather than the real device ones. That was fine, until virtio-mmio came along, which could be talking to a real device (a non-SMP CPU). Unfortunately, just putting back the real barriers (reverting d57ed95d) causes a performance regression on virtio-pci. In particular, Amos reports netbench's TCP_RR over virtio_net CPU utilization

virtio is communicating with a virtual "device" that actually runs on another host processor. Thus SMP barriers can be used to control memory access ordering. Where possible, we should use SMP barriers which are more lightweight than mandatory barriers, because mandatory barriers also control MMIO effects on accesses through relaxed memory I/O windows (which virtio does not use)

Hello everyone, This RFC implements packed ring support in virtio driver. Some simple functional tests have been done with Jason's packed ring implementation in vhost (RFC v4): https://lkml.org/lkml/2018/5/16/501 Both of ping and netperf worked as expected w/ EVENT_IDX disabled. Ping worked as expected w/ EVENT_IDX enabled, but netperf didn't (A hack has been added in the driver to

Hello everyone, This RFC implements packed ring support in virtio driver. Some simple functional tests have been done with Jason's packed ring implementation in vhost (RFC v4): https://lkml.org/lkml/2018/5/16/501 Both of ping and netperf worked as expected w/ EVENT_IDX disabled. Ping worked as expected w/ EVENT_IDX enabled, but netperf didn't (A hack has been added in the driver to

Hello everyone, This RFC implements packed ring support in virtio driver. Some simple functional tests have been done with Jason's packed ring implementation in vhost: https://lkml.org/lkml/2018/4/23/12 Both of ping and netperf worked as expected (with EVENT_IDX disabled). TODO: - Refinements (for code and commit log); - More tests; - Bug fixes; RFC v3 -> RFC v4: - Make ID allocation

Hello everyone, This RFC implements packed ring support in virtio driver. Some simple functional tests have been done with Jason's packed ring implementation in vhost: https://lkml.org/lkml/2018/4/23/12 Both of ping and netperf worked as expected (with EVENT_IDX disabled). But there are below known issues: 1. Reloading the guest driver will break the Tx/Rx; 2. Zeroing the flags when

Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/virtio/virtio_ring.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 79e1b29..78428a8 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -35,7 +35,7 @@ #define virtio_rmb(vq) \ do { if

Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/virtio/virtio_ring.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c index 79e1b29..78428a8 100644 --- a/drivers/virtio/virtio_ring.c +++ b/drivers/virtio/virtio_ring.c @@ -35,7 +35,7 @@ #define virtio_rmb(vq) \ do { if

Hello everyone, This RFC implements packed ring support for virtio driver. The code was tested with DPDK vhost (testpmd/vhost-PMD) implemented by Jens at http://dpdk.org/ml/archives/dev/2018-January/089417.html Minor changes are needed for the vhost code, e.g. to kick the guest. TODO: - Refinements and bug fixes; - Split into small patches; - Test indirect descriptor support; - Test/fix event

Hello everyone, This RFC implements packed ring support for virtio driver. The code was tested with DPDK vhost (testpmd/vhost-PMD) implemented by Jens at http://dpdk.org/ml/archives/dev/2018-January/089417.html Minor changes are needed for the vhost code, e.g. to kick the guest. TODO: - Refinements and bug fixes; - Split into small patches; - Test indirect descriptor support; - Test/fix event