Thu, Apr 18, 2024 at 02:59:41AM CEST, jasowang@xxxxxxxxxx wrote: >On Wed, Apr 17, 2024 at 3:23 PM Jiri Pirko <jiri@xxxxxxxxxxx> wrote: >> >> Wed, Apr 17, 2024 at 06:37:30AM CEST, jasowang@xxxxxxxxxx wrote: >> >On Tue, Apr 16, 2024 at 5:37 PM Jiri Pirko <jiri@xxxxxxxxxxx> wrote: >> >> >> >> Tue, Apr 16, 2024 at 05:52:41AM CEST, jasowang@xxxxxxxxxx wrote: >> >> >On Tue, Apr 16, 2024 at 12:25 AM Jiri Pirko <jiri@xxxxxxxxxxx> wrote: >> >> >> >> >> >> From: Jiri Pirko <jiri@xxxxxxxxxx> >> >> >> >> >> >> Currently there is no way for user to set what features the driver >> >> >> should obey or not, it is hard wired in the code. >> >> >> >> >> >> In order to be able to debug the device behavior in case some feature is >> >> >> disabled, introduce a debugfs infrastructure with couple of files >> >> >> allowing user to see what features the device advertises and >> >> >> to set filter for features used by driver. >> >> >> >> >> >> Example: >> >> >> $cat /sys/bus/virtio/devices/virtio0/features >> >> >> 1110010111111111111101010000110010000000100000000000000000000000 >> >> >> $ echo "5" >/sys/kernel/debug/virtio/virtio0/filter_feature_add >> >> >> $ cat /sys/kernel/debug/virtio/virtio0/filter_features >> >> >> 5 >> >> >> $ echo "virtio0" > /sys/bus/virtio/drivers/virtio_net/unbind >> >> >> $ echo "virtio0" > /sys/bus/virtio/drivers/virtio_net/bind >> >> >> $ cat /sys/bus/virtio/devices/virtio0/features >> >> >> 1110000111111111111101010000110010000000100000000000000000000000 >> >> >> >> >> >> Note that sysfs "features" know already exists, this patch does not >> >> >> touch it. >> >> >> >> >> >> Signed-off-by: Jiri Pirko <jiri@xxxxxxxxxx> >> >> >> --- >> >> > >> >> >Note that this can be done already with vp_vdpa feature provisioning: >> >> > >> >> >commit c1ca352d371f724f7fb40f016abdb563aa85fe55 >> >> >Author: Jason Wang <jasowang@xxxxxxxxxx> >> >> >Date: Tue Sep 27 15:48:10 2022 +0800 >> >> > >> >> > vp_vdpa: support feature provisioning >> >> > >> >> >For example: >> >> > >> >> >vdpa dev add name dev1 mgmtdev pci/0000:02:00.0 device_features 0x300020000 >> >> >> >> Sure. My intension was to make the testing possible on any virtio >> >> device. >> > >> >It did that actually, vp_vdpa bridge virtio-pci device into vDPA bus >> >with mediation layer (like feature filtering etc). So it can only run >> >on top of standard virtio-pci device. >> > >> >> Narrowing the testing for vpda would be limitting. >> > >> >Unless you want to use other transport like virtio-mmio. >> >> Also, the goal is to test virtio_net emulated devices. >> There are couple >> of implementation. Non-vdpa. > >So what I want to say is, vp_vdpa works for all types of virtio-pci >devices no matter if it is emulated or hardware. Sure, but I wanted to have a simple generic way, working on all virtio devices, even the ones backed by a different transport, and without need of extra vdpa layer. > >Thanks > >> >> >> > >> >Thanks >> > >> >> >> >> >> >> > >> >> >Thanks >> >> > >> >> >> > >> >
