On Mon, Mar 18, 2024 at 7:06 PM Xuan Zhuo wrote:
>
> As the spec
> https://github.com/oasis-tcs/virtio-spec/commit/42f389989823039724f95bbbd243291ab0064f82
>
> make virtio-net support getting the stats from the device by ethtool -S
> .
>
> Due to the numerous descriptors stats, an organization me
On Mon, Mar 18, 2024 at 7:06 PM Xuan Zhuo wrote:
>
> Now, we have this:
>
> tx_queue_0_tx_timeouts
>
> This is used to record the tx schedule timeout.
> But this has two "tx". I think the below is enough.
>
> tx_queue_0_timeouts
>
> So I rename this field.
>
> Signed-off-by: Xuan Zhuo
> R
On Mon, Mar 18, 2024 at 7:06 PM Xuan Zhuo wrote:
>
> The key size of ethtool -S is controlled by this macro.
>
> ETH_GSTRING_LEN 32
>
> That includes the \0 at the end. So the max length of the key name must
> is 31. But the length of the prefix "rx_queue_0_" is 11. If the queue
> num is larger th
On Mon, Mar 18, 2024 at 7:06 PM Xuan Zhuo wrote:
>
> As the spec
> https://github.com/oasis-tcs/virtio-spec/commit/42f389989823039724f95bbbd243291ab0064f82
>
> The virtnet cvq supports to get result from the device.
>
> Signed-off-by: Xuan Zhuo
> ---
> drivers/net/virtio_net.c | 47
On Mon, Mar 18, 2024 at 7:06 PM Xuan Zhuo wrote:
>
> The virtio-net device stats spec:
>
> https://github.com/oasis-tcs/virtio-spec/commit/42f389989823039724f95bbbd243291ab0064f82
>
> We introduce the relative feature and structures.
>
> Signed-off-by: Xuan Zhuo
Acked-by: Jason Wang
Thanks
在 2024/4/10 上午9:44, Jakub Kicinski 写道:
On Tue, 9 Apr 2024 20:03:21 +0800 Heng Qi wrote:
+/**
+ * coalesce_put_profile - fill reply with a nla nest with four child nla nests.
+ * @skb: socket buffer the message is stored in
+ * @attr_type: nest attr type ETHTOOL_A_COALESCE_*X_*QE_PROFILE
+ *
在 2024/4/10 上午9:40, Jakub Kicinski 写道:
On Tue, 9 Apr 2024 20:03:24 +0800 Heng Qi wrote:
+ /* DIM profile list */
+ struct dim_cq_moder rx_eqe_conf[NET_DIM_PARAMS_NUM_PROFILES];
Can you please wrap this into a structure with other necessary
information and add a pointer in struct
On Tue, 9 Apr 2024 20:03:21 +0800 Heng Qi wrote:
> +/**
> + * coalesce_put_profile - fill reply with a nla nest with four child nla
> nests.
> + * @skb: socket buffer the message is stored in
> + * @attr_type: nest attr type ETHTOOL_A_COALESCE_*X_*QE_PROFILE
> + * @profile: data passed to userspa
On Tue, 9 Apr 2024 20:03:24 +0800 Heng Qi wrote:
> + /* DIM profile list */
> + struct dim_cq_moder rx_eqe_conf[NET_DIM_PARAMS_NUM_PROFILES];
Can you please wrap this into a structure with other necessary
information and add a pointer in struct net_device instead.
What's the point of eve
Virtio-net has different types of back-end device
implementations. In order to effectively optimize
the dim library's gains for different device
implementations, let's use the new interface params
to fine-tune the profile list.
Signed-off-by: Heng Qi
---
drivers/net/virtio_net.c | 41 +++
Extract the initialization and destruction actions
of dim for use in the next patch.
Signed-off-by: Heng Qi
---
drivers/net/virtio_net.c | 37 ++---
1 file changed, 26 insertions(+), 11 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
The NetDIM library, currently leveraged by an array of NICs, delivers
excellent acceleration benefits. Nevertheless, NICs vary significantly
in their dim profile list prerequisites.
Specifically, virtio-net backends may present diverse sw or hw device
implementation, making a one-size-fits-all par
Place profiles into dim.h file so that the subsequent patch can use
it to initialize driver's custom profiles.
Signed-off-by: Heng Qi
---
include/linux/dim.h | 38 ++
lib/dim/net_dim.c | 38 --
2 files changed, 38 insertio
The NetDIM library provides excellent acceleration for many modern
network cards. However, the default profiles of DIM limits its maximum
capabilities for different NICs, so providing a way which the NIC can
be custom configured is necessary.
Currently, interaction with the driver is still based o
Will modify things as you sugguested.
Thanks!
在 2024/4/9 上午3:13, Jakub Kicinski 写道:
On Mon, 8 Apr 2024 11:44:27 +0800 Heng Qi wrote:
+ nla_for_each_nested(nest, nests, rem) {
+ if (WARN_ONCE(nla_type(nest) !=
ETHTOOL_A_MODERATIONS_MODERATION,
+
15 matches
Mail list logo