operation code. For example, we can
add new uAPIs in virtio-net which could be something like:
struct virtio_net_rss_config_header {
__le32 hash_types;
__le16 indirection_table_mask;
__le16 unclassified_queue;
__le16 indirection_table[];
}
struct virtio_net_rss_confi
On 3/8/25 10:40 PM, Mina Almasry wrote:
> From: Stanislav Fomichev
>
> Add bind-tx netlink call to attach dmabuf for TX; queue is not
> required, only ifindex and dmabuf fd for attachment.
>
> Signed-off-by: Stanislav Fomichev
> Signed-off-by: Mina Almasry
>
> ---
>
> v3:
> - Fix ynl-regen.s
On 3/8/25 10:40 PM, Mina Almasry wrote:
> diff --git a/Documentation/networking/net_cachelines/net_device.rst
> b/Documentation/networking/net_cachelines/net_device.rst
> index 6327e689e8a8..8c0334851b45 100644
> --- a/Documentation/networking/net_cachelines/net_device.rst
> +++ b/Documentation/ne
On Thu, Mar 13, 2025 at 11:21:00AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> Similarly to net.mptcp.available_schedulers, this patch adds a new one
> net.mptcp.available_path_managers to list the available path managers.
>
> Signed-off-by: Geliang Tang
> Reviewed-by: Matthie
On Thu, Mar 13, 2025 at 11:20:54AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> This patch adds a "struct_group(reset, ...)" in struct mptcp_pm_data to
> simplify the reset, and make sure we don't miss any.
>
> Suggested-by: Matthieu Baerts
> Signed-off-by: Geliang Tang
> Revi
On Thu, Mar 13, 2025 at 11:20:52AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> The variable "pm" has been defined in mptcp_pm_fully_established()
> and mptcp_pm_data_reset() as "msk->pm", so use "pm" directly instead
> of using "msk->pm".
>
> Signed-off-by: Geliang Tang
> Revi
On Thu, Mar 13, 2025 at 11:20:50AM +0100, Matthieu Baerts (NGI0) wrote:
> The registration of mptcp_genl_family is useful for both the in-kernel
> and the userspace PM. It should then be done in pm_netlink.c.
>
> On the other hand, the registration of the in-kernel pernet subsystem is
> specific t
On Thu, Mar 13, 2025 at 11:21:01AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> This patch checks if the newly added net.mptcp.path_manager is mapped
> successfully from or to the old net.mptcp.pm_type in userspace_pm.sh.
>
> Signed-off-by: Geliang Tang
> Reviewed-by: Matthieu
x27;t need
> >>>>> to bother.
> >>>>
> >>>> Yes, it is mandated. 5.1.6.4.3 Hash calculation for incoming packets
> >>>> says:
> >>>>> A device attempts to calculate a per-packet hash in the following
> >>>
On Thu, Mar 13, 2025 at 11:20:57AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> Similar to net.mptcp.scheduler, a new net.mptcp.path_manager sysctl knob
> is added to determine which path manager will be used by each newly
> created MPTCP socket by setting the name of it.
>
> De
On Thu, Mar 13, 2025 at 11:20:53AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> id_avail_bitmap of struct mptcp_pm_data is currently only used by the
> in-kernel PM, so this patch moves its initialization operation under
> the "if (pm_type == MPTCP_PM_TYPE_KERNEL)" condition.
>
On Thu, Mar 13, 2025 at 11:20:56AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> This patch defines the original in-kernel netlink path manager as a
> new struct mptcp_pm_ops named "mptcp_pm_kernel", and register it in
> mptcp_pm_kernel_register(). And define the userspace path ma
On Thu, Mar 13, 2025 at 11:20:55AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> In order to allow users to develop their own BPF-based path manager,
> this patch defines a struct ops "mptcp_pm_ops" for an MPTCP path
> manager, which contains a set of interfaces. Currently only in
On Thu, Mar 13, 2025 at 11:20:59AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> This patch adds a new proc_handler "proc_pm_type" for "pm_type" to
> map old path manager sysctl "pm_type" to the newly added "path_manager".
>
> path_manager pm_type
>
> M
On Thu, Mar 13, 2025 at 11:20:51AM +0100, Matthieu Baerts (NGI0) wrote:
> From: Geliang Tang
>
> Instead of using kmalloc() or kzalloc() to allocate an entry and
> then immediately duplicate another entry to the newly allocated
> one, kmemdup() helper can be used to simplify the code.
>
> Signed
From: Takuma Watanabe
Corrected a typo in the mseal documentation.
Signed-off-by: Takuma Watanabe
---
Documentation/userspace-api/mseal.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/Documentation/userspace-api/mseal.rst
b/Documentation/userspace-api/mseal.rst
index 4
On 2025/03/17 10:15, Jason Wang wrote:
On Wed, Mar 12, 2025 at 1:59 PM Akihiko Odaki wrote:
On 2025/03/12 12:36, Jason Wang wrote:
On Tue, Mar 11, 2025 at 2:24 PM Akihiko Odaki wrote:
On 2025/03/11 9:42, Jason Wang wrote:
On Mon, Mar 10, 2025 at 3:04 PM Akihiko Odaki wrote:
On 2025/03/
on't think there is an advantage to split ioctls to follow the spec
after all. It makes sense if we can pass-through virtio commands to
tuntap, but it is not possible as ioctl operation codes are different
from virtio commands.
I don't see a connection with the operation code. For example, we ca
Adding Kuniyuki.
On 3/8/25 10:40 PM, Mina Almasry wrote:
> @@ -931,10 +932,67 @@ int netdev_nl_bind_rx_doit(struct sk_buff *skb, struct
> genl_info *info)
> return err;
> }
>
> -/* stub */
> int netdev_nl_bind_tx_doit(struct sk_buff *skb, struct genl_info *info)
> {
> - return 0;
>
19 matches
Mail list logo
