On 2018年09月07日 02:00, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:25PM +0800, Jason Wang wrote:
This patch implement TUN_MSG_PTR msg_control type. This type allows
the caller to pass an array of XDP buffs to tuntap through ptr field
of the tun_msg_control. Tap will build skb
On 2018年09月07日 00:54, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:23PM +0800, Jason Wang wrote:
This patch introduces to a new tun/tap specific msg_control:
#define TUN_MSG_UBUF 1
#define TUN_MSG_PTR 2
struct tun_msg_ctl {
int type;
void *ptr;
};
This allows us
On 2018年09月07日 01:48, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:22PM +0800, Jason Wang wrote:
This will allow adding batch flushing on top.
Signed-off-by: Jason Wang
---
drivers/net/tun.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git
On 2018年09月07日 01:21, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:21PM +0800, Jason Wang wrote:
This patch split out XDP logic into a single function. This make it to
be reused by XDP batching path in the following patch.
Signed-off-by: Jason Wang
---
drivers/net/tun.c | 84
On 2018年09月07日 01:16, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:20PM +0800, Jason Wang wrote:
If we're sure not to go native XDP, there's no need for several things
like bh and rcu stuffs.
Signed-off-by: Jason Wang
True...
---
drivers/net/tun.c | 9 ++---
1 file
On 2018年09月07日 01:14, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:19PM +0800, Jason Wang wrote:
There's no need to duplicate page get logic in each action. So this
patch tries to get page and calculate the offset before processing XDP
actions, and undo them when meet errors (we
On 2018年09月07日 00:57, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:17PM +0800, Jason Wang wrote:
Signed-off-by: Jason Wang
Acked-by: Michael S. Tsirkin
any idea why didn't we do this straight away?
Dunno, but git grep told me not all XDP capable driver use this (e.g
On 2018年09月07日 00:56, Michael S. Tsirkin wrote:
On Thu, Sep 06, 2018 at 12:05:16PM +0800, Jason Wang wrote:
This patch introduces a new sock flag - SOCK_XDP. This will be used
for notifying the upper layer that XDP program is attached on the
lower socket, and requires for extra headroom.
TUN
On Mon, Aug 27, 2018 at 05:00:40PM +0300, Michael S. Tsirkin wrote:
> Are there still plans to test the performance with vost pmd?
> vhost doesn't seem to show a performance gain ...
>
I tried some performance tests with vhost PMD. In guest, the
XDP program will return XDP_DROP directly. And in
On Thu, Sep 06, 2018 at 12:05:25PM +0800, Jason Wang wrote:
> This patch implement TUN_MSG_PTR msg_control type. This type allows
> the caller to pass an array of XDP buffs to tuntap through ptr field
> of the tun_msg_control. Tap will build skb through those XDP buffers.
>
> This will avoid lots
On Thu, Sep 06, 2018 at 12:05:24PM +0800, Jason Wang wrote:
> This patch implement TUN_MSG_PTR msg_control type. This type allows
> the caller to pass an array of XDP buffs to tuntap through ptr field
> of the tun_msg_control. If an XDP program is attached, tuntap can run
> XDP program directly.
On Thu, Sep 06, 2018 at 12:05:22PM +0800, Jason Wang wrote:
> This will allow adding batch flushing on top.
>
> Signed-off-by: Jason Wang
> ---
> drivers/net/tun.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/tun.c b/drivers/net/tun.c
> index
On Thu, Sep 06, 2018 at 12:05:21PM +0800, Jason Wang wrote:
> This patch split out XDP logic into a single function. This make it to
> be reused by XDP batching path in the following patch.
>
> Signed-off-by: Jason Wang
> ---
> drivers/net/tun.c | 84
On Thu, Sep 06, 2018 at 12:05:20PM +0800, Jason Wang wrote:
> If we're sure not to go native XDP, there's no need for several things
> like bh and rcu stuffs.
>
> Signed-off-by: Jason Wang
True...
> ---
> drivers/net/tun.c | 9 ++---
> 1 file changed, 6 insertions(+), 3 deletions(-)
>
>
On Thu, Sep 06, 2018 at 12:05:19PM +0800, Jason Wang wrote:
> There's no need to duplicate page get logic in each action. So this
> patch tries to get page and calculate the offset before processing XDP
> actions, and undo them when meet errors (we don't care the performance
> on errors). This
On Thu, Sep 06, 2018 at 12:05:18PM +0800, Jason Wang wrote:
> This patch move the bh enabling a little bit earlier, this will be
> used for factoring out the core XDP logic of tuntap.
>
> Signed-off-by: Jason Wang
Acked-by: Michael S. Tsirkin
no reason to disable bh for non-xdp things.
> ---
On Thu, Sep 06, 2018 at 12:05:17PM +0800, Jason Wang wrote:
> Signed-off-by: Jason Wang
Acked-by: Michael S. Tsirkin
any idea why didn't we do this straight away?
> ---
> drivers/net/tun.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/drivers/net/tun.c
On Thu, Sep 06, 2018 at 12:05:16PM +0800, Jason Wang wrote:
> This patch introduces a new sock flag - SOCK_XDP. This will be used
> for notifying the upper layer that XDP program is attached on the
> lower socket, and requires for extra headroom.
>
> TUN will be the first user.
>
>
On Thu, Sep 06, 2018 at 12:05:23PM +0800, Jason Wang wrote:
> This patch introduces to a new tun/tap specific msg_control:
>
> #define TUN_MSG_UBUF 1
> #define TUN_MSG_PTR 2
> struct tun_msg_ctl {
>int type;
>void *ptr;
> };
>
> This allows us to pass different kinds of
On Thu, Sep 06, 2018 at 12:05:26PM +0800, Jason Wang wrote:
> This patch implements XDP batching for vhost_net. The idea is first to
> try to do userspace copy and build XDP buff directly in vhost. Instead
> of submitting the packet immediately, vhost_net will batch them in an
> array and submit
On Thu, Sep 06, 2018 at 10:18:42AM +0100, David Howells wrote:
> The virtio_net_ctrl_hdr struct uses a C++ keyword as structural members. Fix
> this by inserting an anonymous union that provides an alternative name and
> then hide the reserved name in C++.
>
> Signed-off-by: David Howells
> cc:
On Thu, Sep 06, 2018 at 08:09:19AM +0100, David Howells wrote:
> Michael S. Tsirkin wrote:
>
> > As long as you do not intend to use any classes, how about
> > simply adding
> >
> > -Dclass=_class
> >
> > to your command line?
>
> That kind of misses the point;-). It's not reasonable to
On 07/23/2018 10:36 PM, Dr. David Alan Gilbert wrote:
* Michael S. Tsirkin (m...@redhat.com) wrote:
On Fri, Jul 20, 2018 at 04:33:00PM +0800, Wei Wang wrote:
This patch series is separated from the previous "Virtio-balloon
Enhancement" series. The new feature, VIRTIO_BALLOON_F_FREE_PAGE_HINT,
Here's a set of patches that inserts a step into the build process to make
sure that the UAPI headers can all be built together with C++ (if the
compiler being used supports C++).
Note that it's based on a commit from the sound tree to fix usage of u32
and co..
Most of the patches perform
The virtio_net_ctrl_hdr struct uses a C++ keyword as structural members. Fix
this by inserting an anonymous union that provides an alternative name and
then hide the reserved name in C++.
Signed-off-by: David Howells
cc: "Michael S. Tsirkin"
cc: Jason Wang
cc:
Hi,
> > You can probably get rid of this one if you're refactoring even more. The
> > generic fb_probe implementation (already merged) plus gem-shmem support
> > for it (still in flight) from Noralf should be able to pull that off. That
> > gives you the fb_mmap implementation, but with 100%
Michael S. Tsirkin wrote:
> As long as you do not intend to use any classes, how about
> simply adding
>
> -Dclass=_class
>
> to your command line?
That kind of misses the point;-). It's not reasonable to expect all userspace
C++ users to do this.
David
27 matches
Mail list logo