On Mon, 2016-05-09 at 10:39 -0700, Eric Dumazet wrote:
> On Mon, 2016-05-09 at 10:22 -0700, Alexander Duyck wrote:
>
> > One thought that just occurred to me based on this would be to
> > configure inner headers on the way up, and to configure the outer
> > headers on the way down. Then that way
On Mon, 2016-05-09 at 10:22 -0700, Alexander Duyck wrote:
> One thought that just occurred to me based on this would be to
> configure inner headers on the way up, and to configure the outer
> headers on the way down. Then that way we could go through and be
> guaranteed that the inner headers re
On Sun, May 8, 2016 at 10:35 AM, Eric Dumazet wrote:
> On Sun, 2016-05-08 at 09:08 -0700, Eric Dumazet wrote:
>
>
>> So we probably need to make sure the network header is properly set for
>> the segments. Then skb_reset_mac_len(nskb); would work as intended.
>>
>> Since skb_segment() is called fr
From: Eric Dumazet
Date: Fri, 06 May 2016 18:19:59 -0700
> From: Eric Dumazet
>
> When using ifb+netem on ingress on SIT/IPIP/GRE traffic,
> GRO packets are not properly processed.
>
> Segmentation should not be forced, since ifb is already adding
> quite a performance hit.
>
> Signed-off-by:
On Sun, 2016-05-08 at 09:08 -0700, Eric Dumazet wrote:
> So we probably need to make sure the network header is properly set for
> the segments. Then skb_reset_mac_len(nskb); would work as intended.
>
> Since skb_segment() is called from the deepest point in GSO path,
> it always see the inner n
On Sat, 2016-05-07 at 22:41 -0700, Eric Dumazet wrote:
> On Fri, 2016-05-06 at 18:19 -0700, Eric Dumazet wrote:
> > From: Eric Dumazet
> >
> > When using ifb+netem on ingress on SIT/IPIP/GRE traffic,
> > GRO packets are not properly processed.
> >
> > Segmentation should not be forced, since ifb
On Fri, 2016-05-06 at 18:19 -0700, Eric Dumazet wrote:
> From: Eric Dumazet
>
> When using ifb+netem on ingress on SIT/IPIP/GRE traffic,
> GRO packets are not properly processed.
>
> Segmentation should not be forced, since ifb is already adding
> quite a performance hit.
>
> Signed-off-by: Eri
From: Eric Dumazet
When using ifb+netem on ingress on SIT/IPIP/GRE traffic,
GRO packets are not properly processed.
Segmentation should not be forced, since ifb is already adding
quite a performance hit.
Signed-off-by: Eric Dumazet
---
drivers/net/ifb.c |3 +++
1 file changed, 3 insertion
