* [RFC PATCH net] udp: check sk for UDP GRO fraglist
[not found] <CGME20210108130414epcas2p3217d7b6ac8a8094c5b3b6c5e52480134@epcas2p3.samsung.com>
@ 2021-01-08 12:52 ` Dongseok Yi
2021-01-08 13:35 ` Steffen Klassert
0 siblings, 1 reply; 5+ messages in thread
From: Dongseok Yi @ 2021-01-08 12:52 UTC (permalink / raw)
To: David S. Miller
Cc: namkyu78.kim, Dongseok Yi, Alexey Kuznetsov, Hideaki YOSHIFUJI,
Jakub Kicinski, Willem de Bruijn, Steffen Klassert, netdev,
linux-kernel
It is a workaround patch.
UDP/IP header of UDP GROed frag_skbs are not updated even after NAT
forwarding. Only the header of head_skb from ip_finish_output_gso ->
skb_gso_segment is updated but following frag_skbs are not updated.
A call path skb_mac_gso_segment -> inet_gso_segment ->
udp4_ufo_fragment -> __udp_gso_segment -> __udp_gso_segment_list
does not try to update any UDP/IP header of the segment list.
It might make sense because each skb of frag_skbs is converted to a
list of regular packets. Header update with checksum calculation may
be not needed for UDP GROed frag_skbs.
But UDP GRO frag_list is started from udp_gro_receive, we don't know
whether the skb will be NAT forwarded at that time. For workaround,
try to get sock always when call udp4_gro_receive -> udp_gro_receive
to check if the skb is for local.
I'm still not sure if UDP GRO frag_list is really designed for local
session only. Can kernel support NAT forward for UDP GRO frag_list?
What am I missing?
Fixes: 9fd1ff5d2ac7 (udp: Support UDP fraglist GRO/GSO.)
Signed-off-by: Dongseok Yi <dseok.yi@samsung•com>
---
net/ipv4/udp_offload.c | 5 ++---
1 file changed, 2 insertions(+), 3 deletions(-)
diff --git a/net/ipv4/udp_offload.c b/net/ipv4/udp_offload.c
index ff39e94..d476216 100644
--- a/net/ipv4/udp_offload.c
+++ b/net/ipv4/udp_offload.c
@@ -457,7 +457,7 @@ struct sk_buff *udp_gro_receive(struct list_head *head, struct sk_buff *skb,
int flush = 1;
NAPI_GRO_CB(skb)->is_flist = 0;
- if (skb->dev->features & NETIF_F_GRO_FRAGLIST)
+ if (sk && (skb->dev->features & NETIF_F_GRO_FRAGLIST))
NAPI_GRO_CB(skb)->is_flist = sk ? !udp_sk(sk)->gro_enabled: 1;
if ((sk && udp_sk(sk)->gro_enabled) || NAPI_GRO_CB(skb)->is_flist) {
@@ -537,8 +537,7 @@ struct sk_buff *udp4_gro_receive(struct list_head *head, struct sk_buff *skb)
NAPI_GRO_CB(skb)->is_ipv6 = 0;
rcu_read_lock();
- if (static_branch_unlikely(&udp_encap_needed_key))
- sk = udp4_gro_lookup_skb(skb, uh->source, uh->dest);
+ sk = udp4_gro_lookup_skb(skb, uh->source, uh->dest);
pp = udp_gro_receive(head, skb, uh, sk);
rcu_read_unlock();
--
2.7.4
^ permalink raw reply related [flat|nested] 5+ messages in thread* Re: [RFC PATCH net] udp: check sk for UDP GRO fraglist
2021-01-08 12:52 ` [RFC PATCH net] udp: check sk for UDP GRO fraglist Dongseok Yi
@ 2021-01-08 13:35 ` Steffen Klassert
2021-01-11 2:02 ` Dongseok Yi
0 siblings, 1 reply; 5+ messages in thread
From: Steffen Klassert @ 2021-01-08 13:35 UTC (permalink / raw)
To: Dongseok Yi
Cc: David S. Miller, namkyu78.kim, Alexey Kuznetsov,
Hideaki YOSHIFUJI, Jakub Kicinski, Willem de Bruijn, netdev,
linux-kernel
On Fri, Jan 08, 2021 at 09:52:28PM +0900, Dongseok Yi wrote:
> It is a workaround patch.
>
> UDP/IP header of UDP GROed frag_skbs are not updated even after NAT
> forwarding. Only the header of head_skb from ip_finish_output_gso ->
> skb_gso_segment is updated but following frag_skbs are not updated.
>
> A call path skb_mac_gso_segment -> inet_gso_segment ->
> udp4_ufo_fragment -> __udp_gso_segment -> __udp_gso_segment_list
> does not try to update any UDP/IP header of the segment list.
>
> It might make sense because each skb of frag_skbs is converted to a
> list of regular packets. Header update with checksum calculation may
> be not needed for UDP GROed frag_skbs.
>
> But UDP GRO frag_list is started from udp_gro_receive, we don't know
> whether the skb will be NAT forwarded at that time. For workaround,
> try to get sock always when call udp4_gro_receive -> udp_gro_receive
> to check if the skb is for local.
>
> I'm still not sure if UDP GRO frag_list is really designed for local
> session only. Can kernel support NAT forward for UDP GRO frag_list?
> What am I missing?
The initial idea when I implemented this was to have a fast
forwarding path for UDP. So forwarding is a usecase, but NAT
is a problem, indeed. A quick fix could be to segment the
skb before it gets NAT forwarded. Alternatively we could
check for a header change in __udp_gso_segment_list and
update the header of the frag_skbs accordingly in that case.
^ permalink raw reply [flat|nested] 5+ messages in thread
* RE: [RFC PATCH net] udp: check sk for UDP GRO fraglist
2021-01-08 13:35 ` Steffen Klassert
@ 2021-01-11 2:02 ` Dongseok Yi
2021-01-11 8:43 ` Steffen Klassert
0 siblings, 1 reply; 5+ messages in thread
From: Dongseok Yi @ 2021-01-11 2:02 UTC (permalink / raw)
To: 'Steffen Klassert'
Cc: 'David S. Miller', namkyu78.kim,
'Alexey Kuznetsov', 'Hideaki YOSHIFUJI',
'Jakub Kicinski', 'Willem de Bruijn', netdev,
linux-kernel
On 2021-01-08 22:35, Steffen Klassert wrote:
> On Fri, Jan 08, 2021 at 09:52:28PM +0900, Dongseok Yi wrote:
> > It is a workaround patch.
> >
> > UDP/IP header of UDP GROed frag_skbs are not updated even after NAT
> > forwarding. Only the header of head_skb from ip_finish_output_gso ->
> > skb_gso_segment is updated but following frag_skbs are not updated.
> >
> > A call path skb_mac_gso_segment -> inet_gso_segment ->
> > udp4_ufo_fragment -> __udp_gso_segment -> __udp_gso_segment_list
> > does not try to update any UDP/IP header of the segment list.
> >
> > It might make sense because each skb of frag_skbs is converted to a
> > list of regular packets. Header update with checksum calculation may
> > be not needed for UDP GROed frag_skbs.
> >
> > But UDP GRO frag_list is started from udp_gro_receive, we don't know
> > whether the skb will be NAT forwarded at that time. For workaround,
> > try to get sock always when call udp4_gro_receive -> udp_gro_receive
> > to check if the skb is for local.
> >
> > I'm still not sure if UDP GRO frag_list is really designed for local
> > session only. Can kernel support NAT forward for UDP GRO frag_list?
> > What am I missing?
>
> The initial idea when I implemented this was to have a fast
> forwarding path for UDP. So forwarding is a usecase, but NAT
> is a problem, indeed. A quick fix could be to segment the
> skb before it gets NAT forwarded. Alternatively we could
> check for a header change in __udp_gso_segment_list and
> update the header of the frag_skbs accordingly in that case.
Thank you for explaining.
Can I think of it as a known issue? I think we should have a fix
because NAT can be triggered by user. Can I check the current status?
Already planning a patch or a new patch should be written?
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFC PATCH net] udp: check sk for UDP GRO fraglist
2021-01-11 2:02 ` Dongseok Yi
@ 2021-01-11 8:43 ` Steffen Klassert
0 siblings, 0 replies; 5+ messages in thread
From: Steffen Klassert @ 2021-01-11 8:43 UTC (permalink / raw)
To: Dongseok Yi
Cc: 'David S. Miller', namkyu78.kim,
'Alexey Kuznetsov', 'Hideaki YOSHIFUJI',
'Jakub Kicinski', 'Willem de Bruijn', netdev,
linux-kernel
On Mon, Jan 11, 2021 at 11:02:42AM +0900, Dongseok Yi wrote:
> On 2021-01-08 22:35, Steffen Klassert wrote:
> > On Fri, Jan 08, 2021 at 09:52:28PM +0900, Dongseok Yi wrote:
> > > It is a workaround patch.
> > >
> > > UDP/IP header of UDP GROed frag_skbs are not updated even after NAT
> > > forwarding. Only the header of head_skb from ip_finish_output_gso ->
> > > skb_gso_segment is updated but following frag_skbs are not updated.
> > >
> > > A call path skb_mac_gso_segment -> inet_gso_segment ->
> > > udp4_ufo_fragment -> __udp_gso_segment -> __udp_gso_segment_list
> > > does not try to update any UDP/IP header of the segment list.
> > >
> > > It might make sense because each skb of frag_skbs is converted to a
> > > list of regular packets. Header update with checksum calculation may
> > > be not needed for UDP GROed frag_skbs.
> > >
> > > But UDP GRO frag_list is started from udp_gro_receive, we don't know
> > > whether the skb will be NAT forwarded at that time. For workaround,
> > > try to get sock always when call udp4_gro_receive -> udp_gro_receive
> > > to check if the skb is for local.
> > >
> > > I'm still not sure if UDP GRO frag_list is really designed for local
> > > session only. Can kernel support NAT forward for UDP GRO frag_list?
> > > What am I missing?
> >
> > The initial idea when I implemented this was to have a fast
> > forwarding path for UDP. So forwarding is a usecase, but NAT
> > is a problem, indeed. A quick fix could be to segment the
> > skb before it gets NAT forwarded. Alternatively we could
> > check for a header change in __udp_gso_segment_list and
> > update the header of the frag_skbs accordingly in that case.
>
> Thank you for explaining.
> Can I think of it as a known issue?
No, it was not known before you reported it.
> I think we should have a fix
> because NAT can be triggered by user. Can I check the current status?
> Already planning a patch or a new patch should be written?
We have to do a new patch to fix that issue. If you want do
do so, go ahead.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [RFC PATCH net] udp: check sk for UDP GRO fraglist
@ 2021-01-11 12:09 Alexander Lobakin
0 siblings, 0 replies; 5+ messages in thread
From: Alexander Lobakin @ 2021-01-11 12:09 UTC (permalink / raw)
To: Steffen Klassert
Cc: Alexander Lobakin, Dongseok Yi, 'David S. Miller',
namkyu78.kim, 'Alexey Kuznetsov',
'Hideaki YOSHIFUJI', 'Jakub Kicinski',
'Willem de Bruijn', netdev, linux-kernel
From: Steffen Klassert <steffen.klassert@secunet•com>
Date: Mon, 11 Jan 2021 09:43:22 +0100
> On Mon, Jan 11, 2021 at 11:02:42AM +0900, Dongseok Yi wrote:
>> On 2021-01-08 22:35, Steffen Klassert wrote:
>>> On Fri, Jan 08, 2021 at 09:52:28PM +0900, Dongseok Yi wrote:
>>>> It is a workaround patch.
>>>>
>>>> UDP/IP header of UDP GROed frag_skbs are not updated even after NAT
>>>> forwarding. Only the header of head_skb from ip_finish_output_gso ->
>>>> skb_gso_segment is updated but following frag_skbs are not updated.
>>>>
>>>> A call path skb_mac_gso_segment -> inet_gso_segment ->
>>>> udp4_ufo_fragment -> __udp_gso_segment -> __udp_gso_segment_list
>>>> does not try to update any UDP/IP header of the segment list.
>>>>
>>>> It might make sense because each skb of frag_skbs is converted to a
>>>> list of regular packets. Header update with checksum calculation may
>>>> be not needed for UDP GROed frag_skbs.
>>>>
>>>> But UDP GRO frag_list is started from udp_gro_receive, we don't know
>>>> whether the skb will be NAT forwarded at that time. For workaround,
>>>> try to get sock always when call udp4_gro_receive -> udp_gro_receive
>>>> to check if the skb is for local.
>>>>
>>>> I'm still not sure if UDP GRO frag_list is really designed for local
>>>> session only. Can kernel support NAT forward for UDP GRO frag_list?
>>>> What am I missing?
>>>
>>> The initial idea when I implemented this was to have a fast
>>> forwarding path for UDP. So forwarding is a usecase, but NAT
>>> is a problem, indeed. A quick fix could be to segment the
>>> skb before it gets NAT forwarded. Alternatively we could
>>> check for a header change in __udp_gso_segment_list and
>>> update the header of the frag_skbs accordingly in that case.
>>
>> Thank you for explaining.
>> Can I think of it as a known issue?
>
> No, it was not known before you reported it.
>
>> I think we should have a fix
>> because NAT can be triggered by user. Can I check the current status?
>> Already planning a patch or a new patch should be written?
>
> We have to do a new patch to fix that issue. If you want do
> do so, go ahead.
This patch is incorrect. I do NAT UDP GRO Fraglists via nftables
(both with and without flow offload) with no issues since March'20.
Packet loss rates are always +/- 0, so I can say it works properly.
I can share any details / dump any runtime data if needed.
Thanks,
Al
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2021-01-11 12:11 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <CGME20210108130414epcas2p3217d7b6ac8a8094c5b3b6c5e52480134@epcas2p3.samsung.com>
2021-01-08 12:52 ` [RFC PATCH net] udp: check sk for UDP GRO fraglist Dongseok Yi
2021-01-08 13:35 ` Steffen Klassert
2021-01-11 2:02 ` Dongseok Yi
2021-01-11 8:43 ` Steffen Klassert
2021-01-11 12:09 Alexander Lobakin
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox