Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1350659 > unrolled thread
| Started by | Bill Sommerfeld <wsommerfeld@google.com> |
|---|---|
| First post | 2016-03-04 23:50 +0100 |
| Last post | 2016-03-07 21:30 +0100 |
| Articles | 3 — 3 participants |
Back to article view | Back to linux.kernel
[PATCH] udp6: fix UDP/IPv6 encap resubmit path Bill Sommerfeld <wsommerfeld@google.com> - 2016-03-04 23:50 +0100
Re: [PATCH] udp6: fix UDP/IPv6 encap resubmit path Josh Hunt <johunt@akamai.com> - 2016-03-07 20:40 +0100
Re: [PATCH] udp6: fix UDP/IPv6 encap resubmit path David Miller <davem@davemloft.net> - 2016-03-07 21:30 +0100
| From | Bill Sommerfeld <wsommerfeld@google.com> |
|---|---|
| Date | 2016-03-04 23:50 +0100 |
| Subject | [PATCH] udp6: fix UDP/IPv6 encap resubmit path |
| Message-ID | <r96HM-5lt-17@gated-at.bofh.it> |
IPv4 interprets a negative return value from a protocol handler as a request to redispatch to a new protocol. In contrast, IPv6 interprets a negative value as an error, and interprets a positive value as a request for redispatch. UDP for IPv6 was unaware of this difference. Change __udp6_lib_rcv() to return a positive value for redispatch. Note that the socket's encap_rcv hook still needs to return a negative value to request dispatch, and in the case of IPv6 packets, adjust IP6CB(skb)->nhoff to identify the byte containing the next protocol. Signed-off-by: Bill Sommerfeld <wsommerfeld@google.com> --- net/ipv6/udp.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 0711f8f..fd25e44 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -922,11 +922,9 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, ret = udpv6_queue_rcv_skb(sk, skb); sock_put(sk); - /* a return value > 0 means to resubmit the input, but - * it wants the return to be -protocol, or 0 - */ + /* a return value > 0 means to resubmit the input */ if (ret > 0) - return -ret; + return ret; return 0; } -- 2.7.0.rc3.207.g0ac5344
[toc] | [next] | [standalone]
| From | Josh Hunt <johunt@akamai.com> |
|---|---|
| Date | 2016-03-07 20:40 +0100 |
| Message-ID | <ra9ay-784-7@gated-at.bofh.it> |
| In reply to | #1350659 |
On 03/04/2016 04:47 PM, Bill Sommerfeld wrote: > IPv4 interprets a negative return value from a protocol handler as a > request to redispatch to a new protocol. In contrast, IPv6 interprets a > negative value as an error, and interprets a positive value as a request > for redispatch. > > UDP for IPv6 was unaware of this difference. Change __udp6_lib_rcv() to > return a positive value for redispatch. Note that the socket's > encap_rcv hook still needs to return a negative value to request > dispatch, and in the case of IPv6 packets, adjust IP6CB(skb)->nhoff to > identify the byte containing the next protocol. > > Signed-off-by: Bill Sommerfeld <wsommerfeld@google.com> > --- > net/ipv6/udp.c | 6 ++---- > 1 file changed, 2 insertions(+), 4 deletions(-) > > diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c > index 0711f8f..fd25e44 100644 > --- a/net/ipv6/udp.c > +++ b/net/ipv6/udp.c > @@ -922,11 +922,9 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable, > ret = udpv6_queue_rcv_skb(sk, skb); > sock_put(sk); > > - /* a return value > 0 means to resubmit the input, but > - * it wants the return to be -protocol, or 0 > - */ > + /* a return value > 0 means to resubmit the input */ > if (ret > 0) > - return -ret; > + return ret; > > return 0; > } > This looks good to me. Thanks Bill! Josh
[toc] | [prev] | [next] | [standalone]
| From | David Miller <davem@davemloft.net> |
|---|---|
| Date | 2016-03-07 21:30 +0100 |
| Message-ID | <ra9WW-7Gx-15@gated-at.bofh.it> |
| In reply to | #1350659 |
From: Bill Sommerfeld <wsommerfeld@google.com> Date: Fri, 4 Mar 2016 14:47:21 -0800 > IPv4 interprets a negative return value from a protocol handler as a > request to redispatch to a new protocol. In contrast, IPv6 interprets a > negative value as an error, and interprets a positive value as a request > for redispatch. > > UDP for IPv6 was unaware of this difference. Change __udp6_lib_rcv() to > return a positive value for redispatch. Note that the socket's > encap_rcv hook still needs to return a negative value to request > dispatch, and in the case of IPv6 packets, adjust IP6CB(skb)->nhoff to > identify the byte containing the next protocol. > > Signed-off-by: Bill Sommerfeld <wsommerfeld@google.com> Applied and queued up for -stable, thanks.
[toc] | [prev] | [standalone]
Back to top | Article view | linux.kernel
csiph-web