[PATCH net-next v2] udp: Avoid post-GRO UDP checksum recalculation

[PATCH net-next v2] udp: Avoid post-GRO UDP checksum recalculation

[Sean Tranchetti]

Currently, when resegmenting an unexpected UDP GRO packet, the full UDP
checksum will be calculated for every new SKB created by skb_segment()
because the netdev features passed in by udp_rcv_segment() lack any
information about checksum offload capabilities.

Usually, we have no need to perform this calculation again, as
  1) The GRO implementation guarantees that any packets making it to the
     udp_rcv_segment() function had correct checksums, and, more
     importantly,
  2) Upon the successful return of udp_rcv_segment(), we immediately pull
     the UDP header off and either queue the segment to the socket or
     hand it off to a new protocol handler.

Unless userspace has set the IP_CHECKSUM sockopt to indicate that they
want the final checksum values, we can pass the needed netdev feature
flags to __skb_gso_segment() to avoid checksumming each segment in
skb_segment().

Fixes: cf329aa42b66 ("udp: cope with UDP GRO packet misdirection")
Cc: Paolo Abeni <pabeni@redhat.com>
Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
Signed-off-by: Sean Tranchetti <stranche@codeaurora.org>
---
 include/net/udp.h | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/include/net/udp.h b/include/net/udp.h
index d8ce937..dbe030d 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -471,12 +471,19 @@ struct udp_iter_state {
 static inline struct sk_buff *udp_rcv_segment(struct sock *sk,
 					      struct sk_buff *skb, bool ipv4)
 {
+	netdev_features_t features = NETIF_F_SG;
 	struct sk_buff *segs;
 
+	/* Avoid csum recalculation by skb_segment unless userspace explicitly
+	 * asks for the final checksum values
+	 */
+	if (!inet_get_convert_csum(sk))
+		features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
+
 	/* the GSO CB lays after the UDP one, no need to save and restore any
 	 * CB fragment
 	 */
-	segs = __skb_gso_segment(skb, NETIF_F_SG, false);
+	segs = __skb_gso_segment(skb, features, false);
 	if (unlikely(IS_ERR_OR_NULL(segs))) {
 		int segs_nr = skb_shinfo(skb)->gso_segs;
 
-- 
1.9.1

Re: [PATCH net-next v2] udp: Avoid post-GRO UDP checksum recalculation

[Paolo Abeni]

On Tue, 2019-05-28 at 12:22 -0600, Sean Tranchetti wrote:
> Currently, when resegmenting an unexpected UDP GRO packet, the full UDP
> checksum will be calculated for every new SKB created by skb_segment()
> because the netdev features passed in by udp_rcv_segment() lack any
> information about checksum offload capabilities.
> 
> Usually, we have no need to perform this calculation again, as
>   1) The GRO implementation guarantees that any packets making it to the
>      udp_rcv_segment() function had correct checksums, and, more
>      importantly,
>   2) Upon the successful return of udp_rcv_segment(), we immediately pull
>      the UDP header off and either queue the segment to the socket or
>      hand it off to a new protocol handler.
> 
> Unless userspace has set the IP_CHECKSUM sockopt to indicate that they
> want the final checksum values, we can pass the needed netdev feature
> flags to __skb_gso_segment() to avoid checksumming each segment in
> skb_segment().
> 
> Fixes: cf329aa42b66 ("udp: cope with UDP GRO packet misdirection")
> Cc: Paolo Abeni <pabeni@redhat.com>
> Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
> Signed-off-by: Sean Tranchetti <stranche@codeaurora.org>
> ---
>  include/net/udp.h | 9 ++++++++-
>  1 file changed, 8 insertions(+), 1 deletion(-)
> 
> diff --git a/include/net/udp.h b/include/net/udp.h
> index d8ce937..dbe030d 100644
> --- a/include/net/udp.h
> +++ b/include/net/udp.h
> @@ -471,12 +471,19 @@ struct udp_iter_state {
>  static inline struct sk_buff *udp_rcv_segment(struct sock *sk,
>  					      struct sk_buff *skb, bool ipv4)
>  {
> +	netdev_features_t features = NETIF_F_SG;
>  	struct sk_buff *segs;
>  
> +	/* Avoid csum recalculation by skb_segment unless userspace explicitly
> +	 * asks for the final checksum values
> +	 */
> +	if (!inet_get_convert_csum(sk))
> +		features |= NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
> +
>  	/* the GSO CB lays after the UDP one, no need to save and restore any
>  	 * CB fragment
>  	 */
> -	segs = __skb_gso_segment(skb, NETIF_F_SG, false);
> +	segs = __skb_gso_segment(skb, features, false);
>  	if (unlikely(IS_ERR_OR_NULL(segs))) {
>  		int segs_nr = skb_shinfo(skb)->gso_segs;

The patch itself LGTM, thanks.

Acked-by: Paolo Abeni <pabeni@redhat.com>

Possibly this can target the 'net' tree as the relevant commit is
already there and the patch itself is not very invasive.

Paolo

Re: [PATCH net-next v2] udp: Avoid post-GRO UDP checksum recalculation

[David Miller]

From: Sean Tranchetti <stranche@codeaurora.org>
Date: Tue, 28 May 2019 12:22:54 -0600

> Currently, when resegmenting an unexpected UDP GRO packet, the full UDP
> checksum will be calculated for every new SKB created by skb_segment()
> because the netdev features passed in by udp_rcv_segment() lack any
> information about checksum offload capabilities.
> 
> Usually, we have no need to perform this calculation again, as
>   1) The GRO implementation guarantees that any packets making it to the
>      udp_rcv_segment() function had correct checksums, and, more
>      importantly,
>   2) Upon the successful return of udp_rcv_segment(), we immediately pull
>      the UDP header off and either queue the segment to the socket or
>      hand it off to a new protocol handler.
> 
> Unless userspace has set the IP_CHECKSUM sockopt to indicate that they
> want the final checksum values, we can pass the needed netdev feature
> flags to __skb_gso_segment() to avoid checksumming each segment in
> skb_segment().
> 
> Fixes: cf329aa42b66 ("udp: cope with UDP GRO packet misdirection")
> Cc: Paolo Abeni <pabeni@redhat.com>
> Cc: Subash Abhinov Kasiviswanathan <subashab@codeaurora.org>
> Signed-off-by: Sean Tranchetti <stranche@codeaurora.org>

I've decided to apply this to 'net', thank you.