Thread (28 messages) 28 messages, 6 authors, 2020-03-23

Re: [PATCH] net: Make skb_segment not to compute checksum if network controller supports checksumming

From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Date: 2020-02-23 02:41:58

On Thu, Feb 20, 2020 at 11:14 PM Yadu Kishore [off-list ref] wrote:
Problem:
TCP checksum in the output path is not being offloaded during GSO
in the following case:
The network driver does not support scatter-gather but supports
checksum offload with NETIF_F_HW_CSUM.

Cause:
skb_segment calls skb_copy_and_csum_bits if the network driver
does not announce NETIF_F_SG. It does not check if the driver
supports NETIF_F_HW_CSUM.
So for devices which might want to offload checksum but do not support SG
there is currently no way to do so if GSO is enabled.

Solution:
In skb_segment check if the network controller does checksum and if so
call skb_copy_bits instead of skb_copy_and_csum_bits.

Testing:
Without the patch, ran iperf TCP traffic with NETIF_F_HW_CSUM enabled
in the network driver. Observed the TCP checksum offload is not happening
since the skbs received by the driver in the output path have
skb->ip_summed set to CHECKSUM_NONE.

With the patch ran iperf TCP traffic and observed that TCP checksum
is being offloaded with skb->ip_summed set to CHECKSUM_PARTIAL.
Did you measure a cycle efficiency improvement? As discussed in the
referred email thread, the kernel uses checksum_and_copy because it is
generally not significantly more expensive than copy alone.

skb_segment already is a very complex function. New code needs to
offer a tangible benefit.
Also tested with the patch by disabling NETIF_F_HW_CSUM in the driver
to cover the newly introduced if-else code path in skb_segment.

In-Reply-To: CABGOaVTY6BrzJTYEtVXwawzP7-D8sb1KASDWFk15v0QFaJVbUg@mail.gmail.com
This does not seem to be a commonly used tag. And indeed differs from
the actual In-Reply-To in the email headers. Perhaps

Link: https://lore.kernel.org/netdev/CABGOaVTY6BrzJTYEtVXwawzP7-D8sb1KASDWFk15v0QFaJVbUg@mail.gmail.com/ (local)
quoted hunk ↗ jump to hunk
Signed-off-by: Yadu Kishore <redacted>
---
 net/core/skbuff.c | 24 ++++++++++++++++--------
 1 file changed, 16 insertions(+), 8 deletions(-)
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 1365a55..82a5b53 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -3926,14 +3926,22 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
                        goto perform_csum_check;

                if (!sg) {
-                       if (!nskb->remcsum_offload)
-                               nskb->ip_summed = CHECKSUM_NONE;
-                       SKB_GSO_CB(nskb)->csum =
-                               skb_copy_and_csum_bits(head_skb, offset,
-                                                      skb_put(nskb, len),
-                                                      len, 0);
-                       SKB_GSO_CB(nskb)->csum_start =
-                               skb_headroom(nskb) + doffset;
+                       if (!csum) {
+                               if (!nskb->remcsum_offload)
+                                       nskb->ip_summed = CHECKSUM_NONE;
+                               SKB_GSO_CB(nskb)->csum =
+                                       skb_copy_and_csum_bits(head_skb, offset,
+                                                              skb_put(nskb,
+                                                                      len),
+                                                              len, 0);
+                               SKB_GSO_CB(nskb)->csum_start =
+                                       skb_headroom(nskb) + doffset;
+                       } else {
+                               nskb->ip_summed = CHECKSUM_PARTIAL;
Is this not already handled by __copy_skb_header above? If ip_summed
has to be initialized, so have csum_start and csum_offset. That call
should have initialized all three.


+                               skb_copy_bits(head_skb, offset,
+                                             skb_put(nskb, len),
+                                             len);
+                       }
                        continue;
                }

--
2.7.4
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help