Skip to contentSkip to conversation

net: fix the features flag in sctp_gso_segment

PATCH Applied Xin Long 3 participants 5 messages Jan 15, 2021 – Jan 17, 2021

Subsystems (3)
lore
Share & tools
lore

Xin Long

author maintainer
#0 lore raw

v3 cover net: fix the features flag in sctp_gso_segment

Patch 1/2 is to improve the code in skb_segment(), and it is needed by Patch 2/2.

v1->v2:
- see Patch 1/2.
v2->v3:
- change Patch 2/2 to the right patch.

Xin Long (2):
net: move the hsize check to the else block in skb_segment
sctp: remove the NETIF_F_SG flag before calling skb_segment

 net/core/skbuff.c  | 11 ++++++-----
 net/sctp/offload.c |  2 +-
 2 files changed, 7 insertions(+), 6 deletions(-)
Signature · 1 line
2.1.0
author maintainer
#1 lore raw

PATCH v3 1/2 net: move the hsize check to the else block in skb_segment

After commit 89319d3801d1 ("net: Add frag_list support to skb_segment"), it goes to process frag_list when !hsize in skb_segment(). However, when using skb frag_list, sg normally should not be set. In this case, hsize will be set with len right before !hsize check, then it won't go to frag_list processing code.

So the right thing to do is move the hsize check to the else block, so that it won't affect the !hsize check for frag_list processing.

v1->v2:
- change to do "hsize <= 0" check instead of "!hsize", and also move "hsize < 0" into else block, to save some cycles, as Alex suggested.

---
 net/core/skbuff.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
net/core/skbuff.c +6 -5
index 6039069..e835193 100644
@@ -3894,12 +3894,8 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 		}
 
 		hsize = skb_headlen(head_skb) - offset;
-		if (hsize < 0)
-			hsize = 0;
-		if (hsize > len || !sg)
-			hsize = len;
 
-		if (!hsize && i >= nfrags && skb_headlen(list_skb) &&
+		if (hsize <= 0 && i >= nfrags && skb_headlen(list_skb) &&
 		    (skb_headlen(list_skb) == len || sg)) {
 			BUG_ON(skb_headlen(list_skb) > len);
 
@@ -3942,6 +3938,11 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 			skb_release_head_state(nskb);
 			__skb_push(nskb, doffset);
 		} else {
+			if (hsize > len || !sg)
+				hsize = len;
+			else if (hsize < 0)
+				hsize = 0;
+
 			nskb = __alloc_skb(hsize + doffset + headroom,
 					   GFP_ATOMIC, skb_alloc_rx_flag(head_skb),
 					   NUMA_NO_NODE);
Signature · 1 line
2.1.0
author maintainer
#2 lore raw

PATCH v3 2/2 sctp: remove the NETIF_F_SG flag before calling skb_segment

It makes more sense to clear NETIF_F_SG instead of set it when calling skb_segment() in sctp_gso_segment(), since SCTP GSO is using head_skb's fraglist, of which all frags are linear skbs.

This will make SCTP GSO code more understandable.

---
 net/sctp/offload.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/sctp/offload.c +1 -1
index ce281a9..eb874e3 100644
@@ -68,7 +68,7 @@ static struct sk_buff *sctp_gso_segment(struct sk_buff *skb,
 		goto out;
 	}
 
-	segs = skb_segment(skb, features | NETIF_F_HW_CSUM | NETIF_F_SG);
+	segs = skb_segment(skb, (features | NETIF_F_HW_CSUM) & ~NETIF_F_SG);
 	if (IS_ERR(segs))
 		goto out;
 
Signature · 1 line
2.1.0
Reviewed-by on v3 1/2
#3 lore raw

Re: [PATCHv3 net-next 1/2] net: move the hsize check to the else block in skb_segment

In reply to Xin Long
Quoted 7 lines · Xin Long

After commit 89319d3801d1 ("net: Add frag_list support to skb_segment"), it goes to process frag_list when !hsize in skb_segment(). However, when using skb frag_list, sg normally should not be set. In this case, hsize will be set with len right before !hsize check, then it won't go to frag_list processing code.

Quoted 8 lines

So the right thing to do is move the hsize check to the else block, so that it won't affect the !hsize check for frag_list processing.

v1->v2:
- change to do "hsize <= 0" check instead of "!hsize", and also move "hsize < 0" into else block, to save some cycles, as Alex suggested.

Looks good to me.

Quoted 37 lines
---
 net/core/skbuff.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
net/core/skbuff.c +6 -6
index 6039069..e835193 100644
@@ -3894,12 +3894,8 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
                }
 
                hsize = skb_headlen(head_skb) - offset;
-               if (hsize < 0)
-                       hsize = 0;
-               if (hsize > len || !sg)
-                       hsize = len;
 
-               if (!hsize && i >= nfrags && skb_headlen(list_skb) &&
+               if (hsize <= 0 && i >= nfrags && skb_headlen(list_skb) &&
                    (skb_headlen(list_skb) == len || sg)) {
                        BUG_ON(skb_headlen(list_skb) > len);
 
@@ -3942,6 +3938,11 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
                        skb_release_head_state(nskb);
                        __skb_push(nskb, doffset);
                } else {
+                       if (hsize > len || !sg)
+                               hsize = len;
+                       else if (hsize < 0)
+                               hsize = 0;
+
                        nskb = __alloc_skb(hsize + doffset + headroom,
                                           GFP_ATOMIC, skb_alloc_rx_flag(head_skb),
                                           NUMA_NO_NODE);
--

2.1.0

maintainer Applied
#4 lore raw

Re: [PATCHv3 net-next 1/2] net: move the hsize check to the else block in skb_segment

Quoted 15 lines · On Fri, 15 Jan 2021 17:39:02 -0800 Alexander Duyck wrote:
Quoted 14 lines · Xin Long

After commit 89319d3801d1 ("net: Add frag_list support to skb_segment"), it goes to process frag_list when !hsize in skb_segment(). However, when using skb frag_list, sg normally should not be set. In this case, hsize will be set with len right before !hsize check, then it won't go to frag_list processing code.

So the right thing to do is move the hsize check to the else block, so that it won't affect the !hsize check for frag_list processing.

v1->v2:
- change to do "hsize <= 0" check instead of "!hsize", and also move "hsize < 0" into else block, to save some cycles, as Alex suggested.

Quoted 5 lines
Quoted 1 line

Looks good to me.

Applied, thanks!