[PATCH] tcp: fix setting csum_start in tcp_gso_segment

Subsystems: networking [general], the rest · Maintainers: "David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, Linus Torvalds

STALE4508d

8 messages, 5 authors, 2014-06-26 · open the first message on its own page

[PATCH] tcp: fix setting csum_start in tcp_gso_segment

From: Tom Herbert <hidden>
Date: 2014-06-25 04:03:54

Dave Jones reported that a crash is occuring in

csum_partial
tcp_gso_segment
inet_gso_segment
? update_dl_migration
skb_mac_gso_segment
__skb_gso_segment
dev_hard_start_xmit
sch_direct_xmit
__dev_queue_xmit
? dev_hard_start_xmit
dev_queue_xmit
ip_finish_output
? ip_output
ip_output
ip_forward_finish
ip_forward
ip_rcv_finish
ip_rcv
__netif_receive_skb_core
? __netif_receive_skb_core
? trace_hardirqs_on
__netif_receive_skb
netif_receive_skb_internal
napi_gro_complete
? napi_gro_complete
dev_gro_receive
? dev_gro_receive
napi_gro_receive

It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
not set correctly when doing non-scatter gather. We are using
offset as opposed to doffset.

Reported-by: Dave Jones <redacted>
Signed-off-by: Tom Herbert <redacted>
---
 net/core/skbuff.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

1 more lines

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 9cd5344..c1a3303 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c

@@ -2993,7 +2993,7 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 							    skb_put(nskb, len),
 							    len, 0);
 			SKB_GSO_CB(nskb)->csum_start =
-			    skb_headroom(nskb) + offset;
+			    skb_headroom(nskb) + doffset;
 			continue;
 		}

-- 
2.0.0.526.g5318336

Re: [PATCH] tcp: fix setting csum_start in tcp_gso_segment

From: Linus Torvalds <torvalds@linux-foundation.org>
Date: 2014-06-25 04:17:28

On Tue, Jun 24, 2014 at 9:03 PM, Tom Herbert [off-list ref] wrote:

It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
not set correctly when doing non-scatter gather. We are using
offset as opposed to doffset.

Reported-by: Dave Jones <redacted>

DaveJ, I think you triggered this in five minutes on your box, and I
don't recall seeing anybody else reporting the oops (and google
doesn't find anything in the last month). So it's presumably somewhat
hw-specific. Does this fix the problem?

               Linus

Re: [PATCH] tcp: fix setting csum_start in tcp_gso_segment

From: Eric Dumazet <hidden>
Date: 2014-06-25 07:19:12

On Tue, 2014-06-24 at 21:03 -0700, Tom Herbert wrote:

quoted hunk

It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
not set correctly when doing non-scatter gather. We are using
offset as opposed to doffset.

Reported-by: Dave Jones <redacted>
Signed-off-by: Tom Herbert <redacted>
---
 net/core/skbuff.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 9cd5344..c1a3303 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c

@@ -2993,7 +2993,7 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 							    skb_put(nskb, len),
 							    len, 0);
 			SKB_GSO_CB(nskb)->csum_start =
-			    skb_headroom(nskb) + offset;
+			    skb_headroom(nskb) + doffset;
 			continue;
 		}

Yes, seems an obvious typo, but please change patch title.

This is not "tcp: fix setting csum_start in tcp_gso_segment"

Maybe "net: fix setting csum_start in skb_segment()"

Re: [PATCH] tcp: fix setting csum_start in tcp_gso_segment

From: Dave Jones <hidden>
Date: 2014-06-25 14:11:54

On Tue, Jun 24, 2014 at 09:17:25PM -0700, Linus Torvalds wrote:
 > On Tue, Jun 24, 2014 at 9:03 PM, Tom Herbert [off-list ref] wrote:
 > >
 > > It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
 > > not set correctly when doing non-scatter gather. We are using
 > > offset as opposed to doffset.
 > >
 > > Reported-by: Dave Jones [off-list ref]
 > 
 > DaveJ, I think you triggered this in five minutes on your box, and I
 > don't recall seeing anybody else reporting the oops (and google
 > doesn't find anything in the last month). So it's presumably somewhat
 > hw-specific. Does this fix the problem?

It's survived routing ~1GB of packets overnight, so I'd call this good.

thanks Tom.

	Dave

Re: [PATCH] tcp: fix setting csum_start in tcp_gso_segment

From: David Miller <davem@davemloft.net>
Date: 2014-06-25 18:53:04

From: Dave Jones <redacted>
Date: Wed, 25 Jun 2014 10:10:52 -0400

On Tue, Jun 24, 2014 at 09:17:25PM -0700, Linus Torvalds wrote:
 > On Tue, Jun 24, 2014 at 9:03 PM, Tom Herbert [off-list ref] wrote:
 > >
 > > It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
 > > not set correctly when doing non-scatter gather. We are using
 > > offset as opposed to doffset.
 > >
 > > Reported-by: Dave Jones [off-list ref]
 > 
 > DaveJ, I think you triggered this in five minutes on your box, and I
 > don't recall seeing anybody else reporting the oops (and google
 > doesn't find anything in the last month). So it's presumably somewhat
 > hw-specific. Does this fix the problem?

It's survived routing ~1GB of packets overnight, so I'd call this good.

Tom, please adjust the Subject line as suggested by Eric Dumazet and add
a Tested-by: for Dave.

Thanks!

[PATCH] net: fix setting csum_start in skb_segment()

From: Eric Dumazet <hidden>
Date: 2014-06-25 19:51:08

From: Tom Herbert <redacted>

Dave Jones reported that a crash is occurring in

csum_partial
tcp_gso_segment
inet_gso_segment
? update_dl_migration
skb_mac_gso_segment
__skb_gso_segment
dev_hard_start_xmit
sch_direct_xmit
__dev_queue_xmit
? dev_hard_start_xmit
dev_queue_xmit
ip_finish_output
? ip_output
ip_output
ip_forward_finish
ip_forward
ip_rcv_finish
ip_rcv
__netif_receive_skb_core
? __netif_receive_skb_core
? trace_hardirqs_on
__netif_receive_skb
netif_receive_skb_internal
napi_gro_complete
? napi_gro_complete
dev_gro_receive
? dev_gro_receive
napi_gro_receive

It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
not set correctly when doing non-scatter gather. We are using
offset as opposed to doffset.

Reported-by: Dave Jones <redacted>
Tested-by: Dave Jones <redacted>
Signed-off-by: Tom Herbert <redacted>

6 more lines

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 9cd5344fad73..c1a33033cbe2 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c

@@ -2993,7 +2993,7 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
 							    skb_put(nskb, len),
 							    len, 0);
 			SKB_GSO_CB(nskb)->csum_start =
-			    skb_headroom(nskb) + offset;
+			    skb_headroom(nskb) + doffset;
 			continue;
 		}

Re: [PATCH] net: fix setting csum_start in skb_segment()

From: Tom Herbert <hidden>
Date: 2014-06-26 02:15:55

On Wed, Jun 25, 2014 at 12:51 PM, Eric Dumazet [off-list ref] wrote:

From: Tom Herbert <redacted>

Dave Jones reported that a crash is occurring in

csum_partial
tcp_gso_segment
inet_gso_segment
? update_dl_migration
skb_mac_gso_segment
__skb_gso_segment
dev_hard_start_xmit
sch_direct_xmit
__dev_queue_xmit
? dev_hard_start_xmit
dev_queue_xmit
ip_finish_output
? ip_output
ip_output
ip_forward_finish
ip_forward
ip_rcv_finish
ip_rcv
__netif_receive_skb_core
? __netif_receive_skb_core
? trace_hardirqs_on
__netif_receive_skb
netif_receive_skb_internal
napi_gro_complete
? napi_gro_complete
dev_gro_receive
? dev_gro_receive
napi_gro_receive

It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
not set correctly when doing non-scatter gather. We are using
offset as opposed to doffset.

Acked-by: Tom Herbert <redacted>

quoted hunk

Reported-by: Dave Jones <redacted>
Tested-by: Dave Jones <redacted>
Signed-off-by: Tom Herbert <redacted>
Signed-off-by: Eric Dumazet <redacted>
Fixes: 7e2b10c1e52ca ("net: Support for multiple checksums with gso")
---
 net/core/skbuff.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 9cd5344fad73..c1a33033cbe2 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c

@@ -2993,7 +2993,7 @@ struct sk_buff *skb_segment(struct sk_buff *head_skb,
                                                            skb_put(nskb, len),
                                                            len, 0);
                        SKB_GSO_CB(nskb)->csum_start =
-                           skb_headroom(nskb) + offset;
+                           skb_headroom(nskb) + doffset;
                        continue;
                }

Re: [PATCH] net: fix setting csum_start in skb_segment()

From: David Miller <davem@davemloft.net>
Date: 2014-06-26 03:46:15

From: Eric Dumazet <redacted>
Date: Wed, 25 Jun 2014 12:51:01 -0700

From: Tom Herbert <redacted>

Dave Jones reported that a crash is occurring in

...

It looks like a likely culprit is that SKB_GSO_CB()->csum_start is
not set correctly when doing non-scatter gather. We are using
offset as opposed to doffset.

Reported-by: Dave Jones <redacted>
Tested-by: Dave Jones <redacted>
Signed-off-by: Tom Herbert <redacted>
Signed-off-by: Eric Dumazet <redacted>
Fixes: 7e2b10c1e52ca ("net: Support for multiple checksums with gso")

Applied, thanks everyone.

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help