ah_output_done, as the callback function for the
digest function, receives the exit status in err,
and calls xfrm_output_resume, which also takes a
parameter called err.
Assuming that the digest function exited successfully,
the value of err is 0, and prevents xfrm_output_resume
from calling ah_output, which calls the digest function,
again. However, setting the value of err to ah->nexthdr
gives err a positive protocol value, resulting in
infinite calls of the digest function, preventing the
sending of the packet to the network.
It appears that, except for the first call,
xfrm_output_resume should take err directly from the
value that ah_output_done receives, like esp_output_done,
and unlike ah_input_done, which calls xfrm_input_resume,
which explicitly takes a parameter called nexthdr.
When the change was made, ICMP would work with ah transport
mode.
Signed-off-by: Yuan Kang <[email protected]>
---
net/ipv4/ah4.c | 2 --
1 files changed, 0 insertions(+), 2 deletions(-)
diff --git a/net/ipv4/ah4.c b/net/ipv4/ah4.c
index c1f4154..33ca186 100644
--- a/net/ipv4/ah4.c
+++ b/net/ipv4/ah4.c
@@ -136,8 +136,6 @@ static void ah_output_done(struct crypto_async_request *base, int err)
memcpy(top_iph+1, iph+1, top_iph->ihl*4 - sizeof(struct iphdr));
}
- err = ah->nexthdr;
-
kfree(AH_SKB_CB(skb)->tmp);
xfrm_output_resume(skb, err);
}
--
1.7.4.4