2019-03-07 01:23:51

by Myungho Jung

[permalink] [raw]
Subject: [PATCH v3] xfrm: Reset secpath in xfrm failure

In esp4_gro_receive() and esp6_gro_receive(), secpath can be allocated
without adding xfrm state to xvec. Then, sp->xvec[sp->len - 1] would
fail and result in dereferencing invalid pointer in esp4_gso_segment()
and esp6_gso_segment(). Reset secpath if xfrm function returns error.

Fixes: 7785bba299a8 ("esp: Add a software GRO codepath")
Reported-by: [email protected]
Signed-off-by: Myungho Jung <[email protected]>
---
Changes in v2:
- Add fixes tag.

Changes in v3:
- Add out_reset label to remove duplicate secpath_reset() calls.

net/ipv4/esp4_offload.c | 8 +++++---
net/ipv6/esp6_offload.c | 8 +++++---
2 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/net/ipv4/esp4_offload.c b/net/ipv4/esp4_offload.c
index 8756e0e790d2..d3170a8001b2 100644
--- a/net/ipv4/esp4_offload.c
+++ b/net/ipv4/esp4_offload.c
@@ -52,13 +52,13 @@ static struct sk_buff *esp4_gro_receive(struct list_head *head,
goto out;

if (sp->len == XFRM_MAX_DEPTH)
- goto out;
+ goto out_reset;

x = xfrm_state_lookup(dev_net(skb->dev), skb->mark,
(xfrm_address_t *)&ip_hdr(skb)->daddr,
spi, IPPROTO_ESP, AF_INET);
if (!x)
- goto out;
+ goto out_reset;

sp->xvec[sp->len++] = x;
sp->olen++;
@@ -66,7 +66,7 @@ static struct sk_buff *esp4_gro_receive(struct list_head *head,
xo = xfrm_offload(skb);
if (!xo) {
xfrm_state_put(x);
- goto out;
+ goto out_reset;
}
}

@@ -82,6 +82,8 @@ static struct sk_buff *esp4_gro_receive(struct list_head *head,
xfrm_input(skb, IPPROTO_ESP, spi, -2);

return ERR_PTR(-EINPROGRESS);
+out_reset:
+ secpath_reset(skb);
out:
skb_push(skb, offset);
NAPI_GRO_CB(skb)->same_flow = 0;
diff --git a/net/ipv6/esp6_offload.c b/net/ipv6/esp6_offload.c
index d46b4eb645c2..cb99f6fb79b7 100644
--- a/net/ipv6/esp6_offload.c
+++ b/net/ipv6/esp6_offload.c
@@ -74,13 +74,13 @@ static struct sk_buff *esp6_gro_receive(struct list_head *head,
goto out;

if (sp->len == XFRM_MAX_DEPTH)
- goto out;
+ goto out_reset;

x = xfrm_state_lookup(dev_net(skb->dev), skb->mark,
(xfrm_address_t *)&ipv6_hdr(skb)->daddr,
spi, IPPROTO_ESP, AF_INET6);
if (!x)
- goto out;
+ goto out_reset;

sp->xvec[sp->len++] = x;
sp->olen++;
@@ -88,7 +88,7 @@ static struct sk_buff *esp6_gro_receive(struct list_head *head,
xo = xfrm_offload(skb);
if (!xo) {
xfrm_state_put(x);
- goto out;
+ goto out_reset;
}
}

@@ -109,6 +109,8 @@ static struct sk_buff *esp6_gro_receive(struct list_head *head,
xfrm_input(skb, IPPROTO_ESP, spi, -2);

return ERR_PTR(-EINPROGRESS);
+out_reset:
+ secpath_reset(skb);
out:
skb_push(skb, offset);
NAPI_GRO_CB(skb)->same_flow = 0;
--
2.17.1



2019-03-11 10:10:08

by Steffen Klassert

[permalink] [raw]
Subject: Re: [PATCH v3] xfrm: Reset secpath in xfrm failure

On Thu, Mar 07, 2019 at 10:23:08AM +0900, Myungho Jung wrote:
> In esp4_gro_receive() and esp6_gro_receive(), secpath can be allocated
> without adding xfrm state to xvec. Then, sp->xvec[sp->len - 1] would
> fail and result in dereferencing invalid pointer in esp4_gso_segment()
> and esp6_gso_segment(). Reset secpath if xfrm function returns error.
>
> Fixes: 7785bba299a8 ("esp: Add a software GRO codepath")
> Reported-by: [email protected]
> Signed-off-by: Myungho Jung <[email protected]>

Applied, thanks a lot!