[PATCH net] gro: Enter slow-path if there is no tailroom

[PATCH net] gro: Enter slow-path if there is no tailroom

[Eric Dumazet]

From: Herbert Xu <herbert@gondor.apana.org.au>

The GRO path has a fast-path where we avoid calling pskb_may_pull
and pskb_expand by directly accessing frag0.  However, this should
only be done if we have enough tailroom in the skb as otherwise
we'll have to expand it later anyway.

This patch adds the check by capping frag0_len with the skb tailroom.

Fixes: cb18978cbf45 ("gro: Open-code final pskb_may_pull")
Reported-by: Slava Shwartsman <slavash@mellanox.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/core/dev.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 8db5a0b4b52061afe9cd44dead33cc69523d15df..88d2907ca2cda3438a7a1c0c179051a5038aa1e1 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4441,7 +4441,8 @@ static void skb_gro_reset_offset(struct sk_buff *skb)
 	    pinfo->nr_frags &&
 	    !PageHighMem(skb_frag_page(frag0))) {
 		NAPI_GRO_CB(skb)->frag0 = skb_frag_address(frag0);
-		NAPI_GRO_CB(skb)->frag0_len = skb_frag_size(frag0);
+		NAPI_GRO_CB(skb)->frag0_len = min(skb_frag_size(frag0),
+						  skb->end - skb->tail);
 	}
 }
 

Re: [PATCH net] gro: Enter slow-path if there is no tailroom

[David Miller]

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 10 Jan 2017 12:24:01 -0800

> From: Herbert Xu <herbert@gondor.apana.org.au>
> 
> The GRO path has a fast-path where we avoid calling pskb_may_pull
> and pskb_expand by directly accessing frag0.  However, this should
> only be done if we have enough tailroom in the skb as otherwise
> we'll have to expand it later anyway.
> 
> This patch adds the check by capping frag0_len with the skb tailroom.
> 
> Fixes: cb18978cbf45 ("gro: Open-code final pskb_may_pull")
> Reported-by: Slava Shwartsman <slavash@mellanox.com>
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
> Signed-off-by: Eric Dumazet <edumazet@google.com>


Applied and queued up for -stable.

[PATCH net] gro: use min_t() in skb_gro_reset_offset()

[Eric Dumazet]

From: Eric Dumazet <edumazet@google.com>

On 32bit arches, (skb->end - skb->data) is not 'unsigned int',
so we shall use min_t() instead of min() to avoid a compiler error.

Fixes: 1272ce87fa01 ("gro: Enter slow-path if there is no tailroom")
Reported-by: kernel test robot <fengguang.wu@intel.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/core/dev.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/net/core/dev.c b/net/core/dev.c
index 88d2907ca2cda3438a7a1c0c179051a5038aa1e1..07b307b0b414730688b64fdb2295b0fa1b721e51 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -4441,8 +4441,9 @@ static void skb_gro_reset_offset(struct sk_buff *skb)
 	    pinfo->nr_frags &&
 	    !PageHighMem(skb_frag_page(frag0))) {
 		NAPI_GRO_CB(skb)->frag0 = skb_frag_address(frag0);
-		NAPI_GRO_CB(skb)->frag0_len = min(skb_frag_size(frag0),
-						  skb->end - skb->tail);
+		NAPI_GRO_CB(skb)->frag0_len = min_t(unsigned int,
+						    skb_frag_size(frag0),
+						    skb->end - skb->tail);
 	}
 }
 

Re: [PATCH net] gro: use min_t() in skb_gro_reset_offset()

[David Miller]

From: Eric Dumazet <eric.dumazet@gmail.com>
Date: Tue, 10 Jan 2017 19:52:43 -0800

> From: Eric Dumazet <edumazet@google.com>
> 
> On 32bit arches, (skb->end - skb->data) is not 'unsigned int',
> so we shall use min_t() instead of min() to avoid a compiler error.
> 
> Fixes: 1272ce87fa01 ("gro: Enter slow-path if there is no tailroom")
> Reported-by: kernel test robot <fengguang.wu@intel.com>
> Signed-off-by: Eric Dumazet <edumazet@google.com>

Applied, thanks Eric.