Re: [PATCH ipsec v2] esp: fix page frag reference leak on skb_to_sgvec failure

[Steffen Klassert]

On Wed, Jun 03, 2026 at 01:45:07PM +0200, Alessandro Schino wrote:
> In esp_output_tail(), when esp->inplace is false, the old skb page frags
> are replaced with a new page from the xfrm page_frag cache. The source
> scatterlist (sg) is built from the old frags before the replacement, and
> esp_ssg_unref() is responsible for releasing the old page references
> after the crypto operation completes.
> 
> However, if the second skb_to_sgvec() call (which builds the destination
> scatterlist from the new page) fails, the code jumps to error_free which
> only calls kfree(tmp). The old page frag references captured in the
> source scatterlist are never released:
> 
>   1. sg[] is built from old frags via skb_to_sgvec() (no extra get_page)
>   2. nr_frags is set to 1 and frag[0] is replaced with the new page
>   3. Second skb_to_sgvec() fails -> goto error_free
> 
> Fix this by adding a bool parameter to esp_ssg_unref() that, when true,
> unconditionally unrefs the source scatterlist frags. Since req->src is
> not yet initialized by aead_request_set_crypt() at the point of the
> error, the source scatterlist is obtained directly via esp_req_sg().
> Existing callers pass false to preserve the original behavior.
> 
> The same issue exists in both esp4 and esp6 as the code is identical.
> 
> Fixes: cac2661c53f3 ("esp4: Avoid skb_cow_data whenever possible")
> Fixes: 03e2a30f6a27 ("esp6: Avoid skb_cow_data whenever possible")
> Signed-off-by: Alessandro Schino <7991aleschino@xxxxxxxxx>
> ---
> v2: split the long line in esp_ssg_unref() to keep it under 100 columns
> 
>  net/ipv4/esp4.c | 7 +++++--
>  net/ipv6/esp6.c | 7 +++++--
>  2 files changed, 10 insertions(+), 4 deletions(-)
> 
> diff --git a/net/ipv4/esp4.c b/net/ipv4/esp4.c
> index 2429c7845984..dfc81ee969ae 100644
> --- a/net/ipv4/esp4.c
> +++ b/net/ipv4/esp4.c
> @@ -113,10 +113,13 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb,
>  	/* Unref skb_frag_pages in the src scatterlist if necessary.
>  	 * Skip the first sg which comes from skb->data.
>  	 */
> -	if (already_unref || req->src != req->dst)
> -		for (sg = sg_next(already_unref ? esp_req_sg(aead, req) : req->src); sg; sg = sg_next(sg))
> +	if (already_unref || req->src != req->dst) {
> +		struct scatterlist *src = already_unref ? esp_req_sg(aead, req) : req->src;

This looks like a diff beween two versions of your patch.
We don't have already_unref in the code, so it does not
apply.

> +
> +		for (sg = sg_next(src); sg; sg = sg_next(sg))
>  			skb_page_unref(page_to_netmem(sg_page(sg)),
>  				       skb->pp_recycle);
> +	}
>  }
>  
>  #ifdef CONFIG_INET_ESPINTCP
> diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
> index 50af6ab9b8fc..296b57926abb 100644
> --- a/net/ipv6/esp6.c
> +++ b/net/ipv6/esp6.c
> @@ -130,10 +130,13 @@ static void esp_ssg_unref(struct xfrm_state *x, void *tmp, struct sk_buff *skb,
>  	/* Unref skb_frag_pages in the src scatterlist if necessary.
>  	 * Skip the first sg which comes from skb->data.
>  	 */
> -	if (already_unref || req->src != req->dst)
> -		for (sg = sg_next(already_unref ? esp_req_sg(aead, req) : req->src); sg; sg = sg_next(sg))
> +	if (already_unref || req->src != req->dst) {
> +		struct scatterlist *src = already_unref ? esp_req_sg(aead, req) : req->src;

Same here.