tls: rx: strp: don't use GFP_KERNEL in softirq context

commit 74836ec828fe17b63f2006fdbf53311d691396bf upstream.

When receive buffer is small, or the TCP rx queue looks too
complicated to bother using it directly - we allocate a new
skb and copy data into it.

We already use sk->sk_allocation... but nothing actually
sets it to GFP_ATOMIC on the ->sk_data_ready() path.

Users of HW offload are far more likely to experience problems
due to scheduling while atomic. "Copy mode" is very rarely
triggered with SW crypto.

Fixes: 84c61fe1a75b ("tls: rx: do not use the standard strparser")
Tested-by: Shai Amiram <samiram@nvidia.com>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

diff --git a/net/tls/tls_sw.c b/net/tls/tls_sw.c
index 2e5e785..96b4545 100644 (file)
--- a/net/tls/tls_sw.c
+++ b/net/tls/tls_sw.c
@@ -2289,8 +2289,12 @@ static void tls_data_ready(struct sock *sk)
        struct tls_context *tls_ctx = tls_get_ctx(sk);
        struct tls_sw_context_rx *ctx = tls_sw_ctx_rx(tls_ctx);
        struct sk_psock *psock;
+       gfp_t alloc_save;
 
+       alloc_save = sk->sk_allocation;
+       sk->sk_allocation = GFP_ATOMIC;
        tls_strp_data_ready(&ctx->strp);
+       sk->sk_allocation = alloc_save;
 
        psock = sk_psock_get(sk);
        if (psock) {