From: John Fastabend <john.fastabend@gmail.com>
Date: Mon, 23 Apr 2018 15:39:33 -0700
Subject: bpf: sockmap, fix double page_put on ENOMEM error in redirect path
Patch-mainline: v4.17-rc4
Git-commit: 4fcfdfb83391c74e62683469289db42a143440ac
References: bsc#1109837

In the case where the socket memory boundary is hit the redirect
path returns an ENOMEM error. However, before checking for this
condition the redirect scatterlist buffer is setup with a valid
page and length. This is never unwound so when the buffers are
released latter in the error path we do a put_page() and clear
the scatterlist fields. But, because the initial error happens
before completing the scatterlist buffer we end up with both the
original buffer and the redirect buffer pointing to the same page
resulting in duplicate put_page() calls.

To fix this simply move the initial configuration of the redirect
scatterlist buffer below the sock memory check.

Found this while running TCP_STREAM test with netperf using Cilium.

Fixes: fa246693a111 ("bpf: sockmap, BPF_F_INGRESS flag for BPF_SK_SKB_STREAM_VERDICT")
Signed-off-by: John Fastabend <john.fastabend@gmail.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Thomas Bogendoerfer <tbogendoerfer@suse.de>
---
 kernel/bpf/sockmap.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

--- a/kernel/bpf/sockmap.c
+++ b/kernel/bpf/sockmap.c
@@ -524,8 +524,6 @@ static int bpf_tcp_ingress(struct sock *
 	i = md->sg_start;
 
 	do {
-		r->sg_data[i] = md->sg_data[i];
-
 		size = (apply && apply_bytes < md->sg_data[i].length) ?
 			apply_bytes : md->sg_data[i].length;
 
@@ -536,6 +534,7 @@ static int bpf_tcp_ingress(struct sock *
 		}
 
 		sk_mem_charge(sk, size);
+		r->sg_data[i] = md->sg_data[i];
 		r->sg_data[i].length = size;
 		md->sg_data[i].length -= size;
 		md->sg_data[i].offset += size;