netdev - [net-next PATCH v6 5/5] virtio_net: xdp, add slowpath case for non contiguous buffers

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20161215201436.23639.57133.stgit@john-Precision-Tower-5810>
Date:   Thu, 15 Dec 2016 12:14:36 -0800
From:   John Fastabend <john.fastabend@...il.com>
To:     mst@...hat.com
Cc:     daniel@...earbox.net, netdev@...r.kernel.org,
        alexei.starovoitov@...il.com, john.r.fastabend@...el.com,
        brouer@...hat.com, tgraf@...g.ch, davem@...emloft.net
Subject: [net-next PATCH v6 5/5] virtio_net: xdp,
 add slowpath case for non contiguous buffers

virtio_net XDP support expects receive buffers to be contiguous.
If this is not the case we enable a slowpath to allow connectivity
to continue but at a significan performance overhead associated with
linearizing data. To make it painfully aware to users that XDP is
running in a degraded mode we throw an xdp buffer error.

To linearize packets we allocate a page and copy the segments of
the data, including the header, into it. After this the page can be
handled by XDP code flow as normal.

Then depending on the return code the page is either freed or sent
to the XDP xmit path. There is no attempt to optimize this path.

This case is being handled simple as a precaution in case some
unknown backend were to generate packets in this form. To test this
I had to hack qemu and force it to generate these packets. I do not
expect this case to be generated by "real" backends.

Signed-off-by: John Fastabend <john.r.fastabend@...el.com>
---
 drivers/net/virtio_net.c |   75 +++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 74 insertions(+), 1 deletion(-)

diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index 1f8300b..ce4ae7f 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -471,6 +471,64 @@ static struct sk_buff *receive_big(struct net_device *dev,
 	return NULL;
 }
 
+/* The conditions to enable XDP should preclude the underlying device from
+ * sending packets across multiple buffers (num_buf > 1). However per spec
+ * it does not appear to be illegal to do so but rather just against convention.
+ * So in order to avoid making a system unresponsive the packets are pushed
+ * into a page and the XDP program is run. This will be extremely slow and we
+ * push a warning to the user to fix this as soon as possible. Fixing this may
+ * require resolving the underlying hardware to determine why multiple buffers
+ * are being received or simply loading the XDP program in the ingress stack
+ * after the skb is built because there is no advantage to running it here
+ * anymore.
+ */
+static struct page *xdp_linearize_page(struct receive_queue *rq,
+				       u16 num_buf,
+				       struct page *p,
+				       int offset,
+				       unsigned int *len)
+{
+	struct page *page = alloc_page(GFP_ATOMIC);
+	unsigned int page_off = 0;
+
+	if (!page)
+		return NULL;
+
+	memcpy(page_address(page) + page_off, page_address(p) + offset, *len);
+	page_off += *len;
+
+	while (--num_buf) {
+		unsigned int buflen;
+		unsigned long ctx;
+		void *buf;
+		int off;
+
+		ctx = (unsigned long)virtqueue_get_buf(rq->vq, &buflen);
+		if (unlikely(!ctx))
+			goto err_buf;
+
+		/* guard against a misconfigured or uncooperative backend that
+		 * is sending packet larger than the MTU.
+		 */
+		if ((page_off + buflen) > PAGE_SIZE)
+			goto err_buf;
+
+		buf = mergeable_ctx_to_buf_address(ctx);
+		p = virt_to_head_page(buf);
+		off = buf - page_address(p);
+
+		memcpy(page_address(page) + page_off,
+		       page_address(p) + off, buflen);
+		page_off += buflen;
+	}
+
+	*len = page_off;
+	return page;
+err_buf:
+	__free_pages(page, 0);
+	return NULL;
+}
+
 static struct sk_buff *receive_mergeable(struct net_device *dev,
 					 struct virtnet_info *vi,
 					 struct receive_queue *rq,
@@ -491,6 +549,7 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
 	rcu_read_lock();
 	xdp_prog = rcu_dereference(rq->xdp_prog);
 	if (xdp_prog) {
+		struct page *xdp_page;
 		u32 act;
 
 		/* No known backend devices should send packets with
@@ -500,7 +559,15 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
 		 */
 		if (unlikely(num_buf > 1)) {
 			bpf_warn_invalid_xdp_buffer();
-			goto err_xdp;
+
+			/* linearize data for XDP */
+			xdp_page = xdp_linearize_page(rq, num_buf,
+						      page, offset, &len);
+			if (!xdp_page)
+				goto err_xdp;
+			offset = 0;
+		} else {
+			xdp_page = page;
 		}
 
 		/* Transient failure which in theory could occur if
@@ -514,12 +581,18 @@ static struct sk_buff *receive_mergeable(struct net_device *dev,
 		act = do_xdp_prog(vi, rq, xdp_prog, page, offset, len);
 		switch (act) {
 		case XDP_PASS:
+			if (unlikely(xdp_page != page))
+				__free_pages(xdp_page, 0);
 			break;
 		case XDP_TX:
+			if (unlikely(xdp_page != page))
+				goto err_xdp;
 			rcu_read_unlock();
 			goto xdp_xmit;
 		case XDP_DROP:
 		default:
+			if (unlikely(xdp_page != page))
+				__free_pages(xdp_page, 0);
 			goto err_xdp;
 		}
 	}