[PATCH 0/2 v4] Open vSwitch zerocopy upcall

[PATCH 0/2 v4] Open vSwitch zerocopy upcall

[Thomas Graf]

Respin of the zerocopy patches for the openvswitch upcall.

V4: - Daniel Borkmann pointed out that the style in skbuff.h has changed
      in net-next, adapted to no longer using extern in headers.
V3: - Removed unneeded alignment of nlmsg_len after padding 
V2: - Added skb_zerocopy_headlen() to calculate headroom of destination
      buffer. This also takes care of the from->head_frag issue.
    - Attribute alignment for frag_list case
    - API documentation
    - performance data for CHECKSUM_PARTIAL tx case

Thomas Graf (2):
  net: Export skb_zerocopy() to zerocopy from one skb to another
  openvswitch: Use skb_zerocopy() for upcall

 include/linux/skbuff.h               |  3 ++
 net/core/skbuff.c                    | 85 ++++++++++++++++++++++++++++++++++++
 net/netfilter/nfnetlink_queue_core.c | 59 ++-----------------------
 net/openvswitch/datapath.c           | 52 +++++++++++++++++++---
 4 files changed, 137 insertions(+), 62 deletions(-)

-- 
1.8.3.1

[PATCH 1/2 net-next] net: Export skb_zerocopy() to zerocopy from one skb to another

[Thomas Graf]

Make the skb zerocopy logic written for nfnetlink queue available for
use by other modules.

Signed-off-by: Thomas Graf <tgraf@suug.ch>
---
 include/linux/skbuff.h               |  3 ++
 net/core/skbuff.c                    | 85 ++++++++++++++++++++++++++++++++++++
 net/netfilter/nfnetlink_queue_core.c | 59 ++-----------------------
 3 files changed, 92 insertions(+), 55 deletions(-)

diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 036ec7d..9e6a293 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -2372,6 +2372,9 @@ int skb_splice_bits(struct sk_buff *skb, unsigned int offset,
 		    struct pipe_inode_info *pipe, unsigned int len,
 		    unsigned int flags);
 void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to);
+unsigned int skb_zerocopy_headlen(const struct sk_buff *from);
+void skb_zerocopy(struct sk_buff *to, const struct sk_buff *from,
+		  int len, int hlen);
 void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len);
 int skb_shift(struct sk_buff *tgt, struct sk_buff *skb, int shiftlen);
 void skb_scrub_packet(struct sk_buff *skb, bool xnet);
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index 8c5197f..c5cefbe 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -2125,6 +2125,91 @@ __wsum skb_copy_and_csum_bits(const struct sk_buff *skb, int offset,
 }
 EXPORT_SYMBOL(skb_copy_and_csum_bits);
 
+ /**
+ *	skb_zerocopy_headlen - Calculate headroom needed for skb_zerocopy()
+ *	@from: source buffer
+ *
+ *	Calculates the amount of linear headroom needed in the 'to' skb passed
+ *	into skb_zerocopy().
+ */
+unsigned int
+skb_zerocopy_headlen(const struct sk_buff *from)
+{
+	unsigned int hlen = 0;
+
+	if (!from->head_frag ||
+	    skb_headlen(from) < L1_CACHE_BYTES ||
+	    skb_shinfo(from)->nr_frags >= MAX_SKB_FRAGS)
+		hlen = skb_headlen(from);
+
+	if (skb_has_frag_list(from))
+		hlen = from->len;
+
+	return hlen;
+}
+EXPORT_SYMBOL_GPL(skb_zerocopy_headlen);
+
+/**
+ *	skb_zerocopy - Zero copy skb to skb
+ *	@to: destination buffer
+ *	@source: source buffer
+ *	@len: number of bytes to copy from source buffer
+ *	@hlen: size of linear headroom in destination buffer
+ *
+ *	Copies up to `len` bytes from `from` to `to` by creating references
+ *	to the frags in the source buffer.
+ *
+ *	The `hlen` as calculated by skb_zerocopy_headlen() specifies the
+ *	headroom in the `to` buffer.
+ */
+void
+skb_zerocopy(struct sk_buff *to, const struct sk_buff *from, int len, int hlen)
+{
+	int i, j = 0;
+	int plen = 0; /* length of skb->head fragment */
+	struct page *page;
+	unsigned int offset;
+
+	BUG_ON(!from->head_frag && !hlen);
+
+	/* dont bother with small payloads */
+	if (len <= skb_tailroom(to)) {
+		skb_copy_bits(from, 0, skb_put(to, len), len);
+		return;
+	}
+
+	if (hlen) {
+		skb_copy_bits(from, 0, skb_put(to, hlen), hlen);
+		len -= hlen;
+	} else {
+		plen = min_t(int, skb_headlen(from), len);
+		if (plen) {
+			page = virt_to_head_page(from->head);
+			offset = from->data - (unsigned char *)page_address(page);
+			__skb_fill_page_desc(to, 0, page, offset, plen);
+			get_page(page);
+			j = 1;
+			len -= plen;
+		}
+	}
+
+	to->truesize += len + plen;
+	to->len += len + plen;
+	to->data_len += len + plen;
+
+	for (i = 0; i < skb_shinfo(from)->nr_frags; i++) {
+		if (!len)
+			break;
+		skb_shinfo(to)->frags[j] = skb_shinfo(from)->frags[i];
+		skb_shinfo(to)->frags[j].size = min_t(int, skb_shinfo(to)->frags[j].size, len);
+		len -= skb_shinfo(to)->frags[j].size;
+		skb_frag_ref(to, j);
+		j++;
+	}
+	skb_shinfo(to)->nr_frags = j;
+}
+EXPORT_SYMBOL_GPL(skb_zerocopy);
+
 void skb_copy_and_csum_dev(const struct sk_buff *skb, u8 *to)
 {
 	__wsum csum;
diff --git a/net/netfilter/nfnetlink_queue_core.c b/net/netfilter/nfnetlink_queue_core.c
index 21258cf..615ee12 100644
--- a/net/netfilter/nfnetlink_queue_core.c
+++ b/net/netfilter/nfnetlink_queue_core.c
@@ -235,51 +235,6 @@ nfqnl_flush(struct nfqnl_instance *queue, nfqnl_cmpfn cmpfn, unsigned long data)
 	spin_unlock_bh(&queue->lock);
 }
 
-static void
-nfqnl_zcopy(struct sk_buff *to, const struct sk_buff *from, int len, int hlen)
-{
-	int i, j = 0;
-	int plen = 0; /* length of skb->head fragment */
-	struct page *page;
-	unsigned int offset;
-
-	/* dont bother with small payloads */
-	if (len <= skb_tailroom(to)) {
-		skb_copy_bits(from, 0, skb_put(to, len), len);
-		return;
-	}
-
-	if (hlen) {
-		skb_copy_bits(from, 0, skb_put(to, hlen), hlen);
-		len -= hlen;
-	} else {
-		plen = min_t(int, skb_headlen(from), len);
-		if (plen) {
-			page = virt_to_head_page(from->head);
-			offset = from->data - (unsigned char *)page_address(page);
-			__skb_fill_page_desc(to, 0, page, offset, plen);
-			get_page(page);
-			j = 1;
-			len -= plen;
-		}
-	}
-
-	to->truesize += len + plen;
-	to->len += len + plen;
-	to->data_len += len + plen;
-
-	for (i = 0; i < skb_shinfo(from)->nr_frags; i++) {
-		if (!len)
-			break;
-		skb_shinfo(to)->frags[j] = skb_shinfo(from)->frags[i];
-		skb_shinfo(to)->frags[j].size = min_t(int, skb_shinfo(to)->frags[j].size, len);
-		len -= skb_shinfo(to)->frags[j].size;
-		skb_frag_ref(to, j);
-		j++;
-	}
-	skb_shinfo(to)->nr_frags = j;
-}
-
 static int
 nfqnl_put_packet_info(struct sk_buff *nlskb, struct sk_buff *packet,
 		      bool csum_verify)
@@ -304,7 +259,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 {
 	size_t size;
 	size_t data_len = 0, cap_len = 0;
-	int hlen = 0;
+	unsigned int hlen = 0;
 	struct sk_buff *skb;
 	struct nlattr *nla;
 	struct nfqnl_msg_packet_hdr *pmsg;
@@ -356,14 +311,8 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 		if (data_len > entskb->len)
 			data_len = entskb->len;
 
-		if (!entskb->head_frag ||
-		    skb_headlen(entskb) < L1_CACHE_BYTES ||
-		    skb_shinfo(entskb)->nr_frags >= MAX_SKB_FRAGS)
-			hlen = skb_headlen(entskb);
-
-		if (skb_has_frag_list(entskb))
-			hlen = entskb->len;
-		hlen = min_t(int, data_len, hlen);
+		hlen = skb_zerocopy_headlen(entskb);
+		hlen = min_t(unsigned int, hlen, data_len);
 		size += sizeof(struct nlattr) + hlen;
 		cap_len = entskb->len;
 		break;
@@ -504,7 +453,7 @@ nfqnl_build_packet_message(struct net *net, struct nfqnl_instance *queue,
 		nla->nla_type = NFQA_PAYLOAD;
 		nla->nla_len = nla_attr_size(data_len);
 
-		nfqnl_zcopy(skb, entskb, data_len, hlen);
+		skb_zerocopy(skb, entskb, data_len, hlen);
 	}
 
 	nlh->nlmsg_len = skb->len;
-- 
1.8.3.1

Re: [PATCH 1/2 net-next] net: Export skb_zerocopy() to zerocopy from one skb to another

[Daniel Borkmann]

On 11/08/2013 11:47 AM, Thomas Graf wrote:
> Make the skb zerocopy logic written for nfnetlink queue available for
> use by other modules.
>
> Signed-off-by: Thomas Graf <tgraf-G/eBtMaohhA@public.gmane.org>

Reviewed-by: Daniel Borkmann <dborkman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>

[PATCH 2/2 net-next] openvswitch: Use skb_zerocopy() for upcall

[Thomas Graf]

Use of skb_zerocopy() avoids the expensive call to memcpy() when
copying the packet data into the Netlink skb. Completes checksum
through skb_checksum_help() if needed.

Netlink messaged must be properly padded and aligned to meet
sanity checks of the user space counterpart.

Cost of memcpy is significantly reduced from:
+   7.48%       vhost-8471  [k] memcpy
+   5.57%     ovs-vswitchd  [k] memcpy
+   2.81%       vhost-8471  [k] csum_partial_copy_generic

to:
+   5.72%     ovs-vswitchd  [k] memcpy
+   3.32%       vhost-5153  [k] memcpy
+   0.68%       vhost-5153  [k] skb_zerocopy

(megaflows disabled)

Signed-off-by: Thomas Graf <tgraf@suug.ch>
Cc: Jesse Gross <jesse@nicira.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
---
 net/openvswitch/datapath.c | 52 +++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 45 insertions(+), 7 deletions(-)

diff --git a/net/openvswitch/datapath.c b/net/openvswitch/datapath.c
index 1408adc..3f170e3 100644
--- a/net/openvswitch/datapath.c
+++ b/net/openvswitch/datapath.c
@@ -381,10 +381,11 @@ static size_t key_attr_size(void)
 }
 
 static size_t upcall_msg_size(const struct sk_buff *skb,
-			      const struct nlattr *userdata)
+			      const struct nlattr *userdata,
+			      unsigned int hdrlen)
 {
 	size_t size = NLMSG_ALIGN(sizeof(struct ovs_header))
-		+ nla_total_size(skb->len) /* OVS_PACKET_ATTR_PACKET */
+		+ nla_total_size(hdrlen) /* OVS_PACKET_ATTR_PACKET */
 		+ nla_total_size(key_attr_size()); /* OVS_PACKET_ATTR_KEY */
 
 	/* OVS_PACKET_ATTR_USERDATA */
@@ -402,6 +403,7 @@ static int queue_userspace_packet(struct net *net, int dp_ifindex,
 	struct sk_buff *nskb = NULL;
 	struct sk_buff *user_skb; /* to be queued to userspace */
 	struct nlattr *nla;
+	unsigned int plen, hlen;
 	int err;
 
 	if (vlan_tx_tag_present(skb)) {
@@ -422,7 +424,13 @@ static int queue_userspace_packet(struct net *net, int dp_ifindex,
 		goto out;
 	}
 
-	user_skb = genlmsg_new(upcall_msg_size(skb, upcall_info->userdata), GFP_ATOMIC);
+	/* Complete checksum if needed */
+	if (skb->ip_summed == CHECKSUM_PARTIAL &&
+	    (err = skb_checksum_help(skb)))
+		goto out;
+
+	hlen = skb_zerocopy_headlen(skb);
+	user_skb = genlmsg_new(upcall_msg_size(skb, upcall_info->userdata, hlen), GFP_ATOMIC);
 	if (!user_skb) {
 		err = -ENOMEM;
 		goto out;
@@ -441,13 +449,43 @@ static int queue_userspace_packet(struct net *net, int dp_ifindex,
 			  nla_len(upcall_info->userdata),
 			  nla_data(upcall_info->userdata));
 
-	nla = __nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, skb->len);
+	/* Only reserve room for attribute header, packet data is added
+	 * in skb_zerocopy() */
+	if (!(nla = nla_reserve(user_skb, OVS_PACKET_ATTR_PACKET, 0)))
+		goto out;
+	nla->nla_len = nla_attr_size(skb->len);
 
-	skb_copy_and_csum_dev(skb, nla_data(nla));
+	skb_zerocopy(user_skb, skb, skb->len, hlen);
 
-	genlmsg_end(user_skb, upcall);
-	err = genlmsg_unicast(net, user_skb, upcall_info->portid);
+	/* OVS user space expects the size of the message to be aligned to
+	 * NLA_ALIGNTO. Aligning nlmsg_len is not enough, the actual bytes
+	 * read must match nlmsg_len.
+	 */
+	plen = NLA_ALIGN(user_skb->len) - user_skb->len;
+	if (plen > 0) {
+		int nr_frags = skb_shinfo(user_skb)->nr_frags;
+
+		if (nr_frags) {
+			skb_frag_t *frag;
+
+			frag = &skb_shinfo(user_skb)->frags[nr_frags -1];
+			skb_frag_size_add(frag, plen);
+			BUG_ON(frag->size > PAGE_SIZE);
+
+			user_skb->truesize += plen;
+			user_skb->len += plen;
+			user_skb->data_len += plen;
+		} else {
+			/* The linear headroom is aligned to NLMSG_ALIGN by
+			 * genlmsg_new(), room must be available.
+			 */
+			memset(skb_put(user_skb, plen), 0, plen);
+		}
+	}
 
+	((struct nlmsghdr *) user_skb->data)->nlmsg_len = user_skb->len;
+
+	err = genlmsg_unicast(net, user_skb, upcall_info->portid);
 out:
 	kfree_skb(nskb);
 	return err;
-- 
1.8.3.1

Re: [PATCH 2/2 net-next] openvswitch: Use skb_zerocopy() for upcall

[Daniel Borkmann]

On 11/08/2013 11:47 AM, Thomas Graf wrote:
> Use of skb_zerocopy() avoids the expensive call to memcpy() when
> copying the packet data into the Netlink skb. Completes checksum
> through skb_checksum_help() if needed.
>
> Netlink messaged must be properly padded and aligned to meet
> sanity checks of the user space counterpart.
>
> Cost of memcpy is significantly reduced from:
> +   7.48%       vhost-8471  [k] memcpy
> +   5.57%     ovs-vswitchd  [k] memcpy
> +   2.81%       vhost-8471  [k] csum_partial_copy_generic
>
> to:
> +   5.72%     ovs-vswitchd  [k] memcpy
> +   3.32%       vhost-5153  [k] memcpy
> +   0.68%       vhost-5153  [k] skb_zerocopy
>
> (megaflows disabled)
>
> Signed-off-by: Thomas Graf <tgraf-G/eBtMaohhA@public.gmane.org>
> Cc: Jesse Gross <jesse-l0M0P4e3n4LQT0dZR+AlfA@public.gmane.org>
> Cc: Eric Dumazet <eric.dumazet-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>

Reviewed-by: Daniel Borkmann <dborkman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>

Re: [PATCH 0/2 v4] Open vSwitch zerocopy upcall

[David Miller]

From: Thomas Graf <tgraf@suug.ch>
Date: Fri,  8 Nov 2013 11:47:16 +0100

> Respin of the zerocopy patches for the openvswitch upcall.
> 
> V4: - Daniel Borkmann pointed out that the style in skbuff.h has changed
>       in net-next, adapted to no longer using extern in headers.
> V3: - Removed unneeded alignment of nlmsg_len after padding 
> V2: - Added skb_zerocopy_headlen() to calculate headroom of destination
>       buffer. This also takes care of the from->head_frag issue.
>     - Attribute alignment for frag_list case
>     - API documentation
>     - performance data for CHECKSUM_PARTIAL tx case
> 
> Thomas Graf (2):
>   net: Export skb_zerocopy() to zerocopy from one skb to another
>   openvswitch: Use skb_zerocopy() for upcall

No objections from me for these changes, but it's too late for the
current merge window.

Jesse can feel free to queue this up in his tree and you can add my:

Acked-by: David S. Miller <davem@davemloft.net>

to these patches.

Thanks!

Re: [PATCH 0/2 v4] Open vSwitch zerocopy upcall

[Jesse Gross]

On Thu, Nov 14, 2013 at 3:13 PM, David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org> wrote:
> From: Thomas Graf <tgraf-G/eBtMaohhA@public.gmane.org>
> Date: Fri,  8 Nov 2013 11:47:16 +0100
>
>> Respin of the zerocopy patches for the openvswitch upcall.
>>
>> V4: - Daniel Borkmann pointed out that the style in skbuff.h has changed
>>       in net-next, adapted to no longer using extern in headers.
>> V3: - Removed unneeded alignment of nlmsg_len after padding
>> V2: - Added skb_zerocopy_headlen() to calculate headroom of destination
>>       buffer. This also takes care of the from->head_frag issue.
>>     - Attribute alignment for frag_list case
>>     - API documentation
>>     - performance data for CHECKSUM_PARTIAL tx case
>>
>> Thomas Graf (2):
>>   net: Export skb_zerocopy() to zerocopy from one skb to another
>>   openvswitch: Use skb_zerocopy() for upcall
>
> No objections from me for these changes, but it's too late for the
> current merge window.
>
> Jesse can feel free to queue this up in his tree and you can add my:
>
> Acked-by: David S. Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
>
> to these patches.
>
> Thanks!

Thanks - there's some discussion related to these patches in the
thread "linux: Signal datapath that unaligned Netlink message can be
received" so I'm waiting until that concludes.