[ewg] [PATCH] IPoIB-UD 4K MTU patch for RC3 against 2.6.24

Shirley Ma mashirle at us.ibm.com
Mon Feb 4 00:24:08 PST 2008


Hello Vlad,

Here is the IPoIB-4K MTU patch for OFED-1.3-RC3 release agains 2.6.24
kernel. I create an attachment as well since my email has some problem.
Regarding the backport, one line is needed to add for priv-stats vs.
dev-stats. I don't have the backport patch, if you could help me that
would be nice. If this is any issue, I will ask Nam to help out.

I have touch tested mthca for 2K MTU for the updated patch. More test
are going on.

thanks
Shirley

Signed-off-by Shirley Ma <xma at us.ibm.com>
---

 drivers/infiniband/ulp/ipoib/ipoib.h           |   28 +++-
 drivers/infiniband/ulp/ipoib/ipoib_ib.c        |  218
+++++++++++++++++-------
 drivers/infiniband/ulp/ipoib/ipoib_main.c      |   19 ++-
 drivers/infiniband/ulp/ipoib/ipoib_multicast.c |    3 +-
 drivers/infiniband/ulp/ipoib/ipoib_verbs.c     |   16 ++-
 5 files changed, 212 insertions(+), 72 deletions(-)

diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h
b/drivers/infiniband/ulp/ipoib/ipoib.h
index 8eb6aa2..cb3aeab 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib.h
+++ b/drivers/infiniband/ulp/ipoib/ipoib.h
@@ -56,11 +56,11 @@
 /* constants */
 
 enum {
-	IPOIB_PACKET_SIZE         = 2048,
-	IPOIB_BUF_SIZE 		  = IPOIB_PACKET_SIZE + IB_GRH_BYTES,
-
 	IPOIB_ENCAP_LEN 	  = 4,
 
+ 	IPOIB_UD_HEAD_SIZE	  = IB_GRH_BYTES + IPOIB_ENCAP_LEN,
+ 	IPOIB_UD_RX_SG		  = 2, /* for 4K MTU */ 
+
 	IPOIB_CM_MTU              = 0x10000 - 0x10, /* padding to align header
to 16 */
 	IPOIB_CM_BUF_SIZE         = IPOIB_CM_MTU  + IPOIB_ENCAP_LEN,
 	IPOIB_CM_HEAD_SIZE 	  = IPOIB_CM_BUF_SIZE % PAGE_SIZE,
@@ -135,9 +135,9 @@ struct ipoib_mcast {
 	struct net_device *dev;
 };
 
-struct ipoib_rx_buf {
+struct ipoib_sg_rx_buf {
 	struct sk_buff *skb;
-	u64		mapping;
+	u64		mapping[IPOIB_UD_RX_SG];
 };
 
 struct ipoib_tx_buf {
@@ -286,7 +286,7 @@ struct ipoib_dev_priv {
 	unsigned int admin_mtu;
 	unsigned int mcast_mtu;
 
-	struct ipoib_rx_buf *rx_ring;
+	struct ipoib_sg_rx_buf *rx_ring;
 
 	spinlock_t           tx_lock;
 	struct ipoib_tx_buf *tx_ring;
@@ -315,6 +315,9 @@ struct ipoib_dev_priv {
 	struct dentry *mcg_dentry;
 	struct dentry *path_dentry;
 #endif
+	int max_ib_mtu;
+	struct ib_sge rx_sge[IPOIB_UD_RX_SG];
+	struct ib_recv_wr rx_wr;
 };
 
 struct ipoib_ah {
@@ -355,6 +358,19 @@ struct ipoib_neigh {
 	struct list_head    list;
 };
 
+#define IPOIB_UD_MTU(ib_mtu)		(ib_mtu - IPOIB_ENCAP_LEN)
+#define IPOIB_UD_BUF_SIZE(ib_mtu)	(ib_mtu + IB_GRH_BYTES)
+static inline int ipoib_ud_need_sg(int ib_mtu)
+{
+	return (IPOIB_UD_BUF_SIZE(ib_mtu) > PAGE_SIZE) ? 1 : 0;
+}
+static inline void ipoib_sg_dma_unmap_rx(struct ipoib_dev_priv *priv,
+					 u64 mapping[IPOIB_UD_RX_SG])
+{
+	ib_dma_unmap_single(priv->ca, mapping[0], IPOIB_UD_HEAD_SIZE,
DMA_FROM_DEVICE);
+	ib_dma_unmap_single(priv->ca, mapping[1], PAGE_SIZE, DMA_FROM_DEVICE);
+}
+
 /*
  * We stash a pointer to our private neighbour information after our
  * hardware address in neigh->ha.  The ALIGN() expression here makes
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 5063dd5..6c9eefe 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -87,32 +87,93 @@ void ipoib_free_ah(struct kref *kref)
 	spin_unlock_irqrestore(&priv->lock, flags);
 }
 
+/* Adjust length of skb with fragments to match received data */
+static void ipoib_ud_skb_put_frags(struct sk_buff *skb, unsigned int
length,
+				   struct sk_buff *toskb)
+{
+	unsigned int size;
+	skb_frag_t *frag = &skb_shinfo(skb)->frags[0];
+
+	/* put header into skb */
+	size = min(length, (unsigned)IPOIB_UD_HEAD_SIZE);
+	skb->tail += size;
+	skb->len += size;
+	length -= size;
+
+	if (length == 0) {
+		/* don't need this page */
+		skb_fill_page_desc(toskb, 0, frag->page, 0, PAGE_SIZE);
+		--skb_shinfo(skb)->nr_frags;
+	} else {
+		size = min(length, (unsigned) PAGE_SIZE);
+		frag->size = size;
+		skb->data_len += size;
+		skb->truesize += size;
+		skb->len += size;
+		length -= size;
+	}
+}
+
+static struct sk_buff *ipoib_sg_alloc_rx_skb(struct net_device *dev,
+					     int id, u64 mapping[IPOIB_UD_RX_SG])
+{
+	struct ipoib_dev_priv *priv = netdev_priv(dev);
+	struct page *page;
+	struct sk_buff *skb;
+
+	skb = dev_alloc_skb(IPOIB_UD_HEAD_SIZE);
+
+	if (unlikely(!skb)) 
+		return NULL;
+
+	mapping[0] = ib_dma_map_single(priv->ca, skb->data,
IPOIB_UD_HEAD_SIZE,
+				       DMA_FROM_DEVICE);
+	if (unlikely(ib_dma_mapping_error(priv->ca, mapping[0]))) {
+		dev_kfree_skb_any(skb);
+		return NULL;
+	}
+
+	page = alloc_page(GFP_ATOMIC);
+	if (!page)
+		goto partial_error;
+
+	skb_fill_page_desc(skb, 0, page, 0, PAGE_SIZE);
+	mapping[1] = ib_dma_map_page(priv->ca, skb_shinfo(skb)->frags[0].page,
+				     0, PAGE_SIZE, DMA_FROM_DEVICE);
+	if (unlikely(ib_dma_mapping_error(priv->ca, mapping[1])))
+		goto partial_error;
+
+	priv->rx_ring[id].skb = skb;
+	return skb;
+
+partial_error:
+	ib_dma_unmap_single(priv->ca, mapping[0], IPOIB_UD_HEAD_SIZE,
DMA_FROM_DEVICE);
+	dev_kfree_skb_any(skb);
+	return NULL;
+}
+
 static int ipoib_ib_post_receive(struct net_device *dev, int id)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
-	struct ib_sge list;
-	struct ib_recv_wr param;
 	struct ib_recv_wr *bad_wr;
 	int ret;
 
-	list.addr     = priv->rx_ring[id].mapping;
-	list.length   = IPOIB_BUF_SIZE;
-	list.lkey     = priv->mr->lkey;
+	priv->rx_wr.wr_id = id | IPOIB_OP_RECV;
+	priv->rx_sge[0].addr = priv->rx_ring[id].mapping[0];
+	priv->rx_sge[1].addr = priv->rx_ring[id].mapping[1];	
 
-	param.next    = NULL;
-	param.wr_id   = id | IPOIB_OP_RECV;
-	param.sg_list = &list;
-	param.num_sge = 1;
-
-	ret = ib_post_recv(priv->qp, &param, &bad_wr);
+	ret = ib_post_recv(priv->qp, &priv->rx_wr, &bad_wr);
 	if (unlikely(ret)) {
+		if (ipoib_ud_need_sg(priv->max_ib_mtu))
+			ipoib_sg_dma_unmap_rx(priv, priv->rx_ring[id].mapping);
+		else 
+			ib_dma_unmap_single(priv->ca, priv->rx_ring[id].mapping[0],
+					    IPOIB_UD_BUF_SIZE(priv->max_ib_mtu),
+					    DMA_FROM_DEVICE);
 		ipoib_warn(priv, "receive failed for buf %d (%d)\n", id, ret);
-		ib_dma_unmap_single(priv->ca, priv->rx_ring[id].mapping,
-				    IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
 		dev_kfree_skb_any(priv->rx_ring[id].skb);
 		priv->rx_ring[id].skb = NULL;
 	}
-
 	return ret;
 }
 
@@ -122,7 +183,7 @@ static int ipoib_alloc_rx_skb(struct net_device
*dev, int id)
 	struct sk_buff *skb;
 	u64 addr;
 
-	skb = dev_alloc_skb(IPOIB_BUF_SIZE + 4);
+	skb = dev_alloc_skb(IPOIB_UD_BUF_SIZE(priv->max_ib_mtu) + 4);
 	if (!skb)
 		return -ENOMEM;
 
@@ -133,7 +194,8 @@ static int ipoib_alloc_rx_skb(struct net_device
*dev, int id)
 	 */
 	skb_reserve(skb, 4);
 
-	addr = ib_dma_map_single(priv->ca, skb->data, IPOIB_BUF_SIZE,
+	addr = ib_dma_map_single(priv->ca, skb->data,
+				 IPOIB_UD_BUF_SIZE(priv->max_ib_mtu),
 				 DMA_FROM_DEVICE);
 	if (unlikely(ib_dma_mapping_error(priv->ca, addr))) {
 		dev_kfree_skb_any(skb);
@@ -141,7 +203,7 @@ static int ipoib_alloc_rx_skb(struct net_device
*dev, int id)
 	}
 
 	priv->rx_ring[id].skb     = skb;
-	priv->rx_ring[id].mapping = addr;
+	priv->rx_ring[id].mapping[0] = addr;
 
 	return 0;
 }
@@ -149,10 +211,15 @@ static int ipoib_alloc_rx_skb(struct net_device
*dev, int id)
 static int ipoib_ib_post_receives(struct net_device *dev)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
-	int i;
+	int i, ret;
 
 	for (i = 0; i < ipoib_recvq_size; ++i) {
-		if (ipoib_alloc_rx_skb(dev, i)) {
+		if (ipoib_ud_need_sg(priv->max_ib_mtu)) 
+			ret = !(ipoib_sg_alloc_rx_skb(dev, i,
+						      priv->rx_ring[i].mapping));
+		else
+			ret = ipoib_alloc_rx_skb(dev, i);
+		if (ret) {
 			ipoib_warn(priv, "failed to allocate receive buffer %d\n", i);
 			return -ENOMEM;
 		}
@@ -170,7 +237,7 @@ static void ipoib_ib_handle_rx_wc(struct net_device
*dev, struct ib_wc *wc)
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 	unsigned int wr_id = wc->wr_id & ~IPOIB_OP_RECV;
 	struct sk_buff *skb;
-	u64 addr;
+	u64 mapping[IPOIB_UD_RX_SG];
 
 	ipoib_dbg_data(priv, "recv completion: id %d, status: %d\n",
 		       wr_id, wc->status);
@@ -182,42 +249,73 @@ static void ipoib_ib_handle_rx_wc(struct
net_device *dev, struct ib_wc *wc)
 	}
 
 	skb  = priv->rx_ring[wr_id].skb;
-	addr = priv->rx_ring[wr_id].mapping;
+	/* duplicate the code to save fast path condition check times */
+	if (ipoib_ud_need_sg(priv->max_ib_mtu)) {
+		struct sk_buff *newskb;
+		if (unlikely(wc->status != IB_WC_SUCCESS)) {
+			if (wc->status != IB_WC_WR_FLUSH_ERR)
+				ipoib_warn(priv, "failed recv event "
+					   "(status=%d, wrid=%d vend_err %x)\n",
+					   wc->status, wr_id, wc->vendor_err);
+			ipoib_sg_dma_unmap_rx(priv, priv->rx_ring[wr_id].mapping);
+			dev_kfree_skb_any(skb);
+			priv->rx_ring[wr_id].skb = NULL;
+			return;
+		}
+		/*
+		 * Drop packets that this interface sent, ie multicast packets
+		 * that the HCA has replicated.
+		 */
+		if (wc->slid == priv->local_lid && wc->src_qp == priv->qp->qp_num)
+			goto repost;
+		newskb = ipoib_sg_alloc_rx_skb(dev, wr_id, mapping);
+		if (unlikely(!newskb)) {
+			++dev->stats.rx_dropped;
+			goto repost;
+		}
+		ipoib_dbg_data(priv, "received %d bytes, SLID 0x%04x\n",
+			       wc->byte_len, wc->slid);
+		ipoib_sg_dma_unmap_rx(priv, priv->rx_ring[wr_id].mapping);
+		memcpy(priv->rx_ring[wr_id].mapping, mapping,
+		       IPOIB_UD_RX_SG * sizeof *mapping);
+		ipoib_ud_skb_put_frags(skb, wc->byte_len, newskb);
+	} else {
+		u64 addr = priv->rx_ring[wr_id].mapping[0];
+		if (unlikely(wc->status != IB_WC_SUCCESS)) {
+			if (wc->status != IB_WC_WR_FLUSH_ERR)
+				ipoib_warn(priv, "failed recv event "
+					   "(status=%d, wrid=%d vend_err %x)\n",
+					   wc->status, wr_id, wc->vendor_err);
+			ib_dma_unmap_single(priv->ca, addr,
+					    IPOIB_UD_BUF_SIZE(priv->max_ib_mtu),
+					    DMA_FROM_DEVICE);
+			dev_kfree_skb_any(skb);
+			priv->rx_ring[wr_id].skb = NULL;
+			return;
+		}
+		/*
+		 * Drop packets that this interface sent, ie multicast packets
+		 * that the HCA has replicated.
+		 */
+		if (wc->slid == priv->local_lid && wc->src_qp == priv->qp->qp_num)
+			goto repost;
+
+		/*
+		 * If we can't allocate a new RX buffer, dump
+		 * this packet and reuse the old buffer.
+		 */
+		if (unlikely(ipoib_alloc_rx_skb(dev, wr_id))) {
+			++dev->stats.rx_dropped;
+			goto repost;
+		}
 
-	if (unlikely(wc->status != IB_WC_SUCCESS)) {
-		if (wc->status != IB_WC_WR_FLUSH_ERR)
-			ipoib_warn(priv, "failed recv event "
-				   "(status=%d, wrid=%d vend_err %x)\n",
-				   wc->status, wr_id, wc->vendor_err);
+		ipoib_dbg_data(priv, "received %d bytes, SLID 0x%04x\n",
+			       wc->byte_len, wc->slid);
 		ib_dma_unmap_single(priv->ca, addr,
-				    IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
-		dev_kfree_skb_any(skb);
-		priv->rx_ring[wr_id].skb = NULL;
-		return;
+				    IPOIB_UD_BUF_SIZE(priv->max_ib_mtu),
+				    DMA_FROM_DEVICE);
+		skb_put(skb, wc->byte_len);
 	}
-
-	/*
-	 * Drop packets that this interface sent, ie multicast packets
-	 * that the HCA has replicated.
-	 */
-	if (wc->slid == priv->local_lid && wc->src_qp == priv->qp->qp_num)
-		goto repost;
-
-	/*
-	 * If we can't allocate a new RX buffer, dump
-	 * this packet and reuse the old buffer.
-	 */
-	if (unlikely(ipoib_alloc_rx_skb(dev, wr_id))) {
-		++dev->stats.rx_dropped;
-		goto repost;
-	}
-
-	ipoib_dbg_data(priv, "received %d bytes, SLID 0x%04x\n",
-		       wc->byte_len, wc->slid);
-
-	ib_dma_unmap_single(priv->ca, addr, IPOIB_BUF_SIZE, DMA_FROM_DEVICE);
-
-	skb_put(skb, wc->byte_len);
 	skb_pull(skb, IB_GRH_BYTES);
 
 	skb->protocol = ((struct ipoib_header *) skb->data)->proto;
@@ -625,15 +723,19 @@ int ipoib_ib_dev_stop(struct net_device *dev, int
flush)
 			}
 
 			for (i = 0; i < ipoib_recvq_size; ++i) {
-				struct ipoib_rx_buf *rx_req;
+				struct ipoib_sg_rx_buf *rx_req;
 
 				rx_req = &priv->rx_ring[i];
 				if (!rx_req->skb)
 					continue;
-				ib_dma_unmap_single(priv->ca,
-						    rx_req->mapping,
-						    IPOIB_BUF_SIZE,
-						    DMA_FROM_DEVICE);
+				if (ipoib_ud_need_sg(priv->max_ib_mtu))
+					ipoib_sg_dma_unmap_rx(priv,
+							      priv->rx_ring[i].mapping);
+				else
+					ib_dma_unmap_single(priv->ca,
+							    rx_req->mapping[0],
+							    IPOIB_UD_BUF_SIZE(priv->max_ib_mtu),
+							    DMA_FROM_DEVICE);
 				dev_kfree_skb_any(rx_req->skb);
 				rx_req->skb = NULL;
 			}
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c
b/drivers/infiniband/ulp/ipoib/ipoib_main.c
index 46b67f8..9f18a9e 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_main.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c
@@ -193,7 +193,7 @@ static int ipoib_change_mtu(struct net_device *dev,
int new_mtu)
 		return 0;
 	}
 
-	if (new_mtu > IPOIB_PACKET_SIZE - IPOIB_ENCAP_LEN)
+	if (new_mtu > IPOIB_UD_MTU(priv->max_ib_mtu))
 		return -EINVAL;
 
 	priv->admin_mtu = new_mtu;
@@ -977,10 +977,6 @@ static void ipoib_setup(struct net_device *dev)
 	dev->tx_queue_len 	 = ipoib_sendq_size * 2;
 	dev->features            = NETIF_F_VLAN_CHALLENGED | NETIF_F_LLTX;
 
-	/* MTU will be reset when mcast join happens */
-	dev->mtu 		 = IPOIB_PACKET_SIZE - IPOIB_ENCAP_LEN;
-	priv->mcast_mtu 	 = priv->admin_mtu = dev->mtu;
-
 	memcpy(dev->broadcast, ipv4_bcast_addr, INFINIBAND_ALEN);
 
 	netif_carrier_off(dev);
@@ -1112,6 +1108,7 @@ static struct net_device *ipoib_add_port(const
char *format,
 					 struct ib_device *hca, u8 port)
 {
 	struct ipoib_dev_priv *priv;
+	struct ib_port_attr attr;
 	int result = -ENOMEM;
 
 	priv = ipoib_intf_alloc(format);
@@ -1120,6 +1117,18 @@ static struct net_device *ipoib_add_port(const
char *format,
 
 	SET_NETDEV_DEV(priv->dev, hca->dma_device);
 
+	if (!ib_query_port(hca, port, &attr))
+		priv->max_ib_mtu = ib_mtu_enum_to_int(attr.max_mtu);
+	else {
+		printk(KERN_WARNING "%s: ib_query_port %d failed\n",
+		       hca->name, port);
+		goto device_init_failed;
+	}
+
+	/* MTU will be reset when mcast join happens */
+	priv->dev->mtu  = IPOIB_UD_MTU(priv->max_ib_mtu);
+	priv->mcast_mtu  = priv->admin_mtu = priv->dev->mtu;
+
 	result = ib_query_pkey(hca, port, 0, &priv->pkey);
 	if (result) {
 		printk(KERN_WARNING "%s: ib_query_pkey port %d failed (ret = %d)\n",
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 9bcfc7a..19b1895 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -567,8 +567,7 @@ void ipoib_mcast_join_task(struct work_struct *work)
 		return;
 	}
 
-	priv->mcast_mtu = ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu) -
-		IPOIB_ENCAP_LEN;
+	priv->mcast_mtu =
IPOIB_UD_MTU(ib_mtu_enum_to_int(priv->broadcast->mcmember.mtu));
 
 	if (!ipoib_cm_admin_enabled(dev))
 		dev->mtu = min(priv->mcast_mtu, priv->admin_mtu);
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
index 3c6e45d..5654ebc 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c
@@ -150,7 +150,7 @@ int ipoib_transport_dev_init(struct net_device *dev,
struct ib_device *ca)
 			.max_send_wr  = ipoib_sendq_size,
 			.max_recv_wr  = ipoib_recvq_size,
 			.max_send_sge = 1,
-			.max_recv_sge = 1
+			.max_recv_sge = IPOIB_UD_RX_SG 
 		},
 		.sq_sig_type = IB_SIGNAL_ALL_WR,
 		.qp_type     = IB_QPT_UD
@@ -204,6 +204,20 @@ int ipoib_transport_dev_init(struct net_device
*dev, struct ib_device *ca)
 	priv->tx_wr.num_sge 	= 1;
 	priv->tx_wr.send_flags 	= IB_SEND_SIGNALED;
 
+	priv->rx_sge[0].lkey = priv->mr->lkey;
+	priv->rx_sge[1].lkey = priv->mr->lkey;
+	if (ipoib_ud_need_sg(priv->max_ib_mtu)) {
+		priv->rx_sge[0].length = IPOIB_UD_HEAD_SIZE;
+		priv->rx_sge[1].length = PAGE_SIZE;
+		priv->rx_wr.num_sge = IPOIB_UD_RX_SG;
+	} else {
+		priv->rx_sge[0].length = IPOIB_UD_BUF_SIZE(priv->max_ib_mtu);
+		priv->rx_sge[1].length = 0;
+		priv->rx_wr.num_sge = 1;
+	}
+	priv->rx_wr.next = NULL;
+	priv->rx_wr.sg_list = priv->rx_sge;
+
 	return 0;
 
 out_free_cq:

-------------- next part --------------
A non-text attachment was scrubbed...
Name: ipoib-4kmtu-2.6.24-rc3.patch
Type: text/x-patch
Size: 15400 bytes
Desc: not available
URL: <http://lists.openfabrics.org/pipermail/ewg/attachments/20080204/3b7916a3/attachment.bin>


More information about the ewg mailing list