bnx2x: Create bnx2x_cmn.* files

Newly created files have no functionality changes,
but includes some functionality from bnx2x_main.c which
is common for PF and coming in the future VF driver.

Signed-off-by: Dmitry Kravkov <dmitry@broadcom.com>
Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

5 files changed, 2943 insertions, 2600 deletions

diff --git a/drivers/net/bnx2x/Makefile b/drivers/net/bnx2x/Makefile
index 46c853b6cc5..ef4eebb3866 100644
--- a/drivers/net/bnx2x/Makefile
+++ b/drivers/net/bnx2x/Makefile
@@ -4,4 +4,4 @@
 
 obj-$(CONFIG_BNX2X) += bnx2x.o
 
-bnx2x-objs := bnx2x_main.o bnx2x_link.o
+bnx2x-objs := bnx2x_main.o bnx2x_link.o bnx2x_cmn.o
diff --git a/drivers/net/bnx2x/bnx2x.h b/drivers/net/bnx2x/bnx2x.h
index 4afd29201a5..260507032d3 100644
--- a/drivers/net/bnx2x/bnx2x.h
+++ b/drivers/net/bnx2x/bnx2x.h
@@ -45,6 +45,7 @@
 #endif
 
 #include <linux/mdio.h>
+#include <linux/pci.h>
 #include "bnx2x_reg.h"
 #include "bnx2x_fw_defs.h"
 #include "bnx2x_hsi.h"
diff --git a/drivers/net/bnx2x/bnx2x_cmn.c b/drivers/net/bnx2x/bnx2x_cmn.c
new file mode 100644
index 00000000000..30d20c7fee0
--- /dev/null
+++ b/drivers/net/bnx2x/bnx2x_cmn.c
@@ -0,0 +1,2251 @@
+/* bnx2x_cmn.c: Broadcom Everest network driver.
+ *
+ * Copyright (c) 2007-2010 Broadcom Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation.
+ *
+ * Maintained by: Eilon Greenstein <eilong@broadcom.com>
+ * Written by: Eliezer Tamir
+ * Based on code from Michael Chan's bnx2 driver
+ * UDP CSUM errata workaround by Arik Gendelman
+ * Slowpath and fastpath rework by Vladislav Zolotarov
+ * Statistics and Link management by Yitchak Gertner
+ *
+ */
+
+
+#include <linux/etherdevice.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include "bnx2x_cmn.h"
+
+#ifdef BCM_VLAN
+#include <linux/if_vlan.h>
+#endif
+
+static int bnx2x_poll(struct napi_struct *napi, int budget);
+
+/* free skb in the packet ring at pos idx
+ * return idx of last bd freed
+ */
+static u16 bnx2x_free_tx_pkt(struct bnx2x *bp, struct bnx2x_fastpath *fp,
+			     u16 idx)
+{
+	struct sw_tx_bd *tx_buf = &fp->tx_buf_ring[idx];
+	struct eth_tx_start_bd *tx_start_bd;
+	struct eth_tx_bd *tx_data_bd;
+	struct sk_buff *skb = tx_buf->skb;
+	u16 bd_idx = TX_BD(tx_buf->first_bd), new_cons;
+	int nbd;
+
+	/* prefetch skb end pointer to speedup dev_kfree_skb() */
+	prefetch(&skb->end);
+
+	DP(BNX2X_MSG_OFF, "pkt_idx %d  buff @(%p)->skb %p\n",
+	   idx, tx_buf, skb);
+
+	/* unmap first bd */
+	DP(BNX2X_MSG_OFF, "free bd_idx %d\n", bd_idx);
+	tx_start_bd = &fp->tx_desc_ring[bd_idx].start_bd;
+	dma_unmap_single(&bp->pdev->dev, BD_UNMAP_ADDR(tx_start_bd),
+			 BD_UNMAP_LEN(tx_start_bd), PCI_DMA_TODEVICE);
+
+	nbd = le16_to_cpu(tx_start_bd->nbd) - 1;
+#ifdef BNX2X_STOP_ON_ERROR
+	if ((nbd - 1) > (MAX_SKB_FRAGS + 2)) {
+		BNX2X_ERR("BAD nbd!\n");
+		bnx2x_panic();
+	}
+#endif
+	new_cons = nbd + tx_buf->first_bd;
+
+	/* Get the next bd */
+	bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+
+	/* Skip a parse bd... */
+	--nbd;
+	bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+
+	/* ...and the TSO split header bd since they have no mapping */
+	if (tx_buf->flags & BNX2X_TSO_SPLIT_BD) {
+		--nbd;
+		bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+	}
+
+	/* now free frags */
+	while (nbd > 0) {
+
+		DP(BNX2X_MSG_OFF, "free frag bd_idx %d\n", bd_idx);
+		tx_data_bd = &fp->tx_desc_ring[bd_idx].reg_bd;
+		dma_unmap_page(&bp->pdev->dev, BD_UNMAP_ADDR(tx_data_bd),
+			       BD_UNMAP_LEN(tx_data_bd), DMA_TO_DEVICE);
+		if (--nbd)
+			bd_idx = TX_BD(NEXT_TX_IDX(bd_idx));
+	}
+
+	/* release skb */
+	WARN_ON(!skb);
+	dev_kfree_skb(skb);
+	tx_buf->first_bd = 0;
+	tx_buf->skb = NULL;
+
+	return new_cons;
+}
+
+int bnx2x_tx_int(struct bnx2x_fastpath *fp)
+{
+	struct bnx2x *bp = fp->bp;
+	struct netdev_queue *txq;
+	u16 hw_cons, sw_cons, bd_cons = fp->tx_bd_cons;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return -1;
+#endif
+
+	txq = netdev_get_tx_queue(bp->dev, fp->index);
+	hw_cons = le16_to_cpu(*fp->tx_cons_sb);
+	sw_cons = fp->tx_pkt_cons;
+
+	while (sw_cons != hw_cons) {
+		u16 pkt_cons;
+
+		pkt_cons = TX_BD(sw_cons);
+
+		/* prefetch(bp->tx_buf_ring[pkt_cons].skb); */
+
+		DP(NETIF_MSG_TX_DONE, "hw_cons %u  sw_cons %u  pkt_cons %u\n",
+		   hw_cons, sw_cons, pkt_cons);
+
+/*		if (NEXT_TX_IDX(sw_cons) != hw_cons) {
+			rmb();
+			prefetch(fp->tx_buf_ring[NEXT_TX_IDX(sw_cons)].skb);
+		}
+*/
+		bd_cons = bnx2x_free_tx_pkt(bp, fp, pkt_cons);
+		sw_cons++;
+	}
+
+	fp->tx_pkt_cons = sw_cons;
+	fp->tx_bd_cons = bd_cons;
+
+	/* Need to make the tx_bd_cons update visible to start_xmit()
+	 * before checking for netif_tx_queue_stopped().  Without the
+	 * memory barrier, there is a small possibility that
+	 * start_xmit() will miss it and cause the queue to be stopped
+	 * forever.
+	 */
+	smp_mb();
+
+	/* TBD need a thresh? */
+	if (unlikely(netif_tx_queue_stopped(txq))) {
+		/* Taking tx_lock() is needed to prevent reenabling the queue
+		 * while it's empty. This could have happen if rx_action() gets
+		 * suspended in bnx2x_tx_int() after the condition before
+		 * netif_tx_wake_queue(), while tx_action (bnx2x_start_xmit()):
+		 *
+		 * stops the queue->sees fresh tx_bd_cons->releases the queue->
+		 * sends some packets consuming the whole queue again->
+		 * stops the queue
+		 */
+
+		__netif_tx_lock(txq, smp_processor_id());
+
+		if ((netif_tx_queue_stopped(txq)) &&
+		    (bp->state == BNX2X_STATE_OPEN) &&
+		    (bnx2x_tx_avail(fp) >= MAX_SKB_FRAGS + 3))
+			netif_tx_wake_queue(txq);
+
+		__netif_tx_unlock(txq);
+	}
+	return 0;
+}
+
+static inline void bnx2x_update_last_max_sge(struct bnx2x_fastpath *fp,
+					     u16 idx)
+{
+	u16 last_max = fp->last_max_sge;
+
+	if (SUB_S16(idx, last_max) > 0)
+		fp->last_max_sge = idx;
+}
+
+static void bnx2x_update_sge_prod(struct bnx2x_fastpath *fp,
+				  struct eth_fast_path_rx_cqe *fp_cqe)
+{
+	struct bnx2x *bp = fp->bp;
+	u16 sge_len = SGE_PAGE_ALIGN(le16_to_cpu(fp_cqe->pkt_len) -
+				     le16_to_cpu(fp_cqe->len_on_bd)) >>
+		      SGE_PAGE_SHIFT;
+	u16 last_max, last_elem, first_elem;
+	u16 delta = 0;
+	u16 i;
+
+	if (!sge_len)
+		return;
+
+	/* First mark all used pages */
+	for (i = 0; i < sge_len; i++)
+		SGE_MASK_CLEAR_BIT(fp, RX_SGE(le16_to_cpu(fp_cqe->sgl[i])));
+
+	DP(NETIF_MSG_RX_STATUS, "fp_cqe->sgl[%d] = %d\n",
+	   sge_len - 1, le16_to_cpu(fp_cqe->sgl[sge_len - 1]));
+
+	/* Here we assume that the last SGE index is the biggest */
+	prefetch((void *)(fp->sge_mask));
+	bnx2x_update_last_max_sge(fp, le16_to_cpu(fp_cqe->sgl[sge_len - 1]));
+
+	last_max = RX_SGE(fp->last_max_sge);
+	last_elem = last_max >> RX_SGE_MASK_ELEM_SHIFT;
+	first_elem = RX_SGE(fp->rx_sge_prod) >> RX_SGE_MASK_ELEM_SHIFT;
+
+	/* If ring is not full */
+	if (last_elem + 1 != first_elem)
+		last_elem++;
+
+	/* Now update the prod */
+	for (i = first_elem; i != last_elem; i = NEXT_SGE_MASK_ELEM(i)) {
+		if (likely(fp->sge_mask[i]))
+			break;
+
+		fp->sge_mask[i] = RX_SGE_MASK_ELEM_ONE_MASK;
+		delta += RX_SGE_MASK_ELEM_SZ;
+	}
+
+	if (delta > 0) {
+		fp->rx_sge_prod += delta;
+		/* clear page-end entries */
+		bnx2x_clear_sge_mask_next_elems(fp);
+	}
+
+	DP(NETIF_MSG_RX_STATUS,
+	   "fp->last_max_sge = %d  fp->rx_sge_prod = %d\n",
+	   fp->last_max_sge, fp->rx_sge_prod);
+}
+
+static void bnx2x_tpa_start(struct bnx2x_fastpath *fp, u16 queue,
+			    struct sk_buff *skb, u16 cons, u16 prod)
+{
+	struct bnx2x *bp = fp->bp;
+	struct sw_rx_bd *cons_rx_buf = &fp->rx_buf_ring[cons];
+	struct sw_rx_bd *prod_rx_buf = &fp->rx_buf_ring[prod];
+	struct eth_rx_bd *prod_bd = &fp->rx_desc_ring[prod];
+	dma_addr_t mapping;
+
+	/* move empty skb from pool to prod and map it */
+	prod_rx_buf->skb = fp->tpa_pool[queue].skb;
+	mapping = dma_map_single(&bp->pdev->dev, fp->tpa_pool[queue].skb->data,
+				 bp->rx_buf_size, DMA_FROM_DEVICE);
+	dma_unmap_addr_set(prod_rx_buf, mapping, mapping);
+
+	/* move partial skb from cons to pool (don't unmap yet) */
+	fp->tpa_pool[queue] = *cons_rx_buf;
+
+	/* mark bin state as start - print error if current state != stop */
+	if (fp->tpa_state[queue] != BNX2X_TPA_STOP)
+		BNX2X_ERR("start of bin not in stop [%d]\n", queue);
+
+	fp->tpa_state[queue] = BNX2X_TPA_START;
+
+	/* point prod_bd to new skb */
+	prod_bd->addr_hi = cpu_to_le32(U64_HI(mapping));
+	prod_bd->addr_lo = cpu_to_le32(U64_LO(mapping));
+
+#ifdef BNX2X_STOP_ON_ERROR
+	fp->tpa_queue_used |= (1 << queue);
+#ifdef _ASM_GENERIC_INT_L64_H
+	DP(NETIF_MSG_RX_STATUS, "fp->tpa_queue_used = 0x%lx\n",
+#else
+	DP(NETIF_MSG_RX_STATUS, "fp->tpa_queue_used = 0x%llx\n",
+#endif
+	   fp->tpa_queue_used);
+#endif
+}
+
+static int bnx2x_fill_frag_skb(struct bnx2x *bp, struct bnx2x_fastpath *fp,
+			       struct sk_buff *skb,
+			       struct eth_fast_path_rx_cqe *fp_cqe,
+			       u16 cqe_idx)
+{
+	struct sw_rx_page *rx_pg, old_rx_pg;
+	u16 len_on_bd = le16_to_cpu(fp_cqe->len_on_bd);
+	u32 i, frag_len, frag_size, pages;
+	int err;
+	int j;
+
+	frag_size = le16_to_cpu(fp_cqe->pkt_len) - len_on_bd;
+	pages = SGE_PAGE_ALIGN(frag_size) >> SGE_PAGE_SHIFT;
+
+	/* This is needed in order to enable forwarding support */
+	if (frag_size)
+		skb_shinfo(skb)->gso_size = min((u32)SGE_PAGE_SIZE,
+					       max(frag_size, (u32)len_on_bd));
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (pages > min_t(u32, 8, MAX_SKB_FRAGS)*SGE_PAGE_SIZE*PAGES_PER_SGE) {
+		BNX2X_ERR("SGL length is too long: %d. CQE index is %d\n",
+			  pages, cqe_idx);
+		BNX2X_ERR("fp_cqe->pkt_len = %d  fp_cqe->len_on_bd = %d\n",
+			  fp_cqe->pkt_len, len_on_bd);
+		bnx2x_panic();
+		return -EINVAL;
+	}
+#endif
+
+	/* Run through the SGL and compose the fragmented skb */
+	for (i = 0, j = 0; i < pages; i += PAGES_PER_SGE, j++) {
+		u16 sge_idx = RX_SGE(le16_to_cpu(fp_cqe->sgl[j]));
+
+		/* FW gives the indices of the SGE as if the ring is an array
+		   (meaning that "next" element will consume 2 indices) */
+		frag_len = min(frag_size, (u32)(SGE_PAGE_SIZE*PAGES_PER_SGE));
+		rx_pg = &fp->rx_page_ring[sge_idx];
+		old_rx_pg = *rx_pg;
+
+		/* If we fail to allocate a substitute page, we simply stop
+		   where we are and drop the whole packet */
+		err = bnx2x_alloc_rx_sge(bp, fp, sge_idx);
+		if (unlikely(err)) {
+			fp->eth_q_stats.rx_skb_alloc_failed++;
+			return err;
+		}
+
+		/* Unmap the page as we r going to pass it to the stack */
+		dma_unmap_page(&bp->pdev->dev,
+			       dma_unmap_addr(&old_rx_pg, mapping),
+			       SGE_PAGE_SIZE*PAGES_PER_SGE, DMA_FROM_DEVICE);
+
+		/* Add one frag and update the appropriate fields in the skb */
+		skb_fill_page_desc(skb, j, old_rx_pg.page, 0, frag_len);
+
+		skb->data_len += frag_len;
+		skb->truesize += frag_len;
+		skb->len += frag_len;
+
+		frag_size -= frag_len;
+	}
+
+	return 0;
+}
+
+static void bnx2x_tpa_stop(struct bnx2x *bp, struct bnx2x_fastpath *fp,
+			   u16 queue, int pad, int len, union eth_rx_cqe *cqe,
+			   u16 cqe_idx)
+{
+	struct sw_rx_bd *rx_buf = &fp->tpa_pool[queue];
+	struct sk_buff *skb = rx_buf->skb;
+	/* alloc new skb */
+	struct sk_buff *new_skb = netdev_alloc_skb(bp->dev, bp->rx_buf_size);
+
+	/* Unmap skb in the pool anyway, as we are going to change
+	   pool entry status to BNX2X_TPA_STOP even if new skb allocation
+	   fails. */
+	dma_unmap_single(&bp->pdev->dev, dma_unmap_addr(rx_buf, mapping),
+			 bp->rx_buf_size, DMA_FROM_DEVICE);
+
+	if (likely(new_skb)) {
+		/* fix ip xsum and give it to the stack */
+		/* (no need to map the new skb) */
+#ifdef BCM_VLAN
+		int is_vlan_cqe =
+			(le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) &
+			 PARSING_FLAGS_VLAN);
+		int is_not_hwaccel_vlan_cqe =
+			(is_vlan_cqe && (!(bp->flags & HW_VLAN_RX_FLAG)));
+#endif
+
+		prefetch(skb);
+		prefetch(((char *)(skb)) + 128);
+
+#ifdef BNX2X_STOP_ON_ERROR
+		if (pad + len > bp->rx_buf_size) {
+			BNX2X_ERR("skb_put is about to fail...  "
+				  "pad %d  len %d  rx_buf_size %d\n",
+				  pad, len, bp->rx_buf_size);
+			bnx2x_panic();
+			return;
+		}
+#endif
+
+		skb_reserve(skb, pad);
+		skb_put(skb, len);
+
+		skb->protocol = eth_type_trans(skb, bp->dev);
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+		{
+			struct iphdr *iph;
+
+			iph = (struct iphdr *)skb->data;
+#ifdef BCM_VLAN
+			/* If there is no Rx VLAN offloading -
+			   take VLAN tag into an account */
+			if (unlikely(is_not_hwaccel_vlan_cqe))
+				iph = (struct iphdr *)((u8 *)iph + VLAN_HLEN);
+#endif
+			iph->check = 0;
+			iph->check = ip_fast_csum((u8 *)iph, iph->ihl);
+		}
+
+		if (!bnx2x_fill_frag_skb(bp, fp, skb,
+					 &cqe->fast_path_cqe, cqe_idx)) {
+#ifdef BCM_VLAN
+			if ((bp->vlgrp != NULL) && is_vlan_cqe &&
+			    (!is_not_hwaccel_vlan_cqe))
+				vlan_gro_receive(&fp->napi, bp->vlgrp,
+						 le16_to_cpu(cqe->fast_path_cqe.
+							     vlan_tag), skb);
+			else
+#endif
+				napi_gro_receive(&fp->napi, skb);
+		} else {
+			DP(NETIF_MSG_RX_STATUS, "Failed to allocate new pages"
+			   " - dropping packet!\n");
+			dev_kfree_skb(skb);
+		}
+
+
+		/* put new skb in bin */
+		fp->tpa_pool[queue].skb = new_skb;
+
+	} else {
+		/* else drop the packet and keep the buffer in the bin */
+		DP(NETIF_MSG_RX_STATUS,
+		   "Failed to allocate new skb - dropping packet!\n");
+		fp->eth_q_stats.rx_skb_alloc_failed++;
+	}
+
+	fp->tpa_state[queue] = BNX2X_TPA_STOP;
+}
+
+/* Set Toeplitz hash value in the skb using the value from the
+ * CQE (calculated by HW).
+ */
+static inline void bnx2x_set_skb_rxhash(struct bnx2x *bp, union eth_rx_cqe *cqe,
+					struct sk_buff *skb)
+{
+	/* Set Toeplitz hash from CQE */
+	if ((bp->dev->features & NETIF_F_RXHASH) &&
+	    (cqe->fast_path_cqe.status_flags &
+	     ETH_FAST_PATH_RX_CQE_RSS_HASH_FLG))
+		skb->rxhash =
+		le32_to_cpu(cqe->fast_path_cqe.rss_hash_result);
+}
+
+int bnx2x_rx_int(struct bnx2x_fastpath *fp, int budget)
+{
+	struct bnx2x *bp = fp->bp;
+	u16 bd_cons, bd_prod, bd_prod_fw, comp_ring_cons;
+	u16 hw_comp_cons, sw_comp_cons, sw_comp_prod;
+	int rx_pkt = 0;
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return 0;
+#endif
+
+	/* CQ "next element" is of the size of the regular element,
+	   that's why it's ok here */
+	hw_comp_cons = le16_to_cpu(*fp->rx_cons_sb);
+	if ((hw_comp_cons & MAX_RCQ_DESC_CNT) == MAX_RCQ_DESC_CNT)
+		hw_comp_cons++;
+
+	bd_cons = fp->rx_bd_cons;
+	bd_prod = fp->rx_bd_prod;
+	bd_prod_fw = bd_prod;
+	sw_comp_cons = fp->rx_comp_cons;
+	sw_comp_prod = fp->rx_comp_prod;
+
+	/* Memory barrier necessary as speculative reads of the rx
+	 * buffer can be ahead of the index in the status block
+	 */
+	rmb();
+
+	DP(NETIF_MSG_RX_STATUS,
+	   "queue[%d]:  hw_comp_cons %u  sw_comp_cons %u\n",
+	   fp->index, hw_comp_cons, sw_comp_cons);
+
+	while (sw_comp_cons != hw_comp_cons) {
+		struct sw_rx_bd *rx_buf = NULL;
+		struct sk_buff *skb;
+		union eth_rx_cqe *cqe;
+		u8 cqe_fp_flags;
+		u16 len, pad;
+
+		comp_ring_cons = RCQ_BD(sw_comp_cons);
+		bd_prod = RX_BD(bd_prod);
+		bd_cons = RX_BD(bd_cons);
+
+		/* Prefetch the page containing the BD descriptor
+		   at producer's index. It will be needed when new skb is
+		   allocated */
+		prefetch((void *)(PAGE_ALIGN((unsigned long)
+					     (&fp->rx_desc_ring[bd_prod])) -
+				  PAGE_SIZE + 1));
+
+		cqe = &fp->rx_comp_ring[comp_ring_cons];
+		cqe_fp_flags = cqe->fast_path_cqe.type_error_flags;
+
+		DP(NETIF_MSG_RX_STATUS, "CQE type %x  err %x  status %x"
+		   "  queue %x  vlan %x  len %u\n", CQE_TYPE(cqe_fp_flags),
+		   cqe_fp_flags, cqe->fast_path_cqe.status_flags,
+		   le32_to_cpu(cqe->fast_path_cqe.rss_hash_result),
+		   le16_to_cpu(cqe->fast_path_cqe.vlan_tag),
+		   le16_to_cpu(cqe->fast_path_cqe.pkt_len));
+
+		/* is this a slowpath msg? */
+		if (unlikely(CQE_TYPE(cqe_fp_flags))) {
+			bnx2x_sp_event(fp, cqe);
+			goto next_cqe;
+
+		/* this is an rx packet */
+		} else {
+			rx_buf = &fp->rx_buf_ring[bd_cons];
+			skb = rx_buf->skb;
+			prefetch(skb);
+			len = le16_to_cpu(cqe->fast_path_cqe.pkt_len);
+			pad = cqe->fast_path_cqe.placement_offset;
+
+			/* If CQE is marked both TPA_START and TPA_END
+			   it is a non-TPA CQE */
+			if ((!fp->disable_tpa) &&
+			    (TPA_TYPE(cqe_fp_flags) !=
+					(TPA_TYPE_START | TPA_TYPE_END))) {
+				u16 queue = cqe->fast_path_cqe.queue_index;
+
+				if (TPA_TYPE(cqe_fp_flags) == TPA_TYPE_START) {
+					DP(NETIF_MSG_RX_STATUS,
+					   "calling tpa_start on queue %d\n",
+					   queue);
+
+					bnx2x_tpa_start(fp, queue, skb,
+							bd_cons, bd_prod);
+
+					/* Set Toeplitz hash for an LRO skb */
+					bnx2x_set_skb_rxhash(bp, cqe, skb);
+
+					goto next_rx;
+				}
+
+				if (TPA_TYPE(cqe_fp_flags) == TPA_TYPE_END) {
+					DP(NETIF_MSG_RX_STATUS,
+					   "calling tpa_stop on queue %d\n",
+					   queue);
+
+					if (!BNX2X_RX_SUM_FIX(cqe))
+						BNX2X_ERR("STOP on none TCP "
+							  "data\n");
+
+					/* This is a size of the linear data
+					   on this skb */
+					len = le16_to_cpu(cqe->fast_path_cqe.
+								len_on_bd);
+					bnx2x_tpa_stop(bp, fp, queue, pad,
+						    len, cqe, comp_ring_cons);
+#ifdef BNX2X_STOP_ON_ERROR
+					if (bp->panic)
+						return 0;
+#endif
+
+					bnx2x_update_sge_prod(fp,
+							&cqe->fast_path_cqe);
+					goto next_cqe;
+				}
+			}
+
+			dma_sync_single_for_device(&bp->pdev->dev,
+					dma_unmap_addr(rx_buf, mapping),
+						   pad + RX_COPY_THRESH,
+						   DMA_FROM_DEVICE);
+			prefetch(((char *)(skb)) + 128);
+
+			/* is this an error packet? */
+			if (unlikely(cqe_fp_flags & ETH_RX_ERROR_FALGS)) {
+				DP(NETIF_MSG_RX_ERR,
+				   "ERROR  flags %x  rx packet %u\n",
+				   cqe_fp_flags, sw_comp_cons);
+				fp->eth_q_stats.rx_err_discard_pkt++;
+				goto reuse_rx;
+			}
+
+			/* Since we don't have a jumbo ring
+			 * copy small packets if mtu > 1500
+			 */
+			if ((bp->dev->mtu > ETH_MAX_PACKET_SIZE) &&
+			    (len <= RX_COPY_THRESH)) {
+				struct sk_buff *new_skb;
+
+				new_skb = netdev_alloc_skb(bp->dev,
+							   len + pad);
+				if (new_skb == NULL) {
+					DP(NETIF_MSG_RX_ERR,
+					   "ERROR  packet dropped "
+					   "because of alloc failure\n");
+					fp->eth_q_stats.rx_skb_alloc_failed++;
+					goto reuse_rx;
+				}
+
+				/* aligned copy */
+				skb_copy_from_linear_data_offset(skb, pad,
+						    new_skb->data + pad, len);
+				skb_reserve(new_skb, pad);
+				skb_put(new_skb, len);
+
+				bnx2x_reuse_rx_skb(fp, skb, bd_cons, bd_prod);
+
+				skb = new_skb;
+
+			} else
+			if (likely(bnx2x_alloc_rx_skb(bp, fp, bd_prod) == 0)) {
+				dma_unmap_single(&bp->pdev->dev,
+					dma_unmap_addr(rx_buf, mapping),
+						 bp->rx_buf_size,
+						 DMA_FROM_DEVICE);
+				skb_reserve(skb, pad);
+				skb_put(skb, len);
+
+			} else {
+				DP(NETIF_MSG_RX_ERR,
+				   "ERROR  packet dropped because "
+				   "of alloc failure\n");
+				fp->eth_q_stats.rx_skb_alloc_failed++;
+reuse_rx:
+				bnx2x_reuse_rx_skb(fp, skb, bd_cons, bd_prod);
+				goto next_rx;
+			}
+
+			skb->protocol = eth_type_trans(skb, bp->dev);
+
+			/* Set Toeplitz hash for a none-LRO skb */
+			bnx2x_set_skb_rxhash(bp, cqe, skb);
+
+			skb->ip_summed = CHECKSUM_NONE;
+			if (bp->rx_csum) {
+				if (likely(BNX2X_RX_CSUM_OK(cqe)))
+					skb->ip_summed = CHECKSUM_UNNECESSARY;
+				else
+					fp->eth_q_stats.hw_csum_err++;
+			}
+		}
+
+		skb_record_rx_queue(skb, fp->index);
+
+#ifdef BCM_VLAN
+		if ((bp->vlgrp != NULL) && (bp->flags & HW_VLAN_RX_FLAG) &&
+		    (le16_to_cpu(cqe->fast_path_cqe.pars_flags.flags) &
+		     PARSING_FLAGS_VLAN))
+			vlan_gro_receive(&fp->napi, bp->vlgrp,
+				le16_to_cpu(cqe->fast_path_cqe.vlan_tag), skb);
+		else
+#endif
+			napi_gro_receive(&fp->napi, skb);
+
+
+next_rx:
+		rx_buf->skb = NULL;
+
+		bd_cons = NEXT_RX_IDX(bd_cons);
+		bd_prod = NEXT_RX_IDX(bd_prod);
+		bd_prod_fw = NEXT_RX_IDX(bd_prod_fw);
+		rx_pkt++;
+next_cqe:
+		sw_comp_prod = NEXT_RCQ_IDX(sw_comp_prod);
+		sw_comp_cons = NEXT_RCQ_IDX(sw_comp_cons);
+
+		if (rx_pkt == budget)
+			break;
+	} /* while */
+
+	fp->rx_bd_cons = bd_cons;
+	fp->rx_bd_prod = bd_prod_fw;
+	fp->rx_comp_cons = sw_comp_cons;
+	fp->rx_comp_prod = sw_comp_prod;
+
+	/* Update producers */
+	bnx2x_update_rx_prod(bp, fp, bd_prod_fw, sw_comp_prod,
+			     fp->rx_sge_prod);
+
+	fp->rx_pkt += rx_pkt;
+	fp->rx_calls++;
+
+	return rx_pkt;
+}
+
+static irqreturn_t bnx2x_msix_fp_int(int irq, void *fp_cookie)
+{
+	struct bnx2x_fastpath *fp = fp_cookie;
+	struct bnx2x *bp = fp->bp;
+
+	/* Return here if interrupt is disabled */
+	if (unlikely(atomic_read(&bp->intr_sem) != 0)) {
+		DP(NETIF_MSG_INTR, "called but intr_sem not 0, returning\n");
+		return IRQ_HANDLED;
+	}
+
+	DP(BNX2X_MSG_FP, "got an MSI-X interrupt on IDX:SB [%d:%d]\n",
+	   fp->index, fp->sb_id);
+	bnx2x_ack_sb(bp, fp->sb_id, USTORM_ID, 0, IGU_INT_DISABLE, 0);
+
+#ifdef BNX2X_STOP_ON_ERROR
+	if (unlikely(bp->panic))
+		return IRQ_HANDLED;
+#endif
+
+	/* Handle Rx and Tx according to MSI-X vector */
+	prefetch(fp->rx_cons_sb);
+	prefetch(fp->tx_cons_sb);
+	prefetch(&fp->status_blk->u_status_block.status_block_index);
+	prefetch(&fp->status_blk->c_status_block.status_block_index);
+	napi_schedule(&bnx2x_fp(bp, fp->index, napi));
+
+	return IRQ_HANDLED;
+}
+
+
+/* HW Lock for shared dual port PHYs */
+void bnx2x_acquire_phy_lock(struct bnx2x *bp)
+{
+	mutex_lock(&bp->port.phy_mutex);
+
+	if (bp->port.need_hw_lock)
+		bnx2x_acquire_hw_lock(bp, HW_LOCK_RESOURCE_MDIO);
+}
+
+void bnx2x_release_phy_lock(struct bnx2x *bp)
+{
+	if (bp->port.need_hw_lock)
+		bnx2x_release_hw_lock(bp, HW_LOCK_RESOURCE_MDIO);
+
+	mutex_unlock(&bp->port.phy_mutex);
+}
+
+void bnx2x_link_report(struct bnx2x *bp)
+{
+	if (bp->flags & MF_FUNC_DIS) {
+		netif_carrier_off(bp->dev);
+		netdev_err(bp->dev, "NIC Link is Down\n");
+		return;
+	}
+
+	if (bp->link_vars.link_up) {
+		u16 line_speed;
+
+		if (bp->state == BNX2X_STATE_OPEN)
+			netif_carrier_on(bp->dev);
+		netdev_info(bp->dev, "NIC Link is Up, ");
+
+		line_speed = bp->link_vars.line_speed;
+		if (IS_E1HMF(bp)) {
+			u16 vn_max_rate;
+
+			vn_max_rate =
+				((bp->mf_config & FUNC_MF_CFG_MAX_BW_MASK) >>
+				 FUNC_MF_CFG_MAX_BW_SHIFT) * 100;
+			if (vn_max_rate < line_speed)
+				line_speed = vn_max_rate;
+		}
+		pr_cont("%d Mbps ", line_speed);
+
+		if (bp->link_vars.duplex == DUPLEX_FULL)
+			pr_cont("full duplex");
+		else
+			pr_cont("half duplex");
+
+		if (bp->link_vars.flow_ctrl != BNX2X_FLOW_CTRL_NONE) {
+			if (bp->link_vars.flow_ctrl & BNX2X_FLOW_CTRL_RX) {
+				pr_cont(", receive ");
+				if (bp->link_vars.flow_ctrl &
+				    BNX2X_FLOW_CTRL_TX)
+					pr_cont("& transmit ");
+			} else {
+				pr_cont(", transmit ");
+			}
+			pr_cont("flow control ON");
+		}
+		pr_cont("\n");
+
+	} else { /* link_down */
+		netif_carrier_off(bp->dev);
+		netdev_err(bp->dev, "NIC Link is Down\n");
+	}
+}
+
+void bnx2x_init_rx_rings(struct bnx2x *bp)
+{
+	int func = BP_FUNC(bp);
+	int max_agg_queues = CHIP_IS_E1(bp) ? ETH_MAX_AGGREGATION_QUEUES_E1 :
+					      ETH_MAX_AGGREGATION_QUEUES_E1H;
+	u16 ring_prod, cqe_ring_prod;
+	int i, j;
+
+	bp->rx_buf_size = bp->dev->mtu + ETH_OVREHEAD + BNX2X_RX_ALIGN;
+	DP(NETIF_MSG_IFUP,
+	   "mtu %d  rx_buf_size %d\n", bp->dev->mtu, bp->rx_buf_size);
+
+	if (bp->flags & TPA_ENABLE_FLAG) {
+
+		for_each_queue(bp, j) {
+			struct bnx2x_fastpath *fp = &bp->fp[j];
+
+			for (i = 0; i < max_agg_queues; i++) {
+				fp->tpa_pool[i].skb =
+				   netdev_alloc_skb(bp->dev, bp->rx_buf_size);
+				if (!fp->tpa_pool[i].skb) {
+					BNX2X_ERR("Failed to allocate TPA "
+						  "skb pool for queue[%d] - "
+						  "disabling TPA on this "
+						  "queue!\n", j);
+					bnx2x_free_tpa_pool(bp, fp, i);
+					fp->disable_tpa = 1;
+					break;
+				}
+				dma_unmap_addr_set((struct sw_rx_bd *)
+							&bp->fp->tpa_pool[i],
+						   mapping, 0);
+				fp->tpa_state[i] = BNX2X_TPA_STOP;
+			}
+		}
+	}
+
+	for_each_queue(bp, j) {
+		struct bnx2x_fastpath *fp = &bp->fp[j];
+
+		fp->rx_bd_cons = 0;
+		fp->rx_cons_sb = BNX2X_RX_SB_INDEX;
+		fp->rx_bd_cons_sb = BNX2X_RX_SB_BD_INDEX;
+
+		/* "next page" elements initialization */
+		/* SGE ring */
+		for (i = 1; i <= NUM_RX_SGE_PAGES; i++) {
+			struct eth_rx_sge *sge;
+
+			sge = &fp->rx_sge_ring[RX_SGE_CNT * i - 2];
+			sge->addr_hi =
+				cpu_to_le32(U64_HI(fp->rx_sge_mapping +
+					BCM_PAGE_SIZE*(i % NUM_RX_SGE_PAGES)));
+			sge->addr_lo =
+				cpu_to_le32(U64_LO(fp->rx_sge_mapping +
+					BCM_PAGE_SIZE*(i % NUM_RX_SGE_PAGES)));
+		}
+
+		bnx2x_init_sge_ring_bit_mask(fp);
+
+		/* RX BD ring */
+		for (i = 1; i <= NUM_RX_RINGS; i++) {
+			struct eth_rx_bd *rx_bd;
+
+			rx_bd = &fp->rx_desc_ring[RX_DESC_CNT * i - 2];
+			rx_bd->addr_hi =
+				cpu_to_le32(U64_HI(fp->rx_desc_mapping +
+					    BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
+			rx_bd->addr_lo =
+				cpu_to_le32(U64_LO(fp->rx_desc_mapping +
+					    BCM_PAGE_SIZE*(i % NUM_RX_RINGS)));
+		}
+
+		/* CQ ring */
+		for (i = 1; i <= NUM_RCQ_RINGS; i++) {
+			struct eth_rx_cqe_next_page *nextpg;
+
+			nextpg = (struct eth_rx_cqe_next_page *)
+				&fp->rx_comp_ring[RCQ_DESC_CNT * i - 1];
+			nextpg->addr_hi =
+				cpu_to_le32(U64_HI(fp->rx_comp_mapping +
+					   BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+			nextpg->addr_lo =
+				cpu_to_le32(U64_LO(fp->rx_comp_mapping +
+					   BCM_PAGE_SIZE*(i % NUM_RCQ_RINGS)));
+		}
+
+		/* Allocate SGEs and initialize the ring elements */
+		for (i = 0, ring_prod = 0;
+		     i < MAX_RX_SGE_CNT*NUM_RX_SGE_PAGES; i++) {
+
+			if (bnx2x_alloc_rx_sge(bp, fp, ring_prod) < 0) {
+				BNX2X_ERR("was only able to allocate "
+					  "%d rx sges\n", i);
+				BNX2X_ERR("disabling TPA for queue[%d]\n", j);
+				/* Cleanup already allocated elements */
+				bnx2x_free_rx_sge_range(bp, fp, ring_prod);
+				bnx2x_free_tpa_pool(bp, fp, max_agg_queues);
+				fp->disable_tpa = 1;
+				ring_prod = 0;
+				break;
+			}
+			ring_prod = NEXT_SGE_IDX(ring_prod);
+		}
+		fp->rx_sge_prod = ring_prod;
+
+		/* Allocate BDs and initialize BD ring */
+		fp->rx_comp_cons = 0;
+		cqe_ring_prod = ring_prod = 0;
+		for (i = 0; i < bp->rx_ring_size; i++) {
+			if (bnx2x_alloc_rx_skb(bp, fp, ring_prod) < 0) {
+				BNX2X_ERR("was only able to allocate "
+					  "%d rx skbs on queue[%d]\n", i, j);
+				fp->eth_q_stats.rx_skb_alloc_failed++;
+				break;
+			}
+			ring_prod = NEXT_RX_IDX(ring_prod);
+			cqe_ring_prod = NEXT_RCQ_IDX(cqe_ring_prod);
+			WARN_ON(ring_prod <= i);
+		}
+
+		fp->rx_bd_prod = ring_prod;
+		/* must not have more available CQEs than BDs */
+		fp->rx_comp_prod = min_t(u16, NUM_RCQ_RINGS*RCQ_DESC_CNT,
+					 cqe_ring_prod);
+		fp->rx_pkt = fp->rx_calls = 0;
+
+		/* Warning!
+		 * this will generate an interrupt (to the TSTORM)
+		 * must only be done after chip is initialized
+		 */
+		bnx2x_update_rx_prod(bp, fp, ring_prod, fp->rx_comp_prod,
+				     fp->rx_sge_prod);
+		if (j != 0)
+			continue;
+
+		REG_WR(bp, BAR_USTRORM_INTMEM +
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(func),
+		       U64_LO(fp->rx_comp_mapping));
+		REG_WR(bp, BAR_USTRORM_INTMEM +
+		       USTORM_MEM_WORKAROUND_ADDRESS_OFFSET(func) + 4,
+		       U64_HI(fp->rx_comp_mapping));
+	}
+}
+static void bnx2x_free_tx_skbs(struct bnx2x *bp)
+{
+	int i;
+
+	for_each_queue(bp, i) {
+		struct bnx2x_fastpath *fp = &bp->fp[i];
+
+		u16 bd_cons = fp->tx_bd_cons;
+		u16 sw_prod = fp->tx_pkt_prod;
+		u16 sw_cons = fp->tx_pkt_cons;
+
+		while (sw_cons != sw_prod) {
+			bd_cons = bnx2x_free_tx_pkt(bp, fp, TX_BD(sw_cons));
+			sw_cons++;
+		}
+	}
+}
+
+static void bnx2x_free_rx_skbs(struct bnx2x *bp)
+{
+	int i, j;
+
+	for_each_queue(bp, j) {
+		struct bnx2x_fastpath *fp = &bp->fp[j];
+
+		for (i = 0; i < NUM_RX_BD; i++) {
+			struct sw_rx_bd *rx_buf = &fp->rx_buf_ring[i];
+			struct sk_buff *skb = rx_buf->skb;
+
+			if (skb == NULL)
+				continue;
+
+			dma_unmap_single(&bp->pdev->dev,
+					 dma_unmap_addr(rx_buf, mapping),
+					 bp->rx_buf_size, DMA_FROM_DEVICE);
+
+			rx_buf->skb = NULL;
+			dev_kfree_skb(skb);
+		}
+		if (!fp->disable_tpa)
+			bnx2x_free_tpa_pool(bp, fp, CHIP_IS_E1(bp) ?
+					    ETH_MAX_AGGREGATION_QUEUES_E1 :
+					    ETH_MAX_AGGREGATION_QUEUES_E1H);
+	}
+}
+
+void bnx2x_free_skbs(struct bnx2x *bp)
+{
+	bnx2x_free_tx_skbs(bp);
+	bnx2x_free_rx_skbs(bp);
+}
+
+static void bnx2x_free_msix_irqs(struct bnx2x *bp)
+{
+	int i, offset = 1;
+
+	free_irq(bp->msix_table[0].vector, bp->dev);
+	DP(NETIF_MSG_IFDOWN, "released sp irq (%d)\n",
+	   bp->msix_table[0].vector);
+
+#ifdef BCM_CNIC
+	offset++;
+#endif
+	for_each_queue(bp, i) {
+		DP(NETIF_MSG_IFDOWN, "about to release fp #%d->%d irq  "
+		   "state %x\n", i, bp->msix_table[i + offset].vector,
+		   bnx2x_fp(bp, i, state));
+
+		free_irq(bp->msix_table[i + offset].vector, &bp->fp[i]);
+	}
+}
+
+void bnx2x_free_irq(struct bnx2x *bp, bool disable_only)
+{
+	if (bp->flags & USING_MSIX_FLAG) {
+		if (!disable_only)
+			bnx2x_free_msix_irqs(bp);
+		pci_disable_msix(bp->pdev);
+		bp->flags &= ~USING_MSIX_FLAG;
+
+	} else if (bp->flags & USING_MSI_FLAG) {
+		if (!disable_only)
+			free_irq(bp->pdev->irq, bp->dev);
+		pci_disable_msi(bp->pdev);
+		bp->flags &= ~USING_MSI_FLAG;
+
+	} else if (!disable_only)
+		free_irq(bp->pdev->irq, bp->dev);
+}
+
+static int bnx2x_enable_msix(struct bnx2x *bp)
+{
+	int i, rc, offset = 1;
+	int igu_vec = 0;
+
+	bp->msix_table[0].entry = igu_vec;
+	DP(NETIF_MSG_IFUP, "msix_table[0].entry = %d (slowpath)\n", igu_vec);
+
+#ifdef BCM_CNIC
+	igu_vec = BP_L_ID(bp) + offset;
+	bp->msix_table[1].entry = igu_vec;
+	DP(NETIF_MSG_IFUP, "msix_table[1].entry = %d (CNIC)\n", igu_vec);
+	offset++;
+#endif
+	for_each_queue(bp, i) {
+		igu_vec = BP_L_ID(bp) + offset + i;
+		bp->msix_table[i + offset].entry = igu_vec;
+		DP(NETIF_MSG_IFUP, "msix_table[%d].entry = %d "
+		   "(fastpath #%u)\n", i + offset, igu_vec, i);
+	}
+
+	rc = pci_enable_msix(bp->pdev, &bp->msix_table[0],
+			     BNX2X_NUM_QUEUES(bp) + offset);
+
+	/*
+	 * reconfigure number of tx/rx queues according to available
+	 * MSI-X vectors
+	 */
+	if (rc >= BNX2X_MIN_MSIX_VEC_CNT) {
+		/* vectors available for FP */
+		int fp_vec = rc - BNX2X_MSIX_VEC_FP_START;
+
+		DP(NETIF_MSG_IFUP,
+		   "Trying to use less MSI-X vectors: %d\n", rc);
+
+		rc = pci_enable_msix(bp->pdev, &bp->msix_table[0], rc);
+
+		if (rc) {
+			DP(NETIF_MSG_IFUP,
+			   "MSI-X is not attainable  rc %d\n", rc);
+			return rc;
+		}
+
+		bp->num_queues = min(bp->num_queues, fp_vec);
+
+		DP(NETIF_MSG_IFUP, "New queue configuration set: %d\n",
+				  bp->num_queues);
+	} else if (rc) {
+		DP(NETIF_MSG_IFUP, "MSI-X is not attainable  rc %d\n", rc);
+		return rc;
+	}
+
+	bp->flags |= USING_MSIX_FLAG;
+
+	return 0;
+}
+
+static int bnx2x_req_msix_irqs(struct bnx2x *bp)
+{
+	int i, rc, offset = 1;
+
+	rc = request_irq(bp->msix_table[0].vector, bnx2x_msix_sp_int, 0,
+			 bp->dev->name, bp->dev);
+	if (rc) {
+		BNX2X_ERR("request sp irq failed\n");
+		return -EBUSY;
+	}
+
+#ifdef BCM_CNIC
+	offset++;
+#endif
+	for_each_queue(bp, i) {
+		struct bnx2x_fastpath *fp = &bp->fp[i];
+		snprintf(fp->name, sizeof(fp->name), "%s-fp-%d",
+			 bp->dev->name, i);
+
+		rc = request_irq(bp->msix_table[i + offset].vector,
+				 bnx2x_msix_fp_int, 0, fp->name, fp);
+		if (rc) {
+			BNX2X_ERR("request fp #%d irq failed  rc %d\n", i, rc);
+			bnx2x_free_msix_irqs(bp);
+			return -EBUSY;
+		}
+
+		fp->state = BNX2X_FP_STATE_IRQ;
+	}
+
+	i = BNX2X_NUM_QUEUES(bp);