bgmac.c

/*
 * Driver for (BCM4706)? GBit MAC core on BCMA bus.
 *
 * Copyright (C) 2012 Rafał Miłecki <zajec5@gmail.com>
 *
 * Licensed under the GNU/GPL. See COPYING for details.
 */

#include "bgmac.h"

#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/delay.h>
#include <linux/etherdevice.h>
#include <linux/mii.h>
#include <linux/phy.h>
#include <linux/phy_fixed.h>
#include <linux/interrupt.h>
#include <linux/dma-mapping.h>
#include <bcm47xx_nvram.h>

static const struct bcma_device_id bgmac_bcma_tbl[] = {
	BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_4706_MAC_GBIT, BCMA_ANY_REV, BCMA_ANY_CLASS),
	BCMA_CORE(BCMA_MANUF_BCM, BCMA_CORE_MAC_GBIT, BCMA_ANY_REV, BCMA_ANY_CLASS),
	{},
};
MODULE_DEVICE_TABLE(bcma, bgmac_bcma_tbl);

static bool bgmac_wait_value(struct bcma_device *core, u16 reg, u32 mask,
			     u32 value, int timeout)
{
	u32 val;
	int i;

	for (i = 0; i < timeout / 10; i++) {
		val = bcma_read32(core, reg);
		if ((val & mask) == value)
			return true;
		udelay(10);
	}
	pr_err("Timeout waiting for reg 0x%X\n", reg);
	return false;
}

/**************************************************
 * DMA
 **************************************************/

static void bgmac_dma_tx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
{
	u32 val;
	int i;

	if (!ring->mmio_base)
		return;

	/* Suspend DMA TX ring first.
	 * bgmac_wait_value doesn't support waiting for any of few values, so
	 * implement whole loop here.
	 */
	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL,
		    BGMAC_DMA_TX_SUSPEND);
	for (i = 0; i < 10000 / 10; i++) {
		val = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS);
		val &= BGMAC_DMA_TX_STAT;
		if (val == BGMAC_DMA_TX_STAT_DISABLED ||
		    val == BGMAC_DMA_TX_STAT_IDLEWAIT ||
		    val == BGMAC_DMA_TX_STAT_STOPPED) {
			i = 0;
			break;
		}
		udelay(10);
	}
	if (i)
		bgmac_err(bgmac, "Timeout suspending DMA TX ring 0x%X (BGMAC_DMA_TX_STAT: 0x%08X)\n",
			  ring->mmio_base, val);

	/* Remove SUSPEND bit */
	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL, 0);
	if (!bgmac_wait_value(bgmac->core,
			      ring->mmio_base + BGMAC_DMA_TX_STATUS,
			      BGMAC_DMA_TX_STAT, BGMAC_DMA_TX_STAT_DISABLED,
			      10000)) {
		bgmac_warn(bgmac, "DMA TX ring 0x%X wasn't disabled on time, waiting additional 300us\n",
			   ring->mmio_base);
		udelay(300);
		val = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS);
		if ((val & BGMAC_DMA_TX_STAT) != BGMAC_DMA_TX_STAT_DISABLED)
			bgmac_err(bgmac, "Reset of DMA TX ring 0x%X failed\n",
				  ring->mmio_base);
	}
}

static void bgmac_dma_tx_enable(struct bgmac *bgmac,
				struct bgmac_dma_ring *ring)
{
	u32 ctl;

	ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL);
	if (bgmac->core->id.rev >= 4) {
		ctl &= ~BGMAC_DMA_TX_BL_MASK;
		ctl |= BGMAC_DMA_TX_BL_128 << BGMAC_DMA_TX_BL_SHIFT;

		ctl &= ~BGMAC_DMA_TX_MR_MASK;
		ctl |= BGMAC_DMA_TX_MR_2 << BGMAC_DMA_TX_MR_SHIFT;

		ctl &= ~BGMAC_DMA_TX_PC_MASK;
		ctl |= BGMAC_DMA_TX_PC_16 << BGMAC_DMA_TX_PC_SHIFT;

		ctl &= ~BGMAC_DMA_TX_PT_MASK;
		ctl |= BGMAC_DMA_TX_PT_8 << BGMAC_DMA_TX_PT_SHIFT;
	}
	ctl |= BGMAC_DMA_TX_ENABLE;
	ctl |= BGMAC_DMA_TX_PARITY_DISABLE;
	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_CTL, ctl);
}

static void
bgmac_dma_tx_add_buf(struct bgmac *bgmac, struct bgmac_dma_ring *ring,
		     int i, int len, u32 ctl0)
{
	struct bgmac_slot_info *slot;
	struct bgmac_dma_desc *dma_desc;
	u32 ctl1;

	if (i == ring->num_slots - 1)
		ctl0 |= BGMAC_DESC_CTL0_EOT;

	ctl1 = len & BGMAC_DESC_CTL1_LEN;

	slot = &ring->slots[i];
	dma_desc = &ring->cpu_base[i];
	dma_desc->addr_low = cpu_to_le32(lower_32_bits(slot->dma_addr));
	dma_desc->addr_high = cpu_to_le32(upper_32_bits(slot->dma_addr));
	dma_desc->ctl0 = cpu_to_le32(ctl0);
	dma_desc->ctl1 = cpu_to_le32(ctl1);
}

static netdev_tx_t bgmac_dma_tx_add(struct bgmac *bgmac,
				    struct bgmac_dma_ring *ring,
				    struct sk_buff *skb)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	struct net_device *net_dev = bgmac->net_dev;
	int index = ring->end % BGMAC_TX_RING_SLOTS;
	struct bgmac_slot_info *slot = &ring->slots[index];
	int nr_frags;
	u32 flags;
	int i;

	if (skb->len > BGMAC_DESC_CTL1_LEN) {
		bgmac_err(bgmac, "Too long skb (%d)\n", skb->len);
		goto err_drop;
	}

	if (skb->ip_summed == CHECKSUM_PARTIAL)
		skb_checksum_help(skb);

	nr_frags = skb_shinfo(skb)->nr_frags;

	/* ring->end - ring->start will return the number of valid slots,
	 * even when ring->end overflows
	 */
	if (ring->end - ring->start + nr_frags + 1 >= BGMAC_TX_RING_SLOTS) {
		bgmac_err(bgmac, "TX ring is full, queue should be stopped!\n");
		netif_stop_queue(net_dev);
		return NETDEV_TX_BUSY;
	}

	slot->dma_addr = dma_map_single(dma_dev, skb->data, skb_headlen(skb),
					DMA_TO_DEVICE);
	if (unlikely(dma_mapping_error(dma_dev, slot->dma_addr)))
		goto err_dma_head;

	flags = BGMAC_DESC_CTL0_SOF;
	if (!nr_frags)
		flags |= BGMAC_DESC_CTL0_EOF | BGMAC_DESC_CTL0_IOC;

	bgmac_dma_tx_add_buf(bgmac, ring, index, skb_headlen(skb), flags);
	flags = 0;

	for (i = 0; i < nr_frags; i++) {
		struct skb_frag_struct *frag = &skb_shinfo(skb)->frags[i];
		int len = skb_frag_size(frag);

		index = (index + 1) % BGMAC_TX_RING_SLOTS;
		slot = &ring->slots[index];
		slot->dma_addr = skb_frag_dma_map(dma_dev, frag, 0,
						  len, DMA_TO_DEVICE);
		if (unlikely(dma_mapping_error(dma_dev, slot->dma_addr)))
			goto err_dma;

		if (i == nr_frags - 1)
			flags |= BGMAC_DESC_CTL0_EOF | BGMAC_DESC_CTL0_IOC;

		bgmac_dma_tx_add_buf(bgmac, ring, index, len, flags);
	}

	slot->skb = skb;
	ring->end += nr_frags + 1;
	netdev_sent_queue(net_dev, skb->len);

	wmb();

	/* Increase ring->end to point empty slot. We tell hardware the first
	 * slot it should *not* read.
	 */
	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_INDEX,
		    ring->index_base +
		    (ring->end % BGMAC_TX_RING_SLOTS) *
		    sizeof(struct bgmac_dma_desc));

	if (ring->end - ring->start >= BGMAC_TX_RING_SLOTS - 8)
		netif_stop_queue(net_dev);

	return NETDEV_TX_OK;

err_dma:
	dma_unmap_single(dma_dev, slot->dma_addr, skb_headlen(skb),
			 DMA_TO_DEVICE);

	while (i > 0) {
		int index = (ring->end + i) % BGMAC_TX_RING_SLOTS;
		struct bgmac_slot_info *slot = &ring->slots[index];
		u32 ctl1 = le32_to_cpu(ring->cpu_base[index].ctl1);
		int len = ctl1 & BGMAC_DESC_CTL1_LEN;

		dma_unmap_page(dma_dev, slot->dma_addr, len, DMA_TO_DEVICE);
	}

err_dma_head:
	bgmac_err(bgmac, "Mapping error of skb on ring 0x%X\n",
		  ring->mmio_base);

err_drop:
	dev_kfree_skb(skb);
	return NETDEV_TX_OK;
}

/* Free transmitted packets */
static void bgmac_dma_tx_free(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	int empty_slot;
	bool freed = false;
	unsigned bytes_compl = 0, pkts_compl = 0;

	/* The last slot that hardware didn't consume yet */
	empty_slot = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_STATUS);
	empty_slot &= BGMAC_DMA_TX_STATDPTR;
	empty_slot -= ring->index_base;
	empty_slot &= BGMAC_DMA_TX_STATDPTR;
	empty_slot /= sizeof(struct bgmac_dma_desc);

	while (ring->start != ring->end) {
		int slot_idx = ring->start % BGMAC_TX_RING_SLOTS;
		struct bgmac_slot_info *slot = &ring->slots[slot_idx];
		u32 ctl1;
		int len;

		if (slot_idx == empty_slot)
			break;

		ctl1 = le32_to_cpu(ring->cpu_base[slot_idx].ctl1);
		len = ctl1 & BGMAC_DESC_CTL1_LEN;
		if (ctl1 & BGMAC_DESC_CTL0_SOF)
			/* Unmap no longer used buffer */
			dma_unmap_single(dma_dev, slot->dma_addr, len,
					 DMA_TO_DEVICE);
		else
			dma_unmap_page(dma_dev, slot->dma_addr, len,
				       DMA_TO_DEVICE);

		if (slot->skb) {
			bytes_compl += slot->skb->len;
			pkts_compl++;

			/* Free memory! :) */
			dev_kfree_skb(slot->skb);
			slot->skb = NULL;
		}

		slot->dma_addr = 0;
		ring->start++;
		freed = true;
	}

	if (!pkts_compl)
		return;

	netdev_completed_queue(bgmac->net_dev, pkts_compl, bytes_compl);

	if (netif_queue_stopped(bgmac->net_dev))
		netif_wake_queue(bgmac->net_dev);
}

static void bgmac_dma_rx_reset(struct bgmac *bgmac, struct bgmac_dma_ring *ring)
{
	if (!ring->mmio_base)
		return;

	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL, 0);
	if (!bgmac_wait_value(bgmac->core,
			      ring->mmio_base + BGMAC_DMA_RX_STATUS,
			      BGMAC_DMA_RX_STAT, BGMAC_DMA_RX_STAT_DISABLED,
			      10000))
		bgmac_err(bgmac, "Reset of ring 0x%X RX failed\n",
			  ring->mmio_base);
}

static void bgmac_dma_rx_enable(struct bgmac *bgmac,
				struct bgmac_dma_ring *ring)
{
	u32 ctl;

	ctl = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL);
	if (bgmac->core->id.rev >= 4) {
		ctl &= ~BGMAC_DMA_RX_BL_MASK;
		ctl |= BGMAC_DMA_RX_BL_128 << BGMAC_DMA_RX_BL_SHIFT;

		ctl &= ~BGMAC_DMA_RX_PC_MASK;
		ctl |= BGMAC_DMA_RX_PC_8 << BGMAC_DMA_RX_PC_SHIFT;

		ctl &= ~BGMAC_DMA_RX_PT_MASK;
		ctl |= BGMAC_DMA_RX_PT_1 << BGMAC_DMA_RX_PT_SHIFT;
	}
	ctl &= BGMAC_DMA_RX_ADDREXT_MASK;
	ctl |= BGMAC_DMA_RX_ENABLE;
	ctl |= BGMAC_DMA_RX_PARITY_DISABLE;
	ctl |= BGMAC_DMA_RX_OVERFLOW_CONT;
	ctl |= BGMAC_RX_FRAME_OFFSET << BGMAC_DMA_RX_FRAME_OFFSET_SHIFT;
	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_CTL, ctl);
}

static int bgmac_dma_rx_skb_for_slot(struct bgmac *bgmac,
				     struct bgmac_slot_info *slot)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	dma_addr_t dma_addr;
	struct bgmac_rx_header *rx;
	void *buf;

	/* Alloc skb */
	buf = netdev_alloc_frag(BGMAC_RX_ALLOC_SIZE);
	if (!buf)
		return -ENOMEM;

	/* Poison - if everything goes fine, hardware will overwrite it */
	rx = buf + BGMAC_RX_BUF_OFFSET;
	rx->len = cpu_to_le16(0xdead);
	rx->flags = cpu_to_le16(0xbeef);

	/* Map skb for the DMA */
	dma_addr = dma_map_single(dma_dev, buf + BGMAC_RX_BUF_OFFSET,
				  BGMAC_RX_BUF_SIZE, DMA_FROM_DEVICE);
	if (dma_mapping_error(dma_dev, dma_addr)) {
		bgmac_err(bgmac, "DMA mapping error\n");
		put_page(virt_to_head_page(buf));
		return -ENOMEM;
	}

	/* Update the slot */
	slot->buf = buf;
	slot->dma_addr = dma_addr;

	return 0;
}

static void bgmac_dma_rx_update_index(struct bgmac *bgmac,
				      struct bgmac_dma_ring *ring)
{
	dma_wmb();

	bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_INDEX,
		    ring->index_base +
		    ring->end * sizeof(struct bgmac_dma_desc));
}

static void bgmac_dma_rx_setup_desc(struct bgmac *bgmac,
				    struct bgmac_dma_ring *ring, int desc_idx)
{
	struct bgmac_dma_desc *dma_desc = ring->cpu_base + desc_idx;
	u32 ctl0 = 0, ctl1 = 0;

	if (desc_idx == ring->num_slots - 1)
		ctl0 |= BGMAC_DESC_CTL0_EOT;
	ctl1 |= BGMAC_RX_BUF_SIZE & BGMAC_DESC_CTL1_LEN;
	/* Is there any BGMAC device that requires extension? */
	/* ctl1 |= (addrext << B43_DMA64_DCTL1_ADDREXT_SHIFT) &
	 * B43_DMA64_DCTL1_ADDREXT_MASK;
	 */

	dma_desc->addr_low = cpu_to_le32(lower_32_bits(ring->slots[desc_idx].dma_addr));
	dma_desc->addr_high = cpu_to_le32(upper_32_bits(ring->slots[desc_idx].dma_addr));
	dma_desc->ctl0 = cpu_to_le32(ctl0);
	dma_desc->ctl1 = cpu_to_le32(ctl1);

	ring->end = desc_idx;
}

static void bgmac_dma_rx_poison_buf(struct device *dma_dev,
				    struct bgmac_slot_info *slot)
{
	struct bgmac_rx_header *rx = slot->buf + BGMAC_RX_BUF_OFFSET;

	dma_sync_single_for_cpu(dma_dev, slot->dma_addr, BGMAC_RX_BUF_SIZE,
				DMA_FROM_DEVICE);
	rx->len = cpu_to_le16(0xdead);
	rx->flags = cpu_to_le16(0xbeef);
	dma_sync_single_for_device(dma_dev, slot->dma_addr, BGMAC_RX_BUF_SIZE,
				   DMA_FROM_DEVICE);
}

static int bgmac_dma_rx_read(struct bgmac *bgmac, struct bgmac_dma_ring *ring,
			     int weight)
{
	u32 end_slot;
	int handled = 0;

	end_slot = bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_STATUS);
	end_slot &= BGMAC_DMA_RX_STATDPTR;
	end_slot -= ring->index_base;
	end_slot &= BGMAC_DMA_RX_STATDPTR;
	end_slot /= sizeof(struct bgmac_dma_desc);

	while (ring->start != end_slot) {
		struct device *dma_dev = bgmac->core->dma_dev;
		struct bgmac_slot_info *slot = &ring->slots[ring->start];
		struct bgmac_rx_header *rx = slot->buf + BGMAC_RX_BUF_OFFSET;
		struct sk_buff *skb;
		void *buf = slot->buf;
		dma_addr_t dma_addr = slot->dma_addr;
		u16 len, flags;

		do {
			/* Prepare new skb as replacement */
			if (bgmac_dma_rx_skb_for_slot(bgmac, slot)) {
				bgmac_dma_rx_poison_buf(dma_dev, slot);
				break;
			}

			/* Unmap buffer to make it accessible to the CPU */
			dma_unmap_single(dma_dev, dma_addr,
					 BGMAC_RX_BUF_SIZE, DMA_FROM_DEVICE);

			/* Get info from the header */
			len = le16_to_cpu(rx->len);
			flags = le16_to_cpu(rx->flags);

			/* Check for poison and drop or pass the packet */
			if (len == 0xdead && flags == 0xbeef) {
				bgmac_err(bgmac, "Found poisoned packet at slot %d, DMA issue!\n",
					  ring->start);
				put_page(virt_to_head_page(buf));
				break;
			}

			if (len > BGMAC_RX_ALLOC_SIZE) {
				bgmac_err(bgmac, "Found oversized packet at slot %d, DMA issue!\n",
					  ring->start);
				put_page(virt_to_head_page(buf));
				break;
			}

			/* Omit CRC. */
			len -= ETH_FCS_LEN;

			skb = build_skb(buf, BGMAC_RX_ALLOC_SIZE);
			skb_put(skb, BGMAC_RX_FRAME_OFFSET +
				BGMAC_RX_BUF_OFFSET + len);
			skb_pull(skb, BGMAC_RX_FRAME_OFFSET +
				 BGMAC_RX_BUF_OFFSET);

			skb_checksum_none_assert(skb);
			skb->protocol = eth_type_trans(skb, bgmac->net_dev);
			napi_gro_receive(&bgmac->napi, skb);
			handled++;
		} while (0);

		bgmac_dma_rx_setup_desc(bgmac, ring, ring->start);

		if (++ring->start >= BGMAC_RX_RING_SLOTS)
			ring->start = 0;

		if (handled >= weight) /* Should never be greater */
			break;
	}

	bgmac_dma_rx_update_index(bgmac, ring);

	return handled;
}

/* Does ring support unaligned addressing? */
static bool bgmac_dma_unaligned(struct bgmac *bgmac,
				struct bgmac_dma_ring *ring,
				enum bgmac_dma_ring_type ring_type)
{
	switch (ring_type) {
	case BGMAC_DMA_RING_TX:
		bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGLO,
			    0xff0);
		if (bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGLO))
			return true;
		break;
	case BGMAC_DMA_RING_RX:
		bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO,
			    0xff0);
		if (bgmac_read(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO))
			return true;
		break;
	}
	return false;
}

static void bgmac_dma_tx_ring_free(struct bgmac *bgmac,
				   struct bgmac_dma_ring *ring)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	struct bgmac_dma_desc *dma_desc = ring->cpu_base;
	struct bgmac_slot_info *slot;
	int i;

	for (i = 0; i < ring->num_slots; i++) {
		int len = dma_desc[i].ctl1 & BGMAC_DESC_CTL1_LEN;

		slot = &ring->slots[i];
		dev_kfree_skb(slot->skb);

		if (!slot->dma_addr)
			continue;

		if (slot->skb)
			dma_unmap_single(dma_dev, slot->dma_addr,
					 len, DMA_TO_DEVICE);
		else
			dma_unmap_page(dma_dev, slot->dma_addr,
				       len, DMA_TO_DEVICE);
	}
}

static void bgmac_dma_rx_ring_free(struct bgmac *bgmac,
				   struct bgmac_dma_ring *ring)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	struct bgmac_slot_info *slot;
	int i;

	for (i = 0; i < ring->num_slots; i++) {
		slot = &ring->slots[i];
		if (!slot->dma_addr)
			continue;

		dma_unmap_single(dma_dev, slot->dma_addr,
				 BGMAC_RX_BUF_SIZE,
				 DMA_FROM_DEVICE);
		put_page(virt_to_head_page(slot->buf));
		slot->dma_addr = 0;
	}
}

static void bgmac_dma_ring_desc_free(struct bgmac *bgmac,
				     struct bgmac_dma_ring *ring)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	int size;

	if (!ring->cpu_base)
	    return;

	/* Free ring of descriptors */
	size = ring->num_slots * sizeof(struct bgmac_dma_desc);
	dma_free_coherent(dma_dev, size, ring->cpu_base,
			  ring->dma_base);
}

static void bgmac_dma_cleanup(struct bgmac *bgmac)
{
	int i;

	for (i = 0; i < BGMAC_MAX_TX_RINGS; i++)
		bgmac_dma_tx_ring_free(bgmac, &bgmac->tx_ring[i]);

	for (i = 0; i < BGMAC_MAX_RX_RINGS; i++)
		bgmac_dma_rx_ring_free(bgmac, &bgmac->rx_ring[i]);
}

static void bgmac_dma_free(struct bgmac *bgmac)
{
	int i;

	for (i = 0; i < BGMAC_MAX_TX_RINGS; i++)
		bgmac_dma_ring_desc_free(bgmac, &bgmac->tx_ring[i]);

	for (i = 0; i < BGMAC_MAX_RX_RINGS; i++)
		bgmac_dma_ring_desc_free(bgmac, &bgmac->rx_ring[i]);
}

static int bgmac_dma_alloc(struct bgmac *bgmac)
{
	struct device *dma_dev = bgmac->core->dma_dev;
	struct bgmac_dma_ring *ring;
	static const u16 ring_base[] = { BGMAC_DMA_BASE0, BGMAC_DMA_BASE1,
					 BGMAC_DMA_BASE2, BGMAC_DMA_BASE3, };
	int size; /* ring size: different for Tx and Rx */
	int err;
	int i;

	BUILD_BUG_ON(BGMAC_MAX_TX_RINGS > ARRAY_SIZE(ring_base));
	BUILD_BUG_ON(BGMAC_MAX_RX_RINGS > ARRAY_SIZE(ring_base));

	if (!(bcma_aread32(bgmac->core, BCMA_IOST) & BCMA_IOST_DMA64)) {
		bgmac_err(bgmac, "Core does not report 64-bit DMA\n");
		return -ENOTSUPP;
	}

	for (i = 0; i < BGMAC_MAX_TX_RINGS; i++) {
		ring = &bgmac->tx_ring[i];
		ring->num_slots = BGMAC_TX_RING_SLOTS;
		ring->mmio_base = ring_base[i];

		/* Alloc ring of descriptors */
		size = ring->num_slots * sizeof(struct bgmac_dma_desc);
		ring->cpu_base = dma_zalloc_coherent(dma_dev, size,
						     &ring->dma_base,
						     GFP_KERNEL);
		if (!ring->cpu_base) {
			bgmac_err(bgmac, "Allocation of TX ring 0x%X failed\n",
				  ring->mmio_base);
			goto err_dma_free;
		}

		ring->unaligned = bgmac_dma_unaligned(bgmac, ring,
						      BGMAC_DMA_RING_TX);
		if (ring->unaligned)
			ring->index_base = lower_32_bits(ring->dma_base);
		else
			ring->index_base = 0;

		/* No need to alloc TX slots yet */
	}

	for (i = 0; i < BGMAC_MAX_RX_RINGS; i++) {
		ring = &bgmac->rx_ring[i];
		ring->num_slots = BGMAC_RX_RING_SLOTS;
		ring->mmio_base = ring_base[i];

		/* Alloc ring of descriptors */
		size = ring->num_slots * sizeof(struct bgmac_dma_desc);
		ring->cpu_base = dma_zalloc_coherent(dma_dev, size,
						     &ring->dma_base,
						     GFP_KERNEL);
		if (!ring->cpu_base) {
			bgmac_err(bgmac, "Allocation of RX ring 0x%X failed\n",
				  ring->mmio_base);
			err = -ENOMEM;
			goto err_dma_free;
		}

		ring->unaligned = bgmac_dma_unaligned(bgmac, ring,
						      BGMAC_DMA_RING_RX);
		if (ring->unaligned)
			ring->index_base = lower_32_bits(ring->dma_base);
		else
			ring->index_base = 0;
	}

	return 0;

err_dma_free:
	bgmac_dma_free(bgmac);
	return -ENOMEM;
}

static int bgmac_dma_init(struct bgmac *bgmac)
{
	struct bgmac_dma_ring *ring;
	int i, err;

	for (i = 0; i < BGMAC_MAX_TX_RINGS; i++) {
		ring = &bgmac->tx_ring[i];

		if (!ring->unaligned)
			bgmac_dma_tx_enable(bgmac, ring);
		bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGLO,
			    lower_32_bits(ring->dma_base));
		bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_TX_RINGHI,
			    upper_32_bits(ring->dma_base));
		if (ring->unaligned)
			bgmac_dma_tx_enable(bgmac, ring);

		ring->start = 0;
		ring->end = 0;	/* Points the slot that should *not* be read */
	}

	for (i = 0; i < BGMAC_MAX_RX_RINGS; i++) {
		int j;

		ring = &bgmac->rx_ring[i];

		if (!ring->unaligned)
			bgmac_dma_rx_enable(bgmac, ring);
		bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGLO,
			    lower_32_bits(ring->dma_base));
		bgmac_write(bgmac, ring->mmio_base + BGMAC_DMA_RX_RINGHI,
			    upper_32_bits(ring->dma_base));
		if (ring->unaligned)
			bgmac_dma_rx_enable(bgmac, ring);

		ring->start = 0;
		ring->end = 0;
		for (j = 0; j < ring->num_slots; j++) {
			err = bgmac_dma_rx_skb_for_slot(bgmac, &ring->slots[j]);
			if (err)
				goto error;

			bgmac_dma_rx_setup_desc(bgmac, ring, j);
		}

		bgmac_dma_rx_update_index(bgmac, ring);
	}

	return 0;

error:
	bgmac_dma_cleanup(bgmac);
	return err;
}

/**************************************************
 * PHY ops
 **************************************************/

static u16 bgmac_phy_read(struct bgmac *bgmac, u8 phyaddr, u8 reg)
{
	struct bcma_device *core;
	u16 phy_access_addr;
	u16 phy_ctl_addr;
	u32 tmp;

	BUILD_BUG_ON(BGMAC_PA_DATA_MASK != BCMA_GMAC_CMN_PA_DATA_MASK);
	BUILD_BUG_ON(BGMAC_PA_ADDR_MASK != BCMA_GMAC_CMN_PA_ADDR_MASK);
	BUILD_BUG_ON(BGMAC_PA_ADDR_SHIFT != BCMA_GMAC_CMN_PA_ADDR_SHIFT);
	BUILD_BUG_ON(BGMAC_PA_REG_MASK != BCMA_GMAC_CMN_PA_REG_MASK);
	BUILD_BUG_ON(BGMAC_PA_REG_SHIFT != BCMA_GMAC_CMN_PA_REG_SHIFT);
	BUILD_BUG_ON(BGMAC_PA_WRITE != BCMA_GMAC_CMN_PA_WRITE);
	BUILD_BUG_ON(BGMAC_PA_START != BCMA_GMAC_CMN_PA_START);
	BUILD_BUG_ON(BGMAC_PC_EPA_MASK != BCMA_GMAC_CMN_PC_EPA_MASK);
	BUILD_BUG_ON(BGMAC_PC_MCT_MASK != BCMA_GMAC_CMN_PC_MCT_MASK);
	BUILD_BUG_ON(BGMAC_PC_MCT_SHIFT != BCMA_GMAC_CMN_PC_MCT_SHIFT);
	BUILD_BUG_ON(BGMAC_PC_MTE != BCMA_GMAC_CMN_PC_MTE);

	if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) {
		core = bgmac->core->bus->drv_gmac_cmn.core;
		phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS;
		phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL;
	} else {
		core = bgmac->core;
		phy_access_addr = BGMAC_PHY_ACCESS;
		phy_ctl_addr = BGMAC_PHY_CNTL;
	}

	tmp = bcma_read32(core, phy_ctl_addr);
	tmp &= ~BGMAC_PC_EPA_MASK;
	tmp |= phyaddr;
	bcma_write32(core, phy_ctl_addr, tmp);

	tmp = BGMAC_PA_START;
	tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT;
	tmp |= reg << BGMAC_PA_REG_SHIFT;
	bcma_write32(core, phy_access_addr, tmp);

	if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) {
		bgmac_err(bgmac, "Reading PHY %d register 0x%X failed\n",
			  phyaddr, reg);
		return 0xffff;
	}

	return bcma_read32(core, phy_access_addr) & BGMAC_PA_DATA_MASK;
}

/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphywr */
static int bgmac_phy_write(struct bgmac *bgmac, u8 phyaddr, u8 reg, u16 value)
{
	struct bcma_device *core;
	u16 phy_access_addr;
	u16 phy_ctl_addr;
	u32 tmp;

	if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT) {
		core = bgmac->core->bus->drv_gmac_cmn.core;
		phy_access_addr = BCMA_GMAC_CMN_PHY_ACCESS;
		phy_ctl_addr = BCMA_GMAC_CMN_PHY_CTL;
	} else {
		core = bgmac->core;
		phy_access_addr = BGMAC_PHY_ACCESS;
		phy_ctl_addr = BGMAC_PHY_CNTL;
	}

	tmp = bcma_read32(core, phy_ctl_addr);
	tmp &= ~BGMAC_PC_EPA_MASK;
	tmp |= phyaddr;
	bcma_write32(core, phy_ctl_addr, tmp);

	bgmac_write(bgmac, BGMAC_INT_STATUS, BGMAC_IS_MDIO);
	if (bgmac_read(bgmac, BGMAC_INT_STATUS) & BGMAC_IS_MDIO)
		bgmac_warn(bgmac, "Error setting MDIO int\n");

	tmp = BGMAC_PA_START;
	tmp |= BGMAC_PA_WRITE;
	tmp |= phyaddr << BGMAC_PA_ADDR_SHIFT;
	tmp |= reg << BGMAC_PA_REG_SHIFT;
	tmp |= value;
	bcma_write32(core, phy_access_addr, tmp);

	if (!bgmac_wait_value(core, phy_access_addr, BGMAC_PA_START, 0, 1000)) {
		bgmac_err(bgmac, "Writing to PHY %d register 0x%X failed\n",
			  phyaddr, reg);
		return -ETIMEDOUT;
	}

	return 0;
}

/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyinit */
static void bgmac_phy_init(struct bgmac *bgmac)
{
	struct bcma_chipinfo *ci = &bgmac->core->bus->chipinfo;
	struct bcma_drv_cc *cc = &bgmac->core->bus->drv_cc;
	u8 i;

	if (ci->id == BCMA_CHIP_ID_BCM5356) {
		for (i = 0; i < 5; i++) {
			bgmac_phy_write(bgmac, i, 0x1f, 0x008b);
			bgmac_phy_write(bgmac, i, 0x15, 0x0100);
			bgmac_phy_write(bgmac, i, 0x1f, 0x000f);
			bgmac_phy_write(bgmac, i, 0x12, 0x2aaa);
			bgmac_phy_write(bgmac, i, 0x1f, 0x000b);
		}
	}
	if ((ci->id == BCMA_CHIP_ID_BCM5357 && ci->pkg != 10) ||
	    (ci->id == BCMA_CHIP_ID_BCM4749 && ci->pkg != 10) ||
	    (ci->id == BCMA_CHIP_ID_BCM53572 && ci->pkg != 9)) {
		bcma_chipco_chipctl_maskset(cc, 2, ~0xc0000000, 0);
		bcma_chipco_chipctl_maskset(cc, 4, ~0x80000000, 0);
		for (i = 0; i < 5; i++) {
			bgmac_phy_write(bgmac, i, 0x1f, 0x000f);
			bgmac_phy_write(bgmac, i, 0x16, 0x5284);
			bgmac_phy_write(bgmac, i, 0x1f, 0x000b);
			bgmac_phy_write(bgmac, i, 0x17, 0x0010);
			bgmac_phy_write(bgmac, i, 0x1f, 0x000f);
			bgmac_phy_write(bgmac, i, 0x16, 0x5296);
			bgmac_phy_write(bgmac, i, 0x17, 0x1073);
			bgmac_phy_write(bgmac, i, 0x17, 0x9073);
			bgmac_phy_write(bgmac, i, 0x16, 0x52b6);
			bgmac_phy_write(bgmac, i, 0x17, 0x9273);
			bgmac_phy_write(bgmac, i, 0x1f, 0x000b);
		}
	}
}

/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/chipphyreset */
static void bgmac_phy_reset(struct bgmac *bgmac)
{
	if (bgmac->phyaddr == BGMAC_PHY_NOREGS)
		return;

	bgmac_phy_write(bgmac, bgmac->phyaddr, MII_BMCR, BMCR_RESET);
	udelay(100);
	if (bgmac_phy_read(bgmac, bgmac->phyaddr, MII_BMCR) & BMCR_RESET)
		bgmac_err(bgmac, "PHY reset failed\n");
	bgmac_phy_init(bgmac);
}

/**************************************************
 * Chip ops
 **************************************************/

/* TODO: can we just drop @force? Can we don't reset MAC at all if there is
 * nothing to change? Try if after stabilizng driver.
 */
static void bgmac_cmdcfg_maskset(struct bgmac *bgmac, u32 mask, u32 set,
				 bool force)
{
	u32 cmdcfg = bgmac_read(bgmac, BGMAC_CMDCFG);
	u32 new_val = (cmdcfg & mask) | set;

	bgmac_set(bgmac, BGMAC_CMDCFG, BGMAC_CMDCFG_SR(bgmac->core->id.rev));
	udelay(2);

	if (new_val != cmdcfg || force)
		bgmac_write(bgmac, BGMAC_CMDCFG, new_val);

	bgmac_mask(bgmac, BGMAC_CMDCFG, ~BGMAC_CMDCFG_SR(bgmac->core->id.rev));
	udelay(2);
}

static void bgmac_write_mac_address(struct bgmac *bgmac, u8 *addr)
{
	u32 tmp;

	tmp = (addr[0] << 24) | (addr[1] << 16) | (addr[2] << 8) | addr[3];
	bgmac_write(bgmac, BGMAC_MACADDR_HIGH, tmp);
	tmp = (addr[4] << 8) | addr[5];
	bgmac_write(bgmac, BGMAC_MACADDR_LOW, tmp);
}

static void bgmac_set_rx_mode(struct net_device *net_dev)
{
	struct bgmac *bgmac = netdev_priv(net_dev);

	if (net_dev->flags & IFF_PROMISC)
		bgmac_cmdcfg_maskset(bgmac, ~0, BGMAC_CMDCFG_PROM, true);
	else
		bgmac_cmdcfg_maskset(bgmac, ~BGMAC_CMDCFG_PROM, 0, true);
}

#if 0 /* We don't use that regs yet */
static void bgmac_chip_stats_update(struct bgmac *bgmac)
{
	int i;

	if (bgmac->core->id.id != BCMA_CORE_4706_MAC_GBIT) {
		for (i = 0; i < BGMAC_NUM_MIB_TX_REGS; i++)
			bgmac->mib_tx_regs[i] =
				bgmac_read(bgmac,
					   BGMAC_TX_GOOD_OCTETS + (i * 4));
		for (i = 0; i < BGMAC_NUM_MIB_RX_REGS; i++)
			bgmac->mib_rx_regs[i] =
				bgmac_read(bgmac,
					   BGMAC_RX_GOOD_OCTETS + (i * 4));
	}

	/* TODO: what else? how to handle BCM4706? Specs are needed */
}
#endif

static void bgmac_clear_mib(struct bgmac *bgmac)
{
	int i;

	if (bgmac->core->id.id == BCMA_CORE_4706_MAC_GBIT)
		return;

	bgmac_set(bgmac, BGMAC_DEV_CTL, BGMAC_DC_MROR);
	for (i = 0; i < BGMAC_NUM_MIB_TX_REGS; i++)
		bgmac_read(bgmac, BGMAC_TX_GOOD_OCTETS + (i * 4));
	for (i = 0; i < BGMAC_NUM_MIB_RX_REGS; i++)
		bgmac_read(bgmac, BGMAC_RX_GOOD_OCTETS + (i * 4));
}

/* http://bcm-v4.sipsolutions.net/mac-gbit/gmac/gmac_speed */
static void bgmac_mac_speed(struct bgmac *bgmac)
{
	u32 mask = ~(BGMAC_CMDCFG_ES_MASK | BGMAC_CMDCFG_HD);
	u32 set = 0;

	switch (bgmac->mac_speed) {
	case SPEED_10:
		set |= BGMAC_CMDCFG_ES_10;
		break;
	case SPEED_100:
		set |= BGMAC_CMDCFG_ES_100;
		break;
	case SPEED_1000:
		set |= BGMAC_CMDCFG_ES_1000;
		break;
	case SPEED_2500:
		set |= BGMAC_CMDCFG_ES_2500;
		break;
	default:
		bgmac_err(bgmac, "Unsupported speed: %d\n", bgmac->mac_speed);
	}

	if (bgmac->mac_duplex == DUPLEX_HALF)
		set |= BGMAC_CMDCFG_HD;

	bgmac_cmdcfg_maskset(bgmac, mask, set, true);
}

static void bgmac_miiconfig(struct bgmac *bgmac)
{
	struct bcma_device *core = bgmac->core;
	struct bcma_chipinfo *ci = &core->bus->chipinfo;
	u8 imode;

	if (ci->id == BCMA_CHIP_ID_BCM4707 ||
	    ci->id == BCMA_CHIP_ID_BCM53018) {
		bcma_awrite32(core, BCMA_IOCTL,
			      bcma_aread32(core, BCMA_IOCTL) | 0x40 |
			      BGMAC_BCMA_IOCTL_SW_CLKEN);
		bgmac->mac_speed = SPEED_2500;
		bgmac->mac_duplex = DUPLEX_FULL;
		bgmac_mac_speed(bgmac);
	} else {
		imode = (bgmac_read(bgmac, BGMAC_DEV_STATUS) &
			BGMAC_DS_MM_MASK) >> BGMAC_DS_MM_SHIFT;
		if (imode == 0 || imode == 1) {
			bgmac->mac_speed = SPEED_100;
			bgmac->mac_duplex = DUPLEX_FULL;