MODULE_DEVICE_TABLE(pci, rtl8169_pci_tbl);
-/*
- * we set our copybreak very high so that we don't have
- * to allocate 16k frames all the time (see note in
- * rtl8169_open()
- */
-static int rx_copybreak = 16383;
+static int rx_buf_sz = 16383;
static int use_dac;
static struct {
u32 msg_enable;
struct RxDesc *RxDescArray; /* 256-aligned Rx descriptor ring */
dma_addr_t TxPhyAddr;
dma_addr_t RxPhyAddr;
- struct sk_buff *Rx_skbuff[NUM_RX_DESC]; /* Rx data buffers */
+ void *Rx_databuff[NUM_RX_DESC]; /* Rx data buffers */
struct ring_info tx_skb[NUM_TX_DESC]; /* Tx data buffers */
- unsigned align;
- unsigned rx_buf_sz;
struct timer_list timer;
u16 cp_cmd;
u16 intr_event;
MODULE_AUTHOR("Realtek and the Linux r8169 crew <netdev@vger.kernel.org>");
MODULE_DESCRIPTION("RealTek RTL-8169 Gigabit Ethernet driver");
-module_param(rx_copybreak, int, 0);
-MODULE_PARM_DESC(rx_copybreak, "Copy breakpoint for copy-only-tiny-frames");
module_param(use_dac, int, 0);
MODULE_PARM_DESC(use_dac, "Enable PCI DAC. Unsafe on 32 bit PCI slot.");
module_param_named(debug, debug.msg_enable, int, 0);
static inline u32 rtl8169_tx_vlan_tag(struct rtl8169_private *tp,
struct sk_buff *skb)
{
- return (tp->vlgrp && vlan_tx_tag_present(skb)) ?
+ return (vlan_tx_tag_present(skb)) ?
TxVlanTag | swab16(vlan_tx_tag_get(skb)) : 0x00;
}
int ret;
if (vlgrp && (opts2 & RxVlanTag)) {
- __vlan_hwaccel_rx(skb, vlgrp, swab16(opts2 & 0xffff), polling);
+ u16 vtag = swab16(opts2 & 0xffff);
+
+ if (likely(polling))
+ vlan_gro_receive(&tp->napi, vlgrp, vtag, skb);
+ else
+ __vlan_hwaccel_rx(skb, vlgrp, vtag, polling);
ret = 0;
} else
ret = -1;
if ((RTL_R8(ChipCmd) & CmdRxEnb) == 0)
return;
- counters = pci_alloc_consistent(tp->pci_dev, sizeof(*counters), &paddr);
+ counters = dma_alloc_coherent(&tp->pci_dev->dev, sizeof(*counters),
+ &paddr, GFP_KERNEL);
if (!counters)
return;
RTL_W32(CounterAddrLow, 0);
RTL_W32(CounterAddrHigh, 0);
- pci_free_consistent(tp->pci_dev, sizeof(*counters), counters, paddr);
+ dma_free_coherent(&tp->pci_dev->dev, sizeof(*counters), counters,
+ paddr);
}
static void rtl8169_get_ethtool_stats(struct net_device *dev,
#ifdef CONFIG_R8169_VLAN
dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_RX;
#endif
+ dev->features |= NETIF_F_GRO;
tp->intr_mask = 0xffff;
- tp->align = cfg->align;
tp->hw_start = cfg->hw_start;
tp->intr_event = cfg->intr_event;
tp->napi_event = cfg->napi_event;
pci_set_drvdata(pdev, NULL);
}
-static void rtl8169_set_rxbufsize(struct rtl8169_private *tp,
- unsigned int mtu)
-{
- unsigned int max_frame = mtu + VLAN_ETH_HLEN + ETH_FCS_LEN;
-
- if (max_frame != 16383)
- printk(KERN_WARNING PFX "WARNING! Changing of MTU on this "
- "NIC may lead to frame reception errors!\n");
-
- tp->rx_buf_sz = (max_frame > RX_BUF_SIZE) ? max_frame : RX_BUF_SIZE;
-}
-
static int rtl8169_open(struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
pm_runtime_get_sync(&pdev->dev);
- /*
- * Note that we use a magic value here, its wierd I know
- * its done because, some subset of rtl8169 hardware suffers from
- * a problem in which frames received that are longer than
- * the size set in RxMaxSize register return garbage sizes
- * when received. To avoid this we need to turn off filtering,
- * which is done by setting a value of 16383 in the RxMaxSize register
- * and allocating 16k frames to handle the largest possible rx value
- * thats what the magic math below does.
- */
- rtl8169_set_rxbufsize(tp, 16383 - VLAN_ETH_HLEN - ETH_FCS_LEN);
-
/*
* Rx and Tx desscriptors needs 256 bytes alignment.
- * pci_alloc_consistent provides more.
+ * dma_alloc_coherent provides more.
*/
- tp->TxDescArray = pci_alloc_consistent(pdev, R8169_TX_RING_BYTES,
- &tp->TxPhyAddr);
+ tp->TxDescArray = dma_alloc_coherent(&pdev->dev, R8169_TX_RING_BYTES,
+ &tp->TxPhyAddr, GFP_KERNEL);
if (!tp->TxDescArray)
goto err_pm_runtime_put;
- tp->RxDescArray = pci_alloc_consistent(pdev, R8169_RX_RING_BYTES,
- &tp->RxPhyAddr);
+ tp->RxDescArray = dma_alloc_coherent(&pdev->dev, R8169_RX_RING_BYTES,
+ &tp->RxPhyAddr, GFP_KERNEL);
if (!tp->RxDescArray)
goto err_free_tx_0;
err_release_ring_2:
rtl8169_rx_clear(tp);
err_free_rx_1:
- pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
- tp->RxPhyAddr);
+ dma_free_coherent(&pdev->dev, R8169_RX_RING_BYTES, tp->RxDescArray,
+ tp->RxPhyAddr);
tp->RxDescArray = NULL;
err_free_tx_0:
- pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
- tp->TxPhyAddr);
+ dma_free_coherent(&pdev->dev, R8169_TX_RING_BYTES, tp->TxDescArray,
+ tp->TxPhyAddr);
tp->TxDescArray = NULL;
err_pm_runtime_put:
pm_runtime_put_noidle(&pdev->dev);
RTL_W8(EarlyTxThres, EarlyTxThld);
- rtl_set_rx_max_size(ioaddr, tp->rx_buf_sz);
+ rtl_set_rx_max_size(ioaddr, rx_buf_sz);
if ((tp->mac_version == RTL_GIGA_MAC_VER_01) ||
(tp->mac_version == RTL_GIGA_MAC_VER_02) ||
RTL_W8(EarlyTxThres, EarlyTxThld);
- rtl_set_rx_max_size(ioaddr, tp->rx_buf_sz);
+ rtl_set_rx_max_size(ioaddr, rx_buf_sz);
tp->cp_cmd |= RTL_R16(CPlusCmd) | PktCntrDisable | INTT_1;
RTL_W8(EarlyTxThres, EarlyTxThld);
- rtl_set_rx_max_size(ioaddr, tp->rx_buf_sz);
+ rtl_set_rx_max_size(ioaddr, rx_buf_sz);
tp->cp_cmd |= rtl_rw_cpluscmd(ioaddr) | PCIMulRW;
rtl8169_down(dev);
- rtl8169_set_rxbufsize(tp, dev->mtu);
-
ret = rtl8169_init_ring(dev);
if (ret < 0)
goto out;
desc->opts1 &= ~cpu_to_le32(DescOwn | RsvdMask);
}
-static void rtl8169_free_rx_skb(struct rtl8169_private *tp,
- struct sk_buff **sk_buff, struct RxDesc *desc)
+static void rtl8169_free_rx_databuff(struct rtl8169_private *tp,
+ void **data_buff, struct RxDesc *desc)
{
struct pci_dev *pdev = tp->pci_dev;
- pci_unmap_single(pdev, le64_to_cpu(desc->addr), tp->rx_buf_sz,
+ dma_unmap_single(&pdev->dev, le64_to_cpu(desc->addr), rx_buf_sz,
PCI_DMA_FROMDEVICE);
- dev_kfree_skb(*sk_buff);
- *sk_buff = NULL;
+ kfree(*data_buff);
+ *data_buff = NULL;
rtl8169_make_unusable_by_asic(desc);
}
rtl8169_mark_to_asic(desc, rx_buf_sz);
}
-static struct sk_buff *rtl8169_alloc_rx_skb(struct pci_dev *pdev,
- struct net_device *dev,
- struct RxDesc *desc, int rx_buf_sz,
- unsigned int align)
+static inline void *rtl8169_align(void *data)
{
- struct sk_buff *skb;
- dma_addr_t mapping;
- unsigned int pad;
+ return (void *)ALIGN((long)data, 16);
+}
- pad = align ? align : NET_IP_ALIGN;
+static struct sk_buff *rtl8169_alloc_rx_data(struct rtl8169_private *tp,
+ struct RxDesc *desc)
+{
+ void *data;
+ dma_addr_t mapping;
+ struct net_device *dev = tp->dev;
+ int node = dev->dev.parent ? dev_to_node(dev->dev.parent) : -1;
- skb = netdev_alloc_skb(dev, rx_buf_sz + pad);
- if (!skb)
- goto err_out;
+ data = kmalloc_node(rx_buf_sz, GFP_KERNEL, node);
+ if (!data)
+ return NULL;
- skb_reserve(skb, align ? ((pad - 1) & (unsigned long)skb->data) : pad);
+ if (rtl8169_align(data) != data) {
+ kfree(data);
+ data = kmalloc_node(rx_buf_sz + 15, GFP_KERNEL, node);
+ if (!data)
+ return NULL;
+ }
- mapping = pci_map_single(pdev, skb->data, rx_buf_sz,
+ mapping = dma_map_single(&tp->pci_dev->dev, rtl8169_align(data), rx_buf_sz,
PCI_DMA_FROMDEVICE);
+ if (unlikely(dma_mapping_error(&tp->pci_dev->dev, mapping)))
+ goto err_out;
rtl8169_map_to_asic(desc, mapping, rx_buf_sz);
-out:
- return skb;
+ return data;
err_out:
- rtl8169_make_unusable_by_asic(desc);
- goto out;
+ kfree(data);
+ return NULL;
}
static void rtl8169_rx_clear(struct rtl8169_private *tp)
unsigned int i;
for (i = 0; i < NUM_RX_DESC; i++) {
- if (tp->Rx_skbuff[i]) {
- rtl8169_free_rx_skb(tp, tp->Rx_skbuff + i,
+ if (tp->Rx_databuff[i]) {
+ rtl8169_free_rx_databuff(tp, tp->Rx_databuff + i,
tp->RxDescArray + i);
}
}
}
-static u32 rtl8169_rx_fill(struct rtl8169_private *tp, struct net_device *dev,
- u32 start, u32 end)
+static inline void rtl8169_mark_as_last_descriptor(struct RxDesc *desc)
{
- u32 cur;
+ desc->opts1 |= cpu_to_le32(RingEnd);
+}
- for (cur = start; end - cur != 0; cur++) {
- struct sk_buff *skb;
- unsigned int i = cur % NUM_RX_DESC;
+static int rtl8169_rx_fill(struct rtl8169_private *tp)
+{
+ unsigned int i;
- WARN_ON((s32)(end - cur) < 0);
+ for (i = 0; i < NUM_RX_DESC; i++) {
+ void *data;
- if (tp->Rx_skbuff[i])
+ if (tp->Rx_databuff[i])
continue;
- skb = rtl8169_alloc_rx_skb(tp->pci_dev, dev,
- tp->RxDescArray + i,
- tp->rx_buf_sz, tp->align);
- if (!skb)
- break;
-
- tp->Rx_skbuff[i] = skb;
+ data = rtl8169_alloc_rx_data(tp, tp->RxDescArray + i);
+ if (!data) {
+ rtl8169_make_unusable_by_asic(tp->RxDescArray + i);
+ goto err_out;
+ }
+ tp->Rx_databuff[i] = data;
}
- return cur - start;
-}
-static inline void rtl8169_mark_as_last_descriptor(struct RxDesc *desc)
-{
- desc->opts1 |= cpu_to_le32(RingEnd);
+ rtl8169_mark_as_last_descriptor(tp->RxDescArray + NUM_RX_DESC - 1);
+ return 0;
+
+err_out:
+ rtl8169_rx_clear(tp);
+ return -ENOMEM;
}
static void rtl8169_init_ring_indexes(struct rtl8169_private *tp)
rtl8169_init_ring_indexes(tp);
memset(tp->tx_skb, 0x0, NUM_TX_DESC * sizeof(struct ring_info));
- memset(tp->Rx_skbuff, 0x0, NUM_RX_DESC * sizeof(struct sk_buff *));
-
- if (rtl8169_rx_fill(tp, dev, 0, NUM_RX_DESC) != NUM_RX_DESC)
- goto err_out;
+ memset(tp->Rx_databuff, 0x0, NUM_RX_DESC * sizeof(void *));
- rtl8169_mark_as_last_descriptor(tp->RxDescArray + NUM_RX_DESC - 1);
-
- return 0;
-
-err_out:
- rtl8169_rx_clear(tp);
- return -ENOMEM;
+ return rtl8169_rx_fill(tp);
}
static void rtl8169_unmap_tx_skb(struct pci_dev *pdev, struct ring_info *tx_skb,
{
unsigned int len = tx_skb->len;
- pci_unmap_single(pdev, le64_to_cpu(desc->addr), len, PCI_DMA_TODEVICE);
+ dma_unmap_single(&pdev->dev, le64_to_cpu(desc->addr), len,
+ PCI_DMA_TODEVICE);
desc->opts1 = 0x00;
desc->opts2 = 0x00;
desc->addr = 0x00;
tx_skb->len = 0;
}
-static void rtl8169_tx_clear(struct rtl8169_private *tp)
+static void rtl8169_tx_clear_range(struct rtl8169_private *tp, u32 start,
+ unsigned int n)
{
unsigned int i;
- for (i = tp->dirty_tx; i < tp->dirty_tx + NUM_TX_DESC; i++) {
- unsigned int entry = i % NUM_TX_DESC;
+ for (i = 0; i < n; i++) {
+ unsigned int entry = (start + i) % NUM_TX_DESC;
struct ring_info *tx_skb = tp->tx_skb + entry;
unsigned int len = tx_skb->len;
tp->dev->stats.tx_dropped++;
}
}
+}
+
+static void rtl8169_tx_clear(struct rtl8169_private *tp)
+{
+ rtl8169_tx_clear_range(tp, tp->dirty_tx, NUM_TX_DESC);
tp->cur_tx = tp->dirty_tx = 0;
}
txd = tp->TxDescArray + entry;
len = frag->size;
addr = ((void *) page_address(frag->page)) + frag->page_offset;
- mapping = pci_map_single(tp->pci_dev, addr, len, PCI_DMA_TODEVICE);
+ mapping = dma_map_single(&tp->pci_dev->dev, addr, len,
+ PCI_DMA_TODEVICE);
+ if (unlikely(dma_mapping_error(&tp->pci_dev->dev, mapping)))
+ goto err_out;
/* anti gcc 2.95.3 bugware (sic) */
status = opts1 | len | (RingEnd * !((entry + 1) % NUM_TX_DESC));
}
return cur_frag;
+
+err_out:
+ rtl8169_tx_clear_range(tp, tp->cur_tx + 1, cur_frag);
+ return -EIO;
}
static inline u32 rtl8169_tso_csum(struct sk_buff *skb, struct net_device *dev)
struct net_device *dev)
{
struct rtl8169_private *tp = netdev_priv(dev);
- unsigned int frags, entry = tp->cur_tx % NUM_TX_DESC;
+ unsigned int entry = tp->cur_tx % NUM_TX_DESC;
struct TxDesc *txd = tp->TxDescArray + entry;
void __iomem *ioaddr = tp->mmio_addr;
dma_addr_t mapping;
u32 status, len;
u32 opts1;
+ int frags;
if (unlikely(TX_BUFFS_AVAIL(tp) < skb_shinfo(skb)->nr_frags)) {
netif_err(tp, drv, dev, "BUG! Tx Ring full when queue awake!\n");
- goto err_stop;
+ goto err_stop_0;
}
if (unlikely(le32_to_cpu(txd->opts1) & DescOwn))
- goto err_stop;
+ goto err_stop_0;
+
+ len = skb_headlen(skb);
+ mapping = dma_map_single(&tp->pci_dev->dev, skb->data, len,
+ PCI_DMA_TODEVICE);
+ if (unlikely(dma_mapping_error(&tp->pci_dev->dev, mapping)))
+ goto err_dma_0;
+
+ tp->tx_skb[entry].len = len;
+ txd->addr = cpu_to_le64(mapping);
+ txd->opts2 = cpu_to_le32(rtl8169_tx_vlan_tag(tp, skb));
opts1 = DescOwn | rtl8169_tso_csum(skb, dev);
frags = rtl8169_xmit_frags(tp, skb, opts1);
- if (frags) {
- len = skb_headlen(skb);
+ if (frags < 0)
+ goto err_dma_1;
+ else if (frags)
opts1 |= FirstFrag;
- } else {
- len = skb->len;
+ else {
opts1 |= FirstFrag | LastFrag;
tp->tx_skb[entry].skb = skb;
}
- mapping = pci_map_single(tp->pci_dev, skb->data, len, PCI_DMA_TODEVICE);
-
- tp->tx_skb[entry].len = len;
- txd->addr = cpu_to_le64(mapping);
- txd->opts2 = cpu_to_le32(rtl8169_tx_vlan_tag(tp, skb));
-
wmb();
/* anti gcc 2.95.3 bugware (sic) */
return NETDEV_TX_OK;
-err_stop:
+err_dma_1:
+ rtl8169_unmap_tx_skb(tp->pci_dev, tp->tx_skb + entry, txd);
+err_dma_0:
+ dev_kfree_skb(skb);
+ dev->stats.tx_dropped++;
+ return NETDEV_TX_OK;
+
+err_stop_0:
netif_stop_queue(dev);
dev->stats.tx_dropped++;
return NETDEV_TX_BUSY;
return (status & (FirstFrag | LastFrag)) != (FirstFrag | LastFrag);
}
-static inline void rtl8169_rx_csum(struct sk_buff *skb, struct RxDesc *desc)
+static inline void rtl8169_rx_csum(struct sk_buff *skb, u32 opts1)
{
- u32 opts1 = le32_to_cpu(desc->opts1);
u32 status = opts1 & RxProtoMask;
if (((status == RxProtoTCP) && !(opts1 & TCPFail)) ||
((status == RxProtoIP) && !(opts1 & IPFail)))
skb->ip_summed = CHECKSUM_UNNECESSARY;
else
- skb->ip_summed = CHECKSUM_NONE;
+ skb_checksum_none_assert(skb);
}
-static inline bool rtl8169_try_rx_copy(struct sk_buff **sk_buff,
- struct rtl8169_private *tp, int pkt_size,
- dma_addr_t addr)
+static struct sk_buff *rtl8169_try_rx_copy(void *data,
+ struct rtl8169_private *tp,
+ int pkt_size,
+ dma_addr_t addr)
{
struct sk_buff *skb;
- bool done = false;
-
- if (pkt_size >= rx_copybreak)
- goto out;
+ data = rtl8169_align(data);
+ dma_sync_single_for_cpu(&tp->pci_dev->dev, addr, pkt_size,
+ PCI_DMA_FROMDEVICE);
+ prefetch(data);
skb = netdev_alloc_skb_ip_align(tp->dev, pkt_size);
- if (!skb)
- goto out;
-
- pci_dma_sync_single_for_cpu(tp->pci_dev, addr, pkt_size,
- PCI_DMA_FROMDEVICE);
- skb_copy_from_linear_data(*sk_buff, skb->data, pkt_size);
- *sk_buff = skb;
- done = true;
-out:
- return done;
+ if (skb)
+ memcpy(skb->data, data, pkt_size);
+ dma_sync_single_for_device(&tp->pci_dev->dev, addr, pkt_size,
+ PCI_DMA_FROMDEVICE);
+ return skb;
}
/*
void __iomem *ioaddr, u32 budget)
{
unsigned int cur_rx, rx_left;
- unsigned int delta, count;
+ unsigned int count;
int polling = (budget != ~(u32)0) ? 1 : 0;
cur_rx = tp->cur_rx;
rtl8169_schedule_work(dev, rtl8169_reset_task);
dev->stats.rx_fifo_errors++;
}
- rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
+ rtl8169_mark_to_asic(desc, rx_buf_sz);
} else {
- struct sk_buff *skb = tp->Rx_skbuff[entry];
+ struct sk_buff *skb;
dma_addr_t addr = le64_to_cpu(desc->addr);
int pkt_size = (status & 0x00001FFF) - 4;
- struct pci_dev *pdev = tp->pci_dev;
/*
* The driver does not support incoming fragmented
if (unlikely(rtl8169_fragmented_frame(status))) {
dev->stats.rx_dropped++;
dev->stats.rx_length_errors++;
- rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
+ rtl8169_mark_to_asic(desc, rx_buf_sz);
continue;
}
- rtl8169_rx_csum(skb, desc);
-
- if (rtl8169_try_rx_copy(&skb, tp, pkt_size, addr)) {
- pci_dma_sync_single_for_device(pdev, addr,
- pkt_size, PCI_DMA_FROMDEVICE);
- rtl8169_mark_to_asic(desc, tp->rx_buf_sz);
- } else {
- pci_unmap_single(pdev, addr, tp->rx_buf_sz,
- PCI_DMA_FROMDEVICE);
- tp->Rx_skbuff[entry] = NULL;
+ skb = rtl8169_try_rx_copy(tp->Rx_databuff[entry],
+ tp, pkt_size, addr);
+ rtl8169_mark_to_asic(desc, rx_buf_sz);
+ if (!skb) {
+ dev->stats.rx_dropped++;
+ continue;
}
+ rtl8169_rx_csum(skb, status);
skb_put(skb, pkt_size);
skb->protocol = eth_type_trans(skb, dev);
if (rtl8169_rx_vlan_skb(tp, desc, skb, polling) < 0) {
if (likely(polling))
- netif_receive_skb(skb);
+ napi_gro_receive(&tp->napi, skb);
else
netif_rx(skb);
}
count = cur_rx - tp->cur_rx;
tp->cur_rx = cur_rx;
- delta = rtl8169_rx_fill(tp, dev, tp->dirty_rx, tp->cur_rx);
- if (!delta && count)
- netif_info(tp, intr, dev, "no Rx buffer allocated\n");
- tp->dirty_rx += delta;
-
- /*
- * FIXME: until there is periodic timer to try and refill the ring,
- * a temporary shortage may definitely kill the Rx process.
- * - disable the asic to try and avoid an overflow and kick it again
- * after refill ?
- * - how do others driver handle this condition (Uh oh...).
- */
- if (tp->dirty_rx + NUM_RX_DESC == tp->cur_rx)
- netif_emerg(tp, intr, dev, "Rx buffers exhausted\n");
+ tp->dirty_rx += count;
return count;
}
free_irq(dev->irq, dev);
- pci_free_consistent(pdev, R8169_RX_RING_BYTES, tp->RxDescArray,
- tp->RxPhyAddr);
- pci_free_consistent(pdev, R8169_TX_RING_BYTES, tp->TxDescArray,
- tp->TxPhyAddr);
+ dma_free_coherent(&pdev->dev, R8169_RX_RING_BYTES, tp->RxDescArray,
+ tp->RxPhyAddr);
+ dma_free_coherent(&pdev->dev, R8169_TX_RING_BYTES, tp->TxDescArray,
+ tp->TxPhyAddr);
tp->TxDescArray = NULL;
tp->RxDescArray = NULL;
projects / mv-sheeva.git / blobdiff