* trademarks of NVIDIA Corporation in the United States and other
* countries.
*
- * Copyright (C) 2003,4 Manfred Spraul
+ * Copyright (C) 2003,4,5 Manfred Spraul
* Copyright (C) 2004 Andrew de Quincey (wol support)
* Copyright (C) 2004 Carl-Daniel Hailfinger (invalid MAC handling, insane
* IRQ rate fixes, bigendian fixes, cleanups, verification)
* 0.45: 18 Sep 2005: Remove nv_stop/start_rx from every link check
* 0.46: 20 Oct 2005: Add irq optimization modes.
* 0.47: 26 Oct 2005: Add phyaddr 0 in phy scan.
+ * 0.48: 24 Dec 2005: Disable TSO, bugfix for pci_map_single
+ * 0.49: 10 Dec 2005: Fix tso for large buffers.
*
* Known bugs:
* We suspect that on some hardware no TX done interrupts are generated.
* DEV_NEED_TIMERIRQ will not harm you on sane hardware, only generating a few
* superfluous timer interrupts from the nic.
*/
-#define FORCEDETH_VERSION "0.47"
+#define FORCEDETH_VERSION "0.49"
#define DRV_NAME "forcedeth"
#include <linux/module.h>
#define NV_TX2_VALID (1<<31)
#define NV_TX2_TSO (1<<28)
#define NV_TX2_TSO_SHIFT 14
+#define NV_TX2_TSO_MAX_SHIFT 14
+#define NV_TX2_TSO_MAX_SIZE (1<<NV_TX2_TSO_MAX_SHIFT)
#define NV_TX2_CHECKSUM_L3 (1<<27)
#define NV_TX2_CHECKSUM_L4 (1<<26)
#define NV_WATCHDOG_TIMEO (5*HZ)
#define RX_RING 128
-#define TX_RING 64
+#define TX_RING 256
/*
* If your nic mysteriously hangs then try to reduce the limits
* to 1/0: It might be required to set NV_TX_LASTPACKET in the
* last valid ring entry. But this would be impossible to
* implement - probably a disassembly error.
*/
-#define TX_LIMIT_STOP 63
-#define TX_LIMIT_START 62
+#define TX_LIMIT_STOP 255
+#define TX_LIMIT_START 254
/* rx/tx mac addr + type + vlan + align + slack*/
#define NV_RX_HEADERS (64)
unsigned int next_tx, nic_tx;
struct sk_buff *tx_skbuff[TX_RING];
dma_addr_t tx_dma[TX_RING];
+ unsigned int tx_dma_len[TX_RING];
u32 tx_flags;
};
} else {
skb = np->rx_skbuff[nr];
}
- np->rx_dma[nr] = pci_map_single(np->pci_dev, skb->data, skb->len,
- PCI_DMA_FROMDEVICE);
+ np->rx_dma[nr] = pci_map_single(np->pci_dev, skb->data,
+ skb->end-skb->data, PCI_DMA_FROMDEVICE);
if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2) {
np->rx_ring.orig[nr].PacketBuffer = cpu_to_le32(np->rx_dma[nr]);
wmb();
else
np->tx_ring.ex[i].FlagLen = 0;
np->tx_skbuff[i] = NULL;
+ np->tx_dma[i] = 0;
}
}
return nv_alloc_rx(dev);
}
-static void nv_release_txskb(struct net_device *dev, unsigned int skbnr)
+static int nv_release_txskb(struct net_device *dev, unsigned int skbnr)
{
struct fe_priv *np = netdev_priv(dev);
- struct sk_buff *skb = np->tx_skbuff[skbnr];
- unsigned int j, entry, fragments;
-
- dprintk(KERN_INFO "%s: nv_release_txskb for skbnr %d, skb %p\n",
- dev->name, skbnr, np->tx_skbuff[skbnr]);
-
- entry = skbnr;
- if ((fragments = skb_shinfo(skb)->nr_frags) != 0) {
- for (j = fragments; j >= 1; j--) {
- skb_frag_t *frag = &skb_shinfo(skb)->frags[j-1];
- pci_unmap_page(np->pci_dev, np->tx_dma[entry],
- frag->size,
- PCI_DMA_TODEVICE);
- entry = (entry - 1) % TX_RING;
- }
+
+ dprintk(KERN_INFO "%s: nv_release_txskb for skbnr %d\n",
+ dev->name, skbnr);
+
+ if (np->tx_dma[skbnr]) {
+ pci_unmap_page(np->pci_dev, np->tx_dma[skbnr],
+ np->tx_dma_len[skbnr],
+ PCI_DMA_TODEVICE);
+ np->tx_dma[skbnr] = 0;
+ }
+
+ if (np->tx_skbuff[skbnr]) {
+ dev_kfree_skb_irq(np->tx_skbuff[skbnr]);
+ np->tx_skbuff[skbnr] = NULL;
+ return 1;
+ } else {
+ return 0;
}
- pci_unmap_single(np->pci_dev, np->tx_dma[entry],
- skb->len - skb->data_len,
- PCI_DMA_TODEVICE);
- dev_kfree_skb_irq(skb);
- np->tx_skbuff[skbnr] = NULL;
}
static void nv_drain_tx(struct net_device *dev)
np->tx_ring.orig[i].FlagLen = 0;
else
np->tx_ring.ex[i].FlagLen = 0;
- if (np->tx_skbuff[i]) {
- nv_release_txskb(dev, i);
+ if (nv_release_txskb(dev, i))
np->stats.tx_dropped++;
- }
}
}
wmb();
if (np->rx_skbuff[i]) {
pci_unmap_single(np->pci_dev, np->rx_dma[i],
- np->rx_skbuff[i]->len,
+ np->rx_skbuff[i]->end-np->rx_skbuff[i]->data,
PCI_DMA_FROMDEVICE);
dev_kfree_skb(np->rx_skbuff[i]);
np->rx_skbuff[i] = NULL;
static int nv_start_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct fe_priv *np = netdev_priv(dev);
+ u32 tx_flags = 0;
u32 tx_flags_extra = (np->desc_ver == DESC_VER_1 ? NV_TX_LASTPACKET : NV_TX2_LASTPACKET);
unsigned int fragments = skb_shinfo(skb)->nr_frags;
- unsigned int nr = (np->next_tx + fragments) % TX_RING;
+ unsigned int nr = (np->next_tx - 1) % TX_RING;
+ unsigned int start_nr = np->next_tx % TX_RING;
unsigned int i;
+ u32 offset = 0;
+ u32 bcnt;
+ u32 size = skb->len-skb->data_len;
+ u32 entries = (size >> NV_TX2_TSO_MAX_SHIFT) + ((size & (NV_TX2_TSO_MAX_SIZE-1)) ? 1 : 0);
+
+ /* add fragments to entries count */
+ for (i = 0; i < fragments; i++) {
+ entries += (skb_shinfo(skb)->frags[i].size >> NV_TX2_TSO_MAX_SHIFT) +
+ ((skb_shinfo(skb)->frags[i].size & (NV_TX2_TSO_MAX_SIZE-1)) ? 1 : 0);
+ }
spin_lock_irq(&np->lock);
- if ((np->next_tx - np->nic_tx + fragments) > TX_LIMIT_STOP) {
+ if ((np->next_tx - np->nic_tx + entries - 1) > TX_LIMIT_STOP) {
spin_unlock_irq(&np->lock);
netif_stop_queue(dev);
return NETDEV_TX_BUSY;
}
- np->tx_skbuff[nr] = skb;
-
- if (fragments) {
- dprintk(KERN_DEBUG "%s: nv_start_xmit: buffer contains %d fragments\n", dev->name, fragments);
- /* setup descriptors in reverse order */
- for (i = fragments; i >= 1; i--) {
- skb_frag_t *frag = &skb_shinfo(skb)->frags[i-1];
- np->tx_dma[nr] = pci_map_page(np->pci_dev, frag->page, frag->page_offset, frag->size,
- PCI_DMA_TODEVICE);
+ /* setup the header buffer */
+ do {
+ bcnt = (size > NV_TX2_TSO_MAX_SIZE) ? NV_TX2_TSO_MAX_SIZE : size;
+ nr = (nr + 1) % TX_RING;
+
+ np->tx_dma[nr] = pci_map_single(np->pci_dev, skb->data + offset, bcnt,
+ PCI_DMA_TODEVICE);
+ np->tx_dma_len[nr] = bcnt;
+
+ if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2) {
+ np->tx_ring.orig[nr].PacketBuffer = cpu_to_le32(np->tx_dma[nr]);
+ np->tx_ring.orig[nr].FlagLen = cpu_to_le32((bcnt-1) | tx_flags);
+ } else {
+ np->tx_ring.ex[nr].PacketBufferHigh = cpu_to_le64(np->tx_dma[nr]) >> 32;
+ np->tx_ring.ex[nr].PacketBufferLow = cpu_to_le64(np->tx_dma[nr]) & 0x0FFFFFFFF;
+ np->tx_ring.ex[nr].FlagLen = cpu_to_le32((bcnt-1) | tx_flags);
+ }
+ tx_flags = np->tx_flags;
+ offset += bcnt;
+ size -= bcnt;
+ } while(size);
+
+ /* setup the fragments */
+ for (i = 0; i < fragments; i++) {
+ skb_frag_t *frag = &skb_shinfo(skb)->frags[i];
+ u32 size = frag->size;
+ offset = 0;
+
+ do {
+ bcnt = (size > NV_TX2_TSO_MAX_SIZE) ? NV_TX2_TSO_MAX_SIZE : size;
+ nr = (nr + 1) % TX_RING;
+
+ np->tx_dma[nr] = pci_map_page(np->pci_dev, frag->page, frag->page_offset+offset, bcnt,
+ PCI_DMA_TODEVICE);
+ np->tx_dma_len[nr] = bcnt;
if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2) {
np->tx_ring.orig[nr].PacketBuffer = cpu_to_le32(np->tx_dma[nr]);
- np->tx_ring.orig[nr].FlagLen = cpu_to_le32( (frag->size-1) | np->tx_flags | tx_flags_extra);
+ np->tx_ring.orig[nr].FlagLen = cpu_to_le32((bcnt-1) | tx_flags);
} else {
np->tx_ring.ex[nr].PacketBufferHigh = cpu_to_le64(np->tx_dma[nr]) >> 32;
np->tx_ring.ex[nr].PacketBufferLow = cpu_to_le64(np->tx_dma[nr]) & 0x0FFFFFFFF;
- np->tx_ring.ex[nr].FlagLen = cpu_to_le32( (frag->size-1) | np->tx_flags | tx_flags_extra);
+ np->tx_ring.ex[nr].FlagLen = cpu_to_le32((bcnt-1) | tx_flags);
}
-
- nr = (nr - 1) % TX_RING;
+ offset += bcnt;
+ size -= bcnt;
+ } while (size);
+ }
- if (np->desc_ver == DESC_VER_1)
- tx_flags_extra &= ~NV_TX_LASTPACKET;
- else
- tx_flags_extra &= ~NV_TX2_LASTPACKET;
- }
+ /* set last fragment flag */
+ if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2) {
+ np->tx_ring.orig[nr].FlagLen |= cpu_to_le32(tx_flags_extra);
+ } else {
+ np->tx_ring.ex[nr].FlagLen |= cpu_to_le32(tx_flags_extra);
}
+ np->tx_skbuff[nr] = skb;
+
#ifdef NETIF_F_TSO
if (skb_shinfo(skb)->tso_size)
- tx_flags_extra |= NV_TX2_TSO | (skb_shinfo(skb)->tso_size << NV_TX2_TSO_SHIFT);
+ tx_flags_extra = NV_TX2_TSO | (skb_shinfo(skb)->tso_size << NV_TX2_TSO_SHIFT);
else
#endif
- tx_flags_extra |= (skb->ip_summed == CHECKSUM_HW ? (NV_TX2_CHECKSUM_L3|NV_TX2_CHECKSUM_L4) : 0);
+ tx_flags_extra = (skb->ip_summed == CHECKSUM_HW ? (NV_TX2_CHECKSUM_L3|NV_TX2_CHECKSUM_L4) : 0);
- np->tx_dma[nr] = pci_map_single(np->pci_dev, skb->data, skb->len-skb->data_len,
- PCI_DMA_TODEVICE);
-
+ /* set tx flags */
if (np->desc_ver == DESC_VER_1 || np->desc_ver == DESC_VER_2) {
- np->tx_ring.orig[nr].PacketBuffer = cpu_to_le32(np->tx_dma[nr]);
- np->tx_ring.orig[nr].FlagLen = cpu_to_le32( (skb->len-skb->data_len-1) | np->tx_flags | tx_flags_extra);
+ np->tx_ring.orig[start_nr].FlagLen |= cpu_to_le32(tx_flags | tx_flags_extra);
} else {
- np->tx_ring.ex[nr].PacketBufferHigh = cpu_to_le64(np->tx_dma[nr]) >> 32;
- np->tx_ring.ex[nr].PacketBufferLow = cpu_to_le64(np->tx_dma[nr]) & 0x0FFFFFFFF;
- np->tx_ring.ex[nr].FlagLen = cpu_to_le32( (skb->len-skb->data_len-1) | np->tx_flags | tx_flags_extra);
+ np->tx_ring.ex[start_nr].FlagLen |= cpu_to_le32(tx_flags | tx_flags_extra);
}
- dprintk(KERN_DEBUG "%s: nv_start_xmit: packet packet %d queued for transmission. tx_flags_extra: %x\n",
- dev->name, np->next_tx, tx_flags_extra);
+ dprintk(KERN_DEBUG "%s: nv_start_xmit: packet %d (entries %d) queued for transmission. tx_flags_extra: %x\n",
+ dev->name, np->next_tx, entries, tx_flags_extra);
{
int j;
for (j=0; j<64; j++) {
dprintk("\n");
}
- np->next_tx += 1 + fragments;
+ np->next_tx += entries;
dev->trans_start = jiffies;
spin_unlock_irq(&np->lock);
np->stats.tx_packets++;
np->stats.tx_bytes += skb->len;
}
- nv_release_txskb(dev, i);
}
} else {
if (Flags & NV_TX2_LASTPACKET) {
np->stats.tx_packets++;
np->stats.tx_bytes += skb->len;
}
- nv_release_txskb(dev, i);
}
}
+ nv_release_txskb(dev, i);
np->nic_tx++;
}
if (np->next_tx - np->nic_tx < TX_LIMIT_START)
* the performance.
*/
pci_unmap_single(np->pci_dev, np->rx_dma[i],
- np->rx_skbuff[i]->len,
+ np->rx_skbuff[i]->end-np->rx_skbuff[i]->data,
PCI_DMA_FROMDEVICE);
{
projects / linux-2.6-omap-h63xx.git / blobdiff