Skip to contentSkip to conversation

via-rhine driver: wicked 2005 problem

DISCUSSION Manfred Spraul 3 participants 4 messages Mar 26, 2001 – Mar 29, 2001 Network drivers

lore
Share & tools
lore
#0 lore raw
Quoted 1 line

[Kernel 2.4.2,

the -ac kernels contain a patch that automatically resets the nic if it dies. I've attached my old patch, it applies to 2.4.2.

Quoted 5 lines

I tried the diagnostic utilities from Donald Becker at Scyld.com, but I don't know what I should be looking for. The text output seems ok to me.

Could you post the output?
And a few more lines from your kernel log.

--
Manfred

drivers/net/via-rhine.c +267 -191
@@ -55,6 +55,9 @@
 	LK1.1.6:
 	- Urban Widmark: merges from Beckers 1.08b version (VT6102 + mdio)
 	                 set netif_running_on/off on startup, del_timer_sync
+	
+	LK1.1.7:
+	- Manfred Spraul: added reset into tx_timeout
 */
 
 
@@ -121,13 +124,14 @@
 #include <linux/etherdevice.h>
 #include <linux/skbuff.h>
 #include <linux/init.h>
+#include <linux/delay.h>
 #include <asm/processor.h>		/* Processor type for cache alignment. */
 #include <asm/bitops.h>
 #include <asm/io.h>
 
 /* These identify the driver base version and may not be removed. */
 static char version1[] __devinitdata =
-"via-rhine.c:v1.08b-LK1.1.6  8/9/2000  Written by Donald Becker\n";
+"via-rhine.c:v1.08b-LK1.1.7  8/9/2000  Written by Donald Becker\n";
 static char version2[] __devinitdata =
 "  http://www.scyld.com/network/via-rhine.html\n";
 
@@ -380,6 +384,7 @@
 	CmdNoTxPoll=0x0800, CmdReset=0x8000,
 };
 
+#define MAX_MII_CNT	4
 struct netdev_private {
 	/* Descriptor rings */
 	struct rx_desc *rx_ring;
@@ -421,7 +426,8 @@
 
 	/* MII transceiver section. */
 	u16 advertising;					/* NWay media advertisement */
-	unsigned char phys[2];				/* MII device addresses. */
+	unsigned char phys[MAX_MII_CNT];			/* MII device addresses. */
+	unsigned int mii_cnt;			/* number of MIIs found, but only the first one is used */
 	u16 mii_status;						/* last read MII status */
 };
 
@@ -431,7 +437,6 @@
 static void via_rhine_check_duplex(struct net_device *dev);
 static void via_rhine_timer(unsigned long data);
 static void via_rhine_tx_timeout(struct net_device *dev);
-static void via_rhine_init_ring(struct net_device *dev);
 static int  via_rhine_start_tx(struct sk_buff *skb, struct net_device *dev);
 static void via_rhine_interrupt(int irq, void *dev_instance, struct pt_regs *regs);
 static void via_rhine_tx(struct net_device *dev);
@@ -443,6 +448,25 @@
 static int  via_rhine_close(struct net_device *dev);
 static inline void clear_tally_counters(long ioaddr);
 
+static void wait_for_reset(struct net_device *dev)
+{
+	long ioaddr = dev->base_addr;
+	int i;
+
+	i = 0;
+	do {
+		udelay(5);
+		i++;
+		if(i > 2000) {
+			printk(KERN_ERR "%s: reset did not complete in 10 ms.\n",
+					dev->name);
+			break;
+		}
+	} while(readw(ioaddr + ChipCmd) & CmdReset);
+	if (debug > 1)
+		printk(KERN_INFO "%s: reset finished after %d microseconds.\n",
+				dev->name, 5*i);
+}
 
 static int __devinit via_rhine_init_one (struct pci_dev *pdev,
 					 const struct pci_device_id *ent)
@@ -451,14 +475,11 @@
 	struct netdev_private *np;
 	int i, option;
 	int chip_id = (int) ent->driver_data;
-	int irq = pdev->irq;
 	static int card_idx = -1;
 	static int did_version = 0;
 	long ioaddr;
 	int io_size;
 	int pci_flags;
-	void *ring;
-	dma_addr_t ring_dma;
 	
 	/* print version once and once only */
 	if (! did_version++) {
@@ -471,6 +492,10 @@
 	io_size = via_rhine_chip_info[chip_id].io_size;
 	pci_flags = via_rhine_chip_info[chip_id].pci_flags;
 
+	if (pci_enable_device (pdev))
+		goto err_out;
+
+
 	/* this should always be supported */
 	if (!pci_dma_supported(pdev, 0xffffffff)) {
 		printk(KERN_ERR "32-bit PCI DMA addresses not supported by the card!?\n");
@@ -484,20 +509,7 @@
 		goto err_out;
 	}
 
-	/* allocate pci dma space for rx and tx descriptor rings */
-	ring = pci_alloc_consistent(pdev, 
-				    RX_RING_SIZE * sizeof(struct rx_desc) +
-				    TX_RING_SIZE * sizeof(struct tx_desc),
-				    &ring_dma);
-	if (!ring) {
-		printk(KERN_ERR "Could not allocate DMA memory.\n");
-		goto err_out;
-	}
-
 	ioaddr = pci_resource_start (pdev, pci_flags & PCI_ADDR0 ? 0 : 1);
-
-	if (pci_enable_device (pdev))
-		goto err_out_free_dma;
 	
 	if (pci_flags & PCI_USES_MASTER)
 		pci_set_master (pdev);
@@ -506,7 +518,7 @@
 	if (dev == NULL) {
 		printk (KERN_ERR "init_ethernet failed for card #%d\n",
 			card_idx);
-		goto err_out_free_dma;
+		goto err_out;
 	}
 	SET_MODULE_OWNER(dev);
 	
@@ -545,23 +557,19 @@
 		dev->dev_addr[i] = readb(ioaddr + StationAddr + i);
 	for (i = 0; i < 5; i++)
 			printk("%2.2x:", dev->dev_addr[i]);
-	printk("%2.2x, IRQ %d.\n", dev->dev_addr[i], irq);
+	printk("%2.2x, IRQ %d.\n", dev->dev_addr[i], pdev->irq);
 
 	/* Reset the chip to erase previous misconfiguration. */
 	writew(CmdReset, ioaddr + ChipCmd);
+	wait_for_reset(dev);
 
 	dev->base_addr = ioaddr;
-	dev->irq = irq;
 
 	np = dev->priv;
 	spin_lock_init (&np->lock);
 	np->chip_id = chip_id;
 	np->drv_flags = via_rhine_chip_info[chip_id].drv_flags;
 	np->pdev = pdev;
-	np->rx_ring = ring;
-	np->tx_ring = ring + RX_RING_SIZE * sizeof(struct rx_desc);
-	np->rx_ring_dma = ring_dma;
-	np->tx_ring_dma = ring_dma + RX_RING_SIZE * sizeof(struct rx_desc);
 
 	if (dev->mem_start)
 		option = dev->mem_start;
@@ -593,7 +601,7 @@
 	if (np->drv_flags & CanHaveMII) {
 		int phy, phy_idx = 0;
 		np->phys[0] = 1;		/* Standard for this chip. */
-		for (phy = 1; phy < 32 && phy_idx < 4; phy++) {
+		for (phy = 1; phy < 32 && phy_idx < MAX_MII_CNT; phy++) {
 			int mii_status = mdio_read(dev, phy, 1);
 			if (mii_status != 0xffff  &&  mii_status != 0x0000) {
 				np->phys[phy_idx++] = phy;
@@ -610,6 +618,7 @@
 					netif_carrier_off(dev);
 			}
 		}
+		np->mii_cnt = phy_idx;
 	}
 
 	return 0;
@@ -628,16 +637,202 @@
 err_out_free_netdev:
 	unregister_netdev (dev);
 	kfree (dev);
-err_out_free_dma:
-	pci_free_consistent(pdev, 
+err_out:
+	return -ENODEV;
+}
+
+static int alloc_ring(struct net_device* dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+	void *ring;
+	dma_addr_t ring_dma;
+
+	ring = pci_alloc_consistent(np->pdev, 
+				    RX_RING_SIZE * sizeof(struct rx_desc) +
+				    TX_RING_SIZE * sizeof(struct tx_desc),
+				    &ring_dma);
+	if (!ring) {
+		printk(KERN_ERR "Could not allocate DMA memory.\n");
+		return -ENOMEM;
+	}
+	np->tx_bufs = pci_alloc_consistent(np->pdev, PKT_BUF_SZ * TX_RING_SIZE,
+								   &np->tx_bufs_dma);
+	if (np->tx_bufs == NULL) {
+		pci_free_consistent(np->pdev, 
 			    RX_RING_SIZE * sizeof(struct rx_desc) +
 			    TX_RING_SIZE * sizeof(struct tx_desc),
 			    ring, ring_dma);
-err_out:
-	return -ENODEV;
+		return -ENOMEM;
+	}
+
+	np->rx_ring = ring;
+	np->tx_ring = ring + RX_RING_SIZE * sizeof(struct rx_desc);
+	np->rx_ring_dma = ring_dma;
+	np->tx_ring_dma = ring_dma + RX_RING_SIZE * sizeof(struct rx_desc);
+
+
+	return 0;
+}
+
+void free_ring(struct net_device* dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+
+	pci_free_consistent(np->pdev, 
+			    RX_RING_SIZE * sizeof(struct rx_desc) +
+			    TX_RING_SIZE * sizeof(struct tx_desc),
+			    np->rx_ring, np->rx_ring_dma);
+
+	pci_free_consistent(np->pdev, PKT_BUF_SZ * TX_RING_SIZE,
+						np->tx_bufs, np->tx_bufs_dma);
+
 }
 
+static void alloc_rbufs(struct net_device *dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+	dma_addr_t next;
+	int i;
+
+	np->dirty_rx = np->cur_rx = 0;
+
+	np->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32);
+	np->rx_head_desc = &np->rx_ring[0];
+	next = np->rx_ring_dma;
+	
+	/* Init the ring entries */
+	for (i = 0; i < RX_RING_SIZE; i++) {
+		np->rx_ring[i].rx_status = 0;
+		np->rx_ring[i].desc_length = cpu_to_le32(np->rx_buf_sz);
+		next += sizeof(struct rx_desc);
+		np->rx_ring[i].next_desc = cpu_to_le32(next);
+		np->rx_skbuff[i] = 0;
+	}
+	/* Mark the last entry as wrapping the ring. */
+	np->rx_ring[i-1].next_desc = cpu_to_le32(np->rx_ring_dma);
+
+	/* Fill in the Rx buffers.  Handle allocation failure gracefully. */
+	for (i = 0; i < RX_RING_SIZE; i++) {
+		struct sk_buff *skb = dev_alloc_skb(np->rx_buf_sz);
+		np->rx_skbuff[i] = skb;
+		if (skb == NULL)
+			break;
+		skb->dev = dev;                 /* Mark as being used by this device. */
+
+		np->rx_skbuff_dma[i] =
+			pci_map_single(np->pdev, skb->tail, np->rx_buf_sz,
+						   PCI_DMA_FROMDEVICE);
+
+		np->rx_ring[i].addr = cpu_to_le32(np->rx_skbuff_dma[i]);
+		np->rx_ring[i].rx_status = cpu_to_le32(DescOwn);
+	}
+	np->dirty_rx = (unsigned int)(i - RX_RING_SIZE);
+}
+
+static void free_rbufs(struct net_device* dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+	int i;
+
+	/* Free all the skbuffs in the Rx queue. */
+	for (i = 0; i < RX_RING_SIZE; i++) {
+		np->rx_ring[i].rx_status = 0;
+		np->rx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */
+		if (np->rx_skbuff[i]) {
+			pci_unmap_single(np->pdev,
+							 np->rx_skbuff_dma[i],
+							 np->rx_buf_sz, PCI_DMA_FROMDEVICE);
+			dev_kfree_skb(np->rx_skbuff[i]);
+		}
+		np->rx_skbuff[i] = 0;
+	}
+}
+
+static void alloc_tbufs(struct net_device* dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+	dma_addr_t next;
+	int i;
+
+	np->dirty_tx = np->cur_tx = 0;
+	next = np->tx_ring_dma;
+	for (i = 0; i < TX_RING_SIZE; i++) {
+		np->tx_skbuff[i] = 0;
+		np->tx_ring[i].tx_status = 0;
+		np->tx_ring[i].desc_length = cpu_to_le32(0x00e08000);
+		next += sizeof(struct tx_desc);
+		np->tx_ring[i].next_desc = cpu_to_le32(next);
+		np->tx_buf[i] = &np->tx_bufs[i * PKT_BUF_SZ];
+	}
+	np->tx_ring[i-1].next_desc = cpu_to_le32(np->tx_ring_dma);
+
+}
+
+static void free_tbufs(struct net_device* dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+	int i;
+
+	for (i = 0; i < TX_RING_SIZE; i++) {
+		np->tx_ring[i].tx_status = 0;
+		np->tx_ring[i].desc_length = cpu_to_le32(0x00e08000);
+		np->tx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */
+		if (np->tx_skbuff[i]) {
+			if (np->tx_skbuff_dma[i]) {
+				pci_unmap_single(np->pdev,
+								 np->tx_skbuff_dma[i],
+								 np->tx_skbuff[i]->len, PCI_DMA_TODEVICE);
+			}
+			dev_kfree_skb(np->tx_skbuff[i]);
+		}
+		np->tx_skbuff[i] = 0;
+		np->tx_buf[i] = 0;
+	}
+}
+
+static void init_registers(struct net_device *dev)
+{
+	struct netdev_private *np = (struct netdev_private *)dev->priv;
+	long ioaddr = dev->base_addr;
+	int i;
+
+	for (i = 0; i < 6; i++)
+		writeb(dev->dev_addr[i], ioaddr + StationAddr + i);
+
+	/* Initialize other registers. */
+	writew(0x0006, ioaddr + PCIBusConfig);	/* Tune configuration??? */
+	/* Configure the FIFO thresholds. */
+	writeb(0x20, ioaddr + TxConfig);	/* Initial threshold 32 bytes */
+	np->tx_thresh = 0x20;
+	np->rx_thresh = 0x60;			/* Written in via_rhine_set_rx_mode(). */
+
+	if (dev->if_port == 0)
+		dev->if_port = np->default_port;
+
+	writel(np->rx_ring_dma, ioaddr + RxRingPtr);
+	writel(np->tx_ring_dma, ioaddr + TxRingPtr);
+
+	via_rhine_set_rx_mode(dev);
+
+	/* Enable interrupts by setting the interrupt mask. */
+	writew(IntrRxDone | IntrRxErr | IntrRxEmpty| IntrRxOverflow| IntrRxDropped|
+		   IntrTxDone | IntrTxAbort | IntrTxUnderrun |
+		   IntrPCIErr | IntrStatsMax | IntrLinkChange | IntrMIIChange,
+		   ioaddr + IntrEnable);
+
+	np->chip_cmd = CmdStart|CmdTxOn|CmdRxOn|CmdNoTxPoll;
+	if (np->duplex_lock)
+		np->chip_cmd |= CmdFDuplex;
+	writew(np->chip_cmd, ioaddr + ChipCmd);
+
+	via_rhine_check_duplex(dev);
 
+	/* The LED outputs of various MII xcvrs should be configured.  */
+	/* For NS or Mison phys, turn on bit 1 in register 0x17 */
+	/* For ESI phys, turn on bit 7 in register 0x17. */
+	mdio_write(dev, np->phys[0], 0x17, mdio_read(dev, np->phys[0], 0x17) |
+			   (np->drv_flags & HasESIPhy) ? 0x0080 : 0x0001);
+}
 /* Read and write over the MII Management Data I/O (MDIO) interface. */
 
 static int mdio_read(struct net_device *dev, int phy_id, int regnum)
@@ -698,68 +893,29 @@
 	/* Reset the chip. */
 	writew(CmdReset, ioaddr + ChipCmd);
 
-	i = request_irq(dev->irq, &via_rhine_interrupt, SA_SHIRQ, dev->name, dev);
+	i = request_irq(np->pdev->irq, &via_rhine_interrupt, SA_SHIRQ, dev->name, dev);
 	if (i)
 		return i;
 
 	if (debug > 1)
 		printk(KERN_DEBUG "%s: via_rhine_open() irq %d.\n",
-			   dev->name, dev->irq);
-
-	np->tx_bufs = pci_alloc_consistent(np->pdev, PKT_BUF_SZ * TX_RING_SIZE,
-									   &np->tx_bufs_dma);
-	if (np->tx_bufs == NULL) {
-		free_irq(dev->irq, dev);
-		return -ENOMEM;
-	}
-
-	via_rhine_init_ring(dev);
-
-	writel(np->rx_ring_dma, ioaddr + RxRingPtr);
-	writel(np->tx_ring_dma, ioaddr + TxRingPtr);
-
-	for (i = 0; i < 6; i++)
-		writeb(dev->dev_addr[i], ioaddr + StationAddr + i);
-
-	/* Initialize other registers. */
-	writew(0x0006, ioaddr + PCIBusConfig);	/* Tune configuration??? */
-	/* Configure the FIFO thresholds. */
-	writeb(0x20, ioaddr + TxConfig);	/* Initial threshold 32 bytes */
-	np->tx_thresh = 0x20;
-	np->rx_thresh = 0x60;			/* Written in via_rhine_set_rx_mode(). */
-
-	if (dev->if_port == 0)
-		dev->if_port = np->default_port;
-
-	netif_start_queue(dev);
-
-	via_rhine_set_rx_mode(dev);
-
-	/* Enable interrupts by setting the interrupt mask. */
-	writew(IntrRxDone | IntrRxErr | IntrRxEmpty| IntrRxOverflow| IntrRxDropped|
-		   IntrTxDone | IntrTxAbort | IntrTxUnderrun |
-		   IntrPCIErr | IntrStatsMax | IntrLinkChange | IntrMIIChange,
-		   ioaddr + IntrEnable);
-
-	np->chip_cmd = CmdStart|CmdTxOn|CmdRxOn|CmdNoTxPoll;
-	if (np->duplex_lock)
-		np->chip_cmd |= CmdFDuplex;
-	writew(np->chip_cmd, ioaddr + ChipCmd);
-
-	via_rhine_check_duplex(dev);
-
-	/* The LED outputs of various MII xcvrs should be configured.  */
-	/* For NS or Mison phys, turn on bit 1 in register 0x17 */
-	/* For ESI phys, turn on bit 7 in register 0x17. */
-	mdio_write(dev, np->phys[0], 0x17, mdio_read(dev, np->phys[0], 0x17) |
-			   (np->drv_flags & HasESIPhy) ? 0x0080 : 0x0001);
-
+			   dev->name, np->pdev->irq);
+	
+	i = alloc_ring(dev);
+	if (i)
+		return i;
+	alloc_rbufs(dev);
+	alloc_tbufs(dev);
+	wait_for_reset(dev);
+	init_registers(dev);
 	if (debug > 2)
 		printk(KERN_DEBUG "%s: Done via_rhine_open(), status %4.4x "
 			   "MII status: %4.4x.\n",
 			   dev->name, readw(ioaddr + ChipCmd),
 			   mdio_read(dev, np->phys[0], 1));
 
+	netif_start_queue(dev);
+
 	/* Set the timer to check for link beat. */
 	init_timer(&np->timer);
 	np->timer.expires = jiffies + 2;
@@ -835,84 +991,37 @@
 	struct netdev_private *np = (struct netdev_private *) dev->priv;
 	long ioaddr = dev->base_addr;
 
-	/* Lock to protect mdio_read and access to stats. A friendly
-       advice to the implementor of the XXXs in this function is to be
-       sure not to spin too long (whatever that means :) */
-	spin_lock_irq (&np->lock);
-
 	printk (KERN_WARNING "%s: Transmit timed out, status %4.4x, PHY status "
 		"%4.4x, resetting...\n",
 		dev->name, readw (ioaddr + IntrStatus),
 		mdio_read (dev, np->phys[0], 1));
 
-	/* XXX Perhaps we should reinitialize the hardware here. */
 	dev->if_port = 0;
 
-	/* Stop and restart the chip's Tx processes . */
-	/* XXX to do */
-
-	/* Trigger an immediate transmit demand. */
-	/* XXX to do */
+	/* protect against concurrent rx interrupts */
+	disable_irq(np->pdev->irq);
 
-	dev->trans_start = jiffies;
-	np->stats.tx_errors++;
-
-	spin_unlock_irq (&np->lock);
-}
+	spin_lock(&np->lock);
 
+	/* Reset the chip. */
+	writew(CmdReset, ioaddr + ChipCmd);
 
-/* Initialize the Rx and Tx rings, along with various 'dev' bits. */
-static void via_rhine_init_ring(struct net_device *dev)
-{
-	struct netdev_private *np = (struct netdev_private *)dev->priv;
-	int i;
-	dma_addr_t next = np->rx_ring_dma;
-
-	np->cur_rx = np->cur_tx = 0;
-	np->dirty_rx = np->dirty_tx = 0;
-
-	np->rx_buf_sz = (dev->mtu <= 1500 ? PKT_BUF_SZ : dev->mtu + 32);
-	np->rx_head_desc = &np->rx_ring[0];
-
-	for (i = 0; i < RX_RING_SIZE; i++) {
-		np->rx_ring[i].rx_status = 0;
-		np->rx_ring[i].desc_length = cpu_to_le32(np->rx_buf_sz);
-		next += sizeof(struct rx_desc);
-		np->rx_ring[i].next_desc = cpu_to_le32(next);
-		np->rx_skbuff[i] = 0;
-	}
-	/* Mark the last entry as wrapping the ring. */
-	np->rx_ring[i-1].next_desc = cpu_to_le32(np->rx_ring_dma);
-
-	/* Fill in the Rx buffers.  Handle allocation failure gracefully. */
-	for (i = 0; i < RX_RING_SIZE; i++) {
-		struct sk_buff *skb = dev_alloc_skb(np->rx_buf_sz);
-		np->rx_skbuff[i] = skb;
-		if (skb == NULL)
-			break;
-		skb->dev = dev;                 /* Mark as being used by this device. */
-
-		np->rx_skbuff_dma[i] =
-			pci_map_single(np->pdev, skb->tail, np->rx_buf_sz,
-						   PCI_DMA_FROMDEVICE);
-
-		np->rx_ring[i].addr = cpu_to_le32(np->rx_skbuff_dma[i]);
-		np->rx_ring[i].rx_status = cpu_to_le32(DescOwn);
-	}
-	np->dirty_rx = (unsigned int)(i - RX_RING_SIZE);
-
-	next = np->tx_ring_dma;
-	for (i = 0; i < TX_RING_SIZE; i++) {
-		np->tx_skbuff[i] = 0;
-		np->tx_ring[i].tx_status = 0;
-		np->tx_ring[i].desc_length = cpu_to_le32(0x00e08000);
-		next += sizeof(struct tx_desc);
-		np->tx_ring[i].next_desc = cpu_to_le32(next);
-		np->tx_buf[i] = &np->tx_bufs[i * PKT_BUF_SZ];
-	}
-	np->tx_ring[i-1].next_desc = cpu_to_le32(np->tx_ring_dma);
+	/* clear all descriptors */
+	free_tbufs(dev);
+	free_rbufs(dev);
+	alloc_tbufs(dev);
+	alloc_rbufs(dev);
+
+	/* Reinitialize the hardware. */
+	wait_for_reset(dev);
+	init_registers(dev);
+	
+	spin_unlock(&np->lock);
+	enable_irq(np->pdev->irq);
 
-	return;
+	dev->trans_start = jiffies;
+	np->stats.tx_errors++;
+	netif_wake_queue(dev);
 }
 
 static int via_rhine_start_tx(struct sk_buff *skb, struct net_device *dev)
@@ -1215,8 +1324,9 @@
 				   "threshold setting to %2.2x.\n", dev->name, np->tx_thresh);
 	}
 	if ((intr_status & ~( IntrLinkChange | IntrStatsMax |
-						  IntrTxAbort | IntrTxAborted)) && debug > 1) {
-		printk(KERN_ERR "%s: Something Wicked happened! %4.4x.\n",
+						  IntrTxAbort | IntrTxAborted))) {
+		if (debug > 1)
+			printk(KERN_ERR "%s: Something Wicked happened! %4.4x.\n",
 			   dev->name, intr_status);
 		/* Recovery for other fault sources not known. */
 		writew(CmdTxDemand | np->chip_cmd, dev->base_addr + ChipCmd);
@@ -1339,12 +1449,10 @@
 {
 	long ioaddr = dev->base_addr;
 	struct netdev_private *np = (struct netdev_private *)dev->priv;
-	int i;
-	unsigned long flags;
 
 	del_timer_sync(&np->timer);
 
-	spin_lock_irqsave(&np->lock, flags);
+	spin_lock_irq(&np->lock);
 
 	netif_stop_queue(dev);
 
@@ -1361,44 +1469,12 @@
 	/* Stop the chip's Tx and Rx processes. */
 	writew(CmdStop, ioaddr + ChipCmd);
 
-	spin_unlock_irqrestore(&np->lock, flags);
-
-	/* Make sure there is no irq-handler running on a different CPU. */
-	synchronize_irq();
-
-	free_irq(dev->irq, dev);
+	spin_unlock_irq(&np->lock);
 
-	/* Free all the skbuffs in the Rx queue. */
-	for (i = 0; i < RX_RING_SIZE; i++) {
-		np->rx_ring[i].rx_status = 0;
-		np->rx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */
-		if (np->rx_skbuff[i]) {
-			pci_unmap_single(np->pdev,
-							 np->rx_skbuff_dma[i],
-							 np->rx_buf_sz, PCI_DMA_FROMDEVICE);
-			dev_kfree_skb(np->rx_skbuff[i]);
-		}
-		np->rx_skbuff[i] = 0;
-	}
-
-	/* Free all the skbuffs in the Tx queue, and also any bounce buffers. */
-	for (i = 0; i < TX_RING_SIZE; i++) {
-		np->tx_ring[i].tx_status = 0;
-		np->tx_ring[i].desc_length = cpu_to_le32(0x00e08000);
-		np->tx_ring[i].addr = cpu_to_le32(0xBADF00D0); /* An invalid address. */
-		if (np->tx_skbuff[i]) {
-			if (np->tx_skbuff_dma[i]) {
-				pci_unmap_single(np->pdev,
-								 np->tx_skbuff_dma[i],
-								 np->tx_skbuff[i]->len, PCI_DMA_TODEVICE);
-			}
-			dev_kfree_skb(np->tx_skbuff[i]);
-		}
-		np->tx_skbuff[i] = 0;
-		np->tx_buf[i] = 0;
-	}
-	pci_free_consistent(np->pdev, PKT_BUF_SZ * TX_RING_SIZE,
-						np->tx_bufs, np->tx_bufs_dma);
+	free_irq(np->pdev->irq, dev);
+	free_rbufs(dev);
+	free_tbufs(dev);
+	free_ring(dev);
 
 	return 0;
 }
#1 lore raw
Quoted 4 lines · Manfred Spraul
Quoted 1 line

[Kernel 2.4.2,

the -ac kernels contain a patch that automatically resets the nic if it dies. I've attached my old patch, it applies to 2.4.2.

Your patch works fine for resetting the NIC. There are some a one or two log messages coming up, but the transfer continues after a small delay.

It doesn't solve the (less urgent) problem of not being able the use the NIC after a warm boot in M$ Windows. As I said, pulling the power cord from the ATX power supply and reinserting it, makes it go away.

Quoted 6 lines
Quoted 4 lines

I tried the diagnostic utilities from Donald Becker at Scyld.com, but I don't know what I should be looking for. The text output seems ok to me.

Could you post the output?
And a few more lines from your kernel log.

It's rather much to post the the mailing list, I think, so I've put it online at [http://win-www.uia.ac.be/u/s965817/via-rhine.report/]. If you prefer otherwise, just tell me. I can always send a tarball of the logs.

Some explanation of the logs:

report: script for creating the device status dumps
hang/: about the "timed out" problem receiving much data
242/: using the default module from 2.4.2 release
before/: device status before the lock, while NIC works well
after/: device status after the lock, during timeout messages
242patch/: using the patched version
before/: device status before the timeout or wicked messages
after/: device status afterwards (NIC keeps working fine)
windoze:/ about the warm-boot-after-windoze problem
winboot/: device info booting after warm boot (problem case) coldboot/: device info booting after total cold boot (no problem)

Thanks a lot for the patch.

Tung

#2 lore raw
In reply to Manfred Spraul
Quoted 3 lines · wing tung Leung

It doesn't solve the (less urgent) problem of not being able the use the NIC after a warm boot in M$ Windows. As I said, pulling the power cord from the ATX power supply and reinserting it, makes it go away.

Would it be possible for you to re-run your tests against kernel 2.4.3-pre8? (ftp://ftp.us.kernel.org/pub/linux/kernel/testing/)

This is the "official" latest version of the via-rhine driver, and it includes Manfred's patch, as well as a pci_enable_device movement might solve your problem.

If the problem is still not solved, could you download via-diag.c and
libmii.c from ftp://www.scyld.com/pub/diag/ Compile instructions are
at the bottom of via-diag.c. I'm interested in seeing two via-diag register snapshots, one from a cold boot (where it is working), and one from a warm boot.

./via-diag -maaavvveef > via-diag-cold.txt
and
./via-diag -maaavvveef > via-diag-warm.txt
then
diff -u v*cold.txt v*warm.txt | send mail...

And to see if the PCI configuration registers change between warm boot and cold boot, run lspci from pciutils:

lspci -vvvxxx > lspci-cold.txt
and
lspci -vvvxxx > lspci-warm.txt
then
diff -u l*cold.txt l*warm.txt | send mail...

Signature · 3 lines
Jeff Garzik       | May you have warm words on a cold evening,
Building 1024     | a full moon on a dark night,
MandrakeSoft      | and a smooth road all the way to your door.
#3 lore raw
In reply to Manfred Spraul
Quoted 13 lines · Jeff Garzik

If the problem is still not solved, could you download via-diag.c and
libmii.c from ftp://www.scyld.com/pub/diag/ Compile instructions are
at the bottom of via-diag.c. I'm interested in seeing two via-diag register snapshots, one from a cold boot (where it is working), and one from a warm boot.

./via-diag -maaavvveef > via-diag-cold.txt
and
./via-diag -maaavvveef > via-diag-warm.txt
then
diff -u v*cold.txt v*warm.txt | send mail...

Quoted 8 lines

And to see if the PCI configuration registers change between warm boot and cold boot, run lspci from pciutils:

lspci -vvvxxx > lspci-cold.txt
and
lspci -vvvxxx > lspci-warm.txt
then
diff -u l*cold.txt l*warm.txt | send mail...

I still have the boot problem, these are the diffs you asked for. I hope they are useful. (using 2.4.3-pre8)

I had to include the driver in the kernel, and could not load it as module, I don't know why.

==
[root@twisted /root]# modprobe via-rhine
/lib/modules/2.4.3-pre8/kernel/drivers/net/via-rhine.o: unresolved symbol alloc_etherdev /lib/modules/2.4.3-pre8/kernel/drivers/net/via-rhine.o: unresolved symbol unregister_netdev /lib/modules/2.4.3-pre8/kernel/drivers/net/via-rhine.o: unresolved symbol register_netdev /lib/modules/2.4.3-pre8/kernel/drivers/net/via-rhine.o: insmod /lib/modules/2.4.3-pre8/kernel/drivers/net/via-rhine.o failed /lib/modules/2.4.3-pre8/kernel/drivers/net/via-rhine.o: insmod via-rhine failed
==

lspci-warm.txt renamed from lspci-cold.txt +2 -2
@@ -65,7 +65,7 @@
 10: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 20: 00 00 00 00 00 00 00 00 00 00 00 00 06 11 00 00
 30: 00 00 00 00 c0 00 00 00 00 00 00 00 00 00 00 00
-40: 08 01 00 00 00 80 60 e6 05 01 84 00 00 00 f0 f3
+40: 08 01 00 00 00 80 60 ee 05 01 84 00 00 00 f0 f3
 50: 00 00 04 00 00 a0 0b f0 00 06 ff 08 50 00 00 00
 60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 70: 00 00 00 00 05 1e 00 02 00 00 f0 40 00 00 00 00
@@ -115,7 +115,7 @@
 20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 30: 00 00 00 00 68 00 00 00 00 00 00 00 00 00 00 00
 40: 20 84 5c 00 ba 70 00 00 01 40 00 00 d8 10 00 00
-50: 00 fe ff 88 50 04 00 00 00 ff ff 00 00 00 00 00
+50: 00 ff ff 88 50 04 00 00 00 ff ff 00 00 00 00 00
 60: 00 00 00 00 00 00 00 00 01 00 02 00 00 00 00 00
 70: 01 60 00 00 01 00 00 00 00 00 00 00 00 00 00 00
 80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
via-diag-warm.txt renamed from via-diag-cold.txt +5 -10
@@ -1,24 +1,19 @@
 via-diag.c:v2.04 7/14/2000 Donald Becker ([email protected])
  http://www.scyld.com/diag/index.html
 Index #1: Found a VIA VT3065 Rhine-II adapter at 0xec00.
- Station address 00:50:ba:1e:91:1b.
+ Station address 00:00:00:00:00:00.
  Tx disabled, Rx disabled, half-duplex (0x0804).
   Receive  mode is 0x00: Unknown/invalid.
   Transmit mode is 0x00: Normal transmit, 128 byte threshold.
 VIA VT3065 Rhine-II chip registers at 0xec00
- 0x000: 1eba5000 00001b91 00000804 00000000 00000000 00000000 00000000 00000000
+ 0x000: 00000000 00000000 00000804 00000000 00000000 00000000 00000000 00000000
  0x020: 00000400 00000000 00000000 00000000 00000000 00000000 00000000 00000000
  0x040: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 ffd7dffa
- 0x060: 00000000 00000000 00000000 0e091108 ffff9f00 08000080 02470000 00000000
+ 0x060: 00000000 00000000 00000000 0e09131f 00008100 08000080 02470000 00000000
  No interrupt sources are pending (0000).
   Access to the EEPROM has been disabled (0x80).
     Direct reading or writing is not possible.
 EEPROM contents (Assumed from chip registers):
-0x100:  00 50 ba 1e 91 1b 00 00 00 00 00 00 00 00 00 00
+0x100:  00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 0x110:  00 00 00 00 00 00 00 00 09 0e 00 00 47 02 73 73
- mdio_read(0xec00, 0, 1).. mdio_read(0xec00, 1, 1).. mdio_read(0xec00, 2, 1).. mdio_read(0xec00, 3, 1).. mdio_read(0xec00, 4, 1).. mdio_read(0xec00, 5, 1).. mdio_read(0xec00, 6, 1).. mdio_read(0xec00, 7, 1).. mdio_read(0xec00, 8, 1).. MII PHY found at address 8, status 0x782d.
- mdio_read(0xec00, 9, 1).. mdio_read(0xec00, 10, 1).. mdio_read(0xec00, 11, 1).. mdio_read(0xec00, 12, 1).. mdio_read(0xec00, 13, 1).. mdio_read(0xec00, 14, 1).. mdio_read(0xec00, 15, 1).. mdio_read(0xec00, 16, 1).. mdio_read(0xec00, 17, 1).. mdio_read(0xec00, 18, 1).. mdio_read(0xec00, 19, 1).. mdio_read(0xec00, 20, 1).. mdio_read(0xec00, 21, 1).. mdio_read(0xec00, 22, 1).. mdio_read(0xec00, 23, 1).. mdio_read(0xec00, 24, 1).. mdio_read(0xec00, 25, 1).. mdio_read(0xec00, 26, 1).. mdio_read(0xec00, 27, 1).. mdio_read(0xec00, 28, 1).. mdio_read(0xec00, 29, 1).. mdio_read(0xec00, 30, 1).. mdio_read(0xec00, 31, 1).. MII PHY #8 transceiver registers: mdio_read(0xec00, 8, 0)..
-   3000 mdio_read(0xec00, 8, 1).. 782d mdio_read(0xec00, 8, 2).. 0016 mdio_read(0xec00, 8, 3).. f880 mdio_read(0xec00, 8, 4).. 01e1 mdio_read(0xec00, 8, 5).. 0021 mdio_read(0xec00, 8, 6).. ffff mdio_read(0xec00, 8, 7).. ffff mdio_read(0xec00, 8, 8)..
-   ffff mdio_read(0xec00, 8, 9).. ffff mdio_read(0xec00, 8, 10).. ffff mdio_read(0xec00, 8, 11).. ffff mdio_read(0xec00, 8, 12).. ffff mdio_read(0xec00, 8, 13).. ffff mdio_read(0xec00, 8, 14).. ffff mdio_read(0xec00, 8, 15).. ffff mdio_read(0xec00, 8, 16)..
-   0022 mdio_read(0xec00, 8, 17).. ff40 mdio_read(0xec00, 8, 18).. 0000 mdio_read(0xec00, 8, 19).. ffc0 mdio_read(0xec00, 8, 20).. 00a0 mdio_read(0xec00, 8, 21).. ffff mdio_read(0xec00, 8, 22).. ffff mdio_read(0xec00, 8, 23).. ffff mdio_read(0xec00, 8, 24)..
-   ffff mdio_read(0xec00, 8, 25).. ffff mdio_read(0xec00, 8, 26).. ffff mdio_read(0xec00, 8, 27).. ffff mdio_read(0xec00, 8, 28).. ffff mdio_read(0xec00, 8, 29).. ffff mdio_read(0xec00, 8, 30).. ffff mdio_read(0xec00, 8, 31).. ffff.
+ mdio_read(0xec00, 0, 1).. mdio_read(0xec00, 1, 1).. mdio_read(0xec00, 2, 1).. mdio_read(0xec00, 3, 1).. mdio_read(0xec00, 4, 1).. mdio_read(0xec00, 5, 1).. mdio_read(0xec00, 6, 1).. mdio_read(0xec00, 7, 1).. mdio_read(0xec00, 8, 1).. mdio_read(0xec00, 9, 1).. mdio_read(0xec00, 10, 1).. mdio_read(0xec00, 11, 1).. mdio_read(0xec00, 12, 1).. mdio_read(0xec00, 13, 1).. mdio_read(0xec00, 14, 1).. mdio_read(0xec00, 15, 1).. mdio_read(0xec00, 16, 1).. mdio_read(0xec00, 17, 1).. mdio_read(0xec00, 18, 1).. mdio_read(0xec00, 19, 1).. mdio_read(0xec00, 20, 1).. mdio_read(0xec00, 21, 1).. mdio_read(0xec00, 22, 1).. mdio_read(0xec00, 23, 1).. mdio_read(0xec00, 24, 1).. mdio_read(0xec00, 25, 1).. mdio_read(0xec00, 26, 1).. mdio_read(0xec00, 27, 1).. mdio_read(0xec00, 28, 1).. mdio_read(0xec00, 29, 1).. mdio_read(0xec00, 30, 1).. mdio_read(0xec00, 31, 1).. ***WARNING***: No MII transceivers found!