Re: [PATCH] Gigabit Ethernet support for forcedeth

[Manfred Spraul <manfred@colorfullife.com>]

[-- Attachment #1: Type: text/plain, Size: 576 bytes --]

Manfred Spraul wrote:

>>
>> your driver needs to call pci_set_dma_mask() and 
>> pci_set_consistent_dma_mask().
>>
> I'll add that.
>
I've checked the pci layer: 32-bit is the default, thus no calls are 
required. What's missing is error handling for pci_map, I'll add that later.

Attached is an incremental patch that fixes the simple issues you 
mentioned. Could you merge both patches together to your tree? They 
improve the driver significantly and I want some test input from users.

I've added message flags and pci_map error handling to my todo list.
--
    Manfred

[-- Attachment #2: patch-forced-candidate-inc --]
[-- Type: text/plain, Size: 2436 bytes --]

// $Header$
// Kernel Version:
//  VERSION = 2
//  PATCHLEVEL = 6
//  SUBLEVEL = 7
//  EXTRAVERSION = -mm2
--- 2.6/drivers/net/forcedeth.c	2004-06-30 18:45:23.000000000 +0200
+++ build-2.6/drivers/net/forcedeth.c	2004-06-30 18:45:05.000000000 +0200
@@ -348,18 +348,22 @@ struct ring_desc {
 
 /* General driver defaults */
 #define NV_WATCHDOG_TIMEO	(5*HZ)
-#define DEFAULT_MTU		1500	/* also maximum supported, at least for now */
 
 #define RX_RING		128
 #define TX_RING		64
-/* limited to 1 packet until we understand NV_TX_LASTPACKET */
+/* 
+ * 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
 
 /* rx/tx mac addr + type + vlan + align + slack*/
-#define RX_NIC_BUFSIZE		(DEFAULT_MTU + 64)
+#define RX_NIC_BUFSIZE		(ETH_DATA_LEN + 64)
 /* even more slack */
-#define RX_ALLOC_BUFSIZE	(DEFAULT_MTU + 128)
+#define RX_ALLOC_BUFSIZE	(ETH_DATA_LEN + 128)
 
 #define OOM_REFILL	(1+HZ/20)
 #define POLL_WAIT	(1+HZ/100)
@@ -988,7 +992,7 @@ static void nv_tx_done(struct net_device
 	while (np->nic_tx != np->next_tx) {
 		i = np->nic_tx % TX_RING;
 
-		Flags = cpu_to_le32(np->tx_ring[i].FlagLen);
+		Flags = le32_to_cpu(np->tx_ring[i].FlagLen);
 
 		dprintk(KERN_DEBUG "%s: nv_tx_done: looking at packet %d, Flags 0x%x.\n",
 					dev->name, np->nic_tx, Flags);
@@ -1077,7 +1081,7 @@ static void nv_rx_process(struct net_dev
 			break;	/* we scanned the whole ring - do not continue */
 
 		i = np->cur_rx % RX_RING;
-		Flags = cpu_to_le32(np->rx_ring[i].FlagLen);
+		Flags = le32_to_cpu(np->rx_ring[i].FlagLen);
 		len = nv_descr_getlength(&np->rx_ring[i], np->desc_ver);
 
 		dprintk(KERN_DEBUG "%s: nv_rx_process: looking at packet %d, Flags 0x%x.\n",
@@ -1193,7 +1197,7 @@ next_pkt:
  */
 static int nv_change_mtu(struct net_device *dev, int new_mtu)
 {
-	if (new_mtu > DEFAULT_MTU)
+	if (new_mtu > ETH_DATA_LEN)
 		return -EINVAL;
 	dev->mtu = new_mtu;
 	return 0;
@@ -1999,7 +2003,7 @@ static void __exit exit_nic(void)
 	pci_unregister_driver(&driver);
 }
 
-MODULE_PARM(max_interrupt_work, "i");
+module_param(max_interrupt_work, int, 0);
 MODULE_PARM_DESC(max_interrupt_work, "forcedeth maximum events handled per interrupt");
 
 MODULE_AUTHOR("Manfred Spraul <manfred@colorfullife.com>");