PATCH: Fibre attached pcnet/32

PATCH: Fibre attached pcnet/32

[Alan Cox]

The current driver does workarounds for errata that do not work with
fibre attached devices. This patch avoids doing the workaround on the
only known fibre attach pcnet/32 hardware. All handling is automated on
pci sub-ids

Patch by: Guido Guenther
Signed-off-by: Alan Cox <alan@redhat.com>

diff -u --new-file --recursive --exclude-from /usr/src/exclude linux.vanilla-2.6.11/drivers/net/pcnet32.c linux-2.6.11/drivers/net/pcnet32.c
--- linux.vanilla-2.6.11/drivers/net/pcnet32.c	2005-03-05 15:15:10.000000000 +0000
+++ linux-2.6.11/drivers/net/pcnet32.c	2005-03-05 16:12:01.000000000 +0000
@@ -1429,26 +1429,32 @@
 	val |= 0x10;
     lp->a.write_csr (ioaddr, 124, val);
 
-    /* 24 Jun 2004 according AMD, in order to change the PHY,
-     * DANAS (or DISPM for 79C976) must be set; then select the speed,
-     * duplex, and/or enable auto negotiation, and clear DANAS */
-    if (lp->mii && !(lp->options & PCNET32_PORT_ASEL)) {
-	lp->a.write_bcr(ioaddr, 32, lp->a.read_bcr(ioaddr, 32) | 0x0080);
-	/* disable Auto Negotiation, set 10Mpbs, HD */
-	val = lp->a.read_bcr(ioaddr, 32) & ~0xb8;
-	if (lp->options & PCNET32_PORT_FD)
-	    val |= 0x10;
-	if (lp->options & PCNET32_PORT_100)
-	    val |= 0x08;
-	lp->a.write_bcr (ioaddr, 32, val);
+    /* Skip PHY selection on AT2701FX, looses link otherwise */
+    if(lp->pci_dev->subsystem_vendor == PCI_VENDOR_ID_AT && 
+       lp->pci_dev->subsystem_device == PCI_SUBDEVICE_ID_AT_2701FX ) {
+    	printk(KERN_DEBUG "pcnet32: Skipping PHY selection.\n");
     } else {
-	if (lp->options & PCNET32_PORT_ASEL) {
-	    lp->a.write_bcr(ioaddr, 32, lp->a.read_bcr(ioaddr, 32) | 0x0080);
-	    /* enable auto negotiate, setup, disable fd */
-	    val = lp->a.read_bcr(ioaddr, 32) & ~0x98;
-	    val |= 0x20;
-	    lp->a.write_bcr(ioaddr, 32, val);
-	}
+        /* 24 Jun 2004 according AMD, in order to change the PHY,
+         * DANAS (or DISPM for 79C976) must be set; then select the speed,
+         * duplex, and/or enable auto negotiation, and clear DANAS */
+        if (lp->mii && !(lp->options & PCNET32_PORT_ASEL)) {
+    	lp->a.write_bcr(ioaddr, 32, lp->a.read_bcr(ioaddr, 32) | 0x0080);
+    	/* disable Auto Negotiation, set 10Mpbs, HD */
+    	val = lp->a.read_bcr(ioaddr, 32) & ~0xb8;
+    	if (lp->options & PCNET32_PORT_FD)
+    	    val |= 0x10;
+    	if (lp->options & PCNET32_PORT_100)
+    	    val |= 0x08;
+    	lp->a.write_bcr (ioaddr, 32, val);
+        } else {
+    	    if (lp->options & PCNET32_PORT_ASEL) {
+    	        lp->a.write_bcr(ioaddr, 32, lp->a.read_bcr(ioaddr, 32) | 0x0080);
+    	        /* enable auto negotiate, setup, disable fd */
+    	        val = lp->a.read_bcr(ioaddr, 32) & ~0x98;
+    	        val |= 0x20;
+    	        lp->a.write_bcr(ioaddr, 32, val);
+    	    }
+        }
     }
 
 #ifdef DO_DXSUFLO

Re: PATCH: Fibre attached pcnet/32

[Jeff Garzik]

Alan Cox wrote:
> The current driver does workarounds for errata that do not work with
> fibre attached devices. This patch avoids doing the workaround on the
> only known fibre attach pcnet/32 hardware. All handling is automated on
> pci sub-ids
> 
> Patch by: Guido Guenther
> Signed-off-by: Alan Cox <alan@redhat.com>


Linus already has this:

ChangeSet@1.1966.107.1, 2005-01-27 15:55:00-05:00, brazilnut@us.ibm.com
   [PATCH] pcnet32: 79c976 with fiber optic fix

   After testing this patch I agree that it should be applied.  The one
   change I made was to print the device name (ethN) instead of 'pcnet32'.
   Tested ia32.

   From: Guido Guenther <agx@sigxcpu.org>,
         Lars Munch <lars@segv.dk>

   Skip PHY selection on Allied Telesyn 2701FX, it looses the link 
otherwise.
   Fix up the AT 2700FX as well.

   Signed-Off-By: Guido Guenther <agx@sigxcpu.org>
   Signed-off-by: Andrew Morton <akpm@osdl.org>
   signed-off-by: Don Fry <brazilnut@us.ibm.com>
   Signed-off-by: Jeff Garzik <jgarzik@pobox.com>