Re: [patch] GNU Mach-1.x support for 3c556 NIC

[Michael Banck]

On Tue, Feb 17, 2004 at 12:06:12PM +0100, Alfred M. Szmidt wrote:
>    Index: linux/src/drivers/net/3c59x.c
>    ===================================================================
>    RCS file: /cvsroot/hurd/gnumach/linux/src/drivers/net/Attic/3c59x.c,v
>    retrieving revision 1.3
>    diff -u -r1.3 3c59x.c
>    --- linux/src/drivers/net/3c59x.c       8 Oct 1999 13:50:16 -0000       1.3
>    +++ linux/src/drivers/net/3c59x.c       15 Feb 2004 20:20:05 -0000
>    @@ -893,7 +899,11 @@
>                  printk("%s: CardBus functions mapped %8.8x->%p (PCMCIA 
> committee "
>                             " brain-damage).\n", dev->name, fn_st_addr, 
> vp->cb_fn _base);
>                  EL3WINDOW(2);
>    -               outw(0x10 | inw(ioaddr + Wn2_ResetOptions), ioaddr + 
> Wn2_ResetOp tions);
>    +               if (pci_tbl[chip_idx].device_id == 0x6055) {
>    +                       outw(0x4010 | inw(ioaddr + Wn2_ResetOptions), 
> ioaddr + W n2_ResetOptions);
> 
> Just curious, shouldn't the space in "W n2_ResetOptions" be removed?
> Or is this some artifact from your mail client?

Good catch, that space (and a couple of others, 'cb_fn _base' for
example) is definetely bogus and not in my tree or the cvs diff output.

I'll append the patch again, having checked it looks good in my editor
and the exim mailspool before sending it. Sorry about that.


thanks,

Michael

Index: 3c59x.c
===================================================================
RCS file: /cvsroot/hurd/gnumach/linux/src/drivers/net/Attic/3c59x.c,v
retrieving revision 1.3
diff -u -r1.3 3c59x.c
--- 3c59x.c     8 Oct 1999 13:50:16 -0000       1.3
+++ 3c59x.c     17 Feb 2004 11:27:29 -0000
@@ -286,6 +286,8 @@
         PCI_USES_IO|PCI_USES_MASTER, IS_CYCLONE, 128, vortex_probe1},
        {"3c555 Laptop Hurricane",      0x10B7, 0x5055, 0xffff,
         PCI_USES_IO|PCI_USES_MASTER, IS_CYCLONE, 128, vortex_probe1},
+       {"3c556 10/100 Mini PCI Adapter",       0x10B7, 0x6055, 0xffff,
+        PCI_USES_IO|PCI_USES_MASTER, IS_CYCLONE|HAS_CB_FNS, 128, 
vortex_probe1},
        {"3c575 Boomerang CardBus",             0x10B7, 0x5057, 0xffff,
         PCI_USES_IO|PCI_USES_MASTER, IS_BOOMERANG|HAS_MII, 64, vortex_probe1},
        {"3CCFE575 Cyclone CardBus",    0x10B7, 0x5157, 0xffff,
@@ -845,7 +847,11 @@
 #ifdef CARDBUS
                outw(0x230 + i, ioaddr + Wn0EepromCmd);
 #else
-               outw(EEPROM_Read + i, ioaddr + Wn0EepromCmd);
+               if (pci_tbl[chip_idx].device_id == 0x6055) {
+                       outw(0x230 + i, ioaddr + Wn0EepromCmd);
+               } else {
+                       outw(EEPROM_Read + i, ioaddr + Wn0EepromCmd);
+               }
 #endif
                /* Pause for at least 162 us. for the read to take place. */
                for (timer = 10; timer >= 0; timer--) {
@@ -893,7 +899,11 @@
                printk("%s: CardBus functions mapped %8.8x->%p (PCMCIA 
committee"
                           " brain-damage).\n", dev->name, fn_st_addr, 
vp->cb_fn_base);
                EL3WINDOW(2);
-               outw(0x10 | inw(ioaddr + Wn2_ResetOptions), ioaddr + 
Wn2_ResetOptions);
+               if (pci_tbl[chip_idx].device_id == 0x6055) {
+                       outw(0x4010 | inw(ioaddr + Wn2_ResetOptions), ioaddr + 
Wn2_ResetOptions);
+               } else {
+                       outw(0x10 | inw(ioaddr + Wn2_ResetOptions), ioaddr + 
Wn2_ResetOptions);
+               }
        }
 
        /* Extract our information from the EEPROM data. */