On Thu, Jul 12, 2012 at 04:11:37PM -0700, Rick Jones wrote:
On 07/12/2012 01:57 PM, Greg KH wrote:
quoted
On Thu, Jul 12, 2012 at 01:21:31PM -0700, Greg KH wrote:
quoted
I have an Apple Thunderbolt Ethernet device here, and running 3.5-rc6 it
isn't detected. It has PCI id 14e4:1682, which seems like it should be
supported by the tg3 driver. Any hints?
I guess I could just go and add the device id to the driver and see what
happens...
And that worked, patch sent.
But, as the patch shows, odds are it has a "real" device type in it,
so the #define I used isn't as descriptive as it should be. Any hints
on how I can figure out what to look at to make it more "correct"?
a long-shot, but maybe there is something in the pci.ids database?
Ah, nice, there is something there:
1682 NetXtreme BCM57762 Gigabit Ethernet PCIe
I'll redo the patch with that information, thanks.
greg k-h
On Thu, Jul 12, 2012 at 04:18:34PM -0700, Greg KH wrote:
On Thu, Jul 12, 2012 at 04:11:37PM -0700, Rick Jones wrote:
quoted
On 07/12/2012 01:57 PM, Greg KH wrote:
quoted
On Thu, Jul 12, 2012 at 01:21:31PM -0700, Greg KH wrote:
quoted
I have an Apple Thunderbolt Ethernet device here, and running 3.5-rc6 it
isn't detected. It has PCI id 14e4:1682, which seems like it should be
supported by the tg3 driver. Any hints?
I guess I could just go and add the device id to the driver and see what
happens...
And that worked, patch sent.
But, as the patch shows, odds are it has a "real" device type in it,
so the #define I used isn't as descriptive as it should be. Any hints
on how I can figure out what to look at to make it more "correct"?
a long-shot, but maybe there is something in the pci.ids database?
Ah, nice, there is something there:
1682 NetXtreme BCM57762 Gigabit Ethernet PCIe
I'll redo the patch with that information, thanks.
Wait, something's a bit "odd" here. We already have:
#define TG3PCI_DEVICE_TIGON3_57762 0x1682
in drivers/net/ethernet/broadcom/tg3.h
But it's only used in the function where we can't figure out what type
of vpd we have in tg3_read_vpd(). One could ask how that codepath
was ever tested, as that device id was not in the MODULE_DEVICE_TABLE();
I wonder if someone reused the device id?
Anyway, I'll respin the patch...
greg k-h