[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Broadcom NetExtreme BCM5722KFB1G and Debian Etch AMD64



On Mon, Jun 02, 2008 at 11:48:44 +0200, Thomas Brandt wrote:
> I have a Dell server with 2 network controllers with a Broadcom 
> BCM5722KFB1G - chip (1 x onboard, 1 x PCI-E).
> Unfortunately Debian Etch AMD64 does not recognize them.
> 
> 
> Somebody did post a workaround on this website:
> 
> http://kerneltrap.org/mailarchive/freebsd-bugs/2008/1/18/581834
> 
> 
> With this workaround these NICs should work with the original drivers that
> shipped with Debian Etch. But unfortunately I can't find the specified 
> file.
> (/usr/src/sys/dev/bge)
> The lines with the + at their beginning should be added to the original 
> file.
> Is there anybody who has a hint for me?

That patch is for the freeBSD driver. Which kernel do you run? 

(The patch assumes that the driver already knows how to work with that
 chipset, it only needs to be told to accept the relevant PCI ID. This
 is not necessarily the case for the tg3 driver in the 2.6.18 linux
 kernel.)

> As an alternative the Linux-tg3 driver could get compilied by myself:
> Dell offers the source code for the driver in their download area (v tg3-
> 3.81c):
> http://ftp.us.dell.com/network/Bcom_LAN_11.0_4.1_Linux_Source_A00.tar.gz
> 
> The according driver source code on the Broadcom website is even a little 
> bit newer (v tg3-3.85l):
> http://www.broadcom.com/docs/driver_download/570x/linux-3.85l.zip
> 
> But unfortunately something seems to be wrong with the MAKEFILE . 
> make does not work:
> 
> ==============================================
> server-deb: tmp/tg3-381C# make
> make -C SUBDIRS= tmp/tg3-381C modules
> make *** SUBDIRS= tmp/tg3-381C: File or folder not found.
> Stop.
> make: ***[default] Error 2
> ==============================================
> 
> Any suggestions? Any tips for compiling the driver? 

I would try the 2.6.22 linux kernel from backports.org; its tg3 module
has PCI_DEVICE_ID_TIGON3_5722 included in its pci_device_id table, so it
should recognize the chip. 

-- 
Regards,            | http://users.icfo.es/Florian.Kulzer
          Florian   |


Reply to: