[eepro100] Simple patch for 815E chipset

[Peter.Stern@weizmann.ac.il]

On Tue, 1 Aug 2000, Donald Becker wrote:

> On Tue, 1 Aug 2000, Robert J.Dunlop wrote:

> > I was recently asked to install Linux on an Intel D815EEA
motherboard (Intel
> > chipset 815E) which contains an intergrated Intel Pro100 network interface.
> > Whilst the driver did not work out the box I found that all that was necessary
> > was to add the new device ID to the list of those supported.
> >
> > This patch is against the driver I found with kernel 2.2.16
> > static const char *version =
> > "eepro100.c:v1.09j-t 9/29/99 Donald Becker ...
> > "eepro100.c: $Revision: 1.20.2.10 $ 2000/05/31 Modified ...

> This ID was reported by Chas Wareing c3w@isp..net and is already in v1.11
>     ftp://www.scyld.com/pub/network/test/eepro100.c

> The new entry uses the ID string
>    "Intel i82562 EEPro100"
> Does this match the chip part number on your implementation?

> >   #define PCI_DEVICE_ID_INTEL_ID1030 0x1030
> >   #endif
> > + #ifndef PCI_DEVICE_ID_INTEL_ID2449
> > + #define PCI_DEVICE_ID_INTEL_ID2449 0x2449
> > + #endif
> ...
> > +     { "Intel D815EEA motherboard Pro100/V",
> > +       PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ID2449,

> I know that this looks like the obvious way to add new entries, but I've
> changed to just using the numeric constant directly.  The text name in the
> table is more descriptive, and there isn't the problem of pci.h changing the
> name.  What is the reason for a symbolic name when the value is constant for
> all time, is only used once and the name carries no information.


> Donald Becker                           becker@scyld.com
> Scyld Computing Corporation             http://www.scyld.com
> 410 Severn Ave. Suite 210               Beowulf Clusters / Linux
> Installations
> Annapolis MD 21403

I am trying to run Linux on a PC with the D815EEA Intel motherboard
which includes the i82562 chip.

I have tried to compile the new linux device driver eepro100.c from
     ftp://www.scyld.com/pub/network/test/eepro100.c
as a module following the instructions to compile pci-scan.c and eepro100.c
(http://www.scyld.com/expert/modules.html).

I copied eepro100.c, kern_compat.h, pci-scan.h, pci-scan.c  to:
     /usr/src/modules
and tried to compile the two C programs.

I am using the Redhat v6.2 distribution of Linux.

I got the following errors:

[root@localhost modules]#  gcc -DMODULE -D__KERNEL__ -DEXPORT_SYMTAB -Wall -Wstrict-prototypes -O6 -c pci-scan.c
In file included from pci-scan.c:62:
kern_compat.h:164: parse error before `0'
kern_compat.h:168: warning: type defaults to `int' in declaration of `mark_bh'
kern_compat.h:168: warning: parameter names (without types) in function declaration
kern_compat.h:168: warning: data definition has no type or storage class
kern_compat.h:168: parse error before `}'

root@localhost modules]# gcc -DMODULE -Wall -Wstrict-prototypes -O6 -c eepro100.c
In file included from eepro100.c:127:
kern_compat.h:164: parse error before `0'
kern_compat.h:168: warning: type defaults to `int' in declaration of `mark_bh'
kern_compat.h:168: warning: parameter names (without types) in function declaration
kern_compat.h:168: conflicting types for `mark_bh'
/usr/include/asm/softirq.h:101: previous declaration of `mark_bh'
kern_compat.h:168: warning: data definition has no type or storage class
kern_compat.h:168: parse error before `}'
gcc: Internal compiler error: program cc1 got fatal signal 11

And this dumps core.

Any help from you experts would be greatly appreciated.

Best wishes,
Peter

Peter Stern
Chemical Physics Department
Weizmann Institute of Science
76100 Rehovot, ISRAEL

email:  Peter.Stern@weizmann.ac.il
phone:  972-8-9342096
fax:    972-8-9344123