Re: Re[2]: Booting with PCI NE2k

[markus@infoscape.com]

--pgp-sign-Multipart_Wed_Jun_11_12:38:49_1997-1
Content-Type: text/plain; charset=US-ASCII

I am very happy that this problem was so easy to fix! Ken, can you
please include a README.PCI file into the doc directory that describes
this little trick.

Thanx


Markus


Harald Brueckner writes:
 > markus@infoscape.com wrote:
 > > Harald Brueckner writes:
 > >  > It is a NE2k clone (and Linux detects it as a NE2k card) and it
 > >  > even has a PROM socket but it gets not detected by any boot loader
 > >  > I tried.  I think the problem comes from the I/O port which gets
 > >  > automatically set to 0xe400 (by the PCI BIOS) with this PCI card
 > >  > (at least in my special case).  With ISA cards I/O is defined from
 > >  > 0x0000 - 0x1fff I think. Just PCI moves that I/O ports to that high
 > >  > addresses. :-/
 > >
 > > It should be very easy to patch the source code for etherboot. If you
 > > are lucky, then all you have to do is edit "src-32/Makefile" and add
 > > the port address to the line that reads:
 > >
 > >    NEFLAGS=	-DINCLUDE_NE -DNE_SCAN=0x280,0x300,0x320,0x340
 > >
 > > Of course, this is not a general solution to the problem and it will
 > > only work with your card. Actually, it might even fail to work when
 > > you switch motherboards and/or add/remove other PCI cards. At the very
 > > least, I would add these ports 0xE000, 0xE100, 0xE200, ..., 0xEF00. To
 > > avoid problems when probing ports on other cards, make sure that you
 > > list 0xE400 as the first address in the list. So this is how you want
 > > to change the line:
 > >
 > >    NEFLAGS=	-DINCLUDE_NE
 > > -DNE_SCAN=0xE400,0x280,0x300,0x320,0x340,0xE000,0xE100,0xE200,0xE300,0xE500,0xE600,0xE700,0xE800,0xE900,0xEA00,0xEB00,0xEC00,0xED00,0xEE00,0xEF00
 > >
 > > Just for the record; I know that other cards use 0x6?00 as their base
 > > address. So anybody trying to patch "etherboot" for PCI NE2k clones,
 > > should first check "/proc/ioports" before blindly entering any port
 > > addresses.
 > >
 > > Harald, can you please let us know, if this patch fixed your problem.
 >
 > Hello,
 >
 > yes, it does!! I just added your suggested line and it works pretty fine. As it
 > seems it was just a problem of the new, unknown I/O ports!
 > Thank you very much for your help!!
 >
 > If there is anything else I can check out for you please let me know!
 >
 > Harry
 >
 >
--
Markus Gutschke                         Internet: markus@infoscape.com
Infoscape, Inc                          Phone:    +1-415-537-3778
657 Mission Street, Suite 200
San Francisco, CA 94105

Disclaimer: The above message represents my personal opinion; It does not
            constitute an offical statement by Infoscape!

--pgp-sign-Multipart_Wed_Jun_11_12:38:49_1997-1
Content-Type: application/pgp-signature
Content-Transfer-Encoding: 7bit

-----BEGIN PGP MESSAGE-----
Version: 2.6.2

iQCVAwUBM57+yhqJqDLErwMxAQEZkgQAidx8O4C4SLDUxbHEifY1GJCuzRS5HHhZ
LYFBtfpGCwGlGqlMyIN9O97Mpirur3vSqeX+WZSgYLyzWks4UY8zX0ckXRopWFPH
BzGZLsMj5RPq/R1q1seFH+lMQoVrMlGapvk7m71np3534lM+MIyrbZ/Uj4XpSEKT
iyTvxwGW18k=
=kWHL
-----END PGP MESSAGE-----

--pgp-sign-Multipart_Wed_Jun_11_12:38:49_1997-1--