Re: FA511 not passing much data

On 9 Jan 2003, Jarl Friis wrote:

> Donald Becker <[email protected]> writes:
> 
> > On 8 Jan 2003, Jarl Friis wrote:
> > 
> > > Subject: Re: [tulip] FA511 not passing much data
> > 
> > > > Indeed 1385 is Netgear, though 1317 is linksys.
> > > > > 02:00.0 Class 0200: 1317:1985 (rev 11)
> > > > >         Subsystem: 1385:511a
> > 
> > Huh?  Where did you get the idea that PCI vendor 1317 is Linksys?
> 
> >From http://pciids.sourceforge.net/iii/?i=1317
> But of course that may be wrong. Do you know the correct PCI ID of
> linksys?

A board reseller isn't required to have an ID, only the people making
the PCI chips.  A PCI device must have an ID, and may optionally have a
subsystem ID.

The board vendor typically programs the subsystem ID with their own
values, although a few chips allow the configuration EEPROM to also
override the chip's standard PCI ID.  The motivation to do this is to
appear to have a unique product, for instance when the vendor doesn't
want the customer to know that "$80 value" NIC is exactly the same a $6
generic rtl8139 board.

Actually, the vendor ID is more general than just PCI devices.  Many
other devices (e.g. USB) now use a 16+16 bit Vendor/Device ID.  Much
like the 2+22+24 bit IEEE station address, this kind of standardization
is a very good thing.

> > This is a ADMtek Comet '985 chip, implemented on a NetGear card.
> 
> Very well, but it seems to be branded under linksys' PCI ID, right?

No. The list that Linux uses was apparently user-submitted guesses.
And once there is an entry, it is never validated.

> > There are several other Ethernet NIC chips in the Centuar/Comet series
> > with 0x1317 ID, generally with the format [01]98[135] or 951?
> 
> I have added the entry 1385:511a on
> http://pciids.sourceforge.net/iii/?i=13171985
> Do you disagree on this?

The 1385:511a is the subsystem ID for the Netgear FA511, which should
always use the 1317:1985 ADMtek Comet/Centaur chip.

This is similar to the 1385:f311 (subsystem ID) Netgear FA311 v1 board,
which always uses the 100B:0020 (primary ID) National Semiconductor
DP83815 chip.

-- 
Donald Becker				[email protected]
Scyld Computing Corporation		http://www.scyld.com
410 Severn Ave. Suite 210		Scyld Beowulf cluster system
Annapolis MD 21403			410-990-9993

_______________________________________________
tulip mailing list, [email protected]
To change to digest mode or unsubscribe visit
http://www.scyld.com/mailman/listinfo/tulip