Date: Thu, 26 Jul 2001 02:01:16 +0200 From: "Jose M. Alcaide" <jose@we.lc.ehu.es> To: Luis Javier Rodriguez <luisja@we.lc.ehu.es> Cc: hardware@FreeBSD.ORG, scott.mitchell@mail.com Subject: Re: Negotiation problems with a Xircom Ethernet 10/100 (RE-100BTX) PCCARD Message-ID: <3B5F5DCC.D40A8FB6@we.lc.ehu.es> References: <3B5EB0B7.2C3B9911@we.lc.ehu.es>
next in thread | previous in thread | raw e-mail | index | archive | help
[CC'ed to Scott Mitchell, as the author of the xe driver]
Luis Javier Rodriguez wrote:
>
> A colleague told me that maybe "the error was testing the pccard with
> Windows Me", because it could write internal registers with different
> values from those suitable for FreeBSD. But the pccard is resetted
> when detected, so this should not be an argument.
I am that colleague (hi, Luis Javier ;-) ), and I want to point out the
following facts:
1. We have two Xircom RealPort 10/100 cards (CE3-10/100): one was
purchased about 18 months ago, and the other one only some days ago.
2. Both cards are *identical*: same DingoID, RevisionID and VendorID.
3. Each card was tested under FreeBSD 4.3-RELEASE on two Dell Inspiron
laptops, models 3700 and 4000.
4. Also, we did the tests using two different Ethernet connections:
to a hub (10 Mbps) and to a switch (100 Mbps half-duplex).
5. The old card works fine, but the new card shows a strange
behavior, as explained in the original message: the media seems
to be correctly reported by ifconfig(8), but a simple ping(8) does
not work; instead, the driver resets the card again, and even once
more, before the packets begin to flow.
6. The old card has been used *only* with FreeBSD. The new card was
first used under Windows Millennium on the Inspiron 4000.
7. The version of the Xircom driver for Windows is 2.05; however,
we did another test, installing an old version of the driver
(1.75), and then the card showed a different behavior when
the machine was rebooted with FreeBSD (in fact, it was worse).
Then, we reinstalled the previous driver (2.05), and we came
back to the original situation. From this test, we could infer
that the Windows driver "touches" something in the card which
survives the reset issued by the xe(4) driver. Hey, it's my
hypothesis ;-)
8. We could test the old card with Windows and then with FreeBSD,
but I don't like the risk of another card being corrupted by the
Windows driver (if my hypothesis is correct). At least, now we
have one card which works fine ;-)
We are really interested in finding the cause of this problem, so
any help will be greatly appreciated.
Cheers,
-- JMA
****** Jose M. Alcaide // jose@we.lc.ehu.es // jmas@FreeBSD.org ******
** "Beware of Programmers who carry screwdrivers" -- Leonard Brandwein **
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hardware" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3B5F5DCC.D40A8FB6>