Problems with .91

[Douglas Eadline]

  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.
  Send mail to mime@docserver.cac.washington.edu for more info.

--8323328-629604385-926257099=:15069
Content-Type: TEXT/PLAIN; charset=US-ASCII

 
Don,

I am having problems with .91.  Here is the background,
I  was using 90q with tulip based NICS from KTI
"http://www.ktinet.com/" (model KF-221TX/2) 21143 tulip.

With .90q things seemed to work OK, although there are some problems
when using 2.0.36 and SCSI modules.  What I found was the 
PCI probing was wrong if I had SCSI enabled in the kernel.
I had two NICs - the addresses were confused.

Anyway, I tried .91 and the NICs will not auto-negotiate correctly.
After booting, debug reports 100Btx-FD as does the switch, but
the TCP performance on netperf is very poor.

I also have tried the "Intel Tulip" NICs - same NIC but
new Intel Tulip chip, but the problem is the same.

I have included tulip_debug output for 4 cases:

dec tulip ruuning .90q
dec tulip running .91
Intel tulip running .90q
Intel tulip running .91A


I can provide more info or samples of the NICs.

Doug

-------------------------------------------------------------------
Paralogic, Inc.           |     PEAK     |      Voice:+610.861.6960
115 Research Drive        |   PARALLEL   |        Fax:+610.861.8247
Bethlehem, PA 18017 USA   |  PERFORMANCE |    http://www.plogic.com
-------------------------------------------------------------------

--8323328-629604385-926257099=:15069
Content-Type: APPLICATION/zip; name="tulip.zip"
Content-Transfer-Encoding: BASE64
Content-ID: <Pine.LNX.4.04.9905090938190.15069@lisa.plogic.com>
Content-Description: diagnostic results
Content-Disposition: attachment; filename="tulip.zip"
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--8323328-629604385-926257099=:15069--