Problems with LinkSys 10/100 cards

[Brian Denheyer]

>>>>> "rgb" == Robert G Brown <rgb@phy.duke.edu> writes:

rgb> I'm curious.  I may have asked these questions before (a couple of
rgb> months ago) but for those of us with brains damaged by a hard youth the
rgb> world is always new, so bear with me.  Do you still get the same general
rgb> KIND of performance as is indicated in the graph?  That is, a region of
rgb> linear growth in speed where the NIC is sending its maximum packet rate
rgb> (so speed is varying linearly with the size of the packet payload),
rgb> followed by a saturation region where the number of packets per second
rgb> decreases as the size of the packets increases?  Where, in the case of a
rgb> normal card, the saturation occurs where the local transmission
rgb> requirements hit wirespeed and in the case of the lite-on they are being
rgb> bottlenecked by something else?

I didn't explain myself too well.  The card tracks the "shape" of the
graph, including those interesting drop-outs, but its' "top-end" is
60-70Mbits/s.  This is not very useful since 2/3 of the time it can't
make it through the test(s).

rgb> I thought that the guy who posted one of the patches (Steven Huang?)
rgb> worked for Linksys.  He ought to be able to send you a current datasheet
rgb> (or provide even more substantial help).  There are a lot of linux users
rgb> out there -- my newspaper (in the article that trumpeted the fact that
rgb> Dell now officially supports linux) pointed out that linux owns 18% of
rgb> the server market and has the fastest growth rate by far of any of the
rgb> major operating systems in all markets.  Linksys appears to be wise
rgb> enough to recognize this and to maybe be making an effort to ensure that
rgb> their cards work with linux.

I agree.  One would think an official datasheet is not a high price to
pay for free work to give a product a broader market.

rgb> At $30/each, this is MY favorite solution;-).  But <sigh>, with 21140's
rgb> getting scarce, I suppose that one day having the lite-on chip work
rgb> properly would be very useful.  I can't help much with the actual
rgb> debugging, though -- I do have one remote box with a lite-on chip in it,
rgb> but it is on a 10-base hub.

Well, as I'm sure you saw on a subsequent post, there are genuine
21140's to be found at a good price.  My e-mail was mistaken, the
1200TX uses a 21140 not a 21141.

Re the lite-on chip, I think it is the 169 version which makes the
difference.  My guess is that cards with 168's work.  And you're
right, it would be useful to have the 169 version work.

Of course, there is always that small possibility that there is
something in my partcular hardware set-up causing the problem, but I
sure would think that my tx1200 experiment proved otherwise.

Brian

-- 

Brian Denheyer             briand@deldotd.com
Deldot Design, Inc.        (503) 788-1607
__              
\/.d = Electronic design and development