[2763] in linux-net channel archive
Re: Tx TCP rates down > 20% - A report.
daemon@ATHENA.MIT.EDU (Alan Cox)
Wed May 1 23:31:56 1996
From: Alan Cox <alan@cymru.net>
To: gpg109@rsphy6.anu.edu.au (Paul Gortmaker)
Date: Wed, 1 May 1996 14:47:50 +0100 (BST)
Cc: linux-net@vger.rutgers.edu, linux-kernel@vger.rutgers.edu
In-Reply-To: <9605010219.AA03829@rsphy6.anu.edu.au> from "Paul Gortmaker" at May 1, 96 12:19:55 pm
> Tx speeds are way down in 1.3.97 as compared to v1.2.13 -- in fact there
> appears to be a "glass-ceiling" effect, where v1.3.97 can't Tx any
> better than sending 820 -> 830kB/s even with decent hardware. However
Its probably generating some stupid retransmits like the current
hacked around code does on PPP links. The other thing that may do it
is windows filling because of ack problems.
> Rx with v1.3.97 is good, as I can consistently jam >1100kB/s into a
> wd8013 and about 1040kB/s into a soft-config Winbond based ne2000 when
> Tx'ing from a 1.2.13 kernel.
Thats nice to know.
> It is also worth noting that similar Tx rates were obtained by ftp'ing
> a large file (cached) to /dev/null which indicates that the TTCP
> measurements are not subjet to some huge systematic error.
Does changing the large windows option affect it. Can a third box log the
link and see what kind of mess you see in duplicates. How does 1.3.59 compare.
We need to pin this down - its I think got to be extra frames or window/ack
problems. The base performance goes well over 40Mbit/second on 100baseT
Alan
