lkml.org 
[lkml]   [2003]   [Mar]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
Date
SubjectRe: 2.5.65-mm1: eth0: Transmit error, Tx status register 90
----- Original Message ----- 
From: Andrew Morton <akpm@digeo.com>
Date: Tue, 18 Mar 2003 19:18:33 -0800
To: "Felipe Alfaro Solana" <felipe_alfaro@linuxmail.org>
Subject: Re: 2.5.65-mm1: eth0: Transmit error, Tx status register 90

> "Felipe Alfaro Solana" <felipe_alfaro@linuxmail.org> wrote:
> >
> > I've been benchmarking file copy operations between 2.5.65-mm1
> > and 2.4.20-2.51 since I have noticed that transferring files from my
> > 2.4.20 machine to 2.5.65-mm1 gives an steady 10MBps throughput
> > but doing the opposite (from 2.5.65-mm1 to 2.4.20) gives me a
> > mere 3MBps throughput.
>
> Is it slow with both scp and NFS? Or just NFS?

Well, it seems the network transport is slow. I have benchmarked
using NFS, FTP and SCP and *all* of them are 4 orders of magnitude
slower when sending data (from 2.5 to 2.4), but behave normally
when receiving data (from 2.4 to 2.5).

> If just NFS then yes, I see this too. Transferring files 2.5->2.4 over NFS
> is several times slower than 2.4->2.4 or 2.5->2.5. Quite repeatable.

The problem is that FTP is also four times slower. Here are my timings
against 2.5.64 vanilla:

Test case: 128MB file (dd if=/dev/zero of=test bs=1024k count=128)

NFS
---
time cp 2.4 -> 2.5: 0m11.759s
time cp 2.5 -> 2.4: 0m39.541s

FTP
---
time get 2.4 -> 2.5: 11.5s
time put 2.5 -> 2.4: 43.3s

So it seems a problem with the network transport. This doesn't
happen with 2.4 on the same hardware: NEC Chrom@ laptop,
TI CardBus bridge, 3Com Corporation 3CCFE575CT Cyclone
CardBus (rev 10) NIC.

Attached is my .config file. It's worth a look since I needed to
build CardBus into the kernel (my RedHat distro doesn't work
with CardBus built as a module).

> That's a transmit underrun. The PCI/memory system was not able to feed data
> into the NIC fast enough.
> Please determine when this started. 2.5.64 would be a good kernel to test
> because it doesn't have the PCI changes.

During benchmarks with 2.5.64 I've also seen this errors,
although they were quite few (only two).

Now, what else? I'm lost...
Thanks!

Felipe
--
______________________________________________
http://www.linuxmail.org/
Now with e-mail forwarding for only US$5.95/yr

Powered by Outblaze
[unhandled content-type:application/octet-stream]
\
 
 \ /
  Last update: 2005-03-22 13:34    [W:0.037 / U:0.284 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site