Re: pre-2.1.90..

Matthew Kirkwood (weejock@ferret.lmh.ox.ac.uk)
Wed, 11 Mar 1998 07:07:29 +0000 (GMT)


On Wed, 11 Mar 1998, G. Sumner Hayes wrote:

> > I just put a pre-90 on ftp.kernel.org, and I'm happy to report that Davem
> > seems to have found and fixed the TCP performance problem, which means
> > that the code-freeze for 2.2 is going to go into effect shortly..
>
> Looks good here; TCP performance is up. I've only done a once-over,
> though, and haven't had a chance to benchmark against pre-2.1.89
> numbers.

It's a bit better, but not what it has been:

Over relatively quiet 10mb/s network, I can get no more than 830k/s ftp,
where 2.1.8x were doing only about 720k/s. 2.0 and previous 2.1 would
often get upwards of 1mb/s.

Local throughput is a lot better -- up to 7.5mb/s from about 3mb/s on .89;
there's always a stall after about the first 25k, though -- is this
related to the bandwidth problem?

> > pre-90 does a few other minor things, like for example getting rid of
> > kerneld because the new kmod thing is a lot simpler in many ways. Let's
[snip]
>
> Works fine here.

And here. Won't it break diald (or whatever the modem/kerneld hack was),
though?

> The scheduler hack seems to be a win, subjectively, but I have no timings to
> back up that feeling.

It didn't affect kernel compilation times, but feels a little more
responsive.

> Some compilation issues with 2.1.89 have been fixed, too.

There's a horrible-looking warning from floppy.c, too. Don't know the
correct fix, though.

Matthew.

-- 
Matthew Kirkwood  |  Mail:  matthew.kirkwood@lmh.ox.ac.uk
LMH JCR,          |  Web:   http://www-jcr.lmh.ox.ac.uk/~weejock/
Oxford OX2 6QA,   |  
England.          |  "To do things badly is a basic human right"

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu