Re: Linux V2.1.128 localhost performance

Andi Kleen (ak@muc.de)
21 Nov 1998 11:30:58 +0100

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: Andi Kleen: "MM tuning, was: Re: Linux-2.1.129.."
Previous message: Gerard Roudier: "Re: Linux-2.1.129.. (Oops in NCR5c8xx)"

In article <m0zgsFg-0007U7C@the-village.bc.nu>,
alan@lxorguk.ukuu.org.uk (Alan Cox) writes:
>> throughput off from chargen via localhost runs at a high speed for
>> about 1/2 second, then dribbles down to 40-50 kilobytes per second.
>> This makes it difficult to test the performance of network thingies.

> Thats a TCP/scheduling bug in 2.1.x - someone modified the algorithms
> used to schedule wakeups without really understanding the "deep magic"
> in the 2.0.x one. Im assuming it'll get fixed

This is supposed to be fixed in 2.1.129. David checked a fix for it
into vger a few days ago. If anyone still sees it on 2.1.129+ please
report.

-Andi

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

Next message: Andi Kleen: "MM tuning, was: Re: Linux-2.1.129.."
Previous message: Gerard Roudier: "Re: Linux-2.1.129.. (Oops in NCR5c8xx)"