lkml.org 
[lkml]   [1998]   [Sep]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: Today Linus redesigns the networking driver interface (was Re: tulip driver in ...)
Date
Hello!

> - when an interrupt handler is in process, that interrupt handler is
> single-threaded. All other interrupts can still happen.

== no fast paths are allowed?

I understand this point.

8)8) When this problem hitted us the first time, I advocated
exactly this viewpoint. Maybe, Alan remembers, I told something
sort of "Let's do not try to jump higher our bottom.
If you want to handle 100Mbit interface, wait for 1GHz cpu.
If you plan to switch to 1Gbit, you need 10GHz cpu"

Actually it would be great relief, if fast path processing were
officially announced heretical. It is too difficult problem. 8)8)

It is much easier to announce, that apps sending udp with full speed
are buggy or malicious (it is close to truth), routers should be purchased
from Cisco (>$20,000 year ago) and forget about this problem.
Is it good solution? It is not ironic note, I am serious now.

Alexey

-
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/

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.215 / U:0.360 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site