lkml.org 
[lkml]   [1999]   [Jun]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Why khttpd is a bad idea (was a pointless argument about
* Bjorn Wesen said:

> > > > uhh.. why do you want an embedded system to serve static webpages?
> > > If you have ever seen 3Com's NetBuilder SuperStack II, or the 3Com's
> [or the hundreds of other networking products on the market with built in
> webservers]
Exactly.

> > And you assert this could not be done in user space at sufficient speed?
>
> The original poster was probably refering to smaller memory footprint
> instead of speed. In an embedded system where flash memory is expensive, a
Also, I think, that implementing everything in the kernel of some sort
requires much simpler file system and thus faster code and smaller memory
requirements than implementing complex servers in the user space.

> large webserver is overkill. But there are smaller servers for that, like
> boa, which is about 40 kb object code. You need the appropriate functions
> from libc as well of course, but you probably need those anyway.
You wouldn't need them if the server is in the kernel. At the very least,
there wouldn't be any need for wrapping the kernel calls with user-space
functions - again, conservation of space and time.

> I've used both an in-kernel http server and boa for an embedded system and
> boa was much cleaner.
It's probably a much cleaner solution, yes, but not always faster and more
efficient both in terms of space and time. But the in-kernel implementation
is suitable mostly for the embedded systems, IMO.

marek
[unhandled content-type:application/pgp-signature]
\
 
 \ /
  Last update: 2005-03-22 13:52    [W:0.638 / U:0.048 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site