Guy Harris <[EMAIL PROTECTED]> writes:
> On Mon, Jul 23, 2001 at 07:10:14PM +0200, Yoann Vandoorselaere wrote:
> > But it would be a good thing to be able to specify it's own buffer allocation
> > policy. I sent a patch a while ago, but nobody answered (note it had a bug).
>
> ...and that it only adds that functionality on one platform.
>
> What should it do on, for example, platforms where a single kernel call
> returns multiple packets (BSD, SunOS 3.x/4.x/5.x, Digital UNIX), or on
> platforms that support a memory-mapped buffer (Linux 2.4[.x] kernel,
> although we haven't yet put in any case for that, and perhaps the BSDs
> in the future)? Should it fail? Or should it, on the first type of
> platform, allocate buffers on each read (rather than on each packet)
> and, on the second type of platform, allocate a buffer for each packet
> and copy into it?
Sure, it will behave differently, but should work without problem on
all architectures... Or am I missing something ?
--
Yoann Vandoorselaere | Tiniest "mesures unities?"
MandrakeSoft | - lenght : millimeter
| - volume : milliliter
| - intelligence : military man
-
This is the TCPDUMP workers list. It is archived at
http://www.tcpdump.org/lists/workers/index.html
To unsubscribe use mailto:[EMAIL PROTECTED]?body=unsubscribe
