> > Such curcumstances just do not exist in the nature.
>
> As far as you know. You have not seen it so do not believe it. I guess I
> have to disagree with you here :-)
>
> "There are more things in heaven and earth than are dreamt of ..."
So now that we've shown that our program does in fact exist, and does lose
significant performance due to nastiness in the linux TCP implementation,
what does this mean? Since this program exists in our universe, does this
mean that we must not exist in order to preserve this pristine, happy-ack
universe you're referring to? Seems asking for an example or such would
be more productive than simply assuming that a set of programs simply
can't exist.
Sorry for the mini-flame, but I think responses to potential bugs or
performance issues with "that doesn't exist" deserves it...
--m
|