[JDEV] Minor 'bug'let ?

Jeremie Miller jeremie at netins.net
Wed Jan 6 10:21:42 CST 1999

> I'll give it a go and send you a diff; all it does is that when the (largish)
> bit array
> is sequentially walked trhough, from bit 0 to bit n-1 it stops at this max. On
> a
> desktop machine; FD_SETSIZE is typically so low (256 or so) that you do not
> incur that much of an overhead; but on a server; it can be set (usually at
> kernel
> level) quite high. We use values upward of 64000 for certain applications. As
> most (BSDish) kernels seem to epect low values; we find that keeping the value
> low is worth the expense of keeping a max performance wise.

Ahh... gotcha.  A patch would be great, otherwise I can work this in later
today, no biggie.

> > What does TCP_NODELAY exactly do?  I've read up on some of this stuff, but
> > it's not always clear.
> [good stuff zapped]

That's kind of the gist I got from what I read about it.  I'll toss it in
today too... 

I did notice when I was playing on FreeBSD, that it was sending an initial
chunk of the first write() incompletly, and this helps to explain that...

> Forgive me for nattering.

Ha, please, natter more, really great stuff and I love hearing about the
background on these things!  Thanks!


More information about the JDev mailing list