Re: Fine grained locking at the socket level?

On Sun, Jun 22, 2003 at 04:58:54PM -0700, George V. Neville-Neil wrote: > Hi, > > It would seem that splnet() and frieds now simply return 0, > which I figure is part of making the code look like it used > to. What I'm wondering is why the Giant lock is still used in > th

Fine grained locking at the socket level?

Hi, It would seem that splnet() and frieds now simply return 0, which I figure is part of making the code look like it used to. What I'm wondering is why the Giant lock is still used in the socket layer? I thought sockets had had fine grained locking appli

Re: Zero Copy Sockets?

On Sun, Jun 22, 2003 at 12:46:26PM -0700, George V. Neville-Neil wrote: > Hi, > > I'm reading over the internals of the network stack in > -CURRENT and I'm wondering if the Zero Copy stuff is actually > in use yet. > > Thanks, > George Yes. But your driver needs to support

Re: nested ipfw dummynet pipes

On Fri, Jun 20, 2003 at 02:58:07PM -0400, Don Bowman wrote: ... > Is there a benefit to having the single wide pipe first, or > the many narrow pipes first, in the ruleset? i'd probably put the narrow pipes first, so that any single flow will not be able to monopolize the entire fat pipe. Still no

Zero Copy Sockets?

Hi, I'm reading over the internals of the network stack in -CURRENT and I'm wondering if the Zero Copy stuff is actually in use yet. Thanks, George ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/free

Re: freebsd-net Digest, Vol 13, Issue 6

Hello, A couple of years ago I was one of the networking/security engineers of a major datacenter company in my country. There goes my $0,02: 1) I find it very trustworthy of yours to share so many info about your net and systems with the whole Internet, but I'm not sure if it's a good policy, ma