Date: Tue, 13 Jun 1995 02:29:53 -0500 (CDT) From: Mike Pritchard <mpp@legarto.minn.net> To: bde@zeta.org.au (Bruce Evans) Cc: roberto@blaise.ibp.fr, wollman@halloran-eldar.lcs.mit.edu, current@FreeBSD.ORG Subject: Re: GENERIC kernel & some basic UNIX pointers Message-ID: <199506130729.CAA03409@mpp.com> In-Reply-To: <199506130350.NAA06987@godzilla.zeta.org.au> from "Bruce Evans" at Jun 13, 95 01:50:07 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> The kernel mode ppp has a hack for low ping latency. This saves an average > of 5ms per end. Again, this should have little effect for large transmissions > if everything is streaming properly. > > kernel ppp is inefficiently programmed. On a 486DX2/66 with a 16550 UART, > the input overheads for 115200 bps are approximately: > > termios(raw) 6.3% > cslip 7% > ppp 9% > > These overheads include delivery of packets to an application that throws > the packets away. About 3% of each overhead is for the lowest level of > the driver (which handles interrupts and stores the input in a buffer). > Thus for ppp, about 66% of the overhead is for the protocol and for > delivery of packets to the user. It might be possibele for a user mode > ppp to improve on this, but not by much since it has to pay for the > termios(raw) protocol and packet delivery. > > Bruce I guess that I will have to run some user-mode/kernel-mode tests myself. If it really does turn out that the kernel-mode ppp interface is poorly implemented, I could probably be talked into trying to make it better, especially since I'm stuck with a ppp-only ISP now. If anyone has any suggestions towards this goal, let me know. -- Mike Pritchard mpp@legarto.minn.net "Go that way. Really fast. If something gets in your way, turn"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199506130729.CAA03409>