Re: SIGIO for UDP?

Andi Kleen (ak@muc.de)
07 Nov 1998 19:16:17 +0100

Messages sorted by: [ date ][ thread ][ subject ][ author ]
Next message: David Woodhouse: "Re: Comments on Microsoft Open Source document"
Previous message: David Woodhouse: "Re: Suggestion for /proc/net/dev: Additional field "Bandwidth""
In reply to: Hugo Mildenberger: "SIGIO for UDP?"
Next in thread: Tim Smith: "Re: Suggestion for /proc/net/dev: Additional field "Bandwidth""
Reply: Tim Smith: "Re: Suggestion for /proc/net/dev: Additional field "Bandwidth""

In article <98110717223900.00545@andromeda>,
Hugo Mildenberger <milden@dialup.nacamar.de> writes:
> Hi there,
> ntp tries to use SIGIO for UDP packets - but it is not implemented just until
> 2.1.126. Does the signal mechanism lead to a quicker response or, at least, to
> a more deterministic behavior compared to the select () system call? ntp
> relies on reading a timestamp directly after packet arrives, so using SIGIO
> could eventually increase accuracy.

SIGIO for UDP sockets should work. If not it is a bug.

-Andi

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

Next message: David Woodhouse: "Re: Comments on Microsoft Open Source document"
Previous message: David Woodhouse: "Re: Suggestion for /proc/net/dev: Additional field "Bandwidth""
In reply to: Hugo Mildenberger: "SIGIO for UDP?"
Next in thread: Tim Smith: "Re: Suggestion for /proc/net/dev: Additional field "Bandwidth""
Reply: Tim Smith: "Re: Suggestion for /proc/net/dev: Additional field "Bandwidth""