Re: A possible bug with non-blocking sockets and SIGIO

[Dennis Ferguson <dennis.c.ferguson%gmail.com@localhost>]

On 7 Mar 2011, at 11:48 , der Mouse wrote:
>> The conclusion is not to use SIGIO.
> 
> If NetBSD no longer supports use of SIGIO, NetBSD should stop
> pretending otherwise: NetBSD should remove the interfaces that appear
> to support it (O_ASYNC, SIGIO, and the associated documentation), to
> prevent this kind of confusion in the future.
> 
> If, on the other hand, NetBSD does still support SIGIO, this kind of
> response is of negative utility.

I agree with this.  A canonical "good" use of SIGIO is in the NTP
daemon, where it is about the only commonly available mechanism to
measure packet arrival times that has a hope of not degrading under
increasing server load (== more time not sitting in poll(), hence
bigger delays between the actual arrival of a packet at the machine
and the taking of a time stamp).  If use of SIGIO is to be deprecated
and the facility allowed to bit rot there should be some consideration
of what to replace it with in situations where SIGIO appears to be the
best game in town.

Dennis Ferguson