I stand (sit, actually) corrected !
When I was (even more) Linux newbie (than now), I briefly surveyed the
methods offered for doing subsecond waits, and as I remember, some of
them had a warning about signal interference attached, as of some older
version of libc docs had it. As I felt this was unclean (sic) and they
actually couldn't do the job either (I wanted 1 to 2 msec waits), I
haven't use them for anything.
> ----------
> From: Glynn Clements[SMTP:[EMAIL PROTECTED]]
> Sent: 29. september 1998 14:47
> To: Niels Hald Pedersen
> Cc: Marin D; [EMAIL PROTECTED]
> Subject: RE: Waiting for 125msecs.
>
>
> Niels Hald Pedersen wrote:
>
> > I would suspect usleep to encapsulate "the SIGALRM method of
> waiting"
> > (set an alarm, sending process a signal after a given time, wait for
> the
> > signal, handle it),
>
> It doesn't appear to. Sending SIGALRM to the following program doesn't
> cause it to terminate any sooner.
>
> #include <unistd.h>
> #include <signal.h>
>
> int main(void)
> {
> signal(SIGALRM, SIG_IGN);
> usleep(10 * 1000000);
> return 0;
> }
>
> --
> Glynn Clements <[EMAIL PROTECTED]>
>
