[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: proposed getitimer and setitimer functions

From: Rob Browning
Subject: Re: proposed getitimer and setitimer functions
Date: 29 Jul 2001 16:35:50 -0500
User-agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7

(Message that postdrop dropped on Date: 12 Jul 2001 10:16:04 -0500.
 Re-sending now.)

Dirk Herrmann <address@hidden> writes:

> For the long term, I'd suggest to use seconds as the base unit.  As
> has been suggested before, using rational numbers or reals allows to
> specify any fraction of the given time unit.  So why choose
> something else than the SI standard, i. e. seconds?

The only arguments I could think of *might* be precision or efficiency
-- but those don't seem all that defensible.  With the interface that
closely matches the unix call interface you can specify exactly what
you can specify from the C level, but I'm not sure that's important.

Note that getitimer (and setitimer) returns values that are intended
to be passed (more or less exactly) back to setitimer later when
you're doing real start/stop profiling...


Rob Browning
rlb,, and
GPG=1C58 8B2C FB5E 3F64 EA5C  64AE 78FE E5FE F0CB A0AD

reply via email to

[Prev in Thread] Current Thread [Next in Thread]