svn commit: r198487 - head/sys/dev/ata

[M. Warner Losh]

In message: <4AE76611.4030104 at FreeBSD.org>
            Alexander Motin <mav at FreeBSD.org> writes:
: M. Warner Losh wrote:
: > In message: <200910261123.n9QBNfJY087028 at svn.freebsd.org>
: >             Alexander Motin <mav at FreeBSD.org> writes:
: > : Author: mav
: > : Date: Mon Oct 26 11:23:41 2009
: > : New Revision: 198487
: > : URL: http://svn.freebsd.org/changeset/base/198487
: > : 
: > : Log:
: > :   Round timeout up when converting CAM milliseconds to ATA seconds.
: > 
: > This implements ceil(timeout) rather than round(timeout).  Is that
: > intended?
: 
: Yes. It is better to turn 400ms to 1s rather then 0. Usually timeouts
: aren't so short to make it important, but I have seen such case once, so
: it is possible to predict another.

OK.  I was curious because the commit message was ambiguous and could
be taken to mean two different things.  If that's the real intent,
then the code does that and I'm happy.

Warner