Re: ata: printf on every spinup/spindown?

On Thu, Mar 26, 2009 at 07:32:36PM +0000, Bruce Cran wrote:
> On Thu, 26 Mar 2009 20:12:07 +0100
> Lars Engels <lars.engels_at_0x20.net> wrote:
> 
> > On Thu, Mar 26, 2009 at 07:54:49PM +1100, Peter Jeremy wrote:
> > > On 2009-Mar-20 19:19:38 +0000, Bruce Cran <bruce_at_cran.org.uk> wrote:
> > > >There's a reluctance to include code like this, I
> > > >think because it bypasses the ATA driver and talks directly to the
> > > >drive. Since the driver doesn't know what the drive's been told to
> > > >do, it can't know to adjust timers etc. to wait for the disk to
> > > >spin back up, for example.
> > > 
> > > This code is no worse than installing sysutils/ataidle - which also
> > > bypasses the driver.
> > > 
> > > As it stands, FreeBSD out-of-the-box behaves in a way that adversely
> > > impacts laptop HDD life - and correcting this requires that the
> > > end-user both be aware of the problem and then find, install and
> > > configure a port to work around this.  I am very uncomfortable with
> > > this and would prefer to see the base system require less user
> > > knowledge/intervention.
> > 
> > AFAIK ataidle is under the BSD license, so we could include it in the
> > base system.
> 
> As the author I can confirm that it is BSD licensed :)
> Either way - including ataidle or patching atacontrol, I'd like to see
> a utility to set APM and AAM features at a minimum in the base system
> because especially with the problems on recent laptops it's something
> we're missing.   
> 
> I mentioned the reluctance to include code like it
> because when I initially wrote ataidle I got a bit of resistance about
> including it, I think from the worry that it would cause problems.
> There are indeed a few PRs reporting panics when spinning drives back
> up, but I suspect the driver is more resiliant to those issues now.

+1 in including it in base and add the according entries to devd.conf
(maybe commented, so people don't get any problems).