Re: [RFC] ata port runtime pm

[Alan Stern]

On Fri, 28 Oct 2011, Rafael J. Wysocki wrote:

> On Friday, October 28, 2011, Lin Ming wrote:
> > On Fri, 2011-10-28 at 11:37 +0800, Jeff Garzik wrote:
> > > On 10/27/2011 11:21 PM, Lin Ming wrote:
> > > > @@ -3208,6 +3209,11 @@ int ata_scsi_queuecmd(struct Scsi_Host *shost, struct scsi_cmnd *cmd)
> > > >
> > > >   	ap = ata_shost_to_port(shost);
> > > >
> > > > +	if (pm_runtime_suspended(&ap->tdev))
> > > > +		pm_runtime_resume(&ap->tdev);
> > > > +	pm_runtime_mark_last_busy(&ap->tdev);
> > > > +	pm_request_autosuspend(&ap->tdev);
> > > > +
> > > >   	spin_lock_irqsave(ap->lock, irq_flags);
> > > >
> > > 
> > > 
> > > Putting this into the core command dispatch fast-path is rather 
> > > disappointing.  That's at least one additional lock, plus some atomic 
> > > instructions and tests.

And it calls pm_runtime_resume(), which requires process context, from 
within a SCSI queuecmd routine, which runs in interrupt context.

> > Maybe move suspend request to the resume function, as below.
> > 
> > static int ata_port_runtime_resume(struct device *dev)
> > {
> >         struct ata_port *ap = to_ata_port(dev);
> >         int rc;
> > 
> >         rc = ata_port_request_pm(ap, PMSG_ON, ATA_EH_RESET,
> >                 ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 1);
> > 
> >         pm_runtime_mark_last_busy(dev);
> >         pm_request_autosuspend(dev);
> > 
> >         return rc;
> > }
> > 
> > Then the fast-path looks like,
> > 
> > @@ -3208,6 +3209,9 @@ int ata_scsi_queuecmd(struct Scsi_Host *shost,
> > struct scsi_cmnd *cmd)
> > 
> >         ap = ata_shost_to_port(shost);
> > 
> > +       if (pm_runtime_suspended(&ap->tdev))
> > +               pm_runtime_resume(&ap->tdev);
> > +

Unfortunately that won't work.  It's got a race; the device might be 
active when the pm_runtime_suspended() test runs and then it might 
suspend immediately after.

> >         spin_lock_irqsave(ap->lock, irq_flags);
> > 
> >         ata_scsi_dump_cdb(ap, cmd);
> > 
> > What do you think?

See the example code in section 9 of 
Documentation/power/runtime_pm.txt for the outline of a general 
approach.

Alan Stern

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/