Messages in this thread Patch in this message | | | Subject | Re: [RFC] ata port runtime pm | From | Lin Ming <> | Date | Fri, 28 Oct 2011 13:36:34 +0800 |
| |
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.
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); + spin_lock_irqsave(ap->lock, irq_flags);
ata_scsi_dump_cdb(ap, cmd); What do you think?
Thanks, Lin Ming
| |