Patchwork [v3,6/7] scsi: sr: balance sr disk events block depth

login
register
mail settings
Submitter Aaron Lu
Date July 26, 2012, 10:05 a.m.
Message ID <1343297129-28174-7-git-send-email-aaron.lu@amd.com>
Download mbox | patch
Permalink /patch/173386/
State Not Applicable
Delegated to: David Miller
Headers show

Comments

Aaron Lu - July 26, 2012, 10:05 a.m.
When the ODD is resumed, disk_unblock_events should be called when:
1 The ODD is runtime resumed;
2 System is resuming from S3 and the ODD is runtime suspended before S3;
But not when the system is resuming from S3 and the ODD is runtime
active before S3.

So seperate the resume calls, one for system resume and one for runtime
resume to do different things accordingly.

Signed-off-by: Aaron Lu <aaron.lu@amd.com>
---
 drivers/scsi/sr.c | 21 +++++++++++++++++++++
 1 file changed, 21 insertions(+)
Sergei Shtylyov - July 26, 2012, 10:54 a.m.
Hello.

On 26-07-2012 14:05, Aaron Lu wrote:

> When the ODD is resumed, disk_unblock_events should be called when:
> 1 The ODD is runtime resumed;
> 2 System is resuming from S3 and the ODD is runtime suspended before S3;
> But not when the system is resuming from S3 and the ODD is runtime
> active before S3.

> So seperate the resume calls, one for system resume and one for runtime
> resume to do different things accordingly.

> Signed-off-by: Aaron Lu <aaron.lu@amd.com>
> ---
>   drivers/scsi/sr.c | 21 +++++++++++++++++++++
>   1 file changed, 21 insertions(+)

> diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
> index fd1c2f6..b8c2f9d 100644
> --- a/drivers/scsi/sr.c
> +++ b/drivers/scsi/sr.c
[...]
> @@ -211,6 +217,21 @@ static int sr_suspend(struct device *dev, pm_message_t msg)
>
>   static int sr_resume(struct device *dev)
>   {
> +	struct scsi_cd *cd = dev_get_drvdata(dev);
> +
> +	/*
> +	 * If ODD is runtime suspended before system pm, unblock disk
> +	 * events now since on system resume, we will fully resume it

    Comma not needed.

> +	 * and set its rumtime status to active.

    s/rumtime/runtime/. Nice typo. %-)

WBR, Sergei

--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Aaron Lu - July 26, 2012, 12:47 p.m.
On Thu, Jul 26, 2012 at 02:54:01PM +0400, Sergei Shtylyov wrote:
> Hello.
> 
> On 26-07-2012 14:05, Aaron Lu wrote:
> 
> >When the ODD is resumed, disk_unblock_events should be called when:
> >1 The ODD is runtime resumed;
> >2 System is resuming from S3 and the ODD is runtime suspended before S3;
> >But not when the system is resuming from S3 and the ODD is runtime
> >active before S3.
> 
> >So seperate the resume calls, one for system resume and one for runtime
> >resume to do different things accordingly.
> 
> >Signed-off-by: Aaron Lu <aaron.lu@amd.com>
> >---
> >  drivers/scsi/sr.c | 21 +++++++++++++++++++++
> >  1 file changed, 21 insertions(+)
> 
> >diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
> >index fd1c2f6..b8c2f9d 100644
> >--- a/drivers/scsi/sr.c
> >+++ b/drivers/scsi/sr.c
> [...]
> >@@ -211,6 +217,21 @@ static int sr_suspend(struct device *dev, pm_message_t msg)
> >
> >  static int sr_resume(struct device *dev)
> >  {
> >+	struct scsi_cd *cd = dev_get_drvdata(dev);
> >+
> >+	/*
> >+	 * If ODD is runtime suspended before system pm, unblock disk
> >+	 * events now since on system resume, we will fully resume it
> 
>    Comma not needed.

Thanks.

> 
> >+	 * and set its rumtime status to active.
> 
>    s/rumtime/runtime/. Nice typo. %-)

Thanks, and I'll use a spell checker next time to avoid such mistakes.

-Aaron
--
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Patch

diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index fd1c2f6..b8c2f9d 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -82,6 +82,11 @@  static int sr_remove(struct device *);
 static int sr_done(struct scsi_cmnd *);
 static int sr_suspend(struct device *dev, pm_message_t msg);
 static int sr_resume(struct device *dev);
+static int sr_runtime_resume(struct device *dev);
+
+static struct dev_pm_ops sr_pm_ops = {
+	.runtime_resume = sr_runtime_resume,
+};
 
 static struct scsi_driver sr_template = {
 	.owner			= THIS_MODULE,
@@ -91,6 +96,7 @@  static struct scsi_driver sr_template = {
 		.remove		= sr_remove,
 		.suspend	= sr_suspend,
 		.resume		= sr_resume,
+		.pm		= &sr_pm_ops,
 	},
 	.done			= sr_done,
 };
@@ -211,6 +217,21 @@  static int sr_suspend(struct device *dev, pm_message_t msg)
 
 static int sr_resume(struct device *dev)
 {
+	struct scsi_cd *cd = dev_get_drvdata(dev);
+
+	/*
+	 * If ODD is runtime suspended before system pm, unblock disk
+	 * events now since on system resume, we will fully resume it
+	 * and set its rumtime status to active.
+	 */
+	if (pm_runtime_suspended(dev))
+		disk_unblock_events(cd->disk);
+
+	return 0;
+}
+
+static int sr_runtime_resume(struct device *dev)
+{
 	struct scsi_cd *cd;
 	struct scsi_sense_hdr sshdr;