Re: [PATCH 1/2] libata: Introduce ata_platform_shutdown_one()

[Tejun Heo]

On Mon, Jul 03, 2017 at 02:02:41PM -0400, Nate Watterson wrote:
> In similar fashion to how ata_platform_remove_one() is used, this
> newly introduced method can be used by platform ata drivers to get
> basic shutdown functionality (stopping host DMA and interrupts).
> 
> Signed-off-by: Nate Watterson 
> ---
>  drivers/ata/libata-core.c | 20 
>  include/linux/libata.h|  1 +
>  2 files changed, 21 insertions(+)
> 
> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
> index e157a0e..537932e 100644
> --- a/drivers/ata/libata-core.c
> +++ b/drivers/ata/libata-core.c
> @@ -6702,6 +6702,25 @@ int ata_platform_remove_one(struct platform_device 
> *pdev)
>   return 0;
>  }
>  
> +/**
> + *   ata_platform_shutdown_one - Platform layer callback for device shutdown
> + *   @pdev: Platform device being shutdown
> + *
> + *   Platform layer indicates to libata via this hook that shutdown is
> + *   in progress and the input device should be quiesced. Functionally this
^
> + *   is equivalent to ata_platform_remove_one(), however devres_release_all()
> + *   is not called on the shutdown path as it is for remove so releasing the
> + *   resources associated with the device must instead be initiated directly.

I can't easily understand the above comment.  Can you please try to
rephrase it so that it's clear why and when it's used?

Thanks.

-- 
tejun

Re: [PATCH 1/2] libata: Introduce ata_platform_shutdown_one()

[Tejun Heo]

On Mon, Jul 03, 2017 at 02:02:41PM -0400, Nate Watterson wrote:
> In similar fashion to how ata_platform_remove_one() is used, this
> newly introduced method can be used by platform ata drivers to get
> basic shutdown functionality (stopping host DMA and interrupts).
> 
> Signed-off-by: Nate Watterson 
> ---
>  drivers/ata/libata-core.c | 20 
>  include/linux/libata.h|  1 +
>  2 files changed, 21 insertions(+)
> 
> diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
> index e157a0e..537932e 100644
> --- a/drivers/ata/libata-core.c
> +++ b/drivers/ata/libata-core.c
> @@ -6702,6 +6702,25 @@ int ata_platform_remove_one(struct platform_device 
> *pdev)
>   return 0;
>  }
>  
> +/**
> + *   ata_platform_shutdown_one - Platform layer callback for device shutdown
> + *   @pdev: Platform device being shutdown
> + *
> + *   Platform layer indicates to libata via this hook that shutdown is
> + *   in progress and the input device should be quiesced. Functionally this
^
> + *   is equivalent to ata_platform_remove_one(), however devres_release_all()
> + *   is not called on the shutdown path as it is for remove so releasing the
> + *   resources associated with the device must instead be initiated directly.

I can't easily understand the above comment.  Can you please try to
rephrase it so that it's clear why and when it's used?

Thanks.

-- 
tejun

[PATCH 1/2] libata: Introduce ata_platform_shutdown_one()

[Nate Watterson]

In similar fashion to how ata_platform_remove_one() is used, this
newly introduced method can be used by platform ata drivers to get
basic shutdown functionality (stopping host DMA and interrupts).

Signed-off-by: Nate Watterson 
---
 drivers/ata/libata-core.c | 20 
 include/linux/libata.h|  1 +
 2 files changed, 21 insertions(+)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index e157a0e..537932e 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -6702,6 +6702,25 @@ int ata_platform_remove_one(struct platform_device *pdev)
return 0;
 }
 
+/**
+ * ata_platform_shutdown_one - Platform layer callback for device shutdown
+ * @pdev: Platform device being shutdown
+ *
+ * Platform layer indicates to libata via this hook that shutdown is
+ * in progress and the input device should be quiesced. Functionally this
+ * is equivalent to ata_platform_remove_one(), however devres_release_all()
+ * is not called on the shutdown path as it is for remove so releasing the
+ * resources associated with the device must instead be initiated directly.
+ */
+void ata_platform_shutdown_one(struct platform_device *pdev)
+{
+   struct ata_host *host = platform_get_drvdata(pdev);
+
+   ata_host_detach(host);
+   devres_release(>dev, ata_host_stop, NULL, NULL);
+   devres_release(>dev, ata_host_release, NULL, NULL);
+}
+
 static int __init ata_parse_force_one(char **cur,
  struct ata_force_ent *force_ent,
  const char **reason)
@@ -7222,6 +7241,7 @@ void ata_print_version(const struct device *dev, const 
char *version)
 #endif /* CONFIG_PCI */
 
 EXPORT_SYMBOL_GPL(ata_platform_remove_one);
+EXPORT_SYMBOL_GPL(ata_platform_shutdown_one);
 
 EXPORT_SYMBOL_GPL(__ata_ehi_push_desc);
 EXPORT_SYMBOL_GPL(ata_ehi_push_desc);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index c9a69fc..6c01f23f 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1240,6 +1240,7 @@ struct pci_bits {
 struct platform_device;
 
 extern int ata_platform_remove_one(struct platform_device *pdev);
+extern void ata_platform_shutdown_one(struct platform_device *pdev);
 
 /*
  * ACPI - drivers/ata/libata-acpi.c
-- 
Qualcomm Datacenter Technologies, Inc. on behalf of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux
Foundation Collaborative Project.

[PATCH 1/2] libata: Introduce ata_platform_shutdown_one()

[Nate Watterson]

In similar fashion to how ata_platform_remove_one() is used, this
newly introduced method can be used by platform ata drivers to get
basic shutdown functionality (stopping host DMA and interrupts).

Signed-off-by: Nate Watterson 
---
 drivers/ata/libata-core.c | 20 
 include/linux/libata.h|  1 +
 2 files changed, 21 insertions(+)

diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index e157a0e..537932e 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -6702,6 +6702,25 @@ int ata_platform_remove_one(struct platform_device *pdev)
return 0;
 }
 
+/**
+ * ata_platform_shutdown_one - Platform layer callback for device shutdown
+ * @pdev: Platform device being shutdown
+ *
+ * Platform layer indicates to libata via this hook that shutdown is
+ * in progress and the input device should be quiesced. Functionally this
+ * is equivalent to ata_platform_remove_one(), however devres_release_all()
+ * is not called on the shutdown path as it is for remove so releasing the
+ * resources associated with the device must instead be initiated directly.
+ */
+void ata_platform_shutdown_one(struct platform_device *pdev)
+{
+   struct ata_host *host = platform_get_drvdata(pdev);
+
+   ata_host_detach(host);
+   devres_release(>dev, ata_host_stop, NULL, NULL);
+   devres_release(>dev, ata_host_release, NULL, NULL);
+}
+
 static int __init ata_parse_force_one(char **cur,
  struct ata_force_ent *force_ent,
  const char **reason)
@@ -7222,6 +7241,7 @@ void ata_print_version(const struct device *dev, const 
char *version)
 #endif /* CONFIG_PCI */
 
 EXPORT_SYMBOL_GPL(ata_platform_remove_one);
+EXPORT_SYMBOL_GPL(ata_platform_shutdown_one);
 
 EXPORT_SYMBOL_GPL(__ata_ehi_push_desc);
 EXPORT_SYMBOL_GPL(ata_ehi_push_desc);
diff --git a/include/linux/libata.h b/include/linux/libata.h
index c9a69fc..6c01f23f 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -1240,6 +1240,7 @@ struct pci_bits {
 struct platform_device;
 
 extern int ata_platform_remove_one(struct platform_device *pdev);
+extern void ata_platform_shutdown_one(struct platform_device *pdev);
 
 /*
  * ACPI - drivers/ata/libata-acpi.c
-- 
Qualcomm Datacenter Technologies, Inc. on behalf of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum, a Linux
Foundation Collaborative Project.