* [PATCH] platform: add new device registration helper
@ 2008-09-04 21:30 Dmitry Baryshkov
2008-09-05 0:20 ` Andrew Morton
2008-09-05 0:23 ` Andrew Morton
0 siblings, 2 replies; 5+ messages in thread
From: Dmitry Baryshkov @ 2008-09-04 21:30 UTC (permalink / raw)
To: linux-kernel; +Cc: Greg Kroah-Hartman, Dmitry Baryshkov
Add a helper that registers simple platform_device
w/o resources but with parent and device data.
This is usefull to cleanup platform code from code that
registers such simple devices as leds-gpio, generic-bl,
etc.
Signed-off-by: Dmitry Baryshkov <dbaryshkov@gmail.com>
---
drivers/base/platform.c | 47 +++++++++++++++++++++++++++++++++++++++
include/linux/platform_device.h | 2 +
2 files changed, 49 insertions(+), 0 deletions(-)
diff --git a/drivers/base/platform.c b/drivers/base/platform.c
index 3f94039..fcd9f97 100644
--- a/drivers/base/platform.c
+++ b/drivers/base/platform.c
@@ -391,6 +391,53 @@ error:
}
EXPORT_SYMBOL_GPL(platform_device_register_simple);
+/**
+ * platform_device_register_data
+ * @parent: parent device for the device we're adding
+ * @name: base name of the device we're adding
+ * @id: instance id
+ * @data: platform specific data for this platform device
+ * @size: size of platform specific data
+ *
+ * This function creates a simple platform device that requires minimal
+ * resource and memory management. Canned release function freeing memory
+ * allocated for the device allows drivers using such devices to be
+ * unloaded without waiting for the last reference to the device to be
+ * dropped.
+ */
+struct platform_device *platform_device_register_data(
+ struct device *parent,
+ const char *name, int id,
+ const void *data, size_t size)
+{
+ struct platform_device *pdev;
+ int retval;
+
+ pdev = platform_device_alloc(name, id);
+ if (!pdev) {
+ retval = -ENOMEM;
+ goto error;
+ }
+
+ pdev->dev.parent = parent;
+
+ if (size) {
+ retval = platform_device_add_data(pdev, data, size);
+ if (retval)
+ goto error;
+ }
+
+ retval = platform_device_add(pdev);
+ if (retval)
+ goto error;
+
+ return pdev;
+
+error:
+ platform_device_put(pdev);
+ return ERR_PTR(retval);
+}
+
static int platform_drv_probe(struct device *_dev)
{
struct platform_driver *drv = to_platform_driver(_dev->driver);
diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h
index 95ac21a..4b8cc6a 100644
--- a/include/linux/platform_device.h
+++ b/include/linux/platform_device.h
@@ -37,6 +37,8 @@ extern int platform_add_devices(struct platform_device **, int);
extern struct platform_device *platform_device_register_simple(const char *, int id,
struct resource *, unsigned int);
+extern struct platform_device *platform_device_register_data(struct device *,
+ const char *, int, const void *, size_t);
extern struct platform_device *platform_device_alloc(const char *name, int id);
extern int platform_device_add_resources(struct platform_device *pdev, struct resource *res, unsigned int num);
--
1.5.6.5
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: [PATCH] platform: add new device registration helper
2008-09-04 21:30 [PATCH] platform: add new device registration helper Dmitry Baryshkov
@ 2008-09-05 0:20 ` Andrew Morton
2008-09-05 10:33 ` Dmitry Baryshkov
2008-09-05 0:23 ` Andrew Morton
1 sibling, 1 reply; 5+ messages in thread
From: Andrew Morton @ 2008-09-05 0:20 UTC (permalink / raw)
To: Dmitry Baryshkov; +Cc: linux-kernel, gregkh, dbaryshkov
On Fri, 5 Sep 2008 01:30:16 +0400
Dmitry Baryshkov <dbaryshkov@gmail.com> wrote:
> Add a helper that registers simple platform_device
> w/o resources but with parent and device data.
>
> This is usefull to cleanup platform code from code that
> registers such simple devices as leds-gpio, generic-bl,
> etc.
>
nits:
> include/linux/platform_device.h | 2 +
> 2 files changed, 49 insertions(+), 0 deletions(-)
>
> diff --git a/drivers/base/platform.c b/drivers/base/platform.c
> index 3f94039..fcd9f97 100644
> --- a/drivers/base/platform.c
> +++ b/drivers/base/platform.c
> @@ -391,6 +391,53 @@ error:
> }
> EXPORT_SYMBOL_GPL(platform_device_register_simple);
>
> +/**
> + * platform_device_register_data
> + * @parent: parent device for the device we're adding
> + * @name: base name of the device we're adding
> + * @id: instance id
> + * @data: platform specific data for this platform device
> + * @size: size of platform specific data
> + *
> + * This function creates a simple platform device that requires minimal
> + * resource and memory management. Canned release function freeing memory
> + * allocated for the device allows drivers using such devices to be
> + * unloaded without waiting for the last reference to the device to be
> + * dropped.
Should document the return value (tested with IS_ERR)
> + */
> +struct platform_device *platform_device_register_data(
> + struct device *parent,
> + const char *name, int id,
> + const void *data, size_t size)
Something like
struct platform_device *platform_device_register_data(struct device *parent,
const char *name, int id, const void *data, size_t size)
would be more typical.
> +{
> + struct platform_device *pdev;
> + int retval;
> +
> + pdev = platform_device_alloc(name, id);
> + if (!pdev) {
> + retval = -ENOMEM;
> + goto error;
> + }
> +
> + pdev->dev.parent = parent;
> +
> + if (size) {
> + retval = platform_device_add_data(pdev, data, size);
> + if (retval)
> + goto error;
> + }
> +
> + retval = platform_device_add(pdev);
> + if (retval)
> + goto error;
> +
> + return pdev;
> +
> +error:
> + platform_device_put(pdev);
Are you sure this can't trigger a !kobj->state_initialized warning in
kobject_put()?
> + return ERR_PTR(retval);
> +}
> +
>
> ...
>
> +extern struct platform_device *platform_device_register_data(struct device *,
> + const char *, int, const void *, size_t);
It's nice (IMO) to include the names of the args, for documentation
purposes. Obviously the surrounding code didn't agree.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] platform: add new device registration helper
2008-09-04 21:30 [PATCH] platform: add new device registration helper Dmitry Baryshkov
2008-09-05 0:20 ` Andrew Morton
@ 2008-09-05 0:23 ` Andrew Morton
1 sibling, 0 replies; 5+ messages in thread
From: Andrew Morton @ 2008-09-05 0:23 UTC (permalink / raw)
To: Dmitry Baryshkov; +Cc: linux-kernel, gregkh, dbaryshkov
also,
On Fri, 5 Sep 2008 01:30:16 +0400
Dmitry Baryshkov <dbaryshkov@gmail.com> wrote:
> }
> EXPORT_SYMBOL_GPL(platform_device_register_simple);
>
> ...
>
> +struct platform_device *platform_device_register_data(
When you start implementing callers, please check whether
platform_device_register_data() should have been exported to modules.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] platform: add new device registration helper
2008-09-05 0:20 ` Andrew Morton
@ 2008-09-05 10:33 ` Dmitry Baryshkov
2008-09-05 18:09 ` Andrew Morton
0 siblings, 1 reply; 5+ messages in thread
From: Dmitry Baryshkov @ 2008-09-05 10:33 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel, gregkh
On Thu, Sep 04, 2008 at 05:20:50PM -0700, Andrew Morton wrote:
> On Fri, 5 Sep 2008 01:30:16 +0400
> Dmitry Baryshkov <dbaryshkov@gmail.com> wrote:
>
> > Add a helper that registers simple platform_device
> > w/o resources but with parent and device data.
> >
> > This is usefull to cleanup platform code from code that
> > registers such simple devices as leds-gpio, generic-bl,
> > etc.
> >
>
> nits:
>
> > + *
> > + * This function creates a simple platform device that requires minimal
> > + * resource and memory management. Canned release function freeing memory
> > + * allocated for the device allows drivers using such devices to be
> > + * unloaded without waiting for the last reference to the device to be
> > + * dropped.
>
> Should document the return value (tested with IS_ERR)
Will the wording "Returns a valid platform device, or a valid IS_ERR
condition containing errno" be appropriate?
> > +{
> > + struct platform_device *pdev;
> > + int retval;
> > +
> > + pdev = platform_device_alloc(name, id);
> > + if (!pdev) {
> > + retval = -ENOMEM;
> > + goto error;
> > + }
> > +
> > + pdev->dev.parent = parent;
> > +
> > + if (size) {
> > + retval = platform_device_add_data(pdev, data, size);
> > + if (retval)
> > + goto error;
> > + }
> > +
> > + retval = platform_device_add(pdev);
> > + if (retval)
> > + goto error;
> > +
> > + return pdev;
> > +
> > +error:
> > + platform_device_put(pdev);
>
> Are you sure this can't trigger a !kobj->state_initialized warning in
> kobject_put()?
No. The kobject is initialised in platform_device_alloc() ->
device_initialise()
> >
> > ...
> >
> > +extern struct platform_device *platform_device_register_data(struct device *,
> > + const char *, int, const void *, size_t);
>
> It's nice (IMO) to include the names of the args, for documentation
> purposes. Obviously the surrounding code didn't agree.
I just copied the surrounding code style. If you wish, I can put the
names for this function back.
BTW: would you prefer the followup "fix" patch or just the replacement
one?
--
With best wishes
Dmitry
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [PATCH] platform: add new device registration helper
2008-09-05 10:33 ` Dmitry Baryshkov
@ 2008-09-05 18:09 ` Andrew Morton
0 siblings, 0 replies; 5+ messages in thread
From: Andrew Morton @ 2008-09-05 18:09 UTC (permalink / raw)
To: Dmitry Baryshkov; +Cc: linux-kernel, gregkh
On Fri, 5 Sep 2008 14:33:18 +0400 Dmitry Baryshkov <dbaryshkov@gmail.com> wrote:
> On Thu, Sep 04, 2008 at 05:20:50PM -0700, Andrew Morton wrote:
> > On Fri, 5 Sep 2008 01:30:16 +0400
> > Dmitry Baryshkov <dbaryshkov@gmail.com> wrote:
> >
> > > Add a helper that registers simple platform_device
> > > w/o resources but with parent and device data.
> > >
> > > This is usefull to cleanup platform code from code that
> > > registers such simple devices as leds-gpio, generic-bl,
> > > etc.
> > >
> >
> > nits:
> >
> > > + *
> > > + * This function creates a simple platform device that requires minimal
> > > + * resource and memory management. Canned release function freeing memory
> > > + * allocated for the device allows drivers using such devices to be
> > > + * unloaded without waiting for the last reference to the device to be
> > > + * dropped.
> >
> > Should document the return value (tested with IS_ERR)
>
> Will the wording "Returns a valid platform device, or a valid IS_ERR
> condition containing errno" be appropriate?
Sounds good.
>
> BTW: would you prefer the followup "fix" patch or just the replacement
> one?
>
Doesn't worry me at this stage, but Greg would probably like a whole
new patch, as he probably didn't merge version 1.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2008-09-05 18:10 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-09-04 21:30 [PATCH] platform: add new device registration helper Dmitry Baryshkov
2008-09-05 0:20 ` Andrew Morton
2008-09-05 10:33 ` Dmitry Baryshkov
2008-09-05 18:09 ` Andrew Morton
2008-09-05 0:23 ` Andrew Morton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox