Hello, Dmitry.
On Wed, Oct 31, 2012 at 02:37:01PM -0700, Dmitry Torokhov wrote:
> Not all drivers use devres for rest of their resources so it makes sense
> to have unmanaged versions (like we have request_irq/devm_request_irq).
> Besides:
>
> [dtor@dtor-ws vmci]$ grep -r input_allocate_device
On Wednesday, October 31, 2012 02:05:32 PM Tejun Heo wrote:
> Hello, Dmitry.
>
> On Mon, Oct 22, 2012 at 10:35:14PM -0700, Dmitry Torokhov wrote:
> > There is a demand from driver's writers to use managed devices framework
> > for their drivers. Unfortunately up to this moment input devices did
Hello, Dmitry.
On Mon, Oct 22, 2012 at 10:35:14PM -0700, Dmitry Torokhov wrote:
> There is a demand from driver's writers to use managed devices framework
> for their drivers. Unfortunately up to this moment input devices did not
> provide support for managed devices and that lead to mixing two
Hello, Dmitry.
On Mon, Oct 22, 2012 at 10:35:14PM -0700, Dmitry Torokhov wrote:
There is a demand from driver's writers to use managed devices framework
for their drivers. Unfortunately up to this moment input devices did not
provide support for managed devices and that lead to mixing two
On Wednesday, October 31, 2012 02:05:32 PM Tejun Heo wrote:
Hello, Dmitry.
On Mon, Oct 22, 2012 at 10:35:14PM -0700, Dmitry Torokhov wrote:
There is a demand from driver's writers to use managed devices framework
for their drivers. Unfortunately up to this moment input devices did not
Hello, Dmitry.
On Wed, Oct 31, 2012 at 02:37:01PM -0700, Dmitry Torokhov wrote:
Not all drivers use devres for rest of their resources so it makes sense
to have unmanaged versions (like we have request_irq/devm_request_irq).
Besides:
[dtor@dtor-ws vmci]$ grep -r input_allocate_device
On Monday, October 29, 2012 10:32:54 PM Henrik Rydberg wrote:
> > > Either way, the code looks functional to me.
> >
> > So is that "reviewed-by"?
>
> I was thinking about this hunk:
> > @@ -1972,7 +2084,20 @@ int input_register_device(struct input_dev *dev)
> >
> >
> > Either way, the code looks functional to me.
>
> So is that "reviewed-by"?
I was thinking about this hunk:
> @@ -1972,7 +2084,20 @@ int input_register_device(struct input_dev *dev)
>
> mutex_unlock(_mutex);
>
> + if (dev->devres_managed) {
> +
On Mon, Oct 29, 2012 at 09:02:26PM +0100, Henrik Rydberg wrote:
> > > > @@ -1766,8 +1830,14 @@ EXPORT_SYMBOL(input_allocate_device);
> > > > */
> > > > void input_free_device(struct input_dev *dev)
> > > > {
> > > > - if (dev)
> > > > + if (dev) {
> > > > + if
> > > @@ -1766,8 +1830,14 @@ EXPORT_SYMBOL(input_allocate_device);
> > > */
> > > void input_free_device(struct input_dev *dev)
> > > {
> > > - if (dev)
> > > + if (dev) {
> > > + if (dev->devres_managed)
> > > + WARN_ON(devres_destroy(dev->dev.parent,
> > > +
Hi Henrik,
On Mon, Oct 29, 2012 at 07:22:53PM +0100, Henrik Rydberg wrote:
> Hi Dmitry Torokhov wrote:
>
> > There is a demand from driver's writers to use managed devices framework
> > for their drivers. Unfortunately up to this moment input devices did not
> > provide support for managed
Hi Dmitry Torokhov wrote:
> There is a demand from driver's writers to use managed devices framework
> for their drivers. Unfortunately up to this moment input devices did not
> provide support for managed devices and that lead to mixing two styles
> of resource management which usually
Either way, the code looks functional to me.
So is that reviewed-by?
I was thinking about this hunk:
@@ -1972,7 +2084,20 @@ int input_register_device(struct input_dev *dev)
mutex_unlock(input_mutex);
+ if (dev-devres_managed) {
+ dev_info(dev-dev.parent,
On Monday, October 29, 2012 10:32:54 PM Henrik Rydberg wrote:
Either way, the code looks functional to me.
So is that reviewed-by?
I was thinking about this hunk:
@@ -1972,7 +2084,20 @@ int input_register_device(struct input_dev *dev)
mutex_unlock(input_mutex);
+
Hi Dmitry Torokhov wrote:
There is a demand from driver's writers to use managed devices framework
for their drivers. Unfortunately up to this moment input devices did not
provide support for managed devices and that lead to mixing two styles
of resource management which usually introduced
Hi Henrik,
On Mon, Oct 29, 2012 at 07:22:53PM +0100, Henrik Rydberg wrote:
Hi Dmitry Torokhov wrote:
There is a demand from driver's writers to use managed devices framework
for their drivers. Unfortunately up to this moment input devices did not
provide support for managed devices and
@@ -1766,8 +1830,14 @@ EXPORT_SYMBOL(input_allocate_device);
*/
void input_free_device(struct input_dev *dev)
{
- if (dev)
+ if (dev) {
+ if (dev-devres_managed)
+ WARN_ON(devres_destroy(dev-dev.parent,
+
On Mon, Oct 29, 2012 at 09:02:26PM +0100, Henrik Rydberg wrote:
@@ -1766,8 +1830,14 @@ EXPORT_SYMBOL(input_allocate_device);
*/
void input_free_device(struct input_dev *dev)
{
- if (dev)
+ if (dev) {
+ if (dev-devres_managed)
+
There is a demand from driver's writers to use managed devices framework
for their drivers. Unfortunately up to this moment input devices did not
provide support for managed devices and that lead to mixing two styles
of resource management which usually introduced more bugs, such as
manually
There is a demand from driver's writers to use managed devices framework
for their drivers. Unfortunately up to this moment input devices did not
provide support for managed devices and that lead to mixing two styles
of resource management which usually introduced more bugs, such as
manually
20 matches
Mail list logo