All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Input: sparse-keymap - add managed version of sparse_keymap_setup()
@ 2017-02-28  9:45 Michał Kępień
  2017-02-28 18:45 ` Dmitry Torokhov
  0 siblings, 1 reply; 4+ messages in thread
From: Michał Kępień @ 2017-02-28  9:45 UTC (permalink / raw)
  To: Dmitry Torokhov; +Cc: linux-input, linux-kernel

Some platform drivers use devm_input_allocate_device() together with
sparse_keymap_setup() in their .probe callbacks.  While using the former
simplifies error handling, using the latter necessitates calling
sparse_keymap_free() in the error path and upon module unloading to
avoid leaking the copy of the keymap allocated by sparse_keymap_setup().

To help prevent such leaks and enable simpler error handling in these
drivers, add a new function which allows automatic freeing of the keymap
copy upon probe failure and on driver detach.

As devm_input_allocate_device() adds its devres to the device owning the
input device, we do the same for managed input devices to ensure freeing
the keymap copy is properly slotted in the devres stack.

The new function can also be used by non-managed input devices, though
in this case the devres is attached to the struct device embedded inside
the input device itself.

Signed-off-by: Michał Kępień <kernel@kempniu.pl>
---
 drivers/input/sparse-keymap.c       | 51 +++++++++++++++++++++++++++++++++++++
 include/linux/input/sparse-keymap.h |  4 +++
 2 files changed, 55 insertions(+)

diff --git a/drivers/input/sparse-keymap.c b/drivers/input/sparse-keymap.c
index e7409c45bdd0..9cb37762f5fd 100644
--- a/drivers/input/sparse-keymap.c
+++ b/drivers/input/sparse-keymap.c
@@ -228,6 +228,57 @@ int sparse_keymap_setup(struct input_dev *dev,
 }
 EXPORT_SYMBOL(sparse_keymap_setup);
 
+struct sparse_keymap_devres {
+	struct input_dev *dev;
+};
+
+static void devm_sparse_keymap_free(struct device *dev, void *res)
+{
+	struct sparse_keymap_devres *devres = res;
+
+	sparse_keymap_free(devres->dev);
+}
+
+/**
+ * devm_sparse_keymap_setup - set up managed sparse keymap for an input device
+ * @dev: Input device
+ * @keymap: Keymap in form of array of &key_entry structures ending
+ *	with %KE_END type entry
+ * @setup: Function that can be used to adjust keymap entries
+ *	depending on device's needs, may be %NULL
+ *
+ * The function calculates size and allocates copy of the original
+ * keymap after which sets up input device event bits appropriately.
+ * The allocated copy of the keymap is automatically freed when it is no
+ * longer needed, thus drivers using this function must not explicitly
+ * call sparse_keymap_free() for the same input device.
+ */
+int devm_sparse_keymap_setup(struct input_dev *dev,
+			     const struct key_entry *keymap,
+			     int (*setup)(struct input_dev *,
+					  struct key_entry *))
+{
+	struct sparse_keymap_devres *devres;
+	int ret;
+
+	devres = devres_alloc(devm_sparse_keymap_free, sizeof(*devres),
+			      GFP_KERNEL);
+	if (!devres)
+		return -ENOMEM;
+
+	ret = sparse_keymap_setup(dev, keymap, setup);
+	if (ret) {
+		devres_free(devres);
+		return ret;
+	}
+
+	devres->dev = dev;
+	devres_add(dev->devres_managed ? dev->dev.parent : &dev->dev, devres);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(devm_sparse_keymap_setup);
+
 /**
  * sparse_keymap_free - free memory allocated for sparse keymap
  * @dev: Input device using sparse keymap
diff --git a/include/linux/input/sparse-keymap.h b/include/linux/input/sparse-keymap.h
index 52db62064c6e..bb8cc7bd6a29 100644
--- a/include/linux/input/sparse-keymap.h
+++ b/include/linux/input/sparse-keymap.h
@@ -51,6 +51,10 @@ struct key_entry *sparse_keymap_entry_from_keycode(struct input_dev *dev,
 int sparse_keymap_setup(struct input_dev *dev,
 			const struct key_entry *keymap,
 			int (*setup)(struct input_dev *, struct key_entry *));
+int devm_sparse_keymap_setup(struct input_dev *dev,
+			     const struct key_entry *keymap,
+			     int (*setup)(struct input_dev *,
+					  struct key_entry *));
 void sparse_keymap_free(struct input_dev *dev);
 
 void sparse_keymap_report_entry(struct input_dev *dev, const struct key_entry *ke,
-- 
2.12.0

^ permalink raw reply related	[flat|nested] 4+ messages in thread

* Re: [PATCH] Input: sparse-keymap - add managed version of sparse_keymap_setup()
  2017-02-28  9:45 [PATCH] Input: sparse-keymap - add managed version of sparse_keymap_setup() Michał Kępień
@ 2017-02-28 18:45 ` Dmitry Torokhov
  2017-03-01 10:37   ` Michał Kępień
  0 siblings, 1 reply; 4+ messages in thread
From: Dmitry Torokhov @ 2017-02-28 18:45 UTC (permalink / raw)
  To: Michał Kępień; +Cc: linux-input, linux-kernel

On Tue, Feb 28, 2017 at 10:45:25AM +0100, Michał Kępień wrote:
> Some platform drivers use devm_input_allocate_device() together with
> sparse_keymap_setup() in their .probe callbacks.  While using the former
> simplifies error handling, using the latter necessitates calling
> sparse_keymap_free() in the error path and upon module unloading to
> avoid leaking the copy of the keymap allocated by sparse_keymap_setup().
> 
> To help prevent such leaks and enable simpler error handling in these
> drivers, add a new function which allows automatic freeing of the keymap
> copy upon probe failure and on driver detach.
> 
> As devm_input_allocate_device() adds its devres to the device owning the
> input device, we do the same for managed input devices to ensure freeing
> the keymap copy is properly slotted in the devres stack.
> 
> The new function can also be used by non-managed input devices, though
> in this case the devres is attached to the struct device embedded inside
> the input device itself.

This is wrong and does not work as input devices are never probed and
never unbound, so the cleanup will never happen.

Either pass device explicitly, or always take input's parent.

Thanks.

-- 
Dmitry

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Input: sparse-keymap - add managed version of sparse_keymap_setup()
  2017-02-28 18:45 ` Dmitry Torokhov
@ 2017-03-01 10:37   ` Michał Kępień
  2017-03-01 18:19     ` Dmitry Torokhov
  0 siblings, 1 reply; 4+ messages in thread
From: Michał Kępień @ 2017-03-01 10:37 UTC (permalink / raw)
  To: Dmitry Torokhov; +Cc: linux-input, linux-kernel

> On Tue, Feb 28, 2017 at 10:45:25AM +0100, Michał Kępień wrote:
> > Some platform drivers use devm_input_allocate_device() together with
> > sparse_keymap_setup() in their .probe callbacks.  While using the former
> > simplifies error handling, using the latter necessitates calling
> > sparse_keymap_free() in the error path and upon module unloading to
> > avoid leaking the copy of the keymap allocated by sparse_keymap_setup().
> > 
> > To help prevent such leaks and enable simpler error handling in these
> > drivers, add a new function which allows automatic freeing of the keymap
> > copy upon probe failure and on driver detach.
> > 
> > As devm_input_allocate_device() adds its devres to the device owning the
> > input device, we do the same for managed input devices to ensure freeing
> > the keymap copy is properly slotted in the devres stack.
> > 
> > The new function can also be used by non-managed input devices, though
> > in this case the devres is attached to the struct device embedded inside
> > the input device itself.
> 
> This is wrong and does not work as input devices are never probed and
> never unbound, so the cleanup will never happen.
> 
> Either pass device explicitly, or always take input's parent.

Thank you for taking a look, Dmitry.  I may be missing something, but
please bear with me.  AFAICT, in the non-managed input device case the
devres release callback is called when the last reference to that input
device is dropped, i.e. after input_unregister_device() is called.
Setting devres.log=1 confirms this, so does putting a WARN_ON() inside
devm_sparse_keymap_free().  Could you please elaborate?  Perhaps I am
misunderstanding your argument.

Thanks,

-- 
Best regards,
Michał Kępień

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Input: sparse-keymap - add managed version of sparse_keymap_setup()
  2017-03-01 10:37   ` Michał Kępień
@ 2017-03-01 18:19     ` Dmitry Torokhov
  0 siblings, 0 replies; 4+ messages in thread
From: Dmitry Torokhov @ 2017-03-01 18:19 UTC (permalink / raw)
  To: Michał Kępień; +Cc: linux-input, linux-kernel

On Wed, Mar 01, 2017 at 11:37:14AM +0100, Michał Kępień wrote:
> > On Tue, Feb 28, 2017 at 10:45:25AM +0100, Michał Kępień wrote:
> > > Some platform drivers use devm_input_allocate_device() together with
> > > sparse_keymap_setup() in their .probe callbacks.  While using the former
> > > simplifies error handling, using the latter necessitates calling
> > > sparse_keymap_free() in the error path and upon module unloading to
> > > avoid leaking the copy of the keymap allocated by sparse_keymap_setup().
> > > 
> > > To help prevent such leaks and enable simpler error handling in these
> > > drivers, add a new function which allows automatic freeing of the keymap
> > > copy upon probe failure and on driver detach.
> > > 
> > > As devm_input_allocate_device() adds its devres to the device owning the
> > > input device, we do the same for managed input devices to ensure freeing
> > > the keymap copy is properly slotted in the devres stack.
> > > 
> > > The new function can also be used by non-managed input devices, though
> > > in this case the devres is attached to the struct device embedded inside
> > > the input device itself.
> > 
> > This is wrong and does not work as input devices are never probed and
> > never unbound, so the cleanup will never happen.
> > 
> > Either pass device explicitly, or always take input's parent.
> 
> Thank you for taking a look, Dmitry.  I may be missing something, but
> please bear with me.  AFAICT, in the non-managed input device case the
> devres release callback is called when the last reference to that input
> device is dropped, i.e. after input_unregister_device() is called.
> Setting devres.log=1 confirms this, so does putting a WARN_ON() inside
> devm_sparse_keymap_free().  Could you please elaborate?  Perhaps I am
> misunderstanding your argument.

Ah, I missed the fact that we drop devres resources when we destroy the
device. OK, in this case why don't we make sparse keymap always use
devm-allocated memory and make sparse_keymap_free() a noop for the time
being? Once we remove it from all the drivers we can kill the stub.

Thanks.

-- 
Dmitry

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2017-03-01 18:30 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-02-28  9:45 [PATCH] Input: sparse-keymap - add managed version of sparse_keymap_setup() Michał Kępień
2017-02-28 18:45 ` Dmitry Torokhov
2017-03-01 10:37   ` Michał Kępień
2017-03-01 18:19     ` Dmitry Torokhov

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.