Re: [RFC 1/2] input: mt: Add method to extract the MT slot state

Hi Henrik,

On Thu, Jan 27, 2011 at 13:06, Henrik Rydberg [off-list ref] wrote:

Hi Benjamin,

On Thu, Jan 27, 2011 at 11:35:46AM +0100, Benjamin Tissoires wrote:

quoted

This patch adds the function input_mt_get_abs_value(), which may be

Ehm, input_mt_get_absinfo?

In fact, the idea was to put this on the table again ;). I have no
idea of what is the right think to do in such case.
So I'll take all comments.

quoted

used to extract the current state of the MT slots.

Signed-off-by: Henrik Rydberg <redacted>

I had rather re-add this myself, thanks.

That was my question. Plus I had a subliminal message in which I
intend you to take the drive on this work. ;)

quoted

Signed-off-by: Benjamin Tissoires <redacted>
---
 drivers/input/input-mt.c |   32 ++++++++++++++++++++++++++++++++
 include/linux/input/mt.h |    1 +
 2 files changed, 33 insertions(+), 0 deletions(-)

diff --git a/drivers/input/input-mt.c b/drivers/input/input-mt.c
index c48c81f..5b95e38 100644
--- a/drivers/input/input-mt.c
+++ b/drivers/input/input-mt.c

@@ -168,3 +168,35 @@ void input_mt_report_pointer_emulation(struct input_dev *dev, bool use_count)

      }
 }
 EXPORT_SYMBOL(input_mt_report_pointer_emulation);
+
+/**
+ * input_mt_get_absinfo - retrieve MT state variables from a device slot
+ * @dev: input device which state is being queried
+ * @code: ABS code to retrieve
+ * @slot: slot to retrieve from
+ *
+ * Return the ABS state of the given MT slot. If code is not an MT
+ * event, the corresponding ABS event is returned.
+ *
+ * This function may be called by anyone interested in extracting the
+ * current MT state. Used by evdev handlers.
+ */
+struct input_absinfo input_mt_get_absinfo(struct input_dev *dev, unsigned int code, int slot)

Why return the struct? Give it a pointer instead, so the return value
can be used for errors.

Quick and dirty think. Maybe the old implementation is better.

Also, I wonder about the usefulness of extracting the absinfo here -
only the value depends on the slot argument, and absinfo does not
require the same lock protection. Why not keep the layout from the
original patch?

same above

quoted

+{
+     unsigned int mtmap = code > ABS_MT_FIRST ? code - ABS_MT_FIRST : 0;

Should rather return -EINVAL on bad argument.

I did not know how to handle the original code:
"unsigned int mtmap = input_mt_abs_map[code];"

Now I know.

quoted

+     unsigned long flags;
+     struct input_absinfo retval;
+
+     spin_lock_irqsave(&dev->event_lock, flags);
+     if (!mtmap)
+             retval = dev->absinfo[code];
+     else if (slot >= 0 && slot < dev->mtsize) {
+             retval = dev->absinfo[code];
+             retval.value = dev->mt[slot].abs[mtmap];

Please use input_mt_get_value().

This is a problem when working with code that was send 6 months ago.

quoted

+     } else
+             retval.value = 0;
+     spin_unlock_irqrestore(&dev->event_lock, flags);
+
+     return retval;
+}
+EXPORT_SYMBOL(input_mt_get_absinfo);

diff --git a/include/linux/input/mt.h b/include/linux/input/mt.h
index b3ac06a..a71a27f 100644
--- a/include/linux/input/mt.h
+++ b/include/linux/input/mt.h

@@ -54,4 +54,5 @@ void input_mt_report_slot_state(struct input_dev *dev,

 void input_mt_report_finger_count(struct input_dev *dev, int count);
 void input_mt_report_pointer_emulation(struct input_dev *dev, bool use_count);

+struct input_absinfo input_mt_get_absinfo(struct input_dev *dev, unsigned int code, int slot);
 #endif

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

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