| Message ID | 1444262062-19247-2-git-send-email-killertofu@gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Delegated to: | Jiri Kosina |
| Headers | show |
On Wed, 7 Oct 2015, Jason Gerecke wrote: > When introduced in commit 1b5d514, the check 'if (hid_data->cc_index >= 0)' > in 'wacom_wac_finger_pre_report' was intended to switch where the driver > got the expected number of contacts from: HID_DG_CONTACTCOUNT if the usage > was present, or 'touch_max' otherwise. Unfortunately, an oversight worthy > of a brown paper bag (specifically, that 'cc_index' could never be negative) > meant that the latter 'else' clause would never be entered. > > The patch prior to this one introduced a way for 'cc_index' to be negative, > but only if HID_DG_CONTACTCOUNT is present in some report _other_ than the > one being processed. To ensure the 'else' clause is also entered for devices > which don't have HID_DG_CONTACTCOUNT on _any_ report, we add the additional > constraint that 'cc_report' be non-zero (which is true only if the usage is > present in some report). > > Signed-off-by: Jason Gerecke <jason.gerecke@wacom.com> > --- > Jiri, > > Could you please queue this patch and the prior for 4.3? They fix issues > with the implementation of my "Ignore contacts in excess of declared > contact count" patch which was pulled in as part of the 4.3 merge window. Alright, I've applied this to for-4.3/upstream-fixes, but it's now very late in the cycle, so I am not 100% sure this patch will make it into 4.3. Hence I've added -stable annotation to it so that it's picked by first 4.3-stable afterwards in such case.
diff --git a/drivers/hid/wacom_wac.c b/drivers/hid/wacom_wac.c index 4140b1d..46bd02b 100644 --- a/drivers/hid/wacom_wac.c +++ b/drivers/hid/wacom_wac.c @@ -1740,8 +1740,8 @@ static void wacom_wac_finger_pre_report(struct hid_device *hdev, } } } - - if (hid_data->cc_index >= 0) { + if (hid_data->cc_report != 0 && + hid_data->cc_index >= 0) { struct hid_field *field = report->field[hid_data->cc_index]; int value = field->value[hid_data->cc_value_index]; if (value)
When introduced in commit 1b5d514, the check 'if (hid_data->cc_index >= 0)' in 'wacom_wac_finger_pre_report' was intended to switch where the driver got the expected number of contacts from: HID_DG_CONTACTCOUNT if the usage was present, or 'touch_max' otherwise. Unfortunately, an oversight worthy of a brown paper bag (specifically, that 'cc_index' could never be negative) meant that the latter 'else' clause would never be entered. The patch prior to this one introduced a way for 'cc_index' to be negative, but only if HID_DG_CONTACTCOUNT is present in some report _other_ than the one being processed. To ensure the 'else' clause is also entered for devices which don't have HID_DG_CONTACTCOUNT on _any_ report, we add the additional constraint that 'cc_report' be non-zero (which is true only if the usage is present in some report). Signed-off-by: Jason Gerecke <jason.gerecke@wacom.com> --- Jiri, Could you please queue this patch and the prior for 4.3? They fix issues with the implementation of my "Ignore contacts in excess of declared contact count" patch which was pulled in as part of the 4.3 merge window. Thanks :) drivers/hid/wacom_wac.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)