On Wed, Dec 7, 2011 at 2:16 PM, Jason Gerecke <[email protected]> wrote: > This patch breaks the Graphire 4. The X driver appears to get confused > about what buttons are being pressed. This happens with all tools, > from what I can tell. Only the right mouse button appears to work > correctly, and can be used to get the driver back in sync with reality > until you make another non-right click.
I think I see the bug in patch. I'll point it out below and try to send an updated patch at another time. > > Jason > > --- > Day xee-nee-svsh duu-'ushtlh-ts'it; > nuu-wee-ya' duu-xan' 'vm-nvshtlh-ts'it. > Huu-chan xuu naa~-gha. > > > > On Thu, Dec 1, 2011 at 8:02 PM, <[email protected]> wrote: >> From: Chris Bagwell <[email protected]> >> >> Don't work so hard to not report events. Let input event >> filtering do its thing for us. Makes it easier to follow >> data parsing vs. event sending features. >> >> Left in event filtering for overlapping events and had >> a purpose for preventing events. >> >> Signed-off-by: Chris Bagwell <[email protected]> >> --- >> drivers/input/tablet/wacom_wac.c | 141 >> ++++++++++++++++++++++--------------- >> 1 files changed, 84 insertions(+), 57 deletions(-) >> >> diff --git a/drivers/input/tablet/wacom_wac.c >> b/drivers/input/tablet/wacom_wac.c >> index 7b69a21..7acc87b 100644 >> --- a/drivers/input/tablet/wacom_wac.c >> +++ b/drivers/input/tablet/wacom_wac.c >> @@ -206,9 +206,10 @@ static int wacom_graphire_irq(struct wacom_wac *wacom) >> struct wacom_features *features = &wacom->features; >> unsigned char *data = wacom->data; >> struct input_dev *input = wacom->input; >> - int prox; >> - int rw = 0; >> int retval = 0; >> + int prox, x = 0, y = 0, p = 0, d = 0, pen = 0, btn1 = 0, btn2 = 0; >> + int btnl = 0, btnm = 0, btnr = 0, btnb = 0, btnf = 0; >> + int rw = 0, aw = 0; >> >> if (data[0] != WACOM_REPORT_PENABLED) { >> dbg("wacom_graphire_irq: received unknown report #%d", >> data[0]); >> @@ -216,86 +217,112 @@ static int wacom_graphire_irq(struct wacom_wac *wacom) >> } >> >> prox = data[1] & 0x80; >> - if (prox || wacom->id[0]) { >> - if (prox) { >> - switch ((data[1] >> 5) & 3) { >> + if (prox) { >> + switch ((data[1] >> 5) & 3) { >> >> - case 0: /* Pen */ >> - wacom->tool[0] = BTN_TOOL_PEN; >> - wacom->id[0] = STYLUS_DEVICE_ID; >> - break; >> + case 0: /* Pen */ >> + wacom->tool[0] = BTN_TOOL_PEN; >> + break; >> >> - case 1: /* Rubber */ >> - wacom->tool[0] = BTN_TOOL_RUBBER; >> - wacom->id[0] = ERASER_DEVICE_ID; >> - break; >> + case 1: /* Rubber */ >> + wacom->tool[0] = BTN_TOOL_RUBBER; >> + break; >> >> - case 2: /* Mouse with wheel */ >> - input_report_key(input, BTN_MIDDLE, data[1] >> & 0x04); >> - /* fall through */ >> + case 2: /* Mouse with wheel */ >> + btnm = data[1] & 0x04; >> + /* fall through */ >> >> - case 3: /* Mouse without wheel */ >> - wacom->tool[0] = BTN_TOOL_MOUSE; >> - wacom->id[0] = CURSOR_DEVICE_ID; >> - break; >> - } >> + case 3: /* Mouse without wheel */ >> + wacom->tool[0] = BTN_TOOL_MOUSE; >> + break; >> } >> - input_report_abs(input, ABS_X, le16_to_cpup((__le16 >> *)&data[2])); >> - input_report_abs(input, ABS_Y, le16_to_cpup((__le16 >> *)&data[4])); >> + x = le16_to_cpup((__le16 *)&data[2]); >> + y = le16_to_cpup((__le16 *)&data[4]); >> if (wacom->tool[0] != BTN_TOOL_MOUSE) { >> - input_report_abs(input, ABS_PRESSURE, data[6] | >> ((data[7] & 0x01) << 8)); >> - input_report_key(input, BTN_TOUCH, data[1] & 0x01); >> - input_report_key(input, BTN_STYLUS, data[1] & 0x02); >> - input_report_key(input, BTN_STYLUS2, data[1] & 0x04); >> + p = data[6] | ((data[7] & 0x01) << 8); >> + pen = data[1] & 0x01; >> + btn1 = data[1] & 0x02; >> + btn2 = data[1] & 0x04; >> } else { >> - input_report_key(input, BTN_LEFT, data[1] & 0x01); >> - input_report_key(input, BTN_RIGHT, data[1] & 0x02); >> + btnl = data[1] & 0x01; >> + btnr = data[1] & 0x02; >> if (features->type == WACOM_G4 || >> features->type == WACOM_MO) { >> - input_report_abs(input, ABS_DISTANCE, >> data[6] & 0x3f); >> + d = data[6] & 0x3f; >> rw = (data[7] & 0x04) - (data[7] & 0x03); >> } else { >> - input_report_abs(input, ABS_DISTANCE, >> data[7] & 0x3f); >> + d = data[7] & 0x3f; >> rw = -(signed char)data[6]; >> } >> - input_report_rel(input, REL_WHEEL, rw); >> } >> + } >> >> - if (!prox) >> - wacom->id[0] = 0; >> - input_report_key(input, wacom->tool[0], prox); >> - input_sync(input); /* sync last event */ >> + input_report_abs(input, ABS_X, x); >> + input_report_abs(input, ABS_Y, y); >> + input_report_abs(input, ABS_PRESSURE, p); >> + input_report_key(input, BTN_TOUCH, pen); >> + input_report_key(input, BTN_STYLUS, btn1); >> + input_report_key(input, BTN_STYLUS2, btn2); >> + input_report_abs(input, ABS_DISTANCE, d); >> + input_report_rel(input, REL_WHEEL, rw); >> + input_report_key(input, BTN_MIDDLE, btnm); >> + >> + /* Buttons on the mouse puck are sending the same events as >> + * buttons located on the tablet. >> + * >> + * The following is to only send these overlapping events >> + * when pressed and once more upon release so that the >> + * two sources are not fighting each other. >> + * >> + * It would be odd for user to click the mouse puck and tablet >> + * buttons at same time. >> + */ >> + if (btnl || btnr || wacom->id[0]) { >> + input_report_key(input, BTN_LEFT, btnl); >> + input_report_key(input, BTN_RIGHT, btnr); >> + wacom->id[0] = 0; >> } That should be something like this. Without it, we were never sending the button release event. if (btnl || btnr || wacom->id[0]) { wacom->id[0] = CURSOR_DEVICE_ID; input_report_key(input, BTN_LEFT, btnl); input_report_key(input, BTN_RIGHT, btnr); if (wacom->id[0]) wacom->id[0] = 0; } Could be more bugs lurking, I'm sure. Chris >> >> + input_report_key(input, wacom->tool[0], prox); >> + input_sync(input); /* sync last event */ >> + >> /* send pad data */ >> switch (features->type) { >> case WACOM_G4: >> - prox = data[7] & 0xf8; >> - if (prox || wacom->id[1]) { >> - wacom->id[1] = PAD_DEVICE_ID; >> - input_report_key(input, BTN_BACK, (data[7] & 0x40)); >> - input_report_key(input, BTN_FORWARD, (data[7] & >> 0x80)); >> - rw = ((data[7] & 0x18) >> 3) - ((data[7] & 0x20) >> >> 3); >> - input_report_rel(input, REL_WHEEL, rw); >> - if (!prox) >> - wacom->id[1] = 0; >> - retval = 1; >> - } >> + btnb = (data[7] & 0x40); >> + btnf = (data[7] & 0x80); >> + aw = ((data[7] & 0x18) >> 3) - ((data[7] & 0x20) >> 3); >> + input_report_key(input, BTN_BACK, (data[7] & 0x40)); >> + input_report_key(input, BTN_FORWARD, (data[7] & 0x80)); >> + input_report_rel(input, REL_WHEEL, aw); >> + retval = 1; >> + >> break; >> >> case WACOM_MO: >> - prox = (data[7] & 0xf8) || data[8]; >> - if (prox || wacom->id[1]) { >> + btnb = (data[7] & 0x08); >> + btnf = (data[7] & 0x10); >> + aw = (data[8] & 0x7f); >> + input_report_key(input, BTN_BACK, btnb); >> + input_report_key(input, BTN_FORWARD, btnf); >> + input_report_abs(input, ABS_WHEEL, aw); >> + >> + /* See comment for mouse puck buttons above */ >> + btnl = 0; >> + btnr = 0; >> + prox = data[7] & 0x60; >> + if (prox) { >> + btnl = (data[7] & 0x20); >> + btnr = (data[7] & 0x40); >> wacom->id[1] = PAD_DEVICE_ID; >> - input_report_key(input, BTN_BACK, (data[7] & 0x08)); >> - input_report_key(input, BTN_LEFT, (data[7] & 0x20)); >> - input_report_key(input, BTN_FORWARD, (data[7] & >> 0x10)); >> - input_report_key(input, BTN_RIGHT, (data[7] & 0x40)); >> - input_report_abs(input, ABS_WHEEL, (data[8] & 0x7f)); >> - if (!prox) >> - wacom->id[1] = 0; >> - retval = 1; >> + }; >> + if (prox || wacom->id[1]) { >> + input_report_key(input, BTN_LEFT, btnl); >> + input_report_key(input, BTN_RIGHT, btnr); >> + wacom->id[1] = 0; >> } >> + retval = 1; >> + >> break; >> } >> exit: >> -- >> 1.7.7.3 >> >> >> ------------------------------------------------------------------------------ >> All the data continuously generated in your IT infrastructure >> contains a definitive record of customers, application performance, >> security threats, fraudulent activity, and more. Splunk takes this >> data and makes sense of it. IT sense. And common sense. >> http://p.sf.net/sfu/splunk-novd2d >> _______________________________________________ >> Linuxwacom-devel mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel ------------------------------------------------------------------------------ Cloud Services Checklist: Pricing and Packaging Optimization This white paper is intended to serve as a reference, checklist and point of discussion for anyone considering optimizing the pricing and packaging model of a cloud services business. Read Now! http://www.accelacomm.com/jaw/sfnl/114/51491232/ _______________________________________________ Linuxwacom-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/linuxwacom-devel
