From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?windows-1252?Q?Eisenberger_Tam=E1s?= Subject: Re: [PATCH] Input: Add support for Fujitsu S762 laptops scroll wheel Date: Mon, 27 May 2013 20:13:30 +0200 Message-ID: <51A3A24A.5060707@gmail.com> References: <1367861750.11289.3.camel@u-foka-laptop.ethome2> <519BA0AA.7080002@gmail.com> <20130521185844.GA23365@core.coreip.homeip.net> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mail-we0-f178.google.com ([74.125.82.178]:43553 "EHLO mail-we0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751017Ab3E0SNd (ORCPT ); Mon, 27 May 2013 14:13:33 -0400 Received: by mail-we0-f178.google.com with SMTP id q56so4498157wes.23 for ; Mon, 27 May 2013 11:13:32 -0700 (PDT) In-Reply-To: <20130521185844.GA23365@core.coreip.homeip.net> Sender: linux-input-owner@vger.kernel.org List-Id: linux-input@vger.kernel.org To: Dmitry Torokhov Cc: linux-input@vger.kernel.org Hi Dimitry, Thanks for your reply! On 2013-05-21 20:58, Dmitry Torokhov wrote: > Hi Tam=E1s, > > On Tue, May 21, 2013 at 06:28:26PM +0200, Eisenberger Tam=E1s wrote: >> Hello, >> >> This were my first contribution, and I received no response at all := ( >> Am I done something wrong, no one else is interested in supporting >> this particular piece of hardware or what? >> >> If anyone has a minute please review it. >> > > Ah, yes, sorry... > >> Thanks: >> Tam=E1s >> >> On 2013-05-06 19:35, Eisenberger Tam=E1s wrote: >>> Detects and makes the Touch Scroll Wheel found on some Fujitsu lapt= ops >>> working. >>> >>> The detection is based on the (hopefully) unique E7 report of the w= heel >>> device. >>> Up and down scrolling is detected by the only byte that has differe= nt >>> values in the devices packets, but it should be fine since it only = able >>> to report that two events. >>> >>> Signed-off-by: Tama's Eisenberger >>> Index: linux/drivers/input/mouse/alps.c >>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >>> --- linux.orig/drivers/input/mouse/alps.c >>> +++ linux/drivers/input/mouse/alps.c >>> @@ -410,6 +410,17 @@ static void alps_process_trackstick_pack >>> if (packet[1] =3D=3D 0x7f && packet[2] =3D=3D 0x7f && packet[4] = =3D=3D 0x7f) >>> return; >>> >>> + if (priv->quirks & ALPS_QUIRK_SCROLL_WHEEL) { >>> + if (packet[3] =3D=3D 0x4c) >>> + input_report_rel(dev, REL_WHEEL, 1); >>> + >>> + if (packet[3] =3D=3D 0x58) >>> + input_report_rel(dev, REL_WHEEL, -1); > > Are these really the only values that emitted by the wheel? Have you > tried spinning it vigorously to see if it produces different results? I've tried to scroll really slow, and really fast with it, also tryed t= o=20 move my finger diagonally and randomly but only that two were sent. Als= o=20 events are really rare when not moving the finger around the edge. > >>> + >>> + input_sync(dev); >>> + return; > > Why are we ignoring the rest of the packet in this case? Does it cont= ain > invalid data? Without my modifications the original driver reports middle button=20 events when I scroll upwards, so I have to return here to prevent this.= =20 I don't think that this scroll well can co-exists with a trackstick. > >>> + } >>> + >>> x =3D (s8)(((packet[0] & 0x20) << 2) | (packet[1] & 0x7f)); >>> y =3D (s8)(((packet[0] & 0x10) << 3) | (packet[2] & 0x7f)); >>> z =3D (packet[4] & 0x7c) >> 2; >>> @@ -1255,6 +1266,7 @@ error: >>> static int alps_setup_trackstick_v3(struct psmouse *psmouse, int >>> reg_base) >>> { >>> struct ps2dev *ps2dev =3D &psmouse->ps2dev; >>> + struct alps_data *priv =3D psmouse->private; >>> int ret =3D 0; >>> unsigned char param[4]; >>> >>> @@ -1277,6 +1289,12 @@ static int alps_setup_trackstick_v3(stru >>> psmouse_dbg(psmouse, "trackstick E7 report: %3ph\n", param); >>> >>> /* >>> + * Detect fujitsu scroll wheel device >>> + */ >>> + if (param[0] =3D=3D 0x34 && param[1] =3D=3D 0x01 && param[2] =3D= =3D 0x14) >>> + priv->quirks |=3D ALPS_QUIRK_SCROLL_WHEEL; >>> + > > I believe this quirk should go into alps_model_data[] table. I've tried to do that but it seems that the exact same (as my primary=20 touchpad) signature is already in the alps_model_data[] table (E7=3D73 = 02=20 64, EC=3D88 07 9d) and the secondary device seems to be independent. Th= is=20 is why I'm trying to use the EC report of the secondary device but it=20 seemed to be overkill to create a new model_data array for attached=20 devices or I don't know what could be the consequences to use the same=20 array also for secondary devices. > >>> + /* >>> * Not sure what this does, but it is absolutely >>> * essential. Without it, the touchpad does not >>> * work at all and the trackstick just emits normal >>> @@ -1798,12 +1816,16 @@ int alps_init(struct psmouse *psmouse) >>> dev2->id.product =3D PSMOUSE_ALPS; >>> dev2->id.version =3D 0x0000; >>> dev2->dev.parent =3D &psmouse->ps2dev.serio->dev; >>> - >>> dev2->evbit[0] =3D BIT_MASK(EV_KEY) | BIT_MASK(EV_REL); >>> dev2->relbit[BIT_WORD(REL_X)] =3D BIT_MASK(REL_X) | BIT_MASK(REL= _Y); >>> dev2->keybit[BIT_WORD(BTN_LEFT)] =3D >>> BIT_MASK(BTN_LEFT) | BIT_MASK(BTN_MIDDLE) | BIT_MASK(BTN_RIGHT)= ; >>> >>> + if (priv->quirks & ALPS_QUIRK_SCROLL_WHEEL) { >>> + dev2->name =3D "DualPoint Scroll Wheel"; > > I'd keep the name as is. No prob. I'l remove the name change. > >>> + dev2->relbit[BIT_WORD(REL_X)] |=3D BIT_MASK(REL_WHEEL); >>> + } >>> + >>> if (input_register_device(priv->dev2)) >>> goto init_fail; >>> >>> Index: linux/drivers/input/mouse/alps.h >>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D >>> --- linux.orig/drivers/input/mouse/alps.h >>> +++ linux/drivers/input/mouse/alps.h >>> @@ -158,6 +158,7 @@ struct alps_data { >>> }; >>> >>> #define ALPS_QUIRK_TRACKSTICK_BUTTONS 1 /* trakcstick buttons in >>> trackstick packet */ >>> +#define ALPS_QUIRK_SCROLL_WHEEL 2 /* secondary device is scroll w= heel >>> */ >>> >>> #ifdef CONFIG_MOUSE_PS2_ALPS >>> int alps_detect(struct psmouse *psmouse, bool set_properties); >>> > > BTW, your mailer line-wrapped the patch so it is not possible to appl= y, > when resending please make sure the long lines stay intact. You're right, I=92ll try to use an other one. > > Thanks. > So basically as I see the main concern is the introduction of a new=20 quirk, but I don't know how to do this using the alps_model_data[] if=20 you have any suggestions please tell me! Thanks: Tam=E1s -- 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