* Re: Elo touch panel [not found] ` <20060207081415.GA6731@suse.cz> @ 2006-02-07 23:41 ` Shaun Jackman 2006-02-08 8:51 ` Vojtech Pavlik 0 siblings, 1 reply; 2+ messages in thread From: Shaun Jackman @ 2006-02-07 23:41 UTC (permalink / raw) To: Vojtech Pavlik; +Cc: lkml On 2/7/06, Vojtech Pavlik <vojtech@suse.cz> wrote: > On Mon, Feb 06, 2006 at 07:30:26PM -0700, Shaun Jackman wrote: ... > > I then run... > > $ inputattach -elo /dev/ttyS3 > > .. at which point dmesg says... > > serio: Serial port ttyS3 > > but it never gives a message indicating that the Elo driver has been > > attached to ttyS3. Is there anything else I must configure? > > Try to look at /proc/bus/input/devices. The elo_connect() function > should get called at the point the port is created. Unfortunately, I > never could test this driver, because I don't have any touchscreens > anymore (ever had one and it broke when I was moving). So there may be > problems in the driver. But I believe others had some success running > it. > > Looking at the driver, one will probably need to set the panel type in > inputattach ... There's a bug in inputattach. The SERIO_RS232 constant is completely mucking the bits that specify the protocol. --- inputattach.c- 2006-02-07 14:37:04.000000000 -0700 +++ inputattach.c 2006-02-07 16:22:07.000000000 -0700 @@ -455,7 +455,7 @@ return 1; } - devt = SERIO_RS232 | input_types[type].type | (id << 8) | (extra << 16); + devt = input_types[type].type | (id << 8) | (extra << 16); if(ioctl(fd, SPIOCSTYPE, &devt)) { fprintf(stderr, "inputattach: can't set device type\n"); Cheers, Shaun APPENDIX A You probably know this better than I do, but the .type above is misnamed. It should really be .proto. It's impossible, as far as I can tell, to specify the type (like SERIO_RS232) with a SPIOCSTYPE call. ^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Elo touch panel 2006-02-07 23:41 ` Elo touch panel Shaun Jackman @ 2006-02-08 8:51 ` Vojtech Pavlik 0 siblings, 0 replies; 2+ messages in thread From: Vojtech Pavlik @ 2006-02-08 8:51 UTC (permalink / raw) To: Shaun Jackman; +Cc: lkml On Tue, Feb 07, 2006 at 04:41:16PM -0700, Shaun Jackman wrote: > There's a bug in inputattach. The SERIO_RS232 constant is completely > mucking the bits that specify the protocol. > > --- inputattach.c- 2006-02-07 14:37:04.000000000 -0700 > +++ inputattach.c 2006-02-07 16:22:07.000000000 -0700 > @@ -455,7 +455,7 @@ > return 1; > } > > - devt = SERIO_RS232 | input_types[type].type | (id << 8) | (extra << 16); > + devt = input_types[type].type | (id << 8) | (extra << 16); > > if(ioctl(fd, SPIOCSTYPE, &devt)) { > fprintf(stderr, "inputattach: can't set device type\n"); > > Cheers, > Shaun > > APPENDIX A > > You probably know this better than I do, but the .type above is > misnamed. It should really be .proto. It's impossible, as far as I can > tell, to specify the type (like SERIO_RS232) with a SPIOCSTYPE call. It's an incompatibility in a later rework of serio.h/serport.c. SERIO_RS232 used to be 0x02000000, now the constant isn't shifted up anymore, which messed up things. Also, the type used to be possible to set, but I don't think that needs to be fixed, since there are so far no ttys that'd need a different serio type. I'll apply your fix to inpuatttach. Thanks for spotting the bug. -- Vojtech Pavlik SuSE Labs, SuSE CR ^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2006-02-08 8:51 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <7f45d9390602061830k6b984ban6fb302a3089cac6c@mail.gmail.com>
[not found] ` <20060207081415.GA6731@suse.cz>
2006-02-07 23:41 ` Elo touch panel Shaun Jackman
2006-02-08 8:51 ` Vojtech Pavlik
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox