[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#617488: [regression] No input device for IR remote on Pinnacle PCTV 110i (saa7133) video board



reassign 617488 src:linux linux-2.6/2.6.37-2
severity 617488 important
tags 617488 + upstream moreinfo
found 617488 linux-2.6/2.6.38-1
quit

Hi Alex,

In March, 2011, Alex Volkov wrote:

> Pinnacle PCTV 110i TV-tuner board has IR remote control, which was recognized
> by kernel as input device before this (http://goo.gl/U3Fal) upstream patch.

(For reference: that's v2.6.37-rc1~64^2~378 "V4L/DVB: Remove the usage
of I2C_HW_B_CX2388x on ir-kbd-i2c.c", 2010-03-09)

> Since then, ir-kbd-i2c module fails to probe it with the following debug
> message:
>
> ir-kbd-i2c: : Unsupported device at address 0x47
>
> which is probably due to some uninitialized structure member. I tried to look
> for fix in the upstream, but cannot find any yet (they did a lot of code
> rearranging). Well, don't know how useful is this patch for upstream, but it
> fixes the problem anyway:
[...]
> --- a/drivers/media/video/saa7134/saa7134-input.c	2011-01-05 05:50:19.000000000 +0500
> +++ b/drivers/media/video/saa7134/saa7134-input.c	2011-03-09 14:17:05.000000000 +0500
> @@ -936,6 +936,7 @@
>  	memset(&info, 0, sizeof(struct i2c_board_info));
>  	memset(&dev->init_data, 0, sizeof(dev->init_data));
>  	strlcpy(info.type, "ir_video", I2C_NAME_SIZE);
> +	dev->init_data.type = IR_TYPE_OTHER;

Thanks for reporting it, and sorry for the long silence.  Am I correct in
guessing that you can still reproduce this with unpatched 3.2.y and newer
kernels?

If so, please report this upstream to linux-media@vger.kernel.org, cc-ing
linux-kernel@vger.kernel.org, Michael Hunold <michael@mihu.de>, and
me.  The format of your message should be:

	Subject: [PATCH/RFC] (short description of the change your patch makes goes here)

	(description of what happened, which kernels you have tested,
	and what your patch does to fix it goes here)

	(sign-off goes here; please see Documentation/SubmittingPatches,
	section 12 "Sign your work" for what that means)
	---	<=== three dashes go here
	(patch goes here)

Then we can fix it for everyone.

Hope that helps,
Jonathan


Reply to: