Hi Hans,
> Can you test whether just changing the audio sampling frequency using
> 'v4l2-ctl -c audio_sampling_frequency=0' will also cause problems? So just
> run this and check if capturing from the TV (no radio involved in this
> test) will work or not. I'm wondering if there is an endian problem when
> the sampling freq. is changed. This code is now called for radio as well,
> so any endian problem is now triggered when using the radio.
>
> Explicitly changing the sampling freq. should trigger such a bug for any
> ivtv version (with or without the radio patch).
>
> I looked through the code and didn't see any obvious endianness problem,
> but this test should reveal whether there are any or not.
I run mplayer /dev/video0 and it was ok. I run 'v4l2-ctl -c
audio_sampling_frequency=0' right after that, which did not affect the
sound, but just video, which became very slow. This did not appear
after restarting mplayer.
> Please download the 0.10 branch again: the development snapshot version
>should read 0.10.3.
Same thing, no changes
[EMAIL PROTECTED]:~/packages/ivtv_0.10.3/0.10# /usr/bin/ivtvctl --version
ivtvctl version 0.10.3 (development snapshot compiled on Di 29 Mai
2007 17:43:05 CEST)
[EMAIL PROTECTED]:~/packages/ivtv_0.10.3/0.10# /usr/bin/ivtvctl --reset-ir
ioctl VIDIOC_INT_RESET failed: Invalid argument
ivtv: ==================== START INIT IVTV ====================
ivtv: version 0.10.3 (development snapshot compiled on Di 29 Mai 2007
17:44:02 CEST) loading
> Anyway, if this still doesn't work, then check ivtvctl.c. Close to the top
> it has this code:
> #if LINUX_VERSION_CODE > KERNEL_VERSION(2, 6, 18)
> #define VIDIOC_INT_RESET _IOW ('d', 102, __u32)
> #else
> #define VIDIOC_INT_RESET _IO ('d', 102)
> #endif
> Make sure the correct #define is used: if the wrong one is chosen then
> there may be a problem with the installed kernel headers where the wrong
> kernel version is returned and so the wrong VIDIOC_INT_RESET ioctl is
> used.
I have no idea how to check which #define is used, but I tried
something else. I replaced #define lines, to see if it will work other
way around. I tried this:
#if LINUX_VERSION_CODE > KERNEL_VERSION(2, 6, 18)
#define VIDIOC_INT_RESET _IO ('d', 102)
#else
#define VIDIOC_INT_RESET _IOW ('d', 102, __u32)
#endif
That caused the error "Bad address":
[EMAIL PROTECTED]:~/packages/ivtv_0.10.3/0.10/utils# ./ivtvctl --reset-ir
ioctl VIDIOC_INT_RESET failed: Bad address
so I suppose that the right #define was used before I changed this
line, but still without success.
M.
_______________________________________________
ivtv-devel mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-devel
