On Thu, 2009-09-10 at 07:23 -0500, Chris Kennedy wrote:
> > Well you can stop until I fix the debug line to print the interrupt
> > status to the log too. I want to know which status bit I need to care
> > about.
>
> Oh I see that line, but just in dmesg and not the logs, here's what
> I'm getting in dmesg at least so you see the normal pattern (seems the
> afc ac97 are the most frequent)...
^^^^
Really, really odd, seeing as we don't configure the AC'97 interface,
nor use it.
> cx25840 0-0044: ... after initial format detection loop completed!
> cx25840 0-0044: Detected audio mode: mono
> cx25840 0-0044: Detected audio standard: BTSC
> cx25840 0-0044: Audio muted: no
> cx25840 0-0044: Audio microcontroller: running
> cx25840 0-0044: Configured audio standard: automatic detection
> cx25840 0-0044: Configured audio system: BTSC
> cx25840 0-0044: Specified audio input: Tuner (In8)
> cx25840 0-0044: Preferred audio mode: stereo
> cx25840 3-0044: Audio status: nber nll ifl fdl afc amc ac97
^^^^^^^^ ^^^
The NICAM (French?) audio status flags seem just plain wrong.
Loosing the SIF (IF Lost: ifl) can't be good in any case.
There are just way too many flags set at once here. Something's screwed
up: either the microcontroller software or the I2C bus maybe?
Also RDS interrupt status isn't set, which my firmware always seems to
have set.
> cx25840 3-0044: Audio mode change occured
> cx25840 3-0044: ... after initial format detection loop completed!
> cx25840 3-0044: Audio format change occured
> cx25840 3-0044: ... after initial format detection loop completed!
> cx25840 3-0044: Audio format detection loop complete
> cx25840 3-0044: Detected audio mode: mono
> cx25840 3-0044: Detected audio standard: BTSC
> cx25840 3-0044: Audio muted: no
> cx25840 3-0044: Audio microcontroller: running
> cx25840 3-0044: Configured audio standard: automatic detection
> cx25840 3-0044: Configured audio system: BTSC
> cx25840 3-0044: Specified audio input: Tuner (In8)
> cx25840 3-0044: Preferred audio mode: stereo
> cx25840 3-0044: Audio status: afc ac97
> cx25840 3-0044: Audio format change occured
> cx25840 3-0044: ... after initial format detection loop completed!
> cx25840 3-0044: Detected audio mode: mono
> cx25840 3-0044: Detected audio standard: BTSC
> cx25840 3-0044: Audio muted: no
> cx25840 3-0044: Audio microcontroller: running
> cx25840 3-0044: Configured audio standard: automatic detection
> cx25840 3-0044: Configured audio system: BTSC
> cx25840 3-0044: Specified audio input: Tuner (In8)
> cx25840 3-0044: Preferred audio mode: stereo
> cx25840 3-0044: Audio status: fdl afc amc ac97
^^^^^^^^^^^^^^^^
Better, but it still seems like too many things are set. Do you have a
long poll interval set?
>
> Is it basically it needs to be logged at the info level instead
> of using that dbg statement, this seems to catch it for me...
>
> /* Debug log on anything other than RDS which is always set
> apparently*/
> if (v & ~AUD_INT_STAT_RDS)
> if (cx25840_debug >= 3)
> v4l_info(client,
> "Audio status: %s%s%s%s%s%s%s%s\n",
> v & AUD_INT_STAT_RDS ? "rds " : " ",
> v & AUD_INT_STAT_NBER ? "nber " : " ",
> v & AUD_INT_STAT_NLL ? "nll " : " ",
> v & AUD_INT_STAT_IFL ? "ifl " : " ",
> v & AUD_INT_STAT_FDL ? "fdl " : " ",
> v & AUD_INT_STAT_AFC ? "afc " : " ",
> v & AUD_INT_STAT_AMC ? "amc " : " ",
> v & AUD_INT_STAT_AC97 ? "ac97 " : " ");
> //v4l_dbg(3, cx25840_debug, client,
>
Yeah, that's all I was going to do.
Well, collect some data. We'll see what shows up.
I'm not hopeful given the number of flags that are always set in the
log. It really indicates to me something is wrong. I'd guess the
firmware version is not actually right for the CX25843 or the firmware
load is getting corrupted.
When I find the time this weekend, I'll port my firmware load
verification routine from the cx18 driver.
Thanks for all the testing so far.
BTW, here are hashes for the firmwares I have in use on my system:
$ md5sum /lib/firmware/v4l-cx25840.fw /lib/firmware/v4l-cx23418-dig.fw
99836e41ccb28c7b373e87686f93712a /lib/firmware/v4l-cx25840.fw
b3704908fd058485f3ef136941b2e513 /lib/firmware/v4l-cx23418-dig.fw
Regards,
Andy
>
> Thanks,
> Chris
_______________________________________________
ivtv-devel mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-devel
