On Wed, Sep 16, 2009 at 12:27 PM, Noah Beck <[email protected]> wrote:
> On Tue, Sep 15, 2009 at 10:54:28PM -0400, Andy Walls wrote:
>> On Tue, 2009-09-15 at 14:27 +1200, Robin Gilks wrote:
>> > > Andy Walls wrote:
>> >
>> > > Given the testing Chris has done with my patches, it looks like using
>> > > CX23418 A/V core firmware with a CX25843 is returning non-sensical
>> > > cominations of interrupt status flags in the audio interrupt status
>> > > register (0x819 IIRC) of the CX25843.
>> > >
>> > > In the above the NICAM related flags don't make sense and neither does
>> > > the AC97 status.  The (S)IF L(ost) flag makes no sense with the format
>> > > detection flags and ormat detetction loop complete flags, etc.   Rarely
>> > > do I get more than 1 flag other than RDS (which is always set for me).
>> > > When I do get multple flags at once, its FDL and AFC (IIRC).
>> > >
>> > > Could you revert the firmware archive to use a known CX2584x specific
>> > > firmware for the cx25840 firmware file in the archive at
>> > > dl.ivtvdriver.org?
>> > >
>> > > I suspect the CX23418 firmware image being distributed as CX2584x
>> > > firmware is causing problems for CX2584x chips.
>> >
>> > I'm seeing more and more of this bad audio problem with a PAL PVR150 using
>> > the line input audio and more recent kernel versions (currently Linux
>> > jupiter 2.6.30-gentoo-r4 #2 SMP Thu Aug 13 23:54:02 NZST 2009 x86_64 AMD
>> > Athlon(tm) Dual Core Processor 5050e AuthenticAMD GNU/Linux) and newer
>> > (faster) hardware.
>> >
>> > Is there an md5sum that would identify a known good firmware? The files in
>> > the archive are not very well described (dates don't mean much to me). I'm
>> > also a bit confused by the device numbers as my dsmeg quotes both cx23416
>> > and cx25840
>>
>> I use this (possibly ancient) firmware with my PVR-150MCE:
>>
>> $ md5sum /lib/firmware/v4l-cx25840.fw
>> 99836e41ccb28c7b373e87686f93712a  /lib/firmware/v4l-cx25840.fw
>>
>> it works wellfor me, but I barely use analog anymore.  IIRC, I pulled
>> the file from the Windows driver CD for my Hauppauge card.  MakoC.rom or
>> something like that.
>>
>>
>> The CX23416 is the MPEG encoder, the CX2584[0123] is a broadcast TV
>> Video and Audio signal decoder and digitizer that feeds digitized video
>> and audio data to the CX23416 encoder for MPEG encoding.
>>
>> Regards,
>> Andy
>
> Looking at md5sums of the firmware files at dl.ivtvdriver.org/ivtv/firmware,
> I see that Andy is using the pre-20080701 version of v4l-cx25840.fw:
>
> 588f081b562f5c653a3db1ad8f65939a  cx18-firmware/v4l-cx23418-apu.fw
> b6c7ed64bc44b1a6e0840adaeac39d79  cx18-firmware/v4l-cx23418-cpu.fw
> b3704908fd058485f3ef136941b2e513  cx18-firmware/v4l-cx23418-dig.fw
> 305dba74bbe5905447add8883f3ecb68  ivtv-firmware-20060701/v4l-cx2341x-dec.fw
> d85cb08382395390dc95ac6ebc2205f9  ivtv-firmware-20060701/v4l-cx2341x-enc.fw
> 99836e41ccb28c7b373e87686f93712a  ivtv-firmware-20060701/v4l-cx25840.fw
> 305dba74bbe5905447add8883f3ecb68  ivtv-firmware-20061007/v4l-cx2341x-dec.fw
> d85cb08382395390dc95ac6ebc2205f9  ivtv-firmware-20061007/v4l-cx2341x-enc.fw
> 99836e41ccb28c7b373e87686f93712a  ivtv-firmware-20061007/v4l-cx25840.fw
> 305dba74bbe5905447add8883f3ecb68  ivtv-firmware-20070217/v4l-cx2341x-dec.fw
> 9b39b3d3bba1ce2da40f82ef0c50ef48  ivtv-firmware-20070217/v4l-cx2341x-enc.fw
> 99836e41ccb28c7b373e87686f93712a  ivtv-firmware-20070217/v4l-cx25840.fw
> 305dba74bbe5905447add8883f3ecb68  ivtv-firmware-20080701/v4l-cx2341x-dec.fw
> 9b39b3d3bba1ce2da40f82ef0c50ef48  ivtv-firmware-20080701/v4l-cx2341x-enc.fw
> b3704908fd058485f3ef136941b2e513  ivtv-firmware-20080701/v4l-cx25840.fw
> 3a4803384f749d644ee1f1ca9dcb12fa  HcwMakoB.ROM
>
> My system has files matching ivtv-firmware-20080701 and the intermittent
> "tinny" audio issue on my PVR-150 (I use line-in).  This install is from
> Nov 2008, and my previous install (~Feb 2007) on the same system did not
> have the audio issue.  I will try reverting to
> ivtv-firmware-20070217/v4l-cx25840.fw and see if my audio issue goes away.

Changing the firmware version on my myth box did not appear to affect
the "tinny" audio issue.

Noah

_______________________________________________
ivtv-devel mailing list
[email protected]
http://ivtvdriver.org/mailman/listinfo/ivtv-devel

Reply via email to