From mboxrd@z Thu Jan 1 00:00:00 1970 From: Takashi Iwai Subject: Re: Re: [alsa-cvslog] CVS: BEFORE REVERT [SUCCESS] C-Media CMI9761 -A- type, WORKING MASTER, PCM ON 1. DXS SLIDER! Date: Thu, 17 Feb 2005 19:43:11 +0100 Message-ID: References: <4213BB44.5060702@gmx.de> <4214DD40.3070102@gmx.de> Mime-Version: 1.0 (generated by SEMI 1.14.5 - "Awara-Onsen") Content-Type: text/plain; charset=US-ASCII In-Reply-To: <4214DD40.3070102@gmx.de> Sender: alsa-devel-admin@lists.sourceforge.net Errors-To: alsa-devel-admin@lists.sourceforge.net List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , List-Archive: To: t.schorpp@gmx.de Cc: alsa-devel@lists.sourceforge.net List-Id: alsa-devel@alsa-project.org At Thu, 17 Feb 2005 19:06:56 +0100, thomas schorpp wrote: > > Takashi Iwai wrote: > > At Wed, 16 Feb 2005 22:29:40 +0100, > > thomas schorpp wrote: > > > >>- arts works with pcm fine > >>- working master vol slider(!) (kmix) > >>- working pcm vol regulation on 1. dxs slider (rest 3 no sound reaction) > > > > > > (snip) > > > > > >>>Revert the last addition for CM9761A support. The codec doesn't > >>>support the real control of Master > >> > >>yes, it does fine for my chip , sorry. > >> > >>/PCM volumes, too. > > > > > > Hmm, I'm really confused. Both the comments above look > > contradictionary. > > me too ;). so let me first excuse for that "thing" i called patch v0.1 > tonight. im sorry i cant afford a prebuild-kernel distributors > requirements and better read your comprehensive docs on the homepage > before coding again. i thank you again for getting my chip to work based > on 1.73 cvs. i can live with custom patched kernels for my own needs. > > although i would gladly help you out by providing verified infos on this > newer chip 1000s of users may have got, so you can work out a common > solution. > > so much for priorities and requirements. > > > > > Let me ask again: In the older version, does PCM volume work, that > > is, it really changes the output volume (not only mute/unmute)? > > Does Master volume work as well on your chip? > > which version? I meant the last working version you tried before (now I assume 2.6.10 kernel?) > i repeat my notices from tonight: > > 2.6.10 kernel.org kernel: > - master vol slider drives real volume at least for looped through > lines. (dont know if a real hardware gain driver exists, i cant find the > list mentioned sw volume code.) The looped-through lines = analog inputs like Line, Mic, CD? And the Master volume does NOT affect the PCM playback volume? > - master pcm slider exists but seems no affect, but typical hardware > switching noise on enable/disable knob OK. > - no output on pcm dac audio (tried workarounds for hours) Err, with 2.6.10, you got no PCM playback output? > patch.c cvs rev 1.73 et al: > - master vol slider drives real volume for looped through lines. no > affect on pcm. So, it has no influence on PCM... Interesting. Possibly there is a mysterious register bit switch to turn this on/off :) (For example, STAC9750/51 controls the PCM-bypass mode by misusing GENERAL_PURPOSE register bit 15.) > -alsa pcm volume working on dxs 1 slider. > -oss pcm volume working on dxs 2 slider. The assignment of DXS channels are quite casual. > - master pcm slider missing. (which code determines this?) In ac97_codec.c:check_volume_resolution(). This means that AC97_PCM register returns only a certain value (0xX808). So, the expected behavior. > rev 1.74: > - master / pcm sliders missing, cant affect any volumes, lines looped > through, input regulateable with cd and aux input sliders. This is intentional. > - pcm not working. This is not :) > > Note that the Master and PCM volumes are _removed_ because they don't > > work and are only confusing. It's intentional. They _must_ disappear > > in the latest version. > > > > If CM9761-A- provides the real PCM/Master volume control, we should > > get them back again. But, this essential information has been > > missing. Please let me know _this_ clearly above all subtle things! > > > > see above. "real" means hardware, yes? Yes. > btw: you got my 2 mails from tonight? windos driver reports rev 8D for > this chip. Hmm? I see a string "REV_50" in the screenshot. > pls answer this question: what are those 32bit xxxxxxxx_rev ids? Which value do you mean? If it's ac97->id, this is just the combination of vendor_id1 (7c) and vondor_id2 (7e) registers as (vendor_id1<<16) | (device_id2). AFAIK, CM9761 has three different models and each of them has different vendor ID2 value: 0x4978, 0x4982 and 0x4983. Yours looks like the last one. > have you defined it? > where do you get it from the chip? > cant find the function asking the chip. ac97->id reads only defines > doesnt it? I don't understand your question here. > > > >>>Instead, fixed the default multi-channel register setting now. > >>> > >> > >>why? > >>not verified. source of info? are you sure thats the -A- type, you got > >>for testing? who verified? > > > > > > Very much confision of debug information... > > ok. you stated you got rev 83 mine should be 8d. > that would be clearer then. The value is just taken from the vendor_id2 register value (the lower byte). > > > >>oops! sound gone after doing this: > > > > > > How about the attached patch? > > would try it, but pls comment those reg bit setting(?) in short. > you stated you got no register specs, is this guessed? Yes, just a guess from cmedia's driver code and trial-and-error. Takashi ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click