On Mon, 23 Oct 2006 18:04:55 -0700 (PDT) Bill Unruh <unruh@xxxxxxxxxxxxxx> wrote: > On Tue, 24 Oct 2006, Sergei Steshenko wrote: > > > On Tue, 24 Oct 2006 02:14:35 +0200 > > Kay Smarczewski <kay@xxxxxxxxxxxxxxxxxx> wrote: > > > >> On Mon, Oct 23, 2006 at 11:04:07AM +0200, Kay Smarczewski wrote: > >>> Hello, > >>> > >>> I have a real strange problem. I don't get my sound working. > >>> > >>> My system is a Acer Aspice 5102 WLMi with: > >>> Debian Etch > >>> 2.6.18 Kernel > >>> Alsa: 1.0.13 > >>> lspci | grep -i audio returns: > >>> 00:14.2 Audio device: ATI Technologies Inc SB450 HDA Audio (rev 01) > >>> (Acer description: Realtek High Definition Audio) > >>> > >>> On boot I get many error messages of this: > >>> hdc_codec: Invalid dep_range_val 0:7fff > >>> > >>> and after that a few of this: > >>> hdc_codec: num_steps = 0 for NID=0xd > >>> > >>> And now the real strange thing: > >>> If I purge alsa-base, alsa-utils, linux-sound-base and lsof everything > >>> works fine after the first reboot. But only then: from the second boot > >>> it just don't works anymore. > >>> > >>> And even more strange: > >>> aplay can play wav files everytimes. On the first time i start it after > >>> booting there is following warning: > >>> hda-intel: invalid position buffer, using LPIB instead > >>> > >>> cat /proc/modules | cut -f1 -d " " | sort > modules.new;diff modules.new > >>> modules.old don't gives any changes between the loaded modules. > >>> > >>> > >>> I've tried the 2.6.19rc2 kernel already, tried to compile Alsa-Driver, > >>> Alsa-Lib and Alsa-Utils for myself but nothing works. > >>> I've read the doc for the ATI-IXP southbridge HD-audio and modem and > >>> some other manuals and followed the instructions but nothing solved my > >>> problem. > >>> > >>> There was even the suggestion for a bios-update. I've done it. Of course > >>> it was unneccessary... > >>> > >>> I have recognized that there is an error before the hda_codec-error: > >>> unknown model for alc833, azx_response... timeout and that it just > >>> looks for something in the bios. > >>> > >>> and then a timeout... > >>> and then the hda_codec errors... > >>> > >>> but i still have no ideas :( > >>> > >>> Can anyone help me, please? > >> > >> Why nobody helps me, hmm? That sucks really! I am sure solving this > >> problem will just take seconds of your time if you are more expierenced > >> than me. But I will probably never get it working. So I have to use MS > >> Windows again because that just works. > >> > >> I know that you know that there is some trouble with some hardware. So, > >> what is the problem? A mailing list should not just be for some experts > >> which can write some freaky stuff. You just should not forget the normal > >> user!!! > >> > >> ------------------------------------------------------------------------- > >> Using Tomcat but need to do more? Need to support web services, security? > >> Get stuff done quickly with pre-integrated technology to make your job easier > >> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > >> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > >> _______________________________________________ > >> Alsa-user mailing list > >> Alsa-user@xxxxxxxxxxxxxxxxxxxxx > >> https://lists.sourceforge.net/lists/listinfo/alsa-user > >> > > > > If your chip appears in the list of supported by ALSA chips and doesn't > > work - file a bug. > > > > ... > > > > The Intel HDA chips appear to be all the same except for routing of > > audio streams. > > > > If it's the case, can ALSA developers publish a cheatsheet on how to get > > the routing info from a running Windows driver ? > > HOw would they do that? Do you think that they have the windows driver > source code they can look at and fugure out what is going on? If they did, > development of drivers would be trivial. They also have no idea how to get > the information. It would really really really help if the manufacturers > gave infomation. But they do not. > > > > > I guess it in the end is reading data from a limited set of registers > > whose addresses and format is already known to ALSA developers. > > And you know that how? > Each incarnation can totally change the registers, their content and the > meaning of their content. > Well, as far as I know the VLSI world (in which I work since late 1990) VLSI designers are as constructively lazy as the ones in other fields. In 2004 I myself dealt with a piece of external IP - there were enough problems with it, so there was no incentive whatsoever to change register addresses and such. I also think that companies simply buy Intel IP - I mean, Intel HDA, so why would a reasonably lazy designer start changing register map ? it's additional effort bringing no gain. That's why I think that the real degree of freedom is PCB layout, and that's why I think that variations in PCB layout are compensated for by rerouting signals inside the HDA chip. I've had a quick look into http://www.alsa-project.org/alsa/ftp/datasheets/intel/30234902_ICH6.pdf http://download.intel.com/design/chipsets/manuals/30701701.pdf - the specs appear to be pretty certain regarding addresses and formats. --Sergei. -- Visit my http://appsfromscratch.berlios.de/ open source project. ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Alsa-user mailing list Alsa-user@xxxxxxxxxxxxxxxxxxxxx https://lists.sourceforge.net/lists/listinfo/alsa-user
