Lord Kamina Posted February 23, 2015 Share Posted February 23, 2015 I've been trying to produce a working AppleHDA for my new board (an MSI P67a-GD55) and so far have been met mostly with failure. What I'm doing is trying to create a custom layout (and pathmap etc etc) from scratch and am using DigitalDJ's method to patch the layout hard-coded size in the Binary (https://digitaldj.net/blog/2013/07/09/take-my-layout-applehda/) So far, it would appear my layout is being indeed read but I have no sound devices and I'm being plastered with assertions. Using IDA Pro I began trying to trace them as explained on that post... and have basically hit a wall with that. I'm really getting A LOT of assertions but the first two are always 22-02-15 14:42:42,000 kernel[0]: Sound assertion in AppleHDAPath at line 135022-02-15 14:42:42,000 kernel[0]: Sound assertion in AppleHDAEngine at line 515 Something tells me I should try to correct them from the top-down as most likely one failure would lead to another. Upon tracing those, I've gotten to the point where I know they're originated in AppleHDAEngine::getCurrentAmplifierGainRange(uint *,uint *,int *,int *) and AppleHDAPath::getWidgetAmplifierGainRange(uchar *,uchar *,uchar *,int *,int *,uint) But I have not been able to get past that into a way to actually figuring out specifically what the issue is. Any help or maybe somebody that has already figured it out? 1 Link to comment Share on other sites More sharing options...
Mirone Posted February 23, 2015 Share Posted February 23, 2015 tell us which is your audio codec, if possible attach your codec_dump. Link to comment Share on other sites More sharing options...
Lord Kamina Posted February 23, 2015 Author Share Posted February 23, 2015 Can't put much more info now, since at work. It's an ALC892. Nothing unusual in the codec dump afaict. Also, using 10.9.5 Link to comment Share on other sites More sharing options...
Mirone Posted February 23, 2015 Share Posted February 23, 2015 try this: ALC892 use Layout_id=9. Link to comment Share on other sites More sharing options...
Lord Kamina Posted February 23, 2015 Author Share Posted February 23, 2015 I'll take a look later but I don't want just a kext that will make it work. I want to understand why it is broken and how to fix it. That's why I'm chasing after these assertions. Link to comment Share on other sites More sharing options...
Mirone Posted February 23, 2015 Share Posted February 23, 2015 even a functional kext can have these statements, is there a way to suppress this look here: http://www.insanelymac.com/forum/topic/292377-how-to-get-rid-of-any-sound-assertions-on-applehda/ Guide To Patch AppleHDA For Your Codec Link to comment Share on other sites More sharing options...
Lord Kamina Posted February 23, 2015 Author Share Posted February 23, 2015 Yes but not all of them are fatal. In my case they clearly are. Link to comment Share on other sites More sharing options...
Mirone Posted February 23, 2015 Share Posted February 23, 2015 to me that's the problem with binary patch. Link to comment Share on other sites More sharing options...
Lord Kamina Posted February 23, 2015 Author Share Posted February 23, 2015 What is the alternative? Voodoo? What is the alternative? Voodoo? Link to comment Share on other sites More sharing options...
Mirone Posted February 23, 2015 Share Posted February 23, 2015 the kext I sent you does not work? Link to comment Share on other sites More sharing options...
Lord Kamina Posted February 23, 2015 Author Share Posted February 23, 2015 I haven't tried it yet, but unless it's a completely new implementation of HDA, it would still require a bin patch. Link to comment Share on other sites More sharing options...
Mirone Posted February 23, 2015 Share Posted February 23, 2015 first of all you should try kext I sent you. Link to comment Share on other sites More sharing options...
Recommended Posts