AppleALC — dynamic AppleHDA patching

[tyufhl]

1 hour ago, polyzargone said:

@Andres ZeroCross

 

Hello,

 

My specs:

 

Acer Aspire V3 772G

Realtek ALC282

Layouts tested : 3, 4, 13, 27, 28, 29, 76, 86, 127

Currently, 28 gives best results but not perfect

 

I don't know for sure but since I'm using 10.14.4 beta, I'm loosing audio after sleep. Everything worked fine before and I never had to use CodecCommander nor SSDT. So I tried using CodecCommander in L/E w/o success. I suspect something changed but I can't tell if it's the OS or AppleALC ALC282 resources.

 

Anyway, mic is not working well, there's noise when it's used. I don't use it often and that's why I didn't paid attention to it until now and this annoying sleep issue. Maybe there's something wrong…

 

Could you please have a look on this?

 

Thanks

 

Files

there is no such patch in your resources 

check kext. 

AppleALC.kext.zip

[polyzargone]

@tyufhl

 

Thanks a lot!

 

Sound after sleep works great now with your AppleALC.kext. Still, there's this annoying noise when I use mic. Do you think you can fix that too?

[Rodion2010]

17 hours ago, insanelyDeepak said:

hmmm , i don't know anything about it 

 

@Rodion2010 knows enough , is it really good ? 

 

It builds working Platforms.xml automatically

some tuning is required for autodetect, multiple speakers, multiple inputs

but it does the biggest part of job

[Andrey1970]

1 hour ago, tyufhl said:

Realtek ALC282

Layouts tested : 3, 4, 13, 27, 28, 29, 76, 86, 127

Currently, 28 gives best results but not perfect

 

I don't know for sure but since I'm using 10.14.4 beta, I'm loosing audio after sleep. Everything worked fine before and I never had to use CodecCommander nor SSDT.

 

Fixed in master.

[tyufhl]

5 hours ago, polyzargone said:

@tyufhl

 

Thanks a lot!

 

Sound after sleep works great now with your AppleALC.kext. Still, there's this annoying noise when I use mic. Do you think you can fix that too?

Check kext. AppleALC.kext.zip

[polyzargone]

@tyufhl

 

That's a lot better!

 

Still a little background noise so if you think you can improve it again that would be great. Otherwise, it will be sufficient enough for me ;).

[insanelyDeepak]

5 hours ago, Rodion2010 said:

 

It builds working Platforms.xml automatically

some tuning is required for autodetect, multiple speakers, multiple inputs

but it does the biggest part of job

thanks alot

[tyufhl]

14 minutes ago, polyzargone said:

@tyufhl

 

That's a lot better!

 

Still a little background noise so if you think you can improve it again that would be great. Otherwise, it will be sufficient enough for me ;).

Check. AppleALC.kext.zip

[BlackPilot]

Somebody can help me???  

 

 

 

[polyzargone]

3 hours ago, tyufhl said:

Check. AppleALC.kext.zip

 

Fantastic!

 

Noise fixed !

 

Thanks a lot and have a good day/night!

[tyufhl]

9 hours ago, BlackPilot said:

Somebody can help me???  

 

 

Check. Layout 2.

AppleALC.kext.zip

[BlackPilot]

20 minutes ago, tyufhl said:

Check. Layout 2.

AppleALC.kext.zip

 

Big Thanks bro  I will try it, but this is intended for IDT Audio Codec, right??? 

[Rodion2010]

12 hours ago, insanelyDeepak said:

thanks alot

it will be great to make Platforms.xml in the similar way from dump, not from VoodooHDA

is it possible?

Edited February 15, 2019 by Rodion2010

[headkaze]

1 hour ago, Rodion2010 said:

it will be great to make Platforms.xml in the similar way from dump, not from VoodooHDA

is it possible?

I've built a working version of Clover that will create a binary dump of the audio config by pressing F4 at the Clover menu. This data can then be parsed directly into Pin Configurator.

 

I'm currently looking at getting Pin Configurator to create a Platforms.xml like GetDumpXML based on the Clover dump.

[Rodion2010]

22 minutes ago, headkaze said:

I've built a working version of Clover that will create a binary dump of the audio config by pressing F4 at the Clover menu. This data can then be parsed directly into Pin Configurator.

I did it manually

fs0:
cd \EFI\Clover\Tools
load AudioDxe-64.efi //if not loaded by default
HdaCodecDump.efi > dump.txt

btw there was an error in dumper https://github.com/Goldfish64/AudioPkg/issues/10

fixed

and here is the dump

dump.txt.zip

 

PS F4 is for DSDT

Edited February 15, 2019 by Rodion2010

[Andrey1970]

18 minutes ago, headkaze said:

I've built a working version of Clover that will create a binary dump of the audio config by pressing F4 at the Clover menu.

 

F4 is occupied.

[headkaze]

3 minutes ago, Andrey1970 said:

F4 is occupied.

I know F4 dumps ACPI data. I didn't see the point in assigning a new key to it. I guess I'll see what @Slice thinks.

[insanelyDeepak]

11 hours ago, Rodion2010 said:

it will be great to make Platforms.xml in the similar way from dump, not from VoodooHDA

is it possible?

making Platforms from voodooDump is pretty easy , but from codec dump it's lengthy work e.g 

Pin Complex -> Audio Mixer -> Audio Output

for Input you have to do reverse  

but if you have more than 1 audio mixer nodes for particular node id could be issue most of IDT Codecs have this and they require all audio/mixer to enable audio 

 

 

if you implement how voodoo connects all nodes to each other then it could be better 

 

[akram.kawamleh]

I'd installed Mac OS Mojave on my Lenovo L560 everything works fine but the Audio i came across forums on internet to chnage the configuration via clover and tried the layout which are detected 11 ,28, 29 and placed the Applealc.kext and Lilu.kext in kext/ other folder but with no luck
please i need your help

debug_9148.zip

[Rodion2010]

7 hours ago, insanelyDeepak said:

making Platforms from voodooDump is pretty easy , but from codec dump it's lengthy work e.g 

Pin Complex -> Audio Mixer -> Audio Output

for Input you have to do reverse  

but if you have more than 1 audio mixer nodes for particular node id could be issue most of IDT Codecs have this and they require all audio/mixer to enable audio 

 

 

if you implement how voodoo connects all nodes to each other then it could be better 

 

i think i understand how to build outputs

 

if node is "PinConplex" and type is "Output"

read the default (first?) connection from connection list

it node is "Audio Output" then ready

else read next connected node

 

inputs are more complicated but similar

 

I have "more than one" too

Node 17 "Headphone Playback Switch" Node 34 [Audio Mixer] - Node 55 [Audio Selector] - Node 3 [Audio Output]

Node 8 [Audio Input] Node 12 [Audio Selector] - Node *60 [Audio Selector] - Node 23 [Jack] Mic at Ext Rear

 

Edited February 16, 2019 by Rodion2010

[Rodion2010]

dump

           nid: 17
           Name: pin: Headphones (Green Front)
     Widget cap: 0x0040018d
        ...
    connections: 1 enabled 1
          |
          + <- nid=34 [audio mixer]

 

nid: 34
           Name: audio mixer
     Widget cap: 0x00200103
    ...
    connections: 2 enabled 2
          |
          + <- nid=55 [audio selector]
          + <- nid=33 [audio selector]

 

nid: 55
           Name: audio selector
     Widget cap: 0x00300101
     ...
    connections: 3 enabled 1
          |
          + <- nid=3 [audio output] (selected)
          + [DISABLED] <- nid=4 [audio output]
          + [DISABLED] <- nid=6 [audio output]
 

nid: 3
           Name: audio output

Edited February 16, 2019 by Rodion2010

[Rodion2010]

and for input

 

          nid: 8
           Name: audio input
     Widget cap: 0x00100501
     ...
    connections: 1 enabled 1
          |
          + <- nid=12 [audio selector]

 

 

          nid: 12
           Name: audio selector
     Widget cap: 0x0030010d
     ...
    connections: 10 enabled 5
          |
          + [DISABLED] <- nid=56 [audio selector] [DISABLED]
          + <- nid=57 [audio selector] (selected)
          + <- nid=58 [audio selector]
          + [DISABLED] <- nid=59 [audio selector] [DISABLED]
          + <- nid=60 [audio selector]
          + <- nid=24 [pin: CD (ATAPI)]
          + [DISABLED] <- nid=36 [pin: Line-out (Orange Rear)]
          + [DISABLED] <- nid=37 [pin: Line-out (Grey Rear)]
          + [DISABLED] <- nid=61 [audio selector] [DISABLED]
          + <- nid=32 [audio mixer]

 

nid: 57
           Name: audio selector
     Widget cap: 0x0030010d
   ...
    connections: 1 enabled 1
          |
          + <- nid=20 [pin: Microphone (Pink Front)]

 

 nid: 20
           Name: pin: Microphone (Pink Front)
     
 

Edited February 16, 2019 by Rodion2010

[insanelyDeepak]

On 2/16/2019 at 4:56 PM, Rodion2010 said:

i think i understand how to build outputs

 

if node is "PinConplex" and type is "Output"

read the default (first?) connection from connection list

it node is "Audio Output" then ready

else read next connected node

 

inputs are more complicated but similar

 

I have "more than one" too

Node 17 "Headphone Playback Switch" Node 34 [Audio Mixer] - Node 55 [Audio Selector] - Node 3 [Audio Output]

Node 8 [Audio Input] Node 12 [Audio Selector] - Node *60 [Audio Selector] - Node 23 [Jack] Mic at Ext Rear

 

only one Audio Mixer is enough don't need audio selector , so it will be 17 ->34 -> 3 , if you have multiple audio mixer for one nodes then you may require to use it too

 

if you read voodoodump it's pretty easy to do so like below 

Headphone  33 -> 3

+-------------------------------+
| DUMPING Playback/Record Paths |
+-------------------------------+

Playback:

    nid=33 [pin: Headphones (Black Front)]
      |
      + <- nid=3 [audio output] [src: pcm] bindSeq=00000001

 

Speakers : 20 -> 2

+-------------------------------+
| DUMPING Playback/Record Paths |
+-------------------------------+

Playback:

    nid=20 [pin: Speaker (Analog)]
      |
      + <- nid=2 [audio output] [src: pcm] bindSeq=00000001

 

[Rodion2010]

9 minutes ago, insanelyDeepak said:

only one Audio Mixer is enough don't need audio selector , so it will be 17 ->34 -> 3 , if you have multiple audio mixer for one nodes then you may require to use it too

 

 

it depends on Codec and Connection List of each node

Headphone - [Audio Mixer] -  [Audio Selector] -  [Audio Output]

Headphone - [Audio Mixer] -  [Audio Output]

Headphone -  [Audio Output]

 

there is no one answer what is needed and what is not needed

Look at          Name: pin: Headphones (Green Front)
     Widget cap: 0x0040018d
        ...
    connections: 1 enabled 1
          |
          + <- nid=34 [audio mixer]

 

and Voodoo shows exactly the same

Playback:

    nid=17 [pin: Headphones (Green Front)]
      |
      + <- nid=34 [audio mixer] bindSeq=00000001
             |
             + <- nid=55 [audio selector] [src: pcm] bindSeq=00000001

                    |
                    + <- nid=3 [audio output] [src: pcm] bindSeq=00000001

 

 

 

 

 

[tyufhl]

11 minutes ago, insanelyDeepak said:

only one Audio Mixer is enough don't need audio selector , so it will be 17 ->34 -> 3 , if you have multiple audio mixer for one nodes then you may require to use it too

 

if you read voodoodump it's pretty easy to do so like below

Headphone 33 -> 3

+-------------------------------+
| DUMPING Playback/Record Paths |
+-------------------------------+

Playback:

    nid=33 [pin: Headphones (Black Front)]
      |
      + <- nid=3 [audio output] [src: pcm] bindSeq=00000001

 

Speakers : 20 -> 2

+-------------------------------+
| DUMPING Playback/Record Paths |
+-------------------------------+

Playback:

    nid=20 [pin: Speaker (Analog)]
      |
      + <- nid=2 [audio output] [src: pcm] bindSeq=00000001

 

Need Connection Select Control ( 701h)  to pinpoint the contact. This is not in HdaCodecDump.efi.