You are not logged in.

#1 2024-03-28 04:52:47

SinoCiCi
Member
Registered: 2024-03-28
Posts: 6

[SOLVED] Internal Mic Fix for IdeaPad Pro 5 w/ AMD chip and ALC257

I just want to post here how I fix my internal Mic issue for my new 83D5 IdeaPad Pro 5 16AHP9 laptop with (amd 8845hs chip in it).
I checked all the past blogs and posts they all mentioned that you have to recompile your kernel by yourself in order to activate the Internal mic . ( see https://bbs.archlinux.org/viewtopic.php?id=278886 )
But now for AMD 7000 chips and its variants (e.g., 7840hs and 8845hs ), you do not have to do that!
Now the AMD 7000 variants uses Pink Sardine platform for sound, so what you need to do is just to enable the module for them

A easy way of doing it is doing the following command

echo -e "snd-pci-ps\nsnd-soc-ps-mach" | sudo tee /etc/modules-load.d/sound.conf > /dev/null

and once you reboot your laptop, your mic should work.

It might be beneficial to create a new post since I spent quite a bit of time searching for solutions, only to discover the incorrect one (compiling Yellow Carp (YC)). Many users purchasing laptops equipped with the new AMD 7840 or 8845 chips are likely to encounter similar issues and seek solutions online.

Last edited by SinoCiCi (2024-03-28 05:36:46)

Offline

#2 2024-03-29 09:53:58

luca020400
Member
Registered: 2015-03-02
Posts: 9

Re: [SOLVED] Internal Mic Fix for IdeaPad Pro 5 w/ AMD chip and ALC257

FYI
I got linux-stable to queue the fixes related to this issue, as it broke every device using the non-YC driver.
Seems like you'll need a more generic quirk based approach to fix this? Could you upload an `acpidump`? Maybe there's an entry there that can be used (they already check for one of them)

https://git.kernel.org/pub/scm/linux/ke … 3031cc63f2

Last edited by luca020400 (2024-03-29 10:19:20)

Offline

#3 2024-03-29 11:04:01

SinoCiCi
Member
Registered: 2024-03-28
Posts: 6

Re: [SOLVED] Internal Mic Fix for IdeaPad Pro 5 w/ AMD chip and ALC257

luca020400 wrote:

I got linux-stable to queue the fixes related to this issue, as it broke every device using the non-YC driver.

I assume you refer to this update https://git.kernel.org/pub/scm/linux/ke … 3031cc63f2  . Well, the thing is the yc driver is compatible with the PCI revision id 0x63 mic hardware. Thus some users actually can use yc driver to activate their 0x63 mic before this upgrade (see https://bbs.archlinux.org/viewtopic.php?id=294293). After this upgrade the yc driver will be disabled by default for those 0x63 mic hardware. And it's our turn to enable the correct

snd-pci-ps
snd-soc-ps-mach

module by ourselves. I do not know why arch install did not automatically activate that for me.

Offline

#4 2024-03-29 11:59:51

luca020400
Member
Registered: 2015-03-02
Posts: 9

Re: [SOLVED] Internal Mic Fix for IdeaPad Pro 5 w/ AMD chip and ALC257

SinoCiCi wrote:
luca020400 wrote:

I got linux-stable to queue the fixes related to this issue, as it broke every device using the non-YC driver.

I assume you refer to this update https://git.kernel.org/pub/scm/linux/ke … 3031cc63f2  . Well, the thing is the yc driver is compatible with the PCI revision id 0x63 mic hardware. Thus some users actually can use yc driver to activate their 0x63 mic before this upgrade (see https://bbs.archlinux.org/viewtopic.php?id=294293). After this upgrade the yc driver will be disabled by default for those 0x63 mic hardware. And it's our turn to enable the correct

snd-pci-ps
snd-soc-ps-mach

module by ourselves. I do not know why arch install did not automatically activate that for me.

Did force probing the module fix it?

The YC driver can be compatible with 0x63 if you manually add a quirk (of force load the module)

On here ACPI doesn't report the required node, and since no quirk was available, the driver was registered but it wouldn't enable the platform device.

Meanwhile the PS driver has support for this revision and works out of the box (
If PS users are having issues it likely means the module isn't getting loaded. And the YC driver taking over would do that (since it can't work without a quirk on PS)

Reading the code 0x60 is YC, and that's supported in the driver, 0x63 is PS and that's also supported.

What seems to be happening is that a lot of YC devices don't have the ACPI node required to probe the driver properly, that's why I asked for a dump of it, maybe there's another entry that would do the job?
(No PS device has the ACPI entry needed by the YC driver, so here's why the revert exists, and also because the PS driver works just fine)

Offline

#5 2024-03-29 13:04:13

SinoCiCi
Member
Registered: 2024-03-28
Posts: 6

Re: [SOLVED] Internal Mic Fix for IdeaPad Pro 5 w/ AMD chip and ALC257

luca020400 wrote:
SinoCiCi wrote:
luca020400 wrote:

I got linux-stable to queue the fixes related to this issue, as it broke every device using the non-YC driver.

I assume you refer to this update https://git.kernel.org/pub/scm/linux/ke … 3031cc63f2  . Well, the thing is the yc driver is compatible with the PCI revision id 0x63 mic hardware. Thus some users actually can use yc driver to activate their 0x63 mic before this upgrade (see https://bbs.archlinux.org/viewtopic.php?id=294293). After this upgrade the yc driver will be disabled by default for those 0x63 mic hardware. And it's our turn to enable the correct

snd-pci-ps
snd-soc-ps-mach

module by ourselves. I do not know why arch install did not automatically activate that for me.

Did force probing the module fix it?

The YC driver can be compatible with 0x63 if you manually add a quirk (of force load the module)

On here ACPI doesn't report the required node, and since no quirk was available, the driver was registered but it wouldn't enable the platform device.

Meanwhile the PS driver has support for this revision and works out of the box (
If PS users are having issues it likely means the module isn't getting loaded. And the YC driver taking over would do that (since it can't work without a quirk on PS)

Reading the code 0x60 is YC, and that's supported in the driver, 0x63 is PS and that's also supported.

What seems to be happening is that a lot of YC devices don't have the ACPI node required to probe the driver properly, that's why I asked for a dump of it, maybe there's another entry that would do the job?
(No PS device has the ACPI entry needed by the YC driver, so here's why the revert exists, and also because the PS driver works just fine)

I see, very useful inforinformation. thanks for making it clear

Offline

Board footer

Powered by FluxBB