OS2 World Community Forum

OS/2, eCS & ArcaOS - Technical => Multimedia => Topic started by: Paul Smedley on May 06, 2020, 11:00:30 am

Title: Updated Uniaud32 build
Post by: Paul Smedley on May 06, 2020, 11:00:30 am
Hi All,

Prompted by Lewis reporting a problem with no sound on a newish laptop, I built uniaud32.sys last night for the first time since December 2015.

After managing to patch the realtek driver to add some additional chipsets which got Lewis sound, I got interested again in updating Uniaud.

As some may recall, way back when, I was updating the uniaud32 core with the latest ALSA code.... I got as far as ALSA 1.0.24 (released January 2011), but could never get ALSA 1.0.25 (released early 2012) to work. After that, the code was incorporated into the linux kernel, and it all got too hard. Plus, at the time, we still had the hangs with hda audio and I started running linux as my main OS.

Anyhow, after helping Lewis last night, I added the OS/2 patches to the linux kernel 3.0 code and got it building. This represents an alsa-kernel approximately 6 months newer that what's in the last uniaud32 release.

I've tested this in virtualbox with the Intel HDA driver and it still makes sound. That's about the level of testing I've done. I can't be bothered walking upstairs to grab one of the Thinkpads and test on other hardware.

Anyway...... link is http://smedley.id.au/tmp/uniaud32-linux-3.0.zip

Feedback appreciated. If there's interest, I'll continue to try and update the code to hopefully fix upstream bugs and support newer hardware.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 06, 2020, 01:41:56 pm
Hi Paul,

First of all, thanks for your ongoing support of OS/2. It is very appreciated.

I have tested your uniaud32.sys driver today at my TP X250. Unfortunately it didn't work well. I got only distorted sound after using your new driver  :(.
The one from AN I am using is working fine so far, but I thought testing the new one will help.
Please see also "https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_X250" for my review of the working one.

Greetings from Potsdam/Germany
Mike
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on May 06, 2020, 03:37:54 pm
 ::)Pau ::)l ::), sadly this driver resulted in no sound.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on May 06, 2020, 05:05:56 pm
Hi Eugene, can you please provide more feedback?
Like the hardware you are running it, and if it is not longer working compared to a xxx version of Uniaud?

Just in case I guess that the latest version of Uniaud was 2.2.5 / 2018-11-26  (Available at hobbes (https://hobbes.nmsu.edu/?dir=&search=Uniaud-20181126) and Arca Noae site).

Regards
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on May 06, 2020, 05:20:30 pm
Hi Paul

Thanks for doing this and checking the code of Uniaud. I was also good to know more about the technical history of the port and know that because of the complexity with the Linux Kernel it was harder to get up to the ALSA latest version.

I tried your updated driver on a VirtualBox ArcaOS 5 guest. Intel HD Audio. I just backup the old uniaud32.sys and replaced with the new one. Rebooted and all desktop sounds seems to be working normally. Nothing fancy on the config.sys (just DEVICE=C:\MMOS2\UNIAUD32.SYS) I will keep testing to see if I can find something else.

Paul, Is there any other recommended test or information that you also need that we can help with?

Regards
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on May 06, 2020, 06:18:39 pm
Hello Martin the subject of the driver my Dell laptop. Model Inspiron 5537. It was using the last availible drive from Arca Noae.
Title: Re: Updated Uniaud32 build
Post by: David Graser on May 06, 2020, 06:55:14 pm
On my Levovo T430S, the sound works as it should.
Title: Re: Updated Uniaud32 build
Post by: roberto on May 06, 2020, 07:04:50 pm
Well, it's not all bad news, it works perfectly for me. On an Hp8760W, the old one also worked. Thank you anyway.
saludos
Title: Re: Updated Uniaud32 build
Post by: David Graser on May 06, 2020, 08:12:59 pm
One of the problems I have noticed is that when adjusting the sound level through the sound object in the System Setup folder, it has no effect on the volume icon.  When I first installed the new driver, I had the volume.exe set to 0 and was adjusting with the sound object.  I had no sound with the new driver.  Although sound object set to 100%, the volume controller still showed 0.  It was only when I adjusted the sound with the volume controller did I get my sound back.
This with ArcaOS 5.04.

Easy to test:

Use volume icon to set sound level to 0.
Go to System Setup folder > Sound icon.
Set the volume on the sound icon to 100% and test sound.  On my machine, I had no sound.

There will not be sound until one goes back to the volume control icon and adjust the sound there.
Title: Re: Updated Uniaud32 build
Post by: Hermit on May 06, 2020, 08:27:25 pm
Hi, Paul.

On my desktop computer with the updated driver, the sound is gone. In archive log files.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on May 06, 2020, 10:57:47 pm
Haven't tried this set yet as my speakers are broke, but one thing I've noticed with some sound chips is I have to use PMUnimix to turn up the front speakers to actually get sound, once they're turned up, the regular volume control works.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 06, 2020, 11:09:49 pm
Hi All,

Thanks for testing, I'll try and post some more specific comments tonight, but for reference, could anyone that's having troubles with the driver based on the linux kernel 3.0 code, try with http://smedley.id.au/tmp/uniaud32-trunk.zip ? This build represents the current code in the svn trunk at netlabs.

Thanks,

Paul
Title: Re: Updated Uniaud32 build
Post by: Hermit on May 07, 2020, 08:27:35 am
Hi Paul,

The uniaud32-trunk.zip driver version works normally, there is sound. The log is attached.
Title: Re: Updated Uniaud32 build
Post by: Lars on May 07, 2020, 08:46:29 am
One of the problems I have noticed is that when adjusting the sound level through the sound object in the System Setup folder, it has no effect on the volume icon.  When I first installed the new driver, I had the volume.exe set to 0 and was adjusting with the sound object.  I had no sound with the new driver.  Although sound object set to 100%, the volume controller still showed 0.  It was only when I adjusted the sound with the volume controller did I get my sound back.
This with ArcaOS 5.04.

Easy to test:

Use volume icon to set sound level to 0.
Go to System Setup folder > Sound icon.
Set the volume on the sound icon to 100% and test sound.  On my machine, I had no sound.

There will not be sound until one goes back to the volume control icon and adjust the sound there.
That's because the volume set by volume.exe sets the MASTER volume whereas the volume lever in the sound object is the APPLICATION volume (in that case, for the PMSHELL.EXE application as the sound object is a WPS object). In the end, the audio driver just multiplies these two values :-)
Title: Re: Updated Uniaud32 build
Post by: Lars on May 07, 2020, 08:51:11 am
Haven't tried this set yet as my speakers are broke, but one thing I've noticed with some sound chips is I have to use PMUnimix to turn up the front speakers to actually get sound, once they're turned up, the regular volume control works.
In the end, this is always true. But in practice, there are some MB providers that are smart enough to crank up the HW volume of the sound device to the max, I suppose by the BIOS/UEFI code already on system start.
If that is not true, unimix.exe/pmunimix.exe will have to do it. That's even one step  before setting MASTER volume and finally APPLICATION volume.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 07, 2020, 09:36:27 am
Hi, Paul.

On my desktop computer with the updated driver, the sound is gone. In archive log files.

Interesting... the key things I note are the extra Uniaud16(?) lines in the log...

U16: WS::ABS InProcess Now=1 Save=3
U16: WS::ABS InProcess Now=1 Save=3
U16: WS::AB: BufferMode=1
U16: WS::AB: BufferMode=1
U16: WS::AB: BufferMode=1

and finally:
U16: StartStream: ConfigDev failed!!
U16: StartStream: ConfigDev failed!!

No real idea why though....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 07, 2020, 09:43:21 am
Hi All,

thanks to those that have tested.

For anyone that has no sound with the linux-3.0 build, can they please provide logs from a debug build?

Debug build of trunk (roughly what is the latest in ArcaOS):
http://smedley.id.au/tmp/uniaud32-trunk-debug.zip

Debug build of linux-3.0 code:
http://smedley.id.au/tmp/uniaud32-linux-3.0-debug.zip

'copy alsahlp$ trunk-debug.log' (for trunk) and 'copy alsahlp$ trunk-linux-3.0.log' for the 3.0 build should suffice

For good measure, there's also a build based on the linux 3.2 kernel code. I must note that this one produces NO audio for me with HDA hardware in Virtualbox, but does for AC97. I'd be interested in results on real hardware...
http://smedley.id.au/tmp/uniaud32-linux-3.2-debug.zip (release)
http://smedley.id.au/tmp/uniaud32-linux-3.2-debug.zip (debug)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 07, 2020, 02:50:29 pm
Hi Paul,

here comes the test report from my Thinkpad X250:

Debug build of trunk (roughly what is the latest in ArcaOS):
http://smedley.id.au/tmp/uniaud32-trunk-debug.zip
=> working as well as the one from AN

Debug build of linux-3.0 code:
http://smedley.id.au/tmp/uniaud32-linux-3.0-debug.zip
=> sound is distorted

http://smedley.id.au/tmp/uniaud32-linux-3.2-debug.zip (release)
http://smedley.id.au/tmp/uniaud32-linux-3.2-debug.zip (debug)
=> both seems to be same files
=> sound distorted

Unfortunately I have no idea how to create the requested debug output files, sorry.

Greetings from Potsdam/Germany,
Mike
Title: Re: Updated Uniaud32 build
Post by: roberto on May 07, 2020, 05:26:11 pm
Unfortunately I have no idea how to create the requested debug output files, sorry.
I think you need this file, to generate the log.
ftp://ftp.netlabs.org/pub/uniaud/UniLog.zip
saludos
Title: Re: Updated Uniaud32 build
Post by: Lars on May 07, 2020, 06:03:48 pm
Unfortunately I have no idea how to create the requested debug output files, sorry.
I think you need this file, to generate the log.
ftp://ftp.netlabs.org/pub/uniaud/UniLog.zip
saludos

'copy alsahlp$ trunk-debug.log' (for trunk) and 'copy alsahlp$ trunk-linux-3.0.log' for the 3.0 build should suffice
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 07, 2020, 10:07:44 pm
 Hi Paul,

  Both the trunk and Linux 3.0 work fine here, but no sound with Linux 3.2. I'll attach the debug alsahlp$. Thanks for tackling Uniaud again!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 08, 2020, 01:03:48 am
Hi Dave,

  Both the trunk and Linux 3.0 work fine here, but no sound with Linux 3.2. I'll attach the debug alsahlp$. Thanks for tackling Uniaud again!

Interesting that linux 3.2 doesn't work for most, but there is one positive report (even if the sound is distorted).

At this stage, I think I'm going to just press on with the re-sync, and worry about trying to fix the issues later. If the hardware is at least being detected, I'm hoping the fixes will be trivial to get system sounds working.

I also want to test with VLC using the KAI interface - which I believe talks directly to Uniaud32 and bypasses Uniaud16 - I'm wondering if the no system sounds symptom is a sign of issues with uniaud16 talking to uniaud32.

I have the patches from uniaud32-linux-3.2 applied to uniaud32-linux-3.5 - just need to copy over the files and get it to compile. With any luck, I'll have something sometime on Saturday my time.

I also need to grab the T60 and T510 Thinkpads from upstairs at some point so I can test on real hardware....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 08, 2020, 11:49:21 am
Update: uniaud32 based on linux 3.5 compiled, much the same behaviour as the 3.2 build here.  Skipping to 3.16.83 now... there's a bit of code reorganisation but not too bad in terms of application of the patches. If it compiles and recognises the hardware, I'll try and fix the audio issues :)
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 08, 2020, 12:01:04 pm
Hi Paul
Please find attached three log-files for my Thinkpad X250.
Thanks to Lars for pointing out how to create the log files.
Again: The trunk sounds normal, the Linux 3.x give distorted sound.
Greetings from Potsdam/Germany
Mike
Title: Re: Updated Uniaud32 build
Post by: Hermit on May 08, 2020, 08:16:19 pm
Paul,  on my desktop PC

uniaud32-trunk-debug.zip   - normal sound,
uniaud32-linux-3.0-debug.zip - there is no sound,
uniaud32-linux-3.2-debug.zip   - normal sound.

The logs I apply.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 08, 2020, 10:00:22 pm
Paul,

 FYI - I tried VLC with the Linux 3.2 driver, but still no sound. No errors at boot though, and it shows up in the hardware manager...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 09, 2020, 03:54:33 am
Hi Dave!

FYI - I tried VLC with the Linux 3.2 driver, but still no sound. No errors at boot though, and it shows up in the hardware manager...

Are you able to test with http://smedley.id.au/tmp/uniaud32-linux-3.1.zip and confirm you still get sound?

3.16.x is still in work, but this might help reduce the codeset to analyse as to why sound stops working. This one still produces sound with HDA in Virtualbox.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 09, 2020, 02:10:07 pm
Paul,

  I gave the Linux 3.1 version a shot and it seems to work fine... I'll upload the debug alsahlp$ for it in case it's interesting...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Pete on May 09, 2020, 03:03:42 pm
Hi Paul

Tested uniaud32-linux-3.2-debug.zip on a Gigabyte 990FXA-UD3 mainboard with this onboard sound

 Bus 0 (PCI Express), Device Number 20, Device Function 2
 Vendor 1002h Advanced Micro Devices, Inc. [AMD/ATI]
 Device 4383h SBx00 Azalia (Intel HDA)
 Subsystem ID A1321458h Unknown
 Subsystem Vendor 1458h Gigabyte Technology Co., Ltd
 System IRQ 16, INT# -


Changed driver, rebooted, startup sound played OK, other system sounds play OK, CD playback fine.
Sound from both (s)mplayer and vlc when using kai:DART but not when using kai:uniaud


Used the following command to generate the list of Controls and settings

   [M:\MMOS2]unimix -list > new_driver.lst

Of interest is that the file generated is larger than the unimixer.txt file created when using the above unimix -list command with an earlier driver - v2.02.02 according to the info at the top of unimixer.txt

The reason?

A quick read through comparing both files reveals that new_driver.lst has these 2 sections which are missing in the unimixer.txt file:-

Control ID: 11
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Headphone Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 59
  value 1: 59

Control ID: 19
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Auto-Mute Mode]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 2, item: 0, name: [Disabled]
  value 0: 1


So, on this real hardware the uniaud32-linux-3.2-debug.zip seems to be working fine.

People testing this build who reported no audio may want to double check the mixer settings.
Make sure that the following are set:

[Master Playback Volume] - set fairly high (adjust down to suit if it works)

[Master Playback Switch] - must be On ie value of 1


Attached the trunk-linux-3.2-log.txt in case it is of any interest.


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on May 10, 2020, 12:21:05 am
 Uniaud32-linux-3.2-debug.zip ended up working great on both my laptop and my desktop. The volume was very low. I used PMmixer to raise it on my laptop. My desktop has a USB audio device. So PMmix doesn't work.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 10, 2020, 02:56:03 am
Hi Pete,

unimix -list output for linux 3.1 (sound works) and 3.2 (sound doesn't work) attached....

Hardware is 'Intel HDA' under Virtualbox....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 10, 2020, 05:12:19 am
Hi Pete,

  Good point to check the mixer settings. I notice a couple of additional settings in my Linux 3.2 mixer as well (compared to earlier ones) - expanded support in that version for my sound chip maybe? Master Playback and Master Switch are set the same in all the ones tried though, so something else is going on with Linux 3.2 here...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 10, 2020, 10:55:19 am
Guys,

Given the linux 3.2 build works for some, here's the linux-3.5 build...
http://smedley.id.au/tmp/uniaud32-linux-3.5.zip - release & debug builds included.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Pete on May 10, 2020, 03:09:48 pm
Hi Paul

The only difference between 3.1.txt and 3.2.txt seems to be:-

   3.1 Control iD: 1 has values set to 0
   3.2  Control iD: 1 has values set to 64.

I would have expected 3.1.txt to show that control with values up to 64 but not 0 as 3.1 produces sound at your end.

It might be worth trying 3.2 with Control ID: 1 values set to 0 - it if works for 3.1...


Regards

Pete



Title: Re: Updated Uniaud32 build
Post by: Pete on May 10, 2020, 05:26:50 pm
Hi Paul

Installed the debug 3.5 build, rebooted, sound seems OK with this build - no obvious problems on limited testing.

Setting vlc to use kai:uniaud results in no sound from vlc; resetting to DART works fine.


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: David Graser on May 10, 2020, 08:19:35 pm
Hi Paul

Installed the debug 3.5 build, rebooted, sound seems OK with this build - no obvious problems on limited testing.

Setting vlc to use kai:uniaud results in no sound from vlc; resetting to DART works fine.


Regards

Pete
Works great on my Lenovo Thinkpad T430S.
 PMView123 1.40, VLC 3.0.10, and system soounds all work/
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 10, 2020, 09:07:06 pm
Paul,

  No sound with this one either - debug log attached...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 10, 2020, 11:31:48 pm
Guys,

For those that have sound with any uniaud32.sys based on linux-3.2 or above, can you please confirm what version of uniaud16.sys you're running:
'bldlevel x:\mmos2\uniaud16.sys' should do it.

FWIW - I'm testing with the version shipped in ArcaOS 5.0.4
1.9.7 - built 25 Jun 2017

Thanks,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 10, 2020, 11:32:56 pm
Hi Dave,

Hi Paul

Installed the debug 3.5 build, rebooted, sound seems OK with this build - no obvious problems on limited testing.

Setting vlc to use kai:uniaud results in no sound from vlc; resetting to DART works fine.

Pete
Works great on my Lenovo Thinkpad T430S.
 PMView123 1.40, VLC 3.0.10, and system soounds all work/

Are you able to supply a log for this?

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Pete on May 11, 2020, 04:34:00 am
Hi Paul

I have attached debug-linux-3.5.-log.txt

The uniaud16.sys in use is:-

[M:\MMOS2]bldlevel uniaud16.sys
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature:       @#Netlabs:1.9.7#@##1## 25 Jun 2017 17:11:40     DAZAR1    ::::7
::@@Universal Audio MMPM/2 Driver for OS/2
Vendor:          Netlabs
Revision:        1.09.7
Date/Time:       25 Jun 2017 17:11:40
Build Machine:   DAZAR1
File Version:    1.9.7
Description:     Universal Audio MMPM/2 Driver for OS/2


The above is from the AN update package Uniaud-20181126.exe


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: David Graser on May 11, 2020, 07:22:37 pm
Hi Dave,

Hi Paul

Installed the debug 3.5 build, rebooted, sound seems OK with this build - no obvious problems on limited testing.

Setting vlc to use kai:uniaud results in no sound from vlc; resetting to DART works fine.

Pete
Works great on my Lenovo Thinkpad T430S.
 PMView123 1.40, VLC 3.0.10, and system soounds all work/

Are you able to supply a log for this?

Cheers,

Paul

Hi Paul
I am not sure if you wanted a log from me since my system works great.  Here is a log using your latest test build.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 11, 2020, 11:06:11 pm
Hi Dave,

I am not sure if you wanted a log from me since my system works great.  Here is a log using your latest test build.

Thanks - I'm hoping the logs of a working system may give some clues about why some systems don't work.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 12, 2020, 05:17:04 am
Update: copying over the 'core' code from linux 3.1 to the linux 3.2 build (with a couple of build breaks) got sound working. Next step to work out which specific change in core broke sound for us and fix it.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on May 12, 2020, 05:32:19 am
I tried the 3.2 Uniaud. Worked well here on real hardware and using headphones (speakers are broke). Got the option to adjust headphone volume that was never there. I'm hoping that it will allow using the headphones without the speakers also outputting sound.
Only problem was testing libkai support by starting SM up with "SET KAI_AUTOMODE=UNIAUD" and playing Youtube videos. If I paused the video, sound turned into clacking. Adjusting the volume down to zero and back up resulted in sound once again working. Now to test the 3.5 one.
 Vendor 8086h Intel Corporation
 Device 1C20h 6 Series/C200 Series Chipset Family High Definition Audio Controller
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 13, 2020, 11:19:22 am
Hi all,

I built uniaud32 based on 3.1.9 today. I almost hoped it wouldn't produce sound on my test system here, as it would have meant that I could use the releases from 3.1 -> 3.1.9 to help identify where things are going wrong for us.

Anyway, for those that have no sound in 3.2; but had sound in 3.1, I'm really interested in confirming that this one still makes sound.

http://smedley.id.au/tmp/uniaud32-linux-3.1.9.zip - debug and release builds included.

I still hope this will help narrow down the breaking (for us) changes in 3.2 and above....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 13, 2020, 11:37:47 am
Hi Paul,

Please find attached the logs for v3.5 and v3.1.9.
As expected both give only distorted sound on my Thinkpad X250.

Greetings form Potsdam/Germany,
Mike
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 13, 2020, 12:16:40 pm
Paul,

  Can't get the computer to boot with either the debug or release version of 3.1.9 - after the disk check there is a long pause, then a bunch of messages flash by and end with 'Not enough memory to run PMShell.exe on the Protshell = statement in Config.sys' (I have 8GB). Never saw that before!

Regards,

Title: Re: Updated Uniaud32 build
Post by: OS4User on May 13, 2020, 12:20:34 pm
'Not enough memory to run PMShell.exe on the Protshell = statement in Config.sys' (I have 8MB). Never saw that before!

It possible that new Uniaud32 consumes more system memory - try to decrease VAL.
Title: Re: Updated Uniaud32 build
Post by: OS4User on May 13, 2020, 12:29:38 pm
http://smedley.id.au/tmp/uniaud32-linux-3.5.zip - release & debug builds included.

3.5 works ok on my CA0106/CA0111 [SB Live!/Audigy/X-Fi Series]
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 13, 2020, 01:03:55 pm
 OS4User - Tried decreasing VAL in increments of 512 down to 1024, but always the same result - 'Not enough memory...'.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Pete on May 13, 2020, 10:21:58 pm
Hi David

I seem to recall seeing a similar message many years ago but am struggling to remember details: I think it was either a failing PSU or a choked up CPU cooling fan.

Might be worth checking inside the system case...


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 13, 2020, 10:41:06 pm
Pete,

 System case is fine, the official UniAud works fine, and Paul's builds before 3.1.9 work fine. I'll upload a 'testlog generic' in case anyone is interested. Must be something else...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 13, 2020, 10:53:58 pm
Try running lxlite on uniaud32.sys...
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 13, 2020, 11:18:51 pm
Hey Paul,

  Tried 'lxlite uniaud32.sys' on both the debug and release versions of 3.1.9 and it made them much smaller, but both still result in 'Not enough memory...' at boot. I am now running the new official version released at http:\\trac.netlabs.org\uniaud (20200506.exe) with no trouble...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on May 14, 2020, 02:05:57 am
Try reducing the THREADS number in config.sys, perhaps 768 or 512.
There's also a related one which I'm having trouble remembering, PROCESSES I think that needs adding to change.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 14, 2020, 07:12:50 am
progress! I copied hda_intel.c from linux 3.1.9 into my linux 3.2.0 source tree, rebuilt the driver and got sound. Now to review the changes and work out which change is the breaking one..

It's in the attached somewhere :)
Title: Re: Updated Uniaud32 build
Post by: Lars on May 14, 2020, 03:35:30 pm
As a test can you undefine CONFIG_X86 (only for the given file if possible) and rebuild?
Somebody has tried to mark memory ( sound data ? DMA accessed memory...) as "write combined". I guess that is flawed and would also explain the weird stuff with PMSHELL.EXE not having enough memory (which is obviously not true).
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 14, 2020, 11:26:08 pm
Hi Lars,

As a test can you undefine CONFIG_X86 (only for the given file if possible) and rebuild?
Somebody has tried to mark memory ( sound data ? DMA accessed memory...) as "write combined". I guess that is flawed and would also explain the weird stuff with PMSHELL.EXE not having enough memory (which is obviously not true).

99.9% certain that CONFIG_X86 is not defined for us... That's a new define in newer linux builds - our config.h is manually created.

I will double check a little later.

My guess at the moment is that it's the changes around buffer sizes that break things for us in 3.2.0.

More to come later today.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 15, 2020, 12:40:55 am
Hi All,

My guess at the moment is that it's the changes around buffer sizes that break things for us in 3.2.0.

Well I was wrong - it was actually these chunks that broke things:
Code: [Select]
--- hda_intel.c 2020-05-13 14:27:46.000000000 +0930
+++ hda_intel.c.32 2020-05-14 14:47:24.000000000 +0930
@@ -378,7 +398,7 @@
  */
  unsigned char stream_tag; /* assigned stream */
  unsigned char index; /* stream index */
- int device; /* last device number assigned to */
+ int assigned_key; /* last device# key assigned to */
 
  unsigned int opened :1;
  unsigned int running :1;
@@ -1582,6 +1673,9 @@
 {
  int dev, i, nums;
  struct azx_dev *res = NULL;
+ /* make a non-zero unique key for the substream */
+ int key = (substream->pcm->device << 16) | (substream->number << 2) |
+ (substream->stream + 1);
 
  if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK) {
  dev = chip->playback_index_offset;
@@ -1593,12 +1687,12 @@
  for (i = 0; i < nums; i++, dev++)
  if (!chip->azx_dev[dev].opened) {
  res = &chip->azx_dev[dev];
- if (res->device == substream->pcm->device)
+ if (res->assigned_key == key)
  break;
  }
  if (res) {
  res->opened = 1;
- res->device = substream->pcm->device;
+ res->assigned_key = key;
  }
  return res;
 }

Reverting these gets me sound in Virtualbox with both the linux 3.2 and linux 3.5 code.

New builds (each containing debug and release builds):
http://smedley.id.au/tmp/uniaud32-linux-3.2.zip
http://smedley.id.au/tmp/uniaud32-linux-3.5.zip

Feedback appreciated!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 15, 2020, 03:41:51 am
...and for good measure, a build based on linux 3.2.102 - http://smedley.id.au/tmp/uniaud32-linux-3.2.102.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 15, 2020, 12:35:44 pm
 Hi Paul,

  Looks like you've been busy! I tried all 3 and it is a mixed bag:

Linux 3.2: Can't boot, get the 'Not enough memory...' message described earlier
Linux 3.2.102: Can't boot, ends with a 'System detected an internal processing error....' type message.
Linux 3.5: Seems to work fine (system sounds and YouTube)

 I'll attach a debug log for 3.5. Thanks again!
Title: Re: Updated Uniaud32 build
Post by: Pete on May 15, 2020, 03:52:37 pm
Hi Paul

Installed the debug build from your latest uniaud32-linux-3.5.zip and can report it works as well as the previous build so far.


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 15, 2020, 04:10:41 pm
Hi Paul,

I've tested all three new builds on my Thinkpad X250.
All gave only distorted sound for me.
Please find attached the log-files.

Wish you a nice weekend.
Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 16, 2020, 09:07:43 am
Hi Mike,

Can you please test with: http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK.zip

Ta,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 16, 2020, 12:59:21 pm
Hi Paul,

I have tested the special version uniaud32-linux-3.2.102-MikeK.zip today.
The result is mixed.

After booting I got only distorted sound from the system sound and from pm123 (v. 142 Beta 1 from May 9, 2020).

After reboot and having the following lines in my startup.cmd enabled:
"unimix -id1 -cnt0 -val87"
"unimix -id1 -cnt1 -val87"
I got clear sound from system sound. Even volume control was working well.
Once I opened UniMixpm.exe the sound become distorted again.
unimixpm.exe is from 2005-08-12, size 218.282 Byte, found in C:\MMOS2\DLL\

Same behaviour after next reboot and starting pm123 instead of UniMixpm.exe.

Greetings from Potsdam/Germany
Mike
Title: Re: Updated Uniaud32 build
Post by: David Graser on May 16, 2020, 08:36:22 pm
Hi Mike,

Can you please test with: http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK.zip

Ta,

Paul

I just dusted off my old Lenova Thinkpad X220 and copied this driver over to it and rebooted.  Sound works fine.  Do not know what it is about the X250 that is different.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 17, 2020, 11:08:22 am
Uggh - linux 3.6.x introduces some quite significant changes... might be a while before we see another bump in kernel source....
Title: Re: Updated Uniaud32 build
Post by: Pete on May 17, 2020, 01:56:06 pm
Hi Mike

There is a later PMunimix here https://hobbes.nmsu.edu/download/pub/os2/apps/mmedia/util/mixers/PMUniMix-bin-0_82.zip

Might also be worth checking which unimix.exe you have installed. This version works OK here with Pauls 3.2 and 3.5 uniaud builds

25-06-17  5:12p        27,585      0 a---  unimix.exe

The current "official" release has this build of unimix.exe included ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20200506.exe

I suggest you place both unimix.exe and PMunimix.exe in \mmos2 rather than \mmos2\dll.


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Jan-Erik Lärka on May 17, 2020, 03:17:38 pm
I've tried the versions mentioned here up to 3.5 and I can notice improvements...
* PM123 play web radio 44.1 kHz just as nice as before
* Unimix 0.82 doesn't crash and can now allow microphone in, though doesn't give any input
* Mplayer stutter and freeze the movie clips I play
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 18, 2020, 01:54:23 am
Hi Jan-Erik,

I've tried the versions mentioned here up to 3.5 and I can notice improvements...
* PM123 play web radio 44.1 kHz just as nice as before
* Unimix 0.82 doesn't crash and can now allow microphone in, though doesn't give any input
* Mplayer stutter and freeze the movie clips I play

Thanks for the report. Re: the MPlayer issues, are you able to try http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK.zip ?
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 18, 2020, 11:18:44 am
Hi Mike

There is a later PMunimix here https://hobbes.nmsu.edu/download/pub/os2/apps/mmedia/util/mixers/PMUniMix-bin-0_82.zip
Might also be worth checking which unimix.exe you have installed. This version works OK here with Pauls 3.2 and 3.5 uniaud builds
25-06-17  5:12p        27,585      0 a---  unimix.exe
The current "official" release has this build of unimix.exe included ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20200506.exe
I suggest you place both unimix.exe and PMunimix.exe in \mmos2 rather than \mmos2\dll.
Regards
Pete

Hi Pete,

I have installed the latest version for Uniaud from David as recommended. Everything was working well. Luckily also the pmUniMix is now working again. Version 0.8.2 crashed always, version 0.7.0 started but did no show any control at all in the past. Now both version are running. With Davids Uniaud also the Mic control is there, but I have no means of testing it.

Next step was to re-test Pauls different Uniaud versions. Unfortunately all new version didn't work well. Mostly there is sound, but never clear, always broken and distorted. pmUniMix is still starting, but doesn't show the Mic control when Pauls Uniaud is loaded.

I created the debug files for tests I have done today and returned to Davids latest version for now.

Greetings from Potsdam/Germany,
Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 18, 2020, 11:56:28 am
Hi Mike!

Next step was to re-test Pauls different Uniaud versions. Unfortunately all new version didn't work well. Mostly there is sound, but never clear, always broken and distorted. pmUniMix is still starting, but doesn't show the Mic control when Pauls Uniaud is loaded.

I added the PCI_QUIRKS listed at line 15077 of patch_realtek.c from http://trac.netlabs.org/uniaud/changeset/609 and rebuilt.

http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK2.zip - note this does not have the change that was included in uniaud32-linux-3.2.102-Mike2.zip
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 18, 2020, 12:16:26 pm
Hi Mike!
I added the PCI_QUIRKS listed at line 15077 of patch_realtek.c from http://trac.netlabs.org/uniaud/changeset/609 and rebuilt.
http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK2.zip - note this does not have the change that was included in uniaud32-linux-3.2.102-Mike2.zip

Hi Paul,

Just tested the provided version. Sound is still disrupted. :(

Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 18, 2020, 11:32:13 pm
Hi Mike,

Last try (for now): http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK3.zip

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David Graser on May 19, 2020, 12:06:51 am
Hi Mike,

Last try (for now): http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK3.zip

Cheers,

Paul

Hi Paul,
I noticed David just released a version of the Uniaudio drivers.  Does your attempts included his fixes?  I noticed KO had a Uniaudio driver on his home page with some kind of patch.  Maybe we need to get all these patches into a unified driver.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 19, 2020, 12:26:12 am
Hi David,

Hi Mike,

Last try (for now): http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK3.zip


I noticed David just released a version of the Uniaudio drivers.  Does your attempts included his fixes?  I noticed KO had a Uniaudio driver on his home page with some kind of patch.  Maybe we need to get all these patches into a unified driver.

My builds are based on a set of diffs I ran against the Netlabs SVN code a couple of weeks ago. You'll notice two recent checkins at Netlabs, one referencing Paul (me) and another referencing Andy.

My changes (to add some PCI IDs to the realtek driver) were included, I mistakenly assumed that Andy's changes were already included in the newer ALSA code from the kernel. The 'Mike3' build above should include the relevant changes from Andy to the realtek driver.

I didn't know KO had built uniaud32, happy to incorporate the changes if they're still relevant.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 19, 2020, 12:28:33 am
Oh, and I forgot to mention, I created a branch in Netlabs SVN and uploaded the 3.2.102 code - http://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-linux-3.2.102
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on May 19, 2020, 06:15:26 am
Hi David,

Hi Mike,

Last try (for now): http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK3.zip


I noticed David just released a version of the Uniaudio drivers.  Does your attempts included his fixes?  I noticed KO had a Uniaudio driver on his home page with some kind of patch.  Maybe we need to get all these patches into a unified driver.

My builds are based on a set of diffs I ran against the Netlabs SVN code a couple of weeks ago. You'll notice two recent checkins at Netlabs, one referencing Paul (me) and another referencing Andy.

My changes (to add some PCI IDs to the realtek driver) were included, I mistakenly assumed that Andy's changes were already included in the newer ALSA code from the kernel. The 'Mike3' build above should include the relevant changes from Andy to the realtek driver.

I didn't know KO had built uniaud32, happy to incorporate the changes if they're still relevant.

Cheers,

Paul

I believe KO's patches are for small buffers, such as VPC/Odin uses for its audio. 
If you look in sound.c, search for:
  /* Change from Andy. If statement added around 5 statements.
   * Andy says: TODO:  determine why small buffers are a problem for this code
   * 20100107: DAZ: This change is questionable.
   * 20180109: DAZ: This changes causes problems for many systems including intermittent
   * sound and unexpected hangs. Reverted by commenting out the if statement.
   */
The if was my attempt to fix the small buffers issue.  I have locally changed the if to:
  if (periodsize > 416)
rather than the orginal:
if (periodsize > 6 * minperiodsize)
It seems to be better but that may again be on limited hardware.  Basically, the if is blocking out code that, when it was added, caused the VPC sound issues when the periodsize is small.
I thought David had applied the below patch to backout an earlier change, you probably should as it seems it breaks more than it fixed:
Code: [Select]
Index: soundmixer.c
===================================================================
--- soundmixer.c (revision 606)
+++ soundmixer.c (working copy)
@@ -408,13 +408,6 @@
       idx = pHandle->controls[OSS_MIXER_FRONT].idxVolume;
       idxMute = pHandle->controls[OSS_MIXER_FRONT].idxMute;
     }
-    /* HDA codecs workaround from Andy */
-    if (idx == -1)
-    {
-      idx = pHandle->controls[OSS_MIXER_SPEAKER].idxVolume;
-      idxMute = pHandle->controls[OSS_MIXER_SPEAKER].idxMute;
-    }
-    /* HDA codecs workaround from Andy */
     break;
   case OSS32_MIX_VOLUME_MIDI:
     idx = pHandle->controls[OSS_MIXER_SYNTH].idxVolume;
I don't recall the exact volume issue it creates on some systems.
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on May 19, 2020, 10:21:15 am
Hi Mike,
Last try (for now): http://smedley.id.au/tmp/uniaud32-linux-3.2.102-MikeK3.zip
Cheers,
Paul

Hi Paul,

I have tested the third version as well. I can only see one change in pmUnimix. There is now a control for the mic. Please see attached picture.
(http://Paul0000.jpg)

The sound is still not clear. Sorry for this bad test result. Attached again the debug log file.

Greetings,
Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 19, 2020, 11:45:32 am
Hi Mike, bugger :( I was hoping that would fix it. I'll try get back to this sometime soon - I'm distracted at the moment looking at Samba Server again.
Title: Re: Updated Uniaud32 build
Post by: roberto on May 19, 2020, 10:39:10 pm
Hello
Excuse me but perhaps this failure may be of your interest.
If I am listening to sound, either from the seamonkey or from the vlc, I select the * find * system
 with the right mouse find ...
 Sometimes only when opening the find, in other cases when opening properties, of the searched and
  found file, I select file.File - properties - filing cabinet. Suddenly it makes a little noise, and from
   there on the *wps* will go crazy, and I have to restart. Although the system still works.

I attach two files, one before the error and the other after the error.
The version of :
 9/07/17  1:52         68.682      0 ----  uniaud16.sys
10/05/20 11:52         37.332    252 a---  uniaud32.sym
10/05/20 11:52        651.742    302 a---  uniaud32.sys

 {0}[c:\mmos2] bldlevel uniaud16.sys
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature:       @#Netlabs:1.9.7#@##1## 25 Jun 2017 17:11:40     DAZAR1    ::::7
::@@Universal Audio MMPM/2 Driver for OS/2
Vendor:          Netlabs
Revision:        1.09.7
Date/Time:       25 Jun 2017 17:11:40
Build Machine:   DAZAR1
File Version:    1.9.7
Description:     Universal Audio MMPM/2 Driver for OS/2

{0}[c:\mmos2] bldlevel uniaud32.sys
Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature:       @#Netlabs:2.02.05#@##1## 10 May 2020 18:22:07     ARCAOS-D0C:3.
5.0:::05::SVN@@Universal Audio Driver for OS/2 (c)  Netlabs
Vendor:          Netlabs
Revision:        2.02.05
Date/Time:       10 May 2020 18:22:07
Build Machine:   ARCAOS-D0C
ASD Feature ID:  3.5.0
FixPak Version:  SVN
File Version:    2.2.5
Description:     Universal Audio Driver for OS/2 (c)  Netlabs

It really seems that the conflict is more of the wps than the uniaud32, since in the rest of the cases
 it works well.

Saludos

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on June 04, 2020, 07:54:15 am
Hey Mike,

Hi Mike, bugger :( I was hoping that would fix it. I'll try get back to this sometime soon - I'm distracted at the moment looking at Samba Server again.

I ended up ordered a X250 Thinkpad for a fairly reasonable price ($A250), that should arrive next week. Hopefully that will allow me to reproduce your issues - and hopefully fix them too :) It's only an i3 but has 8gb ram and a 250gb SSD.

It will also me a much smaller/lighter machine for testing things natively on ArcaOS than the T510 I currently have. It also has Win10 and Office which should be useful for testing as a client against Samba Server on ArcaOS.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on June 04, 2020, 11:04:05 am

I ended up ordered a X250 Thinkpad for a fairly reasonable price ($A250), that should arrive next week. Hopefully that will allow me to reproduce your issues - and hopefully fix them too :) It's only an i3 but has 8gb ram and a 250gb SSD.

Cheers,

Paul

Hi Paul,
That sounds good. Installation of AOS shall be working from USB thumb drive without problems. You just need to set it to USB2 mode in the BIOS. And you have to disable the UEFI as well as long as we don't have it. Even installation of eCS2.2Beta2 shall be working using an external CD_ROM/DVD drive. An older eCS2.1 needs to be copied with all recent drivers to be able to boot.
Regards,
Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on June 04, 2020, 11:19:53 am
Hey Mike!


I ended up ordered a X250 Thinkpad for a fairly reasonable price ($A250), that should arrive next week. Hopefully that will allow me to reproduce your issues - and hopefully fix them too :) It's only an i3 but has 8gb ram and a 250gb SSD.

That sounds good. Installation of AOS shall be working from USB thumb drive without problems. You just need to set it to USB2 mode in the BIOS. And you have to disable the UEFI as well as long as we don't have it. Even installation of eCS2.2Beta2 shall be working using an external CD_ROM/DVD drive. An older eCS2.1 needs to be copied with all recent drivers to be able to boot.

Thanks for the tips - I hadn't gotten as far as thinking about install yet :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 31, 2020, 01:00:57 am
Hi All,

I bumped the code to that based on linux 3.6.11 - based on a reboot, I still get sound in a Virtualbox VM...

http://smedley.id.au/tmp/uniaud32-linux-3.6.11.zip - debug and release builds included.

I'd like to slowly get us to a supported kernel, no idea how difficult this will get as new features are added to the linux kernel that need to be added to uniaud - but as time permits - I'll try and keep looking at this, for as long as there is some interest.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on July 31, 2020, 01:44:58 am
 Hi Paul,

  It's working just fine here - system sounds, WPS MP3 objects and YouTube - all on ArcaOS 5.0.5 on bare metal 8086:A170. Thanks for continuing to progress with this! Any chance for multiple streams (so system sounds don't interrupt something else, they are just mixed in)?... or even surround...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 31, 2020, 02:17:43 am
Hi David,

  It's working just fine here - system sounds, WPS MP3 objects and YouTube - all on ArcaOS 5.0.5 on bare metal 8086:A170. Thanks for continuing to progress with this! Any chance for multiple streams (so system sounds don't interrupt something else, they are just mixed in)?... or even surround...

Thanks for testing. In terms of multiple streams - I believe the SBLive driver supported this so it definitely should be possible to support. Surround is likely more problematic unless there was an interface for apps to talk directly to uniaud32 - like kai?

As time permits, I'll keep plugging away at bumping the kernel version - noting that it's over 2 months between 3.5.x builds and 3.6.x builds :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 31, 2020, 07:44:46 am
http://smedley.id.au/tmp/uniaud32-linux-3.7.9.zip - based on linux 3.7.9 code....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 31, 2020, 11:03:31 am
http://smedley.id.au/tmp/uniaud32-linux-3.8.13.zip <== linux 3.8.13 code
Title: Re: Updated Uniaud32 build
Post by: David McKenna on July 31, 2020, 10:15:28 pm
 Wow... those 2 months just blew by!  :D Both of those versions work well here - same as the 3.6 version. I'll stick with the 3.8 version for now and let you know if I hit any issues. Thanks again!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 01, 2020, 03:18:28 am
Linux 3.10.108 code:
http://smedley.id.au/tmp/uniaud32-linux-3.10.108.zip

With this one, front playback sound was off by default with the HDA hardware emulated by Virtualbox - for a moment, I thought sound was broken :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 01, 2020, 04:06:08 am
Linux 3.12.74: http://smedley.id.au/tmp/uniaud32-linux-3.12.74.zip
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 01, 2020, 07:54:33 am
Hey Guys,

I have 3.13.11 compiling - but it's giving me a TRAP 000E (attached) and also 3.14.79 compiled - but that one isn't detecting any hardware :/
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 01, 2020, 03:31:27 pm
Paul,

  Tried 3.10 and 3.12, but both of those give me a TrapE in UNIAUD32, very similar to the one you show for 3.13. I actually got 3.12 to boot to the desktop once, but there was no sound and it did not respond to unimix commands (they processed as if accepted, but no effect on sound).

Regards,
Title: Re: Updated Uniaud32 build
Post by: roberto on August 01, 2020, 03:48:04 pm
Links to 3.13 are missing. and 3.14 when ready.
saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 01, 2020, 11:34:59 pm
Links to 3.13 are missing. and 3.14 when ready.

Correct - I'm not going to release something that doesn't even work for me....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 01, 2020, 11:36:32 pm
Hey David,

  Tried 3.10 and 3.12, but both of those give me a TrapE in UNIAUD32, very similar to the one you show for 3.13. I actually got 3.12 to boot to the desktop once, but there was no sound and it did not respond to unimix commands (they processed as if accepted, but no effect on sound).

Interesting... I'll try and test on bare metal hardware today and see if I get the same thing.

meanwhile, Steven is trying to help me debug the Trap 000E

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 02, 2020, 01:54:52 am
Hey David,

  Tried 3.10 and 3.12, but both of those give me a TrapE in UNIAUD32, very similar to the one you show for 3.13. I actually got 3.12 to boot to the desktop once, but there was no sound and it did not respond to unimix commands (they processed as if accepted, but no effect on sound).

Interesting... I'll try and test on bare metal hardware today and see if I get the same thing.

Well, the "good" news is that on my Thinkpad X250, I can reproduce the trap in 3.10 & 3.12, and 3.8 works.

Building 3.9 now to try and narrow down where things broke....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 02, 2020, 08:55:01 am
OK more tests..
3.9.11 - http://smedley.id.au/tmp/uniaud32-linux-3.9.11.zip

3.13.11 - http://smedley.id.au/tmp/uniaud32-linux-3.13.11.zip  <== includes hackaround with help from Steven Levine for the TRAP000E I was seeing
3.14.79 - http://smedley.id.au/tmp/uniaud32-linux-3.14.79.zip <== doesn't trap, but fails to find an audio driver

Note: I haven't tried these on bare metal yet
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 02, 2020, 09:28:03 am
Update: same behaviour as above on an X250.
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 02, 2020, 11:16:51 am
Hi, I am way behind, but anyway.. I have a Thinkpad X61 with ArcaOS 5.05 recently installed. Like with any other installation before, I only had sound when docked to the docking station. Outside the docking station sound only worked after adding unimix line -id3 -val1. This made the speaker work, audio can be muted by using the hardware 'knob' for audio on-off, volume control knobs do not work.
Using 3.8.13 now for the first time gives me sound when undocked and without unimix line -id3 -val1. Control by hardware knobs is the same as before
Trying 3.12.74 gives me a trap at boot.

Paul, thank you for all your efforts!



Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 02, 2020, 01:38:28 pm
Paul,

  Tried the latest batch - here is what I see:

3.9: Boots, but no sound. Front Playback Volume (id2) is set to zero, but turning it up using unimix or PMUnimix does not get sound (even though 'unimix -list' shows volumes are turned up). I loaded the debug drivers and created an alsahlp$ log after booting, setting volumes and playing an MP3 file.

3.13: TrapE during boot at location: ##0168:fff0f667 - 000f:e667

3.14: During boot I see this message:
 open ret = -19
 OSS32_QueryNames() ret = -19
 Unable to attach 32 bit driver
 The device driver C:\MMOS2\UNIAUD16.SYS was not installed

 Then the boot proceeds to the desktop with no sound (and no driver loaded)

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 02, 2020, 11:13:08 pm
Hi David,

Thanks for testing. Weird that you see a trap 000E with 3.13 :( Can you take a photo of the whole trap screen with the debug driver?

re: 3.14 - that's exactly what I see here.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 02, 2020, 11:16:34 pm
Hi Holger,

Hi, I am way behind, but anyway.. I have a Thinkpad X61 with ArcaOS 5.05 recently installed. Like with any other installation before, I only had sound when docked to the docking station. Outside the docking station sound only worked after adding unimix line -id3 -val1. This made the speaker work, audio can be muted by using the hardware 'knob' for audio on-off, volume control knobs do not work.
Using 3.8.13 now for the first time gives me sound when undocked and without unimix line -id3 -val1. Control by hardware knobs is the same as before
Trying 3.12.74 gives me a trap at boot.

Paul, thank you for all your efforts!

Great to hear this improves your audio experience with the X61! It does make sense - current ArcaOS audio drivers are based on ALSA 1.0.24 (with a handful of backports from newer kernels), which was released in April 2011 - almost 10 years ago. It's incredible that it still supports some of the hardware in use with ArcaOS.

Even linux 3.8.13 which improves things for you, was released in May 2013 - so still a long time ago!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: roberto on August 03, 2020, 12:48:16 am
Hello. My tests have been the same as David M.

What I'm seeing is that with different color resolutions of the Panorama driver, I get different operating results.
With 1920x1080 64K , With the sound driver that I like the most, I lose the power to increase and decrease the volume with
 Alt + mouse wheel
 At the same time the Manhonj game works better for me

With 1920x1080 16M , It happens to me the other way around, I can activate the volume with Alt + mouse wheel,
 but the Manhonj game, it is a chaos.
 If I cancel the sound the game works fine.

My favorite driver:

 7/05/20 17:28         36.148      0 a---  uniaud32.sym
 7/05/20 17:28        700.316      0 a---  uniaud32.sys

 I attach the photos at 16m and 64k in case they are of any use, from driver 3.13.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 03, 2020, 05:41:30 am
Guys,

For any traps, please state which version and build type the trap corresponds to. Ie 3.13.11 debug or 3.12.74 release.

Otherwise, the trap is very difficult to interpret.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 03, 2020, 07:59:10 am
Tested 3.8.13 on a Dell 790, perhaps 7 years old. Strange thing is my unaud.dll seems to have vanished, breaking the mixer. Once I restored it and turned up the PCM volume, seems to work fine. The headphone volume also works independently of the PCM volume unlike the ArcaOS version.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 04, 2020, 01:19:18 am
Hi Paul

Linux 3.12.74: http://smedley.id.au/tmp/uniaud32-linux-3.12.74.zip
FYI: I tried this version on real hardware (ThinkPad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420)) and it traps on boot.

I will try to test your other releases.

Regards
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 04, 2020, 01:33:19 am
Hi Paul

I also tried these drivers on the Thinkpad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420). I had only tested the "release" versions.

3.9.11 - http://smedley.id.au/tmp/uniaud32-linux-3.9.11.zip
Does not trap. Produces sounds. Volume control works, but it only sounds on the attached headset, not on the laptop speaker. Possible more unimix tweeaking is required from my side.

3.13.11 - http://smedley.id.au/tmp/uniaud32-linux-3.13.11.zip  <== includes hackaround with help from Steven Levine for the TRAP000E I was seeing

Traps.

3.14.79 - http://smedley.id.au/tmp/uniaud32-linux-3.14.79.zip <== doesn't trap, but fails to find an audio driver

Boot error:
Code: [Select]
ERROR: Query names failed.
open ret = -19
      OSS32_QueryNames() ret = -19
             Unable to attach 32-bit driver


Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 04, 2020, 01:34:55 am
Hi Martin,

Linux 3.12.74: http://smedley.id.au/tmp/uniaud32-linux-3.12.74.zip
FYI: I tried this version on real hardware (ThinkPad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420)) and it traps on boot.

Traps on my X250E as well - which I guess is a good thing :)
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 04, 2020, 02:21:47 am
 Paul,

  Attached is the screen shot of the trap using the debug 3.13 drivers you requested....

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 04, 2020, 11:46:21 am
Hi All,

Can anyone experience a TRAP000E with 3.10 -> 3.13 please test with http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200804.zip

This has the hda drivers from 3.10.108 replaced with the code from 3.9.11. This fixes things on my X250. Assuming it works for others too, next steps will be to start trying to identify/fix the code change  between these two versions that broke things for us :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 04, 2020, 12:56:56 pm
 Paul,

  When I download that link, I get a 0 byte file...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 04, 2020, 01:04:41 pm
Paul,

  When I download that link, I get a 0 byte file...

Regards,

Sorry - try again now
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 05, 2020, 12:38:16 am
 Hey Paul,

  This one behaves like the 3.9 version: it boots, but no sound, and unimix doesn't change anything (hardware manager shows it is loaded). I'll upload the alsahlp$ log running the debug drivers...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 05, 2020, 12:50:25 pm
Quick update: I've narrowed down the cause of the trap to changes in hda_codec.c between the 3.9 code and the 3.10 code. Nothing in the diff immediately jumps out as a cause, but I'll try and work through it tomorrow night, and post an update here.

Dave: fwiw I get no sound with 3.9 either in the x250 with Realtek hardware, sound is fine with virtualised sigmatel hardware.

I prefer successful boot with no sound to a trap right now :)
Title: Re: Updated Uniaud32 build
Post by: Heikki Kekki on August 06, 2020, 10:42:52 am
Hi David,

  It's working just fine here - system sounds, WPS MP3 objects and YouTube - all on ArcaOS 5.0.5 on bare metal 8086:A170. Thanks for continuing to progress with this! Any chance for multiple streams (so system sounds don't interrupt something else, they are just mixed in)?... or even surround...

Thanks for testing. In terms of multiple streams - I believe the SBLive driver supported this so it definitely should be possible to support.
Therefore, I use the SBLive driver, it works flawlessly. The downside of RAM is only 2 GB, exactly on this computer Memory Watcher shows 2180716 KB installed memory. If there is 16 GB more, the sound is distorted. Installed at 4 GB, the ramdisk uses almost half.

OS2LDR.CFG
Code: [Select]
RAMDISK = MIN=1836 HPFS Q:
HIDISK = RLOW=128 RHIGH=0
RAMDISKNAME = RAM disk

Browser cache, TMP, TEMP and swapper.dat are on ramdisk Q. Swapping (rarely happens) is pretty imperceptible.
This is Dell OptiPlex 960 and sound card is Creative Labs - SB Live! EMU10k1
Title: Re: Updated Uniaud32 build
Post by: roberto on August 06, 2020, 10:55:08 am
Can anyone experience a TRAP000E with 3.10 -> 3.13 please test with http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200804.zip

This has the hda drivers from 3.10.108 replaced with the code from 3.9.11. This fixes things on my X250. Assuming it works for others too, next steps will be to start trying to identify/fix the code change  between these two versions that broke things for us :)
It doesn't give me any traps, it works perfect even at 64K and 16M panorama colors. The trap I sent earlier was Linux3.13.11 debug.
Saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 06, 2020, 12:01:24 pm
Hi Roberto,

Can anyone experience a TRAP000E with 3.10 -> 3.13 please test with http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200804.zip

This has the hda drivers from 3.10.108 replaced with the code from 3.9.11. This fixes things on my X250. Assuming it works for others too, next steps will be to start trying to identify/fix the code change  between these two versions that broke things for us :)
It doesn't give me any traps, it works perfect even at 64K and 16M panorama colors. The trap I sent earlier was Linux3.13.11 debug.
Saludos

Great to hear - what sound driver is detected?

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: roberto on August 06, 2020, 01:45:45 pm
I attach the log , in debug.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2020, 01:20:28 am
OK, I found the change in hda_codec.c that was causing the TRAP000E in 3.10.108 on some chipsets - I don't understand why yet, but I've reverted part of the code to that from 3.9 and I now get no trap from 3.10.108 on my X250 with an ALC292. I'm also (currently) getting no sound, but that's a different issue.

http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200807.zip
&
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200807.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 07, 2020, 02:11:00 am
 Paul,

  I can boot with both of those release versions, but no sound and no trap, even when playing an MP3. They do show in the hardware manager. Let me know if you need a debug alsahlp$....

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2020, 02:22:48 am
Thanks Dave, that matches behaviour on my X250, so no need for a log at this stage. Let me investigate locally, and once I get sound working here on my X250, if you still have issues, then I'll ask for a log.

I'm also trying to clean up some of the patches to simplify things going forwards.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2020, 05:13:53 am
Does http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200807b.zip give sound for people? The sound is distorted on my X250, but I think that's the problem MikeK reported a few months back with earlier builds.

I'll try and test this evening with the other old Thinkpad I have here.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 07, 2020, 06:21:23 am
Hi Paul, linux-3.10.108-20200807b gives not bad sound here, it doesn't seem quite right but close, like a crappy or old cassette recording or a radio station just very slightly miss tuned.
I only tested with Z!. The volume control on Z! and the volume widget, when adjusted causes static while adjusting and then back to where it was, same with mute, no muting.
Using PMUniMix, the volume sliders all work but if click the on/off button, the sound just loses some bass/gets tinny.
I'd put up with this but given a choice, well, I'm downgrading to your last good driver, Linux-3.6.11.
Thanks for the work.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2020, 06:35:11 am
Hi Dave,

Hi Paul, linux-3.10.108-20200807b gives not bad sound here, it doesn't seem quite right but close, like a crappy or old cassette recording or a radio station just very slightly miss tuned.
I only tested with Z!. The volume control on Z! and the volume widget, when adjusted causes static while adjusting and then back to where it was, same with mute, no muting.
Using PMUniMix, the volume sliders all work but if click the on/off button, the sound just loses some bass/gets tinny.
I'd put up with this but given a choice, well, I'm downgrading to your last good driver, Linux-3.6.11.
Thanks for the work.

Thanks for the comments. I thought your earlier post said 3.8.x was OK for you?

Can you try with:
http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200807c.zip (hda_intel.c backlevelled to linux 3.8.13)
or
http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200807d.zip (hda_intel.c backlevelled to linux 3.6.11)

This is just on a hunch that changes to hda_intel.c are driving the sound issues.

Cheers,

Paul

Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 07, 2020, 06:43:42 am
Hi Paul, I'm going to have to take back my comments after downgrading and then downgrading again to the stock uniaud shipped with AOS 5.05. I think it is these speakers causing the crappy sound, though with my bad hearing it is hard to judge :) PMUniMix acts much the same with all the tested drivers and the volume at first worked on Z!, then after opening the volume control, stopped working along with the volume control not working.
I probably did get mixed up on the tested versions.
I'll try your latest. I am using the Intel HDA I believe, Ivy Lake I5
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 07, 2020, 07:07:26 am
Hi, uniaud32-linux-3.10.108-20200807c works much the same and actually sounds pretty good on headphones, but still not the best, need better speakers :)
About the volume, the Z! volume works fine until I open the volume widget, then it gets staticy when adjusting and stops actually changing the volume. Trick is to not use it.
All your builds I've tried vs the stock one in PMUniMix show PCM volume instead of front volume and also show the Headphone Playback Volume and it works, PCM adjusts the speakers, Headphone adjusts the headphone jack on the front of the computer. There's also front and rear mic volume controls which I haven't tested.
Have to see what others find with newer chipsets.
Once again, thanks.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2020, 07:10:10 am
Hi Dave,

Hi, uniaud32-linux-3.10.108-20200807c works much the same and actually sounds pretty good on headphones, but still not the best, need better speakers :)
About the volume, the Z! volume works fine until I open the volume widget, then it gets staticy when adjusting and stops actually changing the volume. Trick is to not use it.
All your builds I've tried vs the stock one in PMUniMix show PCM volume instead of front volume and also show the Headphone Playback Volume and it works, PCM adjusts the speakers, Headphone adjusts the headphone jack on the front of the computer. There's also front and rear mic volume controls which I haven't tested.
Have to see what others find with newer chipsets.
Once again, thanks.

Thanks for testing! So no significant difference between 20200807b.zip and 20200807c.zip ? (That's good if it's the case, as I'll continue with cleaning up the code).
Title: Re: Updated Uniaud32 build
Post by: Hermit on August 07, 2020, 02:51:58 pm
Hi, Paul.

Does http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200807b.zip give sound for people? The sound is distorted on my X250, but I think that's the problem MikeK reported a few months back with earlier builds.

Versions
uniaud32-linux-3.10.108-20200807b.zip
uniaud32-linux-3.10.108-20200807c.zip
uniaud32-linux-3.10.108-20200807d.zip
they work just as well as
uniaud32-linux-3.5.zip

  driver: HDA-Intel
  longname: HDA ATI SB at 0xfe6f8000 irq 19
  mixer: Realtek ALC892
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 07, 2020, 03:36:35 pm
Paul,

  All 3 of those 3.10 versions work with sound here - system sounds, MP3's, YouTube, and PMUniMix works with them as well. Great progress, thanks!

Regards,
Title: Re: Updated Uniaud32 build
Post by: roberto on August 07, 2020, 04:02:45 pm
All these versions work fine.

 LINUX-3-12-74-20200807-DEBUG   OK SONIDO, OK VOLUMEN, PANTALLA 16M
 LINUX-3-10-108-20200807B-DEBUG OK SONIDO, OK VOLUMEN, PANTALLA 16M
 LINUX-3-10-108-20200807C-DEBUG OK SONIDO, OK VOLUMEN, PANTALLA 16M
 LINUX-3-10-108-20200807D-DEBUG OK SONIDO, OK VOLUMEN, PANTALLA 16M

But when returning to the version  lllinux-3.10.108-20200804-debug

 4/08/20 19:06         38.868      0 a---  uniaud32.sym
 4/08/20 19:06      1.051.468      0 a---  uniaud32.sys


I had just started this video on youtube:
 https://www.youtube.com/watch?v=Tuyy-DMBAwM
 background music
 
 
open TrueSpectra, load any program and close it.
The system sounds are activated, I do not remember if when opening or closing.
The volume suddenly increased without me activating anything
 And the screen disappears but only from the running video, what normally looks like a tree and moon turns black.
 All the rest of the screen works correctly.
 The sound interferes with the screen, but if I try to repeat it I can't,
  although this is the second time it has happened to me.
Luck
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 08, 2020, 09:59:10 am
Code tidyups....
http://smedley.id.au/tmp/uniaud32-linux-3.10.108-20200808.zip
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200808.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 08, 2020, 01:26:38 pm
 Paul,

  3.12 seems to work fine, but I get no sound with 3.10. I'll upload debug Alsahlp$ files...

Regards,
Title: Re: Updated Uniaud32 build
Post by: roberto on August 08, 2020, 03:25:02 pm
Both work well for me with the basics, but the 3-12-74-20200808 is much higher. This is my new favorite. After torturing him a bit.
Saludos
Title: Re: Updated Uniaud32 build
Post by: Remy on August 08, 2020, 10:20:44 pm
I give a try to  3-12-74-20200808
- It found my card
- it found irq
but it only found one time the audio driver "Realtek ALC892"  resulting into a nice working audio (better than originally one) but this occurred only 1 time of 6 reboot !
Strange...

Parms used:    DEVICE=C:\MMOS2\UNIAUD32.SYS /A:1 /V /C /M /L
Title: Re: Updated Uniaud32 build
Post by: Joop on August 08, 2020, 11:36:39 pm
You have to screen your config.sys on the position of this line, it must be one of the last in the device block. Everything uniaud needs has to be started before the line with uniaud.   
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2020, 01:13:36 am
Hi David,,

  3.12 seems to work fine, but I get no sound with 3.10. I'll upload debug Alsahlp$ files...

Most likely this is due to be implementing more of the 'quirks' in patch_realtek.c in 3.12 than are present in 3.10; and one of those quirks happens to fix sound for your hardware.

I still have a bit more work to do to fully implement all quirks in 3.12, and this is pretty tedious work - given 3.12 is working, I'll likely not re-do this work for 3.10.

I'll try and finish 3.12 off today, then move on to working out why 3.14 can't load a driver.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2020, 03:36:42 am
I think all the quirks are now enabled in 3.12.74:
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200809.zip

This might help with joop's issue as there were a bunch of missing workarounds for Realtek chips up until this morning...

I also tidied up patch_sigmatel and now get sound in Virtualbox again :)
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 09, 2020, 05:56:33 am
Paul,

  This one works well too... thanks again!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2020, 06:33:16 am
Thanks Dave, now to try and fix 3.14!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2020, 10:14:17 am
I was sure I'd tried previously, but I tried again 3.14.79 this morning with ac97 hardware (using Virtualbox) and I detected hardware successfully. I then backlevelled hda_intel.c to an older version and got detected hardware with 3.14.79 and HDA hardware.

After some detective work, I've got HDA working with a slight hack whilst I find a better way to implement schedule_work() on OS/2.

Produces sound in Virtualbox with AC97 and HDA hardware for me.

http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200809.zip
Title: Re: Updated Uniaud32 build
Post by: Remy on August 09, 2020, 03:05:35 pm
I give a try to  3-12-74-20200808
- It found my card
- it found irq
but it only found one time the audio driver "Realtek ALC892"  resulting into a nice working audio (better than originally one) but this occurred only 1 time of 6 reboot !
Strange...

Parms used:    DEVICE=C:\MMOS2\UNIAUD32.SYS /A:1 /V /C /M /L

Uniaudio driver had default ArcaOS position into config.sys
I moved it just before the latest device statement which is for the video.
Problem remain the same "Realtek ALC892" not detected

I updated to the latest 3.12 and did a few more reboot to check and I have the result: "realtek ALC892" not detected
(From all boot/reboot I did, it was only one time detected with the /A:1 adapter, using the debug build)

 
Title: Re: Updated Uniaud32 build
Post by: Remy on August 09, 2020, 03:15:12 pm
now tried 3.14.79 and it seems to be worse

See boot messages
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 09, 2020, 04:08:25 pm
 Paul,

 The new 3.14 version is working well here...

Regards,
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 09, 2020, 04:26:18 pm
 Remy.

  What do you see when you type 'unimix -card' on a command prompt? Also, what happens if you change /A:1 to /A:0?

Regards,
Title: Re: Updated Uniaud32 build
Post by: Remy on August 09, 2020, 06:35:25 pm
Remy.

  What do you see when you type 'unimix -card' on a command prompt? Also, what happens if you change /A:1 to /A:0?

Regards,

Hi David,

Using ArcaOS driver, I have:
Code: [Select]
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.06
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xf7410000 irq 22
  mixer: Realtek ALC892
  componenets: HDA:10ec0892,1462d922,00100302

With the new 3.12.74, I have:
ommand line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xf7414000 irq 16
  mixer: Intel ID 2808
  componenets: HDA:80862808,80860101,00100000


Changing /A:1 to /A:0 gives:
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: HDMI
  driver: HDA-Intel
  name: HDA Intel HDMI
  longname: HDA Intel HDMI at 0xf7414000 irq 16
  mixer: Intel ID 2808
  componenets: HDA:80862808,80860101,00100000

Without the /A parameter, same result.

It looks like adapter number isn't taken ! 

Result of unimix -list
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Control list for adapter 0

Control ID: 1
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [IEC958 Playback Con Mask]
Index: 0
Element type: IEC958 (S/PDIF) setup
Count of values: 1
Value:
 AES0=0x0f,AES1=0xff,AES2=0x00,AES3=0x00

Control ID: 2
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [IEC958 Playback Pro Mask]
Index: 0
Element type: IEC958 (S/PDIF) setup
Count of values: 1
Value:
 AES0=0x0f,AES1=0x00,AES2=0x00,AES3=0x00

Control ID: 3
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [IEC958 Playback Default]
Index: 0
Element type: IEC958 (S/PDIF) setup
Count of values: 1
Value:
 AES0=0x04,AES1=0x00,AES2=0x00,AES3=0x00

Control ID: 4
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [IEC958 Playback Switch]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1

Control ID: 5
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [HDMI Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 6
Interface: PCM device
Device(client): 3, subdevice (substream) 0
Name: [Playback Channel Map]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 36, step: 0
  value 0: 0
  value 1: 0
Title: Re: Updated Uniaud32 build
Post by: Remy on August 09, 2020, 07:49:28 pm
Here is unimix -dev1 -list output
Code: [Select]
----
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 2 audio adapter(s)
Control list for adapter 1

Control ID: 1
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Front Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 64
  value 1: 64

Control ID: 2
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Front Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 3
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Surround Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 64
  value 1: 64

Control ID: 4
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Surround Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 5
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Center Playback Volume]
Index: 0
Element type: integer type
Count of values: 1
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 64

Control ID: 6
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [LFE Playback Volume]
Index: 0
Element type: integer type
Count of values: 1
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 64

Control ID: 7
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Center Playback Switch]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1

Control ID: 8
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [LFE Playback Switch]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1

Control ID: 9
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Side Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 64
  value 1: 64

Control ID: 10
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Side Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 11
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Headphone Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 64
  value 1: 64

Control ID: 12
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Headphone Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 13
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Front Mic Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 31, step: 0
  value 0: 0
  value 1: 0

Control ID: 14
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Front Mic Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 15
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Rear Mic Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 31, step: 0
  value 0: 0
  value 1: 0

Control ID: 16
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Rear Mic Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 17
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Line Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 31, step: 0
  value 0: 0
  value 1: 0

Control ID: 18
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Line Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 19
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Auto-Mute Mode]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 2, item: 0, name: [Disabled]
  value 0: 1

Control ID: 20
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Input Source]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 3, item: 0, name: [Front Mic]
  value 0: 0

Control ID: 21
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Input Source]
Index: 1
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 3, item: 0, name: [Front Mic]
  value 0: 0

Control ID: 22
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Capture Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 46, step: 0
  value 0: 0
  value 1: 0

Control ID: 23
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Capture Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 24
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Capture Volume]
Index: 1
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 46, step: 0
  value 0: 0
  value 1: 0

Control ID: 25
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Capture Switch]
Index: 1
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 26
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Front Mic Boost Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 3, step: 0
  value 0: 0
  value 1: 0

Control ID: 27
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Rear Mic Boost Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 3, step: 0
  value 0: 0
  value 1: 0

Control ID: 28
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Line Boost Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 3, step: 0
  value 0: 0
  value 1: 0

Control ID: 29
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Master Playback Volume]
Index: 0
Element type: integer type
Count of values: 1
Value:
 Bounds: min: 0, max: 64, step: 0
  value 0: 0

Control ID: 30
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Master Playback Switch]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 31
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Front Mic Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 32
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Rear Mic Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 33
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Line Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 34
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Line Out Front Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1

Control ID: 35
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Line Out Surround Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 36
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Line Out CLFE Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 37
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Line Out Side Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 38
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Front Headphone Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 39
Interface: PCM device
Device(client): 0, subdevice (substream) 0
Name: [Playback Channel Map]
Index: 0
Element type: integer type
Count of values: 8
Value:
 Bounds: min: 0, max: 36, step: 0
  value 0: 0
  value 1: 0
  value 2: 0
  value 3: 0
  value 4: 0
  value 5: 0
  value 6: 0
  value 7: 0

Control ID: 40
Interface: PCM device
Device(client): 0, subdevice (substream) 0
Name: [Capture Channel Map]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 36, step: 0
  value 0: 0
  value 1: 0

Control ID: 41
Interface: PCM device
Device(client): 2, subdevice (substream) 0
Name: [Capture Channel Map]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 36, step: 0
  value 0: 0
  value 1: 0

For me it looks like the /A:1 isn't recognized. May it be possible ?

Regards
Rémy
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 09, 2020, 09:03:26 pm
 Hi Remy,

 "For me it looks like the /A:1 isn't recognized. May it be possible ?"

 Yes, it looks like the mixer for the HDMI device (/A:0) is mistakenly being detected for the HDA device (/A:1). Hopefully Paul can figure it out...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2020, 11:29:59 pm
Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 10, 2020, 12:33:01 am
I am with ArcaOS 5.0.5 and Uniaud package 20200506. As usual, after installation (and reboot) "Master Front Playback Volume" and "Front Playback Volume" is switched ON but level setted to zero. Volume.exe has no effect. All sound level is controlled only by Unimix settings. I saw that unimix -save <file> not works anymore (always file missing message output). Motherboard interface is High Definition Audio NVidia.

Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 10, 2020, 12:42:29 am
Details of the interface, if helps

[E:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.06
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: NVidia
  driver: HDA-Intel
  name: HDA NVidia
  longname: HDA NVidia at 0xebef8000 irq 23
  mixer: VIA VT1708S
  componenets: HDA:11064397,18490397,00100000
Title: Re: Updated Uniaud32 build
Post by: David Graser on August 10, 2020, 02:59:58 am
Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.

The sound is exceptional using PM123 and VLC. 
Title: Re: Updated Uniaud32 build
Post by: Remy on August 10, 2020, 04:52:41 am
Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.

Same as previous 3.14.79 (it doesn't work)
No mixer found and /A:1 not taken, adapter 0 is always taken despite the /A:1 flag

Could this help you ?
unimix -card:
[C:\mmos2]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Card info for adapter 0:
  num: 0
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xf7414000 irq 16
  mixer:
  componenets: HDA:80862808,80860101,00100000

Then, find why /A:1 is ignored
[ note: adapter 1 is on irq22  while irq16 is for adapter 0 ]
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 10, 2020, 04:59:46 am
Hi Remy,

Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.

Same as previous 3.14.79 (it doesn't work)
No mixer found and /A:1 not taken, adapter 0 is always taken despite the /A:1 flag

Interesting - I tested with my X250 earlier and might be seeing something similar.  I'll investigate as time permits.

Just confirming - 3.12.x work OK?

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 10, 2020, 05:00:17 am
Hi David,

Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.

The sound is exceptional using PM123 and VLC.

Thanks for confirming!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Remy on August 10, 2020, 05:05:01 am
Hi Remy,

Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.

Same as previous 3.14.79 (it doesn't work)
No mixer found and /A:1 not taken, adapter 0 is always taken despite the /A:1 flag

Interesting - I tested with my X250 earlier and might be seeing something similar.  I'll investigate as time permits.

Just confirming - 3.12.x work OK?

Cheers,

Paul

With 3.12, a mixer is found but /A:1 doesn't work too !  (only adapter 0 is taken)
See my post : https://www.os2world.com/forum/index.php/topic,2436.msg27702.html#msg27702

I need to use adapter 1 to get sound  (how to when the /A:1 has no effect)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 10, 2020, 06:00:58 am
Hi Remy,

Hi Remy,

Does http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip help at all? In reviewing the diff from the original sources - something screwy had happened; so I took a fresh copy of the 3.14.79 sources and re-applied the diffs from 3.12.x. The diff looks much cleaner now. Driver still loads/works in Virtualbox, no time before work today to test on real hardware.

Same as previous 3.14.79 (it doesn't work)
No mixer found and /A:1 not taken, adapter 0 is always taken despite the /A:1 flag

Interesting - I tested with my X250 earlier and might be seeing something similar.  I'll investigate as time permits.

Just confirming - 3.12.x work OK?

Cheers,

Paul

With 3.12, a mixer is found but /A:1 doesn't work too !  (only adapter 0 is taken)
See my post : https://www.os2world.com/forum/index.php/topic,2436.msg27702.html#msg27702

I need to use adapter 1 to get sound  (how to when the /A:1 has no effect)

Are you able to confirm the latest one of my builds that works for you? That should help me narrow down the code change that broke things.

On my X250 - 3.10.x seems to identify the mixer correctly, 3.12.74 gives lines like the following in the log, can you confirm you see similar?
hda-intel : Clearing TCSEL

hda-intel : codec_mask = 0x1

hda-intel : azx_get_response timeout, polling the codec once: last cmd=0x000f0000

hda-intel : azx_get_response timeout, polling the codec once: last cmd=0x000f0000

<4>hda-intel : azx_get_response timeout, switching to polling mode: last cmd=0x000f0000

<4>hda-intel : Codec #0 probe error; disabling it...

hda-intel : codec_mask = 0x1

<3>hda_intel: azx_get_response timeout, switching to single_cmd mode: last cmd=0x000f0000

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Remy on August 10, 2020, 10:08:50 am
Quote
Are you able to confirm the latest one of my builds that works for you? That should help me narrow down the code change that broke things.

On my X250 - 3.10.x seems to identify the mixer correctly, 3.12.74 gives lines like the following in the log, can you confirm you see similar?
hda-intel : Clearing TCSEL

hda-intel : codec_mask = 0x1

hda-intel : azx_get_response timeout, polling the codec once: last cmd=0x000f0000

hda-intel : azx_get_response timeout, polling the codec once: last cmd=0x000f0000

<4>hda-intel : azx_get_response timeout, switching to polling mode: last cmd=0x000f0000

<4>hda-intel : Codec #0 probe error; disabling it...

hda-intel : codec_mask = 0x1

<3>hda_intel: azx_get_response timeout, switching to single_cmd mode: last cmd=0x000f0000

Cheers,

Paul

Hi Paul,

- Are you able to confirm the latest one of my builds that works for you?
  uniaud32-linux-3.10.108-20200808.zip  works

- 3.12.74 gives lines like the following in the log, can you confirm you see similar?
 
  Yes, extract follows :

pci_register_driver: matched 29 8086:160c/40300 with 8086:ffffffff/40300 ffffff (hda_intel)
request_irq: irq 16
hda-intel : chipset global capabilities = 0x3001

hda-intel : Clearing TCSEL

hda-intel : codec_mask = 0x1

hda-intel : azx_get_response timeout, polling the codec once: last cmd=0x000f0000

hda-intel : azx_get_response timeout, polling the codec once: last cmd=0x000f0000

<4>hda-intel : azx_get_response timeout, switching to polling mode: last cmd=0x000f0000

<4>hda-intel : Codec #0 probe error; disabling it...

hda-intel : codec_mask = 0x1

<3>hda_intel: azx_get_response timeout, switching to single_cmd mode: last cmd=0x000f0000

<3>hda-codec: No codec parser is available


Regards
Rémy


Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 10, 2020, 10:39:22 am
Thanks for confirming... Are you also able to confirm that http://smedley.id.au/tmp/uniaud32-linux-3.11.10-20200810.zip also works? It seems to recognise the mixer on my X250 - sound quality sucks on this hardware, but that's a separate issue...
Title: Re: Updated Uniaud32 build
Post by: Remy on August 10, 2020, 10:52:34 am
Thanks for confirming... Are you also able to confirm that http://smedley.id.au/tmp/uniaud32-linux-3.11.10-20200810.zip also works? It seems to recognise the mixer on my X250 - sound quality sucks on this hardware, but that's a separate issue...

Hi Paul,

Yes, this build works too.

Regards
Rémy
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 10, 2020, 11:59:19 am
OK, please test with:
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200810.zip
http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip

Fixes my X250 and sound works nicely for the 1st time.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 10, 2020, 01:56:08 pm
 Paul,

These 810 versions of 3.12 and 3.14 both work well here...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Remy on August 10, 2020, 03:01:19 pm
OK, please test with:
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200810.zip
http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip

Fixes my X250 and sound works nicely for the 1st time.

It works for me too (great job Paul)
Sound is very nice
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 10, 2020, 05:18:23 pm

http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip

This version worked fine on the Thinkpad L420. No trap, sound working, sound volume working.
For the moment this is a winner for me.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 10, 2020, 11:58:53 pm
OK, please test with:
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200810.zip
http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip

Fixes my X250 and sound works nicely for the 1st time.

It works for me too (great job Paul)
Sound is very nice

Great - now to work out why the original code fails!
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 11, 2020, 01:23:52 am
OK, please test with:
http://smedley.id.au/tmp/uniaud32-linux-3.12.74-20200810.zip
http://smedley.id.au/tmp/uniaud32-linux-3.14.79-20200810.zip
My tests with:
CardInfo.id: NVidia
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA NVidia
CardInfo.LongName: HDA NVidia at 0xebef8000 irq 23
CardInfo.MixerName: VIA VT1708S
CardInfo.Components: HDA:11064397,18490397,00100000

UNIAUD 20200506 TEST
.No sound output every reboot
 .Master Front Playback Volume is reset to zero (both channels)
 .Front Playback Volume is reset to zero (both channels)
.Volume.exe has no effects. All have to be controlled by mixer (unimix.exe)


UNIAUD32-LINUX-3.12.74-20200810 TEST
.Sound output after boot. Very good!
.Volume.exe all operational. Very good!

 
UNIAUD32-LINUX-3.14.79-20200810 TEST
.Sound output after boot. Very good!
.Volume.exe all operational. Very good!

Good work Paul!

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 11, 2020, 06:02:09 am
Thanks to everyone for their testing and feedback, it's gratifying to hear these are improving things for some users, and so far at least, don't appear to have any negative effects.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 12, 2020, 11:18:29 am
Hey All,

Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip

Works in Virtualbox (Sigmatel HDA), and on a Thinkpad X250 (Realtek HDA) - I get a Trap 000E on a T60 with Analog Devices HDA that I'll investigate in due course.

This one was the biggest challenge since the bump from 3.5 -> 3.8 - lots of changes so I'm bound to have broken something :) so in reality, I'm pleasantly surprised that 2/3 tests work OK.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 12, 2020, 12:11:50 pm
 Paul,

 Get a Trap000e here as well... do you need a screen shot?

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 12, 2020, 12:55:49 pm
Get a Trap000e here as well... do you need a screen shot?

Not at this stage, let me fix my T60 first, if you still get traps once that is resolved, then I'll ask for a screenshot.

Yours is Realtek hardware?

In the morning I'll try my old favourite of backlevelling hda_intel.c, and assuming that fixes things, work through the changes.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 12, 2020, 03:46:22 pm
Hi
Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip

I just tried it on a Thinkpad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420). No traps. Produce sounds. Volume control works.
Seems to be working fine for me. I will report if I find something.

Regards
Title: Re: Updated Uniaud32 build
Post by: roberto on August 12, 2020, 05:39:07 pm
Hello
Another trap ... over here with HDA-Intel in HP
Saludos
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 12, 2020, 07:06:59 pm
The latest wprks out a trap on my Dell Inspiron 15R 5537. But for som reason I am unable to load browser url pages.
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 13, 2020, 02:06:26 am
Hey All,

Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip


More one sound interface tested:

Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: ICH5
  driver: ICH4
  name: Intel ICH5
  longname: Intel ICH5 with CMI9739 at irq 17
  mixer: C-Media Electronics CMI9739
  componenets: AC97a:434d4961

Tests were done with Package 20200506, uniaud32-linux-3.12.74, uniaud32-linux-3.14.79 and the last uniaud32-linux-3.16.85.
In all tests sound hardware is not identified by os/2 (ArcaOS 5.0.5), although uniaud32.sys is normally loading at boot.
Seems that are not any sound hardware. Applications like Klondike Solitaire, Mahjongg Solitaire, with sound enabled by default, show messages of "No device driver found" .

To know if was not an ALSA problem, I boot a live cd of linux Ubuntu 12.10 with ALSA version 1.0.24. All sound works as expected, without problems.

unimix -list resulted from uniaud32-linux-3.16.85 and pci.exe are attached.
Title: Re: Updated Uniaud32 build
Post by: Remy on August 13, 2020, 12:18:48 pm
Hey All,

Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip


More one sound interface tested:

Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: ICH5
  driver: ICH4
  name: Intel ICH5
  longname: Intel ICH5 with CMI9739 at irq 17
  mixer: C-Media Electronics CMI9739
  componenets: AC97a:434d4961

Tests were done with Package 20200506, uniaud32-linux-3.12.74, uniaud32-linux-3.14.79 and the last uniaud32-linux-3.16.85.
In all tests sound hardware is not identified by os/2 (ArcaOS 5.0.5), although uniaud32.sys is normally loading at boot.
Seems that are not any sound hardware. Applications like Klondike Solitaire, Mahjongg Solitaire, with sound enabled by default, show messages of "No device driver found" .

To know if was not an ALSA problem, I boot a live cd of linux Ubuntu 12.10 with ALSA version 1.0.24. All sound works as expected, without problems.

unimix -list resulted from uniaud32-linux-3.16.85 and pci.exe are attached.

What are your driver parameters in the config.sys ?
Did you try with /A:1 ?
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 13, 2020, 07:26:31 pm


What are your driver parameters in the config.sys ?
Did you try with /A:1 ?
[/quote]

With uniaud32.sys /A:1 in config.sys I get the error messages below:

Universal Audio Driver for OS/2 v2.02.05-SVN
Based on Linux 3.16.85
Copyright 2005-2020 Netlabs http://www.netlabs.org
Copyright 2000-2020 The ALSA Project

ERROR: Initialization failed
OSS32_Initialize: FAILEDpci_register-driver: query_device found fd 8086:24d5 class=40100 checking bt87x
OS/2 Universal Audio MMPM/2 Driver v1.9.7
Copyright 2000-2002 InnoTek Systemberatung GmbH
Copyright 2002 IBM Corporation
Mantained by Netlabs http://www.netlabs.org

Unable to attach 32-bit driver
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2020, 12:03:39 am
Hi Renee,

Hey All,

Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip


More one sound interface tested:

Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: ICH5
  driver: ICH4
  name: Intel ICH5
  longname: Intel ICH5 with CMI9739 at irq 17
  mixer: C-Media Electronics CMI9739
  componenets: AC97a:434d4961

Tests were done with Package 20200506, uniaud32-linux-3.12.74, uniaud32-linux-3.14.79 and the last uniaud32-linux-3.16.85.
In all tests sound hardware is not identified by os/2 (ArcaOS 5.0.5), although uniaud32.sys is normally loading at boot.
Seems that are not any sound hardware. Applications like Klondike Solitaire, Mahjongg Solitaire, with sound enabled by default, show messages of "No device driver found" .

To know if was not an ALSA problem, I boot a live cd of linux Ubuntu 12.10 with ALSA version 1.0.24. All sound works as expected, without problems.

unimix -list resulted from uniaud32-linux-3.16.85 and pci.exe are attached.

An alsa log would be interesting.
copy alsahlp$ alsa-cmi9739.log
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2020, 11:36:15 am
For anyone who was getting a TRAP000E with 3.16.85, can you test with http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200814.zip ?

Seems a function I'd stubbed was causing the tarp (thank's to Steven for his help in setting up a dump partition and helping with analysis).
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 14, 2020, 12:08:52 pm
 Paul,

  This 3.16 version is working here... Realtek-ALC887-VD.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2020, 12:12:51 pm
Hi Dave,


  This 3.16 version is working here... Realtek-ALC887-VD.


Thanks for confirming. Now to try and fix it more elegantly, then work out why 3.18.140 is trapping on all hda hardware (but seems ok on the one ac97 system I can test on)

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Guillaume Gay on August 14, 2020, 04:27:52 pm
Bonjour to all,

"uniaud32-linux-3.16.85-20200814" works in a sense: it plays the startup sound in loop!
I reverted back to 2.02.06, since I needed to use my system.

{ uniaudio (until 2.02.06) seems to run out of resource after many days of system uptime. No more notification sounds. Short WAV files (less than 1sec or so) can't be used as system sounds, for example. I wondered if 3.16.85 worked better... }

See you !
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 14, 2020, 06:35:21 pm
Hi Renee,

Hey All,

Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip


More one sound interface tested:

Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: ICH5
  driver: ICH4
  name: Intel ICH5
  longname: Intel ICH5 with CMI9739 at irq 17
  mixer: C-Media Electronics CMI9739
  componenets: AC97a:434d4961

Tests were done with Package 20200506, uniaud32-linux-3.12.74, uniaud32-linux-3.14.79 and the last uniaud32-linux-3.16.85.
In all tests sound hardware is not identified by os/2 (ArcaOS 5.0.5), although uniaud32.sys is normally loading at boot.
Seems that are not any sound hardware. Applications like Klondike Solitaire, Mahjongg Solitaire, with sound enabled by default, show messages of "No device driver found" .

To know if was not an ALSA problem, I boot a live cd of linux Ubuntu 12.10 with ALSA version 1.0.24. All sound works as expected, without problems.

unimix -list resulted from uniaud32-linux-3.16.85 and pci.exe are attached.

An alsa log would be interesting.
copy alsahlp$ alsa-cmi9739.log

Hi, alsa-cmi9739log.txt is attached.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2020, 11:54:21 pm
Hi Guillaume,
"uniaud32-linux-3.16.85-20200814" works in a sense: it plays the startup sound in loop!
I reverted back to 2.02.06, since I needed to use my system.

{ uniaudio (until 2.02.06) seems to run out of resource after many days of system uptime. No more notification sounds. Short WAV files (less than 1sec or so) can't be used as system sounds, for example. I wondered if 3.16.85 worked better... }

See you !

Can you please upload the output of 'copy alsahlp$ alsa.log' - from any uniaud32 version - I just want to see what chipset you're running.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2020, 11:56:24 pm
Hi Renee,

Hey All,

Updated to 3.16.85 - http://smedley.id.au/tmp/uniaud32-linux-3.16.85-20200812.zip


More one sound interface tested:

Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: ICH5
  driver: ICH4
  name: Intel ICH5
  longname: Intel ICH5 with CMI9739 at irq 17
  mixer: C-Media Electronics CMI9739
  componenets: AC97a:434d4961

Tests were done with Package 20200506, uniaud32-linux-3.12.74, uniaud32-linux-3.14.79 and the last uniaud32-linux-3.16.85.
In all tests sound hardware is not identified by os/2 (ArcaOS 5.0.5), although uniaud32.sys is normally loading at boot.
Seems that are not any sound hardware. Applications like Klondike Solitaire, Mahjongg Solitaire, with sound enabled by default, show messages of "No device driver found" .

To know if was not an ALSA problem, I boot a live cd of linux Ubuntu 12.10 with ALSA version 1.0.24. All sound works as expected, without problems.

unimix -list resulted from uniaud32-linux-3.16.85 and pci.exe are attached.

An alsa log would be interesting.
copy alsahlp$ alsa-cmi9739.log

Hi, alsa-cmi9739log.txt is attached.

This is likely the problem:
U16: Uniaud16 v1.9.7
OSS32_QueryDevCaps: SNDRV_PCM_IOCTL_HW_REFINE error -22

OSS32_QueryDevCaps failed

Uniaud16 can't query the device capabilities for whatever reason.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on August 15, 2020, 09:42:19 am
uniaud32-linux-3.16.85-20200812.zip

Works well with a Creative SB Live! 5.1
PMUniMax has also filled out with many sliders including ones for Surround+center playback, Front, Line2, LiveDrive, Coaxial,  headphone, but I have not tested any of them, yet.

Listening to Gustav Holst - The Planets
Very nice, even for a deaf bloke.

Thanks for your continuing work on various OS/2 drivers and software.

(added: undated to uniaud32-linux-3.16.85-20200814.zip and all still working well)
Title: Re: Updated Uniaud32 build
Post by: Guillaume Gay on August 15, 2020, 11:42:05 am
Hi Paul from here as well!

[...]
"uniaud32-linux-3.16.85-20200814" works in a sense: it plays the startup sound in loop!
I reverted back to 2.02.06, since I needed to use my system.
[...]

Can you please upload the output of 'copy alsahlp$ alsa.log' - from any uniaud32 version - I just want to see what chipset you're running.

Cheers,

Paul

Please find attached "alsa-GG_20200814.txt" made with UNiAud v2.02.06 in use on this system.

Have a nice day!
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 15, 2020, 08:29:01 pm
The latest drop is working on my Dell laptop now. I did nothing.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 16, 2020, 03:20:05 am
3.8.140 is now working on my virtualised ArcaOS system, I haven't had time yet to try on bare metal.

http://smedley.id.au/tmp/uniaud32-linux-3.18.140-20200816.zip

Just some background information - the way they apply "quirks" or workarounds for different hardware started changing from 3.16.x onwards. Currently, I haven't implemented the newer quirks as it's a bit of work due to Watcom's lack of C99 support. As a result, there is some hardware that may work better in 3.14.x than in newer builds.

In the fullness of time, I'll fully implement the quirks in 3.18.x

As some more background, ALSA 1.0.24 (which is the basis of the Uniaud driver shipped with ArcaOS 5.0.5) was released on 31st Jan 2011. The Linux 3.18 kernel was released in December 2014, and 3.18.140 in May 2019 - so we've advanced a few years :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 16, 2020, 06:38:47 am
OK... no traps but no sound on my T60 (Analog devices HDA) - will need to investigate...  no traps and perfect sound on my X250 (Realtek HDA)
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 16, 2020, 08:42:20 am
OK... no traps but no sound on my T60 (Analog devices HDA) - will need to investigate...  no traps and perfect sound on my X250 (Realtek HDA)

Hi Paul, works on my X61 Thinkpad, sound is great. Still have the problem that the 2 buttons to control volume do not work, the button to mute volume works. Thanks again for your efforts

CardInfo.card: 0
CardInfo.id: Intel
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA Intel
CardInfo.LongName: HDA Intel at 0xf8220000 irq 17
CardInfo.MixerName: Analog Devices AD1984
CardInfo.Components: HDA:11d41984,17aa20d6,00100400
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 16, 2020, 11:10:46 am
Well, I'm pleased to say that I just broke through a psychological barrier and built the code based on linux kernel 4.0.9

Build is at http://smedley.id.au/tmp/uniaud32-linux-4.0.9-20200816.zip

The linux 4.0 kernel was released in April 2015, and 4.0.9 was released in July 2015.

Next step will be 4.1.52 - from a brief look - this one looks a bit more complex - there's been some code restructuring :/

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 16, 2020, 03:51:45 pm
 Hi Paul,

  I was hoping to try out my new trap dump partition, but no such luck... both the 3.18 and 4.0 versions posted work fine here. Congratulations on moving into the 4.x versions... and thanks again!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 16, 2020, 04:39:40 pm
Working on my Dell laptop! And well I might add.
 
Title: Re: Updated Uniaud32 build
Post by: Pete on August 17, 2020, 02:12:04 am
Hi Paul

uniaud32-linux-3.16.85-20200814.zip (Release) works fine.

uniaud32-linux-4.0.9-20200816.zip (Release) causes system hang at boot when the startup sound should play.

Card info for adapter 0:
  num: 0
  id: SB
  driver: HDA-Intel
  name: HDA ATI SB
  longname: HDA ATI SB at 0xfeb00000 irq 16
  mixer: Realtek ALC892
  componenets: HDA:10ec0892,104384fb,00100302


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: David Graser on August 17, 2020, 02:19:24 am
Works great with my Lenovo T430s.
Title: Re: Updated Uniaud32 build
Post by: Remy on August 17, 2020, 10:28:50 pm
uniaud32-linux-4.0.9-20200816.zip works for me too.
Title: Re: Updated Uniaud32 build
Post by: Pete on August 18, 2020, 04:52:44 pm
Hi Paul

Have now tried uniaud32-linux-3.18.140-20200816.zip (Release) - causes system hang at boot when the startup sound should play.


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on August 20, 2020, 06:55:06 am
> Creative SB Live! 5.1
> PMUniMax has also filled out with many sliders including ones for Surround+center playback, Front, Line2, LiveDrive, Coaxial,  headphone, but I have not tested any of them, yet.

Have now tested, controls and outputs for

  Wave Surround Playback Volume
  Center Playback Volume
  LFE Playback volume
  Front panel
  LiveDrive Playback (Line 1 and 2)
  Coaxial Playback
  Headphone

all work.
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 20, 2020, 10:20:33 pm
Hi All,

Prompted by Lewis reporting a problem with no sound on a newish laptop, I built uniaud32.sys last night for the first time since December 2015.

After managing to patch the realtek driver to add some additional chipsets which got Lewis sound, I got interested again in updating Uniaud.

As some may recall, way back when, I was updating the uniaud32 core with the latest ALSA code.... I got as far as ALSA 1.0.24 (released January 2011), but could never get ALSA 1.0.25 (released early 2012) to work. After that, the code was incorporated into the linux kernel, and it all got too hard. Plus, at the time, we still had the hangs with hda audio and I started running linux as my main OS.

Anyhow, after helping Lewis last night, I added the OS/2 patches to the linux kernel 3.0 code and got it building. This represents an alsa-kernel approximately 6 months newer that what's in the last uniaud32 release.

I've tested this in virtualbox with the Intel HDA driver and it still makes sound. That's about the level of testing I've done. I can't be bothered walking upstairs to grab one of the Thinkpads and test on other hardware.

Anyway...... link is http://smedley.id.au/tmp/uniaud32-linux-3.0.zip

Feedback appreciated. If there's interest, I'll continue to try and update the code to hopefully fix upstream bugs and support newer hardware.

Cheers,

Paul

More tests...

Lenovo Thinkpad T61

Card info for adapter 0:
  num: 0
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xfe020000 irq 11
  mixer: Analog Devices AD1984
  componenets: HDA:11d41984,17aa20d7,00100400

Tests with Uniaud20200506 package:

-Master Playback Volume set to zero and off at boot
-PCM Playback Volume set to zero at boot
-IEC958 Playback Volume set to zero and off at boot
-Volume.exe has no effects
-Sound at phone output ok after mixer adjust
-No sound at notebook speakers

Tests with uniaud32-linux-3.12.74-20200810

-now with a new Headphone Playback Volume control
-now with a new Speaker Playback Volume control
-Volume.exe working as expected
-Sound at phone output ok
-Sound at notebook speakers ok

Tests with uniaud32-linux-3.14.79-20200810

-with a new Headphone Playback Volume control
-with a new Speaker Playback Volume control
-Volume.exe  working as expected
-Sound at phone output ok
-Sound at notebook speakers ok

Tests with uniaud32-linux-3.16.85-20200812

-a TRAP 000e is generated

Tests with uniaud32-linux-3.18.140-20200816

-with a new Headphone Playback Volume control
-with a new Speaker Playback Volume control
-Volume.exe  working as expected
-Sound at phone output ok
-Sound at notebook speakers ok

Tests with uniaud32-linux-4.0.9-20200816

-now with a new Headphone Playback Volume control
-now with a new Speaker Playback Volume control
-Volume.exe working as expected
-Sound at phone output ok
-Sound at notebook speakers ok

Tests with microphone input and winos2/dos generic audio driver (DTA) not already done.

Attached unimix-list and alsa-AD1984log.txt from uniaud32-linux-4.0.9-20200816 and pciexe.txt
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on August 22, 2020, 02:36:09 pm
Hello Paul, everyone...

...uniaud32-linux-4.0.9-20200816.zip (Release) causes system hang at boot when the startup sound should play.

Card info for adapter 0:
  num: 0
  id: SB
  driver: HDA-Intel
  name: HDA ATI SB
  longname: HDA ATI SB at 0xfeb00000 irq 16
  mixer: Realtek ALC892
  componenets: HDA:10ec0892,104384fb,00100302
...

I tried the uniaud32-linux-3.18.140-20200816.zip here, seeing the same result Pete reported, that being a hard lock. In my case the only way to recover is through a power cycle, no other attempts work.

Have not tried any of the other releases, will hopefully get a little time today to do some catch-up. Attached the UniMix log from a working AOS driver config.

Code: [Select]
Number of CPU cores = 5
CPU name Not available
PMUniMix V0.82
Uniaud32.sys version = 202.6
Number of cards detected = 1
Using card number 0
number of controls = 35

CardInfo.card: 0
CardInfo.id: SB
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA ATI SB
CardInfo.LongName: HDA ATI SB at 0xfe8f4000 irq 16
CardInfo.MixerName: Realtek ALC889
CardInfo.Components: HDA:10ec0889,14627576,00100004
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on August 22, 2020, 03:42:38 pm
Alright, tested the remaining two releases:

1) uniaud32-linux-4.0.9-20200816.zip
did NOT work, same HARD stop as with uniaud32-linux-3.18.140-20200816.zip

2) uniaud32-linux-3.16.85-20200812.zip
did work, appears to be fine, this release (as compared to the AOS one) added an extra control to my PMUnixMix, that being the Surround Sound. I have not gone through the unimix CLI yet to see the full list of controls.

I've attached the LOG out of the DEBUG version of this driver, my thinking being Paul that maybe that gives you some additional data capture to see why the newer drivers are failing with this hardware?

Thanks!
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 22, 2020, 11:37:00 pm
Hi Dariusz,

Alright, tested the remaining two releases:

1) uniaud32-linux-4.0.9-20200816.zip
did NOT work, same HARD stop as with uniaud32-linux-3.18.140-20200816.zip

2) uniaud32-linux-3.16.85-20200812.zip
did work, appears to be fine, this release (as compared to the AOS one) added an extra control to my PMUnixMix, that being the Surround Sound. I have not gone through the unimix CLI yet to see the full list of controls.

I've attached the LOG out of the DEBUG version of this driver, my thinking being Paul that maybe that gives you some additional data capture to see why the newer drivers are failing with this hardware?

Thanks for the logs. As I mentioned, there are some hardware quirks missing in the newer builds - I'm 'hoping' there's one for our hardware and I can try enabling it and it fixes things.

I'll try and look at this this afternoon.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 23, 2020, 12:54:45 am
Hi All,

Prompted by Lewis reporting a problem with no sound on a newish laptop, I built uniaud32.sys last night for the first time since December 2015.

After managing to patch the realtek driver to add some additional chipsets which got Lewis sound, I got interested again in updating Uniaud.

As some may recall, way back when, I was updating the uniaud32 core with the latest ALSA code.... I got as far as ALSA 1.0.24 (released January 2011), but could never get ALSA 1.0.25 (released early 2012) to work. After that, the code was incorporated into the linux kernel, and it all got too hard. Plus, at the time, we still had the hangs with hda audio and I started running linux as my main OS.

Anyhow, after helping Lewis last night, I added the OS/2 patches to the linux kernel 3.0 code and got it building. This represents an alsa-kernel approximately 6 months newer that what's in the last uniaud32 release.

I've tested this in virtualbox with the Intel HDA driver and it still makes sound. That's about the level of testing I've done. I can't be bothered walking upstairs to grab one of the Thinkpads and test on other hardware.

Anyway...... link is http://smedley.id.au/tmp/uniaud32-linux-3.0.zip

Feedback appreciated. If there's interest, I'll continue to try and update the code to hopefully fix upstream bugs and support newer hardware.

Cheers,

Paul

More tests...

Aureal Vortex au8820
Chip: Sigmatel STAC9701, 3, 4, 5


Now with Turtle Beach Montego based in Aureal Vortex 8820 ( first Montego soundboard). I have this board working in OS/2 Warp 4 with the driver Aureal_Vortex_AU8820_1-0.zip (more two necessary patches Aureal_Vortex_Patch.zip and Aureal_Vortex_AU8820x_Patch.zip, but working as expected)

I woud like to use this board with ArcaOS but not works with this driver. Really, this is one of OS/2 sound drivers not included in the list of available drivers in installation program of ArcaOS or eCS.

So, I decided to try Uniaud, as in my tests with linux live CD (Ubuntu 12.04-32bit-alsa 1.0.24 based) soundboard works ok.

Unfortunatelly, not good results...

At boot, similar for all versions of uniaud32.sys tried:

Universal Audio Driver for OS/2 v2.02.05-SVN
Based on Linux 3.10.140
Copyright 2005-2020 Netlabs http://;www.netlabs.org
Copyright 2000-2020 The ALSA Project

ERROR: Initialization failed
OSS32_Initialize: FAILEDpci_register_driver: query_device found 257 12eb:1 class=40100 checking bt87x
OS/2 Universal Audio MMPM/2 Driver v1.9.7
Copyright 2000-2002 Innotek Systemberatung GmbH
Copyright 2002 IBM Corporation
Maintained by Netlabs http://;www.netlabs.org

Unable to attach 32-bit driver
SYS1201: The device driver C:\MMOS2\UNIAUD16.SYS specified in the DEVICE statement on line 182 of the CONFIG.SYS file was not installed



Attached is alsa_aureal_log.txt
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 23, 2020, 01:12:38 am
More tests...

Aureal Vortex au8820
Chip: Sigmatel STAC9701, 3, 4, 5


Now with Turtle Beach Montego based in Aureal Vortex 8820 ( first Montego soundboard). I have this board working in OS/2 Warp 4 with the driver Aureal_Vortex_AU8820_1-0.zip (more two necessary patches Aureal_Vortex_Patch.zip and Aureal_Vortex_AU8820x_Patch.zip, but working as expected)

I woud like to use this board with ArcaOS but not works with this driver. Really, this is one of OS/2 sound drivers not included in the list of available drivers in installation program of ArcaOS or eCS.

So, I decided to try Uniaud, as in my tests with linux live CD (Ubuntu 12.04-32bit-alsa 1.0.24 based) soundboard works ok.

Unfortunatelly, not good results...

At boot, similar for all versions of uniaud32.sys tried:

Universal Audio Driver for OS/2 v2.02.05-SVN
Based on Linux 3.10.140
Copyright 2005-2020 Netlabs http://;www.netlabs.org
Copyright 2000-2020 The ALSA Project

ERROR: Initialization failed
OSS32_Initialize: FAILEDpci_register_driver: query_device found 257 12eb:1 class=40100 checking bt87x
OS/2 Universal Audio MMPM/2 Driver v1.9.7
Copyright 2000-2002 Innotek Systemberatung GmbH
Copyright 2002 IBM Corporation
Maintained by Netlabs http://;www.netlabs.org

Unable to attach 32-bit driver
SYS1201: The device driver C:\MMOS2\UNIAUD16.SYS specified in the DEVICE statement on line 182 of the CONFIG.SYS file was not installed



Attached is alsa_aureal_log.txt

Attached PCI.txt.May be useful!
Title: Re: Updated Uniaud32 build
Post by: Renee Senger on August 23, 2020, 03:51:43 am
Another test:

Card info for adapter 0:
  num: 0
  id: ICH5
  driver: ICH4
  name: Intel ICH5
  longname: Intel ICH5 with AD1888 at irq 17
  mixer: Analog Devices AD1888
  componenets: AC97a:41445368

Tested with uniaud32-linux-4.0.9-20200816

Very good results. All is working!

(mic input and winos2 dta not yet tested)

Attached unimix_list.txt and alsa-AD1888log.txt
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 23, 2020, 09:26:51 am
Hi All,

Prompted by Lewis reporting a problem with no sound on a newish laptop, I built uniaud32.sys last night for the first time since December 2015.

After managing to patch the realtek driver to add some additional chipsets which got Lewis sound, I got interested again in updating Uniaud.

As some may recall, way back when, I was updating the uniaud32 core with the latest ALSA code.... I got as far as ALSA 1.0.24 (released January 2011), but could never get ALSA 1.0.25 (released early 2012) to work. After that, the code was incorporated into the linux kernel, and it all got too hard. Plus, at the time, we still had the hangs with hda audio and I started running linux as my main OS.

Anyhow, after helping Lewis last night, I added the OS/2 patches to the linux kernel 3.0 code and got it building. This represents an alsa-kernel approximately 6 months newer that what's in the last uniaud32 release.

I've tested this in virtualbox with the Intel HDA driver and it still makes sound. That's about the level of testing I've done. I can't be bothered walking upstairs to grab one of the Thinkpads and test on other hardware.

Anyway...... link is http://smedley.id.au/tmp/uniaud32-linux-3.0.zip

Feedback appreciated. If there's interest, I'll continue to try and update the code to hopefully fix upstream bugs and support newer hardware.

Cheers,

Paul

More tests...

Aureal Vortex au8820
Chip: Sigmatel STAC9701, 3, 4, 5


Now with Turtle Beach Montego based in Aureal Vortex 8820 ( first Montego soundboard). I have this board working in OS/2 Warp 4 with the driver Aureal_Vortex_AU8820_1-0.zip (more two necessary patches Aureal_Vortex_Patch.zip and Aureal_Vortex_AU8820x_Patch.zip, but working as expected)

I woud like to use this board with ArcaOS but not works with this driver. Really, this is one of OS/2 sound drivers not included in the list of available drivers in installation program of ArcaOS or eCS.

So, I decided to try Uniaud, as in my tests with linux live CD (Ubuntu 12.04-32bit-alsa 1.0.24 based) soundboard works ok.

Unfortunatelly, not good results...

At boot, similar for all versions of uniaud32.sys tried:

Universal Audio Driver for OS/2 v2.02.05-SVN
Based on Linux 3.10.140
Copyright 2005-2020 Netlabs http://;www.netlabs.org
Copyright 2000-2020 The ALSA Project

ERROR: Initialization failed
OSS32_Initialize: FAILEDpci_register_driver: query_device found 257 12eb:1 class=40100 checking bt87x
OS/2 Universal Audio MMPM/2 Driver v1.9.7
Copyright 2000-2002 Innotek Systemberatung GmbH
Copyright 2002 IBM Corporation
Maintained by Netlabs http://;www.netlabs.org

Unable to attach 32-bit driver
SYS1201: The device driver C:\MMOS2\UNIAUD16.SYS specified in the DEVICE statement on line 182 of the CONFIG.SYS file was not installed



Attached is alsa_aureal_log.txt

OK this one is easy to explain.... the au88x0 driver is not enabled in uniaud32....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 23, 2020, 09:55:00 am
For Pete and Dariusz,

Hi Dariusz,

Alright, tested the remaining two releases:

1) uniaud32-linux-4.0.9-20200816.zip
did NOT work, same HARD stop as with uniaud32-linux-3.18.140-20200816.zip

2) uniaud32-linux-3.16.85-20200812.zip
did work, appears to be fine, this release (as compared to the AOS one) added an extra control to my PMUnixMix, that being the Surround Sound. I have not gone through the unimix CLI yet to see the full list of controls.

I've attached the LOG out of the DEBUG version of this driver, my thinking being Paul that maybe that gives you some additional data capture to see why the newer drivers are failing with this hardware?

Thanks for the logs. As I mentioned, there are some hardware quirks missing in the newer builds - I'm 'hoping' there's one for our hardware and I can try enabling it and it fixes things.

I'll try and look at this this afternoon.


Too help narrow things down.... can you test with http://smedley.id.au/tmp/uniaud32-linux-3.17.8-20200823.zip ?

Ta,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on August 23, 2020, 04:52:31 pm
Hi Paul,

For Pete and Dariusz,

Too help narrow things down.... can you test with http://smedley.id.au/tmp/uniaud32-linux-3.17.8-20200823.zip ?
...

This time it's a TRAP 000E in UNIAUD32...see attached screenshot please, this was with the DEBUG version of the driver...should I try the RELEASE one?

Thanks,
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: roberto on August 23, 2020, 10:29:38 pm
uniaud32-linux-3.17.8-20200823.zip        TRAP -Very similar to the photo of Dariusz
uniaud32-linux-3.18.140-20200816.zip    Sound ok, volume ok, screen 16m. okay
uniaud32-linux-4.0.9-20200816.zip       Sound ok, volume ok, screen 16m. okay
All tests in debug mode.
The latest version 409, has a good thing, I can watch a video with background sound, and play the last two versions of the
 bubbels program, without crashing.
 I just need to be able to play a full game of manhjon, without surprises.
  Nice job.
Saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 24, 2020, 03:07:05 am
Sorry about the trap with 3.17 - I'll admit to not testing it locally. I suspect I know what to backport from 3.18.x to fix it.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 24, 2020, 11:14:20 am
Hi All,

Sorry about the trap with 3.17 - I'll admit to not testing it locally. I suspect I know what to backport from 3.18.x to fix it.

What about http://smedley.id.au/tmp/uniaud32-linux-3.17.8-20200824.zip ?
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on August 24, 2020, 02:25:43 pm
Hi Paul,

...What about http://smedley.id.au/tmp/uniaud32-linux-3.17.8-20200824.zip ?...

Oh yeah...now we're cooking with gas!!! lol

Yes sir, no trap, appears to be a clean boot (no warnings, etc.), the WPS comes up fine, sound playback is fine (object moves, as well as PM123 and mplayer). Did not test anything else though.

LOG output from the DEBUG driver is attached.

Thanks!
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: Pete on August 24, 2020, 07:13:39 pm
Hi Paul

uniaud32-linux-3.17.8-20200824.zip (Release) seems to be working fine - limited testing to date.

Interesting that "back porting" a chunk of code from 3.18.x fixes this build as 3.18.x hangs system when startup sound should play.


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 25, 2020, 05:21:34 pm
Build is at http://smedley.id.au/tmp/uniaud32-linux-4.0.9-20200816.zip

Hi Paul.

Version uniaud32-linux-4.0.9-20200816.zip is working on my Thinkpad L420.
No Traps, Sound OK, Volume Control working.

Regards
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on August 25, 2020, 09:50:30 pm
Quote
What about http://smedley.id.au/tmp/uniaud32-linux-3.17.8-20200824.zip ?

I found a few minutes to test this on my Asus A88XM-A machine. No sound.

I attached a photo of the old UniAud (from ArcaOS 5.0.5, which works). It shows the Detected mixer, while the new one shows blank after the colon.
Title: Re: Updated Uniaud32 build
Post by: Neil Waldhauer on September 08, 2020, 05:52:16 am
I'm trying the updated Uniaud32 builds on a Lenovo ThinkCentre M720s, which is a current model. There is an error message at boot time that goes by too quickly to read. No sound, and unimix says there are no cards.

I tried uniaud32-linux-4.0.9-20200816 and uniaud32-linux-3.17.8-20200824

I am guessing from the PCI output that this is Intel High Definition Audio. 8086:A348

The page listed below says that the support is found in Linux kernels: 3.15–3.19, 4.0–4.20, 5.0–5.8, 5.9-rc+HEAD

https://cateee.net/lkddb/web-lkddb/SND_HDA_INTEL.html (https://cateee.net/lkddb/web-lkddb/SND_HDA_INTEL.html)

I am attaching a testlog uniaud in the hope that there is something useful logged.
Title: Re: Updated Uniaud32 build
Post by: Pete on September 08, 2020, 06:01:51 am
Hi Neil

It is possible to pause boot and read the message; simply add this line after the uniaud lines in config.sys:-

DEVICE=C:\PAUSEIT.SYS

The above causes boot to stop and display a message that the "driver cannot be found, press a key to continue...".


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 08, 2020, 06:24:29 am
Hi Neil,

I'm trying the updated Uniaud32 builds on a Lenovo ThinkCentre M720s, which is a current model. There is an error message at boot time that goes by too quickly to read. No sound, and unimix says there are no cards.

I tried uniaud32-linux-4.0.9-20200816 and uniaud32-linux-3.17.8-20200824

I am guessing from the PCI output that this is Intel High Definition Audio. 8086:A348

The page listed below says that the support is found in Linux kernels: 3.15–3.19, 4.0–4.20, 5.0–5.8, 5.9-rc+HEAD

https://cateee.net/lkddb/web-lkddb/SND_HDA_INTEL.html (https://cateee.net/lkddb/web-lkddb/SND_HDA_INTEL.html)

I am attaching a testlog uniaud in the hope that there is something useful logged.

I'll take a look when I get an opportunity. Not all HDA 'backends' are currently enabled - potentially this machine has one of the one's that isn't enabled....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 08, 2020, 11:44:01 am
Hi Neil,

I'm trying the updated Uniaud32 builds on a Lenovo ThinkCentre M720s, which is a current model. There is an error message at boot time that goes by too quickly to read. No sound, and unimix says there are no cards.

I tried uniaud32-linux-4.0.9-20200816 and uniaud32-linux-3.17.8-20200824

I am guessing from the PCI output that this is Intel High Definition Audio. 8086:A348

The page listed below says that the support is found in Linux kernels: 3.15–3.19, 4.0–4.20, 5.0–5.8, 5.9-rc+HEAD

https://cateee.net/lkddb/web-lkddb/SND_HDA_INTEL.html (https://cateee.net/lkddb/web-lkddb/SND_HDA_INTEL.html)

I am attaching a testlog uniaud in the hope that there is something useful logged.

8086:A348 looks like the correct pci id.  The bad news is that this chipset is too new for linux 4.1. That link is misleading - many of those chipsets have only been added to newer kernels.

8086:a348 appears in the linux 4.19.143 code, but not the 4.9.219 code - so seems it was added somewhere between linux 4.9 and 4.19.

It's non-trivial just to add the pci-ids to the current code - it will have to wait until I can spend some more time on this and get the linux code updated to a newer kernel...

Edit: 4.11 is the first time I see the a348 pci id..
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 10, 2020, 08:57:55 am
OK I'm kinda stuck trying to get 4.1 to produce sound, so I've decided to try applying the 4.0.9 patches to the 4.4.x source code. Will see how I go....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 11, 2020, 10:44:41 am
OK I'm kinda stuck trying to get 4.1 to produce sound, so I've decided to try applying the 4.0.9 patches to the 4.4.x source code. Will see how I go....

4.4.238 code is compiling - with a bunch of warnings (some of which will need to be actioned) - and 124 missing symbols - down from 150+ this morning ;)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 17, 2020, 11:06:32 am
OK I'm kinda stuck trying to get 4.1 to produce sound, so I've decided to try applying the 4.0.9 patches to the 4.4.x source code. Will see how I go....

4.4.238 code is compiling - with a bunch of warnings (some of which will need to be actioned) - and 124 missing symbols - down from 150+ this morning ;)

Down to 5 missing symbols...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 18, 2020, 01:29:57 am
OK I'm kinda stuck trying to get 4.1 to produce sound, so I've decided to try applying the 4.0.9 patches to the 4.4.x source code. Will see how I go....

4.4.238 code is compiling - with a bunch of warnings (some of which will need to be actioned) - and 124 missing symbols - down from 150+ this morning ;)

Down to 5 missing symbols...

OK I got it to link - the final missing symbol was a hack, but it will do for now.

Current status is that AC97 hardware boots (at least my virtualbox system does), but HDA traps (on virtualbox).

I'm currently investigating the HDA traps, but if anyone has AC97 hardware and wants to help confirm the core at least works, feel free to try http://smedley.id.au/tmp/uniaud32-linux-4.4.238-20201018.zip

Again: this code traps on Virtualbox with HDA hardware and a Sigmatel mixer.


Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 18, 2020, 09:51:13 pm
Sorry to report my laptop trapped with the new driver.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 18, 2020, 10:37:23 pm
Sorry to report my laptop trapped with the new driver.

What hardware?
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on October 18, 2020, 11:41:43 pm
Hi Paul,

...
I'm currently investigating the HDA traps, but if anyone has AC97 hardware and wants to help confirm the core at least works, feel free to try http://smedley.id.au/tmp/uniaud32-linux-4.4.238-20201018.zip
...

Trap 000e at boot, see attached.

This is on the following hardward (copy/paste from a prior report to simplify):

Code: [Select]
pci_register_driver: matched 29 1002:4383/40300 with 1002:4383/0 0 (hda_intel)
RMCreateDriver rc=0 DriverHandle=f54bdf60
pci_enable_device f344f8d0
...
00-00: ALC889 Analog : ALC889 Analog
 : playback 1
 : capture 1
00-01: ALC889 Digital : ALC889 Digital
 : playback 1
00-02: ALC889 Alt Analog : ALC889 Alt Analog
 : capture 2
OSS32_WaveOpen. streamid f33161d8

GetUniaudPcmCaps: cp1. phandle f33161d8
GetUniaudPcmCaps: cp2. nr of streams: 1
chan: from 2 to 8

HDA audio detected - don't support 5512 - 32000 Hz audio sample rates
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 19, 2020, 12:09:53 am
Similar trap e here too,
Code: [Select]
Bus 0 (PCI Express), Device Number 27, Device Function 0
 Vendor 8086h Intel Corporation
 Device 1C20h 6 Series/C200 Series Chipset Family High Definition Audio Controller
Code: [Select]
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xe1a60000 irq 22
  mixer: Intel ID 2805
  componenets: HDA:10ec0269,102804ad,00100100 HDA:80862805,80860101,00100000
[code]
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 19, 2020, 12:16:49 am
Sorry, perhaps I wasn't clear - I'm *EXPECTING* traps with HDA hardware - I'm more interested if it works with AC97 hardware other than my Virtualised system....
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 19, 2020, 12:59:49 am
My laptop is a DELL Inspiron 15R 5537. The older Uniaudio drivers work.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 19, 2020, 04:42:41 am
My laptop is a DELL Inspiron 15R 5537. The older Uniaudio drivers work.

What does 'unimix -card' report - I'll guess it's Intel HDA in which case the trap is to be expected.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 20, 2020, 12:19:01 am
It reports HDA Intel PCH.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 20, 2020, 12:37:16 am
It reports HDA Intel PCH.

Yep - so the trap is known... I'll post when/if I have HDA hardware working....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 20, 2020, 10:46:51 am
It reports HDA Intel PCH.

Yep - so the trap is known... I'll post when/if I have HDA hardware working....

With some help from Steven Levine, I found the problem with the initial trap, and came across a second trap. I understand where the code is trapping, I just need to work out how to fixit. Hopefully that will be the last issue before I get a booting system again :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 22, 2020, 10:29:15 am
Hi All,

http://smedley.id.au/tmp/uniaud32-linux-4.4.238-20201022.zip

Trap should be fixed - but here at least - no mixer is detected. I still need to investigate this - but curious if it's a Sigmatel issue, or a broader issue.

unimix -card here reports:
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Card info for adapter 0:
  num: 0
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xf0804000 irq 11
  mixer:
  componenets: HDA:83847680,83847680,00103401


Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on October 22, 2020, 12:10:53 pm
Paul,

 Seems to work OK here:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xef320000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

 I'll keep using it and let you know if anything comes up... thanks!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 22, 2020, 02:00:22 pm
No trap here but no sound. Unimix gives me the same exact output as David.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 22, 2020, 10:31:59 pm
Hi Dave,

Seems to work OK here:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xef320000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

 I'll keep using it and let you know if anything comes up... thanks!


Wow! That's encouraging! Maybe my issue is some idiosyncracy with Sigmatel hardware. I'll dust off the Thinkpad tonight or over the weekend and test locally with real hardware.

Are you able to upload a copy of alsahlp$ after a fresh boot?

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 22, 2020, 10:32:43 pm
No trap here but no sound. Unimix gives me the same exact output as David.

Mixer settings may have changed - try pmunimix and make sure all outputs are turned up.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on October 22, 2020, 10:43:44 pm
Hi Dave,

Seems to work OK here:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xef320000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

 I'll keep using it and let you know if anything comes up... thanks!


Wow! That's encouraging! Maybe my issue is some idiosyncracy with Sigmatel hardware. I'll dust off the Thinkpad tonight or over the weekend and test locally with real hardware.

Are you able to upload a copy of alsahlp$ after a fresh boot?

Cheers,

Paul.

 Sure... here ya go...
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on October 22, 2020, 11:14:29 pm
Hi Paul,

the hardware I am using is far to new for the driver, I guess. So I did not expect it to work.

Even though I attach my results, may be I can be of help in a future release.

The hardware is a lenovo Thinkpad T480 with Intel core i7-8550 CPU.

Your driver loads and boots but can not detect the mixer. I guess this is what you would expect with this hardware.

If I can be of any help, please let me know.

Thanks for your great work!!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 22, 2020, 11:33:33 pm
Hi Sigurd,

the hardware I am using is far to new for the driver, I guess. So I did not expect it to work.

Even though I attach my results, may be I can be of help in a future release.

The hardware is a lenovo Thinkpad T480 with Intel core i7-8550 CPU.

Your driver loads and boots but can not detect the mixer. I guess this is what you would expect with this hardware.

If I can be of any help, please let me know.

Thanks for your great work!!

From some googling, seems the T480 has the ALC285 chip. According to https://www.phoronix.com/scan.php?page=news_item&px=Realtek-ALC-215-285-289 support was added in the linux 4.13 kernel.

I'll try and fix Sigmatel support so I can test in Virtualbox, but otherwise, I'll probably push on to newer kernel versions over the weekend.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 23, 2020, 10:52:55 am
I'll try and fix Sigmatel support so I can test in Virtualbox, but otherwise, I'll probably push on to newer kernel versions over the weekend.

I didn't look at Sigmatel, but pushed on. 4.4.238 patches applied pretty cleanly to 4.9.240 - we'll see how hard it is to make it compile/link...
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on October 24, 2020, 01:19:52 am
Hi Paul,

...http://smedley.id.au/tmp/uniaud32-linux-4.4.238-20201022.zip

Trap should be fixed - but here at least - no mixer is detected. I still need to investigate this - but curious if it's a Sigmatel issue, or a broader issue.
...

Here unimix reports:

Code: [Select]
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Card info for adapter 0:
  num: 0
  id: SB
  driver: HDA-Intel
  name: HDA ATI SB
  longname: HDA ATI SB at 0xfe8f4000 irq 16
  mixer:
  componenets: HDA:10ec0889,14627576,00100004

...but no sound is produced.

Invoking UniMix produces a pop-up error message with the following text:

Code: [Select]
uniaud_mixer_get_ctls_number: error $00000000. Close the program?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 24, 2020, 05:10:17 am
4.9.240: http://smedley.id.au/tmp/uniaud32-linux-4.9.240-20201024.zip

Here on Sigmatel hardware (virtualbox), and on a Thinkpad X250 (Realtek HW), I get no sound and output similar to:
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Card info for adapter 0:
  num: 0
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xf0804000 irq 11
  mixer:
  componenets: HDA:83847680,83847680,00103401

YMMV... It's  interesting that Dave get's his mixer identified... I might need more logging...
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 24, 2020, 07:13:37 am
Tried this one, no sound and PMUniMix gives a ctls error number of $0000000. I posted my Intel HDA info earlier in the thread.
Thanks for the work.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 24, 2020, 07:42:30 am
Tried this one, no sound and PMUniMix gives a ctls error number of $0000000. I posted my Intel HDA info earlier in the thread.
Thanks for the work.
Thanks Dave - this will be the mixer not being detected.... weird how it works for Dave Mckenna... the positive is that 4.9.x behaves the same as 4.4.x for me at least, so the fix will likely be similar. Patches also applied pretty cleanly to 4.14.x, so will try get that compiling then work on fixes...
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on October 24, 2020, 10:03:55 am
Hi Paul,

No problems here but thought I would still let you know that all of your latest builds still work well with:

Card info for adapter 0:
  num: 0
  id: Live
  driver: EMU10K1
  name: SB Live! 5.1
  longname: SB Live! 5.1 (rev.7, serial:0x80641102) at 0xd880, irq 22
  mixer: eMicro EM28028
  componenets: AC97a:454d4328

including uniaud32-linux-4.9.240-20201024.zip

Thanks :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 24, 2020, 10:21:45 am
Hi All,

Tried this one, no sound and PMUniMix gives a ctls error number of $0000000. I posted my Intel HDA info earlier in the thread.
Thanks for the work.
Thanks Dave - this will be the mixer not being detected.... weird how it works for Dave... the positive is that 4.9.x behaves the same as 4.4.x for me at least, so the fix will likely be similar. Patches also applied pretty cleanly to 4.14.x, so will try get that compiling then work on fixes...

http://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201024.zip is based on the linux 4.14.x kernel - again - no hda mixer detection here in Virtualbox.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 24, 2020, 10:22:26 am
Hi Ian,

No problems here but thought I would still let you know that all of your latest builds still work well with:

Card info for adapter 0:
  num: 0
  id: Live
  driver: EMU10K1
  name: SB Live! 5.1
  longname: SB Live! 5.1 (rev.7, serial:0x80641102) at 0xd880, irq 22
  mixer: eMicro EM28028
  componenets: AC97a:454d4328

including uniaud32-linux-4.9.240-20201024.zip

Thanks :)

Thanks for confirming things work on non-HDA hardware :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on October 24, 2020, 02:37:34 pm
Hi All,

http://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201024.zip is based on the linux 4.14.x kernel - again - no hda mixer detection here in Virtualbox.

Cheers,

Paul

Hi Paul,

tested this 4.14.202, seems to show the same results as with the older one.

Boots OK, does not detect the mixer.

Great to see 4.14 boot at all!

What can I provide to assist?

Thanks!
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 25, 2020, 05:37:42 pm
New version no trap, no sound and no PMmixer connect.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 25, 2020, 09:11:21 pm
New version no trap, no sound and no PMmixer connect.

Thanks guys - I need to think about how best to proceed. At this stage, I think it's best to resurrect the logging that was broken around kernel 4.0 timeframe - I'm hoping that will elicit some clues as to why the mixer isn't being detected.

I'll post here when I have some news.
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on October 25, 2020, 09:30:21 pm

Thanks guys - I need to think about how best to proceed. At this stage, I think it's best to resurrect the logging that was broken around kernel 4.0 timeframe - I'm hoping that will elicit some clues as to why the mixer isn't being detected.

I'll post here when I have some news.

Hi Paul,

may be you can team up with "abwillis1", he helped me out with getting Uniaud to run on a Thinkpad X250.

Please see: https://mantis.arcanoae.com/view.php?id=2637 (https://mantis.arcanoae.com/view.php?id=2637)

(ArcaOS Bugtracker Bug 2637)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 25, 2020, 10:08:26 pm
OK, so I fixed some logging code....

Firstly, there was a pretty major code re-structuring in relation to the sound code between 4.0 and 4.1. Some of the added api calls I simply stubbed as I didn't think they were important, some I implemented.

What I believe is happening is that whilst the HDA chipset is being recognised, I don't think the code to support the mixers is being loaded at all. I see no logging from (for example) patch_sigmatel.c

What I need to do is step through the code and determine where the hda_intel/hda_codec code is supposed to call the code for the mixer, then work out why it isn't being loaded. I suspect this will come back to some code I stubbed that needs to be at least partially implemented - I just don't want to waste time implementing the wrong code...

No timeframes on updates, I have a busy week coming up...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 26, 2020, 09:51:50 am
OK, so I fixed some logging code....

Firstly, there was a pretty major code re-structuring in relation to the sound code between 4.0 and 4.1. Some of the added api calls I simply stubbed as I didn't think they were important, some I implemented.

What I believe is happening is that whilst the HDA chipset is being recognised, I don't think the code to support the mixers is being loaded at all. I see no logging from (for example) patch_sigmatel.c

What I need to do is step through the code and determine where the hda_intel/hda_codec code is supposed to call the code for the mixer, then work out why it isn't being loaded. I suspect this will come back to some code I stubbed that needs to be at least partially implemented - I just don't want to waste time implementing the wrong code...

No timeframes on updates, I have a busy week coming up...

OK I think I figured out that I need to implement driver_register() that than stubbing it. I got what I think is enough implemented, and got it compiling, however, for my efforts I'm rewarded with a Trap.

I'll try debug that tomorrow.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 31, 2020, 07:21:54 am
Just updating things, with help from Steven Levine, I fixed the trap very quickly. I then found I had to implement some more stuff. I think I'm VERY close, but at this stage, still no working HDA.

I just need to find the last bit of missing glue - hopefully soon - as I've spent way too much time on this over the last couple of weeks...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 31, 2020, 11:06:59 pm
Note: This build will NOT produce sound on HDA hardware....

However, I found some missing 'glue' and made progress. I still have some code wrapped around '#ifndef TARGET_OS2' statements to prevent traps....

What I'm interested in is whether the mixer is correctly detected or not - in particular for Sigurd as his hardware should now be supported.

I'll keep working on fixing the traps so we can actually get sound :)

https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201101.zip

AC97 hardware should continue to work....

Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 31, 2020, 11:41:06 pm
No trap, no sound and no connect to PMmix.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 01, 2020, 12:08:49 am
No trap, no sound and no connect to PMmix.

Is the mixer displayed correctly by unimix -card ?

No trap, no sound and no connect to PMmix is the expected behaviour.
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on November 01, 2020, 01:10:20 am
Note: This build will NOT produce sound on HDA hardware....

However, I found some missing 'glue' and made progress. I still have some code wrapped around '#ifndef TARGET_OS2' statements to prevent traps....

What I'm interested in is whether the mixer is correctly detected or not - in particular for Sigurd as his hardware should now be supported.

I'll keep working on fixing the traps so we can actually get sound :)

https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201101.zip

AC97 hardware should continue to work....

Hi Paul,

yes, it does detect the mixer, REALTEK ALC 257

Good work!!

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 01, 2020, 09:21:51 am
yes, it does detect the mixer, REALTEK ALC 257

Good work!!

Great!!! VERY encouraging. Hopefully by this time next week you'll actually be able to produce sound :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 01, 2020, 11:03:53 am
https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201101b.zip has some updates - still no mixer controls yet - calling the code to set them up hangs the system whilst processing uniaud32 (hence it's commented out right now)

However, 'copy alsahlp$ alsa.log' gives encouraging results that we're VERY close :)
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on November 03, 2020, 05:06:06 pm
Hi Paul

uniaud32-linux-4.14.202-20201101b.zip is not working for me on the Thinkpad L420. (Intel HDA 8086:1C20)
- No traps, no sound after reboot.

Version 2.02.06 / 6 May 2020 14:21:32 was working fine for me.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 03, 2020, 09:51:40 pm
Hi Martin,

uniaud32-linux-4.14.202-20201101b.zip is not working for me on the Thinkpad L420. (Intel HDA 8086:1C20)
- No traps, no sound after reboot.

Version 2.02.06 / 6 May 2020 14:21:32 was working fine for me.


Thanks - that's good to know. I'll only tested here with virtualised sigmatel hardware. As I mentioned - the code to setup controls is currently commented out as it's causing a hang loading uniaud32.sys. I made progress last night understanding why that's occuring. Hopefully soon you'll have no traps AND sound.

The positive is that the ALSA core has detected the hardware and all the appropriate inputs/outputs - we just need to make them work ;)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 05, 2020, 11:28:10 pm
https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201101b.zip has some updates - still no mixer controls yet - calling the code to set them up hangs the system whilst processing uniaud32 (hence it's commented out right now)

However, 'copy alsahlp$ alsa.log' gives encouraging results that we're VERY close :)

OK I found where the code is hanging - I don't yet know why it's working, but am semi-confident if that issue is fixed, I'll have working sound again with a significantly updated core code.

(Not sure anyway is reading/cares about these updates, but just trying to keep people informed on the status of this work)
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on November 06, 2020, 02:48:44 am
(Not sure anyway is reading/cares about these updates, but just trying to keep people informed on the status of this work)

I care  :D I keep looking your Uniaud updates, it is important to bring to the latest level to try to get more sound support for the platform. Keep posting !!
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on November 06, 2020, 03:58:06 am
Hi Paul,

Quote
(Not sure anyway is reading/cares about these updates, but just trying to keep people informed on the status of this work)

updates are great, they serve several purposes including:

1) lets us know something is happening.
2) lets us know there is a lot more to it then simply recompiling code (which some of us really appreciate!).
3) you put a lot of effort into something you don't need to, which once more many of us really appreciate, and I note your karma is high :).
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on November 06, 2020, 08:25:38 am
(Not sure anyway is reading/cares about these updates, but just trying to keep people informed on the status of this work)

Hi Paul,

I am very happy to see your progress and I could just repeat the words of Ianmanners, wich I can confirm in anyway!

To have sound on a Laptop with native ArcaOS/ OS/2 that was build in 2018 is great and with that you keep my motivation to still play around with OS/2 alive!!

All the best and thank you again!
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 06, 2020, 01:31:32 pm
Paul,
As this helps my favorite OS. I appreciate your efforts. And we all need a boost from time to time. I will continue to test.

Title: Re: Updated Uniaud32 build
Post by: ivan on November 06, 2020, 01:47:50 pm
Hi Paul,

As others have said there are some of us that appreciate the work you do for us and the fact that you let us know your progress. 

In my case I hope your work will give sound to my various AMD Ryzen based computer builds that have the Realtek ALC 892 7.1 channel audio as well as HDMI audio since nothing available works at the moment.

BTW I am still using the version of Sigil you produced years ago even if it is rather long in the tooth, so thank you for that also.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 08, 2020, 11:35:30 pm
Hi Ivan,
As others have said there are some of us that appreciate the work you do for us and the fact that you let us know your progress. 

In my case I hope your work will give sound to my various AMD Ryzen based computer builds that have the Realtek ALC 892 7.1 channel audio as well as HDMI audio since nothing available works at the moment.

I'd be interested to see the results of trying to boot with the 4.14.202 build I posted earlier with the Ryzen's - you won't get sound, but the kernel is reasonably up to date. If it doesn't boot, please send me a testlog report so I can see the PCI IDs and give you an indication of the required kernel version to get working sounds.

Status update on 4.14.202:
I fixed the hang at boot by making sure I was using consistent kernel code/headers for regmap/regcache/rbtree - I'd previously shoplifted bits of code from different kernel versions as I'd needed different functions - not a good idea.

For my rewards, I got a trap in krealloc() which I fixed, now another trap setting up jack controls that I'm still investigating. I still hope that once this issue is resolved, the mixer controls will work and we'll be good to go, but there's no guarantee that hasn't been other changes that break things.

Thanks to all who posted in here - I'll keep posting updates as meaningful things happen.

Cheers,

Paul

PS that Sigil build is so long ago, I don't even remember what Sigil is/does :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 10, 2020, 08:17:14 am
OK.... with some more insight from Steven Levine, I now get a successful boot with 4.14.202 and HDA hardware without having to comment out any code! However, on Virtualbox, I'm still getting no sound output. I'll try testing on real hardware as time permits (noting that on my X250 I get a Trap 000E), but any data from other users would be helpful.

A copy of alsahlp$ (ie copy alsahlp$ alsa.log) would be helpful after an initial boot.

Download link is: https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201110.zip

Thanks!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on November 10, 2020, 01:17:51 pm
 Hi Paul,

  Unfortunately, I get a TrapE with this one, I'll attach a screen shot. Also got my first Trap dump if you need it. (Haven't had a chance in a while to try any other versions lately... sorry about that).

Regards,
Title: Re: Updated Uniaud32 build
Post by: Andi B. on November 10, 2020, 01:34:41 pm
Interestingly I had a TRAP 000e yesterday in os2krnl with the same 'location' information. Maybe totally different as EIP is different. But can anyone explain why 'location' is the same?
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 10, 2020, 02:15:42 pm
Got a trap like David.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 10, 2020, 09:12:36 pm
Hi Dave,

  Unfortunately, I get a TrapE with this one, I'll attach a screen shot. Also got my first Trap dump if you need it. (Haven't had a chance in a while to try any other versions lately... sorry about that).

If you can zip the dump up and get it to me that would be helpful. It's probably the same trap as I get on the X250, but I haven't had time to setup a dump partition on it yet, so this may be quicker.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on November 10, 2020, 11:00:49 pm
Paul,

 7zipped it up and it is about 105MB (I limited the trap dump to 512MB - hope that is OK). How should I send?

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 10, 2020, 11:30:46 pm
Hi Dave,

7zipped it up and it is about 105MB (I limited the trap dump to 512MB - hope that is OK). How should I send?

You hopefully have a Dropbox file request from me, that should allow you to upload the dump to my dropbox account.

The other sorta good news, is that I was able to setup a dump partition on my X250 so I have a sample dump from there - it would be good to confirm it's the same trap though.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Lars on November 11, 2020, 12:10:21 am
Interestingly I had a TRAP 000e yesterday in os2krnl with the same 'location' information. Maybe totally different as EIP is different. But can anyone explain why 'location' is the same?
The location is just the exception handler in the kernel. It will always be the same (at least for the same trap number but I even think it is a general handler).
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on November 11, 2020, 01:26:55 pm
Even if it is not much of help, the same trap here on the T480.

Unfortunately I do not have the possibility to grap a log file, but I think it is not necessary as others provided already!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 11, 2020, 09:38:58 pm
Thanks guys - the results are a little deflating given my relative success in Virtualbox. I'm going to focus for the next few days on cleaning up the build and trying to eliminate as many warnings as possible - there are some that could *potentially* explain crashes. Interestingly, Dave's dump indicates a trap in a different section of the code to the trap from my X250 - I'll try and get dump partitions setup on my T60 and T510 as well.

I'll post updates here when there's something to update.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 14, 2020, 08:26:09 am
I'll post updates here when there's something to update.

Still a WIP - but I fixed some warnings - and my T60 now boots (still no sound but it detects the hardware). My X250 still traps, but it's a different trap to Dave's - so who knows, this might boot for some more folks.

https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201114.zip
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on November 16, 2020, 10:01:36 am
Hi Paul,

FYI: unfortunately I will not be able to test for a while, I had to change the internal SSD of the T480 to a NVme, so ARcaOS is gone on this device. I might bring back a SATA SSD but I am not sure when. Maybe meanwhile there will be a NVme Driver from Arca Noae and that would help as well. Sorry.

All the best.

Sigurd
Title: Re: Updated Uniaud32 build
Post by: roberto on November 19, 2020, 09:09:17 pm
Another TRAP with this card:

[C:\]UNIMIX -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xd2520000 irq 22
  mixer: IDT 92HD81B1X5
  componenets: HDA:111d7605,103c1630,00100105 HDA:11c11040,103c3066,00100200

Saludos
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on November 20, 2020, 09:11:48 pm
Hi Paul.

It may be trivial, but I hope it can be possible to do this on the UNIAUD driver:
- http://trac.netlabs.org/uniaud/ticket/246 (http://trac.netlabs.org/uniaud/ticket/246)

I know it is a controversial issue, since people think different on how to show the /V (verbose) information on boot, but I have the dream that maybe someday all drivers may look similar when you boot your OS/2 box (http://www.edm2.com/index.php/Suggestion_for_OS/2_and_eComstation_Drivers_Arguments#Verbose_Mode_.E2.80.9C.2FV.E2.80.9D_or_.E2.80.9C.2Fv.E2.80.9D).

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 20, 2020, 09:47:38 pm
Hi Roberto,

Another TRAP with this card:

Thanks - we think we know what's causing this - just need to work out why it's happening. The positive is that once it's fixed, it's likely to fix the sound as well.

It may be trivial, but I hope it can be possible to do this on the UNIAUD driver:
- http://trac.netlabs.org/uniaud/ticket/246 (http://trac.netlabs.org/uniaud/ticket/246)

I know it is a controversial issue, since people think different on how to show the /V (verbose) information on boot, but I have the dream that maybe someday all drivers may look similar when you boot your OS/2 box (http://www.edm2.com/index.php/Suggestion_for_OS/2_and_eComstation_Drivers_Arguments#Verbose_Mode_.E2.80.9C.2FV.E2.80.9D_or_.E2.80.9C.2Fv.E2.80.9D).

It might happen - but it currently has zero priority until we're at as higher code level as we can get....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 21, 2020, 09:52:20 am
OK, http://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201121.zip doesn't trap (or produce sound) on my X250, T60 or virtualbox systems.

Interested to get feedback from other systems. Pretty sure the lack of audio are due to broken regmap/regcache
Title: Re: Updated Uniaud32 build
Post by: David McKenna on November 21, 2020, 02:58:55 pm
 Hi Paul,

  Sorry to report that I get a TrapE with this one too... I'll attach the serial port log, and I have a trapdump if you want it...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 21, 2020, 04:49:20 pm
No trap E.But no sound. Just like you said.
Title: Re: Updated Uniaud32 build
Post by: roberto on November 22, 2020, 10:12:24 am
The TRAP persist
Saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 22, 2020, 11:08:09 am
The TRAP persist
Saludos

Very interesting - anyway - it's confirmed that regmap/regcache are still a problem. I'll keep digging as time permits.
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on November 23, 2020, 03:20:54 am
Hi Paul,

OK, http://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201121.zip doesn't trap (or produce sound) on my X250, T60 or virtualbox systems...

Small change in hardware here, I went from my MSI 880G-E45 motherboard to a MSI 890FXA-HD70. The advertised specs showed a RealTek chipset:

1) 880G-E45 => Realtek® ALC888S/ACL889

2) 890FXA-GD70 => Realtek® ALC889
Code: [Select]
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Card info for adapter 0:
  num: 0
  id: SB
  driver: HDA-Intel
  name: HDA ATI SB
  longname: HDA ATI SB at 0xfe4f8000 irq 16
  mixer: Realtek ALC889
  componenets: HDA:10ec0889,14627640,00100004

...but your last known good release which was working on the 880G-E45:

Code: [Select]
8-12-20  5:07a       586,198     54 a---  uniaud32.sys

Build Level Display Facility Version 6.12.675 Sep 25 2001
(C) Copyright IBM Corporation 1993-2001
Signature:       @#Netlabs:2.02.05#@##1## 12 Aug 2020 18:37:33     ARCAOS-D0C:3.
16.85:::05::SVN@@Universal Audio Driver for OS/2 (c)  Netlabs
Vendor:          Netlabs
Revision:        2.02.05
Date/Time:       12 Aug 2020 18:37:33
Build Machine:   ARCAOS-D0C
ASD Feature ID:  3.16.85
FixPak Version:  SVN
File Version:    2.2.5
Description:     Universal Audio Driver for OS/2 (c)  Netlabs

...was now trapping on the newer 890FXA-GD70 board.

Anyways, the good news is that this latest drop does NOT trap, but as others have noted, it does not produce output either.

For me it's a positive direction given no TRAP! Thank you!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 30, 2020, 09:05:16 am
Hi All,

Just a quick update to say that there's no real progress - broadly it's understood why the builds don't produce sound, and where the problem lies (the port of the regmap/regcache APIs), but there isn't a clear root cause/solution to this yet. I'm actually taking a break from looking at Uniaud for a short time - with any luck a fresh look after a break may help shed some light on the problem.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mentore on December 07, 2020, 08:10:42 am
Hi All,

Just a quick update to say that there's no real progress - broadly it's understood why the builds don't produce sound, and where the problem lies (the port of the regmap/regcache APIs), but there isn't a clear root cause/solution to this yet. I'm actually taking a break from looking at Uniaud for a short time - with any luck a fresh look after a break may help shed some light on the problem.

Cheers,

Paul

Best of luck Pete. I'm following this and hope to try the new builds soon (have to build one or two machines almost from scratch with some old hardware).
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 07, 2020, 08:47:28 am
Hi Mentore,

Just a quick update to say that there's no real progress - broadly it's understood why the builds don't produce sound, and where the problem lies (the port of the regmap/regcache APIs), but there isn't a clear root cause/solution to this yet. I'm actually taking a break from looking at Uniaud for a short time - with any luck a fresh look after a break may help shed some light on the problem.

Best of luck Pete. I'm following this and hope to try the new builds soon (have to build one or two machines almost from scratch with some old hardware).

Hopefully there will be good news sometime soon. It's extremely frustrating being so close to it working  and it not quite working!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 23, 2020, 02:59:33 am
Quick update:

I built a driver based on the latest 4.19.x kernel - unfortunately I'm getting a TRAP 000E during boot that I need to debug.

I updated the code to 5.4.x - the code is compiling but I have around 10 missing symbols of functions that need to be added to the linux 'kernel'.

Hopefully I can resolve those over the next couple of days and see where we land. I keep hoping that in bumping the code levels things will start working :)

I'll have increased free time over the next couple of weeks over the Christmas/New Year break so will hopefully make some progress.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on December 23, 2020, 09:12:52 am
Quick update:

I built a driver based on the latest 4.19.x kernel - unfortunately I'm getting a TRAP 000E during boot that I need to debug.

I updated the code to 5.4.x - the code is compiling but I have around 10 missing symbols of functions that need to be added to the linux 'kernel'.

Hopefully I can resolve those over the next couple of days and see where we land. I keep hoping that in bumping the code levels things will start working :)

I'll have increased free time over the next couple of weeks over the Christmas/New Year break so will hopefully make some progress.

Cheers,

Paul

Hi Paul,

good news!! I will try to get the T480 to run with OS/2 - ArcaOS  over the Christmas/New Year break as well, so I hopefully will be able to test and support. It depends on only one LCD cable that I expect to be delivered since the beginning of December from China. So it may happen, if not, I will come back with that hardware as soon as possible.

All the best!

Sigurd
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 26, 2020, 09:38:58 pm
OK.... it took a while - but I finally produced sound with HDA hardware (albeit virtualised) with 4.14.202...

Testing appreciated: https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201227.zip

Not sure I fully understand why the change I made to OSS32_WaveClose() fixed things, but I'll worry about that later!!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 26, 2020, 09:49:37 pm
 Just happened to be browsing OS2World when your post appeared, so installed the release 4.14.202 version and it works! Congratulations! I'll keep running with it and let you know if any issues arise, but so far system sounds, PMUniMix and MP3's work.... thanks again!
Title: Re: Updated Uniaud32 build
Post by: roberto on December 26, 2020, 11:34:49 pm
Sorry, another TRAP. The good thing is that I tried it on the 26th, before you created it.
saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 27, 2020, 12:28:19 am
Hi Dave,

Just happened to be browsing OS2World when your post appeared, so installed the release 4.14.202 version and it works! Congratulations! I'll keep running with it and let you know if any issues arise, but so far system sounds, PMUniMix and MP3's work.... thanks again!

Woohoo! Good news! I'll see if I can test with the X250 and T60 later today to get some more data.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 27, 2020, 12:29:33 am
Hi,

Sorry, another TRAP. The good thing is that I tried it on the 26th, before you created it.

I might need a trapdump - let's see first if I can recreate here on bare metal - so far I've only tested ac97 and hda using Virtualbox (Sigmatel hardware)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on December 27, 2020, 02:15:01 am
Hi
Testing appreciated: https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201227.zip

I tried it on the Thinkpad L420.
- No traps on boot, but no sound.
Let me know if something else can be tested.

I returned to "uniaud32-linux-4.0.9-20200816.zip" to get sound back.


Regards
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on December 27, 2020, 03:41:54 am
OK.... it took a while - but I finally produced sound with HDA hardware (albeit virtualised) with 4.14.202...

Testing appreciated: https://smedley.id.au/tmp/uniaud32-linux-4.14.202-20201227.zip

Not sure I fully understand why the change I made to OSS32_WaveClose() fixed things, but I'll worry about that later!!

Sound working on my Thinkpad W530
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 27, 2020, 08:30:01 am
4.19.x is still trapping for me, so meanwhile, here's a build of 4.16.18 - https://smedley.id.au/tmp/uniaud32-linux-4.16.18-20201227.zip
Title: Re: Updated Uniaud32 build
Post by: roberto on December 27, 2020, 11:32:15 am
TRAP 4.16.18
saludos
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 27, 2020, 02:33:00 pm
 Paul,

  After successfully using 4.14 for a few hours yesterday, when I booted up this morning I got a TrapE in UNIAUD32. Tried to boot a few times but no go. I could get a trap dump if you want.

  Installed 4.16 and it works (at least at first) too. It is a little different in that volume is very loud (but not distorted) and I only get 22 controls with Unimix where I got 44 with older versions. PMUnimix only show 3 volume controls where I had 9 before. This has no practical implications (can't use surround anyway) but thought I should mention.

Regards,
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 27, 2020, 02:48:31 pm
 Went ahead and rebooted with 4.16, and now I get no sound (but it does boot). Alsahlp$ attached....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 27, 2020, 09:21:03 pm
Went ahead and rebooted with 4.16, and now I get no sound (but it does boot). Alsahlp$ attached....

Looks like Uniaud16 can't open the device - possibly related to the change I made that made sound work (at least initially).

Was this a hard reboot (power off) or a soft reboot?
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 27, 2020, 09:58:12 pm
 Hi Paul,

  That was a soft reboot. Tried a power off, then restart and now sound is back, although loud with only 3 volume controls in PMUnimix. I'll upload alsahlp$....

Edit: Hmmm... now there are 25 controls listed by Unimix -names and 4 volume controls in PMUnimix. Strange...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 27, 2020, 10:11:45 pm
Hi Paul,

  That was a soft reboot. Tried a power off, then restart and now sound is back, although loud with only 3 volume controls in PMUnimix. I'll upload alsahlp$....

Regards,

OK interesting - I'll have to mess around some more with the code I changed to get sound sorta working. First though, I want to work out why 4.19.x is trapping. Last night I discovered that v4.18 and v4.17 trap in the same place - so now I "just" need to work out what changed between v4.16 and v4.17 :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 27, 2020, 11:36:18 pm
Update on hardware testing:

Thinkpad T60 (Analog Devices codec) - no sound with 4.14.x or 4.16.x

Thinkpad X250 (Realtek codec) - no sound with 4.14; sound with 4.16.x
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 28, 2020, 12:34:05 am
Hi All,

Hi Paul,

  That was a soft reboot. Tried a power off, then restart and now sound is back, although loud with only 3 volume controls in PMUnimix. I'll upload alsahlp$....

Regards,

OK interesting - I'll have to mess around some more with the code I changed to get sound sorta working. First though, I want to work out why 4.19.x is trapping. Last night I discovered that v4.18 and v4.17 trap in the same place - so now I "just" need to work out what changed between v4.16 and v4.17 :)

OK that wasn't too hard - seems krealloc() is causing issues still - I knew it was a dirty hack. For now, I reverted some code in 4.17's pcm_lib.c to the code from 4.16 and I get sound in Virtualbox:
https://smedley.id.au/tmp/uniaud32-linux-4.17.19-20201228.zip

Hopefully the same fix will get 4.19.x working, I'll know shortly....
Title: Re: Updated Uniaud32 build
Post by: David Graser on December 28, 2020, 01:35:35 am
Hi All,

Hi Paul,

  That was a soft reboot. Tried a power off, then restart and now sound is back, although loud with only 3 volume controls in PMUnimix. I'll upload alsahlp$....

Regards,

OK interesting - I'll have to mess around some more with the code I changed to get sound sorta working. First though, I want to work out why 4.19.x is trapping. Last night I discovered that v4.18 and v4.17 trap in the same place - so now I "just" need to work out what changed between v4.16 and v4.17 :)

OK that wasn't too hard - seems krealloc() is causing issues still - I knew it was a dirty hack. For now, I reverted some code in 4.17's pcm_lib.c to the code from 4.16 and I get sound in Virtualbox:
https://smedley.id.au.au/tmp/uniaud32-linux-4.17.19-20201228.zip

Hopefully the same fix will get 4.19.x working, I'll know shortly....

Should be
https://smedley.id.au/tmp/uniaud32-linux-4.17.19-20201228.zip
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 28, 2020, 02:02:25 am
Hopefully the same fix will get 4.19.x working, I'll know shortly....

https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 28, 2020, 02:24:53 am
Paul,

  4.17 works the same as 4.16 here - only get sound from a hard boot. With 4.17 though, I get a slight delay from system sounds - the item cross-hatches, hesitates, then the sound occurs, then the item opens. Also, alsahlp$ doesn't show the typical info.

  4.19 does not give sound no matter what I do, it boots fine, but no sound. I also detect the hesitation with this one, only with no sound. Alsahlp$ files attached...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on December 28, 2020, 02:39:05 am
Hopefully the same fix will get 4.19.x working, I'll know shortly....

https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

Version uniaud32-linux-4.19.163-20201228.zip is working fine on my Thinkpad L420.
- No Traps, sound and volume control working.
For the moment it is a keeper for my Thinkpad L420.

Regards
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 28, 2020, 02:59:42 am
Paul,

  Installed the debug version of 4.19, rebooted and got a TrapE in UNIAUD32. Rebooted again and after disk checks, it booted and I had sound! I'll attach an alsahlp$. The hesitation is there but different now, items cross-hatch and start, then the sound is heard late. Rebooted again and TrapE. Then again and after disk checks it booted...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 28, 2020, 03:44:16 am
Hi Dave,

  Installed the debug version of 4.19, rebooted and got a TrapE in UNIAUD32. Rebooted again and after disk checks, it booted and I had sound! I'll attach an alsahlp$. The hesitation is there but different now, items cross-hatch and start, then the sound is heard late. Rebooted again and TrapE. Then again and after disk checks it booted...

Any chance of a dump after the trap?

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 28, 2020, 12:20:10 pm
 OK, got it... where to send?
Title: Re: Updated Uniaud32 build
Post by: OS4User on December 28, 2020, 01:53:29 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

hi, tried on on my comp
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on December 28, 2020, 03:42:09 pm
No trap here but no sound.
Title: Re: Updated Uniaud32 build
Post by: roberto on December 28, 2020, 05:06:53 pm
It has given me trap in debug and release. In ver 4.19.163
saludos
Title: Re: Updated Uniaud32 build
Post by: TeLLie on December 28, 2020, 05:07:17 pm
Hi Paul
No traps but also no sound on my W530 thats for the  uniaud32-linux-4.14.202-20201227.zip
But the uniaud32-linux-4.19.163-20201228.zip provide nice sound :)

Thankz for all
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on December 28, 2020, 06:34:21 pm
Hi Paul!

https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

No TRAP here, no sound either, debug log attached.

I did notice something though, not sure if this is of significance or not. Once booted, the numbering of controls (as it shows up in UniMix) is different from the current working driver (that being AOS 2.02.06).

I've captured the PMUnixMix debug console output for both the working and this latest drop in case the comparison may tell you something.

Thanks!
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 28, 2020, 09:10:04 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

hi, tried on on my comp

What hardware is this? I presume previous Uniaud builds worked?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 28, 2020, 09:13:38 pm
Hi Dariusz,
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

No TRAP here, no sound either, debug log attached.

I did notice something though, not sure if this is of significance or not. Once booted, the numbering of controls (as it shows up in UniMix) is different from the current working driver (that being AOS 2.02.06).

I've captured the PMUnixMix debug console output for both the working and this latest drop in case the comparison may tell you something.

Thanks!
-Dariusz

Thanks - I'll review later today. I don't *think* the reordering of controls is an issue - I've seen this before when bumping the version of the linux code. Remember, the code in the AOS driver is from ~2011 - the linux 4.19 code was released in October 2018 - so there's been 7 years of code development.

I'm also seeing no sound on a T60, so will do some investigations locally.

Cheers,
Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 28, 2020, 09:14:55 pm
Hi Roberto,

It has given me trap in debug and release. In ver 4.19.163

Does your system have a serial port? It would be interesting to see where in the boot cycle your system is trapping.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: OS4User on December 28, 2020, 09:36:22 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

hi, tried on on my comp

What hardware is this? I presume previous Uniaud builds worked?

CA0106/CA0111 [SB Live!/Audigy/X-Fi Series]       
I did not check previous    but  uniaud   usually works ok
Title: Re: Updated Uniaud32 build
Post by: roberto on December 28, 2020, 09:54:33 pm
Does your system have a serial port? It would be interesting to see where in the boot cycle your system is trapping.
I have no serial port.
Marking with REM the uniaud16.sys, I was able to start the version. 4.19.163, but no sound. Even so, only one start has been possible. Other attempts have failed with a trap. I attach the log - debug.
saludos
Title: Re: Updated Uniaud32 build
Post by: roberto on December 28, 2020, 10:29:04 pm
Unchecking the uniaud16.sys, and marking REM BASEDEV = HD4DISK.ADD (Memory disk) with REM, the system starts up without problems, but without sound.
Saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 05:52:11 am
New build:
Cleanup up diffs as much as reasonably possible to make future code maintenance easier
Updated rbtree  / regmap / regcache to the 4.19.163 code level

I don't expect any change compared to yesterday's build - but you never know :)

https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201229.zip
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 06:36:56 am
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201229b.zip - has a different implementation of krealloc() which allows me to use some unmodified code in pcm_lib.c which was trapping with my old krealloc() implementation.

TODO is to reduce logging in release builds.....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 11:55:19 am
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

hi, tried on on my comp

What hardware is this? I presume previous Uniaud builds worked?

CA0106/CA0111 [SB Live!/Audigy/X-Fi Series]       
I did not check previous    but  uniaud   usually works ok

Can you upload a copy of alsahlp$ ?
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 29, 2020, 12:27:07 pm
 Paul,

 Just tried the 1229b version of 4.19 and I get a TrapE in UNIAUD32 at boot. Also uploaded a trap dump into Dropbox. Let me know if I should also try the previous 1229 version...

Regards,
Title: Re: Updated Uniaud32 build
Post by: roberto on December 29, 2020, 03:57:35 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201229b.zip - has a different implementation of krealloc() which allows me to use some unmodified code in pcm_lib.c which was trapping with my old krealloc() implementation.

TODO is to reduce logging in release builds.....
This version has worked well for me!
But shortly after using it watching a youtube, and clicking on "search" in the seamonkey closed the browser with an ugly noise. I have to try it longer.
saludos
Title: Re: Updated Uniaud32 build
Post by: OS4User on December 29, 2020, 07:39:20 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

hi, tried on on my comp

What hardware is this? I presume previous Uniaud builds worked?

CA0106/CA0111 [SB Live!/Audigy/X-Fi Series]       
I did not check previous    but  uniaud   usually works ok

Can you upload a copy of alsahlp$ ?
Code: [Select]
pci_register_driver: query_device found 1108 1102:7 class=40100 checking intel8x0
pci_register_driver: query_device found 1108 1102:7 class=40100 checking via82xx
pci_register_driver: query_device found 1108 1102:7 class=40100 checking emu10k1
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cmipci
pci_register_driver: query_device found 1108 1102:7 class=40100 checking als4000
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cs4281
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cs46xx
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cs5535audio
pci_register_driver: query_device found 1108 1102:7 class=40100 checking es1938
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ens1370
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ens1370
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ymfpci
pci_register_driver: query_device found 1108 1102:7 class=40100 checking es1968
pci_register_driver: query_device found 1108 1102:7 class=40100 checking Maestro3
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ali5451
pci_register_driver: query_device found 1108 1102:7 class=40100 checking trident
pci_register_driver: query_device found 1108 1102:7 class=40100 checking nm256
pci_register_driver: query_device found 1108 1102:7 class=40100 checking fm801
pci_register_driver: query_device found 1108 1102:7 class=40100 checking atiixp
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ca0106
pci_register_driver: matched 0 1102:7/40100 with 1102:7/0 0 (ca0106)
request_irq: irq 23
<6>snd-ca0106: Model 100a Rev 00000000 Serial 100a1102

ca0106: probe for MIDI channel A ...
 done.

pci_register_driver: query_device found 1108 1102:7 class=40100 checking hda_intel
pci_register_driver: query_device found 1108 1102:7 class=40100 checking bt87x
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 09:00:07 pm
Hi Dave,

Just tried the 1229b version of 4.19 and I get a TrapE in UNIAUD32 at boot. Also uploaded a trap dump into Dropbox. Let me know if I should also try the previous 1229 version...

I'll look at the dump later today, please test with the original 1229 build - that will help confirm that it was the krealloc change which broke things...

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 09:02:10 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201228.zip - works in Virtualbox with ac97 and hda hardware

hi, tried on on my comp

What hardware is this? I presume previous Uniaud builds worked?

CA0106/CA0111 [SB Live!/Audigy/X-Fi Series]       
I did not check previous    but  uniaud   usually works ok

Can you upload a copy of alsahlp$ ?
Code: [Select]
pci_register_driver: query_device found 1108 1102:7 class=40100 checking intel8x0
pci_register_driver: query_device found 1108 1102:7 class=40100 checking via82xx
pci_register_driver: query_device found 1108 1102:7 class=40100 checking emu10k1
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cmipci
pci_register_driver: query_device found 1108 1102:7 class=40100 checking als4000
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cs4281
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cs46xx
pci_register_driver: query_device found 1108 1102:7 class=40100 checking cs5535audio
pci_register_driver: query_device found 1108 1102:7 class=40100 checking es1938
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ens1370
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ens1370
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ymfpci
pci_register_driver: query_device found 1108 1102:7 class=40100 checking es1968
pci_register_driver: query_device found 1108 1102:7 class=40100 checking Maestro3
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ali5451
pci_register_driver: query_device found 1108 1102:7 class=40100 checking trident
pci_register_driver: query_device found 1108 1102:7 class=40100 checking nm256
pci_register_driver: query_device found 1108 1102:7 class=40100 checking fm801
pci_register_driver: query_device found 1108 1102:7 class=40100 checking atiixp
pci_register_driver: query_device found 1108 1102:7 class=40100 checking ca0106
pci_register_driver: matched 0 1102:7/40100 with 1102:7/0 0 (ca0106)
request_irq: irq 23
<6>snd-ca0106: Model 100a Rev 00000000 Serial 100a1102

ca0106: probe for MIDI channel A ...
 done.

pci_register_driver: query_device found 1108 1102:7 class=40100 checking hda_intel
pci_register_driver: query_device found 1108 1102:7 class=40100 checking bt87x

Thanks -it would be helpful if you could try some of my earlier builds to identify when the ca0106 broke.. Meanwhile, I'll do some investigation..
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on December 29, 2020, 09:03:35 pm
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201229b.zip ....

Hi.

I tried uniaud32-linux-4.19.163-20201229b.zip  on VirtualBox on W10 host . Thinkpad Yoga14.
No traps. Sound and volume control working. For the moment it is working fine

Regards
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on December 29, 2020, 09:20:29 pm
Hi, as it need more time to get the T480 to run again I meanwhile tested with my S1 Yoga 12 from Lenovo:

Version 20201228 works like a charm!

Hardware:
Lenovo Thinkpad S1 Yoga 12
Intel core i7 5th Generation
Soundcardinfo:
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf1204000 irq 22
  mixer: Conexant CX20751/2
  componenets: HDA:14f1510f,17aa503e,00100100

Software:
ArcaOS 5.06

Great work, Paul!!

I guess the "b" with the attached files is not right, I downloaded it from the link from OS4Users latest post.

Will try the 20121229b now and report. More information needed?

Hope this helps!

Sigurd

EDIT: 20201229b crashes with trap, See attachment please.

EDIT: 20201229 crashes with trap as well, See attachment please.

Switch back to 20201228.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 29, 2020, 10:32:10 pm
 Paul,

  Version 4.19.163-20201229 also boots with a TrapE in UNIAUD32 here. I sent you the trap dump for that one too...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 10:56:58 pm
Hi Sigurd!

EDIT: 20201229b crashes with trap, See attachment please.

EDIT: 20201229 crashes with trap as well, See attachment please.

Switch back to 20201228.

From the 20201228 build, are you able to load theseus, General -> System -> Modules - find the entry for uniaud32.sys and double click the value in the left held column.

Another window will open - post the output below  'Output Table' which should be something like the below.
Code: [Select]
Object Table:
Obj#   address      size   hob   sel  page map  map size     flags / flags...
0001  FA875000  0000048B  0000  2A88  00000001  00000001  00001023 / readable writeable shared 16-bit
0002  FA874000  000006DB  0000  2A90  00000002  00000001  00001025 / readable executable shared 16-bit
0003  F26E4000  000959F8  0000  2A98  00000003  00000096  00002025 / readable executable shared 32-bit
0004  FA862000  00000007  0000  2AA0  00000099  00000001  00002025 / readable executable shared 32-bit
0005  F2682000  00061FB0  0000  2AA8  0000009A  00000052  00002023 / readable writeable shared 32-bit

This helps understand where the driver is loaded in memory and relates the trap screen output back to uniaud32.map

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 11:01:06 pm
Hi Dave,

  Version 4.19.163-20201229 also boots with a TrapE in UNIAUD32 here. I sent you the trap dump for that one too...

OK damn! Well at least it wasn't the changes to krealloc. The 1229b trap is confusing me as the address of the trap isn't being found - ie I can't find which function is trapping - most likely as it's in a static function - but normally I can at least see the source file the trap is in.

I'm going to do a build with regmap & friends backlevelled to the code from the 1228 build - that's my best guess right now as to the cause.
EDIT: https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201230.zip

Unfortunately, I can't reproduce the trap on my T60 or X250

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 29, 2020, 11:40:56 pm
 Paul,

  Got a TrapE with 1230 too... trap dump uploaded to dropbox. Maybe these problems are specific to Realtek chips...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 29, 2020, 11:58:13 pm
Hi Dave,

  Got a TrapE with 1230 too... trap dump uploaded to dropbox. Maybe these problems are specific to Realtek chips...

It's weird - my X250 uses a Realtek chip too, so it's not that simple...

Will look at your dump shortly, but any chance of a serial log of 1228 vs any of the trapping builds?

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 30, 2020, 07:03:49 am
For those that have a working 1228 build, and a trapping 1229 build... please try the following:
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201230c.zip - alsa-kernel directory backlevelled to the same code as 1228
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201230d.zip - alsa-kernel, include and lib32 directories backlevelled to the same code as 1228

Hopefully this helps narrow things down....
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on December 30, 2020, 12:48:39 pm
Hi Paul,

20201230c works like a charm
20201230d works like a charm (attached files unimix card and unimix list)

Downloaded Theseus from hobbes (never used this program before, afair  ;D ) and attach as you guided with 20201228

All the best!!

Sigurd

Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 30, 2020, 01:08:16 pm
 Paul!

  1230c works great - sound is normal, no hesitations, and PMUnimix shows all 9 controls as before (one boot only).

  1230d gives a Trap E in UNIAUD32.

 Gotta head to work right now, but I'll provide those serial logs when I get back later today...

Regards,
Title: Re: Updated Uniaud32 buildno sound still
Post by: Eugene Tucker on December 30, 2020, 02:51:13 pm
C works like a charm. There are no issues with PMunimix.
D there is nno sound.
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on December 30, 2020, 06:35:00 pm
Paul, everyone...

For those that have a working 1228 build, and a trapping 1229 build... please try the following:
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201230c.zip
https://smedley.id.au/tmp/uniaud32-linux-4.19.163-20201230d.zip

Alright, here are the results:

1) uniaud32-linux-4.19.163-20201230c
PASS - sounds is ON, no issues I can detect, see attached log capture

2) uniaud32-linux-4.19.163-20201230d
FAIL - no sound on bootup, see attached log capture

Further on, not sure if this is going to help or not (I suspect you are probably already using something similar to track so many configurations), but since you've put out so many releases I went back to P1 of this THREAD and built a quick matrix of the various test releases I tested, along with the TRAP (Y/N) and remarks. That includes the hardware snapshot as well ('unimix -card').

Thanks as always!

-Dariusz
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 31, 2020, 12:32:40 am
 Paul,

  Attached are various 4.19 debug version serial logs. All booted with sound except 1229 (boots, no sound) and 1230d (Trap E). This is somewhat different from the release versions...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 06:38:27 am
I'll respond to some of the other posts later, but I figured out why 5.0 was trapping...

VERY lightly tested https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
& https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
& https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip
Title: Re: Updated Uniaud32 build
Post by: OS4User on December 31, 2020, 09:29:56 am
I'll respond to some of the other posts later, but I figured out why 5.0 was trapping...

VERY lightly tested https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
& https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
& https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

Hi, Paul
5.1.21 and 5.2.21  work ok on my  CA0106    (5.0.21  did not test  yet)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 09:53:20 am
5.3 is trapping, will investigate next year (aka tomorrow)

Goal is to get to 5.4.x which is the current Long Term Support kernel then try and stabilise things and upload sources to netlabs svn.
Title: Re: Updated Uniaud32 build
Post by: Lars on December 31, 2020, 11:17:21 am
Hi Paul,
can you produce a MAP file during the build process and ideally also intermixed assembler/c source files (MS called these .cod files)? Watcom solution to the latter is wdis.exe with the /s switch, I don't know if gcc has something comparable.
That will greatly help in identifying the exact trap location in the c source code and also typically the root cause of the trap.
Title: Re: Updated Uniaud32 build
Post by: roberto on December 31, 2020, 01:03:01 pm
5.0.21 and 5.1.21 start but no sound
5.2.21 no start, no trap.
saludos
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 31, 2020, 03:21:57 pm
 Happy New Year!

 All 3 of the 5.x versions could boot, but 5.1 gave a TrapE 1 time (couldn't get it to repeat, but have the dump if you want it). No sound most of the time with any of them, except 1 time 5.2 booted with loud sound and system sounds occurring late. In every case PMUnimix showed only 3 controls.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on December 31, 2020, 05:30:12 pm
Ohh.. version 5... we are getting close.

https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

I tried "5.2.21-20201231" on the ArcaOS 5.0.6 VM on W10 host on the TP Yoga 14.
No Trap. It produce sound, but I lost control of the master volume tool.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 09:16:58 pm
I'll respond to some of the other posts later, but I figured out why 5.0 was trapping...

VERY lightly tested https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
& https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
& https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

Hi, Paul
5.1.21 and 5.2.21  work ok on my  CA0106    (5.0.21  did not test  yet)

Good news! No need to test 5.0.21 given the newer ones work.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 09:18:36 pm
Hi LArs,

can you produce a MAP file during the build process and ideally also intermixed assembler/c source files (MS called these .cod files)? Watcom solution to the latter is wdis.exe with the /s switch, I don't know if gcc has something comparable.
That will greatly help in identifying the exact trap location in the c source code and also typically the root cause of the trap.

Uniaud is built with Watcom - and map files are included in the distros since a week or so ago. I've never had luck with the wdis /s switch - I must try get that working....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 09:19:17 pm
Hi Roberto,

5.0.21 and 5.1.21 start but no sound
5.2.21 no start, no trap.

So 5.2.21 hangs at startup?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 09:20:21 pm
Hi Dave,
All 3 of the 5.x versions could boot, but 5.1 gave a TrapE 1 time (couldn't get it to repeat, but have the dump if you want it). No sound most of the time with any of them, except 1 time 5.2 booted with loud sound and system sounds occurring late. In every case PMUnimix showed only 3 controls.

No need for a dump from 5.1- the focus is on getting to 5.4. The PMUnimix case is interesting - I'll investigate that when I get a moment.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 09:21:54 pm
Hi Martin,

Ohh.. version 5... we are getting close.

https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

I tried "5.2.21-20201231" on the ArcaOS 5.0.6 VM on W10 host on the TP Yoga 14.
No Trap. It produce sound, but I lost control of the master volume tool.

My thinking is that this:
Code: [Select]
Control ID: 1
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: []
Index: 0
Element type: invalid
Count of values: 0
Value:

Is the cause of the Unimixpm issues.....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 31, 2020, 10:23:15 pm
HAPPY NEW YEAR

I have no idea what I did wrong last night, but after running a diff of the linux code between 5.2.x and 5.3.x, and seeing how trivial the changes were, I re-applied my diffs and rebuilt.

I must have been getting tired last night, as now 5.3.19 produces sound for me in Virtualbox with both ac97 and hda hardware!

https://smedley.id.au/tmp/uniaud32-linux-5.3.18-20210101.zip

Will try for 5.4.x after a beach walk with the dogs!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 31, 2020, 10:57:37 pm
 Happy New Year, Paul!

 7 more hours to go here...

 Version 5.3.18 boots, both release and debug versions. Debug has sound, but very loud. Release has no sound. Both only show 3 controls in PMUnimix. Also can't get a complete ALSAHLP$ log - maybe it wraps? (every time, I try immediately after the desktop appears). Example attached...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 02:46:04 am
Time for a glass of wine :)

https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101.zip - produces sound with ac97 and hda hardware in Virtualbox.

Will try on real hardware this afternoon.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 01, 2021, 03:37:00 am
Happy New Year.  I have 2.5 hours left for midnight here.
Time for a glass of wine :)

https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101.zip - produces sound with ac97 and hda hardware in Virtualbox.

Will try on real hardware this afternoon.

I tried "uniaud32-linux-5.4.86-20210101.zip " on the ArcaOS 5.0.6 VM on W10 host on the TP Yoga 14.
No Trap. It produce sound, but I lost control of the master volume tool.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 03:47:26 am
Hi Martin,

Thanks for testing. Next steps will be to upload current sources to Netlabs, then try and fix some of the reported issues.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 05:48:25 am
Source is now uploaded to netlabs: https://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-next

There is definitely a problem with HDA hardware and the mixer list - causing issues with pmunimix.

For example - unimix -names shows:
Code: [Select]
ID: 1, name: [Front Playback Volume]
ID: 2, name: [Front Playback Switch]
ID: 3, name: [Surround Playback Volume]
ID: 4, name: [Surround Playback Switch]
ID: 5, name: [Center Playback Volume]

But unimix -list shows the following for ID: 1:
Code: [Select]
Control ID: 1
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: []
Index: 0
Element type: invalid
Count of values: 0
Value:

This will be the priority issue to resolve .... I've created https://trac.netlabs.org/uniaud/ticket/247 for this issue.
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 01, 2021, 06:09:56 am
uniaud32-linux-5.4.86-20210101.zip  working well (including unimix) with

Sound Blaster SB0100 PCI 2001 SB Live! 5.1
=============================
Card info for adapter 0:
  num: 0
  id: Live
  driver: EMU10K1
  name: SB Live! 5.1
  longname: SB Live! 5.1 (rev.7, serial:0x80641102) at 0xd880, irq 22
  mixer: eMicro EM28028
  componenets: AC97a:454d4328

I thought I had a problem at the start but it was just me adjusting the wrong mixing console balance (real world hardware).
My workshop is rockin :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 06:59:33 am
Thanks Ian - my assumption is that mixer controls are only broken on HDA hardware - ac97 (and now your) hardware is OK.

I have a November build of uniaud32.sys installed on my dev system - which doesn't produce sound, but does seem to have working mixer controls - I wonder what changed?

Edit: https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101b.zip *might* fix the mixer controls.. at least here unimix -list shows sensible output, and pmunimix seems to work.

If others can confirm, I'll commit the fix.
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 01, 2021, 08:02:28 am
uniaud32-linux-5.4.86-20210101.zip
FYI - UNIAUD loads but no sound and nothing can attach to the driver (as in pm123 or other audio apps)

Sound Blaster SB1040 PCIe 2007
==============================
[C:\mmos2]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0    <====
Card info for adapter 0:
  num: 0
  id: Creative
  driver: HDA-Intel
  name: HDA Creative
  longname: HDA Creative at 0xfeafc000 irq 16
  mixer:                                                                                        <====
  componenets: HDA:1102000a,11021006,00100000


[D:\fil\SB\SB0100]unimix -list
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Control list for adapter 0

[D:\fil\SB\SB0100]unimix -id1 -cnt0 -val15 -cnt1 -val16 -list
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Setting Adapter: 0, ID:1, cnt:0, val:15...Failed
Setting Adapter: 0, ID:1, cnt:1, val:16...Failed
Control list for adapter 0

Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 01, 2021, 08:12:43 am
uniaud32-linux-5.4.86-20210101.zip    Works well

Sound Blaster SB0610 PCI 2005  SB Audigy 4
=============================
[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Audigy2
  driver: Audigy2
  name: SB Audigy 4 [SB0610]
  longname: SB Audigy 4 [SB0610] (rev.0, serial:0x10211102) at 0xdc00, irq 22
  mixer: SigmaTel STAC9750,51
  componenets: AC97a:83847650
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 09:37:20 am
Edit: https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101b.zip *might* fix the mixer controls.. at least here unimix -list shows sensible output, and pmunimix seems to work.

If others can confirm, I'll commit the fix.

FWIW - pmunimix works correctly on T60 (Analog Devices HDA), X250 (Realtek HDA) & Virtualbox (Sigmatel HDA) with the 20210101b build.

YMMV
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 01, 2021, 11:19:15 am
uniaud32-linux-5.4.86-20210101b.zip   no work :(
Sound Blaster SB1040 PCIe 2007
==============================
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Card info for adapter 0:
  num: 0
  id: Creative
  driver: HDA-Intel
  name: HDA Creative
  longname: HDA Creative at 0xfeafc000 irq 16
  mixer:
  componenets: HDA:1102000a,11021006,00100000

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 11:48:24 am
I'll check tomorrow, but I don't think the code for that Creative HDA is enabled. It might be easy to enable it....

Edit: I think we need to add http://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-next/alsa-kernel/pci/hda/patch_ca0110.c in order to support that chip.
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 01, 2021, 12:05:00 pm
https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101.zip - produces sound with ac97 and

it works ok on my  CA0106
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 01, 2021, 03:13:04 pm
 Paul,

 A very mixed bag with the 5.4 versions:

0101 release - boots, no sound, only 3 controls in PMUnimix, hesitations where system sounds would occur
0101 debug - boots, loud sound, only 4 controls in PMUnimix, hesitations before system sounds occur
0101b release - TrapE in UNIAUD32, I uploaded a dump to dropbox
0101b debug - boots, proper sound, PMUnimix shows 9 controls, hesitations before system sounds occur

 Tried these all both warm and cold boot and they were consistent. I'll attach alsahlp$ logs for the 3 that booted (all taken immediately after booting to desktop).

Regards,
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 01, 2021, 03:17:38 pm
Hi Martin,

Thanks for testing. Next steps will be to upload current sources to Netlabs, then try and fix some of the reported issues.

Cheers,

Paul

Paul, as an alternative, Adrian (Netlabs) gave me administrator access to Netlabs github account: https://github.com/netlabsorg
If you prefer a git this can be al alternative, otherwise everything is ok at Netlabs trak.

Regards
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 01, 2021, 03:23:32 pm
Hi Paul,

Quote
I'll check tomorrow, but I don't think the code for that Creative HDA is enabled. It might be easy to enable it....

Not a priority unless others want it, I just had that PCIe card in the spares box but if it's easy, thanks :)
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on January 01, 2021, 03:53:22 pm
I would like it. I have a machine that I will switch over to ArcaOS in the near future and retire the another. And it has a built is Sound Blaster Recon.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 01, 2021, 04:50:30 pm

Time for a glass of wine :)

https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101.zip - produces sound with ac97 and hda hardware in Virtualbox.

Will try on real hardware this afternoon.

I tried "uniaud32-linux-5.4.86-20210101.zip " on ArcaOS 5.0.6 on the Thinkpad L420.
No Trap. But it does not produce any sounds.

Regards
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 01, 2021, 04:56:30 pm
 Gave the 5.4 0101b version a try on my laptop:

Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.06
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xd4730000 irq 22
  mixer: Intel ID 2806
  componenets: HDA:111d7605,103c17b7,00100107 HDA:11c11040,103c3066,00100200 HDA:80862806,80860101,00100000

  Get a TrapE in UNIAUD32 with the release version (no dump partition on this), but the debug version seems to work correctly. Only thing is the card does not appear in the hardware manager, and there is an error in alsahlp$... attached.

Regards,
Title: Re: Updated Uniaud32 build
Post by: roberto on January 01, 2021, 06:46:18 pm
This version  https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210101b.zip , in debug,  is working especially well for me, very well.
 But I see discrepancies with what David M. shows us.
 [C:\MMOS2]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05                          ==>2.02.05  David 06 ???
Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 1
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xd2520000 irq 22
  mixer: IDT 92HD81B1X5
  componenets: HDA:111d7605,103c1630,00100105 HDA:11c11040,103c3066,00100200

 I notice a faster functioning of the browser, with many more websites open.
saludos
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 01, 2021, 07:03:41 pm
 Hi roberto,

 I captured that unimix -card with the original version before applying Paul's new version. Wanted to make sure to show the correct info. Sorry to be confusing...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 09:23:32 pm
Hi Dave!
A very mixed bag with the 5.4 versions:

0101 release - boots, no sound, only 3 controls in PMUnimix, hesitations where system sounds would occur
0101 debug - boots, loud sound, only 4 controls in PMUnimix, hesitations before system sounds occur
0101b release - TrapE in UNIAUD32, I uploaded a dump to dropbox
0101b debug - boots, proper sound, PMUnimix shows 9 controls, hesitations before system sounds occur

 Tried these all both warm and cold boot and they were consistent. I'll attach alsahlp$ logs for the 3 that booted (all taken immediately after booting to desktop).
Thanks for the detailed report - this is where I admit that all of my testing has been with the DEBUG build and I've been assuming the only differences is the compiler flags.

Confirming that 0101b release - also gives me a TrapE in UNIAUD32 in Virtualbox, which is a good thing as at least I can reproduce it.

Also good to see that 0101b debug fixes the mixers for you - which confirms we need to fix krealloc.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 09:25:09 pm
Hi Dave,

Just tried the 1229b version of 4.19 and I get a TrapE in UNIAUD32 at boot. Also uploaded a trap dump into Dropbox. Let me know if I should also try the previous 1229 version...

Do you recall if this was with the DEBUG or RELEASE driver? If it was the RELEASE, please test with the DEBUG.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 01, 2021, 09:40:30 pm
 Hi Paul,

  That was with the release version - what I typically try first. Just tried the debug version and that one works perfectly - 9 controls in PMUnimix, no hesitations, and shows up in the hardware manager.

  Guess i didn't try it then because of the flurry of releases at that time. I'll try to be more rigorous about testing in future. Thanks for your Olympian efforts at modernizing UniAud! Alsahlp$ attached...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 09:44:26 pm
Hi Dave,

  That was with the release version - what I typically try first. Just tried the debug version and that one works perfectly - 9 controls in PMUnimix, no hesitations, and shows up in the hardware manager.

  Guess i didn't try it then because of the flurry of releases at that time. I'll try to be more rigorous about testing in future. Thanks for your Olympian efforts at modernizing UniAud! Alsahlp$ attached...

Good news - it means my other code cleanups/simplifications weren't the cause of the trap - it was 'just' the change in krealloc - which fixed one issue and broke another :P
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 01, 2021, 11:34:01 pm
Hey Ian,

Quote
I'll check tomorrow, but I don't think the code for that Creative HDA is enabled. It might be easy to enable it....

Not a priority unless others want it, I just had that PCIe card in the spares box but if it's easy, thanks :)

I think it was easy :) https://smedley.id.au/tmp//uniaud32-linux-5.4.86-20210102-debug.zip

Note: this is otherwise identical to 0101b - I've only included the debug build as the release build is trapping.

I'll remove some of the debug messages in the next build that will help quiet things down a bit in the log, and may also reduce the delays...
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 02, 2021, 12:23:51 am
FYI...

Paul's link below has an extra '/tmp/' in the URL, so in case you're not able to fix it up on your own here is the correct URL => https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip (https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip)

...I think it was easy :) https://smedley.id.au/tmp//tmp/uniaud32-linux-5.4.86-20210102-debug.zip...

Paul,
Here are my results for the last three releases:

1) uniaud32-linux-5.3.18-20210101.zip
FAIL - HARD STOP during boot, Power OFF/ON required

2) uniaud32-linux-5.4.86-20210101b.zip   
FAIL - TRAP 000e

3) uniaud32-linux-5.4.86-20210102-debug.zip
FAIL - HARD STOP during boot, Power OFF/ON required

The HARD STOP during boot behaviour for #1 and #3 above is something new...haven't see this before (or at least I do not remember seeing it), not sure if maybe additional DEBUG stuff caused some kind of memory issue (3072 here), although not likely since I tried both the RELEASE and DEBUG builds of uniaud32-linux-5.3.18-20210101.zip with the same results for both.

Happy New Year to you too!
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 02, 2021, 12:45:12 am
FYI...

Paul's link below has an extra '/tmp/' in the URL, so in case you're not able to fix it up on your own here is the correct URL => https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip (https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip)

...I think it was easy :) https://smedley.id.au/tmp//tmp/uniaud32-linux-5.4.86-20210102-debug.zip...

Oops - fixed :/
Title: Re: Updated Uniaud32 build
Post by: David Graser on January 02, 2021, 12:45:33 am
FYI...

Paul's link below has an extra '/tmp/' in the URL, so in case you're not able to fix it up on your own here is the correct URL => https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip (https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip)

...I think it was easy :) https://smedley.id.au/tmp//tmp/uniaud32-linux-5.4.86-20210102-debug.zip...

Okay, I found what the problem is.  This driver works good with the os/2 kernel.  I does NOT work with the os/4 kernel. 

With the ArcaOs uniaud32 driver, I have sound.

With all these test drivers I have tested, I lose my sound.

Intel Corporation
7 series/C216 Chipset Family High Definition Audio Controller.  There is no trap, only the loss of sound.
Title: Re: Updated Uniaud32 build
Post by: David Graser on January 02, 2021, 01:01:05 am
FYI...

Paul's link below has an extra '/tmp/' in the URL, so in case you're not able to fix it up on your own here is the correct URL => https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip (https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210102-debug.zip)

...I think it was easy :) https://smedley.id.au/tmp//tmp/uniaud32-linux-5.4.86-20210102-debug.zip...
Found the problem.  Driver works great with the OS/2 kernel.  It does not work with OS/4 kernel.

With the ArcaOs uniaud32 driver, I have sound.

With all these test drivers I have tested, I lose my sound.

Intel Corporation
7 series/C216 Chipset Family High Definition Audio Controller.  There is no trap, only the loss of sound.


Okay, I found what the problem is.  This driver works good with the os/2 kernel.  I does NOT work with the os/4 kernel. 
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 02, 2021, 04:39:30 am
Hi Paul,

I think it was easy :) https://smedley.id.au/tmp//uniaud32-linux-5.4.86-20210102-debug.zip

Works very well thanks :)

Sound Blaster SB1040 PCIe 2007
uniaud32-linux-5.4.86-20210102-debug.zip
========================================
[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Creative
  driver: HDA-Intel
  name: HDA Creative
  longname: HDA Creative at 0xfeafc000 irq 16
  mixer: Creative CA0110-IBG
  componenets: HDA:1102000a,11021006,00100000

Currently playing Uriah Heap.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 02, 2021, 06:50:45 am
Hi Ian,

I think it was easy :) https://smedley.id.au/tmp//uniaud32-linux-5.4.86-20210102-debug.zip

Works very well thanks :)

Sound Blaster SB1040 PCIe 2007
uniaud32-linux-5.4.86-20210102-debug.zip
========================================
[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Creative
  driver: HDA-Intel
  name: HDA Creative
  longname: HDA Creative at 0xfeafc000 irq 16
  mixer: Creative CA0110-IBG
  componenets: HDA:1102000a,11021006,00100000

Currently playing Uriah Heap.

Good to hear! (pun intended!)

Cheers,


Paul
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 02, 2021, 09:34:32 am
Okay, I found what the problem is.  This driver works good with the os/2 kernel.  I does NOT work with the os/4 kernel.

Just tested   5.4.86-20210102-debug on my CA0106 (of course  on OS/4)   -  works ok

pls show logs on OS/4  and  ory kernel
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on January 02, 2021, 04:39:05 pm
Happy new year! Tested the 5.4.86-20210102-debug on my S1 Yoga 12. Works well.

But - as with all driver files before, only three controls are enabled, I guess it should be more of them.

Attach a picture and the list file.

Well done!

Sigurd
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 02, 2021, 06:35:35 pm
 Hi Paul,

  Like you say, 5.4.86-20210102 works identically to 0101b-debug here: boots, proper sound, 9 controls in PMUnimix, but hesitates before playing system sounds.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on January 02, 2021, 07:36:10 pm
Testing 5.4.86-20210102-debug here, worked fine, though loud, using PMUniMix (has 5 controls, which seems about right) to turn down the volume results in the system sounds that often start loud then adjust to the correct volume while playing.
Code: [Select]
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xe1b60000 irq 22
  mixer: Realtek ALC269VB
  componenets: HDA:10ec0269,102804ad,00100100 HDA:80862805,80860101,00100000
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 02, 2021, 09:04:44 pm
Hi Sigurd,

Happy new year! Tested the 5.4.86-20210102-debug on my S1 Yoga 12. Works well.

But - as with all driver files before, only three controls are enabled, I guess it should be more of them.

Attach a picture and the list file.

Well done!

What does unimix -list show for controls? I have no idea how pmunimix 'short lists' the output from unimix to determine what it shows. Perhaps it's logic needs a tuneup?

Cheers,

Paul


Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 02, 2021, 10:57:03 pm
Hi All,

I have a rather large diff ready to commit - before I do so, I just want to confirm there are no significant regressions vs the 20210102 builds.

This diff cleans up a lot of the alsa-kernel code which should ease future patching against updated linux code. It also updates some libraries from 4.14.202 level to 4.19.163 level - TODO is to update them to 5.4.86 levels, but that can wait for a rainy day.

Debug only build until the traps with the release code are fixed.

https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210103-debug.zip

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on January 02, 2021, 11:29:46 pm
Hi Paul,

20210103 debug runs without problems.

I attach the unimix -list and unimix - card files.

Still 3 Entries in PM Uniaud as before.

All the best!

Sigurd

Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 03, 2021, 12:03:56 am
 Hi Paul,

 This 0103 version works perfectly on my desktop machine (HDA: ALC887-VD): boots, proper sound, 9 controls in PMUnimix, no hesitations, appears in the hardware manager. I'll keep using this one and let you know if anything comes up - thanks again! Hope to soon see your bust in the OS/2 hall of fame!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on January 03, 2021, 12:27:59 am
For the notes:

Sound in WINOS2 does work on my S1 Yoga 12 with the combination of Pauls 20210103 and the winos2-audio-19980501 driver from hobbes.  ;)

You can grab the Beta Driver from here: https://hobbes.nmsu.edu/download/pub/windows/WinOS2_Audio_19980501.zip (https://hobbes.nmsu.edu/download/pub/windows/WinOS2_Audio_19980501.zip)

Please follow the included readme to install.

Nice to have sound in WINOS2 as well  :)
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 03, 2021, 12:29:33 am
https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210103-debug.zip

it works ok here
Title: Re: Updated Uniaud32 build
Post by: David Graser on January 03, 2021, 02:18:09 am
https://smedley.id.au/tmp/uniaud32-linux-5.4.86-20210103-debug.zip

it works ok here
Found my problem.  For whatever reason, I had the uniaudio32.sys and uniaudio16.sys statements remmed out in the config.os4 file.   One of the earlier test drivers kept trapping my system and this was why the two statements were remmed out.   The latest driver works great.
Title: Re: Updated Uniaud32 build
Post by: David Graser on January 03, 2021, 03:50:03 am
For the notes:

Sound in WINOS2 does work on my S1 Yoga 12 with the combination of Pauls 20210103 and the winos2-audio-19980501 driver from hobbes.  ;)

You can grab the Beta Driver from here: https://hobbes.nmsu.edu/download/pub/windows/WinOS2_Audio_19980501.zip (https://hobbes.nmsu.edu/download/pub/windows/WinOS2_Audio_19980501.zip)

Please follow the included readme to install.

Nice to have sound in WINOS2 as well  :)

Works great here also.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 03, 2021, 07:17:02 am
Thanks all - will check in the latest changes shortly, then set about trying to fix the release builds.
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 03, 2021, 08:22:16 am
uniaud32-linux-5.4.86-20210103-debug.zip
   Sound Blaster SB0100 PCI 2001 SB Live! 5.1
   Sound Blaster SB0610 PCI 2005  SB Audigy 4
   Sound Blaster SB1040 PCIe 2007
All work well thanks :)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 03, 2021, 08:27:34 am
Hi Dave,

This 0103 version works perfectly on my desktop machine (HDA: ALC887-VD): boots, proper sound, 9 controls in PMUnimix, no hesitations, appears in the hardware manager. I'll keep using this one and let you know if anything comes up - thanks again! Hope to soon see your bust in the OS/2 hall of fame!

Glad it works for you! Your comment about the hall of fame gave me a giggle - to be honest - I'm just glad people appreciate my work. The comments (and of course the occasional sponsorship that helps pay for new gadgets) are truly appreciated.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 03, 2021, 10:44:17 am
Got a TrapE

hi, could you show me this trap screen on  OS/4 kernel (it does not trap on my comp)
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 03, 2021, 07:28:40 pm
 Paul,

  The 0103-debug version seems to work well on my laptop, but the mixer is identified as 'IDT 92HD81B1X5'. With the 'official' AOS version of UniAud, it is identified as 'Intel 2806'. Not sure if this even matters, or maybe they are the same thing? Get more controls with PMUnimix and 0103 too. Alsahlp$ attached...

Regards,
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 03, 2021, 07:30:32 pm
Got a TrapE

hi, could you show me this trap screen on  OS/4 kernel (it does not trap on my comp)

  No problems here with OS/4 kernel (at least not any different than with OS/2 kernel). Maybe you are thinking of David Graser (who reported a difference between the 2 kernels)?

Regards,
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 03, 2021, 07:51:47 pm
Got a TrapE

hi, could you show me this trap screen on  OS/4 kernel (it does not trap on my comp)

  No problems here with OS/4 kernel (at least not any different than with OS/2 kernel). Maybe you are thinking of David Graser (who reported a difference between the 2 kernels)?

Regards,

You reported about some trap earlier   I would like  to see that   trap screen on OS/4.
David Graser has already solved his issue https://www.os2world.com/forum/index.php/topic,2436.msg29615.html#msg29615
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 03, 2021, 08:57:14 pm
 OK, here you go... version 4.19.163-20201230.

 Don't use OS/4 much on this machine, mostly on the laptop, but Paul's latest version seems to work fine with OS/4...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 03, 2021, 09:01:17 pm
Hi Dave,

  The 0103-debug version seems to work well on my laptop, but the mixer is identified as 'IDT 92HD81B1X5'. With the 'official' AOS version of UniAud, it is identified as 'Intel 2806'. Not sure if this even matters, or maybe they are the same thing? Get more controls with PMUnimix and 0103 too. Alsahlp$ attached...

There's nothing I've done that would have changed this. Given the 'official' AOS version of Uniaud is based on ALSA 1.0.24 - released in February 2011, and my latest builds are based on the linux 5.4 kernel - originally released in November 2019 - I'll put the change in mixer name down to the ~9 years of ALSA development :)

Thanks for all your testing!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 03, 2021, 10:22:21 pm
OK, here you go... version 4.19.163-20201230.

tnx a lot
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 03, 2021, 11:07:23 pm
OK I read on the weekend that the new LTS kernel will be 5.10.x...

Therefore my OCD wants us to be on the current LTS kernel....

Here's step 1 on the pathway from 5.4.x to 5.10.x

Debug only at this stage.... https://smedley.id.au/tmp/uniaud32-linux-5.5.19-20210104-debug.zip

Seems to work the same as 5.4.x in my limited testing with HDA in Virtualbox
Title: Re: Updated Uniaud32 build
Post by: David Graser on January 04, 2021, 01:13:20 am
Both ArcaOs and Windows sound continues to work as it should with this latest release.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 04, 2021, 03:45:08 am
Both ArcaOs and Windows sound continues to work as it should with this latest release.

Thanks Dave - https://smedley.id.au/tmp/uniaud32-linux-5.6.19-20210104-debug.zip is the next one....
&
https://smedley.id.au/tmp/uniaud32-linux-5.7.19-20210104-debug.zip
&
https://smedley.id.au/tmp/uniaud32-linux-5.8.18-20210104-debug.zip
&
https://smedley.id.au/tmp/uniaud32-linux-5.9.16-20210104-debug.zip
&
https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210104-debug.zip

No need to test all builds - I sugggest starting with the latest - if it works - no need to test the previous ones. If it doesn't work, finding the newest build that works is helpful.

These have only been confirmed to boot with HDA hardware under Virtualbox. Virtualbox is not giving me sound right now (even under my normal VM with a working driver).

I'll try test on real hardware shortly....
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 04, 2021, 04:40:31 am
https://smedley.id.au/tmp/uniaud32-linux-5.8.18-20210104-debug.zip

I just did a quick test with "uniaud32-linux-5.8.18-20210104-debug.zip" on the ArcaOS 5.0.6 VM on W10 host on the TP Yoga 14.
No Trap. It produce sound, and I have control of the volume with the master volume tool. Working good.

Regards
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 04, 2021, 08:15:05 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210104-debug.zip

it works ok for me
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 04, 2021, 09:05:00 am
uniaud32-linux-5.10.4-20210104-debug.zip
Works well for all my Creative boards +
  ASRock G41C-GS Onboard Intel audio
  id: Intel
  driver: HDA-Intel
  name: HDA Intel
  longname: HDA Intel at 0xfeafc000 irq 16
  mixer: VIA VT1705
  componenets: HDA:11064397,18490397,00100000
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 04, 2021, 12:15:40 pm
 Paul,

 5.10 works fine with both my desktop and laptop - thanks!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on January 04, 2021, 12:41:26 pm
Hi Paul,

5.10.4. works very well here with the Yoga.

Still 3 controls, not more.

All the best!
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on January 04, 2021, 01:32:22 pm
Boots with no traps but no sound.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 04, 2021, 09:01:14 pm
Hi Eugene,

Boots with no traps but no sound.

Have you checked mixer controls with pmunimix?

Logs please - 'copy alsahlp$ alsa.log' and 'unimix -list > unimixlist.txt'

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 04, 2021, 11:13:03 pm
Hi Paul, everyone...

...https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210104-debug.zip

No need to test all builds - I sugggest starting with the latest - if it works - no need to test the previous ones. If it doesn't work, finding the newest build that works is helpful...

This last one also produces the same result as several of the most recent releases, namely: hard lock during boot, but no TRAP.

I took a stab at playing with numerous config options, just in case something about my system is off (given that no one else is reporting anything similar to this).

I did the following:

1) moved from os2krnl 202 to 201
2) changed VIRTUALADDRESSLIMIT=3072 to 2048
3) changes JFS cache from 256M to 128M

All of these produced the very same result.

I will try a couple more times with several other changes such as shutting off FAT32 IFS, disabling the VirtualBox driver, etc.

Thanks,
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 04, 2021, 11:46:56 pm
Hi Dariusz,


Hi Paul, everyone...

...https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210104-debug.zip

No need to test all builds - I sugggest starting with the latest - if it works - no need to test the previous ones. If it doesn't work, finding the newest build that works is helpful...

This last one also produces the same result as several of the most recent releases, namely: hard lock during boot, but no TRAP.

I took a stab at playing with numerous config options, just in case something about my system is off (given that no one else is reporting anything similar to this).

I did the following:

1) moved from os2krnl 202 to 201
2) changed VIRTUALADDRESSLIMIT=3072 to 2048
3) changes JFS cache from 256M to 128M

All of these produced the very same result.

I will try a couple more times with several other changes such as shutting off FAT32 IFS, disabling the VirtualBox driver, etc.

Weird.... do you access to a serial port and null modem cable at all? It would be interesting to see how far the boot gets.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 05, 2021, 02:22:58 am
Paul,

Hi Dariusz,

Weird.... do you access to a serial port and null modem cable at all? It would be interesting to see how far the boot gets.

Yup, I've got my APC UPS hooked up to it normally, but could troubleshoot this specific issue otherwise. The NULL cable is the bigger problem...got a few regular cables here, so maybe I could re-wire? I see plenty of on-line references that show the null vs normal pin mapping. Would this require a basic Rx=>Tx and Tx=>Rx conversion, or will this setup need any handshaking provisions?

Do we need to be running a debug kernel for this?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 05, 2021, 03:33:47 am
Hi Dariusz!

Weird.... do you access to a serial port and null modem cable at all? It would be interesting to see how far the boot gets.

Yup, I've got my APC UPS hooked up to it normally, but could troubleshoot this specific issue otherwise. The NULL cable is the bigger problem...got a few regular cables here, so maybe I could re-wire? I see plenty of on-line references that show the null vs normal pin mapping. Would this require a basic Rx=>Tx and Tx=>Rx conversion, or will this setup need any handshaking provisions?

Do we need to be running a debug kernel for this?

OK 1st things - from what I can see - you're the only user with an ATI SB - everyone else seems to have HDA Intel... so that's likely a different code path.

From what I can see - 4.19.x works for you, 5.3.x fails.  Are you able to narrow down which build broke things for you? this will help identify code changes which caused the break.

To be honest, I'm not 100% it needs to be a null-model serial cable - I just capture the serial port in virtualbox and it works. To send log messages to serial, add /P:1 for com1 or /P:2 for com2

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 05, 2021, 03:54:19 am
Paul,

OK 1st things - from what I can see - you're the only user with an ATI SB - everyone else seems to have HDA Intel... so that's likely a different code path.

From what I can see - 4.19.x works for you, 5.3.x fails.  Are you able to narrow down which build broke things for you? this will help identify code changes which caused the break.

Yup, the last good one was: uniaud32-linux-4.19.163-20201230c.zip, following that they were either a TRAP 000e, or the hard-lock. See the attached testing matrix.

For what it's worth, there were two other earlier releases that did actually show the same behaviour, they were:

1) uniaud32-linux-3.18.140-20200816.zip
2) uniaud32-linux-4.0.9-20200816.zip

To be honest, I'm not 100% it needs to be a null-model serial cable - I just capture the serial port in virtualbox and it works. To send log messages to serial, add /P:1 for com1 or /P:2 for com2

OK, may not be as hard then...let me see if I can just hook this thing up to one of my laptops, they are older machines so should have a serial port on one of them...

Thanks!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 05, 2021, 04:21:12 am
Hi - you're missing a few...

https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 05, 2021, 05:15:38 am
Hi - you're missing a few...

https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

Aha...you bet, testing right now...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 05, 2021, 05:19:45 am
Fixes for ksize() from Steven Levine - release builds work for me now....
https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210105.zip (corresponds to svn r631)

&
https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210105b.zip adds an improved krealloc() implementation to the above (corresponds to svn r632)

Both work for me in virtualbox.

Similar to yesterday, if the newer build works, no need to test the earlier one....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on January 05, 2021, 05:45:22 am
Release seems fine and fixes my volume issue
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 05, 2021, 05:48:39 am
Paul,

Hi - you're missing a few...

https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

Aha...you bet, testing right now...

Last one to successfully boot is uniaud32-linux-5.1.21-20201231.zip, but that has no sound. The uniaud32-linux-5.2.21-20201231.zip had the same hard-lock as all the following ones did.

Attached is the updated tracker.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 05, 2021, 06:04:40 am
Thanks, I'll run a diff of 5.1 -> 5.2 and see if anything stands out that may cause a hang.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 05, 2021, 07:17:06 am
Meanwhile, I reverted some diffs in http://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-next/alsa-kernel/hda/hdac_controller.c (http://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-next/alsa-kernel/hda/hdac_controller.c) in snd_hdac_bus_get_response() that were put in place some years ago to avoid hangs on HDA chips on OS/2. There seems to be similar code already so I don't think the OS/2 changes are required anymore.

I haven't committed this one yet until I have some more confirmation it doesn't break things...

https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210105c.zip

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 05, 2021, 12:29:03 pm
 Paul,

  All 3 of the release versions of 5.10 made on 0105 work fine here on my HDA:ALC887-VD desktop. I'll try the laptop later today...

  EDIT: all work on the  HDA:IDT92HD81B1X5 laptop too...

Regards,
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 05, 2021, 03:09:36 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210105c.zip

works ok on my CA0106
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on January 05, 2021, 06:23:50 pm
Working well here on my Dell laptop.
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 06, 2021, 04:39:53 pm
uniaud32-linux-5.10.4-20210105c.zip
Working well on everything I've previously tested, plus another 3 intel and 2 AMD Motherboards.

Only thing it's not working on is a Sound Blaster CT4810 PCI 1999 but I think it's a hardware fault. I bought this in 2000 and I've never been able to get it to work with anything though I keep meaning to try it with Windows, I've never got around to it - think it's time to bin it.
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 06, 2021, 05:27:30 pm
Hi Ian,

uniaud32-linux-5.10.4-20210105c.zip
Working well on everything I've previously tested, plus another 3 intel and 2 AMD Motherboards...

What chipset are the AMD motherboards?

My MSI 890FXA-GD70 is 890FX and SB850, but Paul identified a problem (suspicion of one anyways) with the ATI chipset based Realtek AC889 mixers, which is why all releases past uniaud32-linux-5.1.21-20201231.zip have resulted in a HARD-LOCK during boot for me here.

Therefore I'm curious what AMD combo works on your end.

Thanks!
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 06, 2021, 06:46:13 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.4-20210105c.zip

I just did a quick test with "uniaud32-linux-5.10.4-20210105c.zip" on the ArcaOS 5.0.6 VirtualBox VM (Intel HD Audio) on W10 host on the TP Yoga 14.
No Trap. It produce sound and I have control of the volume with the master volume tool. Working good for the moment.

Regards
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 07, 2021, 06:15:02 am
Hi Darius,

Old(er) AMD motherboards,

What chipset are the AMD motherboards?

Gigabyte GA-M61PME-S2 (AMD) Motherboard
=================================
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: NVidia
  driver: HDA-Intel
  name: HDA NVidia
  longname: HDA NVidia at 0xfbff8000 irq 23
  mixer: Realtek ALC662 rev1
  componenets: HDA:10ec0662,1458a002,00100101

eMachines (ECS) ET1331, MCP61PM-GM Rev 2.4 (AMD)
==========================================
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: NVidia
  driver: HDA-Intel
  name: HDA NVidia
  longname: HDA NVidia at 0xdfe78000 irq 23
  mixer: Realtek ALC1200
  componenets: HDA:10ec0888,10250181,00100101

(Edited spelling)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 07, 2021, 09:22:03 am
Hi Dariusz,

Hi - you're missing a few...

https://smedley.id.au/tmp/uniaud32-linux-5.0.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.1.21-20201231.zip
https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20201231.zip

Aha...you bet, testing right now...

Last one to successfully boot is uniaud32-linux-5.1.21-20201231.zip, but that has no sound. The uniaud32-linux-5.2.21-20201231.zip had the same hard-lock as all the following ones did.

Attached is the updated tracker.

https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20210107.zip has hda_intel.c replaced with the version from 5.1.21 - I'd be interested to hear if this resolves the hang. There's nothing obvious at first glance that changed, but worth a shot.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 07, 2021, 02:29:49 pm
Attached file is from testing using  uniaud32-linux-5.10.4-20210105c.zip.
Collection of Intel, AMD motherboards as well as individual PCI cards. All audio was tested with ears as well as 'unimix -card'.
None of the MB's or cards are recent, as they all work fine, and I'm happy with the cards I use under OS/2 or Linux.
I have not tested with ISA cards due to the time it would take to image OS/2 to a IDE drive as I nolonger have any motherboards with both ISA, and SATA.
Not sure if anyone would still be using ISA audio cards (Old consumer SB, Mediavision, aztech etc),

** UPDATED attached file. **
Title: Re: Updated Uniaud32 build
Post by: Andi B. on January 07, 2021, 03:16:52 pm
Quote
Not sure if anyone would still be using ISA audio cards.
Sometimes I do. But only because the analog audio performance of this Turtle Beach Pinnacle is outstanding. As there were never OS/2 drivers available for this card I only use it with the included Win32 applications in WinOS2. Offtopic for this thread.
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 07, 2021, 10:24:48 pm
Hello Paul,

...https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20210107.zip has hda_intel.c replaced with the version from 5.1.21 - I'd be interested to hear if this resolves the hang. There's nothing obvious at first glance that changed, but worth a shot...

Thank you, but same result. Hard-lock during boot.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 08, 2021, 12:53:28 am
Hi Dariusz,

...https://smedley.id.au/tmp/uniaud32-linux-5.2.21-20210107.zip has hda_intel.c replaced with the version from 5.1.21 - I'd be interested to hear if this resolves the hang. There's nothing obvious at first glance that changed, but worth a shot...

Thank you, but same result. Hard-lock during boot.

Interesting - any chance you can setup a dump partition and generate a system dump during the hang? I'm assuming Ctrl-Alt-Num Lock-Num Lock will will work to trigger a dump....
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 08, 2021, 02:27:15 am
Interesting - any chance you can setup a dump partition and generate a system dump during the hang? I'm assuming Ctrl-Alt-Num Lock-Num Lock will will work to trigger a dump....

That's the thing though Paul...even that "get me out of this jam last resort" approach does not work. It is literally a HARD lock...nothing responds.

I honestly am starting to question whether there is something else off in my system.

I am booting with QSINIT and the AOS 202_SMP kernel...tried the old 201_SMP, no difference there. Let me figure out how to get a boot that's as plain-vanilla as possible going...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 08, 2021, 03:13:44 am
Hi Dariusz,

Interesting - any chance you can setup a dump partition and generate a system dump during the hang? I'm assuming Ctrl-Alt-Num Lock-Num Lock will will work to trigger a dump....

That's the thing though Paul...even that "get me out of this jam last resort" approach does not work. It is literally a HARD lock...nothing responds.

I honestly am starting to question whether there is something else off in my system.

I am booting with QSINIT and the AOS 202_SMP kernel...tried the old 201_SMP, no difference there. Let me figure out how to get a boot that's as plain-vanilla as possible going...
A serial log might still narrow down where in the boot things die....
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 08, 2021, 03:42:54 am
Paul!

That's the thing though Paul...even that "get me out of this jam last resort" approach does not work. It is literally a HARD lock...nothing responds.

I honestly am starting to question whether there is something else off in my system.

I am booting with QSINIT and the AOS 202_SMP kernel...tried the old 201_SMP, no difference there. Let me figure out how to get a boot that's as plain-vanilla as possible going...
A serial log might still narrow down where in the boot things die....

Alright...I've got some interesting results to report after all!  8)

1) I went back to the non-QSINIT loader, same result, HARD-LOCK

2) I then went to OS4 config, which means I ran their APIC drivers
a) uniaud32-linux-5.2.21-20210107.zip => boots fine, no problem, sound is there
b) uniaud32-linux-5.10.4-20210104-debug.zip => boots fine, no problem, sound is there

Sooo...clearly in my case there is something about the non-OS4 config that's causing a problem. Just about the only other thing I need to rule out is the latest ACPI driver, that being 3.23.15.

Therefore, I will revert back to 3.23.14 ACPI driver and re-test. I had some funny things pop up with the 3.23.15 driver before, but it was inconsistent and since it didn't repeat I had written it off as a one-off type of a thing. Maybe there is more to it?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 08, 2021, 10:12:51 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.5-20210108.zip
- fixes to eliminate some warnings from David
- Updated core to linux 5.10.5 level
Title: Re: Updated Uniaud32 build
Post by: Lars on January 08, 2021, 11:42:35 am
Sooo...clearly in my case there is something about the non-OS4 config that's causing a problem. Just about the only other thing I need to rule out is the latest ACPI driver, that being 3.23.15.

Therefore, I will revert back to 3.23.14 ACPI driver and re-test. I had some funny things pop up with the 3.23.15 driver before, but it was inconsistent and since it didn't repeat I had written it off as a one-off type of a thing. Maybe there is more to it?
It is worth giving it a try but keep in the back of your head that a Trap E can occur for a lot of reasons. If you use the OS4 kernel, the memory layout will look different. This is also the case if you load additional drivers etc.

By the way: what I have observed with ACPI 3.23.15 is that cores > 0 can take a long time to be reenabled and if they are, they run at 100 % for some time until everything drops back to normal. Can you check with 3.23.14 if that was different or the same ?
The reason behind my question: I had modified an app for 2 of its threads to NOT execute on core 0 (for specific reasons). But with 3.23.15, the system would not reenable cores > 0 even when these 2 threads signaled that they were ready to execute (but would not be scheduled). I had to change the app back to using all cores for these 2 threads.
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 08, 2021, 12:04:12 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.5-20210108.zip

it works ok here
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 08, 2021, 09:34:06 pm
Lars, Paul, everyone...

It is worth giving it a try but keep in the back of your head that a Trap E can occur for a lot of reasons. If you use the OS4 kernel, the memory layout will look different. This is also the case if you load additional drivers etc.

Indeed. There is more to this than just the ACPI driver difference. Going back to 3.23.14 did NOT improve the situation, still the same failure.

For some happy news though: uniaud32-linux-5.10.5-20210108.zip works just fine with OS4.

NOTE: below is very much non-uniaudio specific, but I figured it was worthwhile sharing my experience in case others may benefit from it.

...By the way: what I have observed with ACPI 3.23.15 is that cores > 0 can take a long time to be reenabled and if they are, they run at 100 % for some time until everything drops back to normal. Can you check with 3.23.14 if that was different or the same ?

Interesting, this is very close to what I was seeing actually. More specifically though, in a certain sequence of events (shutting off all but a single core and re-enabling) caused all the cores to go to 100%. That was my brief experience while attempting (re-visiting actually) the ACPI Power Manager use. I normally do NOT run the ACPI power management on my desktop machine. I simply have found that throttling logic and/or execution to be literaly too lagging. Namely, it took too long for the immediate power-needed NOW to translate to additional cores actually throttling up.

Yes I know that the throtting is supposed to happen fast, but alas here that just did not pan out. This is not for lack of trying. I spent a good amount of time working with acpid.cfg and the various settings never really arriving at what felt like a seamless execution. I suppose (I hope) that on a laptop the momentary inconvenience of a lag may be tollerable given the hopefully longer battery life (due to core throttling and therefore power use), but on my desktop machine that just wasn't a tradeoff I wanted to pursue.

...The reason behind my question: I had modified an app for 2 of its threads to NOT execute on core 0 (for specific reasons). But with 3.23.15, the system would not reenable cores > 0 even when these 2 threads signaled that they were ready to execute (but would not be scheduled). I had to change the app back to using all cores for these 2 threads.

I can not speak to core-affinity requirements such as these, but I will tell you that previously 3.23.13 was trully the last good version of ACPI I had. When 3.23.14 came out I immediately hit a wall in terms of being able to enabled and disable cores, keep in mind this is all with the ACPI PM shut OFF. I had logged this issue for AOS given that my reporting was being done as part of the Testers List and was very specific to the ACPI driver itself as opposed to the whole AOS package (https://mantis.arcanoae.com/view.php?id=3002 (https://mantis.arcanoae.com/view.php?id=3002)). Did not make much progress with that, things went nowhere beyond some technical info exchange and the case was ultimately closed due to being of "unsupported configuration" (I am not running AOS, still Warp 4.5 CP2 with piece-wise upgrades).

My point is, through multiple tests of my own it felt like 3.23.15 finally brought back the 3.23.13 behaviour (core disable & enable now works), however, I did see this weirdness you talked about.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 09, 2021, 06:05:58 pm
Hi

https://smedley.id.au/tmp/uniaud32-linux-5.10.5-20210108.zip

I tried "uniaud32-linux-5.10.5-20210108.zip" on the Thinkpad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420).
- No Trap. Sound is working. Master volume control working. PMUnimix 0.82 master volume working.

Congratulation on reaching this Linux kernel level release.

Regards
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 09, 2021, 07:35:39 pm
 Paul,

 The 5.10.5-20210108 version is working well on both my desktop and laptop. Thanks again!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 09, 2021, 10:47:50 pm
Quote
The 5.10.5-20210108 version

Works okay on:
Lenovo ThinkPad T510
Lenovo ThinkPad L530
Asus M3A78-EM

Doesn't work on my Asus A88XM-A. No traps, but it appears that it doesn't load, at all. The older UniAud (from ArcaOS 5.0.6) works, but the controller has problems, even in windows (apparently it has an open input, that cannot be turned off, without using a special Realtek driver).

Okay, some details, with or without "/A:1" (needed for the ArcaOS driver):
Using the 108 driver, UniMix says:
Code: [Select]
[C:\MMOS2]unimix -dev0 -card -pcms
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Error: invalid number of PCM instances for adapter 1
Card info for adapter 0:
  num: 0
  id: HDMI
  driver: HDA-Intel
  name: HDA ATI HDMI
  longname: HDA ATI HDMI at 0xfeb44000 irq 18
  mixer:
  componenets: HDA:1002aa01,00aa0100,00100300

[C:\MMOS2]unimix -dev1 -card -pcms
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Error: invalid number of PCM instances for adapter 1
Card info for adapter 1:
  num: 1
  id: Generic
  driver: HDA-Intel
  name: HD-Audio Generic
  longname: HD-Audio Generic at 0xfeb40000 irq 16
  mixer:
  componenets: HDA:10ec0887,10438576,00100302

Using the driver from ArcaOS 5.0.6 (with "/A:1) it says:
Code: [Select]
[C:\MMOS2]unimix -dev0 -card -pcms
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.06
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Generic
  driver: HDA-Intel
  name: HD-Audio Generic
  longname: HD-Audio Generic at 0xfeb40000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,10438576,00100302
PCM Instance Info for adapter 0
PCM No 0 info:
PCM dev name: [ALC887-VD Analog] mixer [subdevice #0]
 PLAY
  streams: 1
  channels: from 2 to 2
  rates: from 44100 to 192000
 RECORD
  streams: 1
  channels: from 2 to 2
  rates: from 44100 to 192000

PCM No 1 info:
PCM dev name: [] mixer []
 PLAY
  streams: 0
  channels: from 0 to 0
  rates: from 0 to 0
 RECORD
  streams: 0
  channels: from 0 to 0
  rates: from 0 to 0


[C:\MMOS2]unimix -dev1 -card -pcms
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.06
Detected 1 audio adapter(s)

Now, the "interesting part:
From PCI.EXE, I see two audio devices. First an Intel device, that seems to be attached to HDMI:
Code: [Select]
Bus 0 (PCI Express), Device Number 1, Device Function 1
 Vendor 1002h Advanced Micro Devices, Inc. [AMD/ATI]
 Device 9902h Trinity HDMI Audio Controller
 Command 0007h (I/O Access, Memory Access, BusMaster)
 Status 0010h (Has Capabilities List, Fast Timing)
 Revision 00h, Header Type 80h, Bus Latency Timer 00h
 Self test 00h (Self test not supported)
 Cache line size 64 Bytes (16 DWords)
 PCI Class Multimedia, type Unknown!
 Subsystem ID 85261043h Unknown
 Subsystem Vendor 1043h ASUSTeK Computer Inc.
 Address 0 is a Memory Address (0-4GiB) : FEB44000h..FEB47FFFh
 System IRQ 18, INT# B
 New Capabilities List Present:
   Power Management Capability, Version 1.2
     Supports low power State D1
     Supports low power State D2
     Does not support PME# signalling
     Current Power State : D0 (Device operational, no power saving)
   PCI Express Capability, Version 2
     Device/Port Type :
       Root Complex Integrated Endpoint Device
       Unsupported Request Severity is Fatal
     Device Status :
     Link Capabilities :
       Maximum Link speed : Unknown (00h)!!
       Maximum Link Width : xReserved
       Link Port Number   : 0
     Link Control :
       Asynchronous Clocking in Use
     Link Status :
       Current Link speed : Unknown (00)!!
       Current Link Width : x0
   Message Signalled Interrupt Capability
     MSI is disabled
     MSI function can generate 64-bit addresses

There is also:
Code: [Select]
Bus 0 (PCI Express), Device Number 20, Device Function 2
 Vendor 1022h Advanced Micro Devices, Inc. [AMD]
 Device 780Dh FCH Azalia Controller
 Command 0006h (Memory Access, BusMaster)
 Status 0410h (Has Capabilities List, Slow Timing)
 Revision 01h, Header Type 00h, Bus Latency Timer 20h
 Self test 00h (Self test not supported)
 Cache line size 64 Bytes (16 DWords)
 PCI Class Multimedia, type Unknown!
 Subsystem ID 85761043h Unknown
 Subsystem Vendor 1043h ASUSTeK Computer Inc.
 Address 0 is a Memory Address (64-bit) : FEB40000h..FEB43FFFh
 System IRQ 16, INT# A
 New Capabilities List Present:
   Power Management Capability, Version 1.1
     Does not support low power State D1 or D2
     Supports PME# signalling from mode(s) D0, D3hot, D3cold
     PME# signalling is currently disabled
     3.3v AUX Current required : 55 mA
     Current Power State : D0 (Device operational, no power saving)
which seems to be the one that actually makes sound, with the driver from ArcaOS.

Also, note that the ArcaOS driver is shown as version 2.02.06, while your 108 is shown as version 2.02.05.

Let me know if I can supply more information.

THANKS...
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 10, 2021, 01:09:36 am
Code: [Select]
Hi Doug,

[quote author=Doug Bissett link=topic=2436.msg29806#msg29806 date=1610228870]
Doesn't work on my Asus A88XM-A. No traps, but it appears that it doesn't load, at all. The older UniAud (from ArcaOS 5.0.6) works, but the controller has problems, even in windows (apparently it has an open input, that cannot be turned off, without using a special Realtek driver).
[C:\MMOS2]unimix -dev1 -card -pcms
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 0
Error: invalid number of PCM instances for adapter 1
Card info for adapter 1:
  num: 1
  id: Generic
  driver: HDA-Intel
  name: HD-Audio Generic
  longname: HD-Audio Generic at 0xfeb40000 irq 16
  mixer:
  componenets: HDA:10ec0887,10438576,00100302
[/quote]

Seems the mixer isn't identified for some reason.  I'd be interested in seeing the output of 'copy alsahlp$ alsa.log'

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 02:47:33 am
Quote
I'd be interested in seeing the output of 'copy alsahlp$ alsa.log'

Code: [Select]
Uniaud32 version 2.02.05-SVN
Based on linux 5.10.5
pci_register_driver: query_device found 9 1002:9902 class=40300 checking intel8x0
pci_register_driver: query_device found a2 1022:780d class=40300 checking intel8x0
pci_register_driver: query_device found 9 1002:9902 class=40300 checking via82xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking via82xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking emu10k1
pci_register_driver: query_device found a2 1022:780d class=40300 checking emu10k1
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cmipci
pci_register_driver: query_device found a2 1022:780d class=40300 checking cmipci
pci_register_driver: query_device found 9 1002:9902 class=40300 checking als4000
pci_register_driver: query_device found a2 1022:780d class=40300 checking als4000
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs4281
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs4281
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs46xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs46xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs5535audio
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs5535audio
pci_register_driver: query_device found 9 1002:9902 class=40300 checking es1938
pci_register_driver: query_device found a2 1022:780d class=40300 checking es1938
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ens137x
pci_register_driver: query_device found a2 1022:780d class=40300 checking ens137x
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ymfpci
pci_register_driver: query_device found a2 1022:780d class=40300 checking ymfpci
pci_register_driver: query_device found 9 1002:9902 class=40300 checking es1968
pci_register_driver: query_device found a2 1022:780d class=40300 checking es1968
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Maestro3
pci_register_driver: query_device found a2 1022:780d class=40300 checking Maestro3
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ali5451
pci_register_driver: query_device found a2 1022:780d class=40300 checking ali5451
pci_register_driver: query_device found 9 1002:9902 class=40300 checking trident
pci_register_driver: query_device found a2 1022:780d class=40300 checking trident
pci_register_driver: query_device found 9 1002:9902 class=40300 checking nm256
pci_register_driver: query_device found a2 1022:780d class=40300 checking nm256
pci_register_driver: query_device found 9 1002:9902 class=40300 checking fm801
pci_register_driver: query_device found a2 1022:780d class=40300 checking fm801
pci_register_driver: query_device found 9 1002:9902 class=40300 checking atiixp
pci_register_driver: query_device found a2 1022:780d class=40300 checking atiixp
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ca0106
pci_register_driver: query_device found a2 1022:780d class=40300 checking ca0106
pci_register_driver: query_device found 9 1002:9902 class=40300 checking hda_intel
pci_register_driver: matched 87 1002:9902/40300 with 1002:9902/0 0 (hda_intel)
pci_register_driver: matched 119 1002:9902/40300 with 1002:ffffffff/40300 ffffff (hda_intel)
Using LPIB position fix

Force to snoop mode by module option

Disabling 64bit MSI

chipset global capabilities = 0x0

codec_mask = 0x1

request_irq: irq 18
codec #0 probed OK

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

Unable to bind the codec

pci_register_driver: query_device found 9 1002:9902 class=40300 checking bt87x
pci_register_driver: query_device found a2 1022:780d class=40300 checking bt87x
pcm_info()
OSS32_Initialize: SUCCESS. Cards=1
Init=35/0 Between=0/0 Complete=0/0
U16: Uniaud16 v1.9.7
OSS32_WaveOpen: open error: -19

OSS32_QueryDevCaps: wave open error 3

OSS32_QueryDevCaps failed

Init=35/0 Between=0/0 Complete=0/0
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 10, 2021, 03:12:10 am
Hi Doug, that's a very short log - is that with the Debug or Release driver?
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 05:09:58 am
Quote
Hi Doug, that's a very short log - is that with the Debug or Release driver?

The release driver. I will try it with the debug driver.
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 05:19:30 am
Okay, with the debug driver:
Code: [Select]
Uniaud32 version 2.02.05-SVN-DEBUG
Based on linux 5.10.5
OSS32_Initialize: Start address: ef692000
kobject: 'snd_seq' (0xef4ec3a8): kobject_add_internal: parent: '<NULL>', set: '<NULL>'

kobject: 'devices' (0xef4ec490): kobject_add_internal: parent: 'snd_seq', set: '<NULL>'

kobject: 'drivers' (0xef4ec4e4): kobject_add_internal: parent: 'snd_seq', set: '<NULL>'

bus: 'snd_seq': registered

OSS32_Initialize1: Start address: ef692000
bus: 'snd_seq': add driver opl3_seq

kobject: 'opl3_seq' (0xef4eca5c): kobject_add_internal: parent: 'drivers', set: 'drivers'

kobject: 'hdaudio' (0xef4ecdd4): kobject_add_internal: parent: '<NULL>', set: '<NULL>'

kobject: 'devices' (0xef4ecea8): kobject_add_internal: parent: 'hdaudio', set: '<NULL>'

kobject: 'drivers' (0xef4ecee8): kobject_add_internal: parent: 'hdaudio', set: '<NULL>'

bus: 'hdaudio': registered

bus: 'hdaudio': add driver hda_generic

kobject: 'hda_generic' (0xef4ecf1c): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_analog

kobject: 'patch_analog' (0xef4ecf90): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_ca0110

kobject: 'patch_ca0110' (0xef4ed008): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_cirrus

kobject: 'patch_cirrus' (0xef4ed080): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_cmedia

kobject: 'patch_cmedia' (0xef4ed0f8): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_conexant

kobject: 'patch_conexant' (0xef4ed170): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_realtek

kobject: 'patch_realtek' (0xef4ed1e8): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_sigmatel

kobject: 'patch_sigmatel' (0xef4ed260): kobject_add_internal: parent: 'drivers', set: 'drivers'

bus: 'hdaudio': add driver patch_via

kobject: 'patch_via' (0xef4ed2d8): kobject_add_internal: parent: 'drivers', set: 'drivers'

pci_register_driver: query_device found 9 1002:9902 class=40300 checking intel8x0
pci_register_driver: query_device found a2 1022:780d class=40300 checking intel8x0
pci_register_driver: query_device found 9 1002:9902 class=40300 checking via82xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking via82xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking emu10k1
pci_register_driver: query_device found a2 1022:780d class=40300 checking emu10k1
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cmipci
pci_register_driver: query_device found a2 1022:780d class=40300 checking cmipci
pci_register_driver: query_device found 9 1002:9902 class=40300 checking als4000
pci_register_driver: query_device found a2 1022:780d class=40300 checking als4000
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs4281
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs4281
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs46xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs46xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs5535audio
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs5535audio
pci_register_driver: query_device found 9 1002:9902 class=40300 checking es1938
pci_register_driver: query_device found a2 1022:780d class=40300 checking es1938
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ens137x
pci_register_driver: query_device found a2 1022:780d class=40300 checking ens137x
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ymfpci
pci_register_driver: query_device found a2 1022:780d class=40300 checking ymfpci
pci_register_driver: query_device found 9 1002:9902 class=40300 checking es1968
pci_register_driver: query_device found a2 1022:780d class=40300 checking es1968
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Maestro3
pci_register_driver: query_device found a2 1022:780d class=40300 checking Maestro3
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ali5451
pci_register_driver: query_device found a2 1022:780d class=40300 checking ali5451
pci_register_driver: query_device found 9 1002:9902 class=40300 checking trident
pci_register_driver: query_device found a2 1022:780d class=40300 checking trident
pci_register_driver: query_device found 9 1002:9902 class=40300 checking nm256
pci_register_driver: query_device found a2 1022:780d class=40300 checking nm256
pci_register_driver: query_device found 9 1002:9902 class=40300 checking fm801
pci_register_driver: query_device found a2 1022:780d class=40300 checking fm801
pci_register_driver: query_device found 9 1002:9902 class=40300 checking atiixp
pci_register_driver: query_device found a2 1022:780d class=40300 checking atiixp
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ca0106
pci_register_driver: query_device found a2 1022:780d class=40300 checking ca0106
pci_register_driver: query_device found 9 1002:9902 class=40300 checking hda_intel
pci_register_driver: matched 87 1002:9902/40300 with 1002:9902/0 0 (hda_intel)
pci_register_driver: matched 119 1002:9902/40300 with 1002:ffffffff/40300 ffffff (hda_intel)
RMCreateDriver rc=0 DriverHandle=f937c0f0
pci_enable_device ef62dba8

Using LPIB position fix

Force to snoop mode by module option

Disabling 64bit MSI

chipset global capabilities = 0x0

snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f879b000 (phys eb8000) total alloc size=1000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f879a000 (phys eb7000) total alloc size=2000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8799000 (phys eb6000) total alloc size=3000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8798000 (phys eb5000) total alloc size=4000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8797000 (phys eb4000) total alloc size=5000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8796000 (phys eb3000) total alloc size=6000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8795000 (phys eb2000) total alloc size=7000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8794000 (phys eb1000) total alloc size=8000
snd_dma_alloc_pages with size = 64
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8782000 (phys eb0000) total alloc size=9000
snd_dma_alloc_pages with size = 4096
pci_alloc_consistent 4096 mask 0
get_free_dma_pages: size=1000 adr=f8780000 (phys eae000) total alloc size=a000
codec_mask = 0x1

request_irq: irq 18
RMAllocResource[2] IRQ=18 rc=0
codec #0 probed OK

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

Unable to bind the codec

kobject: 'card0' (0xef4ed684): kobject_cleanup

kobject: 'card0' (0xef4ed684): calling ktype release

kobject: 'card0': free name

RMCreateAdapter rc=0
RMCreateDevice rc=0
RMDone: DriverHandle=f937c0f0 DeviceHandle=f927c3b0
pci_register_driver: query_device found 9 1002:9902 class=40300 checking bt87x
pci_register_driver: query_device found a2 1022:780d class=40300 checking bt87x
pcm_info()
OSS32_Initialize: SUCCESS. Cards=1
Init=35/0 Between=0/0 Complete=0/0
OSS32_ATTACH

U16: Uniaud16 v1.9.7
OSS32_QueryDevCaps
Number of cards=1 dev id=0
OSS32_WaveOpen: open error: -19

OSS32_QueryDevCaps: wave open error 3

OSS32_QueryDevCaps failed

Init=35/0 Between=0/0 Complete=0/0
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 05:37:09 am
Just to complete the picture, here is the log from the driver that works:
Code: [Select]
Uniaud32 version 2.02.06-SVN
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Intel ICH
pci_register_driver: query_device found a2 1022:780d class=40300 checking Intel ICH
pci_register_driver: query_device found 9 1002:9902 class=40300 checking VIA 82xx Audio
pci_register_driver: query_device found a2 1022:780d class=40300 checking VIA 82xx Audio
pci_register_driver: query_device found 9 1002:9902 class=40300 checking EMU10K1_Audigy
pci_register_driver: query_device found a2 1022:780d class=40300 checking EMU10K1_Audigy
pci_register_driver: query_device found 9 1002:9902 class=40300 checking C-Media PCI
pci_register_driver: query_device found a2 1022:780d class=40300 checking C-Media PCI
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ALS4000
pci_register_driver: query_device found a2 1022:780d class=40300 checking ALS4000
pci_register_driver: query_device found 9 1002:9902 class=40300 checking CS4281
pci_register_driver: query_device found a2 1022:780d class=40300 checking CS4281
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Sound Fusion CS46xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking Sound Fusion CS46xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs5535audio
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs5535audio
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ESS ES1938 (Solo-1)
pci_register_driver: query_device found a2 1022:780d class=40300 checking ESS ES1938 (Solo-1)
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ENS1370
pci_register_driver: query_device found a2 1022:780d class=40300 checking ENS1370
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ENS1371
pci_register_driver: query_device found a2 1022:780d class=40300 checking ENS1371
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Yamaha DS-1 PCI
pci_register_driver: query_device found a2 1022:780d class=40300 checking Yamaha DS-1 PCI
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ES1968 (ESS Maestro)
pci_register_driver: query_device found a2 1022:780d class=40300 checking ES1968 (ESS Maestro)
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Maestro3
pci_register_driver: query_device found a2 1022:780d class=40300 checking Maestro3
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ALI 5451
pci_register_driver: query_device found a2 1022:780d class=40300 checking ALI 5451
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Trident4DWaveAudio
pci_register_driver: query_device found a2 1022:780d class=40300 checking Trident4DWaveAudio
pci_register_driver: query_device found 9 1002:9902 class=40300 checking NeoMagic 256
pci_register_driver: query_device found a2 1022:780d class=40300 checking NeoMagic 256
pci_register_driver: query_device found 9 1002:9902 class=40300 checking FM801
pci_register_driver: query_device found a2 1022:780d class=40300 checking FM801
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ATI IXP AC97 controller
pci_register_driver: query_device found a2 1022:780d class=40300 checking ATI IXP AC97 controller
pci_register_driver: query_device found 9 1002:9902 class=40300 checking CA0106
pci_register_driver: query_device found a2 1022:780d class=40300 checking CA0106
pci_register_driver: query_device found 9 1002:9902 class=40300 checking HDA Intel
pci_register_driver: matched 35 1002:9902/40300 with 1002:ffffffff/40300 ffffff (HDA Intel)
pci_register_driver: query_device found a2 1022:780d class=40300 checking HDA Intel
pci_register_driver: matched 36 1022:780d/40300 with 1022:ffffffff/40300 ffffff (HDA Intel)
request_irq: irq 16
SKU: Nid=0x1d sku_cfg=0x4044c601

SKU: port_connectivity=0x1

SKU: enable_pcbeep=0x0

SKU: check_sum=0x00000004

SKU: customization=0x000000c6

SKU: external_amp=0x0

SKU: platform_type=0x0

SKU: swap=0x0

SKU: override=0x1

<6>hda_codec: ALC887-VD: BIOS auto-probing.

autoconfig: line_outs=1 (0x14/0x0/0x0/0x0/0x0)

   speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)

   hp_outs=1 (0x1b/0x0/0x0/0x0/0x0)

   mono: mono_out=0x0

   dig-out=0x11/0x0

   inputs:


realtek: No valid SSID, checking pincfg 0x4044c601 for NID 0x1d

realtek: Enabling init ASM_ID=0xc601 CODEC_ID=10ec0887

pci_register_driver: query_device found 9 1002:9902 class=40300 checking Bt87x
pci_register_driver: query_device found a2 1022:780d class=40300 checking Bt87x
00-00: ALC887-VD Analog : ALC887-VD Analog
 : playback 1
 : capture 1


00-03: ALC887-VD Digital : ALC887-VD Digital
 : playback 1


OSS32_WaveOpen: open error: -19

OSS32_WaveOpen: open error: -19

OSS32_Initialize: SUCCESS. Cards=1
Init=384/0 Between=0/0 Complete=0/0
U16: Uniaud16 v1.9.7
Midi Open 2 failed with error -19
Midi Open 3 failed with error -19
Couldn't find OPL3 device
Init=384/0 Between=27/0 Complete=0/0
snd_pcm_common_ioctl1: unknown ioctl = 0x401c4162
<4>hda-intel: Invalid position buffer, using LPIB read method instead.

Init=384/0 Between=27/0 Complete=312/0
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 10, 2021, 07:05:05 am
Definitely interesting - it might help to know when this chipset stopped working - it which kernel revision broke things - just depends if you can be bothered working through all the builds I've posted to determine when it broke... The only clue in the log is 'Unable to bind the codec'
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 09:00:57 am
Quote
Definitely interesting - it might help to know when this chipset stopped working - it which kernel revision broke things - just depends if you can be bothered working through all the builds I've posted to determine when it broke... The only clue in the log is 'Unable to bind the codec'

I haven't been testing on this machine, because it is my "production" machine. Others were reporting what I saw on the laptops. They seem to be working now, so I tried this one. OOPS!

I will see if I can figure it out, tomorrow. I will try jumping in at various points to try to avoid having to try every release.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 10, 2021, 03:34:37 pm
Doug:
FYI: I have a hardware report on "AMD FCH Azalia Controller  1002:780D" that was not working with UniAud 2.2.6 on the Thinkpad E555 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_E555). Is it working now with uniaud32-linux-5.10.5-20210108.zip?

Regard
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 07:53:06 pm
Doug:
FYI: I have a hardware report on "AMD FCH Azalia Controller  1002:780D" that was not working with UniAud 2.2.6 on the Thinkpad E555 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_E555). Is it working now with uniaud32-linux-5.10.5-20210108.zip?

Regard

A lot of your information is probably wrong, or needs an added explanation. Mine is a desktop, with Asus A88XM-A motherboard and AMD A6 processor. It does work with the UniAud that comes with ArcaOS 5.0.6, but but requires the /A:1 parameter to get it to use the proper adapter (the HDMI audio adapter doesn't work, but I don't have an HDMI device to connect anyway). It did NOT work with some earlier version of UniAud (about 3 or 4 years ago). I don't remember the details. As I posted, it is NOT working with the 108 test version.
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 10, 2021, 09:34:49 pm
Definitely interesting - it might help to know when this chipset stopped working - it which kernel revision broke things - just depends if you can be bothered working through all the builds I've posted to determine when it broke... The only clue in the log is 'Unable to bind the codec'

I tried the release versions:
uniaud32-linux-3.0.zip works
uniaud32-linux-3.1.zip works
uniaud32-trunk.zip works
uniaud32-linux-3.1.9.zip Traps
uniaud32-linux-3.7.9.zip FAILS
AlSA LOG:
Code: [Select]
Uniaud32 version 2.02.05-SVN
pci_register_driver: query_device found 9 1002:9902 class=40300 checking intel8x0
pci_register_driver: query_device found a2 1022:780d class=40300 checking intel8x0
pci_register_driver: query_device found 9 1002:9902 class=40300 checking via82xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking via82xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking emu10k1
pci_register_driver: query_device found a2 1022:780d class=40300 checking emu10k1
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cmipci
pci_register_driver: query_device found a2 1022:780d class=40300 checking cmipci
pci_register_driver: query_device found 9 1002:9902 class=40300 checking als4000
pci_register_driver: query_device found a2 1022:780d class=40300 checking als4000
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs4281
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs4281
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs46xx
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs46xx
pci_register_driver: query_device found 9 1002:9902 class=40300 checking cs5535audio
pci_register_driver: query_device found a2 1022:780d class=40300 checking cs5535audio
pci_register_driver: query_device found 9 1002:9902 class=40300 checking es1938
pci_register_driver: query_device found a2 1022:780d class=40300 checking es1938
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ens1370
pci_register_driver: query_device found a2 1022:780d class=40300 checking ens1370
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ens1370
pci_register_driver: query_device found a2 1022:780d class=40300 checking ens1370
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ymfpci
pci_register_driver: query_device found a2 1022:780d class=40300 checking ymfpci
pci_register_driver: query_device found 9 1002:9902 class=40300 checking es1968
pci_register_driver: query_device found a2 1022:780d class=40300 checking es1968
pci_register_driver: query_device found 9 1002:9902 class=40300 checking Maestro3
pci_register_driver: query_device found a2 1022:780d class=40300 checking Maestro3
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ali5451
pci_register_driver: query_device found a2 1022:780d class=40300 checking ali5451
pci_register_driver: query_device found 9 1002:9902 class=40300 checking trident
pci_register_driver: query_device found a2 1022:780d class=40300 checking trident
pci_register_driver: query_device found 9 1002:9902 class=40300 checking nm256
pci_register_driver: query_device found a2 1022:780d class=40300 checking nm256
pci_register_driver: query_device found 9 1002:9902 class=40300 checking fm801
pci_register_driver: query_device found a2 1022:780d class=40300 checking fm801
pci_register_driver: query_device found 9 1002:9902 class=40300 checking atiixp
pci_register_driver: query_device found a2 1022:780d class=40300 checking atiixp
pci_register_driver: query_device found 9 1002:9902 class=40300 checking ca0106
pci_register_driver: query_device found a2 1022:780d class=40300 checking ca0106
pci_register_driver: query_device found 9 1002:9902 class=40300 checking hda_intel
pci_register_driver: matched 37 1002:9902/40300 with 1002:9902/0 0 (hda_intel)
pci_register_driver: matched 54 1002:9902/40300 with 1002:ffffffff/40300 ffffff (hda_intel)
hda-intel: Using LPIB position fix

request_irq: irq 18
hda-intel: chipset global capabilities = 0x0

hda-intel: codec_mask = 0x1

hda-intel: codec #0 probed OK

hda-intel: Enable sync_write for stable communication

hda_generic: no proper input path found

Skip Digital OUT node 2

hda_generic: no proper output path found

hda_generic: no PCM found

pci_register_driver: query_device found 9 1002:9902 class=40300 checking bt87x
pci_register_driver: query_device found a2 1022:780d class=40300 checking bt87x
OSS32_Initialize: SUCCESS. Cards=1
Init=61/0 Between=0/0 Complete=0/0
U16: Uniaud16 v1.9.7
OSS32_WaveOpen: open error: -19

OSS32_QueryDevCaps: wave open error 3

OSS32_QueryDevCaps failed

Init=61/0 Between=0/0 Complete=0/0

Did I miss any? It is difficult to follow in this forum.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 11, 2021, 09:51:07 am
Did I miss any? It is difficult to follow in this forum.

A couple :) Doing a dir of uniaud32-*202*.zip on my webserver temp directory, I get the following:
Code: [Select]
uniaud32-linux-3.0.zip
uniaud32-linux-3.1.zip
uniaud32-linux-3.1.9.zip
uniaud32-linux-3.2.102.zip
uniaud32-linux-3.2-debug.zip
uniaud32-linux-3.2.zip
uniaud32-linux-3.5.zip
uniaud32-linux-3.6.11.zip
uniaud32-linux-3.7.9.zip
uniaud32-linux-3.8.13.zip
uniaud32-linux-3.9.11.zip
uniaud32-linux-3.10.108-20200804.zip
uniaud32-linux-3.10.108-20200807b.zip
uniaud32-linux-3.10.108-20200807c.zip
uniaud32-linux-3.10.108-20200807d.zip
uniaud32-linux-3.10.108-20200807.zip
uniaud32-linux-3.10.108-20200808.zip
uniaud32-linux-3.11.10-20200810.zip
uniaud32-linux-3.12.74-20200807.zip
uniaud32-linux-3.12.74-20200808.zip
uniaud32-linux-3.12.74-20200809.zip
uniaud32-linux-3.12.74-20200810.zip
uniaud32-linux-3.14.79-20200809.zip
uniaud32-linux-3.14.79-20200810.zip
uniaud32-linux-3.16.85-20200812.zip
uniaud32-linux-3.16.85-20200813.zip
uniaud32-linux-3.16.85-20200814.zip
uniaud32-linux-3.17.8-20200823.zip
uniaud32-linux-3.17.8-20200824.zip
uniaud32-linux-3.18.140-20200816.zip
uniaud32-linux-3.18.140-20201009.zip
uniaud32-linux-4.0.9-20200816.zip
uniaud32-linux-4.0.9-20201008.zip
uniaud32-linux-4.4.238-20201018.zip
uniaud32-linux-4.4.238-20201022.zip
uniaud32-linux-4.9.240-20201024.zip
uniaud32-linux-4.14.202-20201227.zip
uniaud32-linux-4.16.18-20201227.zip
uniaud32-linux-4.17.19-20201228.zip
uniaud32-linux-4.19.163-20201228.zip
uniaud32-linux-4.19.163-20201229b.zip
uniaud32-linux-4.19.163-20201229.zip
uniaud32-linux-4.19.163-20201230b.zip
uniaud32-linux-4.19.163-20201230c.zip
uniaud32-linux-4.19.163-20201230d.zip
uniaud32-linux-4.19.163-20201230.zip
uniaud32-linux-5.0.21-20201231.zip
uniaud32-linux-5.1.21-20201231.zip
uniaud32-linux-5.2.21-20201231.zip
uniaud32-linux-5.2.21-20210107.zip
uniaud32-linux-5.3.18-20210101.zip
uniaud32-linux-5.4.86-20210101b.zip
uniaud32-linux-5.4.86-20210101.zip
uniaud32-linux-5.4.86-20210102-debug.zip
uniaud32-linux-5.4.86-20210103-debug.zip
uniaud32-linux-5.5.19-20210104-debug.zip
uniaud32-linux-5.6.19-20210104-debug.zip
uniaud32-linux-5.7.19-20210104-debug.zip
uniaud32-linux-5.8.18-20210104-debug.zip
uniaud32-linux-5.9.16-20210104-debug.zip
uniaud32-linux-5.10.4-20210104-debug.zip
uniaud32-linux-5.10.4-20210105b.zip
uniaud32-linux-5.10.4-20210105c.zip
uniaud32-linux-5.10.4-20210105.zip
uniaud32-linux-5.10.5-20210108.zip

Add https://smedley.id.au/tmp/ to the front of the filenames to download....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: mike on January 11, 2021, 07:07:06 pm
Hi Dave,

This 0103 version works perfectly on my desktop machine (HDA: ALC887-VD): boots, proper sound, 9 controls in PMUnimix, no hesitations, appears in the hardware manager. I'll keep using this one and let you know if anything comes up - thanks again! Hope to soon see your bust in the OS/2 hall of fame!

Glad it works for you! Your comment about the hall of fame gave me a giggle - to be honest - I'm just glad people appreciate my work. The comments (and of course the occasional sponsorship that helps pay for new gadgets) are truly appreciated.

Big thanks for recent working updates for uniaud, as it was somehow stated that the code base what we had was not usable anymore and needs a complete replacement. Good that the core updates are in place now.

Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 11, 2021, 07:53:02 pm
Big thanks Paul for getting the UniAud drivers to the latest linux kernel level. From may 2020 to today you have generated 65 compilations of the driver. Awesome.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 11, 2021, 09:54:13 pm
Hi Mike,
Hi Dave,

This 0103 version works perfectly on my desktop machine (HDA: ALC887-VD): boots, proper sound, 9 controls in PMUnimix, no hesitations, appears in the hardware manager. I'll keep using this one and let you know if anything comes up - thanks again! Hope to soon see your bust in the OS/2 hall of fame!

Glad it works for you! Your comment about the hall of fame gave me a giggle - to be honest - I'm just glad people appreciate my work. The comments (and of course the occasional sponsorship that helps pay for new gadgets) are truly appreciated.

Big thanks for recent working updates for uniaud, as it was somehow stated that the code base what we had was not usable anymore and needs a complete replacement. Good that the core updates are in place now.
Thanks! As to the comment on the code base being unusable - I'll admit that I once believed that. ALSA used to be a separate download up until ALSA 1.0.25 at which point the code moved into the linux kernel, and code moved around in terms of what directory it was located in. At the time ALSA stopped being able to be downloaded separately, I did look at the linux kernel code and for whatever reason - couldn't piece together how they'd split the code up.

Looking at it again early last year it didn't take me long at all to decipher things, and I can now pretty quickly replicate the necessary code from the linux kernel is a structure that's compatible with Uniaud32.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 11, 2021, 09:56:18 pm
Hi Martin,

Big thanks Paul for getting the UniAud drivers to the latest linux kernel level. From may 2020 to today you have generated 65 compilations of the driver. Awesome.

Thanks :) There's actually many more than 65 builds - what you see is just the ones I uploaded. In particular, between 4.0.9 and the working 4.14.202 I was not idle - I've spent literally HUNDREDS of hours over the past 12 months on Uniaud.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on January 12, 2021, 12:48:54 pm
Paul you are brilliant. Thank you!
Title: Re: Updated Uniaud32 build
Post by: Jochen Schäfer on January 12, 2021, 02:24:52 pm
I tried uniaud32-linux-5.10.5-20210108.zip on an Yoga 11e (1st gen) and now I have no stutters anymore, volume control works and Dosbox makes noise again.
Good job, Paul.

PS: I will also test the test builds on a T420.
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 12, 2021, 10:00:32 pm
Quote
A couple :) Doing a dir of uniaud32-*202*.zip on my webserver temp directory, I get the following:

Sorry, but I have had a serious problem with my cell phone (now fixed).  Then my brother is having a PMMail problem, as well as a desktop problem, so I haven't had time to look at this. I will probably be able to get to it sometime tomorrow, if I can see, after an eye exam.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 13, 2021, 12:24:29 am
Hi Doug,

Quote
A couple :) Doing a dir of uniaud32-*202*.zip on my webserver temp directory, I get the following:

Sorry, but I have had a serious problem with my cell phone (now fixed).  Then my brother is having a PMMail problem, as well as a desktop problem, so I haven't had time to look at this. I will probably be able to get to it sometime tomorrow, if I can see, after an eye exam.

All good - I'm back at work now anyway, so my amount of spare time to investigate things is significantly reduced...

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 15, 2021, 09:36:31 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210115.zip - code bumped to linux 5.10.7 level - small changes as listed in http://trac.netlabs.org/uniaud/changeset/643
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 15, 2021, 11:28:15 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210115.zip

it works fine for me
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 15, 2021, 11:49:11 am
uniaud32-linux-5.10.7-20210115.zip

Works well here :)
Thanks again for your work on bringing uniaud up to-date.
Title: Re: Updated Uniaud32 build
Post by: roberto on January 15, 2021, 04:56:06 pm
uniaud32-linux-5.10.7-20210115.zip
All ok , and all previous version is ok too.
Thanks
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on January 15, 2021, 05:04:21 pm
Another winner Paul on my Dell laptop.
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 15, 2021, 08:28:34 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210115.zip - code bumped to linux 5.10.7 level - small changes as listed in http://trac.netlabs.org/uniaud/changeset/643

BINGO! Now it works, better than any other UniAud that I have working. It seems to have turned off the open input too.

THANK YOU
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 15, 2021, 09:54:34 pm
 5.10.7 Works good here too on both laptop and desktop....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 15, 2021, 11:00:38 pm
Hi Doug,

https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210115.zip - code bumped to linux 5.10.7 level - small changes as listed in http://trac.netlabs.org/uniaud/changeset/643

BINGO! Now it works, better than any other UniAud that I have working. It seems to have turned off the open input too.

THANK YOU

Whilst I'm pleased to hear it's working for you, I'm struggling to understand what's changed in the code to fix things <shrug>

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Lars on January 15, 2021, 11:15:55 pm
The mismatch of using one mutex for locking and unlocking a different mutex. That surely led to a hang.
I am talking about file patch_hdmi.c.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 15, 2021, 11:30:36 pm
Hey Lars,
The mismatch of using one mutex for locking and unlocking a different mutex. That surely led to a hang.
I am talking about file patch_hdmi.c.

Except we're not currently compiling patch_hdmi.c :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on January 16, 2021, 12:29:59 am
Hi Paul,

https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210115.zip - code bumped to linux 5.10.7 level - small changes as listed in http://trac.netlabs.org/uniaud/changeset/643

Same failure as the prior ones, however it does work with OS4.

Thanks!
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 16, 2021, 02:44:31 am
Quote
Whilst I'm pleased to hear it's working for you, I'm struggling to understand what's changed in the code to fix things <shrug>

Well, you can't argue with success. The change had something to do with selecting a second adapter. That seems to be where the problem was. I do need to use /A:1 to make it work.

One more step forward.  :)
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on January 16, 2021, 04:17:01 am
I'm getting weirdness at the start of playing some system sounds sometimes, like sounds staticy at first, other times plays fine.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 16, 2021, 09:00:09 am
HI All,

This hopefully doesn't break anything, and note I haven't committed the changes to SVN yet - but they will again help to simplify the code maintenance (assuming nothing breaks).

https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210116.zip

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 16, 2021, 09:29:50 am
Hi Paul

https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210116.zip

it works ok
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 16, 2021, 09:44:10 am
Hi Paul

https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210116.zip

it works ok

Thanks - if anything - it would break mixer controls. Virtualbox has decided not to give me sound right now, so I can't test effectively at the moment....
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 16, 2021, 10:01:07 am
Hi Paul

https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210116.zip

it works ok
if anything - it would break mixer controls

mixer controls   works ok too
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 16, 2021, 11:20:07 am
uniaud32-linux-5.10.7-20210116.zip

Works well here thanks
Mixer also still works well, as long as I remember to delete PMUniMix.ini after swapping cards..
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on January 17, 2021, 12:35:48 am
Latest 20211016 works well on my Lenovo S12 Yoga, but still only the lack of some controls.

By the way - i noticed that PM123 Version 1.41 does show covers of the Sounds played, see scond picture upper right corner - nice feature  :)



Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 17, 2021, 08:01:50 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210117.zip - just code cleanups - shouldn't change any behaviour....
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 17, 2021, 10:39:36 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210117.zip

it works ok here
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 17, 2021, 01:28:27 pm
uniaud32-linux-5.10.7-20210117.zip      Audio and mixer work well.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 17, 2021, 03:36:02 pm
 The 0117 version works well here too.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on January 17, 2021, 06:58:24 pm
Works the same as previous here, turn the master playback volume down and the beginning of a system sound sounds staticy and loud sometimes.
Leave volume on full and use speaker volume control and it sounds good.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 19, 2021, 08:20:31 am
FYI - no sound related changes in linux 5.10.8
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 19, 2021, 09:34:27 am
Hi All,

I'm trying to look into what's required to support multiple streams.

It appears that almost all modern hardware only supports a single stream in hardware - so other OS have a software mixer to mix the sounds before passing on to the hardware.

Uniaud32 logs record what the number of supported streams is, in a line like:
GetUniaudPcmCaps: pcm 0, cp2. nr of streams: 1

I'm interested to see if anyone has any hardware that has 'nr of streams' > 1 - most likely this would be SB Live! type hardware.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 19, 2021, 12:17:51 pm
 Hi Paul,

  Don't see anything like that in my alsahlp$ log, but maybe you can spot the right message? Alsahlp$ attached...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 19, 2021, 02:06:52 pm
Sorry, needs a debug driver....
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 19, 2021, 03:39:06 pm
Sorry, needs a debug driver....

Yep, I also figured that out on the 2nd card swap when I realised the logs didn't have much :o)
Lots of streams for you after I also swapped to the debug version :)
One Audigy card (Audigy SE [SB0570]) not checked as its in another computer at present.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 19, 2021, 11:33:59 pm
Sorry, needs a debug driver....

 Ah... I should have known. Both my desktop (HDA:ALC887-VD) and laptop (HDA:IDT 92HD81B1X5) have 3 instances of something like this:

GetUniaudPcmCaps: cp1. pcm 0, phandle f00a8af8
GetUniaudPcmCaps: pcm 0, cp2. nr of streams: 1
chan: from 2 to 2

 So 1 stream... Bummer City!

 Can stream mixing be added to the driver or does it need to be implemented thru MMOS/2 (if at all)?

Regards,
Title: Re: Updated Uniaud32 build
Post by: Neil Waldhauer on January 20, 2021, 01:12:13 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.7-20210117.zip - just code cleanups - shouldn't change any behaviour....

This build produces sound output on my Lenovo ThinkCentre M720s with the following sound card built-in.

Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xb1230000 irq 16
  mixer: Realtek ALC233
  componenets: HDA:10ec0235,17aa312a,00100002 HDA:8086280b,80860101,00100000

Line out makes sound. The volume control works.
Speaker does not make sound.
Microphone was not tested.

PMUniMix shows 5 volume controls: master volume and line-out volume work.

unimix - list shows 29 controls
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 20, 2021, 03:11:41 am
Hi Dave,
Can stream mixing be added to the driver or does it need to be implemented thru MMOS/2 (if at all)?

I don't know yet. In my head I'm thinking it would need to be a layer in between uniaud16 and uniaud32. linux seems to use dmix as a software mixer - so in my mind I'm thinking uniaud16 writes multiple streams to dmix, which writes a consolidated single stream to uniaud32.

I have no idea how difficult this would be to implement, I've never written a device driver - I port code, I don't write it :)

For largely academic interest purposes, I'd like to support the multiple streams on hardware that supports it natively - there should be some clues in the sblive driver on how this was don't on OS/2, which I'm hoping can be adapted into uniaud. I'm not sure yet if this will require uniaud16 and/or uniaud32 changes.

I'll explore as time permits.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 20, 2021, 09:44:32 am
For largely academic interest purposes, I'd like to support the multiple streams on hardware that supports it natively - there should be some clues in the sblive driver on how this was don't on OS/2, which I'm hoping can be adapted into uniaud. I'm not sure yet if this will require uniaud16 and/or uniaud32 changes.

As a minimum - changes to uniaud16 would be required for devices that support hardware mixing (ie sblive). This will need some quality time studying the sblive16 code and seeing how that code can be integrated into uniaud16 - and of course, other than making sure it doesn't fail on hardware that only supports a single stream, I can't test it locally.

For a software mixing solution, I have no real idea on how this could be done...
Title: Re: Updated Uniaud32 build
Post by: Andi B. on January 20, 2021, 10:58:21 am
Quote
For a software mixing solution, I have no real idea on how this could be done...
I have. From very simple (>> 1 to make half level and add every sample from both streams. Given they have the same sample rate) to sophisticated complex with filtering and ... :-). You will run into problems like clicking noise very fast. The same as you will when doing it in the analog world. I think you quickly will desire some fading and 'switching at (near) zero level'. And then volume control before mixing and mixing of different sample rates and ....

This all gets rather complex quickly. Given our OS will never be again a compatible sound processing platform and PC hardware in general isn't usable for any serious sound processing stuff, at least not the hardware we have drivers for, IMHO it's not worth to spend much time on this.

Probably I've lost track in this thread. But can someone tell me why we want to have 'mixing of channels'? And which channels do we want to mix?
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 20, 2021, 11:28:32 am
can someone tell me why we want to have 'mixing of channels'?

I'm wondering as well. I use two analog mixing consoles as I find real knobs easier to use and set, and if recent cards no longer have streams it means there is a limited audience for it.

In saying that :)

Paul, I'm happy to post you a SBLive PCI card at no cost to you. Not sure if that would help if your using a VPC.
alsa-sblive.log
 GetUniaudPcmCaps: cp1. pcm 0, phandle f5e0e4bc
  GetUniaudPcmCaps: pcm 0, cp2. nr of streams: 32
  chan: from 1 to 2
 GetUniaudPcmCaps: cp1. pcm 0, phandle f5e0eee8
  GetUniaudPcmCaps: pcm 0, cp2. nr of streams: 1
  chan: from 1 to 2
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 20, 2021, 09:30:15 pm
Hi Andi,

Quote
For a software mixing solution, I have no real idea on how this could be done...
I have. From very simple (>> 1 to make half level and add every sample from both streams. Given they have the same sample rate) to sophisticated complex with filtering and ... :-). You will run into problems like clicking noise very fast. The same as you will when doing it in the analog world. I think you quickly will desire some fading and 'switching at (near) zero level'. And then volume control before mixing and mixing of different sample rates and ....

This all gets rather complex quickly. Given our OS will never be again a compatible sound processing platform and PC hardware in general isn't usable for any serious sound processing stuff, at least not the hardware we have drivers for, IMHO it's not worth to spend much time on this.

Probably I've lost track in this thread. But can someone tell me why we want to have 'mixing of channels'? And which channels do we want to mix?

Sorry - I know there's libraries to merge different streams - dmix for example is used on some linux systems - when I say  'no real idea on how this could be done' I more mean the mechanics of *how* this could be done between uniaud16 and uniaud32.

In terms of the 'why' we might want this - one simple example is to mix in system notifications with a mp3 stream, rather than the annoying pause the mp3, play the system sound, resume the mp3 behaviour.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 20, 2021, 09:32:53 pm
Hi Ian,

can someone tell me why we want to have 'mixing of channels'?

I'm wondering as well. I use two analog mixing consoles as I find real knobs easier to use and set, and if recent cards no longer have streams it means there is a limited audience for it.

In saying that :)

Paul, I'm happy to post you a SBLive PCI card at no cost to you. Not sure if that would help if your using a VPC.

Thanks Ian - at this stage I have no hardware that has a PCI slot - all my physical hardware in the house is either a Thinkpad, or a variant of a Raspberry Pi.

Hold that thought for now - I've half thought to set up some physical hardware running ArcaOS - I just need to work out where I'd set this up with access to a display.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Andi B. on January 21, 2021, 09:03:51 am
Quote
...system notifications with a mp3 stream, rather than the annoying pause the mp3, play the system sound, resume the mp3 behaviour...
Thanks for the reminder. Understood now. I don't mind about such things probably cause I never ever let a system produce useless noise in my environment. But of course this is a personal habit.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 21, 2021, 11:42:53 pm
In terms of the 'why' we might want this - one simple example is to mix in system notifications with a mp3 stream, rather than the annoying pause the mp3, play the system sound, resume the mp3 behaviour.

 Exactly! This has been my pet peeve for OS/2 since the CompuServe days. Even Win 3.1 could handle this. SBLive cards handled it on OS/2 with Sander's driver, but PCI slots were last seen riding into the sunset. Guess audio chips nowadays tend not to support it directly, but rely on the OS to do mixing. Unless there are PCIe cards that support hardware multiple streams?....

Regards,
Title: Re: Updated Uniaud32 build
Post by: Lars on January 22, 2021, 12:14:14 am
But you can only easily mix PCM streams. If you have something else like MPEG you need to decode first. And if you want to do this in real time, you need a block oriented codec with a fixed and fairly small block size. I would think that mixing capable HW will mix the analog output signal. But that would require multiple analog internal outputs.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 22, 2021, 10:52:09 am
Very much thinking out loud.....

Looking at https://www.github.com/komh/kai/tree/master/libkai.txt it seems that KO's libkai 2.0 has a soft mixer.

I wonder if uniaud16.sys could be adapted to call libkai.dll rather than uniaud32 sys? I have no idea if it's even possible for a 16 but mmpm driver to call a 32 bit DLL, as I say, just thinking out loud for now....

Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 22, 2021, 07:28:41 pm
Hi Paul.

FYI: I have cleaned up the UniAud tickets on the netlabs SVN. (Here: http://trac.netlabs.org/uniaud/report/1[/i][/i] (http://trac.netlabs.org/uniaud/report/1))
The very old ones are closed and I kept the ones that had more recent feedback from the submitter. There are only 7 tickets now.

Regards
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 23, 2021, 04:50:07 am
one simple example is to mix in system notifications with a mp3 stream, rather than the annoying pause the mp3, play the system sound, resume the mp3 behaviour.

Interesting, I forgot all about system sounds, do not think I've had those turned on since the later 90's.
I clicked on the 'Enable system sounds', clicked on Play (for test sound) button and got a sound with an MP3 still playin in pm123 ok but the second time I did it there was no sound, and the MP3 though still playing was silent. I could stop, and restart playing the MP3 but nothing I could do from then on resulted in any sounds. I unticked 'Enable system sounds', exited, closed pm123, then restarted pm123 and had audio back again. I find it interesting that I was able to at least play a system sound once without affecting the mp3 stream so there appears to be some sort of mixing happening though broken. This is with the SB Audigy 4 [SB0610] card. Will find the time to do more experimenting later.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 24, 2021, 09:28:41 am
New build - https://smedley.id.au/tmp/uniaud32-linux-5.10.10-20210124.zip

No significant changes - mainly updated some libraries to the latest kernel (rbtree, regmap, regcache) and some code cleanups from AlexT

Full details at http://trac.netlabs.org/uniaud/timeline
Title: Re: Updated Uniaud32 build
Post by: OS4User on January 24, 2021, 02:46:54 pm
Uniaud32 logs record what the number of supported streams is, in a line like:
GetUniaudPcmCaps: pcm 0, cp2. nr of streams: 1

I'm interested to see if anyone has any hardware that has 'nr of streams' > 1 - most likely this would be SB Live! type hardware.

I have  got SB Audigy 1 [SB0092]. It reports:

Code: [Select]
GetUniaudPcmCaps: cp1. pcm 0, phandle e494dc84
GetUniaudPcmCaps: pcm 0, cp2. nr of streams: 32

5.10.10 works ok
Title: Re: Updated Uniaud32 build
Post by: Alex Taylor on January 24, 2021, 06:37:18 pm
No significant changes - mainly updated some libraries to the latest kernel (rbtree, regmap, regcache) and some code cleanups from AlexT

?? ???
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 24, 2021, 07:46:36 pm
No significant changes - mainly updated some libraries to the latest kernel (rbtree, regmap, regcache) and some code cleanups from AlexT

?? ???
Hi Alex.  On the IRC there is a user called AlexT on the some OS/2 channels.  I guess it is not you (Alex Taylor) because you usally go on IRC with the "altsan" nickname, right?
Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 24, 2021, 09:34:26 pm
No significant changes - mainly updated some libraries to the latest kernel (rbtree, regmap, regcache) and some code cleanups from AlexT

?? ???
Hi Alex.  On the IRC there is a user called AlexT on the some OS/2 channels.  I guess it is not you (Alex Taylor) because you usally go on IRC with the "altsan" nickname, right?
Regards

Correct.
Title: Re: Updated Uniaud32 build
Post by: Alexander on January 25, 2021, 02:27:47 am
Hi!
First, thank you very much for your great work on uniaud!

I have sblive CT4830 4-channels card. It supports multistream and it really works with old uniaud.

I tried uniaud32-linux-5.10.10-20210124 , no traps , but no sound at all, pm123 says mci error 5134: no device driver found
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 25, 2021, 03:18:30 am
New build - https://smedley.id.au/tmp/uniaud32-linux-5.10.10-20210124.zip
Hi Paul

I tested it on the Thinkpad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420).
- Works. No trap. Master volume working. Works good here.
Thanks
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 25, 2021, 05:59:45 am
Hi!
First, thank you very much for your great work on uniaud!

I have sblive CT4830 4-channels card. It supports multistream and it really works with old uniaud.

I tried uniaud32-linux-5.10.10-20210124 , no traps , but no sound at all, pm123 says mci error 5134: no device driver found

Check mixer controls with pmunimix.

Card seems to be detected:
Sound card name = SB Live! Value [CT4830]
...
pcm_info()
00-00: emu10k1 : ADC Capture/Standard PCM Playback
 : playback 32
 : capture 1
...

What does unimix -card show?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 25, 2021, 10:00:54 pm
Hi All,

I have one report (and can confirm on a T60 locally) of sound not working after a warm boot (ie ctrl-alt-del or setboot /b). The T60 has Analog Devices HDA hardware. I'm wondering if anyone else is seeing similar behaviour. Sound works 100% of the time after a cold boot.

Thanks,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 25, 2021, 10:18:17 pm
 Paul,

 Version uniaud32-linux-5.10.10-20210124 works just fine here on my laptop and desktop, cold or warm boot...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 25, 2021, 11:36:48 pm
Hi David,
Version uniaud32-linux-5.10.10-20210124 works just fine here on my laptop and desktop, cold or warm boot...

Thanks for confirming - at this stage, it seems to be something unusual with the Analog Devices hardware...

New build - https://smedley.id.au/tmp/uniaud32-linux-5.10.10-20210126.zip has some code cleanups to remove some OS/2 specific workarounds in alsa-kernel/core/oss/pcm_oss.c, relating to calculations for min and max buffer sizes.

Seems to work ok here in minimal testing.

I'm slowly working my way through my //FIXME comments in the code, where in the interests of speed, I reverted to old code rather than spend time implementing kernel functionality.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 26, 2021, 08:24:59 am
Hi Paul,

I have one report (and can confirm on a T60 locally) of sound not working after a warm boot (ie ctrl-alt-del or setboot /b). The T60 has Analog Devices HDA hardware. I'm wondering if anyone else is seeing similar behaviour. Sound works 100% of the time after a cold boot.

Both
  uniaud32-linux-5.10.10-20210124.zip and
  uniaud32-linux-5.10.10-20210126.zip
work fine here from cold and warm boots at least with SB Audigy 4 [SB0610].
Title: Re: Updated Uniaud32 build
Post by: David McKenna on January 26, 2021, 10:29:19 pm
 Paul,

  Uniaud32-linux-5.10.10-20210126 is working great here too...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on January 27, 2021, 10:02:42 pm
20210126 works fine with the Yoga 12.

thumbs up!!
Title: Re: Updated Uniaud32 build
Post by: Alexander on January 28, 2021, 08:18:44 am
I have sblive CT4830 4-channels card. It supports multistream and it really works with old uniaud.
Check mixer controls with pmunimix.
I doubt that mixer controls can lead to msi error

Ok, now i found, whats wrong, its very old uniaud16
I take this...
::@@Universal Audio MMPM/2 Driver for OS/2
Vendor:          Netlabs
Revision:        1.9.7
Date/Time:       25 Jun 2017 17:11:40
Build Machine:   DAZAR1
File Version:    1.9.7
Description:     Universal Audio MMPM/2 Driver for OS/2
...and with it sound works ok with uniaud32-linux-5.10.10-20210126.zip

But mutistream is not working
I attach several logs:
uniaud0 - after boot, no wps sounds
uniaud1 - warpvision playing
uniaud2 - now i started pm123 and in says mci err 5613: insufficient buffer space specified
uniaud3 - warpvision closed and pm123 playing ok

Ps With old uniaud warpvision and pm123 playing simultaneously without any errors
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 28, 2021, 08:31:26 am
Hi Alexander,

But mutistream is not working
I attach several logs:
uniaud0 - after boot, no wps sounds
uniaud1 - warpvision playing
uniaud2 - now i started pm123 and in says mci err 5613: insufficient buffer space specified
uniaud3 - warpvision closed and pm123 playing ok

Ps With old uniaud warpvision and pm123 playing simultaneously without any errors

Can you repeat the above test with https://smedley.id.au/tmp/uniaud32-20210128-Alexander.zip ? It reverts a change which allowed HDA hardware to produce sound, and which (I think) might be the cause of the issues with stream interruptions.

NB: for anyone else following along, the above build will NOT work with HDA hardware.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Alexander on January 28, 2021, 09:22:46 am
Can you repeat the above test with https://smedley.id.au/tmp/uniaud32-20210128-Alexander.zip ? It reverts a change which allowed HDA hardware to produce sound, and which (I think) might be the cause of the issues with stream interruptions.
No, the same result
Title: Re: Updated Uniaud32 build
Post by: Alexander on January 28, 2021, 09:28:20 am
This old uniaud version 1.1
uniaud16 says 1.1.2
uniaud32 - 1.1.0
29.12.04   2:20      61554             0  uniaud16.sys
29.12.04  19:52     479694            54  uniaud32.sys
ftp://icn@79.135.223.10/os2soft/uniaud/uniaud_latest.zi6
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 28, 2021, 09:36:12 am
I see what Alexander means about mutistream not working.
I normally only play multimedia one at a time :)
If I play an mp3, or a movie via mplayer, then try to play something via pm123 I get "MCI Error 5613: Insufficient buffer size specified."
If I stop mplayer then pm123 works fine.
If I start playing music via pm123, then start something in mplayer, mplayer doesnt complain it just doesnt make any sound.

Tried it with uniaud32-20210128-Alexander.zip and there is no difference.
Log attached.

PS, I'm using uniaud16.sys (dated 25th Jun 2017, revision 1.09.7) which I assumed is the latest 16 driver.
PPS, thought I would try system sounds again. They work fine until I use pm123 or another mm application, then they nolonger work - ever. If I warm boot then test the system sounds (works), then start pm123, click on the system sounds Test button, I get no system sound. If I click a second time on the Test button, the audio from pm123 stops. To get audio back for pm123 I need to restart pm123. <== with uniaud32-20210128-Alexander.zip
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on January 28, 2021, 03:20:47 pm
Hi Ian, Alexander.

Do you think this "multistream issue" is related (or the same) as this ticket? "Shared mode doesn't work (http://trac.netlabs.org/uniaud/ticket/210)"
Or are different things?

Regards
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 28, 2021, 04:02:04 pm
Hi Martin,

Speaking only for me, yes, I would say the issue is related if not the same. Though I'm not angry about it :)
I think the system sounds issue is also part of the same issue, ie, there is no input mixing happening.
As to a MM application resuming audio, or having to restart it I think would depend on the programming behind that application.

PS, same issue, installed Daniel's Mahjongg - also 100% CPU after pm123 audio stops about 5 seconds after starting Mahjongg. When I quit Mahjongg, take about another 5 seconds then pm123 audio starts again. Tried it a second time, pm123 is still playing ok start Mahjongg but there is no sound from Mahjongg.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 29, 2021, 08:39:10 am
Hi Guys,

Do you think this "multistream issue" is related (or the same) as this ticket? "Shared mode doesn't work (http://trac.netlabs.org/uniaud/ticket/210)"
Or are different things?

Not to contradict Ian, but I think they're related. Whilst Ian and Alexander have *hardware* that is capable of multiple streams, uniaud16 doesn't currently support it (noting that uniaud 1.1.2 seems to have supported multiple streams, and the code was commented out.

Shared mode is the alternative when multiple streams are NOT supported.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 29, 2021, 10:21:50 am
Hi Paul,

I think I'm looking at it from an Audio path perspective, your looking at it from a programming perspective :)

Programming how multiple streams are used and allocated and/or selecting a shared mode for inputs could very easily get complex in programming. Might be why the code was commented out as I suspect it would be easier to understand using a mixing console as a representation, a bit like how its done under Linux with Jack, Pulse Audio, and the MIDI connection PATCH panel which is neat. In the end it comes down to audio inputs from different physical and/or virtual sources, and what to do with them.

As Paul says, streams are to do with the inputs from the hardware, shared mode is the software side of it which I assume simply throws it altogether into the one L or R audio 'stream' as a basic software mixer. From what you've said this means that the shared mode is what is not working, and I've made the assumption that its all in the same code location.

I was thinking of also remming out Timidity and seeing how it worked with the MIDI drivers but haven't gone that far yet, now I think about the software, not sure if that's even important.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 29, 2021, 10:56:52 am
Actually, I think I disagree with myself now, after re-reading the ticket.

To me, shared mode was describing an brief interruption to an MP3 stream as a result of a system sound being played. There are reports that this is broken currently - ie the MP3 doesn't resume.

When I re-read the ticket, I think they're asking for sound mixing where (per the ticket) the Klondike sounds would play over the MP3 without any interruption a - ie a software version of hardware like the SB live that supports multiple streams. Afaik no version of uniaud has ever supported this on hardware that only supports a single stream (ie 99% of hardware).

I'll look into a uniaud16 build that re-enables the code for multiple streams over the weekend.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 31, 2021, 10:10:12 am
Hi All,

https://smedley.id.au/tmp/uniaud32-linux-5.10.11-20210131.zip

Fixed a problem in hda_codec.c that allowed me to remove a dirty hack from lib32/sound.c

This *might* help fix a reported problem with system sounds interrupting a MP3 stream and not resuming correctly, which is a regression from the 2.02 driver.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on January 31, 2021, 11:45:28 am
Hi Paul,

uniaud32-linux-5.10.11-20210131.zip
Installed, logs attached, one taken directly after a fresh boot, the other after the fault condition.

If I'm playing an MP3 file using pm123, then play a system sound (test was moving emails in pmmail folders) or via the Sound panel in System Setup, the mp3 will keep playing for x number of seconds before going silent. After this event there is no mp3 or other audio output. I can get the mp3 audio back by restarting play (without restarting pm123) but the system sounds are gone for good. Before playing the mp3, system sounds worked fine.

Bit like the pm123 buffer can no longer pull data because OS/2 has control of the queue after attempting to play a system sound.

PS, Apart from that, it works fine :)
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on January 31, 2021, 09:33:16 pm
Quote
This *might* help fix a reported problem with system sounds interrupting a MP3 stream and not resuming correctly, which is a regression from the 2.02 driver.

I find that it takes more system sounds to trigger the problem, but it still happens.

Run PM123, and play something. then open a window that has scroll bars. Click to move out of the window. It makes a sound. try that a few more times, allowing the sound to finish. That seems to work, at least for a few tries. Now, click fast enough to start a new sound, before the first sound has finished playing, and it will lock the desktop. The mouse won't even move. Reset is required.

Now, an interesting thing: If I restart the desktop (without locking it), sounds come back. Sometimes, it seems to be okay (until the next restart). Sometimes, it is worse than before.

When it locks, there is a POPUPLOG associated:
Code: [Select]
01-31-2021  12:51:57  SYS3175  PID 01b1  TID 0002  Slot 0060
C:\OS2\PMSHELL.EXE
c0000005
1f0aa7a5
P1=00000001  P2=00000008  P3=XXXXXXXX  P4=XXXXXXXX 
EAX=00000002  EBX=00000004  ECX=00000000  EDX=004bb82c
ESI=00000004  EDI=00381be8 
DS=0053  DSACC=d0f3  DSLIM=9fffffff 
ES=0053  ESACC=d0f3  ESLIM=9fffffff 
FS=150b  FSACC=00f3  FSLIM=00000030
GS=0000  GSACC=****  GSLIM=********
CS:EIP=005b:1f0aa7a5  CSACC=d0df  CSLIM=9fffffff
SS:ESP=0053:004bb8d8  SSACC=d0f3  SSLIM=9fffffff
EBP=004bb910  FLG=00010206

MMPM.DLL 0001:0007a7a5

There is also a process dump, which shows this:
Code: [Select]
IBM OS/2 Dump Formatter for a retail or an hstrict SMP kernel.
Formatter is --> Internal revision 14.106_SMP
Dump file is --> Internal revision 14.202_SMP (process dump)
 
Symbol (c:\os2\pdpsi\pmdf\warp45_s\os2krnlr.sym) linked
Symbol (c:\os2\pdpsi\pmdf\warp45_s\pmctls.sym) linked
Symbol (c:\os2\pdpsi\pmdf\warp45_s\pmwp.sym) linked
Symbol (c:\os2\pdpsi\pmdf\warp45_s\pmspl.sym) linked
Symbol (c:\os2\pdpsi\pmdf\warp45_s\pmviop.sym) linked
Symbol (c:\os2\pdpsi\pmdf\warp45_s\pmmerge.sym) linked
Symbol (c:\os2\pdpsi\pmdf\warp45_s\doscall1.sym) linked
 
Current slot number: 0060
 
 Slot  Pid  Ppid Csid Ord  Sta Pri  pTSD     pPTDA    pTCB     Disp SG Name
*0060# 01b1 001d 01b1 0002 run 0800 f705f000 fcf477d0 fce91b98 06c4 1a PMSHELL
 
eax=00000002 ebx=00000004 ecx=00000000 edx=004bb82c esi=00000004 edi=00381be8
eip=1f0aa7a5 esp=004bb8d8 ebp=004bb910 iopl=0 rf -- -- nv up ei pl nz na pe nc
cs=005b ss=0053 ds=0053 es=0053 fs=150b gs=0000 cr2=00000000 cr3=00000000 p=00
005b:1f0aa7a5 ff7108         push      dword ptr [ecx+08]   ds:00000008=invalid
#
Loading version data ... please wait
Loading Process Information ... please wait


Loading structure info for WSeB/ACP SMP ... please wait.
....Structure definition file (warp45_s\kernel.sdf) loaded.
....Structure definition file (warp45_s\shell.sdf) loaded.
Done.
 
#
                        TRAP SCREEN INFORMATION


------------------------------------------------------------
Trap screen 1 found at address #70:989c
This screen is most likely related to a previous trap.

P1=00000001  P2=00000008  P3=XXXXXXXX  P4=XXXXXXXX   
CS:EIP=005b:1f0aa7a5  CSACC=d0df  CSLIM=9fffffff
SS:ESP=0053:004bb8d8  SSACC=d0f3  SSLIM=9fffffff
EBP=004bb910  FLG=00010206
EAX=00000002  EBX=00000004  ECX=00000000  EDX=004bb82c
ESI=00000004  EDI=00381be8   
DS=0053  DSACC=d0f3  DSLIM=9fffffff   
ES=0053  ESACC=d0f3  ESLIM=9fffffff   
FS=150b  FSACC=00f3  FSLIM=00000030
GS=0000  GSACC=****  GSLIM=*******
 
No Symbols Found
#

------------------------------------------------------------
Trap screen 2 found at address #70:9c84

P1=00000001  P2=00000008  P3=XXXXXXXX  P4=XXXXXXXX   
EAX=00000002  EBX=00000004  ECX=00000000  EDX=004bb82c
ESI=00000004  EDI=00381be8   
DS=0053  DSACC=d0f3  DSLIM=9fffffff   
ES=0053  ESACC=d0f3  ESLIM=9fffffff   
FS=150b  FSACC=00f3  FSLIM=00000030
GS=0000  GSACC=****  GSLIM=********
CS:EIP=005b:1f0aa7a5  CSACC=d0df  CSLIM=9fffffff
SS:ESP=0053:004bb8d8  SSACC=d0f3  SSLIM=9fffffff
EBP=004bb910  FLG=0001020

ASCII found at #70:9c84-12 is c0000005 (Exception Code?)
c0000005:  System Encountered an access violation

ASCII found at #70:9c84-9 is 1f0aa7a5. (Linear Address?)

ASCII found after trap screen: (Module name Object:Offset?)
MMPM.DLL 0001:0007a7a5
No Symbols Found

 
#

The dump is about 2.5 GB. I can get it to you, if you think it will help.

I am doing this on my Asus A88XN-A machine (AMD A6 processor), but I get similar results on my Lenovo ThinkPad L530 (Intel based).
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 01, 2021, 03:12:52 am
Hi Doug,

Quote
This *might* help fix a reported problem with system sounds interrupting a MP3 stream and not resuming correctly, which is a regression from the 2.02 driver.

I find that it takes more system sounds to trigger the problem, but it still happens.

Thanks for the response and the notes on how to reproduce. I'll try and do some testing locally this week - and also see if I can reproduce with the 2.02.06 driver.

No need for a dump right now - I'm not sure I'd know what to do with it anyway :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 01, 2021, 07:47:00 am
Hi Ian,

uniaud32-linux-5.10.11-20210131.zip
Installed, logs attached, one taken directly after a fresh boot, the other after the fault condition.

Interesting, this bit:
Code: [Select]
channels: 1, period size: 2c0
channels selected: 1
HWP: SR rate 22050, BPS 16, CH 1, PRSZ 160, periods 5c
 Error -77 from first IOCTL HW Parms
channels: 1, period size: 2c0
channels selected: 1
HWP: SR rate 22050, BPS 16, CH 1, PRSZ 160, periods 5c
Error -77 second time.. Bailing
U16: StartStream: ConfigDev failed!!
U16: u32 wrap: BytesProcessed=3EE18C LastPosition=3EE18C Bytes=1F76A4 Delta=1F76A4 InProcess=0 Done=1

Looks 'interesting'!
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on February 01, 2021, 05:43:29 pm
Hi Paul,

https://smedley.id.au/tmp/uniaud32-linux-5.10.11-20210131.zip

Fixed a problem in hda_codec.c that allowed me to remove a dirty hack from lib32/sound.c

This *might* help fix a reported problem with system sounds interrupting a MP3 stream and not resuming correctly, which is a regression from the 2.02 driver...

Whoa!!!

The BIG DEAL here is the fact that this is the 1st newer release which does NOT HANG during boot!!!

Thank you sir...I've been testing each new release with a sense of dread given the results trend so far...but, this is a good one.

Whatever that tweak was, it did fix the failure here!  :-* (because you deserve it - LOL)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 01, 2021, 09:39:30 pm
Hi Dariusz

https://smedley.id.au/tmp/uniaud32-linux-5.10.11-20210131.zip

Fixed a problem in hda_codec.c that allowed me to remove a dirty hack from lib32/sound.c

This *might* help fix a reported problem with system sounds interrupting a MP3 stream and not resuming correctly, which is a regression from the 2.02 driver...

Whoa!!!

The BIG DEAL here is the fact that this is the 1st newer release which does NOT HANG during boot!!!

Thank you sir...I've been testing each new release with a sense of dread given the results trend so far...but, this is a good one.

Whatever that tweak was, it did fix the failure here!  :-* (because you deserve it - LOL)

Wow, that was unexpected! I'm struggling to understand how the change avoided a hang - but if it worked - great :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Alexander on February 02, 2021, 03:33:03 pm
Hi!

With uniaud32-linux-5.10.11-20210131.zip results the same, sound working , but no multistream
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 03, 2021, 03:25:42 am
Hi Alexander

With uniaud32-linux-5.10.11-20210131.zip results the same, sound working , but no multistream

Based on IRC - you had working multistream with an old uniaud16.sys and uniaud32.sys

Are you able build date/version numbers of the 'working' uniaud16.sys and uniaud32.sys so I can try and ascertain the relevant SVN revision and see what changes have been made since.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Alexander on February 03, 2021, 12:19:36 pm
Based on IRC - you had working multistream with an old uniaud16.sys and uniaud32.sys

Are you able build date/version numbers of the 'working' uniaud16.sys and uniaud32.sys so I can try and ascertain the relevant SVN revision and see what changes have been made since.
29.12.04   2:20      61554    uniaud16.sys
29.12.04  19:52     479694  uniaud32.sys

uniaud16:
Signature:       @#"OS/2 Universal Audio Driver Version":1.01.002#@
Vendor:          "OS/2 Universal Audio Driver Ve
Revision:        1.1
File Version:    1.1
Description:

On boot it says:
OS/2 Universal Audio MMPM/2 Driver v1.1.2
Copyright 2000-2002 InnoTek Systemberatung GmbH
Copyright 2002 IBM Corporation
Maintained by http://os2.kiev.ua/en/uniaud.php


uniaud32 have not version strings
on boot says:
OS/2 Universal Audio Core Driver v1.1.0
Copyright 2000-2002 InnoTek Systemberatung GmbH
Copyright 2000-2002 The ALSA Project

It can be downloaded from
https://filebin.net/06736c66fezgbduw/uniaud_latest.zip?t=ouzf69qm



Title: Re: Updated Uniaud32 build
Post by: Remy on February 05, 2021, 11:33:51 am
Hi All,

https://smedley.id.au/tmp/uniaud32-linux-5.10.11-20210131.zip

Fixed a problem in hda_codec.c that allowed me to remove a dirty hack from lib32/sound.c

This *might* help fix a reported problem with system sounds interrupting a MP3 stream and not resuming correctly, which is a regression from the 2.02 driver.

Cheers,

Paul

Hi Paul,

I give a try to your latest build and it is the best off all I could try (now using it in replace of previous one).
No ff crash seen, better switch when using PM123 and running a youtube video and going back to pm123 etc... Very good
Definitively the best build.

Regards
Rémy 
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 05, 2021, 09:15:35 pm
Thanks Remy - great to hear!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 27, 2021, 09:06:17 am
Updated build with more cleanups from AlexT from #os2russian that are not yet committed - diff is included:
https://smedley.id.au/tmp/uniaud32-linux-5.10.17-20210227.zip

Feedback appreciated.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on February 27, 2021, 04:09:49 pm
 Hi Paul,

  Been using 5.10.17 for a couple hours without any issues with my ALC887-VD sound chip.... thanks to you and AlexT!

Regards,
Title: Re: Updated Uniaud32 build
Post by: TeLLie on February 27, 2021, 04:15:21 pm
Hi
Here it also works on my W530

Thankz
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on February 27, 2021, 05:54:03 pm
Paul,

https://smedley.id.au/tmp/uniaud32-linux-5.10.17-20210227.zip

Still rockin' here, no sight of previous boot hang issues, so this would be the 2nd drop that's been alright in that respect.

Thank you!
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on February 27, 2021, 07:43:26 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.17-20210227.zip
Thanks Paul.

It keeps working fine here on the ArcaOS 5.0.6 VirtualBox VM (Intel HD Audio) on W10 host on the TP Yoga 14.
- Sound, Master Volume working, No Trap.

Paul, can you please check this ticket suggestion to change the  /V "boot messages" of the driver: http://trac.netlabs.org/uniaud/ticket/246 (http://trac.netlabs.org/uniaud/ticket/246). I guess it should not be a complex thing, but it is up to you.

Regards
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on February 28, 2021, 06:13:30 am
uniaud32-linux-5.10.17-20210227.zip working well here with pm123, mplayer and timidity on SB cards.
Title: Re: Updated Uniaud32 build
Post by: mauro on February 28, 2021, 08:36:07 am
Hi all, sorry to ask this question, can I get installation tips somwhere or is it just un upgrade to a previous installed version ?  If so, which is the uniaud.diff  placing ?
I'm asking because previously -months ago-  I installed uniaud on my OS2 Warp 4.52 Virtualbox machine trying to solve the issue of no sound coming from youtube video in Firefox 45.9, whith the result of not solving this .
By reading these posts I'm understanding that after the update a new attempt  could be done.
Thank you
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on February 28, 2021, 08:43:10 am
Hi Mauro,

Hi all, sorry to ask this question, can I get installation tips somwhere or is it just un upgrade to a previous installed version ?  If so, which is the uniaud.diff  placing ?
I'm asking because previously -months ago-  I installed uniaud on my OS2 Warp 4.52 Virtualbox machine trying to solve the issue of no sound coming from youtube video in Firefox 45.9, whith the result of not solving this .
By reading these posts I'm understanding that after the update a new attempt  could be done.

The diff is purely to comply with the GPL license - it shows the source code differences between the current code in SVN and this build.

To update a working system:
1) backup \mmos2\uniaud32.sys
2) copy a release\uniaud32.sy* to \mmos2
3) reboot
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on February 28, 2021, 02:29:13 pm
Updated build with more cleanups from AlexT from #os2russian that are not yet committed - diff is included:
https://smedley.id.au/tmp/uniaud32-linux-5.10.17-20210227.zip

Feedback appreciated.

Hi Paul,
it's working on my T430 flawless. Great work. Will test on my X250 tomorrow and let you know.
Greetings from Potsdam / Germany,
Mike
Title: Re: Updated Uniaud32 build
Post by: Neil Waldhauer on February 28, 2021, 04:27:55 pm
Updated build with more cleanups from AlexT from #os2russian that are not yet committed - diff is included:
https://smedley.id.au/tmp/uniaud32-linux-5.10.17-20210227.zip

Feedback appreciated.

Cheers,

Paul

This works on ThinkCentre M720s. The Jan 24 build also works. Thanks!!
Title: Re: Updated Uniaud32 build
Post by: mauro on February 28, 2021, 06:04:44 pm
Hi Mauro,

To update a working system:
1) backup \mmos2\uniaud32.sys
2) copy a release\uniaud32.sy* to \mmos2
3) reboot

Ok thank you Paul, did as above. On both ICH AC97 (first) and Intel HD item (after) selected in Virtualbox audio system, everything like before; audio events are well working except sound from FF video play (Youtube). Almost sure it is not an audio driver matter.
So I would say this Uniaud versio is working on my virtual machine system
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on February 28, 2021, 06:52:51 pm
Hi Mauro, 2 things to try.adding "SET KAI_AUTOMODE=DART" (without quotes) in your config.sys, reboot and test.
Download PMUniMix from Hobbes and play with the volume sliders. Sometimes the mixer is not set up correctly and something like the front volume needs turning up.
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on February 28, 2021, 07:23:32 pm
20210227 works fine with the Thinkpad S12 Yoga, thanks, Paul
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on March 01, 2021, 10:22:01 am
Updated build with more cleanups from AlexT from #os2russian that are not yet committed - diff is included:
https://smedley.id.au/tmp/uniaud32-linux-5.10.17-20210227.zip

Feedback appreciated.

Hi Paul,
it's working on my T430 flawless. Great work. Will test on my X250 tomorrow and let you know.
Greetings from Potsdam / Germany,
Mike

Hi Paul,

I'm happy to report that the driver not only works on my X250, but it is first time that I can can also use the output line from my docking station.
Really great - Thanks a lot for all your work.

Greetings from Potsdam / Germany,
Mike
Title: Re: Updated Uniaud32 build
Post by: mauro on March 01, 2021, 12:35:06 pm
Hi Mauro, 2 things to try.adding "SET KAI_AUTOMODE=DART" (without quotes) in your config.sys, reboot and test.
Download PMUniMix from Hobbes and play with the volume sliders. Sometimes the mixer is not set up correctly and something like the front volume needs turning up.

Hi Dave, SET KAI_AUTOMODE=DART was already active in config.sys, then tested Uniaud mixer again (tried also previously as you gave me this tip), activating all audio device controls including front, video, aux, pcm  and their cursors .  All as before.
Thank you
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 05, 2021, 08:36:36 am
Code update to 5.10.20 - https://smedley.id.au/tmp/uniaud32-linux-5.10.20-20210305.zip
Title: Re: Updated Uniaud32 build
Post by: Lars on March 05, 2021, 12:53:56 pm
Hi Mauro, 2 things to try.adding "SET KAI_AUTOMODE=DART" (without quotes) in your config.sys, reboot and test.
Download PMUniMix from Hobbes and play with the volume sliders. Sometimes the mixer is not set up correctly and something like the front volume needs turning up.

Hi Dave, SET KAI_AUTOMODE=DART was already active in config.sys, then tested Uniaud mixer again (tried also previously as you gave me this tip), activating all audio device controls including front, video, aux, pcm  and their cursors .  All as before.
Thank you

Have you installed libkai ? If not, you should, FF will not play anything if libkai is not installed.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 20, 2021, 10:17:57 pm
https://www.arcanoae.com/uniaud-audio-driver-package-updated-20210320/
Title: Re: Updated Uniaud32 build
Post by: David McKenna on March 20, 2021, 11:54:09 pm
 Paul,

  Thanks for the head's-up... it is working well here. One thing I notice - with your build on my machine, the driver uses IRQ16, but David A's version uses IRQ55. Does he add some secret sauce for that? Is there an advantage to a higher IRQ?

 Regards,

 P.S. Your Karma is up to 50! Can you levitate?
Title: Re: Updated Uniaud32 build
Post by: Olafur Gunnlaugsson on March 21, 2021, 11:59:01 am
Paul,

  Thanks for the head's-up... it is working well here. One thing I notice - with your build on my machine, the driver uses IRQ16, but David A's version uses IRQ55. Does he add some secret sauce for that? Is there an advantage to a higher IRQ?

Yes, in that AT compatibles had 15 interrupts available while early APIC based system had 24 interrupts, with the extra 9 interrupt lines being handled dynamically in a way that could cause problems with certain drivers and hardware, in some cases drivers either did not see any interrupt above 15 or saw any interrupt above 15 a IRQ16 even though the hardware itself supported higher IRQ's. Modern MSI based APIC systems have 224 interrupt lines available, BTW.

A number of old OS/2 drivers will only work correctly with interrupts <16 or <25 and a number of them showing problems with IRQ's higher than 15. While sharing an interrupt is possible both from a hardware and software perspective it can have side effects both on the software side and with the hardware itself such as the interrupt switching modulating the analogue gain stages on a soundcard or a video card with analogue (VGA) outputs. And there are some other issues, drivers like the old Martox ones being quite aggressive in claiming interrupts, clashing with other dirvers and sometimes with the OS itself.

By moving as much of the modern drivers/hardware to interrupts higher than 24 you are simply leaving more space free to legacy drivers and hardware to play with, so to speak, minimising the possibility of software/hardware clashes.
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on March 21, 2021, 03:38:12 pm
Hi Paul,

https://www.arcanoae.com/uniaud-audio-driver-package-updated-20210320/

Since AN has done an official release, does that mean we are at a somewhat "settled/stable" code base?

I ask b/c if you recall I kept on running into boot hang issues. Those went away with the uniaud32-linux-5.10.11-20210131.zip release, however something else appeared.

I figured I would continue to test to see if maybe I was seeing some side-effect specific to another application, a weird conflict, etc. Well, it turns out the same behaviour has been present with the other remaining releases as well, that being:

1) uniaud32-linux-5.10.17-20210227.zip
2) uniaud32-linux-5.10.20-20210305.zip

The same symptom is present in the AN release, but they officially won't work the issue unless you are running AOS as well, so that's a dead end.

Anyways, the problem I am runing into is a complete hard-lock of the system when playing back MP4 videos with mplyer. Haven't narrowed it down any further, it may be a conflict specifically with mplayer, or it may be just the audio driver that's hit during such playback. Each time it requires a power OFF/ON fix, so I'm reluctant to continue to troubleshoot in the dark.

Having said that, on my hardware platform the last good working driver is uniaud32-linux-3.16.85-20200812.zip, so I will continue to use that as it does not appear to have any functionality shortcomings.

As always, appreciate all the work you've done for us here. If you are willing and want to continue to debug my problem I'm happy to do so, but otherwise if that affects only a small subset of hardware configs out there, we can probably pass.

Here is my hardware config:

Code: [Select]
Card info for adapter 0:
  num: 0
  id: SB
  driver: HDA-Intel
  name: HDA ATI SB
  longname: HDA ATI SB at 0xfe3f8000 irq 16
  mixer: Realtek ALC889
  componenets: HDA:10ec0889,14627640,00100004
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 21, 2021, 09:22:12 pm
Hi David,
  Thanks for the head's-up... it is working well here. One thing I notice - with your build on my machine, the driver uses IRQ16, but David A's version uses IRQ55. Does he add some secret sauce for that? Is there an advantage to a higher IRQ?
No idea - he must be doing something different - whatever it is, doesn't appear to be in the GPL code...

P.S. Your Karma is up to 50! Can you levitate?

hahahah not that I'm aware of :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 21, 2021, 09:47:40 pm
Updated build: https://smedley.id.au/tmp/uniaud32-linux-5.10.25-20210322.zip

Updated linux code to 5.10.25, applied David's changes from r680
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on March 22, 2021, 02:09:50 pm
Updated build: https://smedley.id.au/tmp/uniaud32-linux-5.10.25-20210322.zip

Updated linux code to 5.10.25, applied David's changes from r680

Hi Paul,

latest version from you is working on my TP X250. I need to check on my TP T430s later this week.
Thank for your great effort.

Mike
Title: Re: Updated Uniaud32 build
Post by: David Graser on March 22, 2021, 06:17:31 pm
My thinkpad T430S works fine with it.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on March 26, 2021, 12:56:09 am
 Working well here too - desktop and laptop... thanks again!

 Regards,
Title: Re: Updated Uniaud32 build
Post by: Remy on April 06, 2021, 08:10:28 pm
Updated build: https://smedley.id.au/tmp/uniaud32-linux-5.10.25-20210322.zip

Updated linux code to 5.10.25, applied David's changes from r680

Tested, works but found a "regration"
- playing an audio file through PM123 (mp3)
- open firefox and watch a video  (video sound is on and PM123 silent
- stop play on the video (no sound under pm123)
- close firefox = no sound under pm123 (have to kill it and restart)
(Arca build = previous one) at firefox close, pm123 plays again the song, no need to kill it.

!

Regards
 
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on April 14, 2021, 11:26:27 am
Source code update from latest 5.10.x kernel: https://smedley.id.au/tmp/uniaud32-linux-5.10.30-20210414.zip
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on April 14, 2021, 12:18:37 pm
Source code update from latest 5.10.x kernel: https://smedley.id.au/tmp/uniaud32-linux-5.10.30-20210414.zip

Hi Paul,
Seems to be working well on my TP X250 (in the docking station). System sound is there, Firefox can play audio files and pm123 is also working.
Thanks again for your great work.
Greetings from Potsdam / germany,
Mike
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on April 14, 2021, 04:23:53 pm
Seems to be working well on my TP X250 (in the docking station). System sound is there, Firefox can play audio files and pm123 is also working.

Hi Mike, can you please let me know the model of the Lenovo docking station that you are using? If it is not much to ask, can you please take a pci.exe report without the dock and with the dock to notice the difference, so I can add it to the OS2World wiki?

Regards
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on April 14, 2021, 05:32:49 pm
Hi Mike, can you please let me know the model of the Lenovo docking station that you are using? If it is not much to ask, can you please take a pci.exe report without the dock and with the dock to notice the difference, so I can add it to the OS2World wiki?

Hi Martin,

I am using the Thinkpad Pro Dock 90W. Please see attached picture for the type plate.
(http://TPX250Dock.png)

Kindly note that I have the USB3 feature of the X250 disabled, due to problems with the AN USB drivers. Even if only USB2 is enabled I need Lars Erdmann's drivers to use my external keyboard and mouse.

Greetings,
Mike
Title: Re: Updated Uniaud32 build
Post by: Remy on April 14, 2021, 06:34:39 pm
Source code update from latest 5.10.x kernel: https://smedley.id.au/tmp/uniaud32-linux-5.10.30-20210414.zip

Works very well.
No issue when watching a youtube video while PM123 is playing, pm123 plaies again after youtube stop/close
Fine.

Regards
Rémy 
Title: Re: Updated Uniaud32 build
Post by: Andreas Schnellbacher on April 14, 2021, 09:34:28 pm
due to problems with the AN USB drivers. Even if only USB2 is enabled I need Lars Erdmann's drivers to use my external keyboard and mouse.
Apparently there might be differences between 12.07 and 12.08, regarding the messages from others. For some, 12.08 finally works, others get traps. I guess you have tested both versions?
Title: Re: Updated Uniaud32 build
Post by: Neil Waldhauer on April 15, 2021, 01:08:24 am
Hi Mike, can you please let me know the model of the Lenovo docking station that you are using? If it is not much to ask, can you please take a pci.exe report without the dock and with the dock to notice the difference, so I can add it to the OS2World wiki?

Hi Martin,

I am using the Thinkpad Pro Dock 90W. Please see attached picture for the type plate.
(http://TPX250Dock.png)

Kindly note that I have the USB3 feature of the X250 disabled, due to problems with the AN USB drivers. Even if only USB2 is enabled I need Lars Erdmann's drivers to use my external keyboard and mouse.

Greetings,
Mike

I have the same dock. It does not work well with Arca Noae USB. If you must use Arca Noae USB with this dock, set USB to Auto in the X250 BIOS and use USB 3.0. Even then I need to unplug and replug the keyboard and/or mouse to get it to work. Before 12.08, it didn't really work at all. Once in 50 times, it traps. If you get a trap screen for this Dock, post it here or open a ticket at Arca Noae.
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on April 15, 2021, 09:29:14 am
Apparently there might be differences between 12.07 and 12.08, regarding the messages from others. For some, 12.08 finally works, others get traps. I guess you have tested both versions?

Yes, I try always newest from AN to see if it is finally working. The AN drivers are okay as long as I only use the laptop with he track point and the internal keyboard. But having the laptop placed into the docking station I'd like to use my external mouse and keyboard. This configuration works only with Lars' drivers. The AN became even worse over time. In the beginning it helped to unplug the keyboard and plug it in again. With the latest two releases even this work around is not working any longer.

Greetings,
Mike
Title: Re: Updated Uniaud32 build
Post by: Remy on April 15, 2021, 11:13:54 am
Apparently there might be differences between 12.07 and 12.08, regarding the messages from others. For some, 12.08 finally works, others get traps. I guess you have tested both versions?

Yes, I try always newest from AN to see if it is finally working. The AN drivers are okay as long as I only use the laptop with he track point and the internal keyboard. But having the laptop placed into the docking station I'd like to use my external mouse and keyboard. This configuration works only with Lars' drivers. The AN became even worse over time. In the beginning it helped to unplug the keyboard and plug it in again. With the latest two releases even this work around is not working any longer.

Grettings,
Mike

Hi,

I reported an issu on my desktop PC about usb keyboard/mouse not enabled after power off from windows and then boot into ArcaOS. To resolve it, I had to boot, go into bios and do an exit reboot and then, it went ok.
It was told me been an HW issue
(despite linux not windows have this kind of problem)

Your problem seems to be different but may be you can give it a try to go into the bios, and then exit reboot to check usb kb and mouse is available then.

Regards
Title: Re: Updated Uniaud32 build
Post by: Lars on April 15, 2021, 09:35:05 pm
Why would he if it works with my drivers ?
Seriously, it is not the HW, it is the USB drivers failing and there are many error conditions.
It surely can be fixed and I spent months chasing these problems, they mostly are due to USB devices not behaving the way you would expect.

It is completely absurd that people are complaining about Windows reliability but accept about any odd workaround for OS/2.

If you switch to using OS/2 in a virtual machine, you can get rid of that problem anyway: use USB 2.x. The performance gain of USB 3.x is non existent anyways (with the OS/2 drivers at least). And for simple devices as keyboards, mice and even video/audio devices, USB 2.x will do the job just fine.
Using a virtual machine will also get rid of the problem of NVM, new network HW bla bla bla.
Title: Re: Updated Uniaud32 build
Post by: Remy on April 15, 2021, 11:16:07 pm
Nothing to do with your drivers.
He spoke about AN drivers too. (these have a not resolve problem and supposed to be HW - not virtual machine)
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on April 16, 2021, 01:14:32 am
If you switch to using OS/2 in a virtual machine, you can get rid of that problem anyway: use USB 2.x. The performance gain of USB 3.x is non existent anyways (with the OS/2 drivers at least). And for simple devices as keyboards, mice and even video/audio devices, USB 2.x will do the job just fine.
Using a virtual machine will also get rid of the problem of NVM, new network HW bla bla bla.

I get about 5x the performance using USB3 compared to USB2 on my USB3 drive, 20MB/s with USB2 and 110MB/s using a USB3.1 card and transferring an ISO (over 1GB).
I find OS/2 slow in a virtual machine, perhaps as it can't take advantage of all the cores and using Linux, while all hardware works fine, the experience is just lacking, even performance.
A while ago I tried compiling Firefox on Linux with -j5 (5 jobs, recommended on a 4 core machine), and an apt upgrade after making sure nothing that the compile was using was getting updated and then tried to watch a YouTube video in Firefox. The system came to a halt basically, with the mouse even barely moving and it took quite a while to get ALT-CTRL-F1 to switch to a terminal so I could kill stuff. Guess I've just been spoiled by OS/2 when it comes to using the desktop.
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on April 16, 2021, 07:39:37 am
Why would he if it works with my drivers ?
Seriously, it is not the HW, it is the USB drivers failing and there are many error conditions.
It surely can be fixed and I spent months chasing these problems, they mostly are due to USB devices not behaving the way you would expect.

It is completely absurd that people are complaining about Windows reliability but accept about any odd workaround for OS/2.

If you switch to using OS/2 in a virtual machine, you can get rid of that problem anyway: use USB 2.x. The performance gain of USB 3.x is non existent anyways (with the OS/2 drivers at least). And for simple devices as keyboards, mice and even video/audio devices, USB 2.x will do the job just fine.
Using a virtual machine will also get rid of the problem of NVM, new network HW bla bla bla.

Hi Lars,
Yes, I totally agree about the use of your drivers. No need for workarounds like unplugging and re-plugging of devices. Your drivers just work. Great job - Thanks.

I am using OS/ 2 since version 2.1 always on bare metal. And I would like to stay this way as long as possible. By selecting the right hardware I have been able to get it running every time. If the hardware was at latest levels it took me some tweaking in the beginning but it worked out smoothly over time.

Greetings form Potsdam / Germany,
mike
Title: Re: Updated Uniaud32 build
Post by: Lars on April 16, 2021, 12:29:57 pm
If you switch to using OS/2 in a virtual machine, you can get rid of that problem anyway: use USB 2.x. The performance gain of USB 3.x is non existent anyways (with the OS/2 drivers at least). And for simple devices as keyboards, mice and even video/audio devices, USB 2.x will do the job just fine.
Using a virtual machine will also get rid of the problem of NVM, new network HW bla bla bla.

I get about 5x the performance using USB3 compared to USB2 on my USB3 drive, 20MB/s with USB2 and 110MB/s using a USB3.1 card and transferring an ISO (over 1GB).
I find OS/2 slow in a virtual machine, perhaps as it can't take advantage of all the cores and using Linux, while all hardware works fine, the experience is just lacking, even performance.
A while ago I tried compiling Firefox on Linux with -j5 (5 jobs, recommended on a 4 core machine), and an apt upgrade after making sure nothing that the compile was using was getting updated and then tried to watch a YouTube video in Firefox. The system came to a halt basically, with the mouse even barely moving and it took quite a while to get ALT-CTRL-F1 to switch to a terminal so I could kill stuff. Guess I've just been spoiled by OS/2 when it comes to using the desktop.

How often do you need to use USB in a virtual machine ? If you need to copy large amounts of data, then at least for me, shared folders are the option. Plus you can even set up a network between host and guest (I have not tried that).

But agreed, Virtualbox has 2 problems:
1) it only supports one core if you run OS/2 as a guest. Everything else just hangs the guest sooner or later.
2) it does not properly support USB isochronous transfers. That becomes immediately apparent if you try to play a sound file in an OS/2 guest (using a USB audio device). It will exhibit heavy dropouts. If you do the same on a native/"bare metal" installation, it works just fine. I have reported this as a bug in the Virtualbox bug tracker but I think they couldn't care less.



Title: Re: Updated Uniaud32 build
Post by: Lars on April 16, 2021, 12:40:06 pm
Nothing to do with your drivers.
He spoke about AN drivers too. (these have a not resolve problem and supposed to be HW - not virtual machine)

Again, this problem is not about HW. If it were, my USB drivers would also not work. There is a problem with one or more AN USB drivers. It's either about a stalled pipe not being cleared correctly or with data toggle not being set correctly (at least these cover around 90% of what can go wrong). And of course, in this case, the problem will show on real HW as well as in Virtualbox.
Title: Re: Updated Uniaud32 build
Post by: Remy on April 16, 2021, 05:13:39 pm
Nothing to do with your drivers.
He spoke about AN drivers too. (these have a not resolve problem and supposed to be HW - not virtual machine)

Again, this problem is not about HW. If it were, my USB drivers would also not work. There is a problem with one or more AN USB drivers. It's either about a stalled pipe not being cleared correctly or with data toggle not being set correctly (at least these cover around 90% of what can go wrong). And of course, in this case, the problem will show on real HW as well as in Virtualbox.

Hi Lars,

I agree with you...

Regards
Rémy
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on April 18, 2021, 12:45:32 am
Hi Paul,

as I was able to install ArcaOS on Bare metal named Lenovo T480 I am really happy to report, that your driver 20210414 works perfect with this hardware!

Thanks a lot, very well done!!

I attach the files I was able to get.

If you need more information on this hardware, please let me know!

EDIT: one regression here, as mentioned earlier in this thread: Playing PM123 and having system Sounds enabled does not work correct. When the first time system sound appears while PM123 is playing, there is no Volumecontrol with PM123 anymore.

All the best!

Sigurd

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on May 27, 2021, 11:38:27 am
https://smedley.id.au/tmp/uniaud32-linux-5.10.40-20210527.zip - code updated to linux 5.10.40 level...
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on May 27, 2021, 02:38:47 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.40-20210527.zip - code updated to linux 5.10.40 level...

Hi
I tested it with ArcaOS 5.0.6 VM guest on VBox 6.1.14 (Windows 10 host).
- VM Intel HD Audio: Works, Audio control working, no traps.
- VM ICH AC97: Works, Audio control working, no traps.

After that test I updated VirtualBox to 6.1.22 (same Windows 10 host)
- VM Intel HD Audio: Produce audio, but gives looping sound at end. Audio control working, no traps.
- VM ICH AC97: Works, Audio control working, no traps.

I did these because on the past I noticed a strange Audio problems (Looping sounds) when I updated VBox from 6.1.14 and I want to confirm these with your UniAud release. I don't know what the VirtualBox team did with the "Intel HD Audio" in these newer releases. It is not a problem of these Uniaud release, or Uniaud, something was changed on Virtualbox (Ticket 20273) (https://www.virtualbox.org/ticket/20273).

Thanks Paul for this update.

Regards
Title: Re: Updated Uniaud32 build
Post by: David McKenna on May 27, 2021, 09:52:29 pm
 Thanks Paul... works just fine here on my Intel HD audio desktop system....

Regards,
Title: Re: Updated Uniaud32 build
Post by: roberto on May 27, 2021, 11:06:34 pm

Thanks for the new file, it's 90% ok and getting better.

I wanted to ask you if you can tell me what programs or dlls can be used when I press an active screen change with the mouse.
 For more clarification if I am viewing this web page, and with the mouse clicking on a command prompt or any other open window,
  which programs can make that change.
   I have observed that the sound glitches improve when switching screens.

saludos
Title: Re: Updated Uniaud32 build
Post by: Andreas Schnellbacher on May 28, 2021, 01:01:31 am
About switching desktop windows:

There's no program to switch the desktop windows of XPager. You can do this only manually. If you open its context menu and click on Properties, the Screen object opens with the XPager panel. You can there activate hotkeys to do that.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on May 28, 2021, 02:07:18 pm
Not working on my Gigabyte MB with a Sound Blaster Recon chip. My old laptop is gone.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on June 06, 2021, 06:10:04 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.40-20210527.zip - code updated to linux 5.10.40 level...

Hi Paul

Today I also tested it on the Thinkpad L420.
- Works. No trap. Master volume working. Works good here.
Thanks
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on June 07, 2021, 04:52:55 am
Hi Eugene,

Not working on my Gigabyte MB with a Sound Blaster Recon chip. My old laptop is gone.

Did older builds work? If the did, I'd like to see a log - but I do know there is at least one chipset used by Creative/Sound Blaster that is not currently enabled in the Uniaud code (as it won't compile).

Can you post the PCI IDs here?

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on June 10, 2021, 06:55:49 pm
CHKPCI Version 0.02aß

PCIDEVS.TXT   : 1185.7KiB

Notice: line has non-ASCII characters: "D   9990   MCS9990 PCIe to 4ÔÇÉPort USB 2.0 Host Controller"

Found and checked 34961 entries

Vendors       : 2437
Devices       : 10900
OEM IDs       : 5831
Subsystem IDs : 15793
Revision IDs  : 0
X Codes       : 0

Comment Lines : 7
Blank Lines   : 0
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on June 19, 2021, 01:00:57 am
linux code updted to 5.10.45 level...

https://smedley.id.au/tmp/uniaud32-linux-5.10.45-20210619.zip
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on June 19, 2021, 07:49:13 pm
You must have pulled tha latest drop I just got home from vacation and it is not connecting. but thank yo anyway.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on June 19, 2021, 11:37:43 pm
https://smedley.id.au/tmp/uniaud32-linux-5.10.45-20210619.zip

Hi
I tested it with ArcaOS 5.0.6 VM guest on VBox 6.1.22 (Windows 10 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Produce audio, but gives looping sound at end. Audio control working, no traps. (VirtualBox bug Ticket 20273 (https://www.virtualbox.org/ticket/20273) remains)

Thanks Paul for this update.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on June 19, 2021, 11:41:58 pm
You must have pulled tha latest drop I just got home from vacation and it is not connecting. but thank yo anyway.
Bit simpler than that, the router decided to stop working around 1am and was rebooted around 6:30am...
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on June 21, 2021, 06:49:43 pm
No sudden stops or hangs but no sound either.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on June 22, 2021, 12:12:20 am
 Hi Paul,

 5.10.45 is working good here... thanks again!

 @ Eugene - get PMUniMix 0.82 from Hobbes and run it. There may be volume controls that default to zero. I have one HP laptop that needs them turned up to work...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on June 22, 2021, 02:30:17 am
Thanks David but on my system Unimix will not run correctly. I knew that from in the Past and it wil not even display any volume controls.
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on June 22, 2021, 10:28:09 am
linux code updted to 5.10.45 level...

https://smedley.id.au/tmp/uniaud32-linux-5.10.45-20210619.zip

Thinkpad T430:
Sound is there. But if I run VLC/2 and try to change volume during video playing sound stops and does not return until I close VLC/2 and restart it.

Best regards,
Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 16, 2021, 11:39:45 pm
Linux sound code updated to 5.10.50 level - https://smedley.id.au/tmp/uniaud32-linux-5.10.50-202107179.zip

Pretty minor code changes for Realtek HDA hardware, and Intel AC97 hardware.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on July 17, 2021, 05:40:56 am
Linux sound code updated to 5.10.50 level - https://smedley.id.au/tmp/uniaud32-linux-5.10.50-202107179.zip

Hi
I tested it again with ArcaOS 5.0.6 VM guest on VBox 6.1.22 (Windows 10 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Produce audio, but gives looping sound at end. Audio control working, no traps. (VirtualBox bug Ticket 20273 (https://www.virtualbox.org/ticket/20273) supposedly fixed on upcoming 6.1.23 version)

Thanks again Paul for this update.

Regards
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on July 18, 2021, 04:35:49 pm
Well stil not crashes but no sound and unimix no longer wil even come up.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on July 19, 2021, 05:50:23 am
Hi Eugene.
Well stil not crashes but no sound and unimix no longer wil even come up.
Can you confirm the pci.exe report of your mainboard? On an older post you said it is a "Gigabyte MB with a Sound Blaster Recon chip", but can you please post the PCI report  and mainboard model?

Please also provide this log "copy alsahlp$ alsalog.txt".

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 19, 2021, 10:53:26 am
Hi Guys,

Well stil not crashes but no sound and unimix no longer wil even come up.
Can you confirm the pci.exe report of your mainboard? On an older post you said it is a "Gigabyte MB with a Sound Blaster Recon chip", but can you please post the PCI report  and mainboard model?

Please also provide this log "copy alsahlp$ alsalog.txt".

From a quick google, a Sound Blaster Recon sounds like a ca0132 chipset - this chipset isn't currently enabled in the uniaud code.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 20, 2021, 12:00:17 pm
https://uniaud32-linux-5.10.50-20210720.zip (https://smedley.id.au/tmp/uniaud32-linux-5.10.50-20210720.zip) *might* allow the ca_0132 to work - there are still a few hacks - but an alsahlp$ from it might be interesting...
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on July 20, 2021, 03:19:30 pm
I will give them a try. Maybe tomorrow too much to do today.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on July 20, 2021, 04:03:00 pm
Hi Eugene
I will give them a try. Maybe tomorrow too much to do today.
Please let me know also the model of your Sound Blaster Recon card to post is on the wiki as working or non working.  ;D

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 20, 2021, 11:15:16 pm
I will give them a try. Maybe tomorrow too much to do today.
Let me know - after a sleep, I think I know how to fix some of the hacks that were in yesterday's build. I'll try and address those this evening.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 21, 2021, 11:11:29 am
I will give them a try. Maybe tomorrow too much to do today.
Let me know - after a sleep, I think I know how to fix some of the hacks that were in yesterday's build. I'll try and address those this evening.

OK https://smedley.id.au/tmp/uniaud32-linux-5.10.50-20210721.zip should properly support ca0132 hardware
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on July 21, 2021, 02:33:06 pm
Well thank you so very much I will it a try soon the other one lead to a setacular crash.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on July 21, 2021, 02:42:00 pm
Well I got sound. It is loud and the system is irradic. Mouse is very slow. PMMix is slow but comes up. I wil have to play with it but sound and no crash.
Title: Re: Updated Uniaud32 build
Post by: Hajo on July 21, 2021, 07:12:51 pm
My soundblaster z initializes (click sound) but boot stops afterwards. No initilization message. I created this partition on real disk with virtual box, so I can boot without virtual box, but I don't have any input devices after arcaos comes up on physical hardware. In virtualbox this driver works like all the others before  :)

My subscription ended short before USB3.0 drivers were released and I don't think that I will resubscribe to get input devices on physical machine.

Regards

Hajo
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on July 21, 2021, 11:10:40 pm
Well I got sound. It is loud and the system is irradic. Mouse is very slow. PMMix is slow but comes up. I wil have to play with it but sound and no crash.

OK cool - I'll try and check my patches this evening - it's possible I screwed something up when reworking some code that wouldn't compile with openwatcom...
Title: Re: Updated Uniaud32 build
Post by: Andi B. on July 22, 2021, 09:03:37 pm
I installed latest warpin package on a fairly fresh ArcaOS beta. I still have no sound on this system. Think I never had sound. If yes then it must have been a very old uniaud but I don't remember. Attached the logs. Maybe you see what's going wrong.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on July 22, 2021, 10:42:26 pm
 Hi Andi,

  I looked at your files and was surprised to see you have the identical sound hardware as myself (Realtek ALC887-VD). Here it works perfectly. One thing I notice in your testlog file is this:

----- Checking G:\CONFIG.SYS ---------------------------------------------------
ERROR: DEVICE=G:\IBMCOM\MACS\MMRE.OS2
ERROR: DEVICE=G:\IBMCOM\MACS\E100B.OS2
ERROR: DEVICE=G:\IBMCOM\MACS\E1000B.OS2
       Load order violation. All MAC drivers must come after all
       IFS statements and all network drivers

 Maybe correcting those errors (by putting those lines at the end of CONFIG.SYS) might help?

Regards,
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on July 23, 2021, 02:21:40 am
Hi Andi. B

I installed latest warpin package on a fairly fresh ArcaOS beta. I still have no sound on this system. Think I never had sound. If yes then it must have been a very old uniaud but I don't remember. Attached the logs. Maybe you see what's going wrong.

I checked some other hardware reports and there are ones that says that "8086:8C20 - Intel 8 Series/C220 Series Chipset High Definition Audio Controller" works with Uniaud. Maybe is the config.sys that David M says, or some other trick with unimix.exe (https://www.os2world.com/wiki/index.php/Unimix).

It is as "DEVICE=G:\MMOS2\UNIAUD32.SYS /v /A:0" maybe it is "/A:1"

Andi, which is the Brand/model of you mainboard ? or machine and model if it is a retail machine. I want to check if was previously reported and include it on the wiki if applies.

Regards
Title: Re: Updated Uniaud32 build
Post by: Andi B. on July 23, 2021, 09:05:19 am
It's an Asus H87-Pro motherboard.

I saw the config.sys error message too. But this is what the ArcaOS installer made. I had only a quick look at the config.sys (it's included in the testlog) and think it's the ISO ifs line which triggers this error message. Again, this is a rather fresh ArcaOS installation. I didn't change ifs or nif lines in config.sys by myself. Of course worth to check as soon as I'm on this machine again....

Edit: No difference when putting IFS=G:\sys\bin\STUBFSD.IFS ISOFS before the NIF drivers. /A:0 or /A:1 or /A:2 - nothing produces sound.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on July 25, 2021, 03:27:39 pm
Hi Andi,

  Just had another look at your uniaud log file and it looks like UniAud was never properly installed because there are no [UNIAUDxxx] sections in your MMPM2.INI file. Try re-installing using the latest AN version (Uniaud-20210320.wpi) which should create those sections for you. Reboot and see if it works...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Andi B. on July 26, 2021, 09:00:23 am
Thanks David. Before I posted I installed latest .wpi from David from here - ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20210721-beta.exe. As previous install attempts did not work I hoped this wpi would do it. Unfortunately not.

Do you have a clue how to repair MMPM2.INI. Obviously the wpi didn't do it. Neither on my work partition nor on the fairly new ArcaOS beta partition.
Title: Re: Updated Uniaud32 build
Post by: Lars on July 26, 2021, 09:09:53 am
Post it here or attach and I will fix it for you.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on July 26, 2021, 10:40:51 pm
 Hi Andi,

  It's tricky to edit MMPM2.INI properly. I'll attach my attempt to edit yours to include UniAud - rename the attached file MMPM2.INI (can't upload *.INI files here apparently). Would be good if Lars could check it too. Make sure to backup your original MMPM2.INI file, and be prepared to replace it if necessary.

 Might be better to try to un-install UniAud (using WarpIn), reboot, then re-install UniAud using the latest official version. It could be that the install detects UniAud as already installed, so does not update MMPM2.INI.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Lars on July 27, 2021, 09:00:21 am
Looks good. Just to be sure you can remove the Sequencer= lines.
I don't think it will be necessary to install the drivers from scratch once MMPM2.INI is correct (of course, the drivers will need to be loaded via config.sys). MMPM2.INI has always been the weak spot of the video/audio system.
Title: Re: Updated Uniaud32 build
Post by: Andi B. on July 27, 2021, 09:50:14 am
Thanks. Will try as soon as I'm at this system. May take one or two days.

Edit: Thanks David, Lars and Paul. My ArcaOS installation produces sound now. The outputs do not fully match when I change volume with PMUnimix. But that's a task for later time. First I've to repair my main partition on this system so that this works too.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on July 28, 2021, 12:10:13 am
Paul, I have given this driver a good shakedown. Cds wil not play, MP3s will not play. the system tends to lock up and release when you try to do Warpin installs.Youtube will not play sound and is sluggish,the video is sluggish there as well.. I do have system sounds and they work well. I again thank you for your effort.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 07, 2021, 04:14:09 pm
Anoher update. The new arcanoae release drivers work on my system I did have to run LCSS to get it to be more stables with the Desktop and the mouse. but it is great to have system sounds. Thank you Paul for all your hard work on this.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2021, 11:38:45 pm
Hi Eugene,

Anoher update. The new arcanoae release drivers work on my system I did have to run LCSS to get it to be more stables with the Desktop and the mouse. but it is great to have system sounds. Thank you Paul for all your hard work on this.

Good news! I'm not sure what David A does differently when he builds, but I recall on an early version someone had more success with an AN build than mine, even though the core code is identical.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 08, 2021, 04:34:43 pm
Paul AN gave a word of thanks to yo for your work on the news posting.of the release. You and David both deserve credit.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 08, 2021, 07:04:36 pm
Paul AN gave a word of thanks to yo for your work on the news posting.of the release. You and David both deserve credit.
Hi Eugene.

Was that the "Creative Sound Blaster Recon3D PCIe" you own? Can you please post more information about it, like confirm the full model number and harwareID?, I don't know if you have the pci.exe report, and any remark you may share to make it work.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2021, 04:45:08 am
Hi Eugene,
Paul AN gave a word of thanks to yo for your work on the news posting.of the release. You and David both deserve credit.
Yeah I saw that, which was nice :) I spent literally hundreds of hours on Uniaud last year trying to get the code updated to the current level. Thankfully current maintenance time is minimal, although later this year I'll aim to move us to the next stable LTS (Long Term Support) code level.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 09, 2021, 05:06:05 am
Hi

I will also like to remind the audience that you can also show some appreciation to Paul by donating some sponsor units with the Arca Noae shop (https://www.arcanoae.com/shop/os2-ports-paul-smedley/).

Thanks for your efforts Paul.

Regards
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 09, 2021, 10:47:40 pm
Ye3s you are correct on the sound card ID. attached is the results of the PI sniffer.
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 11, 2021, 06:52:44 pm
Hi there,
has anyone tried the latest uniaud32 builds (either Paul's or the one from AN) on a Laptop where suspend and resume works?

Holger
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 11, 2021, 11:46:55 pm
Ye3s you are correct on the sound card ID. attached is the results of the PI sniffer.

Thanks Eugene for sharing.

It seems you are using an old pci.ids, since a lot of it shows like "Unknown".
But I'm now able to find which is the hardware ID for the "Creative Sound Blaster Recon3D PCIe".

If I set aside all the Intel ones (I guess Creative VendorID is not the same as Intel), I only found this ones:

 Vendor 1002h Advanced Micro Devices [AMD] nee ATI
 Device AAB0h Oland/Hainan/Cape Verde/Pitcairn HDMI Audio [Radeon HD 7000 Series]

 Vendor 1B21h ASMedia Technology Inc.
 Device 0612h ASM1062 Serial ATA Controller

 Vendor 1969h Atheros Communications Inc.
 Device E0A1h Killer E2400 Gigabit Ethernet Controller

 Vendor 15B7h Sandisk Corp
 Device 5006h WD Black SN750 / PC SN730 NVMe SSD

I don't know I'm missing something on the report.

Regards
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 12, 2021, 04:38:21 am
Hi there,
has anyone tried the latest uniaud32 builds (either Paul's or the one from AN) on a Laptop where suspend and resume works?

Holger

Just tested the latest from AN on my T42, suspend and resume still work here. Old laptop that uses APM rather then ACPI for power control as Warp supported it.
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 12, 2021, 10:47:21 am
Hi Dave,

does sound still work after suspend and resume? Can you play for example *.wav files after resume?

Regards
Holger
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 12, 2021, 04:03:41 pm
Hi Dave,

does sound still work after suspend and resume? Can you play for example *.wav files after resume?

Regards
Holger

Yes, once the system wakes up, which takes a  few seconds.
What machine are you using?
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 12, 2021, 06:54:36 pm
Hi Dave,

I have a x61 here, which caused a lot of problems in the last months. But right now I have a completely new installation of AN (the second released version). I run this with acpi 3.23.04 because this version allows suspend and resume to work, all newer versions do not, all newer versions of AN neither off course. The strange thing is that when I use the original uniaud supplied with AN I can suspend and resume and sound works after resume. With the new builds sound is much better, I even do not need any unimix line in config.sys to have sound enabled, but after resume sound no longer works, neither when clicking a file or for example with PM123.Not a big deal, I just wondered if someone else has the same problems.

Regards
Holger

 
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 13, 2021, 12:24:30 pm
Hi Holger,

I have a x61 here, which caused a lot of problems in the last months. But right now I have a completely new installation of AN (the second released version). I run this with acpi 3.23.04 because this version allows suspend and resume to work, all newer versions do not, all newer versions of AN neither off course. The strange thing is that when I use the original uniaud supplied with AN I can suspend and resume and sound works after resume. With the new builds sound is much better, I even do not need any unimix line in config.sys to have sound enabled, but after resume sound no longer works, neither when clicking a file or for example with PM123.Not a big deal, I just wondered if someone else has the same problems.

Are you able to send me the output of 'testlog uniaud' ? I'm curious what hardware you have. There were a LOT of changes in the most recent uniaud updates in AN, I'm not terribly surprised suspend/resume doesn't work. There were a lot of power related api's added in newer linux kernels, many of them I stubbed just to get things building.

@DaveYeo are you able to do the same on your T42? I'm assuming the T42 may still be AC97 hardware whereas the x61 will be HDA..
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2021, 07:57:20 am
Lazy Saturday afternoon, so updated uniadu32-next to 5.11.22: https://smedley.id.au/tmp/uniaud32-linux-5.11.22-20210814.zip - next stop 5.12.19
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 14, 2021, 08:32:35 am
Quote
Are you able to send me the output of 'testlog uniaud' ? I'm curious what hardware you have. There were a LOT of changes in the most recent uniaud updates in AN, I'm not terribly surprised suspend/resume doesn't work. There were a lot of power related api's added in newer linux kernels, many of them I stubbed just to get things building.
@DaveYeo are you able to do the same on your T42? I'm assuming the T42 may still be AC97 hardware whereas the x61 will be HDA..
Hi Paul, I will send it to you on monday, I am away right now. Writing this from a windows laptop.

Regards
Holger
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 14, 2021, 08:56:57 am
5.12.19: https://smedley.id.au/tmp/uniaud32-linux-5.12.19-20210814.zip
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 14, 2021, 05:42:05 pm
5.12.19: https://smedley.id.au/tmp/uniaud32-linux-5.12.19-20210814.zip

Hi
I tested the release version with ArcaOS 5.0.6 VM guest on VBox 6.1.26 (Windows 10 host).
- VM ICH AC97: It stopped working. No Audio, no traps.
- VM Intel HD Audio:  It also stopped working. No Audio, no traps.
If I run a .wav on the desktop it will quickly close, it will not sound and will not show the counter on the bottom, which I think it means it does not even try to playback.

uniaud32-linux-5.11.22-20210814.zip  is also not working, the same as 5.12.19

I returned to version "5.10.40" and "5.10.50" and sound is back.

Regards
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 14, 2021, 05:49:15 pm
 Paul,

  My experience is the same as Martin's with both versions (release and debug)- it seems to load fine, PMUniMix works, but no sound. Also can't play any WAV or MP3 file. Alsahlp$ attached...
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 14, 2021, 08:16:08 pm
Paul I had the same exact result with the Creative recon 3D.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 15, 2021, 01:08:38 am
5.13.10: https://smedley.id.au/tmp/uniaud32-linux-5.13.10-20210815.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 15, 2021, 01:24:12 am
Paul,

 Now we're talking - or I can hear it! 5.13 is making sounds here and seems to work well - thanks again, you're aces...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 15, 2021, 02:27:52 am
Hi Dave,

Now we're talking - or I can hear it! 5.13 is making sounds here and seems to work well - thanks again, you're aces...

There's no logic in 5.13 working when 5.11 and 5.12 didn't - but I'll take it :)

I only tested these builds in Virtualbox - I often don't get sound in Virtualbox so I wasn't concerned about not getting sound.

I'll try and do some testing locally with real hardware ASAP.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 15, 2021, 02:50:39 am
5.13.10: https://smedley.id.au/tmp/uniaud32-linux-5.13.10-20210815.zip
Hi
I tested it with ArcaOS 5.0.6 VM guest on VBox 6.1.26 (Windows 10 host).
- VM ICH AC97: Works, Audio control working, no traps. It sounds, but sometimes the volume control goes to the max when I double click from one wav to another.
- VM Intel HD Audio: Works, Audio control working, no traps. But the sound it is lower than "AC97".  If I set the volume to 40%, I can not hear it at all.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 15, 2021, 03:05:54 am
Hi Martin,

5.13.10: https://smedley.id.au/tmp/uniaud32-linux-5.13.10-20210815.zip
Hi
I tested it with ArcaOS 5.0.6 VM guest on VBox 6.1.26 (Windows 10 host).
- VM ICH AC97: Works, Audio control working, no traps. It sounds, but sometimes the volume control goes to the max when I double click from one wav to another.
- VM Intel HD Audio: Works, Audio control working, no traps. But the sound it is lower than "AC97".  If I set the volume to 40%, I can not hear it at all.

Any chances of logs from a 'good' build and from the latest?

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on August 15, 2021, 04:34:10 am
@DaveYeo are you able to do the same on your T42? I'm assuming the T42 may still be AC97 hardware whereas the x61 will be HDA..

Should be in your inbox. Let me know if it doesn't show up
Dave
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 15, 2021, 02:26:44 pm
I too have sound again. they perform much like the AN Uniaud drivers. It seems to interfere with the video driver or WPS on my machine. I am not sure which. screen redraws are very slow. There is hesitency in the populating of windows. The sound is limited to system sounds VLT wil play a Cd but generates no sound. this is just the same as the new stock drivers from AN. But it is getting better.
Title: Re: Updated Uniaud32 build
Post by: Holger Schuett on August 16, 2021, 06:55:39 pm
Hi Holger,
Are you able to send me the output of 'testlog uniaud' ? I'm curious what hardware you have. There were a LOT of changes in the most recent uniaud updates in AN, I'm not terribly surprised suspend/resume doesn't work. There were a lot of power related api's added in newer linux kernels, many of them I stubbed just to get things building.

@DaveYeo are you able to do the same on your T42? I'm assuming the T42 may still be AC97 hardware whereas the x61 will be HDA..

Hi Paul,
which address should I send the log to?

Regards
Holger
Title: Re: Updated Uniaud32 build
Post by: Hajo on August 17, 2021, 07:34:57 pm
Getting closer

HDA Creative at 0xf6404000 irq 5
Detected Mixer: Creative Sound Blaster Z

Output of Unilog is attached

Thanks for the good work :-)
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 18, 2021, 02:55:28 am
Hi

These are the alsahlp logs I got in the VirtualBox guest.

Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 19, 2021, 01:29:04 pm
Just to ask are there any PCIe cards that use supported chipsets for the current releases of Uniaud?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 26, 2021, 10:42:40 am
Linux code update to 5.14.7 level: https://smedley.id.au/tmp/uniaud32-linux-5.14.7-20210926.zip
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on September 27, 2021, 12:53:19 pm
Just to ask are there any PCIe cards that use supported chipsets for the current releases of Uniaud?

Sound Blaster SB1040 PCIe 2007
==============================
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: Creative
  driver: HDA-Intel
  name: HDA Creative
  longname: HDA Creative at 0xfeafc000 irq 16
  mixer: Creative CA0110-IBG
  componenets: HDA:1102000a,11021006,00100000

I think you would need to check the adapters chipsets and ask Paul if they are compiled in. The above card works fine with Uniaud.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 27, 2021, 01:36:07 pm
The new driver works totally on system sounds. but i does not give audio from Youtube.com or VLC played Cds. In fact no CD player works with this driver and I am not sure the problem is not elseware i my system.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 27, 2021, 05:58:25 pm
Linux code update to 5.14.7 level: https://smedley.id.au/tmp/uniaud32-linux-5.14.7-20210926.zip

Hi
I tested it with ArcaOS 5.0.6 VM guest on VBox 6.1.26 (Windows 10 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Works, Audio control working, no traps.

Regards
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on September 27, 2021, 06:36:48 pm
The new driver works totally on system sounds. but i does not give audio from Youtube.com or VLC played Cds. In fact no CD player works with this driver and I am not sure the problem is not elseware i my system.

Have you tried the classic CD player, after making sure it's set to use digital transfer under the View menu?
I take it you are using one of the Mozilla apps to visit Youtube, them and VLC use libkai, which will default to using UNIAUD to play sounds and is somewhat broken. Make sure that you have "SET KAI_AUTOMODE=DART" in your environment. You could also try "SET KAI_AUTOMODE=UNIAUD", you never know.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 27, 2021, 07:59:52 pm
Thank yo very much for the suggestion Dave.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 24, 2021, 08:07:39 pm
Well there is a new drop from the 23rd on Netlabs and the performance is much much improved. Still some little hangs when opening certain folders and a still s slight hang when the folder makes a sound. But it is much more usable now. I will comment further as I run into things. but this is looking much better.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on October 24, 2021, 09:35:35 pm
Well there is a new drop from the 23rd on Netlabs and the performance is much much improved. Still some little hangs when opening certain folders and a still s slight hang when the folder makes a sound. But it is much more usable now. I will comment further as I run into things. but this is looking much better.
Thanks for the heads up.
FYI, Here it is the link: ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe)

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 25, 2021, 03:52:29 am
Well there is a new drop from the 23rd on Netlabs and the performance is much much improved. Still some little hangs when opening certain folders and a still s slight hang when the folder makes a sound. But it is much more usable now. I will comment further as I run into things. but this is looking much better.
Thanks for the heads up.
FYI, Here it is the link: ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe)

Regards

FWIW - this is built from the same code of my last 5.14.x release - there is something about the way David A does the builds that makes a difference...
Title: Re: Updated Uniaud32 build
Post by: Olafur Gunnlaugsson on October 25, 2021, 07:06:40 am
Thanks for the heads up.
FYI, Here it is the link: ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe)

Regards

Is there a parsing bug in the forum software? The link shows up correctly as
ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe
However the actual link is ftp//ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe i.e., the : is dropped
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 25, 2021, 03:41:29 pm
I would like to thank Paul Smedly and Steve Levine for their efforts and work on this driver.
Title: Re: Updated Uniaud32 build
Post by: ivan on October 25, 2021, 09:46:27 pm
Just out of curiosity may I ask a question?  Has anyone been able to get sound to work through an HDMI port and link to a monitor?

I ask this because all of my computing equipment now has an HDMI port on the motherboard as well as all the monitors, only used at the moment for video.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 26, 2021, 03:49:48 am
Hey Ivan,
Just out of curiosity may I ask a question?  Has anyone been able to get sound to work through an HDMI port and link to a monitor?

I ask this because all of my computing equipment now has an HDMI port on the motherboard as well as all the monitors, only used at the moment for video.

I'd have to check the code, but in theory I think this could work - but I've never tested it. I've be interested in a copy of 'testlog uniaud' for a system connected with HDMI

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 26, 2021, 03:50:26 am
I would like to thank Paul Smedly and Steve Levine for their efforts and work on this driver.

Thanks! Steven's help with diagnosing traps/dump files has been invaluable - definitely new territory for me!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on October 26, 2021, 10:33:42 pm
 Hey Paul,

 I attached testlog uniaud for my computer, which has HDMI video and monitor does have speakers (that work with Windows). I currently use the sound jack on the back of the computer for OS/2 sound, but would be nice to just use the monitor via HDMI. No sign of HDMI sound in the log that I can see though... using your latest 5.14.7 driver.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 27, 2021, 10:55:38 am
Hey Dave,

I attached testlog uniaud for my computer, which has HDMI video and monitor does have speakers (that work with Windows). I currently use the sound jack on the back of the computer for OS/2 sound, but would be nice to just use the monitor via HDMI. No sign of HDMI sound in the log that I can see though... using your latest 5.14.7 driver.

OK I checked the makefiles, patch_hdmi.c wasn't being built - I remember why now - it dragged in a bunch of other code that I didn't feel like trying to get working at the time.

Leave it with me and I'll see if I can get it compiling.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 28, 2021, 08:07:42 am
patch_hdmi enabled build of uniaud32: https://smedley.id.au/tmp/uniaud32-linux-5.14.7-20211028.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on October 28, 2021, 12:11:01 pm
 Thanks Paul! This one still works with the Realtek sound jack, but no HDMI - testlog attached. I assume unimix -card should show 2 adapters present if HDMI works?

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 28, 2021, 11:03:02 pm
Hi Dave,
Thanks Paul! This one still works with the Realtek sound jack, but no HDMI - testlog attached. I assume unimix -card should show 2 adapters present if HDMI works?
Yeah I would think it would show two adapters - I'll take a look at the log and see if there are any clues.....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 02, 2021, 01:29:40 pm
The new beta driver works Just like th elast driver. But I have noticed uninstall the old Workplace to upgrade to the new version that uninstlaled the stalls that occur seems to be reduced. I will down load and install the new version  of the Desktop for ArcaOA 1.0.14 to see what happens.
Title: Re: Updated Uniaud32 build
Post by: mike on January 18, 2022, 09:14:20 pm

Yeah I would think it would show two adapters - I'll take a look at the log and see if there are any clues....

here it shows 2 adapter, the monitor is connected with dislayport cable. however no hdmi/dp sound nor onboard sound.

Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 1
Card info for adapter 0:
  num: 0
  id: HDMI
  driver: HDA-Intel
  name: HDA ATI HDMI
  longname: HDA ATI HDMI at 0xfeb64000 irq 45
  mixer: ATI R6xx HDMI
  componenets: HDA:1002aa01,00aa0100,00100500

CardInfo.card: 0
CardInfo.id: HDMI
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA ATI HDMI
CardInfo.LongName: HDA ATI HDMI at 0xfeb64000 irq 45
CardInfo.MixerName: ATI R6xx HDMI
CardInfo.Components: HDA:1002aa01,00aa0100,00100500

CardInfo.card: 1
CardInfo.id: Generic
CardInfo.Driver: HDA-Intel
CardInfo.Name: HD-Audio Generic
CardInfo.LongName: HD-Audio Generic at 0xfeb60000 irq 16
CardInfo.MixerName:
CardInfo.Components: HDA:10ec0671,17341203,00100001

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on January 19, 2022, 09:06:43 am
For onboard sound, you can probably add /A:1 to the uniaud32.sys line in config.sys
Title: Re: Updated Uniaud32 build
Post by: mike on February 06, 2022, 10:21:29 am
That did enable the onboard sound and the mixer is now ok.

Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 1
Card info for adapter 0:
  num: 0
  id: HDMI
  driver: HDA-Intel
  name: HDA ATI HDMI
  longname: HDA ATI HDMI at 0xfeb64000 irq 45
  mixer: ATI R6xx HDMI
  componenets: HDA:1002aa01,00aa0100,00100500
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on March 01, 2022, 07:19:30 pm
Good Morning Guys and Gals!

I'm coming into this conversation very late after about 23 months of continuous response to first CoVid, then Forest Fires, then being called back to service this February, manning the Emergency Operations Centre here in Alberta, dealing with a little occupation disguised as a protest, blocking the international border crossing (I'm starting to think emergencies are the normal state of affairs now).

In January, I read through this thread, catching up on over 700 posts, to understand the nature of what has been accomplished. I would first like to add my compliments as well to both Paul Smedley and Steve Levine for the absolutely massive amount of work undertaken to advance the functionality of the ArcaOS / eCS / OS/2 UniAud32 driver, as well as to the testers who dedicated a LOT of time to provide feedback on the various incarnations of the driver created by Paul, as he advanced the driver closer to the current level of Linux releases.

As I finally have some time to do some testing on my own systems (if, hopefully not being called away yet again), I wanted to add a couple of data points into the mix, based on the sound equipment in my systems. I'm not sure if Paul's updated driver is planned to be added by Lewis to a future ArcaOS update, but, if so, I have found a couple of additional instances of sound not being generated, by the installed hardware.

The level of Pauls driver installed is the 5.14.7-20211028 on the each of two test systems. As I installed the release drive first, and got no sound in certain circumstances, I've substituted the debug driver on each test partition.

Each system has three sound chipsets present in the equipment installed:

HYPERFIRE

Number of CPU cores = 4
CPU name Not available
PMUniMix V0.82
Uniaud32.sys version = 301.1
Number of cards detected = 3
Using card number 0
number of controls = 31


Card 0
F1- A75-V Motherboard
On-Board Audio - ALC892 - Supports 8 channel High Definition Audio

CardInfo.card: 0
CardInfo.id: Generic
CardInfo.Driver: HDA-Intel
CardInfo.Name: HD-Audio Generic
CardInfo.LongName: HD-Audio Generic at 0xfc700000 irq 16
CardInfo.MixerName: Realtek ALC892
CardInfo.Components: HDA:10ec0892,1043841b,00100302

Card 1
On-board HDMI sound - ATI Radeon 5770 Video card

CardInfo.card: 1
CardInfo.id: Generic_1
CardInfo.Driver: HDA-Intel
CardInfo.Name: HD-Audio Generic
CardInfo.LongName: HD-Audio Generic at 0xfc640000 irq 57
CardInfo.MixerName: ATI ID aa01
CardInfo.Components: HDA:1002aa01,00aa0100,00100200

Card 2
Sound Blaster X-Fi Titanium Series Sound Card

CardInfo.card: 2
CardInfo.id: Creative
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA Creative
CardInfo.LongName: HDA Creative at 0xfc200000 irq 56
CardInfo.MixerName: Creative ID d
CardInfo.Components: HDA:1102000d,11020043,00100000


MANTICORE II

Number of CPU cores = 6
CPU name Not available
PMUniMix V0.82
Uniaud32.sys version = 301.1
Number of cards detected = 3
Using card number 0
number of controls = 59

Card 0
M3N-HT Deluxe Motherboard
On-Board ADI AD1988B BH 8 Channel High Definition Audio

CardInfo.card: 0
CardInfo.id: NVidia
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA NVidia
CardInfo.LongName: HDA NVidia at 0xfe020000 irq 22
CardInfo.MixerName: Analog Devices AD1988B
CardInfo.Components: HDA:11d4198b,1043829b,00100400 HDA:10de0002,10de0101,00100000

Card 1

On-board HDMI sound - ASUS Radeon HD 7850 Video Card

CardInfo.card: 1
CardInfo.id: HDMI
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA ATI HDMI
CardInfo.LongName: HDA ATI HDMI at 0xfdafc000 irq 16
CardInfo.MixerName:
CardInfo.Components: HDA:1002aa01,00aa0100,00100300

Card 2

Sound Blaster Recon3D / Z-Series Sound Card

CardInfo.card: 2
CardInfo.id: Creative
CardInfo.Driver: HDA-Intel
CardInfo.Name: HDA Creative
CardInfo.LongName: HDA Creative at 0xfd9fc000 irq 16
CardInfo.MixerName:
CardInfo.Components: HDA:11020011,11020015,00100918

For the system I've named HYPERFIRE above, the following results we're obtained:

Adapter 0 :  No sound, but PMUnimix v0.8.2 reports all 9 normal volume sections - putting sound volumes to the max produces no sound.

Adapter 1: ATI 5770 Video card reports no volume sections under PMUnimix and no sound (NOTE: I do not have an HDMI cable available to connect to my display sound system, but as no sections are present in PMUnimix, I would not expect to have sound)

Adapter 2: Sound Blaster X-Fi Titanium Series Sound Card - No sound but PMUnimix v0.8.2 reports 6 volume sections, as follows:

Headphone + LO Playback Volume (both channels enabled)
Surround Playback Volume (both channels enabled)
Centre Playback Volume (1 channel enabled)
LFE Playback Volume (1 channel enabled)
Side Playback Volume (both channels enabled)
Master Playback Volume (1 channel enabled)
Maximum volume for all channels is 103, if that makes a difference.

For the system I've named MANTICORE II above the following results we're obtained:

Adapter 0 : Sound present and very clear, no problems noted. Sound present in 5 speaker system (Front, rear and bass woofer) which can be adjusted through PMUnimix.

Adapter 1: ATI 5850 Video card reports no volume section settings under PMUnimix and no sound (NOTE: As mentioned previously, I do not have an HDMI cable available to connect to my display sound system, but as no volume sections are present in PMUnimix, I would not expect to have sound)

Adapter 2: Sound Blaster Recon3D / Z-Series Sound Card: Master Playback Volume reported under PMUnimix and sound present but not controlled by Master Playback volume (the only control present), using /a:2 as a switch in config.sys for UniAud32.sys. Required a cold reboot to get sound. Sound is only present from front speakers (sound system attached to this adapter has 2 front, 2 rear and bass speakers - 5 in all) and hardware volume on speakers needs to be turned down as sound is blasting on max (IOW no software sound control).

I don't know, reading about Soudblaster cards in the posts, whether success in getting sound was hit and miss in previous incarnations of Pauls current Uniaud32 driver. I also have not tried the HDMI enabled driver mentioned at the very end of the thread so that may work or not. (still have to test it).

I have attached two testlogs generated, one for each system. For Hyperfire, the testlog was generated with UniAud32 in default mode (using the sound chipset on the motherboard). The second testlog for Manticore II was generated with the /a:2 switch enabled on UniAud32, which pointed to the Soundblaster Recon card. I don't know if you would like me to generate a separate testlog for Hyperfire, with the /a:2 switch enabled.

Anyway, hope this helps development of this driver proceeding in future and Best!

Mark
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on March 11, 2022, 10:32:46 pm
Hi All!

As an update I did download Paul's last update, with HDMI support and installed it. I acquired an HDMI cable and attached the ASUS Radeon 5850 Card to my monitor, which is equipped with a Bang and Olufson sound system. Using the /a:1 switch on Manticore II I attempted to generate sound through the sound chipset reported in the PMUnimix debug report (no volume controls displayed) and got no results. I expected this as no controls were present, but thought I'd be thorough in my testing.

Reading through the posted to this subject, I don't recall seeing any results specifically related to installed sound chipsets on a Radeon card working with UniAud32, but perhaps someone that's got one working can enlighten me as to that, and how to do it and Best!

Mark.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on March 12, 2022, 04:53:20 am
Hi Mark, I think the audio adapters on video cards might be mostly for DRMed content, things like playing a bluray or DRMed streaming like Netflix. They don't want you tapping into the audio (or visual) stream and saving it like you could do with a regular audio adapter. So the audio driver would be part of the video driver for that card on platforms that are supported.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 05, 2022, 12:41:20 am
Hey All,

It's been a while between drinks, but I updated uniaud32 to the 5.15 kernel. I first looked at this almost a year ago, and got lost in trying to implement the missing kernel functionality.

I took a slightly different approach this time, and got a compiled binary, that seems to work in virtualbox. I'll try see if my real hardware works as well over the weekend.

Link is https://smedley.id.au/tmp/uniaud32-linux-5.15-20220805.zip

Next steps will be to get to the current 5.15.x LTS build (5.15.59 at the time of writing)

The 5.15 kernel is a Long Term Support kernel, that has a projected end of life date of October 2023.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 05, 2022, 03:48:30 am
 Hey Paul,

 Wow... thanks for this! It is working fine here so far:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7220000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

 I'll attach alsahlp$ too.

Regards,

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 05, 2022, 04:30:59 am
Updated to 5.15.59 - https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220805.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 05, 2022, 12:10:33 pm
Paul,

  Not getting any sound out of this one, but it seems to be loading:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7220000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

 I'll upload alsahlp$ using the debug version...

Regards,

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 05, 2022, 12:23:20 pm
Thanks Dave, I'll try and investigate over the weekend. I'm also getting a trap with AC97 hardware in VirtualBox.
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on August 05, 2022, 04:01:57 pm
Hi Paul,

Updated to 5.15.59 - https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220805.zip

Similar to David's experience, I am not getting no SOUND with the 5.15.59 release...HOWEVER, the 5.15 kernel update does work quite fine and unless it's my ears playing tricks on me, the sound seems to have higher quality??? ugh...???

Seriously...fired up PM123 and went through some of my music (eg Phil Collins' "In the Air Tonight", 251 kb/s VBR, 44.1 kHz, Joint-Stereo) and it literally kicks ass!!!

Thank you!
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 05, 2022, 05:51:43 pm
same as the others No sound but loads fine.
Title: Re: Updated Uniaud32 build
Post by: Pete on August 05, 2022, 08:21:22 pm
Hi Paul

Updated to 5.15.59 - https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220805.zip

Loads fine but no startup sound.

Settings all look good.

Tried playing Klondike Solitaire but game does not appear onscreen- definitely related to this uniaud update as works fine otherwise.

Also tried Makman/2 which made it to screen and then locked the system requiring a reboot - also related to this uniaud update.

Will have to give uniaud32-linux-5.15-20220805.zip a try later.


Regards

Pete

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 05, 2022, 11:34:52 pm
https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220806.zip might fix the lack of audio...

Or it might not based on my testing in Virtualbox... For those that have no sound with HDA hardware in 5.15.59, does your alsa.log end with something like:
Code: [Select]
channels: 2, period size: b00
channels selected: 2
HWP: SR rate 44100, BPS 16, CH 2, PRSZ 2c0, periods 16
Period size min=2c0 max=2c0
Error -125548448 second time.. Bailing
Unknown error 125548448
U16: StartStream: ConfigDev failed!!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 06, 2022, 02:04:16 am
Paul,

This one is no different - alsahlp$ attached...

Regards, 
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 06, 2022, 02:10:03 am
Hey Dave,
This one is no different - alsahlp$ attached...

Yep...
Code: [Select]
channels: 2, period size: b00
channels selected: 2
HWP: SR rate 44100, BPS 16, CH 2, PRSZ 2c0, periods 16
Period size min=2c0 max=2c0
Error -123398048 second time.. Bailing
Unknown error 123398048
U16: StartStream: ConfigDev failed!!

<sigh> I'll likely end up merging 5.15.20 or so and see if that works, then 5.15.40 and try and isolate the breaking change....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2022, 12:57:32 am
5.15.20 - alsa.log looks OK in Virtualbox - I'm not getting sound, but that's not unusual here when virtualised...
https://smedley.id.au/tmp/uniaud32-linux-5.15.20-20220807.zip

NB: non-HDA hardware will probably still trap with this build - will try investigate that today.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2022, 01:50:16 am
5.15.20 - alsa.log looks OK in Virtualbox - I'm not getting sound, but that's not unusual here when virtualised...
https://smedley.id.au/tmp/uniaud32-linux-5.15.20-20220807.zip

NB: non-HDA hardware will probably still trap with this build - will try investigate that today.

5.15.40 shows the same issue as 5.15.59. Some tweaks appear to resolve the issue...

https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220807.zip
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2022, 03:24:06 am
https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220807b.zip hacks around the issues causing traps with non-HDA hardware (tested with AC97 on Virtualbox).
Title: Re: Updated Uniaud32 build
Post by: David McKenna on August 07, 2022, 05:14:59 am
Hey Paul,

  I went straight for the 20220807b version and it seems to work fine - system sounds, Youtube, PM123, double click an MP3 in the WPS all work great - thanks again! And I agree with Dariusz - the sound seems to be enhanced somehow...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 07, 2022, 08:09:01 am
Hey Dave,
  I went straight for the 20220807b version and it seems to work fine - system sounds, Youtube, PM123, double click an MP3 in the WPS all work great - thanks again! And I agree with Dariusz - the sound seems to be enhanced somehow...

Great to hear! Interesting that the sound 'sounds' better - although I guess that's the point of keeping the code up to date (as far as practicable) with linux :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Sigurd Fastenrath on August 07, 2022, 02:41:58 pm
Hi Paul,

the same here with the b version on my Thinkpad X250:

Perfect Systemsound
Perfect PM123 Version 1.43
Perfect VLC Version 3.0.16 (Audio Setting: K Audio Interface; DART)
Perfect WINOS2 with the Universal WINOS2 Audio Driver

Thank you very much, superb!

And yes, for me it does sound better than with previous versions!

Thank You!!

Sigurd
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on August 07, 2022, 03:58:46 pm
Paul thank you everything except VLC works here too, which may be a problem with VLC
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on August 07, 2022, 05:21:16 pm
Paul, everyone...

Paul thank you everything except VLC works here too, which may be a problem with VLC

The 'b' version boots up fine, regular sounds plays fine, but I did run into a HARD trap during mplayer playback. Specifically, take a look please at the attached mplayer log, looks like some kind of an error in the 'uninit_ao' module, and since I'm using AO: kai:dart config here (as opposed to uniaudio - which plays no sound at all) I'm not quite sure if this points to an actual uniaud driver problem or not.

I re-tested by going back to the earlier 5.15 drop, that works fine, or at least I was NOT able to reproduce the same TRAP with the same video file.

Thanks!
-Dariusz
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 07, 2022, 06:44:11 pm
https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220807b.zip hacks around the issues causing traps with non-HDA hardware (tested with AC97 on Virtualbox).
Hi Paul

I also went straight for the 20220807b and got a trap on VirtualBox.
I tested it with ArcaOS 5.0.7 VM guest on VBox 6.1.36 (Windows 10 host).
- VM ICH AC97: traps
- VM Intel HD Audio: Works, Audio control working, no traps.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 08, 2022, 12:07:40 am
Hey Martin,

https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220807b.zip hacks around the issues causing traps with non-HDA hardware (tested with AC97 on Virtualbox).
I also went straight for the 20220807b and got a trap on VirtualBox.
I tested it with ArcaOS 5.0.7 VM guest on VBox 6.1.36 (Windows 10 host).
- VM ICH AC97: traps
- VM Intel HD Audio: Works, Audio control working, no traps.

I'm pretty interested in the trap on Virtualbox - as I specifically tested AC97 here on Virtualbox (but under Ubuntu). Was this with the debug or release driver? I only tested the debug driver. Are you able to get a serial port log?

Enable serial port in Virtualbox, and there's an option to save the serial port output to a text file.
Make com.sys (or pscom.sys) is loaded
add /p:1 to the uniaud32 driver line in config.sys
reboot

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 08, 2022, 12:54:49 am
Hi Paul

I'm using the "release" version.
- Enabled Serial port and take the output as a RAW file.
- Enabled COM.SYS
- Added /p:1 on Uniaud.
- Rebooted and changed to "ICH AC97" to reproduce the trap.

This is what it showed on the RAW text file.

Quote
Uniaud32 version 3.01.01
Based on linux 5.15.59
pci_register_driver: query_device found 28 8086:2415 class=40100 checking intel8x0
pci_register_driver: matched 0 8086:2415/40100 with 8086:2415/0 0 (intel8x0)

It is something else I need to do?

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 08, 2022, 04:56:13 am
Hey Martin,

Hi Paul

I'm using the "release" version.
- Enabled Serial port and take the output as a RAW file.
- Enabled COM.SYS
- Added /p:1 on Uniaud.
- Rebooted and changed to "ICH AC97" to reproduce the trap.

This is what it showed on the RAW text file.

Quote
Uniaud32 version 3.01.01
Based on linux 5.15.59
pci_register_driver: query_device found 28 8086:2415 class=40100 checking intel8x0
pci_register_driver: matched 0 8086:2415/40100 with 8086:2415/0 0 (intel8x0)

It is something else I need to do?

Thanks - I'll see if I can reproduce with the release version tonight, and try and debug it.

Looks like it's trapping in a similar spot to the initial 5.15.59 build - during the initialisation of the hardware.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: mauro on August 08, 2022, 09:03:02 am


Enable serial port in Virtualbox, and there's an option to save the serial port output to a text file.
Make com.sys (or pscom.sys) is loaded
add /p:1 to the uniaud32 driver line in config.sys
reboot

Cheers,

Paul.

Hi Paul, the uniaud drivers installation is a task I left undone last year (Warp 4.52 on Virtualbox) since I could not obtain sound from video streaming on browser (ie. youtube), which is still my actual issue in soundblaster mode as well. Following these latest posts I feel I could try again now. Where can I find an installation steps how-to, like config.sys editing?
Thank you
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 08, 2022, 11:03:40 am
Hey Martin,

Hey Martin,

Hi Paul

I'm using the "release" version.
- Enabled Serial port and take the output as a RAW file.
- Enabled COM.SYS
- Added /p:1 on Uniaud.
- Rebooted and changed to "ICH AC97" to reproduce the trap.

This is what it showed on the RAW text file.

Quote
Uniaud32 version 3.01.01
Based on linux 5.15.59
pci_register_driver: query_device found 28 8086:2415 class=40100 checking intel8x0
pci_register_driver: matched 0 8086:2415/40100 with 8086:2415/0 0 (intel8x0)

It is something else I need to do?

Thanks - I'll see if I can reproduce with the release version tonight, and try and debug it.

Looks like it's trapping in a similar spot to the initial 5.15.59 build - during the initialisation of the hardware.

I can reproduce the trap with a release build of uniaud32.sys on ac97 hardware. Investigating now.

Edit: really weird why it's trapping where it is with release vs debug....

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on August 08, 2022, 04:10:04 pm


Enable serial port in Virtualbox, and there's an option to save the serial port output to a text file.
Make com.sys (or pscom.sys) is loaded
add /p:1 to the uniaud32 driver line in config.sys
reboot

Cheers,

Paul.

Hi Paul, the uniaud drivers installation is a task I left undone last year (Warp 4.52 on Virtualbox) since I could not obtain sound from video streaming on browser (ie. youtube), which is still my actual issue in soundblaster mode as well. Following these latest posts I feel I could try again now. Where can I find an installation steps how-to, like config.sys editing?
Thank you

Hi Mauro.

I would do it this way.
- Install the full Uniaud using the latest full installer. (ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe))
- Check everything is fine on the VM
- Later install the latest devbuild that Paul's offers here.

Regards
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on August 08, 2022, 09:16:39 pm
Hey Martin,

Thanks - I'll see if I can reproduce with the release version tonight, and try and debug it.

Looks like it's trapping in a similar spot to the initial 5.15.59 build - during the initialisation of the hardware.

Cheers,

Paul.

Hi Paul!

Hope your having a fine day!

I've been off-line for most of the last 4 weeks, as our summer up here is pretty short, and lots of work to do around our quarter acre, as with every summer. But saw you were updating the UniAud32 builds at the moment, and thought I would chime in with my test results as well.

I wasn't one of your original "Merry Men" that have helped out so extensively with the testing, but I did post a full report of my findings on two of my systems back on March 1 when your development build 5.14.7 and the HDMI version of same was last released, and you took a well deserved hiatus (as a result I won't repeat all that, as you can reference back to it in the forum here, with the detailed data and TestLog attachments).

At the time I found that I was able to get that build working perfectly on one of my systems (MANTICORE II) using that version of the driver with a ASUS M3N-HT Motherboard and the on-board ADI AD1988B BH 8 Channel High Definition audio chipset.

However, I was unable to get any sound on the other system (HYPERFIRE) which has an F1- A75V Motherboard equipped with an ALC892 8 channel HDA chipset, using the March releases of UniAud32. On that system, the chipset (Adapter 0) generates no sound but PMUnimix v0.8.2 reports all 9 normal volume sections - putting sound volumes to the max produces no sound. Sound is generated using the UniAud32 version that comes with the ArcaOS package. Tested both the release and debug versions at the time.

Moving on to your latest package (5.15.59b) I have the same results to report - MANTICORE II works well with the latest version, but  5.15.59b still does not generate any sound on HYPERFIRE, using the on-board chipset.

In the March tests I also used the /a:2 switch to address Soundblaster cards I also have installed on each system, which up to this point could only be used in Windows - In MANTICORE II, I have a Soundblaster Recon3D / Z-Series Sound Card, while HYPERFIRE has a Soundblaster X-Fi Titanium Series Sound Card.

I have not tried these cards yet with the new UniAud32 build, but previously HYPERFIRE and the Soundblaster X-Fi reported 6 volume sections under PMUnimix but no sound. MANTICORE II  and the Soundblaster Recon3D reported under PMUnimix the Master Playtback Volume with sound present after a cold reboot, but it did not control the sound. Sound was only present from the front speakers of a speaker set consisting of two front, two rear and one bass speaker. I needed to turn down the hardware speaker volume, as no software sound control seemed to be present.

I have not tested the Soundblasters with 5.15.59b as of yet - will possibly try that later today.

Let me know if you want any further information, Paul, related to the items above, if it would be of any help in development, and thanks for the yeoman's work you've put into advancing development of UniAud32.

As always and Best, Sir!

Mark

 
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on August 08, 2022, 09:59:23 pm
Hi Mauro!


Hi Paul, the uniaud drivers installation is a task I left undone last year (Warp 4.52 on Virtualbox) since I could not obtain sound from video streaming on browser (ie. youtube), which is still my actual issue in soundblaster mode as well. Following these latest posts I feel I could try again now. Where can I find an installation steps how-to, like config.sys editing?
Thank you

Back in March, after going through 46 pages of posts, I put together a Testing Guide for my own use, as a quick reference source.

I spruced it up a bit today and have attached it below - hope you will find it of some use.

If anyone has any comments related to the guide, let me know and I will update it.

Best!

Mark

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on August 09, 2022, 12:37:05 am
Hye Mark,

However, I was unable to get any sound on the other system (HYPERFIRE) which has an F1- A75V Motherboard equipped with an ALC892 8 channel HDA chipset, using the March releases of UniAud32. On that system, the chipset (Adapter 0) generates no sound but PMUnimix v0.8.2 reports all 9 normal volume sections - putting sound volumes to the max produces no sound. Sound is generated using the UniAud32 version that comes with the ArcaOS package. Tested both the release and debug versions at the time.

Are you able to generate a testlog for this system using http://88watts.net/dlfile.php?F=testlog- ? It's quite possible there is a specific 'fix' for this hardware that is currently in code that is commented out (long story, the short version is that openwatcom is old).

Best if you can open a ticket for the hardware at https://trac.netlabs.org/uniaud to save me having to trawl through the forum to find logs.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: mauro on August 09, 2022, 06:59:10 am

Hi Mauro.

I would do it this way.
- Install the full Uniaud using the latest full installer. (ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe))
- Check everything is fine on the VM
- Later install the latest devbuild that Paul's offers here.

Regards

Good, thank you
Title: Re: Updated Uniaud32 build
Post by: mauro on September 01, 2022, 09:13:20 am
Hi Mauro!


Back in March, after going through 46 pages of posts, I put together a Testing Guide for my own use, as a quick reference source.

I spruced it up a bit today and have attached it below - hope you will find it of some use.

If anyone has any comments related to the guide, let me know and I will update it.

Best!

Mark

Thank to you as well Mark,  very useful job
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 03, 2022, 01:08:08 am
Just an update on the issues with non-HDA hardware and the linux 5.15.x based builds.

I now understand what is different between the HDA driver, and the non-HDA drivers - but I'm not clear yet on what I need to do to fix it.

Steven Levine has provided some invaluable assistance, but I'm wagering I'm missing something obvious that is the final piece of the puzzle.

I'll keep working on it as time permits - it's driving me crazy!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 03, 2022, 11:55:20 am
Hey All,

Just an update on the issues with non-HDA hardware and the linux 5.15.x based builds.

I now understand what is different between the HDA driver, and the non-HDA drivers - but I'm not clear yet on what I need to do to fix it.

Steven Levine has provided some invaluable assistance, but I'm wagering I'm missing something obvious that is the final piece of the puzzle.

I'll keep working on it as time permits - it's driving me crazy!

Well after spending way too much time on this yesterday, I had an idea today that seems to have fixed the trap on non-HDA hardware.

New debug and release builds at https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220903.zip

Note: with my AC97 testing on Virtualbox, I'm getting an error initialising the hardware with the release build on a cold boot. If I do a cold boot with the debug driver, then switch to the release driver, then sound works fine - so some secret sauce is still missing :/

Interested in any feedback on this build - even to ensure that I didn't break anything on HDA hardware - but particularly interested in any pci based sound cards, and AC97 - as those use a different code path to the HDA driver.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 03, 2022, 05:56:30 pm
Paul just letting you know the the HDA driver is still working well on my system.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 04, 2022, 03:48:36 am
This one should fix non-HDA hardware (at least in my limited testing):
https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220904.zip
Title: Re: Updated Uniaud32 build
Post by: mauro on September 04, 2022, 10:37:54 am

Enable serial port in Virtualbox, and there's an option to save the serial port output to a text file.
Make com.sys (or pscom.sys) is loaded
add /p:1 to the uniaud32 driver line in config.sys
reboot

Cheers,

Paul.

this is something my VirtualBox prevent me doing.
If I enable Serial Port Com 1 in "file Raw" mode , and set a local folder on my host , then VM machine won't boot : "VERR_ACCESS_DENIED" on folder (it's my Public Document folder on Windows)
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 04, 2022, 11:16:53 am
Minor updates in https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220904b.zip
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 04, 2022, 11:17:39 am

Enable serial port in Virtualbox, and there's an option to save the serial port output to a text file.
Make com.sys (or pscom.sys) is loaded
add /p:1 to the uniaud32 driver line in config.sys
reboot

this is something my VirtualBox prevent me doing.
If I enable Serial Port Com 1 in "file Raw" mode , and set a local folder on my host , then VM machine won't boot : "VERR_ACCESS_DENIED" on folder (it's my Public Document folder on Windows)

Sorry, I have NFI about Virtualbox ona a Windows host... Works a treat on Ubuntu :)
Title: Re: Updated Uniaud32 build
Post by: mauro on September 04, 2022, 11:23:50 am

Enable serial port in Virtualbox, and there's an option to save the serial port output to a text file.
Make com.sys (or pscom.sys) is loaded
add /p:1 to the uniaud32 driver line in config.sys
reboot

this is something my VirtualBox prevent me doing.
If I enable Serial Port Com 1 in "file Raw" mode , and set a local folder on my host , then VM machine won't boot : "VERR_ACCESS_DENIED" on folder (it's my Public Document folder on Windows)

Sorry, I have NFI about Virtualbox ona a Windows host... Works a treat on Ubuntu :)

ok I have same OS2 VM Ubuntu hosted too, I'll try that later on, the fact is that my "OS2 VM for testing" is on Windows.
Thank you
Title: Re: Updated Uniaud32 build
Post by: mauro on September 04, 2022, 11:24:29 am


Hi Mauro.

I would do it this way.
- Install the full Uniaud using the latest full installer. (ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe))
- Check everything is fine on the VM
- Later install the latest devbuild that Paul's offers here.

Regards

Done,
in VM settings Windows Direct Sound as driver host , and ICH AC97 as controller

and also added
DEVICE=C:\MMOS2\UNIAUD32.SYS
DEVICE=C:\MMOS2\UNIAUD16.SYS /C

statement in config.sys as per Mark's how-to indications.

When VM boots, I get the message "The system cannot find the file C:\MMOS2\UNIAUD16.SYS"
infact no such file there. I expected it was at least installed by the Uniaud-20211023.exe execution.
Must be there?
Paul's patch does not have this file in the Zip folders, only 32 version which I copied in C:\MMOS2 .
So, no sound in VM (Warp 4.52) by these steps
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 04, 2022, 12:00:24 pm


Hi Mauro.

I would do it this way.
- Install the full Uniaud using the latest full installer. (ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe (http://ftp://ftp.netlabs.org/pub/uniaud/Uniaud-20211023.exe))
- Check everything is fine on the VM
- Later install the latest devbuild that Paul's offers here.

Regards

Done,
in VM settings Windows Direct Sound as driver host , and ICH AC97 as controller

and also added
DEVICE=C:\MMOS2\UNIAUD32.SYS
DEVICE=C:\MMOS2\UNIAUD16.SYS /C

statement in config.sys as per Mark's how-to indications.

When VM boots, I get the message "The system cannot find the file C:\MMOS2\UNIAUD16.SYS"
infact no such file there. I expected it was at least installed by the Uniaud-20211023.exe execution.
Must be there?
Paul's patch does not have this file in the Zip folders, only 32 version which I copied in C:\MMOS2 .
So, no sound in VM (Warp 4.52) by these steps

I woudl absolutely expect Uniaud-20211023.exe to include uniaud16.sys...

Just confirmed with wic that it does.

try:
wic -x u:Uniaud-20211023.exe 1

...and copy uniaud16.sys to x:\mmos2

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: mauro on September 04, 2022, 01:46:10 pm


try:
wic -x u:Uniaud-20211023.exe 1

...and copy uniaud16.sys to x:\mmos2

Cheers,

Paul

"SYS1041: The name wic is not recognized as an
internal or external command, operable program or batch file."

a missing application to be installed?
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 04, 2022, 03:01:53 pm


try:
wic -x u:Uniaud-20211023.exe 1

...and copy uniaud16.sys to x:\mmos2

Cheers,

Paul

"SYS1041: The name wic is not recognized as an
internal or external command, operable program or batch file."

a missing application to be installed?

Hi Mauro.
wic.exe is located in C:\sys\install\WARPIN. You can add it to your path and libpath to run it on the command line session.

Regards
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 04, 2022, 03:18:54 pm
 Hi Paul,

  I gave these new builds a try, 20220904b gives me a TrapE on boot. 20220903 and 20220904 both seem to work at first, but I lose sound altogether within a few minutes of booting. Haven't determined what might be causing that though. My sound card:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.04
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7220000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

Regards,
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 04, 2022, 03:41:28 pm
Minor updates in https://smedley.id.au/tmp/uniaud32-linux-5.15.59-20220904b.zip
Hi Paul

I tested it (Release version) with ArcaOS 5.0.7 VM guest on VBox 6.1.36 (Windows 10 21H2 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Works, Audio control working, no traps.

Regards
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 04, 2022, 06:57:20 pm
Heh Paul!

Hope your having a fine day.

So have you been looking around for another companion to replace loyal Hudson? A tough time without your friend of many years, but I'm sure walks in that lovely scenery of yours go better with another loyal dog around ......... !

Hye Mark,

However, I was unable to get any sound on the other system (HYPERFIRE) which has an F1- A75V Motherboard equipped with an ALC892 8 channel HDA chipset, using the March releases of UniAud32. On that system, the chipset (Adapter 0) generates no sound but PMUnimix v0.8.2 reports all 9 normal volume sections - putting sound volumes to the max produces no sound. Sound is generated using the UniAud32 version that comes with the ArcaOS package. Tested both the release and debug versions at the time.

Are you able to generate a testlog for this system using http://88watts.net/dlfile.php?F=testlog- ? It's quite possible there is a specific 'fix' for this hardware that is currently in code that is commented out (long story, the short version is that openwatcom is old).

Best if you can open a ticket for the hardware at https://trac.netlabs.org/uniaud to save me having to trawl through the forum to find logs.

Cheers,

Paul.

Sorry for not replying back until today, Sir - was called back to emergency service for a couple of weeks, then trying to catch up on my outside property work - never a dull moment in this Year of Strange.

I saw your latest posts and downloaded the UniAud32 20220903 package and installed it on the two systems I mentioned previously, to test as you requested,

On Manticore II, which has the ADI AD1988B HDA chipset on the motherboard again no problems at all - PMUniMix reports 11 volume control sections, and preliminary testing with PM123 played perfectly.

On Hyperfire, the system I was having problems with, (On-board ALC892 HDA chipset) again no sound but 10 volume sections are reported by PMUniMix (just noted the missing section is CD Playback Volume, which is reported on Manticore II). I tried the release version first, then replaced it with the debug version - no difference.

I generated a new testlog, which is attached below - it was generated with the debug version of your 20220903 test driver - if you want I can still open a ticket on https://trac.netlabs.org/uniaud - didn't do it previously as I was rushing out the door the day you sent the reply - forest fires wait for no one <sic>.

I haven't got around to testing the Soundblaster cards installed on the two systems, but still plan to try them, when I get a free hour or so, somewhere.

Best as always !

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 04, 2022, 11:42:02 pm
Hi Mauro

Hi Mauro!


Back in March, after going through 46 pages of posts, I put together a Testing Guide for my own use, as a quick reference source.

I spruced it up a bit today and have attached it below - hope you will find it of some use.

If anyone has any comments related to the guide, let me know and I will update it.

Best!

Mark

Thank to you as well Mark,  very useful job

Posting to Paul at the moment, I realized that I did not update the Testing Guide to v1.1 at the beginning of August, after a very convoluted month.

Attached is the updated version (v1.1) which corrects a few things, as well as adding references to the latest TestLog (v3.5.0), its links, and references to opening a TRAC ticket, to NetLabs.org, if you want to report a problem with Paul's version of the drivers.

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 05, 2022, 12:00:29 am
Hi Dave,
Hi Paul,

  I gave these new builds a try, 20220904b gives me a TrapE on boot. 20220903 and 20220904 both seem to work at first, but I lose sound altogether within a few minutes of booting. Haven't determined what might be causing that though. My sound card:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.04
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7220000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

Regards,

Can you create a 'testlog uniaud' with 0904, then post the trapscreen from 0904b?

The only change in 0904b is http://trac.netlabs.org/uniaud/changeset/723 which should have been pretty safe, and doesn't trap in VirtualBox or on a T470s.

Interesting stuff!

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 05, 2022, 02:00:06 am
 OK... while the original report was using the 'release' version, these reports are using the 'debug' version.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 05, 2022, 11:59:28 am
Hey all,

Updated the source to 5.15.65 - nothing major in this, but a few updates to cirrus, connexant, realtek HDA code and to the core code.

@David - I'm still pondering your trap - I don't expect this to help...

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220905.zip

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 05, 2022, 05:52:05 pm
https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220905.zip

Hi Paul

Again, I tested it (Release version) with ArcaOS 5.0.7 VM guest on VBox 6.1.36 (Windows 10 21H2 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Works, Audio control working, no traps.

I just noticed that with "ICH AC97" the audio on the VM sounds louder than with "Intel HD Audio". I'm not sure if always worked that way, and I just noticed now.

Regards
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on September 05, 2022, 08:17:02 pm
Hey all,

Updated the source to 5.15.65 - nothing major in this, but a few updates to cirrus, connexant, realtek HDA code and to the core code.

@David - I'm still pondering your trap - I don't expect this to help...

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220905.zip

Cheers,

Paul

Up to this one, I have had no issues, however, I get a trap e with this one on boot (warm or cold).
Back to the original uniaud to boot, unimix shows this about the card:
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf2530000 irq 22
  mixer: Realtek ALC269VC
  componenets: HDA:10ec0269,17aa2204,00100203

 I can put the trapping uniaud32.sys back in and get more trap info (or a dump) but I will probably have to hand write the data as the trap superimposed over the previous boot messages such that I do not think a picture would be very clear.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 05, 2022, 09:23:07 pm
Hi Paul!

Rainy day today, the first in about a month, so had some time for more testing

Hey all,

Updated the source to 5.15.65 - nothing major in this, but a few updates to cirrus, connexant, realtek HDA code and to the core code.

@David - I'm still pondering your trap - I don't expect this to help...

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220905.zip

Cheers,

Paul

Tried your 5.15.65 release on both systems:

On Manticore II, with the AD1988B chipset, again things working perfectly

However, on Hyperfire, with the ALC892 chipset, I seem to get Davids Trap000e (attached the JPG below)

I've tested both the Release and the Debug versions of 5.15.65, and get the same trap and result (or at least the results look the same - some slight differences in the various sections reported).

However, reverting to the 5.15.59 version from 20220903 I still get no sound - but PMUnimix reports 10 volume sections as per my previous post.

So something is happening in 5.15.65, but not on everything - just thought you'd like to know that the trap is reproducible, in this case in 5.15.65, even though its on an HDA system that is not working (ie. no sound produced) - perhaps its the HDA code you added for cirrus, connexant, and realtek that caused the break this time, at least for me (probably not the core code, as the AD1988B HDA chipset is still working perfectly).

Yet more of a puzzler if your not puzzled out already, Paul, and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Jan-Erik Lärka on September 05, 2022, 09:35:35 pm
Errr, that's the other computer.
Tried your 5.15.6 as i've got conextant hardware and can also see a trap here during boot.
Reverted back to 3.01.01 that work with the hardware.
//Jan-Erik

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 05, 2022, 09:42:29 pm
Heh Jan-Erik!

NTried your 5.15.6 as i've got ALC888 hardware (Conexant) and can also see a trap here during boot.
Reverted back to 5.10.7 that work with the hardware.
//Jan-Erik

As my ALC892 chipset is obviously also Conexant, I don't recall if I tried that version of Paul's driver - has you had any issues with 5.10.7 at all and Best!

Mark
Title: Re: Updated Uniaud32 build
Post by: Jan-Erik Lärka on September 05, 2022, 10:43:50 pm
Hello Mark,

Sorry, I mixed the two up, Conexant is present on the Sony Vaio Laptop.

But my guess is that I'll meet the same issue you have with ALC892 as mine is ALC888 on an AMD Quad Core machine.

I think that 3.01.01 was the last one that's ok as newer versions I've tried tend to be a bit sluggish/slow so they can stall/freeze if one use mplayer and jump in short intervals. I skip uninteresting parts of movies/clips.  :)

Heh Jan-Erik!

As my ALC892 chipset is obviously also Conexant, I don't recall if I tried that version of Paul's driver - has you had any issues with 5.10.7 at all and Best!

Mark
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 05, 2022, 10:51:13 pm
For those with traps with 0904b and/or 0905 can you please confirm 0904 works? The key change from 0904 to 0904b was the change in kzalloc so I'll look to revert that change this evening, but some more data would be useful.
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on September 05, 2022, 11:43:24 pm
For those with traps with 0904b and/or 0905 can you please confirm 0904 works? The key change from 0904 to 0904b was the change in kzalloc so I'll look to revert that change this evening, but some more data would be useful.
I had missed 0904b, it too does trap.  0904 does not trap.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 06, 2022, 01:17:59 am
Thanks Andy, I'll revert the kzalloc change this evening my time and provide a new build.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 06, 2022, 02:33:10 am
Thanks Andy, I'll revert the kzalloc change this evening my time and provide a new build.

Same here Paul - no sound still, on Hyperfire, but at least no trap using the 04 build.

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 06, 2022, 11:17:26 am
For those who had traps with 0904b or 0905 - please try with https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220906.zip - this build reverts the changes in https://trac.netlabs.org/uniaud/changeset/723 (https://trac.netlabs.org/uniaud/changeset/723)
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 06, 2022, 04:46:29 pm
Hi Paul!

For those who had traps with 0904b or 0905 - please try with https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220906.zip - this build reverts the changes in https://trac.netlabs.org/uniaud/changeset/723 (https://trac.netlabs.org/uniaud/changeset/723)

The 5.15.65 20220906 build seems to have cleared the trap 000e on Hyperfire with the HDA ALC892 chipset. Still no sound, but exactly the same state as the 5.15.59 20220904 build (no sound but 10 volume sections reported in PMUnimix).

Just tested Manticore II on the 20220906 build as well, with the HDA AD1988B chipset. everything humming along there as well with the new build, as per all the recent previous builds - system sounds working and PM123 plays sound perfectly.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on September 07, 2022, 02:58:08 am
For those who had traps with 0904b or 0905 - please try with https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220906.zip - this build reverts the changes in https://trac.netlabs.org/uniaud/changeset/723 (https://trac.netlabs.org/uniaud/changeset/723)
Works here.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 07, 2022, 04:49:38 am
Hey Guys,

For those who had traps with 0904b or 0905 - please try with https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220906.zip - this build reverts the changes in https://trac.netlabs.org/uniaud/changeset/723 (https://trac.netlabs.org/uniaud/changeset/723)
Works here.

Thanks for confirming - next to figure out why the revised kzalloc implementation wasn't working!
Title: Re: Updated Uniaud32 build
Post by: Lars on September 07, 2022, 09:16:51 am
I can tell you why:
if you allocate more than a page, kmalloc uses vmalloc and returns right away, without applying the zeroing flag.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 07, 2022, 11:12:44 am
Hey Lars!

I can tell you why:
if you allocate more than a page, kmalloc uses vmalloc and returns right away, without applying the zeroing flag.

Thanks!! I haven't commited the change yet, but https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220907.zip will hopefully still work for those that previously had traps :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Lars on September 07, 2022, 11:18:08 am
I also wonder where these umpteen allocation routines come from. They all seem to do about the same thing.
Another thing: "copy_from_user" and "copy_to_user" could make good use of the KEE routines "KernCopyIn" and "KernCopyOut" respectively. These will catch traps if the user  buffer address or the access to it is invalid and will fail gracefully without bringing down the whole system.
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on September 07, 2022, 11:22:26 am
Thanks!! I haven't commited the change yet, but https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220907.zip will hopefully still work for those that previously had traps :)

TRAP is gone on my TP X250 after installing the latest version "uniaud32-linux-5.15.65-20220907.zip".
Thank you Paul.

Greetings, Mike
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 07, 2022, 04:12:58 pm
Hey Lars!

I can tell you why:
if you allocate more than a page, kmalloc uses vmalloc and returns right away, without applying the zeroing flag.

Thanks!! I haven't commited the change yet, but https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220907.zip will hopefully still work for those that previously had traps :)

Cheers,

Paul

Hi Paul!

Running out the door today but just tried 0907:

Manticore II - The new build works fine with the AD1988B chipset, as always

Hyperfire - No Traps with the ALC892 chipset, but still no sound, so same status

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 07, 2022, 04:24:30 pm
Paul latest drop working well here ad in fact soundig quit good. Cleaner sound it seems.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 07, 2022, 11:59:05 pm
Hi Paul,

  Build 20220907 is working fine here... thanks!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 08, 2022, 11:23:14 am
https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220908.zip is now available - adds support for the __GFP_ZERO flag in __vmalloc, which fixes audio on some emu10k1 audio cards.

This may also fix some other stuff too - certainly shouldn't break anything :)
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on September 08, 2022, 12:14:29 pm
https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220908.zip is now available - adds support for the __GFP_ZERO flag in __vmalloc, which fixes audio on some emu10k1 audio cards.

This may also fix some other stuff too - certainly shouldn't break anything :)

Hi Paul,

works on my TP X250. tested with system sound and pm123  :) - Thanks.

Greetings, Mike
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 08, 2022, 12:16:35 pm
Thanks for testing Mike, now that most of the uniaud32 issues appear to be resolved, I'll get back to stabilising smbd - in particular when sharing resources that are Netdrive mounts.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 08, 2022, 08:04:57 pm
Hey Paul!

Thanks for testing Mike, now that most of the uniaud32 issues appear to be resolved, I'll get back to stabilising smbd - in particular when sharing resources that are Netdrive mounts.

Tested UniAud32 5.15.65 20220908 on both systems:

Manticore II - AD1988B - still functioning perfectly - system sounds present and PM123 works perfectly

Hyperfire - still no sound with the ALC892 chipset, neither system sounds nor PM123  - tested with both the Release and Debug drivers. UniMix still reports 10 volume sections - volumes were set to max.

I've attached an updated Testlog for Hyperfire, which was generated today with 0908 and the Debug version you built. If you wish I will generate a TRAC ticket.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 09, 2022, 12:39:32 am
Hey Mark,

I've attached an updated Testlog for Hyperfire, which was generated today with 0908 and the Debug version you built. If you wish I will generate a TRAC ticket.

I just had a quick look at the log, the part that seems strange to me is:
Code: [Select]
codec #1 probed OK

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

Unable to bind the codec

Cannot probe codecs, giving up

I need to check against another system, but I don't believe these lines are normal. I'll do some checks tonight and also chekc the relevant code to see if here is anything enlightening.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 09, 2022, 03:15:07 pm
Hi Paul

Hey Mark,

I've attached an updated Testlog for Hyperfire, which was generated today with 0908 and the Debug version you built. If you wish I will generate a TRAC ticket.

I just had a quick look at the log, the part that seems strange to me is:
Code: [Select]
codec #1 probed OK

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

Unable to bind the codec

Cannot probe codecs, giving up

I need to check against another system, but I don't believe these lines are normal. I'll do some checks tonight and also chekc the relevant code to see if here is anything enlightening.

Cheers,

Paul

I generated a testlog for UniAud on the working system (Manticore II) which I've attached below, if that helps out at all - not the same chipset, but may help for comparison purposes.

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 09, 2022, 11:29:52 pm
I generated a testlog for UniAud on the working system (Manticore II) which I've attached below, if that helps out at all - not the same chipset, but may help for comparison purposes.

It saved me generating one myself :)

As I suspected, your other log has:
Code: [Select]
codec #0 probed OK

codec #3 probed OK

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

Unfortunately, I have no real ideas right now on why this is happening :(
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 10, 2022, 12:16:11 am
@Mark - I don't really expect this to help - but this does update some functions that are mentioned in https://lkml.org/lkml/2021/11/15/2038

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910.zip

I also added a #define for a HDMI codec - which who knows, might help HDMI audio - I don't have an easy way to test this right now.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 10, 2022, 02:06:46 am
Hey Paul

@Mark - I don't really expect this to help - but this does update some functions that are mentioned in https://lkml.org/lkml/2021/11/15/2038

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910.zip

I doubt you claim omnipotence, and I certainly don't, Sir - but I'll try the latest update.

Like always, at least we know what doesn't work ............

In emergency's we always try bright ideas when common sense fails - sometimes the most absurd succeeds ......... why is a mystery to me, even after 42 years doing this ......... !

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 10, 2022, 02:24:57 am
I also just refreshed the zip to try and load the azx driver (aka hda) driver first - previously it would iterate through ~10 other drivers first - given the prevalence of hda hardware these days, I figured it made sense to try HDA first - might shave a few milliseconds off of the boot time on HDA hardware :P
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 10, 2022, 02:26:25 am
Hey Mark!

@Mark - I don't really expect this to help - but this does update some functions that are mentioned in https://lkml.org/lkml/2021/11/15/2038

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910.zip

I doubt you claim omnipotence, and I certainly don't, Sir - but I'll try the latest update.

Like always, at least we know what doesn't work ............

In emergency's we always try bright ideas when common sense fails - sometimes the most absurd succeeds ......... why is a mystery to me, even after 42 years doing this ......... !

I certainly don't claim omnipotence :) I'm just a hacker who enjoys trying to make things work, even though it consumes time I could probably better spend elsewhere :P

edit: https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910b.zip implements pci_intx - which was previously stubbed. This is only used for HDA hardware - hda continues to work for me in Virtualbox and on a T470s (ie bare metal).

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 10, 2022, 01:36:15 pm
edit: https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910b.zip implements pci_intx - which was previously stubbed. This is only used for HDA hardware - hda continues to work for me in Virtualbox and on a T470s (ie bare metal).

Hi Paul
The link is not working for me.
Regards
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 10, 2022, 04:33:08 pm
Hey Paul!

Hey Mark!

@Mark - I don't really expect this to help - but this does update some functions that are mentioned in https://lkml.org/lkml/2021/11/15/2038

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910.zip

I doubt you claim omnipotence, and I certainly don't, Sir - but I'll try the latest update.

Like always, at least we know what doesn't work ............

In emergency's we always try bright ideas when common sense fails - sometimes the most absurd succeeds ......... why is a mystery to me, even after 42 years doing this ......... !

I certainly don't claim omnipotence :) I'm just a hacker who enjoys trying to make things work, even though it consumes time I could probably better spend elsewhere :P

edit: https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910b.zip implements pci_intx - which was previously stubbed. This is only used for HDA hardware - hda continues to work for me in Virtualbox and on a T470s (ie bare metal).

Cheers,

Paul

Got tied up last night.

Ahhhh, a man after my own heart ............

I got into first the crazy business of Hazmat (Hazardous Materials) response, then emergency response in general because I'm a troubleshooter to the core - I want things to work, and in emergencies, you first have to stabilize the situation, and then scratch your head as to how to clean-up sometimes very unusual messes and troubles .........

It takes patient and a very thick skin in some situations ........... but after four decades I still love the craziness ............ !

As I've said many times, I'm not a programmer (but thinking of learning REXX this fall, to see if I can assist in pushing another project forward). But I get wanting to drop more mundane things to take on a challenge. When things start working its always a rush, having accomplished something that potentially no one else has done

Whether we can get ALC892 to work or not is irrelevant - but I appreciate you scratching your head and trying to figure it out ...........

And as Martin mentioned, the 10b link is coming up as a 404 - perhaps you took it down, after having yet another bright crazy idea .........

In the meantime will try the 10 release anyway and report back sometime today.

BTW Anne (my wife) and I were out visiting her cousin, who is a dog breeder, and were absolutely mobbed by a herd of 5 Miniature Schnauzers on Thursday, who took an instant liking to us - nearly took two of those middle weight mustached cuties home - which would probably have upset the current feline household of two .......... !

Best!

M


Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 10, 2022, 07:06:54 pm
Heh Paul!

@Mark - I don't really expect this to help - but this does update some functions that are mentioned in https://lkml.org/lkml/2021/11/15/2038

https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910.zip

I also added a #define for a HDMI codec - which who knows, might help HDMI audio - I don't have an easy way to test this right now.

Tried 5.15.65 20220910 on Hyperfire with the same results - didn't break anything but still no sound.

The IRQ 19 and 16 entries you pointed out are still in the latest log I generated today where "giving up" was mentioned regarding the probe

Skimming down the testlog I also noted these entries - don't know if they have any bearing on the problem but thought I'd mention them: 

realtek: No valid SSID, checking pincfg 0x4005e601 for NID 0x1d

realtek: Enabling init ASM_ID=0xe601 CODEC_ID=10ec0892

The problem may be related somehow specifically to the Realtek HDA codec (not being a programmer I'm not sure If I'm using the correct terminology here), and maybe to the SSID for the ALC892 chipset not being recognized by same. Am I making any sense?

M


Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 10, 2022, 11:44:42 pm
edit: https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910b.zip implements pci_intx - which was previously stubbed. This is only used for HDA hardware - hda continues to work for me in Virtualbox and on a T470s (ie bare metal).

Hi Paul
The link is not working for me.
Regards

oops, fixed now.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 11, 2022, 12:16:56 am
Hey Paul!

edit: https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220910b.zip implements pci_intx - which was previously stubbed. This is only used for HDA hardware - hda continues to work for me in Virtualbox and on a T470s (ie bare metal).

Hi Paul
The link is not working for me.
Regards

oops, fixed now.

Tried 5.15.65 20220910b on Hyperfire and exactly the same as 10 in the previous builds. (no sound - 10 sound sections reported under PMUnimix)

Generated a Testlog and skimmed it - the sections we discussed previously are still exactly the same
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 11, 2022, 02:16:33 am
Hey Mark, I'm unfortunately pretty much out of ideas for Hyperfire. It would be interesting to boot that system from an Ubuntu 22.04.1 live image (https://releases.ubuntu.com/22.04/) - that release runs a 5.15 kernel so it should be pretty similar code.

Edit: I just noticed something interesting.... the output of unimix -card for Hyperfire is:
Card info for adapter 0:
  num: 0
  id: Generic
  driver: HDA-Intel
  name: HD-Audio Generic
  longname: HD-Audio Generic at 0xfc700000 irq 16
  mixer: Realtek ALC892
  componenets: HDA:10ec0892,1043841b,00100302

Other systems (such as Mantiscore and others) have:
Card info for adapter 0:
  num: 0
  id: NVidia
  driver: HDA-Intel
  name: HDA NVidia
  longname: HDA NVidia at 0xfe020000 irq 22
  mixer: Analog Devices AD1988B
  componenets: HDA:11d4198b,1043829b,00100400 HDA:10de0002,10de0101,00100000

I wonder if this is part of the problem?

Edit2: perhaps not... from the code:
   /* AMD/ATI Generic, PCI class code and Vendor ID for HD Audio */
   { PCI_DEVICE(PCI_VENDOR_ID_ATI, PCI_ANY_ID),
     .class = PCI_CLASS_MULTIMEDIA_HD_AUDIO << 8,
     .class_mask = 0xffffff,
     .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_HDMI },
   { PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_ANY_ID),
     .class = PCI_CLASS_MULTIMEDIA_HD_AUDIO << 8,
     .class_mask = 0xffffff,
     .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_HDMI },

So if the PCI Vendor ID matches AMD, it uses AZX_DRIVER_GENERIC - but it could still be something to do with the use of AZX_DRIVER_GENERIC vs the other specific drivers...
Title: Re: Updated Uniaud32 build
Post by: Lars on September 11, 2022, 10:46:22 am
I am not yet out of ideas:
if it is still the "giving up" error, I'd make routine "azx_probe" the same as "azx_probe_continue": call in both routines "pci_set_drvdata(PCI,NULL)" and "snd_card_free(card)" on the out_free error exit point. My suspicion is that the probing is aborted prematurely.
That's because azx_probe allocates a card but if it determines later on that there was an error, it does not reset the card info (the same explanation as for "azx_probe_continue" should be true).
I see another risk (maybe): "azx_,probe_continue" is called from within "azx_probe". But both call "snd_card_free" on error exit. Let's hope that is not a problem.
All that in file hda_intel.c.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 11, 2022, 11:30:45 am
Hi Mark,

Tried 5.15.65 20220910b on Hyperfire and exactly the same as 10 in the previous builds. (no sound - 10 sound sections reported under PMUnimix)

Generated a Testlog and skimmed it - the sections we discussed previously are still exactly the same

Small change in <<link removed cos it traps>> It seems that for HDA Generic, it wants to load a codec via a module. We don't support modules, so it doesn't try and load the code - but the necessary object files are in the binary, so I've tried to workaound this...

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 11, 2022, 03:09:42 pm
tHE uNIAUD32-LINUX-5.15.65-20220911 is show stopper for me. Traps all pratitions.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 11, 2022, 03:50:32 pm
Paul,

 20220910b works fine, but I get a trap with 20220911 too...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 11, 2022, 08:53:17 pm
I get the same exac trap as David.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 12, 2022, 01:25:00 am
Sorry about that, of course it didn't trap here in Virtualbox - unfortunately I didn't test it on bare metal. Are your systems AMD or Intel?

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 12, 2022, 03:01:12 am
My system are all Intel. The sound card is a Creative Recon.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 12, 2022, 03:28:11 am
Hey Eugene,

My system are all Intel. The sound card is a Creative Recon.

Interesting - are you able to provide a 'testlog uniaud' for this system?

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 12, 2022, 04:36:56 am
I could but not with that driver installed. If you want a baseline with say the default drive or one of the working drivers yes I can and wil do so tomorrow.
.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 12, 2022, 05:41:08 am
Hi Paul!

Hi Mark,

Tried 5.15.65 20220910b on Hyperfire and exactly the same as 10 in the previous builds. (no sound - 10 sound sections reported under PMUnimix)

Generated a Testlog and skimmed it - the sections we discussed previously are still exactly the same

Small change in https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220911.zip It seems that for HDA Generic, it wants to load a codec via a module. We don't support modules, so it doesn't try and load the code - but the necessary object files are in the binary, so I've tried to workaound this...

Cheers,

Paul

Busy Sunday - just writing this before I retire to bedlam.

Tried UniAud32 20220911 on both bare metal AMD machines and got the trap both David and Eugene reported. So not much to report.

If your looking for other bare metal machines to test on I have 4 towers and multiple portables, some legacy.

Best!

M

Title: Re: Updated Uniaud32 build
Post by: Andy Willis on September 12, 2022, 05:43:38 am
Hey Eugene,

My system are all Intel. The sound card is a Creative Recon.

Interesting - are you able to provide a 'testlog uniaud' for this system?

Cheers,

Paul.
This data is from 10b, 11 traps here as well:

Code: [Select]
----- Output of: unimix.exe -card ----------------------------------------------
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf2530000 irq 22
  mixer: Realtek ALC269VC
  componenets: HDA:10ec0269,17aa2204,00100203
----- Output of: unimix.exe -list ----------------------------------------------
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Control list for adapter 0

Control ID: 1
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Speaker Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 87, step: 0
  value 0: 0
  value 1: 0

Control ID: 2
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Speaker Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 3
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Headphone Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 87, step: 0
  value 0: 80
  value 1: 80

Control ID: 4
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Headphone Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 5
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Headphone Playback Switch]
Index: 1
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1
  value 1: 1

Control ID: 6
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Loopback Mixing]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 2, item: 0, name: [Disabled]
  value 0: 0

Control ID: 7
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Mic Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 31, step: 0
  value 0: 0
  value 1: 0

Control ID: 8
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Mic Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 9
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Dock Mic Playback Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 31, step: 0
  value 0: 0
  value 1: 0

Control ID: 10
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Dock Mic Playback Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 11
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Auto-Mute Mode]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 2, item: 0, name: [Disabled]
  value 0: 1

Control ID: 12
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Capture Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 63, step: 0
  value 0: 0
  value 1: 0

Control ID: 13
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Capture Switch]
Index: 0
Element type: boolean type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 14
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Mic Boost Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 3, step: 0
  value 0: 0
  value 1: 0

Control ID: 15
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Dock Mic Boost Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 3, step: 0
  value 0: 0
  value 1: 0

Control ID: 16
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Internal Mic Boost Volume]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0
  value 1: 0

Control ID: 17
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Master Playback Volume]
Index: 0
Element type: integer type
Count of values: 1
Value:
 Bounds: min: 0, max: 87, step: 0
  value 0: 87

Control ID: 18
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Master Playback Switch]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 1

Control ID: 19
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Mic Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 20
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Dock Mic Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 21
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Internal Mic Phantom Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 2

Control ID: 22
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Speaker Phantom Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 4

Control ID: 23
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Headphone Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 24
Interface: global control
Device(client): 0, subdevice (substream) 0
Name: [Dock Headphone Jack]
Index: 0
Element type: boolean type
Count of values: 1
Value:
 Bounds: min: 0, max: 1, step: 0
  value 0: 0

Control ID: 25
Interface: PCM device
Device(client): 0, subdevice (substream) 0
Name: [Playback Channel Map]
Index: 0
Element type: integer type
Count of values: 4
Value:
 Bounds: min: 0, max: 36, step: 0
  value 0: 3
  value 1: 4
  value 2: 0
  value 3: 0

Control ID: 26
Interface: PCM device
Device(client): 0, subdevice (substream) 0
Name: [Capture Channel Map]
Index: 0
Element type: integer type
Count of values: 2
Value:
 Bounds: min: 0, max: 36, step: 0
  value 0: 0
  value 1: 0
----- Output of: unimix.exe -pcms ----------------------------------------------
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
PCM Instance Info for adapter 0
PCM No 0 info:
PCM dev name: [ALC269VC Analog] mixer [subdevice #0]
 PLAY
  streams: 1
  channels: from 2 to 4
  rates: from 44100 to 192000
 RECORD
  streams: 1
  channels: from 2 to 2
  rates: from 44100 to 192000

----- Output of: unimix.exe -powerget ------------------------------------------
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Getting power state for adapter 0: ...power state=0
----- Contents of file: alsahlp$ -----------------------------------------------
Uniaud32 version 3.01.01
Based on linux 5.15.65
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking hda_intel
pci_register_driver: matched 2 8086:1e20/40300 with 8086:1e20/0 0 (hda_intel)
Force to snoop mode by module option

chipset global capabilities = 0x4401

Clearing TCSEL

SCH snoop: Enabled

codec_mask = 0x1

request_irq: irq 22
codec #0 probed OK

snd_hda_codec_device_init: entry

snd_hda_codec_device_new: entry

Initializing rbtree cache

ALC269VC: picked fixup Thinkpad T530 for PCI SSID 17aa:21f6

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply pincfg for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

autoconfig for ALC269VC: line_outs=1 (0x14/0x0/0x0/0x0/0x0) type:speaker

   speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)

   hp_outs=2 (0x15/0x1b/0x0/0x0/0x0)

   mono: mono_out=0x0

   inputs:

     Mic=0x18

     Dock Mic=0x19

     Internal Mic=0x12

realtek: Enable default setup for auto mode as fallback

==> lo_type=2, wired=1, mio=1, badness=0x112

multi_outs = 15/1b/0/0 : 2/3/0/0 (type HP)

  out path: depth=3 '02:0c:15'

  out path: depth=3 '03:0d:1b'

spk_outs = 14/0/0/0 : 2/0/0/0

  spk path: depth=3 '02:0c:14'

==> lo_type=2, wired=1, mio=0, badness=0x112

multi_outs = 15/1b/0/0 : 2/3/0/0 (type HP)

  out path: depth=3 '02:0c:15'

  out path: depth=3 '03:0d:1b'

spk_outs = 14/0/0/0 : 2/0/0/0

  spk path: depth=3 '02:0c:14'

==> lo_type=2, wired=0, mio=1, badness=0x112

multi_outs = 15/1b/0/0 : 2/3/0/0 (type HP)

  out path: depth=3 '02:0c:15'

  out path: depth=3 '03:0d:1b'

spk_outs = 14/0/0/0 : 2/0/0/0

  spk path: depth=3 '02:0c:14'

==> lo_type=2, wired=0, mio=0, badness=0x112

multi_outs = 15/1b/0/0 : 2/3/0/0 (type HP)

  out path: depth=3 '02:0c:15'

  out path: depth=3 '03:0d:1b'

spk_outs = 14/0/0/0 : 2/0/0/0

  spk path: depth=3 '02:0c:14'

==> lo_type=1, wired=1, mio=1, badness=0x20

multi_outs = 14/0/0/0 : 2/0/0/0 (type SP)

  out path: depth=3 '02:0c:14'

hp_outs = 15/1b/0/0 : 3/3/0/0

  hp  path: depth=3 '03:0d:15'

  hp  path: depth=3 '03:0d:1b'

==> lo_type=1, wired=1, mio=0, badness=0x20

multi_outs = 14/0/0/0 : 2/0/0/0 (type SP)

  out path: depth=3 '02:0c:14'

hp_outs = 15/1b/0/0 : 3/3/0/0

  hp  path: depth=3 '03:0d:15'

  hp  path: depth=3 '03:0d:1b'

==> lo_type=1, wired=0, mio=1, badness=0x20

multi_outs = 14/0/0/0 : 2/0/0/0 (type SP)

  out path: depth=3 '02:0c:14'

hp_outs = 15/1b/0/0 : 3/3/0/0

  hp  path: depth=3 '03:0d:15'

  hp  path: depth=3 '03:0d:1b'

==> lo_type=1, wired=0, mio=0, badness=0x20

multi_outs = 14/0/0/0 : 2/0/0/0 (type SP)

  out path: depth=3 '02:0c:14'

hp_outs = 15/1b/0/0 : 3/3/0/0

  hp  path: depth=3 '03:0d:15'

  hp  path: depth=3 '03:0d:1b'

==> restoring best_cfg

==> Best config: lo_type=1, wired=1, mio=1

multi_outs = 14/0/0/0 : 2/0/0/0 (type SP)

  out path: depth=3 '02:0c:14'

hp_outs = 15/1b/0/0 : 3/3/0/0

  hp  path: depth=3 '03:0d:15'

  hp  path: depth=3 '03:0d:1b'

loopback path: depth=2 '18:0b'

input path: depth=3 '18:23:08'

input path: depth=3 '18:22:09'

loopback path: depth=2 '19:0b'

input path: depth=3 '19:23:08'

input path: depth=3 '19:22:09'

input path: depth=3 '12:22:09'

Enable HP auto-muting on NID 0x15

Enable HP auto-muting on NID 0x1b

Enable auto-mic switch on NID 0x12/0x19/0x18

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

Syncing rbtree cache

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

ALC269VC: Apply fix-func for Thinkpad T530

pci_register_driver: query_device found d8 8086:1e20 class=40300 checking intel8x0
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking via82xx
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking emu10k1
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking cmipci
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking als4000
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking cs4281
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking cs46xx
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking cs5535audio
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking es1938
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking ens137x
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking ymfpci
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking es1968
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking Maestro3
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking ali5451
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking trident
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking nm256
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking fm801
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking atiixp
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking ca0106
pci_register_driver: query_device found d8 8086:1e20 class=40300 checking bt87x
pcm_info()
00-00: ALC269VC Analog : ALC269VC Analog
 : playback 1
 : capture 1


OSS32_Initialize: SUCCESS. Cards=1
Init=248/0 Between=0/0 Complete=0/0
U16: Uniaud16 v1.9.7
Midi Open 2 failed with error -19
Midi Open 3 failed with error -19
Couldn't find OPL3 device
Init=248/0 Between=10/0 Complete=0/0
hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

unknown ioctl = 0x401c4162

Init=248/0 Between=10/0 Complete=299/0
hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

Init=248/0 Between=10/0 Complete=369/0
hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

Init=248/0 Between=10/0 Complete=439/0
hda_codec_setup_stream: NID=0x2, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

hda_codec_setup_stream: NID=0x3, stream=0x5, channel=0, format=0x4011

Init=248/0 Between=10/0 Complete=509/0
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 12, 2022, 05:49:35 am
Heh Eugene!

My system are all Intel. The sound card is a Creative Recon.

Have you ever got Paul's UniAud32"s to work on your system?

In one of my systems I also have a Creative Recon3D / Z Series Card installed - however it only works on Windows, so I was curious as to whether sound worked for you - testing it with the latest drivers, I only get 5 sound volume sections in PMUnimix, and no sound.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on September 12, 2022, 06:09:38 am
Heh Eugene!

My system are all Intel. The sound card is a Creative Recon.

Have you ever got Paul's UniAud32"s to work on your system?

In one of my systems I also have a Creative Recon3D / Z Series Card installed - however it only works on Windows, so I was curious as to whether sound worked for you - testing it with the latest drivers, I only get 5 sound volume sections in PMUnimix, and no sound.

Best!

M

On some systems, in the past, we've seen that the sound is actually working but is extremely quiet.
Two changes I would recommend to the mmpm.ini until you get sound, to make sure it is just not really really low:

mastervolume=77   to   mastervolume=100
and
PARMSTRING=BALANCE=50,VOL=80,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER
to
PARMSTRING=BALANCE=50,VOL=100,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 12, 2022, 08:26:22 am
I could but not with that driver installed. If you want a baseline with say the default drive or one of the working drivers yes I can and wil do so tomorrow.
Yeah , sorry, I meant baseline with a recent working driver. I'll see if I can reproduce the trap on my T470S, and also workaround it.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 12, 2022, 11:10:09 am
Hey all,
Sorry about that, of course it didn't trap here in Virtualbox - unfortunately I didn't test it on bare metal. Are your systems AMD or Intel?

Well clearly I was either very tired or had drunk too much red wine yesterday. I'm embarassed to say that when I fired up the laptop after work today, my ArcaOS test partition was sat there with a Trap screen on it - so turns out I should have known it was going to fail for others :(

Sorry for the hassles, and I'll have another look at the code.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 12, 2022, 03:10:07 pm
No problem Paul.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 12, 2022, 04:51:12 pm
Heh Paul!

Hey all,
Sorry about that, of course it didn't trap here in Virtualbox - unfortunately I didn't test it on bare metal. Are your systems AMD or Intel?

Well clearly I was either very tired or had drunk too much red wine yesterday. I'm embarassed to say that when I fired up the laptop after work today, my ArcaOS test partition was sat there with a Trap screen on it - so turns out I should have known it was going to fail for others :(

Sorry for the hassles, and I'll have another look at the code.

Cheers,

Paul

What fun would it be if everything just worked, Sir? - no problem with the occasional BSOD here.

As to wines, we are quite partial to Savignon Blancs from the Marlborough region of New Zealand and of course many of the Australian Shiraz's. Oyster Bay and Starborough Sav Blancs, and Yellowtail and Rosemont Shirazs are frequent guests in our home.

Don't know what good stuff you save from export there, but what we get here from you guys and gals is pretty good, and reasonably priced.

M

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 12, 2022, 07:15:12 pm
Heh Andy


On some systems, in the past, we've seen that the sound is actually working but is extremely quiet.
Two changes I would recommend to the mmpm.ini until you get sound, to make sure it is just not really really low:

mastervolume=77   to   mastervolume=100
and
PARMSTRING=BALANCE=50,VOL=80,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER
to
PARMSTRING=BALANCE=50,VOL=100,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER

I didn't see anything about editing mmpm.ini in the past, but will try it and report back.

Thanks for the tip and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 12, 2022, 07:46:46 pm
Mark, Yes I have had the unaud32 driver working with my system. Even with the betas before the last officail release.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 12, 2022, 08:29:28 pm
Heh Eugene!

Mark, Yes I have had the unaud32 driver working with my system. Even with the betas before the last officail release.

Interesting.

In addition to the creative Recon3D, I also have a Sound Blaster X-Fi Titanium Series on one of the metal systems - works on Windows but that one shows only a Master Volume Sound Section under the latest UniAud32 releases.

So many chipsets so little time but Thanks, Sir and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 12, 2022, 08:45:35 pm
Hi Andy!


On some systems, in the past, we've seen that the sound is actually working but is extremely quiet.
Two changes I would recommend to the mmpm.ini until you get sound, to make sure it is just not really really low:

mastervolume=77   to   mastervolume=100
and
PARMSTRING=BALANCE=50,VOL=80,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER
to
PARMSTRING=BALANCE=50,VOL=100,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER

Using an INI editor, I was having a look at the MMPM.INI file in question and couldn't see the lines you mentioned (picture attached below)

Haven't done much hex editing in the past, so perhaps I'm misunderstanding the editing process - there seems to be only five lines present.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on September 12, 2022, 09:43:30 pm
Using an INI editor, I was having a look at the MMPM.INI file in question and couldn't see the lines you mentioned (picture attached below)

Shouldn't that be MMPM2.INI, which is plain text.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 13, 2022, 01:39:45 am
Heh Dave!

Using an INI editor, I was having a look at the MMPM.INI file in question and couldn't see the lines you mentioned (picture attached below)
[/quote>

Shouldn't that be MMPM2.INI, which is plain text.

Your right and Thanks!

Busy afternoon and limited time to have a look at the INI's - MMPM2.INI was on the next row down and didn't notice it - other than copying and testing Pauls drivers I don't spend much time in the MMOS2 directory.

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 13, 2022, 01:47:14 am
Heh Andy!

On some systems, in the past, we've seen that the sound is actually working but is extremely quiet.
Two changes I would recommend to the mmpm.ini until you get sound, to make sure it is just not really really low:

mastervolume=77   to   mastervolume=100
and
PARMSTRING=BALANCE=50,VOL=80,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER
to
PARMSTRING=BALANCE=50,VOL=100,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER

After Dave turned me around and I found text based MMPM2.INI (hiding on the next line of the directory), I made the mods you suggested - no difference, with all sound volumes set at max.

PMUnimix reports max volumes as 64, with the current drivers for most sections - the mic sections and line playback volume at 31.

M
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on September 13, 2022, 04:40:01 am
Heh Andy!

On some systems, in the past, we've seen that the sound is actually working but is extremely quiet.
Two changes I would recommend to the mmpm.ini until you get sound, to make sure it is just not really really low:

mastervolume=77   to   mastervolume=100
and
PARMSTRING=BALANCE=50,VOL=80,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER
to
PARMSTRING=BALANCE=50,VOL=100,INPUT=LINE,GAIN=70,OUTPUT=SPEAKER

After Dave turned me around and I found text based MMPM2.INI (hiding on the next line of the directory), I made the mods you suggested - no difference, with all sound volumes set at max.

PMUnimix reports max volumes as 64, with the current drivers for most sections - the mic sections and line playback volume at 31.

M
Sorry about the wrong filename, hopefully it will help later.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 13, 2022, 04:59:18 am
In addition to the creative Recon3D, I also have a Sound Blaster X-Fi Titanium Series on one of the metal systems - works on Windows but that one shows only a Master Volume Sound Section under the latest UniAud32 releases.

Testlogs from these non-working cards would be interesting.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 13, 2022, 07:38:45 pm
Heh Paul!

In addition to the creative Recon3D, I also have a Sound Blaster X-Fi Titanium Series on one of the metal systems - works on Windows but that one shows only a Master Volume Sound Section under the latest UniAud32 releases.

Testlogs from these non-working cards would be interesting.

Cheers,

Paul.

I've attached testlogs for both Hyperfire and Manticore II below.

Both were generated this morning my time using the Debug version of UniAud32 20220910b. In Config.sys the switch UniAud32 /a:2 was used to switch to the appropriate Sound Blaster card in each system, as described below.

For Manticore II, which contains the Sound Blaster Recon3D, I had an interesting experience.

Suddenly I was getting wonderful clear sound where none was present before. Looking at PMUniAud, only the Master Sound Volume section was present, maximum volume 99, and both system sounds and PM123 would play with no problems whatsoever.

The sound system the Recon3D is connected to is surround sound, but only the front speakers would generate sound, (as a note, the AD1988B chipset on that system is not connected to this surround speaker system - it is connected to sound through the monitor, which has a Bang & Olafsen sound system built in - has always worked fine with your modified drivers as reported)

And then, rebooting, no sound would be generated through the Recon3D ......... ?!?!?!?!

I traced the problem to the form of rebooting ........... if a warm reboot back to the partition I am testing, or to my other boot partition on Manticore II occurs, no sound is generated when I go back to the test partition. However, if I do a cold reboot to the test partition, sound is again generated (I tested it a number of times and it is reproducible).

And as for Hyperfire ..........

Booting to the Sound Blaster X-Fi Titanium using the /a:2 switch, no sound is generated at all - in PMUnimix it shows 6 sound sections jumbled together (I've attached a picture below of what sound sections are present). Andy's sound modifications are still present, as an aside.

The Hyperfire Sound Blaster is also connected to a separate surround sound system than the ALC892 chipset (that is connected to the same sound system in the monitor - I use an audio switch box to switch between the two, depending on which base metal system I am using at the time).

Have fun puzzling over the new testlogs - I'm sure a lot of red wine will be involved <grin>

Best Sir as always!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 14, 2022, 01:35:33 am
Hey Mark,

Interesting... I have a T60 here that is similar in the sense that after a cold boot, sound works fine - but warm boots have no audio.

I don't see a log for Hyperfire with the Sound Blaster X-Fi Titanium?

...and yes, red wine will likely be involved in fixing issues :)

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 14, 2022, 05:14:44 am
Hey Paul!

Hey Mark,

Interesting... I have a T60 here that is similar in the sense that after a cold boot, sound works fine - but warm boots have no audio.

I don't see a log for Hyperfire with the Sound Blaster X-Fi Titanium?

...and yes, red wine will likely be involved in fixing issues :)

Cheers,

Paul.

Interesting as I did attach it - Lets try this again (See below)

Just got in from our deck, at 21:00 on a beautiful 15 C pitch black fall night - watching all our solar lights lighting up - and Shiraz was involved there prior to retiring for the evening ............

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 14, 2022, 11:04:02 am
Interesting.... there's nothing obvious in that log as to why it's not working :(
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 14, 2022, 07:28:04 pm
Heh Paul!

Interesting.... there's nothing obvious in that log as to why it's not working :(

Had a look at the Hyperfire log myself today - the only thing that jumped out at me was near the end of the log:

HDA audio detected - doesn't support 5512 - 32000 Hz audio sample rates

Again, not knowing what is normal that could mean something or nothing at all Paul, but thought I'd point it out - that line appears a number of times at the end of the log.

M

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 14, 2022, 08:50:06 pm
Heh Paul!

Interesting.... there's nothing obvious in that log as to why it's not working :(

Had a wild non common sense idea today, which I didn't think would hurt anything and thought it might help.

As I said, on these two bare metal machines I have two partitions each - one working and one test.

On HYPERFIRE the ALC892 chipset on the working partition works in producing sound - I had a look and the UniAud version in use there is dated 2020-07-05, so packaged with ArcaOS 5.0.6 - I reverted the test partition to that level and it worked fine, albeit some clipping of system sound and on PM123. 8 sound sections reported by PMUnimix.

Attached below is a testlog of the working ALC892 version dated today - I have no idea if it would be of any use, but being a crazy person (you have to be working in the emergency business for 42 years), I thought I'd forward it to you anyway - if its of no use, for whatever reason nothing lost or gained. Just being thorough as a Beta Tester.

I also tried to do the same, using the SB X-Fi and the /a:2 switch - no sound was generated and PMUnimix generated neither an INI file nor a debug.txt, although, generating a Testlog, it seems to have been seen and reported there. I don't know when the switches were implemented in the Linux version you ported, so again nothing ventured nothing gained.

But thought I'd try it anyway since I had an hour or so, and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 15, 2022, 03:18:12 am
Hey Mark!

Funny - I'd had the thought to ask you if any previous versions of Uniaud produced sound with your ALC892 machine and kept forgetting to ask you.

It would be helpful (if you have time) to try and identify which build first broke support. Rather than trawling through this thread, I can send you a consolidated list of builds that are sitting on my webserver. That might help narrow down the specific change that broke things.

re: the creative card and the stock ArcaOS driver - it was relatively recent that I added in the patch_ca0110.c and patch_ca0132.c into the driver - so older drivers would have no chance of supporting this hardware.

Cheers,

Paul.

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 15, 2022, 04:20:08 am
Heh Paul!

Hey Mark!

Funny - I'd had the thought to ask you if any previous versions of Uniaud produced sound with your ALC892 machine and kept forgetting to ask you.

It would be helpful (if you have time) to try and identify which build first broke support. Rather than trawling through this thread, I can send you a consolidated list of builds that are sitting on my webserver. That might help narrow down the specific change that broke things.

re: the creative card and the stock ArcaOS driver - it was relatively recent that I added in the patch_ca0110.c and patch_ca0132.c into the driver - so older drivers would have no chance of supporting this hardware.

Cheers,

Paul.

The list would be useful - do you recall which ArcaOS build came with 5.0.7? - I seem to recall that the sound broke with that build (which is the reason I'm using 5.0.6's version, I think). I used the update process with a USB stick to install 5.0.7, so I don't think I have access to that build, but I can check my archives, if I know the date of the files.

If you have it and can provide that build of UniAud32, I can test it tomorrow,

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 15, 2022, 05:42:07 am
Hey Mark,

No idea which build came with 5.0.7, but your testlog shows:
1.0.24:::06::SVN609

This means the 'working' build is based on ALSA 1.0.24 - circa 2010!

Let me generate a list of builds tonight.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on September 15, 2022, 08:23:30 am
I believe 5.0.7 came with 3.01.02
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 15, 2022, 10:45:56 am
Let me generate a list of builds tonight.

Prepend these filenames with https://smedley.id.au/tmp/
Code: [Select]
uniaud32-linux-3.0-debug.zip
uniaud32-linux-3.0.zip
uniaud32-linux-3.10.108-20200804.zip
uniaud32-linux-3.10.108-20200807b.zip
uniaud32-linux-3.10.108-20200807c.zip
uniaud32-linux-3.10.108-20200807d.zip
uniaud32-linux-3.10.108-20200807.zip
uniaud32-linux-3.10.108-20200808.zip
uniaud32-linux-3.10.108.zip
uniaud32-linux-3.11.10-20200810.zip
uniaud32-linux-3.12.74-20200807.zip
uniaud32-linux-3.12.74-20200808.zip
uniaud32-linux-3.12.74-20200809.zip
uniaud32-linux-3.12.74-20200810.zip
uniaud32-linux-3.12.74.zip
uniaud32-linux-3.13.11.zip
uniaud32-linux-3.14.79-20200809.zip
uniaud32-linux-3.14.79-20200810.zip
uniaud32-linux-3.14.79.zip
uniaud32-linux-3.16.85-20200812.zip
uniaud32-linux-3.16.85-20200813.zip
uniaud32-linux-3.16.85-20200814.zip
uniaud32-linux-3.17.8-20200823.zip
uniaud32-linux-3.17.8-20200824.zip
uniaud32-linux-3.18.140-20200816.zip
uniaud32-linux-3.18.140-20200823-vortex.zip
uniaud32-linux-3.18.140-20201009.zip
uniaud32-linux-3.1.9.zip
uniaud32-linux-3.1.zip
uniaud32-linux-3.2.102-MikeK2.zip
uniaud32-linux-3.2.102-MikeK3.zip
uniaud32-linux-3.2.102-MikeK.zip
uniaud32-linux-3.2.102.zip
uniaud32-linux-3.2-debug.zip
uniaud32-linux-3.2.zip
uniaud32-linux-3.5.zip
uniaud32-linux-3.6.11.zip
uniaud32-linux-3.7.9.zip
uniaud32-linux-3.8.13.zip
uniaud32-linux-3.9.11.zip
uniaud32-linux-4.0.9-20200816.zip
uniaud32-linux-4.0.9-20201008.zip
uniaud32-linux-4.14.202-20201024.zip
uniaud32-linux-4.14.202-20201101b.zip
uniaud32-linux-4.14.202-20201101.zip
uniaud32-linux-4.14.202-20201110.zip
uniaud32-linux-4.14.202-20201114.zip
uniaud32-linux-4.14.202-20201121.zip
uniaud32-linux-4.14.202-20201227.zip
uniaud32-linux-4.14.202-src-20201029.zip
uniaud32-linux-4.14.202-src-20201106.zip
uniaud32-linux-4.14.202-src-20201109.zip
uniaud32-linux-4.14.202-src-20201116.zip
uniaud32-linux-4.14.202-trap-20201029.zip
uniaud32-linux-4.14.202-trap-20201116.zip
uniaud32-linux-4.14.202-trap.zip
uniaud32-linux-4.16.18-20201227.zip
uniaud32-linux-4.17.19-20201228.zip
uniaud32-linux-4.19.163-20201228.zip
uniaud32-linux-4.19.163-20201229b.zip
uniaud32-linux-4.19.163-20201229.zip
uniaud32-linux-4.19.163-20201230b.zip
uniaud32-linux-4.19.163-20201230c.zip
uniaud32-linux-4.19.163-20201230d.zip
uniaud32-linux-4.19.163-20201230.zip
uniaud32-linux-4.4.238-20201018.zip
uniaud32-linux-4.4.238-20201022.zip
uniaud32-linux-4.4.238-src.zip
uniaud32-linux-4.9.240-20201024.zip
uniaud32-linux-5.0.21-20201231.zip
uniaud32-linux-5.2.21-20201231.zip
uniaud32-linux-5.2.21-20210107.zip
uniaud32-linux-5.3.18-20210101.zip
uniaud32-linux-5.4.86-20210101b.zip
uniaud32-linux-5.4.86-20210101.zip
uniaud32-linux-5.4.86-20210102-debug.zip
uniaud32-linux-5.4.86-20210103-debug.zip
uniaud32-linux-5.5.19-20210104-debug.zip
uniaud32-linux-5.6.19-20210104-debug.zip
uniaud32-linux-5.7.19-20210104-debug.zip
uniaud32-linux-5.8.18-20210104-debug.zip
uniaud32-linux-5.9.16-20210104-debug.zip
uniaud32-linux-5.10.10-20210124.zip
uniaud32-linux-5.10.10-20210126.zip
uniaud32-linux-5.10.11-20210131.zip
uniaud32-linux-5.10.17-20210227.zip
uniaud32-linux-5.10.20-20210305.zip
uniaud32-linux-5.10.25-20210322.zip
uniaud32-linux-5.10.30-20210414.zip
uniaud32-linux-5.10.40-20210527.zip
uniaud32-linux-5.10.4-20210104-debug.zip
uniaud32-linux-5.10.4-20210105b.zip
uniaud32-linux-5.10.4-20210105c.zip
uniaud32-linux-5.10.4-20210105.zip
uniaud32-linux-5.10.45-20210619.zip
uniaud32-linux-5.10.50-202107179.zip
uniaud32-linux-5.10.50-20210720.zip
uniaud32-linux-5.10.50-20210721.zip
uniaud32-linux-5.10.5-20210108.zip
uniaud32-linux-5.10.7-20210115.zip
uniaud32-linux-5.10.7-20210116.zip
uniaud32-linux-5.10.7-20210117.zip
uniaud32-linux-5.11.22-20210814.zip
uniaud32-linux-5.1.21-20201231.zip
uniaud32-linux-5.12.19-20210814.zip
uniaud32-linux-5.13.10-20210815.zip
uniaud32-linux-5.14.7-20210926.zip
uniaud32-linux-5.14.7-20211028.zip
uniaud32-linux-5.15.20-20220807.zip
uniaud32-linux-5.15-20211107.zip
uniaud32-linux-5.15-20220805.zip
uniaud32-linux-5.15.59-20220805.zip
uniaud32-linux-5.15.59-20220806.zip
uniaud32-linux-5.15.59-20220807b.zip
uniaud32-linux-5.15.59-20220807.zip
uniaud32-linux-5.15.59-20220903.zip
uniaud32-linux-5.15.59-20220904b.zip
uniaud32-linux-5.15.59-20220904.zip
uniaud32-linux-5.15.65-20220905.zip
uniaud32-linux-5.15.65-20220906.zip
uniaud32-linux-5.15.65-20220907.zip
uniaud32-linux-5.15.65-20220908.zip
uniaud32-linux-5.15.65-20220910b.zip
uniaud32-linux-5.15.65-20220910.zip

Ideally, we want to know the 'newest' build where sound still works.

I'd start about halfway through the list and see if it works, if it doesn't - divide the remaining list in half and try that one, etc

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 15, 2022, 07:21:42 pm
Heh Paul!

Hey Mark,

No idea which build came with 5.0.7, but your testlog shows:
1.0.24:::06::SVN609

This means the 'working' build is based on ALSA 1.0.24 - circa 2010!

Let me generate a list of builds tonight.

Cheers,

Paul.

I'm a packrat when it comes to files and have local archives for a lot of things, including the various ArcaOS versions generated so far.

My recollection was indeed correct - there was an upgrade to the UniAud files in 5.0.7, and yes indeed I did revert to the 5.0.6 UniAud32, because no sound was produced.

Dave was mostly correct - the version included was 3.01.03 not .02:

According to the Testlog generated today:

Revision:        3.01.03
Date/Time:       23 Oct 2021 12:39:40   
Build Machine:   DAZAR1   
ASD Feature ID:  5.14.7
FixPak Version:  SVN707
File Version:    3.1.3
Description:     Universal Audio Driver for OS/2 (c) 2021 D Azarewicz

The UniAud32 files report as being dated 2021-12-7 in the 5.0.7 archive. And if I'm reading the above correctly the files were based on UniAud 5.14.7 (Don't know if David is also using the Linux sources, but I would assume so).

Testlog from today attached below.

We're going out for dinner today, so I don't know whether I will get around to too much testing, Paul - Some of your files precede that version, so perhaps I should test them.

I also have a number of UniAud versions that were generated as official upgrades as Warpin files, if I need to go further back - the problem there is I don't know how you deconstruct a .WPI file to extract the files contain therein, for testing - any suggestions if I need to go that far?

M
Title: Re: Updated Uniaud32 build
Post by: Pete on September 15, 2022, 07:33:47 pm
Hi Mark

A useful tool for investigating and extracting files from WPI packages is WPIView which you should find here ftp://ftp.netlabs.org/pub/vxapps/wpiview/wpiview110.zip


Regards

Pete
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 15, 2022, 08:41:41 pm
Heh Paul!



Ideally, we want to know the 'newest' build where sound still works.

I'd start about halfway through the list and see if it works, if it doesn't - divide the remaining list in half and try that one, etc

Cheers,

Paul

Well that was quick, Gents

As Davids 3.01.03 build didn't work, based on v5.14.7 (if I'm reading that right), I decided to go right back to the first files in your directory, Paul, also based on v5 - I downloaded your 3.0 and 3.1 debugs and releases, and installed each one - unless there is some other variable I'm missing, no sound seems to be generated from any of the 5.x series on the ALC892 chipset.

Still shows the 9 PMUnimix volume sections that the ALSA 1.0.24 based v2.01.02 shows, but there is no sound present in any of the four tests I did this morning.

I also checked what upgrades ArcaOS put out, in WarpIn format, in the interim - they use the file version described in Testlog as the identifier, and the latest versions present in WarpIn are 2.2.6 (the 2.01.02 version from 5.0.6) and 3.1.3 (version 3.01.03).

So by process of elimination, anything that remains must be obvious, to misquote the great sleuth - the ALSA 1.0.24 build must be the last version that worked - hence the testlog I provided from yesterday.

And brooding on it a bit last night over a glass of Hennessy Cognac on a beautiful mid fall summer night on the deck, I suspect its something extremely small, given everything is working, except the sound itself - I guess my question would be was there a wholesale rewrite of UniAud, as things jumped from Alsa v1 to Linux v5?

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 15, 2022, 08:45:41 pm
Heh Pete!

How're things going?

Hi Mark

A useful tool for investigating and extracting files from WPI packages is WPIView which you should find here ftp://ftp.netlabs.org/pub/vxapps/wpiview/wpiview110.zip


Regards

Pete

Actually I do have it, but never have used it - so many files so little time <sic>

Now I don't think I'll need it, but will peg it to Tab LaunchPad - the update files are the same as what I installed - one works, albeit with sound clipping, while the other does not work at all.

Have a fine day Sir and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 15, 2022, 11:23:41 pm
Hey Mark,

Well that was quick, Gents

As Davids 3.01.03 build didn't work, based on v5.14.7 (if I'm reading that right), I decided to go right back to the first files in your directory, Paul, also based on v5 - I downloaded your 3.0 and 3.1 debugs and releases, and installed each one - unless there is some other variable I'm missing, no sound seems to be generated from any of the 5.x series on the ALC892 chipset.

Brief history lesson if I may....

In the early days of ALSA, it was maintained as a separate source repository outside of the linux kernel. This is back in the days when Uniaud was first created. ALSA 1.0.25 was the final release as a standalone project. The Uniaud in ArcaOS 5.0.24 (the v2 series) was based on ALSA code.

Around the time of the release of the linux kernel 3.0; the ALSA code was incorporated into the kernel, and wasn't available as a separate download.

The list of builds I posted (including the 3.0 release) are based on the corresponding sound code from the linux kernel. So (for example) the uniaud32-linux-3.0.zip was based on the 'linux 3.0' kernel.

So if this is the build that broke things for you, something changed between the ALSA 1.0.24 code; and the sound code from the 3.0 kernel release. This certainly should help narrow things down :)

Edit: the 3.0 code didn't make it in to SVN, but the breaking change for Mark's ALC892 should be somewhere in https://trac.netlabs.org/uniaud/changeset/612 :)

Edit2: @Mark - can you please generate a log with the debug driver from https://smedley.id.au/tmp/uniaud32-linux-3.2.102.zip ? I don't expect it to produce audio - but it will give me a log that corresponds with https://trac.netlabs.org/uniaud/changeset/612

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 15, 2022, 11:25:32 pm
Hey Mark,
The UniAud32 files report as being dated 2021-12-7 in the 5.0.7 archive. And if I'm reading the above correctly the files were based on UniAud 5.14.7 (Don't know if David is also using the Linux sources, but I would assume so).
More specifically, the build in AOS 5.0.7 is based on the sound code from linux 5.14.7; and was taken from my 'uniaud32-next' branch at netlabs.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 16, 2022, 08:04:54 pm
Hey Mark,

Brief history lesson if I may....

In the early days of ALSA, it was maintained as a separate source repository outside of the linux kernel. This is back in the days when Uniaud was first created. ALSA 1.0.25 was the final release as a standalone project. The Uniaud in ArcaOS 5.0.24 (the v2 series) was based on ALSA code.

Around the time of the release of the linux kernel 3.0; the ALSA code was incorporated into the kernel, and wasn't available as a separate download.

The list of builds I posted (including the 3.0 release) are based on the corresponding sound code from the linux kernel. So (for example) the uniaud32-linux-3.0.zip was based on the 'linux 3.0' kernel.

So if this is the build that broke things for you, something changed between the ALSA 1.0.24 code; and the sound code from the 3.0 kernel release. This certainly should help narrow things down :)

Edit: the 3.0 code didn't make it in to SVN, but the breaking change for Mark's ALC892 should be somewhere in https://trac.netlabs.org/uniaud/changeset/612 :)

Edit2: @Mark - can you please generate a log with the debug driver from https://smedley.id.au/tmp/uniaud32-linux-3.2.102.zip ? I don't expect it to produce audio - but it will give me a log that corresponds with https://trac.netlabs.org/uniaud/changeset/612

Cheers,

Paul

Thanks for the history lesson - it gives me a better perspective as to how this all evolved, framework wise. When your not part of the development, everything in the black box is a mystery to the user.

As a troubleshooter, its why I've found Beta Testing a fascinating experience. Paul - I've learned a lot about whats under the hood in this OS, and how its all interrelated. And tracing and correcting problems is what I've spent a lifetime doing

It seems a lot of things got fed into the maul of the linux kernel - the other project I've been working on here, on hiatus at the moment (joystick/gamepad support) also seems to be linked directly to the linux kernel in many ways.

As to generating a testlog .............

Downloading and trying 3.2.102 debug - first of all it was a memory hog - had to disable a couple of optional device drivers in CONFIG.SYS to proceed, only to trap 000e in the kernel. Loading the release version got an internal processing error in the current kernel installed - 14.203 SMP.

I noted that you also had a version labelled 3.2.102 MikeK (three versions in fact: MikeK - MikeK3), so I downloaded K3 as a further test and installed it - I assume that was also a debug version as I got the same trap 000e.

So I suspect there are memory usage problems involved with that version of UniAud and my base metal system HYPERFIRE - also don't know what kernel you were testing it on there - I'm loath to revert kernel wise, if that may be needed, even on a test partition, in case thing go wacky.

Do you think installing 3.2 debug would tell you anything? I don't know if that will even load, and I assume that that version preceded 3.2.102 so you may not have committed the changes in the change log. Your next version after that is 3.5.

Best!

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 17, 2022, 11:31:38 am
Hey Mark,
As to generating a testlog .............

Downloading and trying 3.2.102 debug - first of all it was a memory hog - had to disable a couple of optional device drivers in CONFIG.SYS to proceed, only to trap 000e in the kernel. Loading the release version got an internal processing error in the current kernel installed - 14.203 SMP.

I noted that you also had a version labelled 3.2.102 MikeK (three versions in fact: MikeK - MikeK3), so I downloaded K3 as a further test and installed it - I assume that was also a debug version as I got the same trap 000e.

So I suspect there are memory usage problems involved with that version of UniAud and my base metal system HYPERFIRE - also don't know what kernel you were testing it on there - I'm loath to revert kernel wise, if that may be needed, even on a test partition, in case thing go wacky.

Do you think installing 3.2 debug would tell you anything? I don't know if that will even load, and I assume that that version preceded 3.2.102 so you may not have committed the changes in the change log. Your next version after that is 3.5.

Interesting... it's been way too long, but I don't recall others reporting traps in 3.2.102 - otherwise, I doubt I would have checked the code in to SVN... anyways...

Any working version around 3.2 would be useful...

even:
uniaud32-linux-3.0-debug.zip
uniaud32-linux-3.1.zip
uniaud32-linux-3.1.9.zip
uniaud32-linux-3.2-debug.zip

The original linux source is available for these, so they're still possibly helpful in identifying an additional (or mission) log entry which might give clues...

Edit: I also wonder if removing the PCIe sound cards makes any difference...

Cheers,

Paul

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 18, 2022, 12:10:46 am
Minor code cleanups in https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220918.zip
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 18, 2022, 01:43:04 am
 Thanks Paul, it's working well here...

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7220000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 18, 2022, 02:38:57 am
Minor code cleanups in https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220918.zip
Hi Paul

I tested it (Release version) with ArcaOS 5.0.7 VM guest on VBox 6.1.38 (Windows 10 21H2 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Works, Audio control working, no traps.

Regards
Title: Re: Updated Uniaud32 build
Post by: OS4User on September 18, 2022, 07:36:37 am
Minor code cleanups in https://smedley.id.au/tmp/uniaud32-linux-5.15.65-20220918.zip

works ok for me. (EMU10k2)   tnx for your efforts
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 18, 2022, 08:42:17 am
Thanks guys. I spent some time today creating an uniaud32-exp branch, so that I can maintain the Long Term Support 5.15.x codebase in uniaud32-next whilst seeing if I can get the current kernel (5.19.x) working. I worked through the code (5.16.x, 5.17.x, etc) and have the current released kernel (5.19.9) compiling.

AC97 is working; HDA is not detecting hardware for some reason since the update to 5.16.x

More as time permits....
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 18, 2022, 09:12:26 am
Well that was quicker than I thought - I only tested with AC97 and HDA on virtualbox so far.... https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220918.zip

Code is in: http://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-exp

My intent is that once the next LTS kernel is decided (likely 6.0 or maybe 6.1); I'll merge the uniaud32-exp code back into uniaud32-next.

A typical linux kernel release is only supported for about 3 months, by which time another kernel has been released. Whatever kernel is current at the end of a calendar year, gets supported for at least a couple of years.

I'll try test on my T470s a bit later today.
edit: sound works on my T470s
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 18, 2022, 04:09:36 pm
Paul,

 This one seems to work well too - thanks again!

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 18, 2022, 08:16:43 pm
Paul, working well here as well.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 18, 2022, 11:14:06 pm
Heh Paul!


Edit2: @Mark - can you please generate a log with the debug driver from https://smedley.id.au/tmp/uniaud32-linux-3.2.102.zip ? I don't expect it to produce audio - but it will give me a log that corresponds with https://trac.netlabs.org/uniaud/changeset/612

Cheers,

Paul

While you've been playing around with new exp's I haven't been idle - just wanted to make sure what I'm reporting is reproducible.

Thanks for access to your TRAC archives.

I've traced the problem with the ALC892 chipset to something between build 3.8.13 (File version 2.2.5) and build 3.9.11 (File Version 2.2.5). It seems to have broke then. I've attached the testlogs below for both, using the debug versions of both.

I'm now using the release version of 3.8.13 successfully for both system sounds and PM123 - but I don't know what different changes were committed to TRAC at the time.

Hopefully that helps you out and Best!

M

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 19, 2022, 01:01:14 am
I've traced the problem with the ALC892 chipset to something between build 3.8.13 (File version 2.2.5) and build 3.9.11 (File Version 2.2.5). It seems to have broke then. I've attached the testlogs below for both, using the debug versions of both.

Thanks - this should be useful. I'll try and investigate this week. It's a short week here - Thursday is a Public Holiday for QEII, and I'm taking a leave day on Friday to turn it into a long weekend.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 19, 2022, 02:18:26 am
Hey Paul!

I've traced the problem with the ALC892 chipset to something between build 3.8.13 (File version 2.2.5) and build 3.9.11 (File Version 2.2.5). It seems to have broke then. I've attached the testlogs below for both, using the debug versions of both.

Thanks - this should be useful. I'll try and investigate this week. It's a short week here - Thursday is a Public Holiday for QEII, and I'm taking a leave day on Friday to turn it into a long weekend.

Cheers,

Paul.

So you actually take time off?!?!?!?!

WOW, I thought you were active 24 hours a day <sic>

The Funeral will be broadcast here starting 0300 MST, Monday - so will be in bed early - whether you love or hate the monarchy this is indeed the end of an era, and needs to be acknowledged, in this Year of Strange .........

In Alberta it will be a day of mourning but not a holiday - our former Premier (who still thinks he is) is in the UK at the moment saying he's there for Alberta ............ !

My wife is 75, I'm 63 ............ and we've never thought of anyone other than Elizabeth being there, who was actually was older than us all ........... yet another influence that is no longer stable in our lives.

Will raise a glass to the Queen tomorrow ........... I just hope no one tries to disrupt the proceedings with their particular grievance and Best, Sir!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 19, 2022, 03:28:56 am
I've traced the problem with the ALC892 chipset to something between build 3.8.13 (File version 2.2.5) and build 3.9.11 (File Version 2.2.5). It seems to have broke then. I've attached the testlogs below for both, using the debug versions of both.

Thanks - this should be useful. I'll try and investigate this week. It's a short week here - Thursday is a Public Holiday for QEII, and I'm taking a leave day on Friday to turn it into a long weekend.

Cheers,

Paul.

And BTW - 5.19.9 works fine on Manticore II with the AD1988B Chipset and Sound Blaster Recon 3D, only if a cold reboot is involved.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 19, 2022, 11:34:37 am
Hey Mark,

Heh Paul!


Edit2: @Mark - can you please generate a log with the debug driver from https://smedley.id.au/tmp/uniaud32-linux-3.2.102.zip ? I don't expect it to produce audio - but it will give me a log that corresponds with https://trac.netlabs.org/uniaud/changeset/612

Cheers,

Paul

While you've been playing around with new exp's I haven't been idle - just wanted to make sure what I'm reporting is reproducible.

Thanks for access to your TRAC archives.

I've traced the problem with the ALC892 chipset to something between build 3.8.13 (File version 2.2.5) and build 3.9.11 (File Version 2.2.5). It seems to have broke then. I've attached the testlogs below for both, using the debug versions of both.

I'm now using the release version of 3.8.13 successfully for both system sounds and PM123 - but I don't know what different changes were committed to TRAC at the time.

Hopefully that helps you out and Best!

This is VERY helpful! I dug out an old drive with an older copy of my eCS-udrive.vdi and found the source folders for:
uniaud32-linux-3.8.13
uniaud32-linux-3.9.11

A diff of the changes is at https://smedley.id.au/tmp/uniaud32-3.8.13-3.9.11.diff

Unfortunately, this is a large-ish diff - I'll have a look for anything obvious when I get a moment - hopefully it will be something obvious...

Edit: 1st attempt - hda_intel.c from 3.9.11 backlevelled to the 3.8.13 version - https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220919.zip
Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 19, 2022, 11:55:15 am
Hey Mark,

I've traced the problem with the ALC892 chipset to something between build 3.8.13 (File version 2.2.5) and build 3.9.11 (File Version 2.2.5). It seems to have broke then. I've attached the testlogs below for both, using the debug versions of both.

Thanks - this should be useful. I'll try and investigate this week. It's a short week here - Thursday is a Public Holiday for QEII, and I'm taking a leave day on Friday to turn it into a long weekend.

Cheers,

Paul.

So you actually take time off?!?!?!?!

WOW, I thought you were active 24 hours a day <sic>

The Funeral will be broadcast here starting 0300 MST, Monday - so will be in bed early - whether you love or hate the monarchy this is indeed the end of an era, and needs to be acknowledged, in this Year of Strange .........

In Alberta it will be a day of mourning but not a holiday - our former Premier (who still thinks he is) is in the UK at the moment saying he's there for Alberta ............ !

My wife is 75, I'm 63 ............ and we've never thought of anyone other than Elizabeth being there, who was actually was older than us all ........... yet another influence that is no longer stable in our lives.

Will raise a glass to the Queen tomorrow ........... I just hope no one tries to disrupt the proceedings with their particular grievance and Best, Sir!

I most definitely take time off :) We don't have skin kids - and just the one dog these days, so my method of relaxation in the evening is to look at code. The funeral is on at a much more civilised time here - started at around 7pm :) But we will record it and watch at our leisure - so we can fast forward the boring bits :) Whilst I was born in the UK ( we emigrated in 1977 when I was two) I wouldn't say I'm a monarchist - but at the same time - why change a system of government that isn't broken?

Anyway I digress - alcohol free day for us today - will have a glass with dinner tomorrow :P

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on September 19, 2022, 06:02:28 pm
Hi Paul,

...This is VERY helpful! I dug out an old drive with an older copy of my eCS-udrive.vdi and found the source folders for:
uniaud32-linux-3.8.13
uniaud32-linux-3.9.11

A diff of the changes is at https://smedley.id.au/tmp/uniaud32-3.8.13-3.9.11.diff

Unfortunately, this is a large-ish diff - I'll have a look for anything obvious when I get a moment - hopefully it will be something obvious...

Edit: 1st attempt - hda_intel.c from 3.9.11 backlevelled to the 3.8.13 version - https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220919.zip

Not sure if what I'm seeing here may be in some way related to what Mark is reporting. However, the uniaud32-linux-5.15.65-20220910b.zip drop results in a HARD trap while doing MPLAYER playback. So basically while watching any video and attempting to move back or forward causes the sync-up function to eventually freeze and the whole system is non-responsive.

I've attached my current test-matrix, bottom line: I always go back to the 3.16.85 release which has been rock-solid here (this is what's prompting me to ask whether this is similar situation to what Mark is seeing).

I will try this test version next. Thanks, as always!
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 19, 2022, 07:51:25 pm
Heh Dariusz

Long time no talk. Hows it going?

Hi Paul,

...This is VERY helpful! I dug out an old drive with an older copy of my eCS-udrive.vdi and found the source folders for:
uniaud32-linux-3.8.13
uniaud32-linux-3.9.11

A diff of the changes is at https://smedley.id.au/tmp/uniaud32-3.8.13-3.9.11.diff

Unfortunately, this is a large-ish diff - I'll have a look for anything obvious when I get a moment - hopefully it will be something obvious...

Edit: 1st attempt - hda_intel.c from 3.9.11 backlevelled to the 3.8.13 version - https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220919.zip

Not sure if what I'm seeing here may be in some way related to what Mark is reporting. However, the uniaud32-linux-5.15.65-20220910b.zip drop results in a HARD trap while doing MPLAYER playback. So basically while watching any video and attempting to move back or forward causes the sync-up function to eventually freeze and the whole system is non-responsive.

I've attached my current test-matrix, bottom line: I always go back to the 3.16.85 release which has been rock-solid here (this is what's prompting me to ask whether this is similar situation to what Mark is seeing).

I will try this test version next. Thanks, as always!

Actually, the problem we're trying to solve is with the RealTek ALC892 chipset in one of my base metal systems - something seems to have broken way back when, with Paul's drivers - between the official release of ArcaOS 5.0.6 (based on old ALSA code) and 5.0.7 (where the UniAud 32 is based on Linux 5.14.7 - release v3.01.02)

PMUnimix shows 9 volume sections, but no sound is generated - trying to correct that problem at the moment as it appears in all versions of Pauls 5.x series UniAud drivers since.

Those traps I saw were related solely to the old 3.2.102 release Paul wanted me to test - I since have narrowed down that problem to later specific Linux Kernal 3.x versions, so not affecting Paul's Current releases.

Testing a reversion that he just did to 3.9.11, to see if it solves the problem.

Best and have a great week!

M

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 19, 2022, 08:18:57 pm
Heh Paul!


This is VERY helpful! I dug out an old drive with an older copy of my eCS-udrive.vdi and found the source folders for:
uniaud32-linux-3.8.13
uniaud32-linux-3.9.11

A diff of the changes is at https://smedley.id.au/tmp/uniaud32-3.8.13-3.9.11.diff

Unfortunately, this is a large-ish diff - I'll have a look for anything obvious when I get a moment - hopefully it will be something obvious...

Edit: 1st attempt - hda_intel.c from 3.9.11 backlevelled to the 3.8.13 version - https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220919.zip
Cheers,

Paul

Tried your edit of 3.9.11, and no problems loading - but no sound generated.

New Testlog attached and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 19, 2022, 08:44:17 pm
Hey Mark,

I most definitely take time off :) We don't have skin kids - and just the one dog these days, so my method of relaxation in the evening is to look at code. The funeral is on at a much more civilised time here - started at around 7pm :) But we will record it and watch at our leisure - so we can fast forward the boring bits :) Whilst I was born in the UK ( we emigrated in 1977 when I was two) I wouldn't say I'm a monarchist - but at the same time - why change a system of government that isn't broken?

Anyway I digress - alcohol free day for us today - will have a glass with dinner tomorrow :P

Cheers,

Paul

Actually we found it quite an interesting show of pagentry and tradition.

I was particularly impressed as to those 8 strapping lads from the Royal Grenadiers hoisting that coffin 5 times, sometimes standing in place for minutes at a time, or under slow march for minutes - it was mentioning here that the Queens Coffin is made of oak ......... and is totally lead lined, as she will be in one of the Vaults under St. Georges Cathedral, in I assume a niche beside Philip.

Moving that coffin with all that weight on level ground is bad enough, manhandling wise - moving it up three steep flights of stairs on your shoulders at St. Georges takes a LOT of ultra coordinated effort and immense strength.

Also interesting as to how they fastened the Crown, Sceptre and Orb to the top - there must have been a plate fastened under the Royal Ensign, as at least two of the objects had pipes or spikes fastened to the bottom, which fitted into niches - the Sceptre was held in place by a bracket. I doubt they wanted to drill holes into the top of the coffin ..........

And finally, I didn't envy those soldiers who were required, by tradition to carry their weapons backwards as a sign of respect - to first of all have to slow march an hour and twenty minutes, then tie yourself up in knots at the same time, is something to behold - I suspect a lot of them will be glad to return to operational duty after today .......... !

But we've never seen anything like that in most of our lifetimes, Sir - Anne was born the year Elizabeth and Philip married ............ !

So a stately village style funeral, people behaving themselves in silence most of the time (except for the occasional cheer and throwing flowers) and a fitting sendoff to a long life well lived - and now with yet another sense of permanency gone ........... what's next?

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on September 19, 2022, 08:49:35 pm
Hi Mark,

Long time no talk. Hows it going?

Things are alright here, busy and slowing down at the same time LOL. Both kids are away for University studies, so home is a little "slower"...but even with fewer people around it just seems like we have more stuff to chase after...ugh??? LOL

...Actually, the problem we're trying to solve is with the RealTek ALC892 chipset in one of my base metal systems - something seems to have broken way back when, with Paul's drivers - between the official release of ArcaOS 5.0.6 (based on old ALSA code) and 5.0.7 (where the UniAud 32 is based on Linux 5.14.7 - release v3.01.02)...

Yup, understood that and given that my chipset here is the ATI SB using ALC889 I figured there may be an odd-chance that the two are linked:

Code: [Select]
HARDWARE INFO ('unimix -card')
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 2.02.05
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: SB
  driver: HDA-Intel
  name: HDA ATI SB
  longname: HDA ATI SB at 0xfe3f8000 irq 16
  mixer: Realtek ALC889
  componenets: HDA:10ec0889,14627640,00100004
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 20, 2022, 12:17:56 am
Hey Mark,

This is VERY helpful! I dug out an old drive with an older copy of my eCS-udrive.vdi and found the source folders for:
uniaud32-linux-3.8.13
uniaud32-linux-3.9.11

A diff of the changes is at https://smedley.id.au/tmp/uniaud32-3.8.13-3.9.11.diff

Unfortunately, this is a large-ish diff - I'll have a look for anything obvious when I get a moment - hopefully it will be something obvious...

Edit: 1st attempt - hda_intel.c from 3.9.11 backlevelled to the 3.8.13 version - https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220919.zip
Cheers,

Tried your edit of 3.9.11, and no problems loading - but no sound generated.


:) Not surprised - swapping out that source file only was a bit of a WAG... Tonight I'll aim to build the following:
1) ALSA 3.8.13 core code; with hda code from 3.9.11
2) ALSA 3.9.11 core code; with hda code from 3.8.13

One of those should still produce sound - which should quickly split whether the issue is in the ALSA 3.9.11 core code; or the hda code.

More tonight...

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 20, 2022, 10:44:47 am
Me again,


This is VERY helpful! I dug out an old drive with an older copy of my eCS-udrive.vdi and found the source folders for:
uniaud32-linux-3.8.13
uniaud32-linux-3.9.11

A diff of the changes is at https://smedley.id.au/tmp/uniaud32-3.8.13-3.9.11.diff

Unfortunately, this is a large-ish diff - I'll have a look for anything obvious when I get a moment - hopefully it will be something obvious...

Edit: 1st attempt - hda_intel.c from 3.9.11 backlevelled to the 3.8.13 version - https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220919.zip
Cheers,

Tried your edit of 3.9.11, and no problems loading - but no sound generated.


:) Not surprised - swapping out that source file only was a bit of a WAG... Tonight I'll aim to build the following:
1) ALSA 3.8.13 core code; with hda code from 3.9.11
2) ALSA 3.9.11 core code; with hda code from 3.8.13

One of those should still produce sound - which should quickly split whether the issue is in the ALSA 3.9.11 core code; or the hda code.

OK....
1) https://smedley.id.au/tmp/uniaud32-linux-3.8.13-20220920.zip <== sound 'core' from 3.8.13; pci/hda code from 3.9.11
2) https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220920.zip <== sound 'core' from 3.9.11; pci/hda code from 3.8.13

My 'gut' feel is that the problem lies in the hda code, and that 2) above will continue to produce sound.... but we'll see....

I do see one interesting thing in the 3.9.13 log you posted....
Code: [Select]
hda-codec: enabling ADC switching
Not sure if it's part of the problem or not yet though...

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 20, 2022, 05:35:03 pm
Neither of the two latest drivers are working on my system. No sound at all in HDA.
Title: Re: Updated Uniaud32 build
Post by: mauro on September 20, 2022, 07:37:56 pm
Hi Mauro

I spruced it up a bit today and have attached it below - hope you will find it of some use.

If anyone has any comments related to the guide, let me know and I will update it.

Best!

Mark

Thank to you as well Mark,  very useful job
[/quote]

Ok Mark, applied suggested editing on config.sys and also downloaded-executed UniaudMix.exe v 0.82 ; result is "no Uniaud device found"  This should have something to do with no uniaud16.sys driver available in the MMOS2 folder.  At this point it is clear I'm missing something very trivial, I feel have done all suggested steps:  downloaded-executed the Uniaud installer, replaced the existing files in MMOS2 with the Paul's updated ones, edited config.sys as suggested in your Testing Guide, but at boot the warning complains that uniaud16.sys is not found as per config.sys call (obvious, I know it is missing).  And of course no sound in the system.
Hope to learn next step from some of you.
Thank you meantime.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 20, 2022, 08:42:07 pm
Heh Paul!


OK....
1) https://smedley.id.au/tmp/uniaud32-linux-3.8.13-20220920.zip <== sound 'core' from 3.8.13; pci/hda code from 3.9.11
2) https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220920.zip <== sound 'core' from 3.9.11; pci/hda code from 3.8.13

My 'gut' feel is that the problem lies in the hda code, and that 2) above will continue to produce sound.... but we'll see....

I do see one interesting thing in the 3.9.13 log you posted....
Code: [Select]
hda-codec: enabling ADC switching
Not sure if it's part of the problem or not yet though...

Cheers,

Paul

Well your gut feeling was correct, Sir:

Loading UniAud32.sys 3.9.11 Debug 20220920:

Lovely clear sound produced in both system sounds and PM123 (Played The Poppy Family - Where Evil Grows ......... seemed appropriate  8).)

Loading UniAud32.sys 3.8.13 Debug 20220920:

No sound produced

So it seems to be narrowed down to the pci/hda code ........... !

Testlogs for both attached below.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 20, 2022, 08:59:21 pm

Things are alright here, busy and slowing down at the same time LOL. Both kids are away for University studies, so home is a little "slower"...but even with fewer people around it just seems like we have more stuff to chase after...ugh??? LOL


Good Job, Sir - Busy and slowing down is exactly how I would describe this Year of Strange - antithetical to the extreme.


Yup, understood that and given that my chipset here is the ATI SB using ALC889 I figured there may be an odd-chance that the two are linked:


Has it ever worked before? And if so, what version of UniAud32 was it?

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 20, 2022, 09:16:25 pm
Heh Mauro!


Ok Mark, applied suggested editing on config.sys and also downloaded-executed UniaudMix.exe v 0.82 ; result is "no Uniaud device found"  This should have something to do with no uniaud16.sys driver available in the MMOS2 folder.  At this point it is clear I'm missing something very trivial, I feel have done all suggested steps:  downloaded-executed the Uniaud installer, replaced the existing files in MMOS2 with the Paul's updated ones, edited config.sys as suggested in your Testing Guide, but at boot the warning complains that uniaud16.sys is not found as per config.sys call (obvious, I know it is missing).  And of course no sound in the system.
Hope to learn next step from some of you.
Thank you meantime.


Paul would know better than I would, but I think having UniAud16 in X:/MMOS2 is still required, in all cases. I have it still in all the partitions I've tested - whatever was installed originally is fine, and to my knowledge never has to change or be removed to use the UniAud32 drivers:

I also always leave the UniAud16.sys statement in place as well, under the UniAud32.sys statement (ie. not removed or REM'd out):

DEVICE=X:\MMOS2\UNIAUD16.SYS /C

The only item really required is changing UniAud32 to the version being tested, in X;\MMOS2 - I usually install all the files Paul provides but don't know if that is really required - testing the legacy UniAud32 files Paul provided at the moment, to solve the ALC892 problem, I have only installed the appropriate UniAud32.sys, and it seems to work - for the newest versions I don't know if .sym and .map are also needed, but never has hurt yet to include them.

I will say that when I've been generating testlogs I always do it with the debug version, as opposed to release.

So I suspect not having UniAud16.sys (and possibly UniAud16.sym as well) or removing the  UniAud16.sys statement in Config.sys, or both, is probably your problem.

Let us know - yet another item possible for v1.2 of the guide, along with Andy Willis's suggestion related to volume.

BTW include the sound chipset if you could as well (testlog would probably show that if you don't know) - I'm currently going through the posts to see if I can compile a list of chipsets that have been tested with the new UniAud drivers, and work when I get some free moments - might be useful to include in the guide in future, or to have Martin post as a Wiki.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 21, 2022, 12:59:58 am
Hi Eugene,
Neither of the two latest drivers are working on my system. No sound at all in HDA.

If you're referring to the 0920 builds - I should probably have made it clearer that these are directed towards Mark - and trying to diagnose what change in the linux source code broke sound on his ALC892. They are based on VERY old linux kernels - probably pre-dating the release of the chipsets in your hardware.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 21, 2022, 01:02:32 am
Hey Mark,

OK....
1) https://smedley.id.au/tmp/uniaud32-linux-3.8.13-20220920.zip <== sound 'core' from 3.8.13; pci/hda code from 3.9.11
2) https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220920.zip <== sound 'core' from 3.9.11; pci/hda code from 3.8.13

My 'gut' feel is that the problem lies in the hda code, and that 2) above will continue to produce sound.... but we'll see....

I do see one interesting thing in the 3.9.13 log you posted....
Code: [Select]
hda-codec: enabling ADC switching
Not sure if it's part of the problem or not yet though...


Well your gut feeling was correct, Sir:

Loading UniAud32.sys 3.9.11 Debug 20220920:

Lovely clear sound produced in both system sounds and PM123 (Played The Poppy Family - Where Evil Grows ......... seemed appropriate  8).)

Loading UniAud32.sys 3.8.13 Debug 20220920:

No sound produced

So it seems to be narrowed down to the pci/hda code ........... !

Testlogs for both attached below.

Thanks - tonight, I'll aim to produce several builds, slowly replacing the 3.8.x pci/hda/* code with the 3.9.x pci/hda/* code - ideally one source file at a time.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 21, 2022, 11:50:05 am
Hi Mark,
Thanks - tonight, I'll aim to produce several builds, slowly replacing the 3.8.x pci/hda/* code with the 3.9.x pci/hda/* code - ideally one source file at a time.

Note: these are specifically directed at trying to understand what broken with his hardware between the sound code in linux 3.8, and linux 3.9.

Code: [Select]
hda_intel.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921.zip

hda_auto_parser.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921b.zip
hda_auto_parser.h
hda_codec.c
hda_codec.h
hda_local.h
hda_proc.c
hda_generic.c
hda_generic.h
patch_realtek.c

hda_beep.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921c.zip
hda_beep.h


hda_eld.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921d.zip

hda_hwdep.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921e.zip

hda_jack.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921f.zip
hda_jack.h



hda_intel_trace.h https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921g.zip
hda_patch.h
hda_trace.h

If I had to bet, uniaud32-linux-3.9.11-20220921b.zip is the one that will break things. It's both the most obvious source files impacted, but also it's the largest code change - unfortunately there are interdependencies between those source files, so changing one at a time wasn't possible.

Hopefully I'm wrong, and it will be once of the smaller code changes.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 21, 2022, 04:56:25 pm
Heh Paul!

Had our first frost here last night - temps hit -2 C,. so we're now into the major change that is our fall (three weeks ago in Alberta we hit 35 C, and it gets a LOT colder than this in the dead of our winter  :(

Hi Mark,
Thanks - tonight, I'll aim to produce several builds, slowly replacing the 3.8.x pci/hda/* code with the 3.9.x pci/hda/* code - ideally one source file at a time.

Note: these are specifically directed at trying to understand what broken with his hardware between the sound code in linux 3.8, and linux 3.9.

Code: [Select]
hda_intel.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921.zip

hda_auto_parser.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921b.zip
hda_auto_parser.h
hda_codec.c
hda_codec.h
hda_local.h
hda_proc.c
hda_generic.c
hda_generic.h
patch_realtek.c

hda_beep.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921c.zip
hda_beep.h


hda_eld.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921d.zip

hda_hwdep.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921e.zip

hda_jack.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921f.zip
hda_jack.h



hda_intel_trace.h https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921g.zip
hda_patch.h
hda_trace.h

If I had to bet, uniaud32-linux-3.9.11-20220921b.zip is the one that will break things. It's both the most obvious source files impacted, but also it's the largest code change - unfortunately there are interdependencies between those source files, so changing one at a time wasn't possible.

Hopefully I'm wrong, and it will be once of the smaller code changes.

Cheers,

Paul

All of them (a to g) on my computer and unzipped - just needed to make sure I got them all - will go methodically from top to bottom to see whats shaking and generate testlogs for each.

Thanks for this tremendous amount of detail and work, Sir - I'm hoping this may solve other peoples problems as well (ie. not just a personal tilt at the windmill) - if other chipsets haven't worked in the past this may be very useful.

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 21, 2022, 07:56:38 pm
Hi Paul!

Hi Mark,
Thanks - tonight, I'll aim to produce several builds, slowly replacing the 3.8.x pci/hda/* code with the 3.9.x pci/hda/* code - ideally one source file at a time.

Note: these are specifically directed at trying to understand what broken with his hardware between the sound code in linux 3.8, and linux 3.9.

Code: [Select]
hda_intel.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921.zip

hda_auto_parser.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921b.zip
hda_auto_parser.h
hda_codec.c
hda_codec.h
hda_local.h
hda_proc.c
hda_generic.c
hda_generic.h
patch_realtek.c

hda_beep.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921c.zip
hda_beep.h


hda_eld.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921d.zip

hda_hwdep.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921e.zip

hda_jack.c https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921f.zip
hda_jack.h



hda_intel_trace.h https://smedley.id.au/tmp/uniaud32-linux-3.9.11-20220921g.zip
hda_patch.h
hda_trace.h

If I had to bet, uniaud32-linux-3.9.11-20220921b.zip is the one that will break things. It's both the most obvious source files impacted, but also it's the largest code change - unfortunately there are interdependencies between those source files, so changing one at a time wasn't possible.

Hopefully I'm wrong, and it will be once of the smaller code changes.

Cheers,

Paul

Unfortunately Paul, your multichoice answer on this test was again correct (your on a roll, just in the hardest way  :P).

Things broke @ hda_auto_parser.c (your build uniaud-linux-3.9.11-2022-09-21b)

I did run ALL 7 builds, in sequence, and no sound was generated at all after I ran "b" - "a" (or the first file as you didn't label it as such) was fine.

To be thorough (just the troubleshooter in me  8) ), I ran all 7 both as a warm boot and a cold boot, just to make sure, like the Recon3D, that wasn't causing silences - same situation with all builds, "b" and beyond.

Like when I was helping Wim on the other project I've been actively involved in, I assume a few things, operating independently - obvious once one source code change is found that causes the silences, the subsequent source code changes would not work, but I tested all of them just in case.

I've generated testlogs for all, on the off chance more than one may tell you something (attached below). They are marked SCa to SCg

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 22, 2022, 12:20:38 am
Thanks Mark - you can keep your frost and -2C - we're heading for a a partly cloud 18C here today :)

Thanks for that testing - I'll take a look at the differences between HYPERFIRE-20220921-uniaud-3.9.11-SCa-ALC892.log and HYPERFIRE-20220921-uniaud-3.9.11-SCb-ALC892.log this afternoon, in the hope it will shed some light. I too, hope that this will unlock the reason that some machines don't product sound. Who knows, it might even explain why HDMI doesn't product sound, even though it all 'looks' to be working from log files.

I almost wish I had a system here that didn't work.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 22, 2022, 04:11:07 am

I almost wish I had a system here that didn't work.

Cheers,

Paul

Thats because you probably have too orderly a mind, Paul (short of the bouts of idea inspiring red wine  ;D). To have your systems not working perfectly would just not be on ..........

Philosophically, I like to think there are only two type of thinkers in the world - linear and lateral thinkers.

Linear thinkers try to troubleshoot on the basis of moving from a to b to c, in logical progression. Emergency experience has taught me to think laterally - some times things are solved by taking seemly unconnected things, from different places, and trying to put them together until they, intuitively, just seem like the proper decision and/or way to proceed.

Being a hardware geek, I use Virtual Box on a number of systems but have built over 10 systems from the bare metal box up - my office is strewn with new and legacy systems that didn't work originally but do now, through a lot of head scratching, in addition to numerous portables ........ and in various exotic configurations with several operating systems.

My network is like a Hydra ........... and expanding constantly ........... and I even have run a business on it, for 28 years ...........

As to HDMI If I can help there let me know .......... both of the test system partitions are attached to graphics cards which also report HDMI sound capabilities, in PMUnimix (whether they are indeed there I don't know) - I have one connected to the Bang and Olufson guts of the monitor, but can't get it to work at the moment, under ArcaOS, with your drivers ......... the other could also be connected .........

Jury_Rig re'Us you might say, over here .......... but one impossibility at a time, as my emergency experience always brings me back down to earth, to refocus on the problems at hand.

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 22, 2022, 08:30:12 am
Hey Mark - slight change of tack - I patched the linux 3.9 code based on the 3.8.13 code - there's a build at https://smedley.id.au/tmp/uniaud32-linux-3.9-20220922.zip
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 22, 2022, 11:28:46 am
Anyone other than Mark seeing no audio with the 5.x builds (5.15.x or 5.19.x) after a warm boot? I have an old T60 that does this, but it would be interesting to see if there are any common themes in hardware that does this.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 22, 2022, 03:37:56 pm
Heh Paul!

Hey Mark - slight change of tack - I patched the linux 3.9 code based on the 3.8.13 code - there's a build at https://smedley.id.au/tmp/uniaud32-linux-3.9-20220922.zip

Latest patch from legacy 3.8.13 to 3.9 generated no sound - tried both warm and cold boot.

Testlog generated and attached below.

A fine day to you and Best!
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 22, 2022, 03:51:26 pm
Anyone other than Mark seeing no audio with the 5.x builds (5.15.x or 5.19.x) after a warm boot? I have an old T60 that does this, but it would be interesting to see if there are any common themes in hardware that does this.

Just a little further to Paul's post, the condition I am seeing is sound is working perfectly when doing a cold boot or reboot. However, after a warm reboot the sound disappears. To get it back I have to perform a cold boot again.

Discovered this rather strange little anomaly only when testing my Sound Blaster Recon3D Z series sound card last week, so if anyone has similar they might want to test. But it would be interesting to see if anyone else has experienced the same with other sound cards or chipsets that have been found to work, then seem not to.

It is reproducible on Pauls latest recent build 5.19.9 here, but I suspect I would see it if I back leveled as well.

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 23, 2022, 03:14:59 am
Hey Mark,

Heh Paul!

Hey Mark - slight change of tack - I patched the linux 3.9 code based on the 3.8.13 code - there's a build at https://smedley.id.au/tmp/uniaud32-linux-3.9-20220922.zip

Latest patch from legacy 3.8.13 to 3.9 generated no sound - tried both warm and cold boot.

Testlog generated and attached below.

A fine day to you and Best!

Whilst I'm not suprised by this result, the amount of code restructuring that took place makes it hard to nail things down. Does this machine have multiple output jacks? Including front panel? Can you try plugging speakers into different jacks to see if it's a jack configuration issue?
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 23, 2022, 11:02:53 pm

Whilst I'm not surprised by this result, the amount of code restructuring that took place makes it hard to nail things down. Does this machine have multiple output jacks? Including front panel? Can you try plugging speakers into different jacks to see if it's a jack configuration issue?


HYPERFIRE is one of my custom built machines, built from the ground up, so I know it rather intimately.

It contains an ASUS F1A75-V Pro Motherboard, with an AMD quad core CPU (forget the specific model). The ALC892 chipset is built-in on the motherboard.

I have liked using ASUS motherboards for years, because of their excellent, well laid out User manuals, both graphically and verbally (something highly appreciated by method engineering professionals like myself  8) ). Its very difficult to go wrong following an ASUS manual, for assembly or other purposes.

As to audio connectors the motherboard has a standard set of High Definition Audio connectors (coloured orange, black, grey, light blue, lime and pink) on the back of the motherboard only, and was connected to the Bang and Olufsen stereo system on the monitor with a single cable plugged into the standard green connector. Therefore it was relatively easy to swap the single cable for the 5:1 surround sound cables (orange, black and lime), attaching them to the appropriate connectors on the motherboard.

Using the modified 3.9.11 0220 driver, I was able to generate surround sound, something I was not able to do with the legacy ArcaOS 5.0.6 driver based on the legacy ALSA code - I tested your latest 0922 driver, Paul, and was still not able to generate any sound.

So I don't really think, Sir that connection to the wrong connectors is the problem, but it pays to be thorough, so I've now confirmed that.

As you know I consider myself a lateral thinker, so I thought I'd spend some time coming at this from a different angle - namely, have we ever asked the questions as to whether the Linux community had problems with this chipset before?

Well, they did .......... and 54 others as well.

I spent some time on search engines, which sent me to various Debian and Fedora and other Linux sites. I can send you the links if you wish, Paul, but I attached a couple of text files down below, that you and others might be interested in.

I will say that trolling quietly (fishing wise) through a linux forum reminded me of Physics lectures of 40 years ago - a lot of esoteric information, technical references, some arguments, and a lot of debate (and I'm not in anyway dissing any of you who are part of said community - its just that, looking in, the Linux forums are an experience unto thyself  ;D )

One item that struck me, referenced four times was mention of "auto-mute" - is there any reference to that in any of the PCI/HDA source code?

It seemed that switching "auto-mute" off (the default seemed to be "on") seemed to solve the no sound problems for various people - the problem is, the forum contributors really didn't say where?

There were references to "AlsaMix", "PulseSound" and "Audacity" of which I've only heard of the last, but I assume they are mixer programs similar to PMUnimix. But not sure if the discussions were recommending adjusting "auto-mute" through there or not.

But it also gave me the possible mad hypothesis that, because this was such a problem, perhaps it was solved in a later release of the Linux kernels v3 or above, so downloaded and tried your 3.14.9 and 4.0.9 builds and installed them - sadly no change and no sound.

Its difficult for me to tell how long ago in the Linux development all that occurred, Paul, so perhaps I caught a red herring <sic>, but you may have a better idea as to where you might scrounge up some linux based changelogs to see whether and when linux developers might have done something, which may not be included in a source file your using.

I can't believe that it would still bee a problem they have not addressed in their kernel for possibly 55 sound chipsets. As an aside, it also might be worthwhile, if their is anyone who has not gotten their sound to work with your drivers to try the latest V3 kernel derived driver that worked for me - that would be contained Uniaud32-3.9.11-20220920.zip - the list of chipsets that had problems is part of what is in the first text file below.

Anyway just another of my mad musings on the subject of lateral thinking Sir and a fine weekend to you!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 12:20:52 am
Hey Mark,

Sorry - didn't mean to imply you were hooking things up wrong, some of my searches suggested that there could be configuration issues within the driver relating to the jack config. Interesting to see that you could get surround sound working...

re: auto-mute - you triggered a memory about https://trac.netlabs.org/uniaud/ticket/248

If this was a 'thing' you should have a mixer control similar to:
Code: [Select]
Control ID: 18
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Auto-Mute Mode]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 2, item: 0, name: [Disabled]
  value 0: 1

Looking at HYPERFIRE-20220921-uniaud-3.9.11-SCb-ALC892.log, it seems you do.
Code: [Select]
Control ID: 19
Interface: virtual mixer device
Device(client): 0, subdevice (substream) 0
Name: [Auto-Mute Mode]
Index: 0
Element type: enumerated type
Count of values: 1
Value:
 Bounds: items: 2, item: 0, name: [Disabled]
  value 0: 1

It would be interesting to see what setting this mixer control to 0 does.

HYPERFIRE-20220921-uniaud-3.9.11-SCa-ALC892.log also has the same value for auto-mute - but sound works <shrug>

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 01:09:37 am
I just stumbled across https://www.os2world.com/forum/index.php?topic=2436.75;msg=26471 - specifically the message from Andy...

Seems I missed this message and didn't back out the code he suggested.

On a whim... please try https://smedley.id.au/tmp/uniaud32-linux-3.9-20220924.zip and see if it changes anything....

Meanwhile, I'm also interested in the results of setting auto-mute to 0.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 24, 2022, 02:44:41 am
Hey Paul!

Hey Mark,

Sorry - didn't mean to imply you were hooking things up wrong, some of my searches suggested that there could be configuration issues within the driver relating to the jack config. Interesting to see that you could get surround sound working...


I've been gone over by experts (read: premiers, presidents and prime ministers), so I may be the one person you've ever met that apologies are not necessary - I think literally - if you have a brainstorm, fire away ......... because out of strange ideas, great things may come .......

This is not a shaming match, Sir, at least from what I've seen in the past - I look at everything you say as literal and worthy of consideration - I wish the rest of the world would be as considerate.

Fire away anytime - out of that great things may come!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 24, 2022, 03:04:47 am
I just stumbled across https://www.os2world.com/forum/index.php?topic=2436.75;msg=26471 - specifically the message from Andy...

Seems I missed this message and didn't back out the code he suggested.

On a whim... please try https://smedley.id.au/tmp/uniad32-linux-3.9-20220924.zip and see if it changes anything....

Meanwhile, I'm also interested in the results of setting auto-mute to 0.

Couldn't get to your link, Paul - says its not found (404)

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 03:06:11 am
Hey Mark,
I just stumbled across https://www.os2world.com/forum/index.php?topic=2436.75;msg=26471 - specifically the message from Andy...

Seems I missed this message and didn't back out the code he suggested.

On a whim... please try https://smedley.id.au/tmp/uniad32-linux-3.9-20220924.zip and see if it changes anything....

Meanwhile, I'm also interested in the results of setting auto-mute to 0.

Couldn't get to your link, Paul - says its not found (404)

I can't believe you didn't spot the deliberate typo? :) try https://smedley.id.au/tmp/uniaud32-linux-3.9-20220924.zip
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 24, 2022, 03:22:06 am
Hey!


I can't believe you didn't spot the deliberate typo? :) try https://smedley.id.au/tmp/uniaud32-linux-3.9-20220924.zip

Too much good Savignon Blanc my friend, and too nice a lovely Beef Stew courtesy of too lovely a wife, on a Friday night - need I say more?

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 24, 2022, 04:42:09 am
Hey Paul!


I can't believe you didn't spot the deliberate typo? :) try https://smedley.id.au/tmp/uniaud32-linux-3.9-20220924.zip


Unfortunately, executed a hard Trap 000e in UniAud32, in what was otherwise a very good week.

If any other bright ideas come to mind, something for the 24th in my neck of the woods - a fine day to you, Paul and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 05:13:38 am
Hey!


I can't believe you didn't spot the deliberate typo? :) try https://smedley.id.au/tmp/uniaud32-linux-3.9-20220924.zip

Too much good Savignon Blanc my friend, and too nice a lovely Beef Stew courtesy of too lovely a wife, on a Friday night - need I say more?

M

Well that was unexpected! The only other bright idea is to use unimjx from the command line to toggle auto mute to 0.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 10:23:05 am
...for Hyperfire with the driver from 3.9.11.... something like:
unimix -id19 -val0=0

19 is the control number for auto-mute from 'unimix -list'
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 11:37:54 am
Hey All,

I just stumbled across https://www.os2world.com/forum/index.php?topic=2436.75;msg=26471 - specifically the message from Andy...

Seems I missed this message and didn't back out the code he suggested.

On a whim... please try https://smedley.id.au/tmp/uniaud32-linux-3.9-20220924.zip and see if it changes anything....

Meanwhile, I'm also interested in the results of setting auto-mute to 0.

Wondering if there is any feedback from users running HDA hardware on differences (good or bad), between:
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220924.zip
&
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220918.zip

The difference is the 0924 build has the patch that @Andy Willis suggested removing two years ago removed. I haven't committed the change yet, pending feedback.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 24, 2022, 11:57:23 am
Separate to the above post, I also updated the uniaud32-next branch to the 5.15.70 code - builds at https://smedley.id.au/tmp/uniaud32-linux-5.15.70-20220924.zip
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on September 24, 2022, 12:24:59 pm
I'm hoping that I will have a house to move into in the coming months, then I can unpack my stuff in storage and start using OS/2 again and be a bit more helpful :)
Either that or I'm going to start driving around Australia again!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on September 24, 2022, 03:00:06 pm
Hi Paul,

  Both the latest 5.15.70 and 5.19.9 builds work just fine here. No obvious difference from the last versions. No problem with cold vs. warm boot either - they both just work.

Regards,
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 24, 2022, 04:39:21 pm
Paul, I have thew same to report as Dave. Still working fine here.
Title: Re: Updated Uniaud32 build
Post by: Jan-Erik Lärka on September 24, 2022, 08:06:22 pm
Same here with 5.x (on conexant).
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 25, 2022, 12:47:02 am
Hey Ian,
I'm hoping that I will have a house to move into in the coming months, then I can unpack my stuff in storage and start using OS/2 again and be a bit more helpful :)
Either that or I'm going to start driving around Australia again!

Where abouts are you these days? still in WA? Thanks again for os2site - I was chasing down a copy of ndcrypt and the link at ecomstation.ru was borked - but sure enough os2site had a copy :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Ian Manners on September 25, 2022, 07:51:43 am
Hi Paul,

Drove to Melbourne from Perth mid January, drove back to Perth with a week in Kalgoorlie early August with my dog in the back seat, due to ageing family members I think I will be staying in Perth for a couple more years.

Looking forward to getting my server out of storage and back on the internet, never did get SFTP/wget etc working to update the public website, will look at networking my server to Linux and update it that way. Maybe even put the server on VirtualBox and use the shared directory option. I stopped updating my servers version of os2site.com late last november when I sold my house but have kept a basic structure on my Linux laptop to put updates in that can be transfered to os2site.com later.

Glad os2site.com is helpful to others :)
Title: Re: Updated Uniaud32 build
Post by: roberto on September 25, 2022, 12:17:19 pm
Quote
Wondering if there is any feedback from users running HDA hardware on differences (good or bad), between:
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220924.zip
&
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220918.zip
The ...924.zip is GOOD
The ...918.zip is Bad
Saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 26, 2022, 12:07:47 am
Hi roberto,

Quote
Wondering if there is any feedback from users running HDA hardware on differences (good or bad), between:
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220924.zip
&
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220918.zip
The ...924.zip is GOOD
The ...918.zip is Bad

Are you able to be a little more descriptive on what GOOD and Bad mean?

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 26, 2022, 12:19:14 am
Paul in the two drops you asked to comment on. 20229018 is extremely buggy. PM123 really acts strange with this driver. It does work though. 20220924 acts much better and is stable for me.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 26, 2022, 02:42:49 am
Hi Eugene,
Paul in the two drops you asked to comment on. 20229018 is extremely buggy. PM123 really acts strange with this driver. It does work though. 20220924 acts much better and is stable for me.

That's really interesting - it does support what Andy commented - in that his change caused more problems than it solved.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: OS4User on September 26, 2022, 12:18:46 pm
please try https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220924.zip

it works ok fo me (Sound card name = SB Audigy 1 [SB0092])
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 26, 2022, 04:56:59 pm
Heh Paul!

...for Hyperfire with the driver from 3.9.11.... something like:
unimix -id19 -val0=0

19 is the control number for auto-mute from 'unimix -list'

In case your wondering I haven't dropped off the face of the earth - got knocked off-line due to Hurricane (now Tropical Depression) Fiona, for three days.

And if people are wondering how Alberta, a solely landlocked province was so affected, I'm on Satellite Internet out here in the country and the ground station for my satellite service provider is roughly 5000 km away, in New Brunswick, in what we call "The Maritimes" on the Canadian East Coast.

It was a bad storm - power, telephone, cell and internet service were all affected severely or gone completely - in some cases whole houses (in one case with a person in it) were simply removed and ended up destroyed out in the ocean, off the shore of Newfoundland.  :(

But I do have some news.

First of all auto-mute in Unimix was id20, not id19 (at least on HYPERFIRE .... I used the command "unimix -names" from a prompt, to find the correct setting - it varies, as Manticore II reported it at id23)

But I can report that I disabled it and ......... SOUND WORKS with the ALC892 chipset!!!!!!

command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

There is another benefit, as well - that was tested with your 5.19.9 20220924 driver debug version - I will switch it later, but I doubt the release version will give any problems. IOW no back pedaling involved  8)

@Edit 1 - the release version has now been tested and also works perfectly

So the puzzle of ALC892 is solved - and lovely clear sound at that, at least in PM123.

Something to add to the testing guide I prepared, if someone is having problems with any of the other 54 chipsets which seem to have or have had this problem.

Best as always!

M

Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 26, 2022, 05:20:32 pm
Mark there have been times when we in West Virginia have suffered fro the effects of a hurricane. So we pray for you And all of Canada that was affected.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 26, 2022, 05:31:48 pm
Heh Eugene!

Mark there have been times when we in West Virginia have suffered fro the effects of a hurricane. So we pray for you And all of Canada that was affected.

Thanks for your thoughts - Anne, my wife, has a sister and her husband, as well as their daughter and grandchildren in Dartmouth, Nova Scotia, across from Halifax - haven't been able to get a hold of them yet - all power and telephone service still out in that area.

You would well know the drill related to Hurricanes, Sir - Fiona hit as a Cat4, then Cat3 over PEI and Newfoundland then moved inland as a Tropical Storm - that persisted for 2 days - still assessing the damage and making repairs.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Tom on September 26, 2022, 08:04:58 pm
command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

Just put that unimix command in the file startup.cmd , and it will be run every time you start your system.
Title: Re: Updated Uniaud32 build
Post by: roberto on September 26, 2022, 08:56:50 pm
The ...924.zip is GOOD
The ...918.zip is Bad
Quote
Are you able to be a little more descriptive on what GOOD and Bad mean?

The two versions work. But with 924, when jumping during the vision of a video on YouTube, they are faster and with less interference. If I visualize a video from beginning to end both work the same.
saludos
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 26, 2022, 09:24:25 pm
Heh Paul!

Separate to the above post, I also updated the uniaud32-next branch to the 5.15.70 code - builds at https://smedley.id.au/tmp/uniaud32-linux-5.15.70-20220924.zip

Tried your latest branch port this morning on Manticore II:

The AD1988B chipset works perfectly

Switched to the SB Recon3D on the same system and have excellent sound as per the previous builds - however still only Master Playback Control section is present in PMUnimix, so can only get sound in stereo from the front speakers - however running unimix -names, it lists 45 ID sections - 18 of which are switches (there is no auto-mute function on that card).

I might try enabling some of those switches to see what happens, and whether PMUnimix would recognize them after being enabled.

And also - the Recon3D is still only activated and recognized from a cold boot.

Will try Hyperfire with the new 70 branch later today.

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 26, 2022, 09:27:13 pm
Heh Tom!

command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

Just put that unimix command in the file startup.cmd , and it will be run every time you start your system.

I was thinking about that after I posted, but haven't worked much with Startup.CMD in the past - will have a look at that tonight or tomorrow.

But thanks for the tip and Best!

Mark
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 27, 2022, 12:28:35 am
Hey Mark,

First of all auto-mute in Unimix was id20, not id19 (at least on HYPERFIRE .... I used the command "unimix -names" from a prompt, to find the correct setting - it varies, as Manticore II reported it at id23)

But I can report that I disabled it and ......... SOUND WORKS with the ALC892 chipset!!!!!!

command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

Very brief reply for now... re: 'Auto Mute' I'm going to see if I can automatically disable it within soundmixer.c - I just need to spend some quality time reading the code and understanding how it works..

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 27, 2022, 02:03:51 am
Hey Paul!

Hey Mark,

First of all auto-mute in Unimix was id20, not id19 (at least on HYPERFIRE .... I used the command "unimix -names" from a prompt, to find the correct setting - it varies, as Manticore II reported it at id23)

But I can report that I disabled it and ......... SOUND WORKS with the ALC892 chipset!!!!!!

command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

Very brief reply for now... re: 'Auto Mute' I'm going to see if I can automatically disable it within soundmixer.c - I just need to spend some quality time reading the code and understanding how it works..

Cheers,

Paul.

Changing things slightly in this post, as it sounded a bit churlish last time, which was not my intention. Paul

Will be great if the next version works for everyone - I'm just happy that I finally have a solution  personally......... but will test of course if you find a solution within the driver code.

BTW - 5.15.70 works fine on HYPERFIRE with  ALC892 - as long as I re-enable Unimix auto-mute disabling, after a boot.

Off for my ride this evening - and a fine day to you Sir!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 27, 2022, 11:49:01 am
Hey Mark,

I took a look at soundmixer.c this evening - I don't see a way to easily set auto-mute to 0 by default. For now, the workaround will need to be using unimix in startup.cmd

I'll look to add something to the readme as well

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 27, 2022, 04:34:38 pm
Good Morning, Paul!
(my time)

Hey Mark,

I took a look at soundmixer.c this evening - I don't see a way to easily set auto-mute to 0 by default. For now, the workaround will need to be using unimix in startup.cmd

I'll look to add something to the readme as well

Cheers,

Paul

I think that is best for the moment, Sir - your on a roll with stable builds, so no use screwing things up just to see if you can help me out. I plan to spend a few minutes either creating Startup.cmd or at least a command file for testing today.

It would be interesting if we can find anyone else who has any of the other 54 chipsets in their system that were mentioned in one of those text files I posted a few days ago, and had PMUnimix recognizing the card but not having any sound - I can't believe I was the only one with the problem (especially when it crept into the UniAud build posted with the ArcaOS v5.0.7 release) - perhaps they could test disabling auto-mute, if present, as well.

I'm also going to update the UniAud testing guide I created, as well, adding the Auto-Mute reference and solution in addition to a couple of other things, so should have v1.2 posted this week.Your more than welcome to include that with your test builds, if you think it is useful.

I started looking into the Soundblaster Recon3D to see what might be done there - as I mentioned it only has one sound section under PMUnimix (Master Playback volume), which as I mentioned works on cold reboot, but only in stereo.

I've attached the output from unimix -names for the Recon3D below if your interested - 45 id's with 18 switches present, but almost all their designators have rather non-standard names so I don't know if that is part of the problem (being a non-programmer, I don't know how those volume control lines are addressed in the code)

I'm going to play around with unimix today - I noted id19 is the "equalizer playback switch" - perhaps that is off by default and not seen by PMUnimix (don't know if Master Playback Volume needs the equalizer enabled).

I'm also going to go out fishing again on the linux forums to see if there is any mention of either a cold reboot problem or how to enable surround sound with this card. Perhaps I'll catch a Pickeral (a very good edible whitefish native to Canadian lakes, especially in the province of Manitoba, where I originally come from).  8)

M

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 27, 2022, 05:47:09 pm
Hey Paul!

Anyone other than Mark seeing no audio with the 5.x builds (5.15.x or 5.19.x) after a warm boot? I have an old T60 that does this, but it would be interesting to see if there are any common themes in hardware that does this.

It seems there is a Bugzilla report about this problem from 2020:

https://bugzilla.kernel.org/show_bug.cgi?id=205863

Bug 205863
Summary: Sometimes no sound at cold boot using Sound Blaster Recon3D / Z-Series

Reading the report, I think the person reporting it had the subject line backwards - it should read "Sometimes no sound in a warm boot using the Recon3D / Z-series" as he claimed that shutting down the system totally, and rebooting corrected the problem (which is what I define as a cold reboot).

So it may be specific to my sound card only - I don't know, Paul, as to where you would look to see what might have been done, to attempt correct the problem - it seems from reading what was tried, it was merely a troubleshooting report, diagnosing the problem, but providing no defined solution, but perhaps I'm reading it incorrectly.

@Edit 1 - One other curiosity I noted today - warm boot for the Recon3D does work - but only if you cold boot to another operating partition first, then I do a warm boot over to the partition I'm testing the Recon3D on.

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 27, 2022, 08:33:13 pm
Heh Paul!

As a final test today (before getting back out into the yard), I did try to enable other sections in the unimix -list for the Recon3D using the command prompt, but was unsuccessful.

However it seems this too was a problem encountered before in Linux - so I did indeed reel in a Pickeral!

From ArchLinus.org ( https://bbs.archlinux.org/viewtopic.php?id=236364: )

<snip>
________________________
Update: I have everything working now! SBZ card now works PERFECTLY under Linux thanks to Conmanx360's patch! Headphone output, S/PDIF etc. ALSA, PA... it all works.

If anyone's interested, my PKGBUILD: https://paste.pound-python.org/show/hJv … XHMHRFDFy/
Custom patch (rename it patch_ca0132.patch) : https://paste.pound-python.org/show/wH7 … 1SNR9z4ra/

Big thanks to Conmanx360 for tracking down the issue I had (I had a board with a different revision number).

Now... how does one get this patch into the Kernel?
____________________

<quote ends>

It seems it was incorporated into Linux kernel 4.18-rc1 (I've attached the change log below - can be accessed on-line at https://lkml.iu.edu/hypermail/linux/kernel/1806.0/04044.html )

Commitments there were significant, after reviewing the change log, related to getting sound chipsets to work, so thought you might be interested

The important changes related to this Recon3D sound card, I think:

HD-audio:
- Finally better support for some CA0132 boards, allowing Windows firmware
- Blacklisting power save default some known boards reported on Fedora

It seems the Recon3D is a CA0132 board, and not addressing the  Windows based firmware through Linux was the problem.

I included the second line because I don't know if the problem related to "power save" may be the problem related to warm boot.

I also have attached a second text file attached below which contains the CA0132 board patch, which is also available at:

https://paste.pound-python.org/show/wH7XZd6Nldo1SNR9z4ra/

But enough trolling and chumming, fish wise, for the day - off to restructuring our rose beds and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 28, 2022, 01:59:41 am
We don't currently support firmware loading... so if this chip needs firmware - it won't work. Warm boot from windows/linux works as the firmware is already loaded....
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 28, 2022, 05:03:56 am
Heh Paul!

We don't currently support firmware loading... so if this chip needs firmware - it won't work. Warm boot from windows/linux works as the firmware is already loaded....

So at least we can check this one off as nothing more that can be done, reasonably - but at least we know why  :)

Off to bed and a fine day to you!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 28, 2022, 07:13:39 pm
Heh Paul!

We don't currently support firmware loading... so if this chip needs firmware - it won't work. Warm boot from windows/linux works as the firmware is already loaded....

Just one further comment - I have two ArcaOS partitions - one which is my working partition (D:), the other is my ArcaOS test partition (F:) - the warm boot also has worked when switching from D: to F:, so I'm not sure if that explains the firmware already being loaded.

If Master Playback volume is present would that be addressed in the firmware as well, or is some other process involved (like generic HDA Audio?)

Don't know if that question makes any sense but trying to understand why the Recon3D "partially" works ie. can play stereo perfectly.

BTW, ran across this site:

https://forums.linuxmint.com/viewtopic.php?t=341022

It had someone there who is a moderator (alias Mr. Een) who is quite experienced in sound problems and has a blog related to fixing same in Linux:

http://mreen.epizy.com/SoundFixTips.html?i=1

Related to HDMI he was mentioning that many of the HDMI problems relate to the OS not redirecting sound through the HDMI cable to the monitor, instead of through the other speaker systems which may be present.

He added a command line to the post to assist the person asking for help, which did solve his problem:

d set-card-profile alsa_card.pci-0000_00_0e.0 output:hdmi-stereo

Doubt it would be of help to us, but at least an explanation as to what may be causing you grief with HDMI - most of his advice is related to settings in the Linux mixer programs, which is of no help with ArcaOS, as it seems the mixers in linux are much more feature rich.

But a good resource that might help explain things encountered in future, when audio problems are present.

Best!
 
M

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 29, 2022, 10:21:35 am
Hi Mark,
We don't currently support firmware loading... so if this chip needs firmware - it won't work. Warm boot from windows/linux works as the firmware is already loaded....

Just one further comment - I have two ArcaOS partitions - one which is my working partition (D:), the other is my ArcaOS test partition (F:) - the warm boot also has worked when switching from D: to F:, so I'm not sure if that explains the firmware already being loaded.

If Master Playback volume is present would that be addressed in the firmware as well, or is some other process involved (like generic HDA Audio?)

I'm assuming that sound only works if linux or windows is booted first, then I'd expect it to keep working until a cold boot occurs.

ie Linux/Windows - then D: then F: would work. If this isn't what you're describing, please let me know.

No idea what causes the Master Playback to show - what driver versions are in play? ca0132 support was added relatively recently...
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 29, 2022, 05:24:23 pm
Heh Paul!

Hi Mark,

I'm assuming that sound only works if linux or windows is booted first, then I'd expect it to keep working until a cold boot occurs.

ie Linux/Windows - then D: then F: would work. If this isn't what you're describing, please let me know.


Actually I have no Linux partitions, and the last time I booted my Windows partition (C:) was over six months ago on Manticore II.

D: is an ArcaOS partition running v5.0.7 - when I need Windows support on Manticore II I usually run VirtualBox on that partition. I use that partition exclusively for tasks like Internet search, messaging, playing games through ArcaOS (and DosBox running under ArcaOS), writing a lot of internal documents using Lotus Word Pro, Lotus Spreadsheets, etc. I also run VLC, SMPlayer, PM123, and various other multimedia software (including a rather neat legacy program call Warped Wave Editor (WaWE) and of course Tonigy). That partition is connected to the built-in AD1988B sound chipset as standard.

F: is a test partition running ArcaOS v5.0.7 and is dedicated solely to testing, although it has all the capabilities of the D: partition also installed - if I needed to (ie. something has screwed up with my working partition) I can switch to F: and run it successfully for days, until I can get the other partition working. Currently, I've left partition F: using the Recon3D through the 5:1 surround sound system, as I know AD1988B has no problems

Hyperfire is configured in exactly the same way, with two ArcaOS partitions, only it has the ALC892 sound chipset and the Sound Blaster Xi-Fi Titanium card. It also has a Windows partition (C:), which is rarely booted, and which I originally set up only to play games.

Just the emergency planner in me, Paul - I hate situations occurring where they can be planned for, but were not, and something serious occurred - my office is filled with battery back-ups, multiple surge protectors, etc., as well as a NetGear Server, for archive and back-up purposes.

So as to warm boot and it working on F: after cold booting to D: and then rebooting to F:,  - the reason I mentioned it is that I can't see it enabling the firmware, the way the system is configured, as no Linux or Windows partition is involved at all.

The warm boot phenomenon is reproducible every time - so obviously something else is involved.

No idea what causes the Master Playback to show - what driver versions are in play? ca0132 support was added relatively recently...

As you know, I got into this sound testing late (only started in March) and at the time I wrote a long post which I had printed out and referred back to today - the first version I tested was 5.14.7-20211028. At the time, referring back, I wrote that Master Playback volume was reported under PMUnimix but was not controlling the sound, stereo sound was only present from the front speakers, and I had to turn down the sound using the surround sound hardware controls as sound was blasting out at max. It also would only work after a cold reboot.

The behavior with this card is the same right up to 5.19.9, which I tested last week, as well as 5.15.70. Having access to your archives I can test further back, if you wish.

It was the reason that I had great hopes that the ArchLinux post a few posts back now may have had a solution, as the patch for CA0132 (the Recon3D is one such board) was actually published in detail and the changelog at the time also indicated they committed that, and many other sound fixes to kernel 4.18 rc-1.

Best as always!

M



Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 30, 2022, 02:02:42 am
Hey Mark!!

Wow! That's super interesting! I guess it's not firmware load after all - as switching between two AOS partitions could not make the firmware load.

The patch you provided was interesting - but that code is from linux 4.18 - we're using linux 5.19.x code which would already include that code (or a later iteration of it).

Does it matter is you boot D: first then F: or f: first, then d:? What uniaud32 versions are in play on each partition? Just looking for clues..

Perhaps a log file from the 1st cold boot, then the 2nd warm boot where sound works. I'm not confident it will show anything, but who knows?

I'm also going to review my patches to patch_ca0132.c - ie from https://github.com/psmedley/uniaud32/commit/12e7ce146924b3e77056e75095e7b0a9a7571399 (I did an import of the netlabs SVN into github - it's easier to review code via the github gui, and it's also a backup)

With the surround speakers - I'm not surprised you have issues controlling volume - take a look at https://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-exp/lib32/soundmixer.c#L398 and note the 'todo' comment. Adjusting the rear speakers is actually adjusting the overall mixer volume. I noticed this the other day looking at where I might be able to manipulate the 'auto mute' setting.

I'll be honest - I don't really understand how this code works - as I like to tell people, I'm not a programmer, I'm just a hacker :)

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on September 30, 2022, 02:44:59 am
Heh Paul!

Hey Mark!!

Wow! That's super interesting! I guess it's not firmware load after all - as switching between two AOS partitions could not make the firmware load.

The patch you provided was interesting - but that code is from linux 4.18 - we're using linux 5.19.x code which would already include that code (or a later iteration of it).

Does it matter is you boot D: first then F: or f: first, then d:? What uniaud32 versions are in play on each partition? Just looking for clues..

Perhaps a log file from the 1st cold boot, then the 2nd warm boot where sound works. I'm not confident it will show anything, but who knows?

I'm also going to review my patches to patch_ca0132.c - ie from https://github.com/psmedley/uniaud32/commit/12e7ce146924b3e77056e75095e7b0a9a7571399 (I did an import of the netlabs SVN into github - it's easier to review code via the github gui, and it's also a backup)

With the surround speakers - I'm not surprised you have issues controlling volume - take a look at https://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-exp/lib32/soundmixer.c#L398 and note the 'todo' comment. Adjusting the rear speakers is actually adjusting the overall mixer volume. I noticed this the other day looking at where I might be able to manipulate the 'auto mute' setting.

I'll be honest - I don't really understand how this code works - as I like to tell people, I'm not a programmer, I'm just a hacker :)

Cheers,

Paul

Your a hacker, not a programmer - I'm a troubleshooter, not a programmer - out of that may come great things or not, Sir <grin>

We're both asking the crazy questions, Paul - sometimes out of that may come interesting solutions.

Have never been the person to merely confirm - I want to know W-H-Y ...... because in the past lives have depended on it

A busy day where I accomplished a number of important things albeit minor, society wise, then our new statutory holiday and long weekend, in Canada - National Truth and Indigenous Reconciliation day tomorrow - probably will will not be much different than today  ???

Off for my ride - will see if I can provide any clues as to what is happening here later or tomorrow, after perusing your complete post and Best!

M


Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on September 30, 2022, 04:28:51 am
Updated the uniaud32-exp branch to the linux 5.19.12 code... https://smedley.id.au/tmp/uniaud32-linux-5.19.12-20220930.zip Pretty minor changes since 5.19.9 but always good to be on the latest code :)
Title: Re: Updated Uniaud32 build
Post by: Dariusz Piatkowski on September 30, 2022, 02:50:55 pm
Mark, everyone...

Heh Tom!

command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

Just put that unimix command in the file startup.cmd , and it will be run every time you start your system.

I was thinking about that after I posted, but haven't worked much with Startup.CMD in the past - will have a look at that tonight or tomorrow.

But thanks for the tip and Best!

Mark

Tom's suggestion is a great one, you can take that a step further and basically control the FULL behaviour of the audio card as far as the various individual controls go.

Basically, manually adjust the controls to what you want them to be first. By the sounds of it you already know how to pull up the different values and set them up.

Once done, you then save the complete set of settings to a file:

'unimix -savesomefilename'

The very next time you want to restore ALL of these settings you execute:

'unimix -loadsomefilename'

I have that load statement in my startup.cmd, which is a super easy way to re-set all the controls to a pre-defined set of values.

This is how you can handle this particular settings, as well as any others that can be adjusted by referencing their ID. In my case, I have some 53 controllable IDs ('unimix -names').
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 30, 2022, 07:32:04 pm
Paul, This driver release is working great. Just played an Eagles CD and it worked greatas far as sound goes. A little hang here and there on system sounds. But it doesn't crash the system and unlocks itself. Only when the mouse is involved.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on September 30, 2022, 10:09:37 pm
https://smedley.id.au/tmp/uniaud32-linux-5.19.9-20220924.zip

Hi Paul

I tested it (Release version) with OS/2 Warp 4.52 (14.106) guest on VBox 6.1.38 (Windows 10 21H2 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Works, Audio control working, no traps.

Regards
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 01, 2022, 04:11:37 am
Hey Mark!!

Wow! That's super interesting! I guess it's not firmware load after all - as switching between two AOS partitions could not make the firmware load.

Does it matter is you boot D: first then F: or f: first, then d:? What uniaud32 versions are in play on each partition? Just looking for clues..


I was knocked off the Internet today - more problems in the Maritimes obviously, so just replying tonight, my time.

In regards to this question, cold booting to D: then warm booting to F: produces sound on F:.

Cold booting to F; produces sound - Warm booting to D: from there, then back to F: produces no sound on F:


The patch you provided was interesting - but that code is from linux 4.18 - we're using linux 5.19.x code which would already include that code (or a later iteration of it).


Perhaps I should have a look at that and test it, if its in your archive - I doubt it will tell us anything, but it never hurts to see if it was committed or not

Perhaps a log file from the 1st cold boot, then the 2nd warm boot where sound works. I'm not confident it will show anything, but who knows?

If your talking about cold boot D: then F:, then cold boot f: to D: and back that can be done.

I'm also going to review my patches to patch_ca0132.c - ie from https://github.com/psmedley/uniaud32/commit/12e7ce146924b3e77056e75095e7b0a9a7571399 (I did an import of the netlabs SVN into github - it's easier to review code via the github gui, and it's also a backup)


Perhaps the  rather long patch (over 3000 lines was not incorporated in full - I'm not sure if you can ascertain that, Sir.


With the surround speakers - I'm not surprised you have issues controlling volume - take a look at https://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-exp/lib32/soundmixer.c#L398 and note the 'todo' comment. Adjusting the rear speakers is actually adjusting the overall mixer volume. I noticed this the other day looking at where I might be able to manipulate the 'auto mute' setting.


Based on what I said here in March I did test it again - and the section in PMUnimix (Master Volume Control) was present, but did NOT change the sound volume - I guess you could say its seen but not heard.

So based on that it may be a red herring (yet more fish analogies) as I don't think anything is being vetted through PMUnimix - it just looks that way ........

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 01, 2022, 10:00:30 am
hmmmmm interesting....

So cold boot to F: works..... what uniaud32 version is running there? and I assume D: is running the latest uniaud32....

How about a testlog from a cold boot on D: and a testlog from a cold boot on F:

If it used to work from a cold boot - surely we can fix it :)
Title: Re: Updated Uniaud32 build
Post by: OS4User on October 01, 2022, 10:58:49 am
https://smedley.id.au/tmp/uniaud32-linux-5.19.12-20220930.zip

it works ok for me
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 01, 2022, 08:10:43 pm
Heh Paul!

hmmmmm interesting....

So cold boot to F: works..... what uniaud32 version is running there? and I assume D: is running the latest uniaud32....

How about a testlog from a cold boot on D: and a testlog from a cold boot on F:

If it used to work from a cold boot - surely we can fix it :)

All partitions are now running 5.19.12 20220930, which I tested today:

Manticore II

AD1988B - sound perfect
Recon 3D (ca0132) - sound perfect - both cold boot to F: directly, and warm boot from D: to F:

Hyperfire

ALC982 - sound perfect, as long as I disable auto-mute (Haven't got around to creating a command file yet
Xi-Fi Titanium - still no sound although 6 sound sections are present

I've attached two testlogs below:

The first is a testlog cold booting F: with the Recon3D
The second is a testlog of F:, cold booting to D: first, then doing a warm boot from there to F:

Both methods produce sound on the Recon 3D.

I'm not sure that providing a testlog of cold booting to D: will tell you much - while running the same version of UniAud32, that partition is running the latest version. with the AD1988 chipset.

Will edit this and attach that cold boot D: testlog a little later!
@Edit - Testlog Cold Boot D: now attached

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 02, 2022, 12:45:20 am
Heh Eugene!

Paul, This driver release is working great. Just played an Eagles CD and it worked greatas far as sound goes. A little hang here and there on system sounds. But it doesn't crash the system and unlocks itself. Only when the mouse is involved.

Heard that Tropical Depression Ian was roaming through Virginia and West Virginia today - not a hurricane but still can blow over a few things - like whole trees and houses - and lost a friend as a result.

Hopefully not disrupting your life - just the emergency planner and responder in me and Best!

Mark
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 02, 2022, 01:02:19 am
Heh Darisuz

Tom's suggestion is a great one, you can take that a step further and basically control the FULL behaviour of the audio card as far as the various individual controls go.

Basically, manually adjust the controls to what you want them to be first. By the sounds of it you already know how to pull up the different values and set them up.

Once done, you then save the complete set of settings to a file:

'unimix -savesomefilename'

The very next time you want to restore ALL of these settings you execute:

'unimix -loadsomefilename'

I have that load statement in my startup.cmd, which is a super easy way to re-set all the controls to a pre-defined set of values.

This is how you can handle this particular settings, as well as any others that can be adjusted by referencing their ID. In my case, I have some 53 controllable IDs ('unimix -names').

Will have a tilt at your suggestions - as I've said, I'm not a programmer so that is a great suggestion for me.

At the moment I've been assessing why the Sound Blaster Xi-Fi Titanium is not producing sound - I noted that it also had an "auto-mute" entry but so far, disabling auto-mute has done nothing, but maybe somthing else may twig ...............

Drifting between issues like this and winterizing my fountains and water features and helping Anne with cleaning our windows on the house and in the yard - Oct. 1 and 22 C today and sunny in Collingwood Cove - but could be snowing in a week or two ...........

Best as always, Sir!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 02, 2022, 02:04:46 am
Tom's suggestion is a great one, you can take that a step further and basically control the FULL behaviour of the audio card as far as the various individual controls go.

Basically, manually adjust the controls to what you want them to be first. By the sounds of it you already know how to pull up the different values and set them up.

Once done, you then save the complete set of settings to a file:

'unimix -savesomefilename'

The very next time you want to restore ALL of these settings you execute:

'unimix -loadsomefilename'

I have that load statement in my startup.cmd, which is a super easy way to re-set all the controls to a pre-defined set of values.

This is how you can handle this particular settings, as well as any others that can be adjusted by referencing their ID. In my case, I have some 53 controllable IDs ('unimix -names').

Worked for me, Dariusz so thanks for the tip ...........

Just had to do one thing - at the command prompt used unimix.exe, then the switches:

ie: 'unimix.exe -savesomefilename' and

'unimix.exe -loadsomefilename' which I added to the STARTUP.CMD file I created in the F: root partition. (used a Command Line.CMD template and added that line). Saved it as STARTUP.CMD in the root partition.

Didn't need to do that previously, I think but whether you add "unimix" to the command line or "unimix.exe" it still got me to the same place - IOW a file that was saved, and a command file that loads those settings automatically.

Yet another item for Testing Guide v1.2 and thanks, Sir!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 03, 2022, 04:26:42 am
Hey All,

For those that follow linux kernel development, the 6.0 kernel was released today. So what's a guy to do on a Public Holiday? Update the uniaud32-exp code to the 6.0 kernel of course ;)

New builds at https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 03, 2022, 03:05:35 pm
Paul, Sytem sounds work well. But 123PM is broken.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 03, 2022, 04:31:55 pm
Heh Paul!

Hey All,

For those that follow linux kernel development, the 6.0 kernel was released today. So what's a guy to do on a Public Holiday? Update the uniaud32-exp code to the 6.0 kernel of course ;)

New builds at https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip

Cheers,

Paul

The Linus 6 UniAud32 is installed on both systems and:

Manticore II - AD1988B - sound perfect - both system and PM123 - all sound channels present in PMUnimix

Manticore II - SB Recon3D (ca0132) - sound perfect - both system and PM123 - still just shows one sound section in PMUnimix, which is non-functional, as per previous tests

Hyperfire - ALC892 - sound perfect - both system and PM123 - all channels present - auto-mute disabled directly from STARTUP.CMD

@Eugene - Heh Sir! Which PM123 version are you using? Tried it and worked on v1.37 and v1.41.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 03, 2022, 04:39:12 pm
Mark I am using the latest 1.43. on one computer it will not make any sounds and on the other it just quits after 26 seconds. No error messages and unimix -card is the same.PMuimix is fine too.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 03, 2022, 07:13:15 pm
Heh Eugene!

Mark I am using the latest 1.43. on one computer it will not make any sounds and on the other it just quits after 26 seconds. No error messages and unimix -card is the same.PMuimix is fine too.

Downloaded v1.43 and have been playing it for about 10 minutes no problem. (AC/DC Hells Bells - love that song!)

However reading the changes in v1.43 in the Warpin file I did note this:

- you can now mute audio via playback menu, hotkeys or a uPnP client.

I had a look at the playback menu and there is a Mute setting at the bottom. Ypu might want to see if for some reason it enabled itself (you know some programs now have a mind of their own <sic>)

Also you might want to have a look at unimix -names

If there is an id for "Auto-Mute" present for your sound card or chipset, you might want to try to disable that:

unimix -idx -val0=0 where "x" is the ID of the Auto-Mute switch - I'm assuming your sound adapter is 0.

The Auto-Mute issue might have crept in to affect other chipsets this time around with the Linux 6 tree.

As to sound cutting out - have not encountered it here with v1.43, at least so far, so not sure what may be a problem there.

Best!

M

 
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 03, 2022, 08:00:25 pm
Mark, thank you! I will try your suggestions.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 03, 2022, 08:10:36 pm
Mark, thank you! I will try your suggestions.

One thing I didn't mention previously, Eugene, is that when I had the problem, I could see PM123 running and operating correctly (sound bar moving across the PM123 display) - just had no sound.

If your hearing system sounds, then its probably not Auto-Mute as that would affect the system sounds as well - but that new mute function seems a likely candidate to at least try.

M
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 03, 2022, 08:15:07 pm
For comparison, there is also the latest mpg123, basically the same audio backend but simpler, maybe a better decoder. Cmdline.
Enable netlabs-exp, do yum install mpg123, disable netlabs-exp, mpg123 path\to\foo.mp3
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 03, 2022, 08:58:01 pm
Heh All!

Prior to going out in the yard to try to finalize my winterization (see cold weather on the horizon), I had time to complete v1.2 of the UniAud32 testing guide, which is attached below:

the major changes are:

- added a Table of Contents at the front
- clarified that UniAud16 files and lines do not need to be removed to run UniAud32
- Added a Workarounds section, compiling the Sound Volume fix suggested by Andy, the workaround related to Cold Boot as opposed to Warm Boot, and what we discovered with Auto-Mute being enabled by default and how to disable it (a special thanks to Tom and to Dariusz, related to both using and explaining how to add Unimix statements to STARTUP.CMD).

Any other suggestions for getting these drivers to work, or suggestions related to the guide always welcome and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 03, 2022, 10:29:17 pm
To all, I got PM123 working. Thanks again to Mark f9rt his suggestions.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 03, 2022, 11:24:52 pm
Heh Eugene!

To all, I got PM123 working. Thanks again to Mark f9rt his suggestions.

So what worked?

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 03, 2022, 11:45:51 pm
Mark, I am not sure it just started running again. I learned in IT to hate these things that fix themselves as they would have a tendency to return.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 04, 2022, 12:28:44 am
Mark, I am not sure it just started running again. I learned in IT to hate these things that fix themselves as they would have a tendency to return.

My systems have personalities ............. some days they just work, some days they work, under protest .......... and then they just don't want to work somedays ............. and are normal the next day.

In one of the Indian cycles called the Upanishads there is a passage which refers to the Goddess of Night, Ratri - "who knows the limits of night, or the moods of a woman" - I like to believe that can apply to computer systems as well .........

I've not been in IT but I have learned from 42 years of emergency response, some problems, and their solutions simply ........... A-R-E

M

As your desire is, so is your will.
As your will is, so is your deed.
As your deed is, so is your destiny.

-- Brihadaranyaka Upanishad (IV.4.5)
Title: Re: Updated Uniaud32 build
Post by: OS4User on October 04, 2022, 07:38:25 am
https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip

it works ok
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 05, 2022, 10:44:19 pm
Heh Paul!

Hope your having a fine week.

Temps today are into fall normals here (ie. cold) so had some time this morning to spend a little more time on having a look at the SB Recon3D.

Again thanks for giving us some access to your historic TRAC files.

I've narrowed down when it started to work as to within the 5.10.x UniAud32 series.

5.10.45 20210619 produces no sound and you get the window attached below, when accessing PMUnimix, which states:

uniaud_mixer_get_ctls_number:error $00000000

Testlog for this version is also attached

5.10.50 20210721, your next build, is where you get the Recon3D to work, with the current state we've observed (cold boot / warm boot from other partitions, a single Master Playback sound section, which does not control the mixer).

Testlog for this version also attached.

As the GIT changelog I originally attached said 4.18 RC1 was released in June 2018, and thats where the ca0132 changes were supposedly committed, I'm wondering whether that patch I attached previously was ever committed to the source files you used, for some strange reason - I noted that  a 4.17.19 build of UniAud32 and a 4.19.163 build were released - I tested both and got the same window as described about, moving upward until 5.10.50 - I suspect some other support function is causing it to function there - hopefully there were not too many commitments between 5.10.45 and 5.10.50.

Anyway, just thought I'd let you know what I discovered, as it was interesting and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 05, 2022, 10:57:56 pm
Heh Eugene!

To all, I got PM123 working. Thanks again to Mark f9rt his suggestions.

Hope your having a fine day.  8)

As you know I only started using PM123 v1.43 a couple of days ago - but I may have run into the condition you described, and it seems to be related to v1.43 (although I haven't tested it on v1.41 or v1.37 yet).

I was running the music player, and then opened the MahJongg game that came with ArcaOS. Switching to PM123 to load a new song, which started playing, then switching back to Mahjongg, the sound cut out - upon closing Mahjongg the sound returned, but then cut out after a few seconds.

I think something similar happened with Daves latest version of Seamonkey and PM123 when both were running - but it seems changing focus to another ArcaOS program and then back may trigger it.

Will have to test this a little more with other PM123 versions and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 06, 2022, 04:39:28 am
Mark, yep i have gone a few days with out it happening. but I bet it strikes again.
Title: Re: Updated Uniaud32 build
Post by: Doug Bissett on October 06, 2022, 07:55:29 am
Quote
I was running the music player, and then opened the MahJongg game that came with ArcaOS. Switching to PM123 to load a new song, which started playing, then switching back to Mahjongg, the sound cut out - upon closing Mahjongg the sound returned, but then cut out after a few seconds.

Mahjong, and Klondike solitaire, use sound to time the game, even when sound is turned off. Depending on the capabilities of the sound driver, that can, or might not, cause sound problems.  Usually, the "music" that you are listening to stops. When you close the game, the "music" should continue. It is very common for the sound to have trouble getting started again. That has been a serious flaw in UniAud, since day one.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 06, 2022, 10:51:43 am
Hey Doug,
Mahjong, and Klondike solitaire, use sound to time the game, even when sound is turned off. Depending on the capabilities of the sound driver, that can, or might not, cause sound problems.  Usually, the "music" that you are listening to stops. When you close the game, the "music" should continue. It is very common for the sound to have trouble getting started again. That has been a serious flaw in UniAud, since day one.

Agreed. Unfortunately the problem there is likely to lie in the uniaud16 code which is voodoo to me.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Lars on October 06, 2022, 01:19:27 pm
There are two aspects to this "sound stop / sound restart" problem:

1) if the sound driver and HW are capable of playing multiple sound streams concurrently: I do not know of UNIAUD but for the USB sound drivers I had to define a configuration and had to implement the runtime capability check correspondingly (there is an IOCTL to the audio driver to check how many streams it can play concurrently), so that only one stream can be played at any time because that is all that the USB audio HW supports. Consequently, if I play a sound file and switch to another one while the first one is still playing, the first will get interrupted until the second has been played to completion or aborted. It will then continue to play the first. At least for the USB sound drivers, that now works very well. The UNIAUD MMPM config claims it can handle up to 32 streams concurrently but the implementation of the sound driver (yes: very likely a part of UNIAUD16.SYS) might not properly report back its capabilities (if at all capable to play multiple streams concurrently).

2) "hiccup" of Chris Wohlgemuths audio classes: if you have these classes installed and if you double click on a sound file to have it play, if something goes wrong (for example, your default sound driver is set to USB driver but you have no USB audio device attached), you will not be able to have that sound file play again until you have done at least a WPS reset. That's a problem in how these audio classes integrate into the WPS but not with any sound driver as such. Easy enough to test: if you load that sound file into say PM123, it will play even if it failed before.

Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 06, 2022, 05:11:03 pm
Hey gentlemeen!

Hey Doug,
Mahjong, and Klondike solitaire, use sound to time the game, even when sound is turned off. Depending on the capabilities of the sound driver, that can, or might not, cause sound problems.  Usually, the "music" that you are listening to stops. When you close the game, the "music" should continue. It is very common for the sound to have trouble getting started again. That has been a serious flaw in UniAud, since day one.

Agreed. Unfortunately the problem there is likely to lie in the uniaud16 code which is voodoo to me.

Cheers,

Paul.

Interesting discussion.

Before reading this this morning I tried PM123 v1.41 last night and found the same issue, so as you say it may be endemic - I found it interesting mainly because it seemed to be a new problem based on Eugene's report, but perhaps it has been there all along, and I never noticed it.

These are the Glassman versions, and I seem to recall back in one of the change logs that he had put a fix in allowing concurrent sound to occur (ie. more than one application using sound at the same time - I have no idea how he would have done that, if it was even possible, and my dusty memory may be faulty).

I don't have sound enabled in either Mahjongg or Solitaire, so it didn't think it would be an issue - but if Doug is correct and its used as a timer in the apps, that would explain it.

To be thorough I will try v1.35 and v1.37 as well later today - taking my lovely wife out for breakfast at a new place for us - Headquarters Restaurant located in Sherwood Park, our nearest large centre, west of us and east of Edmonton.

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 06, 2022, 08:12:23 pm
I agree with Mark thank yo for the discussion. I will only expect so much for the uniaud32.sys.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 07, 2022, 04:19:34 am
Hey Lars!!
There are two aspects to this "sound stop / sound restart" problem:

1) if the sound driver and HW are capable of playing multiple sound streams concurrently: I do not know of UNIAUD but for the USB sound drivers I had to define a configuration and had to implement the runtime capability check correspondingly (there is an IOCTL to the audio driver to check how many streams it can play concurrently), so that only one stream can be played at any time because that is all that the USB audio HW supports. Consequently, if I play a sound file and switch to another one while the first one is still playing, the first will get interrupted until the second has been played to completion or aborted. It will then continue to play the first. At least for the USB sound drivers, that now works very well. The UNIAUD MMPM config claims it can handle up to 32 streams concurrently but the implementation of the sound driver (yes: very likely a part of UNIAUD16.SYS) might not properly report back its capabilities (if at all capable to play multiple streams concurrently).

My understanding with respect to Uniaud and multiple streams is that whilst uniaud16 in theory allows up to 32 streams; in practice; only the SB32 cards  using the emu10k1 code actually support multiple streams. All HDA hardware only supports a single stream.

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Lars on October 07, 2022, 09:41:40 am
It's a general shortcoming of the audio driver design that makes a "one size fits all" solution difficult to implement. The design started off in the 90's where an audio driver was created for exactly one specific audio device (honestly, there is a "device ID" that can take values "PAS16", "SOUND_BLASTER" etc., but also "MINIDD". The first two indicate what I said, the latter already shows that the concept was falling short of reality, it just stands for "mini device driver", whatever that means in MS talk, they obviously where implementing much of the MMPM subsytem).

So, for UNIAUD, the "largest common denominator" is to report back capability for 1 playback and 1 recording stream or for UNIAUD32.SYS to report back to UNIAUD16.SYS if the specific card supports more than one stream and have UNIAUD16.SYS report back the proper number during runtime (on the "IOCTL_Capability" call).

But if the specific HW/FW only supports 1 playback stream, then it is natural for the MMPM subsystem to interrupt a playing stream if another sound file is selected.

Thank god, I have never seen any USB audio HW that would be able to support more than 1 concurrent (playback) stream. I think, the USB audio spec does not even allow that.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 07, 2022, 05:15:53 pm
Heh Paul!


New builds at https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip

Cheers,

Paul

BTW, related to the SB Recon3D working starting at build 5.10.50, I was curious so I pulled the change log from Linux for kernel build 5.10.50 and skimmed through it.

Like every Linux change log, its a long document, but near the bottom I noted the ALSA changes - there were 13 committed of which 3 or 4 were USB audio related.

It included a number of changes related to the codecs for the ALC236 and 285 chipsets, and a number of general changes - nothing directly related to ca0132 boards.

So I suspect that the Recon3D working as of that build in ArcaOS is a bit of a fluke, which would explain why it shows up and works but no mixer support directly through Unimix. Reading up on it today, supposedly it uses a quad core sound processor, which I suspect is a much more complex technology to program for.

I hope to get around to seeing when the SB Xi-Fi Titanium started showing up sometime this month as well but not this weekend - Canadian Thanksgiving this Monday, and we're feasting with the stepson and girlfriend (Anne's son), and friends coming in from Saskatchewan (next province to the east) on Saturday.

Best of the weekend to you!

M
Title: Re: Updated Uniaud32 build
Post by: roberto on October 07, 2022, 09:02:39 pm
Hello
Yesterday testing this version I was very pleasantly surprised.
Version:
https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip


On the other hand, this same version of a PC at work was installed and when restarting it,
 he gave me this failure. I attach a photo. Return to the original version.In a Lenovo i3.
I think I put the release version.
Saludos
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 07, 2022, 11:01:01 pm
Hey Mark,
BTW, related to the SB Recon3D working starting at build 5.10.50, I was curious so I pulled the change log from Linux for kernel build 5.10.50 and skimmed through it.

Like every Linux change log, its a long document, but near the bottom I noted the ALSA changes - there were 13 committed of which 3 or 4 were USB audio related.

It included a number of changes related to the codecs for the ALC236 and 285 chipsets, and a number of general changes - nothing directly related to ca0132 boards.

So I suspect that the Recon3D working as of that build in ArcaOS is a bit of a fluke, which would explain why it shows up and works but no mixer support directly through Unimix. Reading up on it today, supposedly it uses a quad core sound processor, which I suspect is a much more complex technology to program for.

Much simpler explanation... See https://github.com/psmedley/uniaud32/commits/uniaud32-next?after=716c19a32ab8e49891508fdf0a92dcb769ad69d9+34&branch=uniaud32-next&qualified_name=refs%2Fheads%2Funiaud32-next

(this is a github mirror of the netlabs svn - the netlabs trac interface is a bit clunky). Simply put, the patch_ca0132 backend code was only compiled in to uniaud AFTER the update to 5.10.50.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on October 09, 2022, 05:08:17 am
New builds at https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip
Hi Paul

I tested it (Release version) with OS/2 Warp 4.52 (14.106) guest on VBox 6.1.38 (Windows 10 21H2 host).
- VM ICH AC97: Works, Audio control working, no traps.
- VM Intel HD Audio: Works, Audio control working, no traps.

Regards
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on October 09, 2022, 05:13:15 am
New builds at https://smedley.id.au/tmp/uniaud32-linux-6.0-20221003.zip
Hi Paul

Today I also tested it on the Thinkpad L420 (https://www.os2world.com/wiki/index.php/Lenovo_ThinkPad_L420).
- Works. No trap. Master volume working. Works good here.
Thanks
Title: Re: Updated Uniaud32 build
Post by: David McKenna on October 09, 2022, 11:35:10 pm
Hey Paul,

 A little late, but the 20221003 version 6 build works fine here...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 11, 2022, 04:36:37 am
Heh Paul!

Hey Mark,
Much simpler explanation... See https://github.com/psmedley/uniaud32/commits/uniaud32-next?after=716c19a32ab8e49891508fdf0a92dcb769ad69d9+34&branch=uniaud32-next&qualified_name=refs%2Fheads%2Funiaud32-next

(this is a github mirror of the netlabs svn - the netlabs trac interface is a bit clunky). Simply put, the patch_ca0132 backend code was only compiled in to uniaud AFTER the update to 5.10.50.

Cheers,

Paul

Interesting

Do you recall if anything other than the backend was ever committed after that?

Testing for the Recon3D has yielded the same results after that, up to the kernel 6 versions - works, but no mixer control and no surround sound.

Also, had some time tonight, after Thanksgiving festivities, to have a third look at the SB Xi-Fi Titanium, which still does not produce any sound.

I've been able to trace back to where the card was not recognized, and then the build where it was:

Up to build 5.9.16 20210104, I got the same window running PMUnimix as was attached previously, with the Recon3D - IOW it recognized none of the controls.

However, with build 5.10.10 20210124 , PMUnimix started reporting 6 sound sections - I've attached the testlogs for both 5.9.16 and 5.10.10 below.

Interestingly, the Xi-Fi is exhibiting much of the same behavior as with the Recon3D - running PM123, the sound bar is moving on the player and the displays look like something is playing, but no sound is produced. Having a look at unimix -names there is an auto-mute section present at -id11, but disabling it similar to the ALC892 chipset has not produced any effect. I have tried enabling various switches present (I assume the syntax would be -val0=1) but still no sound.

PMUnimix reports the 6 sound sections are on and at maximum value (shown as 103) but don't know if that can be taken as working.

The linux forums indicated a workaround, but haven't actually looked in detail - it is listed as a ca0106 board, so don't know if you committed anything specifically related to that type of card.

Plan to try a couple of other bright ideas with the Xi-Fi this week and Best!

M



Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 11, 2022, 07:31:25 am
Hey Mark,

Heh Paul!

Hey Mark,
Much simpler explanation... See https://github.com/psmedley/uniaud32/commits/uniaud32-next?after=716c19a32ab8e49891508fdf0a92dcb769ad69d9+34&branch=uniaud32-next&qualified_name=refs%2Fheads%2Funiaud32-next

(this is a github mirror of the netlabs svn - the netlabs trac interface is a bit clunky). Simply put, the patch_ca0132 backend code was only compiled in to uniaud AFTER the update to 5.10.50.

Cheers,

Paul

Interesting

Do you recall if anything other than the backend was ever committed after that?

Testing for the Recon3D has yielded the same results after that, up to the kernel 6 versions - works, but no mixer control and no surround sound.

Also, had some time tonight, after Thanksgiving festivities, to have a third look at the SB Xi-Fi Titanium, which still does not produce any sound.

I've been able to trace back to where the card was not recognized, and then the build where it was:

Up to build 5.9.16 20210104, I got the same window running PMUnimix as was attached previously, with the Recon3D - IOW it recognized none of the controls.

However, with build 5.10.10 20210124 , PMUnimix started reporting 6 sound sections - I've attached the testlogs for both 5.9.16 and 5.10.10 below.

Interestingly, the Xi-Fi is exhibiting much of the same behavior as with the Recon3D - running PM123, the sound bar is moving on the player and the displays look like something is playing, but no sound is produced. Having a look at unimix -names there is an auto-mute section present at -id11, but disabling it similar to the ALC892 chipset has not produced any effect. I have tried enabling various switches present (I assume the syntax would be -val0=1) but still no sound.

PMUnimix reports the 6 sound sections are on and at maximum value (shown as 103) but don't know if that can be taken as working.

The linux forums indicated a workaround, but haven't actually looked in detail - it is listed as a ca0106 board, so don't know if you committed anything specifically related to that type of card.

Plan to try a couple of other bright ideas with the Xi-Fi this week and Best!

Looks like I did work on patch_ca0132.c in Jan 2021 - https://github.com/psmedley/uniaud32/commit/029bdd758e302aeb481eaa118af3cd2cbeb81621 - however you reference ca0106.... I'll have to have a look at that code when I get a chance....
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 11, 2022, 08:52:01 pm
Heh Paul!


Looks like I did work on patch_ca0132.c in Jan 2021 - https://github.com/psmedley/uniaud32/commit/029bdd758e302aeb481eaa118af3cd2cbeb81621 - however you reference ca0106.... I'll have to have a look at that code when I get a chance....


Did a little more sniffing around today and not a lot of info related to the Xi-Fi Titanium on the Linux forums - a number of reports of no sound or no sound related to headphones, but not a lot of assistance for those problems.

I did note that the Xi Fi Titanium uses the EMU20K2 sound chipset and confirmed that's what is on the card I have installed, via PCI.exe. I've seen varying reports - most consensus is that it uses the Linux ca0106 driver, but there are so many varying versions of the Xi Fi its hard to say that definitively.

So I'm not sure what triggered it being recognized in build 5.10.10 20210104 - I hooked it up to the 5:1 surround sound, tried disabling auto-mute again and enabling Master Playback and trying to up the volume, just in case - no joy with any of the tests.

Anyway, a fine day to you and Best!
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 12, 2022, 02:11:39 am
Hey Mark,

Looks like I did work on patch_ca0132.c in Jan 2021 - https://github.com/psmedley/uniaud32/commit/029bdd758e302aeb481eaa118af3cd2cbeb81621 - however you reference ca0106.... I'll have to have a look at that code when I get a chance....


Did a little more sniffing around today and not a lot of info related to the Xi-Fi Titanium on the Linux forums - a number of reports of no sound or no sound related to headphones, but not a lot of assistance for those problems.

I did note that the Xi Fi Titanium uses the EMU20K2 sound chipset and confirmed that's what is on the card I have installed, via PCI.exe. I've seen varying reports - most consensus is that it uses the Linux ca0106 driver, but there are so many varying versions of the Xi Fi its hard to say that definitively.

So I'm not sure what triggered it being recognized in build 5.10.10 20210104 - I hooked it up to the 5:1 surround sound, tried disabling auto-mute again and enabling Master Playback and trying to up the volume, just in case - no joy with any of the tests.

Anyway, a fine day to you and Best!
Sorry, I'm a little confused by the status of the Xi Fi. When you say it was 'recognised' in build 5.10.10 20210104, do you mean it produced sound, or just starting being recognised by uniaud32. I agree that it should be the ca0106 driver for the Xi Fi.

I'm struggling to keep track of what's still not working on your systems, and am having trouble trawling through the forums trying to find the right logs. Can I ask you open a ticket at https://trac.netlabs.org/uniaud for any residual issues, and upload logs there?

Thanks in advance,

Paul.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 12, 2022, 03:56:21 am
Hi Paul!

Hey Mark,

Looks like I did work on patch_ca0132.c in Jan 2021 - https://github.com/psmedley/uniaud32/commit/029bdd758e302aeb481eaa118af3cd2cbeb81621 - however you reference ca0106.... I'll have to have a look at that code when I get a chance....


Did a little more sniffing around today and not a lot of info related to the Xi-Fi Titanium on the Linux forums - a number of reports of no sound or no sound related to headphones, but not a lot of assistance for those problems.

I did note that the Xi Fi Titanium uses the EMU20K2 sound chipset and confirmed that's what is on the card I have installed, via PCI.exe. I've seen varying reports - most consensus is that it uses the Linux ca0106 driver, but there are so many varying versions of the Xi Fi its hard to say that definitively.

So I'm not sure what triggered it being recognized in build 5.10.10 20210104 - I hooked it up to the 5:1 surround sound, tried disabling auto-mute again and enabling Master Playback and trying to up the volume, just in case - no joy with any of the tests.

Anyway, a fine day to you and Best!
Sorry, I'm a little confused by the status of the Xi Fi. When you say it was 'recognised' in build 5.10.10 20210104, do you mean it produced sound, or just starting being recognised by uniaud32. I agree that it should be the ca0106 driver for the Xi Fi.

I'm struggling to keep track of what's still not working on your systems, and am having trouble trawling through the forums trying to find the right logs. Can I ask you open a ticket at https://trac.netlabs.org/uniaud for any residual issues, and upload logs there?

Thanks in advance,

Paul.

As to SB Xi-Fi Titanium - no sound as of 5.10.10 20210104, but recognized and 6 sound sections present, and has been the same since then. Driver ca0106 and sound chip EMU20k2.

As to Recon3D - sound first occurred at 5.10.50, but no mixer, although 1 section (Master Volume Control) is reported in PMUnimix and no surround sound, The same up to the current version you have ported (linux Kernel 6). Driver ca0132.

I can file TRAC reports if you'd like for both and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 13, 2022, 11:05:49 pm
I can file TRAC reports if you'd like for both and Best!

Yes please
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 16, 2022, 04:35:36 pm
Heh Paul

I can file TRAC reports if you'd like for both and Best!

Yes please

It seems opening a ticket on TRAC is not an easy process.

Following your link, I got to the UniAud TRAC website, where it indicated I need a Netlabs TRAC login ID and password. I went to the webpage to do so, filled in the necessary information, and received the following bounce back:

fCMS v5.0 FATAL ERROR
Internal Error - Unhandled exception caught!
Message: Function name must be a string

Tried that three times with the same result.

It indicated that if I was having troubles acquiring that I should send an e-mail to the Netlabs Community Mailing List. Doing that I received the following:

--------------------

Hi. This is the qmail-send program at r2-d2.netlabs.org.
I'm afraid I wasn't able to deliver your message to the following addresses.
This is a permanent error; I've given up. Sorry it didn't work out.

<community@netlabs.org>:
Sorry, only subscribers may post. If you are a subscriber, please forward this message to community-owner@netlabs.org to get your new address included (#5.7.2)

-------------------

So I'm not sure if the webpage you provided the link for has not been updated for awhile (the latest build mentioned is rather old), or if TRAC administration for new users is not being maintained.

If you have a different link, let me know - it looks like if I cannot get a logon ID and password, I cannot open any tickets and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 16, 2022, 05:32:09 pm
Hi Mark, registering at Netlabs has been broken for a long time. Perhaps subscribe to the community mailing list and ask on there. community-subscribe (at) netlabs.org
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 16, 2022, 06:36:52 pm
Heh Dave!

Hi Mark, registering at Netlabs has been broken for a long time. Perhaps subscribe to the community mailing list and ask on there. community-subscribe (at) netlabs.org

Will try that out, hopefully a little later today.

Hows your fall going in BC? Finished winterizing here, now nearly time to get up some Christmas lights - I'm one of those ones who for some reason, just don't want to put up lights at -20 C ........ can't understand why ........ <sic> ;D

A fine day and week!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 16, 2022, 06:57:39 pm
Heh again Dave

Hi Mark, registering at Netlabs has been broken for a long time. Perhaps subscribe to the community mailing list and ask on there. community-subscribe (at) netlabs.org

Well the Bot there seems to have subscribed me me to the mailing list - a message was sent so now I wait to see what comes back.

Thanks for the tip and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 16, 2022, 07:29:23 pm
Hi Mark, once you post your request, Adrian will likely get back to you, they're on Swiss time so...
As for the weather, the weirdness continues, 28C yesterday, likely the same today if the smoke doesn't get too thick, normal is 14C with the record being 22C, so very unseasonable along with the drought, 10 mm of rain since July 8, normal is closer to 300 mm, the local reservoir is looking low with Hydro struggling to keep enough water running for the salmon, which are dying like crazy. If you ignore the dying trees and such, it has been a beautiful autumn. Am worried about when the rain shows up, especially after the floods last autumn, if it comes hard there is no way the ground can absorb it.
Enjoy the weather while it lasts. 
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 16, 2022, 09:24:51 pm
Hi Mark, once you post your request, Adrian will likely get back to you, they're on Swiss time so...
As for the weather, the weirdness continues, 28C yesterday, likely the same today if the smoke doesn't get too thick, normal is 14C with the record being 22C, so very unseasonable along with the drought, 10 mm of rain since July 8, normal is closer to 300 mm, the local reservoir is looking low with Hydro struggling to keep enough water running for the salmon, which are dying like crazy. If you ignore the dying trees and such, it has been a beautiful autumn. Am worried about when the rain shows up, especially after the floods last autumn, if it comes hard there is no way the ground can absorb it.
Enjoy the weather while it lasts.

We got your rain this year, at least up to August - 505 mm then the taps shut off which meant our tomatoes could finally develop in some hot weather - in fact some days the hottest on record in the dog days. Crops were great out here, and the grain is in the barn so to speak - early harvest - had time to even get a lot of wood chopped and corded, so Anne and I will have some lovely fires this winter.

In fact haven't had a hard frost yet, so even the Petunias and Snapdragons are still in bloom, mid-October - I suspect that will end next week.

But yeah - the Year of Strange continues, Sir - hope you guys and gals get a break soon and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 19, 2022, 04:12:55 am
HDMI Audio update:
As people may have seen, I bought a mini PC that it seems only have HDMI Audio. Currently, with the latest builds, the audio isn't detected. I'm slowly picking my way through the code to determine where things may be falling over, and if those issues can be addressed. Having working HDMI audio would be a good thing (IMNSHO).
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 19, 2022, 08:20:40 am
Wonder if DRM might block HDMI audio from ever working on our system?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 19, 2022, 08:57:09 am
Wonder if DRM might block HDMI audio from ever working on our system?

It might..... but I'll keep persisting for now ;)

Current errors:
Code: [Select]
trying to bring up adev

Looking for component u

find_component failed
master has incomplete components

couldn't bind with audio component

snd_hdac_i915_init err = -19

The component 'u' is a mismomer - due to the ported code using %zu which watcom doesn't support. I'll generate a fresh log tomorrow that might be more meaningful...
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 19, 2022, 05:39:44 pm
Heh Paul!

HDMI Audio update:
As people may have seen, I bought a mini PC that it seems only have HDMI Audio. Currently, with the latest builds, the audio isn't detected. I'm slowly picking my way through the code to determine where things may be falling over, and if those issues can be addressed. Having working HDMI audio would be a good thing (IMNSHO).

Trying some lateral thinking on this problem, so asking a couple of stupid questions of the non-programming variety (anyone can jump in):

1) Currently can we run anything connecting an HDMI cable to a monitor (an obvious example - can we run video that way under ArcaOS / eCS / OS/2 using the current video drivers, Panorama or otherwise)?

2) Is there any way in our OS to redirect audio output from the standard audio connections to the HDMI connection?

I ask because, while fishing around in the Linux forums a few weeks ago, a lot of the problems related to no sound and HDMI in various incarnations of Linux were related to redirecting sound output to the HDMI connection, to get sound to run.

An example is found in this link - https://itsfoss.com/how-to-fix-no-sound-through-hdmi-in-external-monitor-in-ubuntu/

I know that PMUnimix in both my test systems reports the audio chipsets on my video cards in PMUnimixDebug.txt as "On-board HDMI sound"  (one an ATI Radeon 5770, the other an ASUS Radeon HD 7850). I've connected one of them directly to my monitor (and its Bang and Olafson sound system) via an HDMI cable, but could not produce sound there through UniAud32, switching the audio to their designation (Card 1 - switch /a:1).

So it may be another case, Paul, where either we cannot "see" HDMI or perhaps cannot redirect to it. I hope I'm wrong, but perhaps someone else with programming experience could jump in to indicate if it is possible and how, or perhaps at least know what to look for, coding wise.

Most of the things I have read are related to using the Volume Control (PulseAudio) or Mixer programs to solve the problem in Linux, through redirection, something which may not be possible through Unimix.

May be a stupid question but relevant to the conversation and Best!

M
Title: Re: Updated Uniaud32 build
Post by: ivan on October 19, 2022, 11:44:03 pm
Hi Mark,

To answer your first question I have two ArcaOs, one OS/2 and one Linux Mint computers connected via HDMI cables to a 4Port KVM switch to my 28 inch 1920x1200 HDMI monitor - all work without problems.

Sound works with the Mint unit but I always get conflicts with the others - apparently too many sound systems on the motherboards and Ryzen processors.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 20, 2022, 12:40:44 am
Paul did state that on his new machine, only one HDMI port out of the 2 works.
I have had a couple of ATI video cards that have their own audio hardware to drive HDMI etc, whether that is true of builtin graphics or as hinted at in the link, one audio system with the capability to change the output port. I note here, it was only recently that I got control of the headphone port, older uniaud would see the output going to the speakers and headphone port unless Linux was booted first whereas newer drivers allow me to turn off the speakers and use headphones.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 20, 2022, 03:46:13 am
Heh Dave!

Paul did state that on his new machine, only one HDMI port out of the 2 works.
I have had a couple of ATI video cards that have their own audio hardware to drive HDMI etc, whether that is true of builtin graphics or as hinted at in the link, one audio system with the capability to change the output port. I note here, it was only recently that I got control of the headphone port, older uniaud would see the output going to the speakers and headphone port unless Linux was booted first whereas newer drivers allow me to turn off the speakers and use headphones.

My headphones, and 5:1 surround are also working in the latest builds, with the AD1988B chipset built into my M3N-HT Deluxe motherboard - haven't tried it with the ALC892 chipset on HYPERFIRE.

That's why I thought the possibility of redirection may be the problem, Dave - its seems that what was being talked about in that link and in other posts I read is getting the output to the correct port - only from what I'm getting its using a separate volume control or mixer program in Linux to do that (anybody know how that is accomplished? - addressing the kernel?, creating conf files? - I have no idea how programs like Linux PulseAudio or the various mixers work).

So what I'm suggesting is the HDMI problem here may not be fixable directly within the UniAud32 code.

But the question I'm asking is, if that fails, and redirection is the problem, can anything related to redirection to HDMI be accomplished in some other manner, if Paul finds that the UniAud code cannot be modified to accomplish that.

I was thinking, like the auto-mute problem we discovered, having the capability to change settings through something like PMUnimix, for auto-mute or other settings like output redirection would be useful - however PMUnimix development itself was terminated long ago, I think - and I assume the capabilities of Unimix.exe only extend so far.

But if, as Ivan stated, OS/2 video can be directed through HDMI, there must be a way to direct HDMI audio in a similar manner. Perhaps we need to ask the question as to how that was done in Panorama or otherwise.

Just me and my wacky lateral thinking, out loud, and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 20, 2022, 04:00:55 am
Heh again Dave!


Hi Mark, once you post your request, Adrian will likely get back to you, they're on Swiss time so...


Both Adrian and Steve Levine posted back when I wrote  - I've sent Adrian my preferred UserID as he requested - have not received anything back yet.

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 20, 2022, 06:02:37 pm
Heh Ivan

Hope your having a fine day!

Hi Mark,

Sound works with the Mint unit but I always get conflicts with the others - apparently too many sound systems on the motherboards and Ryzen processors.

I was intrigued by you getting sound "conflicts with the others" - I was curious as to what you meant by that.

As an example, on each of my ArcaOS test systems, I actually have three sound sources recognized - one on the  motherboard, one on the video card (HDMI) and a PCI Sound Blaster card plugged in. However, using the latest builds of UniAud32, only the default sound source is recognized by UniAud32 (it defaults to the sound source on the motherboard). To access either of the others, I have to add the /a:x switch to the statement in config.sys to redirect to the alternate sources (designated as Card 1 or Card 2), to try to run sound through them - have never run into a conflict, despite having all three installed.

As Paul mentioned, he can't get HDMI sound to work through HDMI - I have the same problem with the Radeon sources on the video cards - is the conflict related to having both an HDMI cable and another sound cable plugged into the same system, or switching between your Linux and OS/2 systems using the KVM?

I'm also using KVM's but a couple of older 8 port Belkins - I don't think I ever tried switching between an active Windows OS on one system and ArcaOS on the other (don't run Linux), to see what may happen sound wise, as I have a number of speaker systems (stereo / surround) plugged directly and separately into various systems.

As I've said before, I'm a hardware geek - my office looks like a Hydra most days to the despair of my wife, and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 22, 2022, 10:14:24 am
Hey All,

I backed out some of the HDMI code for now to avoid breaking things for others, and have built a new driver based on the linux 6.0.3 code. The binary is at https://smedley.id.au/tmp/uniaud32-linux-6.0.3-20221022.zip - code is in SVN.

I think we may be out of luck with HDMI audio for now - per my comment in the other thread, I was reading https://askubuntu.com/questions/694677/snd-hda-intel-0000001f-3-failed-to-add-i915-component-master-19 and it says "In order to support sound output over HDMI, video drivers for Linux register an "audio component" for the sound card driver (in my case Intel HD Audio a.k.a. snd_hda_intel) to connect to. Once this connection is made, the sound hardware can produce an audio stream for the GPU to multiplex onto the HDMI cable."

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 22, 2022, 05:36:45 pm
I can report that the drives are working with system sounds. Still seems to hang a bit with the mouse but could be related to Amouse PM123 sounds good to these ears, but has some quirks. In like it doesn't want to shut down right away but does.VLC2 does not work at all. these may not be attributal to your build of Uniaud.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on October 22, 2022, 06:54:44 pm
Hi Paul,

  6.0.3 working well here, thanks! Too bad about HDMI - would be great if that could be made to work. I have a Chuwi Larkbox (mini PC) that would be cool to put OS/2 on...

Regards,
Title: Re: Updated Uniaud32 build
Post by: Andy Willis on October 22, 2022, 10:54:12 pm
Does anyone know if Display Port sound works?
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on October 22, 2022, 11:01:21 pm
Does anyone know if Display Port sound works?

Pretty sure it's in the same boast as HDMI sound....
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 23, 2022, 12:41:11 am
Does anyone know if Display Port sound works?

Wiki says,
Quote
Audio data is transmitted across the main link during the video blanking intervals (short pauses between each line and frame of video data).
So going to need video adapter support.
I notice Linux Mint sees my Display Port as HDMI.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 24, 2022, 08:37:29 pm
Heh Paul!

Hey All,

I backed out some of the HDMI code for now to avoid breaking things for others, and have built a new driver based on the linux 6.0.3 code. The binary is at https://smedley.id.au/tmp/uniaud32-linux-6.0.3-20221022.zip - code is in SVN.

Cheers,

Paul

Busy weekend entertaining visitors from afar (at least from a different province)

Tested 6.0.3 and everything still relatively the same:

ALC892 chipset (Hyperfire) - Sound perfect up to 5.1 Surround Sound, as long as auto-mute is disabled.

Sound Blaster Xi-Fi (Hyperfire) - no sound but 6 sound sections still reported in PMUnimix. PM123 looks like it is playing and tried disabling auto-mute, but still no sound.

AD1988B chipset (Manticore II) - Sound perfect up to 5:1 Surround Sound - no workarounds or adjustments required.

Soundblaster Recon 3D (Manticore II) - still shows only one sound section under PMUnimix (Master Playback) and produces perfect Stereo sound only (2 speakers) - cannot use Surround Sound.

I still can't access TRAC to file tickets, as Adrian has not given me credentials yet, and the automatic registration system is broken for TRAC. Sent to the mailing list and privately last week - will try again this week.

Best!

M
 
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 24, 2022, 08:56:13 pm
Hey Again Paul!

\
Hey All,

I think we may be out of luck with HDMI audio for now - per my comment in the other thread, I was reading https://askubuntu.com/questions/694677/snd-hda-intel-0000001f-3-failed-to-add-i915-component-master-19 and it says "In order to support sound output over HDMI, video drivers for Linux register an "audio component" for the sound card driver (in my case Intel HD Audio a.k.a. snd_hda_intel) to connect to. Once this connection is made, the sound hardware can produce an audio stream for the GPU to multiplex onto the HDMI cable."

Cheers,

Paul

While testing build 6.0.3, I tried addressing the HDMI card on both test systems and received the same results, as March - a PMUnimix mixer screen is displayed, but no sound sections present.

PMUnimixDebug.txt does show the card info for both, but no other BuildfForm information for either.

That obviously would make sense if the Audio Component of HDMI is missing.

If ArcaOS video drivers can address HDMI currently (at least Ivan reported he could), is it possible to add it there? Or through USB? (don't know how or if HDMI and USB have some synergies) - just asking another dumb lateral thinking question) and Best!

M
Title: Re: Updated Uniaud32 build
Post by: ivan on October 25, 2022, 05:42:50 pm
Hi Mark,

Further my statement that HDMI works without problems on all 8 of the computers I have tried it on maybe I should have stated that the video chip is part of the main processor (all Ryzen 3 variants with built in video).  I don't know if that changes anything but I haven't had a video card in a computer since AMD brought out their Vega graphics on the processor.
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on October 25, 2022, 10:01:05 pm
Hi Mark,

Further my statement that HDMI works without problems on all 8 of the computers I have tried it on maybe I should have stated that the video chip is part of the main processor (all Ryzen 3 variants with built in video).  I don't know if that changes anything but I haven't had a video card in a computer since AMD brought out their Vega graphics on the processor.

Hi Ivan, sorry but I got lost on the conversation and I want to understand it right.

Is the Audio output of the HDMI working on the 8 computers you tried? Since this is the UniAud32 thread I think we are discussion the audio output on HDMI with this driver, right?
If you are experiencing Audio output on the HDMI with ArcaOS and UniAud, what do you use in your settings?

Regards
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 26, 2022, 02:09:23 am
Heh Martin!

Hi Ivan, sorry but I got lost on the conversation and I want to understand it right.

Is the Audio output of the HDMI working on the 8 computers you tried? Since this is the UniAud32 thread I think we are discussion the audio output on HDMI with this driver, right?
If you are experiencing Audio output on the HDMI with ArcaOS and UniAud, what do you use in your settings?

Regards

The question that Ivan was answering about video may have some bearing on whether HDMI audio is possible, so may not really be off topic.

I asked the question as to whether we can use HDMI at all under ArcaOS.

As it is used mostly for video first in all systems, having HDMI video working for at least one person shows that transmitting using an HDMI cable is at least possible under ArcaOS. As Paul mentioned a few posts back, the ability to transmit the audio component may be missing at the moment under ArcaOS for whatever reason.

I know HDMI video is very high resolution, but does that also apply to the audio fidelity as well? Or is the audio the same as through the normal cables and speakers, and just piggybacks in the HDMI cable for convenience? I don't believe Ivan was saying he could access audio via HDMI, but I will be interested in his answer.

I'm also curious as to how that is implemented USB wise, as USB for years has functioned universally to transmit a variety of signals and functions. As it seems transmitting both video and audio is done by the same cable in HDMI, I was wondering whether it is a matter of programming, and if so, where should the code reside. I will be interested in seeing what Lars is up to in the USB Audio thread (which I have nothing to participate with - no USB Audio based equipment here at the moment)

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 26, 2022, 02:13:27 am
Hi Ivan!

Hi Mark,

Further my statement that HDMI works without problems on all 8 of the computers I have tried it on maybe I should have stated that the video chip is part of the main processor (all Ryzen 3 variants with built in video).  I don't know if that changes anything but I haven't had a video card in a computer since AMD brought out their Vega graphics on the processor.

That's interesting

Most of the ASUS motherboards I've bought have on-board video, but not on-board the processor - as most are Radeon variants I probably don't need a video card either, but I still use them.

Like the audio functions, I probably could deep six most of my other cards - too old fashion in my thinking I guess.

M
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on October 26, 2022, 02:45:37 am
Mark, Some have claimed the onboard video has better performance than the video cards. Especially with the display port used.
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on October 26, 2022, 03:49:01 am
For HDMI video to work with our drivers likely depends on the hardware/firmware and seems to usually work though Paul reported only one of his 2 HDMI ports worked for him.
Display Port is similar to HDMI and only needs a passive connector to convert.
As the audio is carried on the same wires as the video. On Display Port it is inserted into the video blanking part of the signal. Video blanking on a CRT was when the electron beam shut off and went from the bottom of the screen to the top to draw the next frame, so a brief pause in video every frame. Games and such (our Firefox is supposed to but doesn't) also use that period for doing calculations etc.
USB C has an alternative mode which amongst other things, can act like a Display Port cable, this is how monitors are hooked up via USB 3+, so not really USB, but rather Display Port (or HDMI) transmitted over the USB C cable. Sound would work the same as normal Display Port and likely HDMI, multiplexed into the video signal, so the video hardware/sofware needs to be involved. This is different from USB audio that Lars works on by my understanding.
HDMI audio seems quite diverse according to wiki, it is all optional but to quote wiki, https://en.wikipedia.org/wiki/HDMI (https://en.wikipedia.org/wiki/HDMI)
Quote
For digital audio, if an HDMI device has audio, it is required to implement the baseline format: stereo (uncompressed) PCM. Other formats are optional, with HDMI allowing up to 8 channels of uncompressed audio at sample sizes of 16 bits, 20 bits, or 24 bits, with sample rates of 32 kHz, 44.1 kHz, 48 kHz, 88.2 kHz, 96 kHz, 176.4 kHz, or 192 kHz.[5]: §7  HDMI also carries any IEC 61937-compliant compressed audio stream, such as Dolby Digital and DTS, and up to 8 channels of one-bit DSD audio (used on Super Audio CDs) at rates up to four times that of Super Audio CD.[5]: §7  With version 1.3, HDMI allows lossless compressed audio streams Dolby TrueHD and DTS-HD Master Audio.[5]: §7  As with the Y′CBCR video, audio capability is optional. Audio return channel (ARC) is a feature introduced in the HDMI 1.4 standard.[41] "Return" refers to the case where the audio comes from the TV and can be sent "upstream" to the AV receiver using the HDMI cable connected to the AV receiver.[41] An example given on the HDMI website is that a TV that directly receives a terrestrial/satellite broadcast, or has a video source built in, sends the audio "upstream" to the AV receiver.[41]

Arca Noae does have updated video drivers (Intel at first) on their roadmap, when they find time I guess. At that point they may well add audio support to HDMI and Display Port.
BTW, my 1900x1200 monitor looks much better then VGA (mostly sharper, especially full screen text like during boot) using the Display Port and a DVI converter to hook into the DVI monitor plug. It is slightly broken compared to VGA, X doesn't see the VGA bios and while the screensaver can turn off my monitor, it won't turn back on except by rebooting.
Title: Re: Updated Uniaud32 build
Post by: Andi B. on October 26, 2022, 05:17:42 pm
Quote
Display Port is similar to HDMI and only needs a passive connector to convert.
You mixed up DVI with DP (Display Port). DP is quite different compared to DVI and HDMI. HDMI uses the same electrical signals as DVI has used before. So a simple passive converter can be used to interchange them. For video signals that is. DVI does not support audio officially.

For DP to HDMI the the graphic chip/motherboard has to detect a external adapter and switch different (HDMI) output signals to the DP connector. Not all motherboard can do this. Don't count on this special function. It's not mandatory.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 26, 2022, 06:52:33 pm
Heh Andy!

Quote
Display Port is similar to HDMI and only needs a passive connector to convert.
You mixed up DVI with DP (Display Port). DP is quite different compared to DVI and HDMI. HDMI uses the same electrical signals as DVI has used before. So a simple passive converter can be used to interchange them. For video signals that is. DVI does not support audio officially.

For DP to HDMI the the graphic chip/motherboard has to detect a external adapter and switch different (HDMI) output signals to the DP connector. Not all motherboard can do this. Don't count on this special function. It's not mandatory.

As Martin would say, perhaps we're getting a little off topic as the discussion was really about audio and HDMI, given what Paul is trying to accomplish in adding HDMI audio support in UniAud32.

I was curious as to what you think may need to be accomplished to access audio through HDMI - I know Arca Noae has a version of SNAP that needs updating on the bucket list and of course Panorama - do you think either of these drivers could be updated to interleave audio support through HDMI, as it seems is being done in Linux. Or would there need to be a separate solution, using a different approach? Or is it not really possible in ArcaOS, through video drivers or otherwise?

The discussion seems to have turned to the fine points of video design (which I find interesting being a hardware geek), but its because one cable is being used to carry all the signals, similar I assume to USB design. IOW HDMI video and associated audio, if we want both delivered, seem to be inextricably linked.

I'm not a programmer, but I have been a senior project manager, so I'm taking that approach to this discussion - it seems from what Ivan said that he can access HDMI video signals so by corollary I'm assuming that it should be possible to add audio support as well through HDMI. I think the discussion is really about how, assuming it is possible.

Best!

M
Title: Re: Updated Uniaud32 build
Post by: Andi B. on October 27, 2022, 09:45:21 am
Quote
I was curious as to what you think may need to be accomplished to access audio through HDMI - I know Arca Noae has a version of SNAP that needs updating on the bucket list and of course Panorama - do you think either of these drivers could be updated to interleave audio support through HDMI,
I don't know how the Linux guys do it but given the system design thinking about the basics. The system design is as following
- audio card/chip is digitally connected via PCI(e) to the CPU and the output analog audio goes straight to the attached connectors (yes there's also a digital stream to S/PDIF but that's a completely different thing than digital PCI(e)
- video card/chip is connected via PCI(e) to the CPU and the outputs of these chips go straight to there associated connectors

Video chip is controlled by the video driver, audio chip by the audio driver (which have to work together with the multimedia subsystem). Now you ask for a cooperative video driver and a cooperative audio driver. Audio drivers should send audio data to the video chip at the same the the video driver operates it. Both data streams have to be synchronized and proper mixed together in the video chip (to be more precise feed into the video output DVI/HDMI LVDS driver which today most probably is integrated into the video chip).

I'm mostly a hardware designer and system designer but did professional software development too. I'm not the Linux expert and not the in dept expert of PC hardware I was years ago. But to me this all looks way to complex for us (OS/2 developers) to spent a lot of thinking about. I know some things about our video drivers and think David has a lot of more important things to do than thinking about the quirks of mixing some audio stream into some video chip. I'm pretty sure there is NIL BIOS support for that so I see no chance to make such thing happen with Panorama. In theory with SNAP and a special video chip driver there would be more options. But if you think about the huge problems to improve only small pieces in SNAP, it's pretty clear that even if someone could reanimate SNAP there would be tons of other more important problems to solve before.

Btw. personally I circumvent these crappy consumer interface HDMI as much as I can. I participated the invention of this interface during my active career in consumer electronics. I know the reason why it was made and for which use cases it is good for. And so for the few times I use my PC for audio reproduction I prefer my AV-Receiver connected via S/PDIF with the attached big speakers over any monitor built in lowest end speaker. So take my thoughts above as strongly biased ;-)
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on October 27, 2022, 08:35:17 pm
Heh Andy!

Btw. personally I circumvent these crappy consumer interface HDMI as much as I can. I participated the invention of this interface during my active career in consumer electronics. I know the reason why it was made and for which use cases it is good for. And so for the few times I use my PC for audio reproduction I prefer my AV-Receiver connected via S/PDIF with the attached big speakers over any monitor built in lowest end speaker. So take my thoughts above as strongly biased ;-)

A most complete and eloquent statement of the facts, Sir!

I don't see a lot of synergies that we can draw on here, considering that most of the developments we've drawn upon are based on a Linux base which is being actively developed in a different direction, and from which we are attempting to adopt and adapt a patchwork of fixes for OS/2 / eCS / ArcaOS which has been ported by a lot of very talented people ...... but they are not miracle workers.  :P

If someone comes up with a solution, I can now see it will be very interesting to observe - but I'm not holding my breath. And having a strong but informed bias is merely stating the facts that you know - having used this OS for going on over 30 years and running my own company for 28 years, its a trait I appreciate!  8)

Best and Thanks!

M
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 06, 2022, 05:08:45 am
Not major code changes, but I updated the uniaud32 code to 6.0.7 - https://smedley.id.au/tmp/uniaud32-linux-6.0.7-20221106.zip
Title: Re: Updated Uniaud32 build
Post by: OS4User on November 06, 2022, 09:50:43 am
Not major code changes, but I updated the uniaud32 code to 6.0.7 - https://smedley.id.au/tmp/uniaud32-linux-6.0.7-20221106.zip

it works ok for me
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 06, 2022, 02:18:02 pm
Paul this is the best drop yet. Sounds very good. No new issues.
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on November 08, 2022, 04:32:24 am
Heh Paul!

Not major code changes, but I updated the uniaud32 code to 6.0.7 - https://smedley.id.au/tmp/uniaud32-linux-6.0.7-20221106.zip

Nothing broke here with the new release, as expected, Sir.

The various quirks with the Soundblaster cards remain the same - I've given up for the moment on filing tickets, as the automated system for signing up to gaining access to TRAC is broken and I've had no follow-up from Adrian after three weeks.

Hope your having a more pleasant start to the week - have had 50 cm of snow since November 2, high winds and wind chills, and expecting lows of -28 C overnight. So much for sunny, bare ground and 14 C on Halloween, only 7 days ago ...........  ???

Best!

M

Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 08, 2022, 09:35:43 am
Hi Mark!

That weather sounds hideous! We've just had our first tast of summer, with a couple of days above 30C - it's been glorious!

Sorry about the issues with Trac - I've now created a Uniaud32 project at my own bug tracker - https://mantis.smedley.id.au - please raise issues there.

Cheers,

Paul
Title: Re: Updated Uniaud32 build
Post by: Mark Szkolnicki on November 08, 2022, 11:57:05 pm
Hi Mark!

That weather sounds hideous! We've just had our first tast of summer, with a couple of days above 30C - it's been glorious!

Sorry about the issues with Trac - I've now created a Uniaud32 project at my own bug tracker - https://mantis.smedley.id.au - please raise issues there.

Cheers,

Paul

Thats normal January weather here, Paul ........... only this is beginning of November  ???

Thanks for the link between being out for supplies, lunch and more shovelling, no time today - will have a look tomorrow and Best!

M
Title: Re: Updated Uniaud32 build
Post by: Remy on November 09, 2022, 03:27:24 pm
Hi Mark!

That weather sounds hideous! We've just had our first tast of summer, with a couple of days above 30C - it's been glorious!

Sorry about the issues with Trac - I've now created a Uniaud32 project at my own bug tracker - https://mantis.smedley.id.au - please raise issues there.

Cheers,

Paul

Thats normal January weather here, Paul ........... only this is beginning of November  ???

Thanks for the link between being out for supplies, lunch and more shovelling, no time today - will have a look tomorrow and Best!

M

Out of topic: link to extreme weather https://electroverse.co/
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 17, 2022, 08:46:28 am
New Uniaud32 - based on linux 6.1 code - https://smedley.id.au/tmp/uniaud32-linux-6.1-20221217.zip
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on December 17, 2022, 02:13:29 pm
New Uniaud32 - based on linux 6.1 code - https://smedley.id.au/tmp/uniaud32-linux-6.1-20221217.zip

Thanks Paul

I tested it (Release version) with ArcaOS 5.0.7 guest on VBox 7.0.4  (Windows 10 22H2 host).
- VM ICH AC97: Works, Audio control working, no traps. I have an issue that the first time I play a wav, I hear it loud (100%) and the next times I play it I hear it lower (Volume setting - 40%). If I drag a file to force the drag sound it sound 100% again, I play another wav  and it is 100% again and the next times respects the volume control (40%). I reverted back to 3.01.03 (5.14.7) and got the same issue.
- VM Intel HD Audio: no traps. Does not produce audio. I reverted back to 3.01.03 (5.14.7) and didn't get sound. Maybe it was something changed on VirtualBox that I didn't noticed before.

Regards
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on December 17, 2022, 09:52:58 pm
Thanks Martin - HDA sound has always been temperamental for me in Virtualbox. I just tested on a baremetal T470s and sound is working fine for me.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on December 17, 2022, 11:35:16 pm
 Thanks Paul, this new 6.1 version works well here:

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7220000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 11, 2023, 11:22:45 pm
Hi All,

Longest time between builds for a while :/

I updated the source to linux 6.1.18 this morning - https://smedley.id.au/tmp/uniaud32-linux-6.1.18-20230312.zip

Minimally tested under virtualbox (ac97 & hda) and with HDA on a Thinkpad T470s.

Code will be in netlabs SVN shortly, for anyone that wants to build (in the uniaud32-dev branch)

YMMV :)

Cheers,

Paul.
Title: Re: Updated Uniaud32 build
Post by: David McKenna on March 11, 2023, 11:36:56 pm
 Thanks Paul, working well here on my new MoBo...

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xa9230000 irq 16
  mixer: Realtek ALC892
  componenets: HDA:10ec0892,1458a182,00100302 HDA:8086280b,80860101,00100000

Regards,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 11, 2023, 11:37:52 pm
Thanks David, that was quick!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on March 11, 2023, 11:39:18 pm
 Just happened to be online here when you posted... :-)
Title: Re: Updated Uniaud32 build
Post by: Mike Kölling on March 12, 2023, 12:46:04 pm
I updated the source to linux 6.1.18 this morning - https://smedley.id.au/tmp/uniaud32-linux-6.1.18-20230312.zip

Hi Paul,

it's working fine on my TP 430s. Thanks.

Greetings from Potsdam / Germany,
Mike

Edit:
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf2530000 irq 22
  mixer: Realtek ALC269VC
  componenets: HDA:10ec0269,17aa21fb,00100203 HDA:80862806,80860101,00100000
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on March 12, 2023, 03:16:46 pm
I updated the source to linux 6.1.18 this morning - https://smedley.id.au/tmp/uniaud32-linux-6.1.18-20230312.zip

Thanks Paul

I tested it (Release version) with ArcaOS 5.0.7 guest on VBox 7.0.6  (Windows 10 22H2 host).
- VM ICH AC97: Works, Audio control working, no traps. I still have an issue that the first time I play a wav, I hear it loud (100%) and the next times I play it I hear it lower (Volume setting - 40%). If I drag a file to force the drag sound it sound 100% again, I play another wav  and it is 100% again and the next times respects the volume control (40%).
- VM Intel HD Audio: no traps. Does not produce audio.

Regards
Title: Re: Updated Uniaud32 build
Post by: mauro on March 19, 2023, 08:46:52 am
Hi, occasionally follow this thread of Pauls's Uniaud updates work which I read is giving positive results in most cases.
In my case not exactly, meaning that my last attempts to replace SB16 drivers with Uniaud ones, system sounds ok but no sound in multimedia streaming files and randomly no sound in multimedia local files. SB16 at least ok in both system sound and multimedia local files.
This thread makes me believe that it depends on some wrong configuration I'm not understanding, also based on the fact that I'm not that smart as many of you, but much encouraged by the personal progresses reached just after joined this forum years ago.
Said this, I think that I will dedicate a new virtual machine to such task in Windows.
I actually run Warp 4.52 Vm VirtualBox both in Windows host, were I'm keeping 1 main guest and other 3 dedicated to different testings which if found positive/stable then are implemented on the main one,  and Ubuntu 22.04 host which receives the last stable updates from the guest machine in Windows.
Then I will ask a bit of basic support here, thinking that will begin whith the same issues above described.
Thank you
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on March 19, 2023, 09:41:38 am
Thanks for testing, but SB16 isn't really the target hardware here - the focus is on modern HDA hardware. If SB16 works, great, but given there would be so few people still using that hardware, I have no real interest in investigating the issues.
Title: Re: Updated Uniaud32 build
Post by: mauro on March 19, 2023, 04:03:13 pm
If SB16 works, great, but given there would be so few people still using that hardware, I have no real interest in investigating the issues.

well, not hardware, just emulation.  The VM actually works as explained with SB16 audio filter configuration/drivers, but if I could swhitch to Uniaud fully working would be best, my target.  Believe I'm missing some configuration klowledge.
Tks
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on March 24, 2023, 03:21:25 am
Well I can report that your March 11 drivers work fine. Now that I have my no sound issue worked out. Something was wrong with the MMPM.ini. So it was not the drivers fault. The sound card is on an ASRock B450M-HDV R4.0.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on April 14, 2023, 03:22:46 am
Hey All,

Some news - firstly I've imported the Uniaud source into the Netlabs git repository at https://github.com/netlabsorg/uniaud16 and https://github.com/netlabsorg/uniaud32 respectively.

Note the uniaud16 repo is marked private as it is only available to people who had/hold an IBM DDK license. When PM me if you have that license and want/need access to the source.

Under the uniaud32 repo, I have created two branches:
- this contains the latest 6.1.x (current Long Term Support version) code - currently 6.1.24
Current binaries are:
https://smedley.id.au/tmp/uniaud32-linux-6.1.24-20230414.zip
https://smedley.id.au/tmp/uniaud32-linux-6.2.12-20230414.zip
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on April 14, 2023, 09:00:00 pm
And the 6.2.12 drivers work well just like tha last drop! Clean and ful sound.Thanks Paul!
Title: Re: Updated Uniaud32 build
Post by: David McKenna on April 14, 2023, 09:47:56 pm
 Thanks Paul.. both are working well here...

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xa9230000 irq 16
  mixer: Realtek ALC892
  componenets: HDA:10ec0892,1458a182,00100302 HDA:8086280b,80860101,00100000
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 21, 2023, 03:44:19 pm
Zhould the attached card unimix.txt be working?
Title: Re: Updated Uniaud32 build
Post by: Pete on September 22, 2023, 06:15:26 am
Hi Eugene

Possibly...

Try this command in an os/2 window opened in \mmos2:-

unimix - card > unilist.txt
The above will create the file unilist.txt and the contents should tell you if there is more than 1 audio device present eg:-

Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 2 audio adapter(s)
Error: invalid number of PCM instances for adapter 1
Control list for adapter 0


If there is only 1 audio adapter then the rest of the file contains the Control IDs and their settings. Worth checking Master Playback is on and Master Playback Volume is at a reasonable setting.

If there are 2 adapters then it could be worth using the /A:1 switch with uniaud.sys
 

The below uniaud and unimix settings may come in handy.


Regards

Pete


UNIAUD32.SYS
The following options are supported in CONFIG.SYS:

DEVICE=x:\MMOS2\UNIAUD32.SYS /V /C:CARDNAME
  - /A:n - Use audio adapter n where the first adapter is 0. (Really skip n adapters).
  - /V - Verbose
  - /C:DRIVER - Force audio hardware to *only* use the specified driver.
        Where DRIVER is:
      - SBLIVE      : Sound Blaster Live!
      - CS4281      : Cirrus Logic CS4281
      - ALS4000     : Avance Logic ALS4000
      - CMEDIA      : C-Media CMI8738, CMI8338
      - ICH         : Intel ICHx, NFORCE1/2/3
      - CS46XX      : Cirrus Logic CS4280, CS46XX
      - VIA82XX     : VIA VT82C686A/B/C, VT8233A/C, VT8235
      - ESS1938     : ESS Solo-1 (ESS 1938)
      - MAESTRO     : ESS Maestro1/2/2E, TerraTec DMX
      - MAESTRO3    : ESS Maestro3/Allegro/Canyon3D-2
      - YAMAHA      : Yamaha YMF 724/724F/740/740C/744/754
      - ENSONIQ     : Ensoniq AudioPCI ES1371/73, Creative Labs SB PCI64/128
      - ALI5451     : ALI 5451
      - TRIDENT     : Trident 4D Wave, SI7018
Default: Fully automatic operation. Use of switch is not recommended.

DEVICE=x:\MMOS2\UNIAUD16.SYS /V /C /M /L
  - /V - Verbose
  - /C - Enable CD output
  - /M - Enable Microphone output
  - /L - Enable Line-In output
Default: /C ((verbose messages; CD output enabled)


UNIMIX.EXE
Command line mixer for UNIAUD32. Version 1.1
Usage:
 -dev<num>       - card to work (default=0)
 -id<num>        - use control num (for list or set value). Begins from 1
 -cnt<num>       - set value for count number <num> in control. Begins from 0
 -val<num>       - set value for control
 -card           - show card info
 -pcms           - show PCM instances info
 -list           - full (id,name,bounds,value) control(s) list
 -names          - just control(s) name(s) and ID
 -get            - get value switch. uses with -id and/or -cnt
 -powerget       - get power state
 -powerset<num>  - set power state
 -save<file>     - save dump of all values of all controls to file
 -load<file>     - load dump from file to all values of all controls

All arguments are processed in order. -dev, -id, and -cnt simply set a value
that is retained until changed. All other arguments perform a command.
Example: unimix -id5 -cnt0 -val15 -cnt1 -val16 -list
sets id5 cnt0 to 15, then sets id5 cnt1 to 16, then lists id5


setting Master Playback Volume MSI B450M PRO-VDH MAX - Realtek audio device ( uniaud.sys /V /A:1 )
unimix -id29 -cnt0 -val60 -list > unimix_test_id1.txt
unimix_test_id1.txt shows volume set to 60

Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on September 23, 2023, 12:18:08 am
Thanks Pete,
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 10, 2023, 08:45:17 am
Updated linux code to 6.1.62  -  patches applied pretty cleanly, so I don't anticipate  any significant changes in behaviour over the last 6.1.x build.  https://smedley.id.au/tmp/uniaud32-linux-6.1.62-20231110.zip

Code in https://trac.netlabs.org/uniaud/browser/GPL/branches/uniaud32-exp
Title: Re: Updated Uniaud32 build
Post by: Martin Iturbide on November 10, 2023, 01:23:21 pm
Updated linux code to 6.1.62  -  patches applied pretty cleanly, so I don't anticipate  any significant changes in behaviour over the last 6.1.x build.  https://smedley.id.au/tmp/uniaud32-linux-6.1.62-20231110.zip

Thanks Paul

I tested it (Release version) with ArcaOS 5.0.7 guest on VBox 7.0.10  (Windows 10 22H2 host).
- VM ICH AC97: Works, Audio control working, no traps. Just like the include ArcaOS 5.1 Uniaud, I have an issue that the first time I play a wav, I hear it loud (100%) and the next times I play it I hear it lower respecting the volume setting. If I drag a file to force the drag sound it sound 100% again, I play another wav  and it is 100% again and the next times respects the volume control (40%). It's the same issue from 3.01.03 (5.14.7), not a new issue.
- VM Intel HD Audio: Works, Audio control working, no traps. But it produce sounds in a lower volume. I had to rise it to a 100% to be able to hear something in my speakers, it is the same issue as the 3.01.03 (5.14.7).

For me it is working like the 3.01.03 (5.14.7) release. I'm keeping it on my VM.

Regards
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 10, 2023, 03:39:37 pm
I am having an issue with downloading from Paul's site. Will try again later.
Title: Re: Updated Uniaud32 build
Post by: David Graser on November 10, 2023, 04:45:04 pm
Reading world news, they appear to be having all kinds of internet problems in Australia.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 10, 2023, 09:16:49 pm
My issues are much simpler than Wednesday's Optus outage (which impacted my work mobile, but nothing else here). Seems I have a failing mesh router upstairs - twice in the last week, it has dropped off the network. I'll swap it out with a spare today, but meanwhile, the server should be back online.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 11, 2023, 03:59:58 am
This driver works very well on my Asus TUF GAMING X570-PLUS (WI-FI). I can even listen to music from YouTube in Dooble. With the last drop I got sound and could listen to wave and MP3s and CD's, but no Internet music. Not working on my ASUS FA7 Notebook. which no driver is working on. but I use my desktop more anyway. Good job Paul.
Title: Re: Updated Uniaud32 build
Post by: Paul Smedley on November 11, 2023, 05:41:01 am
Hey Eugene,  have I seen a 'testlog uniaud' from the ASUS FA7 Notebook?  (If I have,  I've forgotten :P)
Title: Re: Updated Uniaud32 build
Post by: David McKenna on November 11, 2023, 01:01:31 pm
 Works well here Paul, thanks!

[C:\]unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7320000 irq 16
  mixer: Realtek ALC887-VD
  componenets: HDA:10ec0887,1458a182,00100302 HDA:8086280b,80860101,00100000
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 11, 2023, 01:19:02 pm
Paul, I reframed due to the fact that you were working on QT.so in short I hav never posted those files.
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.01
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: NVidia
  driver: HDA-Intel
  name: HDA NVidia
  longname: HDA NVidia at 0xfc080000 irq 25
  mixer: Nvidia GPU 94 HDMI/DP
  componenets: HDA:10de0094,104316fd,00100100
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 11, 2023, 01:23:04 pm
Testlog says the driver is unided.
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on November 25, 2023, 06:58:17 pm
Here is the testlog from my Asus laptop
Title: Re: Updated Uniaud32 build
Post by: Dave Yeo on November 26, 2023, 04:00:29 am
After updating my computer (Thinkcenter 910s), when Thunderbird gets a message, it plays the system warning sound, currently garden\gar_warn.wav. Now the sound keeps repeating until I reboot. Other sounds sound fine. Downgraded to
Code: [Select]
Signature:       @#D Azarewicz:3.01.04#@##1## 7 Aug 2022 12:11:44      DAZAR1
 :5.15.59:::04::SVN717@@Universal Audio Driver for OS/2 (c) 2022 D Azarewicz
Vendor:          D Azarewicz
Revision:        3.01.04
Date/Time:       7 Aug 2022 12:11:44
Build Machine:   DAZAR1
ASD Feature ID:  5.15.59
FixPak Version:  SVN717
File Version:    3.1.4
Description:     Universal Audio Driver for OS/2 (c) 2022 D Azarewicz
and sound is fine.
Code: [Select]
W:\MMOS2>unimix -card
Command line mixer for UNIAUD32. Version 1.1
Detected UNIAUD version 3.01.04
Detected 1 audio adapter(s)
Card info for adapter 0:
  num: 0
  id: PCH
  driver: HDA-Intel
  name: HDA Intel PCH
  longname: HDA Intel PCH at 0xf7140000 irq 55
  mixer: Realtek ALC662 rev3
  componenets: HDA:10ec0662,17aa3106,00100300 HDA:8086280b,80860101,00100000
Title: Re: Updated Uniaud32 build
Post by: Martin Vieregg on February 29, 2024, 12:49:51 pm
Paul,

where do I find a compatibility list? I am planning to switch from onboard sound to a spearate sound card. Your driver is already used, it works fine! I did not find a text file in the archive. Thanks for your work!
Title: Re: Updated Uniaud32 build
Post by: Eugene Tucker on February 29, 2024, 04:01:52 pm
Here is a link that may be of some help https://www.alsa-project.org/wiki/Matrix:Main. Although  it may be as it is the ALSA project which the Uniaud is based on.