macnb

Thanks for the reply. 

The released version 1.8.8 of AppleALC on Acidanthera's GitHub has layout ID 11 but that only works on 10.11 and not on 10.13.

When you say you made modifications to those XML files, what did you change?

That is, what's the difference between what they released and what you have ?

 

The reason I am asking is that the info could be fed back to the Devs so that "vanilla" AppleALC can be used by others.

On 1/11/2024 at 3:58 AM, hidematu5310 said:

I recently built an Inspiron 1520 as an archiving machine.
It may be useful as the specs are similar except for the GPU.

 

The AppleALC that has been released does not support 10.4 Tiger and 10.7 Lion, 10.11 and later.
10.4 still needs to be investigated, but there is my mod that has made it compatible with all of 10.5 and later.
I also fixed the mute button issue, the slider sync issue, and a lot of sound assertion errors.
It has been tested on 10.5-10.13.

 

For TrackPad, my 32/64 compiled version of the latest merge of Rehabman VoodooPS2 and Fraxul's merge in Pull requests so you can try it out.

 

I have also moved from Clover to the release version 0.9.6 OpenDuet and OpenCore no acpi mod binary 0.9.7.
The no acpi mod does not use ACPI when booting Windows, so you can use Windows just like Clover.
If you want to try it, use 0.9.6 because OpenDuet 0.9.7 is broken.

mod_kexts.zip 223.9 kB · 0 downloads

@hidematu5310, I have a Dell Vostro 1500 which is essentially same as the Inspiron 1520 and I need it to run El Capitan (for a Matrox Video capture App + ExpressCard H/W).

I have it working...almost using standard OpenCore 0.9.6 (OpenDuet).

I also have High Sierra booting (I can dual boot either from the same drive).

Graphics works fine with a modified DSDT (to inject Nvidia info).

 

For the audio, the latest AppleALC.kext 1.8.8 works as-is for me with El Capitan but does not work with High Sierra.

So I tried your modded AppleALC.kext (one you posted in the other archived thread) and it works with High Sierra (and El Capitan).

 

What I would like to know is what modifications did you make to get it to work ?

 

For the Trackpad, I used Acidanthera's VoodooPS2Controller.kext 2.3.5 as-is and it basically works.

I cannot scroll (either at the trackpad edge nor 2-finger scroll). 

With El Capitan, I can single-tap with a finger but cannot do the same with High Sierra (single tap does not work).

 

With your VoodooPS2, can you scroll from the trackpad ?

Can you single tap with a finger to select ?

Did you make any changes ?

 

Thanks

 

 

 

Yes I think things have changed quite a bit in BS 11.5.2 (and no doubt with the latest 11.6).

Without changing SMBIOS model ID, the only way I could get the Installer to accept the "update" was to set the VMM Flag Bit in the OC config Root::Kernel::Emulate::Cpuid1Data.

That is:

 

This basically tells macOS that it is running in a VM and it's OK to install on that.

 

Install did not go well...lots of hangs & panic...mainly to do with USB.

Then I notice you were using USBInjectAll.kext which I was not as I never needed it on this system.

After installing that in OC, the Install went a bit more smoother and booted fine:

 

BTW, that VMM Flag also enables macOS to detect OTA OS updates. Without it ON, System Preference says that there are no Updates.

With the VMM Flag ON I see:

 

Also, I believe the updates are only the required Delta updates and does not download full blown 12GB+.

That VMM Flag should only used during Installs & updates and turned OFF in normal use as there's a small performance drop (~5%).

 

The biggest issue is USB 1.1 devices.

BS does not recognise any USB 1.1 peripherals when hot plugged.

E.g. for test purposes, plugged in a USB 1.1 Audio dongle and nothing is detected.

If I run IOREGistryExplorer App (or Hackintool), they hang until I unplug the USB 1.1 device.

I have the _isSingleUser kext patch applied to the IOHIDFamily kext.

If the USB 1.1 device (such as a wired mouse) is plugged in before booting, then that device works fine as long as you do not unplug it.

So hot plugging USB 1.1 devices does not work.

USB 2.0 devices work fine.

 

My bluetooth device is an old Apple USB device that is USB 1.1 and that works fine as it is always plugged in.

But...and it's a big but....it breaks sleep. The system does not fully sleep anymore in BS.

Screen goes off and the power light flashes as normal but the fans are still ON.

And, does not wake up on hitting a key on the Apple wireless keyboard or mouse. That implies the USB 1.1 BT device is not detected on wake.

I have a USB 2.0 Apple wired keyboard plugged in and even that cannot wake the system.

 

I booted back into Catalina and used corpnewt's USBMap command to properly I identify all the USB ports and created a USBMap.kext.

This was to replace USBInjectAll.kext as that is only meant to be a temporary kext for identifying the USB ports.

The generated USBMap.kext works well and I can now boot BS everytime. Attached it if you are interested.

 

So the hunt is on to find a Legacy USB 1.1 fix for this rig.

 

USBMap.kext.zip

 

 

On 9/12/2021 at 6:58 PM, Hervé said:

On the CPU power management front, does a table generated by Pike R Alpha's actually make any change? With Clover, you would probably notice better CPU power management with MacPro3,1 SMBIOS; I certainly do on my Vostro200 with the E8600 (and did too when I ran it with a Xeon X5270).

 

I get the same behaviour using Pike R Alpha's SSDT.aml as with Clover's Generate=YES.

With this E5450 Xeon there are not many P-states. I get two: 2Ghz or 3Ghz switching.

 

Running Geekbench 5 on Catalina booted via Clover using Generate = YES:

Running Geekbench 5 on Catalina booted via Clover with Generate=NO and Pike's SSDT.aml power management:

 

 

In both cases, the system idles at 2Ghz as there's low frequency mode on this CPU.

It is the same with OpenCore.

Hi @Hervé

Many thanks for the quick reply.

 

First, I indeed installed a fresh copy of Catalina.

Then, I created a new Container. 

I cloned that working Catalina Container to the new Container using CCC. This was so that I could keep Catalina.

The working Catalina drive is called CL-SSD.

The cloned Catalina drive is now called BS-SSD.

 

It's obviously formatted both CL-SSD & BS-SSD as APFS as that's all Catalina does when freshly installed.

Not sure why/how Catalina would work if I reformatted it to HFS+.

If the original partition was HFS+, when yo install Catalina, it will reformat it to APFS as part of the install.

 

Next, I booted cloned drive (called BS-CCD) which works fine and on that booted drive, I am running the Big Sur Installer......i.e. upgrading an existing Catalina build.

So do not understand why BS would complain.

 

4 hours ago, Hervé said:

You may install 11.5.2 whilst retaining the SMBIOS of an unsupported iMac alongside the -no_compat_check boot arg. However, you won't be able to subsequently receive Big Sur updates as this requires the SMBIOS of a supported model. I therefore suggest you create a separate config with, say, iMacPro1,1 SMBIOS for the sole purpose of updates.

 

Yes I am aware of the issues with Updates. Apart from temp iMacPro1,1 SMBIOS config, one could download the complete new BS Update when available and overwrite the existing OS.

 

4 hours ago, Hervé said:

For daily use you may return to iMac14,2 SMBIOS but it's of little/limited use given that OpenCore does not support CPU power management for these kind of Wolfdale/Harpertown C2D/C2Q/Xeon CPUs (no generation of P-States/C-States, unlike Chameleon/Clover). This is the sort of things I do on the Vostro200: SMBIOS iMac10,1 or MacPro3,1 for daily usage and iMacPro1,1 to update Big Sur.

 

Yes power management is not the same as Clover in OpenCore. I used PikerApha's ssdtgen script to create and SSDT for the Xeon E5450.

It's limited to low and High frequencies of 2Ghz and 3Ghz but that's all I got from Clover too.

 

4 hours ago, Hervé said:

You're using old boot arg kext-dev-mode=1; most people still ignore that this was only ever required for Yosemite and has been of absolutely no use in subsequent OS X/macOS versions ever since. You can get rid of this but, of course, this deprecated boot arg causes no harm whatsoever.

 

kext-dev-mode arg is a hangover from the I used to but many different macOS versions.

I recently removed Maverick, Yosemite & El-Capitan and just kept Snow Leopard, High Sierra & Catalina - for nostalgia 

It does not harm however.

 

Anyway, good to know that you did no patching but use vanilla BS Installer.

I just need to figure why my upgrade from Catalina to BS is not working.

@Hervé I looked at your guide here for your Vostro 200 regarding Big Sur installation.

My system is a Dell Inspiron 530 with G33M03 motherboard upgraded with Quad Core Xeon E5450 with nVidia GT730 GPU.

All works great for many many years as an iMac14,2 (Snow Leopard, High Sierra & Catalina) and booting via OpenCore 0.7.2 with emulated NVRAM.

 

I downloaded Big Sur 11.5.2 (using gibMacOS) directly from Apple.

However, when running the Big Sur Installer from booted Catalina 10.15.7 (19H1323), Installer complains "The update cannot be installed on this computer" :

 

All the drives are greyed out.

 

Looking at your config.plist for your Vostro, you use iMac10,1 which also is an unsupported model for Big Sur (just like my iMac14,2).

 

My boot args are set to :

 

-no_compat_check keepsyms=1 dart=0 debug=0x100 npci=0x2000 alcid=1 kext-dev-mode=1

 

-no_compat_check arg does not seem to help.

So how did you manage to install Big Sur ?

Did you patch the installer before hand ?

 

TIA

With 10.11.4, audio will break.

I had to re-run my script to regenerate my dummy HDA kext on two of my notebooks. One has ALC269 and the other has Constant CX20588.

 

I did not have to modify my script but just re-run it.

The new AppleHDA kext Info.plist file has new bundle library dependency. The old dummy kext will not include that dependancy so will not load.

These two entries need to be added to the Info.plist of the dummy had kext (in the OSBundlelibraries key):

 

com.apple.iokit.IOACPIFamily 1.2.0a1

com.apple.iokit.IOPCIFamily 1.1

I'm working on a fix as i got the issue. EAPDHIKeyboardDevice is meant for jack sense fix in old versions  (still keeping it as reference and might be useful in some cases for some one) to simulate Mute/Un-mute bezel pop up to start the audio stream.

 

EDIT: Check your PM for beta test please.

I just PM'ed you.

Tried your Debug#5 version in /L/E and it works.

Audio OK on boot and after sleep.

It seems to fail finding some object in AppleHDA. Check logs I sent. Thanks.

i tested all eapdfix since made eapdfix with dinesh long time ago. eapd 2.2 is no problem with me and other tester

 Well, v2.2 does NOT work for at least two people.

Just because it works for you does not mean there is no problem with v2.2.

 

codeccommander support l/e folder

I know. I told you it supports installation in /L/E -- like all non-Apple kexts should support.

 

eapdfix officially not support l/e.

- why not ???? 

All non-apple kext should support installation in /L/E

 

only can clover or s/l/e.

- well, as I said it does not work in /S/L/E. Even though audio works from boot, there's no audio after wake.

 

see attached system.log and codec dump

 

BTW, I have solved my problem by using codeccommander.kext but I am replying here to help Dinesh fix his kext.

system.log.zip

codec_dump.txt.zip

Sorry sherlock thanks for your comments/help but I don't understand your english.
Also, why do you post your HDA injector kext without your system spec, or which nodes you have ?
 
I have been doing hackingtoshes or over 5 years so I know what I am doing (most of the time) and know how to install/uninstall kexts (manually via terminal command line and via automated utilities like kext utility).
 
Why do I install kexts in /Library/Extensions ?
Because, since Yosemite, you can and should install NON-Apple kexts there.
I have several hacks and in all them running El Capitan, I install FakeSMC, AppleHDAxxx, USB Injectors, etc, etc in /L/E as they are all non-Apple kexts and they all work.
 
There's no need to put kexts in Clover/kexts folder - saves time (slightly) at boot time as there's no need to inject kexts into the kernel cache (that's how Clover kext injection works).
 
CodecCommander.kext works very well in /L/E folder (I did not read anywhere that states it should NOT be installed in /L/E nor that it does NOT support installation in /L/E).
 
Anyway, as a test, I uninstalled codeccommander.kext from /L/E and installed epadfix.kext (V2.2) into /S/L/E. On reboot, I have sound like before. HP works fine. BUT...after waking from sleep, I have no sound from speakers. HP's work fine.
 
Here's relevant system logs (there's nothing relevant in Clover logs):

29/02/2016 14:16:32.343 com.apple.kextd[45]: kext-dev-mode allowing invalid signature -67062 0xFFFFFFFFFFFEFA0A for kext "/System/Library/Extensions/EAPDFix.kext"
29/02/2016 14:16:33.274 com.apple.kextd[45]: kext signature failure override allowing invalid signature -67062 0xFFFFFFFFFFFEFA0A for kext "/System/Library/Extensions/EAPDFix.kext"
29/02/2016 14:16:33.000 kernel[0]: EAPDFix v2.2 Final Copyright (c) EMlyDinEsH <www.osxlatitude.com> 2013-2016.
29/02/2016 14:16:33.000 kernel[0]: EAPDFix: OS X Version 10.11
29/02/2016 14:16:33.461 com.apple.kextd[45]: kext-dev-mode allowing invalid signature -67062 0xFFFFFFFFFFFEFA0A for kext "/System/Library/Extensions/EAPDFix.kext"
29/02/2016 14:16:33.000 kernel[0]: EAPDFix: EAPD Capabale for output Node 0x14
29/02/2016 14:16:33.830 com.apple.kextd[45]: kext signature failure override allowing invalid signature -67062 0xFFFFFFFFFFFEFA0A for kext "/System/Library/Extensions/EAPDFix.kext"
29/02/2016 14:16:34.000 kernel[0]: EAPDFix: Using Fix type 1, will send the codec verb to enable EAPD/Jack Sense
29/02/2016 14:16:34.000 kernel[0]: EAPDFix: Headphone Node 0x21, Speaker Node 0x1e, External Mic Node 0x1b, Internal Mic Node 0x12
29/02/2016 14:16:34.000 kernel[0]: EAPDFix: Speakers has EAPD, Headphones has no EAPD, Jack sense Fix Disabled
29/02/2016 14:16:34.000 kernel[0]: EAPDFix: Speakers EAPD is down, Trying to power up...
29/02/2016 14:16:34.000 kernel[0]: EAPDFix: Codec verb sent successfully to power up EAPD.
29/02/2016 14:18:23.000 kernel[0]: considerRebuildOfPrelinkedKernel org.emlydinesh.driver.EAPDFix triggered rebuild
29/02/2016 14:23:07.000 kernel[0]: EAPDFix: Speakers EAPD is down, Trying to power up...
29/02/2016 14:23:07.000 kernel[0]: EAPDFix: Codec verb sent successfully to power up EAPD.
29/02/2016 14:23:20.858 com.apple.kextcache[460]: kext-dev-mode allowing invalid signature -67062 0xFFFFFFFFFFFEFA0A for kext EAPDFix.kext

Two issues I see with eapdfix.kext:
1. It should be installable in /L/E (as well as /S/L/E) - should be an easy fix. Also, why is it using IOProviderClass EAPDHIKeyboardDevice and not IOHDACodecFunction ?
2. it's auto detecting nodes incorrectly (see log and speaker node 0x14 and 0x1E).

Can you confirm old version of my kext ? i have not seen this issue in my testing. Please send me your system log with my kext installed so i can fix it.

 

Yes your version 2.1 works fine (installed into /Library/Extensions).

I installed version 2.2 (again into /Library/Extensions) and audio stops working.

 

Relevant logs entries:

28/02/2016 17:16:52.000 kernel[0]: EAPDFix v2.2 Final Copyright (c) EMlyDinEsH <www.osxlatitude.com> 2013-2016.
28/02/2016 17:16:52.000 kernel[0]: EAPDFix: OS X Version 10.11
28/02/2016 17:16:52.000 kernel[0]: EAPDFix: EAPD Capabale for output Node 0x14
28/02/2016 17:16:53.000 kernel[0]: EAPDFix: Using Fix type 1, will send the codec verb to enable EAPD/Jack Sense
28/02/2016 17:16:53.000 kernel[0]: EAPDFix: Headphone Node 0x21, Speaker Node 0x1e, External Mic Node 0x1b, Internal Mic Node 0x12
28/02/2016 17:16:53.000 kernel[0]: EAPDFix: Speakers has EAPD, Headphones has no EAPD, Jack sense Fix Disabled
28/02/2016 17:16:53.000 kernel[0]: EAPDFix: Speakers EAPD is down, Trying to power up...
28/02/2016 17:16:53.000 kernel[0]: EAPDFix: Codec verb sent successfully to power up EAPD.
28/02/2016 17:17:23.000 kernel[0]: EAPDFix: Failed to find AppleHDADriver entry.
28/02/2016 17:17:53.000 kernel[0]: EAPDFix: Failed to find AppleHDADriver entry.

It appears that certain nodes are not being scanned correctly.

My nodes are as follows:

 

0x12 - internal mic

0x14 - Internal speakers (with EAPD)

0x18 - external mic / Line-In

0x21 - HP (no EAPD)

 

0x1e - SPDIF - disabled in the PinConfigs (with 01EF00040)

 

See attached HDA injector.

 

Note: replacing eapdfix.kext (v2.2) with codeccommander.kext (in /library/extensions) and updating the kernel cache (using sudo kextcache -u /) immediately restores audio without restarting the OS.

MacNB-AppleHDA269.kext.zip

 

hello guys.

 

where did you install location?

 

EFI?

 

or S/L/E?

 

 

If you aren't try it,  try it EFI or S/L/E

 

and feedback with log

 

 

 

my system 

 

alc270

 

10.11.3 official and latest 10.11.4 beta4 are no problem

 

i install EFI folder.

 

i want to know your location and log

 

 

I already mentioned that I installed epadfix.kext into /Library/Extensions.

On my Acer 3820T running El Capitan 10.11.3, V2.2 EAPDFix.kext broke audio for me too.

After installing EAPDfix kext (in /Library/Extensions) I get no sound device (in or out).

I removed the EAPDfix.kext and audio is back.

 

BTW, I used your excellent audio patching guide (from codec dump to verbs commands to .zlib files) to kext create an audio injector kext.

 

UPDATE: I fixed my issue by using the EAPD CodecCommander kext by Timewalker and Rehabman (here). Default profile did enable the speakers on wake, but it disabled the HP's. I created my own profile that had a custom command to re-enable the HP's on wake as well. All good now.

Here's a compiled Version of the ALPS code by Son Vu (sontrg).

It doesn't have any of my fixes or updates yet. Only fixed the compile errors.

 

V5 users, please tell me if this works.

 

Jake Lo, please give this a try too.

 

If this works, I'll be gradually updating it to Linux 4.2, carefully documenting my changes to avoid breaking things like I did before.

 

Edit: SonVu2 has an extra device id for V5 (they have multiple device ids). Thats the only difference from the last build. Please try this instead.

 

Just tried your SonVu2 version on an Acer 3820T (running OSX 10.11.3).

Works very well compared to Rehabman's version which has limited ALPS support (jumpy scrolling).

Your version gives me very smooth scrolling vertically and horizontally.

 

Not sure what version of h/w I have. This is what i see in the logs:

ApplePS2ALPSGlidePoint::identify: Found ALPS Device with ID E7=0x73 0x02 0x64, EC=0x88 0x07 0x8a

Two finger scroll works very well.

But I am used to one finger side scroll.

I cannot find any settings in the SysPrefences->Trackpad for enabling one finger scroll.

Cannot find many settings in many of the Info.plist's within the kexts (except for profile of Default or HPQOEM).

Where are the gesture settings (specifically one finger side scrolling) ?

 

UPDATE:

Never mind. I found another solution...using bpedman's solution https://osxlatitude.com/index.php?/topic/2545-new-touchpad-driver-for-e6520-alps/&do=findComment&comment=19137- using rehabmanmerge.zip

I don't know how it enables one-finger scroll but it works. In that version, there's no inertia scroll with one finger. Ah well.

So I did not give up --- found the problem and made a solution.

 

See here.

 

Yosemite now wakes from sleep on Inspiron 530.

 

Enjoy.

OK. Sorry to hear of your drive failure.

 

Will give up for now.

Wait and see if any other brave soul tries 10.10 on their old dell.

The last version you tried was DP5 but not tried the "final" Released version ?

Thx Herve I found your boot pack (I was looking in post #1 where I expected it).

I tried your DSDT. The 530 boots fine but after logging in, it would KP in the AICPUPM kext.

Kind of expected it.

 

Interesting that you use iMAC10,1 platform type in SMBIOS. I use MACPRO3,1.

Anyway, I tried imac10,1 but made no difference.

 

If you ever add Yosemite to your Vostro 200, it would be good to see what result you get with sleep/wake.

My Vostro200 has a DSDT with working sleep in SL, L, ML and Mav. Try it, it should work on the Inspiron 530, they're more our less identical machines. From memory, sleep was not working in my Yosmite DP installations, but then my HDD failed and I haven't replaced it yet. So no mainstream Yosemite tests on this one yet.

yes I know...I originally fixed that DSDT to get the sleep working in SL on the 530 (at TM site and then at IM site).

Same DSDT worked on L, ML then MV.

Something has changed in Yosemite in the way power management and ACPI are handled.

 

Would be good to see if you can replicate the sleep issue on Vostro 200 when you get Yosemite installed.

 

Will try you DSDT but cannot find any download on this site.

Your Vostro 200 topic does not have any links to your boot pack.

There no downloads on this site apart from the edp app.

My Inspiron 530 with stock bios 1.0.18 recognised the Xeon E5450 out of the box - no modded bios required.

I did look at my bios and examined the versions of microcode included.

The Xeon microcodes as are slightly older but no issues at all.

Have a a pretty good working Mavericks build on Inspiron 530. It goes to sleep OK (well it take 30 seconds). Wakes instantly.

Graphics card is GT240 that works out of the box.

 

Did a fresh install of Yosemite on a a spare partition (using latest Chameleon).

Used the existing DSDT.aml. All seems to work well EXCEPT wake from sleep.

 

Putting to sleep is same as Mavericks ...takes ˜30 seconds... the fans turn off and the power light flashes.

 

BUT...touching the keyboard or pressing the power button starts up the power but the screen stays black (no signal from DVI) and the CAPS LOCK light does nothing. Basically it's crashed and the only option is to restart.

 

So I did a new install on another drive but this time using Clover boot loader. No change. Still the same problem ...no wake from sleep.

 

I checked the logs and there's no mention of sleep starting or any Wake Reason. Not sure how to debug this problem.

 

Anyone tried Yosemite on an Inspiron 530 or Vostro 200 with successful wake from sleep ??

Brilliant guide.

Bought a Xeon E5450 and 4 adapters for £26 (even included a Xeon Inside sticker).

 

Followed this guide for my trusty old Dell Inspiron 530 running Mavericks.

All works well. 

 

18 to 20% improvement in benchmarks compare to Q9300 but was fun modding.

 

Thx

do you have the duallink=yes option enabled in chameleon?

I tried duallink enabled and disabled. Made no difference.

It just not loading the frambuffer.

The driver probably has hard coded device id's and when it see 8086:041e, it refuses to load the frame buffer.

 

I curious if the other guy really managed to get the HD4400 working with full acceleration or was it just hi-res (1080p) as none else seems to have any success get QE/CI working.

Like to know how to get my graphics to work with QE/CI.

 

I am toying with my Sony Tap 11 tablet.

The CPU is i5-4210Y which has HD4200 - same family as HD4400 - just slower/low power.

The vendor:device id is 8086:041E (same as HD4400).

 

Installed 10.9.3 on a USB HDD.

 

What works:

• hi-res 1920x1080x32 screen (even in boot loader)
• Sound (ALC283).
• Battery mgr.
• Speedstep (9 p-states).
• USB 3

 

What does NOT work:

• CI works as I have full 1080p resolution but acceleration does not work.
• WiFi (as it's Intel).
• Bluetooth (works sometimes - again, Intel)
• Sleep/wakeup

I have DSDT and SSDT files.

I tried almost almost all combinations of IntelAzulFB=xx on the boot loader options.

I tried SMBIOS.plist with MacBookAir6,2 and iMac14,2.

I using chameleon r2277

 

Anyone any pointers as to what else I can try?