Jump to content

[GUIDE] Making a DSDT.aml for Dell XPS M1330, XPS M1530, and XPS M1730


immo
 Share

2,030 posts in this topic

Recommended Posts

Sorry, it's NOT. Notice my sig - it is M1530! An older model in fact (early 2008), that's why I'm using MacBookPro3,1 in my SMBios.plist.

 

What I have already posted, nothing special. My S/L/E is of course (aside from the modified AppleACPIPlatform.kext) is all vanilla. I have to test upgrades on a real Mac sometimes using a external USB.

 

One thing I have experienced is if I use both HDMI and VGA simultaneously, of course connecting to 2 separate external displays, it locks up as well. One-at-a-time works just fine. I have also connected a DVI display using HDMI->DVI adapter and that works fine, too.

I have similar results to the others in that my machine freezes when connecting the laptop to the HDMI port of my BenQ E2420HD monitor via an HDMI-HDMI cable.

 

The variables I can think of:

  • the tv/monitor you are connecting to is obviously a different model to mine
  • the cable itself maybe different in some way (did you use a HDMI-DVI cable on your Toshiba TV?)
  • SMBIOS (MBP 3.1 vs 5.1)

Maybe anyone who wants to try and gets it working might like to indicate in more detail the TV/Monitor, cable & SMBIOS used?

 

According to this post http://www.insanelymac.com/forum/topic/254195-hdmi-on-xps-m1530-update/ , HDMI to a DVI monitor works.

 

Anyway good to see at least one of us has it working :) .

Link to comment
Share on other sites

I have similar results to the others in that my machine freezes when connecting the laptop to the HDMI port of my BenQ E2420HD monitor via an HDMI-HDMI cable.

 

The variables I can think of:

  • the tv/monitor you are connecting to is obviously a different model to mine
  • the cable itself maybe different in some way (did you use a HDMI-DVI cable on your Toshiba TV?)
  • SMBIOS (MBP 3.1 vs 5.1)

Maybe anyone who wants to try and gets it working might like to indicate in more detail the TV/Monitor, cable & SMBIOS used?

 

 

I use a HDMI->DVI adapter to connect to a HP w2007 flat panel monitor. It only has Digital (DVI-D) and VGA inputs but no HDMI. So HDMI out from M1530 to DVI in to the monitor through the adapter. Works fine.

 

Connection to the Toshiba REGZA 52RV530U HDTV is HDMI->HDMI using a straight through HDMI cable. This cable is more permanently used in a PS3 playstation. Thus, I just unplug the HDMI cable from the PS3 and plug it into my M1530. This is the setup in my posted pic in #1494. Works fine, too.

  • Like 1
Link to comment
Share on other sites

I use a HDMI->DVI adapter to connect to a HP w2007 flat panel monitor. It only has Digital (DVI-D) and VGA inputs but no HDMI. So HDMI out from M1530 to DVI in to the monitor through the adapter. Works fine.

 

Connection to the Toshiba REGZA 52RV530U HDTV is HDMI->HDMI using a straight through HDMI cable. This cable is more permanently used in a PS3 playstation. Thus, I just unplug the HDMI cable from the PS3 and plug it into my M1530. This is the setup in my posted pic in #1494. Works fine, too.

Thanks for the info SQueek.

 

I can confirm that HDMI to DVI definately works and at full resolution.  When I connect my XPSm1530's HDMI port to the BenQ E2420HD's DVI port with a HDMI-DVI cable, there are no hangs/freezes at all, and the output is in the monitor's native HD 1920 x 1080.

 

Whether the connection to a monitor/TV via HDMI cable will work probably depends on whether the TV/monitor treats the connection like DVI (ie accepts video signal without audio) when the XPS is connected.  I suspect in SQeek's case, his TV is more "compatible" with the XPS and can "acknowledge" to the computer that it is OK to supply only the video signal.

 

Just my 2c

 

Edit

Tried to directly connect to my Samsung LCD TV (LA32S7) via an hdmi-hdmi cable and the computer freezes again booting up into OSX Mavericks.

 

However, when I boot the XPS up in Windows 7 X64, everything works perfectly with hdmi-hdmi into the Samsung, including sound (via HDMI playback device in control panel of Windows 7).

 

I think its the luck of the draw whether you can connect via hdmi-hdmi to the XPS in OSX and it depends on how your TV/monitor handles the connection.  As long as audio is ignored, it seems happy.  It probably can be fixed by further dsdt edits or hacking the drivers (evidenced by the hardware working perfectly in windows) but this is beyond my level of expertise.....

Link to comment
Share on other sites

This is stupid, I connect an external monitor with a HDMI to TVI adapter and it works, I connect my TV which has HDMI and it freezes, I use the TV connected with the XPS a lot to watch movies and TV series but booting with the windows partition.
 
Here is a pic:

 

 

IMG_8324.JPG

Link to comment
Share on other sites

 

 
 
 

 

DSDT is now in a ACPI folder?  Can it reside in the root of Extra?  What had to be patched in AICPUPM?  Im guessing from the Zip you provided AppleACPIPlatform.kext within that zip has a newer version 2?

 

Thanks

 

EDITED:

 

I am really having a hard time installing 10.9 using MyHack.  Is there any special boot parameters needed prior to starting the install?  I have 2 XPS 1530's, one with a 2.4GHZ and another with 2.6GHz doing the same thing.  One time I get the PCI Configuration Begins and verbose stops at that point and the installer seems to freeze.  Another time I get into the installer but 17 to 21 minutes along it freezes.

 

I am thinking I am going to have to install it another way or from another PC.

 

Has anyone successfully completed a fresh install from the XXPS M1530 and if so what installer and procedure?

 

I hop I can use the Same DSDT for both?

 

Thanks

Link to comment
Share on other sites

@oSxFr33k, DSDT.aml still resides at the root of the partition, the one in the ACPI folder is a backup. 
 
I never patched the AppleIntelCPUPowerManagement.kext, only the AppleGraphicsPowerManagement.kext required patching or deletion. 
 
Installing the 10.9 is similar to installing all other OS X iterations. Although, unlike with the previous OS X versions, the AppleACPIPlatform.kext regression is no longer required; use the native one for installation. However, if your Sleep mode is faulty post installation, simply replace your native AppleACPIPlatform.kext with this particular one provided earlier in this thread by SQeek:
 
AppleACPIPlatform.kext-v2.0.zip
 

Also, ensure to migrate the mach_kernel to the root of the installed partition.

I've never used MyHack and I really fail to see its purpose and necessity, especially for our systems.
Generally speaking, please avoid third party OS X installers as they tend to unnecessarily complicate installations.

  • Like 2
Link to comment
Share on other sites

Has anyone successfully completed a fresh install from the XXPS M1530 and if so what installer and procedure?

 

I hop I can use the Same DSDT for both?

 

Thanks

 

Generally speaking, please avoid third party OS X installers as they tend to unnecessarily complicate installations.

+1 for DarwinX's recommendation - the best way (and the best way to learn) is to manually create an install USB (see http://www.insanelymac.com/forum/topic/280756-guide-the-all-in-one-guide-to-vanilla-os-x-including-chameleon-dsdt-for-beginners-updated-for-mavericks/ ).

 

I basically followed the guide with some minor changes to create an install USB:

  • In part 2.0, I used an mbr formatted USB with 1 partition Mac OS X Extended (Journaled) - some older BIOS based computers can't boot off GUID USB.
  • In /S/L/E, place FakeSMC v3.3.1 & VoodooPS2Controller.kext v1.1.0 (no need for NullCPUPowerManagement.kext with DSDT in /extra).  Install Chameleon 2.2 r2266 into the USB and make sure that the mach kernel is on the root.
  • Since I had a fully functioning ML install, I omitted most of 2.2.  Use your old /Extra folder (including smbios & org.chameleon.boot.plist & DSDT) and copy onto USB.

Boot flags - I used only -v & -f.

More details for my post install here http://www.insanelymac.com/forum/topic/190780-guide-making-a-dsdtaml-for-dell-xps-m1330-xps-m1530-and-xps-m1730/page-73?do=findComment&comment=1960163 (page 73 of this thread).

 

Maybe immo should update the first post for changes in Mavericks (esp no need for AppleACPIPlatform.kext regression except for XPSM1530 sleep issues, DarwinX's patched DSDT for 10.8.5/10.9 etc)- the same questions tend to get repeated.

  • Like 3
Link to comment
Share on other sites

I noticed that Use Kernel Cache, Generate P and C states, CST Using SystemIO (what ever that is for) and Graphics Enabler are defaulted to no or unchecked in Chameleon Wizard for the org.chameleon.boot.

 

Should they be enabled?

 

I replaced the  Mavericks vanilla AppleACPIPlatform.kext with the one DarwinX provided AppleACPIPlatform.kext-v2.0 but still get a back screen when I wake from sleep?

 

Thanks

Link to comment
Share on other sites

Installing the 10.9 is similar to installing all other OS X iterations. Although, unlike with the previous OS X versions, the AppleACPIPlatform.kext regression is no longer required; use the native one for installation. However, if your Sleep mode is faulty post installation, simply replace your native AppleACPIPlatform.kext with this particular one provided earlier in this thread by SQeek:

 

A bit off-topic but how do I know if I need to replace AppleACPIPlatform.kext on my other Hackintoshes, I am using dsdt on them.

 

Also discover that using VoodooPS2Controller.kext version 1.11 gives a 2 keys with the same keycode so I can only have whith both keys, maybe this does not happen on US layouts, so I am sing the RehabMan-Voodoo-2013-0918 and now have and also | º

Link to comment
Share on other sites

A bit off-topic but how do I know if I need to replace AppleACPIPlatform.kext on my other Hackintoshes, I am using dsdt on them.

 

Also discover that using VoodooPS2Controller.kext version 1.11 gives a 2 keys with the same keycode so I can only have < > whith both keys, maybe this does not happen on US layouts, so I am sing the RehabMan-Voodoo-2013-0918 and now have < > and also | º

I have the exact same question to.  I have a Asus G51Jx sandy bridge first generation that suffers from the PCI configuration Begins then no verbose messages.  I am unsure how this will work in 10.9 but in 10.7 I had this issue until I replaced it with the AppleACPIPlatform.kext from about a year ago's version, which took care of the PCI Configuration Begins issue in 10.7.

 

It would be nice to know if this kext is specific to the XPS Latitude series or can also be used on other brand of Laptops as the one I mentioned.

Link to comment
Share on other sites

Re AppleACPIPlatform.kext - I think the most sensible approach is to try the native 10.9 kext first and if you have no problems with boot or sleep, then keep using it.  The regression is really only necessary if you have boot problems or discover sleep issues (eg reboot with second sleep on the XPSM1530).

 

@oSxFr33k

The kext v2 found by SQeek and also uploaded by DarwinX originated on http://www.osx86.net/files/file/92-appleacpiplatformkext-for-os-x-109/ by an author named crunk.  He labels it as compatible with Mavericks but does not say it is specific for the XPS (if you read his comment, infact he says "only Gigabyte and ASUS will work"....so certainly try it on your Asus G51Jx if the native Mavericks AppleACPIPlatform.kext gives you issues.

 

Also SQeek, DarwinX, lulighttec & I reported no issues with sleep after using the regression so not sure what is causing your black screen on wake.  Did you set hibernate mode to "suspend to RAM only"  ie open terminal and type sudo pmset -a hibernatemode 0 ?

  • Like 1
Link to comment
Share on other sites

A bit off-topic but how do I know if I need to replace AppleACPIPlatform.kext on my other Hackintoshes, I am using dsdt on them.

 

Also discover that using VoodooPS2Controller.kext version 1.11 gives a 2 keys with the same keycode so I can only have < > whith both keys, maybe this does not happen on US layouts, so I am sing the RehabMan-Voodoo-2013-0918 and now have < > and also | º

The AppleACPIPltaform.kext regression was implemented to mitigate a specific issue. If you are experiencing no issues with your current native kext, the regression is unnecessary. 

 

As far as the RehabMan's branch of the VoodooPS2Controller which is coded for the Synaptics touchpad, it is sadly completely incompatible with my {censored}ware ALPS GlidePoint v10.115.

  • Like 1
Link to comment
Share on other sites

Re AppleACPIPlatform.kext - I think the most sensible approach is to try the native 10.9 kext first and if you have no problems with boot or sleep, then keep using it.  The regression is really only necessary if you have boot problems or discover sleep issues (eg reboot with second sleep on the XPSM1530).

 

@oSxFr33k

The kext v2 found by SQeek and also uploaded by DarwinX originated on http://www.osx86.net/files/file/92-appleacpiplatformkext-for-os-x-109/ by an author named crunk.  He labels it as compatible with Mavericks but does not say it is specific for the XPS (if you read his comment, infact he says "only Gigabyte and ASUS will work"....so certainly try it on your Asus G51Jx if the native Mavericks AppleACPIPlatform.kext gives you issues.

 

Also SQeek, DarwinX, lulighttec & I reported no issues with sleep after using the regression so not sure what is causing your black screen on wake.  Did you set hibernate mode to "suspend to RAM only"  ie open terminal and type sudo pmset -a hibernatemode 0 ?

Thanks I forgot to set hibernate-mode to 0.  What about the org.chameleon.boot settings?  See my post above.  Thanks

Link to comment
Share on other sites

Thanks I forgot to set hibernate-mode to 0.  What about the org.chameleon.boot settings?  See my post above.  Thanks

GenerateCStates=Yes/No - Enable auto generation of processor idle sleep states.
GeneratePStates=Yes/No - Enable auto generation of processor power performance states.
CSTUsingSystemIO=Yes/No - New C-State _CST generation method using SystemIO registers instead of FixedHW.
 

 

<?xml version="1.0" encoding="UTF-8"?>

<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">

<plist version="1.0">

<dict>

<key>CSTUsingSystemIO</key>

<string>Yes</string>

<key>Default Partition</key>

<string>hd(0,3)</string>

<key>EthernetBuiltIn</key>

<string>Yes</string>

<key>GenerateCStates</key>

<string>Yes</string>

<key>GeneratePStates</key>

<string>Yes</string>

<key>GenerateTStates</key>

<string>Yes</string>

<key>Graphics Mode</key>

<string>1680x1050x32</string>

<key>GraphicsEnabler</key>

<string>Yes</string>

<key>Kernel</key>

<string>mach_kernel</string>

<key>Kernel Flags</key>

<string>-v</string>

<key>RebootOnPanic</key>

<string>Yes</string>

<key>ShowInfo</key>

<string>Yes</string>

<key>SystemType</key>

<string>2</string>

<key>Theme</key>

<string>Mavs</string>

<key>Timeout</key>

<string>1</string>

<key>UseKernelCache</key>

<string>Yes</string>

</dict>

</plist>

 

Link to comment
Share on other sites

The AppleACPIPltaform.kext regression was implemented to mitigate a specific issue. If you are experiencing no issues with your current native kext, the regression is unnecessary. 

 

As far as the RehabMan's branch of the VoodooPS2Controller which is coded for the Synaptics touchpad, it is sadly completely incompatible with my {censored}ware ALPS GlidePoint v10.115.

Thanks. :D

Link to comment
Share on other sites

The question of c states on our machines is a vexed one and the major reason for temps in OSX running higher than in Windows.

 

I find that setting P & C states in chameleon to not make any difference like drummerboybac at Project OS X forums found here http://www.projectosx.com/forum/index.php?showtopic=2188 .  Despite the fact that AppleLPC.kext loads with our dsdts, it remains non functional, hence the warning "WARNING - LPC device initialization failed: C-state power management not initialized" in the boot messages.  In the chameleon documentation is a prerequisite "to enable C-States you must have LPC working on your system".

 

ab_73 in this post http://www.insanelymac.com/forum/topic/190780-guide-making-a-dsdtaml-for-dell-xps-m1330-xps-m1530-and-xps-m1730/page-22?do=findComment&comment=1519236 a long way back in the thread mentions a difference in memory address (AppleLPC.kext uses PM_BASE of 0x400 vs 0x1000 in our machines).  According to him, a simple test for c-states is to view IOREG with IORegistryExplorer.app and search if "CSTInfo" exists - unfortunately not for me.

 

Edit: From Maldon post http://www.insanelymac.com/forum/topic/235523-dsdt-auto-patcher/page-278?do=findComment&comment=1737553

to determine whether c-states working on your machine, in terminal type

10.6/10.6.7
ioreg -lw0 | grep CSTInfoIN

 

10.6.8 AND LION/ML/Mavericks
ioreg -lw0 | grep AICPMVers

and see if there is any output - no output = no c-states :(

 

Slice mentions that "I think with Dell notebooks it is impossible because of LPC device is differ from Apple's".

  • Like 3
Link to comment
Share on other sites

In-built Ricoh SDHC MMC/SD card on Dell XPS M1530 works on Mavericks 10.9.


I am using Mac OS X 10.7 VoodooSDHC.kext Version 1.1.2b1 to enable Ricoh SDHC MMC/SD card on Dell XPS M1530. 


 


 


s-mbp:~ sudoadmin$ sudo systemstats 


Password:


 


Summary


=======


      System Version: 13A603


          Total Time: 24:17:31


 


Usage


=====


         Time on A/C: 14:00:49


     Time on Battery: 01:03:50


 

s-mbp:~ sudoadmin$ uname -a


Darwin s-mbp 13.0.0 Darwin Kernel Version 13.0.0: Thu Sep 19 22:22:27 PDT 2013; root:xnu-2422.1.72~6/RELEASE_X86_64 x86_64


s-mbp:~ sudoadmin$ kextstat |grep SDHC


  119    0 0xffffff7f825ea000 0x5000     0x5000     org.voodoo.VoodooSDHC (1.1.2b1) <18 11 5 4 3 1>


s-mbp:~ sudoadmin$ ioreg |grep SDHC


    | |   |   +-o SDHC@9,1  <class IOPCIDevice, id 0x10000018e, registered, matched, active, busy 0 (2364 ms), retain 12>


    | |   |   | +-o VoodooSDHC  <class VoodooSDHC, id 0x100000412, registered, matched, active, busy 0 (2343 ms), retain 7>


s-mbp:~ sudoadmin$ 


 


I have attached VoodooSDHC.kext Version 1.1.2b1 for convenience.

VoodooSDHC.kext.zip

  • Like 1
Link to comment
Share on other sites

Any plans for implementing gestures, tabs and scroll on Alps touchpad?.  :)

Something like that would indubitably require someone with significant programming expertise and is therefore unlikely to happen.

 

At least we have the two finger scrolling.

  • Like 1
Link to comment
Share on other sites

Something like that would indubitably require someone with significant programming expertise and is therefore unlikely to happen.

 

At least we have the two finger scrolling.

Ofcourse. :)

 

Looks like Synaptics code and Alps Linux code is being modified by bdepman to make it work with Alps on OS X.

Still work under progress.

https://github.com/RehabMan/OS-X-Voodoo-PS2-Controller/pull/2

 

and another effort of VoodooPS2Controller V1.10 for Alps Touchpad by RehabMan, Slice and Andy Vandijck.

http://www.insanelymac.com/forum/files/file/49-voodoopscontroller-modified-version/

 

Edited: 

Download link for VoodooPS2Controller V1.10 for Alps Touchpad by RehabMan, Slice and Andy Vandijck.

  • Like 2
Link to comment
Share on other sites

The question of c states on our machines is a vexed one and the major reason for temps in OSX running higher than in Windows.

 

I find that setting P & C states in chameleon to not make any difference like drummerboybac at Project OS X forums found here http://www.projectosx.com/forum/index.php?showtopic=2188 .  Despite the fact that AppleLPC.kext loads with our dsdts, it remains non functional, hence the warning "WARNING - LPC device initialization failed: C-state power management not initialized" in the boot messages.  In the chameleon documentation is a prerequisite "to enable C-States you must have LPC working on your system".

 

ab_73 in this post http://www.insanelymac.com/forum/topic/190780-guide-making-a-dsdtaml-for-dell-xps-m1330-xps-m1530-and-xps-m1730/page-22?do=findComment&comment=1519236 a long way back in the thread mentions a difference in memory address (AppleLPC.kext uses PM_BASE of 0x400 vs 0x1000 in our machines).  According to him, a simple test for c-states is to view IOREG with IORegistryExplorer.app and search if "CSTInfo" exists - unfortunately not for me.

 

Slice mentions that "I think with Dell notebooks it is impossible because of LPC device is differ from Apple's".

 

Its too bad that this one minor thing still Plagues the XPS M1330 and XPS M1530.  Not even a DSDT injection of any kind will fix this.  Other than this one minor imperfection, they are still known to be one of the best MacBook Pro clones around to this day!!

  • Like 1
Link to comment
Share on other sites

 Share

×
×
  • Create New...