DSDT for hp DV7-6B32US -AIO patch on page 3

[stuck]

Hoping someone could help me or point me in the right direction in editing DSDT for my Laptop.

I have tried to patch it using the DV6 sandy bridge generic patch and DV6T sandy bridge generic patch, both had errors, I would delete the lines that cause the errors to compile.

These experiments failed. I'm not sure if I should just start patching each of the different patches for sandy bridge eg. DTGP, EHCI etc.

Patching them one at a time, compile and try to boot with it. If it boots fine then go on and patch the next...

Is this the method that the experts would recommend?

DSDT.zip

Edited January 5, 2012 by Guest

[Cassio]

Your unedited DSDT has ony one error, just remove the line External (\TNOT).

You can apply various patches and compile only once in the end, unless one of the patches is causing compilation errors, then you need to find the culprit.

Don't apply all the Sandy Bridge patches, just the ones you actually need http://Olarila.com/forum/viewtopic.php?f=19&t=634

[stuck]

Thanks for responding Cassio,

How do I know what i actually need?

Do I check the kernel.log to try and figure out where the problem is? Or Is ther some other log file to check?

[Cassio]

See the link in my previous reply. Keep reading after "usual patches".

[stuck]

I followed each of the suggestion in the section after usual patches.

1. kextstat shows AppleLPC is loaded - so I guess LPC is not needed?

2. I don't have any HPET problems in boot and no mention of it, in kernel.log - so I guess HPET patch is not needed?

3. I did patch the AICPUPM but did not notice any differences except sleep actually turns off the screen now, but the power light is still on and I can't wake it back up. Kernel log shows "ACPI_SMC_PlatformPluging::startForReal - waitForService(resourceMatching(AppleIntelCPUPowerManagement) timed out" -- before or after the patch.

4. Could not find any references to EHCI or UHCI in kernel.log until I patched AICPUPM, so I applied the EHCI patch, didn't seem to make any difference...

5. Could not find any references to Wake or Wake reason

6. Audio: I couldn't figure out if that applies to intel HD audio or just the Azalia audio?

7. DTGP applied, didn't seem to make a difference?

8. SMBUS, I don't know what the smbus kext does, and didn't read anywhere about whether I need it or not?

9. Shutdown, works fine - so I'm guessing I don't need it...

other errors in kernel.log:

nstat_lookup_entry failed: 2

errors relating to applehdadriver eg. Sound assertion "false ==codecFound" failed in AppleHDADriver at line 1449 goto handler

SNB Framebuffer didnot show up, unload

virtual bool IOHIDEvenSystemUserClient::initWithTask .....: Client task not privileged to open IOHIDSystem for mapping memory

AppleUSBCDC: start - initDevice failed (showed up after patching EHCI)

In the meantime, I'm only able to boot into my Lion disk using: -f cpus=1 busratio=22 arch=i386

[Cassio]

Then you don't need to patch your DSDT.

In "3", after patching AICPUPM you have to remove NullCPUPM, or it will surely make no difference.

You may use "DualLink" patch for your video.

Use "lapic kernel" (Taptun) to be able to boot without cpus=1

http://www.osx86.net/view/1945-os_x_lion_kernel_lapic_native.html

[stuck]

Cassio, you are a genius!

Thank you so much.

Just patching DTGP and DualLink on my dsdt got the laptop to boot to full resolution and HDMI properly detects a secondary monitor.

Since I've already patched the AICPUPM sleep works as well. It does seem to take 10-15 seconds for the lights to turn off though. Pressing any key wakes it up immediately.

I am now seeing:

AppleUSBEHCI [0xfff....]: unable to get filterInterruptEventSource and

AppleUSBEHCI [0xfff....]::UIMInitialize - error occurred (0x0...)

Patching EHCI for sandy bridge laptops doesn't seem to fix this.

and the USB ports are useless at this point.

Also I'm not sure if any of this is important?

Transcript Offline - Buffer Pool Allocate [181000] failed

VoodooPS2SynapticsTouchPad loaded

No interval found for . Using 8000000

VoodooPS2Trackpad: Synaptics TouchPad v7.5

I'm off to try the kernel patch to get rid of the cpus=1 problem and finding proper drivers for all wifi and Battery stats etc.

Thanks again.

[Cassio]

I am now seeing:

AppleUSBEHCI [0xfff....]: unable to get filterInterruptEventSource and

AppleUSBEHCI [0xfff....]::UIMInitialize - error occurred (0x0...)

Patching EHCI for sandy bridge laptops doesn't seem to fix this.

That's strange, it should.

Try this patch

into device name_adr 0x001D0000 insert
begin
OperationRegion (EOWN, PCI_Config, 0x68, 0x04)\n
Field (EOWN, ByteAcc, NoLock, Preserve)\n
{\n
           Offset (0x02), \n
   HCBO,   1, \n
           Offset (0x03), \n
   HCOO,   1\n
}
end;
into device name_adr 0x001A0000 insert
begin
OperationRegion (EOWN, PCI_Config, 0x68, 0x04)\n
Field (EOWN, ByteAcc, NoLock, Preserve)\n
{\n
           Offset (0x02), \n
   HCBO,   1, \n
           Offset (0x03), \n
   HCOO,   1\n
}
end;
into definitionblock code_regex . insert
begin
Method (PINI, 0, NotSerialized)\n
{\n
   Store (Zero, \_SB.PCI0.EHC1.HCBO)\n
   Store (One, \_SB.PCI0.EHC1.HCOO)\n
   Store (Zero, \_SB.PCI0.EHC2.HCBO)\n
   Store (One, \_SB.PCI0.EHC2.HCOO)\n
}
end;
into method label _INI insert
begin
PINI ()
end;
into method label _WAK code_regex Return replace_matched
begin
PINI ()\n
Return
end

[oldnapalm]

Try to add more IRQs in device HPET

http://www.insanelymac.com/forum/index.php?showtopic=206313

It's like the "IRQs" patch, but insert 4 IRQs instead of 2. Check patches for other HP laptops

http://Olarila.com/Packs/HP/DV6%20Sandy%20Bridge%20GENERIC.txt

into device name_hid PNP0000 code_regex IRQNoFlags\s\(\)\n\s+\{(\d+)\} remove_matched;
into device name_hid PNP0100 code_regex IRQNoFlags\s\(\)\n\s+\{(\d+)\} remove_matched;
into device name_hid PNP0B00 code_regex IRQNoFlags\s\(\)\n\s+\{(\d+)\} remove_matched;
into device name_hid PNP0103 code_regex IRQNoFlags\s\(\)\n\s+\{(\d+)\} removeall_matched;
into device name_hid PNP0103 code_regex Name\s\(([^,]+),\sResourceTemplate\s\(\)\n\s+\{((?:.|\n)*)\}\) replace_matched
begin
Name (%1, ResourceTemplate ()\n
                   {\n
                       IRQNoFlags ()\n
                           {0}\n
                       IRQNoFlags ()\n
                           {8}\n
                       IRQNoFlags ()\n
                           {11}\n
                       IRQNoFlags ()\n
                           {15}\n
%2
})
end

[stuck]

Thanks for the suggestions Cassio and Oldnapalm,

I will try each of these patches. But I have a question on the HPET patch. My DSDT doesn't have IRQs in the HPET section at all so do I use this "new HPET" patch:

into device name_hid PNP0103 parent_adr 0x001F0000 remove_entry;
into device name_adr 0x001F0000 insert
begin
Device (HPET)\n
{\n
   Name (_HID, EisaId ("PNP0103"))\n
   Name (CRS, ResourceTemplate ()\n
   {\n
       IRQNoFlags ()\n
           {0}\n
       IRQNoFlags ()\n
           {8}\n
       Memory32Fixed (ReadOnly,\n
           0xFED00000,         // Address Base\n
           0x00000400,         // Address Length\n
           _Y09)\n
   })\n
   Method (_STA, 0, NotSerialized)\n
   {\n
       Return (0x0F)\n
   }\n
   Method (_CRS, 0, NotSerialized)\n
   {\n
       Return (CRS)\n
   }\n
}
end

and just add this section into that patch?

        IRQNoFlags ()\n
           {11}\n
       IRQNoFlags ()\n
           {15}\n

 

Or just create a new patch using the link you provided? His code seems to be just a section of the entire patch.

[Cassio]

The "new HPET" patch is for those who don't have the device HPET. If you have it, just it doesn't have IRQs, use the patch posted by oldnapalm (copy/paste into patch - new), it removes IRQs from devices RTC, PIC and TMR, and adds to device HPET.

[stuck]

Awesome, oldnapalms patch worked. But only the USB ports on the right side of the machines work. And now when I put the machine to sleep, it wakes up automatically

Wake reason: GLEN

[Cassio]

Are you using EHCI patch for Sandy Bridge as well?

There's no GLEN device in your DSDT, isn't it GLAN? Try removing method _PRW from it.

[stuck]

I am not using EHCI patch for sandy bridge, just oldnapalm's patch, should I be?

You're right it was GLAN, it was really late last night when I posted, sorry.

I tried to remove the line below GLAN device:

Method (_PRW, 0, NotSerialized)

but tons of errors popped up when compiling.

[Cassio]

Try to apply the EHCI patch for Sandy Bridge, if it still doesn't work, remove all the method _PRW from GLAN

            Method (_PRW, 0, NotSerialized)
           {
               If (WOLE)
               {
                   Return (Package (0x02)
                   {
                       0x0D, 
                       0x04
                   })
               }
               Else
               {
                   Return (Package (0x02)
                   {
                       0x0D, 
                       Zero
                   })
               }
           }

GLAN will be only

        Device (GLAN)
       {
           Name (_ADR, 0x00190000)
       }

[stuck]

Once again Cassio, you are awesome.

Thank you very much for your help and Oldnapalm's help.

Removing the method (_PRW section as you suggested got arid of the GLAN reason for wake up. But EHC1 and EHC2 popped up as wake reasons. So I followed your advice and patched the EHCI patch for sandy bridge laptop and Voila the sleep works again.

I still don't have the left side USB's working. I don't know if that is a DSDT thing or a Hardware thing?

[Cassio]

Are all your USB ports EHCI or do you have UHCI too? If some are UHCI you can try this patch as well (from Sandy Bridge section).

You should apply USB patches before removing _PRW from GLAN, maybe it's not necessary.

[stuck]

Are all your USB ports EHCI or do you have UHCI too?.

How would I know?

[Cassio]

http://Olarila.com/forum/viewtopic.php?f=18&t=63

or just open the UHCI patch and click "preview", if you see something else than comments in green, you have UHCI ports.

[stuck]

or just open the UHCI patch and click "preview", if you see something else than comments in green, you have UHCI ports.

Tried it, unfortunately no changes with this patch.

BTW, this post here http://www.insanelymac.com/forum/index.php?showtopic=247900&view=findpost&p=1746833

seem to suggest that my wifi will work with this enabler

http://www.osx86.net/view/556-enabler_wireless_adapter_bcm43xx.html

if DSDT was fixed. But I'm not sure what fix he is referring to?

[Cassio]

Check in IORegistryExplorer if you have any other EHCI hub than the 2 affected by the EHCI patch.

About the Broadcom wireless I don't know about any DSDT patch for it.

[stuck]

Thanks for your continued help Cassio..

And HAPPY NEW YEAR to everyone here.

IORegistryExplorer show both hubs affected by the EHCI patch,

Both physical ports on the right are under:

EHC1@1D
 EHCI Root Hub Simulation@1D
 HubDevice@1d100000
   AppleUSBHub
   IOUSBInterface@0
             @1d110000
             @1d120000

There's another device here:

EHC2@1A
 EHCI Root Hub Simulation@1A
 HubDevice@1a110000
   AppleUSBHub
   IOUSBInterface@0
 VendorSpecificDevice@1a110000

But no matter which USB port I plug into, nothing appears under this device...

the USB related messages in kernel.log are:

AppleUSBCDC: start - initDevice failed

and occasionally

USBMSC Identifier (non-unique): 3S9B0A44004C 0x3f0 0x3207 0x100

 

Also, my wifi card does not show up anywhere in IORegistryExplorer nor System report. It shows up in ubuntu with:

lspci -n
07:00.0 0280: 14e4:4727 (rev 01)
lspci 
07:00.0 Network controller: Broadcom Corporation BCM4313 802.11b/g/n Wireless LAN Controller (rev 01)

So I'm guessing the OS doesn't even see the card. Would that be a DSDT problem?

[oldnapalm]

There's no solution for 14e4:4727 in Mac OS, AFAIK.

You can try forcing to enable the EHCI device using Function Disable Register

http://www.projectosx.com/forum/index.php?showtopic=851

into all code_regex Offset\s*\(0x341A\) replace_matched
begin
,   8, \n
EH2D,   1, \n
Offset (0x341A)
end;
into method label _INI insert
begin
Store (Zero, EH2D)
end

[stuck]

Thank you Oldnapalm,

the patch didn't work... Still have the same USB error messages in kernel log.

I wonder if "VendorSpecificDevice@1a110000"

could be connected to the fingerprint reader that is built in to the laptop?

Now in IORegistryExplorer it shows 2 of what ever was there before under EHC2

HP TrueVision HD@1a12000
HP TrueVision HD@1a12000
VenderSpecificDevice@1a110000
VenderSpecificDevice@1a110000

[stuck]

Hi Cassio and Oldnapalm,

I'm interested in putting everything I've done into a single patch and submit it here to help others with my same laptop.

The existing patches are simple enough, I'll just put them into a single txt file separated by ;

but few other changes were required that I don't have a patch for. Could you help me with that?

ORIGINAL UNMODIFIED CODE;

           Method (PAPR, 0, NotSerialized)
           {
               If (ECON) {}
               Else
               {
                   Return (Zero)
               }
           }


FIXED CODE;

Method (PAPR, 0, NotSerialized)
           {
               If (ECON) {}
               Else
               {
                   Return (Zero)
               }
           Return (Zero)
           }


THE ALTERNATIVE FIXED CODE BELOW ALSO WORKS TOO;

Method (PAPR, 0, NotSerialized)
                     {
                            Return (Zero)
                     }

And this one

ORIGINAL UNMODIFIED CODE;

               Method (_CRS, 0, NotSerialized)
               {
                   If (IGDS)
                   {
                       Return (CRS)
                   }
               }
           }
       }
   }


FIXED CODE;

     Method (_CRS, 0, NotSerialized)
               {
                   If (IGDS)
                   {
                       Return (CRS)
                   }

                   Return (Buffer (One)
                   {
                       0x00
                   })
               }
           }
       }
   }

Then there's also the Sleep GLAN issue discussed earlier

            Method (_PRW, 0, NotSerialized)
           {
               If (WOLE)
               {
                   Return (Package (0x02)
                   {
                       0x0D,
                       0x04
                   })
               }
               Else
               {
                   Return (Package (0x02)
                   {
                       0x0D,
                       Zero
                   })
               }
           }

Fixed code

       Device (GLAN)
       {
           Name (_ADR, 0x00190000)
       }