Lenovo Yoga 2 13 Mavericks Install

[dudenesspa]

It is a bit weird. I've never seen "Waiting for DSMOS" *after* "DSMOS has arrived"...

Did you try plugging in an HDMI monitor?

Unfortunately I don't have that kind of cable. I will have to order one.

 

[dudenesspa]

Hi again,

Just doing some research and found that my hd4200 graphics with id 0a1e uses GT1 while the fakeid I've ben using 0a16 is GT2. According to Pike, there are two other GT1 device ids within appleintelhd5000Graphics - 0c06 and 0406. I've tried both of these and they work much like 0a16 did prior to azul patches. Is it possible to create patches for Azul and hd5000graphics kexts for either of these to test them out? Unlike 0a1e, these device ids do exist in hd5000graphics kext per Pike.

https://pikeralpha.wordpress.com/2013/06/page/2/

Is thee anywhere I can learn how to make these patches? I don't get where things like "QYjE6yM=" come from. Is this base 64?

 

[RehabMan]

Hi again,

Just doing some research and found that my hd4200 graphics with id 0a1e uses GT1 while the fakeid I've ben using 0a16 is GT2. According to Pike, there are two other GT1 device ids within appleintelhd5000Graphics - 0c06 and 0406. I've tried both of these and they work much like 0a16 did prior to azul patches. Is it possible to create patches for Azul and hd5000graphics kexts for either of these to test them out? Unlike 0a1e, these device ids do exist in hd5000graphics kext per Pike.

https://pikeralpha.wordpress.com/2013/06/page/2/

I wondered about gt1/gt2/gt3. It is possible to construct a patch (for hd5000.kext) that would allow you to substitute any id.

find: 418b86e40e0000
repl: b88680xxxx9090

Where xxxx is the id you want to fake (byte reversed).

For example (for your 0c06)...

Find: 418b86e40e0000
Repl: b88680060c9090

Is thee anywhere I can learn how to make these patches? I don't get where things like "QYjE6yM=" come from. Is this base 64?

If you understand x86 assembly, you can create your own patches... 'otool -tVj' is useful.

If you view the plist in a plist editor, you can see the hex associated with the base64.

 

[dudenesspa]

Yeah I just rechecked those kexts and neither of the ids are in the primary match section......I also saw that Pikes post was from June of last year so maybe those ids were in the kextSo in the dp of Macericks? Maybe a rollback would work?

 

[RehabMan]

Yeah I just rechecked those kexts and neither of the ids are in the primary match section......I also saw that Pikes post was from June of last year so maybe those ids were in the kextSo in the dp of Macericks? Maybe a rollback would work?

Yeah, I didn't see it in the code either (sometimes even though not in Info.plist, there can be support in the code), so I don't expect it to work. I was just showing how you can construct a patch for any id.

It could be that Yosemite has stripped all gt1 support...

You might try Mavericks instead, assuming you downloaded it when it was available. Mavericks support for Haswell graphics was much more complete.

 

[RehabMan]

Yeah, I didn't see it in the code either (sometimes even though not in Info.plist, there can be support in the code), so I don't expect it to work. I was just showing how you can construct a patch for any id.

It could be that Yosemite has stripped all gt1 support...

You might try Mavericks instead, assuming you downloaded it when it was available. Mavericks support for Haswell graphics was much more complete.

Here is something to try. Maybe all the gt1 code is not removed... just the code that sets it for the various gt1 devices... ???


fake-id 0xa16, and then force gt1 detection for it...

; fake 0a16
find: 418b86e40e0000
repl: b88680160a9090

; force gt1 for 0a16
find: 41c786ec0e000001000000
repl: 41c786ec0e000000000000

 

[dudenesspa]

Unfortunately these did not work. Fake id patch causes the no progress bar/black screen with cursor. Forcing gt1 patch had no effect.

 

[RehabMan]

Unfortunately these did not work. Fake id patch causes the no progress bar/black screen with cursor. Forcing gt1 patch had no effect.

You may need to test in conjunction with all ig-platform-ids... as it may be a combination.

Also, just to be clear about it... you need both patches at the same time. They work together and are not intended to be tested independently.

Or it may just not work at all.

It would probably be better to start with Mavericks where you know that gt1 devices are supported.

 

[dudenesspa]

Hi RehabMan,

So I've given up on QE/CI for now. I tried replacing the AppleIntelHD5000Graphics and AppleIntelFramebufferAzul kexts with the ones from Mavericks 10.9.2 that contained the 0C06 ID and well, as I'm sure you expected and I kind of knew, it really messed up my system. I will try to swap out different ig-platform-ids soon with the patches and see if I get beyond the black screen/cursor.

I decided to move on to other things. Successes: Speedstepping with SSDT, Display Brightness, Wi-Fi, iMessage/Facetime, audio.

Still need: Sleep and Battery.

I followed the instructions from earlier in this thread to get Battery working - OSYS fix, Yoga 13 battery patch, and OS Check fix added to DSDT. My battery icon still has an X.

As far as sleep, the laptop goes to sleep but then immediately reboots. I did pmset -g assertions and have all zeros in the list but have kernel assertions 0x4=USB with two listed: EHC1 and XHC.

Wondering if you could take a look at my DSDT and/or config.plist and let me know if there is anything that jumps out for either issue. I included fresh disassembled SSDTs and DSDT as well as my edited ones. SSDT-6 contains GFX0. I have also included an ioreg with DSDT installed. Let me know if there would be anything that would help to troubleshoot and of course I appreciate the help.

 

[RehabMan]

Hi RehabMan,

So I've given up on QE/CI for now. I tried replacing the AppleIntelHD5000Graphics and AppleIntelFramebufferAzul kexts with the ones from Mavericks 10.9.2 that contained the 0C06 ID and well, as I'm sure you expected and I kind of knew, it really messed up my system. I will try to swap out different ig-platform-ids soon with the patches and see if I get beyond the black screen/cursor.

FYI: You have CI, just no QE.

I decided to move on to other things. Successes: Speedstepping with SSDT, Display Brightness, Wi-Fi, iMessage/Facetime, audio.

Still need: Sleep and Battery.

I followed the instructions from earlier in this thread to get Battery working - OSYS fix, Yoga 13 battery patch, and OS Check fix added to DSDT. My battery icon still has an X.

Did you install ACPIBatteryManager.kext?

Use the debug version of ACPIBatteryManager.kext to determine which DSDT methods are failing.

As far as sleep, the laptop goes to sleep but then immediately reboots. I did pmset -g assertions and have all zeros in the list but have kernel assertions 0x4=USB with two listed: EHC1 and XHC.

You probably have two problems:
- instant wake
- reboot after wake

For instant wake: https://github.com/RehabMan/Laptop-DSDT-Patch
Apply: "7-series/8-series USB"
Also, install GenericUSBXCHI.kext, and use kernel flag -gux_defer_usb2
https://github.com/RehabMan/OS-X-Generic-USB3

For reboot after wake... it is probably a DSDT issue. Perhaps you're dropping an SSDT that is needed...

Download patchmatic: https://github.com/RehabMan/OS-X-MaciASL-patchmatic.
Extract the 'patchmatic' binary from the ZIP. Copy it to /usr/bin, such that you have the binary at /usr/bin/patchmatic.

In terminal,

rm -Rf ~/Downloads/RehabMan
mkdir ~/Downloads/RehabMan
cd ~/Downloads/RehabMan
patchmatic -extract

Post contents of Downloads/RehabMan directory (as ZIP).