Jump to content

Guide for Installing OS X on Lenovo IdeaPad Y510p

ahmed_ais

By ahmed_ais,
in Installation Guides

 Share
1,428 posts in this topic

Recommended Posts

ahmed_ais
ahmed_ais
Posted
  • Author
Posted

Yeah I expected some bugs to appear from a bios swap. But for everything related to patched kexts, I still don't understand how different bios broke them. It was my own bios, patched to remove the whitelist. Not from someone else's machine. But OSX was installed while I was running v3.08, and freaked out when I was on modified 3.05.

 

I restored my backed up bios as a test, and osx remained broken.

 

Very weird. I understand how ACPI would break, I even expect it. But patched and injected kexts shouldn't care about bios settings, should they?

 

But knowing that intruder16 also runs the modified bios, I think I may give a fresh install a try with the modified bios.

 

First, I did not read much about the floating around modded BIOS. But AFAIK, you replace your BIOS with modded one that have the whitelisting removed. I don't have any idea that you can patch your own and if that was possible then the chance tables are changed is tiny (for a small change like removing the whitelisting). Clover settings should not break nor the patched kexts either. Try a fresh install as you suggested and hopefully it will work fine.

Link to comment
Share on other sites
wikkid
wikkid
Posted
Posted

First, I did not read much about the floating around modded BIOS. But AFAIK, you replace your BIOS with modded one that have the whitelisting removed. I don't have any idea that you can patch your own and if that was possible then the chance tables are changed is tiny (for a small change like removing the whitelisting). Clover settings should not break nor the patched kexts either. Try a fresh install as you suggested and hopefully it will work fine.

 

Yeah it's pretty snazzy. You make a backup of your own bios after downgrading to 3.05, and then patch it to remove whitelist and unlock a bunch of other options that aren't normally there.

 

Will try the fresh install at some point today/tonight.

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

So I reflashed modded bios and installed the new card.

 

Turns out the settings/kext you have mentioned in the guide work on this card as well. Full ac support and speed.

 

Have to say I'm very happy with the results! :)

 

https://www.dropbox.com/s/texgycuzktcon3z/2015-07-16%2020.25.46.jpg?dl=0

 

Once again, thank you for your advise and creating this thread. :)

Great! Enjoy your mac!

Link to comment
Share on other sites
wikkid
wikkid
Posted
Posted

Dang it. I thought I was done with questions, but alas, I do have one more.

 

The auto patcher and ssdtPRGen.sh boith produce an SSDT.aml file. (no -0 or -1, etc).

 

Would I rename the one from ssdtPRGen.sh, since I've already done the auto patcher, to like SSDT-2.aml? Or did auto patcher do what ssdtPRGen.sh does and the file is essentially the same?

 

I'm not sure if the ssdt numbering has any importance,

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

 

 

Dang it. I thought I was done with questions, but alas, I do have one more.

 

The auto patcher and ssdtPRGen.sh boith produce an SSDT.aml file. (no -0 or -1, etc).

 

Would I rename the one from ssdtPRGen.sh, since I've already done the auto patcher, to like SSDT-2.aml? Or did auto patcher do what ssdtPRGen.sh does and the file is essentially the same?

 

I'm not sure if the ssdt numbering has any importance,

Auto-patcher creates this SSDT for you so you may not need to use the script again unless you want a new SSDT then it should overwrite the one you get from auto-patcher. You can open both with MaciASL to learn about them and you will see that both are defined as CPUPM (CPU Power Management).

 

SSDT numbering is, sort of, important because this is how the tables are ordered. Clover should read the tables in the correct order and this is controlled using SortedOrder section in config.plist (check the attached file in post #1). You can mix the numbers, however, as long as you put the correct order in config.plist.

Link to comment
Share on other sites
  • 2 weeks later...
TarvoMaesepp
TarvoMaesepp
Posted
Posted

Heey, today I got my copy of windows 10 and now after install I can't see clover boot manager. What whould be the best thing to do? Boot with USB and reinstall Clover? Or is my boot order wrong? Which one should be the first one? I have so many options in the bootlist.

Link to comment
Share on other sites
wikkid
wikkid
Posted
Posted

I haven't done the Windows 10 upgrade yet, but I'm guessing reinstalling clover from usb should do the trick. I know I've blown out my clover boot menu a few times tinkering and was always able to bring it back with everything in tact. :)

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

Heey, today I got my copy of windows 10 and now after install I can't see clover boot manager. What whould be the best thing to do? Boot with USB and reinstall Clover? Or is my boot order wrong? Which one should be the first one? I have so many options in the bootlist.

 

I have not upgraded yet to Windows 10 (still waiting) so take this with a grain of salt. I'm not sure what Windows 10 installer do with the EFI partition but certainly it will override Windows 8 EFI boot files. But usually the newest EFI entry will be listed first so try to check BIOS to see the available entries maybe Clover is still there and you have to set it first. If not, you may need to reinstall Clover again.

 

I was aware that I may have some hiccups with bootloader (although not sure as I have not tried any build of Windows 10). Therefor, I backed-up my EFI folder and restored the original partition data (without Clover) so Windows 10 installer do not complain (if it was ever going to). Then I was planning to just copy Clover folder into EFI partition after Windows 10 is installed and hopefully it would work fine.

Link to comment
Share on other sites
TarvoMaesepp
TarvoMaesepp
Posted
Posted

I have not upgraded yet to Windows 10 (still waiting) so take this with a grain of salt. I'm not sure what Windows 10 installer do with the EFI partition but certainly it will override Windows 8 EFI boot files. But usually the newest EFI entry will be listed first so try to check BIOS to see the available entries maybe Clover is still there and you have to set it first. If not, you may need to reinstall Clover again.

 

I was aware that I may have some hiccups with bootloader (although not sure as I have not tried any build of Windows 10). Therefor, I backed-up my EFI folder and restored the original partition data (without Clover) so Windows 10 installer do not complain (if it was ever going to). Then I was planning to just copy Clover folder into EFI partition after Windows 10 is installed and hopefully it would work fine.

Huhh, thank you. I set my clover to first and can boot again.

 

I also wanted to ask one thing. I have atheros ar5b125 wireless card in my drawer can I use this in our Lenovo laptop and get wifi working on OS X. Also is it better than 802.11n?

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

Huhh, thank you. I set my clover to first and can boot again.

Good to know its fine!

 

... I have atheros ar5b125 wireless card in my drawer can I use this in our Lenovo laptop and get wifi working on OS X ...

Sorry, this card is unsupported and will not work with OS X (according to this).

Link to comment
Share on other sites
  • 3 weeks later...
  • 3 weeks later...
notten
notten
Posted
Posted

 

I bought the same card from the same eBay link and I can't get it to work with the FakePCIID kext from the first page. I put it in S/L/E and Clover EFI locations and booted with cleared cache and with and without injected kexts and nothing works. The wireless chip isn't recognized. Any suggestions on how I can diagnose the issue? OS X dmesg is a lot less helpful than Linux's.

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

I bought the same card from the same eBay link and I can't get it to work with the FakePCIID kext from the first page. I put it in S/L/E and Clover EFI locations and booted with cleared cache and with and without injected kexts and nothing works. The wireless chip isn't recognized. Any suggestions on how I can diagnose the issue? OS X dmesg is a lot less helpful than Linux's.

 

First, congratulations!

Now you don't need to put the kext in two locations. Putting it in S/L/E only is enough. Can you confirm you installed both FakePCIID.kext and FakePCIID_BCM94352Z_as_BCM94360CS2.kext? it will not work with only one of them.

Link to comment
Share on other sites
notten
notten
Posted
Posted

First, congratulations!

Now you don't need to put the kext in two locations. Putting it in S/L/E only is enough. Can you confirm you installed both FakePCIID.kext and FakePCIID_BCM94352Z_as_BCM94360CS2.kext? it will not work with only one of them.

Randomly started working. I had both Kexts. I was trying to use the Wireless Manager that came with another wireless card, and that doesn't work with this one. The built-in OS X one works fine. I'm back on again. Thank you!

Link to comment
Share on other sites
notten
notten
Posted
Posted

The only thing I think is missing from this laptop being a complete Hackintosh is the fact that I have to remove my 2nd GPU to run OS X. I would prefer to obviously have both in and use an NVIDIA GPU in OS X, but I would take having both in and using the Intel GPU in OS X. However, I have to take my 2nd GPU out every time I want to boot OS X, and put it back in when I want to boot Linux or Windows to play games. I guess I could play around with the BIOS hack to see if some option gets the Intel enabled with both GPUs in or something.

 

 

EDIT: In this video at 2:57 I see Primary Display settable to Auto, IGFX, PEG, PCI, SG. I wonder if OS X + Primary Display set to PEG = working NVIDIA in Mac OS X.

 

EDIT2: Ooops, didn't know it would embed the video. I moved the video above the edits.

Link to comment
Share on other sites
notten
notten
Posted
Posted

Well, I guess getting the NVIDIA card is never going to happen. However, it would be nice to at least get the Intel card working 100%. That is, it would be nice if I didn't have to take apart my computer to switch OSes. Also, it wouldn't hurt if I could actually use 4K output. I'd say Y510P Hackintosh is like 80% video. It would be 90% with 4K output. 100% with Intel working without having to remove the 2nd NVIDIA. and 150% if one of the NVIDIA cards worked. 175% if the 2nd NVIDIA worked as CUDA processing. 200% if both NVIDIAs worked in SLI.

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

Well, I guess getting the NVIDIA card is never going to happen. However, it would be nice to at least get the Intel card working 100%. That is, it would be nice if I didn't have to take apart my computer to switch OSes. Also, it wouldn't hurt if I could actually use 4K output. I'd say Y510P Hackintosh is like 80% video. It would be 90% with 4K output. 100% with Intel working without having to remove the 2nd NVIDIA. and 150% if one of the NVIDIA cards worked. 175% if the 2nd NVIDIA worked as CUDA processing. 200% if both NVIDIAs worked in SLI.

 

SLI is never going to happen as Apple never had this configuration (even if OS X can work with multiple GPUs they will work individually). Using the 2nd Nvidia GPU for CUDA can only possible if the 1st is connected directly to display which does not appear to be the case here. Since we have Optimus GPU, don't expect that to work with OSX at all. So now we are left with how to improve OSX on Y510p from 80% to 100% as you put it.

 

To have 4K output I believe you need to use the proper cable (HDMI 1.4a) and not just any HDMI cable. I'm not saying that it will work though, but I have read somewhere that HD4600 can output to 4K under OSX at least with SwitchResX but proper cable is required. Now for removing the 2nd Nvidia GPU, the following does not apply if you use it for CUDA programming. AFAIK, there are not much games out there that benefit from SLI so I toke mine out. I bought a HDD caddy and put my HDD in there. I bought SSD and put it inside to be main disk for operating systems. Now I don't need to do any switching unless I decided to play a game that can significantly get more FPS using SLI which is rare anyway.

Link to comment
Share on other sites
ahmed_ais
ahmed_ais
Posted
  • Author
Posted

By the way, I still have hopes of getting both Nvidia GPUs working with OSX. I read many times that when the 2nd Nvidia GPU is plugged-in, the 1st Nvidia GPU can no longer work as Optimus. The Optimus technology only work with Intel HD for general purpose graphics tasks and Nvidia GPU only for graphic insensitive tasks. But when Intel HD is absent (which is the case when the 2nd GPU is plugged-in) there is no place for Optimus as Nvidia have to be activated all the time. Windows and Linux agree with that but there is something mysterious with OSX.

 

Deep inside of my mind I still believe we can get it to work. But I lack knowledge and most of the experienced guys ignores that case and quickly jump to conclusions like "HEY .. this is Optimus, dead end!". Maybe one day we will figure out something but for now I have no clue ... :(  

  • Like 1
Link to comment
Share on other sites
notten
notten
Posted
Posted

SLI is never going to happen as Apple never had this configuration (even if OS X can work with multiple GPUs they will work individually). Using the 2nd Nvidia GPU for CUDA can only possible if the 1st is connected directly to display which does not appear to be the case here. Since we have Optimus GPU, don't expect that to work with OSX at all. So now we are left with how to improve OSX on Y510p from 80% to 100% as you put it.

 

To have 4K output I believe you need to use the proper cable (HDMI 1.4a) and not just any HDMI cable. I'm not saying that it will work though, but I have read somewhere that HD4600 can output to 4K under OSX at least with SwitchResX but proper cable is required. Now for removing the 2nd Nvidia GPU, the following does not apply if you use it for CUDA programming. AFAIK, there are not much games out there that benefit from SLI so I toke mine out. I bought a HDD caddy and put my HDD in there. I bought SSD and put it inside to be main disk for operating systems. Now I don't need to do any switching unless I decided to play a game that can significantly get more FPS using SLI which is rare anyway.

 

So, I have the proper cable. I use Linux and Windows at 4K using the NVIDIA GPUs all the time on this monitor. It's my work config. I actually have two identical Y510Ps, one for work, and one for home. Anyway, I tried the following link:

 

http://www.tonymacx86.com/graphics/117665-intel-hd-4600-4k-3840-2160-seiki-39-4k-display-3.html

 

See post #30. Using the mac-pixel-clock-patch on my system caused my HDMI display to cease working altogether. Booting with it plugged in or unplugged and plugging it in afterwards both failed. It did work once, but it started failing after I started to use SwitchResX. I tried their SwitchResX config with 3840x2160 at 14Hz and 24Hz and 30Hz. All fail for me, even without the patch. With the patch, even 1080p failed. Also, yeah, I wouldn't expect SLI to ever work. However, I would like CUDA to work. I do use my 2nd GPU for SLI on Windows and CUDA development on Linux. If we ever get NVIDIA to work on OS X, it would be nice to get CUDA and/or PhysX offloading on the 2nd GPU. BTW, this Laptop is amazing for Linux CUDA development. Since the 2nd GPU isn't actively drawing, you can do CUDA development with breakpoints in Linux. Obviously, this would never work on the main GPU or if the 2nd GPU was drawing to the screen since adding breakpoints to an active GPU would pause the display driver. However, in Linux with 1 GPU drawing I can do full CUDA development on the 2nd GPU. And when I want to run the code, I can take full advantage of two GPUs running CUDA code. If anyone needs a CUDA development machine, Linux + dual NVIDIA on this machine is king. 

By the way, I still have hopes of getting both Nvidia GPUs working with OSX. I read many times that when the 2nd Nvidia GPU is plugged-in, the 1st Nvidia GPU can no longer work as Optimus. The Optimus technology only work with Intel HD for general purpose graphics tasks and Nvidia GPU only for graphic insensitive tasks. But when Intel HD is absent (which is the case when the 2nd GPU is plugged-in) there is no place for Optimus as Nvidia have to be activated all the time. Windows and Linux agree with that but there is something mysterious with OSX.

 

Deep inside of my mind I still believe we can get it to work. But I lack knowledge and most of the experienced guys ignores that case and quickly jump to conclusions like "HEY .. this is Optimus, dead end!". Maybe one day we will figure out something but for now I have no clue ... :(  

Maybe more information on the display configuration in Linux will explain what's happening. I see when booting with OS X and NVIDIA enabled that all the displays that are attached are not usable. And OS X won't boot. Can we inject fake EDID information? Maybe try to dump the EDID for the internal display from Linux and compare to the EDID we get in OS X and Windows? I'm also very hopeful that this configuration will eventually work. Unlike all the other laptops that can't disable their IGPU, we have that ability with the SLI config. I really hope that we can take advantage of that so we can get the internal NVIDIA GPU to run.

Link to comment
Share on other sites
 Share
Go to topic listing
×
×
  • Create New...