Jump to content

Intel HD4400 Full Qe/Ci in Yosemite 10.10.2/10.10.3


Mirone
 Share

99 posts in this topic

Recommended Posts

Mirone. Do you know any patch that fixes the problem of "progress bar" in the second stage of bootlogo Yosemite?

 
Here, where the "bar" appears in the lower right corner. And this patch there, is for the apple logo and does not affect the "progress bar".
 
This problem is only in the 2nd stage of the boot.
 
Thanks al friend. For any correction, if any.
Link to comment
Share on other sites

 

Mirone. Do you know any patch that fixes the problem of "progress bar" in the second stage of bootlogo Yosemite?

 
Here, where the "bar" appears in the lower right corner. And this patch there, is for the apple logo and does not affect the "progress bar".
 
This problem is only in the 2nd stage of the boot.
 
Thanks al friend. For any correction, if any.

 

 

You try set your resolution in config.plist?

Link to comment
Share on other sites

 

Mirone. Do you know any patch that fixes the problem of "progress bar" in the second stage of bootlogo Yosemite?

 
Here, where the "bar" appears in the lower right corner. And this patch there, is for the apple logo and does not affect the "progress bar".
 
This problem is only in the 2nd stage of the boot.
 
Thanks al friend. For any correction, if any.

 

Off topic!!

look here for this solution: http://www.insanelymac.com/forum/topic/298808-can-we-get-back-the-apple-logo-on-boot/page-9

is only partly because I can add a second is fake pciid

I think this method is quite different from using a FakePciID.
you think would be a good idea to mix the two?

for me it would confuse people.

  • Like 2
Link to comment
Share on other sites

Mirone,

 

Replacing the device ID in AppleIntelFramebufferAzul and AppleIntelHD5000Graphics is not enough for proper QE/CI.

You will notice that Safari crashes on flash websites and preview has issues with large images.

 

Additionally some programs will 'lag' from time to time, this is due to the OpenCL compiler crashing.

It turns out the OpenCL compiler also reads the device ID from the PCI configuration space.

But even after patching that Safari will still experience issues.

 

Together with Rehabman I have developed FakePCIID, which allows 'spoofing' the result values from the PCI configuration space.

Due to this no more patching at all is needed, once spoofed the device is actually appearing as 0x04128086 to OS X, hence all drivers will work natively.

 

More information is here:

https://github.com/the-darkvoid/OS-X-Fake-PCI-ID

  • Like 3
Link to comment
Share on other sites

darkvoid,

 

Thanks for all the information, I did some testing in

my laptop everything works fine,
I do not have any of these problems you reported.
I will give a chance to FakePCIID and let u know the results.
 
it seems that your kext is based on the same principle of clover.
Link to comment
Share on other sites

Mirone,

 

On a Dell XPS 9333 with Intel HD4400, crashes were experienced. However on a Lenovo U430 with Intel HD4400 all was fine.

So for Intel HD4400 it does not always crash, however for Intel HD4600 it certainly does.

 

To elaborate on the link you posted:

 

Both use the term FakePCIID, but the concept is different.

FakeID on clover injects a device ID into the DSDT to make a device match against a driver in OS X.
This works in some of the cases, however there are the number of drivers which read the vendor and device ID directly from the device PCI configuration space.

 

Since the PCI configuration space is a read-only memory mapped region directly on the PCI device Clover is unable to change these results.

This has resulted in kext patching the ID's in kexts like you are doing in the first post. The patching has been done for Intel HD4600, Broadcom LAN cards, Intel HDMI Audio etc etc.

 

I previously had a chat with Slice if Clover would be able to do such a "spoofing", but he said as a bootloader it would not be possible and would have to be done in OS X.

 

This is where FakePCIID comes in. It attaches to the IOPCIDevice and overrides its C++ virtual methods, allowing it to intercept communication between the original driver and the IOPCIDevice.

When the information we want to fake is requested, its returning the fake information instead of the information from the IOPCIDevice memory space.

Link to comment
Share on other sites

darkvoid,

 

I understand everything you said, until about memory mapping,

but this tutorial is based on my experience with my hardware,

applying only to HD4400, although FakePCIID.kext might have a better outcome,

my method for me is satisfying the other user should try both and decide which one gives the best results.

Link to comment
Share on other sites

  • 2 weeks later...
  • 2 weeks later...

my method for me is satisfying the other user should try both and decide which one gives the best results.

 

Until a simpler procedure for the fix comes along, this works for me on my HD4400 (TP Yoga S1).  Only one problem, I miss having HDMI out - connecting an external display causes crashes. 

Link to comment
Share on other sites

  • 2 weeks later...

Is anyone having any issues with Google Chrome glitching out on them?

 

Everything is working fine, especially with the patches and the DSDT edits, but for some reason , I get screen tearing on Google Chrome, and on Chromium - and then Canary shows just a white screen, but the links are hidden there if you click around

 

 

Is this a Chrome- specific issue --- or does anyone have any light they can shed on the subject?

 

EDIT:

 

it appears I'm getting the tearing on AirMail Beta as well..

 

Also- Neither the AppleIntelHD4000/5000 kexts are loading and neither is AppleIntelAzulCapri

Link to comment
Share on other sites

  • 3 weeks later...

 

With the update to version Yosemite 10.10.2, some users It has a IntelHD4400 lost their Qe/Ci the steps outlined below describe how to activate these  features again.

 

         Change Log

   

             v1.0  1/27/2015 - Yosemite 10.10.2 - Release

             v1.1  2/10/2015 - attachicon.gifPatched_IntelHD4400.sh.zip script, added, (correction in binaries+OpenCL only)

 

 

 

Thanks for your work,

 Unfortunately my HD  4400 only partly function,  is recognized (Intel hd graphics 4400 1536 MB), but with starts as in the normal enterprise flickers again and again the picture???

 

With films the picture is also always transferred shown (picture lag) ??

 

On 10.10.1 the graphics card showed everything correctly.

 

Intel i5- 4210u, HD 4400, Yosemite 10.10.3, Enoch

Link to comment
Share on other sites

 Share

×
×
  • Create New...