OS X Mavericks 10.9.3 to dev

[Wayang-NT]

OS X Mavericks 10.9.3 (13D12) seeded today 

[mnfesq]

 

No problems installing for me.  Unfortunately, I do not use the drivers which are being modified.

[theconnactic]

Which are... ?

[mnfesq]

Which are... ?

 

The focus areas mentioned in the post above - graphics drivers and audio.  I can't use my AMD Radeon 7850M GPU because HP doesn't let you enable it in the BIOS and I use the VoodooHDA driver so that I can get my Beats audio subwoofer working.  No one has yet managed to get the subwoofer recognized using the AppleHDA.kext.

[andrek]

Just updated, everything works as intended.

[wh1sker]

Interested to know what gpu drivers are being modified? i have a 7950 that works great (10.9.1)

[Riley Freeman]

Interested to know what gpu drivers are being modified? i have a 7950 that works great (10.9.1)

 

AMD7000Controller is 1.22.12 13759 if that's any help.

[nawcom]

The new NVIDIA kexts are causing some interesting issues. I've verified it on my PC with my 2 GTX 770s (GK104, uses NVDAStartup.kext to inject needed values, NVCAP, etc) which didn't surprise me since I've had to deal with issues with it in 10.9.2, though that has to do with the new IOPCIFamily kext and was able to get around that and get them working again. But when I installed it on my laptop (GT 525m) which uses Chameleon's GraphicsEnabler which has always worked fine, I had the same issue - the laptop display was gone by the time the desktop loaded up. For GTX 770, DVI #1 is detected, DVI #2 is not detected at all, and HDMI is detected fine. This behavior matches both GTX 770 cards. Before all three get picked up just fine for both cards.

 

I could plug in my second DVI display to DVI #1 on my second GTX 770 and have them detected fine with the new kexts, but that will screw up using SLI in Windows for gaming, and I'm not willing to move around cables in the back of my PC whenever I boot OS X or Windows for the sake of things working. And of course there's no fix like this with the case of my laptop with GT 525m losing its screen.

 

Anyways, this kind of behavior usually happens when the injected NVCAP value is incorrect and as a result not all display ports are correctly identified. I verified that the NVCAP values as well as the rest of the injected values match up with GTX 770 and GT 525m in both 10.9.2 and 10.9.3 beta. Rolling back the NVIDIA kexts to the 10.9.2 versions fixes the display issues.

 

If anyone is having similar issues, the easy fix is to roll back all the NVDA* kexts your card uses to 10.9.2 release versions, though that isn't a good long-term solution. I made sure performance wasn't affected by rolling back. OpenCL and VDA still work fine as before. I'll see if this persists in the upcoming beta releases. Also be sure and mention it here if someone is having similar NVIDIA issues, because if I'm the only one having issues, then I guess I'm lucky, hehe.

[Dr. Hurt]

Does anyone notice faster transitions between workspaces, and switching to fullscreen?

[PoisonApple666]

The new NVIDIA kexts are causing some interesting issues. I've verified it on my PC with my 2 GTX 770s (GK104, uses NVDAStartup.kext to inject needed values, NVCAP, etc) which didn't surprise me since I've had to deal with issues with it in 10.9.2, though that has to do with the new IOPCIFamily kext and was able to get around that and get them working again. But when I installed it on my laptop (GT 525m) which uses Chameleon's GraphicsEnabler which has always worked fine, I had the same issue - the laptop display was gone by the time the desktop loaded up. For GTX 770, DVI #1 is detected, DVI #2 is not detected at all, and HDMI is detected fine. This behavior matches both GTX 770 cards. Before all three get picked up just fine for both cards.

 

I could plug in my second DVI display to DVI #1 on my second GTX 770 and have them detected fine with the new kexts, but that will screw up using SLI in Windows for gaming, and I'm not willing to move around cables in the back of my PC whenever I boot OS X or Windows for the sake of things working. And of course there's no fix like this with the case of my laptop with GT 525m losing its screen.

 

Anyways, this kind of behavior usually happens when the injected NVCAP value is incorrect and as a result not all display ports are correctly identified. I verified that the NVCAP values as well as the rest of the injected values match up with GTX 770 and GT 525m in both 10.9.2 and 10.9.3 beta. Rolling back the NVIDIA kexts to the 10.9.2 versions fixes the display issues.

 

If anyone is having similar issues, the easy fix is to roll back all the NVDA* kexts your card uses to 10.9.2 release versions, though that isn't a good long-term solution. I made sure performance wasn't affected by rolling back. OpenCL and VDA still work fine as before. I'll see if this persists in the upcoming beta releases. Also be sure and mention it here if someone is having similar NVIDIA issues, because if I'm the only one having issues, then I guess I'm lucky, hehe.

Do you know which kexts specifically?  My GTX 580 is really jerky now.  Kind of frustrating...

Thanks

[nawcom]

Do you know which kexts specifically?  My GTX 580 is really jerky now.  Kind of frustrating...

Thanks

 

GTX 580 uses a GF110 core, so you're most likely using NVDAGF100Hal.kext with NVDAResman.kext for framebuffer. I didn't have to roll back GeForce.kext - right now I'm using 8.2.6 version from the beta with 8.2.4 versions (10.9.2) of NVDAGK100Hal.kext and NVDAResman.kext for my GTX 770s without any problems. I tried only changing one kext or the other of those two but dependency issues arise.

 

Run: 

kextstat | grep NVDA

to find out which ones you're using if I'm wrong.

[PoisonApple666]

Thanks Nawcom...I'll try it.....on my second display off of this card I can't even scroll or swipe to the other screens....kinda frustrating.

no luck....a little smoother but the second monitor still can't scroll between different desktops.

any ideas are greatly appreciated.

 

thanks.

[rhcp011235]

Hello All,

 

Well I found out the hard way, that the nvidia kexts do not play nice on 10.9.3. My old Acer 1920x1080 monitors work fine. But, my PLS 2560x1440 ones don't even get recognized.

 

So, I am assume I am in the boat of installing all 10.9.2 displays drivers. I have the Nvidia GTX 660

 

Would anyone happen to have a .zip of all the drivers for nvidia from 10.9.2? I am kind of stuck here.

Thanks!

 

Also: Is there a way to reversion 10.9.3, so I can just install the 10.9.2 DMG over the 10.9.3 install. Like in the old beta days? When it would fake the version to install a new beta.

[cwestpha]

So far it just looks like a beta update to the graphics drivers, new and expanded support for 4K displays (you can now set a Retina resolution for many 3rd party 4K LCDs), and thats mostly it. I don't see new support for Hawaii (I soooo want to get a 290), though I could be missing something. AppleHDA is unchanged so any audio fixes seem to be elsewhere (10.9.2 patches are fine).

[AREOS]

The new NVIDIA kexts are causing some interesting issues. I've verified it on my PC with my 2 GTX 770s (GK104, uses NVDAStartup.kext to inject needed values, NVCAP, etc) which didn't surprise me since I've had to deal with issues with it in 10.9.2, though that has to do with the new IOPCIFamily kext and was able to get around that and get them working again. But when I installed it on my laptop (GT 525m) which uses Chameleon's GraphicsEnabler which has always worked fine, I had the same issue - the laptop display was gone by the time the desktop loaded up. For GTX 770, DVI #1 is detected, DVI #2 is not detected at all, and HDMI is detected fine. This behavior matches both GTX 770 cards. Before all three get picked up just fine for both cards.

 

I could plug in my second DVI display to DVI #1 on my second GTX 770 and have them detected fine with the new kexts, but that will screw up using SLI in Windows for gaming, and I'm not willing to move around cables in the back of my PC whenever I boot OS X or Windows for the sake of things working. And of course there's no fix like this with the case of my laptop with GT 525m losing its screen.

 

Anyways, this kind of behavior usually happens when the injected NVCAP value is incorrect and as a result not all display ports are correctly identified. I verified that the NVCAP values as well as the rest of the injected values match up with GTX 770 and GT 525m in both 10.9.2 and 10.9.3 beta. Rolling back the NVIDIA kexts to the 10.9.2 versions fixes the display issues.

 

If anyone is having similar issues, the easy fix is to roll back all the NVDA* kexts your card uses to 10.9.2 release versions, though that isn't a good long-term solution. I made sure performance wasn't affected by rolling back. OpenCL and VDA still work fine as before. I'll see if this persists in the upcoming beta releases. Also be sure and mention it here if someone is having similar NVIDIA issues, because if I'm the only one having issues, then I guess I'm lucky, hehe.

 

Hello, you are not alone on this, my GTX 770M started to having this issues since 10.9.2 betas, for me this causes a restart before I can reach the desktop. So I had to roll back NVDAGK100Hal and NVDAResman kexts to 10.9 - 10.9.1 version. This fix the issue in 10.9.2, the new nvidia web drivers for 10.9.2 also works ok.

 

How do you injected NVCAP to your GTX 770? I tried EFI string and DSDT way but the info injected just get ignored.

 

In this new 10.9.3 beta, I have the same issue. We need to investigate this in order to fix it so we can use Apple native driver because this new way to declare ports by the driver is here to stay.

 

Good Luck 

[nawcom]

Hello, you are not alone on this, my GTX 770M started to having this issues since 10.9.2 betas, for me this causes a restart before I can reach the desktop. So I had to roll back NVDAGK100Hal and NVDAResman kexts to 10.9 - 10.9.1 version. This fix the issue in 10.9.2, the new nvidia web drivers for 10.9.2 also works ok.

 

How do you injected NVCAP to your GTX 770? I tried EFI string and DSDT way but the info injected just get ignored.

 

 

The native kext NVDAStartup is actually an injector kext among other things from Apple. (I say other things because I notice that if I prevent it from loading on my laptop with the GT525M Fermi card which also requires GraphicsEnabler to boot up right, I get good ol 1024x768 VESA, so it's depended on by other kexts like the one for Fermi architecture cards.) If you open the kext's binary up in a disassembler app like IDA Pro you can see where it probes your video card for its architecture (Kepler, Maxwell, Tesla, Fermi) and in other functions, it injects its determined NVCAP, NVDA,Display-*, etc. So I don't use GraphicsEnabler or any other form, mainly because

 

a.) It only supports one card, and I have 2 cards I want enabled. I of course can't use SLI in OS X but nonetheless I want both of my cards working separately and enabled.

b.) As of now GraphicsEnabler for Chameleon (I haven't checked on Clover) only supports single cards with the old limit of 2 displays enabled, NVDA,Display-A and NVDA,Display-B. Kepler cards (GK1xx) extend to NVDA,Display-C and NVDA,Display-D, which is how you're able to get more than 2 displays going. 

 

I could inject via DSDT and include NVDA,Display-C and D but why should I if NVDAStartup injects it already? Same for NVCAP values - NVDAStartup.kext generates it. Pretty much if you have a Kepler architecture NVIDIA card you shouldn't have to use any injector for everything to work normally (for the most part, I'm sure there are situations where it isn't perfect, though I haven't run into any myself). And like I said before, the kext is still generating the same NVCAP from 10.9.0 in 10.9.3, it's just deciding to ignore one of my DVI ports.

 

 

 I tried EFI string and DSDT way but the info injected just get ignored.

 

That's because NVDAStartup.kext is overriding those injected entries with its own. Pretty much what would happen if you had one NVCAP value injected via DSDT, but also used an injector kext like good ol' Natit.kext at the same time with a different NVCAP - Natit.kext will override since it loads after DSDT/EFI strings are injected and recognized. In the case for my 2 GTX 770s, if I try and boot with DSDT overrides (which I originally did before I understood NVDAStartup) I end up with a black screen, so using both at the same time can cause issues in the end.

 

Now for my laptop with GT525M that uses GraphicsEnabler - the NVCAP is still the same as before and it worked fine with it, detecting the display and HDMI. Now with the 10.9.3 kexts its screen goes blank, as if the NVCAP is wrong. I roll back the kexts and the screen is back. There's something new in these updated kexts that's doing it. 

 

GTX770M uses some newer core than desktop GTX770, mine is GK104 while yours is GK106 so there's some differences, though I wouldn't expect you to be having issues because of it.

 

There's still a lot to figure out as of now. I'll be sure and post here about any updates while I {censored} around with this.

[rhcp011235]

I rolled back the following from 10.9.2 to my 10.9.3 install.

 

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAGF100Hal.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAGK100Hal.kext

drwxr-xr-x    3 root  wheel   102 Oct  1 22:03 NVDANV50HalTesla.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAResman.kext

drwxr-xr-x    3 root  wheel   102 Oct  1 22:03 NVDAResmanTesla.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAStartup.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 22:35 NVSMU.kext

 

All it does it cause an instant reboot. 

 

I am guessing I am just going to do a full re-install and this time setup time machine for such cases as this. 

 

Unless someone else has a better idea quick

[nawcom]

 

I rolled back the following from 10.9.2 to my 10.9.3 install.

 

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAGF100Hal.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAGK100Hal.kext

drwxr-xr-x    3 root  wheel   102 Oct  1 22:03 NVDANV50HalTesla.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAResman.kext

drwxr-xr-x    3 root  wheel   102 Oct  1 22:03 NVDAResmanTesla.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 23:03 NVDAStartup.kext

drwxr-xr-x    3 root  wheel   102 Sep 29 22:35 NVSMU.kext

 

All it does it cause an instant reboot. 

 

I am guessing I am just going to do a full re-install and this time setup time machine for such cases as this. 

 

Unless someone else has a better idea quick

 

NVSMU has nothing to do with NVIDIA graphics - it's a driver for the NVIDIA MCP79 and MCP89 chipset co-processor. Note I said NVDA* kexts, not NV* kexts. If you're actually using it for your board, then bring back the original from 10.9.3, though to be honest I would be surprised if you do use one of those chipsets, it would cause a reboot by using the 10.9.2 version. I don't have a motherboard that uses it so I wouldn't know its effects of rolling it back. If you don't have either of those chipsets then hell, you could delete that kext and it wouldn't make a difference. A simple kextstat command will tell you if you're using it. What are the specs for your motherboard or laptop?

 

Now let's say you don't actually use that kext at all - I'm skeptical that rolling back graphics drivers would cause an instareboot at the moment you boot the kernel (I assume that's what you mean) instead of a kernel panic  - that's usually kernel and/or bootloader related regarding your CPU's support, but I wouldn't be too surprised as I've seen stranger things happen.

[rhcp011235]

NVSMU has nothing to do with NVIDIA graphics - it's a driver for the NVIDIA MCP79 and MCP89 chipset co-processor. Note I said NVDA* kexts, not NV* kexts. If you're actually using it for your board, then bring back the original from 10.9.3, though to be honest I would be surprised if you do use one of those chipsets, it would cause a reboot by using the 10.9.2 version. I don't have a motherboard that uses it so I wouldn't know its effects of rolling it back. If you don't have either of those chipsets then hell, you could delete that kext and it wouldn't make a difference. A simple kextstat command will tell you if you're using it. What are the specs for your motherboard or laptop?

 

Now let's say you don't actually use that kext at all - I'm skeptical that rolling back graphics drivers would cause an instareboot at the moment you boot the kernel (I assume that's what you mean) instead of a kernel panic  - that's usually kernel and/or bootloader related regarding your CPU's support, but I wouldn't be too surprised as I've seen stranger things happen.

 

Yea, I did notice what you said just now about NVDA* , But I already reinstalled the hack with 10.9.2 until things are a bit better dug into

 

I should have known not to install this beta due to the release notes saying "Graphics" and my monitors are very picky.

 

Dual 30" 2560X1440 Korean Monitors like Conti.

[CooSee]

everything works for me   

 

greetings

[Picasso]

The new NVIDIA kexts are causing some interesting issues. I've verified it on my PC with my 2 GTX 770s (GK104, uses NVDAStartup.kext to inject needed values, NVCAP, etc) which didn't surprise me since I've had to deal with issues with it in 10.9.2, though that has to do with the new IOPCIFamily kext and was able to get around that and get them working again. But when I installed it on my laptop (GT 525m) which uses Chameleon's GraphicsEnabler which has always worked fine, I had the same issue - the laptop display was gone by the time the desktop loaded up. For GTX 770, DVI #1 is detected, DVI #2 is not detected at all, and HDMI is detected fine. This behavior matches both GTX 770 cards. Before all three get picked up just fine for both cards.

 

I could plug in my second DVI display to DVI #1 on my second GTX 770 and have them detected fine with the new kexts, but that will screw up using SLI in Windows for gaming, and I'm not willing to move around cables in the back of my PC whenever I boot OS X or Windows for the sake of things working. And of course there's no fix like this with the case of my laptop with GT 525m losing its screen.

 

Anyways, this kind of behavior usually happens when the injected NVCAP value is incorrect and as a result not all display ports are correctly identified. I verified that the NVCAP values as well as the rest of the injected values match up with GTX 770 and GT 525m in both 10.9.2 and 10.9.3 beta. Rolling back the NVIDIA kexts to the 10.9.2 versions fixes the display issues.

 

If anyone is having similar issues, the easy fix is to roll back all the NVDA* kexts your card uses to 10.9.2 release versions, though that isn't a good long-term solution. I made sure performance wasn't affected by rolling back. OpenCL and VDA still work fine as before. I'll see if this persists in the upcoming beta releases. Also be sure and mention it here if someone is having similar NVIDIA issues, because if I'm the only one having issues, then I guess I'm lucky, hehe.

 I have the same card.

You say don't use the files from 10.9.3, and replace for the 10.9.2?

Or you have problems with 10.9.2 and replace for 10.9.1

Thanks.

[AREOS]

 

The focus areas mentioned in the post above - graphics drivers and audio. I can't use my AMD Radeon 7850M GPU because HP doesn't let you enable it in the BIOS and I use the VoodooHDA driver so that I can get my Beats audio subwoofer working. No one has yet managed to get the subwoofer recognized using the AppleHDA.kext.

I have audio working including the subwoofer using AppleHDA, check this post for more info http://www.insanelymac.com/forum/topic/296261-applehdakext-and-laptop-subwoofer-aka-21-audio/?p=2002197

 

Good Luck

 

 

The native kext NVDAStartup is actually an injector kext among other things from Apple. (I say other things because I notice that if I prevent it from loading on my laptop with the GT525M Fermi card which also requires GraphicsEnabler to boot up right, I get good ol 1024x768 VESA, so it's depended on by other kexts like the one for Fermi architecture cards.) If you open the kext's binary up in a disassembler app like IDA Pro you can see where it probes your video card for its architecture (Kepler, Maxwell, Tesla, Fermi) and in other functions, it injects its determined NVCAP, NVDA,Display-*, etc. So I don't use GraphicsEnabler or any other form, mainly because

 

a.) It only supports one card, and I have 2 cards I want enabled. I of course can't use SLI in OS X but nonetheless I want both of my cards working separately and enabled.

b.) As of now GraphicsEnabler for Chameleon (I haven't checked on Clover) only supports single cards with the old limit of 2 displays enabled, NVDA,Display-A and NVDA,Display-B. Kepler cards (GK1xx) extend to NVDA,Display-C and NVDA,Display-D, which is how you're able to get more than 2 displays going.

 

I could inject via DSDT and include NVDA,Display-C and D but why should I if NVDAStartup injects it already? Same for NVCAP values - NVDAStartup.kext generates it. Pretty much if you have a Kepler architecture NVIDIA card you shouldn't have to use any injector for everything to work normally (for the most part, I'm sure there are situations where it isn't perfect, though I haven't run into any myself). And like I said before, the kext is still generating the same NVCAP from 10.9.0 in 10.9.3, it's just deciding to ignore one of my DVI ports.

 

That's because NVDAStartup.kext is overriding those injected entries with its own. Pretty much what would happen if you had one NVCAP value injected via DSDT, but also used an injector kext like good ol' Natit.kext at the same time with a different NVCAP - Natit.kext will override since it loads after DSDT/EFI strings are injected and recognized. In the case for my 2 GTX 770s, if I try and boot with DSDT overrides (which I originally did before I understood NVDAStartup) I end up with a black screen, so using both at the same time can cause issues in the end.

 

Now for my laptop with GT525M that uses GraphicsEnabler - the NVCAP is still the same as before and it worked fine with it, detecting the display and HDMI. Now with the 10.9.3 kexts its screen goes blank, as if the NVCAP is wrong. I roll back the kexts and the screen is back. There's something new in these updated kexts that's doing it.

 

GTX770M uses some newer core than desktop GTX770, mine is GK104 while yours is GK106 so there's some differences, though I wouldn't expect you to be having issues because of it.

 

There's still a lot to figure out as of now. I'll be sure and post here about any updates while I {censored} around with this.

Thanks for the info, Im checking, I´ll try it out.

 

Good luck

[TheHacker66]

Every update ran fine except this one, seems to have fked the IOATAFamily kext my Desktop hackintosh needs to see my hdd..

 

Will do a rollback with time machine and also update my Laptop which is running 10.9.2 fine so far.

[tle88]

13D12  Edit: 13D17 seems to works well in this old machine.

 

I am impressed.  I peacefully now wait consumer

level lga 8 cores with iris pro, ddr4 and zfs...    

 

T  -.-

[Bandes]

New Nvidia Web drivers does not work anymore.

Opencl bug still there

Last cuda works fine

If you change the NVDARequiredOS string to the newer beta build number in the NVDAStartup.kext/Info.plist file, the Web driver will load again.

It worked for me.

 

 

I am having problems with this driver though as well as with the original ones.

Can't get rid of OpenGL Channel Exceptions on my NVidia 9600 GT.

I am trying all the beta updates since 10.9. to see if the graphics crashes finally go away.

But it's not looking like that.

There must be something else. Maybe i should rollback something. Everything was ok until 10.8.4 i guess.

IOGraphicsFamily or some other kext may cause the crashes perhaps.

 

One Note : Changing that string causes the signature of the kext to become invalid, but you still can load it with Kext Utility setting up the Exception List file for that..