Jump to content

Mavericks kernel testing on AMD (formerly Mountain Lion kernel testing on AMD)

theconnactic

By theconnactic,
in OS X 10.9 (Mavericks)

 Share
6,414 posts in this topic

Recommended Posts

Konner
Konner
Posted
Posted

i'm not sure they must publish it.. they certainly don't with iOS. The XNU is not under GPL. Now the thing is that you can just use the 10.9 kernel and it will work anyway. There's not many changes among OS X subversions usually.

I figure they'll publish the 10.9.2 kernel, since it's a different kernel version.

Link to comment
Share on other sites
Bitcore
Bitcore
Posted
Posted

Hey guys,

 

I've just finished a new build for AMD osx development.
It has a fx-6350 cpu ,Gigabyte 970-ud3 mobo, 8GB ram , HD 5450 1GB GPU and a 200GB HDD.

I heard there is a fully working kernel by bronya? Is it true and where can I download it?

Thanks

Bitcore

Link to comment
Share on other sites
spakk
spakk
Posted
Posted

Hey guys,

 

I've just finished a new build for AMD osx development.

It has a fx-6350 cpu ,Gigabyte 970-ud3 mobo, 8GB ram , HD 5450 1GB GPU and a 200GB HDD.

I heard there is a fully working kernel by bronya? Is it true and where can I download it?

Thanks

Bitcore

 

 

see here http://www.insanelymac.com/forum/topic/281450-mavericks-kernel-testing-on-amd-formerly-mountain-lion-kernel-testing-on-amd/page-255

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

Thhat is my experience:

Finally I have Mavericks 10.9.2 almost fully functional. I have this machine:

 

-Procesor: Phenom II 965 BE 

-Motherboard: Asus M5a88 v-Evo

-Sound integrated: ALC892

-Lan integrated: RTL8111E

-DD: Hitachi Ultrastar 1TB

-Graphics card: GTX 650 ti oc

 

I have tested almost all kernels posted here, but the ones that giveme best resutls are Sinetek and AndysV:

 

Sinetek 10.9.0 v6  //github.com/sinetek/xnu/releases  Works very well without graphic glitches or icon bugs, except:

   -Photoshop cc still crashes on Black and white or 3d.

   -When I shut down the computer, it makes all the shutdown procces, in fact power off, but inmediatly power on again.

 

Andys most recent kernel http://www.dropbox.com/s/taf38zdm8va6l4h/BSA_kernel.zip Works very well, shutdown and restart function 100%, no pink icons, except:

-Icon bugs especially in the name of some icons and Chrome graphic bugs too.

-Photoshop cc still crashes on Black and white or 3d.

 

Other kernels have pink icons issue, or graphic glithes at least for my Hackintosh rig.

 

Would be amazing Sinetek or Andy makes a release solving above issues. Especially shutdown (Sinetek) or Graphics glitches solved (Andys).

 

Congratulations to all in this amazing forum, and special  thanks to people working in the development of hackintosh.

  • Like 1
Link to comment
Share on other sites
tragediana150
tragediana150
Posted
Posted

Great test! :)

Would be amazing Sinetek or Andy makes a release solving above issues. Especially shutdown (Sinetek) or Graphics glitches solved (Andys).

I think there is no work today for Sinetek, Andy and Bronya. Today is New Year so they deserve for some free time and fun tonight :) Cheers guys, have a good party!! :)

 

:guitar: :guitar: :guitar::hyper: :hyper: :hyper: :hyper: :guitar: :guitar: :guitar:

  • Like 4
Link to comment
Share on other sites
spakk
spakk
Posted
Posted

Thhat is my experience:

Finally I have Mavericks 10.9.2 almost fully functional. I have this machine:

 

-Procesor: Phenom II 965 BE 

-Motherboard: Asus M5a88 v-Evo

-Sound integrated: ALC892

-Lan integrated: RTL8111E

-DD: Hitachi Ultrastar 1TB

-Graphics card: GTX 650 ti oc

 

I have tested almost all kernels posted here, but the ones that giveme best resutls are Sinetek and AndysV:

 

Sinetek 10.9.0 v6  //github.com/sinetek/xnu/releases  Works very well without graphic glitches or icon bugs, except:

   -Photoshop cc still crashes on Black and white or 3d.

   -When I shut down the computer, it makes all the shutdown procces, in fact power off, but inmediatly power on again.

 

Andys most recent kernel http://www.dropbox.com/s/taf38zdm8va6l4h/BSA_kernel.zip Works very well, shutdown and restart function 100%, no pink icons, except:

-Icon bugs especially in the name of some icons and Chrome graphic bugs too.

-Photoshop cc still crashes on Black and white or 3d.

 

Other kernels have pink icons issue, or graphic glithes at least for my Hackintosh rig.

 

Would be amazing Sinetek or Andy makes a release solving above issues. Especially shutdown (Sinetek) or Graphics glitches solved (Andys).

 

Congratulations to all in this amazing forum, and special  thanks to people working in the development of hackintosh.

you should update your adobe tools. Since the update all adobe tools work to 100%, and no crash, etc. but it can also be dependent on the system, possibly to the correct tuning of the graphics card??

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

Very quickly answers. It's right, Sinetek, Andy and Bronya have a time off, happy new year and party time :D 

 

you should update your adobe tools. Since the update all adobe tools work to 100%, and no crash, etc. but it can also be dependent on the system, possibly to the correct tuning of the graphics card??

 

  My adobe is fully updated, the interesting thing is how I tunne my graphic card, I dont't DSDT... Is there a tool to do it or where can I modify the functionallity?

Link to comment
Share on other sites
spakk
spakk
Posted
Posted

Very quickly answers. It's right, Sinetek, Andy and Bronya have a time off, happy new year and party time :D

 

 

  My adobe is fully updated, the interesting thing is how I tunne my graphic card, I dont't DSDT... Is there a tool to do it or where can I modify the functionallity?

see here http://www.insanelymac.com/forum/topic/285760-help-topic-further-help-in-chatosx86hulegacykernel-irc/page-95

Link to comment
Share on other sites
Diazz72
Diazz72
Posted
Posted

i'm not sure they must publish it.. they certainly don't with iOS. The XNU is not under GPL. Now the thing is that you can just use the 10.9 kernel and it will work anyway. There's not many changes among OS X subversions usually.

There is no problem whit 10.9.1, but there will be with 10.9.2+.

Link to comment
Share on other sites
Bitcore
Bitcore
Posted
Posted

Dude this kernel works like a charm. 

Running the most stable and fast 10.8.2 that I ever had!

And no graphical problems at all :D I even got my old HD4650 working full QE/CI.

This is suitable for daily usage for sure!

Bitcore

  • Like 4
Link to comment
Share on other sites
toledanogus
toledanogus
Posted
Posted

Dude this kernel works like a charm. 

Running the most stable and fast 10.8.2 that I ever had!

And no graphical problems at all :D I even got my old HD4650 working full QE/CI.

This is suitable for daily usage for sure!

Bitcore

 

Ok I will test Mountain Lion 10.8.2 with this Sinetek release V6 and my PC. I'll install at this moment and comment my results. Thanks. 

Mmm... firs check what DMG of mountain Lion works... let's begin again... wll if you recommend one would be ok.  :yes:  

I`ll begin with iatkos too see what happen.

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

Ok I will test Mountain Lion 10.8.2 with this Sinetek release V6 and my PC. I'll install at this moment and comment my results. Thanks. 

Mmm... firs check what DMG of mountain Lion works... let's begin again... wll if you recommend one would be ok.  :yes:  

I`ll begin with iatkos too see what happen.

I used iATKOS ML2 aswell, and this is where I found the kernel: http://wiki.osx86project.org/wiki/index.php/Patched_Kernels (scroll down to OS X 10.8 Mountain Lion and find Bronya Fully working Kernel ) 

Another thing that is really cool is that audio timings are good now. Logic Pro used to give AudioBuffer errors and samplerate problems with my usb soundcard all the time, but with this kernel I can finally use it :D

Bitcore

  • Like 2
Link to comment
Share on other sites
Andy Vandijck
Andy Vandijck
Posted
Posted

@Sinetek: I did some more work on the opemu.

I noticed you started to implement SSE4.2 instructions (only one was present).

I made 5/7 instructions. Still needed: CRC32 instruction and POPCNT (I believe SSE4a has POPCNT).

The string comparing functions: PCMPESTRI, PCMPESTRM, PCMPISTRI and PCMPISTRM are implemented.

Also packed comparing (PCMPGTQ) is implemented.

2 minor fixes needed, they are flagged.

 

New opemu is attached below.

All warnings fixed, like before.

Also note the return value on the kernel trap.

I use this to check wether to return from the trap or fall through...

OPEMU.zip

  • Like 7
Link to comment
Share on other sites
spakk
spakk
Posted
Posted

@Sinetek: I did some more work on the opemu.

I noticed you started to implement SSE4.2 instructions (only one was present).

I made 5/7 instructions. Still needed: CRC32 instruction and POPCNT (I believe SSE4a has POPCNT).

The string comparing functions: PCMPESTRI, PCMPESTRM, PCMPISTRI and PCMPISTRM are implemented.

Also packed comparing (PCMPGTQ) is implemented.

2 minor fixes needed, they are flagged.

 

New opemu is attached below.

All warnings fixed, like before.

Also note the return value on the kernel trap.

I use this to check wether to return from the trap or fall through...

 

Unfortunately, I have no idea about programming in , as far as I 've taken a contribution from a Linux programmer, should the

POPCNT statement contribute to the cause of performence of NVIDIA based Graphics cards provided that the CPU uses the instruction POPCNT,

should not that be will get an error message " invalid instruction " . I have taken out of the report , the AMD Barcelona includes this statement and

AMD Phenom II X6 also . Now please do not laugh me out if you already know . should indeed by a performance increase or a better graphics yield

can be realized , it would be a wonderful thing . :wink_anim:

Link to comment
Share on other sites
Sinetek
Sinetek
Posted
Posted

@Sinetek: I did some more work on the opemu.

I noticed you started to implement SSE4.2 instructions (only one was present).

I made 5/7 instructions. Still needed: CRC32 instruction and POPCNT (I believe SSE4a has POPCNT).

The string comparing functions: PCMPESTRI, PCMPESTRM, PCMPISTRI and PCMPISTRM are implemented.

Also packed comparing (PCMPGTQ) is implemented.

2 minor fixes needed, they are flagged.

 

New opemu is attached below.

All warnings fixed, like before.

Also note the return value on the kernel trap.

I use this to check wether to return from the trap or fall through...

Commited. Awesome job. To address a few issues:

 

KTrap's return value is not used for anything. Are you suggesting we should make use of a return value?

Basically if you return from ktrap that means you resume code execution with the given context.

The formatting looks a bit uneven, are you using Xcode? It seems to mess with the formatting. Not a big deal.

 

I added in WRMSR. That should fix quite a few CPU's methinks.

I can't test anything because my motherboard went the way of the dodo, but i could build a bin if nothing else is fixed soon.

 

 

X86_64 cpu's support at a minimum SSE2, so in _theory_ we're running on pretty much anything now. There's still some glitches in the SSSE3 portion, but because we're bypassing via SSE42, it shouldn't matter for general usage.

 

lastly, Happy new year to you too!

 

And to the community :sorcerer:

  • Like 5
Link to comment
Share on other sites
Andy Vandijck
Andy Vandijck
Posted
Posted

Commited. Awesome job. To address a few issues:

 

KTrap's return value is not used for anything. Are you suggesting we should make use of a return value?

Basically if you return from ktrap that means you resume code execution with the given context.

The formatting looks a bit uneven, are you using Xcode? It seems to mess with the formatting. Not a big deal.

 

I added in WRMSR. That should fix quite a few CPU's methinks.

I can't test anything because my motherboard went the way of the dodo, but i could build a bin if nothing else is fixed soon.

 

 

X86_64 cpu's support at a minimum SSE2, so in _theory_ we're running on pretty much anything now. There's still some glitches in the SSSE3 portion, but because we're bypassing via SSE42, it shouldn't matter for general usage.

 

lastly, Happy new year to you too!

 

And to the community :sorcerer:

in my kernel that return value in the kernel trap part is used.

if valid -> return

if not -> continue in the trap code so that it will panic exactly at the point it was running.

this way its easier to trace things.

  • Like 3
Link to comment
Share on other sites
Sinetek
Sinetek
Posted
Posted

 

in my kernel that return value in the kernel trap part is used.

if valid -> return

if not -> continue in the trap code so that it will panic exactly at the point it was running.

this way its easier to trace things.

good point, that's brilliant!

 

by the way ppl, this year's 30c3 ( @ youtube ) is amazing. speaking as a hardware guy, that is. I'm getting the sense i should have gone there to talk about Opcode Emulators :hysterical:

  • Like 5
Link to comment
Share on other sites
DEgorov
DEgorov
Posted
Posted

Got my Athlon II X4 / Gigabyte 760G / GeForce 210 working with Sinetek's v6 kernel and 10.9.1. About my previous message here - yep, that was a hardware failure, sorry :(

 

Everything seems to work fine, except reboot - the PC does sort of shutdown without actually turning off the power. And as I can see this behaviour is not consistent - sometimes reboot works ok, sometimes not. Seems that this issue was fixed by one of the latest commits?

 

Yet, sometimes my PC stops without any particular reason. Already happened 2 or 3 times during couple of days. It just stops and even mouse cursor doesn't move, the image on display stands still, there is no disk activity and so on. I don't see any kernel messages on the screen, there is nothing suspicious in Console.app logs. Where else can I look to find the reason?

 

Thanks for all developers for their hard work. Bronya, с наступившим!

  • Like 2
Link to comment
Share on other sites
hzhjy123
hzhjy123
Posted
Posted

I compiled a new kernel, using the latest sinetek source code. (added AnVSee3 and See42 opemu)


Use it to boot system normally.


But at compile time has been suggested that "xcrun: error: unable to find utility"ctfmerge", not a developer tool or inPATH"and "xcrun: error: unable to find utility "ctfconvert", not a developertool or in PATH".


It affects the compiler？


Thanks！


mach_kernel V7 test.zip


  • Like 4
Link to comment
Share on other sites
 Share
Go to topic listing
×
×
  • Create New...