Jump to content

10.6.3 legacy kernel for INTEL/AMD ready to download by qoopz/nawcom - a few new features


nawcom
 Share

79 posts in this topic

Recommended Posts

not sure if this is an issue with the kernel or not but i find photo booth kp always on start but the camera works in all other things, i was thinking was somthing stopped that shouldnt or is this a cpuid issue

 

 

btw is on an amd laptop

Link to comment
Share on other sites

Apple Released MacBook Pro Software Update 1.3 for the new core i7, i5 MacBook Pros which also has GPU Nvidia GT 330M

http://support.apple.com/kb/DL1026

 

looking at the contents it includes a new kernel and graphics drivers

 

So we NEED legacy kernel for that kernel specially for unsupported i7 & i5 M processors

 

pleeeeeeeeeeeeease

Link to comment
Share on other sites

Hi Guys,

 

have an Intel T2300 Core Duo and GMA950 and am suffering the mouse lags / cpu stuttering syndrom and expected healing from this new kernel. Tried fsb=667000000, busratio=10, busratiopath=4, tscsync=1, and cpus=1 kernel flags, but no way. Just found another option, idlehalt=0, which I'll try today.

 

Could anyone heal the stuttering syndrom on 10.6.3? Any suggestions welcome!

 

By the way: how to remove the legacy_kernel (how to tell the bootloader which kernel to load)?

 

Thanks,

 

Chris.

Link to comment
Share on other sites

How about 64 bit kernel support?

I've made it possible to make a 32/64bit kernel.

 

The patches I made from your previous sources are here:

http://rapidshare.com/files/365470184/AnV_...u_rel2.diff.zip

 

It involved changing some assembly code with __LP64__ compiler check for passing 32/64 bit code (little 64bit code port).

After that it compiled up fine...

 

Feel free to change your code so you can do the same (I build it with make ARCH_CONFIGS="I386 X86_64" KERNEL_CONFIGS=RELEASE install)

 

Keep up the great work :)

Link to comment
Share on other sites

It works with Phoenom II X4, 8GB if -force64 option is specified, though some apps would not launch.

 

Looking into /Library/Logs, there i found the same repetative crash ( usbmuxd ).

As it always show something like:

 

Exception Type:  EXC_BAD_INSTRUCTION (SIGILL)
Exception Codes: 0x0000000000000001, 0x0000000000000000
Crashed Thread:  0

Thread 0 Crashed:

0   dyld                          	0x8fe2b7af _sysenter_trap + 3
1   dyld                          	0x8fe2d5da mach_init + 47
2   dyld                          	0x8fe01766 dyldbootstrap::start(macho_header 

 

there might be a a bug in SYS_ENTTER emulation in -force64 mode.

Link to comment
Share on other sites

It works with Phoenom II X4, 8GB if -force64 option is specified, though some apps would not launch.

 

Looking into /Library/Logs, there i found the same repetative crash ( usbmuxd ).

As it always show something like:

 

Exception Type:  EXC_BAD_INSTRUCTION (SIGILL)
Exception Codes: 0x0000000000000001, 0x0000000000000000
Crashed Thread:  0

Thread 0 Crashed:

0   dyld                          	0x8fe2b7af _sysenter_trap + 3
1   dyld                          	0x8fe2d5da mach_init + 47
2   dyld                          	0x8fe01766 dyldbootstrap::start(macho_header 

 

there might be a a bug in SYS_ENTTER emulation in -force64 mode.

 

 

Hhehe I see that error all the time since hmm a couple of months ago(booting with -force64)... I really don't see the use of this *kernel... ANV 10.2.0 kernel works just fine for me on myAMD updated to 10.6.3, never needed to patch anything.. all apps worked... is only like you said using -force64 tons of apps Crash giving that exact same error.

 

Now i have not yet tried using any cpuid patcher or even marvin's amd utility to see if that fixes the issue for 64bits... hehee but I'm with you we need a fix for: -force64

Link to comment
Share on other sites

How about 64 bit kernel support?

I've made it possible to make a 32/64bit kernel.

 

The patches I made from your previous sources are here:

http://rapidshare.com/files/365470184/AnV_...u_rel2.diff.zip

 

It involved changing some assembly code with __LP64__ compiler check for passing 32/64 bit code (little 64bit code port).

After that it compiled up fine...

 

Feel free to change your code so you can do the same (I build it with make ARCH_CONFIGS="I386 X86_64" KERNEL_CONFIGS=RELEASE install)

 

Keep up the great work ;)

 

 

Your kernel has made wonders to my AMD, ehehehe still working lovely on 10.6.3... now my question... I can boot fine doing -force64.... do you have any ideas how to fix all the apps from crashing? Can you provide Steps how to do it.. if you have made it successful fixing that issue, that would be great... because I have just bought 16gb of Ram, got tired of my 8gb on Windows lol... even though I can only work fine on maxmem=4096... I want full support

 

Anyways thank you AnV

Link to comment
Share on other sites

Apple Released MacBook Pro Software Update 1.3 for the new core i7, i5 MacBook Pros which also has GPU Nvidia GT 330M

http://support.apple.com/kb/DL1026

 

looking at the contents it includes a new kernel and graphics drivers

 

So we NEED legacy kernel for that kernel specially for unsupported i7 & i5 M processors

 

pleeeeeeeeeeeeease

 

MacBook Pro Software Update 1.3 include Darwin kernel 10.3.1. it support Clarkdale/Arrandale core CPU.

 

The method of installing MacBook Pro Software Update 1.3 besides Core i5/i7 MacBook Pro is as follows.

1. rewrite SMBIOS to MacBookPro (replase SMproductname to MacBookPro6,1 or MacBookPro6,2)

2. rewire installed OS verion information to 10.6.2 (build 102063a)

 

If you wanna get more information, see below post.

http://www.tonymacx86.com/viewtopic.php?f=...402&start=0.

 

I use Darwin 10.3.1 vanilla kernel, run Carkdale Core i5 and 16GB memory on MSI P55M-GD45.

 

Compatibility

-cpu recognition

 work (L3406=Dual-Core Intel Xeon)

-powermanagement (TB is OK)

 work

-shutdown&reboot

 work

-s3 sleep & wakeup

 work

-memory (4GB over is OK)

 work

-disk controller Intel 5 Series

 work

-audio (AppleHDA.kext patched Codec ALC885)

 work

 

Bencmark Score Vanilla kernel 10.3.1 V.S. Legacy kernel 10.3.0

-Xbench 1.3

 CPU Test 173.10 / 156.10

 Thread Test 456.54 / 415.36

 Memory Test 301.75 / 276.76

 Quartz Graphics Test 248.00 / 224.61

 OpenGL Graphics Test 187.21 / 161.84

-iBench

 3.41 / 3.16

-OpenMark

 3252 FPS / 3266 FPS

img20100423085259.jpg

Link to comment
Share on other sites

How about 64 bit kernel support?

I've made it possible to make a 32/64bit kernel.

 

The patches I made from your previous sources are here:

http://rapidshare.com/files/365470184/AnV_...u_rel2.diff.zip

 

It involved changing some assembly code with __LP64__ compiler check for passing 32/64 bit code (little 64bit code port).

After that it compiled up fine...

 

Feel free to change your code so you can do the same (I build it with make ARCH_CONFIGS="I386 X86_64" KERNEL_CONFIGS=RELEASE install)

 

Keep up the great work :wacko:

 

is this patch for kernel 10.2.0 ? xnu-1486.2.11 is found.

Link to comment
Share on other sites

Alright, xnu src was released so that means patching up that source code so everyone else can use the latest and greatest kernel.

* fsb detection

 

 

can i pass my FSB speed and ratio to kernel over boot flags? "busratio"? "busspeed"?

Link to comment
Share on other sites

Your kernel has made wonders to my AMD, ehehehe still working lovely on 10.6.3... now my question... I can boot fine doing -force64.... do you have any ideas how to fix all the apps from crashing? Can you provide Steps how to do it.. if you have made it successful fixing that issue, that would be great... because I have just bought 16gb of Ram, got tired of my 8gb on Windows lol... even though I can only work fine on maxmem=4096... I want full support

 

Anyways thank you AnV

It also needs sysenter patcher still + custom dyld...

It doesn't have to be full cpuid patcher, just sysenter emu...

Then it should function in 64bit properly to run those 32bit apps :)

I haven't had the time to port the sysenter emu from 10.5.8 kernel...

Link to comment
Share on other sites

is this patch for kernel 10.2.0 ? xnu-1486.2.11 is found.

Goes in cleanly for 10.3.0 too... :)

Link to comment
Share on other sites

It works with Phoenom II X4, 8GB if -force64 option is specified, though some apps would not launch.

 

Looking into /Library/Logs, there i found the same repetative crash ( usbmuxd ).

As it always show something like:

 

Exception Type:  EXC_BAD_INSTRUCTION (SIGILL)
Exception Codes: 0x0000000000000001, 0x0000000000000000
Crashed Thread:  0

Thread 0 Crashed:

0   dyld                          	0x8fe2b7af _sysenter_trap + 3
1   dyld                          	0x8fe2d5da mach_init + 47
2   dyld                          	0x8fe01766 dyldbootstrap::start(macho_header 

 

there might be a a bug in SYS_ENTTER emulation in -force64 mode.

sysenter emu unfortunetely hasn't been ported yet... :)

Link to comment
Share on other sites

The instruction at ( _sysenter_trap + 3 ) is "sysenter".

So, the code to trap undefined instruction ( sysenter is undefined in AMD compatible mode ) is not working correctly.

See previous post... sysenter is unavailable for EVERY AMD cpu... too bad :)

Link to comment
Share on other sites

Yeah I've heard of some issues regarding some intel processors. There are some little changes in some of the code regarding cpuid.h and .c as as well as lapic.c which might relate to how the added code gets incorrectly handled in regards to intels. I should have this fixed by the end of today.

 

i want to post my little contributions, it's a patch fixing the kernel panic

Local Apic Error , ESR 64\n"@/sourcecache/xnu/xnu-1228.15.4/osmfk/i386/lapic.c

that occur when booting without the cpus=1 boot flags in Pavilion dv6 and dv7 Series Laptops.

This is a BIOS bug specific to those serie and may be some other Models too !

unlike the Apocolypse fix that just redirect the output message for the error which is still there with this patch there is no more error.

 

Some explanation:

I have discovered that's the error ESR 64 is due to improper setting for LVT_INT0 and LVT_INT1 (32-bit registre of LVT Local Vector Table in the Local apic,

you can found the reference in Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 3A: System Programming Guide, Part 1, Chapter 10.5: Handling Local Interrupts, available for free download from intel website)

LVT_INT0, LVT_INT1 are both reported as being Vector=0x00, Delivery Mode=Fixed, Polarity=Hight, Trigger Mode=Edge, Mask=1 for ExtINT and 0 for NMI

browsing throught some apic source code for Linux, OpenSolaris and FreeBSD i have found that

Trigger Mode should be set to ExtINT for LVT_INT0 and to NMI for LVT_INT1.

i have added the proper setting in the lapic_configure() function of the lapic.c file

in the attachement you have the patched lapic.c and the diff file against the original lapic.c

NB: lapic.c file is from xnu-1228.15.4 source code for Leopard 10.5.8

and here is the link for a patched kernel for leopard 10.5.8 (for testing purpose)

http://www.megaupload.com/?d=8HA6VVI9

lapic_patch.zip

Link to comment
Share on other sites

can i pass my FSB speed and ratio to kernel over boot flags? "busratio"? "busspeed"?

example:

fsb=166 busratio=25

Will do for an FSB of 166Mhz Quad Pumped to 667Mhz multiplied with 2,5 so processor speed = 1.66Ghz

Link to comment
Share on other sites

I replaced my Phenom II X4 (920) with X6 (1055T) , then it stoppped working - it resets immediately after boot stage ( Chameleon RC4 ).

There is no change other than CPU.

I tested both -force64 and maxmem=4096, neither succeeds.

 

10.5.8 still boots with maxmem=4096.

Link to comment
Share on other sites

i want to post my little contributions, it's a patch fixing the kernel panic

Local Apic Error , ESR 64\n"@/sourcecache/xnu/xnu-1228.15.4/osmfk/i386/lapic.c

that occur when booting without the cpus=1 boot flags in Pavilion dv6 and dv7 Series Laptops.

This is a BIOS bug specific to those serie and may be some other Models too !

unlike the Apocolypse fix that just redirect the output message for the error which is still there with this patch there is no more error.

 

Some explanation:

I have discovered that's the error ESR 64 is due to improper setting for LVT_INT0 and LVT_INT1 (32-bit registre of LVT Local Vector Table in the Local apic,

you can found the reference in Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 3A: System Programming Guide, Part 1, Chapter 10.5: Handling Local Interrupts, available for free download from intel website)

LVT_INT0, LVT_INT1 are both reported as being Vector=0x00, Delivery Mode=Fixed, Polarity=Hight, Trigger Mode=Edge, Mask=1 for ExtINT and 0 for NMI

browsing throught some apic source code for Linux, OpenSolaris and FreeBSD i have found that

Trigger Mode should be set to ExtINT for LVT_INT0 and to NMI for LVT_INT1.

i have added the proper setting in the lapic_configure() function of the lapic.c file

in the attachement you have the patched lapic.c and the diff file against the original lapic.c

NB: lapic.c file is from xnu-1228.15.4 source code for Leopard 10.5.8

and here is the link for a patched kernel for leopard 10.5.8 (for testing purpose)

 

although i don't understand so much, but i am happy that there are some people who are trying to fix errors related to Pavilion dv6 and dv7 Series Laptops

but what about SL

Is there any hope to run with the vanilla kernel, or can we fix this using DSDT or even a way to edit the bios

 

please take a look here

[GUIDE] Snow Leopard on HP dvx Quad Core i7 QM, Installing Method, Kexts, and DSDT fixes

Link to comment
Share on other sites

I replaced my Phenom II X4 (920) with X6 (1055T) , then it stoppped working - it resets immediately after boot stage ( Chameleon RC4 ).

There is no change other than CPU.

I tested both -force64 and maxmem=4096, neither succeeds.

 

10.5.8 still boots with maxmem=4096.

 

 

well its a new CPU what did you expect... at least its working on Leopard..... now I don't use this Legacy 10.6.3 for Snow, I simply use the AnV's 10.2.0 no clue if that would help you on yours.

 

I am planning on buying one of those amd 6-cores, but for another build ofcourse.

Link to comment
Share on other sites

 Share

×
×
  • Create New...