Jump to content

OSXRestart.kext


Master Chief
 Share

173 posts in this topic

Recommended Posts

GA-P35-DS3L REV:2

BootThink 2.3.18 Snow Leopard 10.6.2

 

Restart does not work! Tried several time with no success.

OSXRestart.kext won't fix a broken Snow Leopard installation so join the other GigaByte users by following the provided links.

 

What "FACP table data" you used for this reset fix?

Just wait for the source code if you need more hints.

 

Note: My publisher needs time to print my book, and thus people here will have to wait for them to be ready. That is also the reason for me to be so darn tight lip about some things I am working on. Something has to pay the bills here after all ;)

Link to comment
Share on other sites

What "FACP table data" you used for this reset fix?

Well if he using the reset data from the FADT, then it is common for all tables. It provides a register address and a value you can write to that address. I don't see why that method is better than any other tough. There are some other common ways to reset the cpu from software, the "kdreboot" which is writing some data to the keyboard port, and disabling interrupt and doing a fault, also possible to jump to real mode and use the bios code for reset. You can use the pci registers (cf9) and efi if you have that, and some others.

 

http://lxr.linux.no/linux+v2.6.31/include/acpi/actbl.h#L176

Link to comment
Share on other sites

OSXRestart.kext doesn't work for me. But It seems really get a big breakthrough because when the computer shutdown it checks optical driver which is a symptom before rebooting. But still my computer doesn't reboot. Instead,it keeps silent with its energy light on.

 

And any other alternative kext does nothing!

 

 

Mine is a lenovo laptop. And i have no problem of shutting down.

 

In ubuntu,

cat /proc/acpi/fadt | more

shows FACPt

cat /proc/acpi/fadt | less

FACPt^@^@^@^A<84>LENOVOCB-01 ^@^@^D^FLOHRZ^@^@^@<C0><DF><EE><9F><EE>y<EE><9F>^@^@ ^@<B2>^@^@^@<F0><F1><F2><80>^@^P^@^@^@^@^@^@^D^P^@^@^@^@^@^@ ^P^@^@^H^P^@^@(^P^@^@^@^@^@^@^D^B^A^D^H^@^@<85>^A^@9^@^@^@^@^@^A^C^M^@2^@^@^@<A5><82>^@^@

 

I hope this information is helpful!

 

 

My machine:

Screen%20shot%202009-11-25%20at%2012.55.16%20PM.png

Link to comment
Share on other sites

Well if he using the reset data from the FADT, then it is common for all tables. It provides a register address and a value you can write to that address. I don't see why that method is better than any other tough. There are some other common ways to reset the cpu from software, the "kdreboot" which is writing some data to the keyboard port, and disabling interrupt and doing a fault, also possible to jump to real mode and use the bios code for reset. You can use the pci registers (cf9) and efi if you have that, and some others.

The keyboard controller route – and thus kdreboot – has proven to be a problem for many motherboards, and since they can reboot now...

Link to comment
Share on other sites

Master Chief great work...

I think i have discovered why some people might have reported that your kext isn't working...

I tryed it on both my Hacks (check signature) and it didn't work on neither of them, so the idea popped in my mind that it could be a conflict with the VoodooPowerMini.kext - 10.6.x and so it was, after removing the voodoo kext the restart after sleep worked...

 

I hope i was helpful and that this 2 kexts will be able to coexist in the future releases...

Continue the great work u are doing on the P5K (waiting for the DVD - autosleep fix :D )and Hack scene...

THX

Link to comment
Share on other sites

The kext doesn't fix issue for wake after deep sleep and fan continues spinning after HDD stopped; but after messing around here is a statement

 

 

On my computer (see specs in sig):

 

 

8400GS, wake after deep sleep => OK

HD4870, wake after sleep => KO

 

 

10.6 with HD4870 and cpunull...kext => enter in deep sleep and fan stop to spin. Wake: stucked on a black screen and fan is spinning.

10.6.2 with HD4870 and cpunull... (overall perfs decrease by 50%) with a disabler (full perfs) => enter in deep sleep but fan continue spinning (with your kext).

 

 

I don't know if it will help but the 4870 here, seems to bug me for the sleep/wake up state, despite i've tried many solutions.

Link to comment
Share on other sites

The kext doesn't fix issue for wake after deep sleep and fan continues spinning after HDD stopped; but after messing around here is a statement

 

On my computer (see specs in sig):

 

8400GS, wake after deep sleep => OK

HD4870, wake after sleep => KO

 

10.6 with HD4870 and cpunull...kext => enter in deep sleep and fan stop to spin. Wake: stucked on a black screen and fan is spinning.

10.6.2 with HD4870 and cpunull... (overall perfs decrease by 50%) with a disabler (full perfs) => enter in deep sleep but fan continue spinning (with your kext).

 

I don't know if it will help but the 4870 here, seems to bug me for the sleep/wake up state, despite i've tried many solutions.

Hello? This has nothing to do with my restart solution. Please read post #1 again. Thank you.

Link to comment
Share on other sites

Target audience

OSXRestart.kext is developed for OS X 10.6 and 10.6 only. For both 32 and 64 bit (single combo kext). And also (not specifically) for people where restart-after-sleep is broken; no more keyboard controller hacks �" I am using the FACP table data instead.

 

I resume..does not work here. Bye. :)

 

 

ps: english is not my mothertongue. If i misunderstood something: my mistake.

Link to comment
Share on other sites

tried on my hack, breaks shutdown though, been following gigabyte dsdt thread, also p5k pro thread...

_PTS method fixes shutdown? only way to get sleep here is using the setting in energy prefs, when that setting is enabled, doesn't it make os x ignore _PTS and _WAK completely?

seems like it would be a loop, unless i'm missing something...

Link to comment
Share on other sites

The keyboard controller route – and thus kdreboot – has proven to be a problem for many motherboards, and since they can reboot now...

 

Yes I know, but also this method may fail. If you look at the darwin code and bsd code, you will see they use different approaches after each other until they succeed. Linux has a more complicated approach, but it is basically the same thing as well.

 

BSD http://fxr.watson.org/fxr/source/amd64/amd...igexcerpts#L556

BSD x86_64 http://fxr.watson.org/fxr/source/i386/i386...igexcerpts#L633

 

Linux http://lxr.linux.no/linux+v2.6.31/arch/x86...l/reboot.c#L504

Link to comment
Share on other sites

Yes I know, but also this method may fail. If you look at the darwin code and bsd code, you will see they use different approaches after each other until they succeed. Linux has a more complicated approach, but it is basically the same thing as well.

 

BSD http://fxr.watson.org/fxr/source/amd64/amd...igexcerpts#L556

BSD x86_64 http://fxr.watson.org/fxr/source/i386/i386...igexcerpts#L633

 

Linux http://lxr.linux.no/linux+v2.6.31/arch/x86...l/reboot.c#L504

Thank you Superhai. Great links. And I am aware of the limitations of OSXRestart.kext but this is only a start. Still pretty new to OS X development, but I am loving and fully enjoying all the luxury of todays development platforms – I started with a 1MHz CPU like 25 years ago :(

 

Back to work now.

Link to comment
Share on other sites

OSXRestart.kext won't fix a broken Snow Leopard installation so join the other GigaByte users by following the provided links.

;)

 

 

What do you mean? My hack is not broken i just replaced EVOx86 kext with yours and it killed my restart. With your kext my hack just log off and stayed there and would neither restart nor shut down or hung, there was just coming plenty of messages

PIC still valid

about to call reboot

pic still valid

....

....

....

about to call reboot

 

and so on!

Link to comment
Share on other sites

 Share

×
×
  • Create New...