Jump to content

GA-EX58 and GA-X58A DSDT native power management modifications

d00d

By d00d,
in Power Management

 Share
1,771 posts in this topic

Recommended Posts

aikidoka25
aikidoka25
Posted
Posted
Post 1 updated;

 

18. (added 20100121) Network port IOBuiltin=Yes options A, B or C.

Thanks to Mackenstein who alerted me to a problem with some applications such as Korg that need en0 to be recognized as IOBuiltin=Yes.

There are three ways to fix this, the second and third also eliminate the need for using PlatformUUID.kext to stop the `unable to determine UUID for host. Error: 35' errors from continuing after boot, for those still using C2RC3.

C2RC4 doesn't require PlatformUUID.kext.

 

 

...

 

thanks for the update d00d

 

now i only have these on /E/E

drwxr-xr-x@ 3 root wheel 102 14 Jan 19:26 EvOreboot.kext

drwxr-xr-x@ 3 root wheel 102 14 Jan 19:26 IOAHCIBlockStorageInjector.kext

drwxr-xr-x@ 3 root wheel 102 14 Jan 19:26 JMicronATA.kext

drwxr-xr-x@ 3 root wheel 102 18 Jan 20:27 fakesmc.kext

 

nvram -p also shows the platform-uuid and EFIBluetoothDelay

Link to comment
Share on other sites
d00d
d00d
Posted
  • Author
Posted

Post 1 updated;

 

15. Having the driver for Temperature Monitor installed prevents idle sleep, and I've read reports of various DVD drives preventing idle sleep.

The solution for the former is to uninstall, and for the latter to either uncheck `Put the hard disk(s) to sleep when possible' in Energy Saver, or to leave a DVD in the drive.

Another solution is to add the PenntNeu-script RIP.app in Accounts:user:Login Items.

To make this work I had to change main.scpt, although this has worked for others unmodified.

...

Link to comment
Share on other sites
kdawg
kdawg
Posted
Posted
Post 1 updated;

 

15. Having the driver for Temperature Monitor installed prevents idle sleep, and I've read reports of various DVD drives preventing idle sleep.

The solution for the former is to uninstall, and for the latter to either uncheck `Put the hard disk(s) to sleep when possible' in Energy Saver, or to leave a DVD in the drive.

Another solution is to add the PenntNeu-script RIP.app in Accounts:user:Login Items.

To make this work I had to change main.scpt, although this has worked for others unmodified.

...

Try unchecking "Monitor S.M.A.R.T drive temps." in the sensor section of iStat. That should fix that problem. I only monitor CPU temps in iStat and I've eliminated the sleep problem.
Link to comment
Share on other sites
d00d
d00d
Posted
  • Author
Posted
Try unchecking "Monitor S.M.A.R.T drive temps." in the sensor section of iStat. That should fix that problem. I only monitor CPU temps in iStat and I've eliminated the sleep problem.
Thanks, I updated post 1 with this information.

 

Untouched DSDTs from my EX58-UD4P, extracted from Linux.

 

dsdt9d.aml

dsdt10d.aml

dsdt10h.aml

dsdt10k.aml

 

I'll upload patched one later.

Thanks, it looks like there's differences between GA-EX58-UD4 and 5 with equivalent BIOS versions, so DSDT files wouldn't be interchangeable between the two.
Link to comment
Share on other sites
ryansimms
ryansimms
Posted
Posted
18. (added 20100121) Network port IOBuiltin=Yes options A, B or C.

The GA-EX58-UD3 MB uses Realtek RTL8111C, which is recognized as IOBuiltin=Yes by the AppleRTL8169Ethernet.kext supplied with the OS.

The other GA-EX58 MBs and all of the GA-X58A MBs use Realtek RTL8111D.

This needs Bit Shoveler's RealtekR1000SL.kext or similar, and a fix to be recognized as IOBuiltin=Yes.

Actually, only rev 1.6 and 1.7 of the GA-EX58-UD3R use the RTL8111C. The rev 1.0 model uses RTL8111D. I have the rev 1.0 model and I wasn't able to get ethernet to work without a kext.

 

Here's a link to the rev 1.0 model:

http://www.gigabyte.com.tw/Products/Mother...?ProductID=2989

 

Ryan!

Link to comment
Share on other sites
ttgolf
ttgolf
Posted
Posted

I have the UD4P board with i7 920. I have been using d00d's DSDT file in post #1. I decided to build up my own DSDT using the instruction in POST#1. I see some difference here. I need help in understanding what is going on.

 

My issue has been using a USB Bluetooth Dongle with Magic Mouse. Anytime I have Magic Mouse connected with "Allow Bluetooth devices to wake computer" option ON, my hackintosh will sleep but wake immediately. This happens every time with d00d's DSDT. When I use my own DSDT the computer will sleep properly once every 3~4 times. I looked into the system log and discovered when the system properly sleep (not waking immediately) it will show

 

Jan 29 10:13:31 Mac-Pro configd[15]: Sleep: Success - AC - Software Sleep
Jan 29 10:13:31 Mac-Pro configd[15]: Wake: Success - AC - USB4
Jan 29 10:13:31 Mac-Pro configd[15]: Hibernate Statistics

If the system wake immediately I get:

 

Jan 29 10:13:04 Mac-Pro configd[15]: Sleep: Success - AC - Software Sleep
Jan 29 10:13:04 Mac-Pro configd[15]: Wake: Success - AC - RTC USB4
Jan 29 10:13:04 Mac-Pro configd[15]: Hibernate Statistics

 

or

 

Jan 29 10:11:21 Mac-Pro configd[15]: Sleep: Success - AC - Maintenance Sleep
Jan 29 10:11:21 Mac-Pro configd[15]: Wake: Success - AC - USB4
Jan 29 10:11:21 Mac-Pro configd[15]: Hibernate Statistics

 

My question is what is the difference between Maintenance Sleep and Software Sleep? What does RTC mean? And the ultimate question is what do I need to do with my DSDT file to make sure I can properly sleep with Bluetooth Magic Mouse?

 

Thanks.

Link to comment
Share on other sites
d00d
d00d
Posted
  • Author
Posted
I have the UD4P board with i7 920. I have been using d00d's DSDT file in post #1. I decided to build up my own DSDT using the instruction in POST#1. I see some difference here. I need help in understanding what is going on.

 

My issue has been using a USB Bluetooth Dongle with Magic Mouse. Anytime I have Magic Mouse connected with "Allow Bluetooth devices to wake computer" option ON, my hackintosh will sleep but wake immediately. This happens every time with d00d's DSDT. When I use my own DSDT the computer will sleep properly once every 3~4 times. I looked into the system log and discovered when the system properly sleep (not waking immediately) it will show

 

Jan 29 10:13:31 Mac-Pro configd[15]: Sleep: Success - AC - Software Sleep
Jan 29 10:13:31 Mac-Pro configd[15]: Wake: Success - AC - USB4
Jan 29 10:13:31 Mac-Pro configd[15]: Hibernate Statistics

If the system wake immediately I get:

 

Jan 29 10:13:04 Mac-Pro configd[15]: Sleep: Success - AC - Software Sleep
Jan 29 10:13:04 Mac-Pro configd[15]: Wake: Success - AC - RTC USB4
Jan 29 10:13:04 Mac-Pro configd[15]: Hibernate Statistics

 

or

 

Jan 29 10:11:21 Mac-Pro configd[15]: Sleep: Success - AC - Maintenance Sleep
Jan 29 10:11:21 Mac-Pro configd[15]: Wake: Success - AC - USB4
Jan 29 10:11:21 Mac-Pro configd[15]: Hibernate Statistics

 

My question is what is the difference between Maintenance Sleep and Software Sleep? What does RTC mean? And the ultimate question is what do I need to do with my DSDT file to make sure I can properly sleep with Bluetooth Magic Mouse?

 

Thanks.

You shouldn't use a DSDT built from a GA-X58-UD5 BIOS on a GA-X58-UD4 or other MB unless you have confirmed that the base DSDT is the same (it's not);

[mac05:~] me% diff dsdt_ud5.dsl dsdt_ud4.dsl

...

 

The base DSDT from GA-X58-UD5 BIOS F9m or F9p is attached to post 1 so that you can compare with your's.

 

I'm unaware of anything that can be done in the DSDT to assure proper sleep with bluetooth devices.

LocusOfControl and others have recommended some things to try a few pages back.

Link to comment
Share on other sites
LocusOfControl
LocusOfControl
Posted
Posted
I have the UD4P board with i7 920. I have been using d00d's DSDT file in post #1. I decided to build up my own DSDT using the instruction in POST#1. I see some difference here. I need help in understanding what is going on.

 

My issue has been using a USB Bluetooth Dongle with Magic Mouse. Anytime I have Magic Mouse connected with "Allow Bluetooth devices to wake computer" option ON, my hackintosh will sleep but wake immediately. This happens every time with d00d's DSDT. When I use my own DSDT the computer will sleep properly once every 3~4 times. I looked into the system log and discovered when the system properly sleep (not waking immediately) it will show

 

Jan 29 10:13:31 Mac-Pro configd[15]: Sleep: Success - AC - Software Sleep
Jan 29 10:13:31 Mac-Pro configd[15]: Wake: Success - AC - USB4
Jan 29 10:13:31 Mac-Pro configd[15]: Hibernate Statistics

If the system wake immediately I get:

 

Jan 29 10:13:04 Mac-Pro configd[15]: Sleep: Success - AC - Software Sleep
Jan 29 10:13:04 Mac-Pro configd[15]: Wake: Success - AC - RTC USB4
Jan 29 10:13:04 Mac-Pro configd[15]: Hibernate Statistics

 

or

 

Jan 29 10:11:21 Mac-Pro configd[15]: Sleep: Success - AC - Maintenance Sleep
Jan 29 10:11:21 Mac-Pro configd[15]: Wake: Success - AC - USB4
Jan 29 10:11:21 Mac-Pro configd[15]: Hibernate Statistics

 

My question is what is the difference between Maintenance Sleep and Software Sleep? What does RTC mean? And the ultimate question is what do I need to do with my DSDT file to make sure I can properly sleep with Bluetooth Magic Mouse?

 

Thanks.

 

Nice one

 

I have similar issues to yourself as you know.

 

Personally I'm wondering if it is some thing to do with the way we model USB ports

This is what mine look like

 

 
            Device (UHC6)
           {
               Name (_ADR, 0x001A0002)
               Name (_PRW, Package (0x02)
               {
                   0x20, 
                   0x03
               })
           }

           Device (EHC1)
           {
               Name (_ADR, 0x001D0007)
               Name (_PRW, Package (0x02)
               {
                   0x0D, 
                   0x03
               })
           }

 

Here is an interesting apple developer page - do we need an additional _PTS (Prepare To Sleep) method?

 

http://developer.apple.com/mac/library/qa/qa2007/qa1517.html

 

A traditional USB device indicates that it is connected by pulling either the D+ or the D- line to the VBUS power rail. An ExpressCard device should do so by pulling either the USBD+ or USBD- line to the +3.3 VAUX rail. If it instead pulls the line to the +3.3 V rail, when the computer goes to sleep, the USBD+ or USBD- line will be floating. The computer will interpret this as a device disconnect and will wake up immediately.

 

RTC = Real-Time Clock

 

Software sleep = via apple menu - it can not be overridden in software ie ignore all software events

idle sleep = sleep after specified idle times unless something is keeping it awake eg BT

 

never heard of maintenance sleep

 

PS got my DSDT down to 1870 lines ~ 4.5K now ; fakesmc only

Link to comment
Share on other sites
Freexer
Freexer
Posted
Posted

hey your dsdt working nice but with your dsdt my harddisk are shown as external harddrive. I have tried an other dsdt there are my harddrives shown as internal harddrive but there is sleep an other things doesn't work.

 

I have the EX58-UD5 Mobo and the Core i7 920... it should be nice when you can help me because i like it when it's perfect :D

Link to comment
Share on other sites
ttgolf
ttgolf
Posted
Posted
You shouldn't use a DSDT built from a GA-X58-UD5 BIOS on a GA-X58-UD4 or other MB unless you have confirmed that the base DSDT is the same (it's not);

[mac05:~] me% diff dsdt_ud5.dsl dsdt_ud4.dsl

...

 

The base DSDT from GA-X58-UD5 BIOS F9m or F9p is attached to post 1 so that you can compare with your's.

 

I'm unaware of anything that can be done in the DSDT to assure proper sleep with bluetooth devices.

LocusOfControl and others have recommended some things to try a few pages back.

 

Before I used the magic mouse I really did not see any difference using your DSDT for UD5 vs. my UD4P. I am now using the one I build up from clean UD4P DSDT.

 

LocusOfControl,

 

I also think it has something to do with USB. Unfortunately not too many people are using the BT magic mouse with the Gigabyte board. I have read through multiple threads on Gigabyte board DSDT mod but haven't gotten anything to work.

 

What do you think Wake - RTC USB4 means?

Link to comment
Share on other sites
d00d
d00d
Posted
  • Author
Posted
PS got my DSDT down to 1870 lines ~ 4.5K now ; fakesmc only
You're using a natively supported add on NIC for network.

Are you using the no additional kext audio option 14A?

Do you live with the orange drive icons, or are you using IDE1 device-id 2681 that shows ESB2 instead of ICH10 in System Profiler:Hardware:Serial-ATA?

Two people reported drive failure and corruption using that solution, start with post 1011;

http://www.insanelymac.com/forum/index.php...518&st=1010

Link to comment
Share on other sites
LocusOfControl
LocusOfControl
Posted
Posted
Before I used the magic mouse I really did not see any difference using your DSDT for UD5 vs. my UD4P. I am now using the one I build up from clean UD4P DSDT.

 

LocusOfControl,

 

I also think it has something to do with USB. Unfortunately not too many people are using the BT magic mouse with the Gigabyte board. I have read through multiple threads on Gigabyte board DSDT mod but haven't gotten anything to work.

 

What do you think Wake - RTC USB4 means?

 

I get this in kernel log for idle sleep - wake by keyboard attached to EHCI port 5

 

Jan 29 16:55:57 LocusOfControl kernel[0]: System Sleep
Jan 29 17:34:11 LocusOfControl kernel[0]: Wake reason = EHC1 HDEF EHC2
Jan 29 17:34:11 LocusOfControl kernel[0]: System Wake
Jan 29 17:34:11 LocusOfControl kernel[0]: Previous Sleep Cause: 0
Jan 29 17:34:11 LocusOfControl kernel[0]: USB (EHCI):Port 5 on bus 0xfa has remote wakeup from some device

and in system log

Jan 29 17:34:20 LocusOfControl configd[13]: Sleep: Success - AC - Idle Sleep
Jan 29 17:34:20 LocusOfControl configd[13]: Wake: Success - AC - EHC1 HDEF EHC2
Jan 29 17:34:20 LocusOfControl configd[13]: Hibernate Statistics

 

 

Not sure what yours means, I'll turn wake using BT on and see what happens

 

 

 

 

You're using a natively supported add on NIC for network.

Are you using the no additional kext audio option 14A?

Do you live with the orange drive icons, or are you using IDE1 device-id 2681 that shows ESB2 instead of ICH10 in System Profiler:Hardware:Serial-ATA?

Two people reported drive failure and corruption using that solution, start with post 1011;

http://www.insanelymac.com/forum/index.php...518&st=1010

 

 

Yes using native NIC so a little cheat :D

 

Yes using native audio as I only have front speakers aka mac

 

No I have real icons for HD's - Yes I used the SATA fix with ESB2 - they were using Seagate whereas I prefer WD

(this was never actually proven and certain significant others were doubting - I've been using over a month without

issue)

 

Using DSDT shutdown fix with RC boot fix for restart and UUID

 

Your Processor details for 920 i7 f9m

Link to comment
Share on other sites
LocusOfControl
LocusOfControl
Posted
Posted
I get this in kernel log for idle sleep - wake by keyboard attached to EHCI port 5

 

Jan 29 16:55:57 LocusOfControl kernel[0]: System Sleep
Jan 29 17:34:11 LocusOfControl kernel[0]: Wake reason = EHC1 HDEF EHC2
Jan 29 17:34:11 LocusOfControl kernel[0]: System Wake
Jan 29 17:34:11 LocusOfControl kernel[0]: Previous Sleep Cause: 0
Jan 29 17:34:11 LocusOfControl kernel[0]: USB (EHCI):Port 5 on bus 0xfa has remote wakeup from some device

and in system log

Jan 29 17:34:20 LocusOfControl configd[13]: Sleep: Success - AC - Idle Sleep
Jan 29 17:34:20 LocusOfControl configd[13]: Wake: Success - AC - EHC1 HDEF EHC2
Jan 29 17:34:20 LocusOfControl configd[13]: Hibernate Statistics

 

I get the same log entries if I turn BT wake on, however it doesn't sleep for about 1.5 hrs then goes to sleep but

instead powers off (not crashes)

Link to comment
Share on other sites
ttgolf
ttgolf
Posted
Posted
I get the same log entries if I turn BT wake on, however it doesn't sleep for about 1.5 hrs then goes to sleep but

instead powers off (not crashes)

 

Hi LocusOfControl,

 

Interesting. How come your BT port is on EHC not UHC port? Did you attach the BT dongle in the USB port back of the computer or another hub? I have a 24" Apple Display. If I attach the BT dongle on the USB port of the screen, i can't wake the computer at all. It sometime will reboot upon waking the computer.

 

Also when you say the computer stays on for 1.5hr, does your computer just stay on for 1.5hr and finally sleep? Or it will try to idle sleep but it will wake itself up like the issue I have and eventually idle sleep after 1.5hr?

 

BTW, do you have Wake on Ethernet turn on? I believe that was the reason why I was getting RTC USB wake. But not all the time. I still do not know what maintenance sleep is. I also think it might have something to do with Wake on Ethernet. I am using a Netgear GA311 running natively without the Realtek1000SL.kext.

 

Lastly, how did you get your DSDT down to 1800ish lines? What did you remove? The last copy you sent me still have over 3000 lines.

 

d00d,

 

I tried Audio option A and I now don't have any popping sound when the computer wake from sleep. When I was using ALC889a there will be an annoying popping sound when the computer wakes up. Not sure if anyone else is experiencing the same thing. I don't mind having more sound assertion messages.

 

This is an awesome thread. Thanks.

Link to comment
Share on other sites
LocusOfControl
LocusOfControl
Posted
Posted
Hi LocusOfControl,

 

Interesting. How come your BT port is on EHC not UHC port? Did you attach the BT dongle in the USB port back of the computer or another hub? I have a 24" Apple Display. If I attach the BT dongle on the USB port of the screen, i can't wake the computer at all. It sometime will reboot upon waking the computer.

 

Also when you say the computer stays on for 1.5hr, does your computer just stay on for 1.5hr and finally sleep? Or it will try to idle sleep but it will wake itself up like the issue I have and eventually idle sleep after 1.5hr?

 

BTW, do you have Wake on Ethernet turn on? I believe that was the reason why I was getting RTC USB wake. But not all the time. I still do not know what maintenance sleep is. I also think it might have something to do with Wake on Ethernet. I am using a Netgear GA311 running natively without the Realtek1000SL.kext.

 

Lastly, how did you get your DSDT down to 1800ish lines? What did you remove? The last copy you sent me still have over 3000 lines.

 

d00d,

 

I tried Audio option A and I now don't have any popping sound when the computer wake from sleep. When I was using ALC889a there will be an annoying popping sound when the computer wakes up. Not sure if anyone else is experiencing the same thing. I don't mind having more sound assertion messages.

 

This is an awesome thread. Thanks.

 

 

I was playing around with the _PTS settings in the DSDT and I was getting the computer to switch off every 10 mins

with BT wake ON, unfortunately instead of sleeping it would switch off (which I think is S5 state?). The first time

I tried this it slept after 10 mins and I was able to wake the computer with MM, then 10 mins later it slept, but it wouldn't

wake because it was switched off (not KP). After rebooting I checked various lspci outputs, doing a diff on before

and after sleep outputs I was getting some differences from lspci -d 8086:3a16 -xxxvvv

 

My naive assumption is that certain things that should be saved prior to sleep are not being saved, unfortunately

I don't know enough yet to say what. Confusingly it is not always reproducible indicating that several factors maybe

at work

 

With BT wake OFF, sleep was working normally

 

I assume the sections in the DSDT marked S0 to S5 are sleep states, I also noted that what I expect to be S1 is

marked SS1 on the EX58 (same in D00d's file)

 

Wrt to my DSDT I just followed the naming convention on the other MC DSDT thread's followed with experimental

removal of what was left e.g. all the LNK and BUF stuff etc laboriously taking 1 section at a time and retesting, checking

my addressed with lspci etc. Probably another 100 lines or so to go

 

No I don't have WOL on, I thought it was stopping sleep when I booted my windows laptop up, but I'm not sure

now

 

ttyskeepawake - prevent idle system sleep when any tty (e.g. remote login

session) is ’active’. A tty is ’inactive’ only when its idle time exceeds

the system sleep timer. (value = 0/1)

 

so I set pmset ttyskeepawake 0

Link to comment
Share on other sites
LocusOfControl
LocusOfControl
Posted
Posted
I was playing around with the _PTS settings in the DSDT and I was getting the computer to switch off every 10 mins

with BT wake ON, unfortunately instead of sleeping it would switch off (which I think is S5 state?). The first time

I tried this it slept after 10 mins and I was able to wake the computer with MM, then 10 mins later it slept, but it wouldn't

wake because it was switched off (not KP). After rebooting I checked various lspci outputs, doing a diff on before

and after sleep outputs I was getting some differences from lspci -d 8086:3a16 -xxxvvv

 

My naive assumption is that certain things that should be saved prior to sleep are not being saved, unfortunately

I don't know enough yet to say what. Confusingly it is not always reproducible indicating that several factors maybe

at work

 

With BT wake OFF, sleep was working normally

 

I assume the sections in the DSDT marked S0 to S5 are sleep states, I also noted that what I expect to be S1 is

marked SS1 on the EX58 (same in D00d's file)

 

Wrt to my DSDT I just followed the naming convention on the other MC DSDT thread's followed with experimental

removal of what was left e.g. all the LNK and BUF stuff etc laboriously taking 1 section at a time and retesting, checking

my addressed with lspci etc. Probably another 100 lines or so to go

 

No I don't have WOL on, I thought it was stopping sleep when I booted my windows laptop up, but I'm not sure

now

 

ttyskeepawake - prevent idle system sleep when any tty (e.g. remote login

session) is ’active’. A tty is ’inactive’ only when its idle time exceeds

the system sleep timer. (value = 0/1)

 

so I set pmset ttyskeepawake 0

 

Going back to a copy of the dsdt that provided the above results didn't allow me to reproduce this. Hence I assume

it was just a fluke (or some initial conditions I am now missing)

Link to comment
Share on other sites
ryansimms
ryansimms
Posted
Posted

I just found out something interesting today. When my computer is asleep and I use the Apple's "Remote" app on my iPhone (app which controls iTunes) the computer wakes up. I was pleasantly surprised by this behavior and wonder if it may lead to a proper implementation of "Wake for network administrator access," which isn't even showing up in my Energy Saver preferences.

 

Ryan!

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