Jump to content

Wake from Sleep - Freeze - Due to ejected HDD's


Blah101
 Share

54 posts in this topic

Recommended Posts

I used to use orangeiconfix.kext. Before that ,it was years ago, I used different injector which I don't remember it's name.

When I use that kext ,there wasn't eject button in finder. but orangeiconfix.kext made the eject button in finder.

Maybe if you use that kext which makes hd internal, problem might disappear.

Is anyone using that kext which shows no eject button in finder?

Link to comment
Share on other sites

Did you try the more recent, simpler workaround from post #20?

 

 

Try the more recent, simpler workaround from post #20. That way you can keep with DP1's vanilla kexts.

 

 

Interesting find kdtt, thanks for sharing. I'll have to check my BIOS to see if I have an option for this also. Hot-plug is one of the features of ICH10 and that feature and the problem disappears when I match ESB2 to my deviceID, so it's definitely related. So maybe that's the issue.! :)

 

For ref, I'm using two Samsing HDD's

 

 

No sleepenabler and just DSDT with device (HPET). See above as to why you shouldn't need the 10.7.3 kexts.

 

Not works with me :) i tested your file as before :)

Link to comment
Share on other sites

I used to use orangeiconfix.kext. Before that ,it was years ago, I used different injector which I don't remember it's name.

When I use that kext ,there wasn't eject button in finder. but orangeiconfix.kext made the eject button in finder.

Maybe if you use that kext which makes hd internal, problem might disappear.

Is anyone using that kext which shows no eject button in finder?

 

The OrangeIconFix.kext doesn't affect the problem. It's cosmetic and only makes OS X use internal icons for external devices therefore leaving ICH10's hot-plug feature still active and the eject buttons still appear in the Finder. Using this, my drives are still lost after a sleep/wake cycle.

 

The other way which doesn't change the deviceID's yet allows the drives appear as internal drives and not ejectable in the Finder is using the patched IOAHCIBlockStorage.kext. This doesn't make any difference for this problem either.

 

Your BIOS allows you to disable the hot-plug feature - which looks like being the reason for this problem here, where as my BIOS doesn't so the only fix I've so far found is to make OS X think I am not using ICH10 by changing my deviceID, which in essence removes the hot-plug feature.

 

Not works with me :) i tested your file as before :)

What southbridge does your board use?

Link to comment
Share on other sites

I saw in google that there was a similar problem in windows 7, quite a lot people reported that they have disappearing hard disks after wake.

and microsoft released hot fix for that. They said the reason was new eco drives and big sata drives need more time to initialize.

I guess hot plug enabled ich10 needs more time.

Link to comment
Share on other sites

Well my two HDD's are not new and only 500GB each. And they work perfectly well in every other OS type and version, just not when waking from sleep with 10.8 DP1's AHCI kexts and my ICH10.

 

I guess I've got to find out if this issue happens for every user using ICH10?

Link to comment
Share on other sites

Same wake from sleep problem on ICH9, the ESB2 fix works fine here.

Hi STLVNUB - Thanks for the report that it's not only ICH10 that's affected.

So any thoughts on this issue - Could it be a bug or a product of design maybe?

 

Now I have a slightly different problem. Although I can wake from sleep, shut down gives me a CMOS reset when I turn on. Any ideas?

For that, you need to read the CMOS reset thread here (which by the way is incorrectly placed in the installation sub forum and not the post installation sub forum - mods?)

 

ps. BTW blah101 - can you maybe rename this thread to something more descriptive of this problem? - perhaps something like... Wake from Sleep - Freeze - Due to ejected HDD's

Link to comment
Share on other sites

Hi STLVNUB - Thanks for the report that it's not only ICH10 that's affected.

So any thoughts on this issue - Could it be a bug or a product of design maybe?

 

 

For that, you need to read the CMOS reset thread here (which by the way is incorrectly placed in the installation sub forum and not the post installation sub forum - mods?)

 

ps. BTW blah101 - can you maybe rename this thread to something more descriptive of this problem? - perhaps something like... Wake from Sleep - Freeze - Due to ejected HDD's

Renamed, also I will move the solutions to the first post this weekend. Thank you, I will check out that thread.

Link to comment
Share on other sites

Hi efimac

 

NOTE: This is only for testing and not something I'd recommend as a permanent fix.

I used an old trick to change the device-id - You can find read about it here: http://www.insanelym...dpost&p=1253758

 

It's just another step towards trying to pinpoint, or at least to help with trying to understand why I experience loss of drives after a sleep/wake cycle. I now know at least that it's something to do with using my native ICH10 with 10.8 DP1's AppleAHCIPort.kext and IOAHCIFamily.kext.

 

Thanks , can you please make an CodeBox with some more code around the patch (ESB2) , maybe orig, vs. patched code section, to see better were i must do the changes?

I have an ICH9 Chipset with Gigabyte EP35 board.

Link to comment
Share on other sites

Hi mitch_de

 

When booting with Chameleon I now use the LegacyAppleAHCIPort.kext in /Extra/10.8/Extensions. But I still use the DSDT patch when I'm booting with Clover as the kext somehow clahses with Clovers GraphicsEnabler.

 

This is what I used in the DSDT...

 

Before:

Device (SATA)
		{
			Name (_ADR, 0x001F0002)
		}

 

After:

Device (SATA)
		{
			Name (_ADR, 0x001F0002)
			Method (_DSM, 4, NotSerialized)
				{
					Store (Package (0x02)
					{
					"device-id",
					Buffer (0x04)
					{
						0x81, 0x26, 0x00, 0x00
					}
				}, Local0)
				DTGP (Arg2, RefOf (Local0))
				Return (Local0)
			}
		}

  • Like 1
Link to comment
Share on other sites

Thanks for returning with your details hiphopboy.

You're the first to report ICH7 also suffers this issue. So that's ICH10R, ICH9 and ICH7 now.

 

So what happened when you tried the temporary workaround of using the LegacyAppleAHCIPort.kext ?

Did you edit the kext to match your deviceID against the ESB2 AHCI wording as shown in post #20?

I ask because if it works for other I/O controllers I wonder why it didn't work for yours.

  • Like 1
Link to comment
Share on other sites

i think we need to add code for _EJT for port 0 ?

Good thinking LatinMcG. It's definitely worth an investigation. I see a few _EJ.. options in the ACPI spec.

 

I've been thinking about other possibilities too and came up with the question of could it be something to do with the fact that a lot of new Macs use AHCI Version 1.3?. My mobo uses v1.2 and I'm guess that'll apply to most of the users in this thread. Maybe there's a connection between Apple's IOAHCIFamily.kext and AHCI v1.3? or maybe a connection between AHCI v1.3 and the DSDT? I've got the AHCI specifications of both versions here so I'll try to find anything that jumps out as a possible.

 

As you say, we need to find a few system details from user who are not affected by this and then try to draw some common facts.

 

EDIT:

LatinMcG - I guess you were thinking something like this:

_EJ0 indicates the device supports hot removal; _EJ1–EJ4 indicate the device supports warm removal.

For warm removal, the _EJ1–_EJ4 control methods do not cause the device to be immediately ejected.
Instead, they set proprietary registers to prepare the hardware to eject when the system goes into the
given sleep state. The hardware ejects the device only after OSPM has put the system in a sleep state
by writing to the SLP_EN register. After the system resumes, OSPM calls _STA to determine if the eject
succeeded.

6.3.3 _EJx (Eject) - Page 243 - ACPIspec40a

  • Like 1
Link to comment
Share on other sites

im on mountain on G31m-ES2L rev 1.x with fakesmc , AppleHDA for alc883 with dsdt and no hdaenabler , lnx2mac v.90 Lan driver.

wake seems to work fine so far.. except i had a bios reset on 2 sleeps then restart.. (updating chameleon from buildbot as possible solution ? or rtc patch time.. i got single ram bank 128MB message)

but my wake is a ok so far with heavily modded (trimmed fat) dsdt

Link to comment
Share on other sites

Thanks for returning with your details hiphopboy.

You're the first to report ICH7 also suffers this issue. So that's ICH10R, ICH9 and ICH7 now.

 

So what happened when you tried the temporary workaround of using the LegacyAppleAHCIPort.kext ?

Did you edit the kext to match your deviceID against the ESB2 AHCI wording as shown in post #20?

I ask because if it works for other I/O controllers I wonder why it didn't work for yours.

 

yep ! I tested your solution butbut it's not works with me :( ! Hope for your works with my mobo, it's works on Lion but can't on M. Lion :(

Link to comment
Share on other sites

I found a thread about an app called hdapm in Apple support communities page.

This app is for adjusting the level of hard disk's power saving mode.

I tried this app ,but if fails to start on my board. maybe because value of alpm in my ioregistery is false.

if there's anyone who want to try this app for their board, this is link.

http://mckinlay.net.nz/hdapm/

and this link is to discussion page of macbook users about this app.

https://discussions.apple.com/thread/3137281?start=0&tstart=0

Link to comment
Share on other sites

  • 2 weeks later...

I am running DP2 on a Gigabyte P55 board. Using blackosx's recomemdation resolved my Disk Eject issue after wake from sleep.

 

AppleAHCIPort.kext v2.4.0 from 10.8 DP1 replaced with version 2.2.0 from 10.7.3

IOAHCIFamily.kext v2.2.0 from 10.8 DP1 replaced with version 2.0.7 from 10.7.3

Link to comment
Share on other sites

  • 1 month later...
  • 1 month later...

I saw in google that there was a similar problem in windows 7, quite a lot people reported that they have disappearing hard disks after wake.

and microsoft released hot fix for that. They said the reason was new eco drives and big sata drives need more time to initialize.

I guess hot plug enabled ich10 needs more time.

 

I have WD Black drive and OCZ Vertex 3 SSD, and they eject on wake from sleep in ML DP 4. I don't think this issue is rooted in "eco" or green drives.

 

An easier away to change your SATA to ESB2 is by using the LegacyAppleAHCIPort.kext as mentioned by MasterChief here. Far easier and cleaner than editing the DSDT.

 

Edit the info.plist and change the Chipset Name of the device matching yours to EBS2 AHCI.

So for me, searching for 0x3A228086 brings me to the ICH10RAHCI key. Then changing the subsequent:

<key>Chipset Name</key>
<string>ICH10-R AHCI</string>

to

<key>Chipset Name</key>
<string>ESB2 AHCI</string>

is all I need.

 

Simply place the modded kext in to /Extra/Extensions and reboot.

EDIT: to make it 10.8 only, place the kext in /Extra/10.8/Extensions then it won't get loaded for 10.7 or lower.

LegacyAppleAHCIPort.kext.zip

There are many versions of this dummy kext around but I think original credits go to cyclonefr and netkas.

 

Again though. This is a workaround until a proper solution is found.

 

BlackOSX, this works great for me if the boot drive is connected to SATA II on the motherboard (ep45-ud3p rev. 1.6 with Cartri MacBIOS). All of my 3 drives connected to the motherboard do not eject on wake, and ML DP4 works beautifully.

 

However, if the OS drive is connected to a generic controller, such as ASMedia (asm1061) for my SSD with SATA III, and the Generic AHCI and ICH10-R AHCI changed to ESB2 AHCI, the OS drive does not mount after wake (can't tell if the other drives on the motherboard's SATA II aren't ejected since the system is locked up).

 

The temporary solution for me, using ASMedia controller and booting with Vertex 3 SSD with ML DP4, is to revert back AppleAHCIPort and IOAHCIFamily kexts.

 

Has there been any progress on finding a more permanent solution to this ejecting on wake issue?

Link to comment
Share on other sites

 Share

×
×
  • Create New...