Jump to content

USB 2.0 for Leopard 10.5.7

Slice

By Slice,
in Other Peripheral Devices

 Share
390 posts in this topic

Recommended Posts

Slice
Slice
Posted
  • Author
Posted
Hi Slice,

 

I have an Asus A8V deluxe mobo and 10.5.5 with voodoo kernel installed.

 

I'm triying your kext becouse I've read that other people with the same mobo managed to have EHCI working with the PCGenUSBEHCI-261.4.2a.zip but in tiger.

 

Is there any way of porting this "a" variant of the kext to leopard?

 

With your kext I can see devices conected in the system profiler under high speed bus, but they don't mount.

 

Thanks in advance for your great job.

I made a port of PCGen... for Leopard. Here it is.

I don't know what is your problem. Try different kext's version's combinations.

Link to comment
Share on other sites
mitch_de
mitch_de
Posted
Posted
I made a port of PCGen... for Leopard. Here it is.

I don't know what is your problem. Try different kext's version's combinations.

 

Hi slice, my USB 2.0 working at GA-EP35-DS3 ICH9 Chipset.

 

But i have some ??? when i look in my IoregistryExplorer about USB3:

 

USB3 didnt have such IOregEntries as USB0-5, US31,... have (right screenshoot)

Only 2 lines (ledt screenshoot)

 

I use dsdt, but part od USB isnt patched.

 

I cant see any differences in the dsdt part between USB 2 (for example, DSDT.dsl if you have) and USB3(wich isnt registered as the others).

Device (USB2)

{

Name (_ADR, 0x001D0002)

Method (_S3D, 0, NotSerialized)

{

If (LEqual (OSFL, 0x02))

{

Return (0x02)

}

Return (0x03)

}

Name (_PRW, Package (0x02)

{

0x0C,

0x01

})

}

Device (USB3)

{

Name (_ADR, 0x001D0003)

Method (_S3D, 0, NotSerialized)

{

If (LEqual (OSFL, 0x02))

{

Return (0x02)

}

Return (0x03)

}

Name (_PRW, Package (0x02)

{

0x0E,

0x01

})

}

Device (US31)

{

Name (_ADR, 0x001A0000)

Method (_S3D, 0, NotSerialized)

{

If (LEqual (OSFL, 0x02))

{

Return (0x02)

}

Return (0x03)

}

Name (_PRW, Package (0x02)

{

0x0E,

0x01

})

}

Device (USB4)

{

Name (_ADR, 0x001A0001)

Method (_S3D, 0, NotSerialized)

{

If (LEqual (OSFL, 0x02))

{

Return (0x02)

}

Return (0x03)

}

 

Name (_PRW, Package (0x02)

{

0x05,

0x01

})

}

How are your USB Devs seen (in the IORegistry Exploerer) / or other Users with an ICH9® Chipset ?

 

added 2 screenshoots

Bild_232.jpg

Bild_231.jpg

Link to comment
Share on other sites
josu
josu
Posted
Posted

OK, in the "a" release of Orbyte's PCgen... he said.

 

NEW VERSION RELEASE NOTE

 

I am pleased to announce a new version of the PCGenUSBEHCI series. Also I would like to thank to mac fly for suggesting a solution for cases where PCGen could not acquire OS Ownership due to completely buggy (i.e., ignoring OS reset request) SMI handling. The solution was to make a forced BIOS ownership reset, after all other approaches had failed. The code does therefore retain backward compatibility.

 

In essence, this release brings us in line with the Linux 'pci-quirks' handling, and we should now be able to get OWNERSHIP after reboot on all (edit: I should say 'almost all' :-) ) systems. However, please note that there other other documented causes for non-functional EHCI, and these remain. In my archives, I have listed an 'Unable to get device memory error' with ATI SB200 - this is a completely different kind of error. Also there appear problems with the nVidia nf4 under DFI LANPARTY - which are not related to OS Ownership. These remain under investigation, with the hope that we will one day have as broad an EHCI driver as possible.

 

Installation and usage instructions are included in the download archives. The download area is as usual at the beginning of this post.

 

There is now in the Info.plist file, a parameter called

CODE

<key>AlwaysHardBIOSReset</key>

<true/>

 

This should be fine on most systems. However, if the system hangs after reboot, switch this to <false/>. Very likely, you WILL need to do this on 10.4.6

 

 

I think that this parameter is what I need to solve my problem and I think that this is not implemented in your kext, is it?

 

Thanks

Link to comment
Share on other sites
relikwie
relikwie
Posted
Posted
SLICE:

One thing the YOUR great USB drivers FIXED for me (at 10.5.6, i posted before some IOMASS.. problems) is to go to sleep with hit on USB Keyboard. :)

Before since 10.5.6 i must short push power button on pc case. USB Key hit didnt wake up. THATS FIXED using your usb drivers.

Only that USB Mouse (also waked up system in 10.5.5) didnt wake up system is the difference to 10.5.5 before, for me not a problem because USB key hit now works.

THANKS Merry Christmas !!!!

 

 

Out of interest, did you also have the GA-EP35-DS3P board?

I've never had any issues with mass storage or any USB devices, but also never been able to wake from sleep via USB.

 

If it works for you and we share the same hardware I should try this out.

Link to comment
Share on other sites
mitch_de
mitch_de
Posted
Posted
Out of interest, did you also have the GA-EP35-DS3P board?

I've never had any issues with mass storage or any USB devices, but also never been able to wake from sleep via USB.

 

If it works for you and we share the same hardware I should try this out.

 

I have an EP-DS3 no P board. If you have also same bios (F3, F4a beta ) we may have near same hardware.

If GA bios list shows complete differnet bios versions , we didnt have same hardware (P, not P).

 

What do you means "..but also never been able to wake from sleep via USB."

I can press an Key on USB Keyboard or push Mouse Button (USB) to wake up.

Moving mouse only does not wake up - i must press any mouse button.

Perhaps check also bios settings (Power), perhaps there are some wake up settings (wake up by...) which can help enabled.

Link to comment
Share on other sites
firstcolle
firstcolle
Posted
Posted

i update to 10.5.6 and used this las kext with UUID.kext and my old AppleACPIPlatform.kext and IoACPIFamily.kext. Now my sleep work but at sturtup i have this in yello:

Jan  8 17:52:15 localhost kernel[0]: Class "AppleEHCIedMemoryBlock" is duplicate
Jan  8 17:52:15 localhost kernel[0]: Duplicate class
Jan  8 17:52:15 localhost kernel[0]: kmod_control/start failed for com.orByte.driver.PCGenUSBEHCI; destroying kmod
Jan  8 17:52:15 localhost kernel[0]: ApplePS2Keyboard: probe failed
Jan  8 17:52:15 localhost kernel[0]: ^[[33mFailed to load extension com.orByte.driver.PCGenUSBEHCI.
Jan  8 17:52:15 localhost kernel[0]: Class "AppleEHCIedMemoryBlock" is duplicate
Jan  8 17:52:15 localhost kernel[0]: Duplicate class
Jan  8 17:52:15 localhost kernel[0]: kmod_control/start failed for com.orByte.driver.PCGenUSBEHCI; destroying kmod
Jan  8 17:52:15 localhost kernel[0]: ^[[33mFailed to load extension com.orByte.driver.PCGenUSBEHCI.

 

and my usb work in 1.0 mode... what ca i try??

 

thanks

Link to comment
Share on other sites
relikwie
relikwie
Posted
Posted
I have an EP-DS3 no P board. If you have also same bios (F3, F4a beta ) we may have near same hardware.

If GA bios list shows complete differnet bios versions , we didnt have same hardware (P, not P).

 

What do you means "..but also never been able to wake from sleep via USB."

I can press an Key on USB Keyboard or push Mouse Button (USB) to wake up.

Moving mouse only does not wake up - i must press any mouse button.

Perhaps check also bios settings (Power), perhaps there are some wake up settings (wake up by...) which can help enabled.

 

Think they differ then. I use BIOS F6a (beta). The latest stabe is F5.

But I replaced AppleUSB(U,E,O)HCI with the those found on the first page and yes (couldnt believe it) but my pc wakes by clicking USB mouse button. NICE!!

 

I have the aluminium Apple keyboard and pressing keys on it doesnt wake it. So thats bad luck.

 

In BIOS the only option is to use PS2 keyb/mouse to start the pc and such also simulate a puh on the powerbutton (which always wakes pc from sleep).

 

Anyhow, a bit closer.. Thanks!

Link to comment
Share on other sites
Slice
Slice
Posted
  • Author
Posted
OK, in the "a" release of Orbyte's PCgen... he said.

 

AlwaysHardBIOSReset

 

This should be fine on most systems. However, if the system hangs after reboot, switch this to . Very likely, you WILL need to do this on 10.4.6

I think that this parameter is what I need to solve my problem and I think that this is not implemented in your kext, is it?

 

Thanks

Yes, this parameter is implemented but not written in info.plist because I never see how it works. You may try to add it. Report please!

 

i update to 10.5.6 and used this las kext with UUID.kext and my old AppleACPIPlatform.kext and IoACPIFamily.kext. Now my sleep work but at sturtup i have this in yello:

Jan  8 17:52:15 localhost kernel[0]: Class "AppleEHCIedMemoryBlock" is duplicate
Jan  8 17:52:15 localhost kernel[0]: Duplicate class
Jan  8 17:52:15 localhost kernel[0]: kmod_control/start failed for com.orByte.driver.PCGenUSBEHCI; destroying kmod
Jan  8 17:52:15 localhost kernel[0]: ApplePS2Keyboard: probe failed
Jan  8 17:52:15 localhost kernel[0]: ^[[33mFailed to load extension com.orByte.driver.PCGenUSBEHCI.
Jan  8 17:52:15 localhost kernel[0]: Class "AppleEHCIedMemoryBlock" is duplicate
Jan  8 17:52:15 localhost kernel[0]: Duplicate class
Jan  8 17:52:15 localhost kernel[0]: kmod_control/start failed for com.orByte.driver.PCGenUSBEHCI; destroying kmod
Jan  8 17:52:15 localhost kernel[0]: ^[[33mFailed to load extension com.orByte.driver.PCGenUSBEHCI.

 

and my usb work in 1.0 mode... what ca i try??

 

thanks

Duplicate class

com.orByte.driver.PCGenUSBEHCI -- erase it!

Link to comment
Share on other sites
keftes
keftes
Posted
Posted
First of all thank to orByte for the method to correct USBEHCI driver

http://forum.insanelymac.com/index.php?showtopic=28559

He is disappeared making only version 303 that is not enough.

I open new topic because I can't update orByte's topic with new version.

 

Previous my correction USB308 works!

Now sources 315.4.1 is available at Apple's site and I can easy rewrite it by orByte's method.

This is corrected sources and my compilation. All targets can be compiled except IOUSBFamily and IOUSBLib. Use stock similar.

 

EDITED 03.08.08

My kexts are not tested yet so be ready to return to version 308.

 

EDITED 21.11.08

I corrected EHCI and UHCI plugins for sleep support

Instructions how to install

 

That's all!

 

EDITED 19-09-2008

Apple publish 10.5.5 sources. IOUSBFamily is the same but IOUSBMassStorageClass is updated!

There is my compilation with small corrections.

 

EDITED 19-12-2008

New compilation from 10.5.6 sources. I change version to 3.2.8 to differ from Apple's kexts.

What new? I add support for Intel ICH4-5, 9 (ICH10 added by Apple), NForce (MCP79 by Apple, MCP78 and other by me), SIS7001.

Some internal mistakes corrected.

All files tested in 10.5.5 so it compatible with old systems (Leopard only!).

You also may try my IOPCIFamily.kext to improve interrupts handling.

AppleUSBEHCI328.kext.zip

AppleUSBOHCI328.kext.zip

AppleUSBUHCI328.kext.zip

IOUSBMassStorageClass205.kext.zip

 

OK!!!!!!

 

I installed your kexts....SLEEP & WAKE both work. I can even use my keyboard to wake. PERFECT!!!!

 

I noticed that when I came out of sleep and tried to shutdown, the system would panic. I restored from TM and installed each file 1 at a time. AppleUSBUHCI.kext did the trick. Now it sleeps, wakes, shuts down, and restarts.

 

 

Thank you.

 

P5k Deluxe Wifi (not using wifi. using ethernet via access point)

Q6600 @ 3.34 GHz

Nvidia 8600GT 512

4GB Ram

Retail 10.5.6

 

I can honestly say.....FINALLY......the system is 100% functional and blazing fast!

 

Thank you for your hard work.

Link to comment
Share on other sites
Onion_Skin
Onion_Skin
Posted
Posted

Hi Slice,

 

My hardware configuration is like this

MB: Asus P5LD2-VM intel 945+ICH7

CPU: Intel C2D 6320

Memory: 2x2GB

Video: Sapphire HD2600XT

HD: 1x IDE Seagate 250GB 1x SATA Seagate 500GB and 1x SATA Seagate 1TB

 

I installed iDenib 10.5.5, and upgrade it to 10.5.6 with PC-EFI v9. My problems are:

1. I have 8 USB ports in total, and all of them function perfectly in Windows XP. However, 2 of them can't be used in Leopard;

2. USB bus is in UCHI mode. File copying to flash drive is very slow at only ~2MB/s. I think this is typical USB 1.1 performance.

 

I tried your driver version 328. ECHI driver loaded but with some errors in system.log. please see my system log

systemlog.txt

and here is my dmesg info

dmesg.txt

Here are all USB drivers loaded

bash-3.2# kextstat | grep USB

41 9 0x49e0e000 0x29000 0x28000 com.apple.iokit.IOUSBFamily (3.2.7) <7 6 5 4 2>

42 0 0x49e7a000 0xe000 0xd000 com.apple.driver.AppleUSBUHCI (3.2.8) <41 17 7 6 5 4>

43 0 0x49f23000 0x12000 0x11000 com.apple.driver.AppleUSBEHCI (3.2.8) <41 17 7 6 5 4>

57 0 0x44c11000 0x2000 0x1000 com.apple.iokit.IOUSBUserClient (3.2.4) <41 6 5 4>

69 0 0x4a0e4000 0xc000 0xb000 com.apple.driver.AppleUSBHub (3.2.7) <41 6 5 4>

80 0 0x49ea5000 0x4000 0x3000 com.apple.driver.AppleUSBComposite (3.2.0) <41 5 4>

82 0 0x4a288000 0x5000 0x4000 com.apple.iokit.IOUSBHIDDriver (3.2.2) <41 24 6 5 4>

84 0 0x49cb6000 0x3000 0x2000 com.microsoft.driver.MicrosoftMouseUSB (6.2.2) <83 41 12>

86 0 0x4a516000 0xa000 0x9000 com.apple.iokit.IOUSBMassStorageClass (2.0.5) <48 44 41 12>

 

However, USB bus is still in UCHI mode. The file copying speed is still the same.

post-325099-1231475757_thumb.jpg

 

Can you help see what's wrong?

 

Thanks for your great effort!

Link to comment
Share on other sites
Onion_Skin
Onion_Skin
Posted
Posted

Hi Slice,

 

I did an additional performance test on the new USB 2.0 driver. I have a 4GB 150X SDHC card. In windows XP, copying a 700M video file to this card takes 45 seconds. In comparison, copying the same video file to this card takes twice the time(~1.5 minutes) in Leopard. Is this the expected performance or my Leopard installation is still in sub-optimal status?

 

MB: Asus P5LD2-VM intel 945+ICH7

CPU: Intel C2D 6320

Memory: 2x2GB

Video: Sapphire HD2600XT

HD: 1x IDE Seagate 250GB 1x SATA Seagate 500GB and 1x SATA Seagate 1TB

iDenib 10.5.5 upgraded to 10.5.6 with PC-EFI v9

Link to comment
Share on other sites
Beerkex'd
Beerkex'd
Posted
Posted

IOPCIFamily254 restored sleep and shutdown on my hackintosh. :)

I can wake up from my wireless Logitech USB keyboard/mouse now, this never worked in the past, had to use the power button.

What's more, my USB thumb drive is correctly re-mounted on wake...most of the time.

I've got your 328 USB kexts installed as well, inside vanilla 10.5.6 USBFamily. No need for EHCISleepEnabler.

 

WIN!

 

I've been going out of my mind trying to find a fix for shutdown/sleep the past few days.

 

Thanks Slice.

 

EDIT

 

Drat, it's not 100% there yet. Reboot doesn't work if the computer has been sleeping.

It goes through the reboot process but then it hangs, PC speaker beeping repeatedly.

At least it doesn't cause any disk corruption, fsck in single user mode has no complaints and the system boots right up again.

 

2ND EDIT

 

I've discovered that it only works if SuperHai's VoodooPower is installed as well.

http://forum.insanelymac.com/index.php?s=&...t&p=1021550

If I uninstall VoodooPower only restart works - I lose wake and shutdown.

 

3RD EDIT

 

I'm on a different installation now, but on the same hardware - and can't get wake and shutdown to work again no matter what I do.

Link to comment
Share on other sites
thestevo
thestevo
Posted
Posted

Hi Slice,

 

Just wanted to confirm that deleting AppleHPET brought my USB 2.0 back (Latitude D830), with vanilla USB kexts. Thanks for the suggestion!

 

Just hypothetically, do we have a potential solution for this problem? If AppleHPET became necessary in future versions, do we have an idea of how we would go about fixing this? DSDT? However, I understand that it is not needed for now. Thanks.

 

theStevo

Link to comment
Share on other sites
mitch_de
mitch_de
Posted
Posted

I am not sure if AppleHPET isnt used NOW !

HPET is an add to older RTC way but has some big advantages for timers (often used!).

HPET takes much less CPU cyles and is much more accuarte then the old/default RTC!

If HPET isnt available OS X will use RTC i think, so you dont see any error - but i believe its not good !

Link to comment
Share on other sites
josu
josu
Posted
Posted
Yes, this parameter is implemented but not written in info.plist because I never see how it works. You may try to add it. Report please!

Duplicate class

com.orByte.driver.PCGenUSBEHCI -- erase it!

 

Now I've been able to use my usb 2.0 devices but I have a strange behavior.

 

The ipod and the usb scanner work perfectly, but the usb pen-drives don’t. But if I have the ipod or the scanner plugged in previous plugging the pen-drive, they also work.

 

Any idea why does it happen?

Thanks for your help

Link to comment
Share on other sites
jonninja
jonninja
Posted
Posted

I am running iATKOS 10.5.5 .

Even after doing all of those kexts I am still having problems. If I boot with a storage device on it works and as long as I keep that on connected I can disconect and reconect other devices at will. However when I try to delete I get an error code "61".

Also if I do boot without anything connected when I do connect it will be in system profiler and growl notifier will tell me its connected. Right now being able to delete is more important for me. Anyone have any ideas?

 

Edit: If I don't boot with anything then no usb works.

Link to comment
Share on other sites
Beerkex'd
Beerkex'd
Posted
Posted

Mitch:

 

According to MercurySquad, AppleHPET is only needed if you're running "the apple supplied speedstep driver":

 

"The issue is with the new clock initialization routine. David Elliot (otherwise known as 'dfe') has a nice fix to disable hpet on unsupported systems, so we decided to remove the hpet emulation entirely since its only consumer - the apple supplied speedstep driver - is replaced by our speedstep kext and we replaced the clock initialization routine to read the bus ratio directly from the processor. This is implemented by all AMDs and most intel processors from pentium M and pentium 4 model 3 and above but the old willamette and northwood core p4's don't have this register. We had the choice to put a clock timing routine in the kernel to calculate the cpu frequency / bus ratio without which xnu will panic, but we learnt that the chameleon bootloader already does this so instead of reinventing the wheel we simply asked the chameleon devs to also export the cpu frequency in addition to the fsb frequency in the fake EFI tree. For users who cannot or do not want to upgrade their bootloader, there is a boot-time flag they can set to explicitly pass their bus ratio to the kernel."

 

http://forum.insanelymac.com/index.php?showtopic=127896

 

That could be outdated information though or I could be reading it wrong..I probably am because AppleHPET is loaded on my system.

Now I wonder if deleting it will cure my restart after sleep issue...maybe it doesn't work because the timing gets messed up after wake.

Link to comment
Share on other sites
mitch_de
mitch_de
Posted
Posted

Yeah, normaly an driver which isnst needed also isnt loaded !

I dont think that Apples speedstep (must be AppleIntelCPU.....) is the only thing which uses HPET.

HPET is also used by Linux or Win i think.

 

For me, the superhais speedstep isnt really good working without AppleHPET - trottling was bad !

 

I have AppleHPET loaded and my USB 2.0 is running - so people whoch have problems with HPET may have other problems (BIOS ?) and the delete of HPET does not solve that behind/really problem.

 

Better to try to find out if usage of dsdt patching solves the problem - maybe much more than only USB 2.0 troubles.

 

I for myself do:

- HDEF , AppleHDA Audio by dsdt

- ATI 3850 CI/QE by dsdt (insted of natit or efi gfx string)

- LAN / Timemachine fix by dsdt

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