Showing posts with label Core i3. Show all posts
Showing posts with label Core i3. Show all posts
Monday, July 11, 2011
Minimal DSDT Edits for Gigabyte Socket 1155 Motherboards
Posted by
MacManx86
on Monday, July 11, 2011
Labels: Core i3, Core i5, Core i7, DSDT, Gigabyte, H55, H67, Mac, MacManx86, OS X, OSx86, P55, P67, Sandy Bridge, snow leopard, Socket 1155, Z68
Comments: (10)
Labels: Core i3, Core i5, Core i7, DSDT, Gigabyte, H55, H67, Mac, MacManx86, OS X, OSx86, P55, P67, Sandy Bridge, snow leopard, Socket 1155, Z68
Comments: (10)
I started researching and testing what edits from my New Minimal DSDT Edits for Gigabyte Socket 1156 Motherboards were or weren't needed on the Socket 1155 boards. The good news is that the majority of the edits carry over, but some are not needed. For example, we don't need to inject anything for the USB ports to be fully recognized. But there are a couple of required Socket 1155 specific edits.
The following is a guide for editing the extracted DSDT. The first thing you will need to do is the required edits so that the DSDT will compile. They are:
1) "Reserved method must return a value"
Add to the end of _WAK
Return (Package (0x02)
{
0x00,
0x00
})
2) "Method local variable is not initialized"
In Scope (\_SI)
Change both:
Store (Local0, Local0)
To:
Store (0x00, Local0)
3) "Use of compiler reserved name"
Replace all _T_ with T_
4) "Effective AML package length is zero"
Change:
Name (RPMV, Package (0x00) {})
To:
Name (RPMV, Package (Zero) {})
5) "Result is not used, operator has no effect"
Change:
Or (0x03, PARM)
To:
Or (PARM, 0x03, PARM)
The first required edit is to include Method DTGP for passing configuration parameters to OS X. Next up is renaming device names to follow Apple's conventions. They are renaming all PX40 to LPCB, PX43 to SBUS, all AZAL to HDEF, Device (IDE1) to Device (SATA) and it's sub-device ports from Device (PRIM) to Device (PRT0) and Device (SECD) to Device (PRT1), deleting Device (IDE2) (unused) and finally renaming the USB device names, USBE to EHC1 and USE2 to EHC2
Now that we have all of the correct device names now we need to add some missing information. For the SMBus we need to add the following to Device (SBUS) as a nested sub-device:
Device (BUS0)
{
Name (_CID, "smbus")
Name (_ADR, 0x00)
Device (DVL0)
{
Name (_ADR, 0x57)
Name (_CID, "diagsvault")
Method (_DSM, 4, NotSerialized)
{
Store (Package (0x02)
{
"address",
0x57
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}
}
}
For ALC8xxHDA to work correctly we need to inject into Device (HDEF) codec information. Here is an example of what needs to be added for a board using and an ALC889 codec:
Method (_DSM, 4, NotSerialized)
{
Store (Package (0x04)
{
"layout-id",
Buffer (0x04)
{
0x79, 0x03, 0x00, 0x00 // decimal 889
},
"PinConfigurations",
Buffer (Zero)
{
}
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}
For other codecs you will need to change the layout-id to match your codec.
There is a problem with Audio break ups that can be corrected in the DSDT. The first is in Device (PIC) remove these two lines:
IRQNoFlags ()
{2}
The other fix for audio break ups is in Device (TMR) where we need to change:
Method (_CRS, 0, NotSerialized)
{
If (LGreaterEqual (OSFX, 0x03))
{
If (HPTF)
{
Return (ATT6)
}
Else
{
Return (ATT5)
}
}
Else
{
Return (ATT5)
}
}
To:
Method (_CRS, 0, NotSerialized)
{
Return (ATT6)
}
The next fix is to prevent kernel panics in AppleIntelCPUPowerManagement at boot time. This fix is easy; just change Method (_STA, 0, NotSerialized) to Return (0x0F) and to change Method (_CRS, 0, NotSerialized) to Return (ATT3).
The final edit is to allow you to use the power button to put the computer to sleep. In Device (PWRB) change Name (_HID, EisaId ("PNP0C0C")) to Name (_CID, EisaId ("PNP0C0C")).
If you don't want to bother with these edits, you can download pre-edited DSDTs from the tonymacx86.com DSDT Database.
Comments: (10)
Tuesday, July 5, 2011
Creating a Bootable “DOS” USB Stick on a Hackintosh Without Windows or MS-DOS
Posted by
MacManx86
on Tuesday, July 5, 2011
Labels: Core i3, Core i5, Core i7, DOS, guide, Hackintosh, Mac, Mac OSX, OS X, USB
Comments: (1)
Labels: Core i3, Core i5, Core i7, DOS, guide, Hackintosh, Mac, Mac OSX, OS X, USB
Comments: (1)
A lot of manufacturers provide DOS only firmware updaters. If you have a dual boot system with Windows it’s a piece of cake. But the issue is how do you do this on a Hackintosh without Windows installed? Well I have found a solution, FreeDOS, a free DOS-compatible operating system for IBM-PC compatible systems.
What you need:
- USB drive greater than 16 MB
- FreeDOS Base CD
- Blank CD or CD-RW
Installation procedure:
- Format the USB drive as a MBR/FAT volume
- Burn the FreeDOS Base CD
- Shut down OS X and disconnect all disk drives
- Boot computer while ensuring that Legacy USB support is enabled in the BIOS and USB drive is connected
- Boot from FreeDOS Base CD and follow instructions to install. It will install by default to the 1st drive it finds. In this case it’s the USB drive as it’s the only one connected.
- Shutdown and reconnect all disk drives.
- You now have a bootable “DOS” drive.
Comments: (1)
Monday, December 6, 2010
Graphics Card Support and Mac OS X
Posted by
MacManx86
on Monday, December 6, 2010
Labels: Core i3, Core i5, Core i7, Hackintosh, Mac, Mac OSX, MacManx86, OS X, OSx86, snow leopard, tonymacx86
Comments: (17)
Labels: Core i3, Core i5, Core i7, Hackintosh, Mac, Mac OSX, MacManx86, OS X, OSx86, snow leopard, tonymacx86
Comments: (17)
The issue with graphics cards in OS X is that there are what is called a "Reference Design" which is based on what ATI/AMD or NVIDIA provides to manufactures for a baseline hardware design. Then there is what cards manufacturers ship, either a "Reference Design" or their own unique spin on that hardware. They do this unique spin to set their card part from the "Reference Design" cards. That is fine in Windows, but can create problems in OS X. Apple typically uses a "Reference Design" or very close to one in their graphics cards.
So where am I going with this. Apple ships and supports a limited set of ATI/AMD "Reference Design" cards in OS X, they do not provide support for all models or any major variations from the "Reference Design". The reason being that the ATI/AMD architecture requires specific model driver support. For example in 10.6 Apple only supports some 46xx and 48xx cards, there is no OOB support for the 45xx cards. So that is why you see patches and/or modifications for Apple's drivers to support some of these other cards.
The ATI/AMD 5xxx cards are finally easier to install. This has more to do with the GraphicsEnabler technology used within Hackintoshing, Kabyl's boot loader and ATY_Init. Even then you are limited to what cards are supported Out Of BOX (OOB) by Apple. For example the following are the only desktop ATI 5xxx cards Apple supports OOB:
ATI Radeon HD 5800 Series: Device ID 0x6898
ATI Radeon HD 5800 Series: Device ID 0x6899
ATI Radeon HD 5700 Series: Device ID 0x68B8
ATI Radeon HD 5600 Series: Device ID 0x68D8
ATI Radeon HD 5500 Series: Device ID 0x68D9
NVIDIA cards use a different architecture and allows for a more generic driver than can support more models and variations. The new 4xx cards are based on a new architecture nicknamed "Fermi" and require new drivers. Check out this article on Tony's blog for our solutions for NVIDIA 4xx cards.
So until Apple ships support for more cards or allows AMD and NVIDIA to ship generic drivers, graphics card support is a challenge.
So where am I going with this. Apple ships and supports a limited set of ATI/AMD "Reference Design" cards in OS X, they do not provide support for all models or any major variations from the "Reference Design". The reason being that the ATI/AMD architecture requires specific model driver support. For example in 10.6 Apple only supports some 46xx and 48xx cards, there is no OOB support for the 45xx cards. So that is why you see patches and/or modifications for Apple's drivers to support some of these other cards.
The ATI/AMD 5xxx cards are finally easier to install. This has more to do with the GraphicsEnabler technology used within Hackintoshing, Kabyl's boot loader and ATY_Init. Even then you are limited to what cards are supported Out Of BOX (OOB) by Apple. For example the following are the only desktop ATI 5xxx cards Apple supports OOB:
ATI Radeon HD 5800 Series: Device ID 0x6898
ATI Radeon HD 5800 Series: Device ID 0x6899
ATI Radeon HD 5700 Series: Device ID 0x68B8
ATI Radeon HD 5600 Series: Device ID 0x68D8
ATI Radeon HD 5500 Series: Device ID 0x68D9
If your card's device id is not in this list, then you need to add your device ID to ATI5000Controller.kext and possibly ATIRadeonX3000.kext. This is great workaround but will be needed to be done after each and any Apple software update that touches the ATI drivers. Ouch.
So until Apple ships support for more cards or allows AMD and NVIDIA to ship generic drivers, graphics card support is a challenge.
Comments: (17)
Sunday, July 18, 2010
Typo in New Minimal DSDT Edits for Gigabyte Socket 1156 Motherboards
Posted by
MacManx86
on Sunday, July 18, 2010
Labels: Core i3, Core i5, Core i7, DSDT, Gigabyte, H55, Mac, MacManx86, OS X, OSx86, P55, snow leopard
Comments: (2)
Labels: Core i3, Core i5, Core i7, DSDT, Gigabyte, H55, Mac, MacManx86, OS X, OSx86, P55, snow leopard
Comments: (2)
I was notified that there is a typo in the USBE/EHCI and USE2/EHC2 device edits section of my article New Minimal DSDT Edits for Gigabyte Socket 1156 Motherboards.
In that section:
"AAPL,current-extra",
0x04C,
Should be:
"AAPL,current-extra",
0x044C,
So that it matches what an iMac with a Core i5/i7 reports.
I have corrected the original article and I would suggest using this correction in your edits.
Many thanks to rekursor for finding this.
In that section:
"AAPL,current-extra",
0x04C,
Should be:
"AAPL,current-extra",
0x044C,
So that it matches what an iMac with a Core i5/i7 reports.
I have corrected the original article and I would suggest using this correction in your edits.
Many thanks to rekursor for finding this.
Comments: (2)
Wednesday, June 9, 2010
New Minimal DSDT Edits for Gigabyte Socket 1156 Motherboards
Posted by
MacManx86
on Wednesday, June 9, 2010
Labels: Core i3, Core i5, Core i7, DSDT, Gigabyte, H55, Mac, MacManx86, OS X, OSx86, P55, snow leopard
Comments: (29)
Labels: Core i3, Core i5, Core i7, DSDT, Gigabyte, H55, Mac, MacManx86, OS X, OSx86, P55, snow leopard
Comments: (29)
We know that using a DSDT is the best way to create a vanilla Hackintosh build. The problem is that there have been numerous articles and forum posts on editing that are based on old versions of Gigabyte BIOS extracts. Many of them are retreads of the same information and are more complex than necessary. Since I am now doing the DSDT edits for the tonymacx86.com DSDT Database, I wanted to figure out what was the least amount of edits I can do, while trying to make the edits match a real Mac as much as possible.
Gigabyte's latest release cycle of P55/H55 BIOS updates in May 2010 includes a few Hackintosh-friendly enhancements. They've included the CMOS reset fix and some other changes. However, these modifications are not complete and some edits still need to be made. The way to tell if you have the Hackintosh-friendly BIOS is to search the decompiled DSDT for AMAC. If you find "Name (AMAC, 0x00)" then you have the new BIOS.
The following is a guide for editing the extracted DSDT from the new Hackintosh-friendly BIOSes only. Some of these edits can be used with the older BIOS extracts, but has not been tested or is the complete list for those versions.
First here are some cosmetic changes that will make your system look more Mac like. Gigabyte uses generic device names in it's DSDT instead of descriptive names like Apple does. One example is Gigabyte use PX40 instead of LPCB for the LPC Interface Bridge. So the first thing we can do is to search for PX40 and do a global replace with LPCB. The next is to rename the System Management Bus (SMBus) device PX43 to SBUS. Next would be to rename Device (IDE1) to Device (SATA) and then within it rename Device (PRIM) to Device (PRT0) and Device (SECD) to Device (PRT1). The final cosmetic edits deal with USB. You can rename the devices to more appropriate ones as shown within this table:
Rename USB0 to UHC1
Rename USB1 to UHC2
Rename USB2 to UHC3
Rename USB3 to UHC4
Rename USB4 to UHC5
Rename USB5 to UHC6
Rename USB6 to UHC7
Rename USBE to EHC1
Rename USE2 to EHC2
There are three required edits needed regardless of BIOS version. They are to include Method DTGP, add Return Package at the end of _WAK and fix the "Method local variable is not initialized (Local0)" error.
Now let’s review the new edits. One of the common edits has been to add a Device (BUS0). The problem I found is that sometimes it was added to the incorrect place or was coded incorrectly. The proper place for this device is as a sub-device within Device (SBUS). Here is what the correct device code should look like:
Device (BUS0)
{
Name (_CID, "smbus")
Name (_ADR, 0x00)
Device (DVL0)
{
Name (_ADR, 0x57)
Name (_CID, "diagsvault")
Method (_DSM, 4, NotSerialized)
{
Store (Package (0x02)
{
"address",
0x57
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}
}
}
The next edit is for audio. Gigabyte uses a device name of AZAL, so we need to do a global replace with HDEF. Then we need to add codec information to Device (HDEF). Here I found that a lot of edits were injecting a lot of information that wasn't need. All that is required is to inject layout-id and PinConfigurations. Here is an example of what needs to be added for a board using and an ALC889 codec and using our Legacy889HDA.kext:
Method (_DSM, 4, NotSerialized)
{
Store (Package (0x04)
{
"layout-id",
Buffer (0x04)
{
0x79, 0x03, 0x00, 0x00 // decimal 889
},
"PinConfigurations",
Buffer (Zero)
{
}
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}
There is a problem with Audio break ups in 10.6.3 that can be corrected in the DSDT. The first is in Device (PIC) remove these two lines:
IRQNoFlags ()
{2}
The other fix for audio break ups is in Device (TMR) where Gigabyte has not yet added Hackintosh logic. We need to change:
Method (_CRS, 0, NotSerialized)
{
If (LGreaterEqual (OSFX, 0x03))
{
If (HPTF)
{
Return (ATT6)
}
Else
{
Return (ATT5)
}
}
Else
{
Return (ATT5)
}
}
To:
Method (_CRS, 0, NotSerialized)
{
Return (ATT6)
}
Gigabyte has not yet added Hackintosh logic to Device (HPET), so you will get kernel panics in AppleIntelCPUPowerManagement. The fix is easy; just change Method (_STA, 0, NotSerialized) to Return (0x0F) and to change Method (_CRS, 0, NotSerialized) to Return (ATT3).
tonymacx86.com forum user Lnx2Mac brought this next edit to my attention. This edit allows use of the power button to put the computer to sleep. In Device (PWRB) change Name (_HID, EisaId ("PNP0C0C")) to Name (_CID, EisaId ("PNP0C0C")).
It seems that if the USB devices have no device-id or it's incorrect you won't have any USB 1.1 devices. So you have to add the following code to each USBx/UHCx device to enable it in Snow Leopard.
Method (_DSM, 4, NotSerialized)
{
Store (Package (0x02)
{
"device-id",
Buffer (0x04)
{
0x34, 0x3A, 0x00, 0x00
}
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}
I have found that the same device-id can be used for each USBx/UHCx device. They don't have to be unique, just valid.
Likewise for the USBE/EHC1 and USE2/EHC2 devices we need to add the following code:
Method (_DSM, 4, NotSerialized)
{
Store (Package (0x08)
{
"AAPL,current-available",
0x05DC,
"AAPL,current-extra",
0x044C,
"AAPL,current-in-sleep",
0x09C4,
"device-id",
Buffer (0x04)
{
0x3A, 0x3A, 0x00, 0x00
}
}, Local0)
DTGP (Arg0, Arg1, Arg2, Arg3, RefOf (Local0))
Return (Local0)
}
The "AAPL, ..." insertion is for compatibility and for native USB power savings. Again, it is safe to use duplicate device-ids.
If you have a Gigabyte P55/H55 motherboard whose current DSDT is not in the tonymacx86.com DSDT Database, you can submit it on the tonymacx86.com forum by clicking here. The instructions are at the top of the topic. After we receive it and verify that it is vanilla, we will do the edits and upload it to the database.
Comments: (29)
Sunday, June 6, 2010
JMicron SATA Support
Posted by
MacMan
on Sunday, June 6, 2010
Labels: Core i3, Core i5, Core i7, H55, Mac, MacManx86, OS X, OSx86, P55, snow leopard
Comments: (6)
Labels: Core i3, Core i5, Core i7, H55, Mac, MacManx86, OS X, OSx86, P55, snow leopard
Comments: (6)
Are you tired of seeing this in System Profiler?
I was, so that’s why I created JMicron36xeSATA.kext and JMicron36xSATA.kext.
Now in System Profiler you will see this if you use JMicron36xeSATA.kext:
Or this if you use JMicron36xSATA.kext:
So what’s the difference other than the name display? When using JMicron36xSATA.kext all drives connected to the JMicron controllers will be displayed on the desktop as internal disks, just like the Intel connected drives. But if you use the JMicron36xeSATA.kext, then they show up on the desktop as external (orange) disks and can be hot swapped.
They are available in iBoot 2.0 and in MultiBeast and are installed in /Extra/Extensions.
For discussions on this and other topics, register today at tonymacx86.com!
Comments: (6)








