Project OS X Forums > [help] How To Enable Dual Core In Dsdt?

abogar

Mar 29 2010, 06:52 PM

Hello,

I have dual core HP nx7300 notebook. I managed to install Snow Leopard without having access to a Mac by creating an usb install stick in Linux, patched DSDT for LCD screen display, USB/HPET irq-s and native SATA on ICH7 (thanks to this forum).

I have to boot with cpus=1 flag, if I boot with 2 cpus, I get a kernel crash (cpus not responding to irqs) as soon as I move the usb mouse during boot.

Can someone more experienced on this topic give me some hints, what to check? How would you approach this problem.

My dsdt, com.apple.Boot.plist and smbios.plist are attached.

Thanks is advance!

Slice

Apr 5 2010, 06:51 AM

This is a problem of bootloader. Try another one.

Gringo Vermelho

Apr 7 2010, 04:17 PM

Try Chameleon 2.0 RC5:
http://www.projectosx.com/forum/index.php?showtopic=1337

topic 1337 lol

yehia2amer

Apr 8 2010, 04:08 AM

QUOTE (abogar @ Mar 29 2010, 08:52 PM)

Hello,

I have dual core HP nx7300 notebook. I managed to install Snow Leopard without having access to a Mac by creating an usb install stick in Linux, patched DSDT for LCD screen display, USB/HPET irq-s and native SATA on ICH7 (thanks to this forum).

I have to boot with cpus=1 flag, if I boot with 2 cpus, I get a kernel crash (cpus not responding to irqs) as soon as I move the usb mouse during boot.

Can someone more experienced on this topic give me some hints, what to check? How would you approach this problem.

My dsdt, com.apple.Boot.plist and smbios.plist are attached.

Thanks is advance!

the common DSDT fix for this problem is to get rid of the cpu Alias.

See those Alias (P00x, CPUx) etc, lines? Get rid of them.

CODE

Scope (_PR)
{
Processor (P001, 0x01, 0x00000810, 0x06) {}
Alias (P001, CPU0)
}

Scope (_PR)
{
Processor (P002, 0x02, 0x00000810, 0x06) {}
Alias (P002, CPU1)
}

Scope (_PR)
{
Processor (P003, 0x03, 0x00000810, 0x06) {}
Alias (P003, CPU2)
}

Scope (_PR)
{
Processor (P004, 0x04, 0x00000810, 0x06) {}
Alias (P004, CPU3)
}

so it will be

CODE

Scope (_PR)
{
Processor (P001, 0x01, 0x00000810, 0x06) {}
}

Scope (_PR)
{
Processor (P002, 0x02, 0x00000810, 0x06) {}
}

Scope (_PR)
{
Processor (P003, 0x03, 0x00000810, 0x06) {}
}

Scope (_PR)
{
Processor (P004, 0x04, 0x00000810, 0x06) {}
}

abogar

Apr 8 2010, 06:41 PM

QUOTE (yehia2amer @ Apr 8 2010, 05:08 AM)

the common DSDT fix for this problem is to get rid of the cpu Alias.

This must be the uncommon case. Look at my current DSDT it it helps:
HP nx7400 DSDT

abogar

Apr 12 2010, 09:24 PM

Thanks for the comments.

I tried some alternative boot loaders.

With aserebln 1.1.9 I am getting the same "cpu(s) failing to respond to interrupts" kernel crash.

With PC-EFI 10.6 I can sometimes run for some time with 2 cpus. Sometimes it just hangs. I need to use VoodooTSCSync kext in order this to work. Sometimes I am getting a "Failed to start CPU 01, rebooting, CPU halted" message during boot. It is printed in place of the IOAPIC line.

So I think theoretically it is possible to run with 2 cpus on this machine, but some DSDT polishments needs to be done to make it stable.

Please give some hints what to look at (compared to Genuine Mac DSDT), which might be interesting.

I am emulating MacPro3,1 in SMBIOS. Do you think I should try MacBook(Pro) for example?

Thanks,
A.

Gringo Vermelho

Apr 12 2010, 09:51 PM

This probably won't fix your CPU issue but yes, you should definitely use a laptop model identifier if your PC is a laptop.

apocolipse269

Apr 13 2010, 07:33 AM

Please give me a picture of your panic. It may be a problem that i've experienced and fixed with bad LAPIC on some proc's. KP will tell all, but i needz it! IF so i can give you a modified kernel that will work perfectly (its 99.9% vanilla, not perfect but it works). Otherwise, have you tried VoodooTSCSync? That may also help it.

18seven

Apr 22 2010, 06:30 AM

QUOTE (apocolipse269 @ Apr 13 2010, 02:33 AM)

IF so i can give you a modified kernel that will work perfectly (its 99.9% vanilla, not perfect but it works).

I don't have this issue but I would like to see/have the patch just for my info's.

abogar

Apr 22 2010, 09:57 AM

I will post a screenshot of the halt this evening.

abogar

Apr 22 2010, 09:50 PM

The crash screen is here

.

The "Failed to start CPU" message is printed by the kernel.
This is the place where VoodooTSCSync should load and the IOAPIC message should be printed.

Any ideas?

Slice

Apr 25 2010, 06:47 AM

Clean up your system from SleepEnabler (good bootloader do this), CPUInjector (same), NullCPUPM (why you are needed it?).

Gringo Vermelho

Apr 26 2010, 02:44 AM

Yes get rid of all those, if you use Chameleon 2.0 RC5 you shouldn't need a CPUInjector.

XPbIM3

Sep 19 2010, 12:20 PM

Same sh*t here with my HP.
Got random "Failed to start CPU01, rebooting"
Sometimes i am lucky enough to boot up with two cores, sometimes it hangs in beginning of boot, sometimes in middle with KP.
looks like all old HP's suffer from this.

any ideas how to fix?

Slice

Sep 20 2010, 07:07 AM

QUOTE (XPbIM3 @ Sep 19 2010, 04:20 PM)

Same sh*t here with my HP.
Got random "Failed to start CPU01, rebooting"
Sometimes i am lucky enough to boot up with two cores, sometimes it hangs in beginning of boot, sometimes in middle with KP.
looks like all old HP's suffer from this.

any ideas how to fix?

Looks like TSC sync problem.

XPbIM3

Sep 20 2010, 11:20 AM

QUOTE (Slice @ Sep 20 2010, 01:07 PM)

Looks like TSC sync problem.

VoodooTSCsync.kext make no change. The problem appears much earlier, maybe when AppleACPIPlatform is loaded.

XPbIM3

Sep 20 2010, 04:56 PM

ok, i think i found some "traces" of this problem.

when booting with "DropSSDT=Yes" it's high probability to meet "failed to start CPU1" error
if boot with SSDT, there is KP (CPU failed to respond to interrupts) in the middle of the boot ... just after "CPUPowerManagment: initialization complete"(!!!).
So it looks like one of two cores falls to deep C-state or asynchronous state switching r something....

(NullCPUPM leads to "failed to start CPU", so I think I will dig into my SSDT and look for errors/issues _INI methods and ways to simplify overall code.)

Edit:
partially confirmed: I was lucky enough to boot with NullCpuPm, overall very slooooow performance, but 2 cores and no KP.

WRONG.
Still have random chance to KP "CPU failed to respond to interrupts". With or without DSDT/NullCPUPM. What's wrong with these HP's?

yeehaa

Sep 20 2010, 07:12 PM

are you guys using the latest bios and iasl? because HP dsdts are usually much cleaner without any device masking and stuff

XPbIM3

Sep 20 2010, 07:19 PM

QUOTE (yeehaa @ Sep 21 2010, 01:12 AM)

are you guys using the latest bios and iasl? because HP dsdts are usually much cleaner without any device masking and stuff

The lastes Bios and one of the lastest iasl (the one from "iaslme" app, from 28/05/2010)

XPbIM3

Sep 22 2010, 08:16 PM

ok, next suggestion: "maybe there is the two problems? the problem with initializing CPU1 itself and the second problem with interrupts?"
lets see if i can approve this (or disprove).
EDIT:

Still no progress.
Writing to share my observation with you.
1)First of all a fresh install of 10.5.6 Leopard was surprising. System reboots and halts properly without any OpenHaltRestart or OSXRestart. oO
2)"Unable to initialize CPU1, rebooting" appears only when Laptop previously was hardly switched off by holding power-off button for a 5 sec. Leopard auto-reboots after this error so i can do with this issue for a while.
3)Leopard is much more survivable. A much greater success rate to boot with dual-cores enabled. It hangs on CPU-heavy applications after 3-5 minutes after boot. (no chance to complete Cinebench test. Xbench hits 170 points if lucky.)
4)The way it hangs: graphics and mouse freeze, sometimes you can alt-tab between windows. When power button pressed mouse moves little bit and freezes again.

found this old thread:
http://www.insanelymac.com/forum/index.php?showtopic=137237
cannot confirm yet.

Any help appreciated.... :despair:

JuiCe

Sep 24 2010, 07:57 PM

Hi there. We are in the same boat.

Currently i'm running 10.6.4 with AnV Loader and FakeSMC.kext in /E/E only (external mouse and kb) with cpus=1

I have tried numerous DSDT modifications but i still get the pmp_flush_tlbs() error. Sometimes I was able to run both cores for about 12hrs just to reboot and get the KP on start.

I've even tried a brute force approach (DSDT from macbook2,1 and a modified apic.aml to fix lapic cpu references) just to kp on boot with "probe HPET failed"

A question to ppl with compaq/hp laptops with i945 - do You have interrupts (0,1) assigned to cpus in IOREG under AppleCPU?

I have ioregs from MB2,1 and there are no irq assigned to CPUs.

XPbIM3

Sep 24 2010, 09:24 PM

QUOTE (JuiCe @ Sep 25 2010, 01:57 AM)

Hi there. We are in the same boat.

Currently i'm running 10.6.4 with AnV Loader and FakeSMC.kext in /E/E only (external mouse and kb) with cpus=1

I have tried numerous DSDT modifications but i still get the pmp_flush_tlbs() error. Sometimes I was able to run both cores for about 12hrs just to reboot and get the KP on start.

I've even tried a brute force approach (DSDT from macbook2,1 and a modified apic.aml to fix lapic cpu references) just to kp on boot with "probe HPET failed"

A question to ppl with compaq/hp laptops with i945 - do You have interrupts (0,1) assigned to cpus in IOREG under AppleCPU?

I have ioregs from MB2,1 and there are no irq assigned to CPUs.

What Laptop exactly are you running on?

stay in touch, i am in the middle of something....

JuiCe

Sep 25 2010, 10:58 AM

QUOTE (XPbIM3 @ Sep 24 2010, 11:24 PM)

What Laptop exactly are you running on?

stay in touch, i am in the middle of something....

I have an nx7300 with 4 GB Ram and T7200

Yesterday I have been reading ACPI specification PDF and i'm beginning that our problem has to do something with c-states declaration (inspecific C3) and flushing (Section 8.2 Flushing Caches).

What really bugs me is that upon using acpidump in linux i get two FACP files - with different C-state, c2 dalay and c3 dalay values. Could it be that one of FACPs gets randomly loaded upon boot and then we get KP if the other c2, c3 delay is being read by the system.

I've attached my acpidump

Other issue - is it normal that a dumped RSDT (or XSDT) has different ACPI table addresses the one i dump using DSDTSE running on snow? Does it get randomly generated? Or maybe it gets loaded and corrected by the bootloader in such a way that our ssdts are not accessible by the system?

Click to view attachment

XPbIM3

Sep 25 2010, 12:42 PM

Yesterday i had achieved a great success, 2 cores under Leopard, Snow leopard, on battery or on AC, no matter, it works!
I don't know why this fix works, didn't analyzed yet, posting to share with you as soon as possible.
And don't even ask how I've managed to find this solution.

the fixed parts of code:
Added Darwin OSI case, looks like it make sleep possible, still need to confirm:

CODE

                    If (_OSI ("Darwin"))
                    {
                        Store (0x04, C014)
                    }

Added this to Sb._INI method:

CODE

    Method (_SB._INI, 0, NotSerialized)
    {
        \C008 ()
        C01B ()

//code added my XPbIM3 below:
    _WAK (0x03)
        Notify (C1B4, 0x81)
        Notify (C1B3, 0x81)
        Notify (C1B2, 0x81)
    }

Removed these _PRS Recourses of LNKA, LNKB..commented the whole section so code looks like this:

CODE

/*            Name (C10D._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            })
            Name (C10E._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            })
            Name (C10F._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            })
            Name (C110._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            })
            Name (C123._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            })
            Name (C124._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            })
            Name (C125._PRS, ResourceTemplate ()
            {
                IRQ (Level, ActiveLow, Shared, )
                    {11}
            }) */

Added come IRQs to HPET (just to be sure HPET does not interfere with EHCI, I am using external USB HDD to boot, so this is essential), dont forget to remove them from RTC and TMR

CODE

                    Name (C01F, ResourceTemplate ()
                    {
                        IRQNoFlags () //added by XPbIM3
                            {0}
                        IRQNoFlags ()
                            {8}
                        Memory32Fixed (ReadOnly,
                            0xFED00000,         // Address Base
                            0x00000400,         // Address Length
                            _Y0E)
                    })

Changed mouse and keyboard IRQs to be with flags, looks like it change nothing, but 99% Laptops have this.

CODE

IRQ (Edge, ActiveHigh, Exclusive, )
{1}

and

CODE

IRQ (Edge, ActiveHigh, Exclusive, )
{12}

and added the DTGP method and DTGP injection for GMA950. Thats all. Sleep working, 2 cores working. have some cooling issues and still sometimes "Failed to initialize CPU1", but it looks like this is temp/cooling issue too.
See DSDT's and IOregs in attach.
See screenshot with Cinebench working

Click to view attachment
Click to view attachment
Click to view attachment

XPbIM3

Sep 25 2010, 12:46 PM

Aand one little thing:
My machine is TC4400, T7200, 2gb (of sloooww) RAM.

CODE

Results    177.57
    System Info
        Xbench Version        1.3
        System Version        10.6.3 (10D575)
        Physical RAM        2048 MB
        Model        MacBook5,1
        Drive Type        Generic External
    CPU Test    141.05
        GCD Loop    235.75    12.43 Mops/sec
        Floating Point Basic    112.20    2.67 Gflop/sec
        vecLib FFT    93.06    3.07 Gflop/sec
        Floating Point Library    224.26    39.05 Mops/sec
    Thread Test    161.01
        Computation    179.03    3.63 Mops/sec, 4 threads
        Lock Contention    146.29    6.29 Mlocks/sec, 4 threads
    Memory Test    129.25
        System    141.70
            Allocate    221.86    814.74 Kalloc/sec
            Fill    116.28    5653.76 MB/sec
            Copy    124.01    2561.31 MB/sec
        Stream    118.82
            Copy    111.10    2294.71 MB/sec
            Scale    112.65    2327.26 MB/sec
            Add    126.64    2697.70 MB/sec
            Triad    126.73    2711.07 MB/sec
    Quartz Graphics Test    173.65
        Line    137.29    9.14 Klines/sec [50% alpha]
        Rectangle    170.01    50.76 Krects/sec [50% alpha]
        Circle    146.18    11.92 Kcircles/sec [50% alpha]
        Bezier    154.22    3.89 Kbeziers/sec [50% alpha]
        Text    434.28    27.17 Kchars/sec
    OpenGL Graphics Test    364.71
        Spinning Squares    364.71    462.65 frames/sec
    User Interface Test    235.23
        Elements    235.23    1.08 Krefresh/sec

JuiCe

Sep 25 2010, 09:03 PM

If You have commented out the _PRS Recourses of LNKA... - didn't you make Your USB ports useless at the same time?

XPbIM3

Sep 25 2010, 09:54 PM

QUOTE (JuiCe @ Sep 26 2010, 03:03 AM)

If You have commented out the _PRS Recourses of LNKA... - didn't you make Your USB ports useless at the same time?

Nope, working flawless.It may cause IRQ shifts but no inability to use.
Just give it a try.
Or post your DSDT, i will try to correct it the same way.

JuiCe

Sep 25 2010, 10:57 PM

QUOTE (XPbIM3 @ Sep 25 2010, 11:54 PM)

Nope, working flawless.It may cause IRQ shifts but no inability to use.
Just give it a try.
Or post your DSDT, i will try to correct it the same way.

My DSDT is in my last attachment (out.zip)

I will try some of Your fixes and let You know.

XPbIM3

Sep 25 2010, 11:27 PM

QUOTE (JuiCe @ Sep 26 2010, 04:57 AM)

My DSDT is in my last attachment (out.zip)

I will try some of Your fixes and let You know.

Try this one. Click to view attachment

JuiCe

Sep 26 2010, 09:08 PM

QUOTE (XPbIM3 @ Sep 26 2010, 01:27 AM)

Try this one. Click to view attachment

Thanks for Your efforts. After using this DSDT i get CPU failet to initialize an so on.... Hpets not configured.

Back to square one

Why do I have interrupts under AppleACPICPU? You do not have any...

Could You please send me a complete dump of tables from Your system (FACP, RSDT and so on....)

Attached is my IOREGClick to view attachment

XPbIM3

Sep 26 2010, 09:16 PM

QUOTE (JuiCe @ Sep 27 2010, 03:08 AM)

Thanks for Your efforts. After using this DSDT i get CPU failet to initialize an so on.... Hpets not configured.

Back to square one

thats "fine", just reboot again.

This "cpu1 failed to initialize" appears only after hard "5 second hold" poweroff. (Or when CPU too hot)

JuiCe

Sep 26 2010, 09:33 PM

QUOTE (XPbIM3 @ Sep 26 2010, 11:16 PM)

thats "fine", just reboot again.

This "cpu1 failed to initialize" appears only after hard "5 second hold" poweroff. (Or when CPU too hot)

Booted again numerous times - still same error.

XPbIM3

Sep 26 2010, 10:15 PM

QUOTE (JuiCe @ Sep 27 2010, 03:33 AM)

Booted again numerous times - still same error.

not good.
Just tried to install sound codec patching appleHDA on SL, got that bug returned. Which means the fix is closely rely on current system configuration/loaded kexts/active devices.
Need to be analyzed.

EDIT:
Posting some observations. Found some differences between "1cpu", "before" and "after" ioregs.

there is a four devices matched as "IOPCI2PCIBridge", take a look on their InterruptSpecifiers

when on 1core, stable:
C111@1c, Interrupts array 0, Data = <10 00 00 00 07 00 00 00>
on 2 cores, stable:
C111@1c, INterrupts array 0, Data = <10 00 00 00 07 00 00 00>
C111@1c, INterrupts array 1, Data = <00 00 00 00 00 00 01 00>
on 2 cores, UNstable:
C111@1c, INterrupts array 0, Data = <10 00 00 00 07 00 00 00>
C111@1c, INterrupts array 1, Data = <01 00 00 00 00 00 01 00>

devices C111@1C,1, C117@1C,3 behave the same way.
I think there is the key of our problem.

There is _PRT section - PCI routings table, maybe we have something to do with that. I'll continue working this out.

XPbIM3

Oct 3 2010, 06:41 PM

QUOTE (JuiCe @ Sep 27 2010, 03:33 AM)

...

JuiCe, another DSDT to try.
_{4.10.2010, 4:00, minor changes, try it, HPET irqs, OSI, DTGP and sb.INI fixes inside, it may not boot from first time, keep trying at least 5 times so if it boots then 2 cores will work, still not sure why this boot-bug appears}
Click to view attachment

JuiCe

Oct 13 2010, 10:17 PM

QUOTE (XPbIM3 @ Oct 3 2010, 08:41 PM)

JuiCe, another DSDT to try.
_{4.10.2010, 4:00, minor changes, try it, HPET irqs, OSI, DTGP and sb.INI fixes inside, it may not boot from first time, keep trying at least 5 times so if it boots then 2 cores will work, still not sure why this boot-bug appears}
Click to view attachment

Thx. Will try it during the weekend.

dawntodon

Nov 1 2010, 12:03 AM

hi. I stumbled upon this post while looking for a way to enable both cores on my HP Compaq NC4400. I'm fairly new to DSDT patching and was wondering if the earlier DSDT.dsl would be worth trying for my setup. My DSDTSE extracted DSDT file can be found here. Thanks! This is a wonderful thread.

XPbIM3

Nov 2 2010, 09:29 PM

QUOTE (dawntodon @ Nov 1 2010, 06:03 AM)

hi. I stumbled upon this post while looking for a way to enable both cores on my HP Compaq NC4400. I'm fairly new to DSDT patching and was wondering if the earlier DSDT.dsl would be worth trying for my setup. My DSDTSE extracted DSDT file can be found here. Thanks! This is a wonderful thread.

Here, try this one.
Added GMA fix also.
Notice that coldboot will be okay, but boot after long run on heated machine will fail.

Click to view attachment

dawntodon

Nov 3 2010, 01:53 PM

QUOTE (XPbIM3 @ Nov 2 2010, 10:29 PM)

Here, try this one.
Added GMA fix also.
Notice that coldboot will be okay, but boot after long run on heated machine will fail.

Click to view attachment

Oh wow. That did it. Thank you very much.

XPbIM3

Nov 4 2010, 10:28 AM

QUOTE (dawntodon @ Nov 3 2010, 07:53 PM)

Oh wow. That did it. Thank you very much.

one more thing, you need a HPET Irq's patch to add, or you will suffer from USB 2.0 problems.

dawntodon

Nov 6 2010, 08:50 AM

QUOTE (XPbIM3 @ Nov 4 2010, 10:28 AM)

one more thing, you need a HPET Irq's patch to add, or you will suffer from USB 2.0 problems.

Oh wow, this is SO true. I've been finding that my internal bluetooth, which is detected as a usb device isn't detected anymore. Also, I've noticed that when I plug in my usb wireless-n adapter while booting, I either get a kernel panic or once the os loads, the wireless-n adapter isn't detected. I'm sorry about being such a noob in the dsdt scene but how would I go about adding the appropriate patches? Is this something that can be done with a flag in my com.apple.boot.plist in /Extra? Thanks again!

dawntodon

Nov 7 2010, 09:17 AM

so I got another dsdt.aml that seems to fix my usb 2.0 issue.

EDIT: so I futzed around with the two dsdts using Textwrangler's compare function while consulting this thread about IRQs and dual booting and came up with this one.

Ultimately, I was having issues with this chunk:

CODE

                Device (C216)
                {
                    Name (_HID, EisaId ("PNP0100"))
                    Name (_CRS, ResourceTemplate ()
                    {
                        IO (Decode16,
                            0x0040,             // Range Minimum
                            0x0040,             // Range Maximum
                            0x01,               // Alignment
                            0x04,               // Length
                            )
                        IRQNoFlags ()
                            {0}

namely with the value in the last line as well as for Device {C218} which had for its IRQNoFlags () value {8}.

I switched the value to 9 for C216 and the value to 10 for C218 while applying values 0 and 8 to the HPET chunk, which was needed for USB 2.0 to work. It's worked so far one boot and hopefully it will for boots to come. In the event that it doesn't, I'll report back and hopefully I can enlist your help in solving this issue that I've just spent a whole day on.

XPbIM3

Nov 8 2010, 05:55 PM

QUOTE (dawntodon @ Nov 7 2010, 03:17 PM)

.....

Click to view attachment

dawntodon

Nov 11 2010, 07:15 AM

QUOTE (XPbIM3 @ Nov 8 2010, 05:55 PM)

Click to view attachment

Thanks for the response and the dsdt file. I tried it out and for some boots it works, for others I get a kernel panic talking about interrupt handlers or the cpu halted error. I can give a more precise report, but I don't know how to log verbose boots. Again, I appreciate your time and smarts.

XPbIM3

Nov 12 2010, 04:11 PM

QUOTE (dawntodon @ Nov 11 2010, 01:15 PM)

Thanks for the response and the dsdt file. I tried it out and for some boots it works, for others I get a kernel panic talking about interrupt handlers or the cpu halted error. I can give a more precise report, but I don't know how to log verbose boots. Again, I appreciate your time and smarts.

That's what i'am talking about.

Cold boot will be okay.
Boot on heated machine have random(~50%) probability to fail with these kind of errors.

I Have found a way how to enable second core. Next step is to get rid of this "cold boot" bug. Have no free time for this so far. =(

XPbIM3

Nov 17 2010, 02:01 PM

10.6.5 contains new version of AppleACPIPlatform.
Sleep and startup behavior changed.

dawntodon

Nov 18 2010, 08:10 AM

QUOTE (XPbIM3 @ Nov 17 2010, 03:01 PM)

10.6.5 contains new version of AppleACPIPlatform.
Sleep and startup behavior changed.

I installed it too but I haven't noticed much of a difference. I still get intermittent panics while booting or I get through.

ricom2ger

Feb 7 2011, 03:46 PM

nx7400 here, meklort's netbookinstaller bootloader solved the dual core problem. Tested XPbIM3 DSDT out, but it KP too often at startup.

timbob

Feb 9 2011, 05:27 PM

QUOTE (ricom2ger @ Feb 7 2011, 03:46 PM)

nx7400 here, meklort's netbookinstaller bootloader solved the dual core problem. Tested XPbIM3 DSDT out, but it KP too often at startup.

Hi - you have a nx7400 running cpus=2 & stable? I thought netbookinstaller was just for dells mini's? Can you elaborate a bit more please? how did the process go & If the other hardware running OK?

been running my nx7400 for ages on a single core - didn't think there was a solution on the horizon...

Thanks!

ricom2ger

Feb 9 2011, 06:37 PM

QUOTE (timbob @ Feb 9 2011, 05:27 PM)

Hi - you have a nx7400 running cpus=2 & stable? I thought netbookinstaller was just for dells mini's? Can you elaborate a bit more please? how did the process go & If the other hardware running OK?

been running my nx7400 for ages on a single core - didn't think there was a solution on the horizon...

Thanks!

I took the boot File, and the /Extra/modules/ Folder from NBI (0.8.5). I don't specify the core count with cpu=x Boot Option.

timbob

Feb 10 2011, 11:32 AM

QUOTE (ricom2ger @ Feb 9 2011, 06:37 PM)

I took the boot File, and the /Extra/modules/ Folder from NBI (0.8.5). I don't specify the core count with cpu=x Boot Option.

Thanks for this post - think I'll be trying this out soon - never used NBI before, I had a quick read up on it - can I just check I've got the right idea:

1. create usb restore disk from retail disk 10.6
2. run NetbookBootMaker 0.8.4 RC1 to modify boot disk
3. install etc. - NBI patches are automatically applied during install - I guess at this point the CPU issue is patched?
4. run 0.8.5pre?

So:

- how does NetbookInstaller 0.8.5pre fit into this process? - there's a modules folder in the zip, do I copy that onto the book disk prior to install? Or am I just doing as above, running the app afterwards?

Help much appreciated...