Lenovo IdeaPad U310

[Jos1984]

Guess what... i do a stupid thing and i can go down to 9x (not 8x), i think that have something to do with bus speed 99 Mhz, not 100. And maybe when i put to 8x, it goes down 800 Mhz, and it ends in a glorious Kernel Panic.

I changed the line in the script to look like this...

i3-3217U,17,900,1800,0,2,4

Another people says that U processors are giving problems with the script, not only i3.

 

[RehabMan]

Guess what... i do a stupid thing and i can go down to 9x (not 8x), i think that have something to do with bus speed 99 Mhz, not 100. And maybe when i put to 8x, it goes down 800 Mhz, and it ends in a glorious Kernel Panic.

I changed the line in the script to look like this...

i3-3217U,17,900,1800,0,2,4

Another people says that U processors are giving problems with the script, not only i3.

How did you achieve the output in post #271? What SSDT was in use then? I'd be interested in seeing the ioreg in that scenario.

I think that Ivy PM, especially on Ivy CPUs without turbo (i3), and especially those 'U' CPUs is not quite understood.

 

[Jos1984]

How did you achieve the output in post #271? What SSDT was in use then? I'd be interested in seeing the ioreg in that scenario.

The dump was made when I'm using the SSDT inside x12.SSDT.zip from post #272. For that output i used the instructions you give to me in the post #1 from AppleIntelCPUPowerManagementInfo.kext (MSRDumper successor) .

Truly IOReg don't export the information and corrupt the final file. I used IOJones, that can open her own files.

 

[RehabMan]

The dump was made when I'm using the SSDT inside x12.SSDT.zip from post #272. For that output i used the instructions you give to me in the post #1 from AppleIntelCPUPowerManagementInfo.kext (MSRDumper successor) .

Thanks for the clarification, on the ioreg dump (which I can't really use), but I was more interested in what SSDT was in use when you had the x8 multiplier from the AICPMI kext.

Truly IOReg don't export the information and corrupt the final file. I used IOJones, that can open her own files.

I prefer IORegistryExplorer over IOJones. Searching is broken in IOJones.

If you use the correct version of IORegistryExplorer, it works.

Follow the instructions here for gathering ioreg dump: http://www.tonymacx86.com/audio/58368-guide-how-make-copy-ioreg.html

 

[Jos1984]

Thanks for the clarification, on the ioreg dump (which I can't really use), but I was more interested in what SSDT was in use when you had the x8 multiplier from the AICPMI kext.



I prefer IORegistryExplorer over IOJones. Searching is broken in IOJones.

If you use the correct version of IORegistryExplorer, it works.

Follow the instructions here for gathering ioreg dump: http://www.tonymacx86.com/audio/58368-guide-how-make-copy-ioreg.html

Finally make it, i have that dump!!

Well, ill try to stay awake a little more, but here are the 1:35 AM. Thanks for everything, i discover that if i used the values of a i5-3317U, without the Turbo states, i can make the 8x, but i lose 0,1 GHz due to the base speed of i5 800-1700 vs 800-1800.

I have to say that in some ways, this is fun.

Edit: The SSDT can be found in post #272, as x12.SSDT.zip.

 

[RehabMan]

Finally make it, i have that dump!!

Well, ill try to stay awake a little more, but here are the 1:35 AM. Thanks for everything, i discover that if i used the values of a i5-3317U, without the Turbo states, i can make the 8x, but i lose 0,1 GHz due to the base speed of i5 800-1700 vs 800-1800.

I have to say that in some ways, this is fun.

Edit: The SSDT can be found in post #272, as x12.SSDT.zip.

So with this SSDT (I can extract the SSDT in use from the ioreg), I see:

    Store ("ssdtPRGen version: 6.6", Debug)
    Store ("baseFrequency    : 1200", Debug)
    Store ("frequency        : 1800", Debug)
    Store ("busFrequency     : 100", Debug)
    Store ("logicalCPUs      : 4", Debug)
    Store ("tdp              : 17", Debug)
    Store ("packageLength    : 7", Debug)
    Store ("turboStates      : 0", Debug)
    Store ("maxTurboFrequency: 1800", Debug)
    Scope (\_PR.CPU0)
    {
        Name (APLF, 0x04)
        Name (APSN, One)
        Name (APSS, Package (0x0C)

What pstates were you getting according to AICPMI?

Ideally, it is good to get an ioreg dump at the same time as AICPMI output from system.log, so they are easily correlated. Otherwise it is easy to get confused if they are spread across multiple posts...

Edit: Other people have had difficulty with this type of SSDT(APLF=4), and i3 CPU, resulting in panic on boot.

 

[ikkoku]

Still looking at the details of ikkok's DSDT...

FYI: One of the things that his DSDT does it it injects the 6-series device-ids for EHC (USB2) devices. This means you'll get the 6-series drivers for USB2 instead of the proper 7-series drivers. Not really proper, but might explain something. It is possible that ikkoku's machine is 6-series+Sandy Bridge, not 7-series+Ivy Bridge.

actually mine is the 7 series+ivy. This is going to sound bad, but I just experimented with a lot of ivy patches and this combo worked best and as plain vanilla as I can make it. Probably not the most optimal, but I'm not a tech guy either

 

[RehabMan]

actually mine is the 7 series+ivy.

Hi ... OK..

What I was saying is that using your DSDT will make the system think you have 6-series EHC chip.

This is going to sound bad, but I just experimented with a lot of ivy patches and this combo worked best and as plain vanilla as I can make it. Probably not the most optimal, but I'm not a tech guy either

What is "this combo"?

 

[ikkoku]

What is "this combo"?

all the patches that was applied to my dsdt.

reading through many many threads and a lot of your help got me this far. Ive been using it for a few months as my primary machine and everything is working great.

I think the hardest bit for me was trying to figure out how to patch my own applehda. I hope I can remember what I did when 10.9 rolls out and we all try to upgrade.

 

[RehabMan]

all the patches that was applied to my dsdt.

reading through many many threads and a lot of your help got me this far. Ive been using it for a few months as my primary machine and everything is working great.

I was able to "decode" most of the patches that you made. Looks like at least one person has a more native/fresh approach (eg. not using foreign DSDT) working on a U410. See: http://www.tonymacx86.com/mountain-lion-laptop-support/111458-u410-kp-after-boot.html

I think the hardest bit for me was trying to figure out how to patch my own applehda. I hope I can remember what I did when 10.9 rolls out and we all try to upgrade.

It is pretty easy to take an existing patched AppleHDA and pull the right parts out to patch a new version of AppleHDA. I have such a script that does the two ProBook (4x30s, 4x40s... both IDT codecs). And I have one that does a flavor of ALC269. If you remember what binary patch you used, it is easy...