Setup & Software Discussions
[GUIDE] DSDT override eGPU error 12 fix
 

[GUIDE] DSDT override eGPU error 12 fix  

 of  7
  RSS

nu_ninja
(@nu_ninja)
Estimable Member
Joined: 1 year ago
 

For a Mac, instead of using Clover you should try loading the DSDT override with the automate-eGPU EFI. Read the top post on that thread for more info.

Mid-2012 13" Macbook Pro (MacBookPro9,2) TB1 -> RX 460/560 (AKiTiO Node/Thunder2)
+ macOS 10.14+Win10
+ Linux Mint 19.1


ReplyQuote
pasiamoon
(@pasiamoon)
Active Member
Joined: 8 months ago
 

@nu_ninja WOW thank you a lot for that tip!  Awesome, everything works now.

This post was modified 7 months ago

Macbook pro 2015 15" - Razer core X + ASUS GeForce GTX 1070 ROG Strix 8GB


ReplyQuote
(@luca_parisi)
New Member
Joined: 6 months ago
 

Hi, I'm trying using a GTX 750 Ti on a Lenovo T420 with the dock EXP GDC Beast.
The problem is: I can see the large memory only in test mode!! Test mode is very limited and I can't play lots of games. PLEASE, HELP ME!

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 
Posted by: veriestVarlet

Another potential Code 12 fix. For my uncooporative 2015 13" MBP, running "bcdedit.exe -set PCIEXPRESS FORCEDISABLE" (with the Graphics Card totally uninstalled) then installing the card fixed allocation (after a DSDT override.)

(Someone knowledgable should probably try this, discover pitfalls, explain why it works, all that.)

Hi @veriestvarlet! Could you please elaborate:  What exactly is "uncooperative" in the 2015 13"MBP? Thank you.

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
madjester
(@veriestvarlet)
Eminent Member
Joined: 3 years ago
 

Hey @OliverB, just that none of the other DSDT fixes would get the memory to show up in the right place and the device working properly. Only the forcedisable finally got the eGPU to show up and work (but after the first ding another dsdt override.) Otherwise it kept showing the error code 12 in the device manager. 

I don’t have the machine anymore but it did work with Sierra bootcamped into Windows 10 when last used.

Early 2015 13" MBP with Thunder2 RX 470 on Windows 10 and MacOS 10.1 (Formerly)

Lenovo Thinkpad X1 Tablet 3rd Gen and 2070 RTX Aorus Gaming Box (Custom)
Web development, Video Editing, 2D + 3D animation


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 
Posted by: veriestVarlet

Hey @OliverB, just that none of the other DSDT fixes would get the memory to show up in the right place and the device working properly. Only the forcedisable finally got the eGPU to show up and work (but after the first ding another dsdt override.) Otherwise it kept showing the error code 12 in the device manager. 

I don’t have the machine anymore but it did work with Sierra bootcamped into Windows 10 when last used.

Hi @veriestvarlet, This is strange. You have the Early 2015 MBP 13-inch, don't you? Because I have exactly this model and I never get Error 12. How is this possible?

(I am mostly using Bootcamp on this machine and never needed any fix like DSDT Override at all)

This post was modified 6 months ago

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
madjester
(@veriestvarlet)
Eminent Member
Joined: 3 years ago
 

I don’t have them anymore no, sold and replaced with a TB3 setup. 

If you didn’t need the fix great! It might have been the card (RX 470) or enclosure (Node) I was using. Every setup is different. Even my current Gigabyte card to Gigabyte enclosure setup needed a patch so compatibility isn’t predictable.

Early 2015 13" MBP with Thunder2 RX 470 on Windows 10 and MacOS 10.1 (Formerly)

Lenovo Thinkpad X1 Tablet 3rd Gen and 2070 RTX Aorus Gaming Box (Custom)
Web development, Video Editing, 2D + 3D animation


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 
Posted by: veriestVarlet

I don’t have them anymore no, sold and replaced with a TB3 setup. 

If you didn’t need the fix great! It might have been the card (RX 470) or enclosure (Node) I was using. Every setup is different. Even my current Gigabyte card to Gigabyte enclosure setup needed a patch so compatibility isn’t predictable.

It's unlikely that it's the GPU or the enclosure, because they behave all the same in this matter and I tried quite a large number of different ones. At this moment I am running a RX 570 which doesn't make any trouble at all.

On theory of mine is that Windows has been improved in this matter since then. @itsage, what do you think?

This post was modified 6 months ago

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
elpadron
(@elpadron)
New Member
Joined: 6 months ago
 

Hi Guys can anyone help me, i have a Acer Aspire 7750G with a i7 2670QM and 16GB RAM. eGPU is RX 580 connected with exp gdc beast v8.0, im getting the error code 12 so i decided to follow the instructions on the first page, but i am stuck at the Step iv. (using Intel method) i want to create the file named "dsdt-modified.dsl" but i get this error:

Microsoft Windows [Version 6.1.7601] Copyright (c) 2009 Microsoft Corporation. Alle Rechte vorbehalten.

C:\Windows\system32>c: & cd \dsdt

C:\dsdt>iasl -ve dsdt-modified.dsl

Intel ACPI Component Architecture
ASL+ Optimizing Compiler version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation

dsdt-modified.dsl 7977: (Local0 + 0x02)
Error 6114 - ^ Result is not used, operator
has no effect

ASL Input: dsdt-modified.dsl - 10094 lines, 366868 bytes, 3835 keywords

Compilation complete. 1 Errors, 7 Warnings, 10 Remarks, 308 Optimizations

C:\dsdt>

CAN SOMEONE PLS HELP ? 

Acer Aspire 7750G i7-2670QM 16GB RAM | eGPU Amd Raedon RX580 Sapphire Nitro+ | Exp Gdc Beast v8.0


ReplyQuote
Light_from_the_sun
(@light_from_the_sun)
New Member
Joined: 1 year ago
 

is there any way to turn testsigning mode off without stopping the egpu from working? Testsigning makes a lot of anticheat programs stop you playing games.

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
DagonX
(@dagonx)
New Member
Joined: 10 months ago
 

Hi,
Ok so I have made a modified dsdt file, and am able to successfully run an rx580 4gb(express card) on an HP EliteBook 8470p via registry override. And it scores only about 700 lower on superposition than being in a desktop. so not to terrible. Although I'm running into an issue where I cant play games using EAC and Battle eye. due to test signing being enabled. So I've tried the clover method. I've followed the steps exactly, cmd prompt said everything copied/transferred fine. 7zip didn't give any errors. but I can't for the life of me, find where to reboot via clover. can anyone please give me some guidance as to how I'm supposed to do that?

Hp elitebook 8470p: 
 i7-3720m
8 gigs ram
windows 10 pro
Beast 8.5c express gdc adapter
sapphire nitro+ rx 580

This post was modified 6 months ago

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
mox
 mox
(@mox)
New Member
Joined: 5 months ago
 

Hey there DagonX- I am still stuck with compiling the dsdt file. I can't make it over the first line which states an unexpected {. Will try again tomorrow.

Best,
mox

My setup: HP Elitebook 8570W | Windows 10 | EXP GDC Expresscard | GIGABYTE Radeon RX 580 | bequiet System Power 9 500 W - not working yet


ReplyQuote
abelviking
(@abelviking)
Eminent Member
Joined: 5 months ago
 

I'm completely confused about this part of the guide because the dsl and the asl don't match at all...

It is unlikely this will succeed first time as the compiler is very strict. Errors reported will need to be looked at with Notepad++ on the line they occur on. Search for a unique error keyword from your dsdt-modified.dsl error line within dsdt-modified.asl (it may look a bit different) and the just swap the lines above/below into your dsdt-modified.dsl file and compile. This was sufficient to get a Dell XPS 9350, Dell E6540, 2016 15" Macbook Pro DSDT override all done perfectly.

Output from compiler:

c:\dsdt>iasl -ve dsdt-modified.dsl

Intel ACPI Component Architecture
ASL+ Optimizing Compiler version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation

Compiler aborting due to parser-detected syntax error(s)
dsdt-modified.dsl 5765: SGOV (0x01070004, Zero) = NH14 (RH14, If ((BICM == One))
Error 6126 -                                                                                                     ^ syntax error, unexpected PARSEOP_IF, expecting PARSEOP_CLOSE_PAREN or ','

dsdt-modified.dsl 5766: {
Error 6126 - syntax error, unexpected '{' ^

dsdt-modified.dsl 5779: })
Error 6126 - ^ syntax error, unexpected PARSEOP_CLOSE_PAREN

dsdt-modified.dsl 5830:
Error 6126 - syntax error, unexpected PARSEOP_SCOPE, expecting $end and premature End-Of-File

ASL Input: dsdt-modified.dsl - 5830 lines, 220959 bytes, 1798 keywords

Compilation complete. 4 Errors, 0 Warnings, 0 Remarks, 0 Optimizations

Attached is the dsl and asl file. The 8- Copy versions are version BEFORE changes...of course the asl is the old one.

What  should I be looking for in the .asl file. There's no PARSEOP methods there.

Lastly my dsl creation threw some warnings. I'm wondering if it is related.

This post was modified 5 months ago

Sytem InfoiMAC Late 2015. dGPU AMD M390, OSX 10.12.6, Windows 17134.rs4_release.180410-1804(1803). GYGABYTE AORUS RX580 miniITX eGPU my signature


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 

It would be very helpful to specify the exact setups where this fix is necessary and useful, because this is not always the case.
@mac_editor, could you do this?

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
mac_editor
(@mac_editor)
Noble Member Moderator
Joined: 2 years ago
 

@oliverb the guide clearly states what it solves, and what to do before attempting this. I don't see a problem. Also, since this guide can apply to any Windows system, it's not possible to list all configurations. Future eGPU users need to read up what they need to do before they read how to.

purge-wrangler.shpurge-nvda.shset-eGPU.shautomate-eGPU EFI Installer


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 
Posted by: mac_editor

@oliverb the guide clearly states what it solves, and what to do before attempting this. I don't see a problem. Also, since this guide can apply to any Windows system, it's not possible to list all configurations. Future eGPU users need to read up what they need to do before they read how to.

@mac_editor
this is a topic I am disagreeing. This guide sounds like this fix is a necessity for eGPU in general. This happens to other guides on this site as well. People are following this guide (and other guides) even they have setups where it is just not fitting. I agree that they don't read it well but this is exactly the problem. We go from complex to simple. "Simple" is called "lucky", but it's rather not. I claim that on most setups Error 12 is not a problem anymore, not vice versa.

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
itsage
(@itsage)
Famed Member Admin
Joined: 3 years ago
 

@oliverb We can update the OP to list the exact setups that don't require this error 12 fix. Feel free collect that information yourself or  report more host systems that work without encountering error 12.

I've stated this before and I say it again. Solutions made avaiable in our community arose from the needs of people who encountered issues. If you don't encounter any of these issues, that's great. You should move on to things that are more productive.

Best ultrabooks for eGPU use

eGPU enclosure buying guide


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 

@itsage, of course I do not have unlimited access to MacBooks (I could test 4 different so far, 2 owning myself, none of them had Error 12 issues), but there are a lot of unanswered questions.

E.g. I asked you personally in this thread how it is possible that one user experienced Error 12 troubles in a setup that is exactly the same as one of mine where Error 12 was never an issue. I even suggested a possible reason for it. No answer... still.

For me, and you may personally ban me from this site for my opinion, this is looking like hanging on old problems, making things complicated and thus confusing people more than necessary.
I don't know why. It is because it was so much work to solve all this problems? I understand this, but it's not time to move on? To say, "Ok, those old problems are gone, our great solutions are not needed anymore".

I know we have complete disagreement on this topic.

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
Eightarmedpet
(@eightarmedpet)
Noble Member
Joined: 3 years ago

ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 
Posted by: Eightarmedpet

I wouldn’t of got past error 12 without this guide...

@eightarmedpet, which exact setup made the problems? A 13-inch MBP 2017?! I guess it was an AMD eGPU?!

If it is this setup, booting with the eGPU plugged in => No Error 12. (This is a rule for any MBP 2017, 2018 and for 13-inch 2015, MacMini 2018. Others fall under this rule,  too. I claim it's almost every newer mac, 15-inch 2016 may be an exception, have to get one yet).

You can even plug two (2) AMD eGPUs, this means Windows 10 (Build 1803) does allocate the necessary resources when the eGPU is detected on startup.

This post was modified 4 months ago

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
OliverB
(@oliverb)
Noble Member
Joined: 10 months ago
 

I have constructive suggestion in this matter: I believe, you may need an Error 12 fix like in this guide, if either:

1) You must/want/need to hotplug an AMD eGPU
2) You have an older or broken Windows Version

Of course 1) is a good reason, especially if you want to avoid internal black screen. This actually hits me on my 15-inch 2018 system, too (Not on the 2015 Macbook), but there are a couple of other ways to avoid internal black screen. I find each of them more applicable than disabling left Thunderbolt ports.

2018 15" MBP & 2015 13" MBP connected to RTX2080Ti GTX1080Ti GTX1080 Vega56 RX580 R9-290 GTX680


ReplyQuote
speedyconzales
(@speedyconzales)
Active Member
Joined: 4 months ago
 
Posted by: MikejeffWong

Hey guys I am wondering if anyone can help me out here

 

I stuck am on Step 1 using Option 2: Use the Microsoft method  

  • I added the required code needed for the DSDT override
  • I am stuck on one last line of code that I do not understand
  • ERROR: Expecting Symbol ‘{‘
  • This error is literally the very last line of the .asl document 

 

As easy as that error sounds, I have not found a solution on resolving this issue and need some insight from anyone who has encountered this 

I have linked a screenshot of the error code, the original DSDT.asl file, and also the modified DSDT.asl file with the error code

Much appreciated if anybody can help me with this! 🙂

 

same problem here. is there a solution for this problem? checked syntax of code. no obvious mistake there. 

seems to be the last error, because it appears also at the end of the code. all other errors were fixed by me... stuck at this point sadly. also the last option for me to try to bypass error 12

UPDATE: 
I somehow managed to create an *.aml file using the intel method. How can I add that to the public repository? I got my eGPU running under windows with this method!

This post was modified 4 months ago

iMac Retina 5k, 27-inch, Late 2015; TB2>TB3; Razer Core X; Radeon VII


itsage liked
ReplyQuote
Tod4
 Tod4
(@tod4)
New Member
Joined: 3 months ago
 

Hello everyone

My eGPU appears in Device Manager but I get an error 12. I have used these instructions but I  get a BSOD everytime I load the .aml file with eGPU Setup 1.35.

I then have to do a start-up repair to boot into windows.

The "Large Memory" just does not come up in Device Manager at all.

It seems like all the DSDT Override files I keep compiling successfully (no errors) do not want to load on Windows bootup.

Has anyone had the same problem and found a fix to this?

Thanks

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
(@dan_wes)
New Member
Joined: 3 months ago
 

I have a Macbook Pro early-2011. Triple booted with Win10, Linux and OSX - all 64 bit editions.

In Windows 10-64x, the audio will not work with UEFI booting on MacBooks older than 2012.

This DSDT work around fixed my audio device on my laptop under Windows. Thanks!! 🙂

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


itsage liked
ReplyQuote
albenex
(@albenex)
Active Member
Joined: 3 months ago
 

Hello,
my machine:
Lenovo t540p
Intel HD 4600 (igpu)
Nvidia GT 730m (dgpu)
Amd RX 590 (egpu) + corsair VS350 + exp gdc 8.5 (set with ATX power on) + 12V DC adapter 
16gb RAM
Windows 10 Pro (UEFI)
I successfully got large memory by following this guide and at the end using Clover v4077 (newer versions stuck at "scanning entries")

in order to pass POST i have to not attach any egpu .

for test purpose i initially tried a very old Radeon x1550 card which was working properly, then bought RX590

RX590 however shows error #12.
i tried:
-disabling dgpu (currently disabled)
-removing RX590 and related PCI Express root ports and drivers, then re attaching card and reinstalling drivers
-setting forcedisable on bcdedit
-removing PCI to PCI bridge
-even removing igpu from resources list

no success.
i was wondering if nando's script would help but in this guide i read it is just for MBR partition scheme, i have GPT.

i don't know what to try now..

Lenovo Thinkpad t540p 16GB/250GB Intel HD 4600 / Nvidia GT 730m / W10 pro MBR
EXP GDC 8.5 / Corsair VS350 / RX 590 8GB


ReplyQuote
albenex
(@albenex)
Active Member
Joined: 3 months ago
 
Posted by: albenex

Hello,
my machine:
Lenovo t540p
Intel HD 4600 (igpu)
Nvidia GT 730m (dgpu)
Amd RX 590 (egpu) + corsair VS350 + exp gdc 8.5 (set with ATX power on) + 12V DC adapter 
16gb RAM
Windows 10 Pro (UEFI)
I successfully got large memory by following this guide and at the end using Clover v4077 (newer versions stuck at "scanning entries")

in order to pass POST i have to not attach any egpu .

for test purpose i initially tried a very old Radeon x1550 card which was working properly, then bought RX590

RX590 however shows error #12.
i tried:
-disabling dgpu (currently disabled)
-removing RX590 and related PCI Express root ports and drivers, then re attaching card and reinstalling drivers
-setting forcedisable on bcdedit
-removing PCI to PCI bridge
-even removing igpu from resources list

no success.
i was wondering if nando's script would help but in this guide i read it is just for MBR partition scheme, i have GPT.

i don't know what to try now..

"Solved" by reinstalling W10 in MBR and using Setup 1.35
however now i'm facing error 43...

Lenovo Thinkpad t540p 16GB/250GB Intel HD 4600 / Nvidia GT 730m / W10 pro MBR
EXP GDC 8.5 / Corsair VS350 / RX 590 8GB


ReplyQuote
nu_ninja
(@nu_ninja)
Estimable Member
Joined: 1 year ago
 

I looked up your specs and it appears there's a recent bios update for your laptop. I wonder if that's messing with anything...

Mid-2012 13" Macbook Pro (MacBookPro9,2) TB1 -> RX 460/560 (AKiTiO Node/Thunder2)
+ macOS 10.14+Win10
+ Linux Mint 19.1


ReplyQuote
Alnitak
(@alnitak)
Active Member
Joined: 2 months ago
 

Hello all. I recently purchased the EXP GDC V8.5c (ExpressCard version) along with a GTX 1660 to go along with my Lenovo ThinkPad L540.  When I connected the setup I first received  the common "code 12." To resolve this I tried to do a DSDT override. First I tried using the Microsoft method as this is recommended Lenovo laptops, but I was unsuccessful. Even though I changed "Buffer(0x1D4)" to "Buffer(0x202)" along with replacing the 46 hex values making sure this was done in the correct working area of my "dsdt-modified.asl" file I always ended up getting this error in CMD: error: AR00 is not a method. My knowledge of programming is really quite limited so I did not know how to resolve this.

I then gave the Intel method a shot, but problems arose when trying to decompile dsdt.dat as dsdt.dsl. Here I received the following warning on the command prompt :

"iASL Warning: There were 23 external control methods found during
disassembly, but only 16 were resolved (7 unresolved). Additional
ACPI tables may be required to properly disassemble the code. This
resulting disassembler output file may not compile because the
disassembler did not know how many arguments to assign to the
unresolved methods. Note: SSDTs can be dynamically loaded at
runtime and may or may not be available via the host OS. "

 

I did, however, get a dsdt-modified.dsl file which I was able to make the necessary edits to. The file compiled first try with 14 warnings but was nonetheless successful.  So then I tried loading my dsdt-modified.aml as a registry override and this seemed to be successful as I now saw a Large memory entry in the device manager, and sure enough error 12 was gone. The only issue now is that I don't want to permanently run Windows 10  in test mode as there are many drawbacks associated with running windows this way. Therefore I just recently bought Setup 1.35  in order to load my dsdt-modified.aml as an in-memory substitution instead. However, I cannot do this as dsdt-modified.aml (62.2kb) file is slightly bigger than the dsdt.dat file (61.9kb), and setup 1.35 checks the size difference before loading the aml file. I cannot make my dsdt-modified.aml  file any smaller by editing the .dsl file with my current programming knowledge. I have removed all the obvious stuff that nando4 posted in the error 12 guide, such as unused PEG0, PEG1, PEG2 along with removing old windows and Linux entries in the _OSI section of the .dsl file. But every time I compile the .dsl file the output is still 62.2kb so I have really given up at this point. 

So the questions I am left with are: 

  • Is the warning that I received when decompiling the dsdt.dsl file normal? And will this have any effect on my system stability such as random BSOD's etc?

 

  • Can anyone help me get my in-memory substitution up and running so I don't need to run windows in test mode?

Any help will be greatly appreciated!! 🙂

This post was modified 1 month ago

Thinkpad L540, 120hz IPS internal display( N156HHE-GA1), I7 4712MQ, EXP GDC 8.5C (Expresscard), Palit GTX1660 Storm OC eGPU , Adata SU800 1TB SSD, 16GB 2133MHZ HyperX DDR3L.


ReplyQuote
thisguyiknow
(@thisguyiknow)
New Member
Joined: 1 month ago
 

I found this thread several months back and it happened to also fix the 'no audio in pre 2013 Macs' in Windows when installed/booted as UEFI - the downside of this mod is that the PCI bridges for thunderbolt are not 'exposed' to windows -- I wonder if there is a qword that allows for full 64bit address range instead of 36bit or is 36bit our only choice -- Thanks for this post as it did at least get the builtin audio working on pre 2013 macs in UEFI mode Windows 

-Derek

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 

@thisguyiknow, the OP instructions create a window in the 48.5MB-56.25MB range. It's 36-bit so as to be compatible with PAE extended 32-bit Windows. The only good reason to go beyond this range is if you have physical memory occupying that space. We are seeing some 32GB RAM systems. Not so much 64GB. When we do, just expand the window to be beyond 64GB.

 

QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite,
   0x0000000000000000, // Granularity
   0x0000000C20000000, // Range Minimum,  set it to 48.5GB
   0x0000000E0FFFFFFF, // Range Maximum,  set it to 56.25GB
   0x0000000000000000, // Translation Offset
   0x00000001F0000000, // Length calculated by Range Max - Range Min.
   ,, , AddressRangeMemory, TypeStatic)
})
This post was modified 1 month ago

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table    •    Several builds
2015 15" Dell Precision 7510 M1000M + GTX 1080 Ti @ 32Gbps-M2 (ADT-Link R43SG) + Win10


ReplyQuote
thisguyiknow
(@thisguyiknow)
New Member
Joined: 1 month ago
 

Okay I understand -- I am very happy that this dsdt mod does indeed expose the internal audio but am struggling to get the thunderbolt bridges exposed - perhaps removing them in device manager and rebooting will force windows to re-allocate correctly - if not I am still happy but audio working as I don't really use thunderbolt when in Windows

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
dbj1780
(@dbj1780)
New Member
Joined: 10 months ago
 

hello everyone, I try to use this guide for make my egpu working (error 12) , I do anything but I have 2 errors with my dsdt-modified.dsl file 🙁 . Who can help me please ? thank you very much

 

Intel ACPI Component Architecture
ASL+ Optimizing Compiler version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation

Compiler aborting due to parser-detected syntax error(s)
dsdt-modified.dsl 2488: ECRW (If (PM0H)
Error 6126 - ^ syntax error, unexpected PARSEOP_IF, expecting PARSEOP_CLOSE_PAREN or ','

dsdt-modified.dsl 2492: }) = Zero
Error 6126 - ^ syntax error, unexpected PARSEOP_CLOSE_PAREN

ASL Input: dsdt-modified.dsl - 12673 lines, 400613 bytes, 4627 keywords

Compilation complete. 2 Errors, 0 Warnings, 0 Remarks, 0 Optimizations

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
(@carlos_orozco)
New Member
Joined: 1 month ago
 

HP Elitebook 8770W
Microsoft Windows 10 Pro 64
RAM 16G
i7-3840qm
Legacy Bios
Internal graphic QUADRO K3000M
EGPU XPRESS CARD Radeon 280X 

Hello,I want the support about the DSDT Override but I can not generate because at the time of executing the commands I get this 

C:\dsdt>c: & cd \dsdt
C:\dsdt>acpidump -b -z
Intel ACPI Component Architecture
ACPI Binary Table Dump Utility version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation
Writing [DSDT] to binary file: dsdt.dat 0x20987 (133511) bytes
Writing [XSDT] to binary file: xsdt.dat 0xA4 (164) bytes
Writing [FACS] to binary file: facs.dat 0x40 (64) bytes
Writing [FACP] to binary file: facp.dat 0x10C (268) bytes
Writing [SSDT] to binary file: ssdt.dat 0x64F (1615) bytes
Writing [SSDT] to binary file: ssdt1.dat 0x692 (1682) bytes
Writing [SSDT] to binary file: ssdt2.dat 0x48A (1162) bytes
Writing [SSDT] to binary file: ssdt3.dat 0x19EF (6639) bytes
Writing [SSDT] to binary file: ssdt4.dat 0x9B6 (2486) bytes
Writing [SSDT] to binary file: ssdt5.dat 0xAAD (2733) bytes
C:\dsdt>asl /u dsdt.dat
Microsoft ACPI Source Language Assembler Version 5.0.0NT
Copyright (c) 1996,2014 Microsoft Corporation
Compliant with the ACPI 5.0 Specification
asl_ERR: UnAsmSuperName: invalid SuperName - 0x83
asl_ERR: UnAsmSuperName: invalid SuperName - 0x0a

 

 
 
THANKS

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
Asuforce
(@asuforce)
New Member
Joined: 3 weeks ago
 

I tried to do this method.
But I can't create dsdt.dat file. This file is empty.

Here is my command logs.

C:\dsdt>acpidump -b -z
Intel ACPI Component Architecture
ACPI Binary Table Dump Utility version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation
Writing [DSDT] to binary file: dsdt.dat 0x3D4F7 (251127) bytes
Writing [XSDT] to binary file: xsdt.dat 0x10C (268) bytes
Writing [FACS] to binary file: facs.dat 0x40 (64) bytes
Writing [FACP] to binary file: facp.dat 0x114 (276) bytes
Writing [SSDT] to binary file: ssdt.dat 0x1B1C (6940) bytes
Writing [SSDT] to binary file: ssdt1.dat 0x1663 (5731) bytes
Could not get SSDT registry entry: No more data is available.
Could not get ACPI table at index 6, AE_ERROR
Writing [SSDT] to binary file: ssdt3.dat 0x30D7 (12503) bytes
Could not get SSDT registry entry: No more data is available.
Could not get ACPI table at index 8, AE_ERROR
Could not get SSDT registry entry: No more data is available.
Could not get ACPI table at index 9, AE_ERROR
Could not get SSDT registry entry: No more data is available.
Could not get ACPI table at index 10, AE_ERROR
Writing [SSDT] to binary file: ssdt7.dat 0x7FBF (32703) bytes
Writing [SSDT] to binary file: ssdt8.dat 0x144 (324) bytes

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
nando4
(@nando4)
Noble Member Admin
Joined: 3 years ago
 
Posted by: @asuforce

I tried to do this method.
But I can't create dsdt.dat file. This file is empty.

Here is my command logs.

C:\dsdt>acpidump -b -z
Intel ACPI Component Architecture
ACPI Binary Table Dump Utility version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation
Writing [DSDT] to binary file: dsdt.dat 0x3D4F7 (251127) bytes
..

acpidump output shows a dsdt.dat file was created. Pls ensure you work through the opening post as the dsdt.dat needs to be further processed to get a dsdt.dsl edittable file.

This post was modified 3 weeks ago

eGPU Setup 1.35    •    eGPU Port Bandwidth Reference Table    •    Several builds
2015 15" Dell Precision 7510 M1000M + GTX 1080 Ti @ 32Gbps-M2 (ADT-Link R43SG) + Win10


ReplyQuote
Asuforce
(@asuforce)
New Member
Joined: 3 weeks ago
 

@nando4

Thanks for replying.
I tried to create dsdt.ASL file, but this file is empty.

C:\Windows\system32>c: & cd \dsdt

C:\dsdt>asl /u dsdt.dat
Microsoft ACPI Source Language Assembler Version 5.0.0NT
Copyright (c) 1996,2014 Microsoft Corporation
Compliant with the ACPI 5.0 Specification

asl_ERR: UnAsmOpcode: invalid opcode class 0

then I created dsdt.dsl

C:\dsdt>iasl -da -dl -fe refs.txt dsdt.dat

Intel ACPI Component Architecture
ASL+ Optimizing Compiler version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation

Input file dsdt.dat, Length 0x3D4F7 (251127) bytes
ACPI: DSDT 0x0000000000000000 03D4F7 (v02 DELL CBX3 01072009 INTL 20160527)
refs.txt: Importing method external (1 arguments) MDBG
refs.txt: Importing method external (0 arguments) _GPE.MMTB
refs.txt: Importing method external (2 arguments) _SB_.PCI0.LPCB.H_EC.ECWT
refs.txt: Importing method external (1 arguments) _SB_.PCI0.LPCB.H_EC.ECRD
refs.txt: Importing method external (2 arguments) _SB_.PCI0.PEG0.PEGP.SGPO
refs.txt: Importing method external (1 arguments) _SB.PCI0.GFX0.DD02._BCM
refs.txt: Importing method external (4 arguments) _SB.PCI0.SAT0.SDSM
refs.txt: Importing method external (0 arguments) _SB.PCI0.XHC.PS0X
refs.txt: Importing method external (0 arguments) _SB.PCI0.XHC.PS3X
refs.txt: Importing method external (0 arguments) _SB.PCI0.SDHC.PS0X
refs.txt: Importing method external (0 arguments) _SB.PCI0.I2C0.PS0X
refs.txt: Importing method external (0 arguments) _SB.PCI0.I2C0.PS3X
refs.txt: Importing method external (0 arguments) _SB.PCI0.I2C1.PS0X
refs.txt: Importing method external (0 arguments) _SB.PCI0.I2C1.TPD7.PS0X
refs.txt: Importing method external (0 arguments) _SB.PCI0.I2C1.TPD8.PS0X
refs.txt: Importing method external (0 arguments) _SB.PCI0.SDHC.WI01.PS3X
refs.txt: Imported 16 external method definitions
Pass 1 parse of [DSDT] Pass 2 parse of [DSDT] Parsing Deferred Opcodes (Methods/Buffers/Packages/Regions)

Parsing completed

Found 214 external control methods, reparsing with new information
Pass 1 parse of [DSDT] ACPI Error: [ESPC] Namespace lookup failure, AE_ALREADY_EXISTS (20161222/dsfield-732)
ACPI Error: [IDMN] Namespace lookup failure, AE_ALREADY_EXISTS (20161222/dsfield-732)
ACPI Error: [IDPC] Namespace lookup failure, AE_ALREADY_EXISTS (20161222/dsfield-732)
ACPI Error: [DLPN] Namespace lookup failure, AE_ALREADY_EXISTS (20161222/dsfield-732)
ACPI Error: [PSON] Namespace lookup failure, AE_ALREADY_EXISTS (20161222/dsfield-732)
Pass 2 parse of [DSDT] Parsing Deferred Opcodes (Methods/Buffers/Packages/Regions)

Parsing completed
Disassembly completed
ASL Output: dsdt.dsl - 1887206 bytes

iASL Warning: There were 214 external control methods found during
disassembly, but only 199 were resolved (15 unresolved). Additional
ACPI tables may be required to properly disassemble the code. This
resulting disassembler output file may not compile because the
disassembler did not know how many arguments to assign to the
unresolved methods. Note: SSDTs can be dynamically loaded at
runtime and may or may not be available via the host OS.

In addition, the -fe option can be used to specify a file containing
control method external declarations with the associated method
argument counts. Each line of the file must be of the form:
External (<method pathname>, MethodObj, <argument count>)
Invocation:
iasl -fe refs.txt -d dsdt.aml

Edit dsdt-modified.dsl

QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, NonCacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000000000010000, // Range Minimum
0x000000000001FFFF, // Range Maximum
0x0000000000000000, // Translation Offset
0x0000000000010000, // Length
,, _Y0F, AddressRangeMemory, TypeStatic)
DWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, NonCacheable, ReadWrite,
0x00000000, // Granularity
0xFC800000, // Range Minimum
0xFE7FFFFF, // Range Maximum
0x00000000, // Translation Offset
0x02000000, // Length
,, , AddressRangeMemory, TypeStatic)
QWordMemory (ResourceProducer, PosDecode, MinFixed, MaxFixed, Cacheable, ReadWrite,
0x0000000000000000, // Granularity
0x0000000C20000000, // Range Minimum, set it to 48.5GB
0x0000000E0FFFFFFF, // Range Maximum, set it to 56.25GB
0x0000000000000000, // Translation Offset
0x00000001F0000000, // Length calculated by Range Max - Range Min.
,, , AddressRangeMemory, TypeStatic)
})
Device (SRRE)
{
Name (_HID, EisaId ("PNP0C02") /* PNP Motherboard Resources */) // _HID: Hardware ID
Name (_UID, "SARESV") // _UID: Unique ID
Method (_STA, 0, Serialized) // _STA: Status
{
If ((TLUD >= 0x0404))
{
Return (0x03)
}
Else
{
Return (Zero)
}
}

Method (_CRS, 0, Serialized) // _CRS: Current Resource Settings

Create dsdt-modified.dsl

C:\dsdt>iasl -ve dsdt-modified.dsl

Intel ACPI Component Architecture
ASL+ Optimizing Compiler version 20161222-32
Copyright (c) 2000 - 2016 Intel Corporation

Compiler aborting due to parser-detected syntax error(s)
dsdt-modified.dsl 4717: ECRW (If (PM0H)
Error 6126 - ^ syntax error, unexpected PARSEOP_IF, expecting PARSEOP_CLOSE_PAREN or ','

dsdt-modified.dsl 4721: }) = Zero
Error 6126 - ^ syntax error, unexpected PARSEOP_CLOSE_PAREN

dsdt-modified.dsl 22448: Local0
Error 6126 - ^ syntax error, unexpected PARSEOP_LOCAL0

dsdt-modified.dsl 22736: Local0
Error 6126 - ^ syntax error, unexpected PARSEOP_LOCAL0

dsdt-modified.dsl 22770: Local0
Error 6126 - ^ syntax error, unexpected PARSEOP_LOCAL0

dsdt-modified.dsl 23071: Local2
Error 6126 - ^ syntax error, unexpected PARSEOP_LOCAL2

ASL Input: dsdt-modified.dsl - 57083 lines, 1887637 bytes, 25274 keywords

Compilation complete. 6 Errors, 0 Warnings, 0 Remarks, 0 Optimizations

how to fix this syntax error. It is driving me crazy...

Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts


ReplyQuote
 of  7