Jump to content

60 posts in this topic

Recommended Posts

Testing the waters on my new Z170 build based off the entry level board and the Intel 6th gen i7-6700K. I was intending to complete the build, have it up & running smoothly before beginning this post. But I've run into problems where no amount of reading into Pike R. Alpha's post nor Google is helping. So here I am.

Reading around I got to downloading the 10.11.4 from Apple cause the Skylake build it appears needed that. Made the bootable with Clover-on-it-USB.

Clover strangely would not install. Showed complete progress but at the end I had no EFI partition mounted on my desktop which meant it was not okay. Naturally I was doing a UEFI only install. Learnt of needing to run "nvram -c" command in the terminal without quotes and the command to format the EFI partition as FAT32 & repeated the process & it succeeded.

Now I had the Monitor (LP2065) connected via DVI to the onboard video. I got the display on BIOS and the first thing I did was to flash latest F7 BIOS. Rebooted and set the values I thought were okay.

Among the new ones was the APIC numbered setting that I disabled. But as I later found enabling did not change anything. I already had few edits in the config.plist. (a.) The patch for USB port limit (b.) Skylake graphics patch (c.) IOGraphicsFamly patch (for taking care of the second stage glitch).

In verbose mode the USB Stick booted right up. It was throwing a few unfamiliar messages but the installer booted okay. Without much fuss I had the Samsung 850EVO connected and so installed on it. Went through quite fast compared to my older install on Z97-4790K system.

Rebooted SSD and installed clover on it & it worked right away.

Time to look into the Power Management side of things.

the ssdtprgen Beta script downloads and works and also makes a SSDT.aml file. But the thing appears to do nothing. I placed it in ACPI patched and rebooted. Checked with the Intel Power Gadget idles at about 1.6 Ghz (1.51 Ghz to be lowest) at 25º C. Loading the CPU with the dummy yes command in terminal (single instance) takes CPU to 4.0 Ghz with temps at 40º and 6 more instances make the CPU reach 55º. Subsequently killing all yes instances bring CPU idle back to 1.57 Ghz.

So I tired to install plugins to FaleSMC (HWSensors) and used Istatmenus. It showed identical values. I believe HWmonitor would too since I have seen it behave similarly in the past.

Next option left to check was the AppleIntelInfo kext from Pike R. Alpha. Here is where I get lost The thing does not output anything in the terminal or the console. I swear in the early days of building the 4790K system I had the results in terminal. Now nothing. No error messages nothing.

So as things stand - power management is incomplete. System idles at 1.5 Ghz instead of 800 Mhz and turbo boosts to 4.0Ghz. at least what IntelPowerGadget V 3.0.3 tells me. The presence of SSDT does not change anything. But speed stepping is working regardless of its state in the BIOS, enabled or disabled..

So friends any ideas?

 

EDIT:1 - Just checked - sleep not working yet

Share this post


Link to post
Share on other sites
Advertisement

This Power Management thing definitely is Voodoo / Rocket Science.

Loading piker's kext AppleIntelInfo and then at

cat /var/log/system.log | grep "AICPUPMI:"

gives me

MyMac : TTY=ttys000 ; PWD=/Users/MyMac/Downloads ; USER=root ; COMMAND=/usr/bin/grep AICPUPMI: in the terminal.

Both my 4790K and 6700K systems exhibit same thing.

What am I doing wrong here? SIP settings both systems 0X67.

I swear it used to work okay last time I checked in Yosemite.

10.11.4 seems to have changed something.

Guys?

Share this post


Link to post
Share on other sites

Using Intel Power Gadget 3.0.3 from both my systems:

on the Haswell 4790K it exhibits correct graphs. Idling at 0.8 Ghz and when loading with a dummy yes command from terminal like so

yes > /dev/null &

a single instance takes CPU to 4.4 Ghz dropping down to 4.2 Ghz when several more instances of the yes command is repeated. This is the way it should be. Right guys?

Subsequently

killalll yes

immediately terminates all instances of yes and brings the CPU back down to 0.8 ghz instantly.

Man this has got some really naughty connotations. So many Yeses and hot temperatures and all. What the deuce man it's Monday morning. Sorry! Back to the problems at hand.

On the new Skylake 6700K system I put together it settles down to 1.6 Ghz and hits a ceiling of 4.0 Ghz.

Bios setings for the Turbo is at 42 with 1 core and 40 with others.

Thing is even if I disable Turbo Boost and EIST and all C states except C1 in BIOS it still is the same. I suspect since -xcpm engages requisite PM drivers those are not ready yet, or I am missing something.

BTW using iMac 17,1  SMBIOS in clover

Thank you.

Share this post


Link to post
Share on other sites

When you run a program like yes > /dev/null, it is normal that your CPU is running on its highest freq.

 

It is normal, because we always want that our programs run fast. Right?

 

Check your cpu's freq. with HWMonitor, when you do not run any program what creates heavy load.

Share this post


Link to post
Share on other sites

Friend, if you read my first post here, I know it's a long one, you'll see I tried both hwmonitor and istat which both get data from the plugins in fakesmc. The project goes by the name Hwsensors. This shows only 1.6 ghz or 4.0. Strangely on my Haswell it shows stepping through different ones from 0.8 to 4.36 ghz. So I use Intel power gadget which also shows graph from 1.6 to 4.0. I use the yes command to check the max and min. and throttling response.

This was one of the chief reasons I wanted AppleIntelInfo to work. But it is behaving stangely

Share this post


Link to post
Share on other sites

Reporting Comparative results of the Intel Power Gadget on same hardware between Windows 10 X64 and 10.11.4. -

Windows 10 X64 Booted with all CPU Settings on auto (6700K on Z170M-D3H)

Windows 10 X64 and Used a freeware to load CPU cores selectively.

This is what Intel Power Gadget reports when idle / When loaded one core / When loaded all cores

Followed by 10.11.4 different stages of loading.

Most baffling is why do both versions max out at 4.0 ghz whereas turbo is 4.2 ghz. Never even once for a fraction of a second does it show 4.2.

The only way I got it upto 4.2 is by manually changing turbo boost in BIOS all 4 values to 42, which I believe is incorrect?

 

Can anyone with a 6700K please verify.

Thanks 

post-144165-0-27494900-1461647617_thumb.jpg

post-144165-0-02627900-1461647727_thumb.jpg

post-144165-0-45762400-1461647770_thumb.jpg

post-144165-0-75736600-1461648084_thumb.jpg

post-144165-0-37191900-1461648092_thumb.jpg

post-144165-0-07324000-1461648099_thumb.jpg

post-144165-0-40166500-1461648105_thumb.jpg

Share this post


Link to post
Share on other sites

This one I forgot to put in since I was busy. 

Geekbench scores on the 4790K - Z97 is 17,100++ whereas the 6700K on Z170 is 16,700++ 

Am I flogging a dead horse here? Thinking of throwing in the towel. 200 plus views and no one's interested in touching the PM topic with even a 32 yard pole.

Probably have to shelve this build.

Share this post


Link to post
Share on other sites

Thanks Allan. Here are the files. I keep changing the plist as I keep trying things & I have a million questions for you. But first be so kind as to see these files. (ssdt.aml, ssdt.dsl and config.plist) with the ssdts being generated from ssdtprgen script from piker alpha. Thanks

 

Archive.zip

Share this post


Link to post
Share on other sites

No I do not use a DSDT and I do not use modified kexts. Only Clover on-the-fly kext patching. The only kext I use is in Clover>Kexts>10.11 and they are FakeSMC and IntelE1000.

Today I was trying to get audio to work using piker's alc command script for the ALC892 on board & at first it aborted stating HDEF not found. So I F4ed Clover & got the DSDT which I checked with Maciasl and true to the script's displeasure there was no HDEF but HDAS. This needed renaming and that is what you see in the plist. And now I can type while listening to Tchaikovsky or Bach.

Basically what I can see is that this motherboard needs nothing to work except for the usual kexts and LAN/Audio & its good to go.

What really surprised me is that xcpm is loaded & registered even if it is not selected on the clover plist boot flags. System runs stable, shuts down and restarts stable & repeatably.

Special note for owners of this board - IMPORTANT - Setting ERP Enable in the BIOS is what changes shutdown behaviour which is not OS related. Otherwise after powering down system, pressing spacebar switches back it on. With ERP enabled it stays off no matter what unspeakable things you do to the spacebar.

Unfortunately like all others on the forum with HD530 NO Sleep. Hoping someone will solve this before Apple does.

Most Baffling is why Intel Power Gadget does not go beyond 4.0 ghz both in windows and Mac.

Share this post


Link to post
Share on other sites

Thank you very much. From what I can see you've removed the Drop Tables part which had only one entry for CpuPm and you've enabled the APIC patch just below. You've also enable dart=0 and under devices in USB inject is now ticked as is Fix Ownership.

Have I missed anything?

So I used this plist & booted up. Everything's working the way it was, including Tchaikovsky. I do not perceive any difference. Intel Power Gadget is also between 1.52 Ghz and 4.0 Ghz

Anything else you would want me to report?

Thanks

Share this post


Link to post
Share on other sites

I cannot help  you much but I have banged my head around on this speedstep thing for a long time (ever since elcapitan started supporting Skylake). Just to save you the frustation that I dealt with, I am going to tell you that I have been unsucccessful so far. The low CPU idle is around 1.6 Ghz, like your build. It seems to be more motherboard related to me. Reading other forum posts it seems asus boards idle down to 800 Mhz, even if they have my same CPU But then some gigabyte boards idle down to 800 Mhz as as well. I asked pike about the high idle frequency but I could not get a fix. I have tried changing multiple options on the bios power management, changing frequency vectors to a haswell mac smbios, tweaking config.plist, multiple fresh installs of OSX, updating to new clover builds as they become available but to no avail. 

Regarding sleep, It does work but I have to use NVIDIA card. Even on 10.11.4 with the Intel HD graphics, sleep does not work for me. But I have not been following the forums lately, so some one might have come up with a fix. 

If you have any questions let me know, but I probably can't help fix this problem. May be working together we can find a solution. 

Share this post


Link to post
Share on other sites

@Allan

Geek bench score is 16900. It is what I was expecting as I have 2400 mhz memory which when run at 2133 gives a score of 16700. So nothing has changed as I was expecting.

Also the patching of APIC seems redundant as I see enabling or disabling APIC entires in BIOS also does not do anything.

I am attaching all dumps from Darwin Dumper which should give one the clearest picture of the system layout. I hope that will suffice?

Thank you

@hashis.k

First advice - please stick around so all Gigabyte Z170 issues can be consolidated making it easier for our other friends.

I too went through Pike's Universum and chanced upon his brush with an incorrect version of Intel Power Gadget. It appears Patrick Konsor still missed something. You will see in my screen gabs that same hardware running windows 10 X64 shows in Intel Power Gadget as system idling at 800 mhz. Which leads me to conclude that there is nothing wrong with the motherboard. If you would try something similar this theory would become more solid. Boot a native windows install and check IPG results. Which brings us to the next problem. Either IPG is still broken or Apple still has a lot of work to finish in managing PM on this platform (more likely). I used HWSensors with HWMonitor and iStat Menus both reporting the same thing. Could you please check? Also can you check your Geekbench score too?

From Pike's blog all I can gather is this is a very much work in progress, or like in good old days with Master Chief on P55 we have to get our hands dirty manually editing SSDT and get it t work. I've forgotten everything now.

Whew! Oh God, there's more?

How did you get the GTX950 to work? Clover patching of AppleGraphicsDevicePolicy does not seem to help. more on this here

http://www.insanelymac.com/forum/topic/308449-how-to-fix-second-stage-boot-logo-and-loading-bar-for-some-dedicated-desktop-video-cards/page-15

See my post #289 Also i am content in sacrificing sleep as long as the HD530 works as it now does fully because shutdown and start are so fast that I really don't need sleep. Besides Apple made sleep terrible after ML, it takes so long to sleep. But HD530 should improve with 10.11.5 & later.

There's even more my friend..

You will see that I started a post separately for USB 3.0, will appreciate if you share something there for our friends. Yeah?

Cheerio!

DarwinDumper_2.9.9.2_AMI_X64_3438_Cap_MyMac.zip

Share this post


Link to post
Share on other sites

@Allan

Geek bench score is 16900. It is what I was expecting as I have 2400 mhz memory which when run at 2133 gives a score of 16700. So nothing has changed as I was expecting.

Also the patching of APIC seems redundant as I see enabling or disabling APIC entires in BIOS also does not do anything.

I am attaching all dumps from Darwin Dumper which should give one the clearest picture of the system layout. I hope that will suffice?

Thank you

@hashis.k

First advice - please stick around so all Gigabyte Z170 issues can be consolidated making it easier for our other friends.

I too went through Pike's Universum and chanced upon his brush with an incorrect version of Intel Power Gadget. It appears Patrick Konsor still missed something. You will see in my screen gabs that same hardware running windows 10 X64 shows in Intel Power Gadget as system idling at 800 mhz. Which leads me to conclude that there is nothing wrong with the motherboard. If you would try something similar this theory would become more solid. Boot a native windows install and check IPG results. Which brings us to the next problem. Either IPG is still broken or Apple still has a lot of work to finish in managing PM on this platform (more likely). I used HWSensors with HWMonitor and iStat Menus both reporting the same thing. Could you please check? Also can you check your Geekbench score too?

From Pike's blog all I can gather is this is a very much work in progress, or like in good old days with Master Chief on P55 we have to get our hands dirty manually editing SSDT and get it t work. I've forgotten everything now.

Whew! Oh God, there's more?

How did you get the GTX950 to work? Clover patching of AppleGraphicsDevicePolicy does not seem to help. more on this here

http://www.insanelymac.com/forum/topic/308449-how-to-fix-second-stage-boot-logo-and-loading-bar-for-some-dedicated-desktop-video-cards/page-15

See my post #289 Also i am content in sacrificing sleep as long as the HD530 works as it now does fully because shutdown and start are so fast that I really don't need sleep. Besides Apple made sleep terrible after ML, it takes so long to sleep. But HD530 should improve with 10.11.5 & later.

There's even more my friend..

You will see that I started a post separately for USB 3.0, will appreciate if you share something there for our friends. Yeah?

Cheerio!

The GTX950 did not work with iMac 17,1 Smbios with clover on the fly patching for me. So I modified the kext itself. I found a program on the other popular website which can modify the kext. here is the link

http://www.tonymacx86.com/threads/black-screen-with-macpro-6-1-or-imac-15-or-imac-17-system-definition.183113/. with this the sleep works perfectly and no black screen

 

I have windows dual boot and with windows it idles down to 800 Mhz. somehow the idling does not work properly on our boards on mac.

I am not sure how to check geekbench and have been using windows for now due to the powermanagement. I will give it a shot when time permits. 

Share this post


Link to post
Share on other sites

Came across Pike R Alpha's post that PM is indeed Intel Voodoo and this time around it is so bad that we need a new tool to tell us exact frequencies and now all PM is the domain of CPU Firmware.

While we wait, I have gone ahead and moved the new system from the Bench to the Case and begun using it around 12 hours a day from Sunday 1st May (Labour-day)

I had a pair of 8 GB G-Skill 3200Mhz CL14 RAM (16 GB Total) come in and now have a Geekbench score of 18,328 with Intel HD530 working.

Sleep still eludes but works when I am using GTX950. But the GTX 950 does nothing to the score except add just 45 points (18,373) Nice to be able to check same system for both HD530 vs NVIDIA scores on Geekbench running on 10.11.4. System runs super stable & not a single KP or app misbehaviour so far. Frankly I am surprised because my Z97- 4790K system plays the occasional truant even now.

I have a Dell U2515H running on HDMI at 2560X1440 @60Hz. This puts to rest the doubts I had with HDMI 1.4a support on this monitor. I had run the Mac Pixel Clock Patch before even connecting the monitor. I was using an old HP LP2065.

Will report more features & faults as I come across.

Oh! USB Drives all show as internal. Will deal with the issue in due course.

Cheers!

Share this post


Link to post
Share on other sites

Hey bro test this: attachicon.gifconfig.plist.zip

 

And, what you have in drivers64UEFI folder?

 

@Allan - Thx a ton for this. Used to end up with the prohibited sign with Clover(v2.3k_r3504)

 

This helped me get to the install screen -

El Capitan 10.11.4 is up and running.

 

Really appreciate it  :w00t:

 

@William Parker - Thx much for starting this thread.

 

Had removed the graphics card and unhooked the secondary drive for installation. 

 

 

Current setup:

  • Gigabyte Z170M-D3H (F7 Bios)
  • Intel core i5-6400k
  • Vengeance LPX DDR4 2400Mhz 8Gbx2
  • Sapphire Radeon R7 260X 2Gb
  • Samsung SSD 750 evo 250Gbx2

 

Left with the usual post installation minor woes.

 

Following kexts were used for installation: 

  • FakeSMC - placed in Other folder
  • USBInjectAll - placed in 10.10 & 10.11 folders
  • IntelMausiEthernet - placed in Other folder

 

Currently: 

  1. Sleep not working
  2. Shut down leads to a restart
  3. R7 260X shows only 7Mb
  4. RAM listed as 2133Mhz

Still to fix audio

Share this post


Link to post
Share on other sites

@ Graphire78

You are most welcome.

Are you using the plist Allan gave here? You will see that it contains the edit in "kernel and Kext patches" section for AppleUSBXHCIPCI to remove port USB limits. I am running without the USBinject all kext. I will come to the USB later on as I find time & I have another post for this USB part.

For you issues if I may ...

1. Sleep is dependent on a lot of things like power management and display, e.g. my system sleeps with an NVIDA card but not with the built in HD530. So once you've got all things going okay, sleep should fall in place.

2. Shutdown leads to reboot, this is for 2 reasons. First there is a setting in BIOS called "ErP" this needs to be enabled and secondly enable Clover ACPI Fixes "FixShutdown" and also "SlpSmiAtWake" (I hope you're using Clover Configurator)

3. This is one you got to figure out on different threads here 'cause I did not use one ever so haven't the faintest on it.

4. RAM listed a 2133, Where exactly? In the System profiler? Mine shows correctly as installed. Have you enabled XMP in BIOS? If not it will show correctly as 2133. So please use XMP Profile1 and that should make RAM works at 2400 Mhz as intended. Also while you'e at it please install the latest clover revision 3504.

 

Also if you have a USB 3.0 external Hard Disk Like WD Passport etc., could you connect it & report what it shows as, internal or external.

 

Thank you.

 

P.S. While fixing Audio make use of Toledo's ALC Script from his GitHub repo. Works Like a charm. Ensure that my REN AUDIO patch is there in Clover ACPI patches to rename the HDAS to HDEF otherwise the script fails.

post-144165-0-05537700-1462455235_thumb.jpg

Share this post


Link to post
Share on other sites
  • USB 3.0 external Toshiba HD shows up as internal.
  • Sleep is working with the graphics card.
  • Enabled ErP in Bios. Shut down works now.
  • Enabled XMP in Bios. RAM shows correctly as 2400 Mhz.
  • No longer using USBInjectAll.kext as I'm using Allan's config.plist

 

 

Thx much  :star_sunglasses:  :star_sunglasses:

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.

  • Similar Content

    • By Reel
      All work 10.10, 10.11, 10,12
      camera not work in 10.13, 10.14
      You need SwitchResX that the screen does not freeze
      Custom resolutions: Back porch - lines change to 2
      UEFI boot moded bios https://github.com/Thiblizz/l502x-modded-bios
       
      For 10.14 you need: HD3000 Fix for Mojave.zip
      Installed in S/L/E 
       
      SMBIOS: MacbookPro 8.3 or MacbookPro 9.1
       
       

       

       
       
    • By glasgood
      CLOVER DUAL BOOT MOJAVE & WINDOWS 10 GUIDE 
       

       
       
      INCLUDES  MBR / LEGACY BIOS  TO  GPT / EFI CONVERSION
      USING MBR2GPT TOOL
       
       
      PREREQUISITE: Two physical discs ( SSD’s or HDD’s )
       
       
       
       
       
      STEP 1 - Clover dual boot configuration 
       
      Open config.plist with Clover Configurator
       
      Boot
       Legacy = PBR Timeout = True ( will remove the Timeout countdown, from Clover boot menu)  

       
      GUI 
      Scan / Custom
       Entries = True  Tool = True  Legacy = False ( removes extra Windows 10 entries )  
      Hide Volume
      - Preboot ( macOS Preboot )
      - Recovery ( macOS Recovery )
       

       
      So at boot you will have two options: boot macOS Mojave or Windows 10 
       
       
       
       
       
       
       
      ————————————————————
       
       
      STEP 2 - Using a drive without Windows 10 installed
       
      Disconnect system drive that contains your macOS Mojave install from computer ( This is so that Windows does not overwrite existing macOS Mojave boot loader )
       
      Proceed with a Windows 10 UEFI install.  
      After installation reconnect macOS Mojave Drive, the Windows installation should now be detected and usable in Clover. 
      If Windows 10 is not detected or able to boot,  then verify you installed Windows 10 as UEFI and not MBR ---->  ( Read step 2 - For a drive with Windows 10 installed )
       
       
      OR
       
       
       
      STEP 2 - Using a drive with Windows 10 already installed
       
      Verify your Windows install is  GPT / UEFI or MBR / Legacy BIOS.   
      If Windows install is GPT UEFI then Windows 10 install is ready to use at Clover boot menu, you should be able to boot into Windows directly from Clover boot screen. 
       

       
       
      But if  Windows drive is detected at Clover boot screen, but when booting Windows you get a black screen with a cursor on the top left,
      then this is most likely because Windows drive is MBR ( Legacy BIOS ).  You can easily convert MBR to GPT using  Windows MBR2GPT tool ( this saves hours work having to reinstall Windows 10 and setting up all your applications again  ) 
       
      If Windows 10 install is MBR / Legacy BIOS  then simply convert to GPT / UEFI  following instructions below ( read video summary and view video )
       
       
      ** To use Windows 10  MBR2GPT tool  you must have Windows 10 version 1703 ( creators update  ) or later and less than 3 partitions on 
      the Windows 10 drive **
       
      Video summary:
       
      Confirm Windows 10 drive is MBR Legacy BIOS ( in Windows Disk Management ) Reboot into Windows PE ( Advanced Startup ) Convert from MBR Legacy BIOS to GPT UEFI ( using commands below ) mbr2gpt /validate mbr2gpt /convert Restart Verify Windows 10 drive has changed to GPT UEFI ( in Windows Disk Management )  
       
       
       
      After conversion Windows 10 is ready to use at the Clover boot menu 
       
       
    • By SepehrM
      I'm trying to install High Sierra 10.13.3 on an ASUS N552VW laptop, and I get stuck after this message:
      "could not recover SATA HDD after 5 attempts, terminating" 
      And some other warnings afterwards.
      Specs:
      CPU: Intel Skylake 6700HQ 2.6GHz
      RAM: 16GB DDR4
      2TB HDD + 128GB SSD
      GPU: Intel HD Graphics 530 + Nvidia GTX960m
      I'm installing using a 16GB USB3 flash drive on one of HDD volumes (UEFI Mode)
       
      I uploaded the verbose log and clover config.plist
      config-new.plist

    • By gk12345
      hello to everyone, i managed to install mojave on my asus ux303ub with i5 cpu and now sleep does not work when i press sleep display turns black and pc stays on and then i have to force shut down by holding power button in order to restart pc ,by the way i used the clover folder from a guide on the internet, any help would be appreciated, thanks
    • By vin047
      I have a Kingston HyperX Predator PCIE-SSD (model: SHPM2280P2H) matched with an X58 motherboard (legacy bios). The card has a Marvell 88SS9293 controller with an Option ROM which allows it to be detected on my motherboard as a bootable drive - listed as "IDE: Kingston SHPM". As my BIOS is non-UEFI, I assume the Option ROM presents it in IDE mode instead of AHCI.
       
      I use Clover legacy with BiosBlockIO to get around this, and it works well - Clover detects my drive and allows me to install macOS onto it and use it as a boot drive. macOS detects the drive correctly as a SATA AHCI drive. I guess Clover correctly detects (configures?) the Option ROM into AHCI mode and presents it as such to macOS - getting full speed read Read/Write performance. All is well.
       
      Except waking from sleep. When waking from sleep, depending on clover config options + dsdt, I either get an immediate reboot or hang on black screen that requires hard reset. I eventually realised that nothing after entering sleep mode is recorded in the logs, its completely empty until next boot up. This makes me think that on resume from sleep, macOS isn't seeing the drive anymore (hence can't even write to logs). My assumption here is that the Option ROM is presenting/configuring the card in IDE mode, which is not what macOS is expecting.
       
      This post describes the problem: 
       
      and also has a solution: in dsdt, simply write the correct values to 0x40, 0x41, 0x42 (which must be the BAR0 register?) to configure Option ROM to use AHCI mode. But I have no idea what address BAR0 is on the Marvell 88SS9293 - there appears to be no documentation whatsoever on this controller available online!
       
      I managed to dump the ROM using Linux sysfs, but don't have the skills to disassemble/decipher the binary to find the correct addresses. Anyone with such experience able to help?? I've attached the ROM dump and lspci output.
       
      Thanks in advance!
      rom.bin
      lspci_output.txt
×