No graphics / USB / Audio after wake

[Monq]

Much you are talking about DP, but what about those who use HDMI?

[Mieze]

Maguro did the trick. It only has one connector (this is the reason why I never tried it earlier), but I'm using DP only so it's not a problem for me.

I've patched the single connector in Maguro with the DP entry from Futumaki and the display is now waking correctly. The configuration works for both MacPro5,1 and MacPro6,1

Hmm, interesting but frankly I'm not surprised because Maguro is the personality used on the primary GPU of MacPro6,1. Obviously it's designed for the first GPU in the system which doesn't need help for wakeup. Would be interesting to see if Maguro also wakes up on socket 115x systems with the AMD card as the primary GPU?

 

Mieze

[TheRacerMaster]

Tried patching Maguro and sleep is still broken unfortunately (using iMac13,2).

[Mieze]

Tried patching Maguro and sleep is still broken unfortunately (using iMac13,2).

¡Qué lástima! 

[surfcaster]

Mieze my hard is the signature can I try something for you?

[Mieze]

Mieze my hard is the signature can I try something for you?

Thanks for your offer, but at the moment I'm completely at a loss!

[surfcaster]

I just get a list of all possible personalities of my chart, if I get somewhere'll let you know.

By the way I tried to Ozmosis 1479 and have the same results, no improvement, back to Clover

[Slice]

 

With that hard (GA-H61M i3-2120, Radeon HD 6670) should work as it did for me (Z68-MA d2H-B3 and i5 2400S+HD6850), just choose IGFX and eliminate injection ATI and Intel will have no screen until you start the driver of 6670.

My case is a single monitor with a DVI input, a simple cable in "Y" give me two outputs to one monitor connection, it seems a pretty good temporary solution, because Mieze for showing the way, the rest will be testing, sorry for my English

 

Guys, trust me, I tested hundreds of such tricks. I spend hours trying to do something.

Your builtin graphics from i5-2400s is not the same as my i3-2120. See the difference.

I can't make desktop from the graphics. Using different settings and platform-id I got no more then OSX freeze. I never saw the desktop from this connector.

I can start my Radeon in full working mode by two ways: with and without injections by Clover. Both works good, with full acceleration, with transparency, with good fps in games. Both ways wakeup is not working.

I am tired to tested new tricks, I need full explanation what is happen.

[kkarun88]

@ Meize

 

My build is HD 4600 + R9 380.

When booting with IGPU as primary, injecting proper azul framebuffer (0300220d) boot into el capitan (with or without display attached), but injecting empty framebuffer (04001204 / 0b001204) gives a kernel panic when display is attached. When no display is attached to onboard connector, it can boot into el capitan [sleep works]. All injections are through DSDT. Overcoming this kernel panic will be a feasible solution to TONGA users. Any idea ??

 

 

IOJones output.zip

[surfcaster]

I am tired to tested new tricks, I need full explanation what is happen.

Sorry, you search the bottom of the question, my knowledge do not reach there, the solution provided by mieze is good and at least is teaching the way to go, I'm really sorry not to be able to more, the button reset (due to black screens) my machine has already claimed the life of a HDD and has left the power supply in poor condition, as I say but I can not because I do not, I read and try to learn but sometimes I do not get it

[Mieze]

@ Meize

 

My build is HD 4600 + R9 380.

When booting with IGPU as primary, injecting proper azul framebuffer (0300220d) boot into el capitan (with or without display attached), but injecting empty framebuffer (04001204 / 0b001204) gives a kernel panic when display is attached. When no display is attached to onboard connector, it can boot into el capitan [sleep works]. All injections are through DSDT. Overcoming this kernel panic will be a feasible solution to TONGA users. Any idea ??

 

 

I experienced a similar issue with my test machine too (Asrock kH97M Pro4). With a graphics card installed I got an immediate kernel panic on boot when I used Clover to inject a DSDT although the DSDT was working perfectly without graphics card. Even injecting the vanilla DSDT I extracted with Clover some time before resulted in the kernel panic. Unfortunately I haven't found time for further investigation of the problem, I managed to find a temporary workaround. Using Clover's ability to patch the DSDT and inject graphics properties works fine for me. You might give it a try too.

 

By the way, the platform-id should match the system definition, i.e. 04001204 is for iMac14,2 and 0b001204 is for iMac15,1. As a thumb rule it's best to stay as close as possible to the configuration of the Mac model you are using as the system definition.

 

The real cause of your problem may be located in the IGPU related code of the DSDT. I will take a closer look at it as soon as I find some time.

 

I also can't use a trick with IGPU because it is not working for me: Intel HD2000 without any chance.

The 2011 iMacs have CPUs with HD2000 graphics in combination with a Radeon 6xxx GPU and they are able to wakeup. I'm quite confident that this could be achieved with your hardware too provided you try to mimic the configuration of iMac12,x as close as possible (platform-id for IGPU and framebuffer personality of Radeon GPU). 

I need full explanation what is happen and why Radeon cards on hackintoshes will not wake up. Having such explanation I can make a patch or even a new kext....

I'm afraid but a full explanation is something which probably won't be found in the near future because the graphics subsystem is far too complex for complete reverse engineering with the resources we have. Some of the changes have been obviously made with intention, like making the IGPU the primary one, while others could be just bugs, like the DP issues with some framebuffer personalities.

 

Mieze

[toleda]

By the way, the platform-id should match the system definition, i.e. 04001204 is for iMac14,2 and 0b001204 is for iMac15,1.

Clarification, iMacs 14,2 and 15,1 use AMD graphics only; no HD4600 graphics, no switching.

The noted platform-ids have no display connectors (will not boot); only useful to enable AirPlay Mirroring with discreet graphics as primary display.

 

System definition effects CPU power management; suggest matching system definition to CPU.

[Mieze]

Clarification, iMacs 14,2 and 15,1 use AMD graphics only; no HD4600 graphics, no switching.

The noted platform-ids have no display connectors (will not boot); only useful to enable AirPlay Mirroring with discreet graphics as primary display.

Sorry, but this is not correct. I'm using both system definitions with the mentioned platform-ids and IGPU set as primary in BIOS. Everything is working fine, including sleep/wake. Of course the IGPU is working without active framebuffers (accelerator mode only) which means that the onboard DVI / HDMI / DP connectors are unusable in OS X, which is the specified behavior, but they show a picture until the AMD GPU takes over and shut down after that.

 

Mieze

Edited December 28, 2015 by Mieze

[Slice]

The 2011 iMacs have CPUs with HD2000 graphics in combination with a Radeon 6xxx GPU and they are able to wakeup. I'm quite confident that this could be achieved with your hardware too provided you try to mimic the configuration of iMac12,x as close as possible (platform-id for IGPU and framebuffer personality of Radeon GPU).

I have full reports from 12,2 with Radeon 6770M (deviceID=0x6740) and mimic it as much as possible. Including FakeID=0x67401002

I also have two reports from iMac12,2 before and after sleep. Yes, it wakes.

I see no keys to understand why in my case it is not happen.

 

My HD2000 is produced by i3-2120 never used in iMac.

[kkarun88]

@ Mieze

 

Tried clover patching, but same kernel panic. Kernel panic only when Display is connected to mobo connector during boot. Is there any other method/patch (other than empty framebuffer) to kill/freeze IGPU during boot ??

[Mieze]

@ Mieze

 

Tried clover patching, but same kernel panic. Kernel panic only when Display is connected to mobo connector during boot. Is there any other method/patch (other than empty framebuffer) to kill/freeze IGPU during boot ??

Unfortunately no, at least no other method I know. What setup for the IGPU are you using in BIOS? VRAM, render standby, multi monitor support? Have you tired UEFI boot?

 

Mieze 

[Fl0r!an]

Get MaciASL: http://sourceforge.net/projects/maciasl/

Go to preferences, find ACPI compiler or something like that - it should be set to 4.0 in standard - select 5.1

Choose option "New from ACPI" and save DSDT and all SSDT's in DSL file format. The thing is IOReg might be not enough. I got an idea, that something in PC's DSDT is missing, and lack of that thing prevents AMD Graphics from waking properly. E.g. in MacPro6,1 DSDT, the _WAK method is Notifying the PCI bus and the GPU itself to wake, so it would be very helpful to see how it's working in MacPro with non-stock graphics.

Okay, here we go, sorry for the delay (always the same around christmas: Too much to eat, too little time   )

I made the dump on my MP3,1 in El Capitan 10.11.2 with my R9 280 installed. I booted both with injection as well as without injection but the ACPI tables didn't change from what I can tell, so I've only uploaded on set.

 

I hope it will be helpful!

ACPI Tables.zip

[kkarun88]

@ Mieze

 

VRAM & Render Standby - Auto

IGPU multi-monitor : tried with both Enabled & Disabled. 

Tried both UEFI and legacy clover, enoch also.

 

With all combinations, i am always getting kernel panic and a restart. Don't know whether the problem is with my motherboard. With no display attached to motherboard, it can boot into El capitan using emtpy framebuffer, this scenario can also be achieved using 0300220D, Hence no use to me. Since my mobo has only HDMI, i am left with zero options.

[Mieze]

@ Mieze

 

VRAM & Render Standby - Auto

IGPU multi-monitor : tried with both Enabled & Disabled. 

Tried both UEFI and legacy clover, enoch also.

 

With all combinations, i am always getting kernel panic and a restart. Don't know whether the problem is with my motherboard. With no display attached to motherboard, it can boot into El capitan using emtpy framebuffer, this scenario can also be achieved using 0300220D, Hence no use to me. Since my mobo has only HDMI, i am left with zero options.

I was able to reproduce the issue with the HDMI port on both Asrock boards, with the B85 Killer it happens on every boot, with the H97M Pro4 only sometimes. With DVI everything is fine, no kernel panic on boot when a display is connected. Maybe a DSDT patch might help? Anyway, I will try to find a solution.

 

Mieze

[Bansaku]

For the first time ever, for any OS, I am seriously considering revering back to an earlier version. El Capitan has been nothing but a huge headache for AMD users. Here is my predicament (specs are in my sig).

 

- I have 2x HD7950 (in CFX) with 4 displays connected; Triple monitor setup for the desktop on the first card (I 'used' Eyefinity in Windows) via DVI and 2x miniDP and the second GPU is DVI out to my HDTV that is used exclusively for Kodi.

- In Yosemite I had ZERO issues booting with Enoch from the EFI partition, no garbled graphics before the desktop, sleep/wake worked flawlessly, and most important of all, I did not have to use the HD4000 IGFX for any reason. 

- In El Capitan I in no way can boot WITHOUT enabling the IGFX AND a display connected, AND having it default boot device.

    - If I don't I get black screens

    - If I boot with it enabled but not default I only get one miniDP working and no IGPU despite boot flags (see below)

- In El Capitan using ONLY GraphicsEnabler=y nets me full QI/QE and all GPUs are correctly identified and accelerated however I get no DVI out at all on the AMDs, but works for the HD4000.

    - I have to manually add AtiConfig=Hamachi (default is Futomaki) to enable DVI, AtiPorts=X does nothing!

- In El Capitan I do get sleep! But it wakes after 3 seconds to all displays and USB working, and then the system freezes about 5 seconds later causing me to do a hard restart.

- In El Capitan I get constant random graphics freezes when either watching videos in a window (player or browser) AND playing games windowed (like OpenEmu, Game of Thrones). Even system monitors like iStat and HWMonitor freeze as well as the system clock.

    - To remedy the freeze all I need to do is click to the desktop as the mouse still works.

    - Playing games and movies in 'True' fullscreen (i.e. Diablo III, Kodi) causes no issues, where as 'Windowed' fullscreen still causes random freezes.

 

Apple really dropped the ball on El Capitan. Maybe Tim Cook should stop worrying about SJW and 'diversity' issues and concentrate on their software flaws. Even real Mac users are experiencing USB and graphics issues. Sigh......

[djseban]

Okay, here we go, sorry for the delay (always the same around christmas: Too much to eat, too little time   )

I made the dump on my MP3,1 in El Capitan 10.11.2 with my R9 280 installed. I booted both with injection as well as without injection but the ACPI tables didn't change from what I can tell, so I've only uploaded on set.

 

I hope it will be helpful!

 

Thanks man!

I hate to say it, but your MP3,1 ACPI tables have absolutely no traces of any graphics device. Anyway, I also found iMac15,1 tables somewhere on the forum, and tried few tricks, like Notifying GPU in _WAK method - it did't work. 

 

I made wake working only by enabling IGPU as Primary Graphics Device in BIOS. The solution is quite simple - if you have working AMD Graphics just select IGPU as initial device, in Clover inject it with 0x04120004 ig-platform-id and you are good to go, but you have to do a blind boot. Also, in my R9 390 particular case, when IGP is enabled I get weird screen flickers when moving a window or executing any kind of graphics accelerated action - it isn't as bad as in Yosemite, but it's still there, and it's annoying. I fixed this issue accidentally. The thing is I bought 30ft HDMI cable to connect my TV to PC, today it arrived - booting with both display and TV attached removes the "flickering" issue (both on clone mode or extended displays mode), so if anyone has similar problem here is the "fix". I use patched Baladi framebuffer, because I got all ports working on it - the thing is I have 5 ports defined in VBIOS, when Baladi has 6 - maybe this is a root of a problem.

 

Good thing to know, if you have dual-boot, is possibility of booting Windows by selecting it in "Startup Disk" in SysPrefs. Personally, I set Clover to "Fast" boot mode, and if I want to boot Windows I just go to SysPrefs and reboot OS X into Windows.

 

My best

[Mieze]

@kkarun88: The kernel panic might be related to HMDI audio because it doesn't appear with DVI. Also I managed to boot with HDMI cable connected on an installation without kext patches to enable HDMI audio. Did you install these patches too?

 

Mieze

[kkarun88]

No.. I didn't use Kext patch / DSDT patch for HDMI audio in El Capitan.

     

[Grim Reaper]

I've got an OptiPlex 780 running El Capitan with an AMD/ATI HD 6450 which doesn't wake from sleep (like many other AMD/ATI cards). I also have a usb to HDMI video converter which works but also won't wake from sleep.

 

When I try to wake the OptiPlex, it doesn't, all monitors remain black including the one driven by the usb to HDMI converter.

 

Does anyone have a fix for this problem?

 

I am using Clover boot loader with El Capitan running on a Dell Optiplex 780 with an ATI HD 6450 graphics card that won't wake from sleep.

 

The DSDT I am using is in the Optiplex 780 El Capitan installation guide files, it's the DSDT that isn't edited for use with the HD 5450:

http://www.insanelymac.com/forum/topic/308097-guide-el-capitan-1011x-dell-optiplex-780-760-755-790/

[Fl0r!an]

Hey guys,

 

I just moved on to a new Skylake build (i5-6600K, Asus Z170M-Plus) , so can't stay on Yosemite anymore. As expected, sleep is broken with my R9 280.

 

With ig-platform-id=0x19120000 and iGPU set to primary, it'll wake as long as no screen is connected to Intel Graphics. I guess this is caused by the buggy Skylake drivers.

It's not a headless iGPU though, graphics will work (just with "Skylake artifacts").

 

When booting with PCIE=Primary, audio and USB remains functional for a short period of time (I think this didn't happen on my old P55 build), but the screen stays black.