high sierra (but even sierra) black screen after wake amd

[arjunchawda]

I've got a z270-ud5, 7700k, Vega 64 build which is experiencing this very same issue. It can only resolve with holding the power button on the case to force a shutdown..

I'm running High Sierra btw..

 

[SirDaniel94]

On clover v4223 there is a option called hibernation fix. Maybe this is helping you out?

 

[xelanaiznac]

please, let me know if you find a fix

 

[carpentryplus25]

Kinda have a strange config here. Your using Radeon Graphics and have onboard iGPU enabled. Its either the one or the other. Radeon Graphics is known to have (Black Screen) issues. Other than that I cant see anything else wrong here.

This is incorrect information see here in bold red text in post 1 the iGPU is necessary for waking from sleep and booting to desktop without black screen and they can be both used together that is in a desktop configuration

With that said I would recommend the folks in this thread with issues to try patching the framebuffer connectors and injecting iGPU and if that doesn't work well I don't have this hardware to test with so I can't be of much help other than I have AMD cards and They all wake from sleep and boot to the desktop without the black screen issue and all have had the iGPU enabled and set as primary in UEFI/Bios and had to patch the frambuffer connectors to get it working properly.

 

[xelanaiznac]

This is incorrect information see here in bold red text in post 1 the iGPU is necessary for waking from sleep and booting to desktop without black screen and they can be both used together that is in a desktop configuration

With that said I would recommend the folks in this thread with issues to try patching the framebuffer connectors and injecting iGPU and if that doesn't work well I don't have this hardware to test with so I can't be of much help other than I have AMD cards and They all wake from sleep and boot to the desktop without the black screen issue and all have had the iGPU enabled and set as primary in UEFI/Bios and had to patch the frambuffer connectors to get it working properly.

thank you for the help.
in my case, igpu is injecting properly.
as I said in first post, with my mobo and my setup I can't set igpu as primary.
because my pc won't event post on bios. that's because I have the thunderbolt card connected. otherwise I have to disconnect the thunderbolt ex3 pci card

 

[carpentryplus25]

thank you for the help.
in my case, igpu is injecting properly.
as I said in first post, with my mobo and my setup I can't set igpu as primary.
because my pc won't event post on bios. that's because I have the thunderbolt card connected. otherwise I have to disconnect the thunderbolt ex3 pci card

Strange I imagine you have CSM enabled in UEFI/Bios. Without iGPU as primary it is of no use as secondary really. Now Personally I have found that with lilu and whatevergreen I needed the AMD graphics card set as primary in UEFI/Bios, and it always uses the Radeon framebuffer (default framebuffer) which in my opinion should only be used during a safe boot as a fallback driver or the system has no issues using it. And from looking at your IOReg it is indeed using the Radeon Framebuffer. If it were me or I had this hardware I would attempt to patch the framebuffer either Japura, Kamarang, or Iriri framebuffers then inject ATI in clover and select one of those frame buffers and remove WhateverGreen, I have found very mixed results with WhateverGreen. It was made to make the process simpler but doesn't provide the needed customizing that patching a framebuffer does.

 

[xelanaiznac]

Strange I imagine you have CSM enabled in UEFI/Bios. Without iGPU as primary it is of no use as secondary really. Now Personally I have found that with lilu and whatevergreen I needed the AMD graphics card set as primary in UEFI/Bios, and it always uses the Radeon framebuffer (default framebuffer) which in my opinion should only be used during a safe boot as a fallback driver or the system has no issues using it. And from looking at your IOReg it is indeed using the Radeon Framebuffer. If it were me or I had this hardware I would attempt to patch the framebuffer either Japura, Kamarang, or Iriri framebuffers then inject ATI in clover and select one of those frame buffers and remove WhateverGreen, I have found very mixed results with WhateverGreen. It was made to make the process simpler but doesn't provide the needed customizing that patching a framebuffer does.

if you read the whatevergreen kext faq, it's written to disable csm.
and it is exactly what I do. I don't know if this is a issue about the and driver or whatelse. but you agree with me that it is very funny: wake works trough pressing power button. wake doesn't work trough using mouse or keyboard

 

[carpentryplus25]

if you read the whatevergreen kext faq, it's written to disable csm.
and it is exactly what I do. I don't know if this is a issue about the and driver or whatelse. but you agree with me that it is very funny: wake works trough pressing power button. wake doesn't work trough using mouse or keyboard

Yeah I know WhateverGreen Devs recommend CSM Disabled and have a full UEFI setup but see that's just the problem some systems need it to be enabled just as you do that's why I like patching the framebuffer connectors instead of using some kinda injector kext. I almost always hate having to use some kinda injector kext. Makes me feel like it's not as vanilla as it is if I can patch it on the fly with the help of clover.

 

[hulambre]

I've got a z270-ud5, 7700k, Vega 64 build which is experiencing this very same issue. It can only resolve with holding the power button on the case to force a shutdown..

I'm running High Sierra btw..

I have the same issue with RX 560 in High Sierra 10.3.2. I have the latest Lilu, Whatevergreen with iGPU disabled.

 

[abds69]

Wake up hangs on my X79 -3930K R9 280X in sierra and High Sierra . As 3930k hasn't got igpu on die, my only option is Pcie as primary What iv've tested :
-injection ATI with corrected framebuffer futomaki = no results , so for myself, framebuffer is a wrong track to follow
-got modified DSDT and SSDT
-radeondeinit on, no result
-Pc starts not completely and refuse to reboot with the reboot button !!! Strange, it seems an incomplete bios initialization.

Same config was working under Maverick, so i think Apple modified something somewhere, and i think in a method in the DSDT.