Alan Sojan Posted April 14, 2023 Author Share Posted April 14, 2023 Will try the config2.plist now.. Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 14, 2023 Author Share Posted April 14, 2023 (edited) 1 hour ago, miliuco said: @Alan Sojan With changes by @deeveedee. I cross my fingers. config2.plist 18.45 kB · 2 downloads config1.plist 18.45 kB · 1 download Both of the config.plist's are making my monitor turn off 😓 Not sure what the problem is. Edited April 14, 2023 by Alan Sojan Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 14, 2023 Share Posted April 14, 2023 (edited) @Alan Sojan Let's take a look at your config.plist. Whenever you ask for help or have a problem, post your config.plist. Not that we don't trust you, but it's easy to make mistakes (trust me, I've made many). EDIT: Also, according to this, Skylake graphics is only officially supported up to MacOS 12 (Monterey). If you're trying to test with Ventura, I would recommend that you start testing with a version of MacOS that officially supports your graphics. We can "spoof" another device-id after you confirm that graphics works with a macOS that officially supports it. This is my suggestion. Others may suggest that you spoof the device-id with Ventura as your first test. Edited April 14, 2023 by deeveedee 2 Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 14, 2023 Author Share Posted April 14, 2023 18 minutes ago, deeveedee said: @Alan Sojan Let's take a look at your config.plist. Whenever you ask for help or have a problem, post your config.plist. Not that we don't trust you, but it's easy to make mistakes (trust me, I've made many). Okay, I understand. This is my config.plist config.plist Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 14, 2023 Share Posted April 14, 2023 @Alan Sojan What version of OpenCore are you using? When debugging a new system, I'd recommend sticking with only RELEASE versions (not Beta / nightly build). Just my preference. 1 Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 14, 2023 Author Share Posted April 14, 2023 8 minutes ago, deeveedee said: @Alan Sojan What version of OpenCore are you using? When debugging a new system, I'd recommend sticking with only RELEASE versions (not Beta / nightly build). Just my preference. I will check now. Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 14, 2023 Share Posted April 14, 2023 @Alan Sojan If you're using OC 0.9.2 nightly build, that's ok. It's just that I find that there are typically enough things to diagnose in a new system without worrying about the stability of a DEVELOPER build. If you're using OC 0.9.2, keep it for now. 1 Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 14, 2023 Author Share Posted April 14, 2023 3 minutes ago, deeveedee said: @Alan Sojan If you're using OC 0.9.2 nightly build, that's ok. It's just that I find that there are typically enough things to diagnose in a new system without worrying about the stability of a DEVELOPER build. If you're using OC 0.9.2, keep it for now. Hi, how do I check the version of OC I'm using? When I open it with Opencore configurator it says I have an old version of OC. Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 14, 2023 Share Posted April 14, 2023 @Alan Sojan I'm not an OCAT user - nothing against it, I just don't use it. If you run ocvalidate in the Open Core Utilities folder (in the version of Open Core that you downloaded), it will report the version. I asked your version, because you're using quirk DisableIoMapperMapping which I think is not yet released (will be in OC 0.9.2 when it is released I think). I will be able to look at this again later today and expect that others will have things to add here, too. What I'll be looking at is the connectors in the frame buffers that you are using. Here's where graphics patching gets tricky: just because you're using a frame buffer for HD530 doesn't mean that your graphics ports (e.g., HDMI) are mapped properly. I looked at the EFI that @Rocky12 posted here and see that the OP isn't using HD530 iGPU. I haven't patched Skylake iGPU before, so I'll be learning with you. 4 Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 14, 2023 Share Posted April 14, 2023 (edited) @Alan Sojan I'm not an OCAT user - nothing against it, I just don't use it. If you run ocvalidate in the Open Core Utilities folder (in the version of Open Core that you downloaded), it will report the version. I asked your version, because you're using quirk DisableIoMapperMapping which I think is not yet released. I will be able to look at this again later today and expect that others will have things to add here, too. What I'll be looking at is the connectors in the frame buffers that you are using. Here's where graphics patching gets tricky: just because you're using a frame buffer for HD530 doesn't mean that your graphics ports (e.g., HDMI) are mapped properly. I looked at the EFI that @Rocky12 posted here and see that the OP isn't using HD530 iGPU. I haven't patched Skylake iGPU before, so I'll be learning with you. EDIT: @Alan Sojan If you look at this, you will see that it defines each framebuffer (the definition associated with each AAPL,ig-platform-id). For the skylake framebuffer in the config.plist that you posted (<02001619>, 0x19160002), the definition is: ID: 19160002, STOLEN: 57 MB, FBMEM: 0 bytes, VRAM: 1536 MB, Flags: 0x00830B02 TOTAL STOLEN: 58 MB, TOTAL CURSOR: 1 MB (1572864 bytes), MAX STOLEN: 172 MB, MAX OVERALL: 173 MB (181940224 bytes) Model name: Intel Iris Graphics 540 Camellia: CamelliaV3 (3), Freq: 1388 Hz, FreqMax: 1388 Hz Mobile: 1, PipeCount: 3, PortCount: 3, FBMemoryCount: 3 [0] busId: 0x00, pipe: 8, type: 0x00000002, flags: 0x00000498 - ConnectorLVDS [1] busId: 0x05, pipe: 9, type: 0x00000400, flags: 0x000003C7 - ConnectorDP [2] busId: 0x04, pipe: 10, type: 0x00000400, flags: 0x000003C7 - ConnectorDP 00000800 02000000 98040000 01050900 00040000 C7030000 02040A00 00040000 C7030000 Note that the framebuffer defines the type of each connector. None of the connectors in the framebuffer that you have chosen are HDMI. You have two options after you choose a framebuffer: 1) add patches that modify the connector type and 2) choose a different framebuffer that more closely matches that of your rig. Slice provided you with some mobile framebuffers here. I only see one framebuffer (AAPL,ig-platform-id) that includes an HDMI port: 0x193B0000 mobile 3 56 MB (Has 1 HDMI port): ID: 193B0000, STOLEN: 34 MB, FBMEM: 21 MB, VRAM: 1536 MB, Flags: 0x00001187 TOTAL STOLEN: 56 MB, TOTAL CURSOR: 1 MB (1572864 bytes), MAX STOLEN: 124 MB, MAX OVERALL: 125 MB (131608576 bytes) Model name: Intel HD Graphics SKL CRB Camellia: CamelliaDisabled (0), Freq: 1388 Hz, FreqMax: 1388 Hz Mobile: 1, PipeCount: 3, PortCount: 3, FBMemoryCount: 3 [0] busId: 0x00, pipe: 8, type: 0x00000002, flags: 0x00000098 - ConnectorLVDS [2] busId: 0x04, pipe: 10, type: 0x00000800, flags: 0x00000187 - ConnectorHDMI [3] busId: 0x06, pipe: 10, type: 0x00000400, flags: 0x00000187 - ConnectorDP 00000800 02000000 98000000 02040A00 00080000 87010000 03060A00 00040000 87010000 Please try AAPL,ig-platform-id = <00003B19> and report back. If this framebuffer doesn't work for you, then we'll need to add more patches to define the connector type. Note that <00003B19> is the framebuffer that worked for Slice as he reported here. EDIT2: FYI: My laptop with Kabylake UHD620 has an HDMI port. I use an HDMI -> VGA adapter without any problems. Edited April 14, 2023 by deeveedee 3 Quote Link to comment Share on other sites More sharing options...
miliuco Posted April 15, 2023 Share Posted April 15, 2023 @deeveedee When you try to help, you do it in depth. This is certainly great information for @Alan Sojan to try to fix the issue he is having with the Intel 530. But I'll go back to the first post in the thread. @Alan Sojan if you can try again the original EFI, the one you used when you started the thread and macOS showed the Desktop fine but with 6 MB in iGPU, my idea is to try this same EFI with the only change to remove -igfxvesa from boot args. It is the only change in a first step. If the bug is not resolved, the second step is to add igfxagdc=0 to boot args (without -igfxvesa, of course). The third step is to test the one that @deeveedee has proposed, AAPL,ig-platform-id = 00003B19. Always without -igfxvesa. 4 Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 15, 2023 Share Posted April 15, 2023 (edited) @miliuco Very good idea. Note that the macOS desktop will show when graphics patching is broken. One technique to getting macOS to boot on a new hack before graphics patches are known is to intentionally use an invalid framebuffer. macOS Desktop will likely display, but without graphics acceleration. Looking forward to the OP's test results. EDIT: Note that the original EFI in Post #1 has no DeviceProperties. I will be amazed if it works. Edited April 15, 2023 by deeveedee 3 Quote Link to comment Share on other sites More sharing options...
miliuco Posted April 15, 2023 Share Posted April 15, 2023 2 hours ago, deeveedee said: @miliuco Note that the original EFI in Post #1 has no DeviceProperties. I will be amazed if it works. You are right, forgotten that. Then, @Alan Sojan must add a framebuffer to his first config.plist. Maybe AAPL,ig-platform-id=00003B19 is a good starting point. Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 15, 2023 Author Share Posted April 15, 2023 1 hour ago, miliuco said: You are right, forgotten that. Then, @Alan Sojan must add a framebuffer to his first config.plist. Maybe AAPL,ig-platform-id=00003B19 is a good starting point. Hi, how would I do this? Is it like this? Quote Link to comment Share on other sites More sharing options...
miliuco Posted April 15, 2023 Share Posted April 15, 2023 (edited) @Alan Sojan Yes, it's so. Trying after this (if it fails) others framebuffers, replacing only the value. E.g. those posted by @Slice and others: 00001619 / 02001619 / 00001E19 / 00002619 / 00002719 / 00001B19. You must add the property framebuffer-patch-enable with value of 01000000 as Data, this key says to WhateverGreen to patch the framebuffer. I don't use configurators, only plist files editor, to avoid errors when OpenCore Configurator doesn't match the OpenCore version. But the result must be like this: Spoiler Edited April 16, 2023 by miliuco 1 Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 16, 2023 Author Share Posted April 16, 2023 (edited) 19 hours ago, miliuco said: @Alan Sojan Yes, it's so. Trying after this (if it fails) others framebuffers, replacing only the value. E.g. those posted by @Slice and others: 00001619 / 02001619 / 00001E19 / 00002619 / 00002719 / 00001B19. You must add the property framebuffer-patch-enable with value of 01000000 as Data, this key says to WhateverGreen to patch the framebuffer. I don't use configurators, only plist files editor, to avoid errors when OpenCore Configurator doesn't match the OpenCore version. But the result must be like this: Reveal hidden contents Hi, I tried all of the framebuffers with the framebuffer-patch-enable but it results in ":Whatever green igfx: @TCCE: failed to route the function istypeconlysystem" Maybe I need to try this as I have no option to select DVMT in BIOS. Edited April 16, 2023 by Alan Sojan Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 16, 2023 Share Posted April 16, 2023 @Alan Sojan Please continue to post your current config.plist each time you report a problem and / or ask for help. I could have sworn that frame buffer-stolenmem and framebuffer-fbmem were in one of the config.plist posted in this thread, so it's going to be hard to keep track of the changes unless you post your config.plist each time. 2 Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 16, 2023 Author Share Posted April 16, 2023 13 minutes ago, deeveedee said: @Alan Sojan Please continue to post your current config.plist each time you report a problem and / or ask for help. I could have sworn that frame buffer-stolenmem and framebuffer-fbmem were in one of the config.plist posted in this thread, so it's going to be hard to keep track of the changes unless you post your config.plist each time. Config.plistThis is the config.plist. It makes my monitor screen turn off before the apple logo comes on Quote Link to comment Share on other sites More sharing options...
eSaF Posted April 16, 2023 Share Posted April 16, 2023 24 minutes ago, Alan Sojan said: Please continue to post your current config.plist each time you report a problem and / or ask for help. So many config.plists tips and tricks has been thrown at this project, it's hard to keep track and trace what was tried before therefore creating a possible case of going around in circles. So as @deeveedee suggested, it would be easier to post the current tried config.plist trial with each post so that failed fixes won't be repeated. Although my offers were unsuccessful, I am watching the progress with interest wishing and hoping the help being offered by members will aid you to overcome the problem. 2 Quote Link to comment Share on other sites More sharing options...
miliuco Posted April 16, 2023 Share Posted April 16, 2023 @Alan Sojan 6 hours ago, Alan Sojan said: Hi, I tried all of the framebuffers with the framebuffer-patch-enable but it results in ":Whatever green igfx: @TCCE: failed to route the function istypeconlysystem" Maybe I need to try this as I have no option to select DVMT in BIOS? Yes, try adding also framebuffer-stolenmem and framebuffer-fbmem. These values are fixed, 00003001 and 00009000. AAPL,ig-platform-id is the property to try different values. 1 Quote Link to comment Share on other sites More sharing options...
Rocky12 Posted April 16, 2023 Share Posted April 16, 2023 These are a few examples of the information you could submit to get clarification. Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 17, 2023 Share Posted April 17, 2023 (edited) @Alan Sojan Please try the attached config.plist. It sets the connector types to HDMI (consistent with @Rocky12's post) and does a few other things. I'm assuming that you are testing this with Monterey (not Ventura), you do not have a discrete graphics card (you are using the Intel iGPU) and your BIOS is configured to use the iGPU. Please report your test results with this config.plist. Thank you. EDIT: Please attach your current config.plist to the post with your test results. EDIT2: If config.plist doesn't work, please try the attached config2.plist (you'll need to rename it to config.plist to use it in your OpenCore EFI). config2.plist includes boot-arg -disablegfxfirmware which we might need until graphics are working. I'm not sure about this boot-arg with Skylake, but I needed it when I was first trying to find a working solution for my CoffeeLake hack. Config.plist.zip Config2.plist.zip Edited April 17, 2023 by deeveedee 2 Quote Link to comment Share on other sites More sharing options...
Alan Sojan Posted April 17, 2023 Author Share Posted April 17, 2023 14 hours ago, deeveedee said: @Alan Sojan Please try the attached config.plist. It sets the connector types to HDMI (consistent with @Rocky12's post) and does a few other things. I'm assuming that you are testing this with Monterey (not Ventura), you do not have a discrete graphics card (you are using the Intel iGPU) and your BIOS is configured to use the iGPU. Please report your test results with this config.plist. Thank you. EDIT: Please attach your current config.plist to the post with your test results. EDIT2: If config.plist doesn't work, please try the attached config2.plist (you'll need to rename it to config.plist to use it in your OpenCore EFI). config2.plist includes boot-arg -disablegfxfirmware which we might need until graphics are working. I'm not sure about this boot-arg with Skylake, but I needed it when I was first trying to find a working solution for my CoffeeLake hack. Config.plist.zip 4.17 kB · 5 downloads Config2.plist.zip 4.18 kB · 1 download Hey, I tried both config.plist files and they both result in my monitor turning off. Here is my config2.plist Config.plist Quote Link to comment Share on other sites More sharing options...
deeveedee Posted April 17, 2023 Share Posted April 17, 2023 (edited) @Alan Sojan You're being patient - the most important thing about framebuffer patching. And I've violated my own "rule" which is "provide guidance, but don't do it for you." I hope you're starting to learn enough about the trial-and-error framebuffer patching process to start making some of your own guesses and experiments. Try the attached config.plist which specifies framebuffer 19120000 and changes con1 port type to HDMI. Here are some things that I'm considering if the attached doesn't work: We may be having an issue with your HDMI->VGA adapter. Do you have a monitor with HDMI input and an HDMI-HDMI cable? Note in my post here that I needed to experiment with BusIds. BusIDs in framebuffer patching are still black magic to me. I'll be honest when I tell you that I got lucky by guessing BusIds that worked. If you don't have an HDMI monitor and HDMI-HDMI cable, we may need to experiment with other port types (instead of HDMI). Looking forward to your test results with the attached config.plist. Config.plist.zip Edited April 17, 2023 by deeveedee 3 Quote Link to comment Share on other sites More sharing options...
Rocky12 Posted April 17, 2023 Share Posted April 17, 2023 (edited) some more information Edited April 17, 2023 by Rocky12 2 Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.