TheDaemonBarber
-
Posts
11 -
Joined
-
Last visited
TheDaemonBarber's Achievements
Junior Member (1/3)
0
Reputation
-
@fabiosun Thanks, that stuff helped... GPU passthrough seems to be working as the guest is taking over the GPU for the boot. Now the problem has morphed to the NVME not working, giving me an ASPT error that it wasn't doing last night. This leads the VM to boot to the UEFI Interactive Shell V2.2 with missing items in the mapping table. Is there anything important I need to do like unpackaging deb files like from Pavo's build?
-
@fabiosun I think I've added my hardware to my signature now. Proxmox is on a SATA SSD, attached to a SATA port on the motherboard. I've attached the various settings files. I know what you mean about getting the basics working first. I got a basic VM working many times, but the moment I start passing stuff through, it breaks. Archive 2.zip
-
@fabiosun It seems on this latest install, the VM just wont start now. It's not switching to the GPU for the Guest VM, it stays on the commandline for the host, and then "Watchdog: BUG: Soft lockup" for various CPU cores slowly starts appearing. Then the whole host reboots. Is this a USB controller passthrough issue, and I need the Patch from Pavo or elsewhere to fix it?
-
Re-install Proxmox troubles... I want to do the new Proxmox re-install as manually as possible, so I'm avoiding Pavo's script he has kindly provided on his 3950X build. 2 Main questions. Are there patches I need, and how do I get them and apply them, without using his script? (I want to know so it's not relying on someone else's automatic system). I can't get the GPU on the Host to 'turn off' on boot, to reveal the 2 lines like it should on a working passthrough setup. Here's what I did so far... Installed Proxmox Enabled "IOMMU with amd_iommu=on" in grub dmesg returns IOMMU groups. Added the vfio lines to /etc/modules Found the various Device ID codes and added them to the VM.CONF GPU, USB etc. As per my previous install that was passing them through. Added the same ID codes to /etc/modprobe.d/vfio.conf Again same as the previous Proxmox install. Added the blacklisted radeon, amdgpu, and intel sound drivers from the graphics cards to the blacklist. Added "options kvm ignore_msrs=1 report_ignored_msrs=0" to KVM.conf Updated grub and initramfs Reboot, and still Proxmox goes to command line, instead of the 2 lines which shows GPU passthrough is working... What am I missing? It seems I followed all the steps in the various guides I can see.
-
Toms Hardware says the best overall AIO cooler is the Cooler Master M360R, definitely a contributing reason I got one. https://www.tomshardware.com/uk/reviews/best-cpu-coolers,4181-2.html That one is for AM4 and all the other smaller chips, but they do make a bigger one for Threadripper: https://www.coolermaster.com/catalog/coolers/cpu-liquid-coolers/masterliquid-ml360-rgb-tr4/
-
Tried a ton of different BIOS settings, including BIOS defaults. Nothing is working, and I'm still getting this error. EB|`B:WFDW] Err(0xE), 0 @ LocHB 71B4903C-14EC-42C4-BDC6-CE1449930E49 With 4G encoding, it turns out that it just speeds up the boot of the VM. With it on, it gets to the Proxmox boot screen, and then the Opencore error in about 10seconds. With Above 4G encoding turned off, the black screen sits for about 4minutes before the Proxmox logo pops up, and then it lets me select the Mac drive to boot, and then the error. I think I'm going to try making a new Proxmox install on a separate drive and try from scratch. So weird.
-
Hmm. Still getting the following error that stops me booting any further. EB|`B:WFDW] Err(0xE), 0 @ LocHB 71B4903C-14EC-42C4-BDC6-CE1449930E49 I've put my backed up, working, baremetal install onto the passed through NVME drive. I replaced the EFI on that drive with the one @fabiosun posted just above. I can start the VM, the GPU seems to pass correctly, and Opencore loads on the Guest monitor. I choose to load the Mac up, and Opencore starts its boot sequence and once again gets stuck on the above error. Now, I found this guy talking about how he turned off "Above 4G encoding" in the BIOS, to get past this error. I tried that, but now starting the VM just results in a black screen. https://amd-osx.com/forum/viewtopic.php?t=10503 I've tried it with 4G encoding on and off a few times. Off means the VM boots to a black screen and I can't see anything. No logs generated in the EFI either. On, and I get the error. I'm not sure if it's relevant, but the VM Status shows 97% of the 32GB RAM I assign to it, when it hits this error. I think there's a good chance I'm suffering from a BIOS issue. Does anyone have any tips for an X570 board, or even an ASROCK one? I'm going to be going through the BIOS with a fine tooth comb soon, so if anyone can save me any time with some tips, I'd be very grateful.
-
@fabiosun I've tried with both a VM created virtual drive as per the basic install on your guide/Nick Sherlock's guide. I've also tried with deleting that and passing an empty and waiting NVME drive, which works passing through to a Windows VM, but both methods are failing. I've essentially followed your guide now, from the beginning to create a new VM container, got VM assistant to download and create the Clover ISO, the Catalina ISO... loaded those as IDE drives... "failed to start Boot0003 "EFI Internal Shell". I'm getting really disheartened with all this, but will push forward. Maybe I need to yet again destroy Proxmox and start from complete scratch.
-
Hey Driftwood. Yeah, I tried that, and a bunch of other related Quirks tweaks I found with similar searches on the topic. None of them helped, but I'm going to try this Clover route now and see if it helps. *Trying everything from scratch again, because the Clover route gave me the same error as Opencore*.
-
EFI is in an EFI folder inside an EFI partition... Opencore gets to a point where it offers to 'install MacOS Catalina', then dies on the above line. Oh, hadn't tried Clover as the Opencore Guide and a few other people said Clover doesn't work with Catalina? If it does, I'll definitely give it a shot. Nothing to lose 😛
-
Long story short, I need some help as no matter what I try, I keep getting a fail to boot opencore on my Proxmox VM here: [EB|`B:WFDW] Err(0xE), 0 @ LocHB 71B4903C-14EC-42C4-BDC6-CE1449930E49 I see online lots of people mention trying to change these various parameters in the Config.sys EnableWriteUnprotector, ProtectMemoryRegions, RebuildAppleMemoryMap, SetupVirtualMap, SyncRuntimePermissions But so far I've tried all the combinations people suggest online. I've tried this with creating an EFI based on the Dortania Opencore desktop guide, as well as a couple of slimmed down EFIs suggested to me for loading a VM opencore machine. They all lock up at the same place above. I'm running Proxmox 6.2, and the VM is set to pass through various USB and 2 GPUs. This VM setup works and boots to windows, and all the devices appear, so the VM is passing things through correctly as far as I can tell. I'm just stuck on Opencore right now. The VM is set to the classic Penryn CPU type, and the ARGS in the VM.conf are as per Fabiosun's original post here. Any help would be gratefully received.