aureus

Members
  • Posts

    7
  • Joined

  • Last visited

Everything posted by aureus

  1. Wow, now that’s a coincidence, after working all day to get my more or less custom VM upgraded to Big Sur from Catalina after waiting quite a bit since release I discovered that a new version of Macinabox is out, I guess I could have just used this one. I wonder though, have you done any magic regarding the AMD reset bug, SpaceInvaderOne? As I just watched your video and there was no mention about GPU passthrough issues as far as I could tell. That’s the only issue still left with my custom VM and I wonder if I should still try your Macinabox out. Basically any time my macOS VM restarts / shuts down when I use GPU passthrough I need to reboot the whole server, isn’t that an issue here?
  2. I just wanted to emphasize the OSX-KVM Github repo. After days of experimenting with all the scattered information and tips on this topic with still having issues with my VM (mostly audio dropouts, which made music production impossible, one of the main tasks for my macOS VM), I today decided to basically replace all files and the VM setup with the files in that repo – and now my VM runs butter smooth. So for anyone experiencing performance issues, I strongly advise to check that repo out.
  3. To answer my question myself: I finally succeeded by using the E1000 kext (https://github.com/chris1111/AppleIntelE1000e) with my Intel NIC. I haven’t really done any performance tests yet, but I suspect that with the real NIC in passthrough it should be much better than any virtual NIC. I haven’t completely figured out yet why the IntelMausiEthernet kext is not working yet, but as I read on another forum that the guy who built that kext removed some older chipsets, I guess he removed support for the chipset my NIC is using. Edit: I could verify this assumption. My card seems to be using the 82571 chipset, which is not supported by IntelMausiEthernet anymore. Hopefully the older E1000 kext will prove to be stable, otherwise I might have to upgrade my card.
  4. Although it's been a while, but could you elaborate on that? For some reason I really struggle with the NIC passthrough, although I thought with Intel NICs it shouldn't be much of an issue. macOS doesn't recognize any ethernet devices when I pass the NIC through. I’ve tried passing through only one port or two ports with multifunction on (similar to a GPU passthrough). Have you installed any kexts? I have tried IntelMausiEthernet, I’ll probably remove that and see if it helps...
  5. I rather referred to the original post, I didn’t really read the last ones / pages. But thanks for the confirmation!
  6. In regards to the CPU, are you sure about AVX2 as a necessity to run macOS? I’ve read quite often that AVX is necessary (somewhere else, not on this forum), not the newer version. I’m gonna try this out myself soon and will report back, I just thought I might ask first. Still waiting for my motherboard to run a Xeon E5 on, which only support AVX but not AVX2. It also looks to me in the XML as if you were only passing the AVX instructions set anyway, not AVX2? I have to admit I haven’t checked the original XML yet, but all of the ones I could find in this topic seem to pass AVX only.
  7. Thanks a lot for this awesome tool! It looks like I finally figured out the issues with my server that bugged me pretty much since I’m running it. Not yet sure why the results from one PCIe x8 slot are so much worse than in the other x8 one, but still, at least it finally works in the latter. However, even after "fixing" it this way, I’m still worried about my parity drive (Seagate, I’ve lost so many of them...). For some reason, it is noticeably slower at first, with the rest of the graph looking kinda okay (not exactly linear, but linear-ish). While I can reproduce that the first test is always around 25MBs slower than the second, the rest of the "valleys" and "bumps" fluctuates with each test, that’s why I personally focused on the start of the test, although my other drives look better overall. What do you think, should I be worried? (Oh, by the way, I tested each 5% with this test, I guess the curve necessarily won’t look as linear that way?)