Jump to content
InsanelyMac Forum

Brabbelbla

Members
  • Content count

    50
  • Joined

  • Last visited

About Brabbelbla

  • Rank
    InsanelyMac Protégé
  • Birthday 09/01/1993

Profile Information

  • Gender
    Male
  • Location
    Netherlands

Recent Profile Visitors

1,867 profile views
  1. Brabbelbla

    EFI Variable Store on Aptio V (Haswell-E and up)

    I understood as much, I just wanted to know what those speculations were, so perhaps I could dig a bit myself. Apparently, there are no real speculations at this moment?
  2. Brabbelbla

    EFI Variable Store on Aptio V (Haswell-E and up)

    Being one of the people left without NVRAM under macOS after the Kaby Lake firmware update, I am trying to understand what is being discussed here. Do you guys have an hypothesis about what's the matter? Trying to summarize: you think that the data buffer between DXE and SMM is overwritten/relocated by boot.efi, so that it doesn't function anymore, is that correct?
  3. Good to hear! The L702X is still my daily machine, even after six years it still works and performs well, quite amazing actually. Mobility is the only thing, but well. It is currently still on Yosemite, might try to update to (High) Sierra, but that seems to be quite a hassle. For the Renesas USB3 controller, faking an Intel USB3 controller via FakePCIID does not work?
  4. Okay, after all those years I have finally figured out what this connector patch for the L702X should be to enable the mDP port, with display audio, AFAIK this hasn't been posted earlier. This adds properties to select framebuffer ID 00010000 (mobile Sandy Bridge) and adds display audio properties: <key>Devices</key> <dict> <key>AddProperties</key> <array> <dict> <key>Device</key> <string>IntelGFX</string> <key>Key</key> <string>AAPL,snb-platform-id</string> <key>Value</key> <data>AAABAA==</data> <key>Comment</key> <string>Selects mobile HD3000 framebuffer</string> </dict> <dict> <key>Device</key> <string>IntelGFX</string> <key>Key</key> <string>hda-gfx</string> <key>Value</key> <string>onboard-1</string> <key>Comment</key> <string>hda-gfx=onboard-1 for HDMI audio</string> </dict> <dict> <key>Device</key> <string>HDA</string> <key>Key</key> <string>hda-gfx</string> <key>Value</key> <string>onboard-1</string> <key>Comment</key> <string>hda-gfx=onboard-1 for HDMI audio</string> </dict> </array> </dict> And the framebuffer edit: <key>KernelAndKextPatches</key> <dict> <key>Debug</key> <false/> <key>KextsToPatch</key> <array> <dict> <key>Comment</key> <string>L702X_non3D_connector_patch</string> <key>Find</key> <data>AQIEABAHAAAQBwAABQMAAAIAAAAwAAAAAgUAAAAEAAAHAAAAAwQAAAAEAAAJAAAABAYAAAAEAAAJAAAA</data> <key>Name</key> <string>AppleIntelSNBGraphicsFB</string> <key>Replace</key> <data>AQICABAHAAAQBwAABQMAAAIAAAAwAAAAAwQAAAAEAAAJAAAAAAAAAAEAAABAAAAAAAAAAAEAAABAAAAA</data> </dict> </array> </dict> Hope this still helps someone
  5. I am following with great interest what is happening here, and will certainly go try this out sometime soon, though I am struggling a bit to understand the CPU part. I will read up on the info in this thread but I do have a question; I see almost everyone here going for the simplest vCPU model possible (Penryn for Sierra, right?), but what is the rationale behind this? Less emulation -> better performance? Or just that their CPU is an unsupported (socket 2011?) model to start with? I have a desktop Skylake processor, is there any reason not to use host-passthrough as vCPU model?
  6. Brabbelbla

    NVMe knowledge base

    Sorry for not getting back to this, but I also ditched Mac for quite a while, but was happy to see that with the patches for IONVMeFamily these problems seem to be solved.
  7. Brabbelbla

    ALC1150 - External amplifier and sleep issues

    Update: putting the _DSM part of Toleda's SSDT into the HDAS renamed to HDEF device in the DSDT has fixed the second issue, as well as some distortion on the headphone output.
  8. Brabbelbla

    NVMe Express digging

    So where did the attempts with IONVMeFamily end up? Last thing I read was that it wasn't possible to inject device properties for the NVMe controller per DSDT.
  9. Brabbelbla

    SAS Controllers w/ EFI for Mac OS X / OSx86 Solutions

    Sorry to resurrect, but wanted to report that at the moment at least some and possibly every Areca card (at the very least my card; the ARC-1214-4I) has both a Mac EFI and a UEFI driver.
  10. Brabbelbla

    NVMe experiences

  11. Brabbelbla

    NVMe knowledge base

    Updated.
  12. Brabbelbla

    NVMe experiences

    Looking at it now, this poll is not really useful. People need to know more than chipset series. As I don't see how to remove the poll I made a new thread, which is intended to be more comprehensive. Will try to carry useful info over. @mods: can this thread be closed?
  13. Brabbelbla

    NVMe knowledge base

    EDIT: with the new patches for IONVMeFamily, everything seems to work fine. Hi all, I think it's high time there was a central point for reference on using an NVMe drive under OS X, for new users and also for motherboard buyers. As there is no other place yet, let's start here! Perhaps a wiki page would be a better idea, certainly with respect to the motherboard list that will hopefully appear here. Please correct me if I am wrong anywhere! Basics In principle, it is possible: 1) If your motherboard does not support NVMe booting, you have to start Clover from some other place, and place NVMExpressDxe in drivers64-UEFI so Clover can launch the OS from an NVMe drive.; 2) Fetch the driver from http://www.macvidcards.com/nvme-driver1.html. Extract and place NVMeGeneric.kext in your Clover directory under Kexts/10.11 to have Clover inject it. Note: Installing the driver under System/Library/Extensions or Library/Extensions does NOT work if you want to use any NVMe drive as boot drive. Having it there prevents that, even if you also have it injected by Clover. For data drive usage only it is fine there. 3) For boot drive usage: launch the OS X installer with the injected driver and install OS X to the NVMe drive that should be visible in Disk Utility or anywhere else. Issues For some users this works fine. So far the good news. For others, myself included, it doesn't work fine and there are issues, there being two: 1) Sleep results in delays at wake. After waking there are a couple of lines in the systems logs about aborted requests, and it takes about 15-20 seconds for the system to become responsive again. Harmless? Hypothesis: is an issue only with 100 series boards? Have seen no report from older series having it. 2) Kernel panics at boot/logoff/shutdown/reboot. Potentially more serious and certainly more annoying. From my personal experience: rarely at boot, sometimes at logoff, often at shutdown and most of the time at reboot. Varies with motherboard model. Also with OS X version? Unfortunately, the driver developer, @jimj740 at MacRumors has not responded to these issues for months. Motherboards Brings me to the last section: which boards do work with the NVMe driver and which don't work (well). If your board isn't in the list already or your experience is different, please post. Please include OS X version, motherboard model, motherboard revision (if known, usually printed on the board itself), BIOS version and any issues you encounter. NO ISSUES Gigabyte X58A-UD7 rev. 1 with Lycom DT120 PCIe x4 adapter, Samsung SM951 NVMe, OS X 10.11.3 * Gigabyte Z97X-UD5H rev. 1.2 with Startech M.2 to PCI-E 4 adapter, Samsung 950 Pro, OS X 10.11.4 WAKE DELAY ONLY Gigabyte Z170X-UD5, Samsung 950 Pro, OS X 10.11 * REBOOT ISSUES ASRock Z170 Extreme7+, Samsung 950 Pro, OS X 10.11.5 * ASUS Z170 Pro Gaming rev. 1.04, BIOS 1805, Samsung 950 Pro, OS X 10.11.5 * from earlier poll.
×