Contribute
Register

[SUCCESS] Gigabyte Designare Z390 (Thunderbolt 3) + i7-9700K + AMD RX 580

I don't get it either. For Pro Tools sleep should be turned off according to Avid, and Cubase turns sleep of by itself, just the screen will sleep.
Hot swapping an interface is just weird. Like switching a hardware analog mixer on half way in the recording session.

On the other hand, I will sometimes walk away or get distracted for a while, and if sleep would work properly even with a DAW running, this would lower the electricity bill somewhat.

Cubase/ULN-8/PSI-17m's with a serious project loaded: 270W. My desk with Hack sleeping: 19W.

FWIW, I have sleep working on my 6700K iMac 17,1 and sleep ProTools Ultimate all the time without issue. Avid might recommend against it but it certainly works just fine (with my hardware, YMMV)
 
@CaseySJ Completely off-topic: do you know, whether the Synology E10G18-T1 works (OOB, if possible) with OS X? I've read that it has the AQC107 chip, just not sure, if it sports the AQC107"s", or the normal one. If it works oob, it could be flashed with the Apple FW
Please see @pastrychef's post here (and click the Amazon link):
...and here:
 
@mm2margaret , you found all the right bits, good job. All the changes that need to be made are done in the SSDT-TbtOnPCH file, not the system DSDT. The _GPE hot-plug method you need is indeed _E4C. So all you need to do is change _E2C in the SSDT to _E4C and XE2C to XE4C as per the attached file. Don't mess with your DSDT and remember to have Force Power enabled in the thunderbolt configuration section of your bios. Then use the config.plist ACPI patches I've also attached
@dgsga--what would be the patch for replacing RP21._INI with RP21.XINI?
and I assume that patch and the _GPE patch would work in clover if placed in config.plist/ACPI/DSDT/Patches.

attached is my un-patched DSDT.

thanks!
 

Attachments

  • DSDT.aml
    258.7 KB · Views: 63
Last edited:
Could you please tell me how to clean up so many ICONS after I complete the installation according to the steps?
We can selectively hide volumes as shown below. In Step 3, type in the exact name of the volume to hide (you can get the exact name from Clover Boot Menu).

Screen Shot 2020-05-03 at 5.21.11 AM.png

Also, I installed win10 before I installed the MAC, but I can't get in now. Is there anything wrong?
At Clover Boot Menu press spacebar then use down-arrow key to select Verbose and press spacebar again to turn it on. Then press ESC to return to Clover Boot Menu and boot macOS. What error do you see?
 
Please see @pastrychef's post here (and click the Amazon link):

Funny thing: I actually looked for and in the Aquantia threads (major ones from 10.13.3) and didn't find anything (also on other forums). Thanks a lot, got my hopes up. It would seem the Sonnet cards are not that reliable and the original "Gaming" NICs aquantia sold are not available anymore, so I started looking for alternatives.
 
Several comments/suggestions:
  • I see you're using the beta Intel Bluetooth driver. If so, I would suggest eventually getting a Broadcom-based WiFi/BT card. Doesn't have to be done right away.
  • Do you still have your USB install disk for macOS? If so, try booting from it by pressing F12 during BIOS splash screen to open the BIOS Boot Menu (not Clover Boot Menu) and choose the USB disk. Then boot macOS as usual. Does it work?
  • If not, try resetting CMOS by following the procedure in the Designare manual. This will reset BIOS parameters to their factor-default settings, so configure them again starting with Load Optimized Defaults.
  • Then unlock the MSR 0xE2 register once again.
  • Now try booting macOS.


Hello again, I have the same problem again now but I did the same procedure with no success, I even took out the battery this time and have unlocked the MSR 0xE2. I don't know why it isn't working. Is there a way to boot up from a usb drive with an old EFI-folder or how do I succeed from here?

Thanks again
 

Attachments

  • IMG_1523.jpg
    IMG_1523.jpg
    283.9 KB · Views: 51
Seems like you are already reading the chip successfully @ziggenpuss!

Verify that checksums of test1.bin, test2.bin and test3.bin are identical:
Code:
shasum test*.bin

If they are identical, your clip is attached properly are you are succesfully and consistently reading the chip correctly.
KEEP these testx.bin files, they are your backup of the original firmware!

When three checksums match, then move on to the write command to flash the patched firmware.bin.

If flashing is successfull (message with VERIFIED), make sure to do three read commands again (use different filenames, so you don't overwrite your original backup files!!). Verify that the checksums are identical again using shasum *.bin command. If the checksums of those three reads match with the checksum of the patched firmware .bin file, celebrate! Because then you have succesfully patched your Thunderbolt firmware.

I've listed the commands and output of my succesfull flash in the spoiler below. Hope that makes it easier to follow the flashing procedure and recognize desired outputs.
Rich (BB code):
$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r test-connection-ATX-NoPSU-VCC.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r ReadOriginalROM-01-ATX-PSU-NoVCC-BrightWhite.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r ReadOriginalROM-02-ATX-PSU-NoVCC-BrightWhite.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r ReadOriginalROM-03-ATX-PSU-NoVCC-BrightWhite.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ shasum *.bin
8a07244b1f6ea08643eb144e5dd520e50babef7e  ReadOriginalROM-01-ATX-PSU-NoVCC-BrightWhite.bin
8a07244b1f6ea08643eb144e5dd520e50babef7e  ReadOriginalROM-02-ATX-PSU-NoVCC-BrightWhite.bin
8a07244b1f6ea08643eb144e5dd520e50babef7e  ReadOriginalROM-03-ATX-PSU-NoVCC-BrightWhite.bin
1809647344696d2bdc8a4982d1171bafb1dfd630  tb-firmware-original.bin
edbbe3cbf8e3fa4a9d991e0681f2a5702b248224  tb-firmware-patched.bin
8a07244b1f6ea08643eb144e5dd520e50babef7e  test-connection-ATX-NoPSU-VCC.bin
NOTE: all ReadOriginalROM checksums above are identical

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -w tb-firmware-patched.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading old flash chip contents... done.
Erasing and writing flash chip... Erase/write done.
Verifying flash... VERIFIED.

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r ReadPatchedROM-01.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r ReadPatchedROM-02.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ sudo flashrom -p linux_spi:dev=/dev/spidev0.0,spispeed=4096 -r ReadPatchedROM-03.bin
flashrom  on Linux 4.19.97-v7l+ (armv7l)
flashrom is free software, get the source code at https://flashrom.org

Using clock_gettime for delay loops (clk_id: 1, resolution: 1ns).
Found Winbond flash chip "W25Q80.V" (1024 kB, SPI) on linux_spi.
Reading flash... done.

$ shasum ReadPatched*.bin
edbbe3cbf8e3fa4a9d991e0681f2a5702b248224  ReadPatchedROM-01.bin
edbbe3cbf8e3fa4a9d991e0681f2a5702b248224  ReadPatchedROM-02.bin
edbbe3cbf8e3fa4a9d991e0681f2a5702b248224  ReadPatchedROM-03.bin
NOTE: all ReadPatchedROM checksums above are identical, and match checksum of tb-firmware-patched.bin

Good luck, hope you will be succesfull!
Crumbs this is hard..
IMG_1260.jpg
 
Ouch. Some immediate questions:
  • While flashing modified firmware (with the -w option):
    • Was PSU connected (24-pin motherboard cable) and PSU power switch flipped on (not motherboard power)?
    • Was Vcc (pin 8) disconnected?
  • Did any resistors or other nearby surface mount devices get dislodged from the motherboard?
  • Are you able to read the firmware (with the -r option)?
 
@mm2margaret , you found all the right bits, good job. All the changes that need to be made are done in the SSDT-TbtOnPCH file, not the system DSDT. The _GPE hot-plug method you need is indeed _E4C. So all you need to do is change _E2C in the SSDT to _E4C and XE2C to XE4C as per the attached file. Don't mess with your DSDT and remember to have Force Power enabled in the thunderbolt configuration section of your bios. Then use the config.plist ACPI patches I've also attached

So, well, not so good, actually......

No TB tree, and hot plug is gone.

I used your TB dsl, checked for the parameters you suggested, they seemed corrected (I couldn't find either _E2C or XE2C) and produced an aml and loaded it. Very similar to the first time I tried it. I double checked the my config.plist and both patches are there, and are enabled.

I have GPIO force enabled in my BIOS settings (this is Asus's version of TB Power enabling). Remember that with my old config, cold plug and hot plug worked.

Sadly, no change. After changing my EFI, and loading this TB aml, only Cold plug works, hot plug does not, and no tree in Sysinfo. Again, in PCI it does change the name of TB devices from GC-Titan Ridge to Intel, but no tree.
 

Attachments

  • SSDT-TbtOnPch.aml
    4.1 KB · Views: 77
Back
Top