Rox67er

Small update on my experience with the reduced OSX patches set.

- I used the reduced OSX patches for quite a while seemingly without problems but either due to recent update or certain apps I used and not used before I recently had system hangs where the system would just restart. (Kind of logout user and show the login screen again. I had this when using Affinity Photo / Affinity Designer apps.

- I went back to the full patch set and since then have not had issues anymore. (Also not when using the Affinity apps 😝)

Below my reduced patch list:

Now use the latest from GitHub without the mttr_update and with the Big Sur Aquantia patch.

Yes, now if only apple would release drivers any time soon...

Asrock also list a new beta bios with Re-Size Bar. 1.73

1 hour ago, Ploddles said:

Well UPS say they delivered my RX 6800 XT but there was nobody at home when they say they delivered. Hopefully they just delivered it to a UPS Collection Point but 'forgot' to post the notification through my letterbox.

Well they "lost" my new MP5,1 + Cinema Display back in the days. And you know how big these boxes are... luckily it was insured.

Just tried @Driftwood bios settings and still same result with immediate dark-wake after sleep.

21 hours ago, shutterbug168 said:

So, it still wakes up immediately, and it's caused by DarkWake:

2020-12-14 20:06:16 -0600 Sleep               Entering Sleep state due to 'Software Sleep pid=142': Using AC (Charge:0%)           

2020-12-14 20:06:51 -0600 DarkWake            DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 6 secs    

2020-12-14 20:06:57 -0600 Wake                DarkWake to FullWake from Normal Sleep [CDNVA] : due to UserActivity Assertion Using AC (Charge:0%)        

 

No idea on how to interpret these codes...:-(

Will try driftwood bios but until now I have the exact same Darkwake.

18 hours ago, shutterbug168 said:

No, sleep is still not working... but at least it's NOT stuck in sleep mode if I force it to sleep. Thanks again!

Maybe @Driftwood is willing to share his BIOS settings, he showed that sleep works but apparently it is not related to EFI but must be something in the BIOS settings.

21 hours ago, shutterbug168 said:

@Rox67er, your BIOS settings fixed my shutdown/reboot problem! THANK YOU, THANK YOU!!! Now, I will work on getting the TR TB3 card to work... Do you happen to know any Asrock mobo users using the TR TB3 card with a working SSDT for our mobo?  Many thanks again!!!

Does sleep now also work for you or still no sleep???

I have no info on TR TB3 card, sorry. 

18 hours ago, shutterbug168 said:

@Rox67erYour EFI boots up fine... but my machine still has the same shutdown-restart problem.  Since no Asrock users are experiencing this problem, I am thinking I may be on a roll of getting another defective mobo (after my Designare mobo died)🙄 Do you mind sharing your BIOS (before I harass Asrock support) -- you can save your settings to a file under Tweaker (and the file will be saved to the EFI drive of your USB or SSD drive).  Thanks again!

Attached my current BIOS settings, check your fans settings as I use a silent profile for my AIO water cooler. The rest is basic I think XMP 2.0 enabled and PBO level 3 enabled.

BIOS version = 1.7

ASROCK_BIOS_SILENT_XMP_BAREMETAL.BIN.zip

@shutterbug168 attached my latest EFI with my best config so far for the AsRock TRX40 creator. With USB patching thanks to @Rocket88 I don't have the 99 error anymore and everything works except for sleep. (and sidecar)  CSM disabled, Above 4G enabled. Make sure to add your personal platforminfo. 

EFI.zip

14 hours ago, shutterbug168 said:

@Rox67er, thanks for sharing the link and PCIe power option -- I have set to disabled... but still not shutting down properly.  I plan to install @Driftwood's EFI tomorrow.  Until I try the EFI, I always have CSM set to "enabled" in order to boot up my AMD hack.  Is it the same in your case? Will report back tomorrow...

CSM disabled, Above 4G enabled.

1 hour ago, shutterbug168 said:

Thanks again @DriftwoodGreat to know! It will be my weekend fun project😉

Love to hear if it works for you, I still don't get sleep to work. 🥵

@shutterbug168 The BIOS setting for wake on lan is here: It is called PCIE Devices Power On but in the help text it shows this is also for Wake on Lan.

16 hours ago, Driftwood said:

Big Sur Sleep Video

 

 

Catalina Sleep Working

 

 

 

@Rox67er & other Asrock board owners: My Asrock Creator EFI for both BS & Cat (with two variants of config (ie one with full Algrey patches called 'config1.plist', the other 'Config2.plist' with the only patches we need thanks to *fabiosun testing)  - please update and enter with your own platform info and rename your chosen configs.plist to config.plist). * Please note I swapped out the motherboard's  wifi/BT for a Broadcom 100%  Apple supported version board. However, I dont much use wifi/BT so test away!

Driftwood Cat BS with 2 variant configs EFI.zip 1.83 MB · 8 downloads

I copied your EFI and used the limited patch list, replaced my info. All is well, system boots fine but still no sleep, it wakes up immediately after sleep. It must be somewhere in the BIOS settings.

Could you save your config file to usb and share it?

33 minutes ago, shutterbug168 said:

Thanks again @DriftwoodGreat to know! It will be my weekend fun project😉

This is the one: https://nl.aliexpress.com/item/32464748097.html (The BCM94360NG is a drop in replacement which works OOB with MacOS)

On 12/3/2020 at 12:46 AM, shutterbug168 said:

@Rox67er, how did you resolve the shutdown and immediately restarting problem with the Asrock MB since there's no separate "Wake on Lan" option to turn off? Thanks!

I don't have that issue at all. Will look into my bios settings after reboot, I think there is a wake on lan setting somewhere. (Wake on PCI which is also Lan?)

Some more digging on the sleep - immediate wake issue, I noticed this in verbose boot (RT.GV wake failure, fourth line from top):

At the end of the boot stage I see a lot of getCPUstates - unexpected data for C1 / C2:

2 hours ago, fabiosun said:

 

sorry I do not understand well this:

mind though that 4 addresses in your IOMMU will change based on 4G ON / OFF.

About sleep.are you .in Catalina?

 

Big Sur solved for many of us this problem

Maybe you can test Darkwake boot arg in Catalina...like darkwake=8..I have not tested it

 

No my mistake... should be MMIO whitelist...  well almost the same letters. 😝

I tried dark wake = 0 but not 8, will give it a try. I only use Big Sur now.

Just tried darkwake=8 and still same behaviour

11 hours ago, fabiosun said:

@Rocket88 thank you for your EFI and Description

I would like to say that ABOVE 4G on is the same of adding on boot args npci=0x2000 (and this should not be a "my opinion" but a fact

So if Above 4G off AND npci=0x2000 work 

should also work Above 4G on AND nothing on config boot arg

 

In my MSI I can boot with or without 4G and this with my previous Nvidia >GPU and also now with Vega 64

 

@Rox67er

I see nothing bad in your config.plist and If you are in Big Sur sleep/wake should work

In my case the same set of SSDT and config have different behaviour in Catalina or Big Sur OSX

In Catalina after sleep I have only a dual black screen for my display and all system fans on

and system does not wake, I have to reset

With Big Sur sleep/wake work fine with 0 problems at all

this condition happens also without using any ssdt, so for me it is independent this fact (use them or not I mean)

 

Thanks, can confirm that Above 4G "ON" and nothing on boot arg works fine (this is the config I use, mind though that 4 addresses in your MMIO Whitelist will change based on 4G ON / OFF.

@fabiosun sleep just doesn't work, it always "immediately" like 1-2 seconds after sleep wakes up again. (Dark-wake) and the after touching the keyboard, full wake and screen turns on again.

I see this logging consistently in "pmset -g log":

2020-12-02 01:20:20 +0100 DarkWake            	DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 3 secs    

Thanks @Rocket88 I took your USB settings on board and finally got rid of the 99 error. Now rebooted like 10 times with XMP 2.0 enabled and no 99 error.

11 hours ago, Rocket88 said:

 

My Asrock TRX Creator has been running great once I fixed the MMIO Whitelist and configured the USB ports correctly. Before I configured USB, I would get a code 99 failure on boot approximately 30% of the time. Now USB works on all ports ALL of the time. I never had this much success with my X99 machine. The main problem that I have this this card is I cannot run DisplayPort unless I have CSM enabled. This is a minor inconvenience. Sleep does not work for me, but I think the problem is my Vega 64 video card, which I also use as a "space heater". 

Hi @Rocket88 thanks for sharing your EFI I also had the 99 error and "solved" it by switching off my XMP profile in ram. Since I run it at 2666MHz I had no issue with the 99 error. Also I can't get sleep to work on my Radeon VII.

@fabiosun attached my current EFI (all working except sleep, I removed the audio files to reduce file size) and I have swapped out the wifi board for a MacOS compatible one.

EFI.zip

Did anybody successfully install parallels or VMware Fusion ? As soon as I try to create a virtual machine it errors out.

I guess they rely on intel processor?

Virtualbox does appear to work though but it is slow.

13 hours ago, shutterbug168 said:

@Rox67er, what GPU are you using in your hack?  I have the same sleep problem as you do.  And I am using a Radeon VII card, which is known to have sleeping issues...

Use a Sapphire Radeon VII, I also used it in my old Mac Proc 5,1 where sleep was fine. If I look at the pmset -g log it looks like something CPU related.

11 hours ago, ckrueger99 said:

Are the new versions  Final Cut Pro 10.5, Logic Pro 10.6 and Compressor 4.5 working normally on AMD?

FCP 10.5 is running fine.

Giving sleep one more try on Big Sur. At least the system now doesn't lock up but immediately wakes (DarkWake so without display) after I touch the keyboard the display turns on and I get the login prompt.

In: pmset -g log is see these coming back which seems to be the cause:

2020-11-13 11:01:13 +0100 DarkWake            	DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 3 secs    
2020-11-13 21:06:20 +0100 DarkWake            	DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 20 secs   
2020-11-13 21:23:54 +0100 DarkWake            	DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 4 secs    
2020-11-13 21:25:12 +0100 DarkWake            	DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 8 secs    
2020-11-13 21:30:16 +0100 DarkWake            	DarkWake from Normal Sleep [CDN] : due to D0A1 D0A2 D0A3 D0A4 D0A5 D0A6 D0A7 D0B0 D0B1 D0B2 D0B3 D0B4 D0B5 D0B6 D0B7 D1A0 D1A1 D1A2 D1A3 D1A4 D1A5 D1A6 D1A7 D1B0 D1B1 D/ Using AC (Charge:0%) 1 secs    

And when snooping around in my DSDT I find them back:

            Device (D0A0)
            {
                Name (_ADR, 0x00010001)  // _ADR: Address
                Method (_PRW, 0, NotSerialized)  // _PRW: Power Resources for Wake
                {
                    Return (GPRW (0x08, 0x04))
                }

                Method (_PRT, 0, NotSerialized)  // _PRT: PCI Routing Table
                {
                    If (PICM)
                    {
                        Return (AG04) /* \_SB_.AG04 */
                    }

                    Return (PG04) /* \_SB_.PG04 */
                }

                Device (D0A7)
                {
                    Name (_ADR, 0xFFFF)  // _ADR: Address
                }
            }

            Device (D0A1)
            {
                Name (_ADR, 0x00010002)  // _ADR: Address
                Method (_PRW, 0, NotSerialized)  // _PRW: Power Resources for Wake
                {
                    Return (GPRW (0x08, 0x04))
                }

                Method (_PRT, 0, NotSerialized)  // _PRT: PCI Routing Table
                {
                    If (PICM)
                    {
                        Return (AG05) /* \_SB_.AG05 */
                    }

                    Return (PG05) /* \_SB_.PG05 */
                }

                Device (D0A8)
                {
                    Name (_ADR, 0xFFFF)  // _ADR: Address
                }
            }

In total there are 64 of these, looks like something CPU related??? Does anybody know of CPU BIOS setting related to sleep?

I want to thank all of you for your contribution to this forum on TRX40, I just updated my main system running Catalina to Big Sur and it was smooth as a baby. I only needed to point opencore once to the right disc to continue the install, but other than that zero issue. Big thanks to all of you and especially @fabiosun for starting the Proxmox journey which pulled me into this TRX40 adventure.

On 11/8/2020 at 11:12 AM, fabiosun said:

AMD Power Gadget on TRX40 (CPU PM management)

 

 

Wow 375 W ! That is quite aggressive. With PBO set to level 3 I get up to 345 W and 17631 points in CB20. (@2667Mhz memory, I have to retest but it looks like I have better benchmarks on 2667 than on XMP2.0 3200Mhz)

Just confirmed, only changed my memory to XMP2.0 profile default 3200Mhz and I get 17405 pts. (Power ramps to 375 W too by the way.) 

Very strange.... 😳

I'm starting to think the ASRock TRX40 creator doesn't like my G.Skill Ripjaws memory.