[KERNEL] 6.8.0-RC5 w/ navi reset v2, vega reset, k10temp Ryzen patches, x570 onboard audio fix


Recommended Posts

9 minutes ago, TheTechnoPilot said:

Hey there bud, bit of a noobie here when it comes to this level.  Running a 3900X on an ROG B450-I board with a Strix Vega 64 and still back on 6.7.2b.  Looks like upgrading to your build is my best bet now having just moved to the Vega 64 and discovering it is not a smooth sailing as I hoped (MacOS VMs so needed to keep to the Radeon camp).  Sorry for asking such a basic question, but perhaps it could also help others, what is my best bet for upgrading to your specific build with these fixes as I've not done such an upgrade before.

Whats your issue, specifically? Does your MacOS VM work, just looking for a vega reset fix? I've had no luck getting my 5700 XT working (at all, not reset related) in Catalina despite it having native bare metal support. AMD Mac OS is it's own can of worms I haven't tackled (yet).

Link to comment
1 minute ago, Skitals said:

Whats your issue, specifically? Does your MacOS VM work, just looking for a vega reset fix? I've had no luck getting my 5700 XT working (at all, not reset related) in Catalina despite it having native bare metal support. AMD Mac OS is it's own can of worms I haven't tackled (yet).

I've not yet, but I'm having general VM issues I need to sort through first before diving deeper into working on getting it working in MacOS. 

In trying to find solutions to this weird issue I stumbled upon the reset fix issue and seeing your work also seems to deal with board audio pass-through issues along with Ryzen sensors and I haven't bothered upgrading from 6.7.2 yet, this seemed like a good thing to look at and reason to upgrade.

 

Right now my MacOS VM is still running High Sierra and want to get that going first ideally on the new card, but need to fix this basic pass-through error first before even going there.

Link to comment
23 minutes ago, TheTechnoPilot said:

I've not yet, but I'm having general VM issues I need to sort through first before diving deeper into working on getting it working in MacOS. 

In trying to find solutions to this weird issue I stumbled upon the reset fix issue and seeing your work also seems to deal with board audio pass-through issues along with Ryzen sensors and I haven't bothered upgrading from 6.7.2 yet, this seemed like a good thing to look at and reason to upgrade.

 

Right now my MacOS VM is still running High Sierra and want to get that going first ideally on the new card, but need to fix this basic pass-through error first before even going there.

I replied in that thread, this kernel won't fix your issue. 

Link to comment
5 hours ago, TheTechnoPilot said:

Hey there bud, bit of a noobie here when it comes to this level.  Running a 3900X on an ROG B450-I board with a Strix Vega 64 and still back on 6.7.2.  Looks like upgrading to your build is my best bet now having just moved to the Vega 64 and discovering it is not a smooth sailing as I hoped (MacOS VMs so needed to keep to the Radeon camp).  Sorry for asking such a basic question, but perhaps it could also help others, what is my best bet for upgrading to your specific build with these fixes as I've not done such an upgrade before.

Install nvidia plugin and choose install stock unraid 

Link to comment
On 1/24/2020 at 3:39 AM, Skitals said:

I'm glad it's working for you. On a couple of the 6.8 Release Candidates limetech did include v1 of the navi reset patch (this kernel has v2 which is improved but still not perfect). My understanding is it was causing issues with some hardware setups so they pulled it. I'm not sure what the rest of your hardware is, but it seems to work great in conjunction with amd x570 boards. The patch, as it's written, is all or nothing and applies to everyone with a navi gpu. Perhaps limetech would consider including the patch again if it was written to be exclusively opt-in with a kernel flag. I don't mind building my own kernel (and sharing), but I can understand why it would be frustrating to be dependent on someone like me to keep this updated. 

Bud, love it! My machine is working as its supposed to now with the navi patch worked in. I have however had a few hangs...where the machine would hang and then the screen would turn black....I would then have to force stop the VM and I even had a few hard reset on the box.

 

I have now received a notice that my version of unraid is not secure and I need to update urgently to the latest version - security reasons.

Would you be able to work this patch into the latest version for us ?

 

tx

Link to comment
  • 2 weeks later...

Anyone using this kernel with a Nvidia GPU VM passthrough? A VM will start once, but if restarting the VM, or starting another which uses the GPU, the VM will start but I will get no signal from the card. I need to restart the server to get any signal again.

 

Was trying this kernel for the audio passthrough on the x570. Was this the issue that caused Limetech to pull the NAVI patch from the 6.8.0-RCs?

 

-JesterEE

Link to comment
  • 2 weeks later...
On 2/22/2020 at 3:17 PM, JesterEE said:

Anyone using this kernel with a Nvidia GPU VM passthrough? A VM will start once, but if restarting the VM, or starting another which uses the GPU, the VM will start but I will get no signal from the card. I need to restart the server to get any signal again.

 

Was trying this kernel for the audio passthrough on the x570. Was this the issue that caused Limetech to pull the NAVI patch from the 6.8.0-RCs?

 

-JesterEE

They pulled the navi patch (v1, Limetech never bundled patch v2) because it didn't work or caused more issues with certain hardware configurations. The navi patch itself should do absolutely nothing if you don't have an applicable card, it simply hard codes a special behavior, or quirk, only for a matching pci device. There are tons of these quirks for different pci devices bundled into the linux kernel. Sorry, though, I haven't tested nvidia gpu passthough.

Edited by Skitals
  • Thanks 1
Link to comment
  • 2 weeks later...
  • 1 month later...

Quick note about this kernel build with regards to the fix for onboard audio passthrough on x570.

 

I have been tracking the Linux kernel git source to see if this issue has been addressed on the trunk, but unfortunately, it has not.  There is also the possibility that AMD would address this issue in firmware with an updated AGESA ... but the sparsely documented notes for the latest to-be-released AGESA 1.0.0.5 firmware say nothing about the on-board audio.  So the if we have any hope of using the on-board audio on our boards in Unraid, a patched Unraid kernel with this patch will be required.  Luckily, we have a Skitals!

 

As a user of the NVIDIA Unraid build, I'm going to be forced to choose between my GPU in docker and my audio in my VM, which is unfortunate.  It looks like I'm in the market for a USB audio DAC. 😒

 

EDIT: Looks like some people asked AMD to comment about the USB and audio with the AGESA 1.0.0.5, and others actually tried it ... no dice!  

Edited by JesterEE
Refs
Link to comment
  • 3 weeks later...

Hello I am new to unraid and just read this excellent post, I own a AMD 3800x and Asrock 570x mobo so think I have spotted some of these issues (mainly trying to vfi onboard sound to win10 vm crashing it and crashing unraid), I think I can avoid the onboard audio (I use an usb external soundcard) and can get a usb controller seperate card to solve any usb pass through issues.

 

Is there a problem with unraid and AMD cpus/X570 with hot cpu temps or is this false temp readings ?

 

 

 

 

Link to comment
On 5/30/2020 at 8:58 AM, HectorG said:

Hello I am new to unraid and just read this excellent post, I own a AMD 3800x and Asrock 570x mobo so think I have spotted some of these issues (mainly trying to vfi onboard sound to win10 vm crashing it and crashing unraid), I think I can avoid the onboard audio (I use an usb external soundcard) and can get a usb controller seperate card to solve any usb pass through issues.

 

Is there a problem with unraid and AMD cpus/X570 with hot cpu temps or is this false temp readings ?

 

 

 

 

What temp readings are you referring to? There are no issues with unraid and hot cpu temps on these boards.

  • Like 1
Link to comment

As an aside, 6.8.0 rc5 with this kernel has been rock solid. I'm at 84 days uptime as I type, I use this machine as my daily driver (win10 gaming vm w/ 5700xt) with 10+ docker containers running (plex server and more). I'm still holding off updating until there is a 6.9 release candidate.

Link to comment
On 6/1/2020 at 12:39 AM, Skitals said:

What temp readings are you referring to? There are no issues with unraid and hot cpu temps on these boards.

 

Oh thanks I must be mistaken, thought I heard of hotter temps with AMD cpus and unraid good to hear this is not the case.

Link to comment
  • 1 month later...

10 days with this kernel. Much much better. Just wanted to say Thank You!
I was about to go clean Windows 10 and remove unRaid entirely.

Still have some small issues when rebooting my W10 VM with 5700XT. Sometimes boot gets stuck. 

Does anyone know if these fixes will be implemented in the upcoming 6.9 release? Otherwise I will definitely hold back on upgrading.  

Link to comment
  • 1 month later...
36 minutes ago, righardt.marais said:

Can we not speak nicely to them to include your fixes so we get updates as they come along.

Surely they can understand our dilemma with the AMD Navi GPU passthrough. 

Thanks

Not likely, the line is usually "Buy an Nvidia card" when it's bought up.

Link to comment
  • 1 month later...
On 5/6/2020 at 8:04 AM, Skitals said:

Still waiting for a 6.9 Release Candidate before I update from 6.8 RC5.

I hope it comes soon, you would be doing us a great service if you're willing to take a look for the next major update. Desperately in need of an x570 audio fix for my Asus x570-E. Overall just saying thanks for the work ground work you've done so far even though I cant use it (6.8.3).

Link to comment
5 minutes ago, rachid596 said:

If you want 6.9beta 30 had patch for Audio and usb passthrough on x570.

Envoyé de mon HD1913 en utilisant Tapatalk
 

Really!? well that brings some hope that it will all be ok :D Ultimately I don't think I can do betas. Have a lot riding on my machine and may not have all the know how to get myself out of a jam. But I will consider it thank you :) I haven't been able to find any good release notes for kernel version 5.8.13 included with beta 30 that would indicate that to me. I was inspired by @JesterEE's post above to do my own hunting as he has the same board and has posted otherwise invaluable information about how he got his all going.

  • Thanks 1
Link to comment
5 hours ago, rachid596 said:

Before beta24 i have to manually patch my kernel for usb and audio passthrough now with New kernel it's OK.
You can install beta30 and if it's not work roll back to 6.8.3

Envoyé de mon HD1913 en utilisant Tapatalk
 

Just be careful on rollback, not sure but I read in the notes that you have to do some things before roll back. 6.9 got some major changes. Check notes before proceeding. Not around the server to check atm

Link to comment
  • 1 year later...

Hi, I'm currently having this issue as well. 

I've got a Gigabyte Aorus Extreme X570 with an AMD 5950X processor.

My IOMMU:

image.thumb.png.d905e6375fff17a52066bac327dbe641.png

 

Groups 31, 32 and 33 need to be passed together.

Groups 36 and 37 are the Graphics card split, because of ACS override being 'Both', which it doesn't need to be and I'll set it back.

I cannot pass Group 41, as the ImationFashDrive is my unRaid flash drive.

Group 42 is the HD audio controller, which shows up in the list of my Windows 10 VM as a second audio option, even if I haven't bound it to the vfio here.

 

Irrespectively, it tells me that it cannot function without Group 39, which "isn't owned". If I pass through Group 39, it crashes unRaid when the VM starts. Logs mention that its waiting on FLR for x amount of time (not sure what that is).

If I try to pass through Group 39, 40 and 42 I get a similar crash.

 

It looks like there was a lot of work done on a custom kernal, as its an ongoing issue. Looking at the above posts, it was working in 6.9 beta.

I'm on 6.9 stable and I don't seem to have a fix for this. It also looks like the kernal can be recompiled with the relevant patches.

 

Is there anybody working on this? I'm no expert, having recompiled a linux kernal once in my lifetime. It would be great if somebody could update this patch or tell me if it will work on 6.9 and eventually 6.10. Even better, if it was incorporated into new builds of unRaid, as I can see upgrades becoming complicated.

 

Are there alternative options:

 

I have one PCIE slot available, which I want to use for a thunderbolt 4 card to eventually pass through an external GPU (I shudder to think that will go easily).

I have a USB 2 header on the board (only one available), which is being used, but I could use a splitter. However, I think it will end up being passed with the other controllers.

I have an empty USB 3 header on the board. Once again, I believe it may end up being passed with Group 31 or 32.

 

I thought about adding an external PCI usb bracket or even internally to take the unRaid flash, but looking at the groupings I'm not convinced that these would not be passed together 31, 32 and 33, which leaves me in the same boat.

 

I've attached my diagnostics as well.

 

There are a lot of forum posts about this issue. This one is almost identical to mine.

 

 

 

nexus-diagnostics-20211227-1426.zip

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.