[Support] ich777 - AMD Vendor Reset, CoralTPU, hpsahba,...

[ich777]

20 minutes ago, TechHut said:

No it wasn't, I was actually able to get it working thanks to your reply on another thread. THANK YOU!

You don't have to do anything of this usually...

 

Did something change on your Dashboard or do you now see the IMC utilization?

Usually this options are only needed if you want to enable Intel Low Power encoding but I don't recommend that since it reduces the quality too.

 

Now that I'm on my PC I recognize you, good content that you produce...

[OhAndy]

So i switch cables to the new usb-c and rebooted and i got my  2500mbps.  

Thanks so much.. 😁

[ich777]

2 minutes ago, OhAndy said:

So i switch cables to the new usb-c and rebooted and i got my  2500mbps.  

Thanks so much.. 😁

Did you add the line to the go file and tried it in the first place?

[OhAndy]

No i never add the line to the go file. guess I just got lucky. i had not switched cable to the new nic yet, just thought it would not work if i did not add the line.  was wonder if i add the line would it make my onboard nic work. was nice it even kept IP. strange because i have IPs assigned by mac.  

[TechHut]

2 hours ago, ich777 said:

Did something change on your Dashboard or do you now see the IMC utilization?

 

Nothing really changed it is a fresh install. I think it was actually working all along. Using the commands I mentioned it was make hardware encoding actually start working on Plex. Once the GPU was actually being used I started getting data in the monitors.

 

Thank you for the kind words! I've read a lot of threads on here the last day. Big thanks for helping so many people out.

[Rorororororororo]

Is there any issue with this plugin and 6.9.2? 

saw this issue: 

root@Server:~# intel_gpu_top 
intel_gpu_top: /lib64/libc.so.6: version `GLIBC_2.33' not found (required by intel_gpu_top)

after uninstalling the intel gpu top plugin, which was marked as 'incompleted installation', i couldn't find it again. 

 

is there an updated replacement for 6.9.2? 

[ich777]

13 minutes ago, Rorororororororo said:

is there an updated replacement for 6.9.2? 

No, please update Unraid to at least 6.10.0

[Rorororororororo]

15 minutes ago, ich777 said:

No, please update Unraid to at least 6.10.0

Got it. guess it's time to roll the dice. 

 

edit: i actually now see the 6.10.0 note for this plugin via Fix Common Problems. 

 

thanks for the quick reply!

Edited April 13, 2023 by Rorororororororo

[ich777]

1 hour ago, Rorororororororo said:

Got it. guess it's time to roll the dice. 

Anyways, I would recommend that you upgrade to the latest version (6.11.5).

[madmaster]

Question Regarding Intel-GPU-Top.  Looking for some assistance or insight into recurring syslog error on my Unraid server.   I'm a Linux and Unraid newbie and have been unable to solve a repeating system log error noted below.

 

My system is an Asrock Rack E3C256d4U MB with Xeon E-2276G CPU.  The MB has an AST2500 BMC and I have successfully (or so I thought) configured the MB and Plex to enable use of the intel iGPU for hardware transcoding.  Worked great.  Starting with version 6.11.1 and now on 6.11.5 I get the following error messages repeating every few seconds / minutes filling up the syslog file:   

 

To load the I915 intel driver, I have nothing in the "go" file and and have Intel_GPU_TOP plugin loaded.  System devices shows both the AST2500 and intel UHD 630 graphics devices recognized and available.  The MB has one onboard VGA output.  I run headless on my home network without a dummy plug.  At some random time after boot; sometimes within a few hours, sometimes much longer; up to 17 days, the above error messages start hitting the syslog file.  They repeat every few seconds to every few minutes.  Once they start they do not stop.  The only way I can get them to temporarily stop is by rebooting the server.  Despite the error messages, the server remains functional.  I can log in and out, use Plex with or without hardware transcoding.  To date, I've been unable to recognize a pattern of what may be causing the error messages or "force" the error to occur.  IAny help or insight to get me pointed in the right direction of what I should investigate, including where to post this issue if this is not the correct place would be appreciated.   Attached is my syslog and system diagnostics files if anyone is willing to take a look.  

 

madservr-syslog-20230502-1546.zip madservr-diagnostics-20230502-1048.zip

[ich777]

2 hours ago, madmaster said:

Worked great.

This can be caused by various things, you can try to disable the built in AST adapter (this is done usually by a physical jumper on the motherboard).

 

You can also try to use a different dummy adapter, reason behind this is that some dummy adapters don't provide a valid EDID.

 

The last thing that we can try is to hide this message altogether from the log when the functionality from your iGPU isn't affected.

[madmaster]

Thanks for the reply.  I'm not sure I'm able to disable the AST2500. The MB manual lists it as the primary graphics controller for the VGA out port on the MB which is the only video out on the MB.  And I use the AST2500 BMC controller functionality to remotely manage the server.  So i don't want to disable it if that means losing the BMC functionality.  I've tried using a dummy VGA adaptor and see no difference, I eventually get the Unexpected DP dual mode adaptor errors.  You imply there is a way to "hide" the messages going to the syslog.  How is this done?  Does this mean the system would be silently creating these recurring

errors but not writing them to the syslog?   What I don't understand is why the server works for days or sometimes even weeks before the error messages start.  And after they do start, why do they repeat every few seconds / minutes?  This seems like a software bug of some sort.  But I'm a LINUX / UNRAID beginner so it easily could be somethings wrong with my config.  If the software warned me once or twice, but moved on it would be easier to ignore.  And despite the error messages as far as I can tell I never experience a loss of functionality.       

[EmilS]

Hey,
first of all, thanks for your great work! I really appreciate it.
My problem regards the plugin "Nuvoton NCT6687 Driver".
The driver does not recognize the fan speeds. I use a ASUS W680 ACE (not IPMI).
There is a known bug with ASUS Mainboards and nct6798d.  https://bugzilla.kernel.org/show_bug.cgi?id=204807

And one approach to solve the problem: https://github.com/asus-wmi-boards-sensors/asus-board-dsdt

Unfortunately, I am not advanced enough to fully understand the solution or write a modified kernel. However, this is a problem for which I would like to have a solution as soon as possible.

Can you help me by maybe extending the plugin for ASUS motherboards?

Thanks in advance

 

[ich777]

26 minutes ago, EmilS said:

The driver does not recognize the fan speeds.

Please post your Diagnostics with the plugin installed.

[EmilS]

On 6/2/2023 at 11:02 AM, ich777 said:

Please post your Diagnostics with the plugin installed.

Hope it helps

unraid-diagnostics-20230606-1612.zip

[ich777]

On 6/2/2023 at 10:35 AM, EmilS said:

nct6798d

On 6/2/2023 at 10:35 AM, EmilS said:

Can you help me by maybe extending the plugin for ASUS motherboards?

I've took a quick look at this but I can't extend the plugin since this is basically a completely different chip.

 

However I've looked at the linked repository and your board listed as upstreamed (this means basically that a patch was sent to the Kernel team to be implemented directly into the Kernel).

 

Have you yet tried to add:

acpi_enforce_resources=lax

to your syslinux.conf? I'm not a huge fan from this option but at least you can try.

[EmilS]

On 6/7/2023 at 11:06 AM, ich777 said:

I've took a quick look at this but I can't extend the plugin since this is basically a completely different chip.

However I've looked at the linked repository and your board listed as upstreamed (this means basically that a patch was sent to the Kernel team to be implemented directly into the Kernel).

 

I hope this will happen soon.
As I said, I'm not good enough yet to build a custom kernel.

 

Quote

Have you yet tried to add:

acpi_enforce_resources=lax

to your syslinux.conf? I'm not a huge fan from this option but at least you can try.

 

I have now tried it and it works. I happened to find a post that you already commented on as well.

NCT6775 & Dynamix System Temperature + Dynamix Auto Fan Control Support

Can you explain why you do not like this way?

You mentioned possible hardware damage.

 

Thank you :)

 

 

[ich777]

8 hours ago, EmilS said:

Can you explain why you do not like this way?

This is a pretty controversial thing, please make your own opinion on this by searching something like: "acpi_enforce_resources=lax hardware risks" on your favorites search engine.

 

Hope that helps too.

[renedis]

for the NCT6775 (or any other chip like the NCT6798d variants) it's known that the implementation with Asus is done via their own WMI sensor driver. Since kernel 5.15+ there are more and more boards added to mainline. But since 6.3.3 (yes, this late!) there are 150+ bulk boards added to mainline.

 

Bug kernel link:

https://bugzilla.kernel.org/show_bug.cgi?id=204807

 

I'm one of the unlucky persons to that has a NCT6798d chip and would like to see with added to for Unraid.

- I've already tried to check if a custom kernel update would be possible to 6.3.3.. but no.

- I've also tried looking in to editing the "Nuvoton NCT6687 Driver" plugin, but can't find the solution in there or "how to".

- Using the "acpi_enforce_resources=lax" is not a solution in any way. It also brings security risks!

Edited June 18, 2023 by renedis

[ich777]

1 hour ago, renedis said:

- I've already tried to check if a custom kernel update would be possible to 6.3.3.. but no.

Yes it is of course possible, but I would recommend that you wait a bit longer since the next RC release should be based on Kernel 6.3+

 

1 hour ago, renedis said:

- I've also tried looking in to editing the "Nuvoton NCT6687 Driver" plugin, but can't find the solution in there or "how to".

What are you trying to edit? That's not possible since the plugin downloads precompiled packages for every Unraid release.

 

1 hour ago, renedis said:

- Using the "acpi_enforce_resources=lax" is not a solution in any way. It also brings security risks!

I see it the same way, but have you yet tried it, on Unraid nothing should harm your hardware.

[PsychoRS]

On 9/17/2022 at 10:29 AM, Frank Crawford said:

Folks,

I'm currently maintaining the it87 driver that is now being pulled into your systems, and the current version supports the following chipsets: IT8603E, IT8606E, IT8607E, IT8613E, IT8620E, IT8622E, IT8623E, IT8625E, IT8628E, IT8528E, IT8655E, IT8665E, IT8686E, IT8688E, IT8689E, IT8695E, IT8705F, IT8712F, IT8716F, IT8718F, IT8720F, IT8721F, IT8726F, IT8728F, IT8732F, IT8736F, IT8738E, IT8758E, IT8771E, IT8772E, IT8781F, IT8782F, IT8783E/F, IT8786E, IT8790E, IT8792E, Sis950.

 

If you are have any of these chipsets, then you don't need to specify the "force_id" option any further.  If you do find you need it please contact me and I'll look at adding any missing chipsets, but it will depend on what details I can find.

 

Unfortunately, for some motherboards, which have multiple chipsets on board, you will need to add the option "ignore_resource_conflict=1" option, in place of the option "acpi_enforce_resources=lax" during the modprobe command, e.g.

for similar reasons, although it only affects the it87 module not others.

 

Good luck with the use of this module that various people have been improving over the years.

Hi Frank,

 

I'm using your driver on a Gigabyte B560M DS3H. It's working fine detecting and monitoring the pwm/fans, but I can't control PWM outputs (nor using pwmcontrol, it detects pwm/fans but always fire a "no correlation" issue when testing the PWM function) or manually, echoing i.e. 0 to the corresponding /sys/class/hwmon/hwmon4/pwmX.

 

Any idea of what can be happening? In short, your drivers works fine for monitoring the pwm/fans but don't let me control them manually.

 

Thanks.

 

BTW, I needed to use ignore_resource_conflict=1 option.

[ich777]

@PsychoRS sorry for the late response, I would recommend that you create a issue over here if you want to get in touch with @Frank Crawford: Click

[PsychoRS]

20 hours ago, ich777 said:

@PsychoRS sorry for the late response, I would recommend that you create a issue over here if you want to get in touch with @Frank Crawford: Click

Thanks, done it.

 

@ich777, sorry for this but... may I ask you for an update of the IT87 plugin synced with last commits on @Frank Crawford github?

Edited June 27, 2023 by PsychoRS

[ich777]

18 minutes ago, PsychoRS said:

Thanks, done it.

What version of Unraid are you running?

A new package is built every time a new Unraid version is released where the Kernel has been updated.

[PsychoRS]

Just now, ich777 said:

What version of Unraid are you running?

A new package is built every time a new Unraid version is released where the Kernel has been updated.

6.12.1, latest.

 

Sorry then, I just seen the date/version of your plugin (2022.11.2) and noticed some commits later on on @Frank Crawford github.