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


Recommended Posts

Hi,

 

For OpenRGB Patch, after I installed it. I still cannot see the b450m mortar titanium motherboard(nct6775) in the OpenRGB Docker container. Can you please help on how should I configure it? Thanks!

Link to comment
38 minutes ago, imKrow said:

For OpenRGB Patch, after I installed it. I still cannot see the b450m mortar titanium motherboard(nct6775) in the OpenRGB Docker container.

Please post your Diagnostics and please also open up a Terminal and issue this command and post the output here:

ls /dev/i2*

 

Have you installed the OpenRGB Docker container just now or is this an old installation?

Link to comment
On 5/7/2022 at 4:44 AM, Trunkton said:

hey @HarryMuscle just tried to load this driver into my unRAID QNAP TS-873A by doing the following:

  • Place qnap-ec-1.0.0-x86_64-1.txz into /boot/extra/
installpkg qnap-ec-1.0.0-x86_64-1.txz
depmod --quick
modprobe qnap-ec

 

Getting the following after manually moving qnap-ec.ko.xz from the hard-coded directory

/lib/modules/5.10.28-Unraid/extra/

to the only other directory under /lib/modules:

/lib/modules/5.15.37-Unraid/extra/

Output from 'modprobe qnap-ec' command:

modprobe: ERROR: could not insert 'qnap_ec': Exec format error

 

Casual searches about this kind of error say that the binary must be compiled against the current kernel in order to get past that error. I'm sorry to bother you @ich777, is there any chance to compile this for latest RC7 version of unRAID?

 

Additional information shown during pkgremove which is where I got the idea to move that binary.

  --> Deleting /lib/modules/5.10.28-Unraid/extra/qnap-ec.ko.xz
  --> Deleting /usr/lib/libuLinux_hal.so
  --> Deleting /usr/sbin/qnap-ec
  --> Deleting empty directory /lib/modules/5.10.28-Unraid/extra/
  --> Deleting empty directory /lib/modules/5.10.28-Unraid/

 

 

Hi, sorry for the late reply.  I've moved away from using Unraid, but ich777 and I have connected and he's currently providing this driver as an actual plugin for Unraid so it should be available going forwards for the various versions of Unraid.

 

Thanks,

Harry

  • Like 1
Link to comment
On 5/30/2022 at 5:16 PM, ich777 said:

Please post your Diagnostics and please also open up a Terminal and issue this command and post the output here:

ls /dev/i2*

 

Have you installed the OpenRGB Docker container just now or is this an old installation?

Hi, 

 

Sorry for the late update. I just installed the OpenRGB Docker Container. 

The output is

/dev/i2c-0  /dev/i2c-1

and can you please let me know which part of the Diagnostics do you need? The system.log is here:

 

Jun  6 10:37:12 Nas nginx: 2022/06/06 10:37:12 [error] 9445#9445: *7265049 open() "/usr/local/emhttp/plugins/openrgb-patch/images/openrgb-patch.png" failed (2: No such file or directory) while sending to client, client: , server: , request: "GET /plugins/openrgb-patch/images/openrgb-patch.png HTTP/1.1", host: "", referrer: "/Plugins"
Jun  6 10:38:59 Nas emhttpd: cmd: /usr/local/emhttp/plugins/community.applications/scripts/pluginInstall.php install https://github.com/ich777/unraid-openrgb-patch/raw/master/openrgb-patch.plg
Jun  6 10:39:00 Nas root: plugin: running: anonymous
Jun  6 10:39:00 Nas root: plugin: creating: /usr/local/emhttp/plugins/openrgb-patch/README.md - from INLINE content
Jun  6 10:39:00 Nas root: plugin: running: anonymous
Jun  6 10:39:05 Nas kernel: i2c_dev: i2c /dev entries driver
Jun  6 10:39:05 Nas kernel: piix4_smbus 0000:00:14.0: SMBus Host Controller at 0xb00, revision 0
Jun  6 10:39:05 Nas kernel: piix4_smbus 0000:00:14.0: Using register 0x02 for SMBus port selection
Jun  6 10:39:05 Nas kernel: piix4_smbus 0000:00:14.0: Auxiliary SMBus Host Controller at 0xb20
Jun  6 10:39:05 Nas kernel: piix4_smbus 0000:00:14.0: SMBus region 0xb20 already in use!
Jun  6 10:44:07 Nas kernel: docker0: port 2(veth8db9160) entered blocking state
Jun  6 10:44:07 Nas kernel: docker0: port 2(veth8db9160) entered disabled state
Jun  6 10:44:07 Nas kernel: device veth8db9160 entered promiscuous mode
Jun  6 10:44:07 Nas kernel: eth0: renamed from veth0668b8c
Jun  6 10:44:07 Nas kernel: IPv6: ADDRCONF(NETDEV_CHANGE): veth8db9160: link becomes ready
Jun  6 10:44:07 Nas kernel: docker0: port 2(veth8db9160) entered blocking state
Jun  6 10:44:07 Nas kernel: docker0: port 2(veth8db9160) entered forwarding state
Jun  6 10:44:09 Nas avahi-daemon[14531]: Joining mDNS multicast group on interface veth8db9160.IPv6 with address .
Jun  6 10:44:09 Nas avahi-daemon[14531]: New relevant interface veth8db9160.IPv6 for mDNS.
Jun  6 10:44:09 Nas avahi-daemon[14531]: Registering new address record for  on veth8db9160.*.

 

Link to comment
4 hours ago, imKrow said:

and can you please let me know which part of the Diagnostics do you need? The system.log is here:

Please post the full Diagnostics or send them over via PM if you feel more comfortable doing it that way.

 

How does you template for OpenRGB look like, maybe take a screenshot and post it here too.

Link to comment
Posted (edited)

AMD VENDOR RESET ---have a request

 

Invalid for 5700G apu, ID 0x1002, 0x1638

device-db.h does not contain

https://github.com/gnif/vendor-reset/blob/master/src/device-db.h

 

 

in amdgpu_drv.c apu id

https://github.com/torvalds/linux/blob/master/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c

1872 lines

/* Renoir */

{0x1002, 0x15E7, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU},

{0x1002, 0x1636, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU},

{0x1002, 0x1638, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU},

{0x1002, 0x164C, PCI_ANY_ID, PCI_ANY_ID, 0, 0, CHIP_RENOIR|AMD_IS_APU},

 

Can you add the id of the above apu to device-db.h

Generate a new version in the plugin or send it to me to test on unraid。

 

thank you very much!!!

 

unraid 6.10.0-6.10.2

Edited by wx-Rmt
Link to comment
2 hours ago, wx-Rmt said:

Can you add the id of the above apu to device-db.h

I would recommend that you open up a GitHub PR or at least an Issue and that they include it, since I pull it from their repository.

 

2 hours ago, wx-Rmt said:

Invalid for 5700G apu

Are you sure that this APU is affected by the Reset Bug anyways?

 

Isn't this related: Click

Link to comment

Hi,

 

sorry to interrupt, first of all thank you very much for your effort ich777.

 

I wanted to ask if someone could help me. I installed unRaid on a TS453D. I thought the Plugin QNAP-EC should enable Fan Control because the Drives are overheating. After installation something gets recognized, but I don't know what. Dynamix system Temp states fan 3 and 33 through 36, but Dynamix system fan states fan 1 and fan 33. If i try to detect a fan in Dynamix system fan, I get the message:

 

<br /><b>Warning</b>:  file_get_contents(/sys/devices/platform/qnap-ec/hwmon/hwmon2/pwm1_enable): failed to open stream: No such file or directory in <b>/usr/local/emhttp/plugins/dynamix.system.autofan/include/SystemFan.php</b> on line <b>40</b><br /><br /><b>Warning</b>:  file_put_contents(/sys/devices/platform/qnap-ec/hwmon/hwmon2/pwm1_enable): failed to open stream: Permission denied in <b>/usr/local/emhttp/plugins/dynamix.system.autofan/include/SystemFan.php</b> on line <b>42</b><br /><br /><b>Warning</b>:  file_put_contents(/sys/devices/platform/qnap-ec/hwmon/hwmon2/pwm1_enable): failed to open stream: Permission denied in <b>/usr/local/emhttp/plugins/dynamix.system.autofan/include/SystemFan.php</b> on line <b>50</b><br />/sys/devices/platform/qnap-ec/hwmon/hwmon2/fan1_input

 

on all available fan controllers. I know it isn't the TS453Be, but I thougt the same Hardware would be built in since it is the same series.

I have little to none linux experience, but I am trying to get this thing to work since 30 days. The OS version is 6.10.2, the installation is UEFI, but the previos non-UEFI-installation didn't work either. The only plugins installed are Community applications, Nerd Pack, My Servers, QNAP-EC, NCT6687 Driver, Dynamix system temp and autofan. No data is stored on the NAS.

It would be very nice if someone could help me.

Link to comment
13 hours ago, chris1234 said:

sorry to interrupt, first of all thank you very much for your effort ich777.

Sorry for the late response, completely overlooked your post.. :/

 

13 hours ago, chris1234 said:

I know it isn't the TS453Be, but I thougt the same Hardware would be built in since it is the same series.

It's not that simple from what I understand, QNAP builds different chips for controlling the FAN and temperature readouts into their NAS systems, even if it's the same model series.

 

13 hours ago, chris1234 said:

It would be very nice if someone could help me.

Best would be to create a GitHub issue over here and we can talk about it there, simply mark me with @ich777 and I will be notified.

Stonyx, the maintainer, over on GitHub is very helpful with this and answers really quickly.

 

I also contacted @Trunkton because he actually owns a QNAP NAS.

Link to comment

Dynamix System Autofan doesn't work, but that's on that plugin not the driver. To know if the drivers loaded properly you can check under Settings > System Temp.

My settings are as follows:

Available drivers: k10temp

Processor Temp: k10temp - CPU Temp - (Value)

Mainboard Temp: qnap_ec - MB Temp - (Value)

Array fan speed: qnap_ec - Array Fan - (Value), (I have two checkmarked)

 

To control the fans you would utilize the user scripts plugin then add a script for fan control from something like this.

https://github.com/kmwoley/unRAID-Tools/blob/master/unraid_array_fan.sh

 

Attached is my fan control script. The cron job is set to scan every 5 minutes for changes then apply according to my settings.

5 mins schedule is 'custom': */5 * * * *

 

I have 5 HDD's so the file is crafted around that. You'd of course edit the parameters according to your needs.

unraid_array_fan.sh

  • Like 1
  • Thanks 1
Link to comment
48 minutes ago, agh1701 said:

Should I use the 6687d drivers for 6.10?  and will it support the pwm fan plugin?

Since I don't know what hardware are you using I can't give any advice...

 

I would say, try it out to not install it on 6.10 and you will see if it will work... This is always some kind of lottery...

Link to comment
43 minutes ago, ich777 said:

Since I don't know what hardware are you using I can't give any advice...

 

I would say, try it out to not install it on 6.10 and you will see if it will work... This is always some kind of lottery...

I tried.  Can't get it to work with pwm fan control.  Right now I am running nct6683 module and everything works but pwm fan control as well.  I am running a MSI motherboard with a 6687D chip.  I have system temp plugin running and fan speeds on the main page.

Edited by agh1701
can instead of can't
Link to comment
54 minutes ago, agh1701 said:

Can't get it to work with pwm fan control.

Why not control it from the BIOS?

I really can't help because NCT6687D isn't really supported by NCT6683, you can of course use force=1 but this won't show the real numbers from what I know...

 

Have you yet tried to force your chips ID? like force_id=0xd590 or something like that?

Link to comment
  • 2 weeks later...
4 minutes ago, soulelyhayden said:

My goal is to just be able to cleanly start and stop a Windows 11 VM.

Can you try to boot Unraid with Legacy (CSM) mode instead of UEFI?

 

Please keep in mind that the AMD Vendor Reset Patch is more of a workaround and not a real solution.

I know a few people which actually have success on their VMs and the AMD Vendor Reset Patch but it happens from time to time that the server crashes if they start the VM.

It works sometimes 50 times in a row to start/restart the VM and sometimes after the 3rd boot attempt from the VM it crashes the server.

  • Like 1
Link to comment
On 6/24/2022 at 3:25 AM, ich777 said:

Can you try to boot Unraid with Legacy (CSM) mode instead of UEFI?

 

Thank you for the quick reply! Sorry I could not test until today I had a parity check running.

 

Unfortunately, that change has not seemed to help... I still get the same error in the VM logs:

 

qemu-system-x86_64: vfio: Cannot reset device 0000:0b:00.1, no available reset mechanism

 

Link to comment

Hi there, I'm using the Mellanox Firmware Tools plugin and I'm having some issues getting my ConnectX-3 card out of VPI and in to Ethernet mode. When running the command "mstconfig -d 04:00.0 set LINK_TYPE_P2=2" (same on P1) I get the message after confirming to apply config -E- Failed to set configuration: illegal Phy Type value (shold be 1|2|3).

 

Any suggestions. These are IBM branded versions if that helps. Not sure if they need a specific firmware flashing. Specific model is 00D9552, which I believe is the MCX354A-FCBT. I updated to the same/latest version for both adapters on a Windows machine

 

image.png

 

EDIT: Right, so you need to set PHY_LINK_P1/2 to 1(xaui) 2(XFI) or 3(SGMII) before it will allow you to change the link type. No idea which one of those is appropriate and can't find much about it really. Removed the network and network-rules.cfg files and rebooting to see what happens. It did pick up more NICs but all were showing the same MAC address at first

 

Edited by Shinobu
Link to comment
6 hours ago, Shinobu said:

EDIT: Right, so you need to set PHY_LINK_P1/2 to 1(xaui) 2(XFI) or 3(SGMII) before it will allow you to change the link type. No idea which one of those is appropriate and can't find much about it really. Removed the network and network-rules.cfg files and rebooting to see what happens. It did pick up more NICs but all were showing the same MAC address at first

So everything is now working for you?

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.