[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

@ich777 I appreciate your help.

So what are the steps I actually need to perform?
The issue is still happening on 6.12.1. I removed everything I entered previously (I think) and then I went back through the steps you mentioned earlier:

libkmod: kmod_config_parse: /etc/modprobe.d/i915.conf line 1: ignoring bad line starting with 'enable_dc=0'

Attached is diagnostics from just after the reboot. I have set up a syslog server and mirrored it to flash; also attached. I don't see any other errors though (like power_well).

sakaar-diagnostics-20230622-2129.zip syslog (1)

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

Quote

It seems that you are trying to applying it in multiple places and hope that one is working which is definitely not the way to do it.

Fair. I'm just trying to follow instructions from the thread.

 

38 minutes ago, ich777 said:

Why are you not on 6.12.1 since it is now stable and try it there?

Sorry, I am mistaken. I am on 6.12.0 (I forgot I updated when it became available).

 

38 minutes ago, ich777 said:

1 hour ago, Craig Dennis said:

libkmod: kmod_config_parse: /etc/modprobe.d/i915.conf line 1: ignoring bad line starting with 'enable_dc=0'

Where are the contents from this file? They should be:

options i915 enable_dc=0

(please note that if you are using the default editor from OSX it will destroy the formatting and Linux can't read it)

 

The contents have only ever shown `enable_dc=0` but the OSX formatting makes sense (as the error never occurred until today when I ssh'd in)

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

Running rc8 and for whatever reason I am unable to make `enable_dc=0` stick.

@ich777 I must be doing something wrong. It resets to -1 every reboot.

When I run through the commands again (remove/rescan etc.) I get the following error:

libkmod: kmod_config_parse: /etc/modprobe.d/i915.conf line 1: ignoring bad line starting with 'enable_dc=0'

I have it applied via `/boot/config/modprobe.d/i915.conf`

 

I have it in the `syslinux.config` in the flash page

 

But whenever I reboot and `cat /sys/module/i915/parameters/enable_dc` I get `-1`

 

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

For what it's worth, when I installed a discrete GPU and still used the iGPU (both enabled in bios - GPU passed through to a VM) the system stopped crashing (was running rc-7).

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

1 hour ago, ich777 said:

You can do that by adding this to your syslinux.config

i915.disable_power_well=1

 

If the options do nothing for you (as you've previously mentioned the option enable_dc=0 in your case), you can always blacklist the module so that Unraid doesn't load it automatically on boot and enabling it in the go file like mentioned in my previous post with the line:

modprobe i915 disable_power_well=1

or:

modprobe i915 disable_power_well=1 enable_dc=0

syslinux,config changes did not have any effect.
I tried both modprobe commands and rebooted between; server still crashes at the 20-30 min mark.

I will try updating to rc-7 next

 

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

31 minutes ago, ich777 said:

Attention: If you have my Intel GPU TOP plugin installed it will also activate the iGPU so I would recommend that you uninstall it first before trying these options.

@ich777 Awesome. Thanks. Maybe a silly question, will I be able to use the Intel GPU TOP plugin after I have changed these settings?

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

@ich777 I followed your guide above (still on rc-6) and the system crashed at the usual time. I will try with other DC values on rc-6 before upgrading to rc-7 (although I might skip it if the fix has a regression as reported above).

Can someone confirm this working on rc-7?

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

14 hours ago, vojtagrec said:

@Craig Dennis Eh sorry, I just noticed you posted before rc7 was released, so my comment is probably irrelevant to your case...

Yeah I was on RC6 but there’s a chance I put the flag in the wrong location (not in modprobe). 
 

If I get a chance I’ll test the correct location, then upgrade and test again. 

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

@menos i915.enable_dc=0 did not work for me. Server just crashed with PiKVM connected and no ghost monitor.

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

8 hours ago, menos said:

Have you tried the i915.enable_dc=0 option?

I wanted to try them one at a time to ensure I know what worked. 
 

With the ghost monitor installed I have 9 hours uptime. I will now try the i915 flag and report back.

[6.12-rc3 crashes in ~half hour after boot (i915 related?)]

I am experiencing this on an 11700K (also running a PiKVM) and 6.12.0-rc6 with the error:

 

WARNING: CPU: 6 PID: 8994 at drivers/gpu/drm/i915/display/intel_display_power_well.c:271 hsw_wait_for_power_well_enable+0xc9/0xd8

 

Hangs every 20-30 mins. Sometimes it's just really slow to respond but eventually completely hangs. System is still on.

I don't actually have a monitor that I can easily use but I have plugged in a ghost display into the DisplayPort that I had from a previous build.

 

sakaar-diagnostics-20230601-2112.zip

[Unraid OS version 6.10.0-rc2 available]

I'm experiencing consistent hangs of certain parts of the dashboard stats and terminal using RC2 but no corresponding log messages. In fact, there are no log messages for the past 45 mins. It seems to happen anywhere from 10-30 mins after a reboot.

 

[[6.10.0 RC1] Rocket Lake S iGPU - Hardware Encoding/Decoding]

Confirmed: Using the linuxserver container (running 1.25.0.5220) on 6.10.0-rc2 and hardware transcoding works (including tone-mapping).