i915.ko backported driver for intel 10th gen (DS918+ ver 7.0.1 up3)

[FalseMetal]

Reinstalled all files, but still nothing. May be some motherboard problem or wrong bios settings

[blackmanga]

7 hours ago, FalseMetal said:

Reinstalled all files, but still nothing. May be some motherboard problem or wrong bios settings

Sorry, my bad

 

chmod 755 * in the firmware directory.

could you also give the output of the following after you have inserted the modules. So we can see if there is any error message. 
 

dmesg

cat /var/log/messages. (The last 20-25lines)

 

if it still doesn’t work, let’s take it on PM and work through it. 

Edited July 3, 2022 by blackmanga
edit

[FalseMetal]

14 часа назад, blackmanga сказал:

Sorry, my bad

 

chmod 755 * in the firmware directory.

could you also give the output of the following after you have inserted the modules. So we can see if there is any error message. 
 

dmesg

cat /var/log/messages. (The last 20-25lines)

 

if it still doesn’t work, let’s take it on PM and work through it. 

After chmod 755 * in the firmware directory still doesn’t work

dmesg.txt log.txt

[blackmanga]

Did you run the rm_modules script ? I can’t see anything in the messages saying you have unloaded the existing modules (original .ko) files. 
 

[medric]

This is so off topic I apologies in advanced. 

Do you support 

Device ID 0162, 0166    Intel® HD Graphics 4000    Gen7    Ivybridge

 

[flyride]

41 minutes ago, medric said:

Device ID 0162, 0166    Intel® HD Graphics 4000    Gen7    Ivybridge

 

The kernel that supports QuickSync requires Haswell or later. Sorry.

[medric]

On 4/13/2022 at 6:15 AM, blackmanga said:

Sorry buddy. Looks like a lot of the stuff is missing from the 3622 kernel. It’s going to be a painful process to add one by one.
If you are still interested in testing we can take this on direct and see what other modules need to be complied. 

 

I am in the process of building out an 3622 hardware is i9-9900k motherboard Gigabyte C246M WU4 I am free to test with you. Let me know

[pigr8]

On 4/10/2022 at 2:18 PM, blackmanga said:

i915 drivers for various models  and scripts. I have tested them on DS 918+ and can confirm that /dev/dri exists and HW transcoding works perfectly (even HDR tonemapping works with the linuxserver/plex docker)

 

DS918+ DSM 7

https://www.filehosting.org/file/details/3054890/i915.7z

https://www104.zippyshare.com/v/oPItjx82/file.html

https://dailyuploads.net/41tops2ix61g

 

DS920+  DSM 7

https://dailyuploads.net/iiafmtyxq2si

https://www74.zippyshare.com/v/NPbJM27X/file.html

 

DS3622+ DSM 7

https://dailyuploads.net/avd0bpv3ejw0

https://www100.zippyshare.com/v/i1ha0qjo/file.html

 

Attached the scripts as well to remove original modules (rm_modules.sh) in a running DSM system, and then insert the new modules (in_modules.sh). All modules are for kernel 4.4.180+

 

 

If someone would like to compile their own modules, you would need to patch the files in linux.4.4.x source files.

 

The i915.patch file is to be placed in linux.4.4.x folder of the GPL source.

(you can do a dry run to see if there are any errors first: patch -p0 --verbose --dry-run < i915.patch )

 

If everything looks ok, you can run the following command

patch -p0 --verbose -b < i915.patch

 

after this, you just cross compile the modules as usual.

 

in_modules.sh 203 B · 71 downloads rm_modules.sh 183 B · 71 downloads i915.patch 7.13 kB · 69 downloads

hei @blackmanga do you have the ds3622xs+ and ds920+ version of the patched driver somewhere to share? the links on dailyuploads and zippyshare are expired, and only ds921+ can be downloaded just from filehosting.

appreciate!

[blackmanga]

https://mega.nz/folder/rl8xxT7L#ddCwt1mHR1ctxLL0LVKnXA

[pigr8]

5 hours ago, blackmanga said:

https://mega.nz/folder/rl8xxT7L#ddCwt1mHR1ctxLL0LVKnXA

 

thank you!

[D_G_B]

Just for information purposes:

 

The DVA1622 DMS pat file includes the original i915 driver files. They will work for 2nd Gen to 9th gen Intel CPUs, but not for 10th gen CPUs.

 

So, the situation I ran into is: Baremetal, Asus H410m motherboard, Pentium Gold G6400 (Comet Lake) CPU, DVA1622-7.1.0-42661u3, ARPL loader.

1) When the system boots, the original i915 driver files load, but do not get activated because the IGPU is not compatible.

2) Since the DVA1622 uses a geminilake CPU and the DS920+ is also a geminilake platform, I try the patched files for the DS920+. Of course, because the original drivers are loaded, the new drivers fail to load.

3) run the rm_modules.sh script. Original driver files are unloaded.

4) run the in_modules.sh script. Patched modules are loaded. The screen spring to life and loads the local display software and eventually (a minutes or so later) presents a login screen on the local display.

 

So far, so good... Here is where the issue show up...

 

5) copy the required driver files to the /usr/lib/module directory. (The firmware files are already installed ad part of the DVA1622 install)

6) reboot the machine. Machine reboots and loads DSM, but the screen does not change from the ARPL boot screen.

7) SSH into the machine to discover that the original i915 drivers are loaded and inactive.

 

It turns out that, although the patched drivers are installed, they are not used. The boot process loads the original drivers that are embedded in the initial ramdisk in the loader rather than the drivers installed on the DSM on the HD.

 

The Solution... remove the original driver files from the loaders ramdisk:

 

1) boot ARPL using the "configure loader" option. Run the menu.sh script and go to the update menu

2) If you are running ARPL less that 0.3-alpha3, update arpl. (if you need to reboot, do so and restart)

2) update MODULES then exit the menu back to the shell prompt

3) add dummy drivers to the modules archive. (details given below)

4) run menu.sh

5) Build the loader

6) Boot the loader.

 

As the driver files that are embedded in the loader are now empty files, the loader is not able to load the original drivers, and the boot process will load the drivers installed on the HD when udev or eudev starts. The machine will now load and use the patched drivers, and will operate like a real DVA1622.

 

 

To add the dummy drivers into the modules archive, do the following

1) change directory to the modules directory:

      cd /mnt/p3/modules

2) extract the modules from the geminilake archive:

      mkdir tmp;gunzip geminilake-4.4.180.tgz|tar -C ./tmp -xf -

3) create the empty driver files:

      cd ./tmp; touch backlight.ko; touch drm.ko; touch drm_kms_helper.ko; touch fb.ko; touch i2c-algo-bit.ko; touch i915.ko; touch iosf_mbi.ko; touch video.ko

6) create the new modules archive and replace the original archive:

      tar -cf ../drivers.tar *; cd ..; gzip drivers.tar; rm geminilake-4.4.180.tgz; mv drivers.tar.gz geminilake-4.4.180.tgz

7) clean up and return to the ARPL directory:

      rm -rf ./tmp/;  cd /opt/arpl/

 

 

One final thing to mention... When I first created the loader, I used the default LKM (dev)... After the first boot (when all the packages have been downloaded and installed and the DSM image on the HD is stable)... The boot process is excruciatingly slow.... The H410M/G6400 combo in my machine took 289 seconds (almost 5 minutes)... The development LKM does a LOT of logging... an OBSCENE amount of logging, and that significantly affects performance. SO... Once you have set up your machine and are sure that it works... rebuild the loader with the production LKM... It does more modest logging and is significantly fasted... My machine now takes 96 seconds to boot...

 

 

BTW... Big thanks to @pocopico and  @fbelavenuto for all the work on the loaders and @blackmanga for providing the patched driver

 

 

Edited July 21, 2022 by D_G_B

[D_G_B]

Just to add to my previous post...

 

I checked out the repost of the drivers by @blackmanga a few days ago... The ds920 files now have stripped versions available. My previous method of updating the modules archive with dummy files was due to the unstripped drivers being too large to be embedded into the loader. With the stripped drivers now available, they can be embedded into the modules archive.

 

To add the real drivers into the modules archive, do the following

1) before you boot your machine into ARPL, prepare a FAT formatted USB stick with the drivers:

      a) Download the i915_920_7.0_stripped.7z from the link posted by @blackmanga is post #109 of this thread

      b) extract the files to the root of a FAT formatted USB stick

      c) eject the USB stick

2) On the machine you have booted with your ARPL USB stick, change directory to the modules directory:

      cd /mnt/p3/modules

3) extract the modules from the geminilake archive:

      mkdir tmp;gunzip geminilake-4.4.180.tgz|tar -C ./tmp -xf -
4) copy the drivers onto the ARPL machine

      a) Insert the FAT formatted USB stick with the drivers

      b) determine the device identifier for the USB stick FAT partition:

            dmesg|grep -e '  sd';tail -n1

         The line of output will contain the identifier for the usb stick - IE: "[ 2679.003829]   sdc: sdc1" would mean the device identifier is sdc1

      c) mount the USB stick FAT partition:

            mkdir /mnt/drivers; mount /dev/sdc1 /mnt/drivers

         note: replace "sdc1" with the identifier you determined in step b

      d) copy the drivers onto the ARPL USB stick

      cd ./tmp; cp /mnt/drivers/*.ko .

5) create the new modules archive and replace the original archive:

      tar -cf ../drivers.tar *; cd ..; gzip drivers.tar; rm geminilake-4.4.180.tgz; mv drivers.tar.gz geminilake-4.4.180.tgz

7) clean up and return to the ARPL directory:

      rm -rf ./tmp/;  umount /dev/sdc1; rm -rf /mnt/drivers; cd /opt/arpl/

   Note: replace "sdc1" with the identifier you determined in step 4)b

 

Then complete the build steps is the previous post.

 

Note: remove the drivers USB stick before booting your updated loader

 

 

After the loader is updated, it is possible (with an edit to the /opt/arpl/model-configs/DVA1622+.yml file to remove the blanked out update server addresses) to install DSM completely from the xpenology machine... The first boot of the loader will take several minutes, then display a message to 'plug in a keyboard on DVA1622 and press "Enter" to install surveillance station'... Press enter... It will then say "the local installation will take about 6 minutes. Do not shut down your device"... It is not accurate... It will may take more or less than 6 minutes... let it do it's thing and eventually the screen will go blank for a bit and the machine will reboot... the reboot will bring up the DVA1622 splash screen for several minutes and then display the local installation screen... It will take much longer than 6 minutes... leave it be and let it complete the installation... Eventually, the "Welcome to DSM 7.1" screen will appear, and you can complete the DSM installation. It will ask for the device name, username and password, then load the local DSM Surveillance Station desktop.

 

To edit the configuration file to allow for standalone installation:

1) Exit the ARPL menu

2) Edit the model configuration file:

      nano /opt/arpl/model-config/DVA1622+.yml

3) delete all lines (Ctrl+k) in the [synoinfo] section EXCEPT the line that starts with "buzzeroffen"

4) Save the file using (CTRL+x). Press "y" when asked if you want to save the buffer.

5) run ./menu.sh

6) rebuild the loader

 

 

 

[medric]

17 hours ago, D_G_B said:

Just for information purposes:

 

The DVA1622 DMS pat file includes the original i915 driver files. They will work for 2nd Gen to 9th gen Intel CPUs, but not for 10th gen CPUs.

 

So, the situation I ran into is: Baremetal, Asus H410m motherboard, Pentium Gold G6400 (Comet Lake) CPU, DVA1622-7.1.0-42661u3, ARPL loader.

1) When the system boots, the original i915 driver files load, but do not get activated because the IGPU is not compatible.

2) Since the DVA1622 uses a geminilake CPU and the DS920+ is also a geminilake platform, I try the patched files for the DS920+. Of course, because the original drivers are loaded, the new drivers fail to load.

3) run the rm_modules.sh script. Original driver files are unloaded.

4) run the in_modules.sh script. Patched modules are loaded. The screen spring to life and loads the local display software and eventually (a minutes or so later) presents a login screen on the local display.

 

So far, so good... Here is where the issue show up...

 

5) copy the required driver files to the /usr/lib/module directory. (The firmware files are already installed ad part of the DVA1622 install)

6) reboot the machine. Machine reboots and loads DSM, but the screen does not change from the ARPL boot screen.

7) SSH into the machine to discover that the original i915 drivers are loaded and inactive.

 

It turns out that, although the patched drivers are installed, they are not used. The boot process loads the original drivers that are embedded in the initial ramdisk in the loader rather than the drivers installed on the DSM on the HD.

 

The Solution... remove the original driver files from the loaders ramdisk:

 

1) boot ARPL using the "configure loader" option. Run the menu.sh script and go to the update menu

2) If you are running ARPL less that 0.3-alpha3, update arpl. (if you need to reboot, do so and restart)

2) update MODULES then exit the menu back to the shell prompt

3) add dummy drivers to the modules archive. (details given below)

4) run menu.sh

5) Build the loader

6) Boot the loader.

 

As the driver files that are embedded in the loader are now empty files, the loader is not able to load the original drivers, and the boot process will load the drivers installed on the HD when udev or eudev starts. The machine will now load and use the patched drivers, and will operate like a real DVA1622.

 

 

To add the dummy drivers into the modules archive, do the following

1) change directory to the modules directory:

      cd /mnt/p3/modules

2) extract the modules from the geminilake archive:

      mkdir tmp;gunzip geminilake-4.4.180.tgz|tar -C ./tmp -xf -

3) create the empty driver files:

      cd ./tmp; touch backlight.ko; touch drm.ko; touch drm_kms_helper.ko; touch fb.ko; touch i2c-algo-bit.ko; touch i915.ko; touch iosf_mbi.ko; touch video.ko

6) create the new modules archive and replace the original archive:

      tar -cf ../drivers.tar *; cd ..; gzip drivers.tar; rm geminilake-4.4.180.tgz; mv drivers.tar.gz geminilake-4.4.180.tgz

7) clean up and return to the ARPL directory:

      rm -rf ./tmp/;  cd /opt/arpl/

 

 

One final thing to mention... When I first created the loader, I used the default LKM (dev)... After the first boot (when all the packages have been downloaded and installed and the DSM image on the HD is stable)... The boot process is excruciatingly slow.... The H410M/G6400 combo in my machine took 289 seconds (almost 5 minutes)... The development LKM does a LOT of logging... an OBSCENE amount of logging, and that significantly affects performance. SO... Once you have set up your machine and are sure that it works... rebuild the loader with the production LKM... It does more modest logging and is significantly fasted... My machine now takes 96 seconds to boot...

 

 

BTW... Big thanks to @pocopico and  @fbelavenuto for all the work on the loaders and @blackmanga for providing the patched driver

 

 

 

 

Does the DVA3221 DMS pat file includes the original i915 driver files. I am going to build a out my server this weekend cpu i9-9900k which is the9th gen Coffee Lake-S  UHD Graphics 630. I want to know if can build DVA3221 or if I should build something else. I really dont want to run a GPU unless I have to (I only have gtx 650, 1060 and 1080ti) and i want more than 8 cores so I can run VM's any recommendation would be useful this project has been dragging on for 6 months now and I am looking forward to retiring the old DS3615 on its last leg

[wessu]

I'm new to this. I just got Intel 10th gen working and DSM 7.1 and I need to install the driver for 918+. I could not find an easy tutorial to follow. Can someone help please provide clues on how to do that? Much appreciated in advance.

[D_G_B]

3 hours ago, medric said:

 

 

Does the DVA3221 DMS pat file includes the original i915 driver files.

 

No. The DVA3221 pat does NOT include the i915 drivers.

 

The DVA1622 pat has the i915 drivers and does not include the nVidia drivers.

The DVA3221 pat has the nVidia drivers and does NOT include the i915 drivers.

 

@blackmanga has compiled the i915 drivers for the 10gen cpus (and should work for older), and i'm testing it on a DVA3221 install right now. I'll let you know how it goes....

[medric]

16 hours ago, D_G_B said:

 

No. The DVA3221 pat does NOT include the i915 drivers.

 

The DVA1622 pat has the i915 drivers and does not include the nVidia drivers.

The DVA3221 pat has the nVidia drivers and does NOT include the i915 drivers.

 

@blackmanga has compiled the i915 drivers for the 10gen cpus (and should work for older), and i'm testing it on a DVA3221 install right now. I'll let you know how it goes....

 

Thank you sir, That is greatly greatly appreciated!  Looking forward to getting this done before my old 3615 die.. 

 

build hardware 

Gigabyte C246M-WU4

Core i9-9900K

32GB RAM

LSA 9211-8i IT mode

[D_G_B]

22 hours ago, D_G_B said:

@blackmanga has compiled the i915 drivers for the 10gen cpus (and should work for older), and i'm testing it on a DVA3221 install right now. I'll let you know how it goes....

 

Unfortunately, the DVA3221 drivers do not work. They will not work due to missing components in the linux kernel. Checking other forum threads, this seems to be a known and ongoing issue.

[arvin]

my CPU  is  I9 10900T ES ,in PVE 7.2-4. DSM is 7.0.1-42218 update5.

 

the problem is insmod i915.ko

Here below is the dmesg:

2022-07-26T20:24:48+08:00 nas kernel: [  607.926795] Module [i915] is removed. 
2022-07-26T20:24:54+08:00 nas kernel: [  614.011667] ACPI: PCI Interrupt Link [GSIG] enabled at IRQ 22

then the dsm will restart....

how to fix this?

[blackmanga]

How are you passing the iGPU to the virtual machine ? 

Please check for enabling hw transcoding for PVE first before attempting the above process which is pretty straightforward if it’s baremetal installation.

 

[NooL]

Will this work for previous gens also? I have a 9th gen 9900k 3E98 - Is this supported with this mod?  

[wessu]

On 8/6/2022 at 6:29 AM, NooL said:

Will this work for previous gens also? I have a 9th gen 9900k 3E98 - Is this supported with this mod?  

i  think this is only need for 10th gen devices. 9th gen are already supported out of the box.

[pigr8]

@blackmanga after @FalseMetal got in touch with me i tested the i915 backported drivers and i have the same issues that he has, the igpu sais "killed".. i'm confident i got it working in the past but since i have a lot of test bench i cant really remember it this was working or not, damn.

 

z490-i + i3-10100T, esxi, gpu in passthrough, in dsm it's recognized as "0000:0b:00.0 Class 0300: Device 8086:9bc8 (rev 03)", no /dev/dri on a stock dsm 7.1 install.. after copying the firmware

 

root@TestStation:/usr/lib/firmware/i915# ls -al
total 420
drwxr-xr-x 2 root root   4096 Aug 10 08:22 .
drwxr-xr-x 3 root root   4096 Aug  9 20:02 ..
-rwxr-xr-x 1 root root   8380 Oct  7  2021 bxt_dmc_ver1_07.bin
-rwxr-xr-x 1 root root   8800 Oct  7  2021 glk_dmc_ver1_04.bin
-rwxr-xr-x 1 root root   8840 Aug 10 08:21 kbl_dmc_ver1_04.bin
-rwxr-xr-x 1 root root 147776 Aug 10 08:22 kbl_guc_ver9_39.bin
-rwxr-xr-x 1 root root 218688 Aug 10 08:22 kbl_huc_ver02_00_1810.bin

 

and doing the rm_modules and in_modules script /dev/dri appears
 

root@TestStation:/volume1/storage/i915# sh rm_modules.sh
root@TestStation:/volume1/storage/i915# sh in_modules.sh
root@TestStation:/volume1/storage/i915# cd /dev/dri
root@TestStation:/dev/dri# ls
card0  renderD128

 

root@TestStation:/$ tail -f /var/log/messages
2022-08-10T08:18:53+02:00 TestStation synodbudupdate[18770]: update.cpp:299 Fix syno-abuser-blocklist completed and succeed
2022-08-10T08:19:00+02:00 TestStation logger[20372]: version_change hook event: syno-abuser-blocklist installed successfully.
2022-08-10T08:19:02+02:00 TestStation synoindex[20465]: synoindex.cpp:421 Synoindex reindex by synoindex binary, newpath: media
2022-08-10T08:19:03+02:00 TestStation mediaserver.sh[20468]: Re-index triggered, cmd [/usr/syno/bin/synoindex -R media -P MediaIndex]
2022-08-10T08:19:45+02:00 TestStation kernel: [44277.796941] synobios get empty ttyS current
2022-08-10T08:20:46+02:00 TestStation kernel: [44338.596761] synobios get empty ttyS current
2022-08-10T08:21:47+02:00 TestStation kernel: [44399.389364] synobios get empty ttyS current
2022-08-10T08:22:48+02:00 TestStation kernel: [44460.236546] synobios get empty ttyS current
2022-08-10T08:23:48+02:00 TestStation kernel: [44521.074549] synobios get empty ttyS current
2022-08-10T08:24:49+02:00 TestStation kernel: [44581.902709] synobios get empty ttyS current
2022-08-10T08:25:50+02:00 TestStation kernel: [44642.725378] synobios get empty ttyS current
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.664222] Module [i915] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.672351] Module [video] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.681101] Module [backlight] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.690095] Module [iosf_mbi] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.701156] Module [drm_kms_helper] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.710137] Module [drm] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.718138] Module [i2c_algo_bit] is removed.
2022-08-10T08:25:55+02:00 TestStation kernel: [44647.727103] Module [fb] is removed.
2022-08-10T08:26:05+02:00 TestStation kernel: [44657.346787] mtrr: type mismatch for d0000000,10000000 old: write-back new: write-combining
2022-08-10T08:26:05+02:00 TestStation kernel: [44657.348107] Failed to add WC MTRR for [00000000d0000000-00000000dfffffff]; performance may suffer.
2022-08-10T08:26:05+02:00 TestStation kernel: [44657.369102] [drm:gen9_set_dc_state [i915]] *ERROR* DC state mismatch (0x0 -> 0x2)

 

[44647.664222] Module [i915] is removed.
[44647.672351] Module [video] is removed.
[44647.681101] Module [backlight] is removed.
[44647.690095] Module [iosf_mbi] is removed.
[44647.701156] Module [drm_kms_helper] is removed.
[44647.710137] Module [drm] is removed.
[44647.718138] Module [i2c_algo_bit] is removed.
[44647.727103] Module [fb] is removed.
[44657.344563] i915 0000:0b:00.0: enabling device (0000 -> 0003)
[44657.346787] mtrr: type mismatch for d0000000,10000000 old: write-back new: write-combining
[44657.348107] Failed to add WC MTRR for [00000000d0000000-00000000dfffffff]; performance may suffer.
[44657.352767] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[44657.356851] [drm] Driver supports precise vblank timestamp query.
[44657.360460] i915 0000:0b:00.0: BAR 6: can't assign [??? 0x00000000 flags 0x20000000] (bogus alignment)
[44657.365930] [drm] Failed to find VBIOS tables (VBT)
[44657.369102] [drm:gen9_set_dc_state [i915]] *ERROR* DC state mismatch (0x0 -> 0x2)
[44657.373895] vgaarb: device changed decodes: PCI:0000:0b:00.0,olddecodes=io+mem,decodes=none:owns=none
[44657.380718] [drm] Finished loading DMC firmware i915/kbl_dmc_ver1_04.bin (v1.4)
[44657.386000] [drm] Initialized i915 1.6.0 20171222 for 0000:0b:00.0 on minor 0
[44657.408383] [drm] Cannot find any crtc or sizes
[44657.410220] i915 0000:0b:00.0: fb0: inteldrmfb frame buffer device

 

root@TestStation:/dev/dri# cat /sys/kernel/debug/dri/0/i915_frequency_info
Killed

 

any idea on why this is happening?

[pigr8]

btw same happens on build DS918+, not only DS920+

 

after doing "cat /sys/kernel/debug/dri/0/i915_frequency_info" dmesg reports this, dunno if it's relevant.

 

[  528.649500] Module [i915] is removed.
[  528.662672] Module [video] is removed.
[  528.671464] Module [backlight] is removed.
[  528.680441] Module [iosf_mbi] is removed.
[  528.690424] Module [drm_kms_helper] is removed.
[  528.700498] Module [drm] is removed.
[  528.710492] Module [i2c_algo_bit] is removed.
[  528.720401] Module [fb] is removed.
[  537.772134] i915 0000:0b:00.0: enabling device (0000 -> 0003)
[  537.779777] mtrr: type mismatch for d0000000,10000000 old: write-back new: write-combining
[  537.785354] Failed to add WC MTRR for [00000000d0000000-00000000dfffffff]; performance may suffer.
[  537.792724] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[  537.795749] [drm] Driver supports precise vblank timestamp query.
[  537.797809] i915 0000:0b:00.0: BAR 6: can't assign [??? 0x00000000 flags 0x20000000] (bogus alignment)
[  537.800813] [drm] Failed to find VBIOS tables (VBT)
[  537.802469] vgaarb: device changed decodes: PCI:0000:0b:00.0,olddecodes=io+mem,decodes=none:owns=none
[  537.806473] [drm] Finished loading DMC firmware i915/kbl_dmc_ver1_04.bin (v1.4)
[  537.810299] [drm] Initialized i915 1.6.0 20171222 for 0000:0b:00.0 on minor 0
[  537.831388] [drm] Cannot find any crtc or sizes
[  537.832578] i915 0000:0b:00.0: fb0: inteldrmfb frame buffer device
[  560.067392] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
[  560.073035] IP: [<ffffffffa0d7ab11>] drm_debugfs_open+0x11/0x20 [drm]
[  560.077512] PGD 0
[  560.079047] Oops: 0000 [#1] SMP
[  560.081489] Modules linked in: i915(E) video(E) backlight(E) iosf_mbi(E) drm_kms_helper(E) drm(E) i2c_algo_bit(E) fb(E) fuse 8021q vhost_scsi(O) vhost(O) tcm_loop(O) iscsi_target_mod(O) target_core_user(O) target_core_ep(O) target_core_multi_file(O) target_core_file(O) target_core_iblock(O) target_core_mod(O) syno_extent_pool(PO) rodsp_ep(O) udf isofs loop synoacl_vfs(PO) btrfs ecryptfs zstd_decompress zstd_compress xxhash xor raid6_pq aesni_intel glue_helper lrw gf128mul ablk_helper apollolake_synobios(PO) hid_generic usbhid hid usblp syscopyarea sysfillrect sysimgblt fb_sys_fops cfbfillrect cfbcopyarea cfbimgblt drm_panel_orientation_quirks fbdev uhci_hcd r8168(O) e1000e(O) sg dm_snapshot dm_bufio crc_itu_t crc_ccitt psnap p8022 llc hfsplus md4 hmac sit tunnel4 ipv6 flashcache_syno(O) flashcache(O) syno_flashcache_control(O) dm_mod arc4 crc32c_intel cryptd ecb aes_x86_64 authenc des_generic ansi_cprng cts md5 cbc cpufreq_powersave cpufreq_performance processor cpufreq_stats vxlan ip6_udp_tunnel udp_tunnel ip_tunnel zram sha256_generic synorbd(O) synofsbd(O) igb(O) etxhci_hcd vmxnet3(OE) vmw_vmci(E) button usb_storage xhci_pci xhci_hcd ehci_pci ehci_hcd usbcore usb_common [last unloaded: fb]
[  560.122073] CPU: 3 PID: 19503 Comm: cat Tainted: P           OE   4.4.180+ #42661
[  560.123326] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 11/12/2020
[  560.125104] task: ffff88008477a640 ti: ffff880167164000 task.ti: ffff880167164000
[  560.126366] RIP: 0010:[<ffffffffa0d7ab11>]  [<ffffffffa0d7ab11>] drm_debugfs_open+0x11/0x20 [drm]
[  560.127784] RSP: 0018:ffff880167167c80  EFLAGS: 00010286
[  560.128592] RAX: ffff88016fc28780 RBX: ffff88016fc28780 RCX: ffff88016f7564a0
[  560.129709] RDX: 0000000000000000 RSI: ffff88016fc28780 RDI: ffff88016fc28780
[  560.130825] RBP: ffff880167167c80 R08: 0000000000000000 R09: 0000000000000000
[  560.131950] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88004a62e980
[  560.133043] R13: ffffffffa0d7ab00 R14: 0000000000000000 R15: ffff88016fc28790
[  560.134128] FS:  00007fd69b304500(0000) GS:ffff880172d80000(0000) knlGS:0000000000000000
[  560.135383] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  560.136278] CR2: 0000000000000008 CR3: 0000000168940000 CR4: 00000000003606f0
[  560.137380] Stack:
[  560.137695]  ffff880167167cc0 ffffffff81188f91 ffff880081409cc0 ffff88004db71cc0
[  560.139008]  ffff88016fc28780 ffff880167167efc ffff880085d81a80 ffff880167167dc0
[  560.140229]  ffff880167167ce8 ffffffff8118a63b 0000000000008000 0000000000008000
[  560.141467] Call Trace:
[  560.141856]  [<ffffffff81188f91>] do_dentry_open+0x1d1/0x350
[  560.142755]  [<ffffffff8118a63b>] vfs_open+0x5b/0x90
[  560.143520]  [<ffffffff8119f578>] path_openat+0x328/0x1bd0
[  560.144370]  [<ffffffff8113841a>] ? __alloc_pages_nodemask+0x11a/0xb50
[  560.145368]  [<ffffffff811a2b3e>] do_filp_open+0x7e/0xc0
[  560.146186]  [<ffffffff8118855c>] do_sys_open+0x17c/0x210
[  560.147082]  [<ffffffff8118a8ef>] SyS_openat+0xf/0x20
[  560.147864]  [<ffffffff8157b7ca>] entry_SYSCALL_64_fastpath+0x1e/0x8e
[  560.148875] Code: 5d 5d c3 49 8d 7d 00 e8 de f8 ff ff 49 8b 85 e8 01 00 00 eb d2 0f 1f 44 00 00 55 48 8d 06 48 8b 97 80 02 00 00 48 89 e5 48 89 c7 <48> 8b 4a 08 48 8b 71 08 e8 d2 d3 43 e0 5d c3 55 48 8d 06 48 c7
[  560.153352] RIP  [<ffffffffa0d7ab11>] drm_debugfs_open+0x11/0x20 [drm]
[  560.154400]  RSP <ffff880167167c80>
[  560.154958] CR2: 0000000000000008
[  560.155491] ---[ end trace 346afd34bbcba1e1 ]---
[  560.156289] ------------[ cut here ]------------

 

Edited August 10, 2022 by pigr8

[blackmanga]

On 8/10/2022 at 2:53 PM, pigr8 said:

btw same happens on build DS918+, not only DS920+

 

after doing "cat /sys/kernel/debug/dri/0/i915_frequency_info" dmesg reports this, dunno if it's relevant.

 

[  528.649500] Module [i915] is removed.
[  528.662672] Module [video] is removed.
[  528.671464] Module [backlight] is removed.
[  528.680441] Module [iosf_mbi] is removed.
[  528.690424] Module [drm_kms_helper] is removed.
[  528.700498] Module [drm] is removed.
[  528.710492] Module [i2c_algo_bit] is removed.
[  528.720401] Module [fb] is removed.
[  537.772134] i915 0000:0b:00.0: enabling device (0000 -> 0003)
[  537.779777] mtrr: type mismatch for d0000000,10000000 old: write-back new: write-combining
[  537.785354] Failed to add WC MTRR for [00000000d0000000-00000000dfffffff]; performance may suffer.
[  537.792724] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[  537.795749] [drm] Driver supports precise vblank timestamp query.
[  537.797809] i915 0000:0b:00.0: BAR 6: can't assign [??? 0x00000000 flags 0x20000000] (bogus alignment)
[  537.800813] [drm] Failed to find VBIOS tables (VBT)
[  537.802469] vgaarb: device changed decodes: PCI:0000:0b:00.0,olddecodes=io+mem,decodes=none:owns=none
[  537.806473] [drm] Finished loading DMC firmware i915/kbl_dmc_ver1_04.bin (v1.4)
[  537.810299] [drm] Initialized i915 1.6.0 20171222 for 0000:0b:00.0 on minor 0
[  537.831388] [drm] Cannot find any crtc or sizes
[  537.832578] i915 0000:0b:00.0: fb0: inteldrmfb frame buffer device
[  560.067392] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
[  560.073035] IP: [<ffffffffa0d7ab11>] drm_debugfs_open+0x11/0x20 [drm]
[  560.077512] PGD 0
[  560.079047] Oops: 0000 [#1] SMP
[  560.081489] Modules linked in: i915(E) video(E) backlight(E) iosf_mbi(E) drm_kms_helper(E) drm(E) i2c_algo_bit(E) fb(E) fuse 8021q vhost_scsi(O) vhost(O) tcm_loop(O) iscsi_target_mod(O) target_core_user(O) target_core_ep(O) target_core_multi_file(O) target_core_file(O) target_core_iblock(O) target_core_mod(O) syno_extent_pool(PO) rodsp_ep(O) udf isofs loop synoacl_vfs(PO) btrfs ecryptfs zstd_decompress zstd_compress xxhash xor raid6_pq aesni_intel glue_helper lrw gf128mul ablk_helper apollolake_synobios(PO) hid_generic usbhid hid usblp syscopyarea sysfillrect sysimgblt fb_sys_fops cfbfillrect cfbcopyarea cfbimgblt drm_panel_orientation_quirks fbdev uhci_hcd r8168(O) e1000e(O) sg dm_snapshot dm_bufio crc_itu_t crc_ccitt psnap p8022 llc hfsplus md4 hmac sit tunnel4 ipv6 flashcache_syno(O) flashcache(O) syno_flashcache_control(O) dm_mod arc4 crc32c_intel cryptd ecb aes_x86_64 authenc des_generic ansi_cprng cts md5 cbc cpufreq_powersave cpufreq_performance processor cpufreq_stats vxlan ip6_udp_tunnel udp_tunnel ip_tunnel zram sha256_generic synorbd(O) synofsbd(O) igb(O) etxhci_hcd vmxnet3(OE) vmw_vmci(E) button usb_storage xhci_pci xhci_hcd ehci_pci ehci_hcd usbcore usb_common [last unloaded: fb]
[  560.122073] CPU: 3 PID: 19503 Comm: cat Tainted: P           OE   4.4.180+ #42661
[  560.123326] Hardware name: VMware, Inc. VMware Virtual Platform/440BX Desktop Reference Platform, BIOS 6.00 11/12/2020
[  560.125104] task: ffff88008477a640 ti: ffff880167164000 task.ti: ffff880167164000
[  560.126366] RIP: 0010:[<ffffffffa0d7ab11>]  [<ffffffffa0d7ab11>] drm_debugfs_open+0x11/0x20 [drm]
[  560.127784] RSP: 0018:ffff880167167c80  EFLAGS: 00010286
[  560.128592] RAX: ffff88016fc28780 RBX: ffff88016fc28780 RCX: ffff88016f7564a0
[  560.129709] RDX: 0000000000000000 RSI: ffff88016fc28780 RDI: ffff88016fc28780
[  560.130825] RBP: ffff880167167c80 R08: 0000000000000000 R09: 0000000000000000
[  560.131950] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88004a62e980
[  560.133043] R13: ffffffffa0d7ab00 R14: 0000000000000000 R15: ffff88016fc28790
[  560.134128] FS:  00007fd69b304500(0000) GS:ffff880172d80000(0000) knlGS:0000000000000000
[  560.135383] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  560.136278] CR2: 0000000000000008 CR3: 0000000168940000 CR4: 00000000003606f0
[  560.137380] Stack:
[  560.137695]  ffff880167167cc0 ffffffff81188f91 ffff880081409cc0 ffff88004db71cc0
[  560.139008]  ffff88016fc28780 ffff880167167efc ffff880085d81a80 ffff880167167dc0
[  560.140229]  ffff880167167ce8 ffffffff8118a63b 0000000000008000 0000000000008000
[  560.141467] Call Trace:
[  560.141856]  [<ffffffff81188f91>] do_dentry_open+0x1d1/0x350
[  560.142755]  [<ffffffff8118a63b>] vfs_open+0x5b/0x90
[  560.143520]  [<ffffffff8119f578>] path_openat+0x328/0x1bd0
[  560.144370]  [<ffffffff8113841a>] ? __alloc_pages_nodemask+0x11a/0xb50
[  560.145368]  [<ffffffff811a2b3e>] do_filp_open+0x7e/0xc0
[  560.146186]  [<ffffffff8118855c>] do_sys_open+0x17c/0x210
[  560.147082]  [<ffffffff8118a8ef>] SyS_openat+0xf/0x20
[  560.147864]  [<ffffffff8157b7ca>] entry_SYSCALL_64_fastpath+0x1e/0x8e
[  560.148875] Code: 5d 5d c3 49 8d 7d 00 e8 de f8 ff ff 49 8b 85 e8 01 00 00 eb d2 0f 1f 44 00 00 55 48 8d 06 48 8b 97 80 02 00 00 48 89 e5 48 89 c7 <48> 8b 4a 08 48 8b 71 08 e8 d2 d3 43 e0 5d c3 55 48 8d 06 48 c7
[  560.153352] RIP  [<ffffffffa0d7ab11>] drm_debugfs_open+0x11/0x20 [drm]
[  560.154400]  RSP <ffff880167167c80>
[  560.154958] CR2: 0000000000000008
[  560.155491] ---[ end trace 346afd34bbcba1e1 ]---
[  560.156289] ------------[ cut here ]------------

 

Hi, looks like this is in virtual machine and not baremetal.

There could be something specific with regards to how the virtual machine is showing the iGPU.  Instead of my backported drivers, have you tried the patched drivers that ig88 or someone posted in the forums. See if that works.

 

 

[pigr8]

3 hours ago, blackmanga said:

Hi, looks like this is in virtual machine and not baremetal.

There could be something specific with regards to how the virtual machine is showing the iGPU.  Instead of my backported drivers, have you tried the patched drivers that ig88 or someone posted in the forums. See if that works.

 

yes it's VMWare, mm i have to research on those drivers you mention, @IG-88 were those drivers yours?