New motherboard woes

Questions about hardware, drivers and peripherals
Forum rules
Before you post read how to get help. Topics in this forum are automatically closed 6 months after creation.
Locked
Paul L
Level 1
Level 1
Posts: 32
Joined: Fri Jul 13, 2007 7:09 pm

New motherboard woes

Post by Paul L »

Eventually "bit the bullet" and upgraded my ageing PC. Now I find that, while I can install Mint, I can't access the internet, don't have sound and can't change graphic details (I have 3 monitors). I'm assuming that this is a driver issue, but don't how, or if, I can update these.

I have an Asrock Fatality Z170 Gaming K6+ motherboard and Intel Skylake i5 chip - am using Mint 17.2

Has anyone had any success with a similar set up?

Thanks in advance for any help/advice.

Paul
Last edited by LockBot on Wed Dec 28, 2022 7:16 am, edited 1 time in total.
Reason: Topic automatically closed 6 months after creation. New replies are no longer allowed.
AlbertP
Level 16
Level 16
Posts: 6701
Joined: Sun Jan 30, 2011 12:38 pm
Location: Utrecht, The Netherlands

Re: New motherboard woes

Post by AlbertP »

Please open Terminal from the menu, type the following command:

Code: Select all

inxi -Fx
Then please press Enter, wait for the output and copy/paste it to the forum. (Note: inside the terminal window, use Ctrl-Shift-C or right-click > Copy, instead of Ctrl-C)

That will show us your hardware and drivers. Supposedly you need some driver upgrades as the hardware is very new- if I can see the details from the inxi command I can probably tell you which you need.
Registered Linux User #528502
Image
Feel free to correct me if I'm trying to write in Spanish, French or German.
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@ AlbertP I have the same problems namely I have just received a new machine with Skylake i5-6200U processor and it will only run in software rendering mode with Mint 17.2 Cinnamon and Mint xfce but runs perfectly with hardware acceleration in Ubuntu 15.04, 15.10 and MATE. Kernels up to 3.2.5 make no difference so I have provisionally ruled that out. It looks as if Mint has differences in how it identifies hardware to vanilla Ubuntu which is causing the problem.

The inxi output from Mint 17.2 where I only have software rendering is

Code: Select all

$ inxi -Fx
System:    Host: Helios Kernel: 4.2.5-040205-generic x86_64 (64 bit, gcc: 5.2.1) 
           Desktop: Cinnamon 2.6.13  Distro: Linux Mint 17.2 Rafaela
Machine:   System: Intel product: Skylake Platform version: 0.1
           Mobo: Topstar model: WhiteTip Mountain1 Fab2 version: RVP7
           Bios: American Megatrends version: 5.11 date: 09/10/2015
CPU:       Dual core Intel Core i5-6200U CPU (-HT-MCP-) cache: 3072 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 9595.74 
           Clock Speeds: 1: 620.156 MHz 2: 615.093 MHz 3: 529.312 MHz 4: 2699.906 MHz
Graphics:  Card: Intel Device 1916 bus-ID: 00:02.0 
           X.Org: 1.15.1 drivers: intel (unloaded: fbdev,vesa) Resolution: 1920x1080@60.0hz, 1920x1080@60.0hz 
           GLX Renderer: Gallium 0.4 on llvmpipe (LLVM 3.4, 256 bits) GLX Version: 2.1 Mesa 10.1.3 Direct Rendering: Yes
Audio:     Card: Intel Device 9d70 driver: snd_hda_intel bus-ID: 00:1f.3 Sound: ALSA ver: k4.2.5-040205-generic
Network:   Card: Intel Wireless 3160 driver: iwlwifi bus-ID: 01:00.0
           IF: wlan0 state: up mac: b4:6d:83:95:8c:b0
Drives:    HDD Total Size: 266.3GB (3.7% used) 1: id: /dev/sda model: Samsung_SSD_850 size: 250.1GB 
           2: USB id: /dev/sdb model: Drive_SM_USB20 size: 16.2GB 
Partition: ID: / size: 31G used: 5.9G (21%) fs: ext4 ID: /home size: 62G used: 150M (1%) fs: ext4 
           ID: swap-1 size: 10.49GB used: 0.00GB (0%) fs: swap 
RAID:      No RAID devices detected - /proc/mdstat and md_mod kernel raid module present
Sensors:   System Temperatures: cpu: 57.0C mobo: N/A 
           Fan Speeds (in rpm): cpu: N/A 
Info:      Processes: 184 Uptime: 0 min Memory: 461.3/7893.5MB Runlevel: 2 Gcc sys: 4.8.4 
           Client: Shell (bash 4.3.11) inxi: 1.9.17 
$ 
And the output from Ubuntu 15.10 Wily which does hardware acceleration perfectly is:

Code: Select all

$ inxi -Fx
System:    Host: Helios-Ubuntu Kernel: 4.2.0-17-generic x86_64 (64 bit gcc: 5.2.1)
           Desktop: Unity (Gtk 3.16.7-0ubuntu3) Distro: Ubuntu 15.10 wily
Machine:   System: Intel product: Skylake Platform v: 0.1
           Mobo: Topstar model: WhiteTip Mountain1 Fab2 v: RVP7
           Bios: American Megatrends v: 5.11 date: 09/10/2015
CPU:       Dual core Intel Core i5-6200U (-HT-MCP-) cache: 3072 KB
           flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 9595
           clock speeds: max: 2800 MHz 1: 499 MHz 2: 499 MHz 3: 2737 MHz
           4: 1275 MHz
Graphics:  Card: Intel Sky Lake Integrated Graphics bus-ID: 00:02.0
           Display Server: X.Org 1.17.2 drivers: intel (unloaded: fbdev,vesa)
           Resolution: 1920x1080@60.00hz, 1920x1080@60.00hz
           GLX Renderer: Mesa DRI Intel Skylake ULT GT2
           GLX Version: 3.0 Mesa 11.0.2 Direct Rendering: Yes
Audio:     Card Intel Device 9d70 driver: snd_hda_intel bus-ID: 00:1f.3
           Sound: Advanced Linux Sound Architecture v: k4.2.0-17-generic
Network:   Card: Intel Wireless 3160 driver: iwlwifi bus-ID: 01:00.0
           IF: wlp1s0 state: up mac: b4:6d:83:95:8c:b0
Drives:    HDD Total Size: 266.3GB (6.7% used)
           ID-1: /dev/sda model: Samsung_SSD_850 size: 250.1GB temp: 0C
           ID-2: USB /dev/sdb model: Drive_SM_USB20 size: 16.2GB temp: 0C
Partition: ID-1: / size: 23G used: 4.3G (20%) fs: ext4 dev: /dev/sda6
           ID-2: swap-1 size: 10.49GB used: 0.00GB (0%) fs: swap dev: /dev/sda4
RAID:      No RAID devices: /proc/mdstat, md_mod kernel module present
Sensors:   System Temperatures: cpu: 50.0C mobo: N/A
           Fan Speeds (in rpm): cpu: N/A
Info:      Processes: 211 Uptime: 1 min Memory: 453.5/7893.5MB
           Init: systemd runlevel: 5 Gcc sys: 5.2.1
           Client: Shell (bash 4.3.421) inxi: 2.2.16 
$ 
Any suggestions on what to try would be welcome as I do not want to use Ubuntu or even Mate if I can help it.
User avatar
jimallyn
Level 19
Level 19
Posts: 9075
Joined: Thu Jun 05, 2014 7:34 pm
Location: Wenatchee, WA USA

Re: New motherboard woes

Post by jimallyn »

Others that have had problems with Skylake computers have found that installing kernel 4.3 fixes the problems. Here's how to do it:

http://www.yourownlinux.com/2015/09/how ... linux.html

I recommended that to another user in this thread:

forums.linuxmint.com/viewtopic.php?f=49&t=208010

and it did the trick for him.
“If the government were coming for your TVs and cars, then you'd be upset. But, as it is, they're only coming for your sons.” - Daniel Berrigan
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@jimallyn I have installed kernel 4.3 and it makes no difference to the software rendering. It seems to be a difference in the way the hardware is detected or in Ubuntu and Mint or perhaps the packages available as it seemed OK on even quite early Ubuntu LiveUSBs.

I have also installed Cinnamon as an addition Desktop over Ubuntu Wily from a PPA and it renders fine although there are some irritations like quit, restart, suspend and hibernate do nothing.

I have also noticed that the Ubuntu and Mint installs use a different mouse driver as in one case scroll is via two fingers and the other has a right of pad single finger scroll which again points to some difference in hardware identification.

I now have dual Ubuntu and Mint booted system with a common /home folder so at least I can use the system and get some work done.
JCSheeron

Re: New motherboard woes

Post by JCSheeron »

I am having a similar issue with hardware acceleration not working with a new Skylake CPU and Motherboard. The CPU is an i5-6600K and the motherboard is an ASUS Z170M-E D3. My inxi -Fx output is below. Forgive me if I did not post this correctly, this is my first post.

I started with Mint 17.2, Cinnamon 2.7, and Kernel 3.16.0-38. I have since updated to Cinnamon 2.8, and Kernel 4.3.0-0403, and still have software acceleration only. The output inxi -Fx is below. Any help would be appreciated, and secondly, is this expected to work in Mint 17.3? I have my doubts, as the underlying Ubuntu base will remain the same in 17.3 as it is in 17.2. I assume hardware acceleration on a Skylake CPU will work in Mint 18.

Code: Select all

$ inxi -Fx
System:    Host: teamhome Kernel: 4.3.0-040300-generic x86_64 (64 bit, gcc: 5.2.1) 
           Desktop: Cinnamon 2.8.2  Distro: Linux Mint 17.2 Rafaela
Machine:   Mobo: ASUSTeK model: Z170M-E D3 version: Rev X.0x Bios: American Megatrends version: 0501 date: 09/24/2015
CPU:       Quad core Intel Core i5-6600K CPU (-MCP-) cache: 6144 KB flags: (lm nx sse sse2 sse3 sse4_1 sse4_2 ssse3 vmx) bmips: 28032.7 
           Clock Speeds: 1: 987.656 MHz 2: 833.164 MHz 3: 799.941 MHz 4: 2041.210 MHz
Graphics:  Card: Intel Device 1912 bus-ID: 00:02.0 
           X.Org: 1.15.1 drivers: intel (unloaded: fbdev,vesa) Resolution: 1280x1024@60.0hz 
           GLX Renderer: Gallium 0.4 on llvmpipe (LLVM 3.4, 256 bits) GLX Version: 2.1 Mesa 10.1.3 Direct Rendering: Yes
Audio:     Card: Intel Device a170 driver: snd_hda_intel bus-ID: 00:1f.3 Sound: ALSA ver: k4.3.0-040300-generic
Network:   Card: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller 
           driver: r8169 ver: 2.3LK-NAPI port: e000 bus-ID: 06:00.0
           IF: eth0 state: up speed: 1000 Mbps duplex: full mac: 30:5a:3a:0d:d4:24
Drives:    HDD Total Size: 500.1GB (50.2% used) 1: id: /dev/sda model: Samsung_SSD_850 size: 500.1GB 
Partition: ID: / size: 43G used: 6.9G (18%) fs: ext4 ID: /home size: 386G used: 227G (63%) fs: ext4 
           ID: swap-1 size: 33.79GB used: 0.00GB (0%) fs: swap 
RAID:      No RAID devices detected - /proc/mdstat and md_mod kernel raid module present
Sensors:   System Temperatures: cpu: 29.8C mobo: 27.8C 
           Fan Speeds (in rpm): cpu: 0 
Info:      Processes: 212 Uptime: 7 days Memory: 2448.1/15941.8MB Runlevel: 2 Gcc sys: 4.8.4 
           Client: Shell (bash 4.3.11) inxi: 1.9.17 

pdcurtis

Re: New motherboard woes

Post by pdcurtis »

I have been looking in the system logs and found in dmesg that firmware is not being loaded for the i915 in Mint 17.2. This is the section in question:

Code: Select all

    1.826482] [drm] Initialized drm 1.1.0 20060810
[    1.826804] scsi host0: ahci
[    1.830910] scsi host1: ahci
[    1.833073] scsi host2: ahci
[    1.833135] ata1: SATA max UDMA/133 abar m2048@0xdf12b000 port 0xdf12b100 irq 124
[    1.833138] ata2: SATA max UDMA/133 abar m2048@0xdf12b000 port 0xdf12b180 irq 124
[    1.833142] ata3: SATA max UDMA/133 abar m2048@0xdf12b000 port 0xdf12b200 irq 124
[    1.855279] i915 0000:00:02.0: Direct firmware load for i915/skl_dmc_ver1.bin failed with error -2
[    1.855299] [drm:i915_firmware_load_error_print [i915]] *ERROR* failed to load firmware i915/skl_dmc_ver1.bin (0)
[    1.855363] [drm] Memory usable by graphics device = 4096M
[    1.855367] checking generic (c0000000 7f0000) vs hw (c0000000 10000000)
[    1.855368] fb: switching to inteldrmfb from EFI VGA
[    1.855385] Console: switching to colour dummy device 80x25
[    1.855494] [drm] Replacing VGA console driver
[    1.862551] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013).
[    1.862554] [drm] Driver supports precise vblank timestamp query.
[    1.866925] vgaarb: device changed decodes: PCI:0000:00:02.0,olddecodes=io+mem,decodes=io+mem:owns=io+mem
[    1.879668] ACPI: Video Device [GFX0] (multi-head: yes  rom: no  post: no)
[    1.880039] input: Video Bus as /devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/LNXVIDEO:00/input/input7
[    1.880264] [drm] Initialized i915 1.6.0 20150731 for 0000:00:02.0 on minor 0
A quick check with dmesg and grep will show if you have this problem. This is from my Mint 17.2 Install:

Code: Select all

@Helios:~$ dmesg | grep -i firmware
[    0.190866] [Firmware Bug]: ACPI: BIOS _OSI(Linux) query ignored
[    1.838316] i915 0000:00:02.0: Direct firmware load for i915/skl_dmc_ver1.bin failed with error -2
[    1.838351] [drm:i915_firmware_load_error_print [i915]] *ERROR* failed to load firmware i915/skl_dmc_ver1.bin (0)
[    3.189501] psmouse serio1: elantech: assuming hardware version 4 (with firmware version 0x361f00)
[    3.881998] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-17.ucode failed with error -2
[    3.882013] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-16.ucode failed with error -2
[    3.882023] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-15.ucode failed with error -2
[    3.882033] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-14.ucode failed with error -2
[    3.884733] iwlwifi 0000:01:00.0: loaded firmware version 25.30.13.0 op_mode iwlmvm
$

This does not occur with ubuntu 15.10

I have found that this may be a known bug which has ben solved in 15.10 but, as yet, not backported to 14.04 which is the basis of Mint 17.2 see which may be the cause https://bugs.launchpad.net/ubuntu/+sour ... ug/1496163

I have also found a PPA at https://launchpad.net/~kranich/+archive/ubuntu/cinnamon which has allowed me to run Cinnamon as an additional desktop with Ubuntu 15.10 Wily and that works very well apart from some minor irritations - it seems there are differences in 'power' handling between Trusty, which Cinnamon was designed to use, and Wily. The result is that all the quit menus (shutdown, hibernate, suspend, restart) do not work, only logout. That applies via Menus, Automatic shutdowns, Function Keys and Lid closing. It is most likely a problem with the change to Systemd and I see no easy way out without breaking the existing compatibility with Trusty.

I have however written a Quit applet based on the existing one by bownz using the new system calls (systemctl suspend etc ) as a workaround which gives me a fully functional system for my wife to use without having to use the terminal to shut down! I would be very happy to provide copies or upload to my web site if it is useful to people whilst the problem is being fixed. It is much better than using Unity and I am now using the Cinnamon 2.8/Ubuntu Wily combination full time.

I have also tried the i915.preliminary_hw_support=1 fix at http://www.phoronix.com/scan.php?page=n ... im-support but it does not seem to make a difference.
AlbertP
Level 16
Level 16
Posts: 6701
Joined: Sun Jan 30, 2011 12:38 pm
Location: Utrecht, The Netherlands

Re: New motherboard woes

Post by AlbertP »

The Mint 17.x versions are all based on Ubuntu 14.04 LTS, that probably explains why the hardware isn't working yet even though it works in Ubuntu 15.x.

You could try installing the newer linux-firmware package from Ubuntu 15.10 which does include the file skl_dmc_ver1.bin: http://mirrors.kernel.org/ubuntu/pool/m ... .2_all.deb
That will solve at least one problem- maybe there is more software to update, however.
Registered Linux User #528502
Image
Feel free to correct me if I'm trying to write in Spanish, French or German.
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@AlbertP Thanks for the suggestion. Unfortunately I got a strange message in red in the package manager "Error: No longer provides linux-firmware" but I got it installed finally using dpkg.

It has not solved the problem and I get even more firmware related errors:

Code: Select all

$ dmesg | grep firmware
[    1.832908] i915 0000:00:02.0: Direct firmware load for i915/skl_dmc_ver1.bin failed with error -2
[    1.832942] [drm:i915_firmware_load_error_print [i915]] *ERROR* failed to load firmware i915/skl_dmc_ver1.bin (0)
[    1.856013] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    1.865978] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    1.870926] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    2.940985] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    2.947613] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    3.064697] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    3.070084] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    3.074945] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    3.079837] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    3.169852] psmouse serio1: elantech: assuming hardware version 4 (with firmware version 0x361f00)
[    3.748103] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-17.ucode failed with error -2
[    3.748115] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-16.ucode failed with error -2
[    3.748125] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-15.ucode failed with error -2
[    3.748136] iwlwifi 0000:01:00.0: Direct firmware load for iwlwifi-3160-14.ucode failed with error -2
[    3.751021] iwlwifi 0000:01:00.0: loaded firmware version 25.30.13.0 op_mode iwlmvm
[    4.150709] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    4.210734] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    4.215938] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    4.220733] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[    4.225037] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[   12.473535] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[   12.478472] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[   13.804562] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[   33.284143] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
[   33.288895] [drm:skl_set_power_well [i915]] *ERROR* CSR firmware not ready (2)
~$ 
Do I need to update anything to do with drm or the xserver as well.
AlbertP
Level 16
Level 16
Posts: 6701
Joined: Sun Jan 30, 2011 12:38 pm
Location: Utrecht, The Netherlands

Re: New motherboard woes

Post by AlbertP »

Which kernel version are you using? You can find that with:

Code: Select all

uname -r
You can use http://forums.linuxmint.com/viewtopic.php?f=42&t=40185 to upgrade if necessary. Ubuntu 15.10 has a 4.2.x kernel. (The x numbers are updates with bugfixes, usually you should install the latest one if you are offered multiple choices)

Also, it might be worth to check if the firmware file is correctly installed. Please check the contents of the /lib/firmware/i915 folder:

Code: Select all

ls /lib/firmware/i915/
Registered Linux User #528502
Image
Feel free to correct me if I'm trying to write in Spanish, French or German.
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@AlbertP
pcurtis@Helios:~$ uname -r
4.3.0-040300-generic
pcurtis@Helios:~$ ls /lib/firmware/i915/
bxt_dmc_ver1_04.bin skl_dmc_ver1_23.bin skl_guc_ver1_1059.bin
bxt_dmc_ver1.bin skl_dmc_ver1.bin skl_guc_ver1.bin
pcurtis@Helios:~$ lspci
00:00.0 Host bridge: Intel Corporation Device 1904 (rev 08)
00:02.0 VGA compatible controller: Intel Corporation Device 1916 (rev 07)
00:14.0 USB controller: Intel Corporation Device 9d2f (rev 21)
00:14.2 Signal processing controller: Intel Corporation Device 9d31 (rev 21)
00:16.0 Communication controller: Intel Corporation Device 9d3a (rev 21)
00:17.0 SATA controller: Intel Corporation Device 9d03 (rev 21)
00:1c.0 PCI bridge: Intel Corporation Device 9d14 (rev f1)
00:1f.0 ISA bridge: Intel Corporation Device 9d48 (rev 21)
00:1f.2 Memory controller: Intel Corporation Device 9d21 (rev 21)
00:1f.3 Audio device: Intel Corporation Device 9d70 (rev 21)
00:1f.4 SMBus: Intel Corporation Device 9d23 (rev 21)
01:00.0 Network controller: Intel Corporation Wireless 3160 (rev 83)
pcurtis@Helios:~$
I have added a lspci - you will note it is not identifying it as Skylake just by the id 1916 unlike running under wily.
I have already moved to kernel 4.3 but it was essentially the same under 4.2.5
Many thanks for the ideas I did not know how to find the available firmware. Any more suggestions welcome!
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

I have installed Mint 17.3 beta and that seems to have solved the problem so I do not plan to follow this up any further
AlbertP
Level 16
Level 16
Posts: 6701
Joined: Sun Jan 30, 2011 12:38 pm
Location: Utrecht, The Netherlands

Re: New motherboard woes

Post by AlbertP »

pdcurtis wrote:I have added a lspci - you will note it is not identifying it as Skylake just by the id 1916 unlike running under wily.
That is just because of an outdated list of device names used by lspci, which does not influence drivers in any way. You can fetch a new one with:

Code: Select all

sudo update-pciids
Registered Linux User #528502
Image
Feel free to correct me if I'm trying to write in Spanish, French or German.
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@AlbertP You were quite correct and they are now the same. The further I go into the details of Linux the less I find I know. Many Thanks
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@AlbertP I spoke too soon when I said everything was working under the Mint 17.3 beta. I have a bizarre problem which I missed in earlier testing.

After first suspend one of the processors starts running almost flat out and system monitor shows to to be a kworker process. Running perf has shown a lot of usb linked activity which does not occur before the suspend which led me to investigate the USB interface. Plugging in a USB memory stick to USB 2.0 or 3.0 port stops it and it stays stopped even when the stick is ejected but restarts when the USB stick is physically removed. Other USB devices do not stop activity including USB hard drive but enabling the built in camera by cheese stops all the kworker activity until it is turned off. I have also discovered this only occurs when using battery during the suspend cycle. A suspend/restore cycle with the mains adapter on puts the machine back to normal.

I hope you or someone can give some suggests on how to debug this as I have run out of ideas other than to order one of the nano memory sticks and leave it plugged in all the time (to hide the problem from my wife for whom the machine was bought)!

I can attach a copy of the output from perf or any other diagnostic if that helps - they are too big to include 'inline'.

Off the topic of motherboard woes. This is also a lesson in testing methodology - I caught this by chance when looking at power consumption and noticed it seemed to have increased after a suspend. I would normally test with mains on and often with a USB stick in place as I do a lot of testing on the LiveUSBs and both inhibited the problem. It was only have a CPU usage monitoring applet which exposed it. Betas are a good idea but it was only when I had installed that this showed up.
AlbertP
Level 16
Level 16
Posts: 6701
Joined: Sun Jan 30, 2011 12:38 pm
Location: Utrecht, The Netherlands

Re: New motherboard woes

Post by AlbertP »

Please run this in terminal, so we can identify the event (file in /sys/firmware/acpi/interrupts/) which is raised too many times and causes your kworker problem.

Code: Select all

tail -n +1 /sys/firmware/acpi/interrupts/*
If you see one of the gpeXX files which is enabled and shows a number over 10,000 you can probably disable it using the following command (replace gpeXX with the actual filename):

Code: Select all

echo disable | sudo tee -a /sys/firmware/acpi/interrupts/gpeXX
If this works we can look for ways to get this done automatically every time your computer boots.
Registered Linux User #528502
Image
Feel free to correct me if I'm trying to write in Spanish, French or German.
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

@AlbertP Many thanks for suggestions.
There was one gpexx file which had 13000 namely gpe50
I had to get a root terminal to get the command in because of permissions but it then disabled it:

Code: Select all

pete@Helios:~$ sudo su
Helios pete # echo disable | sudo tee -a /sys/firmware/acpi/interrupts/gpe50
disable

Code: Select all

==> /sys/firmware/acpi/interrupts/gpe50 <==
   13245   disabled
Unfortunately it has had no effect and one core is still running at 98% which reduces battery time by ~40%.

I have been doing some more investigation of what stops the excess usage.

It only starts when external power is off and the machine is suspended/resumed - I have not so far been able to provoke it any other way.
If power is restored excess usage stops but restarts if power unplugged.
If power is on and a suspend/resume cyle is carried out power can be unplugged without the excess usage.
Any USB 2 device stops the excess usage when plugged into the USB2 port including USB3 devices.
USB 3 devices plugged into a USB3 port do not stop the power usage.
Turning on the webcam stops the usage as long as it is on - the webcam is on the usb2 hub
Turning Bluetooth off and on with the function key stops the high usage as long as it still remains on. Software disabling and enabling does not work. When bluetooth switched by function key it disappears/appears in lsusb

In other words everything seems to point to the USB 2 system. It looks to me as if the code which monitors for changes loops when there is no device attached after a suspend.

Code: Select all

pete@Helios:~$ lsusb -t
/:  Bus 02.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/6p, 5000M
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=xhci_hcd/12p, 480M
    |__ Port 3: Dev 22, If 0, Class=Wireless, Driver=btusb, 12M
    |__ Port 3: Dev 22, If 1, Class=Wireless, Driver=btusb, 12M
    |__ Port 4: Dev 3, If 0, Class=Video, Driver=uvcvideo, 480M
    |__ Port 4: Dev 3, If 1, Class=Video, Driver=uvcvideo, 480M
pete@Helios:~$ 
Is there some way to 'restart' the usb 2 system after a suspend in the way it starts at a boot as a work-round?

Searching for "xhci_hcd kworker" found this https://bugs.launchpad.net/ubuntu/+sour ... ug/1488426 which looks very similar to what I am seeing although does not seem to involve suspend. The top of the perf report looks almost identical and #10 shows the excess usage stopping when a device is plugged into the USB 2 hub in the same way - unfortunately there does not seem to be a solution which works for everyone and I am resisting loading a completely unknown kernel to try out until I understand more.

Code: Select all

pete@Helios:~$ sudo perf record -g -a sleep 2
[ perf record: Woken up 7 times to write data ]
[ perf record: Captured and wrote 2.536 MB perf.data (10503 samples)

pete@Helios:~$ sudo perf report

+   62.29%     0.00%  kworker/0:1      [kernel.kallsyms]                   [k] ret_from_fork           ▒
+   62.29%     0.00%  kworker/0:1      [kernel.kallsyms]                   [k] kthread                 ▒
+   62.27%     0.03%  kworker/0:1      [kernel.kallsyms]                   [k] worker_thread           ▒
+   62.18%     0.14%  kworker/0:1      [kernel.kallsyms]                   [k] process_one_work        ▒
+   61.06%     0.10%  kworker/0:1      [kernel.kallsyms]                   [k] rpm_idle                ▒
+   60.64%     0.12%  kworker/0:1      [kernel.kallsyms]                   [k] rpm_suspend             ▒
+   60.63%     0.05%  kworker/0:1      [kernel.kallsyms]                   [k] pm_runtime_work         ▒
+   60.38%     0.03%  kworker/0:1      [kernel.kallsyms]                   [k] __rpm_callback          ▒
+   60.35%     0.02%  kworker/0:1      [kernel.kallsyms]                   [k] usb_runtime_idle        ▒
+   60.32%     0.02%  kworker/0:1      [kernel.kallsyms]                   [k] __pm_runtime_suspend    ▒
+   60.11%     0.01%  kworker/0:1      [kernel.kallsyms]                   [k] rpm_callback            ▒
+   60.04%     0.02%  kworker/0:1      [kernel.kallsyms]                   [k] usb_runtime_suspend     ▒
+   60.02%     0.04%  kworker/0:1      [kernel.kallsyms]                   [k] usb_suspend_both        ▒
+   58.09%     0.02%  kworker/0:1      [kernel.kallsyms]                   [k] usb_resume_interface.isr▒
+   58.06%     0.01%  kworker/0:1      [kernel.kallsyms]                   [k] hub_resume              ▒
+   58.01%     0.31%  kworker/0:1      [kernel.kallsyms]                   [k] hub_activate            ▒
+   56.80%     0.24%  kworker/0:1      [kernel.kallsyms]                   [k] hub_port_status         ▒
+   56.11%     0.27%  kworker/0:1      [kernel.kallsyms]                   [k] usb_control_msg         ▒
+   53.31%     0.38%  kworker/0:1      [kernel.kallsyms]                   [k] usb_start_wait_urb      ▒
+   46.41%     0.08%  kworker/0:1      [kernel.kallsyms]                   [k] usb_submit_urb          ▒
+   46.31%     0.45%  kworker/0:1      [kernel.kallsyms]                   [k] usb_submit_urb.part.6   ▒
+   44.42%     1.81%  kworker/0:1      [kernel.kallsyms]                   [k] usb_hcd_submit_urb      ▒
+   31.54%    30.69%  kworker/0:1      [kernel.kallsyms]                   [k] xhci_hub_control        ▒
+   16.62%     0.00%  ksoftirqd/0      [kernel.kallsyms]                   [k] ret_from_fork           ▒
+   16.62%     0.00%  ksoftirqd/0      [kernel.kallsyms]                   [k] kthread                 ▒
+   16.30%     1.08%  ksoftirqd/0      [kernel.kallsyms]                   [k] smpboot_thread_fn       ▒
+    8.57%     0.54%  kworker/0:1      [kernel.kallsyms]                   [k] usb_hcd_giveback_urb    ▒
+    7.71%     0.22%  ksoftirqd/0      [kernel.kallsyms]                   [k] run_ksoftirqd           ▒
+    7.45%     0.22%  ksoftirqd/0      [kernel.kallsyms]                   [k] schedule                ▒
+    7.44%     0.73%  kworker/0:1      [kernel.kallsyms]                   [k] __tasklet_schedule      ▒
+    7.27%     0.07%  kworker/0:1      [kernel.kallsyms]                   [k] wakeup_softirqd         ▒
+    7.26%     0.87%  ksoftirqd/0      [kernel.kallsyms]                   [k] __do_softirq            ▒
+    7.18%     0.13%  kworker/0:1      [kernel.kallsyms]                   [k] wake_up_process         ▒
+    7.11%     1.46%  ksoftirqd/0      [kernel.kallsyms]                   [k] __schedule              ▒
+    6.70%     0.53%  kworker/0:1      [kernel.kallsyms]                   [k] try_to_wake_up          ▒
+    6.33%     0.18%  kworker/0:1      [kernel.kallsyms]                   [k] _cond_resched           ▒
+    6.14%     0.41%  kworker/0:1      [kernel.kallsyms]                   [k] preempt_schedule_common ▒
+    6.01%     0.20%  kworker/0:1      [kernel.kallsyms]                   [k] wait_for_completion_time▒
+    5.98%     1.17%  ksoftirqd/0      [kernel.kallsyms]                   [k] tasklet_action          ▒
+    5.63%     1.96%  kworker/0:1      [kernel.kallsyms]                   [k] __schedule              ▒
+    5.43%     0.08%  kworker/0:1      [kernel.kallsyms]                   [k] ttwu_do_activate.constpr▒
+    4.51%     0.46%  ksoftirqd/0      [kernel.kallsyms]                   [k] usb_giveback_urb_bh     ▒
+    3.98%     0.05%  kworker/0:1      [kernel.kallsyms]                   [k] activate_task           ▒
+    3.87%     0.07%  kworker/0:1      [kernel.kallsyms]                   [k] enqueue_task            ▒
+    3.49%     0.94%  ksoftirqd/0      [kernel.kallsyms]                   [k] __usb_hcd_giveback_urb  ▒
+    3.31%     0.74%  kworker/0:1      [kernel.kallsyms]                   [k] enqueue_task_fair       ▒
+    2.97%     0.02%  ksoftirqd/0      [kernel.kallsyms]                   [k] deactivate_task         ▒
+    2.90%     0.11%  ksoftirqd/0      [kernel.kallsyms]                   [k] dequeue_task            ▒
+    2.43%     0.40%  kworker/0:1      [kernel.kallsyms]                   [k] pick_next_task_fair     ▒
+    2.33%     0.82%  ksoftirqd/0      [kernel.kallsyms]                   [k] dequeue_task_fair       ▒
+    2.28%     0.64%  kworker/0:1      [kernel.kallsyms]                   [k] enqueue_entity          ▒
+    1.87%     0.00%  cinnamon         [kernel.kallsyms]                   [k] entry_SYSCALL_64_fast
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

clem suggested " We might differ to other distributions on disk polling, it might be worth checking if it happens in Ubuntu as well just to confirm that."

As far as I can tell It does not happen in Ubuntu 15.10 Wily kernel 4.2.0.18 however if I add a Cinnamon 2.8 desktop from a Eric Kranich's PPA to Wily the problem returns so it unfortunately now looks as if it is Cinnamon related.

@AlbertP I can not find anything about Mint's disk polling all searches are dominated by the Polls on Mint's popularity! I have asked for hints but he was also very flattering about your abilities in "tinkering with exotic hardware" although I think Skylake will not count as exotic for long.
pdcurtis

Re: New motherboard woes

Post by pdcurtis »

I have seen kworker spinning once or twice under Ubuntu Wily but totally unpredictably unlike Mint 17.3. It looks more and more to me like a race condition in xhci_hcd or usbcore and I have found a very new thread https://www.spinics.net/lists/linux-usb/msg133143.html which shows similar symptoms on a Skylake machine.

On doing echo 'module xhci-hcd =p' >/sys/kernel/debug/dynamic_debug/control as was requested (and I have no idea what that would do ) but I found a very similar output in dmesg to that in the posting namely:

Code: Select all

$sudo su
$echo 'module xhci-hcd =p' >/sys/kernel/debug/dynamic_debug/control
................................
[16440.310449] usb usb1-port3: status 0107 change 0000
[16440.310454] usb usb1-port4: status 0507 change 0000
[16440.310482] hub 1-0:1.0: state 7 ports 12 chg 0000 evt 0000
[16440.310485] hub 1-0:1.0: hub_suspend
[16440.310489] usb usb1: bus auto-suspend, wakeup 1
[16440.310490] usb usb1: bus suspend fail, err -16
[16440.310492] hub 1-0:1.0: hub_resume
...................
I have not yet worked out what this means or even if it is being looked at in a relevant place but I hope it may offer some further enlightenment.
Locked

Return to “Hardware Support”