Few issues with Asus UX501VW

Epoxyc · 2016-03-01 00:58:19

Hi !
I just installed arch on my new Asus Zenbook Pro UX501VW (i7 6700HQ Skylake, GTX 960m), it work well, but I have some issues.

1. Touchpad :
Fixed by kernel update.

2: The Nvidia chipset:
I installed the intel / nvidia drivers and bumblebee, everything works.
However, when I try to install bbswitch, the laptop freezes on boot, like here : https://bbs.archlinux.org/viewtopic.php?id=205091
I tried with the two (stable and mainline), with intel/nouveau, same problem.
Fixed by adding

acpi_osi=! acpi_osi="Windows 2009"

in kernel parameters.

3: Wifi:
The wireless connection works fine, but I get an error at startup:

iwlwifi 0000:03:00.0: Unsupported splx structure

lspci | grep Wireless gives me :

03:00.0 Network controller: Intel Corporation Wireless 7265 (rev 59)

4: Media Keys (fixed):
All media keys work except for brightness.
I tried adding "acpi_osi =" kernel option, the keys work but xbacklight doesn't work anymore.
Add kernel option acpi_osi= and acpi_backlight=native fix the issue.
But it not compatible with the solution for the Nvidia.
Fixed by kernel update.

5: SD Reader:
It doesn't work, I don't know what to install for it to be recognized.
lspci gives me this line that I suppose to be SD reader :

02:00.0 Unassigned class [ff00]: Alcor Micro Device 6621

6: Headphones noise (like whistle) (fixed) :
Patch you kernel with https://raw.githubusercontent.com/ngkah … 01vw.patch
(Thanks to ngkaho1234)
Fixed by Kernel update.

This link can help : https://wiki.archlinux.org/index.php/AS … _Pro_UX501

Thank you in advance.
(And sorry for my bad english)

Last edited by Epoxyc (2017-08-20 19:48:44)

cafe · 2016-03-01 12:28:54

I can only help with item #4. There are a few things you can try.

First, check how many devices are listed under /sys/class/backlight/
If you have more than one device listed, try to change manually the brightness of the screen to find out witch one control the brightness of the screen.

% cat /sys/class/backlight/acpi_video0/max_brightness
9
% su
Password: 
# echo 5 > /sys/class/backlight/acpi_video0/brightness

In the above example, I'm only testing device acpi_video0. I fist check the maximum brightness accepted then I change the register value to something else. If the screen brightness changes, you have found the correct device.

I'm not 100% sure, but I think most DE's like gnome will pick one backlight device arbitrarily. You can either program your key to launch a script that will change the backlight, or you can try using the kernel option acpi_backlight=vendor. This will remove one of the devices listed and probably help your DE to find the correct device under /sys/class/backlight/*.

Hope it helps.

Last edited by cafe (2016-03-01 12:29:15)

Epoxyc · 2016-03-01 17:04:00

I have exactly one device listed (intel_backlight) in /sys/class/backlight.
But when I boot with acpi_osi=, there are two devices (intel_backlight and acpi_video0) and I have not rights to write in intel_backlight/brightness file (I can write in without kernel option).
Add the kernel option acpi_backlight=vendor doesn't work but I tried acpi_backlight=native with acpi_osi= and it works !
Thank you !

EDIT : Key to enable/disable sreen doesn't work, but it's a smaller issue.
EDIT 2: I have no problem with bbswitch if I start manually bumblebee service after boot.

Last edited by Epoxyc (2016-03-02 00:17:22)

zkanda · 2016-03-03 09:21:22

Do you guys have problem with static noise when the headphone is plugged in?

Epoxyc · 2016-03-03 17:39:52

I have this problem, I noticed it yesterday, I will investigate and edit my first post.

Epoxyc · 2016-03-06 12:12:24

I had an error on boot

iwlwifi 0000:03:00.0: Direct firmware load for iwlwifi-7265D-20.ucode failed with error -2
iwlwifi 0000:03:00.0: Direct firmware load for iwlwifi-7265D-19.ucode failed with error -2
iwlwifi 0000:03:00.0: Direct firmware load for iwlwifi-7265D-18.ucode failed with error -2
iwlwifi 0000:03:00.0: Direct firmware load for iwlwifi-7265D-17.ucode failed with error -2

Fixed by downloading and copying iwlwifi-7265D-20.ucode in /lib/firmware.
However I still have :

iwlwifi 0000:03:00.0: Unsupported splx structure

TheHebes · 2016-03-12 01:01:38

Hopping on the train for getting issues on this laptop all worked out.
When I load the bbswitch module, my fans kick on to 100% after a short delay; has anyone else observed this behavior?

Also, I've found that the trackpad works fine with the version of linux in the repos ATM (4.4.3-1), no need to compile mainline.
Installing xf86-input-synaptics was enough.

Epoxyc · 2016-03-12 11:51:29

I had the same behaviour when I used nvidia package with linux-mainline, with nvidia mainline, if I enable bumblebeed service, gdm freeze when I'm loggin in, but if I start bumblebeed after login, all works correctly !

zkanda · 2016-04-06 03:35:58

Did anyone of you tried the latest 4.6 rc? There's a sound fixes for Skylake[0], unfortunately I need to revert back to windows to work done so I cannot test it for now.

0 - http://lkml.iu.edu/hypermail/linux/kern … 02047.html

TheHebes · 2016-04-11 02:03:29

I just compiled and tested 4.6rc1. There's no improvement on the ringing problem.

Epoxyc · 2016-04-11 14:39:36

There is :

      ALSA: hda - Asus N750JV external subwoofer fixup
      ALSA: hda - Fix white noise on Asus N750JV headphone
      ALSA: hda - Apply fix for white noise on Asus N550JV, too

in the 4.6rc2 changelog, but i compiled and there is still the headphones noise issue.

bobby.m · 2016-04-11 15:28:20

Epoxyc wrote:

There is :
      ALSA: hda - Asus N750JV external subwoofer fixup
      ALSA: hda - Fix white noise on Asus N750JV headphone
      ALSA: hda - Apply fix for white noise on Asus N550JV, too
in the 4.6rc2 changelog, but i compiled and there is still the headphones noise issue.

That patch is only for N750JV and N550JV.
You may fix yours by patching hda to apply mute by amp for your laptop.

edit: Let me know if you need help with that.

Last edited by bobby.m (2016-04-11 15:33:16)

Epoxyc · 2016-04-11 17:20:49

I didn't know that it was possible to patch hda for this laptop, how to do it ?

bobby.m · 2016-04-15 18:34:31

I have a dkms module witch i use for some time, if you send your souncard id I can make one for your laptop to test and see if it works.
To get the sound card id install `alsa-utils` and run `alsa-info.sh --stdout | grep Subsystem`

Epoxyc · 2016-04-16 13:17:39

Okay, thankyou. My card sound ID are 0x10431080 (Realtek ALC668) and 0x80860101 (Intel Skylake HDMI).

ngkaho1234 · 2016-05-05 13:48:50

I also had problem with the white noise coming out from the headset when running Linux on the machine.
After fiddling around with the kernel code, I made a kernel patch which made my life easier.

https://raw.githubusercontent.com/ngkah … 01vw.patch

ngkaho1234 · 2016-05-05 19:21:30

You may also need this:
https://github.com/ngkaho1234/ux501vw-c … -base.conf

asus_gd · 2016-05-08 17:01:30

Hi, sorry for my low kernel literacy,
how to include this patch for Realtek into kernel i'am currently on, 4.4.0-22-generic (ubuntu)?
tnks

Epoxyc · 2016-05-09 06:39:32

Thank you for this patch, I will try today.
To include this patch, you have to download the sources of you kernel, patch and rebuild it (the ubuntu documentation should say how to do this).

EDIT : works well \o/

Last edited by Epoxyc (2016-05-09 13:49:17)

zkanda · 2016-05-16 02:29:16

Seems like the headphone noise issue is fix upstream as well with the release of 4.6.
https://lwn.net/Articles/687511/

Thanks ngkaho1234.

Epoxyc · 2016-05-22 09:57:30

Perfect, it remains only one problem, SD card reader, I saw that the driver is already included in kernel, but it won't work. Does anyone have more information ?

Last edited by Epoxyc (2016-05-22 10:07:01)

frix · 2016-06-25 12:53:27

Recently bought the UX501VW (i7, 960M). I don't think we have a wiki page on that laptop model yet, or do we? Some options in the UX501 page seem out of date as they either do nothing or make the issue worse ; the sound noise fix actually made more noise!

TheHebes wrote:

When I load the bbswitch module, my fans kick on to 100% after a short delay; has anyone else observed this behavior?

I've noticed that a few times but not anymore. There's something about fans and acpid in the UX501 wiki page, you might want to try that.

Epoxyc wrote:

However, when I try to install bbswitch, the laptop freezes on boot

For me it works with the bbswitch loaded, just not with bumblebeed on boot : GDM shows up but I get a hard freeze when it tries to load Gnome Shell.

I actually think that there's a race condition or a bug of some kind in the drivers that make the system crash if you try to access the NVIDIA drivers at the wrong time. For instance, trying to run nvidia-smi after having stopped bumblebeed invariably results in a full freeze without anything in the logs. Alternatively this could possibly be an ACPI issue given the few warnings I consistently find in the logs.

A few log lines before a hard freeze after I enabled bumblebeed manually, tried a game, exited it then tried it again :

Jun 25 13:03:34 gnome-shell[732]: GNOME Shell started at Sat Jun 25 2016 13:03:32 GMT+0200 (CEST)
...couple of unrelated ACPI warnings...
Jun 25 13:04:25 bumblebeed[1878]: [   72.630760] [INFO]/usr/bin/bumblebeed 3.2.1 started
Jun 25 13:04:25 kernel: bbswitch: disabling discrete graphics
Jun 25 13:04:25 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 13:05:23 kernel: bbswitch: enabling discrete graphics
...hard freeze...

These are the last lines before a crash when I enabled bumblebee manually after X startup, tried a game with optirun, ran nvidia-smi to check that it was indeed using the discrete GPU, exited it then tried running nvidia-smi again which resulted in a hard freeze :

Jun 25 02:20:35 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 02:20:35 kernel: nvidia-modeset: Unloading
Jun 25 02:20:35 kernel: nvidia-nvlink: Unregistered the Nvlink Core, major device number 243
Jun 25 02:20:35 kernel: bbswitch: disabling discrete graphics
Jun 25 02:20:35 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 02:20:35 kernel: pci 0000:01:00.0: Refused to change power state, currently in D0
Jun 25 02:20:35 acpid[416]: client 10934[0:56] has disconnected
Jun 25 02:20:35 acpid[416]: client 10934[0:56] has disconnected

This is when I don't load bumblebeed on startup but run nvidia-smi once Gnome is up (this one hasn't crashed yet) :

Jun 25 14:39:34 kernel: nvidia: module license 'NVIDIA' taints kernel.
Jun 25 14:39:34 kernel: Disabling lock debugging due to kernel taint
Jun 25 14:39:34 kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 243
Jun 25 14:39:34 kernel: NVRM: loading NVIDIA UNIX x86_64 Kernel Module  367.27  Thu Jun  9 18:53:27 PDT 2016
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)
Jun 25 14:39:34 kernel: ACPI Warning: \_SB.PCI0.PEG0.PEGP._DSM: Argument #4 type mismatch - Found [Buffer], ACPI requires [Package] (20160108/nsarguments-95)

MaikuMori · 2016-06-25 14:12:08

From my experience having the GFX off and then querying information about the GFX results in full kernel freeze. There are no logs because they don't get flushed to disc.

It's very easy to reproduce:

# Turn off GFX card.
lspci -vnn
# Full freeze.

For now I've just removed the bbswitch so my video card is always on and that solved the crashes.

Did you manage to fix the audio issue? The last kernel improved the situation - on boot headphones have almost no noise (almost unnoticeable little hiss, but that's probably just background electronic noise), but after a while it comes back (I think after suspend/closed lid, but haven't checked yet).

I also have problem with the right speaker, but I think it's hardware defect. It just sounds like if you put too much volume on small speaker (highs/bass distorted) except it happens on medium volume and the left speaker works perfectly fine even on 2-3x the volume.

Other random issues I've noticed, but haven't tried to fix:

- Hybernate doesn't work (could be my setup since I have encrypted disc and swap)
- On rare cases resume from suspend doesn't happen, I only get black screen and nothing works.

BlissfulChoreography · 2016-07-13 15:08:53

MaikuMori wrote:

For now I've just removed the bbswitch so my video card is always on and that solved the crashes.

Turning off the discrete card can be stable if you don't use X, I've disabled bumblebeed and loaded bbswitch on startup with load_state=0 unload_state=1

cat /proc/acpi/bbswitch

gives

0000:01:00.0 OFF

Using gdm + Gnome 3.20 on a Wayland session, I only have freezes on vlc and smplayer so far. Chrome works, as do Qt/KDE apps, though the HiDPI support is not great.

Edit: lspci -vnn does not cause a freeze either

Last edited by BlissfulChoreography (2016-07-13 15:09:55)

MaikuMori · 2016-07-13 15:25:10

I find the touchpad experience on Wayland lacking. That's because it uses libinput.

Arch Linux

#1 2016-03-01 00:58:19

Few issues with Asus UX501VW

#2 2016-03-01 12:28:54

Re: Few issues with Asus UX501VW

#3 2016-03-01 17:04:00

Re: Few issues with Asus UX501VW

#4 2016-03-03 09:21:22

Re: Few issues with Asus UX501VW

#5 2016-03-03 17:39:52

Re: Few issues with Asus UX501VW

#6 2016-03-06 12:12:24

Re: Few issues with Asus UX501VW

#7 2016-03-12 01:01:38

Re: Few issues with Asus UX501VW

#8 2016-03-12 11:51:29

Re: Few issues with Asus UX501VW

#9 2016-04-06 03:35:58

Re: Few issues with Asus UX501VW

#10 2016-04-11 02:03:29

Re: Few issues with Asus UX501VW

#11 2016-04-11 14:39:36

Re: Few issues with Asus UX501VW

#12 2016-04-11 15:28:20

Re: Few issues with Asus UX501VW

#13 2016-04-11 17:20:49

Re: Few issues with Asus UX501VW

#14 2016-04-15 18:34:31

Re: Few issues with Asus UX501VW

#15 2016-04-16 13:17:39

Re: Few issues with Asus UX501VW

#16 2016-05-05 13:48:50

Re: Few issues with Asus UX501VW

#17 2016-05-05 19:21:30

Re: Few issues with Asus UX501VW

#18 2016-05-08 17:01:30

Re: Few issues with Asus UX501VW

#19 2016-05-09 06:39:32

Re: Few issues with Asus UX501VW

#20 2016-05-16 02:29:16

Re: Few issues with Asus UX501VW

#21 2016-05-22 09:57:30

Re: Few issues with Asus UX501VW

#22 2016-06-25 12:53:27

Re: Few issues with Asus UX501VW

#23 2016-06-25 14:12:08

Re: Few issues with Asus UX501VW

#24 2016-07-13 15:08:53

Re: Few issues with Asus UX501VW

#25 2016-07-13 15:25:10

Re: Few issues with Asus UX501VW

Board footer