Arch Linux

pl

Member

Registered: 2014-02-10

Posts: 32

Re: Dell XPS 15 9560 (Early 2017)

Hey, what's the situation right now? I have been using a frozen working setup for a while, but just updated to 4.11.2.

My setup is still compiling a custom kernel with ACPI_REV_OVERRIDE_POSSIBLE enabled and acpi_rev_override as my only boot flag. I have enabled power optimizations suggested by Powertop except for GPU power management, that is taken care of by bbswitch by turning it off altogether when not in use. Since 380+ is no longer in beta, basic nvidia-dkms works for using the discrete gpu when needed.

Using this setup my battery consumption is reported at around 7W with half screen brightness and the battery estimate over 12 hours. All features including suspend seems to be working.

So, anything I'm missing? Scrolling recent posts, this kind of stuff makes me think I'm out of the loop:

4.11 seems to work fine. Previously I was using linux-nvme, with the samsung pm961 SSD. Now I've switched from nvidia-dkms and bbswitch-dkms to just nvidia and bbswitch. My battery life has gone from good to straight up ridiculous since getting 4.11 on stock kernel. I get 6-7W with firefox open and basic web browsing on wifi, at half brightness leading to 11~ hours of battery. What other tweaks are you guys using? I think everyone should use the repository tlp package instead of tlp-git because the AUR package seems abandoned and it outdated, while the repo package is up to date. TLP enabled with default settings gets my battery down to 5W, but has a few annoying problems.

Is bbswitch/gpu power management/suspend working for you without the acpi_rev_override boot flag?

I didn't compile the kernel, I'm just using the stock one. Here are the boot parameters I'm using:

  GRUB_CMDLINE_LINUX_DEFAULT='pcie_port_pm=off acpi_backlight=vendor acpi_osi=Linux acpi_osi=! acpi_osi="Windows 2009" rcutree.rcu_idle_gp_delay=1'

Is using the acpi_osi boot parameters not breaking the touchpad anymore?

tj8892

Member

Registered: 2017-01-29

Posts: 14

Re: Dell XPS 15 9560 (Early 2017)

Is bbswitch/gpu power management/suspend working for you without the acpi_rev_override boot flag?

Yep. I've never needed a custom kernel for bumblebee. On a side note, 4.11 seems to have broken one thing for me. GNOME says the fn+brightness key combo is changing the brightness, but it stays the same. xbacklight doesn't do anything, and the only way to change brightness seems to be using light-git from the AUR. Not sure why it's giving me problems.... related to systemd's backlight stuff maybe? same problem on xorg and wayland sessions.

simo8989

Member

Registered: 2015-12-22

Posts: 72

Re: Dell XPS 15 9560 (Early 2017)

I can report that at least for me I don't exclude anymore the Nvidia card from the power management in Tlp. As far as I have seen bbswitch works correctly at every boot now. Do you think that excluding the card in the tlp configuration gives some watts less in power consumption ?

pl

Member

Registered: 2014-02-10

Posts: 32

Re: Dell XPS 15 9560 (Early 2017)

Thanks a lot for your efforts hedgepigdaniel! I updated the original post to point to the wiki page as the main source from now.

Is bbswitch/gpu power management/suspend working for you without the acpi_rev_override boot flag?

Yep. I've never needed a custom kernel for bumblebee. On a side note, 4.11 seems to have broken one thing for me. GNOME says the fn+brightness key combo is changing the brightness, but it stays the same. xbacklight doesn't do anything, and the only way to change brightness seems to be using light-git from the AUR. Not sure why it's giving me problems.... related to systemd's backlight stuff maybe? same problem on xorg and wayland sessions.

Tried with stock kernel and system hardlocked 5 seconds after starting X so I did not get as far as even trying to switch GPU on and off to see if it hangs... what boot parameters are you using?

Last edited by pl (2017-05-24 10:43:34)

gilbertw1

Member

Registered: 2016-04-27

Posts: 14

Re: Dell XPS 15 9560 (Early 2017)

I didn't compile the kernel, I'm just using the stock one. Here are the boot parameters I'm using:

  GRUB_CMDLINE_LINUX_DEFAULT='pcie_port_pm=off acpi_backlight=vendor acpi_osi=Linux acpi_osi=! acpi_osi="Windows 2009" rcutree.rcu_idle_gp_delay=1'

Is using the acpi_osi boot parameters not breaking the touchpad anymore?

I've been using the above boot parameters for roughly two months now and have never seen any issues with the trackpad or any issues with suspend resume (I frequently go a week without a reboot). Additionally, the bumblebee power settings work perfectly, it does the right thing and properly disables the card when I'm not optirunning anything.

tj8892

Member

Registered: 2017-01-29

Posts: 14

Re: Dell XPS 15 9560 (Early 2017)

Tried with stock kernel and system hardlocked 5 seconds after starting X so I did not get as far as even trying to switch GPU on and off to see if it hangs... what boot parameters are you using?

I used to use a whole bunch ( pcie_port_pm=off acpi_backlight=vendor acpi_osi=Linux acpi_osi=! acpi_osi="Windows 2009" rcutree.rcu_idle_gp_delay=1). It seems like all of them except pcie_port_pm are unnecessary and don't do anything so I haven't used them for a while. Pci power management needs to be disabled, or the nvidia card refuses to power off, and can't use optirun. I can still startx on the intel card and don't have any instability.
Anyone have an idea how to fix gnome backlight changing? It's driving me mad.

sbeanie

Member

Registered: 2014-07-01

Posts: 8

Re: Dell XPS 15 9560 (Early 2017)

For those of you with backlight issues after upgrading to 4.11.2, removing the acpi_backlight=vendor boot parameter fixed it.  I'm currently using nvidia-dkms, bbswitch-dkms on a custom 4.11.2 build with the rev_override config flag set.  Current boot parameters: "pcie_port_pm=off enable_fbc=1 enable_psr=1 disable_power_well=0 acpi_rev_override=1".  Note Xorg wouldn't start until nvidia and bbswitch was installed and would freeze the system.  Wayland worked fine without these.

j9ac9k

Member

Registered: 2017-03-27

Posts: 12

Re: Dell XPS 15 9560 (Early 2017)

For those of you with backlight issues after upgrading to 4.11.2, removing the acpi_backlight=vendor boot parameter fixed it.  I'm currently using nvidia-dkms, bbswitch-dkms on a custom 4.11.2 build with the rev_override config flag set.  Current boot parameters: "pcie_port_pm=off enable_fbc=1 enable_psr=1 disable_power_well=0 acpi_rev_override=1".  Note Xorg wouldn't start until nvidia and bbswitch was installed and would freeze the system.  Wayland worked fine without these.

Removed acpi_backlight=vendor boot parameter and definitely still not able to adjust the backlight, but this could also be a X vs. wayland thing?

osho741

Member

Registered: 2017-04-19

Posts: 9

Re: Dell XPS 15 9560 (Early 2017)

I used pcie_port_pm=off  and build the kernel with the rev overide config set on an arch build with Openbox and it won't boot once I install bbswitch. Does the card need to be on.during boot up?

Ahh hell well F##k me, I see where I went wrong now, I didn't use the acpi-overide parameter just the pcie_port_pm=off boot parameter!

Last edited by osho741 (2017-05-24 16:52:31)

sbeanie

Member

Registered: 2014-07-01

Posts: 8

Re: Dell XPS 15 9560 (Early 2017)

For those of you with backlight issues after upgrading to 4.11.2, removing the acpi_backlight=vendor boot parameter fixed it.  I'm currently using nvidia-dkms, bbswitch-dkms on a custom 4.11.2 build with the rev_override config flag set.  Current boot parameters: "pcie_port_pm=off enable_fbc=1 enable_psr=1 disable_power_well=0 acpi_rev_override=1".  Note Xorg wouldn't start until nvidia and bbswitch was installed and would freeze the system.  Wayland worked fine without these.

Removed acpi_backlight=vendor boot parameter and definitely still not able to adjust the backlight, but this could also be a X vs. wayland thing?

Tried acpi_backlight=vendor with Xorg and did not work and locked up the system for a few seconds when trying to change brightness.  Removing the option, regenerating the grub config and then rebooting works on both Xorg and Wayland.

gnunn

Member

Registered: 2013-10-21

Posts: 96

Re: Dell XPS 15 9560 (Early 2017)

Is anyone having issues with VirtualBox and kernel panics on this machine? KVM works fine, but as soon as I try to start up the VirtualBox GUI the laptop hangs and the caps lock key starts flashing indicating a kernel panic.

After much googling I found that it is caused by the TB16 dock. Unplugging the dock and starting virtualbox works fine. See the post here where it was originally reported:

http://en.community.dell.com/techcenter … 5/20995840

Kernel 4.11 seems to have fixed this issue for me.

sbeanie

Member

Registered: 2014-07-01

Posts: 8

Re: Dell XPS 15 9560 (Early 2017)

For those of you with backlight issues after upgrading to 4.11.2, removing the acpi_backlight=vendor boot parameter fixed it.  I'm currently using nvidia-dkms, bbswitch-dkms on a custom 4.11.2 build with the rev_override config flag set.  Current boot parameters: "pcie_port_pm=off enable_fbc=1 enable_psr=1 disable_power_well=0 acpi_rev_override=1".  Note Xorg wouldn't start until nvidia and bbswitch was installed and would freeze the system.  Wayland worked fine without these.

After contemplating why my power usage was higher than it was before reinstalling today, I remembered i915 should be prepended to the start of some of the boot parameters.   Achieving 6-7 watts idle at minimum brightness now:

i915.enable_fbc=1 i915.enable_psr=1 i915.disable_power_well=0 acpi_rev_override=1 nvme_core.default_ps_max_latency_us=170000 pcie_port_pm=off 

hedgepigdaniel

Member

Registered: 2014-01-07

Posts: 42

Re: Dell XPS 15 9560 (Early 2017)

thanks @pl and everyone who has contributed! always great when people participate in a public discussion

@simo8989 wow, I only get more jealous of people with samsung SSDs. Don't have data to prove it but I would imagine that a lower latency to transition to those states would result in significantly reduced power consumption. In my case the first entry in the nvme get-feature state list is 1500ms (from a enlat+exlat of 30ms for power consumption of 0.012) and the deepest state happens after 8500ms (enlat+exlat=170ms). So I guess if the pm961 deepest state only takes 8.2ms to transition in and out of then it will probably transition into that state after only 410ms, and presumably  faster still for the more shallow non operational states. So at least an extra 1 second where the Toshiba SSD is using 1.9W instead of 0.012W. How a piece of circuitry the size of 3 coins with no moving parts can take 170ms to transition between states is not clear to me...

I've found that my power consumption is higher by about 2W on 4.11 than on 4.10+nvme patch. Looking at the powertop report everything looks similar except that the processor spends alot more time in the C3(pc3) package state (11% vs 3%). I don't see much differences in the processes, wakeups etc. I heard that processor frequency governors changed in 4.11, maybe that's the cause...? Same results with the screen off aswell

If I understand correctly,

i915.enable_fbc

is no longer necessary on 4.11: https://www.phoronix.com/scan.php?page= … BC-Skylake

tj8892

Member

Registered: 2017-01-29

Posts: 14

Re: Dell XPS 15 9560 (Early 2017)

I fixed my brightness by removing several kernel parameters- now my only param is pcie_port_pm=off. Without it, everything is stable, but the card can't power off. I'm glad it doesn't require a custom kernel to fix. Is acpi_rev_override the better solution? If it is, I hope the arch maintainers will enable acpi_rev_override in the default kernel.

j9ac9k

Member

Registered: 2017-03-27

Posts: 12

Re: Dell XPS 15 9560 (Early 2017)

I fixed my brightness by removing several kernel parameters- now my only param is pcie_port_pm=off. Without it, everything is stable, but the card can't power off. I'm glad it doesn't require a custom kernel to fix. Is acpi_rev_override the better solution? If it is, I hope the arch maintainers will enable acpi_rev_override in the default kernel.

I didn't remove all my parameters, but after removing i915.enable_fbc=1 acpi_backlight=vendor acpi_osi=Linux, acpi_osi=! and acpi_osi='windows 2009' boot parameters, (and added pcie_port_pm=off) I'm now able to adjust the backlight.

tj8892

Member

Registered: 2017-01-29

Posts: 14

Re: Dell XPS 15 9560 (Early 2017)

I fixed my brightness by removing several kernel parameters- now my only param is pcie_port_pm=off. Without it, everything is stable, but the card can't power off. I'm glad it doesn't require a custom kernel to fix. Is acpi_rev_override the better solution? If it is, I hope the arch maintainers will enable acpi_rev_override in the default kernel.

Actually removing them broke bumblebee. I  think it needs acpi_osi=windows(year) and acpi_osi=linux. with just pcie_port_pm, lspci and screenfetch both cause an unrecoverable hardlock. fun...

j9ac9k

Member

Registered: 2017-03-27

Posts: 12

Re: Dell XPS 15 9560 (Early 2017)

I fixed my brightness by removing several kernel parameters- now my only param is pcie_port_pm=off. Without it, everything is stable, but the card can't power off. I'm glad it doesn't require a custom kernel to fix. Is acpi_rev_override the better solution? If it is, I hope the arch maintainers will enable acpi_rev_override in the default kernel.

Actually removing them broke bumblebee. I  think it needs acpi_osi=windows(year) and acpi_osi=linux. with just pcie_port_pm, lspci and screenfetch both cause an unrecoverable hardlock. fun...

I aded acpi_osi='Windows 2009' and acpi_osi=linux and am able to adjust my backlight brightness still.

EDIT: it looks like for backlight adjustment to be functional, you cannot have acpi_backlight=vendor and acpi_osi=! in your boot parameters.

Last edited by j9ac9k (2017-05-24 23:28:47)

tj8892

Member

Registered: 2017-01-29

Posts: 14

Re: Dell XPS 15 9560 (Early 2017)

Finally found the best combination of parameters (in my opinion). Works with stock 4.11 kernel, optirun works fine, battery is good, and brightness works in gnome. acpi_osi=Linux acpi_osi="Windows 2009" acpi_backlight=native acpi_osi=! . acpi_backlight=native seems to be the key here, as it lets the backlight work despite having acpi_osi=!, Linux and Windows 2009. With any other combination my backlight doesn't work in gnome. This also lets you avoid having a custom kernel, and you can still use pcie_port_pm for other pci devices. (Not sure if it makes a difference, but I blacklisted the 1050 for PCI powersaving in tlp settings with RUNTIME_PM_BLACKLIST="01:00.0" .) I was previously using  rcutree.rcu_idle_gp_delay=1 as well, but it seems to make no difference to anything and I'm not able to find out what it does.

simo8989

Member

Registered: 2015-12-22

Posts: 72

Re: Dell XPS 15 9560 (Early 2017)

thanks @pl and everyone who has contributed! always great when people participate in a public discussion

@simo8989 wow, I only get more jealous of people with samsung SSDs. Don't have data to prove it but I would imagine that a lower latency to transition to those states would result in significantly reduced power consumption. In my case the first entry in the nvme get-feature state list is 1500ms (from a enlat+exlat of 30ms for power consumption of 0.012) and the deepest state happens after 8500ms (enlat+exlat=170ms). So I guess if the pm961 deepest state only takes 8.2ms to transition in and out of then it will probably transition into that state after only 410ms, and presumably  faster still for the more shallow non operational states. So at least an extra 1 second where the Toshiba SSD is using 1.9W instead of 0.012W. How a piece of circuitry the size of 3 coins with no moving parts can take 170ms to transition between states is not clear to me...

I've found that my power consumption is higher by about 2W on 4.11 than on 4.10+nvme patch. Looking at the powertop report everything looks similar except that the processor spends alot more time in the C3(pc3) package state (11% vs 3%). I don't see much differences in the processes, wakeups etc. I heard that processor frequency governors changed in 4.11, maybe that's the cause...? Same results with the screen off aswell

If I understand correctly,

i915.enable_fbc

is no longer necessary on 4.11: https://www.phoronix.com/scan.php?page= … BC-Skylake

Hi Daniel, yes from the outputs of the nvme-cli seems the samsung ssd goes to the deeper states very fast.
Here the output:

[simone@xps15 ~]$ sudo nvme get-feature -f 0x0c -H /dev/nvme0n1
get-feature:0xc (Autonomous Power State Transition), Current value:0x000001
	Autonomous Power State Transition Enable (APSTE): Enabled
	Auto PST Entries	.................
	Entry[ 0]   
	.................
	Idle Time Prior to Transition (ITPT): 86 ms
	Idle Transition Power State   (ITPS): 3
	.................
	Entry[ 1]   
	.................
	Idle Time Prior to Transition (ITPT): 86 ms
	Idle Transition Power State   (ITPS): 3
	.................
	Entry[ 2]   
	.................
	Idle Time Prior to Transition (ITPT): 86 ms
	Idle Transition Power State   (ITPS): 3
	.................
	Entry[ 3]   
	.................
	Idle Time Prior to Transition (ITPT): 410 ms
	Idle Transition Power State   (ITPS): 4
	.................
	Entry[ 4]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[ 5]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[ 6]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[ 7]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[ 8]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[ 9]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[10]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[11]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[12]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[13]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[14]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[15]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[16]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[17]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[18]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[19]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[20]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[21]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[22]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[23]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[24]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[25]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[26]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[27]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[28]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[29]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[30]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................
	Entry[31]   
	.................
	Idle Time Prior to Transition (ITPT): 0 ms
	Idle Transition Power State   (ITPS): 0
	.................

And also the other here:

ps    0 : mp:7.60W operational enlat:0 exlat:0 rrt:0 rrl:0
          rwt:0 rwl:0 idle_power:- active_power:-
ps    1 : mp:5.00W operational enlat:0 exlat:0 rrt:1 rrl:1
          rwt:1 rwl:1 idle_power:- active_power:-
ps    2 : mp:3.60W operational enlat:0 exlat:0 rrt:2 rrl:2
          rwt:2 rwl:2 idle_power:- active_power:-
ps    3 : mp:0.0400W non-operational enlat:210 exlat:1500 rrt:3 rrl:3
          rwt:3 rwl:3 idle_power:- active_power:-
ps    4 : mp:0.0050W non-operational enlat:2200 exlat:6000 rrt:4 rrl:4
          rwt:4 rwl:4 idle_power:- active_power:-

However my power consumption on idle does never go below 9.5 W (when I am really idle) what numbers do you now report with this new kernel? Also I just installed archlinux on the ssd and I found out that the power consumption is the same as the case arch is installed in the usb stick.. still get 9.5 W.
Also do you have any idea or clue for what we can do to further optimize this power discharge? in particular to solve this problem of the PCI devices stuck at 100%?

The battery reports a discharge rate of 9.69 W
The power consumed was 203 J

              Usage     Device name
             21.2%        CPU misc
             21.2%        DRAM
             21.2%        CPU core
             43.3 ops/s   GPU core
             43.3 ops/s   GPU misc
             50.1%        Display backlight
            100.0%        I2C Device (i2c-DLL07BE:01): DLL07BE:01
            100.0%        USB device: USB Receiver (Logitech)
            100.0%        PCI Device: Intel Corporation Sunrise Point-H PCI Express Root Port #1
            100.0%        USB device: xHCI Host Controller
            100.0%        Radio device: ath10k_pci
            100.0%        PCI Device: Intel Corporation Sunrise Point-H PCI Express Root Port #15
            100.0%        PCI Device: Intel Corporation Skylake Processor Thermal Subsystem
            100.0%        PCI Device: Qualcomm Atheros QCA6174 802.11ac Wireless Network Adapter
            100.0%        PCI Device: Samsung Electronics Co Ltd NVMe SSD Controller SM961/PM961
            100.0%        PCI Device: Intel Corporation Device 591b
            100.0%        PCI Device: Intel Corporation Sunrise Point-H Thermal subsystem
            100.0%        PCI Device: Intel Corporation Sunrise Point-H PCI Express Root Port #9
            100.0%        PCI Device: Realtek Semiconductor Co., Ltd. RTS525A PCI Express Card Reader
            100.0%        PCI Device: Intel Corporation Sunrise Point-H USB 3.0 xHCI Controller
            100.0%        PCI Device: Intel Corporation Sunrise Point-H PCI Express Root Port #2
            100.0%        PCI Device: Intel Corporation Sunrise Point-H SATA Controller [AHCI mode]
              0.3 pkts/s  Network interface: wlp2s0 (ath10k_pci)
              0.0%        USB device: usb-device-0cf3-e300
              0.0%        Audio codec hwC0D2: Intel
              0.0%        Audio codec hwC0D0: Realtek
              0.0%        Radio device: btusb
              0.0%        PCI Device: Intel Corporation Device a171
              0.0%        PCI Device: Intel Corporation Device 5910
              0.0%        PCI Device: NVIDIA Corporation GP107M [GeForce GTX 1050 Mobile]
              0.0%        PCI Device: Intel Corporation Sunrise Point-H Serial IO I2C Controller #0

Last edited by simo8989 (2017-05-25 16:30:10)

smb1234

Member

Registered: 2017-03-22

Posts: 3

Re: Dell XPS 15 9560 (Early 2017)

Kernel 4.11 seems to have fixed the suspend / resume issues I had with 4.10. I am still compiling my own kernel, enabling the acpi override flag. The only relevant boot parameter I set is

acpi_rev_override=1

Also, using nvidia-dkms now bumblebee works as expected, the only thing I needed was to add

RUNTIME_PM_BLACKLIST="01:00.0"

to /etc/default/tlp

With a minimum screen brightness I can get the reported wattage down to 9-10 Watt during normal web browsing with wifi. This is the 4k model, I am quite pleased with the performance now.

Orbital_sFear

Member

Registered: 2014-10-13

Posts: 43

Re: Dell XPS 15 9560 (Early 2017)

Kernel 4.11 and 4.10.13 prevent my system from booting.  4.10.10 is okay.

I'm running:
Systemd boot
NVE harddrives
Xorg
GDM
Gnome 3.xx

When I run these kernels GDM does start.  For a few seconds the computer will take keyboard input.  I can change my terminal with Ctrl+Alt+F2 and get into a terminal, from there everything will work fine in that terminal.  If I don't change into a terminal, after a few seconds mouse clicks and all keyboard input is ignored.  The Mouse will continue to move just fine, the backlight on the keyboard acts correctly.

I don't even know where to start debugging.  As a temp "fix" I'm running linux-lts kernel, which runs fine.

Here is my boot config:
https://pastebin.com/4Rn5nkbh

Here is a a journalctl -b of a failed boot (Gdm debug is on):
https://pastebin.com/ykuzkpP8

Here is a journalctl -b of a successful boot:
https://pastebin.com/pb1TXhbX

Anyone else hit by this?  Any other information I can provide?  Thanks.

dglastonbury

Member

Registered: 2017-05-26

Posts: 2

Re: Dell XPS 15 9560 (Early 2017)

Signed up to thank hedgepigdaniel for his work. I'm running his custom kernel + nvidia drivers. I wasn't able to get much below 10-11W when idle though. Installing a recent BIOS update from DELL got me below 10W. Currently I'm "idling" with cinnamon + powertop at 7.20W.

I also have the same problem with 4.11 (built from git via yaourt) with the brightness keys not adjusting the screen brightness.

zzzad

Member

Registered: 2017-02-23

Posts: 4

Re: Dell XPS 15 9560 (Early 2017)

I am running a custom built kernel 4.11 (acpi_rev_override=1) with a Samsung SSD and power consumption is about 4-5 W with screen dimmed to 5%.

Things are running quite smooth, but I am experiencing screen flickering from time to time. Has this been observed and solved before? I think its not related to the kernel as it has been there before my recent updates, but it is starting to annoy me.

Edit: It is not the 4K panel and setting brightness to 100% doesn't seem to fix it.

Last edited by zzzad (2017-05-26 11:26:42)

sbeanie

Member

Registered: 2014-07-01

Posts: 8

Re: Dell XPS 15 9560 (Early 2017)

Just to note installing the current tlp version (updated to 1.0 on the 25th) no longer requires the blacklist of the nvidia card for bumblebee to work.

https://github.com/linrunner/TLP/blob/master/changelog

Have tested and can confirm it works.

simo8989

Member

Registered: 2015-12-22

Posts: 72

Re: Dell XPS 15 9560 (Early 2017)

Just to note installing the current tlp version (updated to 1.0 on the 25th) no longer requires the blacklist of the nvidia card for bumblebee to work.

https://github.com/linrunner/TLP/blob/master/changelog

Have tested and can confirm it works.

I can confirm this too. I never used to backlist the Nvidia card what was happening with the previous version of tlp is that bbswitch could aswell put the discrete card off on boot, dmesg was successfully reporting this but with this configuration from time to time the laptop was running hotter and the fans where spinning and stopping at rather constant intervals. Now with the new version of tlp the output of dmesg regarding bbswitch is also the same but the laptop is much .. much more cool and when on battery the fans almost never spin even when watching 1080p videos on YouTube.
The difference now is that if you have a look at the tunable options in powertop, the Nvidia card is automatically excluded from the runtime management. Therefore, it compares as a tunable option now. I live simply don't touching it and the experience using this laptop now is much better with battery life that in my case (4K model) is even better that windows 10.

Last edited by simo8989 (2017-05-26 12:59:46)