You are not logged in.

Pages: 1 … 49 50 51 52 53 54

Topic closed

#1251 2017-01-25 07:50:04

OdinEidolon
Member
From: Belluno - Italy
Registered: 2011-01-31
Posts: 498

Re: Dell XPS 13 9350 Late 2015

vtrac wrote:

Anyone have an issue coming back from sleep recently?  Mine goes to sleep just fie (per the logs), but on wakeup something crashes (I think) and it goes straight to a fresh boot.  It has been working fine for weeks.. I guess some package updated because I've tried on a few saved kernels I had that I know were working fine.

I had this happen 100% of the time for a while. No fix. It's a Skylake bug. I fixed mine completely by resetting the BIOS.

Hardware: 2016 Dell XPS15 - matte FullHD - i5-6300HQ - 32GB DDR4 - Nvidia GTX960M - Samsung 840EVO 250GB SSD - 56Wh
Software: Plasma 5 - rEFInd - linux-ck - preload - prelink - verynice - psd - bumblebee

Offline

#1252 2017-01-25 08:19:36

xpto
Member
Registered: 2017-01-25
Posts: 1

Re: Dell XPS 13 9350 Late 2015

vtrac wrote:

Anyone have an issue coming back from sleep recently?  Mine goes to sleep just fie (per the logs), but on wakeup something crashes (I think) and it goes straight to a fresh boot.  It has been working fine for weeks.. I guess some package updated because I've tried on a few saved kernels I had that I know were working fine.

Mine would never wake up from sleep on battery power with BIOS 1.4.10. On AC, it would wake up fine. Reverted to 1.4.4 and fixed it. Didn't try 1.4.12 yet.

Offline

#1253 2017-01-25 15:06:45

vtrac
Member
Registered: 2016-01-28
Posts: 42

Re: Dell XPS 13 9350 Late 2015

Welp, just confirmed that rebooting to windows once "fixes" my sleep resume issues.

@xpto: I'm on 1.4.12.

Offline

#1254 2017-01-25 15:26:53

OdinEidolon
Member
From: Belluno - Italy
Registered: 2011-01-31
Posts: 498

Re: Dell XPS 13 9350 Late 2015

vtrac wrote:

Welp, just confirmed that rebooting to windows once "fixes" my sleep resume issues.

@xpto: I'm on 1.4.12.

In my case the issues were present on both Windows and Linux.

Hardware: 2016 Dell XPS15 - matte FullHD - i5-6300HQ - 32GB DDR4 - Nvidia GTX960M - Samsung 840EVO 250GB SSD - 56Wh
Software: Plasma 5 - rEFInd - linux-ck - preload - prelink - verynice - psd - bumblebee

Offline

#1255 2017-01-26 10:57:04

robsmith11
Member
Registered: 2016-09-10
Posts: 23

Re: Dell XPS 13 9350 Late 2015

It turns out the drm-intel-nightly kernel does *not* include the nvme power saving patches.. I'm not sure why I thought it did.  After using the patch from damige's github (which applies cleanly to drm-intel-nightly), my idle power usage is down to 1.94 watts! smile

Last edited by robsmith11 (2017-01-26 11:17:03)

Offline

#1256 2017-01-26 14:37:00

damige
Member
Registered: 2016-09-02
Posts: 40

Re: Dell XPS 13 9350 Late 2015

robsmith11 wrote:

It turns out the drm-intel-nightly kernel does *not* include the nvme power saving patches.. I'm not sure why I thought it did.  After using the patch from damige's github (which applies cleanly to drm-intel-nightly), my idle power usage is down to 1.94 watts! smile

I have built a linuxnvme+inteldrm kernel and i am not getting anywhere near your idle usage. Strange!

[damige@nanobot ~]$ sudo powerstat -d0
Running for 480.0 seconds (48 samples at 10.0 second intervals).
Power measurements will start in 0 seconds time.

  Time    User  Nice   Sys  Idle    IO  Run Ctxt/s  IRQ/s Fork Exec Exit  Watts
15:33:23   0.1   0.0   0.1  99.9   0.0    2    157     29    0    0    0   3.94 
15:33:33   0.2   0.0   0.1  99.7   0.0    1    226     40    1    0    0   3.71 
15:33:43   0.1   0.0   0.1  99.8   0.0    1    147     28    0    0    1   3.78 
15:33:53   0.1   0.0   0.1  99.8   0.1    1    145     28    0    0    0   4.47 
15:34:03   0.1   0.0   0.1  99.8   0.0    1    148     30    0    0    0   3.58 
15:34:13   0.1   0.0   0.1  99.9   0.0    1    142     24    0    0    0   3.66 
15:34:23   0.0   0.0   0.1  99.8   0.0    1    137     24    0    0    0   3.59 
^C-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
 Average   0.1   0.0   0.1  99.8   0.0  1.1  157.5   29.0  0.1  0.0  0.1   3.82 
  StdDev   0.1   0.0   0.0   0.1   0.0  0.3   28.6    4.9  0.3  0.0  0.3   0.29 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
 Minimum   0.0   0.0   0.1  99.7   0.0  1.0  137.0   24.3  0.0  0.0  0.0   3.58 
 Maximum   0.2   0.0   0.1  99.9   0.1  2.0  226.3   40.1  1.0  0.0  1.0   4.47 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
Summary:
System:   3.82 Watts on average with standard deviation 0.29  
[damige@nanobot ~]$ uname -a
Linux nanobot 4.10.0-1-nvme-drm-intel-nightly #1 SMP PREEMPT Thu Jan 26 15:15:58 CET 2017 x86_64 GNU/Linux
[damige@nanobot ~]$

Have you done anything else? how are you testing?

XPS 15 7590 | i9-9980HK | 32GB RAM | 512G NVME | Intel AX200 | NVIDIA 1650 | OLED 4K

Offline

#1257 2017-01-26 14:47:59

OdinEidolon
Member
From: Belluno - Italy
Registered: 2011-01-31
Posts: 498

Re: Dell XPS 13 9350 Late 2015

damige wrote:
robsmith11 wrote:

It turns out the drm-intel-nightly kernel does *not* include the nvme power saving patches.. I'm not sure why I thought it did.  After using the patch from damige's github (which applies cleanly to drm-intel-nightly), my idle power usage is down to 1.94 watts! smile

I have built a linuxnvme+inteldrm kernel and i am not getting anywhere near your idle usage. Strange!

[damige@nanobot ~]$ sudo powerstat -d0
Running for 480.0 seconds (48 samples at 10.0 second intervals).
Power measurements will start in 0 seconds time.

  Time    User  Nice   Sys  Idle    IO  Run Ctxt/s  IRQ/s Fork Exec Exit  Watts
15:33:23   0.1   0.0   0.1  99.9   0.0    2    157     29    0    0    0   3.94 
15:33:33   0.2   0.0   0.1  99.7   0.0    1    226     40    1    0    0   3.71 
15:33:43   0.1   0.0   0.1  99.8   0.0    1    147     28    0    0    1   3.78 
15:33:53   0.1   0.0   0.1  99.8   0.1    1    145     28    0    0    0   4.47 
15:34:03   0.1   0.0   0.1  99.8   0.0    1    148     30    0    0    0   3.58 
15:34:13   0.1   0.0   0.1  99.9   0.0    1    142     24    0    0    0   3.66 
15:34:23   0.0   0.0   0.1  99.8   0.0    1    137     24    0    0    0   3.59 
^C-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
 Average   0.1   0.0   0.1  99.8   0.0  1.1  157.5   29.0  0.1  0.0  0.1   3.82 
  StdDev   0.1   0.0   0.0   0.1   0.0  0.3   28.6    4.9  0.3  0.0  0.3   0.29 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
 Minimum   0.0   0.0   0.1  99.7   0.0  1.0  137.0   24.3  0.0  0.0  0.0   3.58 
 Maximum   0.2   0.0   0.1  99.9   0.1  2.0  226.3   40.1  1.0  0.0  1.0   4.47 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
Summary:
System:   3.82 Watts on average with standard deviation 0.29  
[damige@nanobot ~]$ uname -a
Linux nanobot 4.10.0-1-nvme-drm-intel-nightly #1 SMP PREEMPT Thu Jan 26 15:15:58 CET 2017 x86_64 GNU/Linux
[damige@nanobot ~]$

Have you done anything else? how are you testing?

With a standard SATA ssd, which uses very little power, I cannot get below 3W despite having all the known tweaks in action: forced aspm, usb autosuspend, rc6, fbc, psr, guc...
I have the i5 FHD, so pretty much the model with the lowest power consumption.

EDIT: then of course I keep confusing the 9350 and 9550 threads... with my 9550 I find it hard to go below 4W and I never go below 3W. These are idle stats, no DE open, no application open, display at min brightness, no wifi.

Last edited by OdinEidolon (2017-01-26 15:00:34)

Hardware: 2016 Dell XPS15 - matte FullHD - i5-6300HQ - 32GB DDR4 - Nvidia GTX960M - Samsung 840EVO 250GB SSD - 56Wh
Software: Plasma 5 - rEFInd - linux-ck - preload - prelink - verynice - psd - bumblebee

Offline

#1258 2017-01-26 16:58:12

robsmith11
Member
Registered: 2016-09-10
Posts: 23

Re: Dell XPS 13 9350 Late 2015

I measure power usage using a status bar script that samples

/sys/class/power_supply/BAT0/current_now
/sys/class/power_supply/BAT0/voltage_now

Here's the output of the powerstat script with minimum brightness and nothing besides X, dwm, and st running:

Running for 480.0 seconds (48 samples at 10.0 second intervals).
Power measurements will start in 0 seconds time.

  Time    User  Nice   Sys  Idle    IO  Run Ctxt/s  IRQ/s  Watts
16:40:17   0.0   0.0   0.0 100.0   0.0    1     10      9   1.89
16:40:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.92
16:40:37   0.0   0.0   0.0 100.0   0.0    1      9      9   1.91
16:40:47   0.0   0.0   0.0 100.0   0.0    1      8      9   1.85
16:40:57   0.0   0.0   0.0 100.0   0.0    1      9      9   1.84
16:41:07   0.0   0.0   0.0 100.0   0.0    1      9     10   1.87
16:41:17   0.0   0.0   0.0 100.0   0.0    1      9      9   1.85
16:41:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.88
16:41:37   0.0   0.0   0.0 100.0   0.0    1     10     11   1.88
16:41:47   0.0   0.0   0.0 100.0   0.0    1     10     10   1.89
16:41:57   0.0   0.0   0.0 100.0   0.0    1     10     12   1.84
16:42:07   0.0   0.0   0.0 100.0   0.0    1      9      9   1.84
16:42:17   0.0   0.0   0.0 100.0   0.0    1     11     12   1.88
16:42:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.86
16:42:37   0.0   0.0   0.0 100.0   0.0    1      9      9   1.86
16:42:47   0.0   0.0   0.0 100.0   0.0    1     10     10   1.84
16:42:57   0.0   0.0   0.0  99.9   0.1    1     13     10   1.84
16:43:07   0.0   0.0   0.0 100.0   0.0    1      9      9   1.88
16:43:17   0.0   0.0   0.0 100.0   0.0    1      9     10   1.87
16:43:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.85
16:43:37   0.0   0.0   0.0 100.0   0.0    1      9      9   1.85
16:43:47   0.0   0.0   0.0 100.0   0.0    1      8      9   1.81
16:43:57   0.0   0.0   0.0 100.0   0.0    1      9      9   1.87
16:44:07   0.0   0.0   0.0 100.0   0.0    1     10     12   1.87
16:44:17   0.0   0.0   0.0 100.0   0.0    1      9      9   1.88
16:44:27   0.0   0.0   0.0 100.0   0.0    3     11     10   1.86
16:44:37   0.1   0.0   0.1  99.7   0.2    1     68     33   1.91
16:44:47   0.0   0.0   0.0 100.0   0.0    1     11     11   1.85
16:44:57   0.0   0.0   0.0  99.9   0.1    1     10     10   1.88
16:45:07   0.0   0.0   0.0 100.0   0.0    1      9      9   1.86
16:45:17   0.0   0.0   0.0 100.0   0.0    1     10     10   1.86
16:45:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.84
16:45:37   0.0   0.0   0.1  99.9   0.0    1     31     20   1.88
16:45:47   0.0   0.0   0.0  99.9   0.1    1     12     10   1.95
-------- ----- ----- ----- ----- ----- ---- ------ ------ ------
 Average   0.0   0.0   0.0 100.0   0.0  1.1   11.9   10.7   1.87
  StdDev   0.0   0.0   0.0   0.1   0.0  0.3   10.5    4.4   0.03
-------- ----- ----- ----- ----- ----- ---- ------ ------ ------
 Minimum   0.0   0.0   0.0  99.7   0.0  1.0    8.2    8.6   1.81
 Maximum   0.1   0.0   0.1 100.0   0.2  3.0   67.9   33.1   1.95
-------- ----- ----- ----- ----- ----- ---- ------ ------ ------
Summary:
System:   1.87 Watts on average with standard deviation 0.03

I'm using the NVME power saving patches, the intel DRM parameters from my prior post, the suggested settings from powertop (shown below), and an underclocked GPU (also shown below):

$ cat ~/.xinitrc
echo '0' | sudo tee '/proc/sys/kernel/nmi_watchdog'
echo '1500' | sudo tee '/proc/sys/vm/dirty_writeback_centisecs'
echo '1' | sudo tee '/sys/module/snd_hda_intel/parameters/power_save'
echo 'auto' | sudo tee '/sys/bus/i2c/devices/i2c-1/device/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:14.2/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:00.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:04.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:16.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1c.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1f.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1f.2/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1f.3/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:14.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:3a:00.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:3b:00.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1c.4/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1d.0/power/control'

sudo intel_gpu_frequency -e

The hardware is 9350 FHD, i5-6200U, 8GB RAM, 250GB Samsung NVMe SSD, and Broadcom BCM4350.

Last edited by robsmith11 (2017-01-26 17:16:28)

Offline

#1259 2017-01-26 19:39:11

OdinEidolon
Member
From: Belluno - Italy
Registered: 2011-01-31
Posts: 498

Re: Dell XPS 13 9350 Late 2015

robsmith11 wrote:

I measure power usage using a status bar script that samples

/sys/class/power_supply/BAT0/current_now
/sys/class/power_supply/BAT0/voltage_now

Here's the output of the powerstat script with minimum brightness and nothing besides X, dwm, and st running:

Running for 480.0 seconds (48 samples at 10.0 second intervals).
Power measurements will start in 0 seconds time.

  Time    User  Nice   Sys  Idle    IO  Run Ctxt/s  IRQ/s  Watts
16:40:17   0.0   0.0   0.0 100.0   0.0    1     10      9   1.89
16:40:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.92
16:40:37   0.0   0.0   0.0 100.0   0.0    1      9      9   1.91
16:40:47   0.0   0.0   0.0 100.0   0.0    1      8      9   1.85
16:40:57   0.0   0.0   0.0 100.0   0.0    1      9      9   1.84
16:41:07   0.0   0.0   0.0 100.0   0.0    1      9     10   1.87
16:41:17   0.0   0.0   0.0 100.0   0.0    1      9      9   1.85
16:41:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.88
16:41:37   0.0   0.0   0.0 100.0   0.0    1     10     11   1.88
16:41:47   0.0   0.0   0.0 100.0   0.0    1     10     10   1.89
16:41:57   0.0   0.0   0.0 100.0   0.0    1     10     12   1.84
16:42:07   0.0   0.0   0.0 100.0   0.0    1      9      9   1.84
16:42:17   0.0   0.0   0.0 100.0   0.0    1     11     12   1.88
16:42:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.86
16:42:37   0.0   0.0   0.0 100.0   0.0    1      9      9   1.86
16:42:47   0.0   0.0   0.0 100.0   0.0    1     10     10   1.84
16:42:57   0.0   0.0   0.0  99.9   0.1    1     13     10   1.84
16:43:07   0.0   0.0   0.0 100.0   0.0    1      9      9   1.88
16:43:17   0.0   0.0   0.0 100.0   0.0    1      9     10   1.87
16:43:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.85
16:43:37   0.0   0.0   0.0 100.0   0.0    1      9      9   1.85
16:43:47   0.0   0.0   0.0 100.0   0.0    1      8      9   1.81
16:43:57   0.0   0.0   0.0 100.0   0.0    1      9      9   1.87
16:44:07   0.0   0.0   0.0 100.0   0.0    1     10     12   1.87
16:44:17   0.0   0.0   0.0 100.0   0.0    1      9      9   1.88
16:44:27   0.0   0.0   0.0 100.0   0.0    3     11     10   1.86
16:44:37   0.1   0.0   0.1  99.7   0.2    1     68     33   1.91
16:44:47   0.0   0.0   0.0 100.0   0.0    1     11     11   1.85
16:44:57   0.0   0.0   0.0  99.9   0.1    1     10     10   1.88
16:45:07   0.0   0.0   0.0 100.0   0.0    1      9      9   1.86
16:45:17   0.0   0.0   0.0 100.0   0.0    1     10     10   1.86
16:45:27   0.0   0.0   0.0 100.0   0.0    1      8      9   1.84
16:45:37   0.0   0.0   0.1  99.9   0.0    1     31     20   1.88
16:45:47   0.0   0.0   0.0  99.9   0.1    1     12     10   1.95
-------- ----- ----- ----- ----- ----- ---- ------ ------ ------
 Average   0.0   0.0   0.0 100.0   0.0  1.1   11.9   10.7   1.87
  StdDev   0.0   0.0   0.0   0.1   0.0  0.3   10.5    4.4   0.03
-------- ----- ----- ----- ----- ----- ---- ------ ------ ------
 Minimum   0.0   0.0   0.0  99.7   0.0  1.0    8.2    8.6   1.81
 Maximum   0.1   0.0   0.1 100.0   0.2  3.0   67.9   33.1   1.95
-------- ----- ----- ----- ----- ----- ---- ------ ------ ------
Summary:
System:   1.87 Watts on average with standard deviation 0.03

I'm using the NVME power saving patches, the intel DRM parameters from my prior post, the suggested settings from powertop (shown below), and an underclocked GPU (also shown below):

$ cat ~/.xinitrc
echo '0' | sudo tee '/proc/sys/kernel/nmi_watchdog'
echo '1500' | sudo tee '/proc/sys/vm/dirty_writeback_centisecs'
echo '1' | sudo tee '/sys/module/snd_hda_intel/parameters/power_save'
echo 'auto' | sudo tee '/sys/bus/i2c/devices/i2c-1/device/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:14.2/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:00.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:04.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:16.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1c.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1f.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1f.2/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1f.3/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:14.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:3a:00.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:3b:00.0/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1c.4/power/control'
echo 'auto' | sudo tee '/sys/bus/pci/devices/0000:00:1d.0/power/control'

sudo intel_gpu_frequency -e

The hardware is 9350 FHD, i5-6200U, 8GB RAM, 250GB Samsung NVMe SSD, and Broadcom BCM4350.

I also use both the voltage*current files above and powerstat and they agree.

Out of curiosity, what's the output of

sudo intel_gpu_frequency -g

for your machine?

Hardware: 2016 Dell XPS15 - matte FullHD - i5-6300HQ - 32GB DDR4 - Nvidia GTX960M - Samsung 840EVO 250GB SSD - 56Wh
Software: Plasma 5 - rEFInd - linux-ck - preload - prelink - verynice - psd - bumblebee

Offline

#1260 2017-01-26 19:54:18

damige
Member
Registered: 2016-09-02
Posts: 40

Re: Dell XPS 13 9350 Late 2015

OdinEidolon wrote:

Out of curiosity, what's the output of

sudo intel_gpu_frequency -g

for your machine?

cur: 1050 MHz
min: 300 MHz
RP1: 300 MHz
max: 300 MHz

Not sure why it says cur 1050, its wrong. I force the gpu on lowest freq on boot.
I measured from a booted desktop (plasma, wifi connected, lowest brightness)

XPS 15 7590 | i9-9980HK | 32GB RAM | 512G NVME | Intel AX200 | NVIDIA 1650 | OLED 4K

Offline

#1261 2017-01-26 20:23:45

OdinEidolon
Member
From: Belluno - Italy
Registered: 2011-01-31
Posts: 498

Re: Dell XPS 13 9350 Late 2015

damige wrote:
OdinEidolon wrote:

Out of curiosity, what's the output of

sudo intel_gpu_frequency -g

for your machine?

cur: 1050 MHz
min: 300 MHz
RP1: 300 MHz
max: 300 MHz

Not sure why it says cur 1050, its wrong. I force the gpu on lowest freq on boot.
I measured from a booted desktop (plasma, wifi connected, lowest brightness)

Similar here:

╰─[XPS15af]─>  sudo intel_gpu_frequency -g
cur: 950 MHz
min: 350 MHz
RP1: 350 MHz
max: 350 MHz
╰─[XPS15af]─>  sudo intel_gpu_frequency -m; sudo intel_gpu_frequency -g
cur: 950 MHz
min: 350 MHz
RP1: 350 MHz
max: 950 MHz
╰─[XPS15af]─>  sudo intel_gpu_frequency -e; sudo intel_gpu_frequency -g
cur: 950 MHz
min: 350 MHz
RP1: 350 MHz
max: 350 MHz

Hardware: 2016 Dell XPS15 - matte FullHD - i5-6300HQ - 32GB DDR4 - Nvidia GTX960M - Samsung 840EVO 250GB SSD - 56Wh
Software: Plasma 5 - rEFInd - linux-ck - preload - prelink - verynice - psd - bumblebee

Offline

#1262 2017-01-27 01:02:47

robsmith11
Member
Registered: 2016-09-10
Posts: 23

Re: Dell XPS 13 9350 Late 2015

I see almost the same (also not sure why current is reported so high):

$ sudo intel_gpu_frequency -g
cur: 1000 MHz
min: 300 MHz
RP1: 300 MHz
max: 300 MHz

For me, the GPU frequency setting has no measurable impact on power usage at idle.  If I run a 3d benchmark, however, the performance and power usage difference is very apparent.

I can only guess that the i7 uses more power than the i5 at idle?  I wouldn't expect the difference to be that large though.  You can also see from the powerstat output that my system was more idle than yours; that probably explains at least 0.5W.

Offline

#1263 2017-01-27 08:43:07

OdinEidolon
Member
From: Belluno - Italy
Registered: 2011-01-31
Posts: 498

Re: Dell XPS 13 9350 Late 2015

robsmith11 wrote:

I see almost the same (also not sure why current is reported so high):

$ sudo intel_gpu_frequency -g
cur: 1000 MHz
min: 300 MHz
RP1: 300 MHz
max: 300 MHz

For me, the GPU frequency setting has no measurable impact on power usage at idle.  If I run a 3d benchmark, however, the performance and power usage difference is very apparent.

I can only guess that the i7 uses more power than the i5 at idle?  I wouldn't expect the difference to be that large though.  You can also see from the powerstat output that my system was more idle than yours; that probably explains at least 0.5W.

Maybe the fact is that you are getting ~ 10 IRQ/s which is CRAZY LOW, I can't see how that is even possible!
I get about 50 at idle even just from the TTY (not logged in in KDE). I may have more services and daemons running, that's for sure. I tend to enable thermald, irqbalance and all that stuff...

Keep in mind I have a different laptop so I would easily expect power differences, but 1.5 VS 5W at idle is really huge. Powertop really does not help much in identifying the culprit here.

I will try to investigate further.

Hardware: 2016 Dell XPS15 - matte FullHD - i5-6300HQ - 32GB DDR4 - Nvidia GTX960M - Samsung 840EVO 250GB SSD - 56Wh
Software: Plasma 5 - rEFInd - linux-ck - preload - prelink - verynice - psd - bumblebee

Offline

#1264 2017-01-27 09:56:52

robsmith11
Member
Registered: 2016-09-10
Posts: 23

Re: Dell XPS 13 9350 Late 2015

Well I did notice that even seemingly lightweight background services can impact power usage because they wake up the CPU.  For example, I switched my status bar from a shell script on 1-second poll to a compiled C binary on a 5-second poll and it significantly reduced my idle power usage.

Offline

#1265 2017-01-27 19:33:00

Asriel
Member
Registered: 2016-11-28
Posts: 46

Re: Dell XPS 13 9350 Late 2015

Andrew had refreshed his nvme patches - they apply well on 4.10-rc series now. Try - it may help.

I have patched my kernel and compiled it but my SSD is SATA not NVME - I've replaced 256 samsung with 1Tb Transcend so can not say if the patches work or not, but they do compile.

Offline

#1266 2017-02-02 10:01:22

skioda86
Member
Registered: 2015-05-19
Posts: 13

Re: Dell XPS 13 9350 Late 2015

Sorry guys....
(I have the configuration in signature)

I installed the linux-nvme 4.9.7-1 today and I start the powerstat. The result is:

$ sudo powerstat
Running for 300.0 seconds (30 samples at 10.0 second intervals).
Power measurements will start in 180 seconds time.

  Time    User  Nice   Sys  Idle    IO  Run Ctxt/s  IRQ/s Fork Exec Exit  Watts
10:40:21   0.7   0.0   0.5  98.6   0.3    1    159     90    3    2    2   2.55 
10:40:31   0.6   0.0   0.2  99.2   0.1    1    105     77    2    2    2   2.52 
10:40:41   0.6   0.0   0.3  98.9   0.3    1    136     80    2    2    2   2.56 
10:40:51   0.5   0.0   0.3  99.1   0.2    1    102     77    2    2    2   2.54 
10:41:01   0.8   0.0   0.3  98.6   0.3    1    153     89    2    2    2   2.57 
10:41:11   0.4   0.0   0.2  99.3   0.1    1     90     70    2    2    2   2.52 
10:41:21   0.7   0.0   0.2  98.7   0.4    1    145     83    2    2    2   2.59 
10:41:31   0.6   0.0   0.3  99.1   0.1    1     97     71    2    2    2   2.51 
10:41:41   0.7   0.0   0.3  98.7   0.3    1    154     86    2    2    2   2.71 
10:41:51   0.4   0.0   0.3  99.3   0.1    1     91     67    2    2    2   2.47 
10:42:01   0.9   0.0   0.3  98.5   0.4    1    148     84    2    2    2   2.52 
10:42:11   0.4   0.0   0.2  99.3   0.1    1     88     64    2    2    2   2.54 
10:42:21   0.8   0.0   0.3  98.6   0.3    1    142     78    2    2    2   2.65 
10:42:31   0.5   0.0   0.3  99.2   0.1    1     77     58    2    2    2   2.45 
10:42:41   0.7   0.0   0.3  98.7   0.3    1    122     66    2    2    2   2.54 
10:42:51   0.4   0.0   0.3  99.2   0.1    1     89     61    2    2    2   2.50 
10:43:01   0.7   0.0   0.3  98.7   0.3    1    134     77    2    2    2   3.46 
10:43:11   0.4   0.0   0.4  99.0   0.2    1     93     70    2    2    2   2.52 
10:43:21   0.7   0.0   0.4  98.6   0.4    1    141     80    2    2    2   2.57 
10:43:31   0.5   0.0   0.2  99.2   0.1    1     91     65    2    2    2   2.53 
10:43:41   0.9   0.0   0.5  98.2   0.4    1    165     86    4    4    4   2.58 
10:43:51   0.6   0.0   0.3  99.1   0.1    1     89     68    2    2    2   2.50 
10:44:01   0.6   0.0   0.3  98.9   0.2    1    125     76    2    2    2   2.56 
10:44:11   0.6   0.0   0.3  99.1   0.1    2     92     73    2    2    2   2.92 
10:44:21   0.7   0.0   0.3  98.7   0.3    1    148     85    3    2    2   2.53 
10:44:31   0.4   0.0   0.3  99.2   0.1    1     91     67    3    2    3   2.52 
10:44:41   0.7   0.0   0.3  98.7   0.3    1    133     79    2    2    3   2.59 
10:44:51   0.6   0.0   0.3  98.9   0.2    1    100     70    2    2    3   2.53 
10:45:01   0.7   0.0   0.3  98.7   0.3    1    127     80    2    2    2   2.47 
10:45:11   1.5   0.0   0.3  98.0   0.2    1    149     83    6    1    5   4.61 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
 Average   0.6   0.0   0.3  98.9   0.2  1.0  119.1   75.3  2.3  2.0  2.3   2.65 
  StdDev   0.2   0.0   0.1   0.3   0.1  0.2   26.9    8.5  0.8  0.4  0.7   0.41 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
 Minimum   0.4   0.0   0.2  98.0   0.1  1.0   77.4   57.7  2.0  1.0  2.0   2.45 
 Maximum   1.5   0.0   0.5  99.3   0.4  2.0  164.7   90.1  6.0  4.0  5.0   4.61 
-------- ----- ----- ----- ----- ----- ---- ------ ------ ---- ---- ---- ------ 
Summary:
System:   2.65 Watts on average with standard deviation 0.41

I try to check also:

$ sudo intel_gpu_frequency -g
cur: 1000 MHz
min: 300 MHz
RP1: 300 MHz
max: 1000 MHz

If I lock the most efficent frequency by

$ sudo intel_gpu_frequency -e


$ sudo intel_gpu_frequency -g
cur: 1000 MHz
min: 300 MHz
RP1: 300 MHz
max: 300 MHz

I configured my laptop with the robsmith11 in this post: #1250

Why I consume 2,6W in IDLE?
In the test, the pc was not used by me.
I must change another parameters?
Thanks

Last edited by skioda86 (2017-02-02 10:16:15)

Sorry for my BAD english but I'm a very Italian goat student!
Sorry for my poor information on Arch system but..... I'm a very Italian goat student! smile

[Hardware] = XPS13 (9350) - intel i5-6200U - 8GB Ram - Intel 8260 wifi - FHD - 256SSD Nvme  ||  [DE] = Gnome

Offline

#1267 2017-02-02 10:43:13

robsmith11
Member
Registered: 2016-09-10
Posts: 23

Re: Dell XPS 13 9350 Late 2015

skioda86, I'd say that 2.6 watts is pretty good given that your average CPU usage was 1.1% vs the <0.1% I had for my run.  I would guess that my system would have similar power usage at your CPU activity level.

If you want to get the idle power usage down further, then try disabling background services or switch to a lighter desktop environment.  But in real-world usage, it probably doesn't matter that much.  I never see sub-2W when I'm actually using my laptop.

Offline

#1268 2017-02-02 11:00:40

damige
Member
Registered: 2016-09-02
Posts: 40

Re: Dell XPS 13 9350 Late 2015

I have never seen 2.6w myself, not even <3w. I see you have the i5 (i have the i7) but i cant really imagen the cpu making such a big difference. Maybe it has to do with the iris 540 and its dedicated on die memory using more power?
Could you elaborate on you testing method? What are / aren't you running?

XPS 15 7590 | i9-9980HK | 32GB RAM | 512G NVME | Intel AX200 | NVIDIA 1650 | OLED 4K

Offline

#1269 2017-02-04 09:17:05

skioda86
Member
Registered: 2015-05-19
Posts: 13

Re: Dell XPS 13 9350 Late 2015

robsmith11 wrote:

skioda86, I'd say that 2.6 watts is pretty good given that your average CPU usage was 1.1% vs the <0.1% I had for my run.  I would guess that my system would have similar power usage at your CPU activity level.

If you want to get the idle power usage down further, then try disabling background services or switch to a lighter desktop environment.  But in real-world usage, it probably doesn't matter that much.  I never see sub-2W when I'm actually using my laptop.

Ok perfect...thanks!!!

damige wrote:

I have never seen 2.6w myself, not even <3w. I see you have the i5 (i have the i7) but i cant really imagen the cpu making such a big difference. Maybe it has to do with the iris 540 and its dedicated on die memory using more power?
Could you elaborate on you testing method? What are / aren't you running?

Yes...i have the i5 with the i915....
In my test the pc was just turned on (after modification wrote on previous post), I didn't touch any services in background and no open apps. The screen light was about 30%,

Sorry for my BAD english but I'm a very Italian goat student!
Sorry for my poor information on Arch system but..... I'm a very Italian goat student! smile

[Hardware] = XPS13 (9350) - intel i5-6200U - 8GB Ram - Intel 8260 wifi - FHD - 256SSD Nvme  ||  [DE] = Gnome

Offline

#1270 2017-02-04 11:12:39

belette
Member
Registered: 2014-11-17
Posts: 121

Re: Dell XPS 13 9350 Late 2015

anyone who use Dell Dock WD15 and has system hangs after upgrading to 4.9.6?
after updating to 4.9.6 when I connect the dock my system is unresponsive I have to hard shutdown the system...

EDIT: 4.9.8 seems a little bit better (at least it doesn't hang the system), need to do further tests..

Last edited by belette (2017-02-06 12:31:42)

Offline

#1271 2017-02-07 06:50:57

kang
Member
Registered: 2010-08-07
Posts: 83

Re: Dell XPS 13 9350 Late 2015

belette wrote:

anyone who use Dell Dock WD15 and has system hangs after upgrading to 4.9.6?
after updating to 4.9.6 when I connect the dock my system is unresponsive I have to hard shutdown the system...

EDIT: 4.9.8 seems a little bit better (at least it doesn't hang the system), need to do further tests..

pretty sure i did though im not sure which 4.9.x that was - it was very easy to trigger so im guessing same version as you. already upgraded also, did not have any more lockups after update.
ninja edit: eventually happened again on 4.9.8-nvme. going to upgrade bios from 1.4.12 to 1.4.13 in a  minute though ;-)

Last edited by kang (2017-02-13 23:59:56)

Offline

#1272 2017-02-10 18:59:31

Asriel
Member
Registered: 2016-11-28
Posts: 46

Re: Dell XPS 13 9350 Late 2015

Bios 1.4.13 is released.  Promised improvements with WD15 there. Still no help for DA200 though

Offline

#1273 2017-02-16 22:17:03

kang
Member
Registered: 2010-08-07
Posts: 83

Re: Dell XPS 13 9350 Late 2015

still having the same issue on 1.4.13 here (WD15) though i started being suspicious about the display port cable, ive to replace that  ;-)

Offline

#1274 2017-02-17 09:29:45

Stegamjay
Member
Registered: 2017-01-16
Posts: 8

Re: Dell XPS 13 9350 Late 2015

Hello,

For your information about Wifi card 802.11a/b/g/n  without binary blob/firmware on XPS
1) buy this card
https://www.thinkpenguin.com/gnu-linux/ … combo-card

2) just do it smile
https://www.laro.se/the-quest-for-an-xp … wifi-card/

regards...

Offline

#1275 2017-02-17 21:35:45

damige
Member
Registered: 2016-09-02
Posts: 40

Re: Dell XPS 13 9350 Late 2015

Just to let you guys know, I have made a repo for linux-nvme. Add to /etc/pacman.conf incase you wish to use this:

[linuxnvme]
SigLevel = Never
Server = http://linuxnvme.damige.net/repo

Github link: https://github.com/damige/linux-nvme

XPS 15 7590 | i9-9980HK | 32GB RAM | 512G NVME | Intel AX200 | NVIDIA 1650 | OLED 4K

Offline

Pages: 1 … 49 50 51 52 53 54

Topic closed

Board footer

Powered by FluxBB