Dell XPS 13 9350 Late 2015

Alad · 2016-07-09 23:45:05

davidgf wrote:

After updating to Fedora 24

Ask on the Fedora boards.

https://wiki.archlinux.org/index.php/Co … .2Aonly.2A

Last edited by Alad (2016-07-09 23:45:50)

ewaller · 2016-07-10 03:27:39

davidgf wrote:

H...After updating to Fedora 24, I get a "no devices detected" on Xorg....
Any ideas?

Ask on the Fedora Forums ? O.o

Edit: Missed Alad's response on the next page

Last edited by ewaller (2016-07-10 03:28:24)

davidgf · 2016-07-10 11:32:16

Nevermind guys, reinstalled it. There was something wrong with the f*ckin update.

Baba Tong I'm super interested in understanding what your laptop is doing. Mind pasting a bit more?

#> cd /sys/bus/pci/drivers/nvme/0000:3c:00.0/power (you should have another number like XXXX;XX:XX.X)

Then do a grep:

#> grep "" *

I'm sure you can go to a proper power state (like D3Hot)

Thanks!!!!

kgizdov · 2016-07-11 21:00:54

So, I tried 4.7-rc6 and the touchpad issue is still there for me. I just tried to build and install 4.7-rc7 and I could no longer boot. I don't recommend people trying it unless they have setup a backup boot option.

Baba Tong · 2016-07-11 21:40:51

davidgf wrote:

Nevermind guys, reinstalled it. There was something wrong with the f*ckin update.
Baba Tong I'm super interested in understanding what your laptop is doing. Mind pasting a bit more?
#> cd /sys/bus/pci/drivers/nvme/0000:3c:00.0/power (you should have another number like XXXX;XX:XX.X)
Then do a grep:
#> grep "" *
I'm sure you can go to a proper power state (like D3Hot)
Thanks!!!!

async:enabled
grep: autosuspend_delay_ms: Input/output error
control:auto
runtime_active_kids:0
runtime_active_time:281356
runtime_enabled:enabled
runtime_status:active
runtime_suspended_time:0
runtime_usage:1
wakeup:disabled
wakeup_abort_count:
wakeup_active:
wakeup_active_count:
wakeup_count:
wakeup_expire_count:
wakeup_last_time_ms:
wakeup_max_time_ms:
wakeup_prevent_sleep_time_ms:
wakeup_total_time_ms:

Haven't done any proper system updating since we've been discussing it in this thread. When I have time sometime this week, I'll do so and see whether that changes anything, in case it's a regression missed somewhere.

However, qball, on the previous page, reports roughly the same Wattage. Perhaps there is a hardware-component to this. Have we done any meaningful comparison between those who got the Dev edition and those who got the regular one?

Last edited by Baba Tong (2016-07-11 21:42:50)

frank604 · 2016-07-12 00:40:35

kgizdov wrote:

So, I tried 4.7-rc6 and the touchpad issue is still there for me. I just tried to build and install 4.7-rc7 and I could no longer boot. I don't recommend people trying it unless they have setup a backup boot option.

I use systemd boot. I have a stock arch kernel entry. Then my custom kernel entries. Alternatively I have my handy arch iso bootable usb and drive droid just in case. Then I can arch-chroot in and fix. Hope this helps for your future endeavors.

coconutoil · 2016-07-12 14:56:48

kgizdov wrote:

So, I tried 4.7-rc6 and the touchpad issue is still there for me. I just tried to build and install 4.7-rc7 and I could no longer boot. I don't recommend people trying it unless they have setup a backup boot option.

Are you using libinput? I've had no issues since 4.7rc6

c0re · 2016-07-14 10:22:39

Just want to share my impression that power consumption seems to be mainly bios related. My arch install is not tweaked, and I do not run tlp since it kills my wifi. Back when bios ver. 1.1.7 (iirc) was out, power usage reported by powerstat dropped to 3-4 watt. After an upgrade to 1.3.x, I was not able to get this below 6.8 watt, and I experienced rapid battery drain and horrible fan activity. Now with 1.4.4, at least the termperature and fan issues seem fixed, and battery life has significantly improved. Power consumption is down to 5.1 watt idle. Dell is guinea pig'ing us!

Edit: For testing, I now accepted all changes recommended by powertop. With wifi connected and screen brighness relatively low (but not lowest), several open ssh connections and chromium and firefox open, I am down to 4.3 watt again.

Last edited by c0re (2016-07-14 10:31:18)

cabellicar123 · 2016-07-15 16:51:55

For those having issues with the USB-C to HDMI output. Per this comment I purchased this simple USB-C to HDMI adapter and now HDMI output works fine for me even at high resolutions.

sender · 2016-07-15 17:41:56

I'm using this USB-C to HDMI cable at 2560x1440@60Hz, no problem at all.

kgizdov · 2016-07-16 12:17:31

Any idea why after the Mesa 12 and Vulkan updates I still get OpenGL 3 instead of 4.3?

glxinfo | grep -i opengl
OpenGL vendor string: Intel Open Source Technology Center
OpenGL renderer string: Mesa DRI Intel(R) HD Graphics 520 (Skylake GT2) 
OpenGL core profile version string: 4.3 (Core Profile) Mesa 12.0.1
OpenGL core profile shading language version string: 4.30
OpenGL core profile context flags: (none)
OpenGL core profile profile mask: core profile
OpenGL core profile extensions:
OpenGL version string: 3.0 Mesa 12.0.1
OpenGL shading language version string: 1.30
OpenGL context flags: (none)
OpenGL extensions:
OpenGL ES profile version string: OpenGL ES 3.1 Mesa 12.0.1
OpenGL ES profile shading language version string: OpenGL ES GLSL ES 3.10

fommil · 2016-07-16 18:03:13

I got a new XPS 13 this week, upgrading from a 2 or 3 year old model.

Although the console starts up fine, when I go to any of the other consoles (e.g. alt-2 through alt-5) the screen is completely corrupted. I can log in on each of these screens but it's like every character I type is just coming through as noise.

Is anybody else experiencing this?

It's quite painful because if I do anything bad on the 1st console, then I have no way to recover except to reboot.

set92 · 2016-07-17 12:31:19

c0re wrote:

Just want to share my impression that power consumption seems to be mainly bios related. My arch install is not tweaked, and I do not run tlp since it kills my wifi. Back when bios ver. 1.1.7 (iirc) was out, power usage reported by powerstat dropped to 3-4 watt. After an upgrade to 1.3.x, I was not able to get this below 6.8 watt, and I experienced rapid battery drain and horrible fan activity. Now with 1.4.4, at least the termperature and fan issues seem fixed, and battery life has significantly improved. Power consumption is down to 5.1 watt idle. Dell is guinea pig'ing us!
Edit: For testing, I now accepted all changes recommended by powertop. With wifi connected and screen brighness relatively low (but not lowest), several open ssh connections and chromium and firefox open, I am down to 4.3 watt again.

What changes recommended by powertop? I tried to apply all the things from https://wiki.archlinux.org/index.php/Po … _hibernate but Im still at 8W without chrome running.

zetxx · 2016-07-17 17:01:35

i talked with a guy from dell support concerning power consuption ... it matters resetting bios .. here it is the thread https://plus.google.com/+DougIpperciel/ … FiTJueb9Sd

Last edited by zetxx (2016-07-17 17:02:01)

frank604 · 2016-07-17 18:54:11

fommil wrote:

Although the console starts up fine, when I go to any of the other consoles (e.g. alt-2 through alt-5) the screen is completely corrupted. I can log in on each of these screens but it's like every character I type is just coming through as noise.

NMI, locale? console font? Did this happen after base install? Does this happen before/after X? Is this on your laptop screen or external? etc etc.. Also, this issue is not device specific so I recommend you open a thread about it including as much information as possible about your system and your environment setup in the Newbie Corner for more exposure to this issue.

set92 wrote:

What changes recommended by powertop? I tried to apply all the things from https://wiki.archlinux.org/index.php/Po … _hibernate but Im still at 8W without chrome running.

That link is about different suspend/hibernate. If you want powertop to autotune at startup you can use this systemd service to do so. Just cp to /etc/systemd/system and enable/start it.

Last edited by frank604 (2016-07-17 18:55:30)

diraimondo · 2016-07-18 11:39:26

zetxx wrote:

i talked with a guy from dell support concerning power consuption ... it matters resetting bios .. here it is the thread https://plus.google.com/+DougIpperciel/ … FiTJueb9Sd

Just did a BIOS settings reset. After switching on AHCI mode (mandatory to boot Linux), I can't see any difference. I don't know what is talking about the Dell developer...
I can't believe that Dell was not able to fix this weird problem with the NVMe drives.

$ sudo smartctl -c /dev/nvme0
The NVMe drive is not able to exploit the last two (more power-saving) power states: 
smartctl 6.5 2016-05-07 r4318 [x86_64-linux-4.7.0-rc6-mainline] (local build)
Copyright (C) 2002-16, Bruce Allen, Christian Franke, www.smartmontools.org

=== START OF INFORMATION SECTION ===
Firmware Updates (0x06):            3 Slots
Optional Admin Commands (0x0017):   Security Format Frmw_DL *Other*
Optional NVM Commands (0x001f):     Comp Wr_Unc DS_Mngmt Wr_Zero Sav/Sel_Feat
Maximum Data Transfer Size:         32 Pages

Supported Power States
St Op     Max   Active     Idle   RL RT WL WT  Ent_Lat  Ex_Lat
 0 +     6.00W       -        -    0  0  0  0        5       5
 1 +     4.20W       -        -    1  1  1  1       30      30
 2 +     3.10W       -        -    2  2  2  2      100     100
 3 -   0.0700W       -        -    3  3  3  3      500    5000
 4 -   0.0050W       -        -    4  4  4  4     2000   22000

Supported LBA Sizes (NSID 0x1)
Id Fmt  Data  Metadt  Rel_Perf
 0 +     512       0         0

Powertop shows that I'm able to reach the PC8 package status but this doesn't help to get a lower power discharge rate.

ksim · 2016-07-18 11:57:04

diraimondo wrote:

I can't believe that Dell was not able to fix this weird problem with the NVMe drives.

May be because this should be done by Samsung or Intel or someone who designed hardware.

kang · 2016-07-18 12:25:33

I suspect the DELL guy means that the CPU package C states (as in pc8) weren't reached due to the NVME bug, not that NVME will reach it's own lower power states (1/2/3/4).
Note that PC8 is the deepest state for Skylake according to http://www.intel.com/content/www/us/en/ … vol-1.html
According to the same doc, things like PSR affect the deepest package state you can reach. With a QHD laptop and PSR turned off you cannot go lower than PC6 (that's the case for a lot of us since PSR cause flickering still on Linux).

Aside from that being able to reach lower NVME power states would be nice.

diraimondo · 2016-07-18 12:40:32

kang wrote:

I suspect the DELL guy means that the CPU package C states (as in pc8) weren't reached due to the NVME bug, not that NVME will reach it's own lower power states (1/2/3/4).
Note that PC8 is the deepest state for Skylake according to http://www.intel.com/content/www/us/en/ … vol-1.html
According to the same doc, things like PSR affect the deepest package state you can reach. With a QHD laptop and PSR turned off you cannot go lower than PC6 (that's the case for a lot of us since PSR cause flickering still on Linux).

I'm not sure why, but I can reach PC8 on powertop using a 3200x1800 resolution and without any PSR-related options on kernel command-line (only 'pcie_aspm=force i915.enable_rc6=7 i915.enable_fbc=0') with kernel 4.7rc6.

zetxx · 2016-07-18 13:44:06

i dont know what resetting bios did with the xps but it is much cooler after that.

frank604 · 2016-07-19 00:54:32

I added a new bug report for the intel drm nightly branch here. This is in regards to the ghosting/trailing/artifacts of mouse cursor when using dual monitors off the WD15. Just in case anyone else is experiencing it and wants to follow/add on top.

On a side note, I have at the recommendation of others removed xf86-video-intel and am now using native modesetting provided by the kernel. Reddit thread here.

zetxx · 2016-07-19 10:40:18

frank604 wrote:

I added a new bug report for the intel drm nightly branch here. This is in regards to the ghosting/trailing/artifacts of mouse cursor when using dual monitors off the WD15. Just in case anyone else is experiencing it and wants to follow/add on top.
On a side note, I have at the recommendation of others removed xf86-video-intel and am now using native modesetting provided by the kernel. Reddit thread here.

did you try it ?
forget about this question

edit: tried it but lots of tear and chromium rendering problem .... and wrong monitor resolutions calculation ... went back to intel drv

Last edited by zetxx (2016-07-19 18:31:44)

frank604 · 2016-07-19 19:35:09

zetxx wrote:

frank604 wrote:
I added a new bug report for the intel drm nightly branch here. This is in regards to the ghosting/trailing/artifacts of mouse cursor when using dual monitors off the WD15. Just in case anyone else is experiencing it and wants to follow/add on top.
On a side note, I have at the recommendation of others removed xf86-video-intel and am now using native modesetting provided by the kernel. Reddit thread here.
did you try it ?
forget about this question
edit: tried it but lots of tear and chromium rendering problem .... and wrong monitor resolutions calculation ... went back to intel drv

Strange. I didn't experience any of the above. I'm still testing the native modesetting to gauge if this method is more stable. Just remember to completely remove xf86-video-intel and its conf files to give this a go.

zetxx · 2016-07-19 20:31:32

uep, but

[  268.716471] ------------[ cut here ]------------
[  268.716497] WARNING: CPU: 2 PID: 2298 at drivers/gpu/drm/i915/intel_uncore.c:804 __unclaimed_reg_debug+0x6b/0x80 [i915]
[  268.716499] Unclaimed register detected after reading register 0x65f10
[  268.716501] Modules linked in: rfcomm sha256_ssse3 sha256_generic hmac drbg ansi_cprng ctr ccm uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_v4l2 videobuf2_core videodev btusb btrtl media fuse usbhid joydev mousedev hid_multitouch bnep arc4 nvram msr snd_hda_codec_hdmi dell_led iTCO_wdt iTCO_vendor_support i2c_designware_platform i2c_designware_core dell_wmi nls_iso8859_1 snd_soc_skl snd_soc_skl_ipc snd_soc_sst_ipc snd_soc_sst_dsp dell_laptop snd_hda_codec_realtek x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel iwlmvm nls_cp437 dell_smbios snd_hda_ext_core snd_hda_codec_generic kvm vfat mac80211 fat irqbypass dcdbas snd_soc_sst_match crct10dif_pclmul crc32_pclmul snd_soc_core crc32c_intel ghash_clmulni_intel snd_compress iwlwifi snd_pcm_dmaengine ac97_bus snd_hda_intel snd_hda_codec
[  268.716541]  aesni_intel snd_hda_core input_leds aes_x86_64 led_class lrw snd_hwdep gf128mul glue_helper ablk_helper cryptd serio_raw pcspkr snd_pcm cfg80211 snd_timer rtsx_pci_ms snd memstick soundcore i2c_i801 mei_me mei shpchp idma64 intel_lpss_pci intel_pch_thermal processor_thermal_device intel_soc_dts_iosf fan thermal wmi i2c_hid hid hci_uart btbcm btqca pinctrl_sunrisepoint btintel pinctrl_intel bluetooth rfkill intel_lpss_acpi intel_lpss int3403_thermal int340x_thermal_zone battery intel_hid int3400_thermal acpi_thermal_rel acpi_als sparse_keymap evdev kfifo_buf mac_hid industrialio acpi_pad ac tpm_tis fjes tpm sch_fq_codel vboxnetflt(O) vboxnetadp(O) pci_stub vboxpci(O) vboxdrv(O) ip_tables x_tables ext4 crc16 jbd2 mbcache rtsx_pci_sdmmc mmc_core atkbd libps2 ahci libahci libata xhci_pci
[  268.716586]  rtsx_pci xhci_hcd scsi_mod usbcore usb_common i8042 serio i915 video button intel_gtt i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm nvme nvme_core

frank604 · 2016-07-19 21:07:14

zetxx wrote:

uep, but

[  268.716471] ------------[ cut here ]------------
[  268.716497] WARNING: CPU: 2 PID: 2298 at drivers/gpu/drm/i915/intel_uncore.c:804 __unclaimed_reg_debug+0x6b/0x80 [i915]
[  268.716499] Unclaimed register detected after reading register 0x65f10
[  268.716501] Modules linked in: rfcomm sha256_ssse3 sha256_generic hmac drbg ansi_cprng ctr ccm uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_v4l2 videobuf2_core videodev btusb btrtl media fuse usbhid joydev mousedev hid_multitouch bnep arc4 nvram msr snd_hda_codec_hdmi dell_led iTCO_wdt iTCO_vendor_support i2c_designware_platform i2c_designware_core dell_wmi nls_iso8859_1 snd_soc_skl snd_soc_skl_ipc snd_soc_sst_ipc snd_soc_sst_dsp dell_laptop snd_hda_codec_realtek x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel iwlmvm nls_cp437 dell_smbios snd_hda_ext_core snd_hda_codec_generic kvm vfat mac80211 fat irqbypass dcdbas snd_soc_sst_match crct10dif_pclmul crc32_pclmul snd_soc_core crc32c_intel ghash_clmulni_intel snd_compress iwlwifi snd_pcm_dmaengine ac97_bus snd_hda_intel snd_hda_codec
[  268.716541]  aesni_intel snd_hda_core input_leds aes_x86_64 led_class lrw snd_hwdep gf128mul glue_helper ablk_helper cryptd serio_raw pcspkr snd_pcm cfg80211 snd_timer rtsx_pci_ms snd memstick soundcore i2c_i801 mei_me mei shpchp idma64 intel_lpss_pci intel_pch_thermal processor_thermal_device intel_soc_dts_iosf fan thermal wmi i2c_hid hid hci_uart btbcm btqca pinctrl_sunrisepoint btintel pinctrl_intel bluetooth rfkill intel_lpss_acpi intel_lpss int3403_thermal int340x_thermal_zone battery intel_hid int3400_thermal acpi_thermal_rel acpi_als sparse_keymap evdev kfifo_buf mac_hid industrialio acpi_pad ac tpm_tis fjes tpm sch_fq_codel vboxnetflt(O) vboxnetadp(O) pci_stub vboxpci(O) vboxdrv(O) ip_tables x_tables ext4 crc16 jbd2 mbcache rtsx_pci_sdmmc mmc_core atkbd libps2 ahci libahci libata xhci_pci
[  268.716586]  rtsx_pci xhci_hcd scsi_mod usbcore usb_common i8042 serio i915 video button intel_gtt i2c_algo_bit drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops drm nvme nvme_core

Not sure what's going on with that. That is what you are seeing when you remove xf86-video-intel and switch to native? I'm on a different kernel branch... maybe that's why. The Intel branch is more smoother than mainline (at least for my setup). This is because they push their fixes into their branch first and it takes a few major version bump on mainline (regular linux kernel) to see these changes.

edit: native modesetting... not sure if it is in fact better atm. I don't notice the diff.

Last edited by frank604 (2016-07-19 21:42:35)

Arch Linux

#776 2016-07-09 23:45:05

Re: Dell XPS 13 9350 Late 2015

#777 2016-07-10 03:27:39

Re: Dell XPS 13 9350 Late 2015

#778 2016-07-10 11:32:16

Re: Dell XPS 13 9350 Late 2015

#779 2016-07-11 21:00:54

Re: Dell XPS 13 9350 Late 2015

#780 2016-07-11 21:40:51

Re: Dell XPS 13 9350 Late 2015

#781 2016-07-12 00:40:35

Re: Dell XPS 13 9350 Late 2015

#782 2016-07-12 14:56:48

Re: Dell XPS 13 9350 Late 2015

#783 2016-07-14 10:22:39

Re: Dell XPS 13 9350 Late 2015

#784 2016-07-15 16:51:55

Re: Dell XPS 13 9350 Late 2015

#785 2016-07-15 17:41:56

Re: Dell XPS 13 9350 Late 2015

#786 2016-07-16 12:17:31

Re: Dell XPS 13 9350 Late 2015

#787 2016-07-16 18:03:13

Re: Dell XPS 13 9350 Late 2015

#788 2016-07-17 12:31:19

Re: Dell XPS 13 9350 Late 2015

#789 2016-07-17 17:01:35

Re: Dell XPS 13 9350 Late 2015

#790 2016-07-17 18:54:11

Re: Dell XPS 13 9350 Late 2015

#791 2016-07-18 11:39:26

Re: Dell XPS 13 9350 Late 2015

#792 2016-07-18 11:57:04

Re: Dell XPS 13 9350 Late 2015

#793 2016-07-18 12:25:33

Re: Dell XPS 13 9350 Late 2015

#794 2016-07-18 12:40:32

Re: Dell XPS 13 9350 Late 2015

#795 2016-07-18 13:44:06

Re: Dell XPS 13 9350 Late 2015

#796 2016-07-19 00:54:32

Re: Dell XPS 13 9350 Late 2015

#797 2016-07-19 10:40:18

Re: Dell XPS 13 9350 Late 2015

#798 2016-07-19 19:35:09

Re: Dell XPS 13 9350 Late 2015

#799 2016-07-19 20:31:32

Re: Dell XPS 13 9350 Late 2015

#800 2016-07-19 21:07:14

Re: Dell XPS 13 9350 Late 2015

Board footer