You are not logged in.

Pages: 1

#1 2012-08-11 01:09:46

TheJJ
Member
Registered: 2011-09-29
Posts: 11
Website

3.3.5+ power regression still present

This thread shall hunt down the power regression bug first mentioned here:
https://bugs.archlinux.org/task/29850

The load average bug has been fixed, but power usage is still ~double than 3.3.4.

As the doubled power usage really is annoying, maybe we can track back the origin somehow.
The origin may be in the kernel (https://lwn.net/Articles/496154/) or at a complete different location we haven't thought of yet.

To discover whether this is hardware related, we should first collect and compare specs of affected and non-affected laptops.


Please report your status in the following form:

Affected?, Laptop brand, CPU-vendor, CPU-name, GPU-name (+kernel driver), chipset, current kernel version

Power usage statistics (wattage before -> after) would also be nice.


Ask friends/etc if they have the same issues, and share this please.
Enough attention is necessary to get this fixed!

Last edited by TheJJ (2012-08-11 01:13:17)

Offline

#2 2012-08-11 01:10:26

TheJJ
Member
Registered: 2011-09-29
Posts: 11
Website

Re: 3.3.5+ power regression still present

Hardware listing

AFFECTED:

  • Lenovo X220t, Intel Core i5-2520M, HD3000 (i915), Intel QM67, kernel 3.5.1-1

  • Lenovo T420i, Intel Core i7 Sandy Bridge 2620M, Intel HD 3000 & Nvidia NVS4200 Optimus, kernel 3.3.7-1-ck

  • Lenovo G560, Intel Core i5-430M, nVidia 310M non-optimus (nvidia 304.32-1), kernel 3.4.7-1 ARCH

  • Gigabyte M720-US3 MB, Phenom II X4 965, ATI HD4670 (driver unknown), kernel unknown

NOT Affected:

  • none known yet

Last edited by TheJJ (2012-08-11 12:22:20)

Offline

#3 2012-08-11 02:37:59

donniezazen
Member
From: Salt Lake City
Registered: 2011-06-24
Posts: 671
Website

Re: 3.3.5+ power regression still present

Lenovo T420i, Intel Core i7 Sandy Bridge 2620M, Intel HD 3000 & Nvidia NVS4200 Optimus, 3.3.7-1-ck

Any kernel post 3.3 increases power consumption by about 5W and temperature is increased by at least 10C.

github | blog

Offline

#4 2012-08-11 08:37:15

Cristos
Member
From: Poland
Registered: 2011-10-22
Posts: 3
Website

Re: 3.3.5+ power regression still present

Lenovo G560, Intel Core i5-430M, nVidia 310M non-optimus, 4GB RAM.
kernel 3.4-7-1 ARCH, nvidia proprietary driver 304.32-1

lspci output:

00:00.0 Host bridge: Intel Corporation Core Processor DRAM Controller (rev 02)
00:01.0 PCI bridge: Intel Corporation Core Processor PCI Express x16 Root Port (rev 02)
00:16.0 Communication controller: Intel Corporation 5 Series/3400 Series Chipset HECI Controller (rev 06)
00:1a.0 USB controller: Intel Corporation 5 Series/3400 Series Chipset USB2 Enhanced Host Controller (rev 05)
00:1b.0 Audio device: Intel Corporation 5 Series/3400 Series Chipset High Definition Audio (rev 05)
00:1c.0 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 1 (rev 05)
00:1c.1 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 2 (rev 05)
00:1c.2 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 3 (rev 05)
00:1c.4 PCI bridge: Intel Corporation 5 Series/3400 Series Chipset PCI Express Root Port 5 (rev 05)
00:1d.0 USB controller: Intel Corporation 5 Series/3400 Series Chipset USB2 Enhanced Host Controller (rev 05)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev a5)
00:1f.0 ISA bridge: Intel Corporation Mobile 5 Series Chipset LPC Interface Controller (rev 05)
00:1f.2 SATA controller: Intel Corporation 5 Series/3400 Series Chipset 4 port SATA AHCI Controller (rev 05)
00:1f.3 SMBus: Intel Corporation 5 Series/3400 Series Chipset SMBus Controller (rev 05)
00:1f.6 Signal processing controller: Intel Corporation 5 Series/3400 Series Chipset Thermal Subsystem (rev 05)
01:00.0 VGA compatible controller: NVIDIA Corporation GT218 [GeForce 310M] (rev a2)
01:00.1 Audio device: NVIDIA Corporation High Definition Audio Controller (rev a1)
06:00.0 Network controller: Broadcom Corporation BCM4313 802.11b/g/n Wireless LAN Controller (rev 01)
07:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast Ethernet controller (rev 02)
ff:00.0 Host bridge: Intel Corporation Core Processor QuickPath Architecture Generic Non-core Registers (rev 02)
ff:00.1 Host bridge: Intel Corporation Core Processor QuickPath Architecture System Address Decoder (rev 02)
ff:02.0 Host bridge: Intel Corporation Core Processor QPI Link 0 (rev 02)
ff:02.1 Host bridge: Intel Corporation Core Processor QPI Physical 0 (rev 02)
ff:02.2 Host bridge: Intel Corporation Core Processor Reserved (rev 02)
ff:02.3 Host bridge: Intel Corporation Core Processor Reserved (rev 02)

lsusb output:

Bus 001 Device 002: ID 8087:0020 Intel Corp. Integrated Rate Matching Hub
Bus 002 Device 002: ID 8087:0020 Intel Corp. Integrated Rate Matching Hub
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
Bus 001 Device 003: ID 10ab:0816 USI Co., Ltd 
Bus 002 Device 003: ID 5986:0148 Acer, Inc

additional boot options:

nmi_watchdog=0 pcie_ports=native acpi_osi=Linux snd_hda_intel.position_fix=1 snd_hda_intel.enable_msi=1

Idle temp. is about 10*C higher than with pre 3.3.2 kernel.

Offline

#5 2012-08-11 10:14:35

Roken
Member
From: South Wales, UK
Registered: 2012-01-16
Posts: 1,290

Re: 3.3.5+ power regression still present

Self build - Gigabyte M720-US3 MB, Phenom II 965, ATI HD4670.

from lshw:

H/W path      Device      Class       Description
=================================================
                          system      M720-US3 ()
/0                        bus         M720-US3
/0/0                      memory      128KiB BIOS
/0/3                      processor   AMD Phenom(tm) II X4 965 Processor
/0/3/a                    memory      128KiB L1 cache
/0/3/b                    memory      512KiB L2 cache
/0/1a                     memory      8GiB System Memory
/0/1a/0                   memory      2GiB DIMM 800 MHz (1.2 ns)
/0/1a/1                   memory      2GiB DIMM 800 MHz (1.2 ns)
/0/1a/2                   memory      2GiB DIMM 800 MHz (1.2 ns)
/0/1a/3                   memory      2GiB DIMM 800 MHz (1.2 ns)
/0/4                      memory      RAM memory
/0/1                      bridge      MCP78S [GeForce 8200] LPC Bridge
/0/1.1                    bus         MCP78S [GeForce 8200] SMBus
/0/1.2                    memory      RAM memory
/0/1.4                    memory      RAM memory
/0/2                      bus         MCP78S [GeForce 8200] OHCI USB 1.1 Controller
/0/2.1                    bus         MCP78S [GeForce 8200] EHCI USB 2.0 Controller
/0/5                      bus         MCP78S [GeForce 8200] OHCI USB 1.1 Controller
/0/4.1                    bus         MCP78S [GeForce 8200] EHCI USB 2.0 Controller
/0/6                      storage     MCP78S [GeForce 8200] IDE
/0/7                      multimedia  MCP72XE/MCP72P/MCP78U/MCP78S High Definition Audio
/0/8                      bridge      MCP78S [GeForce 8200] PCI Bridge
/0/8/7        wlan0       network     RT2561/RT61 802.11g PCI
/0/8/8                    multimedia  CA0106 Soundblaster
/0/9                      storage     MCP78S [GeForce 8200] AHCI Controller
/0/10                     bridge      MCP78S [GeForce 8200] PCI Express Bridge
/0/10/0                   display     RV730XT [Radeon HD 4670]
/0/10/0.1                 multimedia  RV710/730 HDMI Audio [Radeon HD 4000 series]
/0/14                     bridge      MCP78S [GeForce 8200] PCI Bridge
/0/14/0                   network     RTL8111/8168B PCI Express Gigabit Ethernet controller
/0/100                    bridge      Family 10h Processor HyperTransport Configuration
/0/101                    bridge      Family 10h Processor Address Map
/0/102                    bridge      Family 10h Processor DRAM Controller
/0/103                    bridge      Family 10h Processor Miscellaneous Control
/0/104                    bridge      Family 10h Processor Link Control
/0/a          scsi0       storage     
/0/a/0.0.0    /dev/sda    disk        250GB HDT722525DLA380
/0/a/0.0.0/1  /dev/sda1   volume      94MiB EXT4 volume
/0/a/0.0.0/2  /dev/sda2   volume      10GiB Linux swap volume
/0/a/0.0.0/3  /dev/sda3   volume      163GiB EXT4 volume
/0/a/0.0.0/4  /dev/sda4   volume      59GiB EXT4 volume
/0/b          scsi1       storage     
/0/b/0.0.0    /dev/cdrom  disk        DVDRAM GSA-H66N
/0/c          scsi2       storage     
/0/c/0.0.0    /dev/sdb    disk        500GB Hitachi HDP72505
/0/c/0.0.0/1  /dev/sdb1   volume      100MiB Windows NTFS volume
/0/c/0.0.0/2  /dev/sdb2   volume      149GiB Windows NTFS volume
/0/c/0.0.0/3  /dev/sdb3   volume      97GiB EXT4 volume
/0/c/0.0.0/4  /dev/sdb4   volume      217GiB HPFS/NTFS partition
/0/d          scsi5       storage     
/0/d/0.0.0    /dev/sdc    disk        500GB Hitachi HDP72505
/0/d/0.0.0/1  /dev/sdc1   volume      445GiB Windows NTFS volume
/0/d/0.0.0/2  /dev/sdc2   volume      20GiB Windows NTFS volume

Ryzen 5900X 12 core/24 thread - RTX 3090 FE 24 Gb, Asus B550-F Gaming MB, 128Gb Corsair DDR4, Cooler Master N300 chassis, 5 HD (2 NvME PCI, 4SSD) + 1 x optical.
Linux user #545703

/ is the root of all problems.

Offline

#6 2012-08-14 08:20:03

donniezazen
Member
From: Salt Lake City
Registered: 2011-06-24
Posts: 671
Website

Re: 3.3.5+ power regression still present

I thought to give stock kernel a try, heck I am running nothing out of ordinary. My temp is over 70C and power is 24W.

Vq0M6s.png

github | blog

Offline

#7 2012-08-15 11:02:10

niklaas
Member
From: Vienna
Registered: 2012-07-24
Posts: 26

Re: 3.3.5+ power regression still present

donniezazen wrote:

I thought to give stock kernel a try, heck I am running nothing out of ordinary. My temp is over 70C and power is 24W.

I had this problem before running arch on a Lenovo Thinkpad T420. After a fresh install of arch one month ago, I had a battery life of round about 8 hours. I achieved this with the tweaks mentioned here: https://wiki.archlinux.org/index.php/Le … _arguments After upgrading the kernel I had a quite higher power consumption of up to 24W -- I think before it must have been between 10W and 15W. I'm not very sure about the exact value.

Since then I tried to tweak battery life with powerdown ( https://aur.archlinux.org/packages.php?ID=57421 ) by taylorchu. This helped a bit.

Now, I just figured out that adding

drm.vblankoffdelay = 1

as described under https://wiki.ubuntu.com/Kernel/PowerMan … vingTweaks really helped a lot and reduced the power consumption of 17W I had one hour before to 9W - 12W.

So this is my configuration:

niklaas@len-t420 ~ % grep i915 /boot/grub/menu.lst
kernel /vmlinuz-linux root=/dev/mapper/arch-root cryptdevice=/dev/sda5:luks ro i915.i915_enable_rc6=1 i915.i915_enable_fbc=1 i915.lvds_downclock=1 drm.vblankoffdelay=1 resume=/dev/mapper/arch-root resume_offset=40960
niklaas@len-t420 ~ % sudo powernow
Password: 
[powerdown] BAT0 using 10.235 watts
Battery 0: Discharging, 73%, 06:31:49 remaining

 :: Detail
 -> /proc/sys/kernel/nmi_watchdog: 0
 -> /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor: powersave
 -> /sys/devices/system/cpu/cpu1/cpufreq/scaling_governor: powersave
 -> /sys/devices/system/cpu/cpu2/cpufreq/scaling_governor: powersave
 -> /sys/devices/system/cpu/cpu3/cpufreq/scaling_governor: powersave
 -> /sys/module/pcie_aspm/parameters/policy: [default] performance powersave 
 -> /proc/sys/vm/laptop_mode: 5
 -> /proc/sys/vm/dirty_ratio: 90
 -> /proc/sys/vm/dirty_background_ratio: 1
 -> /proc/sys/vm/dirty_expire_centisecs: 60000
 -> /proc/sys/vm/dirty_writeback_centisecs: 60000
 -> /dev/sda3 readahead: 4096
 -> /sys/class/scsi_host/host0/link_power_management_policy: min_power
 -> /sys/class/scsi_host/host1/link_power_management_policy: min_power
 -> /sys/class/scsi_host/host2/link_power_management_policy: min_power
 -> /sys/class/scsi_host/host3/link_power_management_policy: min_power
 -> /sys/class/scsi_host/host4/link_power_management_policy: min_power
 -> /sys/class/scsi_host/host5/link_power_management_policy: min_power
 -> /sys/module/snd_hda_intel/parameters/power_save: 1
 -> /sys/module/snd_hda_intel/parameters/power_save_controller: Y
 -> /sys/class/backlight/acpi_video0/brightness: 11
 -> /sys/class/backlight/acpi_video1/brightness: 9
 -> /sys/module/i915/parameters/i915_enable_rc6: 1
 -> wlan0 power management: on
 -> uvcvideo not loaded
 -> nouveau not loaded
 -> btusb not loaded
 -> bluetooth loaded
niklaas@len-t420 ~ % uname -a
Linux len-t420 3.4.8-1-ARCH #1 SMP PREEMPT Thu Aug 9 21:21:22 CEST 2012 x86_64 GNU/Linux
niklaas@len-t420 ~ % glxinfo | grep -i render
direct rendering: Yes
OpenGL renderer string: Mesa DRI Intel(R) Sandybridge Mobile 
    GL_NV_conditional_render, GL_ARB_ES2_compatibility,
niklaas@len-t420 ~ % optirun glxinfo | grep -i render
direct rendering: Yes
OpenGL renderer string: NVS 4200M/PCIe/SSE2
    GL_NV_conditional_render, GL_NV_copy_depth_to_color, GL_NV_copy_image, 
    GL_NV_parameter_buffer_object2, GL_NV_path_rendering, 
    GL_NVX_conditional_render, GL_NVX_gpu_memory_info, 
    GL_OES_depth_texture, GL_OES_element_index_uint, GL_OES_fbo_render_mipmap,

So using the combination of powerdown and the "i915-tweaks" gives quite a good result. (But still I think it could be optimized by resolving the issue in this thread.)

Maybe this helps a bit.

Offline

#8 2012-08-18 15:32:08

donniezazen
Member
From: Salt Lake City
Registered: 2011-06-24
Posts: 671
Website

Re: 3.3.5+ power regression still present

@niklaas I do use boot kernel parameters. I have not tried powerdown. I will give it a try. What DE/WM do you use? Power consumption tends to be high on DEs then simple WMs. Does powerdown interferes with powertop and laptop-mode-tool?

UPDATE:- linux-lts has brought down my system temperature and power usage to less than 50C and 9-10W respectively. It's working great.

Last edited by donniezazen (2012-08-18 22:09:23)

github | blog

Offline

#9 2012-08-19 09:12:56

niklaas
Member
From: Vienna
Registered: 2012-07-24
Posts: 26

Re: 3.3.5+ power regression still present

donniezazen wrote:

@niklaas I do use boot kernel parameters. I have not tried powerdown. I will give it a try. What DE/WM do you use? Power consumption tends to be high on DEs then simple WMs. Does powerdown interferes with powertop and laptop-mode-tool?

I'm using Xfce. Not with powertop but I think with laptop-mode-tools.

You could also try domac's scripts for pm-utils (git: https://github.com/domac/archlinux/tree … pm/power.d).

Last edited by niklaas (2012-08-19 09:14:42)

Offline

Pages: 1

Board footer

Powered by FluxBB