No C-State Deeper than C3 utilized by Kaby Lake 7820HQ in Precision 5520
- Xenial (16.04)
- Bug #1672439
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
linux (Ubuntu) |
Fix Released
|
Medium
|
Unassigned | ||
Xenial |
Fix Released
|
Medium
|
Kai-Heng Feng |
Bug Description
My processor does not appear to be utilizing and idling states deeper than C3.
ProblemType: Bug
DistroRelease: Ubuntu 16.04
Package: linux-image-
ProcVersionSign
Uname: Linux 4.4.0-66-generic x86_64
ApportVersion: 2.20.1-0ubuntu2.5
Architecture: amd64
AudioDevicesInUse:
USER PID ACCESS COMMAND
/dev/snd/
CurrentDesktop: Unity
Date: Mon Mar 13 08:57:06 2017
DistributionCha
# This is a distribution channel descriptor
# For more information see http://
canonical-
HibernationDevice: RESUME=
InstallationDate: Installed on 2017-03-09 (3 days ago)
InstallationMedia: Ubuntu 16.04 "Xenial" - Build amd64 LIVE Binary 20160624-10:47
IwConfig:
lo no wireless extensions.
wlp2s0 no wireless extensions.
MachineType: Dell Inc. Precision 5520
ProcFB: 0 inteldrmfb
ProcKernelCmdLine: BOOT_IMAGE=
RelatedPackageV
linux-
linux-
linux-firmware 1.157.8
SourcePackage: linux
UpgradeStatus: No upgrade log present (probably fresh install)
dmi.bios.date: 01/18/2017
dmi.bios.vendor: Dell Inc.
dmi.bios.version: 1.1.3
dmi.board.name: 0X41RR
dmi.board.vendor: Dell Inc.
dmi.board.version: A00
dmi.chassis.type: 10
dmi.chassis.vendor: Dell Inc.
dmi.modalias: dmi:bvnDellInc.
dmi.product.name: Precision 5520
dmi.sys.vendor: Dell Inc.
CVE References
Carl Mueller (cmuell89) wrote : | #1 |
- AlsaInfo.txt Edit (37.9 KiB, text/plain; charset="utf-8")
- CRDA.txt Edit (392 bytes, text/plain; charset="utf-8")
- CurrentDmesg.txt Edit (69.8 KiB, text/plain; charset="utf-8")
- Dependencies.txt Edit (2.3 KiB, text/plain; charset="utf-8")
- JournalErrors.txt Edit (19.8 KiB, text/plain; charset="utf-8")
- Lspci.txt Edit (14.0 KiB, text/plain; charset="utf-8")
- Lsusb.txt Edit (342 bytes, text/plain; charset="utf-8")
- ProcCpuinfo.txt Edit (9.0 KiB, text/plain; charset="utf-8")
- ProcEnviron.txt Edit (103 bytes, text/plain; charset="utf-8")
- ProcInterrupts.txt Edit (5.2 KiB, text/plain; charset="utf-8")
- ProcModules.txt Edit (7.2 KiB, text/plain; charset="utf-8")
- PulseList.txt Edit (21.0 KiB, text/plain; charset="utf-8")
- RfKill.txt Edit (113 bytes, text/plain; charset="utf-8")
- UdevDb.txt Edit (195.3 KiB, text/plain; charset="utf-8")
- WifiSyslog.txt Edit (100.9 KiB, text/plain; charset="utf-8")
Carl Mueller (cmuell89) wrote : | #2 |
Brad Figg (brad-figg) wrote : Status changed to Confirmed | #3 |
This change was made by a bot.
Changed in linux (Ubuntu): | |
status: | New → Confirmed |
Carl Mueller (cmuell89) wrote : | #4 |
~$ sudo powertop (Idle Stats tab)
Package | Core | CPU 0 CPU 4
POLL 0.0% | POLL 0.0% | POLL 0.0% 0.0 ms 0.0% 0.0 ms
C1 0.8% | C1 0.3% | C1 0.3% 0.2 ms 0.4% 0.6 ms
C2 12.7% | C2 7.2% | C2 10.4% 1.6 ms 4.0% 4.1 ms
C3 81.7% | C3 89.4% | C3 84.6% 6.6 ms 94.0% 8.9 ms
Carl Mueller (cmuell89) wrote : | #5 |
~$ sudo turbostat
cpu0: Guessing tjMax 100 C, Please use -T to specify
CPU Avg_MHz Busy% Bzy_MHz TSC_MHz
- 100 6.63 1500 2904
0 109 7.27 1500 2904
4 80 5.33 1501 2904
1 133 8.85 1498 2904
5 93 6.21 1501 2904
2 116 7.73 1500 2904
6 86 5.73 1500 2904
3 116 7.70 1500 2904
7 64 4.25 1500 2904
CPU Avg_MHz Busy% Bzy_MHz TSC_MHz
- 158 10.52 1501 2904
0 174 11.60 1501 2904
4 105 7.02 1502 2904
1 178 11.87 1499 2904
5 128 8.53 1501 2904
2 191 12.71 1501 2904
6 145 9.65 1500 2904
3 205 13.67 1501 2904
7 137 9.11 1502 2904
Carl Mueller (cmuell89) wrote : | #6 |
Installed:
intel-microcode
Carl Mueller (cmuell89) wrote : | #7 |
~$ dmesg |grep idle
[ 0.000000] clocksource: refined-jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 7645519600211568 ns
[ 0.000000] clocksource: hpet: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 79635855245 ns
[ 0.044934] process: using mwait in idle threads
[ 0.153005] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 7645041785100000 ns
[ 0.162345] cpuidle: using governor ladder
[ 0.178334] cpuidle: using governor menu
[ 0.343661] clocksource: acpi_pm: mask: 0xffffff max_cycles: 0xffffff, max_idle_ns: 2085701024 ns
[ 0.899900] intel_idle: does not run on family 6 model 158
[ 0.901622] ACPI: acpi_idle registered with cpuidle
[ 1.894828] clocksource: tsc: mask: 0xffffffffffffffff max_cycles: 0x29dc020bb13, max_idle_ns: 440795273180 ns
Carl Mueller (cmuell89) wrote : | #8 |
~$ sudo i7z
sudo i7z
i7z DEBUG: i7z version: svn-r93-
i7z DEBUG: Found Intel Processor
i7z DEBUG: Stepping 9
i7z DEBUG: Model e
i7z DEBUG: Family 6
i7z DEBUG: Processor Type 0
i7z DEBUG: Extended Model 9
i7z DEBUG: msr = Model Specific Register
i7z DEBUG: Unknown processor, not exactly based on Nehalem, Sandy bridge or Ivy Bridge
i7z DEBUG: msr device files exist /dev/cpu/*/msr
i7z DEBUG: You have write permissions to msr device files
-------
--[core id]--- Other information
-------
--[0] Processor number 0
--[0] Socket number/
--[0] Core id number 0
--[0] Display core in i7z Tool: Yes
--[1] Processor number 1
--[1] Socket number/
--[1] Core id number 1
--[1] Display core in i7z Tool: Yes
--[2] Processor number 2
--[2] Socket number/
--[2] Core id number 2
--[2] Display core in i7z Tool: Yes
--[3] Processor number 3
--[3] Socket number/
--[3] Core id number 3
--[3] Display core in i7z Tool: Yes
--[4] Processor number 4
--[4] Socket number/
--[4] Core id number 0
--[4] Display core in i7z Tool: No
--[5] Processor number 5
--[5] Socket number/
--[5] Core id number 1
--[5] Display core in i7z Tool: No
--[6] Processor number 6
--[6] Socket number/
--[6] Core id number 2
--[6] Display core in i7z Tool: No
--[7] Processor number 7
--[7] Socket number/
--[7] Core id number 3
--[7] Display core in i7z Tool: No
Socket-0 [num of cpus 4 physical 4 logical 8] 0,1,2,3,
Socket-1 [num of cpus 0 physical 0 logical 0]
GUI has been Turned ON
i7z DEBUG: Single Socket Detected
i7z DEBUG: In i7z Single_Socket()
i7z DEBUG: guessing Nehalem
--------
AND THEN
--------
Cpu speed from cpuinfo 2903.00Mhz
cpuinfo might be wrong if cpufreq is enabled. To guess correctly try estimating via tsc
Linux's inbuilt cpu_khz code emulated now
True Frequency (without accounting Turbo) 2903 MHz
CPU Multiplier 29x || Bus clock frequency (BCLK) 100.10 MHz
Socket [0] - [physical cores=4, logical cores=8, max online cores ever=4]
TURBO ENABLED on 4 Cores, Hyper Threading ON
Max Frequency without considering Turbo 3003.10 MHz (100.10 x [30])
Max TURBO Multiplier (if Enabled) with 1/2/3/4 Cores is 39x/37x/36x/35x
Real Current Frequency 1557.86 MHz [100.10 x 15.56] (Max of below)
Core [core-id] :Actual Freq (Mult.) C0% Halt(C1)% C3 % C6 % Temp VCore
Core 1 [0]: 1516.07 (15.15x) 2.86 98.5 0 0 39 0.7573
Core 2 [1]: 1557.86 (15.56x) 2.16 98.8 0 0 37 0.7573
Core 3 [2]: 1533.17 (15.32x) 2.15 98.9 0 0 39 0.7550
Core 4 [3]: 1554.47 (15.53x) 1.51 98.2 1 0 38 0.7552
C0 = Processor running without halting
C1 = Processor running with halts (States >C0 are power saver modes with cores idling)
C3 = Cores running with PLL turned off and core cache...
Carl Mueller (cmuell89) wrote : | #9 |
Let me know what other information I could supply.
Joseph Salisbury (jsalisbury) wrote : | #10 |
Did this issue start happening after an update/upgrade? Was there a prior kernel version where you were not having this particular problem?
Would it be possible for you to test the latest upstream kernel? Refer to https:/
If this bug is fixed in the mainline kernel, please add the following tag 'kernel-
If the mainline kernel does not fix this bug, please add the tag: 'kernel-
Once testing of the upstream kernel is complete, please mark this bug as "Confirmed".
Thanks in advance.
Changed in linux (Ubuntu): | |
importance: | Undecided → Medium |
Changed in linux (Ubuntu Xenial): | |
status: | New → Incomplete |
importance: | Undecided → Medium |
Changed in linux (Ubuntu): | |
status: | Confirmed → Incomplete |
Kai-Heng Feng (kaihengfeng) wrote : | #11 |
I tested on an KBL laptop and it did have the issue. Backport a patch can make the machine idle to C10.
commit 3ce093d4de753d6
Author: Len Brown <email address hidden>
Date: Wed Apr 6 17:00:59 2016 -0400
intel_idle: Add KBL support
KBL is similar to SKL
Signed-off-by: Len Brown <email address hidden>
Signed-off-by: Rafael J. Wysocki <email address hidden>
Please test the Xenial kernel: http://
Carl Mueller (cmuell89) wrote : | #12 |
@Kai-Heng Feng, What are the odds of this backport patch being included with the next update from Canonical for 16.04. Since your recommended kernel is 4.4.0-68 do you think it will arrive as part of the regular release schedule? I can report if the patch worked when it arrives. As much as I love to support the work on these bugs, jumping kernels has left sour taste in my mouth. Switching between two kernels has sometimes created issues (bluetooth drivers not loading etc,.) that carry over into the previously stable kernel.
Mostly this is my production/work machine and I am resistant to messing with it too much. :(
Sorry to sound like a poor sport.
C
Kai-Heng Feng (kaihengfeng) wrote : | #13 |
No, that's why I built a slightly newer kernel, 4.4.0-68, so it won't mess with the current release kernel, 4.4.0-66.
It's okay if don't want to test it, I already tested it on Precision 5520 at my hand.
The problem lies here is, 5520 has several different configs, I want to make sure the patch works on your machine. Otherwise what's the point?
I assume the bluetooth symptom you saw mainly relates to the non-matching firmware is still in the chip, a cold boot instead of warm reboot should circumvent it.
Carl Mueller (cmuell89) wrote : | #14 |
Okay. I'll try it out.
Carl Mueller (cmuell89) wrote : | #15 |
Should I install all four .deb packages in the link you provided?
Carl Mueller (cmuell89) wrote : | #16 |
Just an FYI, I do see this warning every time I've tried to build a kernel on my 5520. Not sure if this is an issue.
: Unable to find an initial ram disk that I know how to handle.
Will not try to make an initrd.
Carl Mueller (cmuell89) wrote : | #17 |
Okay. Right away I notice a difference with the program i7z:
Cpu speed from cpuinfo 2875.00Mhz
cpuinfo might be wrong if cpufreq is enabled. To guess correctly try estimating via tsc
Linux's inbuilt cpu_khz code emulated now
True Frequency (without accounting Turbo) 2895 MHz
CPU Multiplier 29x || Bus clock frequency (BCLK) 99.83 MHz
Socket [0] - [physical cores=4, logical cores=8, max online cores ever=4]
TURBO ENABLED on 4 Cores, Hyper Threading ON
Max Frequency without considering Turbo 2994.83 MHz (99.83 x [30])
Max TURBO Multiplier (if Enabled) with 1/2/3/4 Cores is 39x/37x/36x/35x
Real Current Frequency 2881.62 MHz [99.83 x 28.87] (Max of below)
Core [core-id] :Actual Freq (Mult.) C0% Halt(C1)% C3 % C6 % Temp VCore
Core 1 [0]: 2865.41 (28.70x) 1.77 96.2 1 1 46 1.0009
Core 2 [1]: 2874.93 (28.80x) 1.74 86.3 1 11 46 0.9927
Core 3 [2]: 2881.62 (28.87x) 1.34 96.1 1 1.55 46 1.0076
Core 4 [3]: 2877.37 (28.82x) 1.4 95 1 2.62 46 1.0226
C0 = Processor running without halting
C1 = Processor running with halts (States >C0 are power saver modes with cores idling)
C3 = Cores running with PLL turned off and core cache turned off
C6, C7 = Everything in C3 + core state saved to last level cache, C7 is deeper than C6
Above values in table are in percentage over the last 1 sec
WAY MORE activity in C6 state. Although I don't see any listings for C10
Carl Mueller (cmuell89) wrote : | #18 |
sudo powertop (idel stats)
Package | Core | CPU 0 CPU 4
POLL 0.0% | POLL 0.0% | POLL 0.0% 0.0 ms 0.1% 3.1 ms
C1E-SKL 0.5% | C1E-SKL 0.2% | C1E-SKL 0.4% 0.6 ms 0.0% 0.1 ms
C3-SKL 0.2% | C3-SKL 0.1% | C3-SKL 0.2% 0.4 ms 0.0% 0.1 ms
C6-SKL 2.8% | C6-SKL 1.5% | C6-SKL 1.8% 0.7 ms 1.2% 1.4 ms
C7s-SKL 0.2% | C7s-SKL 0.0% | C7s-SKL 0.0% 1.7 ms 0.0% 0.0 ms
C8-SKL 24.2% | C8-SKL 19.7% | C8-SKL 29.2% 3.1 ms 10.2% 3.9 ms
C9-SKL 0.0% | C9-SKL 0.0% | C9-SKL 0.0% 0.0 ms 0.0% 1.1 ms
C10-SKL 67.5% | C10-SKL 75.3% | C10-SKL 63.9% 11.5 ms 86.7% 17.5 ms
Carl Mueller (cmuell89) wrote : | #19 |
However, I notcied powertop is now not automatically choosing to adjust many of the bad parameters to good (not sure if that is related...). Most of the PCI devices are not set to Bad when previously they were automatically at Good.
Bad Enable SATA link power management for host0
Bad Enable SATA link power management for host1
Bad Enable Audio codec power management
Bad Runtime PM for I2C Adapter i2c-2 (i915 gmbus dpd)
Bad Autosuspend for USB device USB Receiver [Logitech]
Bad Runtime PM for I2C Adapter i2c-1 (i915 gmbus dpb)
Bad Autosuspend for USB device Touchscreen [ELAN]
Bad Runtime PM for I2C Adapter i2c-0 (i915 gmbus dpc)
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H PCI Express Root Por
Bad Runtime PM for PCI Device Realtek Semiconductor Co., Ltd. RTS525A PCI Express Ca
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H PCI Express Root Por
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H PCI Express Root Por
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H PCI Express Root Por
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H PMC
Bad Runtime PM for PCI Device Intel Corporation Device a171
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H SMBus
Bad Runtime PM for PCI Device NVIDIA Corporation Device 13b6
Bad Runtime PM for PCI Device Intel Corporation Device 24fd
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H LPC Controller
Bad Runtime PM for PCI Device Samsung Electronics Co Ltd Device a804
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H SATA controller [AHC
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H CSME HECI #1
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H LPSS I2C Controller
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H LPSS I2C Controller
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H Thermal subsystem
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H USB 3.0 xHCI Control
Bad Runtime PM for PCI Device Intel Corporation Device 591b
Bad Runtime PM for PCI Device Intel Corporation Sky Lake PCIe Controller (x16)
Bad Runtime PM for PCI Device Intel Corporation Device 5910
Bad Runtime PM for PCI Device Intel Corporation Sunrise Point-H PCI Express Root Por
Bad Runtime PM for PCI Device Intel Corporation Skylake Processor Thermal Subsystem
Good Bluetooth device interface status
Good NMI watchdog should be turned off
Good VM writeback timeout
Good Autosuspend for USB device xHCI Host Controller [usb1]
Good Autosuspend for USB device xHCI Host Controller [usb2]
Good I2C Device i2c-DLL07BF:01 has no runtime power management
Good ...
Carl Mueller (cmuell89) wrote : | #20 |
Also trying to run turbostat:
sudo turbostat --debug
WARNING: turbostat not found for kernel 4.4.0-68
You may need to install the following packages for this specific kernel:
linux-
linux-
You may also want to install one of the following packages to keep up to date:
linux-
linux-
Carl Mueller (cmuell89) wrote : | #21 |
Autotune with powertop works fine. No interference. Still showing heavy usage of C10 state when idling :) :)
I think this is a worthy path to incorporate for us 5520 users.
Kai-Heng Feng (kaihengfeng) wrote : | #22 |
Thanks for your test! I'll backport this patch to Ubuntu Xenial's kernel.
From the log, I think your machine is preloaded with Ubuntu, right? It should come with TLP installed. TLP only enables runtime PM when power cord is unplugged, those "bad" should become "good" once the laptop run on battery.
Carl Mueller (cmuell89) wrote : | #23 |
Kai-Feng, you were right about Powertop. I am a big fan of presuming th worst. For some reason, my trackpad stopped tracking (I could click but no movement). However upon restart it seems to be fully restored and operational.
Changed in linux (Ubuntu Xenial): | |
status: | Incomplete → In Progress |
assignee: | nobody → Kai-Heng Feng (kaihengfeng) |
Carl Mueller (cmuell89) wrote : | #24 |
@Kai-Heng Feng, I am not sure if this patch was intended for 4.4.0-67 (which recently was updated via Software Updater) but that kernel version reverted my CPU behavior back to only using C3 states and nothing deeper. Just FYI.
Kai-Heng Feng (kaihengfeng) wrote : | #25 |
It should be included in version >= 4.4.0-68.
Carl Mueller (cmuell89) wrote : | #26 |
Excuse my ignorance, but does that mean I should continue to use your 4.4.0-68 kernel and as Canonical releases kernel updates, I should naturally phase into the regular update stream again? Or should I revert to the newest updated (4.4.0-67) and remove your custom kernel?
Thanks!
Kai-Heng Feng (kaihengfeng) wrote : | #27 |
You should remove the custom kernel and use the official Ubuntu kernel.
The deeper C state patch should be included soon.
Carl Mueller (cmuell89) wrote : | #28 |
Kai-Heng Feng, I also noticed that by default the scaling driver used is acpi-cpufreq and not p-state. I've read that p-state is favorable for kernels >= 4.4. Is this something that is up to the user to decide or should this be considered an incorrect default?
analyzing CPU 0:
driver: acpi-cpufreq
CPUs which run at the same hardware frequency: 0
CPUs which need to have their frequency coordinated by software: 0
maximum transition latency: 10.0 us.
hardware limits: 800 MHz - 2.80 GHz
available frequency steps: 2.80 GHz, 2.80 GHz, 2.70 GHz, 2.60 GHz, 2.40 GHz, 2.30 GHz, 2.00 GHz, 1.90 GHz, 1.80 GHz, 1.70 GHz, 1.50 GHz, 1.40 GHz, 1.20 GHz, 1.10 GHz, 900 MHz, 800 MHz
available cpufreq governors: conservative, ondemand, userspace, powersave, performance
current policy: frequency should be within 800 MHz and 2.80 GHz.
current CPU frequency is 900 MHz (asserted by call to hardware).
cpufreq stats: 2.80 GHz:2.85%, 2.80 GHz:0.00%, 2.70 GHz:0.25%, 2.60 GHz:0.64%, 2.40 GHz:0.65%, 2.30 GHz:1.42%, 2.00 GHz:2.07%, 1.90 GHz:1.27%, 1.80 GHz:1.64%, 1.70 GHz:3.14%, 1.50 GHz:3.57%, 1.40 GHz:5.39%, 1.20 GHz:6.85%, 1.10 GHz:13.39%, 900 MHz:29.31%, 800 MHz:27.53% (2074191)
boost state support:
Supported: yes
Active: yes
Kai-Heng Feng (kaihengfeng) wrote : | #29 |
You are right. Can you file another bug report?
Carl Mueller (cmuell89) wrote : | #30 |
Changed in linux (Ubuntu Xenial): | |
status: | In Progress → Fix Committed |
Carl Mueller (cmuell89) wrote : | #31 |
Fix not showing up in 4.4.0-70. Not sure if it was intended for this release.
Carl Mueller (cmuell89) wrote : | #32 |
Fix not showing in 4.4.0-71 either. Just FYI in case it was intended.
Carl Mueller (cmuell89) wrote : | #33 |
@Kai-Heng Feng, the fix is not showing up in 4.4.0-72
carl@carl-5520:~$ uname -a
Linux carl-5520 4.4.0-72-generic #93-Ubuntu SMP Fri Mar 31 14:07:41 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux
carl@carl-5520:~$ sudo powertop
Package | Core | CPU 0 CPU 4
POLL 0.0% | POLL 0.0% | POLL 0.0% 0.0 ms 0.0% 0.0 ms
C1 2.7% | C1 2.2% | C1 4.3% 1.3 ms 0.1% 0.3 ms
C2 7.3% | C2 7.3% | C2 7.0% 1.6 ms 7.7% 4.1 ms
C3 87.9% | C3 88.4% | C3 85.2% 14.8 ms 91.5% 23.1 ms
carl@carl-5520:~$ sudo i7z
Cpu speed from cpuinfo 2902.00Mhz
cpuinfo might be wrong if cpufreq is enabled. To guess correctly try estimating via tsc
Linux's inbuilt cpu_khz code emulated now
True Frequency (without accounting Turbo) 2902 MHz
CPU Multiplier 29x || Bus clock frequency (BCLK) 100.07 MHz
Socket [0] - [physical cores=4, logical cores=8, max online cores ever=4]
TURBO ENABLED on 4 Cores, Hyper Threading ON
Max Frequency without considering Turbo 3002.07 MHz (100.07 x [30])
Max TURBO Multiplier (if Enabled) with 1/2/3/4 Cores is 39x/37x/36x/35x
Real Current Frequency 1395.28 MHz [100.07 x 13.94] (Max of below)
Core [core-id] :Actual Freq (Mult.) C0% Halt(C1)% C3 % C6 % Temp VCore
Core 1 [0]: 1351.45 (13.51x) 5.21 97.6 0 0 44 0.8995
Core 2 [1]: 1395.28 (13.94x) 1.37 99.3 0 0 43 0.8981
Core 3 [2]: 1310.44 (13.10x) 2.49 98.9 0 0 44 0.8981
Core 4 [3]: 1328.40 (13.27x) 1 99.6 0 0 44 0.8954
C0 = Processor running without halting
C1 = Processor running with halts (States >C0 are power saver modes with cores idling)
C3 = Cores running with PLL turned off an...
Carl Mueller (cmuell89) wrote : | #34 |
Sorry to be so persistent but Intel has posted a warning about not using lower power states. See page 64 of this data sheet:
Kai-Heng Feng (kaihengfeng) wrote : | #35 |
It's ok ;)
The new kernel in xenial-proposed should fix this issue.
Kleber Sacilotto de Souza (kleber-souza) wrote : | #36 |
This bug is awaiting verification that the kernel in -proposed solves the problem. Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-
If verification is not done by 5 working days from today, this fix will be dropped from the source code, and this bug will be closed.
See https:/
tags: | added: verification-needed-xenial |
tags: |
added: verification-done-xenial removed: verification-needed-xenial |
Launchpad Janitor (janitor) wrote : | #37 |
This bug was fixed in the package linux - 4.4.0-75.96
---------------
linux (4.4.0-75.96) xenial; urgency=low
* linux: 4.4.0-75.96 -proposed tracker (LP: #1684441)
* [Hyper-V] hv: util: move waiting for release to hv_utils_transport itself
(LP: #1682561)
- Drivers: hv: util: move waiting for release to hv_utils_transport itself
linux (4.4.0-74.95) xenial; urgency=low
* linux: 4.4.0-74.95 -proposed tracker (LP: #1682041)
* [Hyper-V] hv: vmbus: Raise retry/wait limits in vmbus_post_msg()
(LP: #1681893)
- Drivers: hv: vmbus: Raise retry/wait limits in vmbus_post_msg()
linux (4.4.0-73.94) xenial; urgency=low
* linux: 4.4.0-73.94 -proposed tracker (LP: #1680416)
* CVE-2017-6353
- sctp: deny peeloff operation on asocs with threads sleeping on it
* vfat: missing iso8859-1 charset (LP: #1677230)
- [Config] NLS_ISO8859_1=y
* Regression: KVM modules should be on main kernel package (LP: #1678099)
- [Config] powerpc: Add kvm-hv and kvm-pr to the generic inclusion list
* linux-lts-xenial 4.4.0-63.84~14.04.2 ADT test failure with linux-lts-xenial
4.4.
- SAUCE: apparmor: fix link auditing failure due to, uninitialized var
* regession tests failing after stackprofile test is run (LP: #1661030)
- SAUCE: fix regression with domain change in complain mode
* Permission denied and inconsistent behavior in complain mode with 'ip netns
list' command (LP: #1648903)
- SAUCE: fix regression with domain change in complain mode
* unexpected errno=13 and disconnected path when trying to open /proc/1/ns/mnt
from a unshared mount namespace (LP: #1656121)
- SAUCE: apparmor: null profiles should inherit parent control flags
* apparmor refcount leak of profile namespace when removing profiles
(LP: #1660849)
- SAUCE: apparmor: fix ns ref count link when removing profiles from policy
* tor in lxd: apparmor="DENIED" operation=
namespace=
name=
- SAUCE: apparmor: Fix no_new_privs blocking change_onexec when using stacked
namespaces
* apparmor oops in bind_mnt when dev_path lookup fails (LP: #1660840)
- SAUCE: apparmor: fix oops in bind_mnt when dev_path lookup fails
* apparmor auditing denied access of special apparmor .null fi\ le
(LP: #1660836)
- SAUCE: apparmor: Don't audit denied access of special apparmor .null file
* apparmor label leak when new label is unused (LP: #1660834)
- SAUCE: apparmor: fix label leak when new label is unused
* apparmor reference count bug in label_merge_
- SAUCE: apparmor: fix reference count bug in label_merge_
* apparmor's raw_data file in securityfs is sometimes truncated (LP: #1638996)
- SAUCE: apparmor: fix replacement race in reading rawdata
* unix domain socket cross permission check failing with nested namespaces
(LP: #1660832)
- SAUCE: apparmor: fix cross ns perm of unix domain sockets
* Xenial update to v4.4.59 stable release (LP: #1678960)
- xfrm: policy: init locks early
- virtio_balloon: init ...
Changed in linux (Ubuntu Xenial): | |
status: | Fix Committed → Fix Released |
Carl Mueller (cmuell89) wrote : | #38 |
Kernel 4.4.75 appears to have fixed the problem for C states.
Package | Core | CPU 0 CPU 4
POLL 0.1% | POLL 0.0% | POLL 0.0% 0.0 ms 0.0% 0.0 ms
C1E-SKL 1.4% | C1E-SKL 2.9% | C1E-SKL 3.5% 0.3 ms 2.3% 0.4 ms
C3-SKL 0.9% | C3-SKL 1.1% | C3-SKL 1.5% 0.2 ms 0.8% 0.2 ms
C6-SKL 8.4% | C6-SKL 12.6% | C6-SKL 19.0% 0.6 ms 6.1% 0.4 ms
C7s-SKL 0.2% | C7s-SKL 0.5% | C7s-SKL 0.6% 1.5 ms 0.3% 0.6 ms
C8-SKL 57.6% | C8-SKL 54.1% | C8-SKL 67.1% 1.7 ms 41.0% 3.2 ms
C9-SKL 0.0% | C9-SKL 0.0% | C9-SKL 0.0% 0.0 ms 0.0% 0.0 ms
C10-SKL 27.9% | C10-SKL 24.9% | C10-SKL 3.2% 4.4 ms 46.6% 8.0 ms
carl@carl-5520:~$ sudo tlp stat -p
--- TLP 0.8 --------...
Carl Mueller (cmuell89) wrote : | #39 |
@Kai-Feng. The other bug regarding p-state as disappeared (I cannot access it on Ubuntu bugs). This bug also appears to be fixed as p-state is now the default scaling driver. However tlp stat is showing unusual max/min cpufreq settings after kernel installation. Do these settings seem incorrect and/or are they even used by the pstate driver?
$ sudo tlp-stat -p
--- TLP 0.8 -------
+++ Processor
CPU Model = Intel(R) Core(TM) i7-7820HQ CPU @ 2.90GHz
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
x86_energy_
x86_energy_
x86_energy_
x86_energy_
Kai-Heng Feng (kaihengfeng) wrote : | #40 |
It's at LP: #1674390.
Kai-Heng Feng (kaihengfeng) wrote : | #41 |
@Carl, I burrowed a Precision 5520 and seems to be normal. Not sure what happened on your system...
--- TLP 0.8 -------
+++ Processor
CPU Model = Intel(R) Core(TM) i7-7820HQ CPU @ 2.90GHz
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
/sys/devices/
x86_energy_
x86_energy_
x86_energy_
x86_energy_
x86_energy_
x86_energy_
x86_energy_
x86_energy_
Changed in linux (Ubuntu): | |
status: | Incomplete → Fix Released |
cat /sys/module/ intel_idle/ parameters/ max_cstate
~$ cat /sys/module/ intel_idle/ parameters/ max_cstate
9