New: Under heavy load, the CPU speed suddenly and irreversibly drops from 3500 to 400 MHz
https://bugzilla.kernel.org/show_bug.cgi?id=173361
Bug ID: 173361
Summary: Under heavy load, the CPU speed suddenly and
irreversibly drops from 3500 to 400 MHz
Product: Power Management
Version: 2.5
Kernel Version: 4.8-rc1
Hardware: All
OS: Linux
Tree: Mainline
Status: NEW
Severity: normal
Priority: P1
Component: cpufreq
Assignee: [email protected]
Reporter: [email protected]
Regression: No
My Toshiba Tecra A50-A laptop that a CPU described as 'Model: 6.60.3 "Intel(R) Core(TM) i7-4600M CPU @ 2.90GHz"'. Under heavy load, this dual-core unit with hyperthreading will suddenly drop from a frequency of 3500 to 410 MHz. The only way to recover is to reboot. See
http://lkml.iu.edu/hypermail/linux/kernel/1609.3/00720.html for some details and other discussion.
This bug is being filed under power management because the 'sensors' command shows the following:
finger@linux-1t8h:~> sensors
coretemp-isa-0000
Adapter: ISA adapter Physical id 0: +88.0°C (high = +84.0°C, crit = +100.0°C) Core 0: +88.0°C (high = +84.0°C, crit = +100.0°C) Core 1: +77.0°C (high = +84.0°C, crit = +100.0°C)
Routinely, at least one of the CPUs has a temperature higher than the "high" value. As Rafael Wysocki states "It looks like in 4.8-rc we made a change that caused the "high" trip point to be acted on." In that case, the bug would be that the frequency is never restored.
Signature · 2 lines
You are receiving this mail because: You are the assignee for the bug.
[Bug 173361] Under heavy load, the CPU speed suddenly and irreversibly drops from 3500 to 400 MHz
https://bugzilla.kernel.org/show_bug.cgi?id=173361
Doug Smythies <[email protected]> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected]
--- Comment #1 from Doug Smythies <[email protected]> --- For what it is worth, some data from my computer:
. I get the exact same steady state temperature and package power under full load with both kernel 4.7 and 4.8-rc8.
. I messed with the cooling so that it could exceed the high limit, and when it did nothing tripped (as expected):
doug@s15:~/temp2$ sensors coretemp-isa-0000 Adapter: ISA adapter Physical id 0: +81.0°C (high = +80.0°C, crit = +98.0°C) Core 0: +77.0°C (high = +80.0°C, crit = +98.0°C) Core 1: +81.0°C (high = +80.0°C, crit = +98.0°C) Core 2: +77.0°C (high = +80.0°C, crit = +98.0°C) Core 3: +78.0°C (high = +80.0°C, crit = +98.0°C)
For this comment from the e-mail thread: "Hmm, I would not expect the CPU to drop from 80 to 40 degrees in a few seconds if the fan is not spinning. I wouldn't even expect it if the fan was spinning. I would think at least 30 to 60 seconds if not more."
I from a steady state, full load, temperature of 78 degrees C to 0 load I see:
15 degrees drop in 1 second; 18 degrees drop in 2 seconds; 22 degrees drop in 10 seconds; 25 degrees droop in 20 seconds.
For the original post comment: "In that case, the bug would be that the frequency is never restored."
It isn't supposed to restore. I do not know why in this case it is kicking in at 50%, usually it is less. Regardless, the current control algorithm in the intel_pstate driver is fundamentally incompatible with clock modulation, and will always drive the CPU down to the minimum * the modulation %, regardless of load. Other drivers typically drive the CPU frequency to what would normally be desired * modulation % (and for the most part users don't even notice).
Signature · 2 lines
You are receiving this mail because: You are the assignee for the bug.
[Bug 173361] Under heavy load, the CPU speed suddenly and irreversibly drops from 3500 to 400 MHz
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=173361
--- Comment #2 from Larry Finger <[email protected]> --- I have not measured the drop in temperature with change in load, but I know it happens very quickly, and that the fan speed is very sensitive to the load. Even when running 4 infinite loops, sending a E-mail with Thunderbird will decrease the CPU utilization enough that the fan will momentarily slow.
Signature · 2 lines
You are receiving this mail because: You are the assignee for the bug.
[Bug 173361] Under heavy load, the CPU speed suddenly and irreversibly drops from 3500 to 400 MHz
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=173361
--- Comment #3 from Doug Smythies <[email protected]> --- Created attachment 240221
--> https://bugzilla.kernel.org/attachment.cgi?id=240221&action=edit
Script to continuously monitor package temperature
I guess the point is that one can not come after the event trigger and obtain anything close to related temperatures. Perhaps the attached script could be used to monitor package temperature from well before to well after the event.
Signature · 2 lines
You are receiving this mail because: You are the assignee for the bug.
[Bug 173361] Under heavy load, the CPU speed suddenly and irreversibly drops from 3500 to 400 MHz
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=173361
--- Comment #4 from Larry Finger <[email protected]> --- Thanks. I'm running the script with the continuing test of 4.7, which has been running for a little over 30 hours without triggering the event.
Signature · 2 lines
You are receiving this mail because: You are the assignee for the bug.
[Bug 173361] Under heavy load, the CPU speed suddenly and irreversibly drops from 3500 to 400 MHz
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=173361
Rafael J. Wysocki <[email protected]> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |[email protected] Assignee|[email protected] |[email protected]
Signature · 2 lines
You are receiving this mail because: You are the assignee for the bug.