Closing.
]]>I'm experiencing the same issue on my laptop (i5-5200U CPU). One interesting thing: most of the time I'm plugged in and using an external screen and freezes NEVER happen. I have started experiencing them when I unplug (not sure if relevant) and don't use an external screen. Odd. Will do a bit more testing over the weekend and report back. Any updates from others?
]]>I've experienced similar issues and tried to set my c_state=0 as well but with no effect. Next thing I did was to take the LTS kernel and see and that worked for the last 3 days.
I decided to remove the c_state kernel paramter and see how that is going.
My LTS kernel version is: 4.14.79-1-lts
The kernel I used previously was linux-zen: 4.18.16.zen1-1
You stated the freezes were more frequent with cstate_max=0?
I'm really not sure. I did experience two freezes in an hour, but this would also happen now and then when I had a clean kernel cmdline. I didn't stick to cstate_max=0 for more than this one hour, so I don't really have much statistical evidence to be sure that "freezes are more frequent with cstate_max=0".
If that's the case have you looked at what scaling_driver you're using? Are you using intel_pstate or acpi_cpufreq?
$ sudo cat /sys/devices/system/cpu/cpu*/cpufreq/scaling_driver
intel_pstate
intel_pstate
intel_pstate
intel_pstate
What's your actual processor?
I actually don't know much about CPUs, so I don't know exactly what you're asking. lscpu says "Model name: Intel(R) Core(TM) i5-5200U CPU @ 2.20GHz".
Do you have turbo enabled?
Never heard of this. I guess I'll take a look at the wiki. Thanks
EDIT: i7z gives "TURBO ENABLED on 2 Cores, Hyper Threading ON"
I did read the info on disabling cstates not being a good long-term solution, but those freezes are really inconvenient. What are the harms (besides high power consumption)?
]]>whites11 wrote:I am having the same exact issue on my dell inspiron 15 (2015 version).
I replaced my SSD because it was unhealthy and I thought it was the cause of the (recently appeared) hangs, but unluckily it was not.
I just tried to add the intel_idle.max_cstate=1 kernel parameter, let's see what happens.Also the exact same issue here, and also a Dell Inspiron 15 5000 Series (5548) with an i5. Freezing can happen with a few minutes of uptime, or sometimes a few days. I tried
intel_idle.max_cstate=0
as suggested here, but if anything there the freezing was more frequent. I'll try
intel_idle.max_cstate=1
now.
Please note those steps were really only recommended as steps to try and rule out (or in) cstate(s) as the cause of the issue with the goal of then taking further steps to address as needed; running a laptop with the max_cstate locked at 0 or 1 is not a long term solution.
You stated the freezes were more frequent with cstate_max=0? If that's the case have you looked at what scaling_driver you're using? Are you using intel_pstate or acpi_cpufreq? What's your actual processor? Do you have turbo enabled? In my case, after much back and forth, and a ton of help from kernel bugtracker - my issue was narrowed down to something with turbo states. If I disable turbo on my box (i5-6440HQ) I no longer experience the freezes anymore. I've since disabled turbo via my BIOS and have had zero issues (albeit I'm stuck at 2.6 base/max clock!). I haven't really had time to drill down much further since then, but it's on my list.
]]>I am having the same exact issue on my dell inspiron 15 (2015 version).
I replaced my SSD because it was unhealthy and I thought it was the cause of the (recently appeared) hangs, but unluckily it was not.
I just tried to add the intel_idle.max_cstate=1 kernel parameter, let's see what happens.
Also the exact same issue here, and also a Dell Inspiron 15 5000 Series (5548) with an i5. Freezing can happen with a few minutes of uptime, or sometimes a few days. I tried
intel_idle.max_cstate=0
as suggested here, but if anything there the freezing was more frequent. I'll try
intel_idle.max_cstate=1
now.
]]>I am having the same exact issue on my dell inspiron 15 (2015 version).
I replaced my SSD because it was unhealthy and I thought it was the cause of the (recently appeared) hangs, but unluckily it was not.
I just tried to add the intel_idle.max_cstate=1 kernel parameter, let's see what happens.
16 days without a freeze now. What I still am missing is how and when this problem arose. My laptop worked just fine for about 3 years, then suddenly started freezing. Was the problem triggered by a kernel update?
]]>I replaced my SSD because it was unhealthy and I thought it was the cause of the (recently appeared) hangs, but unluckily it was not.
I just tried to add the intel_idle.max_cstate=1 kernel parameter, let's see what happens.
No freezes since a week of uptime now. I saw there were multiple BIOS updates for my Laptop recently so I will try to update my BIOS again and maybe try removing the intel_idle.max_cstate=0 line from my boot parameters, to see if only the BIOS update fixes it.
The initial issue seems fixed, however I have new errors as stated above. But I think a BIOS update might fix it. And then I'll try switching back to the latest kernel (no LTS).
I will write a more detailed answer later, to help people who had the same issue.
Thanks again everyone for your help.
]]>[71843.696211] ACPI Error: [\_PR_.CPU0._CST] Namespace lookup failure, AE_NOT_FOUND (20170728/psargs-364)
[71843.696216] ACPI Error: Method parse/execution failed \_PR.CPU3._CST, AE_NOT_FOUND (20170728/psparse-550)
(This message appears at least once a hour, so it's filling my dmesg logs)
I will try if I get some more time to play with the intel_idle.max_cstate option (try to increment it and see if it freezes or not).
Thanks again! (Will mark as Solved if I don't get any freeze for at least 2 weeks).
]]>Computer Information:
CPU: Intel Core i5-7600k @ 4.6GHz (3.8 GHz Cache, 1.26V, Integrated Video Disabled)
GPU: NVidia GTX 1050 Ti (thee monitors connected)
RAM: 2 x 8GB DDR4 2666 (OC from 2400), 17-17-17-39 (Memtest86 report no errors in 20 tests)
SSD: 64Gb Toshiba for / (SMART shows good status)
HDD: 1Tb for /home and /srv (SMART shows good status too)
First time system freezes was 3 weeks ago, keyboard and mouse have not respond to any action. Since that time no such issues was catched but in last 3 days I getting lot of freezes.
No errors in journal nor Xorg.
If video played in browser it repeat last 2 seconds, but not when in spotify.
But today I got pc freeze when worked in phpStorm, two of 3 screens have become black and in journal last lines are:
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: #
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # A fatal error has been detected by the Java Runtime Environment:
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: #
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # SIGSEGV (0xb) at pc=0x00007ffc1412ca57, pid=4932, tid=0x00007fd1fc3c9700
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: #
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # JRE version: OpenJDK Runtime Environment (8.0_152-b39) (build 1.8.0_152-release-1136-b39)
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # Java VM: OpenJDK 64-Bit Server VM (25.152-b39 mixed mode linux-amd64 compressed oops)
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # Problematic frame:
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # C [linux-vdso.so.1+0xa57]
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: #
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # Core dump written. Default location: /home/mkbodanu4/core or core.4932
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: #
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # An error report file with more information is saved as:
чер 19 15:24:02 mkwork3 jetbrains-phpstorm.desktop[4877]: # /home/mkbodanu4/java_error_in_PHPSTORM_4932.log
File
/home/mkbodanu4/java_error_in_PHPSTORM_4932.log
was not created.
I tried disable overclocking, tested ram and all disks, disabled all energy settings in UEFI Setup and installed linux-lts version - nothing help.
So please, how can I debug this on PC?
P.S. Sorry for bad English
]]>intel_idle.max_cstate= [KNL,HW,ACPI,X86]
0 disables intel_idle and fall back on acpi_idle.
1 to 9 specify maximum depth of C-state.
My suggestion is to start with intel_idle.max_cstate=0 which completely disables intel_idle (also dramatically increases power consumption btw). If you're able to run with that for a while with no freezes then continue to increment the max cstate until you find the highest number at which your box freezes. Once you find that you can either continue to boot limiting the max c-state to that number, or you can try and selectively disable the specific cstates that may be troublesome.
What's the output of
$ ls -ld /sys/devices/system/cpu/cpu0/cpuidle/state*
and
$ cat /sys/devices/system/cpu/cpu0/cpuidle/state*/name
I also have this problem on a Thinkpad X1 Carbon 3rd Gen.
I tried Linux and Linux-CK kernel, but I had to "hard reboot" after the freeze.
Now, I boot on Linux LTS, and no more freeze. But if the problem comes from the kernel, we will have the freeze again, when the LTS kernel will include the current options of the kernel...
And like you, I don't know where to investigate, as there is no log at all.
I will continue to follow this thread.
]]>