Can I avoid OS freezes on compute intense tasks

AlmostSurelyRob · 2021-04-26 13:51:12

Hello,

I am looking for some advice on improving my current setup on one of my laptops. I have been working on this computer for around 5 years. Last year though I started experiencing freezes during various compute intense tasks e.g.

1. compilation,
2. video conferencing, particularly when I share screen.

I would like to blame aging hardware, but I am not sure how to prove this. During hotter days last year it was definitely more frequent and I had a crazy idea of a thorough hoovering of the laptop, which I think helped, but I may be suffering from placebo effect.

With compilation I can get away by limiting the concurrent jobs to 2. With video conferencing the freezing is random. Screen-sharing though is quite high probability, I would even say 1 in 2. It may be related to me flicking through workspaces.

I was never able to debug the freeze itself as there's no way to interact with the computer. I tried to ssh, but sshd was timing out. Only hard reset brings it back to life.

The laptop is a Lenovo W541 with four cores and an NVIDIA Quadro K1100M. My system is updated weekly. I am running xmonad on top of ~~SLiM~~ LightDM. I start my session with:

exec dbus-launch ~/.startSession > $HOME/.xmonad.log 2>&1

Please let me know if you can think of anything. Perhaps there's a way to benchmark the hardware or maybe some way to limit software access to resources.

EDIT: SLiM -> LightDM

Last edited by AlmostSurelyRob (2021-04-27 07:45:21)

Head_on_a_Stick · 2021-04-26 15:35:23

CPU µcode?

EDIT: SLiM is dead upstream and it doesn't support systemd login sessions correctly. Have you tried any other display managers (or plain 'startx')?

Last edited by Head_on_a_Stick (2021-04-26 15:41:29)

AlmostSurelyRob · 2021-04-26 21:12:42

Many thanks for a quick reply.

I am so sorry - I made a mistake. I have actually migrated from SLiM over to lightdm some time ago, precisely when I learnt that it doesn't support systemd login session correctly. Thanks for pointing it out.

Microcode is 0x1c. Below I am posting other information about my processor. I have not configured any power management at OS level. Can this be a problem?

processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 60
model name	: Intel(R) Core(TM) i7-4810MQ CPU @ 2.80GHz
stepping	: 3
microcode	: 0x1c
cpu MHz		: 2419.724
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 0
cpu cores	: 4
apicid		: 0
initial apicid	: 0
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm abm cpuid_fault epb invpcid_single pti tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid xsaveopt dtherm ida arat pln pts
vmx flags	: vnmi preemption_timer invvpid ept_x_only ept_ad ept_1gb flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest ple shadow_vmcs
bugs		: cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit srbds
bogomips	: 5587.32
clflush size	: 64
cache_alignment	: 64
address sizes	: 39 bits physical, 48 bits virtual
power management:

jonno2002 · 2021-04-27 01:41:37

what temp is the cpu at when these freezes occur ? it could be heat related, i know my 4700mq toshiba laptop runs far too hot with the nvidia chip enabled which is why i just use the intel gpu, its like they designed the heatsink and fan for the cpu and just threw the nvidia gpu in as an after thought. so yea if its getting hot try disabling the nvidia gpu (i use bbswitch).

also replacing the thermal paste on the cpu/gpu is a good idea too, i do that every few years, and i regularly blast compressed air through the heatsink/fan too.

but in saying all of that i have never experienced freezing due to heat, BUT like i said im always aware of my temps and take steps minimize heat.

AlmostSurelyRob · 2021-04-27 08:11:36

Thanks jonno2002. I will start monitoring the temperature then and report back if it happens this week.

The freezes, I think, were never graphics related. I do some 3D rendering on the GPU for work and I never experienced a freeze then, but it may be all coupled so I definitely will give a go to bbswitch. Thanks.

AlmostSurelyRob · 2021-04-28 09:29:46

I have now added a temperature to my conky configuration:

^fg(\#ebac54) ${hwmon 4 temp 1}°C

where hwmon 4 is the temperature on core 0.

tree /sys/class/hwmon   
/sys/class/hwmon
├── hwmon0 -> ../../devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/device:08/PNP0C09:00/ACPI0003:00/power_supply/AC/hwmon0
├── hwmon1 -> ../../devices/virtual/thermal/thermal_zone0/hwmon1
├── hwmon2 -> ../../devices/LNXSYSTM:00/LNXSYBUS:00/PNP0A08:00/device:08/PNP0C09:00/PNP0C0A:00/power_supply/BAT0/hwmon2
├── hwmon3 -> ../../devices/platform/thinkpad_hwmon/hwmon/hwmon3
└── hwmon4 -> ../../devices/platform/coretemp.0/hwmon/hwmon4

I am also using `watch sensors` that displays temperature of couple of different things although apart from core I am not really sure what other items represent.

coretemp-isa-0000
Adapter: ISA adapter
Package id 0:  +52.0°C  (high = +84.0°C, crit = +100.0°C)
Core 0:        +52.0°C  (high = +84.0°C, crit = +100.0°C)
Core 1:        +52.0°C  (high = +84.0°C, crit = +100.0°C)
Core 2:        +51.0°C  (high = +84.0°C, crit = +100.0°C)
Core 3:        +49.0°C  (high = +84.0°C, crit = +100.0°C)

BAT0-acpi-0
Adapter: ACPI interface
in0:          12.21 V

thinkpad-isa-0000
Adapter: ISA adapter
fan1:        1607 RPM
temp1:        +53.0°C
temp2:         +0.0°C
temp3:         +0.0°C
temp4:         +0.0°C
temp5:         +0.0°C
temp6:         +0.0°C
temp7:         +0.0°C
temp8:         +0.0°C

acpitz-acpi-0
Adapter: ACPI interface
temp1:        +53.0°C  (crit = +200.0°C)

but I am not piping the output anywhere so if the shell isn't open during the freeze I can't really see the readings.

I have suffered a single freeze yesterday during a Webex call. The temperature shot to above 80C before the world stopped.

Arch Linux

#1 2021-04-26 13:51:12

Can I avoid OS freezes on compute intense tasks

#2 2021-04-26 15:35:23

Re: Can I avoid OS freezes on compute intense tasks

#3 2021-04-26 21:12:42

Re: Can I avoid OS freezes on compute intense tasks

#4 2021-04-27 01:41:37

Re: Can I avoid OS freezes on compute intense tasks

#5 2021-04-27 08:11:36

Re: Can I avoid OS freezes on compute intense tasks

#6 2021-04-28 09:29:46

Re: Can I avoid OS freezes on compute intense tasks

Board footer