Skylake hard locks seeminly when IDLE / CStates ON

[GGlyn]

Myself and others have been struggling with a different hard lock problem with Skylake and we have ran out of options but to assume it is a CPU fault/issue. Does anyone have any comments or feedback?

The long thread is over here, but I will summarise. http://www.tomshardware.co.uk/forum/id-2830772/skylake-build-randomly-freezing-crashing/page-7.html New Skylake Build Randomly Freezing/Crashing - Page 7 - windows 10 - Windows 10

Our systems hard lock randomly but appears on the most part to be when IDLE or next to IDLE. The hardlock can come when playing video, browsing (reading) or mostly when not interacting at all. Some (like myself) noted that they seem to occur more often if you are doing a large amount of disk IO (partition move for example or copying 40+ GB of data disk to disk).

Hard locks happen in either Windows (7,8,10) or Linux.

With CStates on, the crashes are often and usually within 10 minutes of IDLE. With CStates disabled completely most find that the system remains "mainly" stable for days. Generally, no one has reported locks when gaming or under high CPU load (so I do not think this is the Prime95 example but then, since compilers / drivers use the latest instructions (according to CPUID) I cant be sure).

The symptoms happen on many different manufacturers motherboards, RAM types and setups. We have successfully removed or replaced all components and the only change that seems to make a difference is CPU for a few. CStates off in the BIOS is absolutely a guarenteed way to stabilize significantly but not fix.

I changed CPU (after trying everything else) and find it now mostly stable with CStates ON. I must be clear that my system is 100% identical, CPU # 1 crashes all the time and must have CStates off, CPU # 2 does not crash so often even with CStates ON. One may argue this is a problem with my PC, a fault, but with so many people having the exact same issues I cant believe it is a defect. And with Cstates/SpeedStep so specifically (a known new part of Skylakes architecture), I remain convinced that this is something real.

Others were not so lucky, a replacement CPU did not make the situation better. Batches from different factories as well etc etc

My new CPU has hard locked but not when IDLE like the last one (not yet anyway). Mine can lock when under full load, video transcoding, but not more than 1 in 10 sessions. Cstates ON seems equally stable on this CPU as OFF.

Thanks for any suggestions you can give.

[satas]

it could be the only solution for many of us who wasted so much time troubleshooting.

getting a stable system has higher priority than finding out what was actually wrong at the end of the day.

[AThey]

To Ronin and Sahafiec

You both correct... But I need my desktop for work, games and studies; and after 4 months of troubleshlooting and fiddling I needed to move on. Also the forum was pointing to the CPU and the Asus Z170 mainboards... So i went for broke. With hindsight I should have dropped my i5 6600K into the new B150 main board and that would have informed on the fault... Maybe I'll still do this when I get some time...

[CCris2]

Well that B150 doesn't have all the features that z170 has, but at the end you may end up with 2 computers in case you find what component is the culprit, and you go with it in service, and fix it or replace.

[satas]

I can understand that, from what I read on the forums mostly 6600k and 6700K are affected by the issue.

bareley other models, I saw only a couple of 6500 user reporting that.

[LMonk]

I think i have stable system now. Disabled c-states, disabled svid, set min cpu fequency in windows to around 50%, changed bclk to 102mhz (i tried everything but not changing bclk, so i did it), disabled sleep mode in system, have no any other overclock. 8 days uptime already and had heavy stress tests at start.

[CCris2]

Hi,

I think that's not a solution, your cpu should be able to run untouched ...

[LMonk]

Yes i know, specialy version K which must be covered with special tests for stability i think. But i don't like idea to send cpu from my computer and wait 20-40 days and maybe recieve it back.

[CCris2]

What mainboard and memory do you have ?

[satas]

the issue is unfortunately not solved Ronin, I updated my thread with some bad news.

[Ronny_G_Intel]

Hello sahafiec,

I am really sorry to hear that your system is still exhibiting issues,

• Is the latest BIOS being used? According to this Asus Z 170 A motherboard URL: http://www.asus.com/us/Motherboards/Z170-A/HelpDesk_Download/ http://www.asus.com/us/Motherboards/Z170-A/HelpDesk_Download/, the latest BIOS is 1702 that is dated 02/29/2016

• Would you please also check (once latest BIOS is loaded) what is the MCU version? This information should be available in BIOS.• Would you also please elaborate with more details on the failing scenario?

Regards,

Ronny G

[CCris2]

Hello,

His mainboard is Asus z170 Maximus Gene and he has latest bios dated this month. He tested 2 CPU already, and his problem is random freeze or errors while idling, but after few days or a week.

All the problems his system has, makes me believe are from mainboard.

More details are here

[satas]

Hi rguevara,

as Ronin already mentioned my bios is the latest 1601, the MCU version I'll have to check.

the failing scenario is mostly the same, system runs on stock setting with only rams at XMP.

no issues under load but after some random time, 1 day to 1 week, the system crashes with BugCheck error and resets.

last crash was on idle, I was even not at home and the BugCheck was "IRQ_NOT_LESS_OR_EQUAL".

I updated my thread, the link was already given by Ronin.

if you need any further information we can also use the private conversation we newly started.

Greetings,

Sahin

[WGord1]

Hello, I recently resolved my idle freeze issue and I wanted to share b/c it was the most frustrating PC issue I've ever had in my 20+ years of being a "computer nerd". I have been having the same idle issue since I first got my i7-6700k in Nov 2015 (for $430 so getting it fixed when it's down to $345 is super frustrating) but back then I couldn't find any forums or support anywhere so I switched back to my old rig until I came across this forum and implemented the c-state work-around which did stabilize it. For reference, my configuration was the 6700k, 250GB Samsung 950 Pro x4 PCI-E SSD, ASRock Z170 Fatal1ty K4, ASRock vendor approved G.Skill TridentZ 2x8GB DDR4 3200 , XFX XTR 1050W Gold, Windows 10 Home, and EVGA 980ti. I beat myself up trying to figure out what I screwed up. I tried multiple OS reinstalls with minimal components, every BIOS/UEFI update, and swapping components, so I was super relieved to finally hear that others were experiencing my issue but equally furious that Intel hadn't acknowledged an issue it Skylake idle freeze issue. After reading some posts, I decided to try the Intel Processor Check program again (had tried it before but it passed) and it randomly failed for AVX so I saved the log and emailed Intel. I tried the CPU check program again a few times over the next couple days while I awaited a reply from Intel and it passed every time so it failing that one time was random and thank gosh I saved the log from the failure. Intel exchanged the CPU and I installed the replacement a week ago and I have had no freezes (knock on wood) so far with c-states on, where as before I couldn't go more than 1 hour without an idle freeze (the one where the screen locks without blue screen or errors and the audio locks on whatever tone was coming out). The batch S/N's from my bad one was "L537B171" and my new one is "L547B548" is that matters or helps anyone. I had read people saying that a new CPU didn't fix their issue so I was super nervous but in my case it seems like that was the issue. Intel, if you're reading these posts still, thank you for exchanging my CPU but you really need to stand behind you products better by giving the community more feed back about this issue. It seems that there is a significant amount of Skylake CPUs that have a defect.

[satas]

well, there is really not much to add here - glad it worked for you GordonW.

pitty that there are not many user reporting the same good results as you.

I've been going crazy in the cpu-motherboard-ram triangle for months.

what was the exact description in the log files for the freezes you got previously?

was it a BugCheck?

[WGord1]

Please excuse my ignorance, I'm not familiar with how to associate a reason with a restart in the system log. There are 15 instances of the computer having to be reset in logs with the critical event "The system has rebooted without cleanly shutting down first". All 15 of those instances are followed by "The previous system shutdown at xx:xx:xx PM on ‎x/‎xx/‎xxxx was unexpected" but of those, only 5 are followed by bug checks errors. The other 10 don't have reasons I can see, but that might be due to a lack of knowledge on my part. Please let me know if there is a system log somewhere that you are interested in seeing. I am more than willing to do anything I can to help to identify what exactly is causing this issue.

[satas]

no problem GordonW, can you see the BugCheck description and copy/paste it here?

for example I mostly get the "MEMORY_MANAGEMENT" or the "IRQ_NOT_LESS_OR_EQUAL" ones.

[CCris2]

Hi GordonW

From what I understand you have the folowing configuration:

i7 6700k

ASRock Z170 Fatal1ty K4

G.Skill TridentZ 2x8GB DDR4 3200

EVGA 980ti.

250GB Samsung 950 Pro x4 PCI-E SSD

XFX XTR 1050W Gold

Windows 10 Home

What is the cooler you have and do you have updated mainboard bios to the last version which is 3.10 ?

How is that mainboard is it ok ? It seems to me or i might be wrong but many people have problems on mainboards from asus and msi ?

[WGord1]

Hello, I remember seeing "MEMORY_MANAGEMENT" or the "IRQ_NOT_LESS_OR_EQUAL" before when I looked at event viewer but now I only see the following...

The computer has rebooted from a bugcheck. The bugcheck was: 0x00000119 (0x0000000000000001, 0x000000000002cced, 0x000000000002ccef, 0xffffe0015c9f83a0). A dump was saved in: C:\WINDOWS\MEMORY.DMP. Report Id: 78ff150b-6001-45c7-bfa3-e92ef67388ee.

All the bugchecks are the same but with different address entries. I started researching how to work with the memory.dmp file (I've never messed with it before) but seems like it'll take a little bit to understand the basics and it's late so I can reengage it another night if there's data in there you would like to see.

[WGord1]

Hello, that is correct for my setup. I have a Corsair H70 water cooler with Artic Silver paste (applied like their anal-retentive directions recommend) and I haven't noticed any unusual temps, seems to be cooling well. The last BIOS I tried was 2.90 (still froze at idle), then I shipped my CPU back to Intel. I didn't realize a new BIOS was out, but I wasn't looking as closely now that it seems to be fixed. I have no complaints about the board. I'd always been an ASUS guy but I built a little cheap mini-ITX and I went with ASRock b/c I'd heard good things about them and it was very affordable, comparatively. I didn't have any issues with it so I bought another ASRock for my Skylake upgrade and I still have no issues with it (assuming the CPU was actually the culprit, which it looks like it was so far).

[satas]

hi GordonW,

you can have a look at my thread over at Windows 10 forums just to compare the BugCheck codes and possible causes.

there is a lot of useful information in the thread as my dump files were analyzed by the BSOD analists there.

http://www.tenforums.com/bsod-crashes-debugging/40664-bsod-attempted_write_to_readonly_memory-windows-10-pro-64bit.html Windows 10 Forums

[CCris2]

Hi sahafiec

He doesn't have any problem now, on the last bios asrock did more improvements to the memory compatibility, he just conformed his config. I saw that memory improvement on my system also, it decreased latency

PS: Wanted to check the asus forums, just to see what's going one there, and surprise ... Like in the last weekend the forums not working. Is like they make it on purpose, so that people don't post more complains, or they upset someone very badly