Test 6 errors on specific CPU.

[ruifung]

When running memtest86 in order to diagnose a string of random crashes and BSoD, I noticed that regardless of RAM used (I alternated between 2), Test 6 would consistently produce errors when ran on a specific CPU core, which is CPU 6 aka Physical Core 4. The same test, when run using other cores, does not produce any errors. Therefore, is this indicative of a bad CPU core? Also all tests prior to test 6 and does not produce errors on any cores.

Log: https://pastebin.com/GMkidSNG

[David (PassMark)]

Looking at the log it is clear that something is wrong.
Despite the CPU looking like it is at fault, RAM sticks are easier to change, compared to the CPU. So I would still be tempted to remove one of the RAM sticks at a time and retest.

[ruifung]

I have tested each stick individually, removing one at a time. And they still show the same results, no errors up to test 6, and no errors on all but core 4.

[patricklang]

Wow, this is an interesting coincidence. I'm hitting the same problem - one specific core is failing block moves. Have you resolved it yet?

Here's what I've done so far:
- first memory kit worked in Windows for about a week, but showed some bluescreens after that
- memtest86+ failed with first memory kit at this point (single cpu mode)
- reseated CPU & RAM, no different
- replaced RAM, different vendor also on motherboard QVL
- passed ~12 hours of single threaded tests with memtest86+
- wouldn't boot Windows without bugchecking
- switched to Memtest86, ran multicore tests
- core 2 failing a bunch of stuff on block move & moving inversions and eventually hang. Core 0 or 4 will pass alone
- Switched motherboard settings to use just 1 CPU core - Windows boots!

This is the strangest issue I've ever dealt with.

I'm about to try swapping the power supply because I have a known good one available. If that fails, I'll try CPU tomorrow since I can't get another of the same board motherboard locally.

[David (PassMark)]

For the record the CPU in the initial case as a Intel Core i7-7700HQ @ 2.80GHz and the errors where generally one or two bit errors,

[MEM ERROR - Data] Test: 6, CPU: 6, Address: 2FCD4028, Expected: EFFFFFFF, Actual: E7FFFFFF
[MEM ERROR - Data] Test: 6, CPU: 6, Address: 30D74F08, Expected: 00FFFF05, Actual: 0000FFBF
[MEM ERROR - Data] Test: 6, CPU: 6, Address: 31483048, Expected: 00FFFF00, Actual: 0000BB7F
[MEM ERROR - Data] Test: 6, CPU: 6, Address: 3154CF08, Expected: FFFE7FFF, Actual: FFFFFFFF
[MEM ERROR - Data] Test: 6, CPU: 6, Address: 3172A188, Expected: 00FFFFF7, Actual: 0000FFFF

[ruifung]

Wow, this is an interesting coincidence. I'm hitting the same problem - one specific core is failing block moves. Have you resolved it yet?

Here's what I've done so far:
- first memory kit worked in Windows for about a week, but showed some bluescreens after that
- memtest86+ failed with first memory kit at this point (single cpu mode)
- reseated CPU & RAM, no different
- replaced RAM, different vendor also on motherboard QVL
- passed ~12 hours of single threaded tests with memtest86+
- wouldn't boot Windows without bugchecking
- switched to Memtest86, ran multicore tests
- core 2 failing a bunch of stuff on block move & moving inversions and eventually hang. Core 0 or 4 will pass alone
- Switched motherboard settings to use just 1 CPU core - Windows boots!

This is the strangest issue I've ever dealt with.

I'm about to try swapping the power supply because I have a known good one available. If that fails, I'll try CPU tomorrow since I can't get another of the same board motherboard locally.

Well, regrettably for me, this is a laptop. And most of my crashing issues was worked around by configuring windows to only use the first 3 cores. So I'd say there's probably something wrong with the fourth core?

[patricklang]

Well I guess it may be possible for certain cores to fail now. I swapped power supplies - no change. I swapped CPUs today leaving the RAM alone and it's passing with the new CPU. I have never seen a CPU fail that wasn't overclocked and overvolted.

Before swap (Intel i7-8700)

Last 10 Errors

[Data Error] Test: 5, CPU: 0, Address: 2B8301074, Expected: EE56B51A, Actual: EE56B53A

[Data Error] Test: 5, CPU: 0, Address: 2B830105C, Expected: EE56B51A, Actual: EE56B50A

[Data Error] Test: 5, CPU: 0, Address: 2B1F41174, Expected: A6EF1BCF, Actual: A6EF1B86

[Data Error] Test: 5, CPU: 0, Address: 2B1F41164, Expected: A6EF1BCF, Actual: A6EF1BA6

[Data Error] Test: 5, CPU: 0, Address: 2B1F41154, Expected: A6EF1BCF, Actual: A6EF1B59

[Data Error] Test: 5, CPU: 0, Address: 2B1F41144, Expected: A6EF1BCF, Actual: A6EF1B59

[Data Error] Test: 5, CPU: 0, Address: 2AC561174, Expected: B711BCD2, Actual: B711BCF2

[Data Error] Test: 5, CPU: 0, Address: 2AC56115C, Expected: B711BCD2, Actual: B711BCC2

[Data Error] Test: 5, CPU: 0, Address: 2AFE51174, Expected: 18D96198, Actual: 18D961B8

[Data Error] Test: 5, CPU: 0, Address: 2AFE5115C, Expected: 18D96198, Actual: 18D96188

After swap (Intel i3-8100 - only thing in stock)

Test	# Tests Passed	Errors
Test 0 [Address test, walking ones, 1 CPU]	2/2 (100%)	0
Test 1 [Address test, own address, 1 CPU]	2/2 (100%)	0
Test 2 [Address test, own address]	2/2 (100%)	0
Test 3 [Moving inversions, ones & zeroes]	2/2 (100%)	0
Test 4 [Moving inversions, 8-bit pattern]	2/2 (100%)	0
Test 5 [Moving inversions, random pattern]	2/2 (100%)	0
Test 6 [Block move, 64-byte blocks]	2/2 (100%)	0
Test 7 [Moving inversions, 32-bit pattern]	2/2 (100%)	0
Test 8 [Random number sequence]	2/2 (100%)	0
Test 9 [Modulo 20, ones & zeros]	2/2 (100%)	0
Test 10 [Bit fade test, 2 patterns, 1 CPU]	2/2 (100%)	0
Test 13 [Hammer test]	1/1 (100%)	0

[David (PassMark)]

Here's the funny thing. I've been building and fixing PCs for 30 years. I've never seen a CPU failure either. Lots of RAM, disk failures. motherboard and PSU failures. Never a CPU. But in a single week we have 3 people in the forums saying they have CPU failures. Maybe the quality assurance processes are getting worse. Or maybe it was AMD pushing Intel into doing releases for new CPUs before they were properly tested.

[Damjan]

Hi,
This morning I stared to get browsers crashes and blue screen in Win10 (64). Three times in a row after system reboot.
Then I've run memorytest84 7.4 and on 3th pass received CPU related errors on test 6 - attached.
Is this the same thing as above mention or maybe something else?

P.S. there is something like this in Event Viewer:

-

System

- Provider [ Name] Microsoft-Windows-Kernel-Power [ Guid] {331C3B3A-2005-44C2-AC5E-77220C37D6B4}

EventID 41

Version 5

Level 1

Task 63

Opcode 0

Keywords 0x8000400000000002

- TimeCreated [ SystemTime] 2018-01-28T11:16:47.362218800Z

EventRecordID 2404

Correlation

- Execution [ ProcessID] 4 [ ThreadID] 8

Channel System

Computer DESKTOP-FFEFH35

- Security [ UserID] S-1-5-18

-

EventData

BugcheckCode

26

BugcheckParameter1

0x403

BugcheckParameter2

0xffffee0000015ee8

BugcheckParameter3

0x80000001e69dc867

BugcheckParameter4

0x0

SleepInProgress

0

PowerButtonTimestamp

0

BootAppStatus

0

Checkpoint

0

ConnectedStandbyInProgress

false

SystemSleepTransitionsToOn

0

CsEntryScenarioInstanceId

0

BugcheckInfoFromEFI

true

I would appreciate your help on this one.
Thank you in advance.
Damjan

[David (PassMark)]

Damjan,

At the risk of stating the obvious, your RAM errors reported in MemTest86 look like RAM errors. Replace the RAM.

[Axelgrey]

I had the exact same issue as Damjan.
Just want to let you know that it might not necessarily be a memory issue. Had it happen on the insanely binned trident neo 3600 cl14, possibly motherboard was not able to handle it as it acted differently with every bios update.
Icreasing the voltage might have also fixed it but I went instead with increasing the delays slightly so it doesn'to overheat from 14-15-15-15-35 to 15-15-15-15-35 and now it's completely stable with no noticable difference.

[Alecswashere]

I’m not a super experienced person with computers. I started doing some research when my system would just suddenly turn off and reboot (not always successfully). My research lead to me doing a test with memtest86. I can’t give a full report like Damjam but I can say my BugcheckCode was 0 (meaning no error code was recorded). I’m wondering if my CPU has gone bad because of my overclocking (I keep my system cool, usually below 75c tops) any advice as to how I can further diagnose or fix would be appreciated. Also here’s a picture of my errors reported by memtest86

[David (PassMark)]

Start with this page
https://www.memtest86.com/troubleshooting.htm