Intel Xeon W-2145 - hyper threading issue

[PH2]

Hi,

Is this processor (Skylake - Intel Xeon W-2145) affected by issue described as SKZ7/SKW144/SKL150/SKX150/SKZ7/KBL095/KBW095?

https://lists.debian.org/debian-devel/2017/06/msg00308.html https://lists.debian.org/debian-devel/2017/06/msg00308.html

I am using HP Z4 G4 with Ubuntu 16. I am facing issues like suspicious crashes of application on whole system freeze. It happens when I am running single application (in my case pybot - part of robotframework) many times (12 or more).

I did many experiments:

- I have similiar workstation (instead of Skylake there is a Broadwell) with the same software configuration - no issues

- I have installed on my system Ubuntu 18 - still issue appears

- I have upgraded BIOS with the newest downloaded from HP - still issue appears

- Finally I found that information about hyper threading issue from June 2017 and as a suggestion I turned off hyper threading support in BIOS - no issues

This is output from my system:

processor : 7

vendor_id : GenuineIntel

cpu family : 6

model : 85

model name : Intel(R) Xeon(R) W-2145 CPU @ 3.70GHz

stepping : 4

microcode : 0x200004d

[ 0.000000] microcode: microcode updated early to revision 0x200004d, date = 2018-05-15

[ 1.661714] microcode: sig=0x50654, pf=0x2, revision=0x200004d

[ 1.661841] microcode: Microcode Update Driver: v2.2.

It strange because I am using microcode which should contain fixes for that issue. It was fixed here: https://launchpad.net/ubuntu/+source/intel-microcode/3.20170707.1~ubuntu16.04.0 3.20170707.1~ubuntu16.04.0 : intel-microcode package : Ubuntu

Is there any known other issue which is not fixed yet and may cause platform instability described above?

This is signature of my hardware: Hardware name: HP HP Z4 G4 Workstation/81C5, BIOS P61 v01.67 08/15/2018

And there is example of segfaults:

Oct 25 13:50:31 hp-ff-1-HP-Z4-G4-Workstation kernel: [ 5215.553430] pybot[26349]: segfault at 30 ip 000055b4b860acef sp 00007ffe728e7ac0 error 4 in python2[55b4b8516000+2ff000]

Oct 25 13:51:24 hp-ff-1-HP-Z4-G4-Workstation kernel: [ 5269.010635] pybot[25804]: segfault at 10 ip 0000562186892cb2 sp 00007ffc04505d10 error 4 in python2[5621867b1000+2ff000]

Oct 25 13:54:52 hp-ff-1-HP-Z4-G4-Workstation kernel: [ 5476.908069] pybot[24524]: segfault at 7f12eeab9000 ip 0000561827317029 sp 00007ffca9889890 error 4 in python2[561827169000+2ff000]

Oct 25 13:59:39 hp-ff-1-HP-Z4-G4-Workstation kernel: [ 5763.718095] pybot[25023]: segfault at 8 ip 0000561082030197 sp 00007ffc73df4200 error 4 in python2[561081f36000+2ff000]

Regards,

Pawel

[idata]

Hello depablo,

Hope you are doing fine.Please let us research about it and we will get back to you as soon as possible.Thank for your patience.Regards,Charlie_Intel

[idata]

Hello depablo,

Hope you are doing fine.Here is the information:

• Intel released updated microcode addressing the issue for some Skylake based processors already
• There's a perl script in the article: https://lists.debian.org/debian-devel/2017/06/msg00367.html https://lists.debian.org/debian-devel/2017/06/msg00367.html that may help identify if the microcode running on the system is affected by the issue

Regards,Charlie_Intel

[PH2]

Hi,

Thank your for your response.

I run this script on my system and I've got following output:

cpu 0: You may need a BIOS/UEFI update (Kaby Lake, or Skylake-X processor)

Note: Kaby Lake X-series processors (i7-7740X, etc) are not affected

I have the newest possible bios and microcode.

In the meantime I looked into following document:

https://www.intel.co.uk/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf https://www.intel.co.uk/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf

There is written that issue SKL150 has no fix. So I am confused it this issue is fixed or not.

Regards,

Pawel

[idata]

Hello depablo,

Let us go deep on this and a will get back to you as soon as possible.Regards,Charlie_Intel

[idata]

Hello depablo,

This is information for microcode update located at the following site:

https://www.intel.com/content/dam/www/public/us/en/documents/sa00115-microcode-update-guidance.pdf https://www.intel.com/content/dam/www/public/us/en/documents/sa00115-microcode-update-guidance.pdf

and the Microcode update download site for Linux OS is here:

https://downloadcenter.intel.com/product/595/Intel-Xeon-Processors https://downloadcenter.intel.com/product/595/Intel-Xeon-Processors

Please let us know if you have further questions.

Regards,

Charlie_Intel

[PH2]

Hi,

It actually doesn't change anything. I already have the newest version of microcode: 0x200004d. (The same version as in your document)

Question is when this issue SKL150 will be fixed? Because this document says it is not fixed yet.:

https://www.intel.co.uk/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf

Regards,

Pawel