Intel Hyperthreading bug...

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

zir_blazer

Active Member
Dec 5, 2016
356
128
43
Technically it is already fixed in Microcode. AMD had a similar issue with SMT in Ryzen, also patched with Microcode.
 

Tom5051

Active Member
Jan 18, 2017
359
79
28
46
The issue only occurs under very specific circumstances like running prime95. Pretty unlikely the average punter will run into this.
 

Rhinox

Member
May 27, 2013
144
26
18
It is triggered by specific code sequence rather than some specific application, as Intel writes in documentation:

"...short loops of less than 64 instructions that use AH, BH, CH or DH registers as well as their corresponding wider register (e.g. RAX, EAX or AX for AH) may cause unpredictable system behavior..."

BTW, if anyone wants to know why high-end xeon cpus never use the latest architecture, here is the answer...
 

niekbergboer

Active Member
Jun 21, 2016
154
59
28
46
Switzerland
Usually included in new BIOS/UEFI releases.

Can also be applied during OS boot (see Microcode - Debian Wiki for a good overview)
Indeed, two of my Proxmox VE / Ceph nodes run a Core i3-6100T (model 94, stepping 3), and these can be updated using the intel-microcode package and a reboot:

[ 0.000000] microcode: microcode updated early to revision 0xba, date = 2017-04-09
[ 0.000000] Linux version 4.10.15-1-pve (root@stretchbuild) (gcc version 6.3.0 20170516 (Debian 6.3.0-18) ) #1 SMP PVE 4.10.15-12 (Mon, 12 Jun 2017 11:18:07 +0200) ()
...
[ 0.093802] smpboot: CPU0: Intel(R) Core(TM) i3-6100T CPU @ 3.20GHz (family: 0x6, model: 0x5e, stepping: 0x3
...
[ 1.275300] microcode: sig=0x506e3, pf=0x2, revision=0xba
[ 1.282128] microcode: Microcode Update Driver: v2.2.


The update 0xba, with signature 0x506e3 is the fix.