Bug 24849 - kernel crash at boot with a Esxi 6.0 host on an Intel Xeon Silver 4112
Summary: kernel crash at boot with a Esxi 6.0 host on an Intel Xeon Silver 4112
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 6
Hardware: All Linux
Priority: Normal critical
Target Milestone: ---
Assignee: Thomas Backlund
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2019-05-21 14:25 CEST by José Jorge
Modified: 2019-05-23 11:18 CEST (History)
1 user (show)

See Also:
Source RPM: kernel
CVE:
Status comment:

Attachments
dmesg (152.07 KB, text/plain)
2019-05-21 14:26 CEST, José Jorge 		Details
View All Add an attachment (proposed patch, testcase, etc.)

Description José Jorge 2019-05-21 14:25:49 CEST 
We have 3 ESXi 6.0 hosts running VMs. In two of them, Mageia works nicely. In the last, it stops at boot. I join the dmesg read through a virtual serial port.

Mageia 7 has the same problem.
Comment 1 José Jorge 2019-05-21 14:26:15 CEST 
Created attachment 11017 [details]
dmesg
José Jorge 2019-05-21 14:26:20 CEST

CC: (none) => lists.jjorge

Comment 2 José Jorge 2019-05-21 14:26:48 CEST 
It seems a FPU problem as far as I understand the dmesg.
Comment 3 Thomas Backlund 2019-05-22 13:51:45 CEST 
I wonder if the ESXi is exposing a cpuid flag that the underlaying Xeon Silver 4112 is not providing...

Can you provide the log from Cauldron with kernel-5.1.3-6 or later.

Also, you can try to boot with "noxsave" on kernel command line
Comment 5 José Jorge 2019-05-23 11:18:49 CEST 
(In reply to Thomas Backlund from comment #4)
> Is your ESXi 6.0 at U3 ?
> 
> https://www.vmware.com/resources/compatibility/detail.
> php?deviceCategory=cpu&productid=119&vcl=true

You got it, we were in U1... it boots nicely with U3 in this hardware.

Resolution: (none) => FIXED
Status: NEW => RESOLVED
Note You need to log in before you can comment on or make changes to this bug.