+++ This bug was initially created as a clone of Bug #594545 +++
Description of problem:
(Just as the kernel suffers from this problem so does the Xen hypervisor.)
Starting from Westmere, VMX transitions with cr0.cd=1 don't work properly (because of some VMX performance optimizations) and the scenario where one logical cpu is doing VMX activity and another logical cpu coming online can result in system crash.
Fortunately, the only time that cr0.cd is set to one is during MTRR updates so the kernel needs to bring all CPUs to a rendezvous, do the MTRR update and then resume all the CPUs.
The upstream Linux kernel has solved this problem with the patch:
so this patch set needs to be back ported to the RHEL 5.6 kernel.
This request was evaluated by Red Hat Product Management for inclusion in a Red
Hat Enterprise Linux maintenance release. Product Management has requested
further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed
products. This request is not yet committed for inclusion in an Update
About the bug description, the Xen patches are actually at:
Created attachment 417442 [details]
This is the actual Xen hypervisor patch that needs to be backported.
You can download this test kernel from http://people.redhat.com/jwilson/el5
Detailed testing feedback is always welcomed.
Confirmed patch in kernel 2.6.18-212.el5.
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on therefore solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.