We're missing xen-unstable cset 16093: # HG changeset patch # User Keir Fraser <keir> # Date 1192101114 -3600 # Node ID 0d7d6804af2204dece108959f4fa40764bd2734f # Parent 49323c8b8633fc02164c8b949c2b55e167aeca1e x86: AMD Fam10/11 adjustments Signed-off-by: Jan Beulich <jbeulich> Sync up affected files with 2.6.23-rc9. Signed-off-by: Keir Fraser <keir> Without it, Xenoprof cannot work on AMD Barcelona. There might be more problems with those machines.
Bhavana, is there anything else missing from the hypervisor we need for the latest AMD hardware?
Not that I'm aware of.
The trouble with xen-unstable cset 16093 is that it mixes unrelated stuff: Xenoprof hardware enablement, MTRR features, cleanups. We need to extract the relevant pieces.
I discussed this issue we have been battling with Jan Beulich and Keir Fraser. The orginal patch that Jan submitted is a lot smaller than the mess that is c/s 16093. Keir said that syncing up upstream resulted in the massive patch. The patch I'm about to attach is needed in the R5.2 HV. as similar changes have already been made to the bare metal kernel to fix Barcelona.
Created attachment 290031 [details] AMD Family 10/11 fixes for HV
Posted patch to RHML and virt list on 12/19
in 2.6.18-71.el5 You can download this test kernel from http://people.redhat.com/dzickus/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 the 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. http://rhn.redhat.com/errata/RHBA-2008-0314.html