Description of problem: On x86_64, the tick divider patch seems to have multiple bugs, which cause problems on VMware ESX server. In particular, because not all of the clocks are being properly divided, it needs many more interrupts than a kernel re-compiled to 100HZ would, and can cause time drift. The problems identified so far are: 1) It does not set the local APIC to the "divided" value; it leaves it at the undivided value. 2) It does not properly account for lost ticks when using PM-timer (it thinks there are more lost ticks than there actually are). 3) It does not properly account for lost ticks when using TSC. A preliminary patch (based on input from VMware) is attached.
Created attachment 205451 [details] Patch to fix some of the tick divider problems
Looks sane on a first glance
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 release.
A patch for this issue has been included in build 2.6.18-53.1.2
confirmed fix is in the -53.1.2 kernel, looks to have already been verified by customer.
setting back to on_qa as this will go out to customers in a later erratum.
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/RHSA-2007-0993.html