Hide Forgot
Hi team, This report is somehow related to and can be considered a continuation of: https://bugzilla.redhat.com/show_bug.cgi?id=1870660 We would be looking to minimize the number of context switches from the VDO threads this time. Current state: starting VDO volumes that are not in use on an idle system increases the number of context switches with around 8000 context switches / second. On an idle physical system with 4 CPUs running kernel 5.8.14 and kvdo-corp 6.2.4.14: VDO devices are not started: 90 context switches / second VDO devices(2 devices) are started and not used: 8000 context switches / second The threads that consume the CPU seem to be: [kvdo2:packerQ] [kvdo2:ackQ] [kvdo2:cpuQ0] [kvdo2:cpuQ1] [kvdo3:dedupeQ] [kvdo3:journalQ] [kvdo3:hashQ0] [kvdo3:bioQ0] [kvdo3:bioQ1] [kvdo3:bioQ2] [kvdo3:bioQ3] [kvdo3:cpuQ0] [kvdo3:cpuQ1] In an ideal world, having started an idling VDO volume should not increase the number of context switches on the system. Thank you, -Nikolay
Hi folks, any progress update with this?
(In reply to Nikolay Kichukov from comment #1) > Hi folks, > any progress update with this? Sorry, Nikolay, not so far. I've had another work item ahead of it in my queue... I'll be looking into it this week though.
Verified with kmod-kvdo-6.2.5.72-79.el8
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (kmod-kvdo bug fix and enhancement update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHBA-2021:4359