http://kerneltrap.org/node/view/3851 Jens Axboe provided a patch that makes the various IO schedulers fully modular, allowing people to switch between them on the fly. This work is relevant to CKRM to avoid the current approach of directly patching the existing CFQ scheduler. My hope is that there will be a CKRM-enabled CFQ and a regular CFQ, which might have the same source base but are compiled separately.
Yes, this is the plan.
I was keen on avoiding a complete duplication of CFQ and was trying to see if we could make the CKRM-IO a runtime variant of CFQ.
But with the limits changes (not sent out yet), it looks like CFQ and CKRM-I/O will diverge. Most of the code will still be duplicated - abstracting it out should be possible (thats something that CFQ needs to do with deadline as well).
Replacing the CFQ shouldn't be done anymore.
-- Shailabh
------------------------------------------------------- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click _______________________________________________ ckrm-tech mailing list https://lists.sourceforge.net/lists/listinfo/ckrm-tech
