Messages in this thread | | | Date | Thu, 20 Jan 2000 14:15:53 -0500 | From | James Manning <> | Subject | Re: static int's for proc_change_penalty and tlb_flush_penalty |
| |
[ Thursday, January 20, 2000 ] Andrea Arcangeli wrote: > I am not sure if it worth to have the cacheline penality in 2.4.x > consdiering I don't like by-hand settings, and that the scheduler > algorithm should have the best values as default.
I'm hoping that with testing over a large number of SMP systems (dual celeries -> 8-way 2MB xeon's at least) that there's enough effective overlap to make this possible... if the "best" values for a given accepted workload vary largely enough over a range of machines, we could pick the "least damage" value and possible have a CONFIG_ option for "Tunable SMP scheduling"? Or would that still be a poor choice for 2.4.x?
James -- Miscellaneous Engineer --- IBM Netfinity Performance Development
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |