On 03/13/2014 01:57 PM, Alex Shi wrote:
In the cpu_load decay usage, we mixed the long term, short term load with balance bias, randomly pick a big/small value from them according to balance destination or source. This mix is wrong, the balance bias should be based on task moving cost between cpu groups, not on random history or instant load. History load maybe diverage a lot from real load, that lead to incorrect bias.
In fact, the cpu_load decays can be replaced by the sched_avg decay, that also decays load on time. The balance bias part can fullly use fixed bias -- imbalance_pct, which is already used in newly idle, wake, forkexec balancing and numa balancing scenarios.
Currently the only working idx is busy_idx and idle_idx. As to busy_idx: We mix history load decay and bias together. The ridiculous thing is, when all cpu load are continuous stable, long/short term load is same. then we lose the bias meaning, so any minimum imbalance may cause unnecessary task moving. To prevent this funny thing happen, we have to reuse the imbalance_pct again in find_busiest_group(). But that clearly causes over bias in normal time. If there are some burst load in system, it is more worse.
Any comments?
As to idle_idx: Though I have some cencern of usage corretion, https://lkml.org/lkml/2014/3/12/247, but since we are working on cpu idle migration into scheduler. The problem will be reconsidered. We don't need to care it now.
This patch removed the cpu_load idx decay, since it can be replaced by sched_avg feature. and left the imbalance_pct bias untouched, since only idle_idx missed it, but it is fine. and will be reconsidered soon.
V5, 1, remove unify bias patch and biased_load function. Thanks for PeterZ's comments! 2, remove get_sd_load_idx() in the 1st patch as SrikarD's suggestion. 3, remove LB_BIAS feature, it is not needed now.