在 2020/11/17 上午5:57, akpm@xxxxxxxxxxxxxxxxxxxx 写道: > On a large machine with memcg enabled but not used, the page's lruvec > seeking pass a few pointers, that may lead to lru_lock holding time > increase and a bit regression. Hi Andrew, Few pointer walking actually has no clear impact in above mentioned situation, so could you please remove above section? when you merge the 3 'replace lru pgdat lru_lock' patches linux-next/akpm or master branch. 3d0d6a8434e1 mm-lru-replace-pgdat-lru_lock-with-lruvec-lock-fix-2 6bffb6674742 mm-lru-replace-pgdat-lru_lock-with-lruvec-lock-fix 7c3520054e02 mm/lru: replace pgdat lru_lock with lruvec lock Also thanks a lot for couple commit log polish on whole series! Thanks again! Alex
