On 3/16/21 3:24 AM, Michal Hocko wrote: > Andrew, > this patchset is still not ready for mmotm. Hugetlb freeing path needs a > major surgery because it can be called from atomic contexts and a simple > fix http://lkml.kernel.org/r/20210311021321.127500-1-mike.kravetz@xxxxxxxxxx > has been nacked by Peter. Until this is handled properly this has to > wait and build on top. > Unfortunately, I tend to agree. Technically, this patchset does not add anything 'new' WRT challenges with the put_page path. It will add a new condition where that path could sleep. However, handling the possibility is an issue with the existing code. What this patchset does add is additional complexity to the existing code paths. It would be better to wait for modifications that take into account put_page/free_huge_page can be called from any context. I am working on such changes and it is a priority. However, this is going to require quite a bit of code change. -- Mike Kravetz
