On Wed, 24 Jun 2020, Jane Chu wrote: > Hi, David, > > On 6/23/2020 9:32 PM, David Rientjes wrote: > > I don't want to hijack Matthew's thread which is primarily about DAX, but > > did get intrigued by your concerns about hugetlbfs page poisoning. We can > > fork the thread off here to discuss only the hugetlb application of this > > if it makes sense to you or you'd like to collaborate on it as well. > > hugetlbfs is not supported in DAX, are you planning to add support? > No, sorry, I was replying only about Tony's comment about the current page poisoning behavior for hugetlbfs and brainstorming a way that we could reduce the memory failure blast radius for hugetlb pages if we had the doublemap support that would also be useful for post-copy live migration. It's independent of DAX so I'm happy to fork this topic off into a different thread for the hugetlb discussion.
