Re: [RFC v5 03/11] mm: Introduce pte_spinlock for FAULT_FLAG_SPECULATIVE

2017-08-08 Thread Laurent Dufour
On 08/08/2017 12:35, Anshuman Khandual wrote: > On 06/16/2017 11:22 PM, Laurent Dufour wrote: >> When handling page fault without holding the mmap_sem the fetch of the >> pte lock pointer and the locking will have to be done while ensuring >> that the VMA is not touched in our back. > > It does no

Re: [RFC v5 03/11] mm: Introduce pte_spinlock for FAULT_FLAG_SPECULATIVE

2017-08-08 Thread Anshuman Khandual
On 06/16/2017 11:22 PM, Laurent Dufour wrote: > When handling page fault without holding the mmap_sem the fetch of the > pte lock pointer and the locking will have to be done while ensuring > that the VMA is not touched in our back. It does not change things from whats happening right now, where d