On 8/6/19 4:15 PM, Jason Gunthorpe wrote:
From: Jason Gunthorpe jgg@mellanox.com
A prior commit e0f3c3f78da2 ("mm/mmu_notifier: init notifier if necessary") made an attempt at doing this, but had to be reverted as calling the GFP_KERNEL allocator under the i_mmap_mutex causes deadlock, see commit 35cfa2b0b491 ("mm/mmu_notifier: allocate mmu_notifier in advance").
However, we can avoid that problem by doing the allocation only under the mmap_sem, which is already happening.
Since all writers to mm->mmu_notifier_mm hold the write side of the mmap_sem reading it under that sem is deterministic and we can use that to decide if the allocation path is required, without speculation.
The actual update to mmu_notifier_mm must still be done under the mm_take_all_locks() to ensure read-side coherency.
Signed-off-by: Jason Gunthorpe jgg@mellanox.com
Looks good to me. Reviewed-by: Ralph Campbell rcampbell@nvidia.com