[PATCH v3 hmm 02/11] mm/mmu_notifiers: do not speculatively allocate a mmu_notifier_mm

Ralph Campbell rcampbell at nvidia.com
Wed Aug 14 20:32:13 UTC 2019


On 8/6/19 4:15 PM, Jason Gunthorpe wrote:
> From: Jason Gunthorpe <jgg at 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 at mellanox.com>

Looks good to me.
Reviewed-by: Ralph Campbell <rcampbell at nvidia.com>


More information about the dri-devel mailing list