linux-stable.git/include/linux/hmm.h, branch v5.6.2

mm/hmm: remove hmm_range_dma_map and hmm_range_dma_unmap

2019-11-23T23:56:45+00:00

These two functions have never been used since they were added.

Link: https://lore.kernel.org/r/20191113134528.21187-1-hch@lst.de
Signed-off-by: Christoph Hellwig 
Reviewed-by: John Hubbard 
Reviewed-by: Jason Gunthorpe 
Signed-off-by: Jason Gunthorpe

mm/hmm: remove hmm_mirror and related

2019-11-23T23:56:45+00:00

The only two users of this are now converted to use mmu_interval_notifier,
delete all the code and update hmm.rst.

Link: https://lore.kernel.org/r/20191112202231.3856-14-jgg@ziepe.ca
Reviewed-by: Jérôme Glisse 
Tested-by: Ralph Campbell 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Jason Gunthorpe

mm/hmm: define the pre-processor related parts of hmm.h even if disabled

2019-11-23T23:56:44+00:00

Only the function calls are stubbed out with static inlines that always
fail. This is the standard way to write a header for an optional component
and makes it easier for drivers that only optionally need HMM_MIRROR.

Link: https://lore.kernel.org/r/20191112202231.3856-5-jgg@ziepe.ca
Reviewed-by: Jérôme Glisse 
Tested-by: Ralph Campbell 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Jason Gunthorpe

mm/hmm: allow hmm_range to be used with a mmu_interval_notifier or hmm_mirror

2019-11-23T23:56:44+00:00

hmm_mirror's handling of ranges does not use a sequence count which
results in this bug:

         CPU0                                   CPU1
                                     hmm_range_wait_until_valid(range)
                                         valid == true
                                     hmm_range_fault(range)
hmm_invalidate_range_start()
   range->valid = false
hmm_invalidate_range_end()
   range->valid = true
                                     hmm_range_valid(range)
                                          valid == true

Where the hmm_range_valid() should not have succeeded.

Adding the required sequence count would make it nearly identical to the
new mmu_interval_notifier. Instead replace the hmm_mirror stuff with
mmu_interval_notifier.

Co-existence of the two APIs is the first step.

Link: https://lore.kernel.org/r/20191112202231.3856-4-jgg@ziepe.ca
Reviewed-by: Jérôme Glisse 
Tested-by: Philip Yang 
Tested-by: Ralph Campbell 
Reviewed-by: Christoph Hellwig 
Signed-off-by: Jason Gunthorpe

hmm: use mmu_notifier_get/put for 'struct hmm'

2019-08-20T12:35:02+00:00

This is a significant simplification, it eliminates all the remaining
'hmm' stuff in mm_struct, eliminates krefing along the critical notifier
paths, and takes away all the ugly locking and abuse of page_table_lock.

mmu_notifier_get() provides the single struct hmm per struct mm which
eliminates mm->hmm.

It also directly guarantees that no mmu_notifier op callback is callable
while concurrent free is possible, this eliminates all the krefs inside
the mmu_notifier callbacks.

The remaining krefs in the range code were overly cautious, drivers are
already not permitted to free the mirror while a range exists.

Link: https://lore.kernel.org/r/20190806231548.25242-6-jgg@ziepe.ca
Reviewed-by: Christoph Hellwig 
Reviewed-by: Ralph Campbell 
Tested-by: Ralph Campbell 
Signed-off-by: Jason Gunthorpe

mm/hmm: remove the page_shift member from struct hmm_range

2019-08-07T17:58:06+00:00

All users pass PAGE_SIZE here, and if we wanted to support single entries
for huge pages we should really just add a HMM_FAULT_HUGEPAGE flag instead
that uses the huge page size instead of having the caller calculate that
size once, just for the hmm code to verify it.

Link: https://lore.kernel.org/r/20190806160554.14046-8-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Felix Kuehling 
Reviewed-by: Jason Gunthorpe 
Signed-off-by: Jason Gunthorpe

mm/hmm: remove superfluous arguments from hmm_range_register

2019-08-07T17:58:05+00:00

The start, end and page_shift values are all saved in the range structure,
so we might as well use that for argument passing.

Link: https://lore.kernel.org/r/20190806160554.14046-7-hch@lst.de
Signed-off-by: Christoph Hellwig 
Reviewed-by: Jason Gunthorpe 
Reviewed-by: Felix Kuehling 
Signed-off-by: Jason Gunthorpe

mm/hmm: remove the unused vma argument to hmm_range_dma_unmap

2019-08-07T17:58:05+00:00

Link: https://lore.kernel.org/r/20190806160554.14046-6-hch@lst.de
Signed-off-by: Christoph Hellwig 
Reviewed-by: Jason Gunthorpe 
Signed-off-by: Jason Gunthorpe

mm/hmm: remove hmm_range vma

2019-07-26T15:35:29+00:00

Since hmm_range_fault() doesn't use the struct hmm_range vma field, remove
it.

Link: https://lore.kernel.org/r/20190726005650.2566-8-rcampbell@nvidia.com
Suggested-by: Jason Gunthorpe 
Signed-off-by: Ralph Campbell 
Reviewed-by: Christoph Hellwig 
Reviewed-by: Jason Gunthorpe 
Signed-off-by: Jason Gunthorpe

mm/hmm: merge hmm_range_snapshot into hmm_range_fault

2019-07-26T14:10:53+00:00

Add a HMM_FAULT_SNAPSHOT flag so that hmm_range_snapshot can be merged
into the almost identical hmm_range_fault function.

Link: https://lore.kernel.org/r/20190726005650.2566-5-rcampbell@nvidia.com
Signed-off-by: Christoph Hellwig 
Signed-off-by: Ralph Campbell 
Reviewed-by: Jason Gunthorpe 
Signed-off-by: Jason Gunthorpe