RE: [PATCH v3 5/8] mm: Device exclusive memory access

[Ralph Campbell]

> From: Alistair Popple <apopple@xxxxxxxxxx>
> Sent: Thursday, February 25, 2021 11:18 PM
> To: linux-mm@xxxxxxxxx; nouveau@xxxxxxxxxxxxxxxxxxxxx;
> bskeggs@xxxxxxxxxx; akpm@xxxxxxxxxxxxxxxxxxxx
> Cc: linux-doc@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; dri-
> devel@xxxxxxxxxxxxxxxxxxxxx; John Hubbard <jhubbard@xxxxxxxxxx>; Ralph
> Campbell <rcampbell@xxxxxxxxxx>; jglisse@xxxxxxxxxx; Jason Gunthorpe
> <jgg@xxxxxxxxxx>; hch@xxxxxxxxxxxxx; daniel@xxxxxxxx; Alistair Popple
> <apopple@xxxxxxxxxx>
> Subject: [PATCH v3 5/8] mm: Device exclusive memory access
> 
> Some devices require exclusive write access to shared virtual memory (SVM)
> ranges to perform atomic operations on that memory. This requires CPU page
> tables to be updated to deny access whilst atomic operations are occurring.
> 
> In order to do this introduce a new swap entry type (SWP_DEVICE_EXCLUSIVE).
> When a SVM range needs to be marked for exclusive access by a device all page
> table mappings for the particular range are replaced with device exclusive swap
> entries. This causes any CPU access to the page to result in a fault.
> 
> Faults are resovled by replacing the faulting entry with the original mapping. This
> results in MMU notifiers being called which a driver uses to update access
> permissions such as revoking atomic access. After notifiers have been called the
> device will no longer have exclusive access to the region.
> 
> Signed-off-by: Alistair Popple <apopple@xxxxxxxxxx>
> ---
>  Documentation/vm/hmm.rst |  15 ++++
>  include/linux/rmap.h     |   3 +
>  include/linux/swap.h     |   4 +-
>  include/linux/swapops.h  |  44 ++++++++++-
>  mm/hmm.c                 |   5 ++
>  mm/memory.c              | 108 +++++++++++++++++++++++++-
>  mm/mprotect.c            |   8 ++
>  mm/page_vma_mapped.c     |   9 ++-
>  mm/rmap.c                | 163 +++++++++++++++++++++++++++++++++++++++
>  9 files changed, 352 insertions(+), 7 deletions(-)
...
> +int make_device_exclusive_range(struct mm_struct *mm, unsigned long start,
> +				unsigned long end, struct page **pages) {
> +	long npages = (end - start) >> PAGE_SHIFT;
> +	long i;

Nit: you should use unsigned long for 'i' and 'npages' to match start/end.