From: John Hubbard <jhubbard@xxxxxxxxxx> Subject: vfio, mm: pin_user_pages (FOLL_PIN) and put_user_page() conversion 1. Change vfio from get_user_pages_remote(), to pin_user_pages_remote(). 2. Because all FOLL_PIN-acquired pages must be released via put_user_page(), also convert the put_page() call over to put_user_pages_dirty_lock(). Note that this effectively changes the code's behavior in vfio_iommu_type1.c: put_pfn(): it now ultimately calls set_page_dirty_lock(), instead of set_page_dirty(). This is probably more accurate. As Christoph Hellwig put it, "set_page_dirty() is only safe if we are dealing with a file backed page where we have reference on the inode it hangs off." [1] [1] https://lore.kernel.org/r/20190723153640.GB720@xxxxxx Link: http://lkml.kernel.org/r/20200107224558.2362728-20-jhubbard@xxxxxxxxxx Signed-off-by: John Hubbard <jhubbard@xxxxxxxxxx> Tested-by: Alex Williamson <alex.williamson@xxxxxxxxxx> Acked-by: Alex Williamson <alex.williamson@xxxxxxxxxx> Cc: Aneesh Kumar K.V <aneesh.kumar@xxxxxxxxxxxxx> Cc: Björn Töpel <bjorn.topel@xxxxxxxxx> Cc: Christoph Hellwig <hch@xxxxxx> Cc: Daniel Vetter <daniel.vetter@xxxxxxxx> Cc: Dan Williams <dan.j.williams@xxxxxxxxx> Cc: Hans Verkuil <hverkuil-cisco@xxxxxxxxx> Cc: Ira Weiny <ira.weiny@xxxxxxxxx> Cc: Jan Kara <jack@xxxxxxx> Cc: Jason Gunthorpe <jgg@xxxxxxxxxxxx> Cc: Jason Gunthorpe <jgg@xxxxxxxx> Cc: Jens Axboe <axboe@xxxxxxxxx> Cc: Jerome Glisse <jglisse@xxxxxxxxxx> Cc: Jonathan Corbet <corbet@xxxxxxx> Cc: Kirill A. Shutemov <kirill@xxxxxxxxxxxxx> Cc: Leon Romanovsky <leonro@xxxxxxxxxxxx> Cc: Mauro Carvalho Chehab <mchehab@xxxxxxxxxx> Cc: Mike Rapoport <rppt@xxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- drivers/vfio/vfio_iommu_type1.c | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) --- a/drivers/vfio/vfio_iommu_type1.c~vfio-mm-pin_user_pages-foll_pin-and-put_user_page-conversion +++ a/drivers/vfio/vfio_iommu_type1.c @@ -309,9 +309,8 @@ static int put_pfn(unsigned long pfn, in { if (!is_invalid_reserved_pfn(pfn)) { struct page *page = pfn_to_page(pfn); - if (prot & IOMMU_WRITE) - SetPageDirty(page); - put_page(page); + + put_user_pages_dirty_lock(&page, 1, prot & IOMMU_WRITE); return 1; } return 0; @@ -329,7 +328,7 @@ static int vaddr_get_pfn(struct mm_struc flags |= FOLL_WRITE; down_read(&mm->mmap_sem); - ret = get_user_pages_remote(NULL, mm, vaddr, 1, flags | FOLL_LONGTERM, + ret = pin_user_pages_remote(NULL, mm, vaddr, 1, flags | FOLL_LONGTERM, page, NULL, NULL); if (ret == 1) { *pfn = page_to_pfn(page[0]); _
