mm/gup: fix follow_devmap_p[mu]d() on page==NULL handling

This is a bug found not by any report but only by code observations.

When GUP sees a devpmd/devpud and if page==NULL is returned, it means a
fault is probably required.  Here falling through when page==NULL can
cause unexpected behavior.

Fix both cases by catching the page==NULL cases with no_page_table().

Link: https://lkml.kernel.org/r/20231123180222.1048297-1-peterx@redhat.com
Fixes: 3565fce3a6 ("mm, x86: get_user_pages() for dax mappings")
Fixes: 080dbb618b ("mm/follow_page_mask: split follow_page_mask to smaller functions.")
Signed-off-by: Peter Xu <peterx@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
This commit is contained in:
Peter Xu 2023-11-23 13:02:22 -05:00 committed by Andrew Morton
parent ac3f3b0a55
commit e9119fb657

View File

@ -710,6 +710,7 @@ static struct page *follow_pmd_mask(struct vm_area_struct *vma,
spin_unlock(ptl);
if (page)
return page;
return no_page_table(vma, flags);
}
if (likely(!pmd_trans_huge(pmdval)))
return follow_page_pte(vma, address, pmd, flags, &ctx->pgmap);
@ -758,6 +759,7 @@ static struct page *follow_pud_mask(struct vm_area_struct *vma,
spin_unlock(ptl);
if (page)
return page;
return no_page_table(vma, flags);
}
if (unlikely(pud_bad(*pud)))
return no_page_table(vma, flags);