xen/privcmd: Support correctly 64KB page granularity when mapping memory
authorJulien Grall <julien.grall@arm.com>
Wed, 31 May 2017 13:03:57 +0000 (14:03 +0100)
committerJuergen Gross <jgross@suse.com>
Wed, 7 Jun 2017 09:23:14 +0000 (11:23 +0200)
Commit 5995a68 "xen/privcmd: Add support for Linux 64KB page granularity" did
not go far enough to support 64KB in mmap_batch_fn.

The variable 'nr' is the number of 4KB chunk to map. However, when Linux
is using 64KB page granularity the array of pages (vma->vm_private_data)
contain one page per 64KB. Fix it by incrementing st->index correctly.

Furthermore, st->va is not correctly incremented as PAGE_SIZE !=
XEN_PAGE_SIZE.

Fixes: 5995a68 ("xen/privcmd: Add support for Linux 64KB page granularity")
CC: stable@vger.kernel.org
Reported-by: Feng Kan <fkan@apm.com>
Signed-off-by: Julien Grall <julien.grall@arm.com>
Reviewed-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Signed-off-by: Juergen Gross <jgross@suse.com>
drivers/xen/privcmd.c

index 7a92a5e1d40c6f17227936ee2b6925286deab511..feca75b07fddce01e6e121542d3e0b74d321f85f 100644 (file)
@@ -362,8 +362,8 @@ static int mmap_batch_fn(void *data, int nr, void *state)
                                st->global_error = 1;
                }
        }
-       st->va += PAGE_SIZE * nr;
-       st->index += nr;
+       st->va += XEN_PAGE_SIZE * nr;
+       st->index += nr / XEN_PFN_PER_PAGE;
 
        return 0;
 }