The patch below does not apply to the 4.9-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@xxxxxxxxxxxxxxx>.
Possible dependencies:
873aefb376bb ("vfio/type1: Unpin zero pages")
4b6c33b32296 ("vfio/type1: Prepare for batched pinning with struct vfio_batch")
be16c1fd99f4 ("vfio/type1: Change success value of vaddr_get_pfn()")
aae7a75a821a ("vfio/type1: Add proper error unwind for vfio_iommu_replay()")
64019a2e467a ("mm/gup: remove task_struct pointer for all gup code")
bce617edecad ("mm: do page fault accounting in handle_mm_fault")
ed03d924587e ("mm/gup: use a standard migration target allocation callback")
bbe88753bd42 ("mm/hugetlb: make hugetlb migration callback CMA aware")
41b4dc14ee80 ("mm/gup: restrict CMA region by using allocation scope API")
19fc7bed252c ("mm/migrate: introduce a standard migration target allocation function")
d92bbc2719bd ("mm/hugetlb: unify migration callbacks")
b4b382238ed2 ("mm/migrate: move migration helper from .h to .c")
c7073bab5772 ("mm/page_isolation: prefer the node of the source page")
3e4e28c5a8f0 ("mmap locking API: convert mmap_sem API comments")
d8ed45c5dcd4 ("mmap locking API: use coccinelle to convert mmap_sem rwsem call sites")
ca5999fde0a1 ("mm: introduce include/linux/pgtable.h")
420c2091b65a ("mm/gup: introduce pin_user_pages_locked()")
5a36f0f3f518 ("Merge tag 'vfio-v5.8-rc1' of git://github.com/awilliam/linux-vfio")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 873aefb376bbc0ed1dd2381ea1d6ec88106fdbd4 Mon Sep 17 00:00:00 2001
From: Alex Williamson <alex.williamson@xxxxxxxxxx>
Date: Mon, 29 Aug 2022 21:05:40 -0600
Subject: [PATCH] vfio/type1: Unpin zero pages
There's currently a reference count leak on the zero page. We increment
the reference via pin_user_pages_remote(), but the page is later handled
as an invalid/reserved page, therefore it's not accounted against the
user and not unpinned by our put_pfn().
Introducing special zero page handling in put_pfn() would resolve the
leak, but without accounting of the zero page, a single user could
still create enough mappings to generate a reference count overflow.
The zero page is always resident, so for our purposes there's no reason
to keep it pinned. Therefore, add a loop to walk pages returned from
pin_user_pages_remote() and unpin any zero pages.
Cc: stable@xxxxxxxxxxxxxxx
Reported-by: Luboslav Pivarc <lpivarc@xxxxxxxxxx>
Reviewed-by: David Hildenbrand <david@xxxxxxxxxx>
Link: https://lore.kernel.org/r/166182871735.3518559.8884121293045337358.stgit@omen
Signed-off-by: Alex Williamson <alex.williamson@xxxxxxxxxx>
diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
index db516c90a977..8706482665d1 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -558,6 +558,18 @@ static int vaddr_get_pfns(struct mm_struct *mm, unsigned long vaddr,
ret = pin_user_pages_remote(mm, vaddr, npages, flags | FOLL_LONGTERM,
pages, NULL, NULL);
if (ret > 0) {
+ int i;
+
+ /*
+ * The zero page is always resident, we don't need to pin it
+ * and it falls into our invalid/reserved test so we don't
+ * unpin in put_pfn(). Unpin all zero pages in the batch here.
+ */
+ for (i = 0 ; i < ret; i++) {
+ if (unlikely(is_zero_pfn(page_to_pfn(pages[i]))))
+ unpin_user_page(pages[i]);
+ }
+
*pfn = page_to_pfn(pages[0]);
goto done;
}
