From: Ross Zwisler <ross.zwis...@linux.intel.com>

commit d0f0931de936a0a468d7e59284d39581c16d3a73 upstream.

When the pmd_devmap() checks were added by 5c7fb56e5e3f ("mm, dax:
dax-pmd vs thp-pmd vs hugetlbfs-pmd") to add better support for DAX huge
pages, they were all added to the end of if() statements after existing
pmd_trans_huge() checks.  So, things like:

  -       if (pmd_trans_huge(*pmd))
  +       if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd))

When further checks were added after pmd_trans_unstable() checks by
commit 7267ec008b5c ("mm: postpone page table allocation until we have
page to map") they were also added at the end of the conditional:

  +       if (pmd_trans_unstable(fe->pmd) || pmd_devmap(*fe->pmd))

This ordering is fine for pmd_trans_huge(), but doesn't work for
pmd_trans_unstable().  This is because DAX huge pages trip the bad_pmd()
check inside of pmd_none_or_trans_huge_or_clear_bad() (called by
pmd_trans_unstable()), which prints out a warning and returns 1.  So, we
do end up doing the right thing, but only after spamming dmesg with
suspicious looking messages:

  mm/pgtable-generic.c:39: bad pmd ffff8808daa49b88(84000001006000a5)

Reorder these checks in a helper so that pmd_devmap() is checked first,
avoiding the error messages, and add a comment explaining why the
ordering is important.

Fixes: commit 7267ec008b5c ("mm: postpone page table allocation until we have 
page to map")
Link: 
http://lkml.kernel.org/r/20170522215749.23516-1-ross.zwis...@linux.intel.com
Signed-off-by: Ross Zwisler <ross.zwis...@linux.intel.com>
Reviewed-by: Jan Kara <j...@suse.cz>
Cc: Pawel Lebioda <pawel.lebi...@intel.com>
Cc: "Darrick J. Wong" <darrick.w...@oracle.com>
Cc: Alexander Viro <v...@zeniv.linux.org.uk>
Cc: Christoph Hellwig <h...@lst.de>
Cc: Dan Williams <dan.j.willi...@intel.com>
Cc: Dave Hansen <dave.han...@intel.com>
Cc: Matthew Wilcox <mawil...@microsoft.com>
Cc: "Kirill A . Shutemov" <kirill.shute...@linux.intel.com>
Cc: Dave Jiang <dave.ji...@intel.com>
Cc: Xiong Zhou <xz...@redhat.com>
Cc: Eryu Guan <eg...@redhat.com>
Cc: <sta...@vger.kernel.org>
Signed-off-by: Andrew Morton <a...@linux-foundation.org>
Signed-off-by: Linus Torvalds <torva...@linux-foundation.org>
---
 mm/memory.c |   40 ++++++++++++++++++++++++++++++----------
 1 file changed, 30 insertions(+), 10 deletions(-)

diff --git a/mm/memory.c b/mm/memory.c
index e2e68767a373..d2db2c4eb0a4 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -2848,6 +2848,17 @@ static int __do_fault(struct fault_env *fe, pgoff_t 
pgoff,
        return ret;
 }
 
+/*
+ * The ordering of these checks is important for pmds with _PAGE_DEVMAP set.
+ * If we check pmd_trans_unstable() first we will trip the bad_pmd() check
+ * inside of pmd_none_or_trans_huge_or_clear_bad(). This will end up correctly
+ * returning 1 but not before it spams dmesg with the pmd_clear_bad() output.
+ */
+static int pmd_devmap_trans_unstable(pmd_t *pmd)
+{
+       return pmd_devmap(*pmd) || pmd_trans_unstable(pmd);
+}
+
 static int pte_alloc_one_map(struct fault_env *fe)
 {
        struct vm_area_struct *vma = fe->vma;
@@ -2871,18 +2882,27 @@ static int pte_alloc_one_map(struct fault_env *fe)
 map_pte:
        /*
         * If a huge pmd materialized under us just retry later.  Use
-        * pmd_trans_unstable() instead of pmd_trans_huge() to ensure the pmd
-        * didn't become pmd_trans_huge under us and then back to pmd_none, as
-        * a result of MADV_DONTNEED running immediately after a huge pmd fault
-        * in a different thread of this mm, in turn leading to a misleading
-        * pmd_trans_huge() retval.  All we have to ensure is that it is a
-        * regular pmd that we can walk with pte_offset_map() and we can do that
-        * through an atomic read in C, which is what pmd_trans_unstable()
-        * provides.
+        * pmd_trans_unstable() via pmd_devmap_trans_unstable() instead of
+        * pmd_trans_huge() to ensure the pmd didn't become pmd_trans_huge
+        * under us and then back to pmd_none, as a result of MADV_DONTNEED
+        * running immediately after a huge pmd fault in a different thread of
+        * this mm, in turn leading to a misleading pmd_trans_huge() retval.
+        * All we have to ensure is that it is a regular pmd that we can walk
+        * with pte_offset_map() and we can do that through an atomic read in
+        * C, which is what pmd_trans_unstable() provides.
         */
-       if (pmd_trans_unstable(fe->pmd) || pmd_devmap(*fe->pmd))
+       if (pmd_devmap_trans_unstable(fe->pmd))
                return VM_FAULT_NOPAGE;
 
+       /*
+        * At this point we know that our vmf->pmd points to a page of ptes
+        * and it cannot become pmd_none(), pmd_devmap() or pmd_trans_huge()
+        * for the duration of the fault.  If a racing MADV_DONTNEED runs and
+        * we zap the ptes pointed to by our vmf->pmd, the vmf->ptl will still
+        * be valid and we will re-check to make sure the vmf->pte isn't
+        * pte_none() under vmf->ptl protection when we return to
+        * alloc_set_pte().
+        */
        fe->pte = pte_offset_map_lock(vma->vm_mm, fe->pmd, fe->address,
                        &fe->ptl);
        return 0;
@@ -3456,7 +3476,7 @@ static int handle_pte_fault(struct fault_env *fe)
                fe->pte = NULL;
        } else {
                /* See comment in pte_alloc_one_map() */
-               if (pmd_trans_unstable(fe->pmd) || pmd_devmap(*fe->pmd))
+               if (pmd_devmap_trans_unstable(fe->pmd))
                        return 0;
                /*
                 * A regular pmd is established and it can't morph into a huge

Reply via email to