mm/hugetlb.c: fix unnecessary address expansion of pmd sharing

The current code would unnecessarily expand the address range.  Consider
one example, (start, end) = (1G-2M, 3G+2M), and (vm_start, vm_end) =
(1G-4M, 3G+4M), the expected adjustment should be keep (1G-2M, 3G+2M)
without expand.  But the current result will be (1G-4M, 3G+4M).  Actually,
the range (1G-4M, 1G) and (3G, 3G+4M) would never been involved in pmd
sharing.

After this patch, if pud aligned *start across vm_start, then we know the
*start and vm_start are in same pud_index, and vm_start is not pud
aligned, so don't adjust *start.  Same logic applied to *end.

Mike said:

: The 'adjusted range' is used for calls to mmu notifiers and cache(tlb)
: flushing.  Since the current code unnecessarily expands the range in
: some cases, more entries than necessary would be flushed.  This
: would/could result in performance degradation.  However, this is highly
: dependent on the user runtime.  Is there a combination of vma layout
: and calls to actually hit this issue?  If the issue is hit, will those
: entries unnecessarily flushed be used again and need to be
: unnecessarily reloaded?

Link: https://lkml.kernel.org/r/20201229042125.2663029-1-lixinhai.lxh@gmail.com
Fixes: commit 75802ca66354 ("mm/hugetlb: fix calculation of adjust_range_if_pmd_sharing_possible")
Signed-off-by: Li Xinhai <lixinhai.lxh@gmail.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index a2602969873dcf6c6bc989bcccb2136bc45348b0..d3caaeab94edbcc8f56b53fb800c69b7460f35de 100644 (file)
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -5269,11 +5269,16 @@ void adjust_range_if_pmd_sharing_possible(struct vm_area_struct *vma,
        a_end = ALIGN(*end, PUD_SIZE);
 
        /*
-        * Intersect the range with the vma range, since pmd sharing won't be
-        * across vma after all
+        * If the PUD aligned address across vma range, then it means the
+        * vm_start/vm_end is not PUD aligned. In that case, we must don't
+        * adjust range because pmd sharing is not possbile at the start and/or
+        * end part of vma.
         */
-       *start = max(vma->vm_start, a_start);
-       *end = min(vma->vm_end, a_end);
+       if (a_start >= vma->vm_start)
+               *start = a_start;
+
+       if (a_end <= vma->vm_end)
+               *end = a_end;
 }
 
 /*