From: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Date: Thu, 29 Nov 2012 03:17:38 +0000 (+1100)
Subject: thp-implement-splitting-pmd-for-huge-zero-page-v6
X-Git-Tag: next-20121205~1^2~263
X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=8963251cf4dbb392bf7f77b7afd11fb36b67fb01;p=karo-tx-linux.git

thp-implement-splitting-pmd-for-huge-zero-page-v6

We can't split huge zero page itself (and it's bug if we try), but we
can split the pmd which points to it.

On splitting the pmd we create a table with all ptes set to normal zero
page.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@linux.intel.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 4b3f410f249d..62209b018aae 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2643,6 +2643,7 @@ static int khugepaged(void *none)
 static void __split_huge_zero_page_pmd(struct vm_area_struct *vma,
 		unsigned long haddr, pmd_t *pmd)
 {
+	struct mm_struct *mm = vma->vm_mm;
 	pgtable_t pgtable;
 	pmd_t _pmd;
 	int i;
@@ -2650,8 +2651,8 @@ static void __split_huge_zero_page_pmd(struct vm_area_struct *vma,
 	pmdp_clear_flush(vma, haddr, pmd);
 	/* leave pmd empty until pte is filled */
 
-	pgtable = get_pmd_huge_pte(vma->vm_mm);
-	pmd_populate(vma->vm_mm, &_pmd, pgtable);
+	pgtable = pgtable_trans_huge_withdraw(mm);
+	pmd_populate(mm, &_pmd, pgtable);
 
 	for (i = 0; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) {
 		pte_t *pte, entry;
@@ -2659,11 +2660,11 @@ static void __split_huge_zero_page_pmd(struct vm_area_struct *vma,
 		entry = pte_mkspecial(entry);
 		pte = pte_offset_map(&_pmd, haddr);
 		VM_BUG_ON(!pte_none(*pte));
-		set_pte_at(vma->vm_mm, haddr, pte, entry);
+		set_pte_at(mm, haddr, pte, entry);
 		pte_unmap(pte);
 	}
 	smp_wmb(); /* make pte visible before pmd */
-	pmd_populate(vma->vm_mm, pmd, pgtable);
+	pmd_populate(mm, pmd, pgtable);
 }
 
 void __split_huge_page_pmd(struct vm_area_struct *vma, unsigned long address,
@@ -2678,7 +2679,7 @@ void __split_huge_page_pmd(struct vm_area_struct *vma, unsigned long address,
 	BUG_ON(vma->vm_start > haddr || vma->vm_end < haddr + HPAGE_PMD_SIZE);
 
 	mmun_start = haddr;
-	mmun_end   = address + HPAGE_PMD_SIZE;
+	mmun_end   = haddr + HPAGE_PMD_SIZE;
 	mmu_notifier_invalidate_range_start(mm, mmun_start, mmun_end);
 	spin_lock(&mm->page_table_lock);
 	if (unlikely(!pmd_trans_huge(*pmd))) {