[PATCH 2/5] khugepaged: reuse the smp_wmb() inside __SetPageUptodate()

From: Miaohe Lin <linmiaohe@huawei.com>
To: <akpm@linux-foundation.org>
Cc: <riel@redhat.com>, <kirill.shutemov@linux.intel.com>,
	<ebru.akagunduz@gmail.com>, <dan.carpenter@oracle.com>,
	<linux-kernel@vger.kernel.org>, <linux-mm@kvack.org>,
	<linmiaohe@huawei.com>
Subject: [PATCH 2/5] khugepaged: reuse the smp_wmb() inside __SetPageUptodate()
Date: Thu, 4 Mar 2021 07:30:10 -0500	[thread overview]
Message-ID: <20210304123013.23560-3-linmiaohe@huawei.com> (raw)
In-Reply-To: <20210304123013.23560-1-linmiaohe@huawei.com>

smp_wmb() is needed to avoid the copy_huge_page writes to become visible
after the set_pmd_at() write here. But we can reuse the smp_wmb() inside
__SetPageUptodate() to remove this redundant one.

Signed-off-by: Miaohe Lin <linmiaohe@huawei.com>
---
 mm/khugepaged.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)

diff --git a/mm/khugepaged.c b/mm/khugepaged.c
index d43812c5ce16..287e7ecf978c 100644
--- a/mm/khugepaged.c
+++ b/mm/khugepaged.c
@@ -1183,19 +1183,18 @@ static void collapse_huge_page(struct mm_struct *mm,
 	__collapse_huge_page_copy(pte, new_page, vma, address, pte_ptl,
 			&compound_pagelist);
 	pte_unmap(pte);
+	/*
+	 * spin_lock() below is not the equivalent of smp_wmb(), but
+	 * the smp_wmb() inside __SetPageUptodate() can be reused to
+	 * avoid the copy_huge_page writes to become visible after
+	 * the set_pmd_at() write.
+	 */
 	__SetPageUptodate(new_page);
 	pgtable = pmd_pgtable(_pmd);
 
 	_pmd = mk_huge_pmd(new_page, vma->vm_page_prot);
 	_pmd = maybe_pmd_mkwrite(pmd_mkdirty(_pmd), vma);
 
-	/*
-	 * spin_lock() below is not the equivalent of smp_wmb(), so
-	 * this is needed to avoid the copy_huge_page writes to become
-	 * visible after the set_pmd_at() write.
-	 */
-	smp_wmb();
-
 	spin_lock(pmd_ptl);
 	BUG_ON(!pmd_none(*pmd));
 	page_add_new_anon_rmap(new_page, vma, address, true);
-- 
2.19.1

