Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
From:   Yin Fengwei <fengwei.yin@intel.com>
To:     willy@infradead.org, ryan.roberts@arm.com,
        linux-arch@vger.kernel.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org
Cc:     fengwei.yin@intel.com
Subject: [PATCH 3/4] mm: mark PTEs referencing the accessed folio young
Date:   Tue, 30 May 2023 16:07:30 +0800
Message-Id: <20230530080731.1462122-4-fengwei.yin@intel.com>
In-Reply-To: <20230530080731.1462122-1-fengwei.yin@intel.com>
References: <ZBho6Q6Xq/YqRmBT@casper.infradead.org>
 <20230530080731.1462122-1-fengwei.yin@intel.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

To allow using larger TLB entries, it's better to mark the
PTEs of same folio accessed when setup the PTEs.

Reported-by: Ryan Roberts <ryan.roberts@arm.com>
Signed-off-by: Yin Fengwei <fengwei.yin@intel.com>
---
 mm/memory.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/mm/memory.c b/mm/memory.c
index c359fb8643e5..2615ea552613 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -4259,7 +4259,7 @@ void set_pte_range(struct vm_fault *vmf, struct folio *folio,
 	struct vm_area_struct *vma = vmf->vma;
 	bool uffd_wp = pte_marker_uffd_wp(vmf->orig_pte);
 	bool write = vmf->flags & FAULT_FLAG_WRITE;
-	bool prefault = vmf->address != addr;
+	bool prefault = (addr > vmf->address) || ((addr + nr) < vmf->address);
 	pte_t entry;
 
 	flush_icache_pages(vma, page, nr);
-- 
2.30.2