[PATCH v5 1/8] mm/huge_memory: only split PMD mapping when necessary in unmap_folio()

From: Zi Yan <zi.yan@sent.com>
To: "Pankaj Raghav (Samsung)" <kernel@pankajraghav.com>, linux-mm@kvack.org
Cc: "Zi Yan" <ziy@nvidia.com>,
	"Matthew Wilcox (Oracle)" <willy@infradead.org>,
	"David Hildenbrand" <david@redhat.com>,
	"Yang Shi" <shy828301@gmail.com>, "Yu Zhao" <yuzhao@google.com>,
	"Kirill A . Shutemov" <kirill.shutemov@linux.intel.com>,
	"Ryan Roberts" <ryan.roberts@arm.com>,
	"Michal Koutný" <mkoutny@suse.com>,
	"Roman Gushchin" <roman.gushchin@linux.dev>,
	"Zach O'Keefe" <zokeefe@google.com>,
	"Hugh Dickins" <hughd@google.com>,
	"Luis Chamberlain" <mcgrof@kernel.org>,
	"Andrew Morton" <akpm@linux-foundation.org>,
	linux-kernel@vger.kernel.org, cgroups@vger.kernel.org,
	linux-fsdevel@vger.kernel.org, linux-kselftest@vger.kernel.org
Subject: [PATCH v5 1/8] mm/huge_memory: only split PMD mapping when necessary in unmap_folio()
Date: Mon, 26 Feb 2024 15:55:27 -0500	[thread overview]
Message-ID: <20240226205534.1603748-2-zi.yan@sent.com> (raw)
In-Reply-To: <20240226205534.1603748-1-zi.yan@sent.com>

From: Zi Yan <ziy@nvidia.com>

As multi-size THP support is added, not all THPs are PMD-mapped, thus
during a huge page split, there is no need to always split PMD mapping
in unmap_folio(). Make it conditional.

Signed-off-by: Zi Yan <ziy@nvidia.com>
---
 mm/huge_memory.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 28341a5067fb..b20e535e874c 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2727,11 +2727,14 @@ void vma_adjust_trans_huge(struct vm_area_struct *vma,
 
 static void unmap_folio(struct folio *folio)
 {
-	enum ttu_flags ttu_flags = TTU_RMAP_LOCKED | TTU_SPLIT_HUGE_PMD |
-		TTU_SYNC | TTU_BATCH_FLUSH;
+	enum ttu_flags ttu_flags = TTU_RMAP_LOCKED | TTU_SYNC |
+		TTU_BATCH_FLUSH;
 
 	VM_BUG_ON_FOLIO(!folio_test_large(folio), folio);
 
+	if (folio_test_pmd_mappable(folio))
+		ttu_flags |= TTU_SPLIT_HUGE_PMD;
+
 	/*
 	 * Anon pages need migration entries to preserve them, but file
 	 * pages can simply be left unmapped, then faulted back on demand.
-- 
2.43.0


