Received-SPF: pass (google.com: domain of linux-kernel+bounces-158217-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.80.249 as permitted sender) client-ip=147.75.80.249;
From: Lance Yang <ioworker0@gmail.com>
To: david@redhat.com,
	ziy@nvidia.com
Cc: ioworker0@gmail.com,
	21cnbao@gmail.com,
	akpm@linux-foundation.org,
	fengwei.yin@intel.com,
	linux-kernel@vger.kernel.org,
	linux-mm@kvack.org,
	maskray@google.com,
	mhocko@suse.com,
	minchan@kernel.org,
	peterx@redhat.com,
	ryan.roberts@arm.com,
	shy828301@gmail.com,
	songmuchun@bytedance.com,
	wangkefeng.wang@huawei.com,
	willy@infradead.org,
	xiehuan09@gmail.com,
	zokeefe@google.com
Subject: Re: [PATCH v2 1/1] mm/vmscan: avoid split PMD-mapped THP during shrink_folio_list()
Date: Thu, 25 Apr 2024 16:50:51 +0800
Message-Id: <20240425085051.74889-1-ioworker0@gmail.com>
In-Reply-To: <CAK1f24nb6FkipH3OZa0uwbBWkefS3f2BrJ_GTxkS2j6+6bgODQ@mail.gmail.com>
References: <CAK1f24nb6FkipH3OZa0uwbBWkefS3f2BrJ_GTxkS2j6+6bgODQ@mail.gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

Hey Zi, David,

How about this change(diff against mm-unstable) as follows?

I'd like to add __try_to_unmap_huge_pmd() as a new internal function
specifically for unmapping PMD-mapped folios. If, for any reason, we cannot
unmap the folio, then we'll still split it as previously done.

Currently, __try_to_unmap_huge_pmd() only handles lazyfree THPs, but it
can be extended to support other large folios that are PMD-mapped in the
future if needed.

diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 670218f762c8..0f906dc6d280 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -100,8 +100,6 @@ enum ttu_flags {
 					 * do a final flush if necessary */
 	TTU_RMAP_LOCKED		= 0x80,	/* do not grab rmap lock:
 					 * caller holds it */
-	TTU_LAZYFREE_THP	= 0x100, /* avoid splitting PMD-mapped THPs
-					  * that are marked as lazyfree. */
 };
 
 #ifdef CONFIG_MMU
diff --git a/mm/rmap.c b/mm/rmap.c
index a7913a454028..879c8923abfc 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -1606,6 +1606,19 @@ void folio_remove_rmap_pmd(struct folio *folio, struct page *page,
 #endif
 }
 
+static bool __try_to_unmap_huge_pmd(struct vm_area_struct *vma,
+				    unsigned long addr, struct folio *folio)
+{
+	VM_WARN_ON_FOLIO(!folio_test_pmd_mappable(folio), folio);
+
+#ifdef CONFIG_TRANSPARENT_HUGEPAGE
+	if (folio_test_anon(folio) && !folio_test_swapbacked(folio))
+		return discard_trans_pmd(vma, addr, folio);
+#endif
+
+	return false;
+}
+
 /*
  * @arg: enum ttu_flags will be passed to this argument
  */
@@ -1631,14 +1644,11 @@ static bool try_to_unmap_one(struct folio *folio, struct vm_area_struct *vma,
 	if (flags & TTU_SYNC)
 		pvmw.flags = PVMW_SYNC;
 
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	if (flags & TTU_LAZYFREE_THP)
-		if (discard_trans_pmd(vma, address, folio))
+	if (flags & TTU_SPLIT_HUGE_PMD) {
+		if (__try_to_unmap_huge_pmd(vma, address, folio))
 			return true;
-#endif
-
-	if (flags & TTU_SPLIT_HUGE_PMD)
 		split_huge_pmd_address(vma, address, false, folio);
+	}
 
 	/*
 	 * For THP, we have to assume the worse case ie pmd for invalidation.
diff --git a/mm/vmscan.c b/mm/vmscan.c
index e2686cc0c037..49bd94423961 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1277,13 +1277,6 @@ static unsigned int shrink_folio_list(struct list_head *folio_list,
 
 			if (folio_test_pmd_mappable(folio))
 				flags |= TTU_SPLIT_HUGE_PMD;
-
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-			if (folio_test_anon(folio) && !was_swapbacked &&
-			    (flags & TTU_SPLIT_HUGE_PMD))
-				flags |= TTU_LAZYFREE_THP;
-#endif
-
 			/*
 			 * Without TTU_SYNC, try_to_unmap will only begin to
 			 * hold PTL from the first present PTE within a large
-- 

Thanks,
Lance