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;
Date:   Tue, 30 May 2023 21:02:51 +0000
Mime-Version: 1.0
Message-ID: <20230530210251.493194-1-yosryahmed@google.com>
Subject: [PATCH] mm: zswap: support exclusive loads
From:   Yosry Ahmed <yosryahmed@google.com>
To:     Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Seth Jennings <sjenning@redhat.com>,
        Dan Streetman <ddstreet@ieee.org>,
        Vitaly Wool <vitaly.wool@konsulko.com>
Cc:     Johannes Weiner <hannes@cmpxchg.org>,
        Nhat Pham <nphamcs@gmail.com>,
        Domenico Cerasuolo <cerasuolodomenico@gmail.com>,
        Yu Zhao <yuzhao@google.com>, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, Yosry Ahmed <yosryahmed@google.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Commit 71024cb4a0bf ("frontswap: remove frontswap_tmem_exclusive_gets")
removed support for exclusive loads from frontswap as it was not used.

Bring back exclusive loads support to frontswap by adding an
exclusive_loads argument to frontswap_ops. Add support for exclusive
loads to zswap behind CONFIG_ZSWAP_EXCLUSIVE_LOADS.

Refactor zswap entry invalidation in zswap_frontswap_invalidate_page()
into zswap_invalidate_entry() to reuse it in zswap_frontswap_load().

With exclusive loads, we avoid having two copies of the same page in
memory (compressed & uncompressed) after faulting it in from zswap. On
the other hand, if the page is to be reclaimed again without being
dirtied, it will be re-compressed. Compression is not usually slow, and
a page that was just faulted in is less likely to be reclaimed again
soon.

Suggested-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
---
 include/linux/frontswap.h |  1 +
 mm/Kconfig                | 13 +++++++++++++
 mm/frontswap.c            |  7 ++++++-
 mm/zswap.c                | 23 +++++++++++++++--------
 4 files changed, 35 insertions(+), 9 deletions(-)

diff --git a/include/linux/frontswap.h b/include/linux/frontswap.h
index a631bac12220..289561e12cad 100644
--- a/include/linux/frontswap.h
+++ b/include/linux/frontswap.h
@@ -13,6 +13,7 @@ struct frontswap_ops {
 	int (*load)(unsigned, pgoff_t, struct page *); /* load a page */
 	void (*invalidate_page)(unsigned, pgoff_t); /* page no longer needed */
 	void (*invalidate_area)(unsigned); /* swap type just swapoff'ed */
+	bool exclusive_loads; /* pages are invalidated after being loaded */
 };
 
 int frontswap_register_ops(const struct frontswap_ops *ops);
diff --git a/mm/Kconfig b/mm/Kconfig
index 7672a22647b4..92c30879bf67 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -46,6 +46,19 @@ config ZSWAP_DEFAULT_ON
 	  The selection made here can be overridden by using the kernel
 	  command line 'zswap.enabled=' option.
 
+config ZSWAP_EXCLUSIVE_LOADS
+	bool "Invalidate zswap entries when pages are loaded"
+	depends on ZSWAP
+	help
+	  If selected, when a page is loaded from zswap, the zswap entry is
+	  invalidated at once, as opposed to leaving it in zswap until the
+	  swap entry is freed.
+
+	  This avoids having two copies of the same page in memory
+	  (compressed and uncompressed) after faulting in a page from zswap.
+	  The cost is that if the page was never dirtied and needs to be
+	  swapped out again, it will be re-compressed.
+
 choice
 	prompt "Default compressor"
 	depends on ZSWAP
diff --git a/mm/frontswap.c b/mm/frontswap.c
index 279e55b4ed87..e5d6825110f4 100644
--- a/mm/frontswap.c
+++ b/mm/frontswap.c
@@ -216,8 +216,13 @@ int __frontswap_load(struct page *page)
 
 	/* Try loading from each implementation, until one succeeds. */
 	ret = frontswap_ops->load(type, offset, page);
-	if (ret == 0)
+	if (ret == 0) {
 		inc_frontswap_loads();
+		if (frontswap_ops->exclusive_loads) {
+			SetPageDirty(page);
+			__frontswap_clear(sis, offset);
+		}
+	}
 	return ret;
 }
 
diff --git a/mm/zswap.c b/mm/zswap.c
index 59da2a415fbb..fba80330afd1 100644
--- a/mm/zswap.c
+++ b/mm/zswap.c
@@ -1329,6 +1329,16 @@ static int zswap_frontswap_store(unsigned type, pgoff_t offset,
 	goto reject;
 }
 
+static void zswap_invalidate_entry(struct zswap_tree *tree,
+				   struct zswap_entry *entry)
+{
+	/* remove from rbtree */
+	zswap_rb_erase(&tree->rbroot, entry);
+
+	/* drop the initial reference from entry creation */
+	zswap_entry_put(tree, entry);
+}
+
 /*
  * returns 0 if the page was successfully decompressed
  * return -1 on entry not found or error
@@ -1403,6 +1413,8 @@ static int zswap_frontswap_load(unsigned type, pgoff_t offset,
 		count_objcg_event(entry->objcg, ZSWPIN);
 freeentry:
 	spin_lock(&tree->lock);
+	if (!ret && IS_ENABLED(CONFIG_ZSWAP_EXCLUSIVE_LOADS))
+		zswap_invalidate_entry(tree, entry);
 	zswap_entry_put(tree, entry);
 	spin_unlock(&tree->lock);
 
@@ -1423,13 +1435,7 @@ static void zswap_frontswap_invalidate_page(unsigned type, pgoff_t offset)
 		spin_unlock(&tree->lock);
 		return;
 	}
-
-	/* remove from rbtree */
-	zswap_rb_erase(&tree->rbroot, entry);
-
-	/* drop the initial reference from entry creation */
-	zswap_entry_put(tree, entry);
-
+	zswap_invalidate_entry(tree, entry);
 	spin_unlock(&tree->lock);
 }
 
@@ -1472,7 +1478,8 @@ static const struct frontswap_ops zswap_frontswap_ops = {
 	.load = zswap_frontswap_load,
 	.invalidate_page = zswap_frontswap_invalidate_page,
 	.invalidate_area = zswap_frontswap_invalidate_area,
-	.init = zswap_frontswap_init
+	.init = zswap_frontswap_init,
+	.exclusive_loads = IS_ENABLED(CONFIG_ZSWAP_EXCLUSIVE_LOADS),
 };
 
 /*********************************
-- 
2.41.0.rc0.172.g3f132b7071-goog