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:   Nhat Pham <nphamcs@gmail.com>
To:     akpm@linux-foundation.org
Cc:     hannes@cmpxchg.org, linux-mm@kvack.org,
        linux-kernel@vger.kernel.org, minchan@kernel.org,
        ngupta@vflare.org, senozhatsky@chromium.org, sjenning@redhat.com,
        ddstreet@ieee.org, vitaly.wool@konsulko.com, kernel-team@meta.com
Subject: [PATCH] zsmalloc: move LRU update from zs_map_object() to zs_malloc()
Date:   Fri,  5 May 2023 11:50:54 -0700
Message-Id: <20230505185054.2417128-1-nphamcs@gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Under memory pressure, we sometimes observe the following crash:

[ 5694.832838] ------------[ cut here ]------------
[ 5694.842093] list_del corruption, ffff888014b6a448->next is LIST_POISON1 (dead000000000100)
[ 5694.858677] WARNING: CPU: 33 PID: 418824 at lib/list_debug.c:47 __list_del_entry_valid+0x42/0x80
[ 5694.961820] CPU: 33 PID: 418824 Comm: fuse_counters.s Kdump: loaded Tainted: G S                5.19.0-0_fbk3_rc3_hoangnhatpzsdynshrv41_10870_g85a9558a25de #1
[ 5694.990194] Hardware name: Wiwynn Twin Lakes MP/Twin Lakes Passive MP, BIOS YMM16 05/24/2021
[ 5695.007072] RIP: 0010:__list_del_entry_valid+0x42/0x80
[ 5695.017351] Code: 08 48 83 c2 22 48 39 d0 74 24 48 8b 10 48 39 f2 75 2c 48 8b 51 08 b0 01 48 39 f2 75 34 c3 48 c7 c7 55 d7 78 82 e8 4e 45 3b 00 <0f> 0b eb 31 48 c7 c7 27 a8 70 82 e8 3e 45 3b 00 0f 0b eb 21 48 c7
[ 5695.054919] RSP: 0018:ffffc90027aef4f0 EFLAGS: 00010246
[ 5695.065366] RAX: 41fe484987275300 RBX: ffff888008988180 RCX: 0000000000000000
[ 5695.079636] RDX: ffff88886006c280 RSI: ffff888860060480 RDI: ffff888860060480
[ 5695.093904] RBP: 0000000000000002 R08: 0000000000000000 R09: ffffc90027aef370
[ 5695.108175] R10: 0000000000000000 R11: ffffffff82fdf1c0 R12: 0000000010000002
[ 5695.122447] R13: ffff888014b6a448 R14: ffff888014b6a420 R15: 00000000138dc240
[ 5695.136717] FS:  00007f23a7d3f740(0000) GS:ffff888860040000(0000) knlGS:0000000000000000
[ 5695.152899] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 5695.164388] CR2: 0000560ceaab6ac0 CR3: 000000001c06c001 CR4: 00000000007706e0
[ 5695.178659] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 5695.192927] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 5695.207197] PKRU: 55555554
[ 5695.212602] Call Trace:
[ 5695.217486]  <TASK>
[ 5695.221674]  zs_map_object+0x91/0x270
[ 5695.229000]  zswap_frontswap_store+0x33d/0x870
[ 5695.237885]  ? do_raw_spin_lock+0x5d/0xa0
[ 5695.245899]  __frontswap_store+0x51/0xb0
[ 5695.253742]  swap_writepage+0x3c/0x60
[ 5695.261063]  shrink_page_list+0x738/0x1230
[ 5695.269255]  shrink_lruvec+0x5ec/0xcd0
[ 5695.276749]  ? shrink_slab+0x187/0x5f0
[ 5695.284240]  ? mem_cgroup_iter+0x6e/0x120
[ 5695.292255]  shrink_node+0x293/0x7b0
[ 5695.299402]  do_try_to_free_pages+0xea/0x550
[ 5695.307940]  try_to_free_pages+0x19a/0x490
[ 5695.316126]  __folio_alloc+0x19ff/0x3e40
[ 5695.323971]  ? __filemap_get_folio+0x8a/0x4e0
[ 5695.332681]  ? walk_component+0x2a8/0xb50
[ 5695.340697]  ? generic_permission+0xda/0x2a0
[ 5695.349231]  ? __filemap_get_folio+0x8a/0x4e0
[ 5695.357940]  ? walk_component+0x2a8/0xb50
[ 5695.365955]  vma_alloc_folio+0x10e/0x570
[ 5695.373796]  ? walk_component+0x52/0xb50
[ 5695.381634]  wp_page_copy+0x38c/0xc10
[ 5695.388953]  ? filename_lookup+0x378/0xbc0
[ 5695.397140]  handle_mm_fault+0x87f/0x1800
[ 5695.405157]  do_user_addr_fault+0x1bd/0x570
[ 5695.413520]  exc_page_fault+0x5d/0x110
[ 5695.421017]  asm_exc_page_fault+0x22/0x30

After some investigation, I have found the following issue: unlike other
zswap backends, zsmalloc performs the LRU list update at the object
mapping time, rather than when the slot for the object is allocated.
This deviation was discussed and agreed upon during the review process
of the zsmalloc writeback patch series:

https://lore.kernel.org/lkml/Y3flcAXNxxrvy3ZH@cmpxchg.org/

Unfortunately, this introduces a subtle bug that occurs when there is a
concurrent store and reclaim, which interleave as follows:

zswap_frontswap_store()            shrink_worker()
  zs_malloc()                        zs_zpool_shrink()
    spin_lock(&pool->lock)             zs_reclaim_page()
    zspage = find_get_zspage()
    spin_unlock(&pool->lock)
                                         spin_lock(&pool->lock)
                                         zspage = list_first_entry(&pool->lru)
                                         list_del(&zspage->lru)
                                           zspage->lru.next = LIST_POISON1
                                           zspage->lru.prev = LIST_POISON2
                                         spin_unlock(&pool->lock)
  zs_map_object()
    spin_lock(&pool->lock)
    if (!list_empty(&zspage->lru))
      list_del(&zspage->lru)
        CHECK_DATA_CORRUPTION(next == LIST_POISON1) /* BOOM */

With the current upstream code, this issue rarely happens. zswap only
triggers writeback when the pool is already full, at which point all
further store attempts are short-circuited. This creates an implicit
pseudo-serialization between reclaim and store. I am working on a new
zswap shrinking mechanism, which makes interleaving reclaim and store
more likely, exposing this bug.

zbud and z3fold do not have this problem, because they perform the LRU
list update in the alloc function, while still holding the pool's lock.
This patch fixes the aforementioned bug by moving the LRU update back to
zs_malloc(), analogous to zbud and z3fold.

Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Nhat Pham <nphamcs@gmail.com>
---
 mm/zsmalloc.c | 36 +++++++++---------------------------
 1 file changed, 9 insertions(+), 27 deletions(-)

diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
index 44ddaf5d601e..02f7f414aade 100644
--- a/mm/zsmalloc.c
+++ b/mm/zsmalloc.c
@@ -1331,31 +1331,6 @@ void *zs_map_object(struct zs_pool *pool, unsigned long handle,
 	obj_to_location(obj, &page, &obj_idx);
 	zspage = get_zspage(page);
 
-#ifdef CONFIG_ZPOOL
-	/*
-	 * Move the zspage to front of pool's LRU.
-	 *
-	 * Note that this is swap-specific, so by definition there are no ongoing
-	 * accesses to the memory while the page is swapped out that would make
-	 * it "hot". A new entry is hot, then ages to the tail until it gets either
-	 * written back or swaps back in.
-	 *
-	 * Furthermore, map is also called during writeback. We must not put an
-	 * isolated page on the LRU mid-reclaim.
-	 *
-	 * As a result, only update the LRU when the page is mapped for write
-	 * when it's first instantiated.
-	 *
-	 * This is a deviation from the other backends, which perform this update
-	 * in the allocation function (zbud_alloc, z3fold_alloc).
-	 */
-	if (mm == ZS_MM_WO) {
-		if (!list_empty(&zspage->lru))
-			list_del(&zspage->lru);
-		list_add(&zspage->lru, &pool->lru);
-	}
-#endif
-
 	/*
 	 * migration cannot move any zpages in this zspage. Here, pool->lock
 	 * is too heavy since callers would take some time until they calls
@@ -1525,9 +1500,8 @@ unsigned long zs_malloc(struct zs_pool *pool, size_t size, gfp_t gfp)
 		fix_fullness_group(class, zspage);
 		record_obj(handle, obj);
 		class_stat_inc(class, ZS_OBJS_INUSE, 1);
-		spin_unlock(&pool->lock);
 
-		return handle;
+		goto out;
 	}
 
 	spin_unlock(&pool->lock);
@@ -1550,6 +1524,14 @@ unsigned long zs_malloc(struct zs_pool *pool, size_t size, gfp_t gfp)
 
 	/* We completely set up zspage so mark them as movable */
 	SetZsPageMovable(pool, zspage);
+out:
+#ifdef CONFIG_ZPOOL
+	/* Add/move zspage to beginning of LRU */
+	if (!list_empty(&zspage->lru))
+		list_del(&zspage->lru);
+	list_add(&zspage->lru, &pool->lru);
+#endif
+
 	spin_unlock(&pool->lock);
 
 	return handle;
-- 
2.34.1