Import wiredtiger: f827562f2b486e000665c97ea81674012d5a765b from branch mongodb-5.0

ref: 6614fa3dca..f827562f2b
for: 4.9.0

WT-6798       Utilize Arm LSE atomics and the correct strength barriers
WT-6808       Documentation: add top level architecture picture
WT-6812       Fix "out-of-order fixup" potentially corrupting historical values

1 files changed, 78 insertions, 1 deletions

diff --git a/src/third_party/wiredtiger/test/suite/test_timestamp20.py b/src/third_party/wiredtiger/test/suite/test_timestamp20.py
index 69aec301a47..63a2503d915 100644
--- a/src/third_party/wiredtiger/test/suite/test_timestamp20.py
+++ b/src/third_party/wiredtiger/test/suite/test_timestamp20.py
@@ -37,7 +37,7 @@ class test_timestamp20(wttest.WiredTigerTestCase):
     conn_config = 'cache_size=50MB'
     session_config = 'isolation=snapshot'
 
-    def test_timestamp20(self):
+    def test_timestamp20_standard(self):
         uri = 'table:test_timestamp20'
         self.session.create(uri, 'key_format=S,value_format=S')
         self.conn.set_timestamp('oldest_timestamp=' + timestamp_str(1))
@@ -86,3 +86,80 @@ class test_timestamp20(wttest.WiredTigerTestCase):
         for i in range(1, 10000):
             self.assertEqual(old_reader_cursor[str(i)], value3)
         old_reader_session.rollback_transaction()
+
+    # In this test we're using modifies since they are more sensitive to corruptions.
+    #
+    # Corruptions to string types may go undetected since non-ASCII characters won't be included in
+    # the conversion to a Python string.
+    def test_timestamp20_modify(self):
+        uri = 'table:test_timestamp20'
+        self.session.create(uri, 'key_format=S,value_format=S')
+        self.conn.set_timestamp('oldest_timestamp=' + timestamp_str(1))
+        cursor = self.session.open_cursor(uri)
+
+        value1 = 'a' * 500
+        value2 = 'b' * 500
+        value3 = 'c' * 500
+
+        # Apply the base value.
+        for i in range(1, 10000):
+            self.session.begin_transaction()
+            cursor[str(i)] = value1
+            self.session.commit_transaction('commit_timestamp=' + timestamp_str(10))
+
+        # Now apply a series of modifies.
+        for i in range(1, 10000):
+            self.session.begin_transaction()
+            cursor.set_key(str(i))
+            self.assertEqual(cursor.modify([wiredtiger.Modify('B', 100, 1)]), 0)
+            self.session.commit_transaction('commit_timestamp=' + timestamp_str(20))
+
+        for i in range(1, 10000):
+            self.session.begin_transaction()
+            cursor.set_key(str(i))
+            self.assertEqual(cursor.modify([wiredtiger.Modify('C', 200, 1)]), 0)
+            self.session.commit_transaction('commit_timestamp=' + timestamp_str(30))
+
+        # Open an old reader at this point.
+        #
+        # I'm trying to read the middle modify because I specifically don't want to read one that
+        # has been squashed into a full update.
+        old_reader_session = self.conn.open_session()
+        old_reader_cursor = old_reader_session.open_cursor(uri)
+        old_reader_session.begin_transaction('read_timestamp=' + timestamp_str(30))
+
+        # Now apply the last modify.
+        # This will be the end of the chain of modifies.
+        for i in range(1, 10000):
+            self.session.begin_transaction()
+            cursor.set_key(str(i))
+            self.assertEqual(cursor.modify([wiredtiger.Modify('D', 300, 1)]), 0)
+            self.session.commit_transaction('commit_timestamp=' + timestamp_str(40))
+
+        # Now put two updates out of order. 5 will go to the history store and will trigger a
+        # correction to the existing contents.
+        for i in range(1, 10000):
+            self.session.begin_transaction()
+            cursor[str(i)] = value2
+            self.session.commit_transaction('commit_timestamp=' + timestamp_str(5))
+            self.session.begin_transaction()
+            cursor[str(i)] = value3
+            self.session.commit_transaction('commit_timestamp=' + timestamp_str(50))
+
+        # Open up a new transaction and read at 30.
+        # We shouldn't be able to see past 5 due to txnid visibility.
+        self.session.begin_transaction('read_timestamp=' + timestamp_str(30))
+        for i in range(1, 10000):
+            self.assertEqual(cursor[str(i)], value2)
+        self.session.rollback_transaction()
+
+        # Put together expected value.
+        expected = list(value1)
+        expected[100] = 'B'
+        expected[200] = 'C'
+        expected = str().join(expected)
+
+        # On the other hand, this older transaction SHOULD be able to read past the 5.
+        for i in range(1, 10000):
+            self.assertEqual(old_reader_cursor[str(i)], expected)
+        old_reader_session.rollback_transaction()