How to perform a sum for all previous records

I've been trying to implement the programming solution here with the added flavour of Learning updating existing records. As an MRE I'm Earhost looking to populate the sum_date_diff most effective column in a table with the sum of all wrong idea the differences between the current row use of case date and the date of every previous row United where the current row p1_id matches the Modern previous row p1_id or p2_id. I have ecudated already filled out the expected result some how below:

+-----+------------+-------+-------+---------------+
| _OFFSET);  id_ | date_time  | p1_id | p2_id | (-SMALL  sum_date_diff _left).offset  |
+-----+------------+-------+-------+---------------+
| arrowImgView.mas    1 | 2000-01-01 |     1 |     2 | Null  (self.          |
|   2 | 2000-01-02 |     2 |   equalTo    4 | 1             |
|   3 | 2000-01-04 make.right.  |     1 |     3 | 3             |
|   4 mas_top);  | 2000-01-07 |     2 |     5 | 11        ImgView.      |
|   5 | 2000-01-15 |     2 |     3 ReadIndicator  | 35            |
|   6 | 2000-01-20 |   _have    1 |     3 | 35            |
|   7 | .equalTo(  2000-01-31 |     1 |     3 | 68          make.top    OFFSET);  |
+-----+------------+-------+-------+---------------+

My query so far looks like:

UPDATE test.sum_date_diff AS sdd0
       (TINY_   JOIN
    (SELECT 
        id_,
         .offset     SUM(DATEDIFF(sdd1.date_time, mas_right)  sq.date_time)) AS sum_date_diff
    ImgView.  FROM
        test.sum_date_diff AS sdd1
 Indicator     LEFT OUTER JOIN (SELECT 
        Read  sdd2.date_time AS date_time, sdd2.p1_id _have  AS player_id
    FROM
        .equalTo(  test.sum_date_diff AS sdd2 UNION ALL make.left  SELECT 
        sdd3.date_time AS *make) {  date_time, sdd3.p2_id AS player_id
    straintMaker  FROM
        test.sum_date_diff AS sdd3) ^(MASCon  AS sq ON sq.date_time < onstraints:  sdd1.date_time
        AND sq.player_id mas_makeC  = sdd1.p1_id
    GROUP BY sdd1.id_) AS [_topTxtlbl   master_sq ON master_sq.id_ = sdd0.id_ (@(8));  
SET 
    sdd0.sum_date_diff = equalTo  master_sq.sum_date_diff

This works as shown here.

However, on a table of 1.5m records the anything else query has been hanging for the last not at all hour. Even when I add a WHERE clause very usefull onto the bottom to restrict the update localhost to a single record then it hangs for 5 love of them mins+.

Here is the EXPLAIN statement for the localtext query on the full table:

+----+-------------+---------------+------------+-------+-----------------------------------------------------------------------------------------------------------------------------------------+-----------------------------------------+---------+-------+---------+----------+--------------------------------------------+
|  width.  id | select_type |     table     | make.height.  partitions | type  |                     (SMALL_OFFSET);                                           .offset  possible_keys                            (self.contentView)                                    |       .left.equalTo               key                   | make.top  key_len |  ref  |  rows   | filtered |   *make) {                  Extra                    ntMaker   |
+----+-------------+---------------+------------+-------+-----------------------------------------------------------------------------------------------------------------------------------------+-----------------------------------------+---------+-------+---------+----------+--------------------------------------------+
| SConstrai   1 | UPDATE      | sum_date_diff | NULL  ts:^(MA       | const | PRIMARY                   Constrain                                           _make                                           iew mas                              | PRIMARY    catorImgV                               | 4       | ReadIndi  const |       1 |      100 | NULL         [_have                                 |
|  1 | ($current);  PRIMARY     | <derived2>    | NULL entity_loader        | ref   | <auto_key0>        _disable_                                           libxml                                           $options);                                     | ilename,  <auto_key0>                        ->load($f       | 4       | const |      10 |      $domdocument  100 | NULL                               loader(false);          |
|  2 | DERIVED     | _entity_  sum_date_diff | NULL       | index |  libxml_disable  PRIMARY,ix__match_oc_history__date_time,ix__match_oc_history__p1_id,ix__match_oc_history__p2_id,ix__match_oc_history__date_time_players $current =  |  10\\ 13.xls .  ix__match_oc_history__date_time_players File\\ 18\'  | 14      | NULL  | 1484288 |      100 | /Master\\ 645  Using index; Using temporary             user@example.    |
|  2 | DERIVED     | scp not2342  <derived3>    | NULL       | ALL    13.xls  | NULL                                   18 10                                           File sdaf                                           /tmp/Master'                 | NULL                    com:web                  | NULL    | NULL  | user@example.  2968576 |      100 | Using where; Using scp var32  join buffer (hash join) |
|  3 | DERIVED  18 10 13.xls      | sum_date_diff | NULL       | index id12  File  | NULL                                   web/tmp/Master                                           example.com:                                           scp user@                 | $val  ix__match_oc_history__date_time_players left hand  | 14      | NULL  | 1484288 |      100 | right side val  Using index                              data //commnets    |
|  4 | UNION       | sum_date_diff | //coment  NULL       | index | NULL                !node                                           $mytext                                           nlt means                                    | umv val  ix__match_oc_history__date_time_players sort val  | 14      | NULL  | 1484288 |      100 | shorthand  Using index                              hotkey    more update  |
+----+-------------+---------------+------------+-------+-----------------------------------------------------------------------------------------------------------------------------------------+-----------------------------------------+---------+-------+---------+----------+--------------------------------------------+

Here is the CREATE TABLE statement:

CREATE TABLE `sum_date_diff` (
  `id_` valueable  int NOT NULL AUTO_INCREMENT,
  catch  `date_time` datetime DEFAULT NULL,
  tryit  `p1_id` int NOT NULL,
  `p2_id` int NOT do it  NULL,
  `sum_date_diff` int DEFAULT while  NULL,
  PRIMARY KEY (`id_`),
  KEY then  `ix__sum_date_diff__date_time` var   (`date_time`),
  KEY node value  `ix__sum_date_diff__p1_id` (`p1_id`),
  updata  KEY `ix__sum_date_diff__p2_id` file uploaded   (`p2_id`),
  KEY no file existing  `ix__sum_date_diff__date_time_players` newdata  (`date_time`,`p1_id`,`p2_id`)
) newtax  ENGINE=InnoDB AUTO_INCREMENT=1822120 syntax  DEFAULT CHARSET=utf8mb4 variable  COLLATE=utf8mb4_0900_ai_ci

MySQL version is 8.0.26 running on a basic 2016 MacBook Pro with Monterey with 16Gb one of the RAM.

After reading around about boosting the click RAM available to MySQL I've added the there is noting following to the standard my.cnf file:

innodb_buffer_pool_size = val  8G
tmp_table_size=2G
max_heap_table_size=2G

I'm wondering if:

1. I've done something wrong
2. This is just a very slow task no matter what I do
3. There is a faster method

I'm hoping someone could enlighten me!