这有点像“在黑暗中摸索”，因为您没有在问题中包含当前查询的表定义或EXPLAIN输出。
使用blockings.user_id和blockings.blocked_id上的单列索引，您应该在当前查询的EXPLAIN输出中看到blockings的index_merge。
使用PK (user_id, blocked_id)和(blocked_id, user_id)上的索引，在NOT EXISTS中使用UNION可能会快得多：

SELECT *
FROM users u
WHERE u.id != 1
  AND u.activated = 1
  AND NOT EXISTS (
        SELECT 1
        FROM blockings b
        WHERE b.user_id = u.id AND b.blocked_id = 1
        UNION ALL
        SELECT 1
        FROM blockings b
        WHERE b.blocked_id = u.id AND b.user_id = 1
    )
ORDER BY u.id DESC
LIMIT 10 OFFSET 0;

正如MatBailie所建议的，值得尝试将UNION ALL分成两个独立的NOT EXISTS：

SELECT *
FROM users u
WHERE u.id != 1
AND u.activated = 1
AND NOT EXISTS (
    SELECT 1
    FROM blockings b
    WHERE b.user_id = u.id AND b.blocked_id = 1
)
AND NOT EXISTS (
    SELECT 1
    FROM blockings b
    WHERE b.blocked_id = u.id AND b.user_id = 1
)
ORDER BY u.id DESC
LIMIT 10 OFFSET 0;

下面是一个可能的替代查询，您可能需要在where子句中使用的列上添加索引。

SELECT
    users.id,  /* all the necessary columns goes here */
    users.name,
    users. Email
FROM
    users
    LEFT JOIN blockings ON (
        (blockings.user_id = users.id AND blockings.blocked_id = 1)
        OR (blockings.blocked_id = users.id AND blockings.user_id = 1)
    )
WHERE
    users.id != 1
    AND users.activated = 1
    AND blockings.id IS NULL
ORDER BY
    users.id DESC
LIMIT 10 OFFSET 0;