Nested subqueries. Once any SQL has nested subqueries, the speed will be greatly reduced. At work, unless I absolutely must use subqueries, I will never write nested subqueries. I see you (select * from b) There is no limit at the end to limit the number of lines? What if there are 10 million data?

You can try join insert into the temporary table first and then use the temporary table to join update

You can update while loop without considering efficiency
http://stackoverflow.com/questions/11430362/update-column-from-another-table-in-large-mysql-db-7-million-rows

It must be the sub-nested query that causes the speed to be too slow. If you are interested, you can send the data set so that you can help adjust the SQL. The IO overhead of retrieving all table data at once is too high. You can try writing like this:
update a set a.country = b.country, a.city = b.city from b where a.iplong >=b.ip1 and a.iplong < b.ip2;

In addition, I can’t see what data type your IP field is stored in. For this comparison, using int will get better performance.

The execution plan does not appear in the index, and iplong cannot enter the index. Can you post the table creation and index statements? For UPDATE with a large amount of data, it will be very slow because the rollback segment is relatively large. You can pass b table ip1, ip2 plus index
create table a_1 as select a.id,a.ipdizhi,a.iplong,(select b.country from b where a.iplong >=b.ip1 and a.iplong < b.ip2) as country,(select b .city from b where a.iplong >=b.ip1 and a.iplong < b.ip2) as city from a;
To implement, then replace the two A tables.