"a long semaphore wait" After Upgrading to 10.1.41

Ever since we upgraded to 10.1.41, we have been having non-stop issues in several of our large servers where the logs start filling up with semaphore errors and working with any InnoDB table seems to lock up.

Doing research, I found people talking about disabling the `innodb_adaptive_hash_index` by setting it to 0 to fix the issue, and I have tried this without success. With it disabled, the issue still crops back up.

Doing some more research, I found issue #16467 (https://jira.mariadb.org/browse/MDEV-16467) which was in the update log for MariaDB 10.1.41, so I am wondering if that is related to our issues.

Any help will be awesome. I have attached the log output we see.