it's crucial to disable Nagle algorithm on client for no-reply commands
(like mysql_stmt_free) to always work fast.
Nagle algorithm instructs the sender to buffer (store) data if any 
unacknowledged data is outstanding and the size of to-send data is less than
the network segment. It was exactly the case with COM_STMT_CLOSE after 
COM_STMT_PREPARE, so the client was waiting for Nagle timer to expire or
 for ACK from the server, while the server was holding up ACK because of 
delayed acknowledgement algorithm.
The tricky part is that we have been already disabling Nagle 
algorithm (by setting TCP_NODELAY)
almost everywhere except Windows (and maybe Netware).
It is the reason why the bug was repeatable only with Windows client.

into serg.mylan:/usr/home/serg/Abk/mysql-4.1

  Bug#7145 Added cast to doublestore() and ARM

into mysql.com:/Users/kent/mysql/bk/mysql-4.1-perl

  Let MYSQL_CLIENT_TEST be set with command line from mysql-test-run script
mysql-test-run.sh:
  Renamed client_test to mysql_client_test
  Support --embedded-server test on installed server
mysql_client_test.test:
  Renamed client_test to mysql_client_test
  Use env var CLIENT_TEST with full name, it differs when run with --embedded-server
  Rename: mysql-test/t/client_test.test -> mysql-test/t/mysql_client_test.test

into serg.mylan:/usr/home/serg/Abk/mysql-4.1

  backported access to ndb_mgm from 5.0

into mysql.com:/Users/kent/mysql/bk/mysql-4.1-perl

into rurik.mysql.com:/home/igor/dev/mysql-4.1-0

  Added a test case for bug #7945.
sql_select.cc:
  Fixed bug #7945. If DISTINCT is used only with constants
  in a query with GROUP BY, we can apply an optimization
  that set LIMIT to 1 only in the case when there is 
  no SQL_CALC_FOUND_ROWS.

  Fix a race condition that could cause the assertion space->n_pending_flushes == 0 to fail in fil0fil.c, in fil_space_free(), in DROP TABLE or in ALTER TABLE

into poseidon.ndb.mysql.com:/home/tomas/mysql-4.1

corrected possible unalignment in NdbRecAttr access methods

into mysql.com:/home/timka/mysql/src/4.1-virgin

  Fix bug #8021 :Windows error number 87 with multiple tablespaces after mysqld restart; still has to be tested on Windows that this now works

  Fix bug #7831: ALTER TABLE ... ADD CONSTRAINT PRIMARY KEY ... complained about bad foreign key definition

into mysql.com:/home/timka/mysql/src/4.1-dbg

into hundin.mysql.fi:/home/heikki/mysql-4.1

SHOW KEYS FROM t1 now displays number of characters in Sub_part,
not number of bytes, to be compatible with SHOW CREATE TABLE.

  Make sure to ship error code in system error signal

into hundin.mysql.fi:/home/heikki/mysql-4.1

- renamed the tests that use the embedded server (client_test -> 
  mysql_client_test_embedded, mysqltest -> mysql_test_embedded
  and changed some Makefiles and scripts so they are installed in $bindir
  (required to be able to run the test suite against the embedded server)

into mysql.com:/Users/kent/mysql/bk/mysql-4.1-perl

into mysql.com:/Users/kent/mysql/bk/mysql-4.1-perl

  Bug#7834 Illegal mix of collations in IN operator
  IN was the first function supporting 
  character set convertion.
  agg_arg_charsets() was written afterwards,
  which is more flexible.
  Now IN just reuses this function.

into mysql.com:/home/jonas/src/mysql-4.1

  release connections when last op was simple read

  environment (using "build" on SUSE Linux)

into mysql.com:/space/my/mysql-4.1

into mysql.com:/home/jonas/src/mysql-4.1

  Minor fix after character set conversion.

into mysql.com:/home/jonas/src/mysql-4.1

  Add SJIS version of Japanese error messages.
  Windows version should include this new file,
  instead of the EUC-JP version.