This is first cset for WL#1394 "Optimize index merge when all involved index ranges include only values with equal keys"
The main idea is to exploit the fact that key scans for "key=const" return ordered sequences of rowids.

(Meaning "... with SELECT v", where v is a local variable.)

into mysql.com:/home/pem/work/mysql-5.0

(This might not be enough, could be more statements that must be detected...)

Found a few more places in the parser (insert, replace and update) where
local stored procedure variables should not be recognized.

(And some minor correction of cursor open)

Initialize and test properly when cleaning up, to avoid crash in some error cases.

Parse column names (and not variables) only in UPDATE ... SET ...

Portability fixes

into mysql.com:/my/mysql-4.1

Fixed bug in ORDER BY on a small column (Bug #2147)
Fixed error from pthread_mutex_destroy() when one had wrong errmsg file

Some code cleanup

into mysql.com:/my/mysql-5.0

(Fixed project files, compiler warnings etc..)

into mysql.com:/home/pem/work/mysql-5.0-merge

  Define macro ULINTPF which can be used in printf statements as the print format of ulint both on Win64 and other platforms

  Check at compilation time on Unix that InnoDB ulint is the same size as void*, and also check it at runtime

A after merge fix for last merge

into mysql.com:/home/mysql_src/mysql-5.0-new-binlog-format

For previous commit I had run only rpl* tests, here the other ones had a 
few surprises. Latest status:
- all tests pass
- all replication tests pass with Valgrind
This is the final-final commit & push.
Doc remains.

into mysql.com:/home/kostja/mysql/mysql-4.1-prep_st_hack

into mysql.com:/home/mysql_src/mysql-5.0-new-binlog-format

into mysql.com:/dbdata/psergey/mysql-5.0-imerge-unique

   * A more dynamic binlog format which allows small changes (1064)
   * Log session variables in Query_log_event (1063)
It contains a few bugfixes (which I made when running the testsuite).
I carefully updated the results of the testsuite (i.e. I checked for every one,
if the difference between .reject and .result could be explained).
Apparently mysql-test-run --manager is broken in 4.1 and 5.0 currently,
so I could neither run the few tests which require --manager, nor check
that they pass nor modify their .result. But for builds, we don't run
with --manager.
Apart from --manager, the full testsuite passes, with Valgrind too (no errors).
I'm going to push in the next minutes. Remains: update the manual.
Note: by chance I saw that (in 4.1, in 5.0) rpl_get_lock fails when run alone;
this is normal at it makes assumptions on thread ids. I will fix this one day
in 4.1.

into mysql.com:/my/mysql-4.1