Remove potential starvation of a full log buffer flush: only flush up to the lsn which was the largest at the time when we requested the full log buffer flush
os0sync.h, os0sync.c:
  Fix a bug in os_event on Unix: even though we signaled the event, some threads could continue waiting if the event became nonsignaled quickly again; this made group commit less efficient than it should be

into deer.(none):/home/hf/work/mysql-4.0.739

check for negative count in RPAD added

into narttu.mysql.fi:/my/mysql-4.0

into narttu.mysql.fi:/my/mysql-4.0

Remove unaligned warnings on Ia64 from client library when using --host
Fix for replication when using many file descriptors 

Fields in key_expr (mysql_ha_read) wasn't linked to tables.
Hmm. How did it work at all?

  Fix bug: group commit still did not work when we had MySQL binlogging on

into deer.(none):/home/hf/work/mysql-4.0

into mysql.com:/home/pz/mysql/mysql-4.0-root

The changes for actual loads and memory sizes are to be done after
text and format are  improved.

into narttu.mysql.fi:/my/mysql-4.0

into deer.(none):/home/hf/work/mysql-4.0.717

into deer.(none):/home/hf/work/mysql-4.0.628

into narttu.mysql.fi:/my/mysql-4.0

Tests cleanup (put drop database first in tests)

Current table is placed into read_tables set of the current join_tab->select
so get_mm_parts function thinks that current table's record is read and
tries to calculate WHERE condition with the fields of the record.
Result of these calculations is unpredictable.
Looks funny - outcome of the SELECT depends on the queries executed before.

Anyway i think we should have testcase on this part of the code.

  Remove outdated comment and the corresponding assertion in debug version code

so I commit again in a fresh tree.
Fix for bug#763 (Relay_log_space too big by 4 bytes),
plus comments and DBUG_PRINT, and we don't start replication
if --bootstrap.

(Needed for platforms with the traditional bzero prototype.)
Fixes BUG#460.

  Remove all nondeterministic tests from innodb.test; but I have to study why on 64 bits we get different estimates

into serg.mylan:/usr/home/serg/Abk/mysql-4.0

into serg.mylan:/usr/home/serg/Abk/mysql-4.0

into eagle.mysql.r18.ru:/home/vva/work/BUG_672/mysql-4.0

into sinisa.nasamreza.org:/mnt/work/mysql-4.0

CREATE / INSERT ... (SELECT ...) UNION (SELECT ...) LIMIT;

  Remove nondeterministic EXPLAIN SELECT which gives on 64-bit platforms a slightly different estimate (sometimes we should study why we get different results on 64-bit and 32-bit)

  Make execution with UNIV_SYNC_DEBUG faster

into narttu.mysql.fi:/my/mysql-4.0

  Fix a latching order violation in the previous (2003-06-15) push: could lead to a hang on the btr0sea.c semaphore