Sun Studio

Those changes are contributed by Sun/MySQL. Two sets of changes
in this patch when Sun Studio is used:
- Explicit inlining of functions
- Prefetch Support

This patch has been tested by Sunny with the plugin statically
built in. Since we've never built the plugin as a dynamically
loaded module on Solaris, it is a separate task to change
plug.in.

rb://142
Approved by: Heikki

Both srv_read_ahead_factor and srv_io_capacity should
be defined as ulong.

Approved by: Sunny

Non-functional change:
put files in noinst_HEADERS and libinnobase_a_SOURCES one per line and sort
alphabetically, so it is easier to find if a file is there or not and
also diffs show exactly the added or removed file instead of surrounding
lines too.

Add include/ut0auxconf.h to noinst_HEADERS

Add @ZLIB_INCLUDES@ so that the InnoDB Plugin picks up the same zlib.h
header file that is eventually used by mysqld.

Remove an extra "with".

This change was suggested by MySQL.

c5419 because it is merge from branches/zip into branches/5.1
c5466 because the source code has been adjusted to match the MySQL
  behavior and the innodb-autoinc test does not fail in branches/zip,
  if c5466 is merged, then innodb-autoinc starts failing, Sunny suggested
  not to merge c5466.
and resolving conflicts in c5410, c5440, c5488:

  ------------------------------------------------------------------------
  r5410 | marko | 2009-06-24 22:26:34 +0300 (Wed, 24 Jun 2009) | 2 lines
  Changed paths:
     M /branches/5.1/include/trx0sys.ic
     M /branches/5.1/trx/trx0purge.c
     M /branches/5.1/trx/trx0sys.c
     M /branches/5.1/trx/trx0undo.c
  
  branches/5.1: Add missing #include "mtr0log.h" to avoid warnings
  when compiling with -DUNIV_MUST_NOT_INLINE.
  ------------------------------------------------------------------------
  r5419 | marko | 2009-06-25 16:11:57 +0300 (Thu, 25 Jun 2009) | 18 lines
  Changed paths:
     M /branches/5.1/handler/ha_innodb.cc
     M /branches/5.1/mysql-test/innodb_bug42101-nonzero.result
     M /branches/5.1/mysql-test/innodb_bug42101-nonzero.test
     M /branches/5.1/mysql-test/innodb_bug42101.result
     M /branches/5.1/mysql-test/innodb_bug42101.test
  
  branches/5.1: Merge r5418 from branches/zip:
  
    ------------------------------------------------------------------------
    r5418 | marko | 2009-06-25 15:55:52 +0300 (Thu, 25 Jun 2009) | 5 lines
    Changed paths:
       M /branches/zip/ChangeLog
       M /branches/zip/handler/ha_innodb.cc
       M /branches/zip/mysql-test/innodb_bug42101-nonzero.result
       M /branches/zip/mysql-test/innodb_bug42101-nonzero.test
       M /branches/zip/mysql-test/innodb_bug42101.result
       M /branches/zip/mysql-test/innodb_bug42101.test
    
    branches/zip: Fix a race condition caused by
    SET GLOBAL innodb_commit_concurrency=DEFAULT. (Bug #45749)
    When innodb_commit_concurrency is initially set nonzero,
    DEFAULT would change it back to 0, triggering Bug #42101.
    rb://139 approved by Heikki Tuuri.
    ------------------------------------------------------------------------
  ------------------------------------------------------------------------
  r5440 | vasil | 2009-06-30 13:04:29 +0300 (Tue, 30 Jun 2009) | 8 lines
  Changed paths:
     M /branches/5.1/fil/fil0fil.c
  
  branches/5.1:
  
  Fix Bug#45814 URL reference in InnoDB server errors needs adjusting to match documentation
  
  by changing the URL from
  http://dev.mysql.com/doc/refman/5.1/en/innodb-troubleshooting.html to
  http://dev.mysql.com/doc/refman/5.1/en/innodb-troubleshooting-datadict.html
  
  ------------------------------------------------------------------------
  r5466 | vasil | 2009-07-02 10:46:45 +0300 (Thu, 02 Jul 2009) | 6 lines
  Changed paths:
     M /branches/5.1/mysql-test/innodb-autoinc.result
     M /branches/5.1/mysql-test/innodb-autoinc.test
  
  branches/5.1:
  
  Adjust the failing innodb-autoinc test to conform to the latest behavior
  of the MySQL code. The idea and the comment in innodb-autoinc.test come
  from Sunny.
  
  ------------------------------------------------------------------------
  r5488 | vasil | 2009-07-09 19:16:44 +0300 (Thu, 09 Jul 2009) | 13 lines
  Changed paths:
     M /branches/5.1/handler/ha_innodb.cc
     A /branches/5.1/mysql-test/innodb_bug21704.result
     A /branches/5.1/mysql-test/innodb_bug21704.test
  
  branches/5.1:
  
  Fix Bug#21704 Renaming column does not update FK definition
  
  by checking whether a column that participates in a FK definition is being
  renamed and denying the ALTER in this case.
  
  The patch was originally developed by Davi Arnaut <Davi.Arnaut@Sun.COM>:
  http://lists.mysql.com/commits/77714
  and was later adjusted to conform to InnoDB coding style by me (Vasil),
  I also added some more comments and moved the bug specific mysql-test to
  a separate file to make it more manageable and flexible.
  
  ------------------------------------------------------------------------

This patch is from Sergey Vojtovich of Sun Microsystems,
to fix run-time symbols clash on Solaris with older C++
compiler:
- when finding out a way to hide symbols, make decision basing
  on compiler, not operating system.
- Sun Studio supports __hidden declaration specifier for this
  purpose.

Fixed warnings on windows where ulint != ib_uint64_t

Revert the flush neighbors patch as it shows regression in
the benchmarks run by Michael.

Restore the original value of innodb_sync_spin_loops at the end, previously
the test assumed that setting it to 20 will do this, but now the default is
30 and MTR's internal check failed.

instruction patch, contributed by Sun Microsystems.

Add ChangeLog entry for 5489.

Change the defaults for
innodb_sync_spin_loops: 20 -> 30
innodb_spin_wait_delay: 5 -> 6

This change was proposed by Sun/MySQL based on their performance testing,
see https://svn.innodb.com/innobase/Release_tasks_for_InnoDB_Plugin_V1.0.4

The original PAUSE instruction patch (r5470) does not
compile on Windows. Also, there is an elegant way of
doing it on Windows - YieldProcessor().

Approved by: Heikki (on IM)

This patch introduces heuristics based flushing rate of dirty pages to
avoid IO bursts at checkpoint.

1) log_capacity / log_generated per second gives us number of seconds
in which ALL dirty pages need to be flushed. Based on this rough
assumption we can say that
n_dirty_pages / (log_capacity / log_generation_rate) = desired_flush_rate

2) We use weighted averages (hard coded to 20 seconds) of
log_generation_rate to avoid resonance.

3) From the desired_flush_rate we subtract the number of pages that have
been flushed due to LRU flushing. That gives us pages that we should
flush as part of flush_list cleanup. And that is the number (capped by
maximum io_capacity) that we try to flush from the master thread.

Knobs:
======

innodb_adaptive_flushing: boolean, global, dynamic, default TRUE.
Since this heuristic is very experimental and has the potential to
dramatically change the IO pattern I think it is a good idea to leave a
knob to turn it off.

Approved by: Heikki

The current implementation is to try to flush the neighbors of every
page that we flush. This patch makes the following distinction:

1) If the flush is from flush_list AND
2) If the flush is intended to move the oldest_modification LSN ahead
(this happens when a user thread sees little space in the log file and
attempts to flush pages from the buffer pool so that a checkpoint can
be made)

THEN

Do not try to flush the neighbors. Just focus on flushing dirty pages at
the end of flush_list

Approved by: Heikki

Based on contribution from Google Inc.
This patch introduces a new parameter innodb_io_capacity to control the
rate at which master threads performs various tasks. The default value
is 200 and higher values imply more aggressive flushing and ibuf merges
from within the master thread.
This patch also changes the ibuf merge from synchronous to asynchronous.
Another minor change is not to force the master thread to wait for a
log flush to complete every second.

Approved by: Heikki

Windows will support PAUSE instruction by default.

1.0.4 contains patches based on contributions from Percona
and Sun Microsystems.

Remove unnecessary quotes and simplify plug.in.

Use PAUSE instruction inside spinloop if it is available.

The patch was originally developed by Mikael Ronstrom <mikael@mysql.com>
and can be found here:
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2768
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2771
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2772
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2774
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2777
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2799
http://bazaar.launchpad.net/%7Emysql/mysql-server/mysql-5.4/revision/2800

Approved by:	Heikki (rb://137)

Non functional change. s/Percona/Percona Inc./

(ISO 8859-1 encoded U+00AD SOFT HYPHEN) with a cheap ASCII substitute.

as this function is used within InnoDB Hot Backup.

with UNIV_HOTBACKUP.

Invoke os_get_os_version() at most once.

very similar to ha_search_and_delete_if_found().