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.
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
or SET GLOBAL innodb_file_format_check=DEFAULT.
innodb_file_format.test: New test for innodb_file_format and
innodb_file_format_check.
innodb_file_format_name_validate(): Store the string in *save.
innodb_file_format_name_update(): Check the string again.
innodb_file_format_check_validate(): Store the string in *save.
innodb_file_format_check_update(): Check the string again.
Issue #282, rb://140 approved by Heikki Tuuri
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.
Change the header file path from ../storage/innobase/include/
to ../include/. In the planned 5.1 + plugin release, the source
directory of the plugin will not be in storage/innobase.
Approved by: Heikki (IM)
