4 bytes, using my_checksum() (the old checksum was one byte and just
a sum of the bytes - that was before I saw we have my_checksum :)
storage/maria/ma_control_file.c:
stronger checksum (4 bytes instead of 1, and using CRC instead of
simple byte sum).
storage/maria/unittest/ma_control_file-t.c:
Checksum is now 4 bytes (total length of control file is now 23),
so LSN and LAST_LOGNO move.
Fixed befaviour when loghandler flags changed from one
run to another one.
Description of maria transaction log and control file
added to the file commandÍs magic number file.
mysys/mf_pagecache.c:
postreview changes
storage/maria/ma_control_file.c:
Postreview changes.
storage/maria/ma_control_file.h:
Postreview changes.
storage/maria/ma_loghandler.c:
Postreview changes.
Fixed befaviour when loghandler flags changed from
one run to another one.
storage/maria/ma_loghandler.h:
Postreview changes.
Functions comment left only near the function body.
storage/maria/ma_loghandler_lsn.h:
Postreview changes.
storage/maria/unittest/ma_test_loghandler-t.c:
Postreview changes.
storage/maria/unittest/ma_test_loghandler_multigroup-t.c:
Postreview changes.
storage/maria/unittest/ma_test_loghandler_multithread-t.c:
Postreview changes.
storage/maria/unittest/ma_test_loghandler_pagecache-t.c:
Postreview changes.
support-files/magic:
Description of maria transaction log and control file
added to the file commandÕs magic number file.
Some debug info and comments added
include/pagecache.h:
postmerge fix
mysys/mf_pagecache.c:
Postmerge fix (including changing type of LSN)
Additional DBUG_ASSERTs added
Comment about pinning mechanism added
storage/maria/ma_control_file.c:
Used the same LSN storing procedure everywhere
Postmerge fix (including changing type of LSN)
storage/maria/ma_control_file.h:
Postmerge fix (including changing type of LSN)
storage/maria/ma_loghandler.c:
Postmerge fix (including changing type of LSN)
storage/maria/ma_loghandler.h:
Postmerge fix (including changing type of LSN)
storage/maria/ma_loghandler_lsn.h:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/Makefile.am:
Postmerge fix
storage/maria/unittest/ma_control_file-t.c:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/ma_test_loghandler-t.c:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/ma_test_loghandler_multigroup-t.c:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/ma_test_loghandler_multithread-t.c:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/ma_test_loghandler_pagecache-t.c:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/mf_pagecache_consist.c:
Postmerge fix (including changing type of LSN)
storage/maria/unittest/mf_pagecache_single.c:
Postmerge fix (including changing type of LSN)
1) on Mac OS X >=10.3, fcntl() is recommended over fsync (from the
man page: "[With fsync()] the disk drive may also re-order the data
so that later writes may be present while earlier writes are not.
Applications such as databases that require a strict ordering of writes
should use F_FULLFSYNC to ensure their data is written in the order
they expect"). I have seen two other pieces of software changing from
fsync to F_FULLFSYNC on Mac OS X.
2) to make a file creation/deletion/renaming durable on Linux (at least
ext2 as I have tested) (see "man fsync"), a fsync() on the directory
is needed: new functions to do that, and a flag MY_SYNC_DIR to do
it in my_create/my_delete/my_rename.
3) now using this directory syncing when creating he frm if
opt_sync_frm, and for Maria's control file when it is created.
include/my_sys.h:
new flag to my_create/my_delete/my_rename, which asks to sync the
directory after the operation is done (currently does nothing except
on Linux)
libmysql/CMakeLists.txt:
my_create() now depends on my_sync() so my_sync is needed for libmysql
libmysql/Makefile.shared:
my_create() now depends on my_sync() so my_sync is needed for libmysql
mysys/my_create.c:
my_create() can now sync the directory if asked for
mysys/my_delete.c:
my_delete() can now sync the directory if asked for
mysys/my_open.c:
it was a bug that my_close() is done on fd but a positive fd would
still be returned, by my_register_filename().
mysys/my_rename.c:
my_rename() can now sync the two directories (the one of "from" and
the one of "to") if asked for.
mysys/my_sync.c:
On recent Mac OS X, fcntl(F_FULLFSYNC) is recommended over fsync()
(see "man fsync" on Mac OS X 10.3).
my_sync_dir(): to sync a directory after a file creation/deletion/
renaming; can be called directly or via MY_SYNC_DIR in my_create/
my_delete/my_rename(). No-op except on Linux (see "man fsync" on Linux).
my_sync_dir_from_file(): same as above, just more practical when the
caller has a file name but no directory name ready.
Should the #warning even be a #error? I mean do we want to release
binaries which don't guarantee any durability?
sql/log.cc:
a TODO for the future.
sql/unireg.cc:
If we sync the frm it makes sense to also sync its creation in the
directory.
storage/maria/ma_control_file.c:
control file is vital, try to make it to disk
last round of fixes to the storage engines' and plugins' unit tests
structure. Will extract a total patch and push it in 5.1 as has been
approved.
Makefile.am:
unittest must be before storage and plugin, because engine and plugin
may have unit tests which link with libtap which is found in
unitttest.
config/ac-macros/plugins.m4:
When enabling an engine/plugin, add its directory to the list
of directories where unit tests should be searched. That is,
its directory will be recursively searched by our unit test framework
which will execute any executable *-t file.
storage/maria/ma_control_file.c:
those my_message pollute the output of unit tests.
storage/maria/plug.in:
When Maria is enabled, add its unittest Makefile.
unittest/Makefile.am:
plugins too
changing pseudocode to use the structures of the Maria pagecache
("pagecache->changed_blocks" etc) and other Maria structures
inherited from MyISAM (THR_LOCK_maria etc).
mysys/mf_pagecache.c:
comment
storage/maria/ma_checkpoint.c:
changing pseudocode to use the structures of the Maria pagecache
("pagecache->changed_blocks" etc) and other Maria structures
inherited from MyISAM (THR_LOCK_maria etc).
storage/maria/ma_checkpoint.h:
copyright
storage/maria/ma_control_file.c:
copyright
storage/maria/ma_control_file.h:
copyright
storage/maria/ma_least_recently_dirtied.c:
copyright
storage/maria/ma_least_recently_dirtied.h:
copyright
storage/maria/ma_recovery.c:
copyright
storage/maria/ma_recovery.h:
copyright
storage/maria/unittest/Makefile.am:
copyright
Fitting ma_control_file_test into the mytap unittest framework:
new directories:
- unittest/storage/ for unit tests of any storage engine
- unittest/storage/maria for ... Maria, containing ma_control_file-t.
Later, older tests like ma_test*, ma_test_all (but which is Unix
dependent in its current form) could move here too.
The plugins macro enable building of unittest/storage/X for any
enabled engine X which has such a directory.
If Falcon wants to have unit tests there too, I may have to merge
this patch into 5.x one day.
config/ac-macros/plugins.m4:
If a storage engine has a directory in unittest/storage, build this
directory.
configure.in:
build storage engines' unit tests.
storage/maria/Makefile.am:
ma_control_file_test moves to unittest/storage/maria
storage/maria/ma_control_file.c:
more error codes when opening the control file fails.
ma_control_file_end() may now return an error if my_close() failed.
storage/maria/ma_control_file.h:
more error codes when opening the control file fails.
unittest/Makefile.am:
adding unit tests for storage engines.
Note that unit.pl simply recurses into "storage", so if a unit test for
storage engine X has been built previously, and now you re-configure
(without making clean) to disable this engine, then the unit test of
X will not be rebuilt but will still be present in storage/X, so will
be run.
unittest/storage/maria/ma_control_file-t.c:
Making the test fit the mytap framework (return all the way up
the stack instead of assert(); use the mytap functions plan(), ok() etc).
Adding test of file too short/long.
unittest/storage/maria/Makefile.am:
a_control_file-t is added to the Maria unit tests.
Later, older tests (ma_test1 etc) could also move here.
unittest/storage/Makefile.am:
New BitKeeper file ``unittest/storage/Makefile.am''
added checksum of the file. Now we have size + magic string + checksum
to detect that all is ok.
Plus misc fixes for "make dist" to work and the resulting tarball to build
include/Makefile.am:
adding pagecache.h to help the tarball build.
The model of pagecache.h, keycache.h, is in pkginclude_HEADERS,
wonder why. Adding pagecache.h to noinst_HEADERS for now.
storage/maria/Makefile.am:
adding ma_control_file.h to help the tarball build
storage/maria/ma_control_file.c:
adding a simple checksum to the control file.
We protect against corruption of this file like this:
- test size
- test magic string at start
- test checksum
I also add some simple my_message() errors (to be changed to a better
reporting later).
storage/maria/ma_control_file.h:
comments
storage/maria/ma_control_file_test.c:
test of wrong checksum in control file
storage/maria/CMakeLists.txt:
just to make "make dist" happy for now.
- fixes to the control file module
- unit test for it
- renames of all Maria files I created to start with ma_
storage/maria/ma_checkpoint.c:
Rename: storage/maria/checkpoint.c -> storage/maria/ma_checkpoint.c
storage/maria/ma_checkpoint.h:
Rename: storage/maria/checkpoint.h -> storage/maria/ma_checkpoint.h
storage/maria/ma_least_recently_dirtied.c:
Rename: storage/maria/least_recently_dirtied.c -> storage/maria/ma_least_recently_dirtied.c
storage/maria/ma_least_recently_dirtied.h:
Rename: storage/maria/least_recently_dirtied.h -> storage/maria/ma_least_recently_dirtied.h
storage/maria/ma_recovery.c:
Rename: storage/maria/recovery.c -> storage/maria/ma_recovery.c
storage/maria/ma_recovery.h:
Rename: storage/maria/recovery.h -> storage/maria/ma_recovery.h
storage/maria/Makefile.am:
control file module and its unit test program
storage/maria/ma_control_file.c:
DBUG_ tags. Fix for gcc warnings.
log_no -> logno (I felt "_no" sounded like a standalone "No" word).
ma_ prefix for some functions.
last_checkpoint_lsn_at_startup -> last_checkpoint_lsn (no need
to make special vars for the values at startup). Same for last_logno.
ma_control_file_write_and_force() now updates last_checkpoint_lsn
and last_logno, the idea being that they belong to the module,
others should not update them.
And thus when the module shuts down, it zeroes those vars.
storage/maria/ma_control_file.h:
importing structs from Sanja to get the control file module to compile;
we'll remove that when Sanja pushes the log handler.
CONTROL_FILE_IMPOSSIBLE_LOGNO is 0, not FFFFFFFF.
storage/maria/ma_control_file_test.c:
Unit test program for the Maria control file module.
Modelled after other ma_test* files in this directory (so, does
not follow the unit test framework recently introduced with libtap;
TODO as a task on all ma_test* programs).
We test that writing to the control file works, and re-reading from it
too, we check (by reading the file by ourselves) that its content
on disk is correct, and check that a corrupted control file is detected.
2006-09-01 17:53:10 +02:00
Renamed from storage/maria/control_file.c (Browse further)
