- developers - lists.mariadb.org

[Maria-developers] Progress (by Sanja): Avoid flushing keycache to disk in case of DROPping table (53)
by worklog-noreply＠askmonty.org 01 Sep '09

01 Sep '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Avoid flushing keycache to disk in case of DROPping table CREATION DATE..: Thu, 27 Aug 2009, 11:12 SUPERVISOR.....: Monty IMPLEMENTOR....: Sanja COPIES TO......: CATEGORY.......: Server-Sprint TASK ID........: 53 (http://askmonty.org/worklog/?tid=53) VERSION........: Server-5.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 3 ESTIMATE.......: 21 (hours remain) ORIG. ESTIMATE.: 24 PROGRESS NOTES: -=-=(Sanja - Tue, 01 Sep 2009, 15:19)=-=- first patch Worked 3 hours and estimate 21 hours remain (original estimate unchanged). -=-=(Sanja - Thu, 27 Aug 2009, 11:15)=-=- High-Level Specification modified. --- /tmp/wklog.53.old.23490 2009-08-27 11:15:24.000000000 +0300 +++ /tmp/wklog.53.new.23490 2009-08-27 11:15:24.000000000 +0300 @@ -1 +1,3 @@ - +To make special handler call 'prepare to drop' which mark table as changed (in +case of server crash just before removing table it force table to be repaired) +then flush keycache without writing on disk. DESCRIPTION: When DROP TABLE executed keycache flushed to disk before removing table which is wasting of resources. So better to have table removed from keycache without writing changes on disk. HIGH-LEVEL SPECIFICATION: To make special handler call 'prepare to drop' which mark table as changed (in case of server crash just before removing table it force table to be repaired) then flush keycache without writing on disk. ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Progress (by Sanja): Backport key cache to 5.0 (52)
by worklog-noreply＠askmonty.org 01 Sep '09

01 Sep '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Backport key cache to 5.0 CREATION DATE..: Thu, 27 Aug 2009, 09:54 SUPERVISOR.....: Monty IMPLEMENTOR....: Monty COPIES TO......: Sanja CATEGORY.......: Server-Sprint TASK ID........: 52 (http://askmonty.org/worklog/?tid=52) VERSION........: Server-5.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 9 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 8 PROGRESS NOTES: -=-=(Sanja - Tue, 01 Sep 2009, 15:18)=-=- testing Monty's patch Worked 3 hours and estimate 0 hours remain (original estimate increased by 1 hour). -=-=(Guest - Fri, 28 Aug 2009, 10:55)=-=- Low Level Design modified. --- /tmp/wklog.52.old.16945 2009-08-28 10:55:40.000000000 +0300 +++ /tmp/wklog.52.new.16945 2009-08-28 10:55:40.000000000 +0300 @@ -1 +1,22 @@ -Just copy files (mf_keycache.c & key=cache.h) and fix what should be fixed. +Here is instructions of how to repeat this if needed: + +- Copy the following files from the latest MySQL 5.1 to MySQL 5.0: + +include/keycache.h +include/my_bit.h +mysys/mf_keycache.c +mysys/mf_keycaches.c +mysys/my_bit.c + +Apply the following change in mysys/mf_keycaches.c +81c81 +< static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, size_t *length, +--- +> static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, uint *length, + +recompile everything. + +Simple test to verify that this is working: +cd mysql-test +mysql-test-run --valgrind key_cache +(the --valgrind argument can be left out if you don't have valgrind installed) -=-=(Guest - Thu, 27 Aug 2009, 21:52)=-=- - Planning code (Email dicussions & discussions with Sanja): 2 hours. - Backport of code (2 hours). Notice that mysys/mf_keycaches.c::safe_hash_entry_get() is different from the code in MySQL 5.1 - Testing code with mysql-test-run (especially running the key_cache.test with valgrind) - Tested the server with sql-bench Worked 6 hours and estimate 2 hours remain (original estimate unchanged). -=-=(Sanja - Thu, 27 Aug 2009, 09:56)=-=- High-Level Specification modified. --- /tmp/wklog.52.old.20094 2009-08-27 09:56:05.000000000 +0300 +++ /tmp/wklog.52.new.20094 2009-08-27 09:56:05.000000000 +0300 @@ -1 +1 @@ - +Backport keycache to 5.0 -=-=(Sanja - Thu, 27 Aug 2009, 09:55)=-=- Low Level Design modified. --- /tmp/wklog.52.old.20035 2009-08-27 09:55:25.000000000 +0300 +++ /tmp/wklog.52.new.20035 2009-08-27 09:55:25.000000000 +0300 @@ -1 +1 @@ - +Just copy files (mf_keycache.c & key=cache.h) and fix what should be fixed. DESCRIPTION: Backport keycache to bring bugfixes made in 5.1. HIGH-LEVEL SPECIFICATION: Backport keycache to 5.0 LOW-LEVEL DESIGN: Here is instructions of how to repeat this if needed: - Copy the following files from the latest MySQL 5.1 to MySQL 5.0: include/keycache.h include/my_bit.h mysys/mf_keycache.c mysys/mf_keycaches.c mysys/my_bit.c Apply the following change in mysys/mf_keycaches.c 81c81 < static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, size_t *length, --- > static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, uint *length, recompile everything. Simple test to verify that this is working: cd mysql-test mysql-test-run --valgrind key_cache (the --valgrind argument can be left out if you don't have valgrind installed) ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] Progress (by Sanja): Backport key cache to 5.0 (52)
by worklog-noreply＠askmonty.org 01 Sep '09

01 Sep '09

----------------------------------------------------------------------- WORKLOG TASK -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- TASK...........: Backport key cache to 5.0 CREATION DATE..: Thu, 27 Aug 2009, 09:54 SUPERVISOR.....: Monty IMPLEMENTOR....: Monty COPIES TO......: Sanja CATEGORY.......: Server-Sprint TASK ID........: 52 (http://askmonty.org/worklog/?tid=52) VERSION........: Server-5.0 STATUS.........: Assigned PRIORITY.......: 60 WORKED HOURS...: 9 ESTIMATE.......: 0 (hours remain) ORIG. ESTIMATE.: 8 PROGRESS NOTES: -=-=(Sanja - Tue, 01 Sep 2009, 15:18)=-=- testing Monty's patch Worked 3 hours and estimate 0 hours remain (original estimate increased by 1 hour). -=-=(Guest - Fri, 28 Aug 2009, 10:55)=-=- Low Level Design modified. --- /tmp/wklog.52.old.16945 2009-08-28 10:55:40.000000000 +0300 +++ /tmp/wklog.52.new.16945 2009-08-28 10:55:40.000000000 +0300 @@ -1 +1,22 @@ -Just copy files (mf_keycache.c & key=cache.h) and fix what should be fixed. +Here is instructions of how to repeat this if needed: + +- Copy the following files from the latest MySQL 5.1 to MySQL 5.0: + +include/keycache.h +include/my_bit.h +mysys/mf_keycache.c +mysys/mf_keycaches.c +mysys/my_bit.c + +Apply the following change in mysys/mf_keycaches.c +81c81 +< static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, size_t *length, +--- +> static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, uint *length, + +recompile everything. + +Simple test to verify that this is working: +cd mysql-test +mysql-test-run --valgrind key_cache +(the --valgrind argument can be left out if you don't have valgrind installed) -=-=(Guest - Thu, 27 Aug 2009, 21:52)=-=- - Planning code (Email dicussions & discussions with Sanja): 2 hours. - Backport of code (2 hours). Notice that mysys/mf_keycaches.c::safe_hash_entry_get() is different from the code in MySQL 5.1 - Testing code with mysql-test-run (especially running the key_cache.test with valgrind) - Tested the server with sql-bench Worked 6 hours and estimate 2 hours remain (original estimate unchanged). -=-=(Sanja - Thu, 27 Aug 2009, 09:56)=-=- High-Level Specification modified. --- /tmp/wklog.52.old.20094 2009-08-27 09:56:05.000000000 +0300 +++ /tmp/wklog.52.new.20094 2009-08-27 09:56:05.000000000 +0300 @@ -1 +1 @@ - +Backport keycache to 5.0 -=-=(Sanja - Thu, 27 Aug 2009, 09:55)=-=- Low Level Design modified. --- /tmp/wklog.52.old.20035 2009-08-27 09:55:25.000000000 +0300 +++ /tmp/wklog.52.new.20035 2009-08-27 09:55:25.000000000 +0300 @@ -1 +1 @@ - +Just copy files (mf_keycache.c & key=cache.h) and fix what should be fixed. DESCRIPTION: Backport keycache to bring bugfixes made in 5.1. HIGH-LEVEL SPECIFICATION: Backport keycache to 5.0 LOW-LEVEL DESIGN: Here is instructions of how to repeat this if needed: - Copy the following files from the latest MySQL 5.1 to MySQL 5.0: include/keycache.h include/my_bit.h mysys/mf_keycache.c mysys/mf_keycaches.c mysys/my_bit.c Apply the following change in mysys/mf_keycaches.c 81c81 < static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, size_t *length, --- > static uchar *safe_hash_entry_get(SAFE_HASH_ENTRY *entry, uint *length, recompile everything. Simple test to verify that this is working: cd mysql-test mysql-test-run --valgrind key_cache (the --valgrind argument can be left out if you don't have valgrind installed) ESTIMATED WORK TIME ESTIMATED COMPLETION DATE ----------------------------------------------------------------------- WorkLog (v3.5.9)

1 0

[Maria-developers] [Branch ~maria-captains/maria/5.1] Rev 2723: Trivial fixes for windows compile failures
by noreply＠launchpad.net 01 Sep '09

01 Sep '09

------------------------------------------------------------ revno: 2723 committer: Sergey Petrunya <psergey(a)askmonty.org> branch nick: maria-5.1-fixwindows timestamp: Tue 2009-09-01 15:59:54 +0400 message: Trivial fixes for windows compile failures - Add strmov_overlapp.c where it's needed - Add libevent where it is required - Fix typo in unittest/mysys/waiting_threads-t.c - fix xtradb C language source so that they have variable declarations before any operators, as it is (or was?) required in C. modified: libmysql/CMakeLists.txt sql/CMakeLists.txt storage/xtradb/btr/btr0cur.c storage/xtradb/buf/buf0buddy.c storage/xtradb/buf/buf0flu.c storage/xtradb/fil/fil0fil.c storage/xtradb/handler/i_s.cc storage/xtradb/srv/srv0srv.c strings/CMakeLists.txt unittest/mysys/waiting_threads-t.c -- lp:maria https://code.launchpad.net/~maria-captains/maria/5.1 Your team Maria developers is subscribed to branch lp:maria. To unsubscribe from this branch go to https://code.launchpad.net/~maria-captains/maria/5.1/+edit-subscription.

1 0

[Maria-developers] Rev 2723: Trivial fixes for windows compile failures in file:///home/psergey/dev/maria-5.1-fixwindows/
by Sergey Petrunya 01 Sep '09

01 Sep '09

At file:///home/psergey/dev/maria-5.1-fixwindows/ ------------------------------------------------------------ revno: 2723 revision-id: psergey(a)askmonty.org-20090901115954-nge2brzh3akkmtlz parent: knielsen(a)knielsen-hq.org-20090831112339-td6t7ljy4w2oamvt committer: Sergey Petrunya <psergey(a)askmonty.org> branch nick: maria-5.1-fixwindows timestamp: Tue 2009-09-01 15:59:54 +0400 message: Trivial fixes for windows compile failures - Add strmov_overlapp.c where it's needed - Add libevent where it is required - Fix typo in unittest/mysys/waiting_threads-t.c - fix xtradb C language source so that they have variable declarations before any operators, as it is (or was?) required in C. === modified file 'libmysql/CMakeLists.txt' --- a/libmysql/CMakeLists.txt 2008-04-28 16:24:05 +0000 +++ b/libmysql/CMakeLists.txt 2009-09-01 11:59:54 +0000 @@ -94,7 +94,7 @@ ../mysys/safemalloc.c ../mysys/sha1.c ../strings/str2int.c ../strings/str_alloc.c ../strings/strcend.c ../strings/strcont.c ../strings/strend.c ../strings/strfill.c ../mysys/string.c ../strings/strinstr.c ../strings/strmake.c - ../strings/strmov.c ../strings/strnlen.c ../strings/strnmov.c ../strings/strtod.c + ../strings/strmov.c ../strings/strmov_overlapp.c ../strings/strnlen.c ../strings/strnmov.c ../strings/strtod.c ../strings/strtoll.c ../strings/strtoull.c ../strings/strxmov.c ../strings/strxnmov.c ../mysys/thr_mutex.c ../mysys/typelib.c ../vio/vio.c ../vio/viosocket.c ../vio/viossl.c ../vio/viosslfactories.c ../strings/xml.c ../mysys/mf_qsort.c === modified file 'sql/CMakeLists.txt' --- a/sql/CMakeLists.txt 2008-11-21 14:21:50 +0000 +++ b/sql/CMakeLists.txt 2009-09-01 11:59:54 +0000 @@ -25,6 +25,7 @@ ${CMAKE_SOURCE_DIR}/sql ${CMAKE_SOURCE_DIR}/regex ${CMAKE_SOURCE_DIR}/zlib + ${CMAKE_SOURCE_DIR}/extra/libevent ) SET_SOURCE_FILES_PROPERTIES(${CMAKE_SOURCE_DIR}/sql/sql_yacc.h @@ -85,7 +86,7 @@ TARGET_LINK_LIBRARIES(mysqld heap myisam myisammrg mysys yassl zlib debug dbug yassl - taocrypt strings vio regex wsock32 ws2_32) + taocrypt strings vio regex libevent wsock32 ws2_32) SET_TARGET_PROPERTIES(mysqld PROPERTIES OUTPUT_NAME mysqld${MYSQLD_EXE_SUFFIX}) @@ -139,7 +140,7 @@ # Gen_lex_hash ADD_EXECUTABLE(gen_lex_hash gen_lex_hash.cc) -TARGET_LINK_LIBRARIES(gen_lex_hash debug dbug mysqlclient wsock32) +TARGET_LINK_LIBRARIES(gen_lex_hash debug dbug mysqlclient strings wsock32) GET_TARGET_PROPERTY(GEN_LEX_HASH_EXE gen_lex_hash LOCATION) ADD_CUSTOM_COMMAND( OUTPUT ${PROJECT_SOURCE_DIR}/sql/lex_hash.h === modified file 'storage/xtradb/btr/btr0cur.c' --- a/storage/xtradb/btr/btr0cur.c 2009-06-25 01:43:25 +0000 +++ b/storage/xtradb/btr/btr0cur.c 2009-09-01 11:59:54 +0000 @@ -3267,6 +3267,7 @@ } while (rec != supremum) { + rec_t* next_rec; /* count recs */ if (stats_method == SRV_STATS_METHOD_IGNORE_NULLS) { n_recs++; @@ -3281,7 +3282,7 @@ } } - rec_t* next_rec = page_rec_get_next(rec); + next_rec = page_rec_get_next(rec); if (next_rec == supremum) { break; } === modified file 'storage/xtradb/buf/buf0buddy.c' --- a/storage/xtradb/buf/buf0buddy.c 2009-06-25 01:43:25 +0000 +++ b/storage/xtradb/buf/buf0buddy.c 2009-09-01 11:59:54 +0000 @@ -461,11 +461,11 @@ actually is a properly initialized buf_page_t object. */ if (size >= PAGE_ZIP_MIN_SIZE) { + mutex_t* mutex; if (!have_page_hash_mutex) mutex_exit(&zip_free_mutex); /* This is a compressed page. */ - mutex_t* mutex; if (!have_page_hash_mutex) { mutex_enter(&LRU_list_mutex); === modified file 'storage/xtradb/buf/buf0flu.c' --- a/storage/xtradb/buf/buf0flu.c 2009-06-25 01:43:25 +0000 +++ b/storage/xtradb/buf/buf0flu.c 2009-09-01 11:59:54 +0000 @@ -1106,11 +1106,13 @@ page_count += buf_flush_try_neighbors( space, offset, flush_type); } else { + mutex_t* block_mutex; + buf_page_t* bpage_tmp; /* Try to flush the page only */ //buf_pool_mutex_enter(); rw_lock_s_lock(&page_hash_latch); - mutex_t* block_mutex = buf_page_get_mutex(bpage); + block_mutex = buf_page_get_mutex(bpage); retry_lock_2: mutex_enter(block_mutex); if (block_mutex != buf_page_get_mutex(bpage)) { @@ -1119,7 +1121,7 @@ goto retry_lock_2; } - buf_page_t* bpage_tmp = buf_page_hash_get(space, offset); + bpage_tmp = buf_page_hash_get(space, offset); if (bpage_tmp) { buf_flush_page(bpage_tmp, flush_type); page_count++; @@ -1238,13 +1240,14 @@ + BUF_FLUSH_EXTRA_MARGIN) && (distance < BUF_LRU_FREE_SEARCH_LEN)) { - if (!bpage->in_LRU_list) { + mutex_t* block_mutex; + if (!bpage->in_LRU_list) { /* reatart. but it is very optimistic */ bpage = UT_LIST_GET_LAST(buf_pool->LRU); continue; } - mutex_t* block_mutex = buf_page_get_mutex(bpage); + block_mutex = buf_page_get_mutex(bpage); retry_lock: mutex_enter(block_mutex); === modified file 'storage/xtradb/fil/fil0fil.c' --- a/storage/xtradb/fil/fil0fil.c 2009-06-25 01:43:25 +0000 +++ b/storage/xtradb/fil/fil0fil.c 2009-09-01 11:59:54 +0000 @@ -3039,6 +3039,13 @@ ulint i; int len; ib_uint64_t current_lsn; + ulint size_low, size_high, size; + ib_int64_t size_bytes; + dict_table_t* table; + dict_index_t* index; + fil_system_t* system; + fil_node_t* node = NULL; + fil_space_t* space; current_lsn = log_get_lsn(); @@ -3060,15 +3067,11 @@ success = os_file_write(filepath, file, page, 0, 0, UNIV_PAGE_SIZE); /* get file size */ - ulint size_low, size_high, size; - ib_int64_t size_bytes; os_file_get_size(file, &size_low, &size_high); size_bytes = (((ib_int64_t)size_high) << 32) + (ib_int64_t)size_low; /* get cruster index information */ - dict_table_t* table; - dict_index_t* index; table = dict_table_get_low(name); index = dict_table_get_first_index(table); ut_a(index->page==3); @@ -3076,9 +3079,9 @@ /* read metadata from .exp file */ n_index = 0; - bzero(old_id, sizeof(old_id)); - bzero(new_id, sizeof(new_id)); - bzero(root_page, sizeof(root_page)); + memset(old_id, 0, sizeof(old_id)); + memset(new_id, 0, sizeof(new_id)); + memset(root_page, 0, sizeof(root_page)); info_file_path = fil_make_ibd_name(name, FALSE); len = strlen(info_file_path); @@ -3128,9 +3131,9 @@ mem_heap_t* heap = NULL; ulint offsets_[REC_OFFS_NORMAL_SIZE]; ulint* offsets = offsets_; + ib_int64_t offset; size = (ulint) (size_bytes / UNIV_PAGE_SIZE); /* over write space id of all pages */ - ib_int64_t offset; rec_offs_init(offsets_); @@ -3284,10 +3287,8 @@ } mem_free(info_file_path); - fil_system_t* system = fil_system; + system = fil_system; mutex_enter(&(system->mutex)); - fil_node_t* node = NULL; - fil_space_t* space; space = fil_space_get_by_id(id); if (space) node = UT_LIST_GET_FIRST(space->chain); === modified file 'storage/xtradb/handler/i_s.cc' --- a/storage/xtradb/handler/i_s.cc 2009-06-25 01:43:25 +0000 +++ b/storage/xtradb/handler/i_s.cc 2009-09-01 11:59:54 +0000 @@ -763,7 +763,7 @@ dict_index_t* index; dulint index_id; - char *p; + const char *p; char db_name_raw[NAME_LEN*5+1], db_name[NAME_LEN+1]; char table_name_raw[NAME_LEN*5+1], table_name[NAME_LEN+1]; @@ -2861,7 +2861,7 @@ } else { rec_per_key = n_rows; } - snprintf(buff, 256, (i == index->n_uniq)?"%llu":"%llu, ", + ut_snprintf(buff, 256, (i == index->n_uniq)?"%llu":"%llu, ", rec_per_key); strncat(row_per_keys, buff, 256 - strlen(row_per_keys)); } === modified file 'storage/xtradb/srv/srv0srv.c' --- a/storage/xtradb/srv/srv0srv.c 2009-07-06 05:47:15 +0000 +++ b/storage/xtradb/srv/srv0srv.c 2009-09-01 11:59:54 +0000 @@ -2693,11 +2693,11 @@ /* defence line (max_checkpoint_age * 1/2) */ ib_uint64_t lsn = log_sys->lsn; - mutex_exit(&(log_sys->mutex)); - ib_uint64_t level, bpl; buf_page_t* bpage; + mutex_exit(&(log_sys->mutex)); + mutex_enter(&flush_list_mutex); level = 0; === modified file 'strings/CMakeLists.txt' --- a/strings/CMakeLists.txt 2008-10-10 15:28:41 +0000 +++ b/strings/CMakeLists.txt 2009-09-01 11:59:54 +0000 @@ -23,7 +23,7 @@ ctype-latin1.c ctype-mb.c ctype-simple.c ctype-sjis.c ctype-tis620.c ctype-uca.c ctype-ucs2.c ctype-ujis.c ctype-utf8.c ctype-win1250ch.c ctype.c decimal.c int2str.c is_prefix.c llstr.c longlong2str.c my_strtoll10.c my_vsnprintf.c r_strinstr.c - str2int.c str_alloc.c strcend.c strend.c strfill.c strmake.c strmov.c strnmov.c + str2int.c str_alloc.c strcend.c strend.c strfill.c strmake.c strmov.c strmov_overlapp.c strnmov.c strtod.c strtol.c strtoll.c strtoul.c strtoull.c strxmov.c strxnmov.c xml.c my_strchr.c strcont.c strinstr.c strnlen.c strappend.c) === modified file 'unittest/mysys/waiting_threads-t.c' --- a/unittest/mysys/waiting_threads-t.c 2009-03-20 11:41:30 +0000 +++ b/unittest/mysys/waiting_threads-t.c 2009-09-01 11:59:54 +0000 @@ -263,6 +263,7 @@ kill_strategy=X; \ do_one_test(); #else +#define test_kill_strategy(X) \ diag("kill strategy: " #X); \ DBUG_PRINT("info", ("kill strategy: " #X)); \ kill_strategy=X; \

1 0

[Maria-developers] Rev 2720: Fix for in file:///Users/hakan/work/monty_program/maria/
by Hakan Kuecuekyilmaz 31 Aug '09

31 Aug '09

At file:///Users/hakan/work/monty_program/maria/ ------------------------------------------------------------ revno: 2720 revision-id: hakan(a)askmonty.org-20090831214839-gsh1r8mnkrdyzc38 parent: knielsen(a)knielsen-hq.org-20090805072137-wg97dcem1cxnzt3p committer: Hakan Kuecuekyilmaz <hakan(a)askmonty.org> branch nick: maria timestamp: Mon 2009-08-31 23:48:39 +0200 message: Fix for mysqlslap: setting --engine does not get replicated http://bugs.mysql.com/bug.php?id=46967 Problem was that a "set storage_engine = <storage_engine>" is not replicated. Fixed by adding an explicit Engine = <storage_engine> to the build_table_string() function. Open question: Should MySQL's replication actually replicate the set storage_engine command or not? === modified file 'client/mysqlslap.c' --- a/client/mysqlslap.c 2009-04-25 10:05:32 +0000 +++ b/client/mysqlslap.c 2009-08-31 21:48:39 +0000 @@ -880,6 +880,12 @@ } dynstr_append(&table_string, ")"); + + if (default_engine) { + dynstr_append(&table_string, " Engine = "); + dynstr_append(&table_string, default_engine); + } + ptr= (statement *)my_malloc(sizeof(statement), MYF(MY_ZEROFILL|MY_FAE|MY_WME)); ptr->string = (char *)my_malloc(table_string.length+1,

1 0

[Maria-developers] Rev 2720: Fix for in file:///Users/hakan/work/monty_program/maria/
by Hakan Kuecuekyilmaz 31 Aug '09

31 Aug '09

At file:///Users/hakan/work/monty_program/maria/ ------------------------------------------------------------ revno: 2720 revision-id: hakan(a)askmonty.org-20090831212214-wg9fwhgnhx1ly6w2 parent: knielsen(a)knielsen-hq.org-20090805072137-wg97dcem1cxnzt3p committer: Hakan Kuecuekyilmaz <hakan(a)askmonty.org> branch nick: maria timestamp: Mon 2009-08-31 23:22:14 +0200 message: Fix for mysqlslap: setting --engine does not get replicated http://bugs.mysql.com/bug.php?id=46967 Problem was that a "set storage_engine = <storage_engine>" is not replicated. Fixed by adding an explicit Engine = <storage_engine> to the build_table_string() function. Open question: Should MySQL's replication actually replicate the set storage_engine command or not? === modified file 'client/mysqlslap.c' --- a/client/mysqlslap.c 2009-04-25 10:05:32 +0000 +++ b/client/mysqlslap.c 2009-08-31 21:22:14 +0000 @@ -880,6 +880,12 @@ } dynstr_append(&table_string, ")"); + + if (default_engine) { + dynstr_append(&table_string, " Engine = "); + dynstr_append(&table_string, default_engine); + } + ptr= (statement *)my_malloc(sizeof(statement), MYF(MY_ZEROFILL|MY_FAE|MY_WME)); ptr->string = (char *)my_malloc(table_string.length+1,

1 0

[Maria-developers] [Branch ~maria-captains/maria/5.1] Rev 2722: Merge Paul's maria-pbxt-rc2 branch into MariaDB.
by noreply＠launchpad.net 31 Aug '09

31 Aug '09

Merge authors: Kristian Nielsen (knielsen) Paul McCullagh (paul-mccullagh) Related merge proposals: https://code.launchpad.net/~paul-mccullagh/maria/maria-pbxt-rc2/+merge/10303 proposed by: Paul McCullagh (paul-mccullagh) ------------------------------------------------------------ revno: 2722 [merge] committer: knielsen(a)knielsen-hq.org branch nick: work timestamp: Mon 2009-08-31 13:23:39 +0200 message: Merge Paul's maria-pbxt-rc2 branch into MariaDB. Includes disabling PBMS (per Paul's recommendation) and test file fixes. Signed-off-by: Kristian Nielsen <knielsen(a)knielsen-hq.org> removed: storage/pbxt/src/streaming_xt.cc storage/pbxt/src/streaming_xt.h added: mysql-test/suite/pbxt/t/lowercase_table_grant-master.opt mysql-test/suite/pbxt/t/lowercase_table_qcache-master.opt mysql-test/suite/pbxt/t/lowercase_view-master.opt mysql-test/suite/pbxt/t/udf-master.opt storage/pbxt/src/pbms_enabled.cc storage/pbxt/src/pbms_enabled.h modified: mysql-test/suite/pbxt/r/alter_table.result mysql-test/suite/pbxt/r/analyze.result mysql-test/suite/pbxt/r/auto_increment.result mysql-test/suite/pbxt/r/delete.result mysql-test/suite/pbxt/r/distinct.result mysql-test/suite/pbxt/r/func_group.result mysql-test/suite/pbxt/r/func_math.result mysql-test/suite/pbxt/r/func_str.result mysql-test/suite/pbxt/r/grant.result mysql-test/suite/pbxt/r/group_min_max.result mysql-test/suite/pbxt/r/join.result mysql-test/suite/pbxt/r/join_nested.result mysql-test/suite/pbxt/r/key.result mysql-test/suite/pbxt/r/key_cache.result mysql-test/suite/pbxt/r/key_diff.result mysql-test/suite/pbxt/r/lowercase_view.result mysql-test/suite/pbxt/r/mysqlshow.result mysql-test/suite/pbxt/r/null.result mysql-test/suite/pbxt/r/null_key.result mysql-test/suite/pbxt/r/partition_pruning.result mysql-test/suite/pbxt/r/pbxt_bugs.result mysql-test/suite/pbxt/r/pbxt_ref_int.result mysql-test/suite/pbxt/r/preload.result mysql-test/suite/pbxt/r/ps_1general.result mysql-test/suite/pbxt/r/range.result mysql-test/suite/pbxt/r/schema.result mysql-test/suite/pbxt/r/select.result mysql-test/suite/pbxt/r/select_safe.result mysql-test/suite/pbxt/r/subselect.result mysql-test/suite/pbxt/r/type_enum.result mysql-test/suite/pbxt/r/type_ranges.result mysql-test/suite/pbxt/r/type_timestamp.result mysql-test/suite/pbxt/r/union.result mysql-test/suite/pbxt/r/view_grant.result mysql-test/suite/pbxt/t/auto_increment.test mysql-test/suite/pbxt/t/delete.test mysql-test/suite/pbxt/t/join_nested.test mysql-test/suite/pbxt/t/null.test mysql-test/suite/pbxt/t/pbxt_bugs.test mysql-test/suite/pbxt/t/rename.test mysql-test/suite/pbxt/t/schema.test mysql-test/suite/pbxt/t/type_enum.test mysql-test/suite/pbxt/t/union.test storage/pbxt/ChangeLog storage/pbxt/src/Makefile.am storage/pbxt/src/cache_xt.cc storage/pbxt/src/cache_xt.h storage/pbxt/src/ccutils_xt.cc storage/pbxt/src/database_xt.cc storage/pbxt/src/datadic_xt.cc storage/pbxt/src/datadic_xt.h storage/pbxt/src/datalog_xt.cc storage/pbxt/src/datalog_xt.h storage/pbxt/src/discover_xt.cc storage/pbxt/src/filesys_xt.cc storage/pbxt/src/filesys_xt.h storage/pbxt/src/ha_pbxt.cc storage/pbxt/src/ha_pbxt.h storage/pbxt/src/ha_xtsys.cc storage/pbxt/src/ha_xtsys.h storage/pbxt/src/hashtab_xt.cc storage/pbxt/src/heap_xt.cc storage/pbxt/src/heap_xt.h storage/pbxt/src/index_xt.cc storage/pbxt/src/index_xt.h storage/pbxt/src/lock_xt.cc storage/pbxt/src/lock_xt.h storage/pbxt/src/locklist_xt.cc storage/pbxt/src/locklist_xt.h storage/pbxt/src/memory_xt.cc storage/pbxt/src/memory_xt.h storage/pbxt/src/myxt_xt.cc storage/pbxt/src/myxt_xt.h storage/pbxt/src/pbms.h storage/pbxt/src/pthread_xt.cc storage/pbxt/src/restart_xt.cc storage/pbxt/src/restart_xt.h storage/pbxt/src/sortedlist_xt.cc storage/pbxt/src/strutil_xt.cc storage/pbxt/src/systab_xt.cc storage/pbxt/src/systab_xt.h storage/pbxt/src/tabcache_xt.cc storage/pbxt/src/tabcache_xt.h storage/pbxt/src/table_xt.cc storage/pbxt/src/table_xt.h storage/pbxt/src/thread_xt.cc storage/pbxt/src/thread_xt.h storage/pbxt/src/trace_xt.cc storage/pbxt/src/trace_xt.h storage/pbxt/src/util_xt.cc storage/pbxt/src/xaction_xt.cc storage/pbxt/src/xaction_xt.h storage/pbxt/src/xactlog_xt.cc storage/pbxt/src/xactlog_xt.h storage/pbxt/src/xt_config.h storage/pbxt/src/xt_defs.h The size of the diff (15366 lines) is larger than your specified limit of 1000 lines -- lp:maria https://code.launchpad.net/~maria-captains/maria/5.1 Your team Maria developers is subscribed to branch lp:maria. To unsubscribe from this branch go to https://code.launchpad.net/~maria-captains/maria/5.1/+edit-subscription.

1 0

[Maria-developers] Rev 2749: MWL#17: Table-elimination in file:///home/psergey/dev/maria-5.1-table-elim-r11/
by Sergey Petrunya 31 Aug '09

31 Aug '09

At file:///home/psergey/dev/maria-5.1-table-elim-r11/ ------------------------------------------------------------ revno: 2749 revision-id: psergey(a)askmonty.org-20090831200209-7gbnkr7omjbgld1x parent: psergey(a)askmonty.org-20090826210140-y2pba3ucdwnvr956 committer: Sergey Petrunya <psergey(a)askmonty.org> branch nick: maria-5.1-table-elim-r11 timestamp: Tue 2009-09-01 00:02:09 +0400 message: MWL#17: Table-elimination - Addressing review feedback, generation 4. Diff too large for email (2658 lines, the limit is 1000).

1 0

[Maria-developers] MariaDB versioning
by Peter Laursen 31 Aug '09

31 Aug '09

I have a concern that I will ask you all to consider while there is time. I think I understand that after MariaDB 5.1 a 5.2 release is planned. I think it should not be versioned like that. Problem is that there is a MySQL 5.2 tree too (it is basically an early 6.0) and it is still available for download from FTP-mirrors like ftp://mirrors.dotsrc.org/mysql/Downloads/ . . For generic and GUI clients (ie. not designed for specific server versions and where SQL is generated and not written by user) it is important that the version string can be used to identify what SQL statements are supported by server. The first thing such clients (like our own SQLyog) does after connection is to SELECT VERSION() .. and only after that we know if there is support for EVENTS, if IF EXISTS is supported for DROP TRIGGER sysntax and so on. If there are two different 5.2 trees (MariaDB based on MySQL 5.1 and MySQL 5.2) it gets more difficult to handle. . I asked Monty about this at one of his sessions at the latest UC and he told like 'we are good citizens - except for the comment field MariaDB will be versioned like the MySQL code it is based upon'. What I also think is the best solution. The *version* should not be the same for different builds with even the slightest difference in what SQL server understands. On the opposite version should be the same as long as SQL- interface (even with possible Engine-specific extensions) is the same no matter how much they differ internally. Maybe in the long run diversification is inevitable. But in that case it should be planned well! Peter Webyog -- This message was sent from Launchpad by the user Peter Laursen (https://launchpad.net/~peter-laursen) using the "Contact this team" link on the Maria Discuss team page to each member directly. For more information see https://help.launchpad.net/YourAccount/ContactingPeople

3 3