Mirrors/php - php - Gitea: Git with a cup of tea

Commit Graph

Author	SHA1	Message	Date
Andrey Hristov	cac5325aac	test 2	17 years ago
Andrey Hristov	af22257ba4	test	17 years ago
Johannes Schlüter	81761e636d	Fix path	17 years ago
Pierre Joye	a68f6c9c28	- fix build	17 years ago
Johannes Schlüter	9485cad688	Revert mysqlnd to RC3 state after issues with pconnects	17 years ago
David Soria Parra	4d7e6edd63	MFH: Fix bug #48644 mysqlnd does not compile with '--enable-mysqlnd-threading'	17 years ago
Johannes Schlüter	86eaa90955	MFH The experimental warning sohuld be in the configure output...	17 years ago
Johannes Schlüter	067276a043	Temporarily disable this optimization causing too much trouble, kept in HEAD	17 years ago
Andrey Hristov	0aada07656	MFH: Fix two problems: - The value of mysqli_get_client_info() has been changed recently and did not include "mysqlnd" anymore thus the test suite was thinking the build is always libmysql. This did not kept the suite from running pconn tests - Going back to the libc allocator because the memory arena could be on a persistent connections. If the build is not debug there will be no error but the memory will be freed and in the second use of this pconn freed memory will be used - not good! For now the arena doesn't take an argument whether it should allocate persistently or not, thus persistent is safe for now. Johannes gave his +1 to commit this.	17 years ago
Andrey Hristov	30800a79b9	MFH: C-comments	17 years ago
Andrey Hristov	10ee06ca48	MFH: Memory usage optimisation. mysqlnd is not libmysql. mysqlnd does use the Zend allocator, which means that is easier to hit memory_limit if you have big stored (buffered) result sets. Before with libmysql you won't hit memory_limit because libmysql uses libc's allocator and nothing is checked. Now, with mysqlnd the situation is stricter and it is easier to hit memory_limit. We try to optimize for big result sets. If a result set is larger than 10 rows we will start freeing some data to keep memory usage after 10 rows constant. This will help in the cases where a buffered result set is scrolled forward only and just only once, or mysqlnd will need to decode data from the network buffers again - yes, it is a trade-off between CPU time and memory size. The best for big result sets is of course using unbuffered queries - for comparison : 3 Million rows with buffered take at least 180MB, with buffered you will stay at 3MB, and unbuffered will be just 7-8% slower.	17 years ago
Andrey Hristov	27270621cb	MFH: Hardwire function call instead of using callbacks. We don't actually need callbacks, it was done for making 2 functions static, not to pollute the global functions space but that had its price of 8 bytes overheat per allocation, which is just too much. Also making the app member 32b instead of 64b, which should save additional 4 byte, to the total of 12 byte per allocation of a row buffer.	17 years ago
Andrey Hristov	a2acaa7fcd	MFH: Use Zend's allocator instead of libc's and also don't try to work on zvals that are NULLs.	17 years ago
Andrey Hristov	22c8786493	MFH: Fix crash when tracing is enabled. Position after buffer was used also direct usage of MYSQLND_STRING pointer instead of the "s" property of the structure.	17 years ago
Andrey Hristov	9a2cc0ce67	Put these in an ifdef because they are not used at all if the zval cache is off.	17 years ago
Pierre Joye	f2fcd8fb29	- silent warning and make it less ambiguous	17 years ago
Johannes Schlüter	11340cf299	MFH: Use a better extension version	17 years ago
Andrey Hristov	28c8d25dca	MFH: Protect the code when variable is NULL, which should never happen, but anyway, you know about "never happens" :)	17 years ago
Andrey Hristov	cff09ba5df	MFH: Check the pointer before calling a function on it, or we will crash. This is a very rare situation where the server is totally broken.	17 years ago
Andrey Hristov	a3b85c28c6	Sync with PHP6 - TSRMLS_FETCH Extract reference decrement in own function Remove PHPAPI of a function and make it static as it is not called externally	17 years ago
Andrey Hristov	ec777b89f4	Merge with HEAD. Someone committed changes to HEAD and did not merge back to the branch. Also switch off the zval cache, for now.	17 years ago
Andrey Hristov	c1d46eb502	MFH:Add support for mysql_stmt_store_result() from libmysql 6.0.8+ and 5.4.x	17 years ago
Andrey Hristov	ccf0665170	MFH:Fix for the failing mysql_stmt_execute() test. Data was cleaned after stmt_reset() but it should stay and be freed as later as next stmt_execute()	17 years ago
Andrey Hristov	133296f962	MFH:Fix a very well hidden error because of not being careful with CPP. I usually don't make this mistake :)	17 years ago
Andrey Hristov	804f6138a6	MFH:Fix a valgrind warning as well as more trace log information	17 years ago
Andrey Hristov	76d091d5f5	MFH:Fix a problem with cursors, which did not happen with unbuffered PS for some reason. Double free of the data, which led to valgrind warnigns. The fix actually optimizes the code in this cases because the old code used copy_ctor while the new one skips it because it is not needed. Transferring data ownership and nulling works best, for PS where we always copy the string from the result set, unlike the text protocol.	17 years ago
Andrey Hristov	6c4e8fa4f9	MFH: Fix a bug with mysqlnd_fetch_field(_direct()). With mysqlnd the optimised function was called, which however, doesn't respect that during store the raw data is not unpacked, to be lazy. The data is unpacked to zvals later, during every row fetch. However, this way max_length won't be calculated correctly. So, if a mysqlnd_fetch_field(_direct) call comes we need to unpack everything and then calculate max_length...and that is expensive, defies our lazy unpacking optimisation.	17 years ago
Andrey Hristov	7ae2505cab	Fix a typo, utf8 is 3 byte max, for us, for now Add a another cset	17 years ago
Johannes Schlüter	501ecb10f4	Fix #47535 Compilation failure in ps_fetch_from_1_to_8_bytes()	17 years ago
Felipe Pena	dcd86aebd2	- MFH: Removed UG(unicode) checks	17 years ago
Johannes Schlüter	bcb40030a5	MFH Fix #47819 (Getting pdo_mysql.so: undefined symbol: mysqlnd_debug_init at startup)	17 years ago
Johannes Schlüter	5942f27d86	MFH: mark mysqlnd's threading (which can be used for background pre-fetching) experimental	17 years ago
Andrey Hristov	df7ebb5d41	MFH: Fix the build because the macro name has changed	17 years ago
Andrey Hristov	8f102c1697	MFH: Add charsets from MySQL 6.0 as a structure to be used later	17 years ago
Johannes Schlüter	ebef020ecc	- MFH: Improve mysqlnd's internal method registration	17 years ago
Johannes Schlüter	074b084852	MFH: export mysql_refresh [DOC] This makes mysql_refresh() as described in http://dev.mysql.com/doc/refman/6.0/en/mysql-refresh.html available as mysqli_Refresh() and mysqli->refresh() MYSQLI_REFRESH_BACKUP_LOG is only available when linking libmysql 6.0 or mysqlnd	17 years ago
Sebastian Bergmann	08659c2dcd	MFH: Bump copyright year, 3 of 3.	17 years ago
Johannes Schlüter	836dd82c8e	MFH: No infinite loop in case the connection broke	17 years ago
Andrey Hristov	d43b7b022f	MFH: Windows fixes for mysqlnd Less warnings + PHPAPI for mysqlnd_poll	17 years ago
Andrey Hristov	e62f5f8f03	MFH: Fix a small bug that mysqlnd::next_result didn't care about an error in a multi-statement. In an inner layer the error has been already set, thus it needed better massage on the top level.	17 years ago
Andrey Hristov	3343619b9a	MFH: Fix Windows build, pure C compiler	17 years ago
Andrey Hristov	825034cea5	MFH: Asynchronous queries for mysqli, when mysqlnd is enabled. Includes 4 tests for mysqli_poll	17 years ago
Andrey Hristov	737c3c89c6	MFB: More Windows types cleanup	17 years ago
Andrey Hristov	e49366850f	MFH: Use win32/php_stdint.h for stdint types on Windows	17 years ago
Andrey Hristov	e099d732a6	MFH: More cleanup of config-win.h	17 years ago
Andrey Hristov	9186528994	MFH: Get rid of many defines (which simplifies the code a lot), as well as uint->unsigned int	17 years ago
Andrey Hristov	a836ea7b71	Fix windows build	17 years ago
Andrey Hristov	9726d6b047	Fix the build	18 years ago
Andrey Hristov	77c1e145e1	MFB: 64bit fixes: - fixes to sprintf modifiers, cleaning warnings - use _t types, like uint64_t instead of uint64, thus skipping series of typedefs.	18 years ago
Johannes Schlüter	04d9bbe65e	MFH: Fix #46285 (lastInsertId() returns "0" when a deferenced PDOStatement is executed)	18 years ago

1 2 3

116 Commits (495a5ab0057504b6719ee5be7462db8fa64a46f2)