Hi! At 10:38 PM 10/23/01 -0700, you wrote: >On Tue, Oct 23, 2001 at 01:43:32PM +0300, Heikki Tuuri wrote: >> Hi! >> >> Thank you for a detailed bug report. This bug is already fixed in >> the source release 3.23.43b available from http://www.innodb.com > >Is there now a test case for it in the MySQL test suite, too?
Sorry, not yet. I test InnoDB with a bunch of multithreaded randomized stress tests which I have written in Perl. The idea behind MySQL's standard tests is that they prevent an old bug from reappearing when source code is changed. But my experience is that old bugs do not reappear. It is completely new bugs which are introduced into the source code when new code is written. Randomized stress tests will catch some of new bugs. There is a lot of validation and assertion code when you compile InnoDB with the debug options switched on in /innobase/include/univ.i. Then InnoDB spends 99 % of CPU time running the validations. These validations will also catch some of new bugs. For example, if UNIV_SYNC_DEBUG is defined, then InnoDB checks the latching order of semaphores during execution. Each semaphore is assigned a rank, and the semaphores must be acquired in the order of a descending rank. >Jeremy >-- >Jeremy D. Zawodny, <[EMAIL PROTECTED]> >Technical Yahoo - Yahoo Finance >Desk: (408) 349-7878 Fax: (408) 349-5454 Cell: (408) 685-5936 > >MySQL 3.23.41-max: up 48 days, processed 1,051,760,607 queries (253/sec. avg) Regards, Heikki http://www.innodb.com --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
