Re: library/3179: bug report

The following reply was made to PR library/3179; it has been noted by GNATS.

From: "Terry" <kewl@compfort.pl>
To: <marc@snafu.org>
Cc: <gnats@openbsd.org>
Subject: RE: library/3179: bug report
Date: Tue, 29 Apr 2003 12:49:46 +0200

 hi,

 > Hmmmm.   Using the port on OpenBSD -current a "make regress" results
 > in all tests passing except the "join" test which has failed for the
 > last several releases.   The failure is such that the test expects
 > a different error code than that being returned.   The port is
 > version 3.23.55.
 

 ive cvs'ed -current tree and havent seen any changes to pthread code,  so i guess it doesnt really matter much whether the tests were run  on -current or 3.3 ?  

 what do you mean by 'make regress' ? i havent found any 'regress' keyword  in any makefiles. i ran 'sh mysql-test-run --skip-innodb' in 'mysql-test' directory.  

 >
 > How was .56 built?
 using following flags and configure options:  

 CFLAGS="-pipe -march=pentium -funroll-loops"  CXXFLAGS="${CFLAGS} -felide-constructors -fno-exceptions -fno-rtti"  

 export CFLAGS CXXFLAGS CPPFLAGS LDFLAGS CC CXX  ./configure --prefix=/mysql --without-perl --without-bench --disable-static --without-docs --with-charset=latin2 --enable-thread-safe-client --enable-libwrap    

Do you need help?

In addition to our free technical library, Pantek provides professional IT services to companies who utilize Open Source software: Learn more about Pantek's Technical Support offerings Place an order for 24/7/365 Technical Support here Learn more about Pantek, who we are, and what we do Chat with a Live Operator Related openbsd.org Links advocacy announce bugs ipv6 misc ports ports-changes security-announce smp source-changes tech Request more information about Pantek

 after your last mail i also tried to do all the tests using ports' version of mysql - the results were the same:  pid 3214: Fatal error '_pq_first: pq_active' at line 219 in file /usr/src/lib/libpthread/uthread/uthread_priority_queue.c (errno = 35)  

 and there is no specific test on which mysql crashes.  

 here is the gdb's backtrace of mysqld:
#0 0x402acfcf in _thread_sys_kill ()
#1 0x401d43b8 in _thread_exit ()
#2 0x401db2f6 in _pq_first ()
#3 0x401d7aea in _thread_kern_poll ()
#4 0x401d74fe in _thread_kern_sched ()
#5 0x401d881c in _thread_sig_handler ()
#6 0x401a5004 in ?? ()
#7 0x401ccc53 in pthread_kill ()
#8 0x150a50 in thr_alarm (alrm=0x320da8, sec=28800, alarm_data=0x320dac) at thr_alarm.c:177
#9 0x4acb6 in my_real_read (net=0x1ec00c, complen=0x320de8) at net_serv.cc:507
#10 0x4afa9 in my_net_read (net=0x1ec00c) at net_serv.cc:639
#11 0x5e9d2 in do_command (thd=0x1ec000) at sql_parse.cc:739
#12 0x5e150 in handle_one_connection (arg=0x1ec000) at sql_parse.cc:558
#13 0x401ce13d in _thread_start ()
#14 0x1f in ?? ()

 Cannot access memory at address 0xffffffff.  

 best regards,
 terry Received on Tue Apr 29 07:05:09 2003