[MDEV-652] LP:611586 - mysql-test crashed on maria.maria-recover with icc compiled binary Created: 2010-07-30  Updated: 2015-05-01  Resolved: 2015-05-01

Status: Closed
Project: MariaDB Server
Component/s: Compiling
Affects Version/s: 5.1.67
Fix Version/s: N/A

Type: Bug Priority: Minor
Reporter: bash99 Assignee: Unassigned
Resolution: Won't Fix Votes: 0
Labels: Launchpad

Attachments: XML File LPexportBug611586.xml    

 Description   

Test fail with signal 11, configure option is

./configure '--prefix=/usr/local/mysql' '--localstatedir=/usr/local/mysql/data'
 
'--with-comment=(MariaDB - http://mariadb.com/)' '--with-system-type=Linux'
 
'--with-machine-type=x86_64' '--disable-shared' '--without-embedded-server'
 
'--without-bench' '--enable-assembler' '--enable-static'
 
'--with-client-ldflags=-static' '--with-mysqld-ldflags=-static'
 
'--enable-thread-safe-client' '--enable-local-infile' '--with-big-tables'
 
'--without-docs' '--with-extra-charsets=complex' '--with-ssl'
 
'--with-readline' '--with-libevent' '--with-zlib-dir=bundled'
 
'--with-partition' '--with-plugins=max-no-ndb' '--with-plugin-maria'
 
'--with-maria-tmp-tables' '--without-plugin-innodb'
 
'--without-plugin-example' '--without-plugin-ibmdb2i'
 
'--without-plugin-pbxt' 'CC=icc -static-intel -static-libgcc'
 
'CFLAGS=-O3 -unroll2 -ip -fp-model source -restrict'
 
'CXX=icpc -static-intel -static-libgcc' 'CXXFLAGS=-O3 -unroll2 -ip -fp-model source -restrict'

change -O3 to -O still fails

maria.maria-recover                      [ fail ]
 
        Test ended at 2010-07-30 15:28:04
 

CURRENT_TEST: maria.maria-recover
 
mysqltest: At line 63: query 'select * from t_corrupted2' failed: 2013: Lost connection to MySQL server during query

The result from queries just before the failure was:

< snip >
 
OFF
 
set global maria_recover=normal;
 
select @@global.maria_recover;
 
@@global.maria_recover
 
NORMAL
 
drop database if exists mysqltest;
 
create database mysqltest;
 
use mysqltest;
 
create table t1 (a varchar(1000), index(a)) engine=maria;
 
insert into t1 values("
 

Server [mysqld.1 - pid: 16160, winpid: 16160, exit: 256] failed during test run
 
Server log from this test:
 
100730 15:28:04 [Note] Plugin 'InnoDB' is disabled.
 
100730 15:28:04 [Warning] /usr/local/src/mariadb-5.1.47/sql/mysqld: unknown variable 'loose-debug-sync-timeout=300'
 
100730 15:28:04 [Note] Event Scheduler: Loaded 0 events
 
100730 15:28:04 [Note] /usr/local/src/mariadb-5.1.47/sql/mysqld: ready for connections.
 
Version: '5.1.47-MariaDB-log'  socket: '/usr/local/src/mariadb-5.1.47/mysql-test/var/tmp/mysqld.1.sock'  port: 13000  (MariaDB - http://mariadb.com/)
 
100730 15:28:04 [ERROR] mysqld: Table './mysqltest/t_corrupted2' is marked as crashed and should be repaired
 
 
 
 - saving '/usr/local/src/mariadb-5.1.47/mysql-test/var/log/maria.maria-recover/' to '/usr/local/src/mariadb-5.1.47/mysql-test/var/log/maria.maria-recover/'
 
 - found 'core.16164' (0/5)
 
 
 
Trying 'dbx' to get a backtrace
 
 
 
Trying 'gdb' to get a backtrace
 
Core generated by '/usr/local/src/mariadb-5.1.47/sql/mysqld'
 
Output from gdb follows. The first stack trace is from the failing thread.
 
The following stack traces are from all threads (so the failing one is
 
duplicated).
 
--------------------------
 
[New Thread 16167]
 
[New Thread 16166]
 
[New Thread 16164]
 
Core was generated by `/usr/local/src/mariadb-5.1.47/sql/mysqld --defaults-group-suffix=.1 --defaults-'.
 
Program terminated with signal 11, Segmentation fault.
 
#0  0x00000000008e0be3 in ha_maria::check_and_repair(THD*) ()
 
#0  0x00000000008e0be3 in ha_maria::check_and_repair(THD*) ()
 
#1  0x00000000007797e0 in handler::ha_check_and_repair(THD*) ()
 
#2  0x000000000068b19e in open_unireg_entry(THD*, st_table*, TABLE_LIST*, char const*, char*, unsigned int, st_mem_root*, unsigned int) ()
 
#3  0x000000000068a893 in open_table(THD*, TABLE_LIST*, st_mem_root*, bool*, unsigned int) ()
 
#4  0x000000000068cdb4 in open_tables(THD*, TABLE_LIST**, unsigned int*, unsigned int) ()
 
#5  0x000000000068bb15 in open_and_lock_tables_derived(THD*, TABLE_LIST*, bool) ()
 
#6  0x000000000064b0d8 in execute_sqlcom_select(THD*, TABLE_LIST*) ()
 
#7  0x00000000006451a5 in mysql_execute_command(THD*) ()
 
#8  0x000000000064b6eb in mysql_parse(THD*, char const*, unsigned int, char const**) ()
 
#9  0x000000000063fa83 in dispatch_command(enum_server_command, THD*, char*, unsigned int) ()
 
#10 0x000000000063e644 in do_command(THD*) ()
 
#11 0x00000000006390ba in handle_one_connection ()
 
#12 0x00002aaaaaee673d in start_thread () from /lib64/libpthread.so.0
 
#13 0x00002aaaabfadd1d in clone () from /lib64/libc.so.6
 
 
 
Thread 4 (Thread 16164):
 
#0  0x00002aaaabfa6fc2 in select () from /lib64/libc.so.6
 
#1  0x000000000062bde3 in main ()
 
 
 
Thread 3 (Thread 16166):
 
#0  0x00002aaaaaeeb150 in pthread_cond_timedwait@@GLIBC_2.3.2 () from /lib64/libpthread.so.0
 
#1  0x000000000094dd36 in ma_checkpoint_background ()
 
#2  0x00002aaaaaee673d in start_thread () from /lib64/libpthread.so.0
 
#3  0x00002aaaabfadd1d in clone () from /lib64/libc.so.6
 
 
 
Thread 2 (Thread 16167):
 
#0  0x00002aaaaaeee838 in do_sigwait () from /lib64/libpthread.so.0
 
#1  0x00002aaaaaeee8dd in sigwait () from /lib64/libpthread.so.0
 
#2  0x000000000062f220 in signal_hand ()
 
#3  0x00002aaaaaee673d in start_thread () from /lib64/libpthread.so.0
 
#4  0x00002aaaabfadd1d in clone () from /lib64/libc.so.6
 
 
 
Thread 1 (Thread 16180):
 
#0  0x00000000008e0be3 in ha_maria::check_and_repair(THD*) ()
 
#1  0x00000000007797e0 in handler::ha_check_and_repair(THD*) ()
 
#2  0x000000000068b19e in open_unireg_entry(THD*, st_table*, TABLE_LIST*, char const*, char*, unsigned int, st_mem_root*, unsigned int) ()
 
#3  0x000000000068a893 in open_table(THD*, TABLE_LIST*, st_mem_root*, bool*, unsigned int) ()
 
#4  0x000000000068cdb4 in open_tables(THD*, TABLE_LIST**, unsigned int*, unsigned int) ()
 
#5  0x000000000068bb15 in open_and_lock_tables_derived(THD*, TABLE_LIST*, bool) ()
 
#6  0x000000000064b0d8 in execute_sqlcom_select(THD*, TABLE_LIST*) ()
 
#7  0x00000000006451a5 in mysql_execute_command(THD*) ()
 
#8  0x000000000064b6eb in mysql_parse(THD*, char const*, unsigned int, char const**) ()
 
#9  0x000000000063fa83 in dispatch_command(enum_server_command, THD*, char*, unsigned int) ()
 
#10 0x000000000063e644 in do_command(THD*) ()
 
#11 0x00000000006390ba in handle_one_connection ()
 
#12 0x00002aaaaaee673d in start_thread () from /lib64/libpthread.so.0
 
#13 0x00002aaaabfadd1d in clone () from /lib64/libc.so.6



 Comments   
Comment by Rasmus Johansson (Inactive) [ 2010-07-30 ]

Re: mysql-test crashed on maria.maria-recover with icc compiled binary
with icc optimize options sets to -O0 -fp-model source -restrict, test passed.

verified on icc 10.1.025 and 11.1.072

Comment by Rasmus Johansson (Inactive) [ 2010-07-30 ]

Re: mysql-test crashed on maria.maria-recover with icc compiled binary
output from a debug binary

Core was generated by `/usr/local/src/mariadb-5.1.47/sql/mysqld -defaults-group-suffix=.1 --defaults'.
Program terminated with signal 11, Segmentation fault.
#0 0x00000000008dfa59 in ha_maria::check_and_repair (this=0x40a8b2e0, thd=0x1d93bab0, $6=<value optimized out>, $7=<value optimized out>) at ha_maria.cc:1938
1938 check_opt.init();
#0 0x00000000008dfa59 in ha_maria::check_and_repair (this=0x40a8b2e0, thd=0x1d93bab0, $6=<value optimized out>, $7=<value optimized out>) at ha_maria.cc:1938
#1 0x00000000007790ae in handler::ha_check_and_repair (this=0x40a8b2e0, thd=0x1d93bab0, $?=<value optimized out>, $?=<value optimized out>) at handler.cc:3274

btw, with -O1 -ip -unroll2 compiled ha_maria.o and other compiled with -O3 -ip -unroll2, test is also passed.

Comment by Rasmus Johansson (Inactive) [ 2010-07-30 ]

Re: mysql-test crashed on maria.maria-recover with icc compiled binary
it's most like a compiler bug, I've try a drity way(tm) :

split function ha_maria::check_and_repair to a separate source file, add it to Makefile and recompile.
everything works fine.

Comment by Rasmus Johansson (Inactive) [ 2010-11-30 ]

Launchpad bug id: 611586

Comment by Elena Stepanova [ 2015-05-01 ]

I don't expect it's going to be fixed (and probably not relevant anymore anyway).
If we get new complaints, we can always reconsider.

Generated at Thu Feb 08 06:30:20 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.