- developers - lists.mariadb.org

[Maria-developers] Can't find messagefile '/usr/share/mysql/errmsg.sys'
by Sylvain Raybaud 24 Jun '15

24 Jun '15

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi, I built mariadb galera cluster 10.0.17 from source. Starting it with mysql_safe fails with: Can't find messagefile '/usr/share/mysql/errmsg.sys' This file does not exist indeed. It seems to not exist at all in mariadb-galera-10.0.17.tar.gz. Is it supposed to be installed separately? Am I missing something? Regards, - -- Sylvain Raybaud www.green-communications.fr -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJVeAYZAAoJEEkkwl4JtJ9yvfoP/jTnvj7V1Ze+E7sFE7Rz4Hsm +4Ga8f9MKCr5VxBZky3wMb8EgWmfcWDHbKqu4httTMH2RS6CCv7aiBeRfrMHMW5I NHl+k9CPUrDs03n4Vg12G2M33IIRfdEBgYf2xZ3sp3kBJ07MUWZPvlVL6ch+Aybn oOVzQ1rbUpxrkX2gj4XZ/wtXmy8vN6s9L3jy+o5HU4JZJcFyp1IpCwCE0OneI81V EiHs8dbpsUstr/32TaQ7gCBN9Hb25i5aVwRCD7UXE6kFt6OD6ifv8mw5KfmdtjHc i/0OotnOOsfj7tgDXKoPOltauGdSADJm2VSp8/nsfd6MR2nxMGZcw/Xbhj5l4FkY jWZf7uRAiKsYlmZEaJlb0WZrCTGSe6pcO82hObhv/vD0O55gKCdOM2rAExRiuxFn IPAuKGAKhzegxXxmmBTQv09tjSY3WTNZfFbDKGDQe4EYmCJ0K14cq37gC+TYI+xb Jr9SI5WBAvVVrCvhRktloEf03jVqXMH3H03DBjxGDB9KRc3Vr6AGhQmg0ixC2ZHC uTbRcx+R5mEn40TVwzLEnsUfiV+0MuJ5YYM5brYYRmXRg4ipwTcts1byKTHpvy69 PrDF1mzStInfex2ntkmw7puX8CQ4uebCLzHWSPODDa1n66D1vaEYCbZV7hEg25dH ox0Iptm4REWGyxgNeQxB =0Ydk -----END PGP SIGNATURE-----

3 4

[Maria-developers] doubt about query cache and write locks
by Roberto Spadim 24 Jun '15

24 Jun '15

hi guys, i'm with a doubt when i execute LOCK TABLE xxx WRITE any query at query cache could be executed without locking (waiting for table lock state) shouldn't the lock table invalidade the table query cache when using a write lock? -- Roberto Spadim

2 2

[Maria-developers] crash - analyze table with histogram
by Roberto Spadim 24 Jun '15

24 Jun '15

Hi guys, i got a crash, could someone help? --- queries: set histogram_size=255,histogram_type=SINGLE_PREC_HB; ANALYZE TABLE 19_org.estoque_itens PERSISTENT FOR ALL; /* Erro SQL (2013): Lost connection to MySQL server during query */ --- log: 150623 12:31:47 [ERROR] mysqld got signal 11 ; This could be because you hit a bug. It is also possible that this binary or one of the libraries it was linked against is corrupt, improperly built, or misconfigured. This error can also be caused by malfunctioning hardware. To report this bug, see http://kb.askmonty.org/en/reporting-bugs We will try our best to scrape up some info that will hopefully help diagnose the problem, but since we have already crashed, something is definitely wrong and this may fail. Server version: 10.0.20-MariaDB-log key_buffer_size=2097152000 read_buffer_size=6291456 max_used_connections=22 max_threads=92 thread_count=9 It is possible that mysqld could use up to key_buffer_size + (read_buffer_size + sort_buffer_size)*max_threads = 3180295 K bytes of memory Hope that's ok; if not, decrease some variables in the equation. Thread pointer: 0x0x7f523b81c008 Attempting backtrace. You can use the following information to find out where mysqld died. If you see no messages after this, something went terribly wrong... stack_bottom = 0x4b1410a0 thread_stack 0x48000 mysys/stacktrace.c:247(my_print_stacktrace)[0xbcb2ee] sql/signal_handler.cc:153(handle_fatal_signal)[0x71fa8c] /lib/libpthread.so.0[0x7f52ccbe98f0] sql/field.h:703(Field::set_notnull(long long))[0x610c70] sql/sql_statistics.cc:533(Stat_table::update_stat())[0x60dd68] sql/sql_admin.cc:759(mysql_admin_table)[0x67adc1] sql/sql_admin.cc:1183(Sql_cmd_analyze_table::execute(THD*))[0x67c66c] sql/sql_parse.cc:5095(mysql_execute_command(THD*))[0x5986c2] sql/sql_parse.cc:6529(mysql_parse(THD*, char*, unsigned int, Parser_state*))[0x59fc86] sql/sql_parse.cc:1310(dispatch_command(enum_server_command, THD*, char*, unsigned int))[0x5a1bb7] sql/sql_parse.cc:999(do_command(THD*))[0x5a22f9] sql/sql_connect.cc:1378(do_handle_one_connection(THD*))[0x66c7d4] sql/sql_connect.cc:1295(handle_one_connection)[0x66c912] /lib/libpthread.so.0[0x7f52ccbe1fe7] /lib/libc.so.6(clone+0x6d)[0x7f52cbb822bd] Trying to get some variables. Some pointers may be invalid and cause the dump to abort. Query (0x7f5190884020): is an invalid pointer Connection ID (thread ID): 304 Status: NOT_KILLED Optimizer switch: index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on,index_merge_sort_intersection=off,engine_condition_pushdown=off,index_condition_pushdown=on,derived_merge=on,derived_with_keys=on,firstmatch=on,loosescan=on,materialization=on,in_to_exists=on,semijoin=on,partial_match_rowid_merge=on,partial_match_table_scan=on,subquery_cache=on,mrr=off,mrr_cost_based=off,mrr_sort_keys=off,outer_join_with_cache=on,semijoin_with_cache=on,join_cache_incremental=on,join_cache_hashed=on,join_cache_bka=on,optimize_join_buffer_size=off,table_elimination=on,extended_keys=on,exists_to_in=on The manual page at http://dev.mysql.com/doc/mysql/en/crashing.html contains information that should help you find out what is causing the crash. -- Roberto Spadim

2 2

Re: [Maria-developers] [Commits] 6050ab6: MDEV-6829 : SELinux/AppArmor policies for Galera server
by Daniel Black 23 Jun '15

23 Jun '15

Nice work. https://mariadb.atlassian.net/browse/MDEV-7637 has some netlink_audit_socket rules that don't appear to be here. Recommend contributing the selinux component to https://github.com/TresysTechnology/refpolicy which distros usually develop their policies from. Does this work for galera multicast? It appears to only allow tcp bind here. note for readme semanage permissive -a mysqld_t - less of a change for enabling just that domain to be permissive. Does any of https://mariadb.com/kb/en/mariadb/what-to-do-if-mariadb-doesnt-start/ need changing? ----- On 18 Jun, 2015, at 11:59 PM, Nirbhay Choubey nirbhay(a)mariadb.com wrote: > revision-id: 6050ab658696925f2a031b901eb398fff65fa92a > parent(s): 9eff9ed5c58e782abf383a52a7e691a55b4798a2 > committer: Nirbhay Choubey > branch nick: 5.5-galera > timestamp: 2015-06-18 09:59:09 -0400 > message: > > MDEV-6829 : SELinux/AppArmor policies for Galera server > > Add SELinux policy and AppArmor profile under policy/. > > --- > policy/apparmor/README | 5 ++ > policy/apparmor/usr.sbin.mysqld | 150 ++++++++++++++++++++++++++++++++++ > policy/apparmor/usr.sbin.mysqld.local | 4 + > policy/selinux/README | 18 ++++ > policy/selinux/mariadb-server.fc | 10 +++ > policy/selinux/mariadb-server.te | 91 +++++++++++++++++++++ > 6 files changed, 278 insertions(+) > > diff --git a/policy/apparmor/README b/policy/apparmor/README > new file mode 100644 > index 0000000..271655f > --- /dev/null > +++ b/policy/apparmor/README > @@ -0,0 +1,5 @@ > +Note: The included AppArmor profiles can be used for MariaDB Galera cluster. > +However, since these profiles had been tested for a limited set of scenarios, > +it is highly recommended to run them in "complain" mode and report any denials > +on mariadb.org/jira. > + > diff --git a/policy/apparmor/usr.sbin.mysqld b/policy/apparmor/usr.sbin.mysqld > new file mode 100644 > index 0000000..307872c > --- /dev/null > +++ b/policy/apparmor/usr.sbin.mysqld > @@ -0,0 +1,150 @@ > +# Last Modified: Fri Mar 1 18:55:47 2013 > +# Based on usr.sbin.mysqld packaged in mysql-server in Ubuntu. > +# This AppArmor profile has been copied under BSD License from > +# Percona XtraDB Cluster, along with some additions. > + > +#include <tunables/global> > + > +/usr/sbin/mysqld flags=(complain) { > + #include <abstractions/base> > + #include <abstractions/mysql> > + #include <abstractions/nameservice> > + #include <abstractions/user-tmp> > + #include <abstractions/winbind> > + > + capability chown, > + capability dac_override, > + capability setgid, > + capability setuid, > + capability sys_rawio, > + capability sys_resource, > + > + network tcp, > + > + /bin/dash rcx, > + /dev/dm-0 r, > + /etc/gai.conf r, > + /etc/group r, > + /etc/hosts.allow r, > + /etc/hosts.deny r, > + /etc/ld.so.cache r, > + /etc/mtab r, > + /etc/my.cnf r, > + /etc/mysql/*.cnf r, > + /etc/mysql/*.pem r, > + /etc/mysql/conf.d/ r, > + /etc/mysql/conf.d/* r, > + /etc/nsswitch.conf r, > + /etc/passwd r, > + /etc/services r, > + /run/mysqld/mysqld.pid w, > + /run/mysqld/mysqld.sock w, > + /sys/devices/system/cpu/ r, > + owner /tmp/** lk, > + /tmp/** rw, > + /usr/lib/mysql/plugin/ r, > + /usr/lib/mysql/plugin/*.so* mr, > + /usr/sbin/mysqld mr, > + /usr/share/mysql/** r, > + /var/lib/mysql/ r, > + /var/lib/mysql/** rwk, > + /var/log/mysql.err rw, > + /var/log/mysql.log rw, > + /var/log/mysql/ r, > + /var/log/mysql/* rw, > + /var/run/mysqld/mysqld.pid w, > + /var/run/mysqld/mysqld.sock w, > + > + > + profile /bin/dash flags=(complain) { > + #include <abstractions/base> > + #include <abstractions/bash> > + #include <abstractions/mysql> > + #include <abstractions/nameservice> > + #include <abstractions/perl> > + > + > + > + /bin/cat rix, > + /bin/dash rix, > + /bin/date rix, > + /bin/grep rix, > + /bin/nc.openbsd rix, > + /bin/netstat rix, > + /bin/ps rix, > + /bin/rm rix, > + /bin/sed rix, > + /bin/sleep rix, > + /bin/tar rix, > + /bin/which rix, > + /dev/tty rw, > + /etc/ld.so.cache r, > + /etc/my.cnf r, > + /proc/ r, > + /proc/*/cmdline r, > + /proc/*/fd/ r, > + /proc/*/net/dev r, > + /proc/*/net/if_inet6 r, > + /proc/*/net/tcp r, > + /proc/*/net/tcp6 r, > + /proc/*/stat r, > + /proc/*/status r, > + /proc/sys/kernel/pid_max r, > + /proc/tty/drivers r, > + /proc/uptime r, > + /proc/version r, > + /sbin/ifconfig rix, > + /sys/devices/system/cpu/ r, > + /tmp/** rw, > + /usr/bin/cut rix, > + /usr/bin/dirname rix, > + /usr/bin/gawk rix, > + /usr/bin/innobackupex rix, > + /usr/bin/mysql rix, > + /usr/bin/perl rix, > + /usr/bin/seq rix, > + /usr/bin/wsrep_sst* rix, > + /usr/bin/wsrep_sst_common r, > + /usr/bin/xtrabackup* rix, > + /var/lib/mysql/ r, > + /var/lib/mysql/** rw, > + /var/lib/mysql/*.log w, > + /var/lib/mysql/*.err w, > + > +# MariaDB additions > + ptrace peer=@{profile_name}, > + > + /bin/hostname rix, > + /bin/ip rix, > + /bin/mktemp rix, > + /bin/ss rix, > + /bin/sync rix, > + /bin/touch rix, > + /bin/uname rix, > + /etc/mysql/*.cnf r, > + /etc/mysql/conf.d/ r, > + /etc/mysql/conf.d/* r, > + /proc/*/attr/current r, > + /proc/*/fdinfo/* r, > + /proc/*/net/* r, > + /proc/locks r, > + /proc/sys/net/ipv4/ip_local_port_range r, > + /run/mysqld/mysqld.sock rw, > + /sbin/ip rix, > + /usr/bin/basename rix, > + /usr/bin/du rix, > + /usr/bin/find rix, > + /usr/bin/lsof rix, > + /usr/bin/my_print_defaults rix, > + /usr/bin/mysqldump rix, > + /usr/bin/pv rix, > + /usr/bin/rsync rix, > + /usr/bin/socat rix, > + /usr/bin/tail rix, > + /usr/bin/timeout rix, > + /usr/bin/xargs rix, > + /usr/bin/xbstream rix, > + } > + # Site-specific additions and overrides. See local/README for details. > + #include <local/usr.sbin.mysqld> > +} > diff --git a/policy/apparmor/usr.sbin.mysqld.local > b/policy/apparmor/usr.sbin.mysqld.local > new file mode 100644 > index 0000000..a0b8a02 > --- /dev/null > +++ b/policy/apparmor/usr.sbin.mysqld.local > @@ -0,0 +1,4 @@ > +# Site-specific additions and overrides for usr.sbin.mysqld.. > +# For more details, please see /etc/apparmor.d/local/README. > +# This AppArmor profile has been copied under BSD License from > +# Percona XtraDB Cluster, along with some additions. > diff --git a/policy/selinux/README b/policy/selinux/README > new file mode 100644 > index 0000000..a8c11c7 > --- /dev/null > +++ b/policy/selinux/README > @@ -0,0 +1,18 @@ > +Note: The included SELinux policy files can be used for MariaDB Galera cluster. > +However, since these policies had been tested for a limited set of scenarios, > +it is highly recommended to run SELinux in "permissive" mode even with these > +policies installed and report any denials on mariadb.org/jira. > + > + > +How to generate and load the policy module of MariaDB Galera cluster ? > + > +* Generate the SELinux policy module. > + # cd <source>/policy/selinux/ > + # make -f /usr/share/selinux/devel/Makefile mariadb-server.pp > + > +* Load the generated policy module. > + # semodule -i /path/to/mariadb-server.pp > + > +* Lastly, run the following command to allow 4568. > + # semanage port -a -t mysqld_port_t -p tcp 4568 > + > diff --git a/policy/selinux/mariadb-server.fc b/policy/selinux/mariadb-server.fc > new file mode 100644 > index 0000000..1a69ecc > --- /dev/null > +++ b/policy/selinux/mariadb-server.fc > @@ -0,0 +1,10 @@ > +# This SELinux file contexts (.fc) file has been copied under BSD License from > +# Percona XtraDB Cluster. > + > +/etc/init\.d/rc\.d/mysql -- > gen_context(system_u:object_r:mysqld_initrc_exec_t,s0) > +/var/lib/mysql/.*\.log -- gen_context(system_u:object_r:mysqld_log_t,s0) > +/var/lib/mysql/.*\.err -- gen_context(system_u:object_r:mysqld_log_t,s0) > +/var/lib/mysql/.*\.pid -- gen_context(system_u:object_r:mysqld_var_run_t,s0) > +/var/lib/mysql/.*\.cnf -- gen_context(system_u:object_r:mysqld_etc_t,s0) > +/usr/bin/xtrabackup.* -- gen_context(system_u:object_r:mysqld_exec_t,s0) > +/usr/bin/wsrep.* -- gen_context(system_u:object_r:mysqld_safe_exec_t,s0) > diff --git a/policy/selinux/mariadb-server.te b/policy/selinux/mariadb-server.te > new file mode 100644 > index 0000000..9c0319c > --- /dev/null > +++ b/policy/selinux/mariadb-server.te > @@ -0,0 +1,91 @@ > +# This SELinux type enforcement (.te) file has been copied under BSD License > +# from Percona XtraDB Cluster, along with some additions. > + > +module mariadb-server 1.0; > + > +require { > + type user_tmp_t; > + type kerberos_port_t; > + type mysqld_safe_t; > + type tmp_t; > + type tmpfs_t; > + type hostname_exec_t; > + type ifconfig_exec_t; > + type sysctl_net_t; > + type proc_net_t; > + type port_t; > + type mysqld_t; > + type var_lib_t; > + type rsync_exec_t; > + type bin_t; > + type shell_exec_t; > + type anon_inodefs_t; > + type fixed_disk_device_t; > + class lnk_file read; > + class process { getattr signull }; > + class unix_stream_socket connectto; > + class capability { sys_resource sys_nice }; > + class tcp_socket { name_bind name_connect }; > + class file { execute setattr read create getattr execute_no_trans write ioctl > open append unlink }; > + class sock_file { create unlink getattr }; > + class blk_file { read write open }; > + class dir { write search getattr add_name read remove_name open }; > + > +# MariaDB additions > + type tram_port_t; > + class process setpgid; > + class netlink_tcpdiag_socket { create nlmsg_read }; > +} > + > + > +#============= mysqld_safe_t ============== > +allow mysqld_safe_t mysqld_t:process signull; > +allow mysqld_safe_t self:capability { sys_resource sys_nice }; > +allow mysqld_safe_t tmp_t:file { create read write open getattr unlink ioctl > setattr }; > +allow mysqld_safe_t tmp_t:dir { write remove_name add_name }; > +allow mysqld_safe_t tmp_t:sock_file { getattr unlink }; > +allow mysqld_safe_t user_tmp_t:sock_file { getattr unlink }; > +allow mysqld_safe_t var_lib_t:dir { write add_name }; > +allow mysqld_safe_t var_lib_t:file { write ioctl setattr create open getattr > append unlink }; > + > +#============= mysqld_t ============== > +allow mysqld_t anon_inodefs_t:file write; > +allow mysqld_t tmp_t:sock_file { create unlink }; > +allow mysqld_t tmpfs_t:dir { write search read remove_name open add_name }; > +allow mysqld_t tmpfs_t:file { write getattr read create unlink open }; > +allow mysqld_t fixed_disk_device_t:blk_file { read write open }; > +allow mysqld_t ifconfig_exec_t:file { read execute open execute_no_trans > getattr }; > + > +#This rule allows connecting on 4444 > +allow mysqld_t kerberos_port_t:tcp_socket { name_bind name_connect }; > + > +allow mysqld_t mysqld_safe_t:dir { getattr search }; > +allow mysqld_t mysqld_safe_t:file { read open }; > +allow mysqld_t self:unix_stream_socket connectto; > +allow mysqld_t port_t:tcp_socket { name_bind name_connect }; > +allow mysqld_t proc_net_t:file { read getattr open }; > +allow mysqld_t sysctl_net_t:dir search; > +allow mysqld_t var_lib_t:file { getattr open append }; > +allow mysqld_t var_lib_t:sock_file { create unlink getattr }; > +allow mysqld_t rsync_exec_t:file { read getattr open execute execute_no_trans > }; > +allow mysqld_t self:process getattr; > +allow mysqld_t hostname_exec_t:file { read getattr execute open > execute_no_trans }; > +allow mysqld_t user_tmp_t:dir { write add_name }; > +allow mysqld_t user_tmp_t:file create; > +allow mysqld_t bin_t:lnk_file read; > +allow mysqld_t tmp_t:file { append create read write open getattr unlink > setattr }; > + > +# Allows too much leeway - the xtrabackup/wsrep rules in fc should fix it, but > +# keep for the moment. > +allow mysqld_t shell_exec_t:file { execute_no_trans getattr read execute open > }; > +allow mysqld_t bin_t:file { getattr read execute open execute_no_trans ioctl }; > + > +# MariaDB additions > +allow mysqld_t self:process setpgid; > +# This rule allows port 4567 > +allow mysqld_t tram_port_t:tcp_socket name_bind; > + > +# Rules related to XtraBackup > +allow mysqld_t self:netlink_tcpdiag_socket { create nlmsg_read }; > +allow mysqld_t sysctl_net_t:file { read getattr open }; > + > _______________________________________________ > commits mailing list > commits(a)mariadb.org > https://lists.askmonty.org/cgi-bin/mailman/listinfo/commits -- -- Daniel Black, Engineer @ Open Query (http://openquery.com.au) Remote expertise & maintenance for MySQL/MariaDB server environments.

2 1

[Maria-developers] Likely a dead optimizer related code
by Alexander Barkov 23 Jun '15

23 Jun '15

Hi Igor, I noticed that during Item_cond::fix_fields() and Item_func_between::fix_fields() update these optimizer related SELECT_LEX members: thd->lex->current_select->cond_count and thd->lex->current_select->between_count The purpose of these members is to allocate optimizer related buffers in: - update_ref_and_keys() in sql_select.cc and - check_func_dependency() in opt_table_elimination,cc and it seems they have no any other purposes.. From my understanding, this is a dead code and the collected values are never used: - cond_count and_between_count are later always initialized to 0 before the walk(count_sargable_conds) call in update_ref_and_keys(). - moreover, using the values collected during fix_fields() would not be correct to allocate optimized related buffers, because the code in fix_fields() does not distinguish between different query parts, so the items can be just a part of the SELECT list rather than a part of WHERE/HAVING/ON. Is the attached patch correct? All tests pass. Thanks.

1 1

Re: [Maria-developers] [Commits] d3b7eb7: MDEV-7528 GIS: Functions return NULL instead of specified -1 for NULL arguments.
by Sergei Golubchik 23 Jun '15

23 Jun '15

Hi, Holyfoot! On Jun 23, holyfoot(a)askmonty.org wrote: > revision-id: d3b7eb7b99ef7dae79c20b5beb63acdfdf6d8046 > parent(s): 3e4126e9d6b21d2f330a365b0847999346485863 > committer: Alexey Botchkov > branch nick: server > timestamp: 2015-06-23 11:57:05 +0500 > message: > > MDEV-7528 GIS: Functions return NULL instead of specified -1 for NULL arguments. > The behaviour required by the standard seems too weird to expect. If the function can never return NULL, please set maybe_null accordingly. Regards, Sergei

1 0

[Maria-developers] GSOC [Adding Redis interface to XtraDB]
by Aalekh Nigam 22 Jun '15

22 Jun '15

Hello, I am Aalekh Nigam, working as a GSOC Summer Strudent for the project "Adding Redis interace to XtraDB". Please follow my blog, for weekly update on my project here: http://mimdl.tumblr.com/ Thank you Aalekh

1 0

[Maria-developers] MXS-2: Week 4
by sriram patil 22 Jun '15

22 Jun '15

Hi All, Please find the link to my blog post of fourth week below. <http://srirampatil.tumblr.com/post/122086304290/mxs-2-week-4> MXS-2: Week 4 <http://srirampatil.tumblr.com/post/122086304290/mxs-2-week-4> Feel free to point out mistakes, if any. Also, suggestions ans/or reviews are welcome! Thanks, Sriram

1 0

Re: [Maria-developers] [Commits] 95a99d8: MDEV-8303; Dumping buffer pool noisy in the logs.
by Daniel Black 20 Jun '15

20 Jun '15

Will only get used if documented: https://mariadb.com/kb/en/mariadb/xtradbinnodb-server-system-variables/ + https://mariadb.com/kb/en/mariadb/server-system-variables/ ----- On 18 Jun, 2015, at 6:45 PM, Jan Lindström jan.lindstrom(a)mariadb.com wrote: > revision-id: 95a99d8dd96be43369d5be9cc23735a1ce86674a > parent(s): b94eaff8d90a60b9dea3b5b68fbde3d35563127b > committer: Jan Lindström > branch nick: 10.1-encryption > timestamp: 2015-06-18 08:38:33 +0300 > message: > > MDEV-8303; Dumping buffer pool noisy in the logs. > > Added new dynamic configuration variable innodb_buf_dump_status_frequency > to configure how often buffer pool dump status is printed in the logs. > A number between [0, 100] that tells how oftern buffer pool dump status > in percentages should be printed. E.g. 10 means that buffer pool dump > status is printed when every 10% of number of buffer pool pages are > dumped. Default is 0 (only start and end status is printed). -- -- Daniel Black, Engineer @ Open Query (http://openquery.com.au) Remote expertise & maintenance for MySQL/MariaDB server environments.

2 1

Re: [Maria-developers] [Commits] d1f5dfe: MDEV-8030 - Apc_target::disable() locks mutex twice
by Sergey Petrunia 20 Jun '15

20 Jun '15

Hi Sergei, Ok to push. I'm sorry for the delay with this. On Wed, May 20, 2015 at 08:09:36AM +0000, svoj(a)mariadb.org wrote: > revision-id: d1f5dfe67703a22f017eb25dd48b1eb5246a9c54 > parent(s): 80333ad847d9f1708fa02fb47a976e96f014bc50 > committer: Sergey Vojtovich > branch nick: mariadb > timestamp: 2015-05-20 12:04:32 +0400 > message: > > MDEV-8030 - Apc_target::disable() locks mutex twice > > Moved Apc_target::destroy(), Apc_target::enable() and Apc_targe::disable() > definitions to my_apc.h so that they can be inlined. > > Apc_targe::disable() now calls Apc_target::process_apc_requests() only if > there're APC requests. This saves one pthread_mutex_lock() call. > > Overhead change: > Apc_target::disable 0.04% -> out of radar > Apc_target::enable 0.03% -> out of radar > Apc_target::process_apc_requests 0.02% -> out of radar > pthread_mutex_lock 0.43% -> 0.42% > pthread_mutex_unlock 0.26% -> 0.25% > > --- > sql/my_apc.cc | 39 --------------------------------------- > sql/my_apc.h | 27 +++++++++++++++++++++++---- > 2 files changed, 23 insertions(+), 43 deletions(-) > > diff --git a/sql/my_apc.cc b/sql/my_apc.cc > index 1766068..91f5cd3 100644 > --- a/sql/my_apc.cc > +++ b/sql/my_apc.cc > @@ -41,45 +41,6 @@ void Apc_target::init(mysql_mutex_t *target_mutex) > } > > > -/* > - Destroy the target. The target must be disabled when this call is made. > -*/ > -void Apc_target::destroy() > -{ > - DBUG_ASSERT(!enabled); > -} > - > - > -/* > - Enter ther state where the target is available for serving APC requests > -*/ > -void Apc_target::enable() > -{ > - /* Ok to do without getting/releasing the mutex: */ > - enabled++; > -} > - > - > -/* > - Make the target unavailable for serving APC requests. > - > - @note > - This call will serve all requests that were already enqueued > -*/ > - > -void Apc_target::disable() > -{ > - bool process= FALSE; > - DBUG_ASSERT(enabled); > - mysql_mutex_lock(LOCK_thd_data_ptr); > - if (!(--enabled)) > - process= TRUE; > - mysql_mutex_unlock(LOCK_thd_data_ptr); > - if (process) > - process_apc_requests(); > -} > - > - > /* [internal] Put request qe into the request list */ > > void Apc_target::enqueue_request(Call_request *qe) > diff --git a/sql/my_apc.h b/sql/my_apc.h > index dfeef5e..20b1ee4 100644 > --- a/sql/my_apc.h > +++ b/sql/my_apc.h > @@ -50,10 +50,29 @@ class Apc_target > ~Apc_target() { DBUG_ASSERT(!enabled && !apc_calls);} > > void init(mysql_mutex_t *target_mutex); > - void destroy(); > - void enable(); > - void disable(); > - > + > + /* Destroy the target. The target must be disabled when this call is made. */ > + void destroy() { DBUG_ASSERT(!enabled); } > + > + /* Enter ther state where the target is available for serving APC requests */ > + void enable() { enabled++; } > + > + /* > + Make the target unavailable for serving APC requests. > + > + @note > + This call will serve all requests that were already enqueued > + */ > + void disable() > + { > + DBUG_ASSERT(enabled); > + mysql_mutex_lock(LOCK_thd_data_ptr); > + bool process= !--enabled && have_apc_requests(); > + mysql_mutex_unlock(LOCK_thd_data_ptr); > + if (unlikely(process)) > + process_apc_requests(); > + } > + > void process_apc_requests(); > /* > A lightweight function, intended to be used in frequent checks like this: > _______________________________________________ > commits mailing list > commits(a)mariadb.org > https://lists.askmonty.org/cgi-bin/mailman/listinfo/commits -- BR Sergei -- Sergei Petrunia, Software Developer MariaDB Corporation | Skype: sergefp | Blog: http://s.petrunia.net/blog

1 0