Your message dated Wed, 25 Jan 2017 02:23:24 +0100
with message-id <b9d9338a-87d7-7f20-a581-2e86177a5...@debian.org>
and subject line Re: rmysql: FTBFS on mips64el
has caused the Debian Bug report #847750,
regarding libmariadb-dev: mariadb_config emits trailing binary bullshit on
mips64el
to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact ow...@bugs.debian.org
immediately.)
--
847750: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=847750
Debian Bug Tracking System
Contact ow...@bugs.debian.org with problems
--- Begin Message ---
Package: libmariadb-dev
Version: 2.3.1-1
Severity: important
Control: affects -1 + src:rmysql
(severity should be serious, will raise after the next britney run)
(sid_mips64el-dchroot)anbe@eller:~$ mariadb_config
Copyright 2011-2015 MariaDB Corporation AB
Get compiler flags for using the MariaDB Connector/C.
Usage: mariadb_config [OPTIONS]
--cflags [-I/usr/include/mariadb -I/usr/include/mariadb/mysql -g -O2
-fdebug-prefix-map=/build/mariadb-connector-c-dopxmU/mariadb-connector-c-2.3.1=.
-fstack-protector-strong -Wformat -Werror=format-security -Wdate-time
-D_FORTIFY_SOURCE=2 -Wunused -Wno-uninitialized]
--include [-I/usr/include/mariadb -I/usr/include/mariadb/mysql]
--libs [-L/usr/lib/mips64el-linux-gnuabi64 -lmariadb]
--libs_r [-L/usr/lib/mips64el-linux-gnuabi64 -lmariadb]
--version [5.5.1]
--socket [/var/run/mysqld/mysqld.sock]
--port [3306]
--plugindir [/usr/lib/mips64el-linux-gnuabi64/mariadb/plugin]
--plugindir [/usr/lib/mips64el-linux-gnuabi64/mariadb/plugin]
ql]
g -O2
-fdebug-prefix-map=/build/mariadb-connector-c-dopxmU/mariadb-connector-c-2.3.1=.
-fstack-protector-strong -Wformat -Werror=format-security -Wdate-time
-D_FORTIFY_SOURCE=2 -Wunused -Wno-uninitialized]
�
Looks like some buffer is printed twice (starting at the second --plugindir)
and this includes all *bytes* from the buffer, ignoring zero termination.
Redirecting this to a file (or piping through hd) does not give the the
duplicate
--plugindir setting, but a lot of \0 bytes (and three non-\0 bytes) at the end.
Running mariadb_config with some arguments emits similar trailing binary crap.
(sid_mips64el-dchroot)anbe@eller:~$ mariadb_config | hd | tail -n 20
00000540 75 73 72 2f 6c 69 62 2f 6d 69 70 73 36 34 65 6c |usr/lib/mips64el|
00000550 2d 6c 69 6e 75 78 2d 67 6e 75 61 62 69 36 34 20 |-linux-gnuabi64 |
00000560 2d 6c 6d 61 72 69 61 64 62 5d 0a 20 20 2d 2d 76 |-lmariadb]. --v|
00000570 65 72 73 69 6f 6e 20 20 20 20 20 20 20 5b 35 2e |ersion [5.|
00000580 35 2e 31 5d 0a 20 20 2d 2d 73 6f 63 6b 65 74 20 |5.1]. --socket |
00000590 20 20 20 20 20 20 20 5b 2f 76 61 72 2f 72 75 6e | [/var/run|
000005a0 2f 6d 79 73 71 6c 64 2f 6d 79 73 71 6c 64 2e 73 |/mysqld/mysqld.s|
000005b0 6f 63 6b 5d 0a 20 20 2d 2d 70 6f 72 74 20 20 20 |ock]. --port |
000005c0 20 20 20 20 20 20 20 5b 33 33 30 36 5d 0a 20 20 | [3306]. |
000005d0 2d 2d 70 6c 75 67 69 6e 64 69 72 20 20 20 20 20 |--plugindir |
000005e0 5b 2f 75 73 72 2f 6c 69 62 2f 6d 69 70 73 36 34 |[/usr/lib/mips64|
000005f0 65 6c 2d 6c 69 6e 75 78 2d 67 6e 75 61 62 69 36 |el-linux-gnuabi6|
00000600 34 2f 6d 61 72 69 61 64 62 2f 70 6c 75 67 69 6e |4/mariadb/plugin|
00000610 5d 0a 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |]...............|
00000620 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
00001310 00 f1 0f 02 00 00 00 00 00 00 00 00 00 00 00 00 |................|
00001320 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
*
00022309
This breaks the rmysql configure script.
While you are it it, please filter out -fdebug-prefix-map=* from the cflags.
Andreas
--- End Message ---
--- Begin Message ---
Version: 2.3.2-2
This seems to be fixed with the latest mariadb-connector-c upload,
although rmysql still segfaults during the build process on a buildd,
while it succeeds on the porterbox.
Andreas
--- End Message ---
