View previous topic :: View next topic |
Author |
Message |
n1ese Tux's lil' helper
Joined: 29 Sep 2019 Posts: 82
|
Posted: Tue Apr 14, 2020 6:25 pm Post subject: Compiling MySQL on 2GB Pi 4B? |
|
|
Anyone successful in compiling MySQL on a Pi 4B with 2GB? I had toned down my MAKEOPTS enough to compile most things. Despite the error message, I think I am failing because of memory. According to htop, mem and swap pegged out at 100% just before failing.
Do I try dropping down to -j2 on MAKEOPTS?
Code: | MAKEOPTS="-j3"
EMERGE_DEFAULT_OPTS="--jobs=1" |
Code: | [ 74%] Building CXX object sql/CMakeFiles/sql_main.dir/auth/sql_user_table.cc.o
cd /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql-8.0.19_build/sql && /usr/bin/aarch64-unknown-linux-gnu-g++ -DHAVE_CONFIG_H -DHAVE_LIBEVENT2 -DHAVE_OPENSSL -DHAVE_TLSv13 -DLZ4_DISABLE_DEPRECATE_WARNINGS -DMYSQL_COMPONENT -DMYSQL_SERVER -DRAPIDJSON_NO_SIZETYPEDEFINE -DRAPIDJSON_SCHEMA_USE_INTERNALREGEX=0 -DRAPIDJSON_SCHEMA_USE_STDREGEX=1 -DUNISTR_FROM_CHAR_EXPLICIT=explicit -DUNISTR_FROM_STRING_EXPLICIT=explicit -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE -D_USE_MATH_DEFINES -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -I/var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql-8.0.19_build -I/var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql-8.0.19_build/include -I/var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql -I/var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/include -isystem /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/extra/rapidjson/include -isystem /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/extra/libedit/editline -isystem /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/extra/zstd/lib -isystem /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/include/boost_1_70_0/patches -isystem /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/boost/boost_1_70_0 -march=armv8-a+crc -mtune=cortex-a72 -ftree-vectorize -O2 -pipe -felide-constructors -fno-strict-aliasing -Wall -Wextra -Wformat-security -Wvla -Wundef -Wmissing-format-attribute -Woverloaded-virtual -Wcast-qual -Wimplicit-fallthrough=2 -Wstringop-truncation -Wlogical-op -DDBUG_OFF -DNDEBUG -fPIC -o CMakeFiles/sql_main.dir/auth/sql_user_table.cc.o -c /var/tmp/portage/dev-db/mysql-8.0.19-r1/work/mysql/sql/auth/sql_user_table.cc
{standard input}: Assembler messages:
{standard input}:21473: Warning: end of file not at end of a line; newline inserted
aarch64-unknown-linux-gnu-g++: fatal error: Killed signal terminated program cc1plus
compilation terminated. |
|
|
Back to top |
|
|
n1ese Tux's lil' helper
Joined: 29 Sep 2019 Posts: 82
|
Posted: Tue Apr 14, 2020 9:01 pm Post subject: |
|
|
Looking like it is not possible on a 2GB Pi 4. I reduced MAKEOPTS to-j2 and made it a bit farther in the build but memory and swap are pegged at 100% once again. Load average showing little individual core usage but numbers of 6.81 4.62 3.79. |
|
Back to top |
|
|
Hu Administrator
Joined: 06 Mar 2007 Posts: 22695
|
Posted: Wed Apr 15, 2020 2:06 am Post subject: |
|
|
How much swap do you have configured? |
|
Back to top |
|
|
n1ese Tux's lil' helper
Joined: 29 Sep 2019 Posts: 82
|
Posted: Wed Apr 15, 2020 9:01 am Post subject: |
|
|
Hu wrote: | How much swap do you have configured? |
Sakaki's image is configured for 1GB swap. |
|
Back to top |
|
|
n1ese Tux's lil' helper
Joined: 29 Sep 2019 Posts: 82
|
Posted: Wed Apr 15, 2020 6:12 pm Post subject: |
|
|
Success! Had to back way down and set MAKEOPTS=-j1 and man it took a long time to compile. |
|
Back to top |
|
|
NeddySeagoon Administrator
Joined: 05 Jul 2003 Posts: 54578 Location: 56N 3W
|
Posted: Wed Apr 15, 2020 8:05 pm Post subject: |
|
|
n1ese,
Well done!
I have an arm64 binhost you can use if it helps. _________________ Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail. |
|
Back to top |
|
|
n1ese Tux's lil' helper
Joined: 29 Sep 2019 Posts: 82
|
Posted: Thu Apr 16, 2020 1:12 am Post subject: |
|
|
Neddy, thanks. After all this MySQL was giving me issues so I switched to MariaDB and it works like a champ and compiled much faster. |
|
Back to top |
|
|
|