Don't use binaries cuz I prefer to roll my own... # c++ -v Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/specs gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release) # gcc -v Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs gcc version 2.96 20000731 (Red Hat Linux 7.0) ]# kgcc -v Reading specs from /usr/lib/gcc-lib/i386-glibc21-linux/egcs-2.91.66/specs gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)
Also, which would be better to use, gcc or kgcc? I know that kgcc is supposed to be for kernel compiles, but other source makes good use of it too. I.E. samba. Which SHOULD I be using? -----Original Message----- From: Benjamin Pflugmann [mailto:[EMAIL PROTECTED]] Sent: Thursday, December 06, 2001 12:39 AM To: Gill, Vern Cc: [EMAIL PROTECTED] Subject: Re: SHOW TABLES; show nothing Hi. On Mon, Dec 03, 2001 at 11:41:38PM -0800, [EMAIL PROTECTED] wrote: > These are....I think....pre-built > Dunno for sure. > > # rpm -qa | grep -i glibc > compat-glibc-6.2-2.1.3.2 > glibc-devel-2.2-5 > glibc-common-2.2.4-13 > glibc-2.2.4-13 > # mysql --version > mysql Ver 11.6 Distrib 3.23.28-gamma, for pc-linux-gnu (i586) > # rpm -qa | grep -i gcc > gcc-2.96-69 > kgcc-1.1.2-40 > > Although I can't tell you if this was the case when it was built... It had > to have been a while ago. I have been trying to build a new mysql, however, > if you check the messages a few days ago, I am having compile problems. I.E; > # cd /usr/src/mysql-3.23.46/ > # ./configure --prefix=/usr/local/mysql > --with-unix-socket-path=/usr/local/mysql/lock/mysql.sock --enable-assembler > --with-libwrap --without-berkeley-db \ > --with-ipv6 --without-bench --with-mysqld-ldflags=-all-static > --disable-shared --with-extra-charset=complex --without-mit-threads > --with-other-libc=/lib > > make > ...blah... > ...blah... > Making all in client > make[2]: Entering directory `/src/mysql-3.23.46/client' > c++ -DUNDEF_THREADS_HACK -I./../include What output gives c++ -v >From the output, I don't see whether kgcc or gcc is used. > -I../include -I./.. -I.. -I.. -O3 > -DDBUG_OFF -fno-implicit-templates -fno-exceptions -fno-rtti -fpermissive > -I/lib/include -c mysql.cc > cc1plus: Invalid option `-fpermissive' > In file included from client_priv.h:19, > from mysql.cc:28: > ../include/global.h:680: abstract declarator used as declaration [...] > Any help there? Well, since I never encountered build problems with MySQL myself, I am probably not of much help. The error above occurs in the follwing part of the code: 679 #if !defined(bool) && !defined(bool_defined) && (!defined(HAVE_BOOL) || !defined(__cplusplus)) 680 typedef char bool; /* Ordinary boolean values 0 1 */ 681 #endif And is probably about declaring "bool" when it is already defined. The question is why the #if evaluates to true. I just checked, and for me (RH 6.2, egcs-2.91.66), HAVE_BOOL and __cplusplus are both defined (the first by configure in config.h and the second by gcc itself, because it compiles a C++ source). So, without knowing, my first bet would be that HAVE_BOOL is not set in config.h. If this really is the case, then the question is, why the detection by configure failed. Most time, when such a simple test by configure fails, it is some missing lib or so (i.e. the compile fails not due to the test itself, but other stuff). On the other hand, I never had a configure finish when such problems occured (because too much tests failed with basic stuff going wrong). Hope that helped somehow. Btw, what is the reason that you don't use the official binaries? Bye, Benjamin. -- [EMAIL PROTECTED] --------------------------------------------------------------------- Before posting, please check: http://www.mysql.com/manual.php (the manual) http://lists.mysql.com/ (the list archive) To request this thread, e-mail <[EMAIL PROTECTED]> To unsubscribe, e-mail <[EMAIL PROTECTED]> Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
