This is the mail archive of the
mailing list for the Cygwin project.
Re: php-mysql-cygwin how to
> 1) Underscore mismatches when using the x86 optimized assembler
> implementation of string functions. Workaround: --disable-assembler
> 2) Missing $(EXEEXT) causes make to invoke an implicit rule, generating a
> huge amount of nonsensical errors. Patch:
> --- mysql-4.0.13/sql/Makefile.am.orig 2003-06-17 12:24:57.000000000 +0100
> +++ mysql-4.0.13/sql/Makefile.am 2003-06-17 12:25:10.000000000 +0100
> @@ -114,8 +114,8 @@
> $(CXXCOMPILE) $(LM_CFLAGS) -c $<
> lex_hash.h: lex.h gen_lex_hash.cc sql_yacc.h
> - $(MAKE) gen_lex_hash
> - ./gen_lex_hash > $@
> + $(MAKE) gen_lex_hash$(EXEEXT)
> + ./gen_lex_hash$(EXEEXT) > $@
> # Hack to ensure that lex_hash.h is built early
> sql_lex.o: lex_hash.h
This one is really the most annoying one...
> 3) The long standing server problem. It seems like the complex C++ is
> confusing gcc?
No, it is a problem with the Cygwin/MySQL socket code. Nothing works
with 1.3.22, similar as it is with Apache. The older versions, e.g.
1.3.10 works better, but still not without crashes.
After Corinna fixed the code, so Apache is running without problems on
1.5 now, MySQL still crashes randomly (same behaviour as with 1.3.10),
maybe it crashes more often as before, well it is completely unstable
and nor usable as database server.
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html