> > make  all-recursive
> > /bin/bash: make: command not found
> > gmake: *** [all] Error 127
> >
> 
> This is bizarre. So you're failing at the very top of the all: target 
> where it recursively invokes the all-recursive target by trying to 
> call make. But gmake seems to be confused about which make it is. 
> According to this: http://www.gnu.org/software/make/manual/html_mono/ 
> make.html#SEC59, and my experience, it should know how to reinvoke 
> itself properly.
> 
> Do you have the variable MAKE set in your environment? That could 
> totally confused gmake, I'd think. Or are you invoking gmake as a 
> shell alias, or something?

No, I don't have it set when that happens, but you can see from the next 
comment, that it works when I do!

> > It's not problem if I create a symlink to gmake and call it "make", 
> > or I
> > export MAKE="gmake", but I don't think users should have to do this.
> >
> 
> You shouldn't need to do this. And in fact I build the code on 
> svn.apache.org, wherein I need to invoke it as gmake, and it works 
> fine.

I agree, but it's happening on all 3 Solaris 10 x86 machines, and it works 
fine on AIX, where I also have to invoke it as gmake. By the way, I do 
have access to several Sun-provided Solaris 10 x86 machines, if you'd like 
to do some investigation.

> > Again, it works if create a symlink to gar and call it "ar", or I 
> > export
> > AR="GAR".  I have the same concerns about this as I do with the make
> > executable issue.
> >
> 
> What does the configure script say? ("checking for ar...")

checking for objdir... .libs
checking for ar... false
checking for ranlib... :

which explain why I see the following during the make:

/bin/bash ../libtool --tag=CC --mode=link gcc  -g -O2   -o libcompat.la 
xerces_compat.lo stricmp.lo strnicmp.lo -lnsl -lsocket  -L/usr/lib -licuuc 
-licudata 
false cru .libs/libcompat.a .libs/xerces_compat.o .libs/stricmp.o 
.libs/strnicmp.o

For whatever reason, it thinks "false" is the name of the archiver.

> > Compiling xercesc/util/XMLWin1252Transcoder.cpp
> > g++: unrecognized option `-pthread'
> >
> 
> What does the configure script say in this section:
> 
> checking for the pthreads library -lpthreads... no
> checking whether pthreads work without any flags... yes
> checking for joinable pthread attribute... PTHREAD_CREATE_JOINABLE
> checking if more special flags are required for pthreads... 
-D_THREAD_SAFE

checking whether the compiler implements L"widestring"... yes
checking whether pthreads work with -pthread... yes
checking for joinable pthread attribute... PTHREAD_CREATE_JOINABLE
checking if more special flags are required for pthreads... -D_REENTRANT

Note there is no line like this:

checking for the pthreads library -lpthreads... no

And from config.log:

configure:23694: checking whether pthreads work with -pthread
configure:23780: gcc -o conftest -g -O2 -pthread   conftest.c  -lnsl 
-lsocket  >&5
gcc: unrecognized option `-pthread'

> > I have no idea if this is an issue with my configuration, or with 
> > the new
> > autoconf implementation.
> >
> 
> I don't know why you're getting multiple inclusions. Do you see 
> multiple inclusions in the command invocation?
> 

Here's the command line that causes the warning:

g++ -g -O2 -D_REENTRANT -pthread -o .libs/CreateDOMDocument 
src/CreateDOMDocument/CreateDOMDocument.o  ../obj/.libs/libxerces.so 
-L/usr/lib 
-L/export/ws1/s10-release/s10_ccd_03/usr/src/cmd/gcc-3.4/gcc-3.4.2-bin/i386-pc-solaris2.10/libstdc++-v3/src
 
-L/export/ws1/s10-release/s10_ccd_03/usr/src/cmd/gcc-3.4/gcc-3.4.2-bin/i386-pc-solaris2.10/libstdc++-v3/src/.libs
 
-L/export/ws1/s10-release/s10_ccd_03/usr/src/cmd/gcc-3.4/gcc-3.4.2-bin/gcc 
/opt/sfw/lib/libstdc++.so -lnsl -lsocket -licuuc -licudata -Wl,-R 
-Wl,/usr/local/lib -Wl,-R -Wl,/opt/sfw/lib
g++: unrecognized option `-pthread'
ld: warning: file 
/opt/sfw/lib/gcc/i386-pc-solaris2.10/3.4.2/../../../libstdc++.so: linked 
to /opt/sfw/lib/libstdc++.so: attempted multiple inclusion of file

Thanks!

Dave

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to