Re: [Reduce-algebra-developers] build problems on mac os x 10.6 (snow leopard)

[mark b. <mbr...@ai...>]

On Jan 8, 2012, at 10:35 AM, Arthur Norman wrote:

> You are leaving me very confused about whether you are looking at a CSL 
> or a PSL build of Reduce. Please note that I am probably the main contact 
> for CSL stuff, but when it gets down to the innards of PSL it will be 
> others.

I'm trying to make sense of this myself. Please understand that I have very little
knowledge of C, so what I thought would be a simple macports tcl script to install 
your software, has turned out to be anything but :-) 

I set up the macports install script similar to the debian rules and install both csl and psl, i.e.

system -W ${worksrcpath} "./configure --prefix=${prefix} --with-csl ; ./configure --with-psl"

make

I have a lion (10.7) machine and was able to install both. On the 10.6 machine both failed
(Xport needs to compile first before downloading the psl binaries). I suspect as a result of that script.

> 
>> 
>> The larger issue is the fact that the sdk was not found, because of this script.
>> For instance, in .../trunk/csl/fox/configure.ac I found these lines:
>> 
> FOX is ONLY wanted or used or even investigated at all in the case that 
> you are building a CSL version.
> 
>> if gcc -isysroot /Developer/SDKs/MacOSX10.7.sdk -arch ix86_64 conftest.c -o conftest > /dev/null 2>&1
>> then
>>  AC_MSG_NOTICE([x96_64 binary creation will probably be possible using 10.7])
>>  else
>>    AC_MSG_ERROR([Need MACOSX SDK version 10.6 or 10.7])
>>  fi
>> fi
>> 
> 
> The code *I* see in csl/fox/configure.in reads (well lines have got 
> truncated a bit, but you can see...). There is NO fox/configure.ac, just a 
> configure.in.
> 
>     if test "x$with_lionSDK" != "xno" &&
>        gcc -isysroot /Developer/SDKs/MacOSX10.7.sdk -arch i386 -arch x86_64 con
>     then
>       AC_MSG_NOTICE("Macintosh x86_64 binaries via 10.7")
> #     enable_dependency_tracking="no"
>       archflags="-arch x86_64 "
>     else
>       if gcc -isysroot /Developer/SDKs/MacOSX10.6.sdk -arch x86_64 conftest.c -
>       then
>         AC_MSG_NOTICE("Macintosh x86_64 binaries via 10.6")
> #       enable_dependency_tracking="no"
>         archflags="-arch x86_64 "
>       else
> #--     if gcc -isysroot /Developer/SDKs/MacOSX10.5.sdk -arch ppc -arch i386 -a
> #--     then
> #--       AC_MSG_NOTICE("Macintosh fat binaries via 10.5")
> #--       enable_dependency_tracking="no"
> #--       archflags="-arch ppc -arch i386 -arch x86_64"
> #--     else
>         ...
> where the commented out bit at the end is looking for 10.5, 10.4, ... 
> versions of the SDK.
> 
> That means I can not match your query up with reality as I see it in the 
> latest trunk sources on sourceforge.

> 
>> any explicit "gcc" calls get rerouted by that script and return error.
>   There may
>> be others in your code, but I haven't looked at them all.
>> 
> 
> So can you please explain to me, as to one who is just plain confused but 
> has maybe missed something, just what sources you are using, and if thay 
> are not the current ones we are maintaining why not.

I used svn r1535, and you're correct, it was my mistake, I meant /trunk/csl/configure.ac. 
I've been looking through so many directories, that I got them confused.
However any instance of an explicit "gcc" call will get rerouted by that script.

> 
> 
> Sorry if I am being obtuse, but your query really looks ODD to me and that 
> makes it a LOT harder for me to know how to help.
> 
> Do I understand that your colleague installs a script names "gcc" that is 
> on his PATH ahead of the regular standard Mac compiler, and that does not 
> behave like ordinary gcc and that THAT is what is causing his macports 
> attempts to fail? Again that sounds WEIRD to me if I have understood it 
> correctly.
> 
The individual who tested the portfile on 10.6 is affiliated with macports and is using
the script referenced in the wiki link I provided. The intent of that script is
to return an error if a program explicity calls "gcc" instead of setting gcc to the
default. 

> I my Mac tests I have had significant pain with the 10.7 SDK (hence the
> "--without-lionSDK" option that the configure scripts support (on the CSL 
> side)) because of changes that Apple had made. Thus even on Lion I 
> typically build with the Snow Leopard SDK.

That was the first instance I found where "gcc" was called explicity. If you used the
explicit call "gcc" to locate the 10.6 SDK, that would also fail. I was told 'a raw reference 
to "gcc" will result in an error, since macports _never_ uses just "gcc"'

I hope this makes enough sense.

> 
> 
>          Arthur
> 
> ------------------------------------------------------------------------------
> Ridiculously easy VDI. With Citrix VDI-in-a-Box, you don't need a complex
> infrastructure or vast IT resources to deliver seamless, secure access to
> virtual desktops. With this all-in-one solution, easily deploy virtual 
> desktops for less than the cost of PCs and save 60% on VDI infrastructure 
> costs. Try it free! http://p.sf.net/sfu/Citrix-VDIinabox
> _______________________________________________
> Reduce-algebra-developers mailing list
> Red...@li...
> https://lists.sourceforge.net/lists/listinfo/reduce-algebra-developers


-Mark