That worked! Thanks, Chun.

> On Feb 22, 2024, at 9:48 PM, Chun Tian (binghe) <[email protected]> wrote:
> 
> Sorry, the environment variable name should be LDFLAGS instead of LD_FLAGS ...
> 
> According to HOL4's FAQ page [1], Poly/ML works on M1 (or later) Macs, 
> including macOS Sonoma (due to Xcode 15, otherwise -ld_classic won't be 
> needed).
> 
> [1] https://hol-theorem-prover.org/faq.html
> 
> On Fri, Feb 23, 2024 at 1:38 PM Steve Arons <[email protected] 
> <mailto:[email protected]>> wrote:
>> Thank you for your reply, Chun; unfortunately the result is the same. I now 
>> see that the Macports folks ran into the same problem with 5.9.1 in 
>> December, which remains unsolved—they’re still distributing 5.9.
>> 
>> Steve
>> 
>> 
>>> On Feb 22, 2024, at 9:22 PM, Chun Tian (binghe) <[email protected] 
>>> <mailto:[email protected]>> wrote:
>>> 
>>> Try putting -ld_classic into your LD_FLAGS, i.e.
>>> 
>>> export LD_FLAGS=-ld_classic
>>> ./configure ...
>>> 
>>> --Chun
>>> 
>>> On Fri, Feb 23, 2024 at 1:09 PM Steve Arons <[email protected] 
>>> <mailto:[email protected]>> wrote:
>>>> The linker fails when writing the object code:
>>>> 
>>>> ******Writing object code******
>>>> /bin/sh ./libtool  --tag=CC   --mode=link gcc  -O3   -o poly  polyexport.o 
>>>>  libpolymain/libpolymain.la <http://libpolymain.la/> 
>>>> libpolyml/libpolyml.la <http://libpolyml.la/>  -lstdc++ 
>>>> libtool: link: gcc -O3 -o .libs/poly polyexport.o  
>>>> libpolymain/.libs/libpolymain.a libpolyml/.libs/libpolyml.dylib -lstdc++
>>>> ld: LINKEDIT overlap of start of LINKEDIT and symbol table in 
>>>> '/Users/susanbeningson/poly/polyml/polyexport.o' in 
>>>> '/Users/susanbeningson/poly/polyml/polyexport.o'
>>>> clang: error: linker command failed with exit code 1 (use -v to see 
>>>> invocation)
>>>> make[2]: *** [poly] Error 1
>>>> make[1]: *** [all-recursive] Error 1
>>>> make: *** [all] Error 2
>>>> 
>>>> This is clang from the Apple dev tools masquerading as gcc; is it 
>>>> necessary to use GNU gcc and the GNU bin tools?
>>>> 
>>>> Thanks,
>>>> Steve
>>>> 
>>>> _______________________________________________
>>>> polyml mailing list
>>>> [email protected] <mailto:[email protected]>
>>>> http://lists.inf.ed.ac.uk/mailman/listinfo/polyml
>> 

_______________________________________________
polyml mailing list
[email protected]
http://lists.inf.ed.ac.uk/mailman/listinfo/polyml

Reply via email to