https://gitlab.com/petsc/petsc/-/merge_requests/3958 
<https://gitlab.com/petsc/petsc/-/merge_requests/3958>


> On May 6, 2021, at 11:57 AM, Satish Balay via petsc-dev 
> <[email protected]> wrote:
> 
> Wack a mole with our approach of configure treating linker warnings as errors 
> [aka -Werror]
> 
> Satish
> 
> -----
> 
> diff --git a/config/BuildSystem/config/framework.py 
> b/config/BuildSystem/config/framework.py
> index 968071d995..689f0c4ee5 100644
> --- a/config/BuildSystem/config/framework.py
> +++ b/config/BuildSystem/config/framework.py
> @@ -547,6 +547,7 @@ class Framework(config.base.Configure, 
> script.LanguageProcessor):
>       lines = [s for s in lines if s.find('may conflict with libgfortran') < 
> 0]
>       # MacOS libraries built for different MacOS versions
>       lines = [s for s in lines if s.find(' was built for newer macOS 
> version') < 0]
> +      lines = [s for s in lines if s.find(' was built for newer OSX 
> version') < 0]
>       if lines: output = '\n'.join(lines)
>       else: output = ''
>       self.log.write("Linker output after filtering:\n"+output+":\n")
> 
> 
> On Thu, 6 May 2021, Jacob Faibussowitsch wrote:
> 
>> Guess who’s back:
>> 
>> ld: warning: object file 
>> (/Users/kongf/projects/petsc1/arch-darwin-c-opt/lib/libfblas.a(ddot.o)) was 
>> built for newer OSX version (10.14) than being linked (10.9)
>> 
>> Best regards,
>> 
>> Jacob Faibussowitsch
>> (Jacob Fai - booss - oh - vitch)
>> 
>>> On May 6, 2021, at 12:43, Fande Kong <[email protected]> wrote:
>>> 
>>> Hi All,
>>> 
>>> I could compile 3.15.0, but could not compile the main branch using the 
>>> same environment.
>>> 
>>> If you need me to do git bisect, please let me know.
>>> 
>>> The log was attached.
>>> 
>>> Fande
>>> 
>>> 
>>> <configure.log>
>> 
>> 

Reply via email to