[petsc-dev] patch for batched config

[Stefano Zampini]

from
https://www.gnu.org/software/autoconf/manual/autoconf-2.69/html_node/Hosts-and-Cross_002dCompilation.html
....
Now, configure enters cross-compilation mode if and only if --host is
passed.
....

So, you just tell configure that the host where you are going to run the
code is different from the building CPU, but you do not specify a string
associated with it, which should be fine


2017-10-10 16:53 GMT+03:00 Matthew Knepley <knepley at gmail.com>:

> On Tue, Oct 10, 2017 at 9:48 AM, Satish Balay <balay at mcs.anl.gov> wrote:
>
>> I never understood --host option of autoconf - and it would rarely
>> work for me.
>>
>
> I do not understand this option either. Jed, what does it do?
>
>   Matt
>
>
>> For packages where '--with-batch --download-package' did not work -
>> our suggstion so far was to install them manually.
>>
>> I'm hesitant to add this patch [it might work on one setup - but not
>> the other?] - but Matt/Jed might have better suggestions.
>>
>> cc:ing petsc-dev
>>
>> Satish
>>
>> On Tue, 10 Oct 2017, Stefano Zampini wrote:
>>
>> > Satish,
>> >
>> > today I had some problems installing PETSc with --download-hypre in
>> batched
>> > mode. Here is my fix
>> >
>> > diff --git a/config/BuildSystem/config/packages/hypre.py
>> > b/config/BuildSystem/config/packages/hypre.py
>> > index e374532..3c607c9 100644
>> > --- a/config/BuildSystem/config/packages/hypre.py
>> > +++ b/config/BuildSystem/config/packages/hypre.py
>> > @@ -49,6 +49,9 @@ class Configure(config.package.GNUPackage):
>> >      libs = ' '.join(libs)
>> >      args.append('--with-MPI-libs="'+libs+'"')
>> >
>> > +    if self.argDB['with-batch']:
>> > +      args.append('--host')
>> > +
>> >      # tell hypre configure not to look for blas/lapack [and not use
>> > hypre-internal blas]
>> >
>> >  args.append('--with-blas-lib="'+self.libraries.toString(self
>> .blasLapack.dlib)+'"')
>> >      args.append('--with-lapack-lib=" "')
>> >
>> >
>> >
>>
>>
>
>
> --
> What most experimenters take for granted before they begin their
> experiments is infinitely more interesting than any results to which their
> experiments lead.
> -- Norbert Wiener
>
> https://www.cse.buffalo.edu/~knepley/ <http://www.caam.rice.edu/~mk51/>
>



-- 
Stefano
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20171010/9da50b38/attachment-0001.html>