[petsc-users] Missing header file in 54f71ec?

[Jay J. Billings]

Jed,

I just tried to rebuild it on another system and it wasn't a problem 
there. If it pops back up on the first machine I used then I will let 
you know and send you the log files. I'm having some configuration 
issues with that machine that may be the problem.

Thanks,
Jay

On 01/21/2014 10:53 PM, Jed Brown wrote:
> "Jay J. Billings" <billingsjj at ornl.gov> writes:
>
>> It looks like this might have been caused by the change to configure.py
>> and petscsys.h in 5affe1d:
>>
>> https://bitbucket.org/petsc/petsc/commits/5affe1da4b5ff6ba9c99ae0e08ad5ae60b099835
> Please send configure.log and make.log.  C89 and others (SVr4, 4.3BSD,
> C99, POSIX.1-2001) say that string.h provides memcpy and memset.
> string.h is included and this change built cleanly on all our test
> machines.
>
> commit a3aaec0ab474bb386780073594fdc8ac7d2aa0be
> Author: Jed Brown <jedbrown at mcs.anl.gov>
> Date:   Thu Jan 9 14:40:33 2014 -0700
>
>      petscsys.h: find memcpy and memset in string.h
>      
>      C89 and later say that memcpy and memset are in string.h.  With glibc,
>      memory.h merely includes string.h, and one user reported an error due to
>      undefined memset (which I cannot reproduce).  So quit bothering with
>      nonstandard memory.h and just include string.h.
>      
>      Reported-by: Alexander Krimm <alex at wirew0rm.de>

-- 
Jay Jay Billings
Oak Ridge National Laboratory
Twitter Handle: @jayjaybillings