[petsc-dev] Possible error running C/C++ src/snes/examples/tutorials/ex19 with 1 MPI process

[Satish Balay]

On Thu, 13 Oct 2016, Antonio Trande wrote:

> New builds:
> 
> Without "/usr/lib/rpm/redhat/redhat-hardened-cc1" flags:
> http://copr-fe.cloud.fedoraproject.org/coprs/sagitter/petsc/build/464687/
> 
> Same configuration for debugging:
> http://copr-fe.cloud.fedoraproject.org/coprs/sagitter/petsc/build/464692/

Why a different debug build? Whats the difference? And what I should be looking for? {in the debug build?}

Wrt 464687 - the only issue i see is the false error check - [the
patch I sent in the previous e-mail should fix this check.]

The other minor issues are:

https://copr-be.cloud.fedoraproject.org/results/sagitter/petsc/fedora-23-i386/00464687-petsc/build.log.gz

<seq>
gcc -o ex19.o -c -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches  -m32 -march=i686 -mtune=atom -fasynchronous-unwind-tables      -I/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/include -I/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/i386/include -I/builddir/build/BUILD/petsc-3.7.4/petsc-3.7.4/include/petsc/mpiuni    `pwd`/ex19.c

<openmp>
/usr/lib/openmpi/bin/mpicc -o ex19.o -c -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches  -m32 -march=i686 -mtune=atom -fasynchronous-unwind-tables -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches  -m32 -march=i686 -mtune=atom -fasynchronous-unwind-tables      -I/builddir/build/BUILD/petsc-3.7.4/buildopenmpi_dir/include -I/builddir/build/BUILD/petsc-3.7.4/buildopenmpi_dir/i386/include -I/usr/include/openmpi-i386/hypre    `pwd`/ex19.c

<mpich>
/usr/lib/mpich/bin/mpicc -o ex19.o -c -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches  -m32 -march=i686 -mtune=atom -fasynchronous-unwind-tables -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches  -m32 -march=i686 -mtune=atom -fasynchronous-unwind-tables -O2 -g -pipe -Wall -Werror=format-security -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector-strong --param=ssp-buffer-size=4 -grecord-gcc-switches  -m32 -march=i686 -mtune=atom -fasynchronous-unwind-tables      -I/builddir/build/BUILD/petsc-3.7.4/buildmpich_dir/include -I/builddir/build/BUILD/petsc-3.7.4/buildmpich_dir/i386/include -I/usr/include/mpich-i386/hypre    `pwd`/ex19.c

The CFLAGS are getting used once in <seq> - twice in <openmp> and
three times in <mpich> builds.  Perhaps the FLAGS cleanup didn't fix
all the builds.. [I won't cause problems - just make it a bit more
unreadable..]

And you appear to remove
-specs=/usr/lib/rpm/redhat/redhat-hardened-cc1 form everywhere
[including CC,CXX,LD] - the buggy behavior was with gfortran
only.. You might want to remove it from FFLAGS - but keep it with
others...

Perhaps the 'gofrtran -specs=/usr/lib/rpm/redhat/redhat-hardened-cc1'
behavior on i386 is a gfortran bug?

Satish