> On Nov 2, 2013, at 5:13 AM, Francesco Ballarin <francesco.balla...@polimi.it>
> wrote:
>
> Hi John,
> thanks.
>
> a) I tried building PETSc with --with-clanguage=cxx as you suggested, my
> configure options are now
Sorry, didn't mean to suggest _trying_ this, it shouldn't be necessary. Was
just wondering if this was what was mucking things up.
> b) I modified m4/compiler.m4 in libMesh to be sure that
> -fno-optimize-sibling-calls CXXFLAGS where really added.
>
> I think -fno-optimize-sibling-calls is not working: when building .o files I
> get the following warning:
> ===
> /opt/ibmcmp/vacpp/bg/12.1/bin/.orig/bgxlC_r: 1501-216 (W) command option
> -fno-optimize-sibling-calls is not recognized - passed to ld
> ===
> and the following error when linking
> ===
> /bgsys/drivers/ppcfloor/gnu-linux/powerpc64-bgq-linux/bin/ld: -f may not be
> used without -shared
> ===
So -fno-optimize-blah isn't relevant to you since you're doing static
linking... disregard that suggestion I guess!
> c) Afterwards, I tried to use -qminimaltoc (which seems to be the xl
> equivalent of -mminimal-toc, see http://hpc.mines.edu/bgq/xlc/mapping.pdf ),
> instead of -fno-optimize-sibling-calls, but however I get the same linking
> errors as in my previous email
> ===
> /bgsys/drivers/ppcfloor/gnu-linux/powerpc64-bgq-linux/bin/ld:
> /bgsys/drivers/V1R2M0/ppc64/comm/sys-fast/lib/libpami.a(component-pami--CAPI.cnk.o)(.text._ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E[_ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E]+0xbc):
> sibling call optimization to
> `_ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E'
> does not allow automatic multiple TOCs; recompile with -mminimal-toc or
> -fno-optimize-sibling-calls, or make
> `_ZNSt8_Rb_treeIjSt4pairIKjjESt10_Select1stIS2_ESt4lessIjESaIS2_EE8_M_eraseEPSt13_Rb_tree_nodeIS2_E'
> extern
> ===
>
> You can find the resulting flags in the patch attached. It also fixes a
> couple of small problems in "contrib/libHilbert/src/hilbert_mpi_ops.C"
> (missing iostream) and "src/utils/timestamp.C" (missing libMesh::out), which
> I forgot to mention previously.
Thanks for the patch, I'll take a look at it soon.
------------------------------------------------------------------------------
Android is increasing in popularity, but the open development platform that
developers love is also attractive to malware creators. Download this white
paper to learn more about secure code signing practices that can help keep
Android apps secure.
http://pubads.g.doubleclick.net/gampad/clk?id=65839951&iu=/4140/ostg.clktrk
_______________________________________________
Libmesh-devel mailing list
Libmesh-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libmesh-devel
