Roland

On Hopper (also a Cray), we switched from using explicit path name to using explicit module load commands. This makes the option list much shorter.

Maybe hopper-intel.cfg could be a good starting point? hopper.ini contains the relevant module commands in the "envsetup" entry.

-erik

On Fri, Mar 9, 2012 at 9:03 PM, Roland Haas roland.haas@physics.gatech.edu wrote:

Hello all,

...

I seem to have trouble compiling Cactus on kraken after NICS's recent (Mar 8th) OS update, see https://portal.xsede.org/web/xup/user-news/-/news/item/5707 .

Right now it will not even mpi.h (most likely since SYS_INC_DIRS=/opt/cray/mpt/5.0.0/xt/seastar/mpich2-intel/include no longer exists).

I assume mentioning that this is with a current checkout of simfactory/trunk and kraken-intel.cfg might help.

Yours, Roland

-- My email is as private as my paper mail. I therefore support encrypting and signing email messages. Get my PGP key from http://keys.gnupg.net.

---

Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users

Scott

I have not seen this error before.

Since you mention running your own thorn before this error appears, and since things changed in Carpet recently, maybe there is a bad interaction and something stomps on memory? You would debug this either by disabling (parts of) your thorn to see whether this makes the problem go away, or by using a tool such a valgrind or another memory checker.

-erik

On Sat, Mar 10, 2012 at 12:11 PM, Scott Hawley scott.hawley@belmont.edu wrote:

I'm running on NICS's nautilus, using the GNU compilers.   And I get a segmentation fault at Line 2767:

while (timer && strcmp(timer->schedule_bin, data->schedule_bin))

"timer" acquires the (bad) address 0xfefefefefefefeff, and the strcmp is still being executed, so it seg faults.

"attribute" seems to have a valid address, and so does attribute->timers (0x55cb6cc0).  But attribute->timers->schedule_bin has a bad address. ("next" is NULL, btw.)

Does anyone have a suggestion?

Thanks, Scott What follows is my .par file The seg fault happens AFTER it exits my thorn TexMEx, btw.

Cactus::cctk_run_title     = "TexMEx, FMR BBH" Cactus::cctk_full_warnings = yes Cactus::cctk_itlast = 0

ActiveThorns = "AEILocalInterp ADMMass Boundary CartGrid3D CoordBase SymBase Slab Carpet CarpetLib CarpetInterp CarpetReduce Ca rpetSlab CarpetRegrid IOUtil SphericalSurface NaNChecker ADMBase ADMCoupling ADMMacros CoordGauge SpaceMask OutsideMask StaticC onformal TexMEx AHFinderDirect ADMAnalysis ADMConstraints LoopControl InitBase CarpetIOBasic CarpetIOScalar CarpetIOASCII Carpe tIOHDF5 "

SphericalSurface::nsurfaces = 3 SphericalSurface::maxntheta = 39 SphericalSurface::maxnphi   = 76 SphericalSurface::ntheta      [0] = 39 SphericalSurface::nphi        [0] = 76 SphericalSurface::nghoststheta[0] = 2 SphericalSurface::nghostsphi  [0] = 2 SphericalSurface::ntheta      [1] = 39 SphericalSurface::nphi        [1] = 76 SphericalSurface::nghoststheta[1] = 2 SphericalSurface::nghostsphi  [1] = 2 SphericalSurface::ntheta      [2] = 39 SphericalSurface::nphi        [2] = 76 SphericalSurface::nghoststheta[2] = 2 SphericalSurface::nghostsphi  [2] = 2

IO::out_dir = $parfile

Carpet::domain_from_coordbase = yes Carpet::max_refinement_levels = 10

driver::ghost_size   = 2

InitBase::initial_data_setup_method="init_all_levels" Carpet::prolongation_order_time  = 1 Carpet::convergence_level = 0 ADMMacros::spatial_order = 4 ADMBase::initial_data = "TexMEx, Kerr-Schild" ADMBase::lapse_timelevels = 3 ADMBase::shift_timelevels = 3 ADMConstraints::constraints_timelevels = 3 ADMBase::metric_timelevels = 3

CoordBase::domainsize = minmax CoordBase::spacing    = numcells CoordBase::xmin = -36.0 CoordBase::ymin = -36.0 CoordBase::zmin = -36.0 CoordBase::xmax =  36.0 CoordBase::ymax =  36.0 CoordBase::zmax =  36.0 CoordBase::ncells_x = 192 CoordBase::ncells_y = 192 CoordBase::ncells_z = 192

CartGrid3D::type         = coordbase CartGrid3D::avoid_origin = no

NaNChecker::check_every     = 524288 NaNChecker::action_if_found = terminate

TexMEx::boundary_type = Robin CarpetRegrid::refinement_levels = 5 CarpetRegrid::refined_regions = manual-coordinate-list CarpetRegrid::smart_outer_boundaries = yes  CarpetRegrid::coordinates = "[

[([-20.25,-20.25,-20.25]:[20.25,20.25,20.25]:[0.1875,0.1875,0.1875])],

[([-12.375,-9.375,-9.375]:[12.375,9.375,9.375]:[0.09375,0.09375,0.09375])],

[([-8.4375,-4.3125,-4.3125]:[-1.6875,4.3125,4.3125]:[0.046875,0.046875,0.04 6875]),

([1.6875,-4.3125,-4.3125]:[8.4375,4.3125,4.3125]:[0.046875,0.046875,0.04687 5])],

[([-6.46875,-1.78125,-1.78125]:[-3.65625,1.78125,1.78125]:[0.0234375,0.0234 375,0.0234375]),

([3.65625,-1.78125,-1.78125]:[6.46875,1.78125,1.78125]:[0.0234375,0.0234375 ,0.0234375])]                       ]" TexMEx::skip_solver = "no" TexMEx::bh1_m = 1.0 TexMEx::bh1_a = 0.5 TexMEx::bh1_theta = 0 TexMEx::bh1_phi = 1.57 TexMEx::bh1_x = -5.00 TexMEx::bh1_y = 0.0 TexMEx::bh1_z = 0.0 TexMEx::bh1_vx = 1.0e-33 TexMEx::bh1_vy = 0.0 TexMEx::bh1_vz = 0.0 TexMEx::bh2_m = 1.0 TexMEx::bh2_a = 0.5 TexMEx::bh2_theta = 0.0 TexMEx::bh2_phi = 0.0 TexMEx::bh2_x = 5.00 TexMEx::bh2_y = 0.0 TexMEx::bh2_z = 0.0 TexMEx::bh2_vx = 1.0e-33 TexMEx::bh2_vy = 0.0 TexMEx::bh2_vz = 0.0 TexMEx::nxyz_c = 25 TexMEx::cgtol = 1e-8 TexMEx::nlevels = 8 TexMEx::r_mask = 0.9 TexMEx::nvcycle = 4 TexMEx::preswp = 3 TexMEx::pstswp = 3 TexMEx::use_esr = "no" TexMEx::sor_factor = 1.7 TexMEx::gft_trace = "no" TexMEx::alt_ib = 0 TexMEx::use_fma = 0 TexMEx::admmass_box = "[-30,-30,-30]:[30,30,30]"

ADMMass::ADMMass_surface_distance[0] = 30 ADMMass::ADMMass_volume_radius[0] = 30 ADMMass::ADMMass_Excise_Horizons = "yes" ADMMass::ADMMass_Debug = "yes"

SpaceMask::use_mask = yes

AHFinderDirect::find_every = 1 AHFinderDirect::geometry_interpolator_name = "Lagrange polynomial interpolation" AHFinderDirect::geometry_interpolator_pars = "order=4" AHFinderDirect::surface_interpolator_name  = "Lagrange polynomial interpolation" AHFinderDirect::surface_interpolator_pars  = "order=4" AHFinderDirect::N_horizons = 2 AHFinderDirect::origin_x                                 [1] =  -5.0 AHFinderDirect::initial_guess__coord_sphere__x_center    [1] =  -5.0 AHFinderDirect::initial_guess__coord_sphere__radius      [1] =  2.0 AHFinderDirect::which_surface_to_store_info              [1] = 0 AHFinderDirect::origin_x                                 [2] =  5.0 AHFinderDirect::initial_guess__coord_sphere__x_center    [2] =  5.0 AHFinderDirect::initial_guess__coord_sphere__radius      [2] =  2.0 AHFinderDirect::which_surface_to_store_info              [2] = 1 AHFinderDirect::set_mask_for_individual_horizon          [2] = no

---

Users mailing list Users@einsteintoolkit.org http://lists.einsteintoolkit.org/mailman/listinfo/users