Replace HLLD Riemann Solver with the one from Athena++

- the Athena++ implementation seems much easier to follow and is noticably faster (for equivalent sets of optimization flags).
- We slightly tweaked some of the expected VL+CT test results to account for this change.
- Unfortunately, we need to drop the -fno-math-errno flag from linux_gcc.cmake, by default. The MHD_shock_tube_test showed that with the fno-math-errno flag, the 1D MHD shock tube problems have slightly different L1 error norms for the VL+CT solver, depending on the position in the transverse direction (In other words, if we ran the 1D shock tube test along the y-direction, the L1 error norm depends on the x,z components of the problem).

Author: mabruzzo

LICENSE_ATHENAPP

@@ -0,0 +1,31 @@
+= Athena++ Software Public License =
+
+BSD 3-Clause License
+
+Copyright (c) 2016, PrincetonUniversity
+All rights reserved.
+
+Redistribution and use in source and binary forms, with or without
+modification, are permitted provided that the following conditions are met:
+
+* Redistributions of source code must retain the above copyright notice, this
+  list of conditions and the following disclaimer.
+
+* Redistributions in binary form must reproduce the above copyright notice,
+  this list of conditions and the following disclaimer in the documentation
+  and/or other materials provided with the distribution.
+
+* Neither the name of the copyright holder nor the names of its
+  contributors may be used to endorse or promote products derived from
+  this software without specific prior written permission.
+
+THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

config/linux_gcc.cmake

@@ -16,7 +16,11 @@ if(NOT __processedUserDefaults)
   # Set some architecture-specific optimization flags
   # (in the future, we might want to excercise more control over the targetted
   # instruction set to a higher level)
-  set(__ARCH_C_OPT_FLAGS "-O3 -fopenmp-simd -march=native -DNDEBUG -funroll-loops -fno-trapping-math -fno-math-errno -fno-signed-zeros")
+
+  # if we use -fno-math-errno, that cause 1D MHD shock tube problems to have
+  # slightly different L1 error norms for the VL+CT solver, depending on the
+  # position in the transverse direction (see MHD_shock_tube_test)
+  set(__ARCH_C_OPT_FLAGS "-O3 -fopenmp-simd -march=native -DNDEBUG -funroll-loops -fno-trapping-math -fno-signed-zeros")
 
   set(CMAKE_C_FLAGS_RELEASE "${__ARCH_C_OPT_FLAGS}")
   set(CMAKE_C_FLAGS_RELWITHDEBINFO "-g ${__ARCH_C_OPT_FLAGS}")

input/vlct/run_MHD_shock_tube_test.py

@@ -78,11 +78,11 @@ def analyze_tests():
     # use 1 block - if we use more than one block, it's unclear to me if it's
     # ok to have round-off errors)
 
-    r.append(err_compare(0.012524502240006011,"x"))
+    r.append(err_compare(0.012523489882320429, "x"))
     r.append(err_compare(0.0, "x", std_dev=True))
-    r.append(err_compare(0.012524502240005972,"y"))
+    r.append(err_compare(0.012523489882320308, "y"))
     r.append(err_compare(0.0, "y", std_dev=True))
-    r.append(err_compare(0.012524502240005921,"z"))
+    r.append(err_compare(0.012523489882320315, "z"))
     r.append(err_compare(0.0, "z", std_dev=True))
 
     n_passed = np.sum(r)

input/vlct/run_dual_energy_cloud_test.py

@@ -81,7 +81,7 @@ def check_cloud_asym(fname, name, max_asym):
 def analyze_tests():
     r = []
     r += check_cloud_asym('hlld_cloud_0.0625/hlld_cloud_0.0625.block_list',
-                          'hlld_cloud', 5.5e-13)
+                          'hlld_cloud', 6.3e-13)
     r += check_cloud_asym('hllc_cloud_0.0625/hllc_cloud_0.0625.block_list',
                           'hllc_cloud', 4.6e-13)
     r += check_cloud_asym('hlle_cloud_0.0625/hlle_cloud_0.0625.block_list',