Allow skipping setting the RHS to zero in MoL

By default, MoL initialises the RHS variables to zero before calling
the CalcRHS routines. This is (a) superfluous in a well-written code,
and (b) makes it impossible to re-use a RHS that has been calculated
ahead of time, e.g. at the end of the previous time step.

This patch adds a parameter to disable this behaviour.


git-svn-id: http://svn.cactuscode.org/arrangements/CactusNumerical/MoL/trunk@173 578cdeb0-5ea1-4b81-8215-5a3b8777ee0b

2 files changed, 12 insertions, 3 deletions

diff --git a/param.ccl b/param.ccl
index 7e29e56..ba43b51 100644
--- a/param.ccl
+++ b/param.ccl
@@ -162,6 +162,12 @@ BOOLEAN disable_prolongation "If Mesh refinement is enabled should we use buffer
 
 
 
+BOOLEAN init_RHS_zero "Initialise the RHS to zero" STEERABLE=recover
+{
+} "yes"
+
+
+
 BOOLEAN adaptive_stepsize "Choose the time step size adaptively"
 {
 } "no"
diff --git a/schedule.ccl b/schedule.ccl
index 9c33bbe..f0aed52 100644
--- a/schedule.ccl
+++ b/schedule.ccl
@@ -484,10 +484,13 @@ if (CCTK_Equals(ODE_Method,"ICN-avg"))
 ###  The time step is initialised to zero  ###
 ##############################################
 
-schedule MoL_InitRHS IN MoL_Step BEFORE MoL_CalcRHS
+if (init_RHS_zero)
 {
-  LANG: C
-} "Initialise the RHS functions"
+  schedule MoL_InitRHS IN MoL_Step BEFORE MoL_CalcRHS
+  {
+    LANG: C
+  } "Initialise the RHS functions"
+}
 
 #####################################################
 ###  The group where all the physics takes place  ###