[SCM] Gerris Flow Solver branch, upstream, updated. b3aa46814a06c9cb2912790b23916ffb44f1f203

[Stephane Popinet]

The following commit has been merged in the upstream branch:
commit 814f5f5a0695803ca70c997bfc5e4c5a86af019c
Author: Stephane Popinet <popinet at users.sf.net>
Date:   Fri Mar 10 15:51:08 2006 +1100

    Time step is set before the first projection
    
    Previously the scaling for the tolerance on the divergence of the
    first projection was based on a timestep of unity (i.e. usually much
    larger than the subsequent timesteps).
    
    darcs-hash:20060310045108-d4795-06ff6ccc15b1f4fafde3260f7d4a8c3e221ce5df.gz

diff --git a/THANKS b/THANKS
index 5ce0892..1cfaf05 100644
--- a/THANKS
+++ b/THANKS
@@ -4,3 +4,5 @@ following people (in random order):
 Dan Kelley
 R Sai Kiran
 David Sterling
+Frank Griessbaum
+Ben Moat
diff --git a/src/simulation.c b/src/simulation.c
index 174d780..fd9ea6c 100644
--- a/src/simulation.c
+++ b/src/simulation.c
@@ -464,6 +464,7 @@ static void simulation_run (GfsSimulation * sim)
     i = i->next;
   }
 
+  gfs_simulation_set_timestep (sim);
   gfs_approximate_projection (domain,
       			      &sim->approx_projection_params,
       			      &sim->advection_params,

-- 
Gerris Flow Solver