Merge branch 'main' into surface_callback_tracking

config/defaults/config.LINUX_INTEL.mk

@@ -7,8 +7,20 @@ PMAKE = make -j 4
 
 # ------- Define the MPI Compilers--------------------------------------
 ifdef I_MPI_ROOT # Using Intel MPI
-  FF90 = mpiifort
-  CC   = mpiicc
+  # Note that ";" is there to avoid make shell optimization, otherwise the shell command may fail
+  ICC_EXISTS := $(shell command -v icc;)
+
+  ifdef ICC_EXISTS
+    # icc only exists on older Intel versions
+    # Assume that we want to use the old compilers
+    FF90 = mpiifort
+    CC = mpiicc
+  else
+    # Use the new compilers
+    FF90 = mpiifx
+    CC = mpiicx
+  endif
+
 else # Using HPE MPI
   FF90 = ifort -lmpi
   CC   = icc -lmpi

config/defaults/config.LINUX_INTEL_AVX2.mk

@@ -7,8 +7,20 @@ PMAKE = make -j 4
 
 # ------- Define the MPI Compilers--------------------------------------
 ifdef I_MPI_ROOT # Using Intel MPI
-  FF90 = mpiifort
-  CC   = mpiicc
+  # Note that ";" is there to avoid make shell optimization, otherwise the shell command may fail
+  ICC_EXISTS := $(shell command -v icc;)
+
+  ifdef ICC_EXISTS
+    # icc only exists on older Intel versions
+    # Assume that we want to use the old compilers
+    FF90 = mpiifort
+    CC = mpiicc
+  else
+    # Use the new compilers
+    FF90 = mpiifx
+    CC = mpiicx
+  endif
+
 else # Using HPE MPI
   FF90 = ifort -lmpi
   CC   = icc -lmpi

doc/introduction.rst

@@ -3,44 +3,37 @@
 Introduction
 ============
 
-ADflow is a multi-block structured flow solver initially developed in 
-the Stanford University under the sponsorship of the Department of
-Energy Advanced Strategic Computing (ASC) Initiative. It solves the 
-compressible Euler, laminar Navier-Stokes and Reynolds-Averaged Navier-Stokes 
-equations. Although its primary objective in this program was to compute the
-flows in the rotating components of jet engines, ADflow has been developed as 
-a completely general solver and it is therefore applicable to a variety of 
-other types of problems, including external aerodynamic flows.
-
-ADflow is a parallel code, suited for running on massively parallel platforms. 
-The parallelization is hidden as much as possible from the end user, i.e. 
-there is only one grid file, one volume solution file and one surface solution file. The
-only thing the end user needs to do is to specify the number of processors 
-he/she wants to use via the mpirun (or equivalent) command.
+ADflow is a multi-block and overset structured flow solver initially developed at Stanford University under the sponsorship of the Department of Energy Advanced Strategic Computing (ASC) Initiative.
+It solves the compressible Euler, laminar Navier-Stokes and Reynolds-Averaged Navier-Stokes (RANS) equations using a second-order finite volume discretization.
+It currently features three solvers: multigrid, approximate Newton-Krylov, and Newton-Krylov, more details of which are under :ref:`solvers <adflow_solvers>`.
+Users control the CFD solver via a solver option dictionary.
+More details are in :ref:`options <adflow_options>`.
+
+Although its primary objective in this program was to compute the flows in the rotating components of jet engines, ADflow has been developed as a completely general solver, and it is therefore applicable to a variety of other types of problems, including external aerodynamic flows.
+ADflow is a parallel code, suited for running on massively parallel platforms.
+The parallelization is hidden as much as possible from the end user, i.e. there is only one grid file, one volume solution file, and one surface solution file.
+The only thing the end user needs to do is to specify the number of processors they want to use via the mpirun (or equivalent) command.
 
 A summary of the various features that can be found in ADflow is given below:
 
-* Compressible, URANS flow solver with various turbulence modeling options (Spalart-Allmaras, k-w, SST, v2-f)
+* Compressible, URANS flow solver with various turbulence modeling options (Spalart-Allmaras, k-:math:`\omega`, SST, v2-f)
 
-* Multiblock structured approach with arbitrary connectivity. One-to-one mesh point matching with subfacing
-  (C-0 multiblock) and point mismatched abutting meshes at block interfaces (C-1 multiblock) are allowed.
-  CGNS I/O (mesh and solution) as well as native, MPI-IO parallel I/O option with back and forth conversion
-  utilities.
+* Multi-block structured approach with arbitrary connectivity.
+  One-to-one mesh point matching with subfacing (C-0 multiblock) and point mismatched abutting meshes at block interfaces (C-1 multiblock) are allowed.
+  CGNS I/O (mesh and solution) as well as native, MPI-IO parallel I/O option with back and forth conversion utilities.
 
 * Massively parallel (both CPU and memory scalable) implementation using MPI.
 
-* ALE Deforming grid implementation using ``pyWarp``
+* ALE Deforming grid implementation using ``IDWarp``
 
 * Interface to conservative and consistent load and displacement transfer for aeroelastic computations.
 
-* Multigrid, Runge-Kutta solver for the mean flow and DD-ADI solution methodology for the turbulence
-  equations.
+* Multigrid, Runge-Kutta solver for the mean flow and DD-ADI solution methodology for the turbulence equations.
 
 * Central difference discretization (second order in space) with various options for artificial dissipation.
-  
+
 * Adaptive wall functions for poor quality meshes.
 
-* Unsteady time integration using second- or third-order (in time) backwards difference formulae (BDF) or a
-  time-spectral approach for time-periodic flows.
+* Unsteady time integration using second- or third-order (in time) backwards difference formulae (BDF) or a time-spectral approach for time-periodic flows.
 
 * Fully parallel, scalable pre-processor responsible load balancing.