Shoplifted version - 2021-07-08

Make/MkFlags.A64FX

@@ -1,6 +1,6 @@
 #Include this file at the end of MkFlags
 
 #Compiler
-CC = mpifccpx
+MPICC = mpifccpx
 CFLAGS = -std=c99 -O3
 

Make/MkFlags.AArch64

@@ -1,6 +1,6 @@
 #Include this file at the end of MkFlags
 
 #Compiler
-CC = mpicc
+MPICC = mpicc
 
 CFLAGS = -std=c99 -O3

Make/MkFlags.OpenPOWER

@@ -1,6 +1,6 @@
 #Include this file at the end of MkFlags
 
 #Compiler
-CC =  mpcc_r
+MPICC =  mpcc_r
 
 CFLAGS = -q64 -O4 -qarch=pwr5 -qtune=pwr5 -qstrict

Make/MkFlags.x86-64

@@ -1,6 +1,6 @@
 #Include this file at the end of MkFlags
 
 #Compiler
-CC = mpicc
+MPICC = mpicc
 CFLAGS = -std=c99 -O3 
 

Make/MkFlags.x86-64_alt

@@ -1,6 +1,6 @@
 #Include this file at the end of MkFlags
 
 #Compiler
-CC = mpicc
+MPICC = mpicc
 CFLAGS = -wd593 -wd188 -wd981 -wd177 -wd869 -std=c99 -O2
 

README.md

@@ -1,7 +1,7 @@
 ![SOMBRERO](sombrero.png)
 
 * [HiRep version: 022f909](https://github.com/sa2c/HiRep/tree/022f909)
-* [SOMBRERO base version: 31eb5fe](https://github.com/sa2c/SOMBRERO/tree/31eb5fe)
+* [SOMBRERO base version: f4d4120](https://github.com/sa2c/SOMBRERO/tree/f4d4120)
 * [shoplifter version: 2ed4c5e](https://github.com/sa2c/shoplifter/tree/2ed4c5e)
 
 A benchmarking utility for high performance computing based on lattice field theory applications.
@@ -10,7 +10,7 @@ A benchmarking utility for high performance computing based on lattice field the
 
 Requires an MPI library and compiler.
 
-## Set Up
+## Setup
 
 1. Clone the repository
 
@@ -29,7 +29,7 @@ make
 ### Setup with Spack
 
 Install and configure the package `sombrero` 
-as you would do with a regular spack package.
+as you would do with a regular Spack package.
 
 ## Usage
 
@@ -39,15 +39,15 @@ To run all benchmarks use the `sombrero.sh` script:
 ./sombrero.sh -n <num-cores>  [ -w ] [ -s small | medium | large | very_large ]
 ```
 
-By default this will run a medium scale problem using `<num-cores>` mpi ranks.
-The problem size independent of the number of processes. 
+By default this will run a medium scale problem using `<num-cores>` MPI ranks.
+The problem size is independent of the number of processes. 
 This is useful for strong scaling.
 
 All output is printed to stdout and error messages to stderr.
 
 ### Usage with Spack 
 
-When using spack, once sombrero has been installed,
+When using Spack, once SOMBRERO has been installed,
 loading the `sombrero` package with `spack load sombrero`
 will make `sombrero.sh` available
 in the PATH environment variable,
@@ -60,7 +60,7 @@ sombrero.sh -n <num-cores>  [ -w ] [ -s small | medium | large | very_large ]
 ## Parameters
 
 `-n <num-cores>`: </br>
-&nbsp;&nbsp;&nbsp; Set the number of mpi ranks. SOMBRERO will attempt to divide the problem accross the ranks automatically. The number of ranks `num-cores` should be a power of 2, `num-cores = 2^n`. This can be multiplied by a factor of 3, `num-cores = 3 * 2^n`.
+&nbsp;&nbsp;&nbsp; Set the number of MPI ranks. SOMBRERO will attempt to divide the problem accross the ranks automatically. The number of ranks `num-cores` should be a power of 2, `num-cores = 2^n`. This can be multiplied by a factor of 3, `num-cores = 3 * 2^n`.
 
 `-s small | medium | large | very_large:`</br>
 &nbsp;&nbsp;&nbsp; Specify the problem size. Compatible with weak or strong scaling.
@@ -165,9 +165,9 @@ Additional parameters `-l` and `-p` exist for more precise control of the proble
 &nbsp;&nbsp;&nbsp; Set the lattice size in each direction. Replace each `N` with an integer larger or equal to 4.
 
 `-p NxNxNxN`:</br>
-&nbsp;&nbsp;&nbsp; Manually partition the lattice among the mpi ranks. The lattice is divided accross `N` ranks in each direction. A positive integer must be used.
+&nbsp;&nbsp;&nbsp; Manually partition the lattice among the MPI ranks. The lattice is divided accross `N` ranks in each direction. A positive integer must be used.
 
-Once compiled, binaries for individual test cases can be found in the sombrero folder. For example the case 3 can be run separately with
+Once compiled, binaries for individual test cases can be found in the `sombrero` folder. For example the case 3 can be run separately with
 
 ```
 mpirun -n Np sombrero/sombrero3 -s medium 
@@ -179,3 +179,18 @@ The verbose mode will produce additional information:
 mpirun -n Np sombrero/sombrero3 -s medium -v verbose 
 ```
 
+In case SOMBRERO was installed with Spack,
+in order to run a specific benchmark separately
+one has to use the full path to the executable program.
+To do so, 
+once SOMBRERO is loaded by Spack with the command
+
+``` 
+spack load sombrero
+```
+
+one can run case 3 with
+
+``` 
+mpirun -n 2 $(which sombrero.sh | xargs dirname)/sombrero/sombrero3 -s small
+```

sombrero/sombrero.c

@@ -345,7 +345,7 @@ int main(int argc, char *argv[]) {
   lprintf("MAIN", 0,
           "SOMBRERO built from HiRep commit 022f909\n");
   lprintf("MAIN", 0,
-          "Base SOMBRERO commit 31eb5fe\n");
+          "Base SOMBRERO commit f4d4120\n");
   lprintf("MAIN", 0,
           "SOMBRERO built using shoplifter commit 2ed4c5e\n");