GPU library updates

lib/gpu/Makefile.linux

@@ -45,6 +45,12 @@ CUDA_ARCH = -arch=sm_60
 #CUDA_ARCH = -arch=sm_80
 #CUDA_ARCH = -arch=sm_86
 
+# Lovelace hardware
+#CUDA_ARCH = -arch=sm_89
+
+# Hopper hardware
+#CUDA_ARCH = -arch=sm_90
+
 # this setting should match LAMMPS Makefile
 # one of LAMMPS_SMALLBIG (default), LAMMPS_BIGBIG and LAMMPS_SMALLSMALL
 

lib/gpu/Makefile.linux_multi

@@ -37,13 +37,17 @@ CUDA_ARCH = -arch=sm_50
 #CUDA_ARCH = -arch=sm_80
 #CUDA_ARCH = -arch=sm_86
 
+# Lovelace hardware
+#CUDA_ARCH = -arch=sm_89
+
 # Hopper hardware
 #CUDA_ARCH = -arch=sm_90
 
 CUDA_CODE = -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52] \
 	    -gencode arch=compute_60,code=[sm_60,compute_60] -gencode arch=compute_61,code=[sm_61,compute_61] \
 	    -gencode arch=compute_70,code=[sm_70,compute_70] -gencode arch=compute_75,code=[sm_75,compute_75] \
-	    -gencode arch=compute_80,code=[sm_80,compute_80] -gencode arch=compute_86,code=[sm_86,compute_86]
+	    -gencode arch=compute_80,code=[sm_80,compute_80] -gencode arch=compute_86,code=[sm_86,compute_86] \
+	  # -gencode arch=compute_89,code=[sm_89,compute_89] -gencode arch=compute_90,code=[sm_90,compute_90]
 
 CUDA_ARCH += $(CUDA_CODE)
 

lib/gpu/geryon/nvd_macros.h

@@ -7,6 +7,13 @@
 
 #define CUDA_INT_TYPE size_t
 
+// Use the primary context for GPU access to enable compatibility with tools
+// like OpenMPI accessing the GPU through the runtime interface.
+// Set to 0 to revert to old behavior
+#ifndef GERYON_NVD_PRIMARY_CONTEXT
+#define GERYON_NVD_PRIMARY_CONTEXT 1
+#endif
+
 #ifdef MPI_GERYON
 #include "mpi.h"
 #define NVD_GERYON_EXIT do {                                               \