Fix arguments to error message, remove tabs and trailing spaces

opal/mca/common/cuda/common_cuda.c

@@ -1750,8 +1750,8 @@ static int mca_common_cuda_is_gpu_buffer(const void *pUserBuf, opal_convertor_t
 
     res = cuFunc.cuPointerGetAttributes(3, attributes, attrdata, dbuf);
     OPAL_OUTPUT_VERBOSE((101, mca_common_cuda_output,
-			"dbuf=%p, memType=%d, ctx=%p, isManaged=%d, res=%d",
+                        "dbuf=%p, memType=%d, ctx=%p, isManaged=%d, res=%d",
-			 (void *)dbuf, (int)memType, (void *)ctx, isManaged, res));
+                         (void *)dbuf, (int)memType, (void *)ctx, isManaged, res));
 
     /* Mark unified memory buffers with a flag.  This will allow all unified
      * memory to be forced through host buffers.  Note that this memory can
@@ -1994,8 +1994,8 @@ int mca_common_cuda_get_address_range(void *pbase, size_t *psize, void *base)
     CUresult result;
     result = cuFunc.cuMemGetAddressRange((CUdeviceptr *)pbase, psize, (CUdeviceptr)base);
     if (OPAL_UNLIKELY(CUDA_SUCCESS != result)) {
-        opal_show_help("help-mpi-common-cuda.txt", "cuMemGetAddressRange failed",
+        opal_show_help("help-mpi-common-cuda.txt", "cuMemGetAddressRange failed 2",
-                       true, result, base);
+                       true, OPAL_PROC_MY_HOSTNAME, result, base);
         return OPAL_ERROR;
     } else {
         opal_output_verbose(50, mca_common_cuda_output,
@@ -2054,7 +2054,8 @@ void mca_common_cuda_get_buffer_id(mca_mpool_base_registration_t *reg)
     res = cuFunc.cuPointerGetAttribute(&bufID, CU_POINTER_ATTRIBUTE_BUFFER_ID,
                                        (CUdeviceptr)dbuf);
     if (OPAL_UNLIKELY(res != CUDA_SUCCESS)) {
-        opal_show_help("help-mpi-common-cuda.txt", "bufferID failed", true, res);
+        opal_show_help("help-mpi-common-cuda.txt", "bufferID failed",
+                       true, OPAL_PROC_MY_HOSTNAME, res);
     }
     reg->gpu_bufID = bufID;
 

opal/mca/common/cuda/help-mpi-common-cuda.txt

@@ -65,6 +65,14 @@ cannot be used.
 Check the cuda.h file for what the return value means. Perhaps a reboot
 of the node will clear the problem.
 #
+[cuMemGetAddressRange failed 2]
+The call to cuMemGetAddressRange failed during the GPU RDMA protocol.
+  Host:  %s
+  cuMemGetAddressRange return value:  %d
+  address:  %p
+Check the cuda.h file for what the return value means. This is highly
+unusual and should not happen. The program will probably abort.
+#
 [Out of cuEvent handles]
 The library has exceeded its number of outstanding event handles.
 For better performance, this number should be increased.