[gqttq] in CODEGEN, improve the comment on nprocesses>2 for issues ma…

…dgraph5/madgraph4gpu#272 and madgraph5/madgraph4gpu#343 (see also PRs madgraph5/madgraph4gpu#619, madgraph5/madgraph4gpu#626, madgraph5/madgraph4gpu#360 and madgraph5/madgraph4gpu#396)
mg5amcnlo · Apr 7, 2023 · 29ade0d · 29ade0d
1 parent de11f92
commit 29ade0d
Showing 1 changed file with 2 additions and 2 deletions.
diff --git a/madgraph/iolibs/template_files/gpu/process_function_definitions.inc b/madgraph/iolibs/template_files/gpu/process_function_definitions.inc
@@ -395,10 +395,10 @@ namespace mg5amcCpu
   {
     mgDebugInitialise();
 
-    // SANITY CHECKS for cudacpp code generation (see issues #272 and #343 and PRs #619 and #626)
+    // SANITY CHECKS for cudacpp code generation (see issues #272 and #343 and PRs #619, #626, #360 and #396)
     // These variable are not used anywhere else in the code and their scope is limited to this sanity check
     {
-      // nprocesses has always been found equal to 1 for all processes generated so far in all code versions
+      // nprocesses>1 was last observed for "mirror processes" in uux_ttx in the 270 branch (see issue #343 and PRs #360 and #396)
       constexpr int nprocesses = %(nproc)i;
       static_assert( nprocesses == 1, "Assume nprocesses == 1" );
       // process_id corresponds to the index of DSIG1 Fortran functions (must be 1 because cudacpp is unable to handle DSIG2)