- simple device printf support added (supports %u, %d, %f)

- add support to call_impl for passing values in through local param space (need to add support for return)
- track local framesize during ptx parsing / update local stack-pointer during call/return
- detect appropriate param space for ld.param (and st.param) depending upon address symbol's scope
- putting back the crazy math for local mem allocation (seems to be required)
- bug fix for global byte array initialization
- force ld and st w/o space specifier to be generic_space
- make type_decode method of type_info_key
- adjust debug printing so listing of code and CFG information requires higher numbers (100 and 50)
- local param ld/st accesses treated like local ld/st for timing model

[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 6878]

1 files changed, 3 insertions, 6 deletions

diff --git a/src/gpgpu-sim/shader.cc b/src/gpgpu-sim/shader.cc
index 3f084f8..05a8b0e 100644
--- a/src/gpgpu-sim/shader.cc
+++ b/src/gpgpu-sim/shader.cc
@@ -1659,8 +1659,7 @@ inline int is_const ( memory_space_t space ) {
 }
 
 inline int is_local ( memory_space_t space ) {
-   assert( space != param_space_local ); // todo: map local param memory to linear address space
-   return((space) == local_space);
+   return (space == local_space) || (space == param_space_local);
 }
 
 inline int is_param ( memory_space_t space ) {
@@ -2481,12 +2480,10 @@ mem_stage_stall_type send_mem_request(shader_core_ctx_t *shader, mem_access_t &a
       access_type = (access.iswrite)? GLOBAL_ACC_W: GLOBAL_ACC_R;   
       break;
    case local_space:
+   case param_space_local:
       code = DCACHE;
       access_type = (access.iswrite)? LOCAL_ACC_W: LOCAL_ACC_R;   
       break;
-   case param_space_local:
-      abort(); // todo: define mapping of local param space to linear memory ?
-      break;
    default:
       assert(0); // NOT A MEM SPACE;
       break; 
@@ -2837,10 +2834,10 @@ void shader_memory_queue(shader_core_ctx_t *shader, shader_queues_t *accessqs)
             break;
          case global_space:
          case local_space:
+         case param_space_local:
             path[i] = GLOBAL_MEM_PATH;
             type_counts[GLOBAL_MEM_PATH]++;
             break;
-         case param_space_local:
          case param_space_unclassified:
             abort(); // todo: define access details
             break;