1. refactoring cuda api code for loading PTX (removing external PTX loading entirely)

2. some bug fixes for warp_inst_t
3. creating a new class, gpgpu_t, which contains the functional "memory" state visible
   to all threads running on a GPU (doing this as part of my continuing effort to hunt 
   down and eradicate every global variable that is not the top level "the gpu")
4. other misc. changes

Almost passing CUDA 3.1 regression?  oclHistogram keeps failing under torque, but
does not fail when run on the command line from the same directory.

[git-p4: depot-paths = "//depot/gpgpu_sim_research/fermi/distribution/": change = 7827]

1 files changed, 0 insertions, 15 deletions

diff --git a/src/cuda-sim/ptx_ir.h b/src/cuda-sim/ptx_ir.h
index 6010caf..612326c 100644
--- a/src/cuda-sim/ptx_ir.h
+++ b/src/cuda-sim/ptx_ir.h
@@ -803,7 +803,6 @@ public:
                     const char *source,
                     unsigned warp_size );
 
-
    void print_insn() const;
    virtual void print_insn( FILE *fp ) const;
    unsigned inst_size() const { return m_inst_size; }
@@ -1404,20 +1403,6 @@ struct textureInfo {
 
 extern std::map<std::string,symbol_table*> g_sym_name_to_symbol_table;
 
-#define GLOBAL_HEAP_START 0x80000000
-   // start allocating from this address (lower values used for allocating globals in .ptx file)
-#define SHARED_MEM_SIZE_MAX (64*1024)
-#define LOCAL_MEM_SIZE_MAX (16*1024)
-#define MAX_STREAMING_MULTIPROCESSORS 64
-#define MAX_THREAD_PER_SM 1024
-#define TOTAL_LOCAL_MEM_PER_SM (MAX_THREAD_PER_SM*LOCAL_MEM_SIZE_MAX)
-#define TOTAL_SHARED_MEM (MAX_STREAMING_MULTIPROCESSORS*SHARED_MEM_SIZE_MAX)
-#define TOTAL_LOCAL_MEM (MAX_STREAMING_MULTIPROCESSORS*MAX_THREAD_PER_SM*LOCAL_MEM_SIZE_MAX)
-#define SHARED_GENERIC_START (GLOBAL_HEAP_START-TOTAL_SHARED_MEM)
-#define LOCAL_GENERIC_START (SHARED_GENERIC_START-TOTAL_LOCAL_MEM)
-#define STATIC_ALLOC_LIMIT (GLOBAL_HEAP_START - (TOTAL_LOCAL_MEM+TOTAL_SHARED_MEM))
-
-
 
 extern bool g_keep_intermediate_files;