diff options
| author | aamir <[email protected]> | 2018-11-02 23:35:50 -0700 |
|---|---|---|
| committer | aamir <[email protected]> | 2018-11-02 23:35:50 -0700 |
| commit | a77c16f1ef6efc0b58eb14273aa52663eb7845b3 (patch) | |
| tree | b10f42d03877ecb45c296ec8120f325ef9ed01d4 | |
| parent | 9807c9393164bf01b14fcda0c35d75b248749c4f (diff) | |
bug in the last commit
| -rw-r--r-- | src/cuda-sim/cuda-sim.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/src/cuda-sim/cuda-sim.cc b/src/cuda-sim/cuda-sim.cc index b2b93e8..d5c5c3d 100644 --- a/src/cuda-sim/cuda-sim.cc +++ b/src/cuda-sim/cuda-sim.cc @@ -1341,7 +1341,7 @@ void ptx_thread_info::ptx_exec_inst( warp_inst_t &inst, unsigned lane_id) } //Tensorcore is warp synchronous operation. So these instructions needs to be executed only once. To make the simulation faster removing the redundant tensorcore operation - if(!tensorcore_op(inst_opcode)||(tensorcore_op)&&(lane_id==0)){ + if(!tensorcore_op(inst_opcode)||(tensorcore_op(inst_opcode))&&(lane_id==0)){ switch ( inst_opcode ) { #define OP_DEF(OP,FUNC,STR,DST,CLASSIFICATION) case OP: FUNC(pI,this); op_classification = CLASSIFICATION; break; #define OP_W_DEF(OP,FUNC,STR,DST,CLASSIFICATION) case OP: FUNC(pI,get_core(),inst); op_classification = CLASSIFICATION; break; |