gpu-compute: fix uninitialized member bug in GPUDynInst

the n_reg field in the GPUDynInst is not currently set in the constructor.
if it is not set externally, there are assertion failures that may occur
if the random value it gets is just right. here we set it to 0 by default.

diff --git a/src/gpu-compute/gpu_dyn_inst.cc b/src/gpu-compute/gpu_dyn_inst.cc
index 83e348dbedf221b9486a8ade098a430f951560c8..594e45907ee6d88fc8bb98faf4e0020b63dfbe4d 100644 (file)
--- a/src/gpu-compute/gpu_dyn_inst.cc
+++ b/src/gpu-compute/gpu_dyn_inst.cc
@@ -43,7 +43,7 @@
 GPUDynInst::GPUDynInst(ComputeUnit *_cu, Wavefront *_wf,
                        GPUStaticInst *_staticInst, uint64_t instSeqNum)
     : GPUExecContext(_cu, _wf), m_op(Enums::MO_UNDEF),
-      memoryOrder(Enums::MEMORY_ORDER_NONE), useContinuation(false),
+      memoryOrder(Enums::MEMORY_ORDER_NONE), n_reg(0) ,useContinuation(false),
       statusBitVector(0), staticInst(_staticInst), _seqNum(instSeqNum)
 {
     tlbHitLevel.assign(VSZ, -1);