diff options
| -rw-r--r-- | src/gpgpu-sim/shader.cc | 14 |
1 files changed, 13 insertions, 1 deletions
diff --git a/src/gpgpu-sim/shader.cc b/src/gpgpu-sim/shader.cc index 9f9a8a3..f02e204 100644 --- a/src/gpgpu-sim/shader.cc +++ b/src/gpgpu-sim/shader.cc @@ -1016,7 +1016,8 @@ void ldst_unit::writeback() } } - for( unsigned c=0; m_next_wb.empty() && (c < m_num_writeback_clients); c++ ) { + unsigned serviced_client = -1; + for( unsigned c = 0; m_next_wb.empty() && (c < m_num_writeback_clients); c++ ) { unsigned next_client = (c+m_writeback_arb)%m_num_writeback_clients; switch( next_client ) { case 0: // shared memory @@ -1024,6 +1025,7 @@ void ldst_unit::writeback() m_next_wb = *m_pipeline_reg[0]; m_core->dec_inst_in_pipeline(m_pipeline_reg[0]->warp_id()); m_pipeline_reg[0]->clear(); + serviced_client = next_client; } break; case 1: // texture response @@ -1031,6 +1033,7 @@ void ldst_unit::writeback() mem_fetch *mf = m_L1T->next_access(); m_next_wb = mf->get_inst(); delete mf; + serviced_client = next_client; } break; case 2: // const cache response @@ -1038,6 +1041,7 @@ void ldst_unit::writeback() mem_fetch *mf = m_L1C->next_access(); m_next_wb = mf->get_inst(); delete mf; + serviced_client = next_client; } break; case 3: // global/local @@ -1047,6 +1051,7 @@ void ldst_unit::writeback() m_core->decrement_atomic_count(m_next_global->get_wid(),m_next_global->get_access_warp_mask().count()); delete m_next_global; m_next_global = NULL; + serviced_client = next_client; } break; case 4: @@ -1054,11 +1059,18 @@ void ldst_unit::writeback() mem_fetch *mf = m_L1D->next_access(); m_next_wb = mf->get_inst(); delete mf; + serviced_client = next_client; } break; default: abort(); } } + // update arbitration priority only if: + // 1. the writeback buffer was available + // 2. a client was serviced + if (serviced_client != (unsigned)-1) { + m_writeback_arb = (serviced_client + 1) % m_num_writeback_clients; + } } unsigned ldst_unit::clock_multiplier() const |