log migrations when affinity is off

2 files changed, 53 insertions, 9 deletions

diff --git a/gpu/gpuspin.cu b/gpu/gpuspin.cu
index f361b86..304d937 100644
--- a/gpu/gpuspin.cu
+++ b/gpu/gpuspin.cu
@@ -26,6 +26,18 @@ using namespace ranlib;
 
 #define ms2s(ms)  ((ms)*0.001)
 
+const unsigned int TOKEN_START = 100;
+const unsigned int TOKEN_END = 101;
+
+const unsigned int EE_START = 200;
+const unsigned int EE_END = 201;
+
+const unsigned int CE_SEND_START = 300;
+const unsigned int CE_SEND_END = 301;
+
+const unsigned int CE_RECV_START = 400;
+const unsigned int CE_RECV_END = 401;
+
 bool SILENT = true;
 inline int xprintf(const char *format, ...)
 {
@@ -382,6 +394,12 @@ static cudaError_t __chunkMemcpy(void* a_dst, const void* a_src, size_t count,
                                         //cudaStreamSynchronize(STREAMS[CUR_DEVICE]);
                                         cudaEventSynchronize(EVENTS[CUR_DEVICE]);
                                         ret = cudaGetLastError();
+
+                                        if (kind == cudaMemcpyDeviceToHost || kind == cudaMemcpyDeviceToDevice)
+                                                inject_action(CE_RECV_END);
+                                        if (kind == cudaMemcpyHostToDevice)
+                                                inject_action(CE_SEND_END);
+
                                         state->unlock();
                                         if(ret != cudaSuccess)
                                                 break;
@@ -395,8 +413,13 @@ static cudaError_t __chunkMemcpy(void* a_dst, const void* a_src, size_t count,
                         }
                 }
 
-                if(state && !state->locked)
+                if(state && !state->locked) {
                         state->lock();
+                        if (kind == cudaMemcpyDeviceToHost || kind == cudaMemcpyDeviceToDevice)
+                                inject_action(CE_RECV_START);
+                        if (kind == cudaMemcpyHostToDevice)
+                                inject_action(CE_SEND_START);
+                }
 
         //ret = cudaMemcpy(dst+i*chunk_size, src+i*chunk_size, bytesToCopy, kind);
                 cudaMemcpyAsync(dst+i*chunk_size, src+i*chunk_size, bytesToCopy, kind, STREAMS[CUR_DEVICE]);
@@ -429,11 +452,23 @@ static cudaError_t chunkMemcpy(void* a_dst, const void* a_src, size_t count,
         else {
                 ce_lock_state state(device_a, kind, count, device_b, migration);
                 state.lock();
+
+                if (kind == cudaMemcpyDeviceToHost || kind == cudaMemcpyDeviceToDevice)
+                        inject_action(CE_RECV_START);
+                if (kind == cudaMemcpyHostToDevice)
+                        inject_action(CE_SEND_START);
+
                 ret = __chunkMemcpy(a_dst, a_src, count, kind, &state);
                 cudaEventSynchronize(cur_event());
                 //              cudaStreamSynchronize(cur_stream());
                 if(ret == cudaSuccess)
                         ret = cudaGetLastError();
+
+                if (kind == cudaMemcpyDeviceToHost || kind == cudaMemcpyDeviceToDevice)
+                        inject_action(CE_RECV_END);
+                if (kind == cudaMemcpyHostToDevice)
+                        inject_action(CE_SEND_END);
+
                 state.unlock();
         }
         return ret;
@@ -1157,6 +1192,7 @@ static void gpu_loop_for(double gpu_sec_time, unsigned int num_kernels, double e
                 goto out;
 
         next_gpu = litmus_lock(TOKEN_LOCK);
+        inject_action(TOKEN_START);
         {
                 MigrateIfNeeded(next_gpu);
                 unsigned int numcycles = ((unsigned int)(cur_hz() * gpu_sec_time))/num_kernels;
@@ -1170,6 +1206,7 @@ static void gpu_loop_for(double gpu_sec_time, unsigned int num_kernels, double e
                 {
                         if(useEngineLocks() && !locked) {
                                 litmus_lock(cur_ee());
+                                inject_action(EE_START);
                                 locked = true;
                         }
 
@@ -1180,6 +1217,7 @@ static void gpu_loop_for(double gpu_sec_time, unsigned int num_kernels, double e
 //                              cudaStreamSynchronize(cur_stream());
                                 cudaEventRecord(cur_event(), cur_stream());
                                 cudaEventSynchronize(cur_event());
+                                inject_action(EE_END);
                                 litmus_unlock(cur_ee());
                                 locked = false;
                         }
@@ -1187,6 +1225,7 @@ static void gpu_loop_for(double gpu_sec_time, unsigned int num_kernels, double e
                 if (locked) {
                         cudaEventRecord(cur_event(), cur_stream());
                         cudaEventSynchronize(cur_event());
+                        inject_action(EE_END);
                         litmus_unlock(cur_ee());
                         locked = false;
                 }
@@ -1198,6 +1237,7 @@ static void gpu_loop_for(double gpu_sec_time, unsigned int num_kernels, double e
                 if (MIGRATE_VIA_SYSMEM)
                         PullState();
         }
+        inject_action(TOKEN_END);
         litmus_unlock(TOKEN_LOCK);
 
         last_gpu() = cur_gpu();
@@ -1696,6 +1736,11 @@ void apply_args(struct Args* args)
         CHUNK_SIZE = args->chunk_size;
         MIGRATE_VIA_SYSMEM = args->use_sysmem_migration;
 
+        if (args->scheduler == LITMUS && !ENABLE_AFFINITY)
+                TRACE_MIGRATIONS = true;
+        else if (args->scheduler == LITMUS)
+                TRACE_MIGRATIONS = false;
+
         // roll back other globals to an initial state
         CUR_DEVICE = -1;
         LAST_DEVICE = -1;
@@ -2089,8 +2134,10 @@ int init_daemon(struct Args* args, int num_total_users, bool is_daemon)
                 pthread_mutex_unlock(daemon_mutex);
         }
 
-        if (!my_run_entry)
+        if (!my_run_entry) {
+                fprintf(stderr, "Could not find task <wcet, gpu_wcet, period>: <%f %f %f>\n", args->wcet_ms, args->gpu_wcet_ms, args->period_ms);
                 return -1;
+        }
         return 0;
 }
 
diff --git a/include/litmus.h b/include/litmus.h
index a6c2b13..e785f92 100644
--- a/include/litmus.h
+++ b/include/litmus.h
@@ -270,19 +270,16 @@ int inject_completion(unsigned int job_no);
 int inject_gpu_migration(unsigned int to, unsigned int from);
 int __inject_action(unsigned int action);
 
-/*
+#if 1
 #define inject_action(COUNT) \
 do { \
-unsigned int temp = (COUNT); \
-printf("%s:%d:%d\n",__FUNCTION__,__LINE__,temp); \
-__inject_action(temp); \
+__inject_action(COUNT); \
 }while(0);
-*/
-
+#else
 #define inject_action(COUNT) \
 do { \
 }while(0);
-
+#endif
 
 /* Litmus signal handling */