|
|
@@ -216,8 +216,7 @@ static void compute_all_performance_predictions(struct starpu_task *task,
|
|
|
double local_data_penalty[STARPU_NMAXWORKERS][STARPU_MAXIMPLEMENTATIONS],
|
|
|
double local_power[STARPU_NMAXWORKERS][STARPU_MAXIMPLEMENTATIONS],
|
|
|
int *forced_worker, unsigned int *forced_impl,
|
|
|
- struct starpu_task_bundle *bundle,
|
|
|
- unsigned int *nimpls)
|
|
|
+ struct starpu_task_bundle *bundle)
|
|
|
{
|
|
|
int calibrating = 0;
|
|
|
double max_exp_end = DBL_MIN;
|
|
|
@@ -233,7 +232,6 @@ static void compute_all_performance_predictions(struct starpu_task *task,
|
|
|
unsigned nimpl;
|
|
|
|
|
|
for (worker = 0; worker < nworkers; worker++) {
|
|
|
- nimpls[worker] = 0;
|
|
|
for (nimpl = 0; nimpl <STARPU_MAXIMPLEMENTATIONS; nimpl++) {
|
|
|
/* Sometimes workers didn't take the tasks as early as we expected */
|
|
|
exp_start[worker] = STARPU_MAX(exp_start[worker], starpu_timing_now());
|
|
|
@@ -295,11 +293,8 @@ static void compute_all_performance_predictions(struct starpu_task *task,
|
|
|
exp_end[worker][nimpl] = exp_start[worker] + exp_len[worker] + local_task_length[worker][nimpl];
|
|
|
|
|
|
if (exp_end[worker][nimpl] < best_exp_end)
|
|
|
- {
|
|
|
/* a better solution was found */
|
|
|
best_exp_end = exp_end[worker][nimpl];
|
|
|
- nimpls[worker] = nimpl;
|
|
|
- }
|
|
|
|
|
|
if (local_power[worker][nimpl] == -1.0)
|
|
|
local_power[worker][nimpl] = 0.;
|
|
|
@@ -330,7 +325,6 @@ static int _heft_push_task(struct starpu_task *task, unsigned prio)
|
|
|
double local_power[nworkers][STARPU_MAXIMPLEMENTATIONS];
|
|
|
double exp_end[nworkers][STARPU_MAXIMPLEMENTATIONS];
|
|
|
double max_exp_end = 0.0;
|
|
|
- unsigned int nimpls[nworkers];
|
|
|
|
|
|
double best_exp_end;
|
|
|
|
|
|
@@ -345,7 +339,7 @@ static int _heft_push_task(struct starpu_task *task, unsigned prio)
|
|
|
&max_exp_end, &best_exp_end,
|
|
|
local_data_penalty,
|
|
|
local_power, &forced_worker, &forced_impl,
|
|
|
- bundle, nimpls);
|
|
|
+ bundle);
|
|
|
|
|
|
/* If there is no prediction available for that task with that arch we
|
|
|
* want to speed-up calibration time so we force this measurement */
|
|
|
@@ -405,8 +399,7 @@ static int _heft_push_task(struct starpu_task *task, unsigned prio)
|
|
|
* length for the entire bundle, but not for the task alone. */
|
|
|
enum starpu_perf_archtype perf_arch = starpu_worker_get_perf_archtype(best);
|
|
|
unsigned memory_node = starpu_worker_get_memory_node(best);
|
|
|
- model_best = starpu_task_expected_length(task, perf_arch,
|
|
|
- _starpu_get_job_associated_to_task(task)->nimpl);
|
|
|
+ model_best = starpu_task_expected_length(task, perf_arch, selected_impl);
|
|
|
transfer_model_best = starpu_task_expected_data_transfer_time(memory_node, task);
|
|
|
|
|
|
/* Remove the task from the bundle since we have made a
|
|
|
@@ -427,7 +420,7 @@ static int _heft_push_task(struct starpu_task *task, unsigned prio)
|
|
|
}
|
|
|
|
|
|
|
|
|
- _starpu_get_job_associated_to_task(task)->nimpl = nimpls[best];
|
|
|
+ _starpu_get_job_associated_to_task(task)->nimpl = selected_impl;
|
|
|
|
|
|
return push_task_on_best_worker(task, best, model_best, transfer_model_best, prio);
|
|
|
}
|
