Fix wrong slice_allocation loop in fairRR scheduler
This commit is contained in:
committed by
Robert Schmidt
parent
70e1841bb1
commit
c1359a103f
@@ -544,7 +544,7 @@ void dlsch_scheduler_pre_processor_fairRR (module_id_t Mod_id,
|
||||
unsigned char MIMO_mode_indicator[MAX_NUM_CCs][N_RBG_MAX];
|
||||
uint8_t slice_allocation[MAX_NUM_CCs][N_RBG_MAX];
|
||||
int UE_id, i;
|
||||
uint16_t j;
|
||||
uint16_t j,c;
|
||||
uint16_t nb_rbs_required[MAX_NUM_CCs][NUMBER_OF_UE_MAX];
|
||||
uint16_t nb_rbs_required_remaining[MAX_NUM_CCs][NUMBER_OF_UE_MAX];
|
||||
// uint16_t nb_rbs_required_remaining_1[MAX_NUM_CCs][NUMBER_OF_UE_MAX];
|
||||
@@ -658,9 +658,9 @@ void dlsch_scheduler_pre_processor_fairRR (module_id_t Mod_id,
|
||||
|
||||
/* slicing support has been introduced into the scheduler. Provide dummy
|
||||
* data so that the preprocessor "simply works" */
|
||||
for (int i = 0; i < MAX_NUM_CCs; ++i)
|
||||
for (int j = 0; i < N_RBG_MAX; ++j)
|
||||
slice_allocation[i][j] = 1;
|
||||
for (c = 0; c < MAX_NUM_CCs; ++c)
|
||||
for (j = 0; j < N_RBG_MAX; ++j)
|
||||
slice_allocation[c][j] = 1;
|
||||
|
||||
LOG_T(MAC,"calling dlsch_scheduler_pre_processor_allocate .. \n ");
|
||||
dlsch_scheduler_pre_processor_allocate (Mod_id,
|
||||
|
||||
Reference in New Issue
Block a user