fixing wrong interpretation of support for 256qam table
This commit is contained in:
@@ -184,31 +184,7 @@ void set_dl_mcs_table(int scs, NR_UE_NR_Capability_t *cap,
|
||||
else if (cap->phy_Parameters.phy_ParametersFR1 && cap->phy_Parameters.phy_ParametersFR1->pdsch_256QAM_FR1)
|
||||
supported = true;
|
||||
|
||||
// check featureSet
|
||||
NR_FeatureSets_t *fs=cap->featureSets;
|
||||
NR_ModulationOrder_t supported_mo = 0;
|
||||
if (fs && supported) {
|
||||
// go through DL feature sets and look for one with current SCS
|
||||
for (int i=0;i<fs->featureSetsDownlinkPerCC->list.count;i++) {
|
||||
int supported_bw;
|
||||
NR_FeatureSetDownlinkPerCC_t *fs_cc = fs->featureSetsDownlinkPerCC->list.array[i];
|
||||
switch (fs_cc->supportedBandwidthDL.present) {
|
||||
case NR_SupportedBandwidth_PR_fr1:
|
||||
supported_bw = fs_cc->supportedBandwidthDL.choice.fr1;
|
||||
break;
|
||||
case NR_SupportedBandwidth_PR_fr2:
|
||||
supported_bw = fs_cc->supportedBandwidthDL.choice.fr2;
|
||||
break;
|
||||
default:
|
||||
AssertFatal(1==0,"Invalid parameter for supported bandwith choice\n");
|
||||
}
|
||||
if (fs_cc->supportedSubcarrierSpacingDL == scs &&
|
||||
supported_bw == bw &&
|
||||
fs_cc->supportedModulationOrderDL)
|
||||
supported_mo = *fs_cc->supportedModulationOrderDL ;
|
||||
}
|
||||
}
|
||||
if (supported && (supported_mo == NR_ModulationOrder_qam256)) {
|
||||
if (supported) {
|
||||
if(bwp_Dedicated->pdsch_Config->choice.setup->mcs_Table == NULL)
|
||||
bwp_Dedicated->pdsch_Config->choice.setup->mcs_Table = calloc(1, sizeof(*bwp_Dedicated->pdsch_Config->choice.setup->mcs_Table));
|
||||
*bwp_Dedicated->pdsch_Config->choice.setup->mcs_Table = NR_PDSCH_Config__mcs_Table_qam256;
|
||||
|
||||
Reference in New Issue
Block a user