No special handling for PNF for RX procedures

See also a later commit "Allow to use multiple PUCCH at once", in which
PUCCH handling in the case of PNF is reworked.

We might squash both commits.
This commit is contained in:
Robert Schmidt
2025-01-27 10:59:15 +01:00
parent a6b40ace23
commit 84d9df501d
@@ -730,8 +730,6 @@ int phy_procedures_gNB_uespec_RX(PHY_VARS_gNB *gNB, int frame_rx, int slot_rx, N
for (int i = 0; i < gNB->max_nb_pucch; i++) {
NR_gNB_PUCCH_t *pucch = &gNB->pucch[i];
if (pucch) {
if (NFAPI_MODE == NFAPI_MODE_PNF)
pucch->frame = frame_rx;
if (pucch->active && (pucch->frame == frame_rx) && (pucch->slot == slot_rx)) {
c16_t **rxdataF = gNB->common_vars.rxdataF[pucch->beam_nb];
pucch_decode_done = 1;