diff --git a/autobuild_mac80211_release/mt7986_bersa_mac80211/package/kernel/mt76/src/bersa/mac.c b/autobuild_mac80211_release/mt7986_bersa_mac80211/package/kernel/mt76/src/bersa/mac.c
new file mode 100644
index 0000000..745ee3a
--- /dev/null
+++ b/autobuild_mac80211_release/mt7986_bersa_mac80211/package/kernel/mt76/src/bersa/mac.c
@@ -0,0 +1,2644 @@
+// SPDX-License-Identifier: ISC
+/* Copyright (C) 2020 MediaTek Inc. */
+
+#include <linux/etherdevice.h>
+#include <linux/timekeeping.h>
+#include "bersa.h"
+#include "../dma.h"
+#include "mac.h"
+#include "mcu.h"
+
+#define to_rssi(field, rxv)	((FIELD_GET(field, rxv) - 220) / 2)
+
+#define HE_BITS(f)		cpu_to_le16(IEEE80211_RADIOTAP_HE_##f)
+#define HE_PREP(f, m, v)	le16_encode_bits(le32_get_bits(v, MT_CRXV_HE_##m),\
+						 IEEE80211_RADIOTAP_HE_##f)
+
+static const struct bersa_dfs_radar_spec etsi_radar_specs = {
+	.pulse_th = { 110, -10, -80, 40, 5200, 128, 5200 },
+	.radar_pattern = {
+		[5] =  { 1, 0,  6, 32, 28, 0,  990, 5010, 17, 1, 1 },
+		[6] =  { 1, 0,  9, 32, 28, 0,  615, 5010, 27, 1, 1 },
+		[7] =  { 1, 0, 15, 32, 28, 0,  240,  445, 27, 1, 1 },
+		[8] =  { 1, 0, 12, 32, 28, 0,  240,  510, 42, 1, 1 },
+		[9] =  { 1, 1,  0,  0,  0, 0, 2490, 3343, 14, 0, 0, 12, 32, 28, { }, 126 },
+		[10] = { 1, 1,  0,  0,  0, 0, 2490, 3343, 14, 0, 0, 15, 32, 24, { }, 126 },
+		[11] = { 1, 1,  0,  0,  0, 0,  823, 2510, 14, 0, 0, 18, 32, 28, { },  54 },
+		[12] = { 1, 1,  0,  0,  0, 0,  823, 2510, 14, 0, 0, 27, 32, 24, { },  54 },
+	},
+};
+
+static const struct bersa_dfs_radar_spec fcc_radar_specs = {
+	.pulse_th = { 110, -10, -80, 40, 5200, 128, 5200 },
+	.radar_pattern = {
+		[0] = { 1, 0,  8,  32, 28, 0, 508, 3076, 13, 1,  1 },
+		[1] = { 1, 0, 12,  32, 28, 0, 140,  240, 17, 1,  1 },
+		[2] = { 1, 0,  8,  32, 28, 0, 190,  510, 22, 1,  1 },
+		[3] = { 1, 0,  6,  32, 28, 0, 190,  510, 32, 1,  1 },
+		[4] = { 1, 0,  9, 255, 28, 0, 323,  343, 13, 1, 32 },
+	},
+};
+
+static const struct bersa_dfs_radar_spec jp_radar_specs = {
+	.pulse_th = { 110, -10, -80, 40, 5200, 128, 5200 },
+	.radar_pattern = {
+		[0] =  { 1, 0,  8,  32, 28, 0,  508, 3076,  13, 1,  1 },
+		[1] =  { 1, 0, 12,  32, 28, 0,  140,  240,  17, 1,  1 },
+		[2] =  { 1, 0,  8,  32, 28, 0,  190,  510,  22, 1,  1 },
+		[3] =  { 1, 0,  6,  32, 28, 0,  190,  510,  32, 1,  1 },
+		[4] =  { 1, 0,  9, 255, 28, 0,  323,  343,  13, 1, 32 },
+		[13] = { 1, 0,  7,  32, 28, 0, 3836, 3856,  14, 1,  1 },
+		[14] = { 1, 0,  6,  32, 28, 0,  615, 5010, 110, 1,  1 },
+		[15] = { 1, 1,  0,   0,  0, 0,   15, 5010, 110, 0,  0, 12, 32, 28 },
+	},
+};
+
+static struct mt76_wcid *bersa_rx_get_wcid(struct bersa_dev *dev,
+					    u16 idx, bool unicast)
+{
+	struct bersa_sta *sta;
+	struct mt76_wcid *wcid;
+
+	if (idx >= ARRAY_SIZE(dev->mt76.wcid))
+		return NULL;
+
+	wcid = rcu_dereference(dev->mt76.wcid[idx]);
+	if (unicast || !wcid)
+		return wcid;
+
+	if (!wcid->sta)
+		return NULL;
+
+	sta = container_of(wcid, struct bersa_sta, wcid);
+	if (!sta->vif)
+		return NULL;
+
+	return &sta->vif->sta.wcid;
+}
+
+void bersa_sta_ps(struct mt76_dev *mdev, struct ieee80211_sta *sta, bool ps)
+{
+}
+
+bool bersa_mac_wtbl_update(struct bersa_dev *dev, int idx, u32 mask)
+{
+	mt76_rmw(dev, MT_WTBL_UPDATE, MT_WTBL_UPDATE_WLAN_IDX,
+		 FIELD_PREP(MT_WTBL_UPDATE_WLAN_IDX, idx) | mask);
+
+	return mt76_poll(dev, MT_WTBL_UPDATE, MT_WTBL_UPDATE_BUSY,
+			 0, 5000);
+}
+
+u32 bersa_mac_wtbl_lmac_addr(struct bersa_dev *dev, u16 wcid, u8 dw)
+{
+	mt76_wr(dev, MT_WTBLON_TOP_WDUCR,
+		FIELD_PREP(MT_WTBLON_TOP_WDUCR_GROUP, (wcid >> 7)));
+
+	return MT_WTBL_LMAC_OFFS(wcid, dw);
+}
+
+static void bersa_mac_sta_poll(struct bersa_dev *dev)
+{
+	static const u8 ac_to_tid[] = {
+		[IEEE80211_AC_BE] = 0,
+		[IEEE80211_AC_BK] = 1,
+		[IEEE80211_AC_VI] = 4,
+		[IEEE80211_AC_VO] = 6
+	};
+	struct ieee80211_sta *sta;
+	struct bersa_sta *msta;
+	struct rate_info *rate;
+	u32 tx_time[IEEE80211_NUM_ACS], rx_time[IEEE80211_NUM_ACS];
+	LIST_HEAD(sta_poll_list);
+	int i;
+
+	spin_lock_bh(&dev->sta_poll_lock);
+	list_splice_init(&dev->sta_poll_list, &sta_poll_list);
+	spin_unlock_bh(&dev->sta_poll_lock);
+
+	rcu_read_lock();
+
+	while (true) {
+		bool clear = false;
+		u32 addr, val;
+		u16 idx;
+		u8 bw;
+
+		spin_lock_bh(&dev->sta_poll_lock);
+		if (list_empty(&sta_poll_list)) {
+			spin_unlock_bh(&dev->sta_poll_lock);
+			break;
+		}
+		msta = list_first_entry(&sta_poll_list,
+					struct bersa_sta, poll_list);
+		list_del_init(&msta->poll_list);
+		spin_unlock_bh(&dev->sta_poll_lock);
+
+		idx = msta->wcid.idx;
+		addr = bersa_mac_wtbl_lmac_addr(dev, idx, 20);
+
+		for (i = 0; i < IEEE80211_NUM_ACS; i++) {
+			u32 tx_last = msta->airtime_ac[i];
+			u32 rx_last = msta->airtime_ac[i + 4];
+
+			msta->airtime_ac[i] = mt76_rr(dev, addr);
+			msta->airtime_ac[i + 4] = mt76_rr(dev, addr + 4);
+
+			tx_time[i] = msta->airtime_ac[i] - tx_last;
+			rx_time[i] = msta->airtime_ac[i + 4] - rx_last;
+
+			if ((tx_last | rx_last) & BIT(30))
+				clear = true;
+
+			addr += 8;
+		}
+
+		if (clear) {
+			bersa_mac_wtbl_update(dev, idx,
+					       MT_WTBL_UPDATE_ADM_COUNT_CLEAR);
+			memset(msta->airtime_ac, 0, sizeof(msta->airtime_ac));
+		}
+
+		if (!msta->wcid.sta)
+			continue;
+
+		sta = container_of((void *)msta, struct ieee80211_sta,
+				   drv_priv);
+		for (i = 0; i < IEEE80211_NUM_ACS; i++) {
+			u8 q = mt76_connac_lmac_mapping(i);
+			u32 tx_cur = tx_time[q];
+			u32 rx_cur = rx_time[q];
+			u8 tid = ac_to_tid[i];
+
+			if (!tx_cur && !rx_cur)
+				continue;
+
+			ieee80211_sta_register_airtime(sta, tid, tx_cur,
+						       rx_cur);
+		}
+
+		/*
+		 * We don't support reading GI info from txs packets.
+		 * For accurate tx status reporting and AQL improvement,
+		 * we need to make sure that flags match so polling GI
+		 * from per-sta counters directly.
+		 */
+		rate = &msta->wcid.rate;
+		addr = bersa_mac_wtbl_lmac_addr(dev, idx, 6);
+		val = mt76_rr(dev, addr);
+
+		switch (rate->bw) {
+		case RATE_INFO_BW_160:
+			bw = IEEE80211_STA_RX_BW_160;
+			break;
+		case RATE_INFO_BW_80:
+			bw = IEEE80211_STA_RX_BW_80;
+			break;
+		case RATE_INFO_BW_40:
+			bw = IEEE80211_STA_RX_BW_40;
+			break;
+		default:
+			bw = IEEE80211_STA_RX_BW_20;
+			break;
+		}
+
+		if (rate->flags & RATE_INFO_FLAGS_HE_MCS) {
+			u8 offs = 24 + 2 * bw;
+
+			rate->he_gi = (val & (0x3 << offs)) >> offs;
+		} else if (rate->flags &
+			   (RATE_INFO_FLAGS_VHT_MCS | RATE_INFO_FLAGS_MCS)) {
+			if (val & BIT(12 + bw))
+				rate->flags |= RATE_INFO_FLAGS_SHORT_GI;
+			else
+				rate->flags &= ~RATE_INFO_FLAGS_SHORT_GI;
+		}
+	}
+
+	rcu_read_unlock();
+}
+
+static void
+bersa_mac_decode_he_radiotap_ru(struct mt76_rx_status *status,
+				 struct ieee80211_radiotap_he *he,
+				 __le32 *rxv)
+{
+	u32 ru_h, ru_l;
+	u8 ru, offs = 0;
+
+	ru_l = FIELD_GET(MT_PRXV_HE_RU_ALLOC_L, le32_to_cpu(rxv[0]));
+	ru_h = FIELD_GET(MT_PRXV_HE_RU_ALLOC_H, le32_to_cpu(rxv[1]));
+	ru = (u8)(ru_l | ru_h << 4);
+
+	status->bw = RATE_INFO_BW_HE_RU;
+
+	switch (ru) {
+	case 0 ... 36:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_26;
+		offs = ru;
+		break;
+	case 37 ... 52:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_52;
+		offs = ru - 37;
+		break;
+	case 53 ... 60:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_106;
+		offs = ru - 53;
+		break;
+	case 61 ... 64:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_242;
+		offs = ru - 61;
+		break;
+	case 65 ... 66:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_484;
+		offs = ru - 65;
+		break;
+	case 67:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_996;
+		break;
+	case 68:
+		status->he_ru = NL80211_RATE_INFO_HE_RU_ALLOC_2x996;
+		break;
+	}
+
+	he->data1 |= HE_BITS(DATA1_BW_RU_ALLOC_KNOWN);
+	he->data2 |= HE_BITS(DATA2_RU_OFFSET_KNOWN) |
+		     le16_encode_bits(offs,
+				      IEEE80211_RADIOTAP_HE_DATA2_RU_OFFSET);
+}
+
+static void
+bersa_mac_decode_he_mu_radiotap(struct sk_buff *skb, __le32 *rxv)
+{
+	struct mt76_rx_status *status = (struct mt76_rx_status *)skb->cb;
+	static const struct ieee80211_radiotap_he_mu mu_known = {
+		.flags1 = HE_BITS(MU_FLAGS1_SIG_B_MCS_KNOWN) |
+			  HE_BITS(MU_FLAGS1_SIG_B_DCM_KNOWN) |
+			  HE_BITS(MU_FLAGS1_CH1_RU_KNOWN) |
+			  HE_BITS(MU_FLAGS1_SIG_B_SYMS_USERS_KNOWN),
+		.flags2 = HE_BITS(MU_FLAGS2_BW_FROM_SIG_A_BW_KNOWN),
+	};
+	struct ieee80211_radiotap_he_mu *he_mu = NULL;
+
+	status->flag |= RX_FLAG_RADIOTAP_HE_MU;
+
+	he_mu = skb_push(skb, sizeof(mu_known));
+	memcpy(he_mu, &mu_known, sizeof(mu_known));
+
+#define MU_PREP(f, v)	le16_encode_bits(v, IEEE80211_RADIOTAP_HE_MU_##f)
+
+	he_mu->flags1 |= MU_PREP(FLAGS1_SIG_B_MCS, status->rate_idx);
+	if (status->he_dcm)
+		he_mu->flags1 |= MU_PREP(FLAGS1_SIG_B_DCM, status->he_dcm);
+
+	he_mu->flags2 |= MU_PREP(FLAGS2_BW_FROM_SIG_A_BW, status->bw) |
+			 MU_PREP(FLAGS2_SIG_B_SYMS_USERS,
+				 le32_get_bits(rxv[2], MT_CRXV_HE_NUM_USER));
+
+	he_mu->ru_ch1[0] = le32_get_bits(rxv[3], MT_CRXV_HE_RU0);
+
+	if (status->bw >= RATE_INFO_BW_40) {
+		he_mu->flags1 |= HE_BITS(MU_FLAGS1_CH2_RU_KNOWN);
+		he_mu->ru_ch2[0] = le32_get_bits(rxv[3], MT_CRXV_HE_RU1);
+	}
+
+	if (status->bw >= RATE_INFO_BW_80) {
+		he_mu->ru_ch1[1] = le32_get_bits(rxv[3], MT_CRXV_HE_RU2);
+		he_mu->ru_ch2[1] = le32_get_bits(rxv[3], MT_CRXV_HE_RU3);
+	}
+}
+
+static void
+bersa_mac_decode_he_radiotap(struct sk_buff *skb, __le32 *rxv, u32 mode)
+{
+	struct mt76_rx_status *status = (struct mt76_rx_status *)skb->cb;
+	static const struct ieee80211_radiotap_he known = {
+		.data1 = HE_BITS(DATA1_DATA_MCS_KNOWN) |
+			 HE_BITS(DATA1_DATA_DCM_KNOWN) |
+			 HE_BITS(DATA1_STBC_KNOWN) |
+			 HE_BITS(DATA1_CODING_KNOWN) |
+			 HE_BITS(DATA1_LDPC_XSYMSEG_KNOWN) |
+			 HE_BITS(DATA1_DOPPLER_KNOWN) |
+			 HE_BITS(DATA1_SPTL_REUSE_KNOWN) |
+			 HE_BITS(DATA1_BSS_COLOR_KNOWN),
+		.data2 = HE_BITS(DATA2_GI_KNOWN) |
+			 HE_BITS(DATA2_TXBF_KNOWN) |
+			 HE_BITS(DATA2_PE_DISAMBIG_KNOWN) |
+			 HE_BITS(DATA2_TXOP_KNOWN),
+	};
+	struct ieee80211_radiotap_he *he = NULL;
+	u32 ltf_size = le32_get_bits(rxv[2], MT_CRXV_HE_LTF_SIZE) + 1;
+
+	status->flag |= RX_FLAG_RADIOTAP_HE;
+
+	he = skb_push(skb, sizeof(known));
+	memcpy(he, &known, sizeof(known));
+
+	he->data3 = HE_PREP(DATA3_BSS_COLOR, BSS_COLOR, rxv[14]) |
+		    HE_PREP(DATA3_LDPC_XSYMSEG, LDPC_EXT_SYM, rxv[2]);
+	he->data4 = HE_PREP(DATA4_SU_MU_SPTL_REUSE, SR_MASK, rxv[11]);
+	he->data5 = HE_PREP(DATA5_PE_DISAMBIG, PE_DISAMBIG, rxv[2]) |
+		    le16_encode_bits(ltf_size,
+				     IEEE80211_RADIOTAP_HE_DATA5_LTF_SIZE);
+	if (le32_to_cpu(rxv[0]) & MT_PRXV_TXBF)
+		he->data5 |= HE_BITS(DATA5_TXBF);
+	he->data6 = HE_PREP(DATA6_TXOP, TXOP_DUR, rxv[14]) |
+		    HE_PREP(DATA6_DOPPLER, DOPPLER, rxv[14]);
+
+	switch (mode) {
+	case MT_PHY_TYPE_HE_SU:
+		he->data1 |= HE_BITS(DATA1_FORMAT_SU) |
+			     HE_BITS(DATA1_UL_DL_KNOWN) |
+			     HE_BITS(DATA1_BEAM_CHANGE_KNOWN);
+
+		he->data3 |= HE_PREP(DATA3_BEAM_CHANGE, BEAM_CHNG, rxv[14]) |
+			     HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
+		break;
+	case MT_PHY_TYPE_HE_EXT_SU:
+		he->data1 |= HE_BITS(DATA1_FORMAT_EXT_SU) |
+			     HE_BITS(DATA1_UL_DL_KNOWN);
+
+		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
+		break;
+	case MT_PHY_TYPE_HE_MU:
+		he->data1 |= HE_BITS(DATA1_FORMAT_MU) |
+			     HE_BITS(DATA1_UL_DL_KNOWN);
+
+		he->data3 |= HE_PREP(DATA3_UL_DL, UPLINK, rxv[2]);
+		he->data4 |= HE_PREP(DATA4_MU_STA_ID, MU_AID, rxv[7]);
+
+		bersa_mac_decode_he_radiotap_ru(status, he, rxv);
+		bersa_mac_decode_he_mu_radiotap(skb, rxv);
+		break;
+	case MT_PHY_TYPE_HE_TB:
+		he->data1 |= HE_BITS(DATA1_FORMAT_TRIG) |
+			     HE_BITS(DATA1_SPTL_REUSE2_KNOWN) |
+			     HE_BITS(DATA1_SPTL_REUSE3_KNOWN) |
+			     HE_BITS(DATA1_SPTL_REUSE4_KNOWN);
+
+		he->data4 |= HE_PREP(DATA4_TB_SPTL_REUSE1, SR_MASK, rxv[11]) |
+			     HE_PREP(DATA4_TB_SPTL_REUSE2, SR1_MASK, rxv[11]) |
+			     HE_PREP(DATA4_TB_SPTL_REUSE3, SR2_MASK, rxv[11]) |
+			     HE_PREP(DATA4_TB_SPTL_REUSE4, SR3_MASK, rxv[11]);
+
+		bersa_mac_decode_he_radiotap_ru(status, he, rxv);
+		break;
+	default:
+		break;
+	}
+}
+
+/* The HW does not translate the mac header to 802.3 for mesh point */
+static int bersa_reverse_frag0_hdr_trans(struct sk_buff *skb, u16 hdr_gap)
+{
+	struct mt76_rx_status *status = (struct mt76_rx_status *)skb->cb;
+	struct ethhdr *eth_hdr = (struct ethhdr *)(skb->data + hdr_gap);
+	struct bersa_sta *msta = (struct bersa_sta *)status->wcid;
+	__le32 *rxd = (__le32 *)skb->data;
+	struct ieee80211_sta *sta;
+	struct ieee80211_vif *vif;
+	struct ieee80211_hdr hdr;
+	__le32 qos_ctrl, ht_ctrl;
+
+	if (FIELD_GET(MT_RXD3_NORMAL_ADDR_TYPE, le32_to_cpu(rxd[3])) !=
+	    MT_RXD3_NORMAL_U2M)
+		return -EINVAL;
+
+	if (!(le32_to_cpu(rxd[1]) & MT_RXD1_NORMAL_GROUP_4))
+		return -EINVAL;
+
+	if (!msta || !msta->vif)
+		return -EINVAL;
+
+	sta = container_of((void *)msta, struct ieee80211_sta, drv_priv);
+	vif = container_of((void *)msta->vif, struct ieee80211_vif, drv_priv);
+
+	/* store the info from RXD and ethhdr to avoid being overridden */
+	hdr.frame_control = FIELD_GET(MT_RXD6_FRAME_CONTROL, rxd[6]);
+	hdr.seq_ctrl = FIELD_GET(MT_RXD8_SEQ_CTRL, rxd[8]);
+	qos_ctrl = FIELD_GET(MT_RXD8_QOS_CTL, rxd[8]);
+	ht_ctrl = FIELD_GET(MT_RXD9_HT_CONTROL, rxd[9]);
+
+	hdr.duration_id = 0;
+	ether_addr_copy(hdr.addr1, vif->addr);
+	ether_addr_copy(hdr.addr2, sta->addr);
+	switch (le16_to_cpu(hdr.frame_control) &
+		(IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS)) {
+	case 0:
+		ether_addr_copy(hdr.addr3, vif->bss_conf.bssid);
+		break;
+	case IEEE80211_FCTL_FROMDS:
+		ether_addr_copy(hdr.addr3, eth_hdr->h_source);
+		break;
+	case IEEE80211_FCTL_TODS:
+		ether_addr_copy(hdr.addr3, eth_hdr->h_dest);
+		break;
+	case IEEE80211_FCTL_TODS | IEEE80211_FCTL_FROMDS:
+		ether_addr_copy(hdr.addr3, eth_hdr->h_dest);
+		ether_addr_copy(hdr.addr4, eth_hdr->h_source);
+		break;
+	default:
+		break;
+	}
+
+	skb_pull(skb, hdr_gap + sizeof(struct ethhdr) - 2);
+	if (eth_hdr->h_proto == cpu_to_be16(ETH_P_AARP) ||
+	    eth_hdr->h_proto == cpu_to_be16(ETH_P_IPX))
+		ether_addr_copy(skb_push(skb, ETH_ALEN), bridge_tunnel_header);
+	else if (eth_hdr->h_proto >= cpu_to_be16(ETH_P_802_3_MIN))
+		ether_addr_copy(skb_push(skb, ETH_ALEN), rfc1042_header);
+	else
+		skb_pull(skb, 2);
+
+	if (ieee80211_has_order(hdr.frame_control))
+		memcpy(skb_push(skb, 2), &ht_ctrl, 2);
+	if (ieee80211_is_data_qos(hdr.frame_control))
+		memcpy(skb_push(skb, 2), &qos_ctrl, 2);
+	if (ieee80211_has_a4(hdr.frame_control))
+		memcpy(skb_push(skb, sizeof(hdr)), &hdr, sizeof(hdr));
+	else
+		memcpy(skb_push(skb, sizeof(hdr) - 6), &hdr, sizeof(hdr) - 6);
+
+	return 0;
+}
+
+static int
+bersa_mac_fill_rx_rate(struct bersa_dev *dev,
+			struct mt76_rx_status *status,
+			struct ieee80211_supported_band *sband,
+			__le32 *rxv)
+{
+	u32 v0, v2;
+	u8 stbc, gi, bw, dcm, mode, nss;
+	int i, idx;
+	bool cck = false;
+
+	v0 = le32_to_cpu(rxv[0]);
+	v2 = le32_to_cpu(rxv[2]);
+
+	idx = i = FIELD_GET(MT_PRXV_TX_RATE, v0);
+	nss = FIELD_GET(MT_PRXV_NSTS, v0) + 1;
+
+	stbc = FIELD_GET(MT_PRXV_HT_STBC, v2);
+	gi = FIELD_GET(MT_PRXV_HT_SHORT_GI, v2);
+	mode = FIELD_GET(MT_PRXV_TX_MODE, v2);
+	dcm = FIELD_GET(MT_PRXV_DCM, v2);
+	bw = FIELD_GET(MT_PRXV_FRAME_MODE, v2);
+
+	switch (mode) {
+	case MT_PHY_TYPE_CCK:
+		cck = true;
+		fallthrough;
+	case MT_PHY_TYPE_OFDM:
+		i = mt76_get_rate(&dev->mt76, sband, i, cck);
+		break;
+	case MT_PHY_TYPE_HT_GF:
+	case MT_PHY_TYPE_HT:
+		status->encoding = RX_ENC_HT;
+		if (gi)
+			status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
+		if (i > 31)
+			return -EINVAL;
+		break;
+	case MT_PHY_TYPE_VHT:
+		status->nss = nss;
+		status->encoding = RX_ENC_VHT;
+		if (gi)
+			status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
+		if (i > 9)
+			return -EINVAL;
+		break;
+	case MT_PHY_TYPE_HE_MU:
+	case MT_PHY_TYPE_HE_SU:
+	case MT_PHY_TYPE_HE_EXT_SU:
+	case MT_PHY_TYPE_HE_TB:
+		status->nss = nss;
+		status->encoding = RX_ENC_HE;
+		i &= GENMASK(3, 0);
+
+		if (gi <= NL80211_RATE_INFO_HE_GI_3_2)
+			status->he_gi = gi;
+
+		status->he_dcm = dcm;
+		break;
+	default:
+		return -EINVAL;
+	}
+	status->rate_idx = i;
+
+	switch (bw) {
+	case IEEE80211_STA_RX_BW_20:
+		break;
+	case IEEE80211_STA_RX_BW_40:
+		if (mode & MT_PHY_TYPE_HE_EXT_SU &&
+		    (idx & MT_PRXV_TX_ER_SU_106T)) {
+			status->bw = RATE_INFO_BW_HE_RU;
+			status->he_ru =
+				NL80211_RATE_INFO_HE_RU_ALLOC_106;
+		} else {
+			status->bw = RATE_INFO_BW_40;
+		}
+		break;
+	case IEEE80211_STA_RX_BW_80:
+		status->bw = RATE_INFO_BW_80;
+		break;
+	case IEEE80211_STA_RX_BW_160:
+		status->bw = RATE_INFO_BW_160;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	status->enc_flags |= RX_ENC_FLAG_STBC_MASK * stbc;
+	if (mode < MT_PHY_TYPE_HE_SU && gi)
+		status->enc_flags |= RX_ENC_FLAG_SHORT_GI;
+
+	return 0;
+}
+
+static int
+bersa_mac_fill_rx(struct bersa_dev *dev, struct sk_buff *skb)
+{
+	struct mt76_rx_status *status = (struct mt76_rx_status *)skb->cb;
+	struct mt76_phy *mphy = &dev->mt76.phy;
+	struct bersa_phy *phy = &dev->phy;
+	struct ieee80211_supported_band *sband;
+	__le32 *rxd = (__le32 *)skb->data;
+	__le32 *rxv = NULL;
+	u32 mode = 0;
+	u32 rxd0 = le32_to_cpu(rxd[0]);
+	u32 rxd1 = le32_to_cpu(rxd[1]);
+	u32 rxd2 = le32_to_cpu(rxd[2]);
+	u32 rxd3 = le32_to_cpu(rxd[3]);
+	u32 rxd4 = le32_to_cpu(rxd[4]);
+	u32 csum_mask = MT_RXD0_NORMAL_IP_SUM | MT_RXD0_NORMAL_UDP_TCP_SUM;
+	bool unicast, insert_ccmp_hdr = false;
+	u8 remove_pad, amsdu_info;
+	bool hdr_trans;
+	u16 hdr_gap;
+	u16 seq_ctrl = 0;
+	u8 qos_ctl = 0;
+	__le16 fc = 0;
+	int idx;
+	u8 band_idx;
+
+	memset(status, 0, sizeof(*status));
+
+	band_idx = FIELD_GET(MT_RXD1_NORMAL_BAND_IDX, rxd1);
+	mphy = mt76_dev_phy_by_band(&dev->mt76, band_idx);
+	phy = mphy->priv;
+	status->phy_idx = mt76_get_phy_id(mphy);
+
+	if (!test_bit(MT76_STATE_RUNNING, &mphy->state))
+		return -EINVAL;
+
+	if (rxd2 & MT_RXD2_NORMAL_AMSDU_ERR)
+		return -EINVAL;
+
+	hdr_trans = rxd2 & MT_RXD2_NORMAL_HDR_TRANS;
+	if (hdr_trans && (rxd1 & MT_RXD1_NORMAL_CM))
+		return -EINVAL;
+
+	/* ICV error or CCMP/BIP/WPI MIC error */
+	if (rxd1 & MT_RXD1_NORMAL_ICV_ERR)
+		status->flag |= RX_FLAG_ONLY_MONITOR;
+
+	unicast = FIELD_GET(MT_RXD3_NORMAL_ADDR_TYPE, rxd3) == MT_RXD3_NORMAL_U2M;
+	idx = FIELD_GET(MT_RXD1_NORMAL_WLAN_IDX, rxd1);
+	status->wcid = bersa_rx_get_wcid(dev, idx, unicast);
+
+	if (status->wcid) {
+		struct bersa_sta *msta;
+
+		msta = container_of(status->wcid, struct bersa_sta, wcid);
+		spin_lock_bh(&dev->sta_poll_lock);
+		if (list_empty(&msta->poll_list))
+			list_add_tail(&msta->poll_list, &dev->sta_poll_list);
+		spin_unlock_bh(&dev->sta_poll_lock);
+	}
+
+	status->freq = mphy->chandef.chan->center_freq;
+	status->band = mphy->chandef.chan->band;
+	if (status->band == NL80211_BAND_5GHZ)
+		sband = &mphy->sband_5g.sband;
+	else
+		sband = &mphy->sband_2g.sband;
+
+	if (!sband->channels)
+		return -EINVAL;
+
+	if ((rxd0 & csum_mask) == csum_mask)
+		skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+	if (rxd1 & MT_RXD3_NORMAL_FCS_ERR)
+		status->flag |= RX_FLAG_FAILED_FCS_CRC;
+
+	if (rxd1 & MT_RXD1_NORMAL_TKIP_MIC_ERR)
+		status->flag |= RX_FLAG_MMIC_ERROR;
+
+	if (FIELD_GET(MT_RXD2_NORMAL_SEC_MODE, rxd2) != 0 &&
+	    !(rxd1 & (MT_RXD1_NORMAL_CLM | MT_RXD1_NORMAL_CM))) {
+		status->flag |= RX_FLAG_DECRYPTED;
+		status->flag |= RX_FLAG_IV_STRIPPED;
+		status->flag |= RX_FLAG_MMIC_STRIPPED | RX_FLAG_MIC_STRIPPED;
+	}
+
+	remove_pad = FIELD_GET(MT_RXD2_NORMAL_HDR_OFFSET, rxd2);
+
+	if (rxd2 & MT_RXD2_NORMAL_MAX_LEN_ERROR)
+		return -EINVAL;
+
+	rxd += 8;
+	if (rxd1 & MT_RXD1_NORMAL_GROUP_4) {
+		u32 v0 = le32_to_cpu(rxd[0]);
+		u32 v2 = le32_to_cpu(rxd[2]);
+
+		fc = cpu_to_le16(FIELD_GET(MT_RXD6_FRAME_CONTROL, v0));
+		qos_ctl = FIELD_GET(MT_RXD8_QOS_CTL, v2);
+		seq_ctrl = FIELD_GET(MT_RXD8_SEQ_CTRL, v2);
+
+		rxd += 4;
+		if ((u8 *)rxd - skb->data >= skb->len)
+			return -EINVAL;
+	}
+
+	if (rxd1 & MT_RXD1_NORMAL_GROUP_1) {
+		u8 *data = (u8 *)rxd;
+
+		if (status->flag & RX_FLAG_DECRYPTED) {
+			switch (FIELD_GET(MT_RXD2_NORMAL_SEC_MODE, rxd2)) {
+			case MT_CIPHER_AES_CCMP:
+			case MT_CIPHER_CCMP_CCX:
+			case MT_CIPHER_CCMP_256:
+				insert_ccmp_hdr =
+					FIELD_GET(MT_RXD2_NORMAL_FRAG, rxd2);
+				fallthrough;
+			case MT_CIPHER_TKIP:
+			case MT_CIPHER_TKIP_NO_MIC:
+			case MT_CIPHER_GCMP:
+			case MT_CIPHER_GCMP_256:
+				status->iv[0] = data[5];
+				status->iv[1] = data[4];
+				status->iv[2] = data[3];
+				status->iv[3] = data[2];
+				status->iv[4] = data[1];
+				status->iv[5] = data[0];
+				break;
+			default:
+				break;
+			}
+		}
+		rxd += 4;
+		if ((u8 *)rxd - skb->data >= skb->len)
+			return -EINVAL;
+	}
+
+	if (rxd1 & MT_RXD1_NORMAL_GROUP_2) {
+		status->timestamp = le32_to_cpu(rxd[0]);
+		status->flag |= RX_FLAG_MACTIME_START;
+
+		if (!(rxd2 & MT_RXD2_NORMAL_NON_AMPDU)) {
+			status->flag |= RX_FLAG_AMPDU_DETAILS;
+
+			/* all subframes of an A-MPDU have the same timestamp */
+			if (phy->rx_ampdu_ts != status->timestamp) {
+				if (!++phy->ampdu_ref)
+					phy->ampdu_ref++;
+			}
+			phy->rx_ampdu_ts = status->timestamp;
+
+			status->ampdu_ref = phy->ampdu_ref;
+		}
+
+		rxd += 4;
+		if ((u8 *)rxd - skb->data >= skb->len)
+			return -EINVAL;
+	}
+
+	/* RXD Group 3 - P-RXV */
+	if (rxd1 & MT_RXD1_NORMAL_GROUP_3) {
+		u32 v3;
+		int ret;
+
+		rxv = rxd;
+		rxd += 4;
+		if ((u8 *)rxd - skb->data >= skb->len)
+			return -EINVAL;
+
+		v3 = le32_to_cpu(rxv[3]);
+
+		status->chains = mphy->antenna_mask;
+		status->chain_signal[0] = to_rssi(MT_PRXV_RCPI0, v3);
+		status->chain_signal[1] = to_rssi(MT_PRXV_RCPI1, v3);
+		status->chain_signal[2] = to_rssi(MT_PRXV_RCPI2, v3);
+		status->chain_signal[3] = to_rssi(MT_PRXV_RCPI3, v3);
+
+		/* RXD Group 5 - C-RXV */
+		if (rxd1 & MT_RXD1_NORMAL_GROUP_5) {
+			rxd += 24;
+			if ((u8 *)rxd - skb->data >= skb->len)
+				return -EINVAL;
+		}
+
+		/* TODO: parse rx rate from rxv */
+		ret = bersa_mac_fill_rx_rate(dev, status, sband, rxv);
+		if (ret < 0)
+			return ret;
+	}
+
+	amsdu_info = FIELD_GET(MT_RXD4_NORMAL_PAYLOAD_FORMAT, rxd4);
+	status->amsdu = !!amsdu_info;
+	if (status->amsdu) {
+		status->first_amsdu = amsdu_info == MT_RXD4_FIRST_AMSDU_FRAME;
+		status->last_amsdu = amsdu_info == MT_RXD4_LAST_AMSDU_FRAME;
+	}
+
+	hdr_gap = (u8 *)rxd - skb->data + 2 * remove_pad;
+	if (hdr_trans && ieee80211_has_morefrags(fc)) {
+		if (bersa_reverse_frag0_hdr_trans(skb, hdr_gap))
+			return -EINVAL;
+		hdr_trans = false;
+	} else {
+		int pad_start = 0;
+
+		skb_pull(skb, hdr_gap);
+		if (!hdr_trans && status->amsdu) {
+			pad_start = ieee80211_get_hdrlen_from_skb(skb);
+		} else if (hdr_trans && (rxd2 & MT_RXD2_NORMAL_HDR_TRANS_ERROR)) {
+			/*
+			 * When header translation failure is indicated,
+			 * the hardware will insert an extra 2-byte field
+			 * containing the data length after the protocol
+			 * type field.
+			 */
+			pad_start = 12;
+			if (get_unaligned_be16(skb->data + pad_start) == ETH_P_8021Q)
+				pad_start += 4;
+
+			if (get_unaligned_be16(skb->data + pad_start) !=
+			    skb->len - pad_start - 2)
+				pad_start = 0;
+		}
+
+		if (pad_start) {
+			memmove(skb->data + 2, skb->data, pad_start);
+			skb_pull(skb, 2);
+		}
+	}
+
+	if (!hdr_trans) {
+		struct ieee80211_hdr *hdr;
+
+		if (insert_ccmp_hdr) {
+			u8 key_id = FIELD_GET(MT_RXD1_NORMAL_KEY_ID, rxd1);
+
+			mt76_insert_ccmp_hdr(skb, key_id);
+		}
+
+		hdr = mt76_skb_get_hdr(skb);
+		fc = hdr->frame_control;
+		if (ieee80211_is_data_qos(fc)) {
+			seq_ctrl = le16_to_cpu(hdr->seq_ctrl);
+			qos_ctl = *ieee80211_get_qos_ctl(hdr);
+		}
+	} else {
+		status->flag |= RX_FLAG_8023;
+	}
+
+	if (rxv && mode >= MT_PHY_TYPE_HE_SU && !(status->flag & RX_FLAG_8023))
+		bersa_mac_decode_he_radiotap(skb, rxv, mode);
+
+	if (!status->wcid || !ieee80211_is_data_qos(fc))
+		return 0;
+
+	/* drop no data frame */
+	if (fc & cpu_to_le16(IEEE80211_STYPE_NULLFUNC))
+		return -EINVAL;
+
+	status->aggr = unicast &&
+		       !ieee80211_is_qos_nullfunc(fc);
+	status->qos_ctl = qos_ctl;
+	status->seqno = IEEE80211_SEQ_TO_SN(seq_ctrl);
+
+	return 0;
+}
+
+static void
+bersa_mac_fill_rx_vector(struct bersa_dev *dev, struct sk_buff *skb)
+{
+#ifdef CONFIG_NL80211_TESTMODE
+	struct bersa_phy *phy = &dev->phy;
+	__le32 *rxd = (__le32 *)skb->data;
+	__le32 *rxv_hdr = rxd + 2;
+	__le32 *rxv = rxd + 4;
+	u32 rcpi, ib_rssi, wb_rssi, v20, v21;
+	u8 phy_idx;
+	s32 foe;
+	u8 snr;
+	int i;
+
+	phy_idx = FIELD_GET(MT_RXV_HDR_BAND_IDX, le32_to_cpu(rxv_hdr[1]));
+	if (phy_idx == MT_EXT_PHY)
+		phy = bersa_ext_phy(dev);
+
+	if (phy_idx == MT_TRI_PHY)
+		phy = bersa_tri_phy(dev);
+
+	rcpi = le32_to_cpu(rxv[6]);
+	ib_rssi = le32_to_cpu(rxv[7]);
+	wb_rssi = le32_to_cpu(rxv[8]) >> 5;
+
+	for (i = 0; i < 4; i++, rcpi >>= 8, ib_rssi >>= 8, wb_rssi >>= 9) {
+		if (i == 3)
+			wb_rssi = le32_to_cpu(rxv[9]);
+
+		phy->test.last_rcpi[i] = rcpi & 0xff;
+		phy->test.last_ib_rssi[i] = ib_rssi & 0xff;
+		phy->test.last_wb_rssi[i] = wb_rssi & 0xff;
+	}
+
+	v20 = le32_to_cpu(rxv[20]);
+	v21 = le32_to_cpu(rxv[21]);
+
+	foe = FIELD_GET(MT_CRXV_FOE_LO, v20) |
+	      (FIELD_GET(MT_CRXV_FOE_HI, v21) << MT_CRXV_FOE_SHIFT);
+
+	snr = FIELD_GET(MT_CRXV_SNR, v20) - 16;
+
+	phy->test.last_freq_offset = foe;
+	phy->test.last_snr = snr;
+#endif
+
+	dev_kfree_skb(skb);
+}
+
+static void
+bersa_mac_write_txwi_tm(struct bersa_phy *phy, __le32 *txwi,
+			 struct sk_buff *skb)
+{
+#ifdef CONFIG_NL80211_TESTMODE
+	struct mt76_testmode_data *td = &phy->mt76->test;
+	const struct ieee80211_rate *r;
+	u8 bw, mode, nss = td->tx_rate_nss;
+	u8 rate_idx = td->tx_rate_idx;
+	u16 rateval = 0;
+	u32 val;
+	bool cck = false;
+	int band;
+
+	if (skb != phy->mt76->test.tx_skb)
+		return;
+
+	switch (td->tx_rate_mode) {
+	case MT76_TM_TX_MODE_HT:
+		nss = 1 + (rate_idx >> 3);
+		mode = MT_PHY_TYPE_HT;
+		break;
+	case MT76_TM_TX_MODE_VHT:
+		mode = MT_PHY_TYPE_VHT;
+		break;
+	case MT76_TM_TX_MODE_HE_SU:
+		mode = MT_PHY_TYPE_HE_SU;
+		break;
+	case MT76_TM_TX_MODE_HE_EXT_SU:
+		mode = MT_PHY_TYPE_HE_EXT_SU;
+		break;
+	case MT76_TM_TX_MODE_HE_TB:
+		mode = MT_PHY_TYPE_HE_TB;
+		break;
+	case MT76_TM_TX_MODE_HE_MU:
+		mode = MT_PHY_TYPE_HE_MU;
+		break;
+	case MT76_TM_TX_MODE_CCK:
+		cck = true;
+		fallthrough;
+	case MT76_TM_TX_MODE_OFDM:
+		band = phy->mt76->chandef.chan->band;
+		if (band == NL80211_BAND_2GHZ && !cck)
+			rate_idx += 4;
+
+		r = &phy->mt76->hw->wiphy->bands[band]->bitrates[rate_idx];
+		val = cck ? r->hw_value_short : r->hw_value;
+
+		mode = val >> 8;
+		rate_idx = val & 0xff;
+		break;
+	default:
+		mode = MT_PHY_TYPE_OFDM;
+		break;
+	}
+
+	switch (phy->mt76->chandef.width) {
+	case NL80211_CHAN_WIDTH_40:
+		bw = 1;
+		break;
+	case NL80211_CHAN_WIDTH_80:
+		bw = 2;
+		break;
+	case NL80211_CHAN_WIDTH_80P80:
+	case NL80211_CHAN_WIDTH_160:
+		bw = 3;
+		break;
+	default:
+		bw = 0;
+		break;
+	}
+
+	if (td->tx_rate_stbc && nss == 1) {
+		nss++;
+		rateval |= MT_TX_RATE_STBC;
+	}
+
+	rateval |= FIELD_PREP(MT_TX_RATE_IDX, rate_idx) |
+		   FIELD_PREP(MT_TX_RATE_MODE, mode) |
+		   FIELD_PREP(MT_TX_RATE_NSS, nss - 1);
+
+	//txwi[2] |= cpu_to_le32(MT_TXD2_FIX_RATE);
+
+	le32p_replace_bits(&txwi[3], 1, MT_TXD3_REM_TX_COUNT);
+	if (td->tx_rate_mode < MT76_TM_TX_MODE_HT)
+		txwi[3] |= cpu_to_le32(MT_TXD3_BA_DISABLE);
+
+	val = MT_TXD6_BW |
+	      FIELD_PREP(MT_TXD6_BW, bw) |
+	      FIELD_PREP(MT_TXD6_TX_RATE, rateval); // |
+	//       FIELD_PREP(MT_TXD6_SGI, td->tx_rate_sgi);
+
+	/* for HE_SU/HE_EXT_SU PPDU
+	 * - 1x, 2x, 4x LTF + 0.8us GI
+	 * - 2x LTF + 1.6us GI, 4x LTF + 3.2us GI
+	 * for HE_MU PPDU
+	 * - 2x, 4x LTF + 0.8us GI
+	 * - 2x LTF + 1.6us GI, 4x LTF + 3.2us GI
+	 * for HE_TB PPDU
+	 * - 1x, 2x LTF + 1.6us GI
+	 * - 4x LTF + 3.2us GI
+	 */
+	// if (mode >= MT_PHY_TYPE_HE_SU)
+	// 	val |= FIELD_PREP(MT_TXD6_HELTF, td->tx_ltf);
+
+	// if (td->tx_rate_ldpc || (bw > 0 && mode >= MT_PHY_TYPE_HE_SU))
+	// 	val |= MT_TXD6_LDPC;
+
+	txwi[3] &= ~cpu_to_le32(MT_TXD3_SN_VALID);
+	txwi[6] |= cpu_to_le32(val);
+	// txwi[7] |= cpu_to_le32(FIELD_PREP(MT_TXD7_SPE_IDX,
+	// 				  phy->test.spe_idx));
+#endif
+}
+
+static void
+bersa_mac_write_txwi_8023(struct bersa_dev *dev, __le32 *txwi,
+			   struct sk_buff *skb, struct mt76_wcid *wcid)
+{
+
+	u8 tid = skb->priority & IEEE80211_QOS_CTL_TID_MASK;
+	u8 fc_type, fc_stype;
+	bool wmm = false;
+	u32 val;
+
+	if (wcid->sta) {
+		struct ieee80211_sta *sta;
+
+		sta = container_of((void *)wcid, struct ieee80211_sta, drv_priv);
+		wmm = sta->wme;
+	}
+
+	val = FIELD_PREP(MT_TXD1_HDR_FORMAT, MT_HDR_FORMAT_802_3) |
+	      FIELD_PREP(MT_TXD1_TID, tid);
+
+	if (be16_to_cpu(skb->protocol) >= ETH_P_802_3_MIN)
+		val |= MT_TXD1_ETH_802_3;
+
+	txwi[1] |= cpu_to_le32(val);
+
+	fc_type = IEEE80211_FTYPE_DATA >> 2;
+	fc_stype = wmm ? IEEE80211_STYPE_QOS_DATA >> 4 : 0;
+
+	val = FIELD_PREP(MT_TXD2_FRAME_TYPE, fc_type) |
+	      FIELD_PREP(MT_TXD2_SUB_TYPE, fc_stype);
+
+	txwi[2] |= cpu_to_le32(val);
+}
+
+static void
+bersa_mac_write_txwi_80211(struct bersa_dev *dev, __le32 *txwi,
+			    struct sk_buff *skb, struct ieee80211_key_conf *key,
+			    bool *mcast)
+{
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)skb->data;
+	struct ieee80211_mgmt *mgmt = (struct ieee80211_mgmt *)skb->data;
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	u8 tid = skb->priority & IEEE80211_QOS_CTL_TID_MASK;
+	__le16 fc = hdr->frame_control;
+	u8 fc_type, fc_stype;
+	u32 val;
+
+	*mcast = is_multicast_ether_addr(hdr->addr1);
+
+	if (ieee80211_is_action(fc) &&
+	    mgmt->u.action.category == WLAN_CATEGORY_BACK &&
+	    mgmt->u.action.u.addba_req.action_code == WLAN_ACTION_ADDBA_REQ)
+		tid = MT_TX_ADDBA;
+	else if (ieee80211_is_mgmt(hdr->frame_control))
+		tid = MT_TX_NORMAL;
+
+	val = FIELD_PREP(MT_TXD1_HDR_FORMAT, MT_HDR_FORMAT_802_11) |
+	      FIELD_PREP(MT_TXD1_HDR_INFO,
+			 ieee80211_get_hdrlen_from_skb(skb) / 2) |
+	      FIELD_PREP(MT_TXD1_TID, tid);
+
+	if (!ieee80211_is_data(fc) || *mcast ||
+	    info->flags & IEEE80211_TX_CTL_USE_MINRATE)
+		val |= MT_TXD1_FIXED_RATE;
+
+	if (key && *mcast && ieee80211_is_robust_mgmt_frame(skb) &&
+	    key->cipher == WLAN_CIPHER_SUITE_AES_CMAC) {
+		val |= MT_TXD1_BIP;
+		txwi[3] &= ~cpu_to_le32(MT_TXD3_PROTECT_FRAME);
+	}
+
+	txwi[1] |= cpu_to_le32(val);
+
+	fc_type = (le16_to_cpu(fc) & IEEE80211_FCTL_FTYPE) >> 2;
+	fc_stype = (le16_to_cpu(fc) & IEEE80211_FCTL_STYPE) >> 4;
+
+	val = FIELD_PREP(MT_TXD2_FRAME_TYPE, fc_type) |
+	      FIELD_PREP(MT_TXD2_SUB_TYPE, fc_stype);
+
+	txwi[2] |= cpu_to_le32(val);
+
+	txwi[3] |= FIELD_PREP(MT_TXD3_BCM, *mcast);
+	if (ieee80211_is_beacon(fc)) {
+		txwi[3] &= ~cpu_to_le32(MT_TXD3_SW_POWER_MGMT);
+		txwi[3] |= cpu_to_le32(MT_TXD3_REM_TX_COUNT);
+	}
+
+	if (info->flags & IEEE80211_TX_CTL_INJECTED) {
+		u16 seqno = le16_to_cpu(hdr->seq_ctrl);
+
+		if (ieee80211_is_back_req(hdr->frame_control)) {
+			struct ieee80211_bar *bar;
+
+			bar = (struct ieee80211_bar *)skb->data;
+			seqno = le16_to_cpu(bar->start_seq_num);
+		}
+
+		val = MT_TXD3_SN_VALID |
+		      FIELD_PREP(MT_TXD3_SEQ, IEEE80211_SEQ_TO_SN(seqno));
+		txwi[3] |= cpu_to_le32(val);
+		txwi[3] &= ~cpu_to_le32(MT_TXD3_HW_AMSDU);
+	}
+}
+
+static u16
+bersa_mac_tx_rate_val(struct mt76_phy *mphy, struct ieee80211_vif *vif,
+		       bool beacon, bool mcast)
+{
+	u8 mode = 0, band = mphy->chandef.chan->band;
+	int rateidx = 0, mcast_rate;
+
+	if (beacon) {
+		struct cfg80211_bitrate_mask *mask;
+
+		mask = &vif->bss_conf.beacon_tx_rate;
+		if (hweight16(mask->control[band].he_mcs[0]) == 1) {
+			rateidx = ffs(mask->control[band].he_mcs[0]) - 1;
+			mode = MT_PHY_TYPE_HE_SU;
+			goto out;
+		} else if (hweight16(mask->control[band].vht_mcs[0]) == 1) {
+			rateidx = ffs(mask->control[band].vht_mcs[0]) - 1;
+			mode = MT_PHY_TYPE_VHT;
+			goto out;
+		} else if (hweight8(mask->control[band].ht_mcs[0]) == 1) {
+			rateidx = ffs(mask->control[band].ht_mcs[0]) - 1;
+			mode = MT_PHY_TYPE_HT;
+			goto out;
+		} else if (hweight32(mask->control[band].legacy) == 1) {
+			rateidx = ffs(mask->control[band].legacy) - 1;
+			goto legacy;
+		}
+	}
+
+	mcast_rate = vif->bss_conf.mcast_rate[band];
+	if (mcast && mcast_rate > 0)
+		rateidx = mcast_rate - 1;
+	else
+		rateidx = ffs(vif->bss_conf.basic_rates) - 1;
+
+legacy:
+	rateidx = mt76_calculate_default_rate(mphy, rateidx);
+	mode = rateidx >> 8;
+	rateidx &= GENMASK(7, 0);
+
+out:
+	return FIELD_PREP(MT_TX_RATE_IDX, rateidx) |
+	       FIELD_PREP(MT_TX_RATE_MODE, mode);
+}
+
+void bersa_mac_write_txwi(struct bersa_dev *dev, __le32 *txwi,
+			   struct sk_buff *skb, struct mt76_wcid *wcid, int pid,
+			   struct ieee80211_key_conf *key, bool beacon)
+{
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(skb);
+	struct ieee80211_vif *vif = info->control.vif;
+	struct bersa_vif *mvif = (struct bersa_vif *)vif->drv_priv;
+	struct mt76_phy *mphy = &dev->mphy;
+	u8 phy_idx = (info->hw_queue & MT_TX_HW_QUEUE_PHY) >> 2;
+	u8 p_fmt, q_idx, omac_idx = 0, wmm_idx = 0, band_idx = 0;
+	bool is_8023 = info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP;
+	bool mcast = false;
+	u16 tx_count = 15;
+	u32 val;
+
+	if (vif) {
+		struct bersa_vif *mvif = (struct bersa_vif *)vif->drv_priv;
+
+		omac_idx = mvif->mt76.omac_idx;
+		wmm_idx = mvif->mt76.wmm_idx;
+		band_idx = mvif->mt76.band_idx;
+	}
+
+	mphy = mt76_dev_phy(&dev->mt76, phy_idx);
+
+	if (beacon) {
+		p_fmt = MT_TX_TYPE_FW;
+		q_idx = MT_LMAC_BCN0;
+	} else if (skb_get_queue_mapping(skb) >= MT_TXQ_PSD) {
+		p_fmt = MT_TX_TYPE_CT;
+		q_idx = MT_LMAC_ALTX0;
+	} else {
+		p_fmt = MT_TX_TYPE_CT;
+		q_idx = wmm_idx * BERSA_MAX_WMM_SETS +
+			mt76_connac_lmac_mapping(skb_get_queue_mapping(skb));
+	}
+
+	val = FIELD_PREP(MT_TXD0_TX_BYTES, skb->len + MT_TXD_SIZE) |
+	      FIELD_PREP(MT_TXD0_PKT_FMT, p_fmt) |
+	      FIELD_PREP(MT_TXD0_Q_IDX, q_idx);
+	txwi[0] = cpu_to_le32(val);
+
+	val = FIELD_PREP(MT_TXD1_WLAN_IDX, wcid->idx) |
+	      FIELD_PREP(MT_TXD1_OWN_MAC, omac_idx);
+
+	if (band_idx &&
+	    q_idx >= MT_LMAC_ALTX0 && q_idx <= MT_LMAC_BCN0)
+		val |= FIELD_PREP(MT_TXD1_TGID, band_idx);
+
+	txwi[1] = cpu_to_le32(val);
+
+	txwi[2] = 0;
+
+	val = MT_TXD3_SW_POWER_MGMT |
+	      FIELD_PREP(MT_TXD3_REM_TX_COUNT, tx_count);
+	if (key)
+		val |= MT_TXD3_PROTECT_FRAME;
+	if (info->flags & IEEE80211_TX_CTL_NO_ACK)
+		val |= MT_TXD3_NO_ACK;
+	if (wcid->amsdu)
+		val |= MT_TXD3_HW_AMSDU;
+
+	txwi[3] = cpu_to_le32(val);
+	txwi[4] = 0;
+
+	val = FIELD_PREP(MT_TXD5_PID, pid);
+	if (pid >= MT_PACKET_ID_FIRST)
+		val |= MT_TXD5_TX_STATUS_HOST;
+	txwi[5] = cpu_to_le32(val);
+
+	/* Todo: check MT_TXD6_DIS_MAT for MLD to Link Address Translation */
+	/* val = MT_TXD6_DIS_MAT | MT_TXD6_VTA | */
+	val = MT_TXD6_DIS_MAT | MT_TXD6_DAS |
+	      FIELD_PREP(MT_TXD6_MSDU_CNT, 1);
+	txwi[6] = cpu_to_le32(val);
+	txwi[7] = 0;
+
+	if (is_8023)
+		bersa_mac_write_txwi_8023(dev, txwi, skb, wcid);
+	else
+		bersa_mac_write_txwi_80211(dev, txwi, skb, key, &mcast);
+
+	if (txwi[1] & cpu_to_le32(MT_TXD1_FIXED_RATE)) {
+		u16 rate = bersa_mac_tx_rate_val(mphy, vif, beacon, mcast);
+
+		/* hardware won't add HTC for mgmt/ctrl frame */
+		/* txwi[2] |= cpu_to_le32(MT_TXD2_HTC_VLD); */
+
+		val = FIELD_PREP(MT_TXD6_BW, 8) |
+		      FIELD_PREP(MT_TXD6_TX_RATE, 1);
+		txwi[6] |= cpu_to_le32(val);
+		//txwi[3] |= cpu_to_le32(MT_TXD3_BA_DISABLE);
+	}
+
+	if (mt76_testmode_enabled(mphy))
+		bersa_mac_write_txwi_tm(mphy->priv, txwi, skb);
+}
+
+int bersa_tx_prepare_skb(struct mt76_dev *mdev, void *txwi_ptr,
+			  enum mt76_txq_id qid, struct mt76_wcid *wcid,
+			  struct ieee80211_sta *sta,
+			  struct mt76_tx_info *tx_info)
+{
+	struct ieee80211_hdr *hdr = (struct ieee80211_hdr *)tx_info->skb->data;
+	struct bersa_dev *dev = container_of(mdev, struct bersa_dev, mt76);
+	struct ieee80211_tx_info *info = IEEE80211_SKB_CB(tx_info->skb);
+	struct ieee80211_key_conf *key = info->control.hw_key;
+	struct ieee80211_vif *vif = info->control.vif;
+	struct mt76_txwi_cache *t;
+	struct bersa_txp *txp;
+	int id, i, nbuf = tx_info->nbuf - 1;
+	bool is_8023 = info->flags & IEEE80211_TX_CTL_HW_80211_ENCAP;
+	u8 *txwi = (u8 *)txwi_ptr;
+	int pid;
+
+	if (unlikely(tx_info->skb->len <= ETH_HLEN))
+		return -EINVAL;
+
+	if (!wcid)
+		wcid = &dev->mt76.global_wcid;
+
+	if (sta) {
+		struct bersa_sta *msta;
+
+		msta = (struct bersa_sta *)sta->drv_priv;
+
+		if (time_after(jiffies, msta->jiffies + HZ / 4)) {
+			info->flags |= IEEE80211_TX_CTL_REQ_TX_STATUS;
+			msta->jiffies = jiffies;
+		}
+	}
+
+	t = (struct mt76_txwi_cache *)(txwi + mdev->drv->txwi_size);
+	t->skb = tx_info->skb;
+
+	id = mt76_token_consume(mdev, &t);
+	if (id < 0)
+		return id;
+
+	pid = mt76_tx_status_skb_add(mdev, wcid, tx_info->skb);
+	memset(txwi_ptr, 0, MT_TXD_SIZE);
+	/* Transmit non qos data by 802.11 header and need to fill txd by host*/
+	if (!is_8023 || pid >= MT_PACKET_ID_FIRST)
+		bersa_mac_write_txwi(dev, txwi_ptr, tx_info->skb, wcid, pid,
+				     key, false);
+
+	txp = (struct bersa_txp *)(txwi + MT_TXD_SIZE);
+	for (i = 0; i < nbuf; i++) {
+		txp->buf[i] = cpu_to_le32(tx_info->buf[i + 1].addr);
+		txp->len[i] = cpu_to_le16(tx_info->buf[i + 1].len);
+	}
+	txp->nbuf = nbuf;
+
+	txp->flags = cpu_to_le16(MT_CT_INFO_FROM_HOST);
+
+	if (!is_8023 || pid >= MT_PACKET_ID_FIRST)
+		txp->flags |= cpu_to_le16(MT_CT_INFO_APPLY_TXD);
+
+	if (!key)
+		txp->flags |= cpu_to_le16(MT_CT_INFO_NONE_CIPHER_FRAME);
+
+	if (!is_8023 && ieee80211_is_mgmt(hdr->frame_control))
+		txp->flags |= cpu_to_le16(MT_CT_INFO_MGMT_FRAME);
+
+	if (vif) {
+		struct bersa_vif *mvif = (struct bersa_vif *)vif->drv_priv;
+
+		txp->bss_idx = mvif->mt76.idx;
+	}
+
+	txp->token = cpu_to_le16(id);
+	if (test_bit(MT_WCID_FLAG_4ADDR, &wcid->flags))
+		txp->rept_wds_wcid = cpu_to_le16(wcid->idx);
+	else
+		txp->rept_wds_wcid = cpu_to_le16(0xfff);
+	tx_info->skb = DMA_DUMMY_DATA;
+
+	/* pass partial skb header to fw */
+	tx_info->buf[1].len = MT_CT_PARSE_LEN;
+	tx_info->buf[1].skip_unmap = true;
+	tx_info->nbuf = MT_CT_DMA_BUF_NUM;
+
+	return 0;
+}
+
+static void
+bersa_tx_check_aggr(struct ieee80211_sta *sta, __le32 *txwi)
+{
+	struct bersa_sta *msta;
+	u16 fc, tid;
+	u32 val;
+
+	if (!sta || !sta->ht_cap.ht_supported)
+		return;
+
+	tid = FIELD_GET(MT_TXD1_TID, le32_to_cpu(txwi[1]));
+	if (tid >= 6) /* skip VO queue */
+		return;
+
+	val = le32_to_cpu(txwi[2]);
+	fc = FIELD_GET(MT_TXD2_FRAME_TYPE, val) << 2 |
+	     FIELD_GET(MT_TXD2_SUB_TYPE, val) << 4;
+	if (unlikely(fc != (IEEE80211_FTYPE_DATA | IEEE80211_STYPE_QOS_DATA)))
+		return;
+
+	msta = (struct bersa_sta *)sta->drv_priv;
+	if (!test_and_set_bit(tid, &msta->ampdu_state))
+		ieee80211_start_tx_ba_session(sta, tid, 0);
+}
+
+static void
+bersa_txp_skb_unmap(struct mt76_dev *dev, struct mt76_txwi_cache *t)
+{
+	struct bersa_txp *txp;
+	int i;
+
+	txp = bersa_txwi_to_txp(dev, t);
+	for (i = 0; i < txp->nbuf; i++)
+		dma_unmap_single(dev->dev, le32_to_cpu(txp->buf[i]),
+				 le16_to_cpu(txp->len[i]), DMA_TO_DEVICE);
+}
+
+static void
+bersa_txwi_free(struct bersa_dev *dev, struct mt76_txwi_cache *t,
+		 struct ieee80211_sta *sta, struct list_head *free_list)
+{
+	struct mt76_dev *mdev = &dev->mt76;
+	struct mt76_wcid *wcid;
+	__le32 *txwi;
+	u16 wcid_idx;
+
+	bersa_txp_skb_unmap(mdev, t);
+	if (!t->skb)
+		goto out;
+
+	txwi = (__le32 *)mt76_get_txwi_ptr(mdev, t);
+	if (sta) {
+		wcid = (struct mt76_wcid *)sta->drv_priv;
+		wcid_idx = wcid->idx;
+
+		if (likely(t->skb->protocol != cpu_to_be16(ETH_P_PAE)))
+			bersa_tx_check_aggr(sta, txwi);
+	} else {
+		wcid_idx = FIELD_GET(MT_TXD1_WLAN_IDX, le32_to_cpu(txwi[1]));
+	}
+
+	__mt76_tx_complete_skb(mdev, wcid_idx, t->skb, free_list);
+
+out:
+	t->skb = NULL;
+	mt76_put_txwi(mdev, t);
+}
+
+static void
+bersa_mac_tx_free(struct bersa_dev *dev, void *data, int len)
+{
+	struct bersa_tx_free *free = (struct bersa_tx_free *)data;
+	struct mt76_dev *mdev = &dev->mt76;
+	struct mt76_phy *mphy_ext = mdev->phy2;
+	struct mt76_phy *mphy_tri = mdev->phy3;
+	struct mt76_txwi_cache *txwi;
+	struct ieee80211_sta *sta = NULL;
+	LIST_HEAD(free_list);
+	struct sk_buff *skb, *tmp;
+	void *end = data + len;
+	bool v3, wake = false;
+	u16 total, count = 0;
+	u32 txd = le32_to_cpu(free->txd);
+	__le32 *cur_info;
+
+	/* clean DMA queues and unmap buffers first */
+	mt76_queue_tx_cleanup(dev, dev->mphy.q_tx[MT_TXQ_PSD], false);
+	mt76_queue_tx_cleanup(dev, dev->mphy.q_tx[MT_TXQ_BE], false);
+	if (mphy_ext) {
+		mt76_queue_tx_cleanup(dev, mphy_ext->q_tx[MT_TXQ_PSD], false);
+		mt76_queue_tx_cleanup(dev, mphy_ext->q_tx[MT_TXQ_BE], false);
+	}
+	if (mphy_tri) {
+		mt76_queue_tx_cleanup(dev, mphy_tri->q_tx[MT_TXQ_PSD], false);
+		mt76_queue_tx_cleanup(dev, mphy_tri->q_tx[MT_TXQ_BE], false);
+	}
+
+	/*
+	 * TODO: MT_TX_FREE_LATENCY is msdu time from the TXD is queued into PLE,
+	 * to the time ack is received or dropped by hw (air + hw queue time).
+	 * Should avoid accessing WTBL to get Tx airtime, and use it instead.
+	 */
+	total = FIELD_GET(MT_TX_FREE_MSDU_CNT, le16_to_cpu(free->ctrl));
+	v3 = (FIELD_GET(MT_TX_FREE_VER, txd) >= 0x4);
+	if (WARN_ON_ONCE((void *)&free->info[total >> v3] > end))
+		return;
+
+	for (cur_info = &free->info[0]; count < total; cur_info++) {
+		u32 msdu, info = le32_to_cpu(*cur_info);
+		u8 i;
+
+		/*
+		 * 1'b1: new wcid pair.
+		 * 1'b0: msdu_id with the same 'wcid pair' as above.
+		 */
+		if (info & MT_TX_FREE_PAIR) {
+			struct bersa_sta *msta;
+			struct mt76_wcid *wcid;
+			u16 idx;
+
+			idx = FIELD_GET(MT_TX_FREE_WLAN_ID, info);
+			wcid = rcu_dereference(dev->mt76.wcid[idx]);
+			sta = wcid_to_sta(wcid);
+			if (!sta)
+				continue;
+
+			msta = container_of(wcid, struct bersa_sta, wcid);
+			spin_lock_bh(&dev->sta_poll_lock);
+			if (list_empty(&msta->poll_list))
+				list_add_tail(&msta->poll_list, &dev->sta_poll_list);
+			spin_unlock_bh(&dev->sta_poll_lock);
+			continue;
+		}
+
+		if (v3 && (info & MT_TX_FREE_MPDU_HEADER))
+			continue;
+
+		for (i = 0; i < 1 + v3; i++) {
+			if (v3) {
+				msdu = (info >> (15 * i)) & MT_TX_FREE_MSDU_ID_V3;
+				if (msdu == MT_TX_FREE_MSDU_ID_V3)
+					continue;
+			} else {
+				msdu = FIELD_GET(MT_TX_FREE_MSDU_ID, info);
+			}
+			count++;
+			txwi = mt76_token_release(mdev, msdu, &wake);
+			if (!txwi)
+				continue;
+
+			bersa_txwi_free(dev, txwi, sta, &free_list);
+		}
+	}
+
+	bersa_mac_sta_poll(dev);
+
+	if (wake)
+		mt76_set_tx_blocked(&dev->mt76, false);
+
+	mt76_worker_schedule(&dev->mt76.tx_worker);
+
+	list_for_each_entry_safe(skb, tmp, &free_list, list) {
+		skb_list_del_init(skb);
+		napi_consume_skb(skb, 1);
+	}
+}
+
+static bool
+bersa_mac_add_txs_skb(struct bersa_dev *dev, struct mt76_wcid *wcid, int pid,
+		       __le32 *txs_data, struct mt76_sta_stats *stats)
+{
+	struct ieee80211_supported_band *sband;
+	struct mt76_dev *mdev = &dev->mt76;
+	struct mt76_phy *mphy;
+	struct ieee80211_tx_info *info;
+	struct sk_buff_head list;
+	struct rate_info rate = {};
+	struct sk_buff *skb;
+	bool cck = false;
+	u32 txrate, txs, mode;
+
+	mt76_tx_status_lock(mdev, &list);
+	skb = mt76_tx_status_skb_get(mdev, wcid, pid, &list);
+	if (!skb)
+		goto out_no_skb;
+
+	txs = le32_to_cpu(txs_data[0]);
+
+	info = IEEE80211_SKB_CB(skb);
+	if (!(txs & MT_TXS0_ACK_ERROR_MASK))
+		info->flags |= IEEE80211_TX_STAT_ACK;
+
+	info->status.ampdu_len = 1;
+	info->status.ampdu_ack_len = !!(info->flags &
+					IEEE80211_TX_STAT_ACK);
+
+	info->status.rates[0].idx = -1;
+
+	txrate = FIELD_GET(MT_TXS0_TX_RATE, txs);
+
+	rate.mcs = FIELD_GET(MT_TX_RATE_IDX, txrate);
+	rate.nss = FIELD_GET(MT_TX_RATE_NSS, txrate) + 1;
+
+	if (rate.nss - 1 < ARRAY_SIZE(stats->tx_nss))
+		stats->tx_nss[rate.nss - 1]++;
+	if (rate.mcs < ARRAY_SIZE(stats->tx_mcs))
+		stats->tx_mcs[rate.mcs]++;
+
+	mode = FIELD_GET(MT_TX_RATE_MODE, txrate);
+	switch (mode) {
+	case MT_PHY_TYPE_CCK:
+		cck = true;
+		fallthrough;
+	case MT_PHY_TYPE_OFDM:
+		mphy = mt76_dev_phy(mdev, wcid->phy_idx);
+
+		if (mphy->chandef.chan->band == NL80211_BAND_5GHZ)
+			sband = &mphy->sband_5g.sband;
+		else
+			sband = &mphy->sband_2g.sband;
+
+		rate.mcs = mt76_get_rate(mphy->dev, sband, rate.mcs, cck);
+		rate.legacy = sband->bitrates[rate.mcs].bitrate;
+		break;
+	case MT_PHY_TYPE_HT:
+	case MT_PHY_TYPE_HT_GF:
+		if (rate.mcs > 31)
+			goto out;
+
+		rate.flags = RATE_INFO_FLAGS_MCS;
+		if (wcid->rate.flags & RATE_INFO_FLAGS_SHORT_GI)
+			rate.flags |= RATE_INFO_FLAGS_SHORT_GI;
+		break;
+	case MT_PHY_TYPE_VHT:
+		if (rate.mcs > 9)
+			goto out;
+
+		rate.flags = RATE_INFO_FLAGS_VHT_MCS;
+		break;
+	case MT_PHY_TYPE_HE_SU:
+	case MT_PHY_TYPE_HE_EXT_SU:
+	case MT_PHY_TYPE_HE_TB:
+	case MT_PHY_TYPE_HE_MU:
+		if (rate.mcs > 11)
+			goto out;
+
+		rate.he_gi = wcid->rate.he_gi;
+		rate.he_dcm = FIELD_GET(MT_TX_RATE_DCM, txrate);
+		rate.flags = RATE_INFO_FLAGS_HE_MCS;
+		break;
+	default:
+		goto out;
+	}
+
+	stats->tx_mode[mode]++;
+
+	switch (FIELD_GET(MT_TXS0_BW, txs)) {
+	case IEEE80211_STA_RX_BW_160:
+		rate.bw = RATE_INFO_BW_160;
+		stats->tx_bw[3]++;
+		break;
+	case IEEE80211_STA_RX_BW_80:
+		rate.bw = RATE_INFO_BW_80;
+		stats->tx_bw[2]++;
+		break;
+	case IEEE80211_STA_RX_BW_40:
+		rate.bw = RATE_INFO_BW_40;
+		stats->tx_bw[1]++;
+		break;
+	default:
+		rate.bw = RATE_INFO_BW_20;
+		stats->tx_bw[0]++;
+		break;
+	}
+	wcid->rate = rate;
+
+out:
+	mt76_tx_status_skb_done(mdev, skb, &list);
+
+out_no_skb:
+	mt76_tx_status_unlock(mdev, &list);
+
+	return !!skb;
+}
+
+static void bersa_mac_add_txs(struct bersa_dev *dev, void *data)
+{
+	struct bersa_sta *msta = NULL;
+	struct mt76_wcid *wcid;
+	__le32 *txs_data = data;
+	u16 wcidx;
+	u32 txs;
+	u8 pid;
+
+	txs = le32_to_cpu(txs_data[0]);
+	if (FIELD_GET(MT_TXS0_TXS_FORMAT, txs) > 1)
+		return;
+
+	txs = le32_to_cpu(txs_data[2]);
+	wcidx = FIELD_GET(MT_TXS2_WCID, txs);
+
+	txs = le32_to_cpu(txs_data[3]);
+	pid = FIELD_GET(MT_TXS3_PID, txs);
+
+	if (pid < MT_PACKET_ID_FIRST)
+		return;
+
+	if (wcidx >= BERSA_WTBL_SIZE)
+		return;
+
+	rcu_read_lock();
+
+	wcid = rcu_dereference(dev->mt76.wcid[wcidx]);
+	if (!wcid)
+		goto out;
+
+	msta = container_of(wcid, struct bersa_sta, wcid);
+
+	bersa_mac_add_txs_skb(dev, wcid, pid, txs_data, &msta->stats);
+
+	if (!wcid->sta)
+		goto out;
+
+	spin_lock_bh(&dev->sta_poll_lock);
+	if (list_empty(&msta->poll_list))
+		list_add_tail(&msta->poll_list, &dev->sta_poll_list);
+	spin_unlock_bh(&dev->sta_poll_lock);
+
+out:
+	rcu_read_unlock();
+}
+
+bool bersa_rx_check(struct mt76_dev *mdev, enum mt76_rxq_id q, void *data, int len)
+{
+	struct bersa_dev *dev = container_of(mdev, struct bersa_dev, mt76);
+	__le32 *rxd = (__le32 *)data;
+	__le32 *end = (__le32 *)&rxd[len / 4];
+	enum rx_pkt_type type;
+
+	type = FIELD_GET(MT_RXD0_PKT_TYPE, le32_to_cpu(rxd[0]));
+	if (q == MT_RXQ_MCU || q == MT_RXQ_MCU_WA ||
+	    q == MT_RXQ_MAIN_WA || q == MT_RXQ_EXT_WA || q == MT_RXQ_TRI_WA) {
+		enum rx_pkt_type sw_type = FIELD_GET(MT_RXD0_SW_PKT_TYPE_MASK, le32_to_cpu(rxd[0]));
+
+		if ((sw_type & MT_RXD0_SW_PKT_TYPE_MAP) == MT_RXD0_SW_PKT_TYPE_FRAME)
+			type = PKT_TYPE_NORMAL;
+	}
+
+	switch (type) {
+	case PKT_TYPE_TXRX_NOTIFY:
+		bersa_mac_tx_free(dev, data, len);
+		return false;
+	case PKT_TYPE_TXS:
+		for (rxd += 4; rxd + 8 <= end; rxd += 8)
+			bersa_mac_add_txs(dev, rxd);
+		return false;
+	case PKT_TYPE_RX_FW_MONITOR:
+		bersa_debugfs_rx_fw_monitor(dev, data, len);
+		return false;
+	default:
+		return true;
+	}
+}
+
+void bersa_queue_rx_skb(struct mt76_dev *mdev, enum mt76_rxq_id q,
+			 struct sk_buff *skb)
+{
+	struct bersa_dev *dev = container_of(mdev, struct bersa_dev, mt76);
+	__le32 *rxd = (__le32 *)skb->data;
+	__le32 *end = (__le32 *)&skb->data[skb->len];
+	enum rx_pkt_type type;
+
+	type = FIELD_GET(MT_RXD0_PKT_TYPE, le32_to_cpu(rxd[0]));
+	if (q == MT_RXQ_MCU || q == MT_RXQ_MCU_WA ||
+	    q == MT_RXQ_MAIN_WA || q == MT_RXQ_EXT_WA || q == MT_RXQ_TRI_WA) {
+		enum rx_pkt_type sw_type = FIELD_GET(MT_RXD0_SW_PKT_TYPE_MASK, le32_to_cpu(rxd[0]));
+
+		if ((sw_type & MT_RXD0_SW_PKT_TYPE_MAP) == MT_RXD0_SW_PKT_TYPE_FRAME)
+			type = PKT_TYPE_NORMAL;
+	}
+
+	switch (type) {
+	case PKT_TYPE_TXRX_NOTIFY:
+		bersa_mac_tx_free(dev, skb->data, skb->len);
+		napi_consume_skb(skb, 1);
+		break;
+	case PKT_TYPE_RX_EVENT:
+		bersa_mcu_rx_event(dev, skb);
+		break;
+	case PKT_TYPE_TXRXV:
+		bersa_mac_fill_rx_vector(dev, skb);
+		break;
+	case PKT_TYPE_TXS:
+		for (rxd += 4; rxd + 8 <= end; rxd += 8)
+			bersa_mac_add_txs(dev, rxd);
+		dev_kfree_skb(skb);
+		break;
+	case PKT_TYPE_RX_FW_MONITOR:
+		bersa_debugfs_rx_fw_monitor(dev, skb->data, skb->len);
+		dev_kfree_skb(skb);
+		break;
+	case PKT_TYPE_NORMAL:
+		if (!bersa_mac_fill_rx(dev, skb)) {
+			mt76_rx(&dev->mt76, q, skb);
+			return;
+		}
+		fallthrough;
+	default:
+		dev_kfree_skb(skb);
+		break;
+	}
+}
+
+void bersa_tx_complete_skb(struct mt76_dev *mdev, struct mt76_queue_entry *e)
+{
+	if (!e->txwi) {
+		dev_kfree_skb_any(e->skb);
+		return;
+	}
+
+	/* error path */
+	if (e->skb == DMA_DUMMY_DATA) {
+		struct mt76_txwi_cache *t;
+		struct bersa_txp *txp;
+
+		txp = bersa_txwi_to_txp(mdev, e->txwi);
+		t = mt76_token_put(mdev, le16_to_cpu(txp->token));
+		e->skb = t ? t->skb : NULL;
+	}
+
+	if (e->skb)
+		mt76_tx_complete_skb(mdev, e->wcid, e->skb);
+}
+
+void bersa_mac_cca_stats_reset(struct bersa_phy *phy)
+{
+	struct bersa_dev *dev = phy->dev;
+	u32 reg = MT_WF_PHY_RX_CTRL1(phy->band_idx);
+
+	mt76_clear(dev, reg, MT_WF_PHY_RX_CTRL1_STSCNT_EN);
+	mt76_set(dev, reg, BIT(11) | BIT(9));
+}
+
+void bersa_mac_reset_counters(struct bersa_phy *phy)
+{
+	struct bersa_dev *dev = phy->dev;
+	int i;
+
+	for (i = 0; i < 4; i++) {
+		mt76_rr(dev, MT_TX_AGG_CNT(phy->band_idx, i));
+		mt76_rr(dev, MT_TX_AGG_CNT2(phy->band_idx, i));
+	}
+
+	i = 0;
+	phy->mt76->survey_time = ktime_get_boottime();
+	if (phy->band_idx)
+		i = ARRAY_SIZE(dev->mt76.aggr_stats) / 2;
+
+	memset(&dev->mt76.aggr_stats[i], 0, sizeof(dev->mt76.aggr_stats) / 2);
+
+	/* reset airtime counters */
+	mt76_set(dev, MT_WF_RMAC_MIB_AIRTIME0(phy->band_idx),
+		 MT_WF_RMAC_MIB_RXTIME_CLR);
+
+	bersa_mcu_get_chan_mib_info(phy, true);
+}
+
+void bersa_mac_set_timing(struct bersa_phy *phy)
+{
+	s16 coverage_class = phy->coverage_class;
+	struct bersa_dev *dev = phy->dev;
+	struct bersa_phy *ext_phy = bersa_ext_phy(dev);
+	struct bersa_phy *tri_phy = bersa_tri_phy(dev);
+	u32 val, reg_offset;
+	u32 cck = FIELD_PREP(MT_TIMEOUT_VAL_PLCP, 231) |
+		  FIELD_PREP(MT_TIMEOUT_VAL_CCA, 48);
+	u32 ofdm = FIELD_PREP(MT_TIMEOUT_VAL_PLCP, 60) |
+		   FIELD_PREP(MT_TIMEOUT_VAL_CCA, 28);
+	int offset;
+	bool is_5ghz = phy->mt76->chandef.chan->band == NL80211_BAND_5GHZ;
+
+	if (!test_bit(MT76_STATE_RUNNING, &phy->mt76->state))
+		return;
+
+	if (ext_phy)
+		coverage_class = max_t(s16, dev->phy.coverage_class,
+				       ext_phy->coverage_class);
+
+	if (tri_phy)
+		coverage_class = max_t(s16, coverage_class,
+				       tri_phy->coverage_class);
+
+	mt76_set(dev, MT_ARB_SCR(phy->band_idx),
+		 MT_ARB_SCR_TX_DISABLE | MT_ARB_SCR_RX_DISABLE);
+	udelay(1);
+
+	offset = 3 * coverage_class;
+	reg_offset = FIELD_PREP(MT_TIMEOUT_VAL_PLCP, offset) |
+		     FIELD_PREP(MT_TIMEOUT_VAL_CCA, offset);
+
+	mt76_wr(dev, MT_TMAC_CDTR(phy->band_idx), cck + reg_offset);
+	mt76_wr(dev, MT_TMAC_ODTR(phy->band_idx), ofdm + reg_offset);
+	mt76_wr(dev, MT_TMAC_ICR0(phy->band_idx),
+		FIELD_PREP(MT_IFS_EIFS_OFDM, is_5ghz ? 84 : 78) |
+		FIELD_PREP(MT_IFS_RIFS, 2) |
+		FIELD_PREP(MT_IFS_SIFS, 10) |
+		FIELD_PREP(MT_IFS_SLOT, phy->slottime));
+
+	mt76_wr(dev, MT_TMAC_ICR1(phy->band_idx),
+		FIELD_PREP(MT_IFS_EIFS_CCK, 314));
+
+	if (phy->slottime < 20 || is_5ghz)
+		val = BERSA_CFEND_RATE_DEFAULT;
+	else
+		val = BERSA_CFEND_RATE_11B;
+
+	mt76_rmw_field(dev, MT_AGG_ACR0(phy->band_idx), MT_AGG_ACR_CFEND_RATE, val);
+	mt76_clear(dev, MT_ARB_SCR(phy->band_idx),
+		   MT_ARB_SCR_TX_DISABLE | MT_ARB_SCR_RX_DISABLE);
+}
+
+void bersa_mac_enable_nf(struct bersa_dev *dev, u8 band)
+{
+	mt76_set(dev, MT_WF_PHY_RXTD12(band),
+		 MT_WF_PHY_RXTD12_IRPI_SW_CLR_ONLY |
+		 MT_WF_PHY_RXTD12_IRPI_SW_CLR);
+
+	mt76_set(dev, MT_WF_PHY_RX_CTRL1(band),
+		 FIELD_PREP(MT_WF_PHY_RX_CTRL1_IPI_EN, 0x5));
+}
+
+static u8
+bersa_phy_get_nf(struct bersa_phy *phy, int idx)
+{
+	static const u8 nf_power[] = { 92, 89, 86, 83, 80, 75, 70, 65, 60, 55, 52 };
+	struct bersa_dev *dev = phy->dev;
+	u32 val, sum = 0, n = 0;
+	int nss, i;
+
+	for (nss = 0; nss < hweight8(phy->mt76->chainmask); nss++) {
+		u32 reg = MT_WF_IRPI_NSS(idx, nss);
+
+		for (i = 0; i < ARRAY_SIZE(nf_power); i++, reg += 4) {
+			val = mt76_rr(dev, reg);
+			sum += val * nf_power[i];
+			n += val;
+		}
+	}
+
+	if (!n)
+		return 0;
+
+	return sum / n;
+}
+
+void bersa_update_channel(struct mt76_phy *mphy)
+{
+	struct bersa_phy *phy = (struct bersa_phy *)mphy->priv;
+	struct mt76_channel_state *state = mphy->chan_state;
+	int nf;
+
+	bersa_mcu_get_chan_mib_info(phy, false);
+
+	nf = bersa_phy_get_nf(phy, phy->band_idx);
+	if (!phy->noise)
+		phy->noise = nf << 4;
+	else if (nf)
+		phy->noise += nf - (phy->noise >> 4);
+
+	state->noise = -(phy->noise >> 4);
+}
+
+static bool
+bersa_wait_reset_state(struct bersa_dev *dev, u32 state)
+{
+	bool ret;
+
+	ret = wait_event_timeout(dev->reset_wait,
+				 (READ_ONCE(dev->reset_state) & state),
+				 BERSA_RESET_TIMEOUT);
+
+	WARN(!ret, "Timeout waiting for MCU reset state %x\n", state);
+	return ret;
+}
+
+static void
+bersa_update_vif_beacon(void *priv, u8 *mac, struct ieee80211_vif *vif)
+{
+	struct ieee80211_hw *hw = priv;
+
+	switch (vif->type) {
+	case NL80211_IFTYPE_MESH_POINT:
+	case NL80211_IFTYPE_ADHOC:
+	case NL80211_IFTYPE_AP:
+		bersa_mcu_add_beacon(hw, vif, vif->bss_conf.enable_beacon);
+		break;
+	default:
+		break;
+	}
+}
+
+static void
+bersa_update_beacons(struct bersa_dev *dev)
+{
+	ieee80211_iterate_active_interfaces(dev->mt76.hw,
+		IEEE80211_IFACE_ITER_RESUME_ALL,
+		bersa_update_vif_beacon, dev->mt76.hw);
+
+	if (!dev->mt76.phy2)
+		return;
+
+	ieee80211_iterate_active_interfaces(dev->mt76.phy2->hw,
+		IEEE80211_IFACE_ITER_RESUME_ALL,
+		bersa_update_vif_beacon, dev->mt76.phy2->hw);
+
+	if (!dev->mt76.phy3)
+		return;
+
+	ieee80211_iterate_active_interfaces(dev->mt76.phy3->hw,
+		IEEE80211_IFACE_ITER_RESUME_ALL,
+		bersa_update_vif_beacon, dev->mt76.phy3->hw);
+}
+
+static void
+bersa_dma_reset(struct bersa_dev *dev)
+{
+	struct mt76_phy *mphy_ext = dev->mt76.phy2;
+	struct mt76_phy *mphy_tri = dev->mt76.phy3;
+	u32 hif1_ofs = MT_WFDMA0_PCIE1(0) - MT_WFDMA0(0);
+	int i;
+
+	mt76_clear(dev, MT_WFDMA0_GLO_CFG,
+		   MT_WFDMA0_GLO_CFG_TX_DMA_EN |
+		   MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+
+	if (dev->hif2)
+		mt76_clear(dev, MT_WFDMA0_GLO_CFG + hif1_ofs,
+			   MT_WFDMA0_GLO_CFG_TX_DMA_EN |
+			   MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+
+	usleep_range(1000, 2000);
+
+	for (i = 0; i < __MT_TXQ_MAX; i++) {
+		mt76_queue_tx_cleanup(dev, dev->mphy.q_tx[i], true);
+		if (mphy_ext)
+			mt76_queue_tx_cleanup(dev, mphy_ext->q_tx[i], true);
+		if (mphy_tri)
+			mt76_queue_tx_cleanup(dev, mphy_tri->q_tx[i], true);
+	}
+
+	for (i = 0; i < __MT_MCUQ_MAX; i++)
+		mt76_queue_tx_cleanup(dev, dev->mt76.q_mcu[i], true);
+
+	mt76_for_each_q_rx(&dev->mt76, i)
+		mt76_queue_rx_reset(dev, i);
+
+	mt76_tx_status_check(&dev->mt76, true);
+
+	/* re-init prefetch settings after reset */
+	bersa_dma_prefetch(dev);
+
+	mt76_set(dev, MT_WFDMA0_GLO_CFG,
+		 MT_WFDMA0_GLO_CFG_TX_DMA_EN | MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+
+	if (dev->hif2)
+		mt76_set(dev, MT_WFDMA0_GLO_CFG + hif1_ofs,
+			 MT_WFDMA0_GLO_CFG_TX_DMA_EN |
+			 MT_WFDMA0_GLO_CFG_RX_DMA_EN);
+}
+
+void bersa_tx_token_put(struct bersa_dev *dev)
+{
+	struct mt76_txwi_cache *txwi;
+	int id;
+
+	spin_lock_bh(&dev->mt76.token_lock);
+	idr_for_each_entry(&dev->mt76.token, txwi, id) {
+		bersa_txwi_free(dev, txwi, NULL, NULL);
+		dev->mt76.token_count--;
+	}
+	spin_unlock_bh(&dev->mt76.token_lock);
+	idr_destroy(&dev->mt76.token);
+}
+
+/* system error recovery */
+void bersa_mac_reset_work(struct work_struct *work)
+{
+	struct bersa_phy *phy2, *phy3;
+	struct mt76_phy *ext_phy, *tri_phy;
+	struct bersa_dev *dev;
+
+	dev = container_of(work, struct bersa_dev, reset_work);
+	ext_phy = dev->mt76.phy2;
+	phy2 = ext_phy ? ext_phy->priv : NULL;
+
+	tri_phy = dev->mt76.phy3;
+	phy3 = tri_phy ? tri_phy->priv : NULL;
+
+	if (!(READ_ONCE(dev->reset_state) & MT_MCU_CMD_STOP_DMA))
+		return;
+
+	ieee80211_stop_queues(mt76_hw(dev));
+	if (ext_phy)
+		ieee80211_stop_queues(ext_phy->hw);
+	if (tri_phy)
+		ieee80211_stop_queues(tri_phy->hw);
+
+	set_bit(MT76_RESET, &dev->mphy.state);
+	set_bit(MT76_MCU_RESET, &dev->mphy.state);
+	wake_up(&dev->mt76.mcu.wait);
+	cancel_delayed_work_sync(&dev->mphy.mac_work);
+	if (phy2) {
+		set_bit(MT76_RESET, &phy2->mt76->state);
+		cancel_delayed_work_sync(&phy2->mt76->mac_work);
+	}
+	if (phy3) {
+		set_bit(MT76_RESET, &phy3->mt76->state);
+		cancel_delayed_work_sync(&phy3->mt76->mac_work);
+	}
+	mt76_worker_disable(&dev->mt76.tx_worker);
+	napi_disable(&dev->mt76.napi[0]);
+	napi_disable(&dev->mt76.napi[1]);
+	napi_disable(&dev->mt76.napi[2]);
+	napi_disable(&dev->mt76.tx_napi);
+
+	mutex_lock(&dev->mt76.mutex);
+
+	mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_DMA_STOPPED);
+
+	if (bersa_wait_reset_state(dev, MT_MCU_CMD_RESET_DONE)) {
+		bersa_dma_reset(dev);
+
+		bersa_tx_token_put(dev);
+		idr_init(&dev->mt76.token);
+
+		mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_DMA_INIT);
+		bersa_wait_reset_state(dev, MT_MCU_CMD_RECOVERY_DONE);
+	}
+
+	clear_bit(MT76_MCU_RESET, &dev->mphy.state);
+	clear_bit(MT76_RESET, &dev->mphy.state);
+	if (phy2)
+		clear_bit(MT76_RESET, &phy2->mt76->state);
+	if (phy3)
+		clear_bit(MT76_RESET, &phy3->mt76->state);
+
+	local_bh_disable();
+	napi_enable(&dev->mt76.napi[0]);
+	napi_schedule(&dev->mt76.napi[0]);
+
+	napi_enable(&dev->mt76.napi[1]);
+	napi_schedule(&dev->mt76.napi[1]);
+
+	napi_enable(&dev->mt76.napi[2]);
+	napi_schedule(&dev->mt76.napi[2]);
+	local_bh_enable();
+
+	tasklet_schedule(&dev->irq_tasklet);
+
+	mt76_wr(dev, MT_MCU_INT_EVENT, MT_MCU_INT_EVENT_RESET_DONE);
+	bersa_wait_reset_state(dev, MT_MCU_CMD_NORMAL_STATE);
+
+	mt76_worker_enable(&dev->mt76.tx_worker);
+
+	napi_enable(&dev->mt76.tx_napi);
+	napi_schedule(&dev->mt76.tx_napi);
+
+	ieee80211_wake_queues(mt76_hw(dev));
+	if (ext_phy)
+		ieee80211_wake_queues(ext_phy->hw);
+	if (tri_phy)
+		ieee80211_wake_queues(tri_phy->hw);
+
+	mutex_unlock(&dev->mt76.mutex);
+
+	bersa_update_beacons(dev);
+
+	ieee80211_queue_delayed_work(mt76_hw(dev), &dev->mphy.mac_work,
+				     BERSA_WATCHDOG_TIME);
+	if (phy2)
+		ieee80211_queue_delayed_work(ext_phy->hw,
+					     &phy2->mt76->mac_work,
+					     BERSA_WATCHDOG_TIME);
+	if (phy3)
+		ieee80211_queue_delayed_work(tri_phy->hw,
+					     &phy3->mt76->mac_work,
+					     BERSA_WATCHDOG_TIME);
+}
+
+void bersa_mac_update_stats(struct bersa_phy *phy)
+{
+	struct bersa_dev *dev = phy->dev;
+	struct mib_stats *mib = &phy->mib;
+	int i, aggr0, cnt;
+	u32 val;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR3(phy->band_idx));
+	mib->fcs_err_cnt +=
+		FIELD_GET(MT_MIB_SDR3_FCS_ERR_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR4(phy->band_idx));
+	mib->rx_fifo_full_cnt += FIELD_GET(MT_MIB_SDR4_RX_FIFO_FULL_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR5(phy->band_idx));
+	mib->rx_mpdu_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR6(phy->band_idx));
+	mib->channel_idle_cnt += FIELD_GET(MT_MIB_SDR6_CHANNEL_IDL_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR7(phy->band_idx));
+	mib->rx_vector_mismatch_cnt += FIELD_GET(MT_MIB_SDR7_RX_VECTOR_MISMATCH_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR8(phy->band_idx));
+	mib->rx_delimiter_fail_cnt += FIELD_GET(MT_MIB_SDR8_RX_DELIMITER_FAIL_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR11(phy->band_idx));
+	mib->rx_len_mismatch_cnt += FIELD_GET(MT_MIB_SDR11_RX_LEN_MISMATCH_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR12(phy->band_idx));
+	mib->tx_ampdu_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR13(phy->band_idx));
+	mib->tx_stop_q_empty_cnt += FIELD_GET(MT_MIB_SDR13_TX_STOP_Q_EMPTY_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR14(phy->band_idx));
+	mib->tx_mpdu_attempts_cnt +=
+		FIELD_GET(MT_MIB_SDR14_TX_MPDU_ATTEMPTS_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR15(phy->band_idx));
+	mib->tx_mpdu_success_cnt +=
+		FIELD_GET(MT_MIB_SDR15_TX_MPDU_SUCCESS_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR22(phy->band_idx));
+	mib->rx_ampdu_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR23(phy->band_idx));
+	mib->rx_ampdu_bytes_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR24(phy->band_idx));
+	mib->rx_ampdu_valid_subframe_cnt +=
+		FIELD_GET(MT_MIB_SDR24_RX_AMPDU_SF_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR25(phy->band_idx));
+	mib->rx_ampdu_valid_subframe_bytes_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR27(phy->band_idx));
+	mib->tx_rwp_fail_cnt += FIELD_GET(MT_MIB_SDR27_TX_RWP_FAIL_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR28(phy->band_idx));
+	mib->tx_rwp_need_cnt += FIELD_GET(MT_MIB_SDR28_TX_RWP_NEED_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR29(phy->band_idx));
+	mib->rx_pfdrop_cnt +=
+		FIELD_GET(MT_MIB_SDR29_RX_PFDROP_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDRVEC(phy->band_idx));
+	mib->rx_vec_queue_overflow_drop_cnt +=
+		FIELD_GET(MT_MIB_SDR30_RX_VEC_QUEUE_OVERFLOW_DROP_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDR31(phy->band_idx));
+	mib->rx_ba_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_SDR32(phy->band_idx));
+	mib->tx_pkt_ebf_cnt += FIELD_GET(MT_MIB_SDR32_TX_PKT_EBF_CNT_MASK, cnt);
+	mib->tx_pkt_ibf_cnt += FIELD_GET(MT_MIB_SDR32_TX_PKT_IBF_CNT_MASK, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_SDRMUBF(phy->band_idx));
+	mib->tx_bf_cnt += FIELD_GET(MT_MIB_MU_BF_TX_CNT, cnt);
+
+	cnt = mt76_rr(dev, MT_MIB_DR8(phy->band_idx));
+	mib->tx_mu_mpdu_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_DR9(phy->band_idx));
+	mib->tx_mu_acked_mpdu_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_MIB_DR11(phy->band_idx));
+	mib->tx_su_acked_mpdu_cnt += cnt;
+
+	cnt = mt76_rr(dev, MT_ETBF_TX_APP_CNT(phy->band_idx));
+	mib->tx_bf_ibf_ppdu_cnt += FIELD_GET(MT_ETBF_TX_IBF_CNT, cnt);
+	mib->tx_bf_ebf_ppdu_cnt += FIELD_GET(MT_ETBF_TX_EBF_CNT, cnt);
+
+	cnt = mt76_rr(dev, MT_ETBF_RX_FB_CNT(phy->band_idx));
+	mib->tx_bf_rx_fb_all_cnt += FIELD_GET(MT_ETBF_RX_FB_ALL, cnt);
+	mib->tx_bf_rx_fb_he_cnt += FIELD_GET(MT_ETBF_RX_FB_HE, cnt);
+	mib->tx_bf_rx_fb_vht_cnt += FIELD_GET(MT_ETBF_RX_FB_VHT, cnt);
+	mib->tx_bf_rx_fb_ht_cnt += FIELD_GET(MT_ETBF_RX_FB_HT, cnt);
+
+	cnt = mt76_rr(dev, MT_ETBF_RX_FB_CONT(phy->band_idx));
+	mib->tx_bf_rx_fb_bw = FIELD_GET(MT_ETBF_RX_FB_BW, cnt);
+	mib->tx_bf_rx_fb_nc_cnt += FIELD_GET(MT_ETBF_RX_FB_NC, cnt);
+	mib->tx_bf_rx_fb_nr_cnt += FIELD_GET(MT_ETBF_RX_FB_NR, cnt);
+
+	cnt = mt76_rr(dev, MT_ETBF_TX_NDP_BFRP(phy->band_idx));
+	mib->tx_bf_fb_cpl_cnt += FIELD_GET(MT_ETBF_TX_FB_CPL, cnt);
+	mib->tx_bf_fb_trig_cnt += FIELD_GET(MT_ETBF_TX_FB_TRI, cnt);
+
+	for (i = 0; i < ARRAY_SIZE(mib->tx_amsdu); i++) {
+		cnt = mt76_rr(dev, MT_PLE_AMSDU_PACK_MSDU_CNT(i));
+		mib->tx_amsdu[i] += cnt;
+		mib->tx_amsdu_cnt += cnt;
+	}
+
+	aggr0 = phy->band_idx ? ARRAY_SIZE(dev->mt76.aggr_stats) / 2 : 0;
+	for (i = 0; i < 2; i++) {
+		/* rts count */
+		val = mt76_rr(dev, MT_MIB_MB_SDR0(phy->band_idx, (i << 2)));
+		mib->rts_cnt += FIELD_GET(GENMASK(15, 0), val);
+		mib->rts_cnt += FIELD_GET(GENMASK(31, 16), val);
+
+		/* rts retry count */
+		val = mt76_rr(dev, MT_MIB_MB_SDR1(phy->band_idx, (i << 2)));
+		mib->rts_retries_cnt += FIELD_GET(GENMASK(15, 0), val);
+		mib->rts_retries_cnt += FIELD_GET(GENMASK(31, 16), val);
+
+		/* ba miss count */
+		val = mt76_rr(dev, MT_MIB_MB_SDR2(phy->band_idx, (i << 2)));
+		mib->ba_miss_cnt += FIELD_GET(GENMASK(15, 0), val);
+		mib->ba_miss_cnt += FIELD_GET(GENMASK(31, 16), val);
+
+		/* ack fail count */
+		val = mt76_rr(dev, MT_MIB_MB_BFTF(phy->band_idx, (i << 2)));
+		mib->ack_fail_cnt += FIELD_GET(GENMASK(15, 0), val);
+		mib->ack_fail_cnt += FIELD_GET(GENMASK(31, 16), val);
+	}
+
+	for (i = 0; i < 8; i++) {
+		val = mt76_rr(dev, MT_TX_AGG_CNT(phy->band_idx, i));
+		dev->mt76.aggr_stats[aggr0++] += FIELD_GET(GENMASK(15, 0), val);
+		dev->mt76.aggr_stats[aggr0++] += FIELD_GET(GENMASK(31, 16), val);
+	}
+}
+
+void bersa_mac_sta_rc_work(struct work_struct *work)
+{
+	struct bersa_dev *dev = container_of(work, struct bersa_dev, rc_work);
+	struct ieee80211_sta *sta;
+	struct ieee80211_vif *vif;
+	struct bersa_sta *msta;
+	u32 changed;
+	LIST_HEAD(list);
+
+	spin_lock_bh(&dev->sta_poll_lock);
+	list_splice_init(&dev->sta_rc_list, &list);
+
+	while (!list_empty(&list)) {
+		msta = list_first_entry(&list, struct bersa_sta, rc_list);
+		list_del_init(&msta->rc_list);
+		changed = msta->changed;
+		msta->changed = 0;
+		spin_unlock_bh(&dev->sta_poll_lock);
+
+		sta = container_of((void *)msta, struct ieee80211_sta, drv_priv);
+		vif = container_of((void *)msta->vif, struct ieee80211_vif, drv_priv);
+
+		if (changed & (IEEE80211_RC_SUPP_RATES_CHANGED |
+			       IEEE80211_RC_NSS_CHANGED |
+			       IEEE80211_RC_BW_CHANGED))
+			bersa_mcu_add_rate_ctrl(dev, vif, sta, true);
+
+		if (changed & IEEE80211_RC_SMPS_CHANGED)
+			bersa_mcu_add_smps(dev, vif, sta);
+
+		spin_lock_bh(&dev->sta_poll_lock);
+	}
+
+	spin_unlock_bh(&dev->sta_poll_lock);
+}
+
+void bersa_mac_work(struct work_struct *work)
+{
+	struct bersa_phy *phy;
+	struct mt76_phy *mphy;
+
+	mphy = (struct mt76_phy *)container_of(work, struct mt76_phy,
+					       mac_work.work);
+	phy = mphy->priv;
+
+	mutex_lock(&mphy->dev->mutex);
+
+	/* TODO: to be checked */
+	/* mt76_update_survey(mphy); */
+	/* if (++mphy->mac_work_count == 5) { */
+	/* 	mphy->mac_work_count = 0; */
+
+	/* 	bersa_mac_update_stats(phy); */
+	/* } */
+
+	mutex_unlock(&mphy->dev->mutex);
+
+	mt76_tx_status_check(mphy->dev, false);
+
+	ieee80211_queue_delayed_work(mphy->hw, &mphy->mac_work,
+				     BERSA_WATCHDOG_TIME);
+}
+
+static void bersa_dfs_stop_radar_detector(struct bersa_phy *phy)
+{
+	struct bersa_dev *dev = phy->dev;
+
+	if (phy->rdd_state & BIT(0))
+		bersa_mcu_rdd_cmd(dev, RDD_STOP, 0,
+				  MT_RX_SEL0, 0);
+	if (phy->rdd_state & BIT(1))
+		bersa_mcu_rdd_cmd(dev, RDD_STOP, 1,
+				  MT_RX_SEL0, 0);
+}
+
+static int bersa_dfs_start_rdd(struct bersa_dev *dev, int chain)
+{
+	int err;
+
+	err = bersa_mcu_rdd_cmd(dev, RDD_START, chain,
+				MT_RX_SEL0, 0);
+	if (err < 0)
+		return err;
+
+	return bersa_mcu_rdd_cmd(dev, RDD_DET_MODE, chain,
+				 MT_RX_SEL0, 1);
+}
+
+static int bersa_dfs_start_radar_detector(struct bersa_phy *phy)
+{
+	struct cfg80211_chan_def *chandef = &phy->mt76->chandef;
+	struct bersa_dev *dev = phy->dev;
+	int err;
+
+	/* start CAC */
+	err = bersa_mcu_rdd_cmd(dev, RDD_CAC_START, phy->band_idx,
+				MT_RX_SEL0, 0);
+	if (err < 0)
+		return err;
+
+	err = bersa_dfs_start_rdd(dev, phy->band_idx);
+	if (err < 0)
+		return err;
+
+	phy->rdd_state |= BIT(phy->band_idx);
+
+	if (is_mt7916(&dev->mt76))
+		return 0;
+
+	if (chandef->width == NL80211_CHAN_WIDTH_160 ||
+	    chandef->width == NL80211_CHAN_WIDTH_80P80) {
+		err = bersa_dfs_start_rdd(dev, 1);
+		if (err < 0)
+			return err;
+
+		phy->rdd_state |= BIT(1);
+	}
+
+	return 0;
+}
+
+static int
+bersa_dfs_init_radar_specs(struct bersa_phy *phy)
+{
+	const struct bersa_dfs_radar_spec *radar_specs;
+	struct bersa_dev *dev = phy->dev;
+	int err, i;
+
+	switch (dev->mt76.region) {
+	case NL80211_DFS_FCC:
+		radar_specs = &fcc_radar_specs;
+		err = bersa_mcu_set_fcc5_lpn(dev, 8);
+		if (err < 0)
+			return err;
+		break;
+	case NL80211_DFS_ETSI:
+		radar_specs = &etsi_radar_specs;
+		break;
+	case NL80211_DFS_JP:
+		radar_specs = &jp_radar_specs;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	for (i = 0; i < ARRAY_SIZE(radar_specs->radar_pattern); i++) {
+		err = bersa_mcu_set_radar_th(dev, i,
+					      &radar_specs->radar_pattern[i]);
+		if (err < 0)
+			return err;
+	}
+
+	return bersa_mcu_set_pulse_th(dev, &radar_specs->pulse_th);
+}
+
+int bersa_dfs_init_radar_detector(struct bersa_phy *phy)
+{
+	struct bersa_dev *dev = phy->dev;
+	enum mt76_dfs_state dfs_state, prev_state;
+	int err;
+
+	prev_state = phy->mt76->dfs_state;
+	dfs_state = mt76_phy_dfs_state(phy->mt76);
+
+	if (prev_state == dfs_state)
+		return 0;
+
+	if (prev_state == MT_DFS_STATE_UNKNOWN)
+		bersa_dfs_stop_radar_detector(phy);
+
+	if (dfs_state == MT_DFS_STATE_DISABLED)
+		goto stop;
+
+	if (prev_state <= MT_DFS_STATE_DISABLED) {
+		err = bersa_dfs_init_radar_specs(phy);
+		if (err < 0)
+			return err;
+
+		err = bersa_dfs_start_radar_detector(phy);
+		if (err < 0)
+			return err;
+
+		phy->mt76->dfs_state = MT_DFS_STATE_CAC;
+	}
+
+	if (dfs_state == MT_DFS_STATE_CAC)
+		return 0;
+
+	err = bersa_mcu_rdd_cmd(dev, RDD_CAC_END,
+				phy->band_idx, MT_RX_SEL0, 0);
+	if (err < 0) {
+		phy->mt76->dfs_state = MT_DFS_STATE_UNKNOWN;
+		return err;
+	}
+
+	phy->mt76->dfs_state = MT_DFS_STATE_ACTIVE;
+	return 0;
+
+stop:
+	err = bersa_mcu_rdd_cmd(dev, RDD_NORMAL_START,
+				phy->band_idx, MT_RX_SEL0, 0);
+	if (err < 0)
+		return err;
+
+	bersa_dfs_stop_radar_detector(phy);
+	phy->mt76->dfs_state = MT_DFS_STATE_DISABLED;
+
+	return 0;
+}
+
+static int
+bersa_mac_twt_duration_align(int duration)
+{
+	return duration << 8;
+}
+
+static u64
+bersa_mac_twt_sched_list_add(struct bersa_dev *dev,
+			      struct bersa_twt_flow *flow)
+{
+	struct bersa_twt_flow *iter, *iter_next;
+	u32 duration = flow->duration << 8;
+	u64 start_tsf;
+
+	iter = list_first_entry_or_null(&dev->twt_list,
+					struct bersa_twt_flow, list);
+	if (!iter || !iter->sched || iter->start_tsf > duration) {
+		/* add flow as first entry in the list */
+		list_add(&flow->list, &dev->twt_list);
+		return 0;
+	}
+
+	list_for_each_entry_safe(iter, iter_next, &dev->twt_list, list) {
+		start_tsf = iter->start_tsf +
+			    bersa_mac_twt_duration_align(iter->duration);
+		if (list_is_last(&iter->list, &dev->twt_list))
+			break;
+
+		if (!iter_next->sched ||
+		    iter_next->start_tsf > start_tsf + duration) {
+			list_add(&flow->list, &iter->list);
+			goto out;
+		}
+	}
+
+	/* add flow as last entry in the list */
+	list_add_tail(&flow->list, &dev->twt_list);
+out:
+	return start_tsf;
+}
+
+static int bersa_mac_check_twt_req(struct ieee80211_twt_setup *twt)
+{
+	struct ieee80211_twt_params *twt_agrt;
+	u64 interval, duration;
+	u16 mantissa;
+	u8 exp;
+
+	/* only individual agreement supported */
+	if (twt->control & IEEE80211_TWT_CONTROL_NEG_TYPE_BROADCAST)
+		return -EOPNOTSUPP;
+
+	/* only 256us unit supported */
+	if (twt->control & IEEE80211_TWT_CONTROL_WAKE_DUR_UNIT)
+		return -EOPNOTSUPP;
+
+	twt_agrt = (struct ieee80211_twt_params *)twt->params;
+
+	/* explicit agreement not supported */
+	if (!(twt_agrt->req_type & cpu_to_le16(IEEE80211_TWT_REQTYPE_IMPLICIT)))
+		return -EOPNOTSUPP;
+
+	exp = FIELD_GET(IEEE80211_TWT_REQTYPE_WAKE_INT_EXP,
+			le16_to_cpu(twt_agrt->req_type));
+	mantissa = le16_to_cpu(twt_agrt->mantissa);
+	duration = twt_agrt->min_twt_dur << 8;
+
+	interval = (u64)mantissa << exp;
+	if (interval < duration)
+		return -EOPNOTSUPP;
+
+	return 0;
+}
+
+void bersa_mac_add_twt_setup(struct ieee80211_hw *hw,
+			      struct ieee80211_sta *sta,
+			      struct ieee80211_twt_setup *twt)
+{
+	enum ieee80211_twt_setup_cmd setup_cmd = TWT_SETUP_CMD_REJECT;
+	struct bersa_sta *msta = (struct bersa_sta *)sta->drv_priv;
+	struct ieee80211_twt_params *twt_agrt = (void *)twt->params;
+	u16 req_type = le16_to_cpu(twt_agrt->req_type);
+	enum ieee80211_twt_setup_cmd sta_setup_cmd;
+	struct bersa_dev *dev = bersa_hw_dev(hw);
+	struct bersa_twt_flow *flow;
+	int flowid, table_id;
+	u8 exp;
+
+	if (bersa_mac_check_twt_req(twt))
+		goto out;
+
+	mutex_lock(&dev->mt76.mutex);
+
+	if (dev->twt.n_agrt == BERSA_MAX_TWT_AGRT)
+		goto unlock;
+
+	if (hweight8(msta->twt.flowid_mask) == ARRAY_SIZE(msta->twt.flow))
+		goto unlock;
+
+	flowid = ffs(~msta->twt.flowid_mask) - 1;
+	le16p_replace_bits(&twt_agrt->req_type, flowid,
+			   IEEE80211_TWT_REQTYPE_FLOWID);
+
+	table_id = ffs(~dev->twt.table_mask) - 1;
+	exp = FIELD_GET(IEEE80211_TWT_REQTYPE_WAKE_INT_EXP, req_type);
+	sta_setup_cmd = FIELD_GET(IEEE80211_TWT_REQTYPE_SETUP_CMD, req_type);
+
+	flow = &msta->twt.flow[flowid];
+	memset(flow, 0, sizeof(*flow));
+	INIT_LIST_HEAD(&flow->list);
+	flow->wcid = msta->wcid.idx;
+	flow->table_id = table_id;
+	flow->id = flowid;
+	flow->duration = twt_agrt->min_twt_dur;
+	flow->mantissa = twt_agrt->mantissa;
+	flow->exp = exp;
+	flow->protection = !!(req_type & IEEE80211_TWT_REQTYPE_PROTECTION);
+	flow->flowtype = !!(req_type & IEEE80211_TWT_REQTYPE_FLOWTYPE);
+	flow->trigger = !!(req_type & IEEE80211_TWT_REQTYPE_TRIGGER);
+
+	if (sta_setup_cmd == TWT_SETUP_CMD_REQUEST ||
+	    sta_setup_cmd == TWT_SETUP_CMD_SUGGEST) {
+		u64 interval = (u64)le16_to_cpu(twt_agrt->mantissa) << exp;
+		u64 flow_tsf, curr_tsf;
+		u32 rem;
+
+		flow->sched = true;
+		flow->start_tsf = bersa_mac_twt_sched_list_add(dev, flow);
+		curr_tsf = __bersa_get_tsf(hw, msta->vif);
+		div_u64_rem(curr_tsf - flow->start_tsf, interval, &rem);
+		flow_tsf = curr_tsf + interval - rem;
+		twt_agrt->twt = cpu_to_le64(flow_tsf);
+	} else {
+		list_add_tail(&flow->list, &dev->twt_list);
+	}
+	flow->tsf = le64_to_cpu(twt_agrt->twt);
+
+	if (bersa_mcu_twt_agrt_update(dev, msta->vif, flow, MCU_TWT_AGRT_ADD))
+		goto unlock;
+
+	setup_cmd = TWT_SETUP_CMD_ACCEPT;
+	dev->twt.table_mask |= BIT(table_id);
+	msta->twt.flowid_mask |= BIT(flowid);
+	dev->twt.n_agrt++;
+
+unlock:
+	mutex_unlock(&dev->mt76.mutex);
+out:
+	le16p_replace_bits(&twt_agrt->req_type, setup_cmd,
+			   IEEE80211_TWT_REQTYPE_SETUP_CMD);
+	twt->control = (twt->control & IEEE80211_TWT_CONTROL_WAKE_DUR_UNIT) |
+		       (twt->control & IEEE80211_TWT_CONTROL_RX_DISABLED);
+}
+
+void bersa_mac_twt_teardown_flow(struct bersa_dev *dev,
+				  struct bersa_sta *msta,
+				  u8 flowid)
+{
+	struct bersa_twt_flow *flow;
+
+	lockdep_assert_held(&dev->mt76.mutex);
+
+	if (flowid >= ARRAY_SIZE(msta->twt.flow))
+		return;
+
+	if (!(msta->twt.flowid_mask & BIT(flowid)))
+		return;
+
+	flow = &msta->twt.flow[flowid];
+	if (bersa_mcu_twt_agrt_update(dev, msta->vif, flow,
+				       MCU_TWT_AGRT_DELETE))
+		return;
+
+	list_del_init(&flow->list);
+	msta->twt.flowid_mask &= ~BIT(flowid);
+	dev->twt.table_mask &= ~BIT(flow->table_id);
+	dev->twt.n_agrt--;
+}