ath10k: enable raw encap mode and software crypto engine

This patch enables raw Rx/Tx encap mode to support software based
crypto engine. This patch introduces a new module param 'cryptmode'.

 cryptmode:

   0: Use hardware crypto engine globally with native Wi-Fi mode TX/RX
      encapsulation to the firmware. This is the default mode.
   1: Use sofware crypto engine globally with raw mode TX/RX
      encapsulation to the firmware.

Known limitation:
   A-MSDU must be disabled for RAW Tx encap mode to perform well when
   heavy traffic is applied.

Testing: (by Michal Kazior <michal.kazior@tieto.com>)

     a) Performance Testing

      cryptmode=1
       ap=qca988x sta=killer1525
        killer1525  ->  qca988x     194.496 mbps [tcp1 ip4]
        killer1525  ->  qca988x     238.309 mbps [tcp5 ip4]
        killer1525  ->  qca988x     266.958 mbps [udp1 ip4]
        killer1525  ->  qca988x     477.468 mbps [udp5 ip4]
        qca988x     ->  killer1525  301.378 mbps [tcp1 ip4]
        qca988x     ->  killer1525  297.949 mbps [tcp5 ip4]
        qca988x     ->  killer1525  331.351 mbps [udp1 ip4]
        qca988x     ->  killer1525  371.528 mbps [udp5 ip4]
       ap=killer1525 sta=qca988x
        qca988x     ->  killer1525  331.447 mbps [tcp1 ip4]
        qca988x     ->  killer1525  328.783 mbps [tcp5 ip4]
        qca988x     ->  killer1525  375.309 mbps [udp1 ip4]
        qca988x     ->  killer1525  403.379 mbps [udp5 ip4]
        killer1525  ->  qca988x     203.689 mbps [tcp1 ip4]
        killer1525  ->  qca988x     222.339 mbps [tcp5 ip4]
        killer1525  ->  qca988x     264.199 mbps [udp1 ip4]
        killer1525  ->  qca988x     479.371 mbps [udp5 ip4]

      Note:
       - only open network tested for RAW vs nwifi performance comparison
       - killer1525 (qca6174 hw2.2) is 2x2 device (hence max 866mbps)
       - used iperf
       - OTA, devices a few cm apart from each other, no shielding
       - tcpX/udpX, X - means number of threads used

      Overview:
       - relative Tx performance drop is seen but is within reasonable and
         expected threshold (A-MSDU must be disabled with RAW Tx)

     b) Connectivity Testing

      cryptmode=1
       ap=iwl6205 sta1=qca988x crypto=open     topology-1ap1sta          OK
       ap=iwl6205 sta1=qca988x crypto=wep1     topology-1ap1sta          OK
       ap=iwl6205 sta1=qca988x crypto=wpa      topology-1ap1sta          OK
       ap=iwl6205 sta1=qca988x crypto=wpa-ccmp topology-1ap1sta          OK
       ap=qca988x sta1=iwl6205 crypto=open     topology-1ap1sta          OK
       ap=qca988x sta1=iwl6205 crypto=wep1     topology-1ap1sta          OK
       ap=qca988x sta1=iwl6205 crypto=wpa      topology-1ap1sta          OK
       ap=qca988x sta1=iwl6205 crypto=wpa-ccmp topology-1ap1sta          OK
       ap=iwl6205 sta1=qca988x crypto=open     topology-1ap1sta2br       OK
       ap=iwl6205 sta1=qca988x crypto=wep1     topology-1ap1sta2br       OK
       ap=iwl6205 sta1=qca988x crypto=wpa      topology-1ap1sta2br       OK
       ap=iwl6205 sta1=qca988x crypto=wpa-ccmp topology-1ap1sta2br       OK
       ap=qca988x sta1=iwl6205 crypto=open     topology-1ap1sta2br       OK
       ap=qca988x sta1=iwl6205 crypto=wep1     topology-1ap1sta2br       OK
       ap=qca988x sta1=iwl6205 crypto=wpa      topology-1ap1sta2br       OK
       ap=qca988x sta1=iwl6205 crypto=wpa-ccmp topology-1ap1sta2br       OK
       ap=iwl6205 sta1=qca988x crypto=open     topology-1ap1sta2br1vlan  OK
       ap=iwl6205 sta1=qca988x crypto=wep1     topology-1ap1sta2br1vlan  OK
       ap=iwl6205 sta1=qca988x crypto=wpa      topology-1ap1sta2br1vlan  OK
       ap=iwl6205 sta1=qca988x crypto=wpa-ccmp topology-1ap1sta2br1vlan  OK
       ap=qca988x sta1=iwl6205 crypto=open     topology-1ap1sta2br1vlan  OK
       ap=qca988x sta1=iwl6205 crypto=wep1     topology-1ap1sta2br1vlan  OK
       ap=qca988x sta1=iwl6205 crypto=wpa      topology-1ap1sta2br1vlan  OK
       ap=qca988x sta1=iwl6205 crypto=wpa-ccmp topology-1ap1sta2br1vlan  OK

      Note:
       - each test takes all possible endpoint pairs and pings
       - each pair-ping flushes arp table
       - ip6 is used

     c) Testbed Topology:

      1ap1sta:
        [ap] ---- [sta]

        endpoints: ap, sta

      1ap1sta2br:
        [veth0] [ap] ---- [sta] [veth2]
           |     |          |     |
        [veth1]  |          \   [veth3]
            \   /            \  /
            [br0]            [br1]

        endpoints: veth0, veth2, br0, br1
        note: STA works in 4addr mode, AP has wds_sta=1

      1ap1sta2br1vlan:
        [veth0] [ap] ---- [sta] [veth2]
           |     |          |     |
        [veth1]  |          \   [veth3]
            \   /            \  /
          [br0]              [br1]
            |                  |
          [vlan0_id2]        [vlan1_id2]

        endpoints: vlan0_id2, vlan1_id2
        note: STA works in 4addr mode, AP has wds_sta=1

Credits:

    Thanks to Michal Kazior <michal.kazior@tieto.com> who helped find the
    amsdu issue, contributed a workaround (already squashed into this
    patch), and contributed the throughput and connectivity tests results.

Signed-off-by: David Liu <cfliu.tw@gmail.com>
Signed-off-by: Michal Kazior <michal.kazior@tieto.com>
Tested-by: Michal Kazior <michal.kazior@tieto.com>
Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com>

drivers/net/wireless/ath/ath10k/core.c

@@ -31,16 +31,19 @@
 #include "wmi-ops.h"
 
 unsigned int ath10k_debug_mask;
+static unsigned int ath10k_cryptmode_param;
 static bool uart_print;
 static bool skip_otp;
 
 module_param_named(debug_mask, ath10k_debug_mask, uint, 0644);
+module_param_named(cryptmode, ath10k_cryptmode_param, uint, 0644);
 module_param(uart_print, bool, 0644);
 module_param(skip_otp, bool, 0644);
 
 MODULE_PARM_DESC(debug_mask, "Debugging mask");
 MODULE_PARM_DESC(uart_print, "Uart target debugging");
 MODULE_PARM_DESC(skip_otp, "Skip otp failure for calibration in testmode");
+MODULE_PARM_DESC(cryptmode, "Crypto mode: 0-hardware, 1-software");
 
 static const struct ath10k_hw_params ath10k_hw_params_list[] = {
 	{
@@ -1073,6 +1076,46 @@ static int ath10k_core_init_firmware_features(struct ath10k *ar)
 		return -EINVAL;
 	}
 
+	ar->wmi.rx_decap_mode = ATH10K_HW_TXRX_NATIVE_WIFI;
+	switch (ath10k_cryptmode_param) {
+	case ATH10K_CRYPT_MODE_HW:
+		clear_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags);
+		clear_bit(ATH10K_FLAG_HW_CRYPTO_DISABLED, &ar->dev_flags);
+		break;
+	case ATH10K_CRYPT_MODE_SW:
+		if (!test_bit(ATH10K_FW_FEATURE_RAW_MODE_SUPPORT,
+			      ar->fw_features)) {
+			ath10k_err(ar, "cryptmode > 0 requires raw mode support from firmware");
+			return -EINVAL;
+		}
+
+		set_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags);
+		set_bit(ATH10K_FLAG_HW_CRYPTO_DISABLED, &ar->dev_flags);
+		break;
+	default:
+		ath10k_info(ar, "invalid cryptmode: %d\n",
+			    ath10k_cryptmode_param);
+		return -EINVAL;
+	}
+
+	ar->htt.max_num_amsdu = ATH10K_HTT_MAX_NUM_AMSDU_DEFAULT;
+	ar->htt.max_num_ampdu = ATH10K_HTT_MAX_NUM_AMPDU_DEFAULT;
+
+	if (test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags)) {
+		ar->wmi.rx_decap_mode = ATH10K_HW_TXRX_RAW;
+
+		/* Workaround:
+		 *
+		 * Firmware A-MSDU aggregation breaks with RAW Tx encap mode
+		 * and causes enormous performance issues (malformed frames,
+		 * etc).
+		 *
+		 * Disabling A-MSDU makes RAW mode stable with heavy traffic
+		 * albeit a bit slower compared to regular operation.
+		 */
+		ar->htt.max_num_amsdu = 1;
+	}
+
 	/* Backwards compatibility for firmwares without
 	 * ATH10K_FW_IE_WMI_OP_VERSION.
 	 */

drivers/net/wireless/ath/ath10k/core.h

@@ -92,6 +92,7 @@ struct ath10k_skb_cb {
 		u8 tid;
 		u16 freq;
 		bool is_offchan;
+		bool nohwcrypt;
 		struct ath10k_htt_txbuf *txbuf;
 		u32 txbuf_paddr;
 	} __packed htt;
@@ -152,6 +153,7 @@ struct ath10k_wmi {
 	const struct wmi_ops *ops;
 
 	u32 num_mem_chunks;
+	u32 rx_decap_mode;
 	struct ath10k_mem_chunk mem_chunks[WMI_MAX_MEM_REQS];
 };
 
@@ -341,6 +343,7 @@ struct ath10k_vif {
 	} u;
 
 	bool use_cts_prot;
+	bool nohwcrypt;
 	int num_legacy_stations;
 	int txpower;
 	struct wmi_wmm_params_all_arg wmm_params;
@@ -382,9 +385,6 @@ struct ath10k_debug {
 	u32 reg_addr;
 	u32 nf_cal_period;
 
-	u8 htt_max_amsdu;
-	u8 htt_max_ampdu;
-
 	struct ath10k_fw_crash_data *fw_crash_data;
 };
 
@@ -453,16 +453,21 @@ enum ath10k_fw_features {
 	ATH10K_FW_FEATURE_WOWLAN_SUPPORT = 6,
 
 	/* Don't trust error code from otp.bin */
-	ATH10K_FW_FEATURE_IGNORE_OTP_RESULT,
+	ATH10K_FW_FEATURE_IGNORE_OTP_RESULT = 7,
 
 	/* Some firmware revisions pad 4th hw address to 4 byte boundary making
 	 * it 8 bytes long in Native Wifi Rx decap.
 	 */
-	ATH10K_FW_FEATURE_NO_NWIFI_DECAP_4ADDR_PADDING,
+	ATH10K_FW_FEATURE_NO_NWIFI_DECAP_4ADDR_PADDING = 8,
 
 	/* Firmware supports bypassing PLL setting on init. */
 	ATH10K_FW_FEATURE_SUPPORTS_SKIP_CLOCK_INIT = 9,
 
+	/* Raw mode support. If supported, FW supports receiving and trasmitting
+	 * frames in raw mode.
+	 */
+	ATH10K_FW_FEATURE_RAW_MODE_SUPPORT = 10,
+
 	/* keep last */
 	ATH10K_FW_FEATURE_COUNT,
 };
@@ -476,6 +481,15 @@ enum ath10k_dev_flags {
 	 * waiters should immediately cancel instead of waiting for a time out.
 	 */
 	ATH10K_FLAG_CRASH_FLUSH,
+
+	/* Use Raw mode instead of native WiFi Tx/Rx encap mode.
+	 * Raw mode supports both hardware and software crypto. Native WiFi only
+	 * supports hardware crypto.
+	 */
+	ATH10K_FLAG_RAW_MODE,
+
+	/* Disable HW crypto engine */
+	ATH10K_FLAG_HW_CRYPTO_DISABLED,
 };
 
 enum ath10k_cal_mode {
@@ -484,6 +498,13 @@ enum ath10k_cal_mode {
 	ATH10K_CAL_MODE_DT,
 };
 
+enum ath10k_crypt_mode {
+	/* Only use hardware crypto engine */
+	ATH10K_CRYPT_MODE_HW,
+	/* Only use software crypto engine */
+	ATH10K_CRYPT_MODE_SW,
+};
+
 static inline const char *ath10k_cal_mode_str(enum ath10k_cal_mode mode)
 {
 	switch (mode) {

drivers/net/wireless/ath/ath10k/debug.c

@@ -128,7 +128,7 @@ void ath10k_print_driver_info(struct ath10k *ar)
 
 	ath10k_core_get_fw_features_str(ar, fw_features, sizeof(fw_features));
 
-	ath10k_info(ar, "%s (0x%08x, 0x%08x%s%s%s) fw %s api %d htt-ver %d.%d wmi-op %d htt-op %d cal %s max-sta %d features %s\n",
+	ath10k_info(ar, "%s (0x%08x, 0x%08x%s%s%s) fw %s api %d htt-ver %d.%d wmi-op %d htt-op %d cal %s max-sta %d raw %d hwcrypto %d features %s\n",
 		    ar->hw_params.name,
 		    ar->target_version,
 		    ar->chip_id,
@@ -144,6 +144,8 @@ void ath10k_print_driver_info(struct ath10k *ar)
 		    ar->htt.op_version,
 		    ath10k_cal_mode_str(ar->cal_mode),
 		    ar->max_num_stations,
+		    test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags),
+		    !test_bit(ATH10K_FLAG_HW_CRYPTO_DISABLED, &ar->dev_flags),
 		    fw_features);
 	ath10k_info(ar, "debug %d debugfs %d tracing %d dfs %d testmode %d\n",
 		    config_enabled(CONFIG_ATH10K_DEBUG),
@@ -1363,12 +1365,8 @@ static ssize_t ath10k_read_htt_max_amsdu_ampdu(struct file *file,
 
 	mutex_lock(&ar->conf_mutex);
 
-	if (ar->debug.htt_max_amsdu)
-		amsdu = ar->debug.htt_max_amsdu;
-
-	if (ar->debug.htt_max_ampdu)
-		ampdu = ar->debug.htt_max_ampdu;
-
+	amsdu = ar->htt.max_num_amsdu;
+	ampdu = ar->htt.max_num_ampdu;
 	mutex_unlock(&ar->conf_mutex);
 
 	len = scnprintf(buf, sizeof(buf), "%u %u\n", amsdu, ampdu);
@@ -1402,8 +1400,8 @@ static ssize_t ath10k_write_htt_max_amsdu_ampdu(struct file *file,
 		goto out;
 
 	res = count;
-	ar->debug.htt_max_amsdu = amsdu;
-	ar->debug.htt_max_ampdu = ampdu;
+	ar->htt.max_num_amsdu = amsdu;
+	ar->htt.max_num_ampdu = ampdu;
 
 out:
 	mutex_unlock(&ar->conf_mutex);
@@ -1905,9 +1903,6 @@ void ath10k_debug_stop(struct ath10k *ar)
 	if (ar->debug.htt_stats_mask != 0)
 		cancel_delayed_work(&ar->debug.htt_stats_dwork);
 
-	ar->debug.htt_max_amsdu = 0;
-	ar->debug.htt_max_ampdu = 0;
-
 	ath10k_wmi_pdev_pktlog_disable(ar);
 }
 

drivers/net/wireless/ath/ath10k/htt.c

@@ -246,12 +246,31 @@ int ath10k_htt_setup(struct ath10k_htt *htt)
 	}
 
 	status = ath10k_htt_verify_version(htt);
-	if (status)
+	if (status) {
+		ath10k_warn(ar, "failed to verify htt version: %d\n",
+			    status);
 		return status;
+	}
 
 	status = ath10k_htt_send_frag_desc_bank_cfg(htt);
 	if (status)
 		return status;
 
-	return ath10k_htt_send_rx_ring_cfg_ll(htt);
+	status = ath10k_htt_send_rx_ring_cfg_ll(htt);
+	if (status) {
+		ath10k_warn(ar, "failed to setup rx ring: %d\n",
+			    status);
+		return status;
+	}
+
+	status = ath10k_htt_h2t_aggr_cfg_msg(htt,
+					     htt->max_num_ampdu,
+					     htt->max_num_amsdu);
+	if (status) {
+		ath10k_warn(ar, "failed to setup amsdu/ampdu limit: %d\n",
+			    status);
+		return status;
+	}
+
+	return 0;
 }

drivers/net/wireless/ath/ath10k/htt.h

@@ -1396,6 +1396,8 @@ struct ath10k_htt {
 	u8 target_version_minor;
 	struct completion target_version_received;
 	enum ath10k_fw_htt_op_version op_version;
+	u8 max_num_amsdu;
+	u8 max_num_ampdu;
 
 	const enum htt_t2h_msg_type *t2h_msg_types;
 	u32 t2h_msg_types_max;
@@ -1558,6 +1560,12 @@ struct htt_rx_desc {
 #define HTT_LOG2_MAX_CACHE_LINE_SIZE 7	/* 2^7 = 128 */
 #define HTT_MAX_CACHE_LINE_SIZE_MASK ((1 << HTT_LOG2_MAX_CACHE_LINE_SIZE) - 1)
 
+/* These values are default in most firmware revisions and apparently are a
+ * sweet spot performance wise.
+ */
+#define ATH10K_HTT_MAX_NUM_AMSDU_DEFAULT 3
+#define ATH10K_HTT_MAX_NUM_AMPDU_DEFAULT 64
+
 int ath10k_htt_connect(struct ath10k_htt *htt);
 int ath10k_htt_init(struct ath10k *ar);
 int ath10k_htt_setup(struct ath10k_htt *htt);

drivers/net/wireless/ath/ath10k/htt_rx.c

@@ -1017,9 +1017,8 @@ static void ath10k_htt_rx_h_undecap_raw(struct ath10k *ar,
 	skb_trim(msdu, msdu->len - FCS_LEN);
 
 	/* In most cases this will be true for sniffed frames. It makes sense
-	 * to deliver them as-is without stripping the crypto param. This would
-	 * also make sense for software based decryption (which is not
-	 * implemented in ath10k).
+	 * to deliver them as-is without stripping the crypto param. This is
+	 * necessary for software based decryption.
 	 *
 	 * If there's no error then the frame is decrypted. At least that is
 	 * the case for frames that come in via fragmented rx indication.
@@ -2066,6 +2065,8 @@ void ath10k_htt_t2h_msg_handler(struct ath10k *ar, struct sk_buff *skb)
 		break;
 	case HTT_T2H_MSG_TYPE_CHAN_CHANGE:
 		break;
+	case HTT_T2H_MSG_TYPE_AGGR_CONF:
+		break;
 	case HTT_T2H_MSG_TYPE_EN_STATS:
 	case HTT_T2H_MSG_TYPE_TX_FETCH_IND:
 	case HTT_T2H_MSG_TYPE_TX_FETCH_CONF:

drivers/net/wireless/ath/ath10k/htt_tx.c

@@ -523,8 +523,12 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 	if ((ieee80211_is_action(hdr->frame_control) ||
 	     ieee80211_is_deauth(hdr->frame_control) ||
 	     ieee80211_is_disassoc(hdr->frame_control)) &&
-	     ieee80211_has_protected(hdr->frame_control))
+	     ieee80211_has_protected(hdr->frame_control)) {
 		skb_put(msdu, IEEE80211_CCMP_MIC_LEN);
+	} else if (!skb_cb->htt.nohwcrypt &&
+		   skb_cb->txmode == ATH10K_HW_TXRX_RAW) {
+		skb_put(msdu, IEEE80211_CCMP_MIC_LEN);
+	}
 
 	skb_cb->paddr = dma_map_single(dev, msdu->data, msdu->len,
 				       DMA_TO_DEVICE);
@@ -595,12 +599,16 @@ int ath10k_htt_tx(struct ath10k_htt *htt, struct sk_buff *msdu)
 			prefetch_len);
 	skb_cb->htt.txbuf->htc_hdr.flags = 0;
 
+	if (skb_cb->htt.nohwcrypt)
+		flags0 |= HTT_DATA_TX_DESC_FLAGS0_NO_ENCRYPT;
+
 	if (!skb_cb->is_protected)
 		flags0 |= HTT_DATA_TX_DESC_FLAGS0_NO_ENCRYPT;
 
 	flags1 |= SM((u16)vdev_id, HTT_DATA_TX_DESC_FLAGS1_VDEV_ID);
 	flags1 |= SM((u16)tid, HTT_DATA_TX_DESC_FLAGS1_EXT_TID);
-	if (msdu->ip_summed == CHECKSUM_PARTIAL) {
+	if (msdu->ip_summed == CHECKSUM_PARTIAL &&
+	    !test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags)) {
 		flags1 |= HTT_DATA_TX_DESC_FLAGS1_CKSUM_L3_OFFLOAD;
 		flags1 |= HTT_DATA_TX_DESC_FLAGS1_CKSUM_L4_OFFLOAD;
 		if (ar->hw_params.continuous_frag_desc)

drivers/net/wireless/ath/ath10k/hw.h

@@ -217,14 +217,16 @@ void ath10k_hw_fill_survey_time(struct ath10k *ar, struct survey_info *survey,
 #define QCA_REV_99X0(ar) ((ar)->hw_rev == ATH10K_HW_QCA99X0)
 
 /* Known pecularities:
- *  - current FW doesn't support raw rx mode (last tested v599)
- *  - current FW dumps upon raw tx mode (last tested v599)
  *  - raw appears in nwifi decap, raw and nwifi appear in ethernet decap
  *  - raw have FCS, nwifi doesn't
  *  - ethernet frames have 802.11 header decapped and parts (base hdr, cipher
  *    param, llc/snap) are aligned to 4byte boundaries each */
 enum ath10k_hw_txrx_mode {
 	ATH10K_HW_TXRX_RAW = 0,
+
+	/* Native Wifi decap mode is used to align IP frames to 4-byte
+	 * boundaries and avoid a very expensive re-alignment in mac80211.
+	 */
 	ATH10K_HW_TXRX_NATIVE_WIFI = 1,
 	ATH10K_HW_TXRX_ETHERNET = 2,
 
@@ -286,10 +288,6 @@ enum ath10k_hw_rate_cck {
 #define TARGET_RX_TIMEOUT_LO_PRI		100
 #define TARGET_RX_TIMEOUT_HI_PRI		40
 
-/* Native Wifi decap mode is used to align IP frames to 4-byte boundaries and
- * avoid a very expensive re-alignment in mac80211. */
-#define TARGET_RX_DECAP_MODE			ATH10K_HW_TXRX_NATIVE_WIFI
-
 #define TARGET_SCAN_MAX_PENDING_REQS		4
 #define TARGET_BMISS_OFFLOAD_MAX_VDEV		3
 #define TARGET_ROAM_OFFLOAD_MAX_VDEV		3
@@ -324,7 +322,6 @@ enum ath10k_hw_rate_cck {
 #define TARGET_10X_RX_CHAIN_MASK		(BIT(0) | BIT(1) | BIT(2))
 #define TARGET_10X_RX_TIMEOUT_LO_PRI		100
 #define TARGET_10X_RX_TIMEOUT_HI_PRI		40
-#define TARGET_10X_RX_DECAP_MODE		ATH10K_HW_TXRX_NATIVE_WIFI
 #define TARGET_10X_SCAN_MAX_PENDING_REQS	4
 #define TARGET_10X_BMISS_OFFLOAD_MAX_VDEV	2
 #define TARGET_10X_ROAM_OFFLOAD_MAX_VDEV	2