Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next

Pull networking changes from David Miller:

 1) GRE now works over ipv6, from Dmitry Kozlov.

 2) Make SCTP more network namespace aware, from Eric Biederman.

 3) TEAM driver now works with non-ethernet devices, from Jiri Pirko.

 4) Make openvswitch network namespace aware, from Pravin B Shelar.

 5) IPV6 NAT implementation, from Patrick McHardy.

 6) Server side support for TCP Fast Open, from Jerry Chu and others.

 7) Packet BPF filter supports MOD and XOR, from Eric Dumazet and Daniel
    Borkmann.

 8) Increate the loopback default MTU to 64K, from Eric Dumazet.

 9) Use a per-task rather than per-socket page fragment allocator for
    outgoing networking traffic.  This benefits processes that have very
    many mostly idle sockets, which is quite common.

    From Eric Dumazet.

10) Use up to 32K for page fragment allocations, with fallbacks to
    smaller sizes when higher order page allocations fail.  Benefits are
    a) less segments for driver to process b) less calls to page
    allocator c) less waste of space.

    From Eric Dumazet.

11) Allow GRO to be used on GRE tunnels, from Eric Dumazet.

12) VXLAN device driver, one way to handle VLAN issues such as the
    limitation of 4096 VLAN IDs yet still have some level of isolation.
    From Stephen Hemminger.

13) As usual there is a large boatload of driver changes, with the scale
    perhaps tilted towards the wireless side this time around.

Fix up various fairly trivial conflicts, mostly caused by the user
namespace changes.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1012 commits)
  hyperv: Add buffer for extended info after the RNDIS response message.
  hyperv: Report actual status in receive completion packet
  hyperv: Remove extra allocated space for recv_pkt_list elements
  hyperv: Fix page buffer handling in rndis_filter_send_request()
  hyperv: Fix the missing return value in rndis_filter_set_packet_filter()
  hyperv: Fix the max_xfer_size in RNDIS initialization
  vxlan: put UDP socket in correct namespace
  vxlan: Depend on CONFIG_INET
  sfc: Fix the reported priorities of different filter types
  sfc: Remove EFX_FILTER_FLAG_RX_OVERRIDE_IP
  sfc: Fix loopback self-test with separate_tx_channels=1
  sfc: Fix MCDI structure field lookup
  sfc: Add parentheses around use of bitfield macro arguments
  sfc: Fix null function pointer in efx_sriov_channel_type
  vxlan: virtual extensible lan
  igmp: export symbol ip_mc_leave_group
  netlink: add attributes to fdb interface
  tg3: unconditionally select HWMON support when tg3 is enabled.
  Revert "net: ti cpsw ethernet: allow reading phy interface mode from DT"
  gre: fix sparse warning
  ...

5 files changed, 159 insertions, 117 deletions

diff --git a/drivers/net/wireless/iwlwifi/pcie/drv.c b/drivers/net/wireless/iwlwifi/pcie/drv.c
index f4c3500b68c..2a467539670 100644
--- a/drivers/net/wireless/iwlwifi/pcie/drv.c
+++ b/drivers/net/wireless/iwlwifi/pcie/drv.c
@@ -263,8 +263,6 @@ MODULE_DEVICE_TABLE(pci, iwl_hw_card_ids);
 /* PCI registers */
 #define PCI_CFG_RETRY_TIMEOUT	0x041
 
-#ifndef CONFIG_IWLWIFI_IDI
-
 static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 {
 	const struct iwl_cfg *cfg = (struct iwl_cfg *)(ent->driver_data);
@@ -282,8 +280,14 @@ static int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (!trans_pcie->drv)
 		goto out_free_trans;
 
+	/* register transport layer debugfs here */
+	if (iwl_trans_dbgfs_register(iwl_trans, iwl_trans->dbgfs_dir))
+		goto out_free_drv;
+
 	return 0;
 
+out_free_drv:
+	iwl_drv_stop(trans_pcie->drv);
 out_free_trans:
 	iwl_trans_pcie_free(iwl_trans);
 	pci_set_drvdata(pdev, NULL);
@@ -301,8 +305,6 @@ static void __devexit iwl_pci_remove(struct pci_dev *pdev)
 	pci_set_drvdata(pdev, NULL);
 }
 
-#endif /* CONFIG_IWLWIFI_IDI */
-
 #ifdef CONFIG_PM_SLEEP
 
 static int iwl_pci_suspend(struct device *device)
@@ -347,15 +349,6 @@ static SIMPLE_DEV_PM_OPS(iwl_dev_pm_ops, iwl_pci_suspend, iwl_pci_resume);
 
 #endif
 
-#ifdef CONFIG_IWLWIFI_IDI
-/*
- * Defined externally in iwl-idi.c
- */
-int iwl_pci_probe(struct pci_dev *pdev, const struct pci_device_id *ent);
-void __devexit iwl_pci_remove(struct pci_dev *pdev);
-
-#endif /* CONFIG_IWLWIFI_IDI */
-
 static struct pci_driver iwl_pci_driver = {
 	.name = DRV_NAME,
 	.id_table = iwl_hw_card_ids,
diff --git a/drivers/net/wireless/iwlwifi/pcie/internal.h b/drivers/net/wireless/iwlwifi/pcie/internal.h
index 4ffc18dc3a5..401178f44a3 100644
--- a/drivers/net/wireless/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/iwlwifi/pcie/internal.h
@@ -184,6 +184,7 @@ struct iwl_queue {
 
 struct iwl_pcie_tx_queue_entry {
 	struct iwl_device_cmd *cmd;
+	struct iwl_device_cmd *copy_cmd;
 	struct sk_buff *skb;
 	struct iwl_cmd_meta meta;
 };
@@ -310,7 +311,7 @@ void iwl_trans_pcie_free(struct iwl_trans *trans);
 ******************************************************/
 void iwl_bg_rx_replenish(struct work_struct *data);
 void iwl_irq_tasklet(struct iwl_trans *trans);
-void iwlagn_rx_replenish(struct iwl_trans *trans);
+void iwl_rx_replenish(struct iwl_trans *trans);
 void iwl_rx_queue_update_write_ptr(struct iwl_trans *trans,
 				   struct iwl_rx_queue *q);
 
diff --git a/drivers/net/wireless/iwlwifi/pcie/rx.c b/drivers/net/wireless/iwlwifi/pcie/rx.c
index d1a61ba6247..17c8e5d8268 100644
--- a/drivers/net/wireless/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/iwlwifi/pcie/rx.c
@@ -35,10 +35,6 @@
 #include "internal.h"
 #include "iwl-op-mode.h"
 
-#ifdef CONFIG_IWLWIFI_IDI
-#include "iwl-amfh.h"
-#endif
-
 /******************************************************************************
  *
  * RX path functions
@@ -181,15 +177,15 @@ void iwl_rx_queue_update_write_ptr(struct iwl_trans *trans,
 }
 
 /**
- * iwlagn_dma_addr2rbd_ptr - convert a DMA address to a uCode read buffer ptr
+ * iwl_dma_addr2rbd_ptr - convert a DMA address to a uCode read buffer ptr
  */
-static inline __le32 iwlagn_dma_addr2rbd_ptr(dma_addr_t dma_addr)
+static inline __le32 iwl_dma_addr2rbd_ptr(dma_addr_t dma_addr)
 {
 	return cpu_to_le32((u32)(dma_addr >> 8));
 }
 
 /**
- * iwlagn_rx_queue_restock - refill RX queue from pre-allocated pool
+ * iwl_rx_queue_restock - refill RX queue from pre-allocated pool
  *
  * If there are slots in the RX queue that need to be restocked,
  * and we have free pre-allocated buffers, fill the ranks as much
@@ -199,7 +195,7 @@ static inline __le32 iwlagn_dma_addr2rbd_ptr(dma_addr_t dma_addr)
  * also updates the memory address in the firmware to reference the new
  * target buffer.
  */
-static void iwlagn_rx_queue_restock(struct iwl_trans *trans)
+static void iwl_rx_queue_restock(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rx_queue *rxq = &trans_pcie->rxq;
@@ -207,6 +203,17 @@ static void iwlagn_rx_queue_restock(struct iwl_trans *trans)
 	struct iwl_rx_mem_buffer *rxb;
 	unsigned long flags;
 
+	/*
+	 * If the device isn't enabled - not need to try to add buffers...
+	 * This can happen when we stop the device and still have an interrupt
+	 * pending. We stop the APM before we sync the interrupts / tasklets
+	 * because we have to (see comment there). On the other hand, since
+	 * the APM is stopped, we cannot access the HW (in particular not prph).
+	 * So don't try to restock if the APM has been already stopped.
+	 */
+	if (!test_bit(STATUS_DEVICE_ENABLED, &trans_pcie->status))
+		return;
+
 	spin_lock_irqsave(&rxq->lock, flags);
 	while ((iwl_rx_queue_space(rxq) > 0) && (rxq->free_count)) {
 		/* The overwritten rxb must be a used one */
@@ -219,7 +226,7 @@ static void iwlagn_rx_queue_restock(struct iwl_trans *trans)
 		list_del(element);
 
 		/* Point to Rx buffer via next RBD in circular buffer */
-		rxq->bd[rxq->write] = iwlagn_dma_addr2rbd_ptr(rxb->page_dma);
+		rxq->bd[rxq->write] = iwl_dma_addr2rbd_ptr(rxb->page_dma);
 		rxq->queue[rxq->write] = rxb;
 		rxq->write = (rxq->write + 1) & RX_QUEUE_MASK;
 		rxq->free_count--;
@@ -230,7 +237,6 @@ static void iwlagn_rx_queue_restock(struct iwl_trans *trans)
 	if (rxq->free_count <= RX_LOW_WATERMARK)
 		schedule_work(&trans_pcie->rx_replenish);
 
-
 	/* If we've added more space for the firmware to place data, tell it.
 	 * Increment device's write pointer in multiples of 8. */
 	if (rxq->write_actual != (rxq->write & ~0x7)) {
@@ -241,15 +247,16 @@ static void iwlagn_rx_queue_restock(struct iwl_trans *trans)
 	}
 }
 
-/**
- * iwlagn_rx_replenish - Move all used packet from rx_used to rx_free
- *
- * When moving to rx_free an SKB is allocated for the slot.
+/*
+ * iwl_rx_allocate - allocate a page for each used RBD
  *
- * Also restock the Rx queue via iwl_rx_queue_restock.
- * This is called as a scheduled work item (except for during initialization)
+ * A used RBD is an Rx buffer that has been given to the stack. To use it again
+ * a page must be allocated and the RBD must point to the page. This function
+ * doesn't change the HW pointer but handles the list of pages that is used by
+ * iwl_rx_queue_restock. The latter function will update the HW to use the newly
+ * allocated buffers.
  */
-static void iwlagn_rx_allocate(struct iwl_trans *trans, gfp_t priority)
+static void iwl_rx_allocate(struct iwl_trans *trans, gfp_t priority)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	struct iwl_rx_queue *rxq = &trans_pcie->rxq;
@@ -328,23 +335,31 @@ static void iwlagn_rx_allocate(struct iwl_trans *trans, gfp_t priority)
 	}
 }
 
-void iwlagn_rx_replenish(struct iwl_trans *trans)
+/*
+ * iwl_rx_replenish - Move all used buffers from rx_used to rx_free
+ *
+ * When moving to rx_free an page is allocated for the slot.
+ *
+ * Also restock the Rx queue via iwl_rx_queue_restock.
+ * This is called as a scheduled work item (except for during initialization)
+ */
+void iwl_rx_replenish(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 	unsigned long flags;
 
-	iwlagn_rx_allocate(trans, GFP_KERNEL);
+	iwl_rx_allocate(trans, GFP_KERNEL);
 
 	spin_lock_irqsave(&trans_pcie->irq_lock, flags);
-	iwlagn_rx_queue_restock(trans);
+	iwl_rx_queue_restock(trans);
 	spin_unlock_irqrestore(&trans_pcie->irq_lock, flags);
 }
 
-static void iwlagn_rx_replenish_now(struct iwl_trans *trans)
+static void iwl_rx_replenish_now(struct iwl_trans *trans)
 {
-	iwlagn_rx_allocate(trans, GFP_ATOMIC);
+	iwl_rx_allocate(trans, GFP_ATOMIC);
 
-	iwlagn_rx_queue_restock(trans);
+	iwl_rx_queue_restock(trans);
 }
 
 void iwl_bg_rx_replenish(struct work_struct *data)
@@ -352,7 +367,7 @@ void iwl_bg_rx_replenish(struct work_struct *data)
 	struct iwl_trans_pcie *trans_pcie =
 	    container_of(data, struct iwl_trans_pcie, rx_replenish);
 
-	iwlagn_rx_replenish(trans_pcie->trans);
+	iwl_rx_replenish(trans_pcie->trans);
 }
 
 static void iwl_rx_handle_rxbuf(struct iwl_trans *trans,
@@ -421,13 +436,23 @@ static void iwl_rx_handle_rxbuf(struct iwl_trans *trans,
 		index = SEQ_TO_INDEX(sequence);
 		cmd_index = get_cmd_index(&txq->q, index);
 
-		if (reclaim)
-			cmd = txq->entries[cmd_index].cmd;
-		else
+		if (reclaim) {
+			struct iwl_pcie_tx_queue_entry *ent;
+			ent = &txq->entries[cmd_index];
+			cmd = ent->copy_cmd;
+			WARN_ON_ONCE(!cmd && ent->meta.flags & CMD_WANT_HCMD);
+		} else {
 			cmd = NULL;
+		}
 
 		err = iwl_op_mode_rx(trans->op_mode, &rxcb, cmd);
 
+		if (reclaim) {
+			/* The original command isn't needed any more */
+			kfree(txq->entries[cmd_index].copy_cmd);
+			txq->entries[cmd_index].copy_cmd = NULL;
+		}
+
 		/*
 		 * After here, we should always check rxcb._page_stolen,
 		 * if it is true then one of the handlers took the page.
@@ -520,7 +545,7 @@ static void iwl_rx_handle(struct iwl_trans *trans)
 			count++;
 			if (count >= 8) {
 				rxq->read = i;
-				iwlagn_rx_replenish_now(trans);
+				iwl_rx_replenish_now(trans);
 				count = 0;
 			}
 		}
@@ -529,9 +554,9 @@ static void iwl_rx_handle(struct iwl_trans *trans)
 	/* Backtrack one entry */
 	rxq->read = i;
 	if (fill_rx)
-		iwlagn_rx_replenish_now(trans);
+		iwl_rx_replenish_now(trans);
 	else
-		iwlagn_rx_queue_restock(trans);
+		iwl_rx_queue_restock(trans);
 }
 
 /**
@@ -713,11 +738,9 @@ void iwl_irq_tasklet(struct iwl_trans *trans)
 		/* Disable periodic interrupt; we use it as just a one-shot. */
 		iwl_write8(trans, CSR_INT_PERIODIC_REG,
 			    CSR_INT_PERIODIC_DIS);
-#ifdef CONFIG_IWLWIFI_IDI
-		iwl_amfh_rx_handler();
-#else
+
 		iwl_rx_handle(trans);
-#endif
+
 		/*
 		 * Enable periodic interrupt in 8 msec only if we received
 		 * real RX interrupt (instead of just periodic int), to catch
diff --git a/drivers/net/wireless/iwlwifi/pcie/trans.c b/drivers/net/wireless/iwlwifi/pcie/trans.c
index 063ecaff5b5..fe0fffd0430 100644
--- a/drivers/net/wireless/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/iwlwifi/pcie/trans.c
@@ -216,7 +216,7 @@ static int iwl_rx_init(struct iwl_trans *trans)
 	rxq->free_count = 0;
 	spin_unlock_irqrestore(&rxq->lock, flags);
 
-	iwlagn_rx_replenish(trans);
+	iwl_rx_replenish(trans);
 
 	iwl_trans_rx_hw_init(trans, rxq);
 
@@ -492,10 +492,11 @@ static void iwl_tx_queue_free(struct iwl_trans *trans, int txq_id)
 	iwl_tx_queue_unmap(trans, txq_id);
 
 	/* De-alloc array of command/tx buffers */
-
 	if (txq_id == trans_pcie->cmd_queue)
-		for (i = 0; i < txq->q.n_window; i++)
+		for (i = 0; i < txq->q.n_window; i++) {
 			kfree(txq->entries[i].cmd);
+			kfree(txq->entries[i].copy_cmd);
+		}
 
 	/* De-alloc circular buffer of TFDs */
 	if (txq->q.n_bd) {
@@ -851,10 +852,8 @@ static int iwl_nic_init(struct iwl_trans *trans)
 
 	iwl_op_mode_nic_config(trans->op_mode);
 
-#ifndef CONFIG_IWLWIFI_IDI
 	/* Allocate the RX queue, or reset if it is already allocated */
 	iwl_rx_init(trans);
-#endif
 
 	/* Allocate or reset and init all Tx and Command queues */
 	if (iwl_tx_init(trans))
@@ -893,6 +892,7 @@ static int iwl_set_hw_ready(struct iwl_trans *trans)
 static int iwl_prepare_card_hw(struct iwl_trans *trans)
 {
 	int ret;
+	int t = 0;
 
 	IWL_DEBUG_INFO(trans, "iwl_trans_prepare_card_hw enter\n");
 
@@ -905,30 +905,25 @@ static int iwl_prepare_card_hw(struct iwl_trans *trans)
 	iwl_set_bit(trans, CSR_HW_IF_CONFIG_REG,
 		    CSR_HW_IF_CONFIG_REG_PREPARE);
 
-	ret = iwl_poll_bit(trans, CSR_HW_IF_CONFIG_REG,
-			   ~CSR_HW_IF_CONFIG_REG_BIT_NIC_PREPARE_DONE,
-			   CSR_HW_IF_CONFIG_REG_BIT_NIC_PREPARE_DONE, 150000);
+	do {
+		ret = iwl_set_hw_ready(trans);
+		if (ret >= 0)
+			return 0;
 
-	if (ret < 0)
-		return ret;
+		usleep_range(200, 1000);
+		t += 200;
+	} while (t < 150000);
 
-	/* HW should be ready by now, check again. */
-	ret = iwl_set_hw_ready(trans);
-	if (ret >= 0)
-		return 0;
 	return ret;
 }
 
 /*
  * ucode
  */
-static int iwl_load_section(struct iwl_trans *trans, u8 section_num,
-			    const struct fw_desc *section)
+static int iwl_load_firmware_chunk(struct iwl_trans *trans, u32 dst_addr,
+				   dma_addr_t phy_addr, u32 byte_cnt)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	dma_addr_t phy_addr = section->p_addr;
-	u32 byte_cnt = section->len;
-	u32 dst_addr = section->offset;
 	int ret;
 
 	trans_pcie->ucode_write_complete = false;
@@ -942,8 +937,8 @@ static int iwl_load_section(struct iwl_trans *trans, u8 section_num,
 			   dst_addr);
 
 	iwl_write_direct32(trans,
-		FH_TFDIB_CTRL0_REG(FH_SRVC_CHNL),
-		phy_addr & FH_MEM_TFDIB_DRAM_ADDR_LSB_MSK);
+			   FH_TFDIB_CTRL0_REG(FH_SRVC_CHNL),
+			   phy_addr & FH_MEM_TFDIB_DRAM_ADDR_LSB_MSK);
 
 	iwl_write_direct32(trans,
 			   FH_TFDIB_CTRL1_REG(FH_SRVC_CHNL),
@@ -962,33 +957,64 @@ static int iwl_load_section(struct iwl_trans *trans, u8 section_num,
 			   FH_TCSR_TX_CONFIG_REG_VAL_DMA_CREDIT_DISABLE	|
 			   FH_TCSR_TX_CONFIG_REG_VAL_CIRQ_HOST_ENDTFD);
 
-	IWL_DEBUG_FW(trans, "[%d] uCode section being loaded...\n",
-		     section_num);
 	ret = wait_event_timeout(trans_pcie->ucode_write_waitq,
 				 trans_pcie->ucode_write_complete, 5 * HZ);
 	if (!ret) {
-		IWL_ERR(trans, "Could not load the [%d] uCode section\n",
-			section_num);
+		IWL_ERR(trans, "Failed to load firmware chunk!\n");
 		return -ETIMEDOUT;
 	}
 
 	return 0;
 }
 
-static int iwl_load_given_ucode(struct iwl_trans *trans,
-				const struct fw_img *image)
+static int iwl_load_section(struct iwl_trans *trans, u8 section_num,
+			    const struct fw_desc *section)
 {
+	u8 *v_addr;
+	dma_addr_t p_addr;
+	u32 offset;
 	int ret = 0;
-		int i;
 
-		for (i = 0; i < IWL_UCODE_SECTION_MAX; i++) {
-			if (!image->sec[i].p_addr)
-				break;
+	IWL_DEBUG_FW(trans, "[%d] uCode section being loaded...\n",
+		     section_num);
+
+	v_addr = dma_alloc_coherent(trans->dev, PAGE_SIZE, &p_addr, GFP_KERNEL);
+	if (!v_addr)
+		return -ENOMEM;
+
+	for (offset = 0; offset < section->len; offset += PAGE_SIZE) {
+		u32 copy_size;
+
+		copy_size = min_t(u32, PAGE_SIZE, section->len - offset);
 
-			ret = iwl_load_section(trans, i, &image->sec[i]);
-			if (ret)
-				return ret;
+		memcpy(v_addr, (u8 *)section->data + offset, copy_size);
+		ret = iwl_load_firmware_chunk(trans, section->offset + offset,
+					      p_addr, copy_size);
+		if (ret) {
+			IWL_ERR(trans,
+				"Could not load the [%d] uCode section\n",
+				section_num);
+			break;
 		}
+	}
+
+	dma_free_coherent(trans->dev, PAGE_SIZE, v_addr, p_addr);
+	return ret;
+}
+
+static int iwl_load_given_ucode(struct iwl_trans *trans,
+				const struct fw_img *image)
+{
+	int i, ret = 0;
+
+	for (i = 0; i < IWL_UCODE_SECTION_MAX; i++) {
+		if (!image->sec[i].data)
+			break;
+
+		ret = iwl_load_section(trans, i, &image->sec[i]);
+		if (ret)
+			return ret;
+	}
 
 	/* Remove all resets to allow NIC to operate */
 	iwl_write32(trans, CSR_RESET, 0);
@@ -1181,9 +1207,8 @@ static void iwl_trans_pcie_stop_device(struct iwl_trans *trans)
 	 */
 	if (test_bit(STATUS_DEVICE_ENABLED, &trans_pcie->status)) {
 		iwl_trans_tx_stop(trans);
-#ifndef CONFIG_IWLWIFI_IDI
 		iwl_trans_rx_stop(trans);
-#endif
+
 		/* Power-down device's busmaster DMA clocks */
 		iwl_write_prph(trans, APMG_CLK_DIS_REG,
 			       APMG_CLK_VAL_DMA_CLK_RQT);
@@ -1454,14 +1479,16 @@ static void iwl_trans_pcie_stop_hw(struct iwl_trans *trans,
 	bool hw_rfkill;
 	unsigned long flags;
 
+	spin_lock_irqsave(&trans_pcie->irq_lock, flags);
+	iwl_disable_interrupts(trans);
+	spin_unlock_irqrestore(&trans_pcie->irq_lock, flags);
+
 	iwl_apm_stop(trans);
 
 	spin_lock_irqsave(&trans_pcie->irq_lock, flags);
 	iwl_disable_interrupts(trans);
 	spin_unlock_irqrestore(&trans_pcie->irq_lock, flags);
 
-	iwl_write32(trans, CSR_INT, 0xFFFFFFFF);
-
 	if (!op_mode_leaving) {
 		/*
 		 * Even if we stop the HW, we still want the RF kill
@@ -1549,9 +1576,8 @@ void iwl_trans_pcie_free(struct iwl_trans *trans)
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
 
 	iwl_trans_pcie_tx_free(trans);
-#ifndef CONFIG_IWLWIFI_IDI
 	iwl_trans_pcie_rx_free(trans);
-#endif
+
 	if (trans_pcie->irq_requested == true) {
 		free_irq(trans_pcie->irq, trans);
 		iwl_free_isr_ict(trans);
@@ -1769,7 +1795,7 @@ void iwl_dump_csr(struct iwl_trans *trans)
 #define DEBUGFS_ADD_FILE(name, parent, mode) do {			\
 	if (!debugfs_create_file(#name, mode, parent, trans,		\
 				 &iwl_dbgfs_##name##_ops))		\
-		return -ENOMEM;						\
+		goto err;						\
 } while (0)
 
 /* file operation */
@@ -2033,6 +2059,10 @@ static int iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans,
 	DEBUGFS_ADD_FILE(fh_reg, dir, S_IRUSR);
 	DEBUGFS_ADD_FILE(fw_restart, dir, S_IWUSR);
 	return 0;
+
+err:
+	IWL_ERR(trans, "failed to create the trans debugfs entry\n");
+	return -ENOMEM;
 }
 #else
 static int iwl_trans_pcie_dbgfs_register(struct iwl_trans *trans,
diff --git a/drivers/net/wireless/iwlwifi/pcie/tx.c b/drivers/net/wireless/iwlwifi/pcie/tx.c
index 6baf8deef51..105e3af3c62 100644
--- a/drivers/net/wireless/iwlwifi/pcie/tx.c
+++ b/drivers/net/wireless/iwlwifi/pcie/tx.c
@@ -521,12 +521,7 @@ static int iwl_enqueue_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd)
 	u16 copy_size, cmd_size;
 	bool had_nocopy = false;
 	int i;
-	u8 *cmd_dest;
-#ifdef CONFIG_IWLWIFI_DEVICE_TRACING
-	const void *trace_bufs[IWL_MAX_CMD_TFDS + 1] = {};
-	int trace_lens[IWL_MAX_CMD_TFDS + 1] = {};
-	int trace_idx;
-#endif
+	u32 cmd_pos;
 
 	copy_size = sizeof(out_cmd->hdr);
 	cmd_size = sizeof(out_cmd->hdr);
@@ -584,15 +579,31 @@ static int iwl_enqueue_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd)
 					 INDEX_TO_SEQ(q->write_ptr));
 
 	/* and copy the data that needs to be copied */
-
-	cmd_dest = out_cmd->payload;
+	cmd_pos = offsetof(struct iwl_device_cmd, payload);
 	for (i = 0; i < IWL_MAX_CMD_TFDS; i++) {
 		if (!cmd->len[i])
 			continue;
 		if (cmd->dataflags[i] & IWL_HCMD_DFL_NOCOPY)
 			break;
-		memcpy(cmd_dest, cmd->data[i], cmd->len[i]);
-		cmd_dest += cmd->len[i];
+		memcpy((u8 *)out_cmd + cmd_pos, cmd->data[i], cmd->len[i]);
+		cmd_pos += cmd->len[i];
+	}
+
+	WARN_ON_ONCE(txq->entries[idx].copy_cmd);
+
+	/*
+	 * since out_cmd will be the source address of the FH, it will write
+	 * the retry count there. So when the user needs to receivce the HCMD
+	 * that corresponds to the response in the response handler, it needs
+	 * to set CMD_WANT_HCMD.
+	 */
+	if (cmd->flags & CMD_WANT_HCMD) {
+		txq->entries[idx].copy_cmd =
+			kmemdup(out_cmd, cmd_pos, GFP_ATOMIC);
+		if (unlikely(!txq->entries[idx].copy_cmd)) {
+			idx = -ENOMEM;
+			goto out;
+		}
 	}
 
 	IWL_DEBUG_HC(trans,
@@ -612,11 +623,6 @@ static int iwl_enqueue_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd)
 	dma_unmap_len_set(out_meta, len, copy_size);
 
 	iwlagn_txq_attach_buf_to_tfd(trans, txq, phys_addr, copy_size, 1);
-#ifdef CONFIG_IWLWIFI_DEVICE_TRACING
-	trace_bufs[0] = &out_cmd->hdr;
-	trace_lens[0] = copy_size;
-	trace_idx = 1;
-#endif
 
 	for (i = 0; i < IWL_MAX_CMD_TFDS; i++) {
 		if (!cmd->len[i])
@@ -635,25 +641,14 @@ static int iwl_enqueue_hcmd(struct iwl_trans *trans, struct iwl_host_cmd *cmd)
 
 		iwlagn_txq_attach_buf_to_tfd(trans, txq, phys_addr,
 					     cmd->len[i], 0);
-#ifdef CONFIG_IWLWIFI_DEVICE_TRACING
-		trace_bufs[trace_idx] = cmd->data[i];
-		trace_lens[trace_idx] = cmd->len[i];
-		trace_idx++;
-#endif
 	}
 
 	out_meta->flags = cmd->flags;
 
 	txq->need_update = 1;
 
-	/* check that tracing gets all possible blocks */
-	BUILD_BUG_ON(IWL_MAX_CMD_TFDS + 1 != 3);
-#ifdef CONFIG_IWLWIFI_DEVICE_TRACING
-	trace_iwlwifi_dev_hcmd(trans->dev, cmd->flags,
-			       trace_bufs[0], trace_lens[0],
-			       trace_bufs[1], trace_lens[1],
-			       trace_bufs[2], trace_lens[2]);
-#endif
+	trace_iwlwifi_dev_hcmd(trans->dev, cmd, cmd_size,
+			       &out_cmd->hdr, copy_size);
 
 	/* start timer if queue currently empty */
 	if (q->read_ptr == q->write_ptr && trans_pcie->wd_timeout)