Merge branch '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/nex

t-queue

Tony Nguyen says:

====================
100GbE Intel Wired LAN Driver Updates 2021-12-15

This series contains updates to ice driver only.

Jake makes changes to flash update. This includes the following:

 * a new shadow-ram region similar to NVM region but for the device shadow
   RAM contents. This is distinct from NVM region because shadow RAM is
   built up during device init and may be different from the raw NVM flash
   data.
 * refactoring of the ice_flash_pldm_image to become the main flash update
   entry point. This is simpler than having both an
   ice_devlink_flash_update and an ice_flash_pldm_image. It will make
   additions like dry-run easier in the future.
 * reducing time to read Option ROM version information.
 * adding support for firmware activation via devlink reload, when
   possible.

The major new work is the reload support, which allows activating firmware
immediately without a reboot when possible. Reload support only supports
firmware activation.

Jesse improves transmit code: utilizing newer netif_tx* API, adding some
prefetch calls, correcting expected conditions when calling ice_vsi_down(),
and utilizing __netdev_tx_sent_queue() call.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Documentation/networking/devlink/ice.rst

@@ -26,8 +26,10 @@ The ``ice`` driver reports the following versions
     * - ``fw.mgmt``
       - running
       - 2.1.7
-      - 3-digit version number of the management firmware that controls the
-        PHY, link, etc.
+      - 3-digit version number of the management firmware running on the
+        Embedded Management Processor of the device. It controls the PHY,
+        link, access to device resources, etc. Intel documentation refers to
+        this as the EMP firmware.
     * - ``fw.mgmt.api``
       - running
       - 1.5.1
@@ -119,6 +121,24 @@ preserving settings, and thus ``DEVLINK_FLASH_OVERWRITE_IDENTIFIERS`` on its
 own will be rejected. If no overwrite mask is provided, the firmware will be
 instructed to preserve all settings and identifying fields when updating.
 
+Reload
+======
+
+The ``ice`` driver supports activating new firmware after a flash update
+using ``DEVLINK_CMD_RELOAD`` with the ``DEVLINK_RELOAD_ACTION_FW_ACTIVATE``
+action.
+
+.. code:: shell
+
+    $ devlink dev reload pci/0000:01:00.0 reload action fw_activate
+
+The new firmware is activated by issuing a device specific Embedded
+Management Processor reset which requests the device to reset and reload the
+EMP firmware image.
+
+The driver does not currently support reloading the driver via
+``DEVLINK_RELOAD_ACTION_DRIVER_REINIT``.
+
 Regions
 =======
 

drivers/net/ethernet/intel/ice/ice.h

@@ -503,6 +503,7 @@ struct ice_pf {
 	struct pci_dev *pdev;
 
 	struct devlink_region *nvm_region;
+	struct devlink_region *sram_region;
 	struct devlink_region *devcaps_region;
 
 	/* devlink port data */
@@ -552,6 +553,7 @@ struct ice_pf {
 	spinlock_t aq_wait_lock;
 	struct hlist_head aq_wait_list;
 	wait_queue_head_t aq_wait_queue;
+	bool fw_emp_reset_disabled;
 
 	wait_queue_head_t reset_wait_queue;
 

drivers/net/ethernet/intel/ice/ice_adminq_cmd.h

@@ -117,6 +117,8 @@ struct ice_aqc_list_caps_elem {
 #define ICE_AQC_CAPS_NET_VER				0x004C
 #define ICE_AQC_CAPS_PENDING_NET_VER			0x004D
 #define ICE_AQC_CAPS_RDMA				0x0051
+#define ICE_AQC_CAPS_PCIE_RESET_AVOIDANCE		0x0076
+#define ICE_AQC_CAPS_POST_UPDATE_RESET_RESTRICT		0x0077
 #define ICE_AQC_CAPS_NVM_MGMT				0x0080
 
 	u8 major_ver;
@@ -1408,6 +1410,11 @@ struct ice_aqc_nvm {
 #define ICE_AQC_NVM_REVERT_LAST_ACTIV	BIT(6) /* Write Activate only */
 #define ICE_AQC_NVM_ACTIV_SEL_MASK	ICE_M(0x7, 3)
 #define ICE_AQC_NVM_FLASH_ONLY		BIT(7)
+#define ICE_AQC_NVM_RESET_LVL_M		ICE_M(0x3, 0) /* Write reply only */
+#define ICE_AQC_NVM_POR_FLAG		0
+#define ICE_AQC_NVM_PERST_FLAG		1
+#define ICE_AQC_NVM_EMPR_FLAG		2
+#define ICE_AQC_NVM_EMPR_ENA		BIT(0) /* Write Activate reply only */
 	__le16 module_typeid;
 	__le16 length;
 #define ICE_AQC_NVM_ERASE_LEN	0xFFFF

drivers/net/ethernet/intel/ice/ice_common.c

@@ -2068,6 +2068,18 @@ ice_parse_common_caps(struct ice_hw *hw, struct ice_hw_common_caps *caps,
 		ice_debug(hw, ICE_DBG_INIT, "%s: max_mtu = %d\n",
 			  prefix, caps->max_mtu);
 		break;
+	case ICE_AQC_CAPS_PCIE_RESET_AVOIDANCE:
+		caps->pcie_reset_avoidance = (number > 0);
+		ice_debug(hw, ICE_DBG_INIT,
+			  "%s: pcie_reset_avoidance = %d\n", prefix,
+			  caps->pcie_reset_avoidance);
+		break;
+	case ICE_AQC_CAPS_POST_UPDATE_RESET_RESTRICT:
+		caps->reset_restrict_support = (number == 1);
+		ice_debug(hw, ICE_DBG_INIT,
+			  "%s: reset_restrict_support = %d\n", prefix,
+			  caps->reset_restrict_support);
+		break;
 	default:
 		/* Not one of the recognized common capabilities */
 		found = false;

drivers/net/ethernet/intel/ice/ice_ethtool.c

@@ -1280,9 +1280,11 @@ static int ice_set_priv_flags(struct net_device *netdev, u32 flags)
 	}
 	if (test_bit(ICE_FLAG_LEGACY_RX, change_flags)) {
 		/* down and up VSI so that changes of Rx cfg are reflected. */
+		if (!test_and_set_bit(ICE_VSI_DOWN, vsi->state)) {
 			ice_down(vsi);
 			ice_up(vsi);
 		}
+	}
 	/* don't allow modification of this flag when a single VF is in
 	 * promiscuous mode because it's not supported
 	 */

drivers/net/ethernet/intel/ice/ice_fw_update.h

@@ -4,9 +4,10 @@
 #ifndef _ICE_FW_UPDATE_H_
 #define _ICE_FW_UPDATE_H_
 
-int ice_flash_pldm_image(struct ice_pf *pf, const struct firmware *fw,
-			 u8 preservation, struct netlink_ext_ack *extack);
-int ice_check_for_pending_update(struct ice_pf *pf, const char *component,
+int ice_devlink_flash_update(struct devlink *devlink,
+			     struct devlink_flash_update_params *params,
+			     struct netlink_ext_ack *extack);
+int ice_get_pending_updates(struct ice_pf *pf, u8 *pending,
 			    struct netlink_ext_ack *extack);
 
 #endif

drivers/net/ethernet/intel/ice/ice_main.c

@@ -6229,14 +6229,15 @@ static void ice_napi_disable_all(struct ice_vsi *vsi)
 /**
  * ice_down - Shutdown the connection
  * @vsi: The VSI being stopped
+ *
+ * Caller of this function is expected to set the vsi->state ICE_DOWN bit
  */
 int ice_down(struct ice_vsi *vsi)
 {
 	int i, tx_err, rx_err, link_err = 0;
 
-	/* Caller of this function is expected to set the
-	 * vsi->state ICE_DOWN bit
-	 */
+	WARN_ON(!test_bit(ICE_VSI_DOWN, vsi->state));
+
 	if (vsi->netdev && vsi->type == ICE_VSI_PF) {
 		netif_carrier_off(vsi->netdev);
 		netif_tx_disable(vsi->netdev);
@@ -6594,6 +6595,14 @@ static void ice_rebuild(struct ice_pf *pf, enum ice_reset_req reset_type)
 
 	dev_dbg(dev, "rebuilding PF after reset_type=%d\n", reset_type);
 
+	if (reset_type == ICE_RESET_EMPR) {
+		/* If an EMP reset has occurred, any previously pending flash
+		 * update will have completed. We no longer know whether or
+		 * not the NVM update EMP reset is restricted.
+		 */
+		pf->fw_emp_reset_disabled = false;
+	}
+
 	err = ice_init_all_ctrlq(hw);
 	if (err) {
 		dev_err(dev, "control queues init failed %d\n", err);

drivers/net/ethernet/intel/ice/ice_nvm.c

@@ -615,7 +615,7 @@ static int
 ice_get_orom_civd_data(struct ice_hw *hw, enum ice_bank_select bank,
 		       struct ice_orom_civd_info *civd)
 {
-	struct ice_orom_civd_info tmp;
+	u8 *orom_data;
 	int status;
 	u32 offset;
 
@@ -623,36 +623,60 @@ ice_get_orom_civd_data(struct ice_hw *hw, enum ice_bank_select bank,
 	 * The first 4 bytes must contain the ASCII characters "$CIV".
 	 * A simple modulo 256 sum of all of the bytes of the structure must
 	 * equal 0.
+	 *
+	 * The exact location is unknown and varies between images but is
+	 * usually somewhere in the middle of the bank. We need to scan the
+	 * Option ROM bank to locate it.
+	 *
+	 * It's significantly faster to read the entire Option ROM up front
+	 * using the maximum page size, than to read each possible location
+	 * with a separate firmware command.
 	 */
-	for (offset = 0; (offset + 512) <= hw->flash.banks.orom_size; offset += 512) {
-		u8 sum = 0, i;
+	orom_data = vzalloc(hw->flash.banks.orom_size);
+	if (!orom_data)
+		return -ENOMEM;
 
-		status = ice_read_flash_module(hw, bank, ICE_SR_1ST_OROM_BANK_PTR,
-					       offset, (u8 *)&tmp, sizeof(tmp));
+	status = ice_read_flash_module(hw, bank, ICE_SR_1ST_OROM_BANK_PTR, 0,
+				       orom_data, hw->flash.banks.orom_size);
 	if (status) {
-			ice_debug(hw, ICE_DBG_NVM, "Unable to read Option ROM CIVD data\n");
+		ice_debug(hw, ICE_DBG_NVM, "Unable to read Option ROM data\n");
 		return status;
 	}
 
+	/* Scan the memory buffer to locate the CIVD data section */
+	for (offset = 0; (offset + 512) <= hw->flash.banks.orom_size; offset += 512) {
+		struct ice_orom_civd_info *tmp;
+		u8 sum = 0, i;
+
+		tmp = (struct ice_orom_civd_info *)&orom_data[offset];
+
 		/* Skip forward until we find a matching signature */
-		if (memcmp("$CIV", tmp.signature, sizeof(tmp.signature)) != 0)
+		if (memcmp("$CIV", tmp->signature, sizeof(tmp->signature)) != 0)
 			continue;
 
+		ice_debug(hw, ICE_DBG_NVM, "Found CIVD section at offset %u\n",
+			  offset);
+
 		/* Verify that the simple checksum is zero */
-		for (i = 0; i < sizeof(tmp); i++)
+		for (i = 0; i < sizeof(*tmp); i++)
 			/* cppcheck-suppress objectIndex */
-			sum += ((u8 *)&tmp)[i];
+			sum += ((u8 *)tmp)[i];
 
 		if (sum) {
 			ice_debug(hw, ICE_DBG_NVM, "Found CIVD data with invalid checksum of %u\n",
 				  sum);
-			return -EIO;
+			goto err_invalid_checksum;
 		}
 
-		*civd = tmp;
+		*civd = *tmp;
+		vfree(orom_data);
 		return 0;
 	}
 
+	ice_debug(hw, ICE_DBG_NVM, "Unable to locate CIVD data within the Option ROM\n");
+
+err_invalid_checksum:
+	vfree(orom_data);
 	return -EIO;
 }
 
@@ -1081,22 +1105,35 @@ int ice_nvm_validate_checksum(struct ice_hw *hw)
 /**
  * ice_nvm_write_activate
  * @hw: pointer to the HW struct
- * @cmd_flags: NVM activate admin command bits (banks to be validated)
+ * @cmd_flags: flags for write activate command
+ * @response_flags: response indicators from firmware
  *
  * Update the control word with the required banks' validity bits
  * and dumps the Shadow RAM to flash (0x0707)
+ *
+ * cmd_flags controls which banks to activate, and the preservation level to
+ * use when activating the NVM bank.
+ *
+ * On successful return of the firmware command, the response_flags variable
+ * is updated with the flags reported by firmware indicating certain status,
+ * such as whether EMP reset is enabled.
  */
-int ice_nvm_write_activate(struct ice_hw *hw, u8 cmd_flags)
+int ice_nvm_write_activate(struct ice_hw *hw, u8 cmd_flags, u8 *response_flags)
 {
 	struct ice_aqc_nvm *cmd;
 	struct ice_aq_desc desc;
+	int err;
 
 	cmd = &desc.params.nvm;
 	ice_fill_dflt_direct_cmd_desc(&desc, ice_aqc_opc_nvm_write_activate);
 
 	cmd->cmd_flags = cmd_flags;
 
-	return ice_aq_send_cmd(hw, &desc, NULL, 0, NULL);
+	err = ice_aq_send_cmd(hw, &desc, NULL, 0, NULL);
+	if (!err && response_flags)
+		*response_flags = cmd->cmd_flags;
+
+	return err;
 }
 
 /**

drivers/net/ethernet/intel/ice/ice_nvm.h

@@ -34,7 +34,7 @@ ice_aq_update_nvm(struct ice_hw *hw, u16 module_typeid, u32 offset,
 int
 ice_aq_erase_nvm(struct ice_hw *hw, u16 module_typeid, struct ice_sq_cd *cd);
 int ice_nvm_validate_checksum(struct ice_hw *hw);
-int ice_nvm_write_activate(struct ice_hw *hw, u8 cmd_flags);
+int ice_nvm_write_activate(struct ice_hw *hw, u8 cmd_flags, u8 *response_flags);
 int ice_aq_nvm_update_empr(struct ice_hw *hw);
 int
 ice_nvm_set_pkg_data(struct ice_hw *hw, bool del_pkg_data_flag, u8 *data,

drivers/net/ethernet/intel/ice/ice_txrx.c

@@ -3,8 +3,9 @@
 
 /* The driver transmit and receive code */
 
-#include <linux/prefetch.h>
 #include <linux/mm.h>
+#include <linux/netdevice.h>
+#include <linux/prefetch.h>
 #include <linux/bpf_trace.h>
 #include <net/dsfield.h>
 #include <net/xdp.h>
@@ -219,6 +220,10 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 	struct ice_tx_desc *tx_desc;
 	struct ice_tx_buf *tx_buf;
 
+	/* get the bql data ready */
+	if (!ice_ring_is_xdp(tx_ring))
+		netdev_txq_bql_complete_prefetchw(txring_txq(tx_ring));
+
 	tx_buf = &tx_ring->tx_buf[i];
 	tx_desc = ICE_TX_DESC(tx_ring, i);
 	i -= tx_ring->count;
@@ -232,6 +237,9 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 		if (!eop_desc)
 			break;
 
+		/* follow the guidelines of other drivers */
+		prefetchw(&tx_buf->skb->users);
+
 		smp_rmb();	/* prevent any other reads prior to eop_desc */
 
 		ice_trace(clean_tx_irq, tx_ring, tx_desc, tx_buf);
@@ -304,8 +312,10 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 
 	ice_update_tx_ring_stats(tx_ring, total_pkts, total_bytes);
 
-	netdev_tx_completed_queue(txring_txq(tx_ring), total_pkts,
-				  total_bytes);
+	if (ice_ring_is_xdp(tx_ring))
+		return !!budget;
+
+	netdev_tx_completed_queue(txring_txq(tx_ring), total_pkts, total_bytes);
 
 #define TX_WAKE_THRESHOLD ((s16)(DESC_NEEDED * 2))
 	if (unlikely(total_pkts && netif_carrier_ok(tx_ring->netdev) &&
@@ -314,11 +324,9 @@ static bool ice_clean_tx_irq(struct ice_tx_ring *tx_ring, int napi_budget)
 		 * sees the new next_to_clean.
 		 */
 		smp_mb();
-		if (__netif_subqueue_stopped(tx_ring->netdev,
-					     tx_ring->q_index) &&
+		if (netif_tx_queue_stopped(txring_txq(tx_ring)) &&
 		    !test_bit(ICE_VSI_DOWN, vsi->state)) {
-			netif_wake_subqueue(tx_ring->netdev,
-					    tx_ring->q_index);
+			netif_tx_wake_queue(txring_txq(tx_ring));
 			++tx_ring->tx_stats.restart_q;
 		}
 	}
@@ -1517,7 +1525,7 @@ int ice_napi_poll(struct napi_struct *napi, int budget)
  */
 static int __ice_maybe_stop_tx(struct ice_tx_ring *tx_ring, unsigned int size)
 {
-	netif_stop_subqueue(tx_ring->netdev, tx_ring->q_index);
+	netif_tx_stop_queue(txring_txq(tx_ring));
 	/* Memory barrier before checking head and tail */
 	smp_mb();
 
@@ -1525,8 +1533,8 @@ static int __ice_maybe_stop_tx(struct ice_tx_ring *tx_ring, unsigned int size)
 	if (likely(ICE_DESC_UNUSED(tx_ring) < size))
 		return -EBUSY;
 
-	/* A reprieve! - use start_subqueue because it doesn't call schedule */
-	netif_start_subqueue(tx_ring->netdev, tx_ring->q_index);
+	/* A reprieve! - use start_queue because it doesn't call schedule */
+	netif_tx_start_queue(txring_txq(tx_ring));
 	++tx_ring->tx_stats.restart_q;
 	return 0;
 }
@@ -1568,6 +1576,7 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first,
 	struct sk_buff *skb;
 	skb_frag_t *frag;
 	dma_addr_t dma;
+	bool kick;
 
 	td_tag = off->td_l2tag1;
 	td_cmd = off->td_cmd;
@@ -1649,9 +1658,6 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first,
 		tx_buf = &tx_ring->tx_buf[i];
 	}
 
-	/* record bytecount for BQL */
-	netdev_tx_sent_queue(txring_txq(tx_ring), first->bytecount);
-
 	/* record SW timestamp if HW timestamp is not available */
 	skb_tx_timestamp(first->skb);
 
@@ -1680,7 +1686,10 @@ ice_tx_map(struct ice_tx_ring *tx_ring, struct ice_tx_buf *first,
 	ice_maybe_stop_tx(tx_ring, DESC_NEEDED);
 
 	/* notify HW of packet */
-	if (netif_xmit_stopped(txring_txq(tx_ring)) || !netdev_xmit_more())
+	kick = __netdev_tx_sent_queue(txring_txq(tx_ring), first->bytecount,
+				      netdev_xmit_more());
+	if (kick)
+		/* notify HW of packet */
 		writel(i, tx_ring->tail);
 
 	return;
@@ -2265,6 +2274,9 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_tx_ring *tx_ring)
 		return NETDEV_TX_BUSY;
 	}
 
+	/* prefetch for bql data which is infrequently used */
+	netdev_txq_bql_enqueue_prefetchw(txring_txq(tx_ring));
+
 	offload.tx_ring = tx_ring;
 
 	/* record the location of the first descriptor for this packet */

drivers/net/ethernet/intel/ice/ice_type.h

@@ -278,6 +278,10 @@ struct ice_hw_common_caps {
 #define ICE_NVM_PENDING_NETLIST			BIT(2)
 	bool nvm_unified_update;
 #define ICE_NVM_MGMT_UNIFIED_UPD_SUPPORT	BIT(3)
+	/* PCIe reset avoidance */
+	bool pcie_reset_avoidance;
+	/* Post update reset restriction */
+	bool reset_restrict_support;
 };
 
 /* IEEE 1588 TIME_SYNC specific info */