Merge tag 'ntb-4.15' of git://github.com/jonmason/ntb

Pull ntb updates from Jon Mason:
 "Support for the switchtec ntb and related changes. Also, a couple of
  bug fixes"

[ The timing isn't great. I had asked people to send me pull requests
  before my family vacation, and this code has not even been in
  linux-next as far as I can tell. But Logan Gunthorpe pleaded for its
  inclusion because the Switchtec driver has apparently been around for
  a while, just never in linux-next - Linus ]

* tag 'ntb-4.15' of git://github.com/jonmason/ntb:
  ntb: intel: remove b2b memory window workaround for Skylake NTB
  NTB: make idt_89hpes_cfg const
  NTB: switchtec_ntb: Update switchtec documentation with notes for NTB
  NTB: switchtec_ntb: Add memory window support
  NTB: switchtec_ntb: Implement scratchpad registers
  NTB: switchtec_ntb: Implement doorbell registers
  NTB: switchtec_ntb: Add link management
  NTB: switchtec_ntb: Add skeleton NTB driver
  NTB: switchtec_ntb: Initialize hardware for doorbells and messages
  NTB: switchtec_ntb: Initialize hardware for memory windows
  NTB: switchtec_ntb: Introduce initial NTB driver
  NTB: Add check and comment for link up to mw_count() and mw_get_align()
  NTB: Ensure ntb_mw_get_align() is only called when the link is up
  NTB: switchtec: Add link event notifier callback
  NTB: switchtec: Add NTB hardware register definitions
  NTB: switchtec: Export class symbol for use in upper layer driver
  NTB: switchtec: Move structure definitions into a common header
  ntb: update maintainer list for Intel NTB driver

Documentation/switchtec.txt

@@ -78,3 +78,15 @@ The following IOCTLs are also supported by the device:
   between PCI Function Framework number (used by the event system)
   and Switchtec Logic Port ID and Partition number (which is more
   user friendly).
+
+
+Non-Transparent Bridge (NTB) Driver
+===================================
+
+An NTB driver is provided for the switchtec hardware in switchtec_ntb.
+Currently, it only supports switches configured with exactly 2
+partitions. It also requires the following configuration settings:
+
+* Both partitions must be able to access each other's GAS spaces.
+  Thus, the bits in the GAS Access Vector under Management Settings
+  must be set to support this.

MAINTAINERS

@@ -9726,12 +9726,11 @@ S:	Supported
 F:	drivers/ntb/hw/idt/
 
 NTB INTEL DRIVER
-M:	Jon Mason <jdmason@kudzu.us>
 M:	Dave Jiang <dave.jiang@intel.com>
 L:	linux-ntb@googlegroups.com
 S:	Supported
-W:	https://github.com/jonmason/ntb/wiki
-T:	git git://github.com/jonmason/ntb.git
+W:	https://github.com/davejiang/linux/wiki
+T:	git https://github.com/davejiang/linux.git
 F:	drivers/ntb/hw/intel/
 
 NTFS FILESYSTEM
@@ -10443,6 +10442,8 @@ F:	Documentation/switchtec.txt
 F:	Documentation/ABI/testing/sysfs-class-switchtec
 F:	drivers/pci/switch/switchtec*
 F:	include/uapi/linux/switchtec_ioctl.h
+F:	include/linux/switchtec.h
+F:	drivers/ntb/hw/mscc/
 
 PCI DRIVER FOR MVEBU (Marvell Armada 370 and Armada XP SOC support)
 M:	Thomas Petazzoni <thomas.petazzoni@free-electrons.com>

drivers/ntb/hw/Kconfig

@@ -1,3 +1,4 @@
 source "drivers/ntb/hw/amd/Kconfig"
 source "drivers/ntb/hw/idt/Kconfig"
 source "drivers/ntb/hw/intel/Kconfig"
+source "drivers/ntb/hw/mscc/Kconfig"

drivers/ntb/hw/Makefile

@@ -1,3 +1,4 @@
 obj-$(CONFIG_NTB_AMD)	+= amd/
 obj-$(CONFIG_NTB_IDT)	+= idt/
 obj-$(CONFIG_NTB_INTEL)	+= intel/
+obj-$(CONFIG_NTB_SWITCHTEC) += mscc/

drivers/ntb/hw/idt/ntb_hw_idt.c

@@ -2628,35 +2628,35 @@ static void idt_pci_remove(struct pci_dev *pdev)
 /*
  * IDT PCIe-switch models ports configuration structures
  */
-static struct idt_89hpes_cfg idt_89hpes24nt6ag2_config = {
+static const struct idt_89hpes_cfg idt_89hpes24nt6ag2_config = {
 	.name = "89HPES24NT6AG2",
 	.port_cnt = 6, .ports = {0, 2, 4, 6, 8, 12}
 };
-static struct idt_89hpes_cfg idt_89hpes32nt8ag2_config = {
+static const struct idt_89hpes_cfg idt_89hpes32nt8ag2_config = {
 	.name = "89HPES32NT8AG2",
 	.port_cnt = 8, .ports = {0, 2, 4, 6, 8, 12, 16, 20}
 };
-static struct idt_89hpes_cfg idt_89hpes32nt8bg2_config = {
+static const struct idt_89hpes_cfg idt_89hpes32nt8bg2_config = {
 	.name = "89HPES32NT8BG2",
 	.port_cnt = 8, .ports = {0, 2, 4, 6, 8, 12, 16, 20}
 };
-static struct idt_89hpes_cfg idt_89hpes12nt12g2_config = {
+static const struct idt_89hpes_cfg idt_89hpes12nt12g2_config = {
 	.name = "89HPES12NT12G2",
 	.port_cnt = 3, .ports = {0, 8, 16}
 };
-static struct idt_89hpes_cfg idt_89hpes16nt16g2_config = {
+static const struct idt_89hpes_cfg idt_89hpes16nt16g2_config = {
 	.name = "89HPES16NT16G2",
 	.port_cnt = 4, .ports = {0, 8, 12, 16}
 };
-static struct idt_89hpes_cfg idt_89hpes24nt24g2_config = {
+static const struct idt_89hpes_cfg idt_89hpes24nt24g2_config = {
 	.name = "89HPES24NT24G2",
 	.port_cnt = 8, .ports = {0, 2, 4, 6, 8, 12, 16, 20}
 };
-static struct idt_89hpes_cfg idt_89hpes32nt24ag2_config = {
+static const struct idt_89hpes_cfg idt_89hpes32nt24ag2_config = {
 	.name = "89HPES32NT24AG2",
 	.port_cnt = 8, .ports = {0, 2, 4, 6, 8, 12, 16, 20}
 };
-static struct idt_89hpes_cfg idt_89hpes32nt24bg2_config = {
+static const struct idt_89hpes_cfg idt_89hpes32nt24bg2_config = {
 	.name = "89HPES32NT24BG2",
 	.port_cnt = 8, .ports = {0, 2, 4, 6, 8, 12, 16, 20}
 };

drivers/ntb/hw/intel/ntb_hw_intel.c

@@ -1742,89 +1742,18 @@ static int skx_setup_b2b_mw(struct intel_ntb_dev *ndev,
 {
 	struct pci_dev *pdev;
 	void __iomem *mmio;
-	resource_size_t bar_size;
 	phys_addr_t bar_addr;
-	int b2b_bar;
-	u8 bar_sz;
 
 	pdev = ndev->ntb.pdev;
 	mmio = ndev->self_mmio;
 
-	if (ndev->b2b_idx == UINT_MAX) {
-		dev_dbg(&pdev->dev, "not using b2b mw\n");
-		b2b_bar = 0;
-		ndev->b2b_off = 0;
-	} else {
-		b2b_bar = ndev_mw_to_bar(ndev, ndev->b2b_idx);
-		if (b2b_bar < 0)
-			return -EIO;
-
-		dev_dbg(&pdev->dev, "using b2b mw bar %d\n", b2b_bar);
-
-		bar_size = pci_resource_len(ndev->ntb.pdev, b2b_bar);
-
-		dev_dbg(&pdev->dev, "b2b bar size %#llx\n", bar_size);
-
-		if (b2b_mw_share && ((bar_size >> 1) >= XEON_B2B_MIN_SIZE)) {
-			dev_dbg(&pdev->dev, "b2b using first half of bar\n");
-			ndev->b2b_off = bar_size >> 1;
-		} else if (bar_size >= XEON_B2B_MIN_SIZE) {
-			dev_dbg(&pdev->dev, "b2b using whole bar\n");
-			ndev->b2b_off = 0;
-			--ndev->mw_count;
-		} else {
-			dev_dbg(&pdev->dev, "b2b bar size is too small\n");
-			return -EIO;
-		}
-	}
-
-	/*
-	 * Reset the secondary bar sizes to match the primary bar sizes,
-	 * except disable or halve the size of the b2b secondary bar.
-	 */
-	pci_read_config_byte(pdev, SKX_IMBAR1SZ_OFFSET, &bar_sz);
-	dev_dbg(&pdev->dev, "IMBAR1SZ %#x\n", bar_sz);
-	if (b2b_bar == 1) {
-		if (ndev->b2b_off)
-			bar_sz -= 1;
-		else
-			bar_sz = 0;
-	}
-
-	pci_write_config_byte(pdev, SKX_EMBAR1SZ_OFFSET, bar_sz);
-	pci_read_config_byte(pdev, SKX_EMBAR1SZ_OFFSET, &bar_sz);
-	dev_dbg(&pdev->dev, "EMBAR1SZ %#x\n", bar_sz);
-
-	pci_read_config_byte(pdev, SKX_IMBAR2SZ_OFFSET, &bar_sz);
-	dev_dbg(&pdev->dev, "IMBAR2SZ %#x\n", bar_sz);
-	if (b2b_bar == 2) {
-		if (ndev->b2b_off)
-			bar_sz -= 1;
-		else
-			bar_sz = 0;
-	}
-
-	pci_write_config_byte(pdev, SKX_EMBAR2SZ_OFFSET, bar_sz);
-	pci_read_config_byte(pdev, SKX_EMBAR2SZ_OFFSET, &bar_sz);
-	dev_dbg(&pdev->dev, "EMBAR2SZ %#x\n", bar_sz);
-
-	/* SBAR01 hit by first part of the b2b bar */
-	if (b2b_bar == 0)
-		bar_addr = addr->bar0_addr;
-	else if (b2b_bar == 1)
-		bar_addr = addr->bar2_addr64;
-	else if (b2b_bar == 2)
-		bar_addr = addr->bar4_addr64;
-	else
-		return -EIO;
-
 	/* setup incoming bar limits == base addrs (zero length windows) */
-	bar_addr = addr->bar2_addr64 + (b2b_bar == 1 ? ndev->b2b_off : 0);
+	bar_addr = addr->bar2_addr64;
 	iowrite64(bar_addr, mmio + SKX_IMBAR1XLMT_OFFSET);
 	bar_addr = ioread64(mmio + SKX_IMBAR1XLMT_OFFSET);
 	dev_dbg(&pdev->dev, "IMBAR1XLMT %#018llx\n", bar_addr);
 
-	bar_addr = addr->bar4_addr64 + (b2b_bar == 2 ? ndev->b2b_off : 0);
+	bar_addr = addr->bar4_addr64;
 	iowrite64(bar_addr, mmio + SKX_IMBAR2XLMT_OFFSET);
 	bar_addr = ioread64(mmio + SKX_IMBAR2XLMT_OFFSET);
 	dev_dbg(&pdev->dev, "IMBAR2XLMT %#018llx\n", bar_addr);

drivers/ntb/hw/mscc/Kconfig

@@ -0,0 +1,9 @@
+config NTB_SWITCHTEC
+	tristate "MicroSemi Switchtec Non-Transparent Bridge Support"
+	select PCI_SW_SWITCHTEC
+	help
+	 Enables NTB support for Switchtec PCI switches. This also
+	 selects the Switchtec management driver as they share the same
+	 hardware interface.
+
+	 If unsure, say N.

drivers/ntb/hw/mscc/Makefile

@@ -0,0 +1 @@
+obj-$(CONFIG_NTB_SWITCHTEC) += ntb_hw_switchtec.o

drivers/ntb/hw/mscc/ntb_hw_switchtec.c

@@ -0,0 +1,1216 @@
+/*
+ * Microsemi Switchtec(tm) PCIe Management Driver
+ * Copyright (c) 2017, Microsemi Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#include <linux/switchtec.h>
+#include <linux/module.h>
+#include <linux/delay.h>
+#include <linux/kthread.h>
+#include <linux/interrupt.h>
+#include <linux/ntb.h>
+
+MODULE_DESCRIPTION("Microsemi Switchtec(tm) NTB Driver");
+MODULE_VERSION("0.1");
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Microsemi Corporation");
+
+static ulong max_mw_size = SZ_2M;
+module_param(max_mw_size, ulong, 0644);
+MODULE_PARM_DESC(max_mw_size,
+	"Max memory window size reported to the upper layer");
+
+static bool use_lut_mws;
+module_param(use_lut_mws, bool, 0644);
+MODULE_PARM_DESC(use_lut_mws,
+		 "Enable the use of the LUT based memory windows");
+
+#ifndef ioread64
+#ifdef readq
+#define ioread64 readq
+#else
+#define ioread64 _ioread64
+static inline u64 _ioread64(void __iomem *mmio)
+{
+	u64 low, high;
+
+	low = ioread32(mmio);
+	high = ioread32(mmio + sizeof(u32));
+	return low | (high << 32);
+}
+#endif
+#endif
+
+#ifndef iowrite64
+#ifdef writeq
+#define iowrite64 writeq
+#else
+#define iowrite64 _iowrite64
+static inline void _iowrite64(u64 val, void __iomem *mmio)
+{
+	iowrite32(val, mmio);
+	iowrite32(val >> 32, mmio + sizeof(u32));
+}
+#endif
+#endif
+
+#define SWITCHTEC_NTB_MAGIC 0x45CC0001
+#define MAX_MWS     128
+
+struct shared_mw {
+	u32 magic;
+	u32 link_sta;
+	u32 partition_id;
+	u64 mw_sizes[MAX_MWS];
+	u32 spad[128];
+};
+
+#define MAX_DIRECT_MW ARRAY_SIZE(((struct ntb_ctrl_regs *)(0))->bar_entry)
+#define LUT_SIZE SZ_64K
+
+struct switchtec_ntb {
+	struct ntb_dev ntb;
+	struct switchtec_dev *stdev;
+
+	int self_partition;
+	int peer_partition;
+
+	int doorbell_irq;
+	int message_irq;
+
+	struct ntb_info_regs __iomem *mmio_ntb;
+	struct ntb_ctrl_regs __iomem *mmio_ctrl;
+	struct ntb_dbmsg_regs __iomem *mmio_dbmsg;
+	struct ntb_ctrl_regs __iomem *mmio_self_ctrl;
+	struct ntb_ctrl_regs __iomem *mmio_peer_ctrl;
+	struct ntb_dbmsg_regs __iomem *mmio_self_dbmsg;
+
+	struct shared_mw *self_shared;
+	struct shared_mw __iomem *peer_shared;
+	dma_addr_t self_shared_dma;
+
+	u64 db_mask;
+	u64 db_valid_mask;
+	int db_shift;
+	int db_peer_shift;
+
+	/* synchronize rmw access of db_mask and hw reg */
+	spinlock_t db_mask_lock;
+
+	int nr_direct_mw;
+	int nr_lut_mw;
+	int direct_mw_to_bar[MAX_DIRECT_MW];
+
+	int peer_nr_direct_mw;
+	int peer_nr_lut_mw;
+	int peer_direct_mw_to_bar[MAX_DIRECT_MW];
+
+	bool link_is_up;
+	enum ntb_speed link_speed;
+	enum ntb_width link_width;
+};
+
+static struct switchtec_ntb *ntb_sndev(struct ntb_dev *ntb)
+{
+	return container_of(ntb, struct switchtec_ntb, ntb);
+}
+
+static int switchtec_ntb_part_op(struct switchtec_ntb *sndev,
+				 struct ntb_ctrl_regs __iomem *ctl,
+				 u32 op, int wait_status)
+{
+	static const char * const op_text[] = {
+		[NTB_CTRL_PART_OP_LOCK] = "lock",
+		[NTB_CTRL_PART_OP_CFG] = "configure",
+		[NTB_CTRL_PART_OP_RESET] = "reset",
+	};
+
+	int i;
+	u32 ps;
+	int status;
+
+	switch (op) {
+	case NTB_CTRL_PART_OP_LOCK:
+		status = NTB_CTRL_PART_STATUS_LOCKING;
+		break;
+	case NTB_CTRL_PART_OP_CFG:
+		status = NTB_CTRL_PART_STATUS_CONFIGURING;
+		break;
+	case NTB_CTRL_PART_OP_RESET:
+		status = NTB_CTRL_PART_STATUS_RESETTING;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	iowrite32(op, &ctl->partition_op);
+
+	for (i = 0; i < 1000; i++) {
+		if (msleep_interruptible(50) != 0) {
+			iowrite32(NTB_CTRL_PART_OP_RESET, &ctl->partition_op);
+			return -EINTR;
+		}
+
+		ps = ioread32(&ctl->partition_status) & 0xFFFF;
+
+		if (ps != status)
+			break;
+	}
+
+	if (ps == wait_status)
+		return 0;
+
+	if (ps == status) {
+		dev_err(&sndev->stdev->dev,
+			"Timed out while peforming %s (%d). (%08x)",
+			op_text[op], op,
+			ioread32(&ctl->partition_status));
+
+		return -ETIMEDOUT;
+	}
+
+	return -EIO;
+}
+
+static int switchtec_ntb_send_msg(struct switchtec_ntb *sndev, int idx,
+				  u32 val)
+{
+	if (idx < 0 || idx >= ARRAY_SIZE(sndev->mmio_self_dbmsg->omsg))
+		return -EINVAL;
+
+	iowrite32(val, &sndev->mmio_self_dbmsg->omsg[idx].msg);
+
+	return 0;
+}
+
+static int switchtec_ntb_mw_count(struct ntb_dev *ntb, int pidx)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+	int nr_direct_mw = sndev->peer_nr_direct_mw;
+	int nr_lut_mw = sndev->peer_nr_lut_mw - 1;
+
+	if (pidx != NTB_DEF_PEER_IDX)
+		return -EINVAL;
+
+	if (!use_lut_mws)
+		nr_lut_mw = 0;
+
+	return nr_direct_mw + nr_lut_mw;
+}
+
+static int lut_index(struct switchtec_ntb *sndev, int mw_idx)
+{
+	return mw_idx - sndev->nr_direct_mw + 1;
+}
+
+static int peer_lut_index(struct switchtec_ntb *sndev, int mw_idx)
+{
+	return mw_idx - sndev->peer_nr_direct_mw + 1;
+}
+
+static int switchtec_ntb_mw_get_align(struct ntb_dev *ntb, int pidx,
+				      int widx, resource_size_t *addr_align,
+				      resource_size_t *size_align,
+				      resource_size_t *size_max)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+	int lut;
+	resource_size_t size;
+
+	if (pidx != NTB_DEF_PEER_IDX)
+		return -EINVAL;
+
+	lut = widx >= sndev->peer_nr_direct_mw;
+	size = ioread64(&sndev->peer_shared->mw_sizes[widx]);
+
+	if (size == 0)
+		return -EINVAL;
+
+	if (addr_align)
+		*addr_align = lut ? size : SZ_4K;
+
+	if (size_align)
+		*size_align = lut ? size : SZ_4K;
+
+	if (size_max)
+		*size_max = size;
+
+	return 0;
+}
+
+static void switchtec_ntb_mw_clr_direct(struct switchtec_ntb *sndev, int idx)
+{
+	struct ntb_ctrl_regs __iomem *ctl = sndev->mmio_peer_ctrl;
+	int bar = sndev->peer_direct_mw_to_bar[idx];
+	u32 ctl_val;
+
+	ctl_val = ioread32(&ctl->bar_entry[bar].ctl);
+	ctl_val &= ~NTB_CTRL_BAR_DIR_WIN_EN;
+	iowrite32(ctl_val, &ctl->bar_entry[bar].ctl);
+	iowrite32(0, &ctl->bar_entry[bar].win_size);
+	iowrite64(sndev->self_partition, &ctl->bar_entry[bar].xlate_addr);
+}
+
+static void switchtec_ntb_mw_clr_lut(struct switchtec_ntb *sndev, int idx)
+{
+	struct ntb_ctrl_regs __iomem *ctl = sndev->mmio_peer_ctrl;
+
+	iowrite64(0, &ctl->lut_entry[peer_lut_index(sndev, idx)]);
+}
+
+static void switchtec_ntb_mw_set_direct(struct switchtec_ntb *sndev, int idx,
+					dma_addr_t addr, resource_size_t size)
+{
+	int xlate_pos = ilog2(size);
+	int bar = sndev->peer_direct_mw_to_bar[idx];
+	struct ntb_ctrl_regs __iomem *ctl = sndev->mmio_peer_ctrl;
+	u32 ctl_val;
+
+	ctl_val = ioread32(&ctl->bar_entry[bar].ctl);
+	ctl_val |= NTB_CTRL_BAR_DIR_WIN_EN;
+
+	iowrite32(ctl_val, &ctl->bar_entry[bar].ctl);
+	iowrite32(xlate_pos | size, &ctl->bar_entry[bar].win_size);
+	iowrite64(sndev->self_partition | addr,
+		  &ctl->bar_entry[bar].xlate_addr);
+}
+
+static void switchtec_ntb_mw_set_lut(struct switchtec_ntb *sndev, int idx,
+				     dma_addr_t addr, resource_size_t size)
+{
+	struct ntb_ctrl_regs __iomem *ctl = sndev->mmio_peer_ctrl;
+
+	iowrite64((NTB_CTRL_LUT_EN | (sndev->self_partition << 1) | addr),
+		  &ctl->lut_entry[peer_lut_index(sndev, idx)]);
+}
+
+static int switchtec_ntb_mw_set_trans(struct ntb_dev *ntb, int pidx, int widx,
+				      dma_addr_t addr, resource_size_t size)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+	struct ntb_ctrl_regs __iomem *ctl = sndev->mmio_peer_ctrl;
+	int xlate_pos = ilog2(size);
+	int nr_direct_mw = sndev->peer_nr_direct_mw;
+	int rc;
+
+	if (pidx != NTB_DEF_PEER_IDX)
+		return -EINVAL;
+
+	dev_dbg(&sndev->stdev->dev, "MW %d: part %d addr %pad size %pap",
+		widx, pidx, &addr, &size);
+
+	if (widx >= switchtec_ntb_mw_count(ntb, pidx))
+		return -EINVAL;
+
+	if (xlate_pos < 12)
+		return -EINVAL;
+
+	rc = switchtec_ntb_part_op(sndev, ctl, NTB_CTRL_PART_OP_LOCK,
+				   NTB_CTRL_PART_STATUS_LOCKED);
+	if (rc)
+		return rc;
+
+	if (addr == 0 || size == 0) {
+		if (widx < nr_direct_mw)
+			switchtec_ntb_mw_clr_direct(sndev, widx);
+		else
+			switchtec_ntb_mw_clr_lut(sndev, widx);
+	} else {
+		if (widx < nr_direct_mw)
+			switchtec_ntb_mw_set_direct(sndev, widx, addr, size);
+		else
+			switchtec_ntb_mw_set_lut(sndev, widx, addr, size);
+	}
+
+	rc = switchtec_ntb_part_op(sndev, ctl, NTB_CTRL_PART_OP_CFG,
+				   NTB_CTRL_PART_STATUS_NORMAL);
+
+	if (rc == -EIO) {
+		dev_err(&sndev->stdev->dev,
+			"Hardware reported an error configuring mw %d: %08x",
+			widx, ioread32(&ctl->bar_error));
+
+		if (widx < nr_direct_mw)
+			switchtec_ntb_mw_clr_direct(sndev, widx);
+		else
+			switchtec_ntb_mw_clr_lut(sndev, widx);
+
+		switchtec_ntb_part_op(sndev, ctl, NTB_CTRL_PART_OP_CFG,
+				      NTB_CTRL_PART_STATUS_NORMAL);
+	}
+
+	return rc;
+}
+
+static int switchtec_ntb_peer_mw_count(struct ntb_dev *ntb)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	return sndev->nr_direct_mw + (use_lut_mws ? sndev->nr_lut_mw - 1 : 0);
+}
+
+static int switchtec_ntb_direct_get_addr(struct switchtec_ntb *sndev,
+					 int idx, phys_addr_t *base,
+					 resource_size_t *size)
+{
+	int bar = sndev->direct_mw_to_bar[idx];
+	size_t offset = 0;
+
+	if (bar < 0)
+		return -EINVAL;
+
+	if (idx == 0) {
+		/*
+		 * This is the direct BAR shared with the LUTs
+		 * which means the actual window will be offset
+		 * by the size of all the LUT entries.
+		 */
+
+		offset = LUT_SIZE * sndev->nr_lut_mw;
+	}
+
+	if (base)
+		*base = pci_resource_start(sndev->ntb.pdev, bar) + offset;
+
+	if (size) {
+		*size = pci_resource_len(sndev->ntb.pdev, bar) - offset;
+		if (offset && *size > offset)
+			*size = offset;
+
+		if (*size > max_mw_size)
+			*size = max_mw_size;
+	}
+
+	return 0;
+}
+
+static int switchtec_ntb_lut_get_addr(struct switchtec_ntb *sndev,
+				      int idx, phys_addr_t *base,
+				      resource_size_t *size)
+{
+	int bar = sndev->direct_mw_to_bar[0];
+	int offset;
+
+	offset = LUT_SIZE * lut_index(sndev, idx);
+
+	if (base)
+		*base = pci_resource_start(sndev->ntb.pdev, bar) + offset;
+
+	if (size)
+		*size = LUT_SIZE;
+
+	return 0;
+}
+
+static int switchtec_ntb_peer_mw_get_addr(struct ntb_dev *ntb, int idx,
+					  phys_addr_t *base,
+					  resource_size_t *size)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (idx < sndev->nr_direct_mw)
+		return switchtec_ntb_direct_get_addr(sndev, idx, base, size);
+	else if (idx < switchtec_ntb_peer_mw_count(ntb))
+		return switchtec_ntb_lut_get_addr(sndev, idx, base, size);
+	else
+		return -EINVAL;
+}
+
+static void switchtec_ntb_part_link_speed(struct switchtec_ntb *sndev,
+					  int partition,
+					  enum ntb_speed *speed,
+					  enum ntb_width *width)
+{
+	struct switchtec_dev *stdev = sndev->stdev;
+
+	u32 pff = ioread32(&stdev->mmio_part_cfg[partition].vep_pff_inst_id);
+	u32 linksta = ioread32(&stdev->mmio_pff_csr[pff].pci_cap_region[13]);
+
+	if (speed)
+		*speed = (linksta >> 16) & 0xF;
+
+	if (width)
+		*width = (linksta >> 20) & 0x3F;
+}
+
+static void switchtec_ntb_set_link_speed(struct switchtec_ntb *sndev)
+{
+	enum ntb_speed self_speed, peer_speed;
+	enum ntb_width self_width, peer_width;
+
+	if (!sndev->link_is_up) {
+		sndev->link_speed = NTB_SPEED_NONE;
+		sndev->link_width = NTB_WIDTH_NONE;
+		return;
+	}
+
+	switchtec_ntb_part_link_speed(sndev, sndev->self_partition,
+				      &self_speed, &self_width);
+	switchtec_ntb_part_link_speed(sndev, sndev->peer_partition,
+				      &peer_speed, &peer_width);
+
+	sndev->link_speed = min(self_speed, peer_speed);
+	sndev->link_width = min(self_width, peer_width);
+}
+
+enum {
+	LINK_MESSAGE = 0,
+	MSG_LINK_UP = 1,
+	MSG_LINK_DOWN = 2,
+	MSG_CHECK_LINK = 3,
+};
+
+static void switchtec_ntb_check_link(struct switchtec_ntb *sndev)
+{
+	int link_sta;
+	int old = sndev->link_is_up;
+
+	link_sta = sndev->self_shared->link_sta;
+	if (link_sta) {
+		u64 peer = ioread64(&sndev->peer_shared->magic);
+
+		if ((peer & 0xFFFFFFFF) == SWITCHTEC_NTB_MAGIC)
+			link_sta = peer >> 32;
+		else
+			link_sta = 0;
+	}
+
+	sndev->link_is_up = link_sta;
+	switchtec_ntb_set_link_speed(sndev);
+
+	if (link_sta != old) {
+		switchtec_ntb_send_msg(sndev, LINK_MESSAGE, MSG_CHECK_LINK);
+		ntb_link_event(&sndev->ntb);
+		dev_info(&sndev->stdev->dev, "ntb link %s",
+			 link_sta ? "up" : "down");
+	}
+}
+
+static void switchtec_ntb_link_notification(struct switchtec_dev *stdev)
+{
+	struct switchtec_ntb *sndev = stdev->sndev;
+
+	switchtec_ntb_check_link(sndev);
+}
+
+static u64 switchtec_ntb_link_is_up(struct ntb_dev *ntb,
+				    enum ntb_speed *speed,
+				    enum ntb_width *width)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (speed)
+		*speed = sndev->link_speed;
+	if (width)
+		*width = sndev->link_width;
+
+	return sndev->link_is_up;
+}
+
+static int switchtec_ntb_link_enable(struct ntb_dev *ntb,
+				     enum ntb_speed max_speed,
+				     enum ntb_width max_width)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	dev_dbg(&sndev->stdev->dev, "enabling link");
+
+	sndev->self_shared->link_sta = 1;
+	switchtec_ntb_send_msg(sndev, LINK_MESSAGE, MSG_LINK_UP);
+
+	switchtec_ntb_check_link(sndev);
+
+	return 0;
+}
+
+static int switchtec_ntb_link_disable(struct ntb_dev *ntb)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	dev_dbg(&sndev->stdev->dev, "disabling link");
+
+	sndev->self_shared->link_sta = 0;
+	switchtec_ntb_send_msg(sndev, LINK_MESSAGE, MSG_LINK_UP);
+
+	switchtec_ntb_check_link(sndev);
+
+	return 0;
+}
+
+static u64 switchtec_ntb_db_valid_mask(struct ntb_dev *ntb)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	return sndev->db_valid_mask;
+}
+
+static int switchtec_ntb_db_vector_count(struct ntb_dev *ntb)
+{
+	return 1;
+}
+
+static u64 switchtec_ntb_db_vector_mask(struct ntb_dev *ntb, int db_vector)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (db_vector < 0 || db_vector > 1)
+		return 0;
+
+	return sndev->db_valid_mask;
+}
+
+static u64 switchtec_ntb_db_read(struct ntb_dev *ntb)
+{
+	u64 ret;
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	ret = ioread64(&sndev->mmio_self_dbmsg->idb) >> sndev->db_shift;
+
+	return ret & sndev->db_valid_mask;
+}
+
+static int switchtec_ntb_db_clear(struct ntb_dev *ntb, u64 db_bits)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	iowrite64(db_bits << sndev->db_shift, &sndev->mmio_self_dbmsg->idb);
+
+	return 0;
+}
+
+static int switchtec_ntb_db_set_mask(struct ntb_dev *ntb, u64 db_bits)
+{
+	unsigned long irqflags;
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (db_bits & ~sndev->db_valid_mask)
+		return -EINVAL;
+
+	spin_lock_irqsave(&sndev->db_mask_lock, irqflags);
+
+	sndev->db_mask |= db_bits << sndev->db_shift;
+	iowrite64(~sndev->db_mask, &sndev->mmio_self_dbmsg->idb_mask);
+
+	spin_unlock_irqrestore(&sndev->db_mask_lock, irqflags);
+
+	return 0;
+}
+
+static int switchtec_ntb_db_clear_mask(struct ntb_dev *ntb, u64 db_bits)
+{
+	unsigned long irqflags;
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (db_bits & ~sndev->db_valid_mask)
+		return -EINVAL;
+
+	spin_lock_irqsave(&sndev->db_mask_lock, irqflags);
+
+	sndev->db_mask &= ~(db_bits << sndev->db_shift);
+	iowrite64(~sndev->db_mask, &sndev->mmio_self_dbmsg->idb_mask);
+
+	spin_unlock_irqrestore(&sndev->db_mask_lock, irqflags);
+
+	return 0;
+}
+
+static u64 switchtec_ntb_db_read_mask(struct ntb_dev *ntb)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	return (sndev->db_mask >> sndev->db_shift) & sndev->db_valid_mask;
+}
+
+static int switchtec_ntb_peer_db_addr(struct ntb_dev *ntb,
+				      phys_addr_t *db_addr,
+				      resource_size_t *db_size)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+	unsigned long offset;
+
+	offset = (unsigned long)sndev->mmio_self_dbmsg->odb -
+		(unsigned long)sndev->stdev->mmio;
+
+	offset += sndev->db_shift / 8;
+
+	if (db_addr)
+		*db_addr = pci_resource_start(ntb->pdev, 0) + offset;
+	if (db_size)
+		*db_size = sizeof(u32);
+
+	return 0;
+}
+
+static int switchtec_ntb_peer_db_set(struct ntb_dev *ntb, u64 db_bits)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	iowrite64(db_bits << sndev->db_peer_shift,
+		  &sndev->mmio_self_dbmsg->odb);
+
+	return 0;
+}
+
+static int switchtec_ntb_spad_count(struct ntb_dev *ntb)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	return ARRAY_SIZE(sndev->self_shared->spad);
+}
+
+static u32 switchtec_ntb_spad_read(struct ntb_dev *ntb, int idx)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (idx < 0 || idx >= ARRAY_SIZE(sndev->self_shared->spad))
+		return 0;
+
+	if (!sndev->self_shared)
+		return 0;
+
+	return sndev->self_shared->spad[idx];
+}
+
+static int switchtec_ntb_spad_write(struct ntb_dev *ntb, int idx, u32 val)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (idx < 0 || idx >= ARRAY_SIZE(sndev->self_shared->spad))
+		return -EINVAL;
+
+	if (!sndev->self_shared)
+		return -EIO;
+
+	sndev->self_shared->spad[idx] = val;
+
+	return 0;
+}
+
+static u32 switchtec_ntb_peer_spad_read(struct ntb_dev *ntb, int pidx,
+					int sidx)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (pidx != NTB_DEF_PEER_IDX)
+		return -EINVAL;
+
+	if (sidx < 0 || sidx >= ARRAY_SIZE(sndev->peer_shared->spad))
+		return 0;
+
+	if (!sndev->peer_shared)
+		return 0;
+
+	return ioread32(&sndev->peer_shared->spad[sidx]);
+}
+
+static int switchtec_ntb_peer_spad_write(struct ntb_dev *ntb, int pidx,
+					 int sidx, u32 val)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+
+	if (pidx != NTB_DEF_PEER_IDX)
+		return -EINVAL;
+
+	if (sidx < 0 || sidx >= ARRAY_SIZE(sndev->peer_shared->spad))
+		return -EINVAL;
+
+	if (!sndev->peer_shared)
+		return -EIO;
+
+	iowrite32(val, &sndev->peer_shared->spad[sidx]);
+
+	return 0;
+}
+
+static int switchtec_ntb_peer_spad_addr(struct ntb_dev *ntb, int pidx,
+					int sidx, phys_addr_t *spad_addr)
+{
+	struct switchtec_ntb *sndev = ntb_sndev(ntb);
+	unsigned long offset;
+
+	if (pidx != NTB_DEF_PEER_IDX)
+		return -EINVAL;
+
+	offset = (unsigned long)&sndev->peer_shared->spad[sidx] -
+		(unsigned long)sndev->stdev->mmio;
+
+	if (spad_addr)
+		*spad_addr = pci_resource_start(ntb->pdev, 0) + offset;
+
+	return 0;
+}
+
+static const struct ntb_dev_ops switchtec_ntb_ops = {
+	.mw_count		= switchtec_ntb_mw_count,
+	.mw_get_align		= switchtec_ntb_mw_get_align,
+	.mw_set_trans		= switchtec_ntb_mw_set_trans,
+	.peer_mw_count		= switchtec_ntb_peer_mw_count,
+	.peer_mw_get_addr	= switchtec_ntb_peer_mw_get_addr,
+	.link_is_up		= switchtec_ntb_link_is_up,
+	.link_enable		= switchtec_ntb_link_enable,
+	.link_disable		= switchtec_ntb_link_disable,
+	.db_valid_mask		= switchtec_ntb_db_valid_mask,
+	.db_vector_count	= switchtec_ntb_db_vector_count,
+	.db_vector_mask		= switchtec_ntb_db_vector_mask,
+	.db_read		= switchtec_ntb_db_read,
+	.db_clear		= switchtec_ntb_db_clear,
+	.db_set_mask		= switchtec_ntb_db_set_mask,
+	.db_clear_mask		= switchtec_ntb_db_clear_mask,
+	.db_read_mask		= switchtec_ntb_db_read_mask,
+	.peer_db_addr		= switchtec_ntb_peer_db_addr,
+	.peer_db_set		= switchtec_ntb_peer_db_set,
+	.spad_count		= switchtec_ntb_spad_count,
+	.spad_read		= switchtec_ntb_spad_read,
+	.spad_write		= switchtec_ntb_spad_write,
+	.peer_spad_read		= switchtec_ntb_peer_spad_read,
+	.peer_spad_write	= switchtec_ntb_peer_spad_write,
+	.peer_spad_addr		= switchtec_ntb_peer_spad_addr,
+};
+
+static void switchtec_ntb_init_sndev(struct switchtec_ntb *sndev)
+{
+	u64 part_map;
+
+	sndev->ntb.pdev = sndev->stdev->pdev;
+	sndev->ntb.topo = NTB_TOPO_SWITCH;
+	sndev->ntb.ops = &switchtec_ntb_ops;
+
+	sndev->self_partition = sndev->stdev->partition;
+
+	sndev->mmio_ntb = sndev->stdev->mmio_ntb;
+	part_map = ioread64(&sndev->mmio_ntb->ep_map);
+	part_map &= ~(1 << sndev->self_partition);
+	sndev->peer_partition = ffs(part_map) - 1;
+
+	dev_dbg(&sndev->stdev->dev, "Partition ID %d of %d (%llx)",
+		sndev->self_partition, sndev->stdev->partition_count,
+		part_map);
+
+	sndev->mmio_ctrl = (void * __iomem)sndev->mmio_ntb +
+		SWITCHTEC_NTB_REG_CTRL_OFFSET;
+	sndev->mmio_dbmsg = (void * __iomem)sndev->mmio_ntb +
+		SWITCHTEC_NTB_REG_DBMSG_OFFSET;
+
+	sndev->mmio_self_ctrl = &sndev->mmio_ctrl[sndev->self_partition];
+	sndev->mmio_peer_ctrl = &sndev->mmio_ctrl[sndev->peer_partition];
+	sndev->mmio_self_dbmsg = &sndev->mmio_dbmsg[sndev->self_partition];
+}
+
+static int map_bars(int *map, struct ntb_ctrl_regs __iomem *ctrl)
+{
+	int i;
+	int cnt = 0;
+
+	for (i = 0; i < ARRAY_SIZE(ctrl->bar_entry); i++) {
+		u32 r = ioread32(&ctrl->bar_entry[i].ctl);
+
+		if (r & NTB_CTRL_BAR_VALID)
+			map[cnt++] = i;
+	}
+
+	return cnt;
+}
+
+static void switchtec_ntb_init_mw(struct switchtec_ntb *sndev)
+{
+	sndev->nr_direct_mw = map_bars(sndev->direct_mw_to_bar,
+				       sndev->mmio_self_ctrl);
+
+	sndev->nr_lut_mw = ioread16(&sndev->mmio_self_ctrl->lut_table_entries);
+	sndev->nr_lut_mw = rounddown_pow_of_two(sndev->nr_lut_mw);
+
+	dev_dbg(&sndev->stdev->dev, "MWs: %d direct, %d lut",
+		sndev->nr_direct_mw, sndev->nr_lut_mw);
+
+	sndev->peer_nr_direct_mw = map_bars(sndev->peer_direct_mw_to_bar,
+					    sndev->mmio_peer_ctrl);
+
+	sndev->peer_nr_lut_mw =
+		ioread16(&sndev->mmio_peer_ctrl->lut_table_entries);
+	sndev->peer_nr_lut_mw = rounddown_pow_of_two(sndev->peer_nr_lut_mw);
+
+	dev_dbg(&sndev->stdev->dev, "Peer MWs: %d direct, %d lut",
+		sndev->peer_nr_direct_mw, sndev->peer_nr_lut_mw);
+
+}
+
+/*
+ * There are 64 doorbells in the switch hardware but this is
+ * shared among all partitions. So we must split them in half
+ * (32 for each partition). However, the message interrupts are
+ * also shared with the top 4 doorbells so we just limit this to
+ * 28 doorbells per partition
+ */
+static void switchtec_ntb_init_db(struct switchtec_ntb *sndev)
+{
+	sndev->db_valid_mask = 0x0FFFFFFF;
+
+	if (sndev->self_partition < sndev->peer_partition) {
+		sndev->db_shift = 0;
+		sndev->db_peer_shift = 32;
+	} else {
+		sndev->db_shift = 32;
+		sndev->db_peer_shift = 0;
+	}
+
+	sndev->db_mask = 0x0FFFFFFFFFFFFFFFULL;
+	iowrite64(~sndev->db_mask, &sndev->mmio_self_dbmsg->idb_mask);
+	iowrite64(sndev->db_valid_mask << sndev->db_peer_shift,
+		  &sndev->mmio_self_dbmsg->odb_mask);
+}
+
+static void switchtec_ntb_init_msgs(struct switchtec_ntb *sndev)
+{
+	int i;
+	u32 msg_map = 0;
+
+	for (i = 0; i < ARRAY_SIZE(sndev->mmio_self_dbmsg->imsg); i++) {
+		int m = i | sndev->peer_partition << 2;
+
+		msg_map |= m << i * 8;
+	}
+
+	iowrite32(msg_map, &sndev->mmio_self_dbmsg->msg_map);
+
+	for (i = 0; i < ARRAY_SIZE(sndev->mmio_self_dbmsg->imsg); i++)
+		iowrite64(NTB_DBMSG_IMSG_STATUS | NTB_DBMSG_IMSG_MASK,
+			  &sndev->mmio_self_dbmsg->imsg[i]);
+}
+
+static int switchtec_ntb_init_req_id_table(struct switchtec_ntb *sndev)
+{
+	int rc = 0;
+	u16 req_id;
+	u32 error;
+
+	req_id = ioread16(&sndev->mmio_ntb->requester_id);
+
+	if (ioread32(&sndev->mmio_self_ctrl->req_id_table_size) < 2) {
+		dev_err(&sndev->stdev->dev,
+			"Not enough requester IDs available.");
+		return -EFAULT;
+	}
+
+	rc = switchtec_ntb_part_op(sndev, sndev->mmio_self_ctrl,
+				   NTB_CTRL_PART_OP_LOCK,
+				   NTB_CTRL_PART_STATUS_LOCKED);
+	if (rc)
+		return rc;
+
+	iowrite32(NTB_PART_CTRL_ID_PROT_DIS,
+		  &sndev->mmio_self_ctrl->partition_ctrl);
+
+	/*
+	 * Root Complex Requester ID (which is 0:00.0)
+	 */
+	iowrite32(0 << 16 | NTB_CTRL_REQ_ID_EN,
+		  &sndev->mmio_self_ctrl->req_id_table[0]);
+
+	/*
+	 * Host Bridge Requester ID (as read from the mmap address)
+	 */
+	iowrite32(req_id << 16 | NTB_CTRL_REQ_ID_EN,
+		  &sndev->mmio_self_ctrl->req_id_table[1]);
+
+	rc = switchtec_ntb_part_op(sndev, sndev->mmio_self_ctrl,
+				   NTB_CTRL_PART_OP_CFG,
+				   NTB_CTRL_PART_STATUS_NORMAL);
+	if (rc == -EIO) {
+		error = ioread32(&sndev->mmio_self_ctrl->req_id_error);
+		dev_err(&sndev->stdev->dev,
+			"Error setting up the requester ID table: %08x",
+			error);
+	}
+
+	return rc;
+}
+
+static void switchtec_ntb_init_shared(struct switchtec_ntb *sndev)
+{
+	int i;
+
+	memset(sndev->self_shared, 0, LUT_SIZE);
+	sndev->self_shared->magic = SWITCHTEC_NTB_MAGIC;
+	sndev->self_shared->partition_id = sndev->stdev->partition;
+
+	for (i = 0; i < sndev->nr_direct_mw; i++) {
+		int bar = sndev->direct_mw_to_bar[i];
+		resource_size_t sz = pci_resource_len(sndev->stdev->pdev, bar);
+
+		if (i == 0)
+			sz = min_t(resource_size_t, sz,
+				   LUT_SIZE * sndev->nr_lut_mw);
+
+		sndev->self_shared->mw_sizes[i] = sz;
+	}
+
+	for (i = 0; i < sndev->nr_lut_mw; i++) {
+		int idx = sndev->nr_direct_mw + i;
+
+		sndev->self_shared->mw_sizes[idx] = LUT_SIZE;
+	}
+}
+
+static int switchtec_ntb_init_shared_mw(struct switchtec_ntb *sndev)
+{
+	struct ntb_ctrl_regs __iomem *ctl = sndev->mmio_peer_ctrl;
+	int bar = sndev->direct_mw_to_bar[0];
+	u32 ctl_val;
+	int rc;
+
+	sndev->self_shared = dma_zalloc_coherent(&sndev->stdev->pdev->dev,
+						 LUT_SIZE,
+						 &sndev->self_shared_dma,
+						 GFP_KERNEL);
+	if (!sndev->self_shared) {
+		dev_err(&sndev->stdev->dev,
+			"unable to allocate memory for shared mw");
+		return -ENOMEM;
+	}
+
+	switchtec_ntb_init_shared(sndev);
+
+	rc = switchtec_ntb_part_op(sndev, ctl, NTB_CTRL_PART_OP_LOCK,
+				   NTB_CTRL_PART_STATUS_LOCKED);
+	if (rc)
+		goto unalloc_and_exit;
+
+	ctl_val = ioread32(&ctl->bar_entry[bar].ctl);
+	ctl_val &= 0xFF;
+	ctl_val |= NTB_CTRL_BAR_LUT_WIN_EN;
+	ctl_val |= ilog2(LUT_SIZE) << 8;
+	ctl_val |= (sndev->nr_lut_mw - 1) << 14;
+	iowrite32(ctl_val, &ctl->bar_entry[bar].ctl);
+
+	iowrite64((NTB_CTRL_LUT_EN | (sndev->self_partition << 1) |
+		   sndev->self_shared_dma),
+		  &ctl->lut_entry[0]);
+
+	rc = switchtec_ntb_part_op(sndev, ctl, NTB_CTRL_PART_OP_CFG,
+				   NTB_CTRL_PART_STATUS_NORMAL);
+	if (rc) {
+		u32 bar_error, lut_error;
+
+		bar_error = ioread32(&ctl->bar_error);
+		lut_error = ioread32(&ctl->lut_error);
+		dev_err(&sndev->stdev->dev,
+			"Error setting up shared MW: %08x / %08x",
+			bar_error, lut_error);
+		goto unalloc_and_exit;
+	}
+
+	sndev->peer_shared = pci_iomap(sndev->stdev->pdev, bar, LUT_SIZE);
+	if (!sndev->peer_shared) {
+		rc = -ENOMEM;
+		goto unalloc_and_exit;
+	}
+
+	dev_dbg(&sndev->stdev->dev, "Shared MW Ready");
+	return 0;
+
+unalloc_and_exit:
+	dma_free_coherent(&sndev->stdev->pdev->dev, LUT_SIZE,
+			  sndev->self_shared, sndev->self_shared_dma);
+
+	return rc;
+}
+
+static void switchtec_ntb_deinit_shared_mw(struct switchtec_ntb *sndev)
+{
+	if (sndev->peer_shared)
+		pci_iounmap(sndev->stdev->pdev, sndev->peer_shared);
+
+	if (sndev->self_shared)
+		dma_free_coherent(&sndev->stdev->pdev->dev, LUT_SIZE,
+				  sndev->self_shared,
+				  sndev->self_shared_dma);
+}
+
+static irqreturn_t switchtec_ntb_doorbell_isr(int irq, void *dev)
+{
+	struct switchtec_ntb *sndev = dev;
+
+	dev_dbg(&sndev->stdev->dev, "doorbell\n");
+
+	ntb_db_event(&sndev->ntb, 0);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t switchtec_ntb_message_isr(int irq, void *dev)
+{
+	int i;
+	struct switchtec_ntb *sndev = dev;
+
+	for (i = 0; i < ARRAY_SIZE(sndev->mmio_self_dbmsg->imsg); i++) {
+		u64 msg = ioread64(&sndev->mmio_self_dbmsg->imsg[i]);
+
+		if (msg & NTB_DBMSG_IMSG_STATUS) {
+			dev_dbg(&sndev->stdev->dev, "message: %d %08x\n", i,
+				(u32)msg);
+			iowrite8(1, &sndev->mmio_self_dbmsg->imsg[i].status);
+
+			if (i == LINK_MESSAGE)
+				switchtec_ntb_check_link(sndev);
+		}
+	}
+
+	return IRQ_HANDLED;
+}
+
+static int switchtec_ntb_init_db_msg_irq(struct switchtec_ntb *sndev)
+{
+	int i;
+	int rc;
+	int doorbell_irq = 0;
+	int message_irq = 0;
+	int event_irq;
+	int idb_vecs = sizeof(sndev->mmio_self_dbmsg->idb_vec_map);
+
+	event_irq = ioread32(&sndev->stdev->mmio_part_cfg->vep_vector_number);
+
+	while (doorbell_irq == event_irq)
+		doorbell_irq++;
+	while (message_irq == doorbell_irq ||
+	       message_irq == event_irq)
+		message_irq++;
+
+	dev_dbg(&sndev->stdev->dev, "irqs - event: %d, db: %d, msgs: %d",
+		event_irq, doorbell_irq, message_irq);
+
+	for (i = 0; i < idb_vecs - 4; i++)
+		iowrite8(doorbell_irq,
+			 &sndev->mmio_self_dbmsg->idb_vec_map[i]);
+
+	for (; i < idb_vecs; i++)
+		iowrite8(message_irq,
+			 &sndev->mmio_self_dbmsg->idb_vec_map[i]);
+
+	sndev->doorbell_irq = pci_irq_vector(sndev->stdev->pdev, doorbell_irq);
+	sndev->message_irq = pci_irq_vector(sndev->stdev->pdev, message_irq);
+
+	rc = request_irq(sndev->doorbell_irq,
+			 switchtec_ntb_doorbell_isr, 0,
+			 "switchtec_ntb_doorbell", sndev);
+	if (rc)
+		return rc;
+
+	rc = request_irq(sndev->message_irq,
+			 switchtec_ntb_message_isr, 0,
+			 "switchtec_ntb_message", sndev);
+	if (rc) {
+		free_irq(sndev->doorbell_irq, sndev);
+		return rc;
+	}
+
+	return 0;
+}
+
+static void switchtec_ntb_deinit_db_msg_irq(struct switchtec_ntb *sndev)
+{
+	free_irq(sndev->doorbell_irq, sndev);
+	free_irq(sndev->message_irq, sndev);
+}
+
+static int switchtec_ntb_add(struct device *dev,
+			     struct class_interface *class_intf)
+{
+	struct switchtec_dev *stdev = to_stdev(dev);
+	struct switchtec_ntb *sndev;
+	int rc;
+
+	stdev->sndev = NULL;
+
+	if (stdev->pdev->class != MICROSEMI_NTB_CLASSCODE)
+		return -ENODEV;
+
+	if (stdev->partition_count != 2)
+		dev_warn(dev, "ntb driver only supports 2 partitions");
+
+	sndev = kzalloc_node(sizeof(*sndev), GFP_KERNEL, dev_to_node(dev));
+	if (!sndev)
+		return -ENOMEM;
+
+	sndev->stdev = stdev;
+	switchtec_ntb_init_sndev(sndev);
+	switchtec_ntb_init_mw(sndev);
+	switchtec_ntb_init_db(sndev);
+	switchtec_ntb_init_msgs(sndev);
+
+	rc = switchtec_ntb_init_req_id_table(sndev);
+	if (rc)
+		goto free_and_exit;
+
+	rc = switchtec_ntb_init_shared_mw(sndev);
+	if (rc)
+		goto free_and_exit;
+
+	rc = switchtec_ntb_init_db_msg_irq(sndev);
+	if (rc)
+		goto deinit_shared_and_exit;
+
+	rc = ntb_register_device(&sndev->ntb);
+	if (rc)
+		goto deinit_and_exit;
+
+	stdev->sndev = sndev;
+	stdev->link_notifier = switchtec_ntb_link_notification;
+	dev_info(dev, "NTB device registered");
+
+	return 0;
+
+deinit_and_exit:
+	switchtec_ntb_deinit_db_msg_irq(sndev);
+deinit_shared_and_exit:
+	switchtec_ntb_deinit_shared_mw(sndev);
+free_and_exit:
+	kfree(sndev);
+	dev_err(dev, "failed to register ntb device: %d", rc);
+	return rc;
+}
+
+void switchtec_ntb_remove(struct device *dev,
+			  struct class_interface *class_intf)
+{
+	struct switchtec_dev *stdev = to_stdev(dev);
+	struct switchtec_ntb *sndev = stdev->sndev;
+
+	if (!sndev)
+		return;
+
+	stdev->link_notifier = NULL;
+	stdev->sndev = NULL;
+	ntb_unregister_device(&sndev->ntb);
+	switchtec_ntb_deinit_db_msg_irq(sndev);
+	switchtec_ntb_deinit_shared_mw(sndev);
+	kfree(sndev);
+	dev_info(dev, "ntb device unregistered");
+}
+
+static struct class_interface switchtec_interface  = {
+	.add_dev = switchtec_ntb_add,
+	.remove_dev = switchtec_ntb_remove,
+};
+
+static int __init switchtec_ntb_init(void)
+{
+	switchtec_interface.class = switchtec_class;
+	return class_interface_register(&switchtec_interface);
+}
+module_init(switchtec_ntb_init);
+
+static void __exit switchtec_ntb_exit(void)
+{
+	class_interface_unregister(&switchtec_interface);
+}
+module_exit(switchtec_ntb_exit);

drivers/ntb/ntb_transport.c

@@ -191,8 +191,6 @@ struct ntb_transport_qp {
 struct ntb_transport_mw {
 	phys_addr_t phys_addr;
 	resource_size_t phys_size;
-	resource_size_t xlat_align;
-	resource_size_t xlat_align_size;
 	void __iomem *vbase;
 	size_t xlat_size;
 	size_t buff_size;
@@ -687,13 +685,20 @@ static int ntb_set_mw(struct ntb_transport_ctx *nt, int num_mw,
 	struct ntb_transport_mw *mw = &nt->mw_vec[num_mw];
 	struct pci_dev *pdev = nt->ndev->pdev;
 	size_t xlat_size, buff_size;
+	resource_size_t xlat_align;
+	resource_size_t xlat_align_size;
 	int rc;
 
 	if (!size)
 		return -EINVAL;
 
-	xlat_size = round_up(size, mw->xlat_align_size);
-	buff_size = round_up(size, mw->xlat_align);
+	rc = ntb_mw_get_align(nt->ndev, PIDX, num_mw, &xlat_align,
+			      &xlat_align_size, NULL);
+	if (rc)
+		return rc;
+
+	xlat_size = round_up(size, xlat_align_size);
+	buff_size = round_up(size, xlat_align);
 
 	/* No need to re-setup */
 	if (mw->xlat_size == xlat_size)
@@ -722,7 +727,7 @@ static int ntb_set_mw(struct ntb_transport_ctx *nt, int num_mw,
 	 * is a requirement of the hardware. It is recommended to setup CMA
 	 * for BAR sizes equal or greater than 4MB.
 	 */
-	if (!IS_ALIGNED(mw->dma_addr, mw->xlat_align)) {
+	if (!IS_ALIGNED(mw->dma_addr, xlat_align)) {
 		dev_err(&pdev->dev, "DMA memory %pad is not aligned\n",
 			&mw->dma_addr);
 		ntb_free_mw(nt, num_mw);
@@ -1104,11 +1109,6 @@ static int ntb_transport_probe(struct ntb_client *self, struct ntb_dev *ndev)
 	for (i = 0; i < mw_count; i++) {
 		mw = &nt->mw_vec[i];
 
-		rc = ntb_mw_get_align(ndev, PIDX, i, &mw->xlat_align,
-				      &mw->xlat_align_size, NULL);
-		if (rc)
-			goto err1;
-
 		rc = ntb_peer_mw_get_addr(ndev, i, &mw->phys_addr,
 					  &mw->phys_size);
 		if (rc)

drivers/ntb/test/ntb_perf.c

@@ -108,8 +108,6 @@ MODULE_PARM_DESC(on_node, "Run threads only on NTB device node (default: true)")
 struct perf_mw {
 	phys_addr_t	phys_addr;
 	resource_size_t	phys_size;
-	resource_size_t	xlat_align;
-	resource_size_t	xlat_align_size;
 	void __iomem	*vbase;
 	size_t		xlat_size;
 	size_t		buf_size;
@@ -472,13 +470,20 @@ static int perf_set_mw(struct perf_ctx *perf, resource_size_t size)
 {
 	struct perf_mw *mw = &perf->mw;
 	size_t xlat_size, buf_size;
+	resource_size_t	xlat_align;
+	resource_size_t	xlat_align_size;
 	int rc;
 
 	if (!size)
 		return -EINVAL;
 
-	xlat_size = round_up(size, mw->xlat_align_size);
-	buf_size = round_up(size, mw->xlat_align);
+	rc = ntb_mw_get_align(perf->ntb, PIDX, 0, &xlat_align,
+			      &xlat_align_size, NULL);
+	if (rc)
+		return rc;
+
+	xlat_size = round_up(size, xlat_align_size);
+	buf_size = round_up(size, xlat_align);
 
 	if (mw->xlat_size == xlat_size)
 		return 0;
@@ -567,11 +572,6 @@ static int perf_setup_mw(struct ntb_dev *ntb, struct perf_ctx *perf)
 
 	mw = &perf->mw;
 
-	rc = ntb_mw_get_align(ntb, PIDX, 0, &mw->xlat_align,
-			      &mw->xlat_align_size, NULL);
-	if (rc)
-		return rc;
-
 	rc = ntb_peer_mw_get_addr(ntb, 0, &mw->phys_addr, &mw->phys_size);
 	if (rc)
 		return rc;

drivers/ntb/test/ntb_tool.c

@@ -753,9 +753,9 @@ static ssize_t tool_peer_mw_trans_read(struct file *filep,
 
 	phys_addr_t base;
 	resource_size_t mw_size;
-	resource_size_t align_addr;
-	resource_size_t align_size;
-	resource_size_t max_size;
+	resource_size_t align_addr = 0;
+	resource_size_t align_size = 0;
+	resource_size_t max_size = 0;
 
 	buf_size = min_t(size_t, size, 512);
 

drivers/pci/switch/switchtec.c

@@ -13,6 +13,7 @@
  *
  */
 
+#include <linux/switchtec.h>
 #include <linux/switchtec_ioctl.h>
 
 #include <linux/interrupt.h>
@@ -20,8 +21,6 @@
 #include <linux/fs.h>
 #include <linux/uaccess.h>
 #include <linux/poll.h>
-#include <linux/pci.h>
-#include <linux/cdev.h>
 #include <linux/wait.h>
 
 MODULE_DESCRIPTION("Microsemi Switchtec(tm) PCIe Management Driver");
@@ -34,265 +33,10 @@ module_param(max_devices, int, 0644);
 MODULE_PARM_DESC(max_devices, "max number of switchtec device instances");
 
 static dev_t switchtec_devt;
-static struct class *switchtec_class;
 static DEFINE_IDA(switchtec_minor_ida);
 
-#define MICROSEMI_VENDOR_ID         0x11f8
-#define MICROSEMI_NTB_CLASSCODE     0x068000
-#define MICROSEMI_MGMT_CLASSCODE    0x058000
-
-#define SWITCHTEC_MRPC_PAYLOAD_SIZE 1024
-#define SWITCHTEC_MAX_PFF_CSR 48
-
-#define SWITCHTEC_EVENT_OCCURRED BIT(0)
-#define SWITCHTEC_EVENT_CLEAR    BIT(0)
-#define SWITCHTEC_EVENT_EN_LOG   BIT(1)
-#define SWITCHTEC_EVENT_EN_CLI   BIT(2)
-#define SWITCHTEC_EVENT_EN_IRQ   BIT(3)
-#define SWITCHTEC_EVENT_FATAL    BIT(4)
-
-enum {
-	SWITCHTEC_GAS_MRPC_OFFSET       = 0x0000,
-	SWITCHTEC_GAS_TOP_CFG_OFFSET    = 0x1000,
-	SWITCHTEC_GAS_SW_EVENT_OFFSET   = 0x1800,
-	SWITCHTEC_GAS_SYS_INFO_OFFSET   = 0x2000,
-	SWITCHTEC_GAS_FLASH_INFO_OFFSET = 0x2200,
-	SWITCHTEC_GAS_PART_CFG_OFFSET   = 0x4000,
-	SWITCHTEC_GAS_NTB_OFFSET        = 0x10000,
-	SWITCHTEC_GAS_PFF_CSR_OFFSET    = 0x134000,
-};
-
-struct mrpc_regs {
-	u8 input_data[SWITCHTEC_MRPC_PAYLOAD_SIZE];
-	u8 output_data[SWITCHTEC_MRPC_PAYLOAD_SIZE];
-	u32 cmd;
-	u32 status;
-	u32 ret_value;
-} __packed;
-
-enum mrpc_status {
-	SWITCHTEC_MRPC_STATUS_INPROGRESS = 1,
-	SWITCHTEC_MRPC_STATUS_DONE = 2,
-	SWITCHTEC_MRPC_STATUS_ERROR = 0xFF,
-	SWITCHTEC_MRPC_STATUS_INTERRUPTED = 0x100,
-};
-
-struct sw_event_regs {
-	u64 event_report_ctrl;
-	u64 reserved1;
-	u64 part_event_bitmap;
-	u64 reserved2;
-	u32 global_summary;
-	u32 reserved3[3];
-	u32 stack_error_event_hdr;
-	u32 stack_error_event_data;
-	u32 reserved4[4];
-	u32 ppu_error_event_hdr;
-	u32 ppu_error_event_data;
-	u32 reserved5[4];
-	u32 isp_error_event_hdr;
-	u32 isp_error_event_data;
-	u32 reserved6[4];
-	u32 sys_reset_event_hdr;
-	u32 reserved7[5];
-	u32 fw_exception_hdr;
-	u32 reserved8[5];
-	u32 fw_nmi_hdr;
-	u32 reserved9[5];
-	u32 fw_non_fatal_hdr;
-	u32 reserved10[5];
-	u32 fw_fatal_hdr;
-	u32 reserved11[5];
-	u32 twi_mrpc_comp_hdr;
-	u32 twi_mrpc_comp_data;
-	u32 reserved12[4];
-	u32 twi_mrpc_comp_async_hdr;
-	u32 twi_mrpc_comp_async_data;
-	u32 reserved13[4];
-	u32 cli_mrpc_comp_hdr;
-	u32 cli_mrpc_comp_data;
-	u32 reserved14[4];
-	u32 cli_mrpc_comp_async_hdr;
-	u32 cli_mrpc_comp_async_data;
-	u32 reserved15[4];
-	u32 gpio_interrupt_hdr;
-	u32 gpio_interrupt_data;
-	u32 reserved16[4];
-} __packed;
-
-enum {
-	SWITCHTEC_CFG0_RUNNING = 0x04,
-	SWITCHTEC_CFG1_RUNNING = 0x05,
-	SWITCHTEC_IMG0_RUNNING = 0x03,
-	SWITCHTEC_IMG1_RUNNING = 0x07,
-};
-
-struct sys_info_regs {
-	u32 device_id;
-	u32 device_version;
-	u32 firmware_version;
-	u32 reserved1;
-	u32 vendor_table_revision;
-	u32 table_format_version;
-	u32 partition_id;
-	u32 cfg_file_fmt_version;
-	u16 cfg_running;
-	u16 img_running;
-	u32 reserved2[57];
-	char vendor_id[8];
-	char product_id[16];
-	char product_revision[4];
-	char component_vendor[8];
-	u16 component_id;
-	u8 component_revision;
-} __packed;
-
-struct flash_info_regs {
-	u32 flash_part_map_upd_idx;
-
-	struct active_partition_info {
-		u32 address;
-		u32 build_version;
-		u32 build_string;
-	} active_img;
-
-	struct active_partition_info active_cfg;
-	struct active_partition_info inactive_img;
-	struct active_partition_info inactive_cfg;
-
-	u32 flash_length;
-
-	struct partition_info {
-		u32 address;
-		u32 length;
-	} cfg0;
-
-	struct partition_info cfg1;
-	struct partition_info img0;
-	struct partition_info img1;
-	struct partition_info nvlog;
-	struct partition_info vendor[8];
-};
-
-struct ntb_info_regs {
-	u8  partition_count;
-	u8  partition_id;
-	u16 reserved1;
-	u64 ep_map;
-	u16 requester_id;
-} __packed;
-
-struct part_cfg_regs {
-	u32 status;
-	u32 state;
-	u32 port_cnt;
-	u32 usp_port_mode;
-	u32 usp_pff_inst_id;
-	u32 vep_pff_inst_id;
-	u32 dsp_pff_inst_id[47];
-	u32 reserved1[11];
-	u16 vep_vector_number;
-	u16 usp_vector_number;
-	u32 port_event_bitmap;
-	u32 reserved2[3];
-	u32 part_event_summary;
-	u32 reserved3[3];
-	u32 part_reset_hdr;
-	u32 part_reset_data[5];
-	u32 mrpc_comp_hdr;
-	u32 mrpc_comp_data[5];
-	u32 mrpc_comp_async_hdr;
-	u32 mrpc_comp_async_data[5];
-	u32 dyn_binding_hdr;
-	u32 dyn_binding_data[5];
-	u32 reserved4[159];
-} __packed;
-
-enum {
-	SWITCHTEC_PART_CFG_EVENT_RESET = 1 << 0,
-	SWITCHTEC_PART_CFG_EVENT_MRPC_CMP = 1 << 1,
-	SWITCHTEC_PART_CFG_EVENT_MRPC_ASYNC_CMP = 1 << 2,
-	SWITCHTEC_PART_CFG_EVENT_DYN_PART_CMP = 1 << 3,
-};
-
-struct pff_csr_regs {
-	u16 vendor_id;
-	u16 device_id;
-	u32 pci_cfg_header[15];
-	u32 pci_cap_region[48];
-	u32 pcie_cap_region[448];
-	u32 indirect_gas_window[128];
-	u32 indirect_gas_window_off;
-	u32 reserved[127];
-	u32 pff_event_summary;
-	u32 reserved2[3];
-	u32 aer_in_p2p_hdr;
-	u32 aer_in_p2p_data[5];
-	u32 aer_in_vep_hdr;
-	u32 aer_in_vep_data[5];
-	u32 dpc_hdr;
-	u32 dpc_data[5];
-	u32 cts_hdr;
-	u32 cts_data[5];
-	u32 reserved3[6];
-	u32 hotplug_hdr;
-	u32 hotplug_data[5];
-	u32 ier_hdr;
-	u32 ier_data[5];
-	u32 threshold_hdr;
-	u32 threshold_data[5];
-	u32 power_mgmt_hdr;
-	u32 power_mgmt_data[5];
-	u32 tlp_throttling_hdr;
-	u32 tlp_throttling_data[5];
-	u32 force_speed_hdr;
-	u32 force_speed_data[5];
-	u32 credit_timeout_hdr;
-	u32 credit_timeout_data[5];
-	u32 link_state_hdr;
-	u32 link_state_data[5];
-	u32 reserved4[174];
-} __packed;
-
-struct switchtec_dev {
-	struct pci_dev *pdev;
-	struct device dev;
-	struct cdev cdev;
-
-	int partition;
-	int partition_count;
-	int pff_csr_count;
-	char pff_local[SWITCHTEC_MAX_PFF_CSR];
-
-	void __iomem *mmio;
-	struct mrpc_regs __iomem *mmio_mrpc;
-	struct sw_event_regs __iomem *mmio_sw_event;
-	struct sys_info_regs __iomem *mmio_sys_info;
-	struct flash_info_regs __iomem *mmio_flash_info;
-	struct ntb_info_regs __iomem *mmio_ntb;
-	struct part_cfg_regs __iomem *mmio_part_cfg;
-	struct part_cfg_regs __iomem *mmio_part_cfg_all;
-	struct pff_csr_regs __iomem *mmio_pff_csr;
-
-	/*
-	 * The mrpc mutex must be held when accessing the other
-	 * mrpc_ fields, alive flag and stuser->state field
-	 */
-	struct mutex mrpc_mutex;
-	struct list_head mrpc_queue;
-	int mrpc_busy;
-	struct work_struct mrpc_work;
-	struct delayed_work mrpc_timeout;
-	bool alive;
-
-	wait_queue_head_t event_wq;
-	atomic_t event_cnt;
-};
-
-static struct switchtec_dev *to_stdev(struct device *dev)
-{
-	return container_of(dev, struct switchtec_dev, dev);
-}
+struct class *switchtec_class;
+EXPORT_SYMBOL_GPL(switchtec_class);
 
 enum mrpc_state {
 	MRPC_IDLE = 0,
@@ -1234,6 +978,49 @@ static const struct file_operations switchtec_fops = {
 	.compat_ioctl = switchtec_dev_ioctl,
 };
 
+static void link_event_work(struct work_struct *work)
+{
+	struct switchtec_dev *stdev;
+
+	stdev = container_of(work, struct switchtec_dev, link_event_work);
+
+	if (stdev->link_notifier)
+		stdev->link_notifier(stdev);
+}
+
+static void check_link_state_events(struct switchtec_dev *stdev)
+{
+	int idx;
+	u32 reg;
+	int count;
+	int occurred = 0;
+
+	for (idx = 0; idx < stdev->pff_csr_count; idx++) {
+		reg = ioread32(&stdev->mmio_pff_csr[idx].link_state_hdr);
+		dev_dbg(&stdev->dev, "link_state: %d->%08x\n", idx, reg);
+		count = (reg >> 5) & 0xFF;
+
+		if (count != stdev->link_event_count[idx]) {
+			occurred = 1;
+			stdev->link_event_count[idx] = count;
+		}
+	}
+
+	if (occurred)
+		schedule_work(&stdev->link_event_work);
+}
+
+static void enable_link_state_events(struct switchtec_dev *stdev)
+{
+	int idx;
+
+	for (idx = 0; idx < stdev->pff_csr_count; idx++) {
+		iowrite32(SWITCHTEC_EVENT_CLEAR |
+			  SWITCHTEC_EVENT_EN_IRQ,
+			  &stdev->mmio_pff_csr[idx].link_state_hdr);
+	}
+}
+
 static void stdev_release(struct device *dev)
 {
 	struct switchtec_dev *stdev = to_stdev(dev);
@@ -1286,6 +1073,7 @@ static struct switchtec_dev *stdev_create(struct pci_dev *pdev)
 	stdev->mrpc_busy = 0;
 	INIT_WORK(&stdev->mrpc_work, mrpc_event_work);
 	INIT_DELAYED_WORK(&stdev->mrpc_timeout, mrpc_timeout_work);
+	INIT_WORK(&stdev->link_event_work, link_event_work);
 	init_waitqueue_head(&stdev->event_wq);
 	atomic_set(&stdev->event_cnt, 0);
 
@@ -1329,6 +1117,9 @@ static int mask_event(struct switchtec_dev *stdev, int eid, int idx)
 	if (!(hdr & SWITCHTEC_EVENT_OCCURRED && hdr & SWITCHTEC_EVENT_EN_IRQ))
 		return 0;
 
+	if (eid == SWITCHTEC_IOCTL_EVENT_LINK_STATE)
+		return 0;
+
 	dev_dbg(&stdev->dev, "%s: %d %d %x\n", __func__, eid, idx, hdr);
 	hdr &= ~(SWITCHTEC_EVENT_EN_IRQ | SWITCHTEC_EVENT_OCCURRED);
 	iowrite32(hdr, hdr_reg);
@@ -1348,6 +1139,7 @@ static int mask_all_events(struct switchtec_dev *stdev, int eid)
 		for (idx = 0; idx < stdev->pff_csr_count; idx++) {
 			if (!stdev->pff_local[idx])
 				continue;
+
 			count += mask_event(stdev, eid, idx);
 		}
 	} else {
@@ -1372,6 +1164,8 @@ static irqreturn_t switchtec_event_isr(int irq, void *dev)
 		iowrite32(reg, &stdev->mmio_part_cfg->mrpc_comp_hdr);
 	}
 
+	check_link_state_events(stdev);
+
 	for (eid = 0; eid < SWITCHTEC_IOCTL_MAX_EVENTS; eid++)
 		event_count += mask_all_events(stdev, eid);
 
@@ -1481,6 +1275,9 @@ static int switchtec_pci_probe(struct pci_dev *pdev,
 	struct switchtec_dev *stdev;
 	int rc;
 
+	if (pdev->class == MICROSEMI_NTB_CLASSCODE)
+		request_module_nowait("ntb_hw_switchtec");
+
 	stdev = stdev_create(pdev);
 	if (IS_ERR(stdev))
 		return PTR_ERR(stdev);
@@ -1498,6 +1295,7 @@ static int switchtec_pci_probe(struct pci_dev *pdev,
 	iowrite32(SWITCHTEC_EVENT_CLEAR |
 		  SWITCHTEC_EVENT_EN_IRQ,
 		  &stdev->mmio_part_cfg->mrpc_comp_hdr);
+	enable_link_state_events(stdev);
 
 	rc = cdev_device_add(&stdev->cdev, &stdev->dev);
 	if (rc)

include/linux/ntb.h

@@ -70,6 +70,7 @@ struct pci_dev;
  * @NTB_TOPO_SEC:	On secondary side of remote ntb.
  * @NTB_TOPO_B2B_USD:	On primary side of local ntb upstream of remote ntb.
  * @NTB_TOPO_B2B_DSD:	On primary side of local ntb downstream of remote ntb.
+ * @NTB_TOPO_SWITCH:	Connected via a switch which supports ntb.
  */
 enum ntb_topo {
 	NTB_TOPO_NONE = -1,
@@ -77,6 +78,7 @@ enum ntb_topo {
 	NTB_TOPO_SEC,
 	NTB_TOPO_B2B_USD,
 	NTB_TOPO_B2B_DSD,
+	NTB_TOPO_SWITCH,
 };
 
 static inline int ntb_topo_is_b2b(enum ntb_topo topo)
@@ -97,6 +99,7 @@ static inline char *ntb_topo_string(enum ntb_topo topo)
 	case NTB_TOPO_SEC:	return "NTB_TOPO_SEC";
 	case NTB_TOPO_B2B_USD:	return "NTB_TOPO_B2B_USD";
 	case NTB_TOPO_B2B_DSD:	return "NTB_TOPO_B2B_DSD";
+	case NTB_TOPO_SWITCH:	return "NTB_TOPO_SWITCH";
 	}
 	return "NTB_TOPO_INVALID";
 }
@@ -730,7 +733,8 @@ static inline int ntb_link_disable(struct ntb_dev *ntb)
  * Hardware and topology may support a different number of memory windows.
  * Moreover different peer devices can support different number of memory
  * windows. Simply speaking this method returns the number of possible inbound
- * memory windows to share with specified peer device.
+ * memory windows to share with specified peer device. Note: this may return
+ * zero if the link is not up yet.
  *
  * Return: the number of memory windows.
  */
@@ -751,7 +755,7 @@ static inline int ntb_mw_count(struct ntb_dev *ntb, int pidx)
  * Get the alignments of an inbound memory window with specified index.
  * NULL may be given for any output parameter if the value is not needed.
  * The alignment and size parameters may be used for allocation of proper
- * shared memory.
+ * shared memory. Note: this must only be called when the link is up.
  *
  * Return: Zero on success, otherwise a negative error number.
  */
@@ -760,6 +764,9 @@ static inline int ntb_mw_get_align(struct ntb_dev *ntb, int pidx, int widx,
 				   resource_size_t *size_align,
 				   resource_size_t *size_max)
 {
+	if (!(ntb_link_is_up(ntb, NULL, NULL) & (1 << pidx)))
+		return -ENOTCONN;
+
 	return ntb->ops->mw_get_align(ntb, pidx, widx, addr_align, size_align,
 				      size_max);
 }

include/linux/switchtec.h

@@ -0,0 +1,373 @@
+/*
+ * Microsemi Switchtec PCIe Driver
+ * Copyright (c) 2017, Microsemi Corporation
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ */
+
+#ifndef _SWITCHTEC_H
+#define _SWITCHTEC_H
+
+#include <linux/pci.h>
+#include <linux/cdev.h>
+
+#define MICROSEMI_VENDOR_ID         0x11f8
+#define MICROSEMI_NTB_CLASSCODE     0x068000
+#define MICROSEMI_MGMT_CLASSCODE    0x058000
+
+#define SWITCHTEC_MRPC_PAYLOAD_SIZE 1024
+#define SWITCHTEC_MAX_PFF_CSR 48
+
+#define SWITCHTEC_EVENT_OCCURRED BIT(0)
+#define SWITCHTEC_EVENT_CLEAR    BIT(0)
+#define SWITCHTEC_EVENT_EN_LOG   BIT(1)
+#define SWITCHTEC_EVENT_EN_CLI   BIT(2)
+#define SWITCHTEC_EVENT_EN_IRQ   BIT(3)
+#define SWITCHTEC_EVENT_FATAL    BIT(4)
+
+enum {
+	SWITCHTEC_GAS_MRPC_OFFSET       = 0x0000,
+	SWITCHTEC_GAS_TOP_CFG_OFFSET    = 0x1000,
+	SWITCHTEC_GAS_SW_EVENT_OFFSET   = 0x1800,
+	SWITCHTEC_GAS_SYS_INFO_OFFSET   = 0x2000,
+	SWITCHTEC_GAS_FLASH_INFO_OFFSET = 0x2200,
+	SWITCHTEC_GAS_PART_CFG_OFFSET   = 0x4000,
+	SWITCHTEC_GAS_NTB_OFFSET        = 0x10000,
+	SWITCHTEC_GAS_PFF_CSR_OFFSET    = 0x134000,
+};
+
+struct mrpc_regs {
+	u8 input_data[SWITCHTEC_MRPC_PAYLOAD_SIZE];
+	u8 output_data[SWITCHTEC_MRPC_PAYLOAD_SIZE];
+	u32 cmd;
+	u32 status;
+	u32 ret_value;
+} __packed;
+
+enum mrpc_status {
+	SWITCHTEC_MRPC_STATUS_INPROGRESS = 1,
+	SWITCHTEC_MRPC_STATUS_DONE = 2,
+	SWITCHTEC_MRPC_STATUS_ERROR = 0xFF,
+	SWITCHTEC_MRPC_STATUS_INTERRUPTED = 0x100,
+};
+
+struct sw_event_regs {
+	u64 event_report_ctrl;
+	u64 reserved1;
+	u64 part_event_bitmap;
+	u64 reserved2;
+	u32 global_summary;
+	u32 reserved3[3];
+	u32 stack_error_event_hdr;
+	u32 stack_error_event_data;
+	u32 reserved4[4];
+	u32 ppu_error_event_hdr;
+	u32 ppu_error_event_data;
+	u32 reserved5[4];
+	u32 isp_error_event_hdr;
+	u32 isp_error_event_data;
+	u32 reserved6[4];
+	u32 sys_reset_event_hdr;
+	u32 reserved7[5];
+	u32 fw_exception_hdr;
+	u32 reserved8[5];
+	u32 fw_nmi_hdr;
+	u32 reserved9[5];
+	u32 fw_non_fatal_hdr;
+	u32 reserved10[5];
+	u32 fw_fatal_hdr;
+	u32 reserved11[5];
+	u32 twi_mrpc_comp_hdr;
+	u32 twi_mrpc_comp_data;
+	u32 reserved12[4];
+	u32 twi_mrpc_comp_async_hdr;
+	u32 twi_mrpc_comp_async_data;
+	u32 reserved13[4];
+	u32 cli_mrpc_comp_hdr;
+	u32 cli_mrpc_comp_data;
+	u32 reserved14[4];
+	u32 cli_mrpc_comp_async_hdr;
+	u32 cli_mrpc_comp_async_data;
+	u32 reserved15[4];
+	u32 gpio_interrupt_hdr;
+	u32 gpio_interrupt_data;
+	u32 reserved16[4];
+} __packed;
+
+enum {
+	SWITCHTEC_CFG0_RUNNING = 0x04,
+	SWITCHTEC_CFG1_RUNNING = 0x05,
+	SWITCHTEC_IMG0_RUNNING = 0x03,
+	SWITCHTEC_IMG1_RUNNING = 0x07,
+};
+
+struct sys_info_regs {
+	u32 device_id;
+	u32 device_version;
+	u32 firmware_version;
+	u32 reserved1;
+	u32 vendor_table_revision;
+	u32 table_format_version;
+	u32 partition_id;
+	u32 cfg_file_fmt_version;
+	u16 cfg_running;
+	u16 img_running;
+	u32 reserved2[57];
+	char vendor_id[8];
+	char product_id[16];
+	char product_revision[4];
+	char component_vendor[8];
+	u16 component_id;
+	u8 component_revision;
+} __packed;
+
+struct flash_info_regs {
+	u32 flash_part_map_upd_idx;
+
+	struct active_partition_info {
+		u32 address;
+		u32 build_version;
+		u32 build_string;
+	} active_img;
+
+	struct active_partition_info active_cfg;
+	struct active_partition_info inactive_img;
+	struct active_partition_info inactive_cfg;
+
+	u32 flash_length;
+
+	struct partition_info {
+		u32 address;
+		u32 length;
+	} cfg0;
+
+	struct partition_info cfg1;
+	struct partition_info img0;
+	struct partition_info img1;
+	struct partition_info nvlog;
+	struct partition_info vendor[8];
+};
+
+enum {
+	SWITCHTEC_NTB_REG_INFO_OFFSET   = 0x0000,
+	SWITCHTEC_NTB_REG_CTRL_OFFSET   = 0x4000,
+	SWITCHTEC_NTB_REG_DBMSG_OFFSET  = 0x64000,
+};
+
+struct ntb_info_regs {
+	u8  partition_count;
+	u8  partition_id;
+	u16 reserved1;
+	u64 ep_map;
+	u16 requester_id;
+} __packed;
+
+struct part_cfg_regs {
+	u32 status;
+	u32 state;
+	u32 port_cnt;
+	u32 usp_port_mode;
+	u32 usp_pff_inst_id;
+	u32 vep_pff_inst_id;
+	u32 dsp_pff_inst_id[47];
+	u32 reserved1[11];
+	u16 vep_vector_number;
+	u16 usp_vector_number;
+	u32 port_event_bitmap;
+	u32 reserved2[3];
+	u32 part_event_summary;
+	u32 reserved3[3];
+	u32 part_reset_hdr;
+	u32 part_reset_data[5];
+	u32 mrpc_comp_hdr;
+	u32 mrpc_comp_data[5];
+	u32 mrpc_comp_async_hdr;
+	u32 mrpc_comp_async_data[5];
+	u32 dyn_binding_hdr;
+	u32 dyn_binding_data[5];
+	u32 reserved4[159];
+} __packed;
+
+enum {
+	NTB_CTRL_PART_OP_LOCK = 0x1,
+	NTB_CTRL_PART_OP_CFG = 0x2,
+	NTB_CTRL_PART_OP_RESET = 0x3,
+
+	NTB_CTRL_PART_STATUS_NORMAL = 0x1,
+	NTB_CTRL_PART_STATUS_LOCKED = 0x2,
+	NTB_CTRL_PART_STATUS_LOCKING = 0x3,
+	NTB_CTRL_PART_STATUS_CONFIGURING = 0x4,
+	NTB_CTRL_PART_STATUS_RESETTING = 0x5,
+
+	NTB_CTRL_BAR_VALID = 1 << 0,
+	NTB_CTRL_BAR_DIR_WIN_EN = 1 << 4,
+	NTB_CTRL_BAR_LUT_WIN_EN = 1 << 5,
+
+	NTB_CTRL_REQ_ID_EN = 1 << 0,
+
+	NTB_CTRL_LUT_EN = 1 << 0,
+
+	NTB_PART_CTRL_ID_PROT_DIS = 1 << 0,
+};
+
+struct ntb_ctrl_regs {
+	u32 partition_status;
+	u32 partition_op;
+	u32 partition_ctrl;
+	u32 bar_setup;
+	u32 bar_error;
+	u16 lut_table_entries;
+	u16 lut_table_offset;
+	u32 lut_error;
+	u16 req_id_table_size;
+	u16 req_id_table_offset;
+	u32 req_id_error;
+	u32 reserved1[7];
+	struct {
+		u32 ctl;
+		u32 win_size;
+		u64 xlate_addr;
+	} bar_entry[6];
+	u32 reserved2[216];
+	u32 req_id_table[256];
+	u32 reserved3[512];
+	u64 lut_entry[512];
+} __packed;
+
+#define NTB_DBMSG_IMSG_STATUS BIT_ULL(32)
+#define NTB_DBMSG_IMSG_MASK   BIT_ULL(40)
+
+struct ntb_dbmsg_regs {
+	u32 reserved1[1024];
+	u64 odb;
+	u64 odb_mask;
+	u64 idb;
+	u64 idb_mask;
+	u8  idb_vec_map[64];
+	u32 msg_map;
+	u32 reserved2;
+	struct {
+		u32 msg;
+		u32 status;
+	} omsg[4];
+
+	struct {
+		u32 msg;
+		u8  status;
+		u8  mask;
+		u8  src;
+		u8  reserved;
+	} imsg[4];
+
+	u8 reserved3[3928];
+	u8 msix_table[1024];
+	u8 reserved4[3072];
+	u8 pba[24];
+	u8 reserved5[4072];
+} __packed;
+
+enum {
+	SWITCHTEC_PART_CFG_EVENT_RESET = 1 << 0,
+	SWITCHTEC_PART_CFG_EVENT_MRPC_CMP = 1 << 1,
+	SWITCHTEC_PART_CFG_EVENT_MRPC_ASYNC_CMP = 1 << 2,
+	SWITCHTEC_PART_CFG_EVENT_DYN_PART_CMP = 1 << 3,
+};
+
+struct pff_csr_regs {
+	u16 vendor_id;
+	u16 device_id;
+	u32 pci_cfg_header[15];
+	u32 pci_cap_region[48];
+	u32 pcie_cap_region[448];
+	u32 indirect_gas_window[128];
+	u32 indirect_gas_window_off;
+	u32 reserved[127];
+	u32 pff_event_summary;
+	u32 reserved2[3];
+	u32 aer_in_p2p_hdr;
+	u32 aer_in_p2p_data[5];
+	u32 aer_in_vep_hdr;
+	u32 aer_in_vep_data[5];
+	u32 dpc_hdr;
+	u32 dpc_data[5];
+	u32 cts_hdr;
+	u32 cts_data[5];
+	u32 reserved3[6];
+	u32 hotplug_hdr;
+	u32 hotplug_data[5];
+	u32 ier_hdr;
+	u32 ier_data[5];
+	u32 threshold_hdr;
+	u32 threshold_data[5];
+	u32 power_mgmt_hdr;
+	u32 power_mgmt_data[5];
+	u32 tlp_throttling_hdr;
+	u32 tlp_throttling_data[5];
+	u32 force_speed_hdr;
+	u32 force_speed_data[5];
+	u32 credit_timeout_hdr;
+	u32 credit_timeout_data[5];
+	u32 link_state_hdr;
+	u32 link_state_data[5];
+	u32 reserved4[174];
+} __packed;
+
+struct switchtec_ntb;
+
+struct switchtec_dev {
+	struct pci_dev *pdev;
+	struct device dev;
+	struct cdev cdev;
+
+	int partition;
+	int partition_count;
+	int pff_csr_count;
+	char pff_local[SWITCHTEC_MAX_PFF_CSR];
+
+	void __iomem *mmio;
+	struct mrpc_regs __iomem *mmio_mrpc;
+	struct sw_event_regs __iomem *mmio_sw_event;
+	struct sys_info_regs __iomem *mmio_sys_info;
+	struct flash_info_regs __iomem *mmio_flash_info;
+	struct ntb_info_regs __iomem *mmio_ntb;
+	struct part_cfg_regs __iomem *mmio_part_cfg;
+	struct part_cfg_regs __iomem *mmio_part_cfg_all;
+	struct pff_csr_regs __iomem *mmio_pff_csr;
+
+	/*
+	 * The mrpc mutex must be held when accessing the other
+	 * mrpc_ fields, alive flag and stuser->state field
+	 */
+	struct mutex mrpc_mutex;
+	struct list_head mrpc_queue;
+	int mrpc_busy;
+	struct work_struct mrpc_work;
+	struct delayed_work mrpc_timeout;
+	bool alive;
+
+	wait_queue_head_t event_wq;
+	atomic_t event_cnt;
+
+	struct work_struct link_event_work;
+	void (*link_notifier)(struct switchtec_dev *stdev);
+	u8 link_event_count[SWITCHTEC_MAX_PFF_CSR];
+
+	struct switchtec_ntb *sndev;
+};
+
+static inline struct switchtec_dev *to_stdev(struct device *dev)
+{
+	return container_of(dev, struct switchtec_dev, dev);
+}
+
+extern struct class *switchtec_class;
+
+#endif