mlx5: Add driver for Mellanox Connect-IB adapters

The driver is comprised of two kernel modules: mlx5_ib and mlx5_core.
This partitioning resembles what we have for mlx4, except that mlx5_ib
is the pci device driver and not mlx5_core.

mlx5_core is essentially a library that provides general functionality
that is intended to be used by other Mellanox devices that will be
introduced in the future.  mlx5_ib has a similar role as any hardware
device under drivers/infiniband/hw.

Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>

[ Merge in coccinelle fixes from Fengguang Wu <fengguang.wu@intel.com>.
  - Roland ]

Signed-off-by: Roland Dreier <roland@purestorage.com>

22 files changed, 5767 insertions, 0 deletions

diff --git a/drivers/net/ethernet/mellanox/Kconfig b/drivers/net/ethernet/mellanox/Kconfig
index bcdbc14aeff0..8cf7563a8d92 100644
--- a/drivers/net/ethernet/mellanox/Kconfig
+++ b/drivers/net/ethernet/mellanox/Kconfig
@@ -19,5 +19,6 @@ config NET_VENDOR_MELLANOX
 if NET_VENDOR_MELLANOX
 
 source "drivers/net/ethernet/mellanox/mlx4/Kconfig"
+source "drivers/net/ethernet/mellanox/mlx5/core/Kconfig"
 
 endif # NET_VENDOR_MELLANOX
diff --git a/drivers/net/ethernet/mellanox/Makefile b/drivers/net/ethernet/mellanox/Makefile
index 37afb9683372..38fe32ef5e5f 100644
--- a/drivers/net/ethernet/mellanox/Makefile
+++ b/drivers/net/ethernet/mellanox/Makefile
@@ -3,3 +3,4 @@
 #
 
 obj-$(CONFIG_MLX4_CORE) += mlx4/
+obj-$(CONFIG_MLX5_CORE) += mlx5/core/
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Kconfig b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
new file mode 100644
index 000000000000..21962828925a
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Kconfig
@@ -0,0 +1,18 @@
+#
+# Mellanox driver configuration
+#
+
+config MLX5_CORE
+	tristate
+	depends on PCI && X86
+	default n
+
+config MLX5_DEBUG
+	bool "Verbose debugging output" if (MLX5_CORE && EXPERT)
+	depends on MLX5_CORE
+	default y
+	---help---
+	  This option causes debugging code to be compiled into the
+	  mlx5_core driver.  The output can be turned on via the
+	  debug_mask module parameter (which can also be set after
+	  the driver is loaded through sysfs).
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/Makefile b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
new file mode 100644
index 000000000000..105780bb980b
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/Makefile
@@ -0,0 +1,5 @@
+obj-$(CONFIG_MLX5_CORE)		+= mlx5_core.o
+
+mlx5_core-y :=	main.o cmd.o debugfs.o fw.o eq.o uar.o pagealloc.o \
+		health.o mcg.o cq.o srq.o alloc.o qp.o port.o mr.o pd.o   \
+		mad.o
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/alloc.c b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
new file mode 100644
index 000000000000..b215742b842f
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/alloc.c
@@ -0,0 +1,238 @@
+/*
+ * Copyright (c) 2013, Mellanox Technologies inc.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/errno.h>
+#include <linux/slab.h>
+#include <linux/mm.h>
+#include <linux/export.h>
+#include <linux/bitmap.h>
+#include <linux/dma-mapping.h>
+#include <linux/vmalloc.h>
+#include <linux/mlx5/driver.h>
+
+#include "mlx5_core.h"
+
+/* Handling for queue buffers -- we allocate a bunch of memory and
+ * register it in a memory region at HCA virtual address 0.  If the
+ * requested size is > max_direct, we split the allocation into
+ * multiple pages, so we don't require too much contiguous memory.
+ */
+
+int mlx5_buf_alloc(struct mlx5_core_dev *dev, int size, int max_direct,
+		   struct mlx5_buf *buf)
+{
+	dma_addr_t t;
+
+	buf->size = size;
+	if (size <= max_direct) {
+		buf->nbufs        = 1;
+		buf->npages       = 1;
+		buf->page_shift   = get_order(size) + PAGE_SHIFT;
+		buf->direct.buf   = dma_zalloc_coherent(&dev->pdev->dev,
+							size, &t, GFP_KERNEL);
+		if (!buf->direct.buf)
+			return -ENOMEM;
+
+		buf->direct.map = t;
+
+		while (t & ((1 << buf->page_shift) - 1)) {
+			--buf->page_shift;
+			buf->npages *= 2;
+		}
+	} else {
+		int i;
+
+		buf->direct.buf  = NULL;
+		buf->nbufs       = (size + PAGE_SIZE - 1) / PAGE_SIZE;
+		buf->npages      = buf->nbufs;
+		buf->page_shift  = PAGE_SHIFT;
+		buf->page_list   = kcalloc(buf->nbufs, sizeof(*buf->page_list),
+					   GFP_KERNEL);
+		if (!buf->page_list)
+			return -ENOMEM;
+
+		for (i = 0; i < buf->nbufs; i++) {
+			buf->page_list[i].buf =
+				dma_zalloc_coherent(&dev->pdev->dev, PAGE_SIZE,
+						    &t, GFP_KERNEL);
+			if (!buf->page_list[i].buf)
+				goto err_free;
+
+			buf->page_list[i].map = t;
+		}
+
+		if (BITS_PER_LONG == 64) {
+			struct page **pages;
+			pages = kmalloc(sizeof(*pages) * buf->nbufs, GFP_KERNEL);
+			if (!pages)
+				goto err_free;
+			for (i = 0; i < buf->nbufs; i++)
+				pages[i] = virt_to_page(buf->page_list[i].buf);
+			buf->direct.buf = vmap(pages, buf->nbufs, VM_MAP, PAGE_KERNEL);
+			kfree(pages);
+			if (!buf->direct.buf)
+				goto err_free;
+		}
+	}
+
+	return 0;
+
+err_free:
+	mlx5_buf_free(dev, buf);
+
+	return -ENOMEM;
+}
+EXPORT_SYMBOL_GPL(mlx5_buf_alloc);
+
+void mlx5_buf_free(struct mlx5_core_dev *dev, struct mlx5_buf *buf)
+{
+	int i;
+
+	if (buf->nbufs == 1)
+		dma_free_coherent(&dev->pdev->dev, buf->size, buf->direct.buf,
+				  buf->direct.map);
+	else {
+		if (BITS_PER_LONG == 64 && buf->direct.buf)
+			vunmap(buf->direct.buf);
+
+		for (i = 0; i < buf->nbufs; i++)
+			if (buf->page_list[i].buf)
+				dma_free_coherent(&dev->pdev->dev, PAGE_SIZE,
+						  buf->page_list[i].buf,
+						  buf->page_list[i].map);
+		kfree(buf->page_list);
+	}
+}
+EXPORT_SYMBOL_GPL(mlx5_buf_free);
+
+static struct mlx5_db_pgdir *mlx5_alloc_db_pgdir(struct device *dma_device)
+{
+	struct mlx5_db_pgdir *pgdir;
+
+	pgdir = kzalloc(sizeof(*pgdir), GFP_KERNEL);
+	if (!pgdir)
+		return NULL;
+
+	bitmap_fill(pgdir->bitmap, MLX5_DB_PER_PAGE);
+	pgdir->db_page = dma_alloc_coherent(dma_device, PAGE_SIZE,
+					    &pgdir->db_dma, GFP_KERNEL);
+	if (!pgdir->db_page) {
+		kfree(pgdir);
+		return NULL;
+	}
+
+	return pgdir;
+}
+
+static int mlx5_alloc_db_from_pgdir(struct mlx5_db_pgdir *pgdir,
+				    struct mlx5_db *db)
+{
+	int offset;
+	int i;
+
+	i = find_first_bit(pgdir->bitmap, MLX5_DB_PER_PAGE);
+	if (i >= MLX5_DB_PER_PAGE)
+		return -ENOMEM;
+
+	__clear_bit(i, pgdir->bitmap);
+
+	db->u.pgdir = pgdir;
+	db->index   = i;
+	offset = db->index * L1_CACHE_BYTES;
+	db->db      = pgdir->db_page + offset / sizeof(*pgdir->db_page);
+	db->dma     = pgdir->db_dma  + offset;
+
+	return 0;
+}
+
+int mlx5_db_alloc(struct mlx5_core_dev *dev, struct mlx5_db *db)
+{
+	struct mlx5_db_pgdir *pgdir;
+	int ret = 0;
+
+	mutex_lock(&dev->priv.pgdir_mutex);
+
+	list_for_each_entry(pgdir, &dev->priv.pgdir_list, list)
+		if (!mlx5_alloc_db_from_pgdir(pgdir, db))
+			goto out;
+
+	pgdir = mlx5_alloc_db_pgdir(&(dev->pdev->dev));
+	if (!pgdir) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
+	list_add(&pgdir->list, &dev->priv.pgdir_list);
+
+	/* This should never fail -- we just allocated an empty page: */
+	WARN_ON(mlx5_alloc_db_from_pgdir(pgdir, db));
+
+out:
+	mutex_unlock(&dev->priv.pgdir_mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL_GPL(mlx5_db_alloc);
+
+void mlx5_db_free(struct mlx5_core_dev *dev, struct mlx5_db *db)
+{
+	mutex_lock(&dev->priv.pgdir_mutex);
+
+	__set_bit(db->index, db->u.pgdir->bitmap);
+
+	if (bitmap_full(db->u.pgdir->bitmap, MLX5_DB_PER_PAGE)) {
+		dma_free_coherent(&(dev->pdev->dev), PAGE_SIZE,
+				  db->u.pgdir->db_page, db->u.pgdir->db_dma);
+		list_del(&db->u.pgdir->list);
+		kfree(db->u.pgdir);
+	}
+
+	mutex_unlock(&dev->priv.pgdir_mutex);
+}
+EXPORT_SYMBOL_GPL(mlx5_db_free);
+
+
+void mlx5_fill_page_array(struct mlx5_buf *buf, __be64 *pas)
+{
+	u64 addr;
+	int i;
+
+	for (i = 0; i < buf->npages; i++) {
+		if (buf->nbufs == 1)
+			addr = buf->direct.map + (i << buf->page_shift);
+		else
+			addr = buf->page_list[i].map;
+
+		pas[i] = cpu_to_be64(addr);
+	}
+}
+EXPORT_SYMBOL_GPL(mlx5_fill_page_array);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
new file mode 100644
index 000000000000..c1c0eef89694
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -0,0 +1,1515 @@
+/*
+ * Copyright (c) 2013, Mellanox Technologies inc.  All rights reserved.
+ *
+ * This software is available to you under a choice of one of two
+ * licenses.  You may choose to be licensed under the terms of the GNU
+ * General Public License (GPL) Version 2, available from the file
+ * COPYING in the main directory of this source tree, or the
+ * OpenIB.org BSD license below:
+ *
+ *     Redistribution and use in source and binary forms, with or
+ *     without modification, are permitted provided that the following
+ *     conditions are met:
+ *
+ *      - Redistributions of source code must retain the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer.
+ *
+ *      - Redistributions in binary form must reproduce the above
+ *        copyright notice, this list of conditions and the following
+ *        disclaimer in the documentation and/or other materials
+ *        provided with the distribution.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <asm-generic/kmap_types.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/dma-mapping.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/random.h>
+#include <linux/io-mapping.h>
+#include <linux/mlx5/driver.h>
+#include <linux/debugfs.h>
+
+#include "mlx5_core.h"
+
+enum {
+	CMD_IF_REV = 3,
+};
+
+enum {
+	CMD_MODE_POLLING,
+	CMD_MODE_EVENTS
+};
+
+enum {
+	NUM_LONG_LISTS	  = 2,
+	NUM_MED_LISTS	  = 64,
+	LONG_LIST_SIZE	  = (2ULL * 1024 * 1024 * 1024 / PAGE_SIZE) * 8 + 16 +
+				MLX5_CMD_DATA_BLOCK_SIZE,
+	MED_LIST_SIZE	  = 16 + MLX5_CMD_DATA_BLOCK_SIZE,
+};
+
+enum {
+	MLX5_CMD_DELIVERY_STAT_OK			= 0x0,
+	MLX5_CMD_DELIVERY_STAT_SIGNAT_ERR		= 0x1,
+	MLX5_CMD_DELIVERY_STAT_TOK_ERR			= 0x2,
+	MLX5_CMD_DELIVERY_STAT_BAD_BLK_NUM_ERR		= 0x3,
+	MLX5_CMD_DELIVERY_STAT_OUT_PTR_ALIGN_ERR	= 0x4,
+	MLX5_CMD_DELIVERY_STAT_IN_PTR_ALIGN_ERR		= 0x5,
+	MLX5_CMD_DELIVERY_STAT_FW_ERR			= 0x6,
+	MLX5_CMD_DELIVERY_STAT_IN_LENGTH_ERR		= 0x7,
+	MLX5_CMD_DELIVERY_STAT_OUT_LENGTH_ERR		= 0x8,
+	MLX5_CMD_DELIVERY_STAT_RES_FLD_NOT_CLR_ERR	= 0x9,
+	MLX5_CMD_DELIVERY_STAT_CMD_DESCR_ERR		= 0x10,
+};
+
+enum {
+	MLX5_CMD_STAT_OK			= 0x0,
+	MLX5_CMD_STAT_INT_ERR			= 0x1,
+	MLX5_CMD_STAT_BAD_OP_ERR		= 0x2,
+	MLX5_CMD_STAT_BAD_PARAM_ERR		= 0x3,
+	MLX5_CMD_STAT_BAD_SYS_STATE_ERR		= 0x4,
+	MLX5_CMD_STAT_BAD_RES_ERR		= 0x5,
+	MLX5_CMD_STAT_RES_BUSY			= 0x6,
+	MLX5_CMD_STAT_LIM_ERR			= 0x8,
+	MLX5_CMD_STAT_BAD_RES_STATE_ERR		= 0x9,
+	MLX5_CMD_STAT_IX_ERR			= 0xa,
+	MLX5_CMD_STAT_NO_RES_ERR		= 0xf,
+	MLX5_CMD_STAT_BAD_INP_LEN_ERR		= 0x50,
+	MLX5_CMD_STAT_BAD_OUTP_LEN_ERR		= 0x51,
+	MLX5_CMD_STAT_BAD_QP_STATE_ERR		= 0x10,
+	MLX5_CMD_STAT_BAD_PKT_ERR		= 0x30,
+	MLX5_CMD_STAT_BAD_SIZE_OUTS_CQES_ERR	= 0x40,
+};
+
+static struct mlx5_cmd_work_ent *alloc_cmd(struct mlx5_cmd *cmd,
+					   struct mlx5_cmd_msg *in,
+					   struct mlx5_cmd_msg *out,
+					   mlx5_cmd_cbk_t cbk,
+					   void *context, int page_queue)
+{
+	gfp_t alloc_flags = cbk ? GFP_ATOMIC : GFP_KERNEL;
+	struct mlx5_cmd_work_ent *ent;
+
+	ent = kzalloc(sizeof(*ent), alloc_flags);
+	if (!ent)
+		return ERR_PTR(-ENOMEM);
+
+	ent->in		= in;
+	ent->out	= out;
+	ent->callback	= cbk;
+	ent->context	= context;
+	ent->cmd	= cmd;
+	ent->page_queue = page_queue;
+
+	return ent;
+}
+
+static u8 alloc_token(struct mlx5_cmd *cmd)
+{
+	u8 token;
+
+	spin_lock(&cmd->token_lock);
+	token = cmd->token++ % 255 + 1;
+	spin_unlock(&cmd->token_lock);
+
+	return token;
+}
+
+static int alloc_ent(struct mlx5_cmd *cmd)
+{
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&cmd->alloc_lock, flags);
+	ret = find_first_bit(&cmd->bitmask, cmd->max_reg_cmds);
+	if (ret < cmd->max_reg_cmds)
+		clear_bit(ret, &cmd->bitmask);
+	spin_unlock_irqrestore(&cmd->alloc_lock, flags);
+
+	return ret < cmd->max_reg_cmds ? ret : -ENOMEM;
+}
+
+static void free_ent(struct mlx5_cmd *cmd, int idx)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&cmd->alloc_lock, flags);
+	set_bit(idx, &cmd->bitmask);
+	spin_unlock_irqrestore(&cmd->alloc_lock, flags);
+}
+
+static struct mlx5_cmd_layout *get_inst(struct mlx5_cmd *cmd, int idx)
+{
+	return cmd->cmd_buf + (idx << cmd->log_stride);
+}
+
+static u8 xor8_buf(void *buf, int len)
+{
+	u8 *ptr = buf;
+	u8 sum = 0;
+	int i;
+
+	for (i = 0; i < len; i++)
+		sum ^= ptr[i];
+
+	return sum;
+}
+
+static int verify_block_sig(struct mlx5_cmd_prot_block *block)
+{
+	if (xor8_buf(block->rsvd0, sizeof(*block) - sizeof(block->data) - 1) != 0xff)
+		return -EINVAL;
+
+	if (xor8_buf(block, sizeof(*block)) != 0xff)
+		return -EINVAL;
+
+	return 0;
+}
+
+static void calc_block_sig(struct mlx5_cmd_prot_block *block, u8 token)
+{
+	block->token = token;
+	block->ctrl_sig = ~xor8_buf(block->rsvd0, sizeof(*block) - sizeof(block->data) - 2);
+	block->sig = ~xor8_buf(block, sizeof(*block) - 1);
+}
+
+static void calc_chain_sig(struct mlx5_cmd_msg *msg, u8 token)
+{
+	struct mlx5_cmd_mailbox *next = msg->next;
+
+	while (next) {
+		calc_block_sig(next->buf, token);
+		next = next->next;
+	}
+}
+
+static void set_signature(struct mlx5_cmd_work_ent *ent)
+{
+	ent->lay->sig = ~xor8_buf(ent->lay, sizeof(*ent->lay));
+	calc_chain_sig(ent->in, ent->token);
+	calc_chain_sig(ent->out, ent->token);
+}
+
+static void poll_timeout(struct mlx5_cmd_work_ent *ent)
+{
+	unsigned long poll_end = jiffies + msecs_to_jiffies(MLX5_CMD_TIMEOUT_MSEC + 1000);
+	u8 own;
+
+	do {
+		own = ent->lay->status_own;
+		if (!(own & CMD_OWNER_HW)) {
+			ent->ret = 0;
+			return;
+		}
+		usleep_range(5000, 10000);
+	} while (time_before(jiffies, poll_end));
+
+	ent->ret = -ETIMEDOUT;
+}
+
+static void free_cmd(struct mlx5_cmd_work_ent *ent)
+{
+	kfree(ent);
+}
+
+
+static int verify_signature(struct mlx5_cmd_work_ent *ent)
+{
+	struct mlx5_cmd_mailbox *next = ent->out->next;
+	int err;
+	u8 sig;
+
+	sig = xor8_buf(ent->lay, sizeof(*ent->lay));
+	if (sig != 0xff)
+		return -EINVAL;
+
+	while (next) {
+		err = verify_block_sig(next->buf);
+		if (err)
+			return err;
+
+		next = next->next;
+	}
+
+	return 0;
+}
+
+static void dump_buf(void *buf, int size, int data_only, int offset)
+{
+	__be32 *p = buf;
+	int i;
+
+	for (i = 0; i < size; i += 16) {
+		pr_debug("%03x: %08x %08x %08x %08x\n", offset, be32_to_cpu(p[0]),
+			 be32_to_cpu(p[1]), be32_to_cpu(p[2]),
+			 be32_to_cpu(p[3]));
+		p += 4;
+		offset += 16;
+	}
+	if (!data_only)
+		pr_debug("\n");
+}
+
+const char *mlx5_command_str(int command)
+{
+	switch (command) {
+	case MLX5_CMD_OP_QUERY_HCA_CAP:
+		return "QUERY_HCA_CAP";
+
+	case MLX5_CMD_OP_SET_HCA_CAP:
+		return "SET_HCA_CAP";
+
+	case MLX5_CMD_OP_QUERY_ADAPTER:
+		return "QUERY_ADAPTER";
+
+	case MLX5_CMD_OP_INIT_HCA:
+		return "INIT_HCA";
+
+	case MLX5_CMD_OP_TEARDOWN_HCA:
+		return "TEARDOWN_HCA";
+
+	case MLX5_CMD_OP_QUERY_PAGES:
+		return "QUERY_PAGES";
+
+	case MLX5_CMD_OP_MANAGE_PAGES:
+		return "MANAGE_PAGES";
+
+	case MLX5_CMD_OP_CREATE_MKEY:
+		return "CREATE_MKEY";
+
+	case MLX5_CMD_OP_QUERY_MKEY:
+		return "QUERY_MKEY";
+
+	case MLX5_CMD_OP_DESTROY_MKEY:
+		return "DESTROY_MKEY";
+
+	case MLX5_CMD_OP_QUERY_SPECIAL_CONTEXTS:
+		return "QUERY_SPECIAL_CONTEXTS";
+
+	case MLX5_CMD_OP_CREATE_EQ:
+		return "CREATE_EQ";
+
+	case MLX5_CMD_OP_DESTROY_EQ:
+		return "DESTROY_EQ";
+
+	case MLX5_CMD_OP_QUERY_EQ:
+		return "QUERY_EQ";
+
+	case MLX5_CMD_OP_CREATE_CQ:
+		return "CREATE_CQ";
+
+	case MLX5_CMD_OP_DESTROY_CQ:
+		return "DESTROY_CQ";
+
+	case MLX5_CMD_OP_QUERY_CQ:
+		return "QUERY_CQ";
+
+	case MLX5_CMD_OP_MODIFY_CQ:
+		return "MODIFY_CQ";
+
+	case MLX5_CMD_OP_CREATE_QP:
+		return "CREATE_QP";
+
+	case MLX5_CMD_OP_DESTROY_QP:
+		return "DESTROY_QP";
+
+	case MLX5_CMD_OP_RST2INIT_QP:
+		return "RST2INIT_QP";
+
+	case MLX5_CMD_OP_INIT2RTR_QP:
+		return "INIT2RTR_QP";
+
+	case MLX5_CMD_OP_RTR2RTS_QP:
+		return "RTR2RTS_QP";
+
+	case MLX5_CMD_OP_RTS2RTS_QP:
+		return "RTS2RTS_QP";
+
+	case MLX5_CMD_OP_SQERR2RTS_QP:
+		return "SQERR2RTS_QP";
+
+	case MLX5_CMD_OP_2ERR_QP:
+		return "2ERR_QP";
+
+	case MLX5_CMD_OP_RTS2SQD_QP:
+		return "RTS2SQD_QP";
+
+	case MLX5_CMD_OP_SQD2RTS_QP:
+		return "SQD2RTS_QP";
+
+	case MLX5_CMD_OP_2RST_QP:
+		return "2RST_QP";
+
+	case MLX5_CMD_OP_QUERY_QP:
+		return "QUERY_QP";
+
+	case MLX5_CMD_OP_CONF_SQP:
+		return "CONF_SQP";
+
+	case MLX5_CMD_OP_MAD_IFC:
+		return "MAD_IFC";
+
+	case MLX5_CMD_OP_INIT2INIT_QP:
+		return "INIT2INIT_QP";
+
+	case MLX5_CMD_OP_SUSPEND_QP:
+		return "SUSPEND_QP";
+
+	case MLX5_CMD_OP_UNSUSPEND_QP:
+		return "UNSUSPEND_QP";
+
+	case MLX5_CMD_OP_SQD2SQD_QP:
+		return "SQD2SQD_QP";
+
+	case MLX5_CMD_OP_ALLOC_QP_COUNTER_SET:
+		return "ALLOC_QP_COUNTER_SET";
+
+	case MLX5_CMD_OP_DEALLOC_QP_COUNTER_SET:
+		return "DEALLOC_QP_COUNTER_SET";
+
+	case MLX5_CMD_OP_QUERY_QP_COUNTER_SET:
+		return "QUERY_QP_COUNTER_SET";
+
+	case MLX5_CMD_OP_CREATE_PSV:
+		return "CREATE_PSV";
+
+	case MLX5_CMD_OP_DESTROY_PSV:
+		return "DESTROY_PSV";
+
+	case MLX5_CMD_OP_QUERY_PSV:
+		return "QUERY_PSV";
+
+	case MLX5_CMD_OP_QUERY_SIG_RULE_TABLE:
+		return "QUERY_SIG_RULE_TABLE";
+
+	case MLX5_CMD_OP_QUERY_BLOCK_SIZE_TABLE:
+		return "QUERY_BLOCK_SIZE_TABLE";
+
+	case MLX5_CMD_OP_CREATE_SRQ:
+		return "CREATE_SRQ";
+
+	case MLX5_CMD_OP_DESTROY_SRQ:
+		return "DESTROY_SRQ";
+
+	case MLX5_CMD_OP_QUERY_SRQ:
+		return "QUERY_SRQ";
+
+	case MLX5_CMD_OP_ARM_RQ:
+		return "ARM_RQ";
+
+	case MLX5_CMD_OP_RESIZE_SRQ:
+		return "RESIZE_SRQ";
+
+	case MLX5_CMD_OP_ALLOC_PD:
+		return "ALLOC_PD";
+
+	case MLX5_CMD_OP_DEALLOC_PD:
+		return "DEALLOC_PD";
+
+	case MLX5_CMD_OP_ALLOC_UAR:
+		return "ALLOC_UAR";
+
+	case MLX5_CMD_OP_DEALLOC_UAR:
+		return "DEALLOC_UAR";
+
+	case MLX5_CMD_OP_ATTACH_TO_MCG:
+		return "ATTACH_TO_MCG";
+
+	case MLX5_CMD_OP_DETACH_FROM_MCG:
+		return "DETACH_FROM_MCG";
+
+	case MLX5_CMD_OP_ALLOC_XRCD:
+		return "ALLOC_XRCD";
+
+	case MLX5_CMD_OP_DEALLOC_XRCD:
+		return "DEALLOC_XRCD";
+
+	case MLX5_CMD_OP_ACCESS_REG:
+		return "MLX5_CMD_OP_ACCESS_REG";
+
+	default: return "unknown command opcode";
+	}
+}
+
+static void dump_command(struct mlx5_core_dev *dev,
+			 struct mlx5_cmd_work_ent *ent, int input)
+{
+	u16 op = be16_to_cpu(((struct mlx5_inbox_hdr *)(ent->lay->in))->opcode);
+	struct mlx5_cmd_msg *msg = input ? ent->in : ent->out;
+	struct mlx5_cmd_mailbox *next = msg->next;
+	int data_only;
+	int offset = 0;
+	int dump_len;
+
+	data_only = !!(mlx5_core_debug_mask & (1 << MLX5_CMD_DATA));
+
+	if (data_only)
+		mlx5_core_dbg_mask(dev, 1 << MLX5_CMD_DATA,
+				   "dump command data %s(0x%x) %s\n",
+				   mlx5_command_str(op), op,
+				   input ? "INPUT" : "OUTPUT");
+	else
+		mlx5_core_dbg(dev, "dump command %s(0x%x) %s\n",
+			      mlx5_command_str(op), op,
+			      input ? "INPUT" : "OUTPUT");
+
+	if (data_only) {
+		if (input) {
+			dump_buf(ent->lay->in, sizeof(ent->lay->in), 1, offset);
+			offset += sizeof(ent->lay->in);
+		} else {
+			dump_buf(ent->lay->out, sizeof(ent->lay->out), 1, offset);
+			offset += sizeof(ent->lay->out);
+		}
+	} else {
+		dump_buf(ent->lay, sizeof(*ent->lay), 0, offset);
+		offset += sizeof(*ent->lay);
+	}
+
+	while (next && offset < msg->len) {
+		if (data_only) {
+			dump_len = min_t(int, MLX5_CMD_DATA_BLOCK_SIZE, msg->len - offset);
+			dump_buf(next->buf, dump_len, 1, offset);
+			offset += MLX5_CMD_DATA_BLOCK_SIZE;
+		} else {
+			mlx5_core_dbg(dev, "command block:\n");
+			dump_buf(next->buf, sizeof(struct mlx5_cmd_prot_block), 0, offset);
+			offset += sizeof(struct mlx5_cmd_prot_block);
+		}
+		next = next->next;
+	}
+
+	if (data_only)
+		pr_debug("\n");
+}
+
+static void cmd_work_handler(struct work_struct *work)
+{
+	struct mlx5_cmd_work_ent *ent = container_of(work, struct mlx5_cmd_work_ent, work);
+	struct mlx5_cmd *cmd = ent->cmd;
+	struct mlx5_core_dev *dev = container_of(cmd, struct mlx5_core_dev, cmd);
+	struct mlx5_cmd_layout *lay;
+	struct semaphore *sem;
+
+	sem = ent->page_queue ? &cmd->pages_sem : &cmd->sem;
+	down(sem);
+	if (!ent->page_queue) {
+		ent->idx = alloc_ent(cmd);
+		if (ent->idx < 0) {
+			mlx5_core_err(dev, "failed to allocate command entry\n");
+			up(sem);
+			return;
+		}
+	} else {
+		ent->idx = cmd->max_reg_cmds;
+	}
+
+	ent->token = alloc_token(cmd);
+	cmd->ent_arr[ent->idx] = ent;
+	lay = get_inst(cmd, ent->idx);
+	ent->lay = lay;
+	memset(lay, 0, sizeof(*lay));
+	memcpy(lay->in, ent->in->first.data, sizeof(lay->in));
+	if (ent->in->next)
+		lay->in_ptr = cpu_to_be64(ent->in->next->dma);
+	lay->inlen = cpu_to_be32(ent->in->len);
+	if (ent->out->next)
+		lay->out_ptr = cpu_to_be64(ent->out->next->dma);
+	lay->outlen = cpu_to_be32(ent->out->len);
+	lay->type = MLX5_PCI_CMD_XPORT;
+	lay->token = ent->token;
+	lay->status_own = CMD_OWNER_HW;
+	if (!cmd->checksum_disabled)
+		set_signature(ent);
+	dump_command(dev, ent, 1);
+	ktime_get_ts(&ent->ts1);
+
+	/* ring doorbell after the descriptor is valid */
+	wmb();
+	iowrite32be(1 << ent->idx, &dev->iseg->cmd_dbell);
+	mlx5_core_dbg(dev, "write 0x%x to command doorbell\n", 1 << ent->idx);
+	mmiowb();
+	if (cmd->mode == CMD_MODE_POLLING) {
+		poll_timeout(ent);
+		/* make sure we read the descriptor after ownership is SW */
+		rmb();
+		mlx5_cmd_comp_handler(dev, 1UL << ent->idx);
+	}
+}
+
+static const char *deliv_status_to_str(u8 status)
+{
+	switch (status) {
+	case MLX5_CMD_DELIVERY_STAT_OK:
+		return "no errors";
+	case MLX5_CMD_DELIVERY_STAT_SIGNAT_ERR:
+		return "signature error";
+	case MLX5_CMD_DELIVERY_STAT_TOK_ERR:
+		return "token error";
+	case MLX5_CMD_DELIVERY_STAT_BAD_BLK_NUM_ERR:
+		return "bad block number";
+	case MLX5_CMD_DELIVERY_STAT_OUT_PTR_ALIGN_ERR:
+		return "output pointer not aligned to block size";
+	case MLX5_CMD_DELIVERY_STAT_IN_PTR_ALIGN_ERR:
+		return "input pointer not aligned to block size";
+	case MLX5_CMD_DELIVERY_STAT_FW_ERR:
+		return "firmware internal error";
+	case MLX5_CMD_DELIVERY_STAT_IN_LENGTH_ERR:
+		return "command input length error";
+	case MLX5_CMD_DELIVERY_STAT_OUT_LENGTH_ERR:
+		return "command ouput length error";
+	case MLX5_CMD_DELIVERY_STAT_RES_FLD_NOT_CLR_ERR:
+		return "reserved fields not cleared";
+	case MLX5_CMD_DELIVERY_STAT_CMD_DESCR_ERR:
+		return "bad command descriptor type";
+	default:
+		return "unknown status code";
+	}
+}
+
+static u16 msg_to_opcode(struct mlx5_cmd_msg *in)
+{
+	struct mlx5_inbox_hdr *hdr = (struct mlx5_inbox_hdr *)(in->first.data);
+
+	return be16_to_cpu(hdr->opcode);
+}
+
+static int wait_func(struct mlx5_core_dev *dev, struct mlx5_cmd_work_ent *ent)
+{
+	unsigned long timeout = msecs_to_jiffies(MLX5_CMD_TIMEOUT_MSEC);
+	struct mlx5_cmd *cmd = &dev->cmd;
+	int err;
+
+	if (cmd->mode == CMD_MODE_POLLING) {
+		wait_for_completion(&ent->done);
+		err = ent->ret;
+	} else {
+		if (!wait_for_completion_timeout(&ent->done, timeout))
+			err = -ETIMEDOUT;
+		else
+			err = 0;
+	}
+	if (err == -ETIMEDOUT) {
+		mlx5_core_warn(dev, "%s(0x%x) timeout. Will cause a leak of a command resource\n",
+			       mlx5_command_str(msg_to_opcode(ent->in)),
+			       msg_to_opcode(ent->in));
+	}
+	mlx5_core_dbg(dev, "err %d, delivery status %s(%d)\n", err,
+		      deliv_status_to_str(ent->status), ent->status);
+
+	return err;
+}
+
+/*  Notes:
+ *    1. Callback functions may not sleep
+ *    2. page queue commands do not support asynchrous completion
+ */
+static int mlx5_cmd_invoke(struct mlx5_core_dev *dev, struct mlx5_cmd_msg *in,
+			   struct mlx5_cmd_msg *out, mlx5_cmd_cbk_t callback,
+			   void *context, int page_queue, u8 *status)
+{
+	struct mlx5_cmd *cmd = &dev->cmd;
+	struct mlx5_cmd_work_ent *ent;
+	ktime_t t1, t2, delta;
+	struct mlx5_cmd_stats *stats;
+	int err = 0;
+	s64 ds;
+	u16 op;
+
+	if (callback && page_queue)
+		return -EINVAL;
+
+	ent = alloc_cmd(cmd, in, out, callback, context, page_queue);
+	if (IS_ERR(ent))
+		return PTR_ERR(ent);
+
+	if (!callback)
+		init_completion(&ent->done);
+
+	INIT_WORK(&ent->work, cmd_work_handler);
+	if (page_queue) {
+		cmd_work_handler(&ent->work);
+	} else if (!queue_work(cmd->wq, &ent->work)) {
+		mlx5_core_warn(dev, "failed to queue work\n");
+		err