/*
 * Copyright (c) 2004, 2005 Topspin Communications.  All rights reserved.
 * Copyright (c) 2005, 2006, 2007, 2008 Mellanox Technologies.
 * All rights reserved.
 * Copyright (c) 2005, 2006, 2007 Cisco Systems, 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/sched.h>
#include <linux/pci.h>
#include <linux/errno.h>
#include <linux/kernel.h>
#include <linux/io.h>
#include <linux/slab.h>
#include <linux/mlx4/cmd.h>
#include <linux/mlx4/qp.h>
#include <linux/if_ether.h>
#include <linux/etherdevice.h>

#include "mlx4.h"
#include "fw.h"
#include "mlx4_stats.h"

#define MLX4_MAC_VALID          (1ull << 63)
#define MLX4_PF_COUNTERS_PER_PORT       2
#define MLX4_VF_COUNTERS_PER_PORT       1

struct mac_res {
        struct list_head list;
        u64 mac;
        int ref_count;
        u8 smac_index;
        u8 port;
};

struct vlan_res {
        struct list_head list;
        u16 vlan;
        int ref_count;
        int vlan_index;
        u8 port;
};

struct res_common {
        struct list_head        list;
        struct rb_node          node;
        u64                     res_id;
        int                     owner;
        int                     state;
        int                     from_state;
        int                     to_state;
        int                     removing;
        const char              *func_name;
};

enum {
        RES_ANY_BUSY = 1
};

struct res_gid {
        struct list_head        list;
        u8                      gid[16];
        enum mlx4_protocol      prot;
        enum mlx4_steer_type    steer;
        u64                     reg_id;
};

enum res_qp_states {
        RES_QP_BUSY = RES_ANY_BUSY,

        /* QP number was allocated */
        RES_QP_RESERVED,

        /* ICM memory for QP context was mapped */
        RES_QP_MAPPED,

        /* QP is in hw ownership */
        RES_QP_HW
};

struct res_qp {
        struct res_common       com;
        struct res_mtt         *mtt;
        struct res_cq          *rcq;
        struct res_cq          *scq;
        struct res_srq         *srq;
        struct list_head        mcg_list;
        spinlock_t              mcg_spl;
        int                     local_qpn;
        atomic_t                ref_count;
        u32                     qpc_flags;
        /* saved qp params before VST enforcement in order to restore on VGT */
        u8                      sched_queue;
        __be32                  param3;
        u8                      vlan_control;
        u8                      fvl_rx;
        u8                      pri_path_fl;
        u8                      vlan_index;
        u8                      feup;
};

enum res_mtt_states {
        RES_MTT_BUSY = RES_ANY_BUSY,
        RES_MTT_ALLOCATED,
};

static inline const char *mtt_states_str(enum res_mtt_states state)
{
        switch (state) {
        case RES_MTT_BUSY: return "RES_MTT_BUSY";
        case RES_MTT_ALLOCATED: return "RES_MTT_ALLOCATED";
        default: return "Unknown";
        }
}

struct res_mtt {
        struct res_common       com;
        int                     order;
        atomic_t                ref_count;
};

enum res_mpt_states {
        RES_MPT_BUSY = RES_ANY_BUSY,
        RES_MPT_RESERVED,
        RES_MPT_MAPPED,
        RES_MPT_HW,
};

struct res_mpt {
        struct res_common       com;
        struct res_mtt         *mtt;
        int                     key;
};

enum res_eq_states {
        RES_EQ_BUSY = RES_ANY_BUSY,
        RES_EQ_RESERVED,
        RES_EQ_HW,
};

struct res_eq {
        struct res_common       com;
        struct res_mtt         *mtt;
};

enum res_cq_states {
        RES_CQ_BUSY = RES_ANY_BUSY,
        RES_CQ_ALLOCATED,
        RES_CQ_HW,
};

struct res_cq {
        struct res_common       com;
        struct res_mtt         *mtt;
        atomic_t                ref_count;
};

enum res_srq_states {
        RES_SRQ_BUSY = RES_ANY_BUSY,
        RES_SRQ_ALLOCATED,
        RES_SRQ_HW,
};

struct res_srq {
        struct res_common       com;
        struct res_mtt         *mtt;
        struct res_cq          *cq;
        atomic_t                ref_count;
};

enum res_counter_states {
        RES_COUNTER_BUSY = RES_ANY_BUSY,
        RES_COUNTER_ALLOCATED,
};

struct res_counter {
        struct res_common       com;
        int                     port;
};

enum res_xrcdn_states {
        RES_XRCD_BUSY = RES_ANY_BUSY,
        RES_XRCD_ALLOCATED,
};

struct res_xrcdn {
        struct res_common       com;
        int                     port;
};

enum res_fs_rule_states {
        RES_FS_RULE_BUSY = RES_ANY_BUSY,
        RES_FS_RULE_ALLOCATED,
};

struct res_fs_rule {
        struct res_common       com;
        int                     qpn;
        /* VF DMFS mbox with port flipped */
        void                    *mirr_mbox;
        /* > 0 --> apply mirror when getting into HA mode      */
        /* = 0 --> un-apply mirror when getting out of HA mode */
        u32                     mirr_mbox_size;
        struct list_head        mirr_list;
        u64                     mirr_rule_id;
};

static void *res_tracker_lookup(struct rb_root *root, u64 res_id)
{
        struct rb_node *node = root->rb_node;

        while (node) {
                struct res_common *res = rb_entry(node, struct res_common,
                                                  node);

                if (res_id < res->res_id)
                        node = node->rb_left;
                else if (res_id > res->res_id)
                        node = node->rb_right;
                else
                        return res;
        }
        return NULL;
}

static int res_tracker_insert(struct rb_root *root, struct res_common *res)
{
        struct rb_node **new = &(root->rb_node), *parent = NULL;

        /* Figure out where to put new node */
        while (*new) {
                struct res_common *this = rb_entry(*new, struct res_common,
                                                   node);

                parent = *new;
                if (res->res_id < this->res_id)
                        new = &((*new)->rb_left);
                else if (res->res_id > this->res_id)
                        new = &((*new)->rb_right);
                else
                        return -EEXIST;
        }

        /* Add new node and rebalance tree. */
        rb_link_node(&res->node, parent, new);
        rb_insert_color(&res->node, root);

        return 0;
}

enum qp_transition {
        QP_TRANS_INIT2RTR,
        QP_TRANS_RTR2RTS,
        QP_TRANS_RTS2RTS,
        QP_TRANS_SQERR2RTS,
        QP_TRANS_SQD2SQD,
        QP_TRANS_SQD2RTS
};

/* For Debug uses */
static const char *resource_str(enum mlx4_resource rt)
{
        switch (rt) {
        case RES_QP: return "RES_QP";
        case RES_CQ: return "RES_CQ";
        case RES_SRQ: return "RES_SRQ";
        case RES_MPT: return "RES_MPT";
        case RES_MTT: return "RES_MTT";
        case RES_MAC: return  "RES_MAC";
        case RES_VLAN: return  "RES_VLAN";
        case RES_EQ: return "RES_EQ";
        case RES_COUNTER: return "RES_COUNTER";
        case RES_FS_RULE: return "RES_FS_RULE";
        case RES_XRCD: return "RES_XRCD";
        default: return "Unknown resource type !!!";
        }
}

static void rem_slave_vlans(struct mlx4_dev *dev, int slave);
static inline int mlx4_grant_resource(struct mlx4_dev *dev, int slave,
                                      enum mlx4_resource res_type, int count,
                                      int port)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct resource_allocator *res_alloc =
                &priv->mfunc.master.res_tracker.res_alloc[res_type];
        int err = -EDQUOT;
        int allocated, free, reserved, guaranteed, from_free;
        int from_rsvd;

        if (slave > dev->persist->num_vfs)
                return -EINVAL;

        spin_lock(&res_alloc->alloc_lock);
        allocated = (port > 0) ?
                res_alloc->allocated[(port - 1) *
                (dev->persist->num_vfs + 1) + slave] :
                res_alloc->allocated[slave];
        free = (port > 0) ? res_alloc->res_port_free[port - 1] :
                res_alloc->res_free;
        reserved = (port > 0) ? res_alloc->res_port_rsvd[port - 1] :
                res_alloc->res_reserved;
        guaranteed = res_alloc->guaranteed[slave];

        if (allocated + count > res_alloc->quota[slave]) {
                mlx4_warn(dev, "VF %d port %d res %s: quota exceeded, count %d alloc %d quota %d\n",
                          slave, port, resource_str(res_type), count,
                          allocated, res_alloc->quota[slave]);
                goto out;
        }

        if (allocated + count <= guaranteed) {
                err = 0;
                from_rsvd = count;
        } else {
                /* portion may need to be obtained from free area */
                if (guaranteed - allocated > 0)
                        from_free = count - (guaranteed - allocated);
                else
                        from_free = count;

                from_rsvd = count - from_free;

                if (free - from_free >= reserved)
                        err = 0;
                else
                        mlx4_warn(dev, "VF %d port %d res %s: free pool empty, free %d from_free %d rsvd %d\n",
                                  slave, port, resource_str(res_type), free,
                                  from_free, reserved);
        }

        if (!err) {
                /* grant the request */
                if (port > 0) {
                        res_alloc->allocated[(port - 1) *
                        (dev->persist->num_vfs + 1) + slave] += count;
                        res_alloc->res_port_free[port - 1] -= count;
                        res_alloc->res_port_rsvd[port - 1] -= from_rsvd;
                } else {
                        res_alloc->allocated[slave] += count;
                        res_alloc->res_free -= count;
                        res_alloc->res_reserved -= from_rsvd;
                }
        }

out:
        spin_unlock(&res_alloc->alloc_lock);
        return err;
}

static inline void mlx4_release_resource(struct mlx4_dev *dev, int slave,
                                    enum mlx4_resource res_type, int count,
                                    int port)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct resource_allocator *res_alloc =
                &priv->mfunc.master.res_tracker.res_alloc[res_type];
        int allocated, guaranteed, from_rsvd;

        if (slave > dev->persist->num_vfs)
                return;

        spin_lock(&res_alloc->alloc_lock);

        allocated = (port > 0) ?
                res_alloc->allocated[(port - 1) *
                (dev->persist->num_vfs + 1) + slave] :
                res_alloc->allocated[slave];
        guaranteed = res_alloc->guaranteed[slave];

        if (allocated - count >= guaranteed) {
                from_rsvd = 0;
        } else {
                /* portion may need to be returned to reserved area */
                if (allocated - guaranteed > 0)
                        from_rsvd = count - (allocated - guaranteed);
                else
                        from_rsvd = count;
        }

        if (port > 0) {
                res_alloc->allocated[(port - 1) *
                (dev->persist->num_vfs + 1) + slave] -= count;
                res_alloc->res_port_free[port - 1] += count;
                res_alloc->res_port_rsvd[port - 1] += from_rsvd;
        } else {
                res_alloc->allocated[slave] -= count;
                res_alloc->res_free += count;
                res_alloc->res_reserved += from_rsvd;
        }

        spin_unlock(&res_alloc->alloc_lock);
        return;
}

static inline void initialize_res_quotas(struct mlx4_dev *dev,
                                         struct resource_allocator *res_alloc,
                                         enum mlx4_resource res_type,
                                         int vf, int num_instances)
{
        res_alloc->guaranteed[vf] = num_instances /
                                    (2 * (dev->persist->num_vfs + 1));
        res_alloc->quota[vf] = (num_instances / 2) + res_alloc->guaranteed[vf];
        if (vf == mlx4_master_func_num(dev)) {
                res_alloc->res_free = num_instances;
                if (res_type == RES_MTT) {
                        /* reserved mtts will be taken out of the PF allocation */
                        res_alloc->res_free += dev->caps.reserved_mtts;
                        res_alloc->guaranteed[vf] += dev->caps.reserved_mtts;
                        res_alloc->quota[vf] += dev->caps.reserved_mtts;
                }
        }
}

void mlx4_init_quotas(struct mlx4_dev *dev)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        int pf;

        /* quotas for VFs are initialized in mlx4_slave_cap */
        if (mlx4_is_slave(dev))
                return;

        if (!mlx4_is_mfunc(dev)) {
                dev->quotas.qp = dev->caps.num_qps - dev->caps.reserved_qps -
                        mlx4_num_reserved_sqps(dev);
                dev->quotas.cq = dev->caps.num_cqs - dev->caps.reserved_cqs;
                dev->quotas.srq = dev->caps.num_srqs - dev->caps.reserved_srqs;
                dev->quotas.mtt = dev->caps.num_mtts - dev->caps.reserved_mtts;
                dev->quotas.mpt = dev->caps.num_mpts - dev->caps.reserved_mrws;
                return;
        }

        pf = mlx4_master_func_num(dev);
        dev->quotas.qp =
                priv->mfunc.master.res_tracker.res_alloc[RES_QP].quota[pf];
        dev->quotas.cq =
                priv->mfunc.master.res_tracker.res_alloc[RES_CQ].quota[pf];
        dev->quotas.srq =
                priv->mfunc.master.res_tracker.res_alloc[RES_SRQ].quota[pf];
        dev->quotas.mtt =
                priv->mfunc.master.res_tracker.res_alloc[RES_MTT].quota[pf];
        dev->quotas.mpt =
                priv->mfunc.master.res_tracker.res_alloc[RES_MPT].quota[pf];
}

static int
mlx4_calc_res_counter_guaranteed(struct mlx4_dev *dev,
                                 struct resource_allocator *res_alloc,
                                 int vf)
{
        struct mlx4_active_ports actv_ports;
        int ports, counters_guaranteed;

        /* For master, only allocate according to the number of phys ports */
        if (vf == mlx4_master_func_num(dev))
                return MLX4_PF_COUNTERS_PER_PORT * dev->caps.num_ports;

        /* calculate real number of ports for the VF */
        actv_ports = mlx4_get_active_ports(dev, vf);
        ports = bitmap_weight(actv_ports.ports, dev->caps.num_ports);
        counters_guaranteed = ports * MLX4_VF_COUNTERS_PER_PORT;

        /* If we do not have enough counters for this VF, do not
         * allocate any for it. '-1' to reduce the sink counter.
         */
        if ((res_alloc->res_reserved + counters_guaranteed) >
            (dev->caps.max_counters - 1))
                return 0;

        return counters_guaranteed;
}

int mlx4_init_resource_tracker(struct mlx4_dev *dev)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        int i, j;
        int t;

        priv->mfunc.master.res_tracker.slave_list =
                kcalloc(dev->num_slaves, sizeof(struct slave_list),
                        GFP_KERNEL);
        if (!priv->mfunc.master.res_tracker.slave_list)
                return -ENOMEM;

        for (i = 0 ; i < dev->num_slaves; i++) {
                for (t = 0; t < MLX4_NUM_OF_RESOURCE_TYPE; ++t)
                        INIT_LIST_HEAD(&priv->mfunc.master.res_tracker.
                                       slave_list[i].res_list[t]);
                mutex_init(&priv->mfunc.master.res_tracker.slave_list[i].mutex);
        }

        mlx4_dbg(dev, "Started init_resource_tracker: %ld slaves\n",
                 dev->num_slaves);
        for (i = 0 ; i < MLX4_NUM_OF_RESOURCE_TYPE; i++)
                priv->mfunc.master.res_tracker.res_tree[i] = RB_ROOT;

        for (i = 0; i < MLX4_NUM_OF_RESOURCE_TYPE; i++) {
                struct resource_allocator *res_alloc =
                        &priv->mfunc.master.res_tracker.res_alloc[i];
                res_alloc->quota = kmalloc_array(dev->persist->num_vfs + 1,
                                                 sizeof(int),
                                                 GFP_KERNEL);
                res_alloc->guaranteed = kmalloc_array(dev->persist->num_vfs + 1,
                                                      sizeof(int),
                                                      GFP_KERNEL);
                if (i == RES_MAC || i == RES_VLAN)
                        res_alloc->allocated =
                                kcalloc(MLX4_MAX_PORTS *
                                                (dev->persist->num_vfs + 1),
                                        sizeof(int), GFP_KERNEL);
                else
                        res_alloc->allocated =
                                kcalloc(dev->persist->num_vfs + 1,
                                        sizeof(int), GFP_KERNEL);
                /* Reduce the sink counter */
                if (i == RES_COUNTER)
                        res_alloc->res_free = dev->caps.max_counters - 1;

                if (!res_alloc->quota || !res_alloc->guaranteed ||
                    !res_alloc->allocated)
                        goto no_mem_err;

                spin_lock_init(&res_alloc->alloc_lock);
                for (t = 0; t < dev->persist->num_vfs + 1; t++) {
                        struct mlx4_active_ports actv_ports =
                                mlx4_get_active_ports(dev, t);
                        switch (i) {
                        case RES_QP:
                                initialize_res_quotas(dev, res_alloc, RES_QP,
                                                      t, dev->caps.num_qps -
                                                      dev->caps.reserved_qps -
                                                      mlx4_num_reserved_sqps(dev));
                                break;
                        case RES_CQ:
                                initialize_res_quotas(dev, res_alloc, RES_CQ,
                                                      t, dev->caps.num_cqs -
                                                      dev->caps.reserved_cqs);
                                break;
                        case RES_SRQ:
                                initialize_res_quotas(dev, res_alloc, RES_SRQ,
                                                      t, dev->caps.num_srqs -
                                                      dev->caps.reserved_srqs);
                                break;
                        case RES_MPT:
                                initialize_res_quotas(dev, res_alloc, RES_MPT,
                                                      t, dev->caps.num_mpts -
                                                      dev->caps.reserved_mrws);
                                break;
                        case RES_MTT:
                                initialize_res_quotas(dev, res_alloc, RES_MTT,
                                                      t, dev->caps.num_mtts -
                                                      dev->caps.reserved_mtts);
                                break;
                        case RES_MAC:
                                if (t == mlx4_master_func_num(dev)) {
                                        int max_vfs_pport = 0;
                                        /* Calculate the max vfs per port for */
                                        /* both ports.                        */
                                        for (j = 0; j < dev->caps.num_ports;
                                             j++) {
                                                struct mlx4_slaves_pport slaves_pport =
                                                        mlx4_phys_to_slaves_pport(dev, j + 1);
                                                unsigned current_slaves =
                                                        bitmap_weight(slaves_pport.slaves,
                                                                      dev->caps.num_ports) - 1;
                                                if (max_vfs_pport < current_slaves)
                                                        max_vfs_pport =
                                                                current_slaves;
                                        }
                                        res_alloc->quota[t] =
                                                MLX4_MAX_MAC_NUM -
                                                2 * max_vfs_pport;
                                        res_alloc->guaranteed[t] = 2;
                                        for (j = 0; j < MLX4_MAX_PORTS; j++)
                                                res_alloc->res_port_free[j] =
                                                        MLX4_MAX_MAC_NUM;
                                } else {
                                        res_alloc->quota[t] = MLX4_MAX_MAC_NUM;
                                        res_alloc->guaranteed[t] = 2;
                                }
                                break;
                        case RES_VLAN:
                                if (t == mlx4_master_func_num(dev)) {
                                        res_alloc->quota[t] = MLX4_MAX_VLAN_NUM;
                                        res_alloc->guaranteed[t] = MLX4_MAX_VLAN_NUM / 2;
                                        for (j = 0; j < MLX4_MAX_PORTS; j++)
                                                res_alloc->res_port_free[j] =
                                                        res_alloc->quota[t];
                                } else {
                                        res_alloc->quota[t] = MLX4_MAX_VLAN_NUM / 2;
                                        res_alloc->guaranteed[t] = 0;
                                }
                                break;
                        case RES_COUNTER:
                                res_alloc->quota[t] = dev->caps.max_counters;
                                res_alloc->guaranteed[t] =
                                        mlx4_calc_res_counter_guaranteed(dev, res_alloc, t);
                                break;
                        default:
                                break;
                        }
                        if (i == RES_MAC || i == RES_VLAN) {
                                for (j = 0; j < dev->caps.num_ports; j++)
                                        if (test_bit(j, actv_ports.ports))
                                                res_alloc->res_port_rsvd[j] +=
                                                        res_alloc->guaranteed[t];
                        } else {
                                res_alloc->res_reserved += res_alloc->guaranteed[t];
                        }
                }
        }
        spin_lock_init(&priv->mfunc.master.res_tracker.lock);
        return 0;

no_mem_err:
        for (i = 0; i < MLX4_NUM_OF_RESOURCE_TYPE; i++) {
                kfree(priv->mfunc.master.res_tracker.res_alloc[i].allocated);
                priv->mfunc.master.res_tracker.res_alloc[i].allocated = NULL;
                kfree(priv->mfunc.master.res_tracker.res_alloc[i].guaranteed);
                priv->mfunc.master.res_tracker.res_alloc[i].guaranteed = NULL;
                kfree(priv->mfunc.master.res_tracker.res_alloc[i].quota);
                priv->mfunc.master.res_tracker.res_alloc[i].quota = NULL;
        }
        return -ENOMEM;
}

void mlx4_free_resource_tracker(struct mlx4_dev *dev,
                                enum mlx4_res_tracker_free_type type)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        int i;

        if (priv->mfunc.master.res_tracker.slave_list) {
                if (type != RES_TR_FREE_STRUCTS_ONLY) {
                        for (i = 0; i < dev->num_slaves; i++) {
                                if (type == RES_TR_FREE_ALL ||
                                    dev->caps.function != i)
                                        mlx4_delete_all_resources_for_slave(dev, i);
                        }
                        /* free master's vlans */
                        i = dev->caps.function;
                        mlx4_reset_roce_gids(dev, i);
                        mutex_lock(&priv->mfunc.master.res_tracker.slave_list[i].mutex);
                        rem_slave_vlans(dev, i);
                        mutex_unlock(&priv->mfunc.master.res_tracker.slave_list[i].mutex);
                }

                if (type != RES_TR_FREE_SLAVES_ONLY) {
                        for (i = 0; i < MLX4_NUM_OF_RESOURCE_TYPE; i++) {
                                kfree(priv->mfunc.master.res_tracker.res_alloc[i].allocated);
                                priv->mfunc.master.res_tracker.res_alloc[i].allocated = NULL;
                                kfree(priv->mfunc.master.res_tracker.res_alloc[i].guaranteed);
                                priv->mfunc.master.res_tracker.res_alloc[i].guaranteed = NULL;
                                kfree(priv->mfunc.master.res_tracker.res_alloc[i].quota);
                                priv->mfunc.master.res_tracker.res_alloc[i].quota = NULL;
                        }
                        kfree(priv->mfunc.master.res_tracker.slave_list);
                        priv->mfunc.master.res_tracker.slave_list = NULL;
                }
        }
}

static void update_pkey_index(struct mlx4_dev *dev, int slave,
                              struct mlx4_cmd_mailbox *inbox)
{
        u8 sched = *(u8 *)(inbox->buf + 64);
        u8 orig_index = *(u8 *)(inbox->buf + 35);
        u8 new_index;
        struct mlx4_priv *priv = mlx4_priv(dev);
        int port;

        port = (sched >> 6 & 1) + 1;

        new_index = priv->virt2phys_pkey[slave][port - 1][orig_index];
        *(u8 *)(inbox->buf + 35) = new_index;
}

static void update_gid(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *inbox,
                       u8 slave)
{
        struct mlx4_qp_context  *qp_ctx = inbox->buf + 8;
        enum mlx4_qp_optpar     optpar = be32_to_cpu(*(__be32 *) inbox->buf);
        u32                     ts = (be32_to_cpu(qp_ctx->flags) >> 16) & 0xff;
        int port;

        if (MLX4_QP_ST_UD == ts) {
                port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
                if (mlx4_is_eth(dev, port))
                        qp_ctx->pri_path.mgid_index =
                                mlx4_get_base_gid_ix(dev, slave, port) | 0x80;
                else
                        qp_ctx->pri_path.mgid_index = slave | 0x80;

        } else if (MLX4_QP_ST_RC == ts || MLX4_QP_ST_XRC == ts || MLX4_QP_ST_UC == ts) {
                if (optpar & MLX4_QP_OPTPAR_PRIMARY_ADDR_PATH) {
                        port = (qp_ctx->pri_path.sched_queue >> 6 & 1) + 1;
                        if (mlx4_is_eth(dev, port)) {
                                qp_ctx->pri_path.mgid_index +=
                                        mlx4_get_base_gid_ix(dev, slave, port);
                                qp_ctx->pri_path.mgid_index &= 0x7f;
                        } else {
                                qp_ctx->pri_path.mgid_index = slave & 0x7F;
                        }
                }
                if (optpar & MLX4_QP_OPTPAR_ALT_ADDR_PATH) {
                        port = (qp_ctx->alt_path.sched_queue >> 6 & 1) + 1;
                        if (mlx4_is_eth(dev, port)) {
                                qp_ctx->alt_path.mgid_index +=
                                        mlx4_get_base_gid_ix(dev, slave, port);
                                qp_ctx->alt_path.mgid_index &= 0x7f;
                        } else {
                                qp_ctx->alt_path.mgid_index = slave & 0x7F;
                        }
                }
        }
}

static int handle_counter(struct mlx4_dev *dev, struct mlx4_qp_context *qpc,
                          u8 slave, int port);

static int update_vport_qp_param(struct mlx4_dev *dev,
                                 struct mlx4_cmd_mailbox *inbox,
                                 u8 slave, u32 qpn)
{
        struct mlx4_qp_context  *qpc = inbox->buf + 8;
        struct mlx4_vport_oper_state *vp_oper;
        struct mlx4_priv *priv;
        u32 qp_type;
        int port, err = 0;

        port = (qpc->pri_path.sched_queue & 0x40) ? 2 : 1;
        priv = mlx4_priv(dev);
        vp_oper = &priv->mfunc.master.vf_oper[slave].vport[port];
        qp_type = (be32_to_cpu(qpc->flags) >> 16) & 0xff;

        err = handle_counter(dev, qpc, slave, port);
        if (err)
                goto out;

        if (MLX4_VGT != vp_oper->state.default_vlan) {
                /* the reserved QPs (special, proxy, tunnel)
                 * do not operate over vlans
                 */
                if (mlx4_is_qp_reserved(dev, qpn))
                        return 0;

                /* force strip vlan by clear vsd, MLX QP refers to Raw Ethernet */
                if (qp_type == MLX4_QP_ST_UD ||
                    (qp_type == MLX4_QP_ST_MLX && mlx4_is_eth(dev, port))) {
                        if (dev->caps.bmme_flags & MLX4_BMME_FLAG_VSD_INIT2RTR) {
                                *(__be32 *)inbox->buf =
                                        cpu_to_be32(be32_to_cpu(*(__be32 *)inbox->buf) |
                                        MLX4_QP_OPTPAR_VLAN_STRIPPING);
                                qpc->param3 &= ~cpu_to_be32(MLX4_STRIP_VLAN);
                        } else {
                                struct mlx4_update_qp_params params = {.flags = 0};

                                err = mlx4_update_qp(dev, qpn, MLX4_UPDATE_QP_VSD, &params);
                                if (err)
                                        goto out;
                        }
                }

                /* preserve IF_COUNTER flag */
                qpc->pri_path.vlan_control &=
                        MLX4_CTRL_ETH_SRC_CHECK_IF_COUNTER;
                if (vp_oper->state.link_state == IFLA_VF_LINK_STATE_DISABLE &&
                    dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_UPDATE_QP) {
                        qpc->pri_path.vlan_control |=
                                MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED |
                                MLX4_VLAN_CTRL_ETH_TX_BLOCK_PRIO_TAGGED |
                                MLX4_VLAN_CTRL_ETH_TX_BLOCK_UNTAGGED |
                                MLX4_VLAN_CTRL_ETH_RX_BLOCK_PRIO_TAGGED |
                                MLX4_VLAN_CTRL_ETH_RX_BLOCK_UNTAGGED |
                                MLX4_VLAN_CTRL_ETH_RX_BLOCK_TAGGED;
                } else if (0 != vp_oper->state.default_vlan) {
                        if (vp_oper->state.vlan_proto == htons(ETH_P_8021AD)) {
                                /* vst QinQ should block untagged on TX,
                                 * but cvlan is in payload and phv is set so
                                 * hw see it as untagged. Block tagged instead.
                                 */
                                qpc->pri_path.vlan_control |=
                                        MLX4_VLAN_CTRL_ETH_TX_BLOCK_PRIO_TAGGED |
                                        MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED |
                                        MLX4_VLAN_CTRL_ETH_RX_BLOCK_PRIO_TAGGED |
                                        MLX4_VLAN_CTRL_ETH_RX_BLOCK_UNTAGGED;
                        } else { /* vst 802.1Q */
                                qpc->pri_path.vlan_control |=
                                        MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED |
                                        MLX4_VLAN_CTRL_ETH_RX_BLOCK_PRIO_TAGGED |
                                        MLX4_VLAN_CTRL_ETH_RX_BLOCK_UNTAGGED;
                        }
                } else { /* priority tagged */
                        qpc->pri_path.vlan_control |=
                                MLX4_VLAN_CTRL_ETH_TX_BLOCK_TAGGED |
                                MLX4_VLAN_CTRL_ETH_RX_BLOCK_TAGGED;
                }

                qpc->pri_path.fvl_rx |= MLX4_FVL_RX_FORCE_ETH_VLAN;
                qpc->pri_path.vlan_index = vp_oper->vlan_idx;
                qpc->pri_path.fl |= MLX4_FL_ETH_HIDE_CQE_VLAN;
                if (vp_oper->state.vlan_proto == htons(ETH_P_8021AD))
                        qpc->pri_path.fl |= MLX4_FL_SV;
                else
                        qpc->pri_path.fl |= MLX4_FL_CV;
                qpc->pri_path.feup |= MLX4_FEUP_FORCE_ETH_UP | MLX4_FVL_FORCE_ETH_VLAN;
                qpc->pri_path.sched_queue &= 0xC7;
                qpc->pri_path.sched_queue |= (vp_oper->state.default_qos) << 3;
                qpc->qos_vport = vp_oper->state.qos_vport;
        }
        if (vp_oper->state.spoofchk) {
                qpc->pri_path.feup |= MLX4_FSM_FORCE_ETH_SRC_MAC;
                qpc->pri_path.grh_mylmc = (0x80 & qpc->pri_path.grh_mylmc) + vp_oper->mac_idx;
        }
out:
        return err;
}

static int mpt_mask(struct mlx4_dev *dev)
{
        return dev->caps.num_mpts - 1;
}

static const char *mlx4_resource_type_to_str(enum mlx4_resource t)
{
        switch (t) {
        case RES_QP:
                return "QP";
        case RES_CQ:
                return "CQ";
        case RES_SRQ:
                return "SRQ";
        case RES_XRCD:
                return "XRCD";
        case RES_MPT:
                return "MPT";
        case RES_MTT:
                return "MTT";
        case RES_MAC:
                return "MAC";
        case RES_VLAN:
                return "VLAN";
        case RES_COUNTER:
                return "COUNTER";
        case RES_FS_RULE:
                return "FS_RULE";
        case RES_EQ:
                return "EQ";
        default:
                return "INVALID RESOURCE";
        }
}

static void *find_res(struct mlx4_dev *dev, u64 res_id,
                      enum mlx4_resource type)
{
        struct mlx4_priv *priv = mlx4_priv(dev);

        return res_tracker_lookup(&priv->mfunc.master.res_tracker.res_tree[type],
                                  res_id);
}

static int _get_res(struct mlx4_dev *dev, int slave, u64 res_id,
                    enum mlx4_resource type,
                    void *res, const char *func_name)
{
        struct res_common *r;
        int err = 0;

        spin_lock_irq(mlx4_tlock(dev));
        r = find_res(dev, res_id, type);
        if (!r) {
                err = -ENONET;
                goto exit;
        }

        if (r->state == RES_ANY_BUSY) {
                mlx4_warn(dev,
                          "%s(%d) trying to get resource %llx of type %s, but it's already taken by %s\n",
                          func_name, slave, res_id, mlx4_resource_type_to_str(type),
                          r->func_name);
                err = -EBUSY;
                goto exit;
        }

        if (r->owner != slave) {
                err = -EPERM;
                goto exit;
        }

        r->from_state = r->state;
        r->state = RES_ANY_BUSY;
        r->func_name = func_name;

        if (res)
                *((struct res_common **)res) = r;

exit:
        spin_unlock_irq(mlx4_tlock(dev));
        return err;
}

#define get_res(dev, slave, res_id, type, res) \
        _get_res((dev), (slave), (res_id), (type), (res), __func__)

int mlx4_get_slave_from_resource_id(struct mlx4_dev *dev,
                                    enum mlx4_resource type,
                                    u64 res_id, int *slave)
{

        struct res_common *r;
        int err = -ENOENT;
        int id = res_id;

        if (type == RES_QP)
                id &= 0x7fffff;
        spin_lock(mlx4_tlock(dev));

        r = find_res(dev, id, type);
        if (r) {
                *slave = r->owner;
                err = 0;
        }
        spin_unlock(mlx4_tlock(dev));

        return err;
}

static void put_res(struct mlx4_dev *dev, int slave, u64 res_id,
                    enum mlx4_resource type)
{
        struct res_common *r;

        spin_lock_irq(mlx4_tlock(dev));
        r = find_res(dev, res_id, type);
        if (r) {
                r->state = r->from_state;
                r->func_name = "";
        }
        spin_unlock_irq(mlx4_tlock(dev));
}

static int counter_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
                             u64 in_param, u64 *out_param, int port);

static int handle_existing_counter(struct mlx4_dev *dev, u8 slave, int port,
                                   int counter_index)
{
        struct res_common *r;
        struct res_counter *counter;
        int ret = 0;

        if (counter_index == MLX4_SINK_COUNTER_INDEX(dev))
                return ret;

        spin_lock_irq(mlx4_tlock(dev));
        r = find_res(dev, counter_index, RES_COUNTER);
        if (!r || r->owner != slave) {
                ret = -EINVAL;
        } else {
                counter = container_of(r, struct res_counter, com);
                if (!counter->port)
                        counter->port = port;
        }

        spin_unlock_irq(mlx4_tlock(dev));
        return ret;
}

static int handle_unexisting_counter(struct mlx4_dev *dev,
                                     struct mlx4_qp_context *qpc, u8 slave,
                                     int port)

{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_common *tmp;
        struct res_counter *counter;
        u64 counter_idx = MLX4_SINK_COUNTER_INDEX(dev);
        int err = 0;

        spin_lock_irq(mlx4_tlock(dev));
        list_for_each_entry(tmp,
                            &tracker->slave_list[slave].res_list[RES_COUNTER],
                            list) {
                counter = container_of(tmp, struct res_counter, com);
                if (port == counter->port) {
                        qpc->pri_path.counter_index  = counter->com.res_id;
                        spin_unlock_irq(mlx4_tlock(dev));
                        return 0;
                }
        }
        spin_unlock_irq(mlx4_tlock(dev));

        /* No existing counter, need to allocate a new counter */
        err = counter_alloc_res(dev, slave, RES_OP_RESERVE, 0, 0, &counter_idx,
                                port);
        if (err == -ENOENT) {
                err = 0;
        } else if (err && err != -ENOSPC) {
                mlx4_err(dev, "%s: failed to create new counter for slave %d err %d\n",
                         __func__, slave, err);
        } else {
                qpc->pri_path.counter_index = counter_idx;
                mlx4_dbg(dev, "%s: alloc new counter for slave %d index %d\n",
                         __func__, slave, qpc->pri_path.counter_index);
                err = 0;
        }

        return err;
}

static int handle_counter(struct mlx4_dev *dev, struct mlx4_qp_context *qpc,
                          u8 slave, int port)
{
        if (qpc->pri_path.counter_index != MLX4_SINK_COUNTER_INDEX(dev))
                return handle_existing_counter(dev, slave, port,
                                               qpc->pri_path.counter_index);

        return handle_unexisting_counter(dev, qpc, slave, port);
}

static struct res_common *alloc_qp_tr(int id)
{
        struct res_qp *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_QP_RESERVED;
        ret->local_qpn = id;
        INIT_LIST_HEAD(&ret->mcg_list);
        spin_lock_init(&ret->mcg_spl);
        atomic_set(&ret->ref_count, 0);

        return &ret->com;
}

static struct res_common *alloc_mtt_tr(int id, int order)
{
        struct res_mtt *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->order = order;
        ret->com.state = RES_MTT_ALLOCATED;
        atomic_set(&ret->ref_count, 0);

        return &ret->com;
}

static struct res_common *alloc_mpt_tr(int id, int key)
{
        struct res_mpt *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_MPT_RESERVED;
        ret->key = key;

        return &ret->com;
}

static struct res_common *alloc_eq_tr(int id)
{
        struct res_eq *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_EQ_RESERVED;

        return &ret->com;
}

static struct res_common *alloc_cq_tr(int id)
{
        struct res_cq *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_CQ_ALLOCATED;
        atomic_set(&ret->ref_count, 0);

        return &ret->com;
}

static struct res_common *alloc_srq_tr(int id)
{
        struct res_srq *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_SRQ_ALLOCATED;
        atomic_set(&ret->ref_count, 0);

        return &ret->com;
}

static struct res_common *alloc_counter_tr(int id, int port)
{
        struct res_counter *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_COUNTER_ALLOCATED;
        ret->port = port;

        return &ret->com;
}

static struct res_common *alloc_xrcdn_tr(int id)
{
        struct res_xrcdn *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_XRCD_ALLOCATED;

        return &ret->com;
}

static struct res_common *alloc_fs_rule_tr(u64 id, int qpn)
{
        struct res_fs_rule *ret;

        ret = kzalloc(sizeof(*ret), GFP_KERNEL);
        if (!ret)
                return NULL;

        ret->com.res_id = id;
        ret->com.state = RES_FS_RULE_ALLOCATED;
        ret->qpn = qpn;
        return &ret->com;
}

static struct res_common *alloc_tr(u64 id, enum mlx4_resource type, int slave,
                                   int extra)
{
        struct res_common *ret;

        switch (type) {
        case RES_QP:
                ret = alloc_qp_tr(id);
                break;
        case RES_MPT:
                ret = alloc_mpt_tr(id, extra);
                break;
        case RES_MTT:
                ret = alloc_mtt_tr(id, extra);
                break;
        case RES_EQ:
                ret = alloc_eq_tr(id);
                break;
        case RES_CQ:
                ret = alloc_cq_tr(id);
                break;
        case RES_SRQ:
                ret = alloc_srq_tr(id);
                break;
        case RES_MAC:
                pr_err("implementation missing\n");
                return NULL;
        case RES_COUNTER:
                ret = alloc_counter_tr(id, extra);
                break;
        case RES_XRCD:
                ret = alloc_xrcdn_tr(id);
                break;
        case RES_FS_RULE:
                ret = alloc_fs_rule_tr(id, extra);
                break;
        default:
                return NULL;
        }
        if (ret)
                ret->owner = slave;

        return ret;
}

int mlx4_calc_vf_counters(struct mlx4_dev *dev, int slave, int port,
                          struct mlx4_counter *data)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_common *tmp;
        struct res_counter *counter;
        int *counters_arr;
        int i = 0, err = 0;

        memset(data, 0, sizeof(*data));

        counters_arr = kmalloc_array(dev->caps.max_counters,
                                     sizeof(*counters_arr), GFP_KERNEL);
        if (!counters_arr)
                return -ENOMEM;

        spin_lock_irq(mlx4_tlock(dev));
        list_for_each_entry(tmp,
                            &tracker->slave_list[slave].res_list[RES_COUNTER],
                            list) {
                counter = container_of(tmp, struct res_counter, com);
                if (counter->port == port) {
                        counters_arr[i] = (int)tmp->res_id;
                        i++;
                }
        }
        spin_unlock_irq(mlx4_tlock(dev));
        counters_arr[i] = -1;

        i = 0;

        while (counters_arr[i] != -1) {
                err = mlx4_get_counter_stats(dev, counters_arr[i], data,
                                             0);
                if (err) {
                        memset(data, 0, sizeof(*data));
                        goto table_changed;
                }
                i++;
        }

table_changed:
        kfree(counters_arr);
        return 0;
}

static int add_res_range(struct mlx4_dev *dev, int slave, u64 base, int count,
                         enum mlx4_resource type, int extra)
{
        int i;
        int err;
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct res_common **res_arr;
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct rb_root *root = &tracker->res_tree[type];

        res_arr = kcalloc(count, sizeof(*res_arr), GFP_KERNEL);
        if (!res_arr)
                return -ENOMEM;

        for (i = 0; i < count; ++i) {
                res_arr[i] = alloc_tr(base + i, type, slave, extra);
                if (!res_arr[i]) {
                        for (--i; i >= 0; --i)
                                kfree(res_arr[i]);

                        kfree(res_arr);
                        return -ENOMEM;
                }
        }

        spin_lock_irq(mlx4_tlock(dev));
        for (i = 0; i < count; ++i) {
                if (find_res(dev, base + i, type)) {
                        err = -EEXIST;
                        goto undo;
                }
                err = res_tracker_insert(root, res_arr[i]);
                if (err)
                        goto undo;
                list_add_tail(&res_arr[i]->list,
                              &tracker->slave_list[slave].res_list[type]);
        }
        spin_unlock_irq(mlx4_tlock(dev));
        kfree(res_arr);

        return 0;

undo:
        for (--i; i >= 0; --i) {
                rb_erase(&res_arr[i]->node, root);
                list_del_init(&res_arr[i]->list);
        }

        spin_unlock_irq(mlx4_tlock(dev));

        for (i = 0; i < count; ++i)
                kfree(res_arr[i]);

        kfree(res_arr);

        return err;
}

static int remove_qp_ok(struct res_qp *res)
{
        if (res->com.state == RES_QP_BUSY || atomic_read(&res->ref_count) ||
            !list_empty(&res->mcg_list)) {
                pr_err("resource tracker: fail to remove qp, state %d, ref_count %d\n",
                       res->com.state, atomic_read(&res->ref_count));
                return -EBUSY;
        } else if (res->com.state != RES_QP_RESERVED) {
                return -EPERM;
        }

        return 0;
}

static int remove_mtt_ok(struct res_mtt *res, int order)
{
        if (res->com.state == RES_MTT_BUSY ||
            atomic_read(&res->ref_count)) {
                pr_devel("%s-%d: state %s, ref_count %d\n",
                         __func__, __LINE__,
                         mtt_states_str(res->com.state),
                         atomic_read(&res->ref_count));
                return -EBUSY;
        } else if (res->com.state != RES_MTT_ALLOCATED)
                return -EPERM;
        else if (res->order != order)
                return -EINVAL;

        return 0;
}

static int remove_mpt_ok(struct res_mpt *res)
{
        if (res->com.state == RES_MPT_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_MPT_RESERVED)
                return -EPERM;

        return 0;
}

static int remove_eq_ok(struct res_eq *res)
{
        if (res->com.state == RES_MPT_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_MPT_RESERVED)
                return -EPERM;

        return 0;
}

static int remove_counter_ok(struct res_counter *res)
{
        if (res->com.state == RES_COUNTER_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_COUNTER_ALLOCATED)
                return -EPERM;

        return 0;
}

static int remove_xrcdn_ok(struct res_xrcdn *res)
{
        if (res->com.state == RES_XRCD_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_XRCD_ALLOCATED)
                return -EPERM;

        return 0;
}

static int remove_fs_rule_ok(struct res_fs_rule *res)
{
        if (res->com.state == RES_FS_RULE_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_FS_RULE_ALLOCATED)
                return -EPERM;

        return 0;
}

static int remove_cq_ok(struct res_cq *res)
{
        if (res->com.state == RES_CQ_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_CQ_ALLOCATED)
                return -EPERM;

        return 0;
}

static int remove_srq_ok(struct res_srq *res)
{
        if (res->com.state == RES_SRQ_BUSY)
                return -EBUSY;
        else if (res->com.state != RES_SRQ_ALLOCATED)
                return -EPERM;

        return 0;
}

static int remove_ok(struct res_common *res, enum mlx4_resource type, int extra)
{
        switch (type) {
        case RES_QP:
                return remove_qp_ok((struct res_qp *)res);
        case RES_CQ:
                return remove_cq_ok((struct res_cq *)res);
        case RES_SRQ:
                return remove_srq_ok((struct res_srq *)res);
        case RES_MPT:
                return remove_mpt_ok((struct res_mpt *)res);
        case RES_MTT:
                return remove_mtt_ok((struct res_mtt *)res, extra);
        case RES_MAC:
                return -EOPNOTSUPP;
        case RES_EQ:
                return remove_eq_ok((struct res_eq *)res);
        case RES_COUNTER:
                return remove_counter_ok((struct res_counter *)res);
        case RES_XRCD:
                return remove_xrcdn_ok((struct res_xrcdn *)res);
        case RES_FS_RULE:
                return remove_fs_rule_ok((struct res_fs_rule *)res);
        default:
                return -EINVAL;
        }
}

static int rem_res_range(struct mlx4_dev *dev, int slave, u64 base, int count,
                         enum mlx4_resource type, int extra)
{
        u64 i;
        int err;
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_common *r;

        spin_lock_irq(mlx4_tlock(dev));
        for (i = base; i < base + count; ++i) {
                r = res_tracker_lookup(&tracker->res_tree[type], i);
                if (!r) {
                        err = -ENOENT;
                        goto out;
                }
                if (r->owner != slave) {
                        err = -EPERM;
                        goto out;
                }
                err = remove_ok(r, type, extra);
                if (err)
                        goto out;
        }

        for (i = base; i < base + count; ++i) {
                r = res_tracker_lookup(&tracker->res_tree[type], i);
                rb_erase(&r->node, &tracker->res_tree[type]);
                list_del(&r->list);
                kfree(r);
        }
        err = 0;

out:
        spin_unlock_irq(mlx4_tlock(dev));

        return err;
}

static int qp_res_start_move_to(struct mlx4_dev *dev, int slave, int qpn,
                                enum res_qp_states state, struct res_qp **qp,
                                int alloc)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_qp *r;
        int err = 0;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[RES_QP], qpn);
        if (!r)
                err = -ENOENT;
        else if (r->com.owner != slave)
                err = -EPERM;
        else {
                switch (state) {
                case RES_QP_BUSY:
                        mlx4_dbg(dev, "%s: failed RES_QP, 0x%llx\n",
                                 __func__, r->com.res_id);
                        err = -EBUSY;
                        break;

                case RES_QP_RESERVED:
                        if (r->com.state == RES_QP_MAPPED && !alloc)
                                break;

                        mlx4_dbg(dev, "failed RES_QP, 0x%llx\n", r->com.res_id);
                        err = -EINVAL;
                        break;

                case RES_QP_MAPPED:
                        if ((r->com.state == RES_QP_RESERVED && alloc) ||
                            r->com.state == RES_QP_HW)
                                break;
                        else {
                                mlx4_dbg(dev, "failed RES_QP, 0x%llx\n",
                                          r->com.res_id);
                                err = -EINVAL;
                        }

                        break;

                case RES_QP_HW:
                        if (r->com.state != RES_QP_MAPPED)
                                err = -EINVAL;
                        break;
                default:
                        err = -EINVAL;
                }

                if (!err) {
                        r->com.from_state = r->com.state;
                        r->com.to_state = state;
                        r->com.state = RES_QP_BUSY;
                        if (qp)
                                *qp = r;
                }
        }

        spin_unlock_irq(mlx4_tlock(dev));

        return err;
}

static int mr_res_start_move_to(struct mlx4_dev *dev, int slave, int index,
                                enum res_mpt_states state, struct res_mpt **mpt)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_mpt *r;
        int err = 0;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[RES_MPT], index);
        if (!r)
                err = -ENOENT;
        else if (r->com.owner != slave)
                err = -EPERM;
        else {
                switch (state) {
                case RES_MPT_BUSY:
                        err = -EINVAL;
                        break;

                case RES_MPT_RESERVED:
                        if (r->com.state != RES_MPT_MAPPED)
                                err = -EINVAL;
                        break;

                case RES_MPT_MAPPED:
                        if (r->com.state != RES_MPT_RESERVED &&
                            r->com.state != RES_MPT_HW)
                                err = -EINVAL;
                        break;

                case RES_MPT_HW:
                        if (r->com.state != RES_MPT_MAPPED)
                                err = -EINVAL;
                        break;
                default:
                        err = -EINVAL;
                }

                if (!err) {
                        r->com.from_state = r->com.state;
                        r->com.to_state = state;
                        r->com.state = RES_MPT_BUSY;
                        if (mpt)
                                *mpt = r;
                }
        }

        spin_unlock_irq(mlx4_tlock(dev));

        return err;
}

static int eq_res_start_move_to(struct mlx4_dev *dev, int slave, int index,
                                enum res_eq_states state, struct res_eq **eq)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_eq *r;
        int err = 0;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[RES_EQ], index);
        if (!r)
                err = -ENOENT;
        else if (r->com.owner != slave)
                err = -EPERM;
        else {
                switch (state) {
                case RES_EQ_BUSY:
                        err = -EINVAL;
                        break;

                case RES_EQ_RESERVED:
                        if (r->com.state != RES_EQ_HW)
                                err = -EINVAL;
                        break;

                case RES_EQ_HW:
                        if (r->com.state != RES_EQ_RESERVED)
                                err = -EINVAL;
                        break;

                default:
                        err = -EINVAL;
                }

                if (!err) {
                        r->com.from_state = r->com.state;
                        r->com.to_state = state;
                        r->com.state = RES_EQ_BUSY;
                }
        }

        spin_unlock_irq(mlx4_tlock(dev));

        if (!err && eq)
                *eq = r;

        return err;
}

static int cq_res_start_move_to(struct mlx4_dev *dev, int slave, int cqn,
                                enum res_cq_states state, struct res_cq **cq)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_cq *r;
        int err;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[RES_CQ], cqn);
        if (!r) {
                err = -ENOENT;
        } else if (r->com.owner != slave) {
                err = -EPERM;
        } else if (state == RES_CQ_ALLOCATED) {
                if (r->com.state != RES_CQ_HW)
                        err = -EINVAL;
                else if (atomic_read(&r->ref_count))
                        err = -EBUSY;
                else
                        err = 0;
        } else if (state != RES_CQ_HW || r->com.state != RES_CQ_ALLOCATED) {
                err = -EINVAL;
        } else {
                err = 0;
        }

        if (!err) {
                r->com.from_state = r->com.state;
                r->com.to_state = state;
                r->com.state = RES_CQ_BUSY;
                if (cq)
                        *cq = r;
        }

        spin_unlock_irq(mlx4_tlock(dev));

        return err;
}

static int srq_res_start_move_to(struct mlx4_dev *dev, int slave, int index,
                                 enum res_srq_states state, struct res_srq **srq)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_srq *r;
        int err = 0;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[RES_SRQ], index);
        if (!r) {
                err = -ENOENT;
        } else if (r->com.owner != slave) {
                err = -EPERM;
        } else if (state == RES_SRQ_ALLOCATED) {
                if (r->com.state != RES_SRQ_HW)
                        err = -EINVAL;
                else if (atomic_read(&r->ref_count))
                        err = -EBUSY;
        } else if (state != RES_SRQ_HW || r->com.state != RES_SRQ_ALLOCATED) {
                err = -EINVAL;
        }

        if (!err) {
                r->com.from_state = r->com.state;
                r->com.to_state = state;
                r->com.state = RES_SRQ_BUSY;
                if (srq)
                        *srq = r;
        }

        spin_unlock_irq(mlx4_tlock(dev));

        return err;
}

static void res_abort_move(struct mlx4_dev *dev, int slave,
                           enum mlx4_resource type, int id)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_common *r;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[type], id);
        if (r && (r->owner == slave))
                r->state = r->from_state;
        spin_unlock_irq(mlx4_tlock(dev));
}

static void res_end_move(struct mlx4_dev *dev, int slave,
                         enum mlx4_resource type, int id)
{
        struct mlx4_priv *priv = mlx4_priv(dev);
        struct mlx4_resource_tracker *tracker = &priv->mfunc.master.res_tracker;
        struct res_common *r;

        spin_lock_irq(mlx4_tlock(dev));
        r = res_tracker_lookup(&tracker->res_tree[type], id);
        if (r && (r->owner == slave))
                r->state = r->to_state;
        spin_unlock_irq(mlx4_tlock(dev));
}

static int valid_reserved(struct mlx4_dev *dev, int slave, int qpn)
{
        return mlx4_is_qp_reserved(dev, qpn) &&
                (mlx4_is_master(dev) || mlx4_is_guest_proxy(dev, slave, qpn));
}

static int fw_reserved(struct mlx4_dev *dev, int qpn)
{
        return qpn < dev->caps.reserved_qps_cnt[MLX4_QP_REGION_FW];
}

static int qp_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
                        u64 in_param, u64 *out_param)
{
        int err;
        int count;
        int align;
        int base;
        int qpn;
        u8 flags;

        switch (op) {
        case RES_OP_RESERVE:
                count = get_param_l(&in_param) & 0xffffff;
                /* Turn off all unsupported QP allocation flags that the
                 * slave tries to set.
                 */
                flags = (get_param_l(&in_param) >> 24) & dev->caps.alloc_res_qp_mask;
                align = get_param_h(&in_param);
                err = mlx4_grant_resource(dev, slave, RES_QP, count, 0);
                if (err)
                        return err;

                err = __mlx4_qp_reserve_range(dev, count, align, &base, flags);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_QP, count, 0);
                        return err;
                }

                err = add_res_range(dev, slave, base, count, RES_QP, 0);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_QP, count, 0);
                        __mlx4_qp_release_range(dev, base, count);
                        return err;
                }
                set_param_l(out_param, base);
                break;
        case RES_OP_MAP_ICM:
                qpn = get_param_l(&in_param) & 0x7fffff;
                if (valid_reserved(dev, slave, qpn)) {
                        err = add_res_range(dev, slave, qpn, 1, RES_QP, 0);
                        if (err)
                                return err;
                }

                err = qp_res_start_move_to(dev, slave, qpn, RES_QP_MAPPED,
                                           NULL, 1);
                if (err)
                        return err;

                if (!fw_reserved(dev, qpn)) {
                        err = __mlx4_qp_alloc_icm(dev, qpn);
                        if (err) {
                                res_abort_move(dev, slave, RES_QP, qpn);
                                return err;
                        }
                }

                res_end_move(dev, slave, RES_QP, qpn);
                break;

        default:
                err = -EINVAL;
                break;
        }
        return err;
}

static int mtt_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
                         u64 in_param, u64 *out_param)
{
        int err = -EINVAL;
        int base;
        int order;

        if (op != RES_OP_RESERVE_AND_MAP)
                return err;

        order = get_param_l(&in_param);

        err = mlx4_grant_resource(dev, slave, RES_MTT, 1 << order, 0);
        if (err)
                return err;

        base = __mlx4_alloc_mtt_range(dev, order);
        if (base == -1) {
                mlx4_release_resource(dev, slave, RES_MTT, 1 << order, 0);
                return -ENOMEM;
        }

        err = add_res_range(dev, slave, base, 1, RES_MTT, order);
        if (err) {
                mlx4_release_resource(dev, slave, RES_MTT, 1 << order, 0);
                __mlx4_free_mtt_range(dev, base, order);
        } else {
                set_param_l(out_param, base);
        }

        return err;
}

static int mpt_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
                         u64 in_param, u64 *out_param)
{
        int err = -EINVAL;
        int index;
        int id;
        struct res_mpt *mpt;

        switch (op) {
        case RES_OP_RESERVE:
                err = mlx4_grant_resource(dev, slave, RES_MPT, 1, 0);
                if (err)
                        break;

                index = __mlx4_mpt_reserve(dev);
                if (index == -1) {
                        mlx4_release_resource(dev, slave, RES_MPT, 1, 0);
                        break;
                }
                id = index & mpt_mask(dev);

                err = add_res_range(dev, slave, id, 1, RES_MPT, index);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_MPT, 1, 0);
                        __mlx4_mpt_release(dev, index);
                        break;
                }
                set_param_l(out_param, index);
                break;
        case RES_OP_MAP_ICM:
                index = get_param_l(&in_param);
                id = index & mpt_mask(dev);
                err = mr_res_start_move_to(dev, slave, id,
                                           RES_MPT_MAPPED, &mpt);
                if (err)
                        return err;

                err = __mlx4_mpt_alloc_icm(dev, mpt->key);
                if (err) {
                        res_abort_move(dev, slave, RES_MPT, id);
                        return err;
                }

                res_end_move(dev, slave, RES_MPT, id);
                break;
        }
        return err;
}

static int cq_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
                        u64 in_param, u64 *out_param)
{
        int cqn;
        int err;

        switch (op) {
        case RES_OP_RESERVE_AND_MAP:
                err = mlx4_grant_resource(dev, slave, RES_CQ, 1, 0);
                if (err)
                        break;

                err = __mlx4_cq_alloc_icm(dev, &cqn);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_CQ, 1, 0);
                        break;
                }

                err = add_res_range(dev, slave, cqn, 1, RES_CQ, 0);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_CQ, 1, 0);
                        __mlx4_cq_free_icm(dev, cqn);
                        break;
                }

                set_param_l(out_param, cqn);
                break;

        default:
                err = -EINVAL;
        }

        return err;
}

static int srq_alloc_res(struct mlx4_dev *dev, int slave, int op, int cmd,
                         u64 in_param, u64 *out_param)
{
        int srqn;
        int err;

        switch (op) {
        case RES_OP_RESERVE_AND_MAP:
                err = mlx4_grant_resource(dev, slave, RES_SRQ, 1, 0);
                if (err)
                        break;

                err = __mlx4_srq_alloc_icm(dev, &srqn);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_SRQ, 1, 0);
                        break;
                }

                err = add_res_range(dev, slave, srqn, 1, RES_SRQ, 0);
                if (err) {
                        mlx4_release_resource(dev, slave, RES_SRQ, 1, 0);
                        __mlx4_srq_free_icm(dev, srqn);
                        break;
                }

                set_param_l(out_param, srqn);
                break;

        default:
                err = -EINVAL;
        }