freebsd-dev/sys/dev/mlx5/mlx5_core/mlx5_cq.c
Hans Petter Selasky e48813009c Widen EPOCH(9) usage in mlx5en(4).
Make completion event path mostly lockless using EPOCH(9).

Implement a mechanism using EPOCH(9) which allows us to make
the callback path for completion events mostly lockless.

Simplify draining callback events using epoch_wait().

While at it make sure all receive completion callbacks are
covered by the network EPOCH(9), because this is required
when calling if_input() and ether_input() after r357012.

Sponsored by:	Mellanox Technologies
2020-01-30 12:35:13 +00:00

264 lines
7.4 KiB
C

/*-
* Copyright (c) 2013-2017, Mellanox Technologies, Ltd. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. 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.
*
* THIS SOFTWARE IS PROVIDED BY AUTHOR AND CONTRIBUTORS `AS IS' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* $FreeBSD$
*/
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/hardirq.h>
#include <dev/mlx5/driver.h>
#include <rdma/ib_verbs.h>
#include <dev/mlx5/cq.h>
#include "mlx5_core.h"
#include <sys/epoch.h>
static void
mlx5_cq_table_write_lock(struct mlx5_cq_table *table)
{
atomic_inc(&table->writercount);
/* make sure all see the updated writercount */
NET_EPOCH_WAIT();
spin_lock(&table->writerlock);
}
static void
mlx5_cq_table_write_unlock(struct mlx5_cq_table *table)
{
spin_unlock(&table->writerlock);
atomic_dec(&table->writercount);
/* drain all pending CQ callers */
NET_EPOCH_WAIT();
}
void mlx5_cq_completion(struct mlx5_core_dev *dev, u32 cqn)
{
struct mlx5_cq_table *table = &dev->priv.cq_table;
struct mlx5_core_cq *cq;
struct epoch_tracker et;
bool do_lock;
NET_EPOCH_ENTER(et);
do_lock = atomic_read(&table->writercount) != 0;
if (unlikely(do_lock))
spin_lock(&table->writerlock);
if (likely(cqn < MLX5_CQ_LINEAR_ARRAY_SIZE))
cq = table->linear_array[cqn].cq;
else
cq = radix_tree_lookup(&table->tree, cqn);
if (unlikely(do_lock))
spin_unlock(&table->writerlock);
if (likely(cq != NULL)) {
++cq->arm_sn;
cq->comp(cq);
} else {
mlx5_core_warn(dev,
"Completion event for bogus CQ 0x%x\n", cqn);
}
NET_EPOCH_EXIT(et);
}
void mlx5_cq_event(struct mlx5_core_dev *dev, u32 cqn, int event_type)
{
struct mlx5_cq_table *table = &dev->priv.cq_table;
struct mlx5_core_cq *cq;
struct epoch_tracker et;
bool do_lock;
NET_EPOCH_ENTER(et);
do_lock = atomic_read(&table->writercount) != 0;
if (unlikely(do_lock))
spin_lock(&table->writerlock);
if (likely(cqn < MLX5_CQ_LINEAR_ARRAY_SIZE))
cq = table->linear_array[cqn].cq;
else
cq = radix_tree_lookup(&table->tree, cqn);
if (unlikely(do_lock))
spin_unlock(&table->writerlock);
if (likely(cq != NULL)) {
cq->event(cq, event_type);
} else {
mlx5_core_warn(dev,
"Asynchronous event for bogus CQ 0x%x\n", cqn);
}
NET_EPOCH_EXIT(et);
}
int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq,
u32 *in, int inlen)
{
struct mlx5_cq_table *table = &dev->priv.cq_table;
u32 out[MLX5_ST_SZ_DW(create_cq_out)] = {0};
u32 din[MLX5_ST_SZ_DW(destroy_cq_in)] = {0};
u32 dout[MLX5_ST_SZ_DW(destroy_cq_out)] = {0};
int err;
MLX5_SET(create_cq_in, in, opcode, MLX5_CMD_OP_CREATE_CQ);
err = mlx5_cmd_exec(dev, in, inlen, out, sizeof(out));
if (err)
return err;
cq->cqn = MLX5_GET(create_cq_out, out, cqn);
cq->cons_index = 0;
cq->arm_sn = 0;
mlx5_cq_table_write_lock(table);
err = radix_tree_insert(&table->tree, cq->cqn, cq);
if (likely(err == 0 && cq->cqn < MLX5_CQ_LINEAR_ARRAY_SIZE))
table->linear_array[cq->cqn].cq = cq;
mlx5_cq_table_write_unlock(table);
if (err)
goto err_cmd;
cq->pid = curthread->td_proc->p_pid;
return 0;
err_cmd:
MLX5_SET(destroy_cq_in, din, opcode, MLX5_CMD_OP_DESTROY_CQ);
MLX5_SET(destroy_cq_in, din, cqn, cq->cqn);
mlx5_cmd_exec(dev, din, sizeof(din), dout, sizeof(dout));
return err;
}
EXPORT_SYMBOL(mlx5_core_create_cq);
int mlx5_core_destroy_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq)
{
struct mlx5_cq_table *table = &dev->priv.cq_table;
u32 out[MLX5_ST_SZ_DW(destroy_cq_out)] = {0};
u32 in[MLX5_ST_SZ_DW(destroy_cq_in)] = {0};
struct mlx5_core_cq *tmp;
mlx5_cq_table_write_lock(table);
if (likely(cq->cqn < MLX5_CQ_LINEAR_ARRAY_SIZE))
table->linear_array[cq->cqn].cq = NULL;
tmp = radix_tree_delete(&table->tree, cq->cqn);
mlx5_cq_table_write_unlock(table);
if (unlikely(tmp == NULL)) {
mlx5_core_warn(dev, "cq 0x%x not found in tree\n", cq->cqn);
return -EINVAL;
} else if (unlikely(tmp != cq)) {
mlx5_core_warn(dev, "corrupted cqn 0x%x\n", cq->cqn);
return -EINVAL;
}
MLX5_SET(destroy_cq_in, in, opcode, MLX5_CMD_OP_DESTROY_CQ);
MLX5_SET(destroy_cq_in, in, cqn, cq->cqn);
return mlx5_cmd_exec(dev, in, sizeof(in), out, sizeof(out));
}
EXPORT_SYMBOL(mlx5_core_destroy_cq);
int mlx5_core_query_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq,
u32 *out, int outlen)
{
u32 in[MLX5_ST_SZ_DW(query_cq_in)] = {0};
MLX5_SET(query_cq_in, in, opcode, MLX5_CMD_OP_QUERY_CQ);
MLX5_SET(query_cq_in, in, cqn, cq->cqn);
return mlx5_cmd_exec(dev, in, sizeof(in), out, outlen);
}
EXPORT_SYMBOL(mlx5_core_query_cq);
int mlx5_core_modify_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq,
u32 *in, int inlen)
{
u32 out[MLX5_ST_SZ_DW(modify_cq_out)] = {0};
MLX5_SET(modify_cq_in, in, opcode, MLX5_CMD_OP_MODIFY_CQ);
return mlx5_cmd_exec(dev, in, inlen, out, sizeof(out));
}
EXPORT_SYMBOL(mlx5_core_modify_cq);
int mlx5_core_modify_cq_moderation(struct mlx5_core_dev *dev,
struct mlx5_core_cq *cq,
u16 cq_period,
u16 cq_max_count)
{
u32 in[MLX5_ST_SZ_DW(modify_cq_in)] = {0};
void *cqc;
MLX5_SET(modify_cq_in, in, cqn, cq->cqn);
cqc = MLX5_ADDR_OF(modify_cq_in, in, cq_context);
MLX5_SET(cqc, cqc, cq_period, cq_period);
MLX5_SET(cqc, cqc, cq_max_count, cq_max_count);
MLX5_SET(modify_cq_in, in,
modify_field_select_resize_field_select.modify_field_select.modify_field_select,
MLX5_CQ_MODIFY_PERIOD | MLX5_CQ_MODIFY_COUNT);
return mlx5_core_modify_cq(dev, cq, in, sizeof(in));
}
int mlx5_core_modify_cq_moderation_mode(struct mlx5_core_dev *dev,
struct mlx5_core_cq *cq,
u16 cq_period,
u16 cq_max_count,
u8 cq_mode)
{
u32 in[MLX5_ST_SZ_DW(modify_cq_in)] = {0};
void *cqc;
MLX5_SET(modify_cq_in, in, cqn, cq->cqn);
cqc = MLX5_ADDR_OF(modify_cq_in, in, cq_context);
MLX5_SET(cqc, cqc, cq_period, cq_period);
MLX5_SET(cqc, cqc, cq_max_count, cq_max_count);
MLX5_SET(cqc, cqc, cq_period_mode, cq_mode);
MLX5_SET(modify_cq_in, in,
modify_field_select_resize_field_select.modify_field_select.modify_field_select,
MLX5_CQ_MODIFY_PERIOD | MLX5_CQ_MODIFY_COUNT | MLX5_CQ_MODIFY_PERIOD_MODE);
return mlx5_core_modify_cq(dev, cq, in, sizeof(in));
}
int mlx5_init_cq_table(struct mlx5_core_dev *dev)
{
struct mlx5_cq_table *table = &dev->priv.cq_table;
memset(table, 0, sizeof(*table));
spin_lock_init(&table->writerlock);
INIT_RADIX_TREE(&table->tree, GFP_ATOMIC);
return 0;
}
void mlx5_cleanup_cq_table(struct mlx5_core_dev *dev)
{
}