numam-dpdk/app/test-crypto-perf/cperf_test_throughput.c
Fan Zhang d4ad392cbb app/crypto-perf: use separate session mempools
This patch uses the two session mempool approach to crypto perf
application. One mempool is for session header objects, and the other
is for session private data.

Signed-off-by: Fan Zhang <roy.fan.zhang@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
2019-01-10 16:57:22 +01:00

332 lines
8.4 KiB
C

/* SPDX-License-Identifier: BSD-3-Clause
* Copyright(c) 2016-2017 Intel Corporation
*/
#include <rte_malloc.h>
#include <rte_cycles.h>
#include <rte_crypto.h>
#include <rte_cryptodev.h>
#include "cperf_test_throughput.h"
#include "cperf_ops.h"
#include "cperf_test_common.h"
struct cperf_throughput_ctx {
uint8_t dev_id;
uint16_t qp_id;
uint8_t lcore_id;
struct rte_mempool *pool;
struct rte_cryptodev_sym_session *sess;
cperf_populate_ops_t populate_ops;
uint32_t src_buf_offset;
uint32_t dst_buf_offset;
const struct cperf_options *options;
const struct cperf_test_vector *test_vector;
};
static void
cperf_throughput_test_free(struct cperf_throughput_ctx *ctx)
{
if (ctx) {
if (ctx->sess) {
rte_cryptodev_sym_session_clear(ctx->dev_id, ctx->sess);
rte_cryptodev_sym_session_free(ctx->sess);
}
if (ctx->pool)
rte_mempool_free(ctx->pool);
rte_free(ctx);
}
}
void *
cperf_throughput_test_constructor(struct rte_mempool *sess_mp,
struct rte_mempool *sess_priv_mp,
uint8_t dev_id, uint16_t qp_id,
const struct cperf_options *options,
const struct cperf_test_vector *test_vector,
const struct cperf_op_fns *op_fns)
{
struct cperf_throughput_ctx *ctx = NULL;
ctx = rte_malloc(NULL, sizeof(struct cperf_throughput_ctx), 0);
if (ctx == NULL)
goto err;
ctx->dev_id = dev_id;
ctx->qp_id = qp_id;
ctx->populate_ops = op_fns->populate_ops;
ctx->options = options;
ctx->test_vector = test_vector;
/* IV goes at the end of the crypto operation */
uint16_t iv_offset = sizeof(struct rte_crypto_op) +
sizeof(struct rte_crypto_sym_op);
ctx->sess = op_fns->sess_create(sess_mp, sess_priv_mp, dev_id, options,
test_vector, iv_offset);
if (ctx->sess == NULL)
goto err;
if (cperf_alloc_common_memory(options, test_vector, dev_id, qp_id, 0,
&ctx->src_buf_offset, &ctx->dst_buf_offset,
&ctx->pool) < 0)
goto err;
return ctx;
err:
cperf_throughput_test_free(ctx);
return NULL;
}
int
cperf_throughput_test_runner(void *test_ctx)
{
struct cperf_throughput_ctx *ctx = test_ctx;
uint16_t test_burst_size;
uint8_t burst_size_idx = 0;
uint32_t imix_idx = 0;
static int only_once;
struct rte_crypto_op *ops[ctx->options->max_burst_size];
struct rte_crypto_op *ops_processed[ctx->options->max_burst_size];
uint64_t i;
uint32_t lcore = rte_lcore_id();
#ifdef CPERF_LINEARIZATION_ENABLE
struct rte_cryptodev_info dev_info;
int linearize = 0;
/* Check if source mbufs require coalescing */
if (ctx->options->segment_sz < ctx->options->max_buffer_size) {
rte_cryptodev_info_get(ctx->dev_id, &dev_info);
if ((dev_info.feature_flags &
RTE_CRYPTODEV_FF_MBUF_SCATTER_GATHER) == 0)
linearize = 1;
}
#endif /* CPERF_LINEARIZATION_ENABLE */
ctx->lcore_id = lcore;
/* Warm up the host CPU before starting the test */
for (i = 0; i < ctx->options->total_ops; i++)
rte_cryptodev_enqueue_burst(ctx->dev_id, ctx->qp_id, NULL, 0);
/* Get first size from range or list */
if (ctx->options->inc_burst_size != 0)
test_burst_size = ctx->options->min_burst_size;
else
test_burst_size = ctx->options->burst_size_list[0];
uint16_t iv_offset = sizeof(struct rte_crypto_op) +
sizeof(struct rte_crypto_sym_op);
while (test_burst_size <= ctx->options->max_burst_size) {
uint64_t ops_enqd = 0, ops_enqd_total = 0, ops_enqd_failed = 0;
uint64_t ops_deqd = 0, ops_deqd_total = 0, ops_deqd_failed = 0;
uint64_t tsc_start, tsc_end, tsc_duration;
uint16_t ops_unused = 0;
tsc_start = rte_rdtsc_precise();
while (ops_enqd_total < ctx->options->total_ops) {
uint16_t burst_size = ((ops_enqd_total + test_burst_size)
<= ctx->options->total_ops) ?
test_burst_size :
ctx->options->total_ops -
ops_enqd_total;
uint16_t ops_needed = burst_size - ops_unused;
/* Allocate objects containing crypto operations and mbufs */
if (rte_mempool_get_bulk(ctx->pool, (void **)ops,
ops_needed) != 0) {
RTE_LOG(ERR, USER1,
"Failed to allocate more crypto operations "
"from the crypto operation pool.\n"
"Consider increasing the pool size "
"with --pool-sz\n");
return -1;
}
/* Setup crypto op, attach mbuf etc */
(ctx->populate_ops)(ops, ctx->src_buf_offset,
ctx->dst_buf_offset,
ops_needed, ctx->sess,
ctx->options, ctx->test_vector,
iv_offset, &imix_idx);
/**
* When ops_needed is smaller than ops_enqd, the
* unused ops need to be moved to the front for
* next round use.
*/
if (unlikely(ops_enqd > ops_needed)) {
size_t nb_b_to_mov = ops_unused * sizeof(
struct rte_crypto_op *);
memmove(&ops[ops_needed], &ops[ops_enqd],
nb_b_to_mov);
}
#ifdef CPERF_LINEARIZATION_ENABLE
if (linearize) {
/* PMD doesn't support scatter-gather and source buffer
* is segmented.
* We need to linearize it before enqueuing.
*/
for (i = 0; i < burst_size; i++)
rte_pktmbuf_linearize(ops[i]->sym->m_src);
}
#endif /* CPERF_LINEARIZATION_ENABLE */
/* Enqueue burst of ops on crypto device */
ops_enqd = rte_cryptodev_enqueue_burst(ctx->dev_id, ctx->qp_id,
ops, burst_size);
if (ops_enqd < burst_size)
ops_enqd_failed++;
/**
* Calculate number of ops not enqueued (mainly for hw
* accelerators whose ingress queue can fill up).
*/
ops_unused = burst_size - ops_enqd;
ops_enqd_total += ops_enqd;
/* Dequeue processed burst of ops from crypto device */
ops_deqd = rte_cryptodev_dequeue_burst(ctx->dev_id, ctx->qp_id,
ops_processed, test_burst_size);
if (likely(ops_deqd)) {
/* Free crypto ops so they can be reused. */
rte_mempool_put_bulk(ctx->pool,
(void **)ops_processed, ops_deqd);
ops_deqd_total += ops_deqd;
} else {
/**
* Count dequeue polls which didn't return any
* processed operations. This statistic is mainly
* relevant to hw accelerators.
*/
ops_deqd_failed++;
}
}
/* Dequeue any operations still in the crypto device */
while (ops_deqd_total < ctx->options->total_ops) {
/* Sending 0 length burst to flush sw crypto device */
rte_cryptodev_enqueue_burst(ctx->dev_id, ctx->qp_id, NULL, 0);
/* dequeue burst */
ops_deqd = rte_cryptodev_dequeue_burst(ctx->dev_id, ctx->qp_id,
ops_processed, test_burst_size);
if (ops_deqd == 0)
ops_deqd_failed++;
else {
rte_mempool_put_bulk(ctx->pool,
(void **)ops_processed, ops_deqd);
ops_deqd_total += ops_deqd;
}
}
tsc_end = rte_rdtsc_precise();
tsc_duration = (tsc_end - tsc_start);
/* Calculate average operations processed per second */
double ops_per_second = ((double)ctx->options->total_ops /
tsc_duration) * rte_get_tsc_hz();
/* Calculate average throughput (Gbps) in bits per second */
double throughput_gbps = ((ops_per_second *
ctx->options->test_buffer_size * 8) / 1000000000);
/* Calculate average cycles per packet */
double cycles_per_packet = ((double)tsc_duration /
ctx->options->total_ops);
if (!ctx->options->csv) {
if (!only_once)
printf("%12s%12s%12s%12s%12s%12s%12s%12s%12s%12s\n\n",
"lcore id", "Buf Size", "Burst Size",
"Enqueued", "Dequeued", "Failed Enq",
"Failed Deq", "MOps", "Gbps",
"Cycles/Buf");
only_once = 1;
printf("%12u%12u%12u%12"PRIu64"%12"PRIu64"%12"PRIu64
"%12"PRIu64"%12.4f%12.4f%12.2f\n",
ctx->lcore_id,
ctx->options->test_buffer_size,
test_burst_size,
ops_enqd_total,
ops_deqd_total,
ops_enqd_failed,
ops_deqd_failed,
ops_per_second/1000000,
throughput_gbps,
cycles_per_packet);
} else {
if (!only_once)
printf("#lcore id,Buffer Size(B),"
"Burst Size,Enqueued,Dequeued,Failed Enq,"
"Failed Deq,Ops(Millions),Throughput(Gbps),"
"Cycles/Buf\n\n");
only_once = 1;
printf("%u;%u;%u;%"PRIu64";%"PRIu64";%"PRIu64";%"PRIu64";"
"%.3f;%.3f;%.3f\n",
ctx->lcore_id,
ctx->options->test_buffer_size,
test_burst_size,
ops_enqd_total,
ops_deqd_total,
ops_enqd_failed,
ops_deqd_failed,
ops_per_second/1000000,
throughput_gbps,
cycles_per_packet);
}
/* Get next size from range or list */
if (ctx->options->inc_burst_size != 0)
test_burst_size += ctx->options->inc_burst_size;
else {
if (++burst_size_idx == ctx->options->burst_size_count)
break;
test_burst_size = ctx->options->burst_size_list[burst_size_idx];
}
}
return 0;
}
void
cperf_throughput_test_destructor(void *arg)
{
struct cperf_throughput_ctx *ctx = arg;
if (ctx == NULL)
return;
cperf_throughput_test_free(ctx);
}