From: Ruifeng Wang <Ruifeng.Wang@arm.com>
To: Adam Dybkowski <adamx.dybkowski@intel.com>,
"dev@dpdk.org" <dev@dpdk.org>,
"fiona.trahe@intel.com" <fiona.trahe@intel.com>,
"Akhil.goyal@nxp.com" <akhil.goyal@nxp.com>
Cc: Fan Zhang <roy.fan.zhang@intel.com>, nd <nd@arm.com>
Subject: Re: [dpdk-dev] [PATCH v2 1/1] crypto/scheduler: rename slave to worker
Date: Mon, 28 Sep 2020 15:12:27 +0000 [thread overview]
Message-ID: <VI1PR0802MB2351C78B98AF0B794538B27D9E350@VI1PR0802MB2351.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <20200928141633.396-2-adamx.dybkowski@intel.com>
> -----Original Message-----
> From: dev <dev-bounces@dpdk.org> On Behalf Of Adam Dybkowski
> Sent: Monday, September 28, 2020 10:17 PM
> To: dev@dpdk.org; fiona.trahe@intel.com; Akhil.goyal@nxp.com
> Cc: Adam Dybkowski <adamx.dybkowski@intel.com>; Fan Zhang
> <roy.fan.zhang@intel.com>
> Subject: [dpdk-dev] [PATCH v2 1/1] crypto/scheduler: rename slave to
> worker
>
> This patch replaces the usage of the word 'slave' with more
> appropriate word 'worker' in QAT PMD and Scheduler PMD
> as well as in their docs. Also the test app was modified
> to use the new wording.
>
> The Scheduler PMD's public API was modified according to the
> previous deprecation notice:
> rte_cryptodev_scheduler_slave_attach is now called
> rte_cryptodev_scheduler_worker_attach,
> rte_cryptodev_scheduler_slave_detach is
> rte_cryptodev_scheduler_worker_detach,
> rte_cryptodev_scheduler_slaves_get is
> rte_cryptodev_scheduler_workers_get.
>
> Also, the configuration value
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES
> was renamed to RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS.
>
> Signed-off-by: Adam Dybkowski <adamx.dybkowski@intel.com>
> Acked-by: Fan Zhang <roy.fan.zhang@intel.com>
> ---
> app/test-crypto-perf/main.c | 2 +-
> app/test/test_cryptodev.c | 20 +-
> doc/guides/cryptodevs/qat.rst | 2 +-
> doc/guides/cryptodevs/scheduler.rst | 40 ++--
> doc/guides/rel_notes/deprecation.rst | 7 -
> doc/guides/rel_notes/release_20_11.rst | 11 +
> .../scheduler/rte_cryptodev_scheduler.c | 114 +++++-----
> .../scheduler/rte_cryptodev_scheduler.h | 35 ++-
> .../rte_cryptodev_scheduler_operations.h | 12 +-
> .../rte_pmd_crypto_scheduler_version.map | 6 +-
> drivers/crypto/scheduler/scheduler_failover.c | 83 +++----
> .../crypto/scheduler/scheduler_multicore.c | 54 ++---
> .../scheduler/scheduler_pkt_size_distr.c | 142 ++++++------
> drivers/crypto/scheduler/scheduler_pmd.c | 54 ++---
> drivers/crypto/scheduler/scheduler_pmd_ops.c | 204 +++++++++---------
> .../crypto/scheduler/scheduler_pmd_private.h | 12 +-
> .../crypto/scheduler/scheduler_roundrobin.c | 87 ++++----
> examples/l2fwd-crypto/main.c | 6 +-
> 18 files changed, 449 insertions(+), 442 deletions(-)
>
> diff --git a/app/test-crypto-perf/main.c b/app/test-crypto-perf/main.c
> index 8f8e580e4..62ae6048b 100644
> --- a/app/test-crypto-perf/main.c
> +++ b/app/test-crypto-perf/main.c
> @@ -240,7 +240,7 @@ cperf_initialize_cryptodev(struct cperf_options *opts,
> uint8_t *enabled_cdevs)
> "crypto_scheduler")) {
> #ifdef RTE_LIBRTE_PMD_CRYPTO_SCHEDULER
> uint32_t nb_slaves =
> -
> rte_cryptodev_scheduler_slaves_get(cdev_id,
> +
> rte_cryptodev_scheduler_workers_get(cdev_id,
> NULL);
>
> sessions_needed = enabled_cdev_count *
> diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
> index 70bf6fe2c..255fb7525 100644
> --- a/app/test/test_cryptodev.c
> +++ b/app/test/test_cryptodev.c
> @@ -479,29 +479,29 @@ testsuite_setup(void)
> char vdev_args[VDEV_ARGS_SIZE] = {""};
> char temp_str[VDEV_ARGS_SIZE] = {"mode=multi-core,"
>
> "ordering=enable,name=cryptodev_test_scheduler,corelist="};
> - uint16_t slave_core_count = 0;
> + uint16_t worker_core_count = 0;
> uint16_t socket_id = 0;
>
> if (gbl_driver_id == rte_cryptodev_driver_id_get(
> RTE_STR(CRYPTODEV_NAME_SCHEDULER_PMD))) {
>
> - /* Identify the Slave Cores
> - * Use 2 slave cores for the device args
> + /* Identify the Worker Cores
> + * Use 2 worker cores for the device args
> */
> RTE_LCORE_FOREACH_SLAVE(i) {
> - if (slave_core_count > 1)
> + if (worker_core_count > 1)
> break;
> snprintf(vdev_args, sizeof(vdev_args),
> "%s%d", temp_str, i);
> strcpy(temp_str, vdev_args);
> strlcat(temp_str, ";", sizeof(temp_str));
> - slave_core_count++;
> + worker_core_count++;
> socket_id = rte_lcore_to_socket_id(i);
> }
> - if (slave_core_count != 2) {
> + if (worker_core_count != 2) {
> RTE_LOG(ERR, USER1,
> "Cryptodev scheduler test require at least "
> - "two slave cores to run. "
> + "two worker cores to run. "
> "Please use the correct coremask.\n");
> return TEST_FAILED;
> }
> @@ -11712,7 +11712,7 @@
> test_chacha20_poly1305_decrypt_test_case_rfc8439(void)
>
> #ifdef RTE_LIBRTE_PMD_CRYPTO_SCHEDULER
>
> -/* global AESNI slave IDs for the scheduler test */
> +/* global AESNI worker IDs for the scheduler test */
> uint8_t aesni_ids[2];
>
> static int
> @@ -11810,7 +11810,7 @@ test_scheduler_attach_slave_op(void)
> ts_params->qp_conf.mp_session_private =
> ts_params->session_priv_mpool;
>
> - ret = rte_cryptodev_scheduler_slave_attach(sched_id,
> + ret = rte_cryptodev_scheduler_worker_attach(sched_id,
> (uint8_t)i);
>
> TEST_ASSERT(ret == 0,
> @@ -11834,7 +11834,7 @@ test_scheduler_detach_slave_op(void)
> int ret;
>
> for (i = 0; i < 2; i++) {
> - ret = rte_cryptodev_scheduler_slave_detach(sched_id,
> + ret = rte_cryptodev_scheduler_worker_detach(sched_id,
> aesni_ids[i]);
> TEST_ASSERT(ret == 0,
> "Failed to detach device %u", aesni_ids[i]);
> diff --git a/doc/guides/cryptodevs/qat.rst b/doc/guides/cryptodevs/qat.rst
> index e5d2cf499..ee0bd3a0d 100644
> --- a/doc/guides/cryptodevs/qat.rst
> +++ b/doc/guides/cryptodevs/qat.rst
> @@ -328,7 +328,7 @@ The "rte_cryptodev_devices_get()" returns the
> devices exposed by either of these
>
> The cryptodev driver name is passed to the dpdk-test-crypto-perf
> tool in the "-devtype" parameter.
>
> - The qat crypto device name is in the format of the slave parameter
> passed to the crypto scheduler.
> + The qat crypto device name is in the format of the worker parameter
> passed to the crypto scheduler.
>
> * The qat compressdev driver name is "compress_qat".
> The rte_compressdev_devices_get() returns the devices exposed by this
> driver.
> diff --git a/doc/guides/cryptodevs/scheduler.rst
> b/doc/guides/cryptodevs/scheduler.rst
> index 7004ca431..565de40f3 100644
> --- a/doc/guides/cryptodevs/scheduler.rst
> +++ b/doc/guides/cryptodevs/scheduler.rst
> @@ -16,12 +16,12 @@ crypto ops among them in a certain manner.
> The Cryptodev Scheduler PMD library (**librte_pmd_crypto_scheduler**)
> acts as
> a software crypto PMD and shares the same API provided by
> librte_cryptodev.
> The PMD supports attaching multiple crypto PMDs, software or hardware, as
> -slaves, and distributes the crypto workload to them with certain behavior.
> +workers, and distributes the crypto workload to them with certain behavior.
> The behaviors are categorizes as different "modes". Basically, a scheduling
> -mode defines certain actions for scheduling crypto ops to its slaves.
> +mode defines certain actions for scheduling crypto ops to its workers.
>
> The librte_pmd_crypto_scheduler library exports a C API which provides an
> API
> -for attaching/detaching slaves, set/get scheduling modes, and
> enable/disable
> +for attaching/detaching workers, set/get scheduling modes, and
> enable/disable
> crypto ops reordering.
>
> Limitations
> @@ -62,7 +62,7 @@ two calls:
> created. This value may be overwritten internally if there are too
> many devices are attached.
>
> -* slave: If a cryptodev has been initialized with specific name, it can be
> +* worker: If a cryptodev has been initialized with specific name, it can be
> attached to the scheduler using this parameter, simply filling the name
> here. Multiple cryptodevs can be attached initially by presenting this
> parameter multiple times.
> @@ -84,13 +84,13 @@ Example:
>
> .. code-block:: console
>
> - ... --vdev "crypto_aesni_mb0,name=aesni_mb_1" --vdev
> "crypto_aesni_mb1,name=aesni_mb_2" --vdev
> "crypto_scheduler,slave=aesni_mb_1,slave=aesni_mb_2" ...
> + ... --vdev "crypto_aesni_mb0,name=aesni_mb_1" --vdev
> "crypto_aesni_mb1,name=aesni_mb_2" --vdev
> "crypto_scheduler,worker=aesni_mb_1,worker=aesni_mb_2" ...
>
> .. note::
>
> * The scheduler cryptodev cannot be started unless the scheduling mode
> - is set and at least one slave is attached. Also, to configure the
> - scheduler in the run-time, like attach/detach slave(s), change
> + is set and at least one worker is attached. Also, to configure the
> + scheduler in the run-time, like attach/detach worker(s), change
> scheduling mode, or enable/disable crypto op ordering, one should stop
> the scheduler first, otherwise an error will be returned.
>
> @@ -111,7 +111,7 @@ operation:
> *Initialization mode parameter*: **round-robin**
>
> Round-robin mode, which distributes the enqueued burst of crypto ops
> - among its slaves in a round-robin manner. This mode may help to fill
> + among its workers in a round-robin manner. This mode may help to fill
> the throughput gap between the physical core and the existing cryptodevs
> to increase the overall performance.
>
> @@ -119,15 +119,15 @@ operation:
>
> *Initialization mode parameter*: **packet-size-distr**
>
> - Packet-size based distribution mode, which works with 2 slaves, the
> primary
> - slave and the secondary slave, and distributes the enqueued crypto
> + Packet-size based distribution mode, which works with 2 workers, the
> primary
> + worker and the secondary worker, and distributes the enqueued crypto
> operations to them based on their data lengths. A crypto operation will be
> - distributed to the primary slave if its data length is equal to or bigger
> + distributed to the primary worker if its data length is equal to or bigger
> than the designated threshold, otherwise it will be handled by the
> secondary
> - slave.
> + worker.
>
> A typical usecase in this mode is with the QAT cryptodev as the primary and
> - a software cryptodev as the secondary slave. This may help applications to
> + a software cryptodev as the secondary worker. This may help applications
> to
> process additional crypto workload than what the QAT cryptodev can
> handle on
> its own, by making use of the available CPU cycles to deal with smaller
> crypto workloads.
> @@ -148,11 +148,11 @@ operation:
>
> *Initialization mode parameter*: **fail-over**
>
> - Fail-over mode, which works with 2 slaves, the primary slave and the
> - secondary slave. In this mode, the scheduler will enqueue the incoming
> - crypto operation burst to the primary slave. When one or more crypto
> + Fail-over mode, which works with 2 workers, the primary worker and the
> + secondary worker. In this mode, the scheduler will enqueue the incoming
> + crypto operation burst to the primary worker. When one or more crypto
> operations fail to be enqueued, then they will be enqueued to the
> secondary
> - slave.
> + worker.
>
> * **CDEV_SCHED_MODE_MULTICORE:**
>
> @@ -167,16 +167,16 @@ operation:
> For mixed traffic (IMIX) the optimal number of worker cores is around 2-3.
> For large packets (1.5 kbytes) scheduler shows linear scaling in
> performance
> up to eight cores.
> - Each worker uses its own slave cryptodev. Only software cryptodevs
> + Each worker uses its own cryptodev. Only software cryptodevs
> are supported. Only the same type of cryptodevs should be used
> concurrently.
>
> The multi-core mode uses one extra parameter:
>
> * corelist: Semicolon-separated list of logical cores to be used as workers.
> - The number of worker cores should be equal to the number of slave
> cryptodevs.
> + The number of worker cores should be equal to the number of worker
> cryptodevs.
> These cores should be present in EAL core list parameter and
> should not be used by the application or any other process.
>
> Example:
> ... --vdev "crypto_aesni_mb1,name=aesni_mb_1" --vdev
> "crypto_aesni_mb_pmd2,name=aesni_mb_2" \
> - --vdev
> "crypto_scheduler,slave=aesni_mb_1,slave=aesni_mb_2,mode=multi-
> core,corelist=23;24" ...
> + --vdev
> "crypto_scheduler,worker=aesni_mb_1,worker=aesni_mb_2,mode=multi-
> core,corelist=23;24" ...
> diff --git a/doc/guides/rel_notes/deprecation.rst
> b/doc/guides/rel_notes/deprecation.rst
> index f9b72acb8..7621210b0 100644
> --- a/doc/guides/rel_notes/deprecation.rst
> +++ b/doc/guides/rel_notes/deprecation.rst
> @@ -241,13 +241,6 @@ Deprecation Notices
> to one it means it represents IV, when is set to zero it means J0 is used
> directly, in this case 16 bytes of J0 need to be passed.
>
> -* scheduler: The functions ``rte_cryptodev_scheduler_slave_attach``,
> - ``rte_cryptodev_scheduler_slave_detach`` and
> - ``rte_cryptodev_scheduler_slaves_get`` will be replaced in 20.11 by
> - ``rte_cryptodev_scheduler_worker_attach``,
> - ``rte_cryptodev_scheduler_worker_detach`` and
> - ``rte_cryptodev_scheduler_workers_get`` accordingly.
> -
> * eventdev: Following structures will be modified to support DLB PMD
> and future extensions:
>
> diff --git a/doc/guides/rel_notes/release_20_11.rst
> b/doc/guides/rel_notes/release_20_11.rst
> index 73ac08fb0..3e06ad7c3 100644
> --- a/doc/guides/rel_notes/release_20_11.rst
> +++ b/doc/guides/rel_notes/release_20_11.rst
> @@ -133,6 +133,17 @@ API Changes
> and the function ``rte_rawdev_queue_conf_get()``
> from ``void`` to ``int`` allowing the return of error codes from drivers.
>
> +* scheduler: Renamed functions ``rte_cryptodev_scheduler_slave_attach``,
> + ``rte_cryptodev_scheduler_slave_detach`` and
> + ``rte_cryptodev_scheduler_slaves_get`` to
> + ``rte_cryptodev_scheduler_worker_attach``,
> + ``rte_cryptodev_scheduler_worker_detach`` and
> + ``rte_cryptodev_scheduler_workers_get`` accordingly.
> +
> +* scheduler: Renamed the configuration value
> + ``RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES`` to
> + ``RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS``.
> +
> * bpf: ``RTE_BPF_XTYPE_NUM`` has been dropped from ``rte_bpf_xtype``.
>
>
> diff --git a/drivers/crypto/scheduler/rte_cryptodev_scheduler.c
> b/drivers/crypto/scheduler/rte_cryptodev_scheduler.c
> index 730504dab..9367a0e91 100644
> --- a/drivers/crypto/scheduler/rte_cryptodev_scheduler.c
> +++ b/drivers/crypto/scheduler/rte_cryptodev_scheduler.c
> @@ -13,31 +13,31 @@
> /** update the scheduler pmd's capability with attaching device's
> * capability.
> * For each device to be attached, the scheduler's capability should be
> - * the common capability set of all slaves
> + * the common capability set of all workers
> **/
> static uint32_t
> sync_caps(struct rte_cryptodev_capabilities *caps,
> uint32_t nb_caps,
> - const struct rte_cryptodev_capabilities *slave_caps)
> + const struct rte_cryptodev_capabilities *worker_caps)
> {
> - uint32_t sync_nb_caps = nb_caps, nb_slave_caps = 0;
> + uint32_t sync_nb_caps = nb_caps, nb_worker_caps = 0;
> uint32_t i;
>
> - while (slave_caps[nb_slave_caps].op !=
> RTE_CRYPTO_OP_TYPE_UNDEFINED)
> - nb_slave_caps++;
> + while (worker_caps[nb_worker_caps].op !=
> RTE_CRYPTO_OP_TYPE_UNDEFINED)
> + nb_worker_caps++;
>
> if (nb_caps == 0) {
> - rte_memcpy(caps, slave_caps, sizeof(*caps) *
> nb_slave_caps);
> - return nb_slave_caps;
> + rte_memcpy(caps, worker_caps, sizeof(*caps) *
> nb_worker_caps);
> + return nb_worker_caps;
> }
>
> for (i = 0; i < sync_nb_caps; i++) {
> struct rte_cryptodev_capabilities *cap = &caps[i];
> uint32_t j;
>
> - for (j = 0; j < nb_slave_caps; j++) {
> + for (j = 0; j < nb_worker_caps; j++) {
> const struct rte_cryptodev_capabilities *s_cap =
> - &slave_caps[j];
> + &worker_caps[j];
>
> if (s_cap->op != cap->op || s_cap-
> >sym.xform_type !=
> cap->sym.xform_type)
> @@ -72,7 +72,7 @@ sync_caps(struct rte_cryptodev_capabilities *caps,
> break;
> }
>
> - if (j < nb_slave_caps)
> + if (j < nb_worker_caps)
> continue;
>
> /* remove a uncommon cap from the array */
> @@ -97,10 +97,10 @@ update_scheduler_capability(struct scheduler_ctx
> *sched_ctx)
> sched_ctx->capabilities = NULL;
> }
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> struct rte_cryptodev_info dev_info;
>
> - rte_cryptodev_info_get(sched_ctx->slaves[i].dev_id,
> &dev_info);
> + rte_cryptodev_info_get(sched_ctx->workers[i].dev_id,
> &dev_info);
>
> nb_caps = sync_caps(tmp_caps, nb_caps,
> dev_info.capabilities);
> if (nb_caps == 0)
> @@ -127,10 +127,10 @@ update_scheduler_feature_flag(struct
> rte_cryptodev *dev)
>
> dev->feature_flags = 0;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> struct rte_cryptodev_info dev_info;
>
> - rte_cryptodev_info_get(sched_ctx->slaves[i].dev_id,
> &dev_info);
> + rte_cryptodev_info_get(sched_ctx->workers[i].dev_id,
> &dev_info);
>
> dev->feature_flags |= dev_info.feature_flags;
> }
> @@ -142,15 +142,15 @@ update_max_nb_qp(struct scheduler_ctx
> *sched_ctx)
> uint32_t i;
> uint32_t max_nb_qp;
>
> - if (!sched_ctx->nb_slaves)
> + if (!sched_ctx->nb_workers)
> return;
>
> - max_nb_qp = sched_ctx->nb_slaves ? UINT32_MAX : 0;
> + max_nb_qp = sched_ctx->nb_workers ? UINT32_MAX : 0;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> struct rte_cryptodev_info dev_info;
>
> - rte_cryptodev_info_get(sched_ctx->slaves[i].dev_id,
> &dev_info);
> + rte_cryptodev_info_get(sched_ctx->workers[i].dev_id,
> &dev_info);
> max_nb_qp = dev_info.max_nb_queue_pairs < max_nb_qp ?
> dev_info.max_nb_queue_pairs : max_nb_qp;
> }
> @@ -160,11 +160,11 @@ update_max_nb_qp(struct scheduler_ctx
> *sched_ctx)
>
> /** Attach a device to the scheduler. */
> int
> -rte_cryptodev_scheduler_slave_attach(uint8_t scheduler_id, uint8_t
> slave_id)
> +rte_cryptodev_scheduler_worker_attach(uint8_t scheduler_id, uint8_t
> worker_id)
> {
> struct rte_cryptodev *dev =
> rte_cryptodev_pmd_get_dev(scheduler_id);
> struct scheduler_ctx *sched_ctx;
> - struct scheduler_slave *slave;
> + struct scheduler_worker *worker;
> struct rte_cryptodev_info dev_info;
> uint32_t i;
>
> @@ -184,30 +184,30 @@ rte_cryptodev_scheduler_slave_attach(uint8_t
> scheduler_id, uint8_t slave_id)
> }
>
> sched_ctx = dev->data->dev_private;
> - if (sched_ctx->nb_slaves >=
> - RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES) {
> - CR_SCHED_LOG(ERR, "Too many slaves attached");
> + if (sched_ctx->nb_workers >=
> + RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS)
> {
> + CR_SCHED_LOG(ERR, "Too many workers attached");
> return -ENOMEM;
> }
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++)
> - if (sched_ctx->slaves[i].dev_id == slave_id) {
> - CR_SCHED_LOG(ERR, "Slave already added");
> + for (i = 0; i < sched_ctx->nb_workers; i++)
> + if (sched_ctx->workers[i].dev_id == worker_id) {
> + CR_SCHED_LOG(ERR, "Worker already added");
> return -ENOTSUP;
> }
>
> - slave = &sched_ctx->slaves[sched_ctx->nb_slaves];
> + worker = &sched_ctx->workers[sched_ctx->nb_workers];
>
> - rte_cryptodev_info_get(slave_id, &dev_info);
> + rte_cryptodev_info_get(worker_id, &dev_info);
>
> - slave->dev_id = slave_id;
> - slave->driver_id = dev_info.driver_id;
> - sched_ctx->nb_slaves++;
> + worker->dev_id = worker_id;
> + worker->driver_id = dev_info.driver_id;
> + sched_ctx->nb_workers++;
>
> if (update_scheduler_capability(sched_ctx) < 0) {
> - slave->dev_id = 0;
> - slave->driver_id = 0;
> - sched_ctx->nb_slaves--;
> + worker->dev_id = 0;
> + worker->driver_id = 0;
> + sched_ctx->nb_workers--;
>
> CR_SCHED_LOG(ERR, "capabilities update failed");
> return -ENOTSUP;
> @@ -221,11 +221,11 @@ rte_cryptodev_scheduler_slave_attach(uint8_t
> scheduler_id, uint8_t slave_id)
> }
>
> int
> -rte_cryptodev_scheduler_slave_detach(uint8_t scheduler_id, uint8_t
> slave_id)
> +rte_cryptodev_scheduler_worker_detach(uint8_t scheduler_id, uint8_t
> worker_id)
> {
> struct rte_cryptodev *dev =
> rte_cryptodev_pmd_get_dev(scheduler_id);
> struct scheduler_ctx *sched_ctx;
> - uint32_t i, slave_pos;
> + uint32_t i, worker_pos;
>
> if (!dev) {
> CR_SCHED_LOG(ERR, "Operation not supported");
> @@ -244,26 +244,26 @@ rte_cryptodev_scheduler_slave_detach(uint8_t
> scheduler_id, uint8_t slave_id)
>
> sched_ctx = dev->data->dev_private;
>
> - for (slave_pos = 0; slave_pos < sched_ctx->nb_slaves; slave_pos++)
> - if (sched_ctx->slaves[slave_pos].dev_id == slave_id)
> + for (worker_pos = 0; worker_pos < sched_ctx->nb_workers;
> worker_pos++)
> + if (sched_ctx->workers[worker_pos].dev_id == worker_id)
> break;
> - if (slave_pos == sched_ctx->nb_slaves) {
> - CR_SCHED_LOG(ERR, "Cannot find slave");
> + if (worker_pos == sched_ctx->nb_workers) {
> + CR_SCHED_LOG(ERR, "Cannot find worker");
> return -ENOTSUP;
> }
>
> - if (sched_ctx->ops.slave_detach(dev, slave_id) < 0) {
> - CR_SCHED_LOG(ERR, "Failed to detach slave");
> + if (sched_ctx->ops.worker_detach(dev, worker_id) < 0) {
> + CR_SCHED_LOG(ERR, "Failed to detach worker");
> return -ENOTSUP;
> }
>
> - for (i = slave_pos; i < sched_ctx->nb_slaves - 1; i++) {
> - memcpy(&sched_ctx->slaves[i], &sched_ctx->slaves[i+1],
> - sizeof(struct scheduler_slave));
> + for (i = worker_pos; i < sched_ctx->nb_workers - 1; i++) {
> + memcpy(&sched_ctx->workers[i], &sched_ctx-
> >workers[i+1],
> + sizeof(struct scheduler_worker));
> }
> - memset(&sched_ctx->slaves[sched_ctx->nb_slaves - 1], 0,
> - sizeof(struct scheduler_slave));
> - sched_ctx->nb_slaves--;
> + memset(&sched_ctx->workers[sched_ctx->nb_workers - 1], 0,
> + sizeof(struct scheduler_worker));
> + sched_ctx->nb_workers--;
>
> if (update_scheduler_capability(sched_ctx) < 0) {
> CR_SCHED_LOG(ERR, "capabilities update failed");
> @@ -459,8 +459,8 @@
> rte_cryptodev_scheduler_load_user_scheduler(uint8_t scheduler_id,
> sched_ctx->ops.create_private_ctx = scheduler->ops-
> >create_private_ctx;
> sched_ctx->ops.scheduler_start = scheduler->ops->scheduler_start;
> sched_ctx->ops.scheduler_stop = scheduler->ops->scheduler_stop;
> - sched_ctx->ops.slave_attach = scheduler->ops->slave_attach;
> - sched_ctx->ops.slave_detach = scheduler->ops->slave_detach;
> + sched_ctx->ops.worker_attach = scheduler->ops->worker_attach;
> + sched_ctx->ops.worker_detach = scheduler->ops->worker_detach;
> sched_ctx->ops.option_set = scheduler->ops->option_set;
> sched_ctx->ops.option_get = scheduler->ops->option_get;
>
> @@ -485,11 +485,11 @@
> rte_cryptodev_scheduler_load_user_scheduler(uint8_t scheduler_id,
> }
>
> int
> -rte_cryptodev_scheduler_slaves_get(uint8_t scheduler_id, uint8_t *slaves)
> +rte_cryptodev_scheduler_workers_get(uint8_t scheduler_id, uint8_t
> *workers)
> {
> struct rte_cryptodev *dev =
> rte_cryptodev_pmd_get_dev(scheduler_id);
> struct scheduler_ctx *sched_ctx;
> - uint32_t nb_slaves = 0;
> + uint32_t nb_workers = 0;
>
> if (!dev) {
> CR_SCHED_LOG(ERR, "Operation not supported");
> @@ -503,16 +503,16 @@ rte_cryptodev_scheduler_slaves_get(uint8_t
> scheduler_id, uint8_t *slaves)
>
> sched_ctx = dev->data->dev_private;
>
> - nb_slaves = sched_ctx->nb_slaves;
> + nb_workers = sched_ctx->nb_workers;
>
> - if (slaves && nb_slaves) {
> + if (workers && nb_workers) {
> uint32_t i;
>
> - for (i = 0; i < nb_slaves; i++)
> - slaves[i] = sched_ctx->slaves[i].dev_id;
> + for (i = 0; i < nb_workers; i++)
> + workers[i] = sched_ctx->workers[i].dev_id;
> }
>
> - return (int)nb_slaves;
> + return (int)nb_workers;
> }
>
> int
> diff --git a/drivers/crypto/scheduler/rte_cryptodev_scheduler.h
> b/drivers/crypto/scheduler/rte_cryptodev_scheduler.h
> index 9a72a90ae..88da8368e 100644
> --- a/drivers/crypto/scheduler/rte_cryptodev_scheduler.h
> +++ b/drivers/crypto/scheduler/rte_cryptodev_scheduler.h
> @@ -10,9 +10,9 @@
> *
> * RTE Cryptodev Scheduler Device
> *
> - * The RTE Cryptodev Scheduler Device allows the aggregation of multiple
> (slave)
> + * The RTE Cryptodev Scheduler Device allows the aggregation of multiple
> worker
> * Cryptodevs into a single logical crypto device, and the scheduling the
> - * crypto operations to the slaves based on the mode of the specified mode
> of
> + * crypto operations to the workers based on the mode of the specified
> mode of
> * operation specified and supported. This implementation supports 3
> modes of
> * operation: round robin, packet-size based, and fail-over.
> */
> @@ -25,8 +25,8 @@ extern "C" {
> #endif
>
> /** Maximum number of bonded devices per device */
> -#ifndef RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES
> -#define RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES (8)
> +#ifndef RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS
> +#define RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS (8)
> #endif
>
> /** Maximum number of multi-core worker cores */
> @@ -106,34 +106,33 @@
> rte_cryptodev_scheduler_load_user_scheduler(uint8_t scheduler_id,
> *
> * @param scheduler_id
> * The target scheduler device ID
> - * @param slave_id
> + * @param worker_id
> * Crypto device ID to be attached
> *
> * @return
> - * - 0 if the slave is attached.
> + * - 0 if the worker is attached.
> * - -ENOTSUP if the operation is not supported.
> * - -EBUSY if device is started.
> - * - -ENOMEM if the scheduler's slave list is full.
> + * - -ENOMEM if the scheduler's worker list is full.
> */
> int
> -rte_cryptodev_scheduler_slave_attach(uint8_t scheduler_id, uint8_t
> slave_id);
> +rte_cryptodev_scheduler_worker_attach(uint8_t scheduler_id, uint8_t
> worker_id);
>
> /**
> * Detach a crypto device from the scheduler
> *
> * @param scheduler_id
> * The target scheduler device ID
> - * @param slave_id
> + * @param worker_id
> * Crypto device ID to be detached
> *
> * @return
> - * - 0 if the slave is detached.
> + * - 0 if the worker is detached.
> * - -ENOTSUP if the operation is not supported.
> * - -EBUSY if device is started.
> */
> int
> -rte_cryptodev_scheduler_slave_detach(uint8_t scheduler_id, uint8_t
> slave_id);
> -
> +rte_cryptodev_scheduler_worker_detach(uint8_t scheduler_id, uint8_t
> worker_id);
>
> /**
> * Set the scheduling mode
> @@ -199,21 +198,21 @@ int
> rte_cryptodev_scheduler_ordering_get(uint8_t scheduler_id);
>
> /**
> - * Get the attached slaves' count and/or ID
> + * Get the attached workers' count and/or ID
> *
> * @param scheduler_id
> * The target scheduler device ID
> - * @param slaves
> - * If successful, the function will write back all slaves' device IDs to it.
> + * @param workers
> + * If successful, the function will write back all workers' device IDs to it.
> * This parameter will either be an uint8_t array of
> - * RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES elements or NULL.
> + * RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS elements or NULL.
> *
> * @return
> - * - non-negative number: the number of slaves attached
> + * - non-negative number: the number of workers attached
> * - -ENOTSUP if the operation is not supported.
> */
> int
> -rte_cryptodev_scheduler_slaves_get(uint8_t scheduler_id, uint8_t *slaves);
> +rte_cryptodev_scheduler_workers_get(uint8_t scheduler_id, uint8_t
> *workers);
>
> /**
> * Set the mode specific option
> diff --git a/drivers/crypto/scheduler/rte_cryptodev_scheduler_operations.h
> b/drivers/crypto/scheduler/rte_cryptodev_scheduler_operations.h
> index c43695894..f8726c009 100644
> --- a/drivers/crypto/scheduler/rte_cryptodev_scheduler_operations.h
> +++ b/drivers/crypto/scheduler/rte_cryptodev_scheduler_operations.h
> @@ -11,10 +11,10 @@
> extern "C" {
> #endif
>
> -typedef int (*rte_cryptodev_scheduler_slave_attach_t)(
> - struct rte_cryptodev *dev, uint8_t slave_id);
> -typedef int (*rte_cryptodev_scheduler_slave_detach_t)(
> - struct rte_cryptodev *dev, uint8_t slave_id);
> +typedef int (*rte_cryptodev_scheduler_worker_attach_t)(
> + struct rte_cryptodev *dev, uint8_t worker_id);
> +typedef int (*rte_cryptodev_scheduler_worker_detach_t)(
> + struct rte_cryptodev *dev, uint8_t worker_id);
>
> typedef int (*rte_cryptodev_scheduler_start_t)(struct rte_cryptodev *dev);
> typedef int (*rte_cryptodev_scheduler_stop_t)(struct rte_cryptodev *dev);
> @@ -36,8 +36,8 @@ typedef int
> (*rte_cryptodev_scheduler_config_option_get)(
> void *option);
>
> struct rte_cryptodev_scheduler_ops {
> - rte_cryptodev_scheduler_slave_attach_t slave_attach;
> - rte_cryptodev_scheduler_slave_attach_t slave_detach;
> + rte_cryptodev_scheduler_worker_attach_t worker_attach;
> + rte_cryptodev_scheduler_worker_attach_t worker_detach;
>
> rte_cryptodev_scheduler_start_t scheduler_start;
> rte_cryptodev_scheduler_stop_t scheduler_stop;
> diff --git
> a/drivers/crypto/scheduler/rte_pmd_crypto_scheduler_version.map
> b/drivers/crypto/scheduler/rte_pmd_crypto_scheduler_version.map
> index ca6f102d9..ab7d50562 100644
> --- a/drivers/crypto/scheduler/rte_pmd_crypto_scheduler_version.map
> +++ b/drivers/crypto/scheduler/rte_pmd_crypto_scheduler_version.map
> @@ -8,9 +8,9 @@ DPDK_21 {
> rte_cryptodev_scheduler_option_set;
> rte_cryptodev_scheduler_ordering_get;
> rte_cryptodev_scheduler_ordering_set;
> - rte_cryptodev_scheduler_slave_attach;
> - rte_cryptodev_scheduler_slave_detach;
> - rte_cryptodev_scheduler_slaves_get;
> + rte_cryptodev_scheduler_worker_attach;
> + rte_cryptodev_scheduler_worker_detach;
> + rte_cryptodev_scheduler_workers_get;
>
> local: *;
> };
> diff --git a/drivers/crypto/scheduler/scheduler_failover.c
> b/drivers/crypto/scheduler/scheduler_failover.c
> index 3a023b8ad..844312dd1 100644
> --- a/drivers/crypto/scheduler/scheduler_failover.c
> +++ b/drivers/crypto/scheduler/scheduler_failover.c
> @@ -8,20 +8,20 @@
> #include "rte_cryptodev_scheduler_operations.h"
> #include "scheduler_pmd_private.h"
>
> -#define PRIMARY_SLAVE_IDX 0
> -#define SECONDARY_SLAVE_IDX 1
> -#define NB_FAILOVER_SLAVES 2
> -#define SLAVE_SWITCH_MASK (0x01)
> +#define PRIMARY_WORKER_IDX 0
> +#define SECONDARY_WORKER_IDX 1
> +#define NB_FAILOVER_WORKERS 2
> +#define WORKER_SWITCH_MASK (0x01)
>
> struct fo_scheduler_qp_ctx {
> - struct scheduler_slave primary_slave;
> - struct scheduler_slave secondary_slave;
> + struct scheduler_worker primary_worker;
> + struct scheduler_worker secondary_worker;
>
> uint8_t deq_idx;
> };
>
> static __rte_always_inline uint16_t
> -failover_slave_enqueue(struct scheduler_slave *slave,
> +failover_worker_enqueue(struct scheduler_worker *worker,
> struct rte_crypto_op **ops, uint16_t nb_ops)
> {
> uint16_t i, processed_ops;
> @@ -29,9 +29,9 @@ failover_slave_enqueue(struct scheduler_slave *slave,
> for (i = 0; i < nb_ops && i < 4; i++)
> rte_prefetch0(ops[i]->sym->session);
>
> - processed_ops = rte_cryptodev_enqueue_burst(slave->dev_id,
> - slave->qp_id, ops, nb_ops);
> - slave->nb_inflight_cops += processed_ops;
> + processed_ops = rte_cryptodev_enqueue_burst(worker->dev_id,
> + worker->qp_id, ops, nb_ops);
> + worker->nb_inflight_cops += processed_ops;
>
> return processed_ops;
> }
> @@ -46,11 +46,12 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> if (unlikely(nb_ops == 0))
> return 0;
>
> - enqueued_ops = failover_slave_enqueue(&qp_ctx->primary_slave,
> + enqueued_ops = failover_worker_enqueue(&qp_ctx-
> >primary_worker,
> ops, nb_ops);
>
> if (enqueued_ops < nb_ops)
> - enqueued_ops += failover_slave_enqueue(&qp_ctx-
> >secondary_slave,
> + enqueued_ops += failover_worker_enqueue(
> + &qp_ctx->secondary_worker,
> &ops[enqueued_ops],
> nb_ops - enqueued_ops);
>
> @@ -79,28 +80,28 @@ schedule_dequeue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> {
> struct fo_scheduler_qp_ctx *qp_ctx =
> ((struct scheduler_qp_ctx *)qp)->private_qp_ctx;
> - struct scheduler_slave *slaves[NB_FAILOVER_SLAVES] = {
> - &qp_ctx->primary_slave, &qp_ctx-
> >secondary_slave};
> - struct scheduler_slave *slave = slaves[qp_ctx->deq_idx];
> + struct scheduler_worker *workers[NB_FAILOVER_WORKERS] = {
> + &qp_ctx->primary_worker, &qp_ctx-
> >secondary_worker};
> + struct scheduler_worker *worker = workers[qp_ctx->deq_idx];
> uint16_t nb_deq_ops = 0, nb_deq_ops2 = 0;
>
> - if (slave->nb_inflight_cops) {
> - nb_deq_ops = rte_cryptodev_dequeue_burst(slave-
> >dev_id,
> - slave->qp_id, ops, nb_ops);
> - slave->nb_inflight_cops -= nb_deq_ops;
> + if (worker->nb_inflight_cops) {
> + nb_deq_ops = rte_cryptodev_dequeue_burst(worker-
> >dev_id,
> + worker->qp_id, ops, nb_ops);
> + worker->nb_inflight_cops -= nb_deq_ops;
> }
>
> - qp_ctx->deq_idx = (~qp_ctx->deq_idx) & SLAVE_SWITCH_MASK;
> + qp_ctx->deq_idx = (~qp_ctx->deq_idx) & WORKER_SWITCH_MASK;
>
> if (nb_deq_ops == nb_ops)
> return nb_deq_ops;
>
> - slave = slaves[qp_ctx->deq_idx];
> + worker = workers[qp_ctx->deq_idx];
>
> - if (slave->nb_inflight_cops) {
> - nb_deq_ops2 = rte_cryptodev_dequeue_burst(slave-
> >dev_id,
> - slave->qp_id, &ops[nb_deq_ops], nb_ops -
> nb_deq_ops);
> - slave->nb_inflight_cops -= nb_deq_ops2;
> + if (worker->nb_inflight_cops) {
> + nb_deq_ops2 = rte_cryptodev_dequeue_burst(worker-
> >dev_id,
> + worker->qp_id, &ops[nb_deq_ops], nb_ops -
> nb_deq_ops);
> + worker->nb_inflight_cops -= nb_deq_ops2;
> }
>
> return nb_deq_ops + nb_deq_ops2;
> @@ -119,15 +120,15 @@ schedule_dequeue_ordering(void *qp, struct
> rte_crypto_op **ops,
> }
>
> static int
> -slave_attach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_attach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
>
> static int
> -slave_detach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_detach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
> @@ -138,8 +139,8 @@ scheduler_start(struct rte_cryptodev *dev)
> struct scheduler_ctx *sched_ctx = dev->data->dev_private;
> uint16_t i;
>
> - if (sched_ctx->nb_slaves < 2) {
> - CR_SCHED_LOG(ERR, "Number of slaves shall no less than 2");
> + if (sched_ctx->nb_workers < 2) {
> + CR_SCHED_LOG(ERR, "Number of workers shall no less than
> 2");
> return -ENOMEM;
> }
>
> @@ -156,12 +157,12 @@ scheduler_start(struct rte_cryptodev *dev)
> ((struct scheduler_qp_ctx *)
> dev->data->queue_pairs[i])->private_qp_ctx;
>
> - rte_memcpy(&qp_ctx->primary_slave,
> - &sched_ctx->slaves[PRIMARY_SLAVE_IDX],
> - sizeof(struct scheduler_slave));
> - rte_memcpy(&qp_ctx->secondary_slave,
> - &sched_ctx-
> >slaves[SECONDARY_SLAVE_IDX],
> - sizeof(struct scheduler_slave));
> + rte_memcpy(&qp_ctx->primary_worker,
> + &sched_ctx-
> >workers[PRIMARY_WORKER_IDX],
> + sizeof(struct scheduler_worker));
> + rte_memcpy(&qp_ctx->secondary_worker,
> + &sched_ctx-
> >workers[SECONDARY_WORKER_IDX],
> + sizeof(struct scheduler_worker));
> }
>
> return 0;
> @@ -198,8 +199,8 @@ scheduler_create_private_ctx(__rte_unused struct
> rte_cryptodev *dev)
> }
>
> static struct rte_cryptodev_scheduler_ops scheduler_fo_ops = {
> - slave_attach,
> - slave_detach,
> + worker_attach,
> + worker_detach,
> scheduler_start,
> scheduler_stop,
> scheduler_config_qp,
> @@ -210,8 +211,8 @@ static struct rte_cryptodev_scheduler_ops
> scheduler_fo_ops = {
>
> static struct rte_cryptodev_scheduler fo_scheduler = {
> .name = "failover-scheduler",
> - .description = "scheduler which enqueues to the primary
> slave, "
> - "and only then enqueues to the secondary
> slave "
> + .description = "scheduler which enqueues to the primary
> worker, "
> + "and only then enqueues to the secondary
> worker "
> "upon failing on enqueuing to primary",
> .mode = CDEV_SCHED_MODE_FAILOVER,
> .ops = &scheduler_fo_ops
> diff --git a/drivers/crypto/scheduler/scheduler_multicore.c
> b/drivers/crypto/scheduler/scheduler_multicore.c
> index 2d6790bb3..1e2e8dbf9 100644
> --- a/drivers/crypto/scheduler/scheduler_multicore.c
> +++ b/drivers/crypto/scheduler/scheduler_multicore.c
> @@ -26,8 +26,8 @@ struct mc_scheduler_ctx {
> };
>
> struct mc_scheduler_qp_ctx {
> - struct scheduler_slave
> slaves[RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES];
> - uint32_t nb_slaves;
> + struct scheduler_worker
> workers[RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS];
> + uint32_t nb_workers;
>
> uint32_t last_enq_worker_idx;
> uint32_t last_deq_worker_idx;
> @@ -132,15 +132,15 @@ schedule_dequeue_ordering(void *qp, struct
> rte_crypto_op **ops,
> }
>
> static int
> -slave_attach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_attach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
>
> static int
> -slave_detach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_detach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
> @@ -154,7 +154,7 @@ mc_scheduler_worker(struct rte_cryptodev *dev)
> struct rte_ring *deq_ring;
> uint32_t core_id = rte_lcore_id();
> int i, worker_idx = -1;
> - struct scheduler_slave *slave;
> + struct scheduler_worker *worker;
> struct rte_crypto_op *enq_ops[MC_SCHED_BUFFER_SIZE];
> struct rte_crypto_op *deq_ops[MC_SCHED_BUFFER_SIZE];
> uint16_t processed_ops;
> @@ -177,15 +177,16 @@ mc_scheduler_worker(struct rte_cryptodev *dev)
> return -1;
> }
>
> - slave = &sched_ctx->slaves[worker_idx];
> + worker = &sched_ctx->workers[worker_idx];
> enq_ring = mc_ctx->sched_enq_ring[worker_idx];
> deq_ring = mc_ctx->sched_deq_ring[worker_idx];
>
> while (!mc_ctx->stop_signal) {
> if (pending_enq_ops) {
> processed_ops =
> - rte_cryptodev_enqueue_burst(slave-
> >dev_id,
> - slave->qp_id,
> &enq_ops[pending_enq_ops_idx],
> + rte_cryptodev_enqueue_burst(worker-
> >dev_id,
> + worker->qp_id,
> + &enq_ops[pending_enq_ops_idx],
> pending_enq_ops);
> pending_enq_ops -= processed_ops;
> pending_enq_ops_idx += processed_ops;
> @@ -195,8 +196,8 @@ mc_scheduler_worker(struct rte_cryptodev *dev)
>
> MC_SCHED_BUFFER_SIZE, NULL);
> if (processed_ops) {
> pending_enq_ops_idx =
> rte_cryptodev_enqueue_burst(
> - slave->dev_id, slave-
> >qp_id,
> - enq_ops,
> processed_ops);
> + worker->dev_id, worker-
> >qp_id,
> + enq_ops, processed_ops);
> pending_enq_ops = processed_ops -
> pending_enq_ops_idx;
> inflight_ops += pending_enq_ops_idx;
> }
> @@ -209,8 +210,9 @@ mc_scheduler_worker(struct rte_cryptodev *dev)
> pending_deq_ops -= processed_ops;
> pending_deq_ops_idx += processed_ops;
> } else if (inflight_ops) {
> - processed_ops =
> rte_cryptodev_dequeue_burst(slave->dev_id,
> - slave->qp_id, deq_ops,
> MC_SCHED_BUFFER_SIZE);
> + processed_ops = rte_cryptodev_dequeue_burst(
> + worker->dev_id, worker->qp_id,
> deq_ops,
> + MC_SCHED_BUFFER_SIZE);
> if (processed_ops) {
> inflight_ops -= processed_ops;
> if (reordering_enabled) {
> @@ -264,16 +266,16 @@ scheduler_start(struct rte_cryptodev *dev)
> qp_ctx->private_qp_ctx;
> uint32_t j;
>
> - memset(mc_qp_ctx->slaves, 0,
> -
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES *
> - sizeof(struct scheduler_slave));
> - for (j = 0; j < sched_ctx->nb_slaves; j++) {
> - mc_qp_ctx->slaves[j].dev_id =
> - sched_ctx->slaves[j].dev_id;
> - mc_qp_ctx->slaves[j].qp_id = i;
> + memset(mc_qp_ctx->workers, 0,
> +
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS *
> + sizeof(struct scheduler_worker));
> + for (j = 0; j < sched_ctx->nb_workers; j++) {
> + mc_qp_ctx->workers[j].dev_id =
> + sched_ctx->workers[j].dev_id;
> + mc_qp_ctx->workers[j].qp_id = i;
> }
>
> - mc_qp_ctx->nb_slaves = sched_ctx->nb_slaves;
> + mc_qp_ctx->nb_workers = sched_ctx->nb_workers;
>
> mc_qp_ctx->last_enq_worker_idx = 0;
> mc_qp_ctx->last_deq_worker_idx = 0;
> @@ -347,7 +349,7 @@ scheduler_create_private_ctx(struct rte_cryptodev
> *dev)
> mc_ctx->sched_enq_ring[i] = rte_ring_lookup(r_name);
> if (!mc_ctx->sched_enq_ring[i]) {
> mc_ctx->sched_enq_ring[i] =
> rte_ring_create(r_name,
> - PER_SLAVE_BUFF_SIZE,
> + PER_WORKER_BUFF_SIZE,
> rte_socket_id(),
> RING_F_SC_DEQ |
> RING_F_SP_ENQ);
> if (!mc_ctx->sched_enq_ring[i]) {
> @@ -361,7 +363,7 @@ scheduler_create_private_ctx(struct rte_cryptodev
> *dev)
> mc_ctx->sched_deq_ring[i] = rte_ring_lookup(r_name);
> if (!mc_ctx->sched_deq_ring[i]) {
> mc_ctx->sched_deq_ring[i] =
> rte_ring_create(r_name,
> - PER_SLAVE_BUFF_SIZE,
> + PER_WORKER_BUFF_SIZE,
> rte_socket_id(),
> RING_F_SC_DEQ |
> RING_F_SP_ENQ);
> if (!mc_ctx->sched_deq_ring[i]) {
> @@ -387,8 +389,8 @@ scheduler_create_private_ctx(struct rte_cryptodev
> *dev)
> }
>
> static struct rte_cryptodev_scheduler_ops scheduler_mc_ops = {
> - slave_attach,
> - slave_detach,
> + worker_attach,
> + worker_detach,
> scheduler_start,
> scheduler_stop,
> scheduler_config_qp,
> diff --git a/drivers/crypto/scheduler/scheduler_pkt_size_distr.c
> b/drivers/crypto/scheduler/scheduler_pkt_size_distr.c
> index 45c8dceb4..57e330a74 100644
> --- a/drivers/crypto/scheduler/scheduler_pkt_size_distr.c
> +++ b/drivers/crypto/scheduler/scheduler_pkt_size_distr.c
> @@ -9,10 +9,10 @@
> #include "scheduler_pmd_private.h"
>
> #define DEF_PKT_SIZE_THRESHOLD (0xffffff80)
> -#define SLAVE_IDX_SWITCH_MASK (0x01)
> -#define PRIMARY_SLAVE_IDX 0
> -#define SECONDARY_SLAVE_IDX 1
> -#define NB_PKT_SIZE_SLAVES 2
> +#define WORKER_IDX_SWITCH_MASK (0x01)
> +#define PRIMARY_WORKER_IDX 0
> +#define SECONDARY_WORKER_IDX 1
> +#define NB_PKT_SIZE_WORKERS 2
>
> /** pkt size based scheduler context */
> struct psd_scheduler_ctx {
> @@ -21,15 +21,15 @@ struct psd_scheduler_ctx {
>
> /** pkt size based scheduler queue pair context */
> struct psd_scheduler_qp_ctx {
> - struct scheduler_slave primary_slave;
> - struct scheduler_slave secondary_slave;
> + struct scheduler_worker primary_worker;
> + struct scheduler_worker secondary_worker;
> uint32_t threshold;
> uint8_t deq_idx;
> } __rte_cache_aligned;
>
> /** scheduling operation variables' wrapping */
> struct psd_schedule_op {
> - uint8_t slave_idx;
> + uint8_t worker_idx;
> uint16_t pos;
> };
>
> @@ -38,13 +38,13 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> {
> struct scheduler_qp_ctx *qp_ctx = qp;
> struct psd_scheduler_qp_ctx *psd_qp_ctx = qp_ctx-
> >private_qp_ctx;
> - struct rte_crypto_op *sched_ops[NB_PKT_SIZE_SLAVES][nb_ops];
> - uint32_t in_flight_ops[NB_PKT_SIZE_SLAVES] = {
> - psd_qp_ctx->primary_slave.nb_inflight_cops,
> - psd_qp_ctx->secondary_slave.nb_inflight_cops
> + struct rte_crypto_op *sched_ops[NB_PKT_SIZE_WORKERS][nb_ops];
> + uint32_t in_flight_ops[NB_PKT_SIZE_WORKERS] = {
> + psd_qp_ctx->primary_worker.nb_inflight_cops,
> + psd_qp_ctx->secondary_worker.nb_inflight_cops
> };
> - struct psd_schedule_op enq_ops[NB_PKT_SIZE_SLAVES] = {
> - {PRIMARY_SLAVE_IDX, 0}, {SECONDARY_SLAVE_IDX, 0}
> + struct psd_schedule_op enq_ops[NB_PKT_SIZE_WORKERS] = {
> + {PRIMARY_WORKER_IDX, 0}, {SECONDARY_WORKER_IDX, 0}
> };
> struct psd_schedule_op *p_enq_op;
> uint16_t i, processed_ops_pri = 0, processed_ops_sec = 0;
> @@ -80,13 +80,13 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> /* stop schedule cops before the queue is full, this shall
> * prevent the failed enqueue
> */
> - if (p_enq_op->pos + in_flight_ops[p_enq_op->slave_idx] ==
> + if (p_enq_op->pos + in_flight_ops[p_enq_op->worker_idx]
> ==
> qp_ctx->max_nb_objs) {
> i = nb_ops;
> break;
> }
>
> - sched_ops[p_enq_op->slave_idx][p_enq_op->pos] = ops[i];
> + sched_ops[p_enq_op->worker_idx][p_enq_op->pos] =
> ops[i];
> p_enq_op->pos++;
>
> job_len = ops[i+1]->sym->cipher.data.length;
> @@ -94,13 +94,13 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> ops[i+1]->sym->auth.data.length;
> p_enq_op = &enq_ops[!(job_len & psd_qp_ctx->threshold)];
>
> - if (p_enq_op->pos + in_flight_ops[p_enq_op->slave_idx] ==
> + if (p_enq_op->pos + in_flight_ops[p_enq_op->worker_idx]
> ==
> qp_ctx->max_nb_objs) {
> i = nb_ops;
> break;
> }
>
> - sched_ops[p_enq_op->slave_idx][p_enq_op->pos] =
> ops[i+1];
> + sched_ops[p_enq_op->worker_idx][p_enq_op->pos] =
> ops[i+1];
> p_enq_op->pos++;
>
> job_len = ops[i+2]->sym->cipher.data.length;
> @@ -108,13 +108,13 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> ops[i+2]->sym->auth.data.length;
> p_enq_op = &enq_ops[!(job_len & psd_qp_ctx->threshold)];
>
> - if (p_enq_op->pos + in_flight_ops[p_enq_op->slave_idx] ==
> + if (p_enq_op->pos + in_flight_ops[p_enq_op->worker_idx]
> ==
> qp_ctx->max_nb_objs) {
> i = nb_ops;
> break;
> }
>
> - sched_ops[p_enq_op->slave_idx][p_enq_op->pos] =
> ops[i+2];
> + sched_ops[p_enq_op->worker_idx][p_enq_op->pos] =
> ops[i+2];
> p_enq_op->pos++;
>
> job_len = ops[i+3]->sym->cipher.data.length;
> @@ -122,13 +122,13 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> ops[i+3]->sym->auth.data.length;
> p_enq_op = &enq_ops[!(job_len & psd_qp_ctx->threshold)];
>
> - if (p_enq_op->pos + in_flight_ops[p_enq_op->slave_idx] ==
> + if (p_enq_op->pos + in_flight_ops[p_enq_op->worker_idx]
> ==
> qp_ctx->max_nb_objs) {
> i = nb_ops;
> break;
> }
>
> - sched_ops[p_enq_op->slave_idx][p_enq_op->pos] =
> ops[i+3];
> + sched_ops[p_enq_op->worker_idx][p_enq_op->pos] =
> ops[i+3];
> p_enq_op->pos++;
> }
>
> @@ -138,34 +138,34 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> ops[i]->sym->auth.data.length;
> p_enq_op = &enq_ops[!(job_len & psd_qp_ctx->threshold)];
>
> - if (p_enq_op->pos + in_flight_ops[p_enq_op->slave_idx] ==
> + if (p_enq_op->pos + in_flight_ops[p_enq_op->worker_idx]
> ==
> qp_ctx->max_nb_objs) {
> i = nb_ops;
> break;
> }
>
> - sched_ops[p_enq_op->slave_idx][p_enq_op->pos] = ops[i];
> + sched_ops[p_enq_op->worker_idx][p_enq_op->pos] =
> ops[i];
> p_enq_op->pos++;
> }
>
> processed_ops_pri = rte_cryptodev_enqueue_burst(
> - psd_qp_ctx->primary_slave.dev_id,
> - psd_qp_ctx->primary_slave.qp_id,
> - sched_ops[PRIMARY_SLAVE_IDX],
> - enq_ops[PRIMARY_SLAVE_IDX].pos);
> - /* enqueue shall not fail as the slave queue is monitored */
> - RTE_ASSERT(processed_ops_pri ==
> enq_ops[PRIMARY_SLAVE_IDX].pos);
> + psd_qp_ctx->primary_worker.dev_id,
> + psd_qp_ctx->primary_worker.qp_id,
> + sched_ops[PRIMARY_WORKER_IDX],
> + enq_ops[PRIMARY_WORKER_IDX].pos);
> + /* enqueue shall not fail as the worker queue is monitored */
> + RTE_ASSERT(processed_ops_pri ==
> enq_ops[PRIMARY_WORKER_IDX].pos);
>
> - psd_qp_ctx->primary_slave.nb_inflight_cops += processed_ops_pri;
> + psd_qp_ctx->primary_worker.nb_inflight_cops +=
> processed_ops_pri;
>
> processed_ops_sec = rte_cryptodev_enqueue_burst(
> - psd_qp_ctx->secondary_slave.dev_id,
> - psd_qp_ctx->secondary_slave.qp_id,
> - sched_ops[SECONDARY_SLAVE_IDX],
> - enq_ops[SECONDARY_SLAVE_IDX].pos);
> - RTE_ASSERT(processed_ops_sec ==
> enq_ops[SECONDARY_SLAVE_IDX].pos);
> + psd_qp_ctx->secondary_worker.dev_id,
> + psd_qp_ctx->secondary_worker.qp_id,
> + sched_ops[SECONDARY_WORKER_IDX],
> + enq_ops[SECONDARY_WORKER_IDX].pos);
> + RTE_ASSERT(processed_ops_sec ==
> enq_ops[SECONDARY_WORKER_IDX].pos);
>
> - psd_qp_ctx->secondary_slave.nb_inflight_cops +=
> processed_ops_sec;
> + psd_qp_ctx->secondary_worker.nb_inflight_cops +=
> processed_ops_sec;
>
> return processed_ops_pri + processed_ops_sec;
> }
> @@ -191,33 +191,33 @@ schedule_dequeue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> {
> struct psd_scheduler_qp_ctx *qp_ctx =
> ((struct scheduler_qp_ctx *)qp)->private_qp_ctx;
> - struct scheduler_slave *slaves[NB_PKT_SIZE_SLAVES] = {
> - &qp_ctx->primary_slave, &qp_ctx-
> >secondary_slave};
> - struct scheduler_slave *slave = slaves[qp_ctx->deq_idx];
> + struct scheduler_worker *workers[NB_PKT_SIZE_WORKERS] = {
> + &qp_ctx->primary_worker, &qp_ctx-
> >secondary_worker};
> + struct scheduler_worker *worker = workers[qp_ctx->deq_idx];
> uint16_t nb_deq_ops_pri = 0, nb_deq_ops_sec = 0;
>
> - if (slave->nb_inflight_cops) {
> - nb_deq_ops_pri = rte_cryptodev_dequeue_burst(slave-
> >dev_id,
> - slave->qp_id, ops, nb_ops);
> - slave->nb_inflight_cops -= nb_deq_ops_pri;
> + if (worker->nb_inflight_cops) {
> + nb_deq_ops_pri = rte_cryptodev_dequeue_burst(worker-
> >dev_id,
> + worker->qp_id, ops, nb_ops);
> + worker->nb_inflight_cops -= nb_deq_ops_pri;
> }
>
> - qp_ctx->deq_idx = (~qp_ctx->deq_idx) &
> SLAVE_IDX_SWITCH_MASK;
> + qp_ctx->deq_idx = (~qp_ctx->deq_idx) &
> WORKER_IDX_SWITCH_MASK;
>
> if (nb_deq_ops_pri == nb_ops)
> return nb_deq_ops_pri;
>
> - slave = slaves[qp_ctx->deq_idx];
> + worker = workers[qp_ctx->deq_idx];
>
> - if (slave->nb_inflight_cops) {
> - nb_deq_ops_sec = rte_cryptodev_dequeue_burst(slave-
> >dev_id,
> - slave->qp_id, &ops[nb_deq_ops_pri],
> + if (worker->nb_inflight_cops) {
> + nb_deq_ops_sec = rte_cryptodev_dequeue_burst(worker-
> >dev_id,
> + worker->qp_id, &ops[nb_deq_ops_pri],
> nb_ops - nb_deq_ops_pri);
> - slave->nb_inflight_cops -= nb_deq_ops_sec;
> + worker->nb_inflight_cops -= nb_deq_ops_sec;
>
> - if (!slave->nb_inflight_cops)
> + if (!worker->nb_inflight_cops)
> qp_ctx->deq_idx = (~qp_ctx->deq_idx) &
> - SLAVE_IDX_SWITCH_MASK;
> + WORKER_IDX_SWITCH_MASK;
> }
>
> return nb_deq_ops_pri + nb_deq_ops_sec;
> @@ -236,15 +236,15 @@ schedule_dequeue_ordering(void *qp, struct
> rte_crypto_op **ops,
> }
>
> static int
> -slave_attach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_attach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
>
> static int
> -slave_detach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_detach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
> @@ -256,9 +256,9 @@ scheduler_start(struct rte_cryptodev *dev)
> struct psd_scheduler_ctx *psd_ctx = sched_ctx->private_ctx;
> uint16_t i;
>
> - /* for packet size based scheduler, nb_slaves have to >= 2 */
> - if (sched_ctx->nb_slaves < NB_PKT_SIZE_SLAVES) {
> - CR_SCHED_LOG(ERR, "not enough slaves to start");
> + /* for packet size based scheduler, nb_workers have to >= 2 */
> + if (sched_ctx->nb_workers < NB_PKT_SIZE_WORKERS) {
> + CR_SCHED_LOG(ERR, "not enough workers to start");
> return -1;
> }
>
> @@ -267,15 +267,15 @@ scheduler_start(struct rte_cryptodev *dev)
> struct psd_scheduler_qp_ctx *ps_qp_ctx =
> qp_ctx->private_qp_ctx;
>
> - ps_qp_ctx->primary_slave.dev_id =
> - sched_ctx-
> >slaves[PRIMARY_SLAVE_IDX].dev_id;
> - ps_qp_ctx->primary_slave.qp_id = i;
> - ps_qp_ctx->primary_slave.nb_inflight_cops = 0;
> + ps_qp_ctx->primary_worker.dev_id =
> + sched_ctx-
> >workers[PRIMARY_WORKER_IDX].dev_id;
> + ps_qp_ctx->primary_worker.qp_id = i;
> + ps_qp_ctx->primary_worker.nb_inflight_cops = 0;
>
> - ps_qp_ctx->secondary_slave.dev_id =
> - sched_ctx-
> >slaves[SECONDARY_SLAVE_IDX].dev_id;
> - ps_qp_ctx->secondary_slave.qp_id = i;
> - ps_qp_ctx->secondary_slave.nb_inflight_cops = 0;
> + ps_qp_ctx->secondary_worker.dev_id =
> + sched_ctx-
> >workers[SECONDARY_WORKER_IDX].dev_id;
> + ps_qp_ctx->secondary_worker.qp_id = i;
> + ps_qp_ctx->secondary_worker.nb_inflight_cops = 0;
>
> ps_qp_ctx->threshold = psd_ctx->threshold;
> }
> @@ -300,9 +300,9 @@ scheduler_stop(struct rte_cryptodev *dev)
> struct scheduler_qp_ctx *qp_ctx = dev->data-
> >queue_pairs[i];
> struct psd_scheduler_qp_ctx *ps_qp_ctx = qp_ctx-
> >private_qp_ctx;
>
> - if (ps_qp_ctx->primary_slave.nb_inflight_cops +
> - ps_qp_ctx-
> >secondary_slave.nb_inflight_cops) {
> - CR_SCHED_LOG(ERR, "Some crypto ops left in slave
> queue");
> + if (ps_qp_ctx->primary_worker.nb_inflight_cops +
> + ps_qp_ctx-
> >secondary_worker.nb_inflight_cops) {
> + CR_SCHED_LOG(ERR, "Some crypto ops left in worker
> queue");
> return -1;
> }
> }
> @@ -399,8 +399,8 @@ scheduler_option_get(struct rte_cryptodev *dev,
> uint32_t option_type,
> }
>
> static struct rte_cryptodev_scheduler_ops scheduler_ps_ops = {
> - slave_attach,
> - slave_detach,
> + worker_attach,
> + worker_detach,
> scheduler_start,
> scheduler_stop,
> scheduler_config_qp,
> diff --git a/drivers/crypto/scheduler/scheduler_pmd.c
> b/drivers/crypto/scheduler/scheduler_pmd.c
> index a1632a2b9..632197833 100644
> --- a/drivers/crypto/scheduler/scheduler_pmd.c
> +++ b/drivers/crypto/scheduler/scheduler_pmd.c
> @@ -18,18 +18,18 @@ uint8_t cryptodev_scheduler_driver_id;
>
> struct scheduler_init_params {
> struct rte_cryptodev_pmd_init_params def_p;
> - uint32_t nb_slaves;
> + uint32_t nb_workers;
> enum rte_cryptodev_scheduler_mode mode;
> char
> mode_param_str[RTE_CRYPTODEV_SCHEDULER_NAME_MAX_LEN];
> uint32_t enable_ordering;
> uint16_t wc_pool[RTE_MAX_LCORE];
> uint16_t nb_wc;
> - char slave_names[RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES]
> + char
> worker_names[RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS]
> [RTE_CRYPTODEV_SCHEDULER_NAME_MAX_LEN];
> };
>
> #define RTE_CRYPTODEV_VDEV_NAME ("name")
> -#define RTE_CRYPTODEV_VDEV_SLAVE ("slave")
> +#define RTE_CRYPTODEV_VDEV_WORKER ("worker")
> #define RTE_CRYPTODEV_VDEV_MODE ("mode")
> #define RTE_CRYPTODEV_VDEV_MODE_PARAM
> ("mode_param")
> #define RTE_CRYPTODEV_VDEV_ORDERING ("ordering")
> @@ -40,7 +40,7 @@ struct scheduler_init_params {
>
> static const char * const scheduler_valid_params[] = {
> RTE_CRYPTODEV_VDEV_NAME,
> - RTE_CRYPTODEV_VDEV_SLAVE,
> + RTE_CRYPTODEV_VDEV_WORKER,
> RTE_CRYPTODEV_VDEV_MODE,
> RTE_CRYPTODEV_VDEV_MODE_PARAM,
> RTE_CRYPTODEV_VDEV_ORDERING,
> @@ -193,31 +193,31 @@ cryptodev_scheduler_create(const char *name,
> break;
> }
>
> - for (i = 0; i < init_params->nb_slaves; i++) {
> - sched_ctx->init_slave_names[sched_ctx->nb_init_slaves] =
> + for (i = 0; i < init_params->nb_workers; i++) {
> + sched_ctx->init_worker_names[sched_ctx-
> >nb_init_workers] =
> rte_zmalloc_socket(
> NULL,
>
> RTE_CRYPTODEV_SCHEDULER_NAME_MAX_LEN, 0,
> SOCKET_ID_ANY);
>
> - if (!sched_ctx->init_slave_names[
> - sched_ctx->nb_init_slaves]) {
> + if (!sched_ctx->init_worker_names[
> + sched_ctx->nb_init_workers]) {
> CR_SCHED_LOG(ERR, "driver %s: Insufficient
> memory",
> name);
> return -ENOMEM;
> }
>
> - strncpy(sched_ctx->init_slave_names[
> - sched_ctx->nb_init_slaves],
> - init_params->slave_names[i],
> + strncpy(sched_ctx->init_worker_names[
> + sched_ctx->nb_init_workers],
> + init_params->worker_names[i],
>
> RTE_CRYPTODEV_SCHEDULER_NAME_MAX_LEN - 1);
>
> - sched_ctx->nb_init_slaves++;
> + sched_ctx->nb_init_workers++;
> }
>
> /*
> * Initialize capabilities structure as an empty structure,
> - * in case device information is requested when no slaves are
> attached
> + * in case device information is requested when no workers are
> attached
> */
> sched_ctx->capabilities = rte_zmalloc_socket(NULL,
> sizeof(struct rte_cryptodev_capabilities),
> @@ -249,12 +249,12 @@ cryptodev_scheduler_remove(struct
> rte_vdev_device *vdev)
>
> sched_ctx = dev->data->dev_private;
>
> - if (sched_ctx->nb_slaves) {
> + if (sched_ctx->nb_workers) {
> uint32_t i;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++)
> - rte_cryptodev_scheduler_slave_detach(dev->data-
> >dev_id,
> - sched_ctx->slaves[i].dev_id);
> + for (i = 0; i < sched_ctx->nb_workers; i++)
> + rte_cryptodev_scheduler_worker_detach(dev-
> >data->dev_id,
> + sched_ctx->workers[i].dev_id);
> }
>
> return rte_cryptodev_pmd_destroy(dev);
> @@ -374,19 +374,19 @@ parse_name_arg(const char *key __rte_unused,
> return 0;
> }
>
> -/** Parse slave */
> +/** Parse worker */
> static int
> -parse_slave_arg(const char *key __rte_unused,
> +parse_worker_arg(const char *key __rte_unused,
> const char *value, void *extra_args)
> {
> struct scheduler_init_params *param = extra_args;
>
> - if (param->nb_slaves >=
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES) {
> - CR_SCHED_LOG(ERR, "Too many slaves.");
> + if (param->nb_workers >=
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS) {
> + CR_SCHED_LOG(ERR, "Too many workers.");
> return -ENOMEM;
> }
>
> - strncpy(param->slave_names[param->nb_slaves++], value,
> + strncpy(param->worker_names[param->nb_workers++], value,
> RTE_CRYPTODEV_SCHEDULER_NAME_MAX_LEN - 1);
>
> return 0;
> @@ -498,8 +498,8 @@ scheduler_parse_init_params(struct
> scheduler_init_params *params,
> if (ret < 0)
> goto free_kvlist;
>
> - ret = rte_kvargs_process(kvlist,
> RTE_CRYPTODEV_VDEV_SLAVE,
> - &parse_slave_arg, params);
> + ret = rte_kvargs_process(kvlist,
> RTE_CRYPTODEV_VDEV_WORKER,
> + &parse_worker_arg, params);
> if (ret < 0)
> goto free_kvlist;
>
> @@ -534,10 +534,10 @@ cryptodev_scheduler_probe(struct
> rte_vdev_device *vdev)
> rte_socket_id(),
>
> RTE_CRYPTODEV_PMD_DEFAULT_MAX_NB_QUEUE_PAIRS
> },
> - .nb_slaves = 0,
> + .nb_workers = 0,
> .mode = CDEV_SCHED_MODE_NOT_SET,
> .enable_ordering = 0,
> - .slave_names = { {0} }
> + .worker_names = { {0} }
> };
> const char *name;
>
> @@ -566,7 +566,7 @@
> RTE_PMD_REGISTER_VDEV(CRYPTODEV_NAME_SCHEDULER_PMD,
>
> RTE_PMD_REGISTER_PARAM_STRING(CRYPTODEV_NAME_SCHEDULER_PM
> D,
> "max_nb_queue_pairs=<int> "
> "socket_id=<int> "
> - "slave=<name>");
> + "worker=<name>");
> RTE_PMD_REGISTER_CRYPTO_DRIVER(scheduler_crypto_drv,
> cryptodev_scheduler_pmd_drv.driver,
> cryptodev_scheduler_driver_id);
> diff --git a/drivers/crypto/scheduler/scheduler_pmd_ops.c
> b/drivers/crypto/scheduler/scheduler_pmd_ops.c
> index 14e5a3712..cb125e802 100644
> --- a/drivers/crypto/scheduler/scheduler_pmd_ops.c
> +++ b/drivers/crypto/scheduler/scheduler_pmd_ops.c
> @@ -12,43 +12,43 @@
>
> #include "scheduler_pmd_private.h"
>
> -/** attaching the slaves predefined by scheduler's EAL options */
> +/** attaching the workers predefined by scheduler's EAL options */
> static int
> -scheduler_attach_init_slave(struct rte_cryptodev *dev)
> +scheduler_attach_init_worker(struct rte_cryptodev *dev)
> {
> struct scheduler_ctx *sched_ctx = dev->data->dev_private;
> uint8_t scheduler_id = dev->data->dev_id;
> int i;
>
> - for (i = sched_ctx->nb_init_slaves - 1; i >= 0; i--) {
> - const char *dev_name = sched_ctx->init_slave_names[i];
> - struct rte_cryptodev *slave_dev =
> + for (i = sched_ctx->nb_init_workers - 1; i >= 0; i--) {
> + const char *dev_name = sched_ctx->init_worker_names[i];
> + struct rte_cryptodev *worker_dev =
>
> rte_cryptodev_pmd_get_named_dev(dev_name);
> int status;
>
> - if (!slave_dev) {
> - CR_SCHED_LOG(ERR, "Failed to locate slave dev %s",
> + if (!worker_dev) {
> + CR_SCHED_LOG(ERR, "Failed to locate worker
> dev %s",
> dev_name);
> return -EINVAL;
> }
>
> - status = rte_cryptodev_scheduler_slave_attach(
> - scheduler_id, slave_dev->data->dev_id);
> + status = rte_cryptodev_scheduler_worker_attach(
> + scheduler_id, worker_dev->data->dev_id);
>
> if (status < 0) {
> - CR_SCHED_LOG(ERR, "Failed to attach slave
> cryptodev %u",
> - slave_dev->data->dev_id);
> + CR_SCHED_LOG(ERR, "Failed to attach worker
> cryptodev %u",
> + worker_dev->data->dev_id);
> return status;
> }
>
> - CR_SCHED_LOG(INFO, "Scheduler %s attached slave %s",
> + CR_SCHED_LOG(INFO, "Scheduler %s attached worker %s",
> dev->data->name,
> - sched_ctx->init_slave_names[i]);
> + sched_ctx->init_worker_names[i]);
>
> - rte_free(sched_ctx->init_slave_names[i]);
> - sched_ctx->init_slave_names[i] = NULL;
> + rte_free(sched_ctx->init_worker_names[i]);
> + sched_ctx->init_worker_names[i] = NULL;
>
> - sched_ctx->nb_init_slaves -= 1;
> + sched_ctx->nb_init_workers -= 1;
> }
>
> return 0;
> @@ -62,17 +62,17 @@ scheduler_pmd_config(struct rte_cryptodev *dev,
> uint32_t i;
> int ret;
>
> - /* although scheduler_attach_init_slave presents multiple times,
> + /* although scheduler_attach_init_worker presents multiple times,
> * there will be only 1 meaningful execution.
> */
> - ret = scheduler_attach_init_slave(dev);
> + ret = scheduler_attach_init_worker(dev);
> if (ret < 0)
> return ret;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
>
> - ret = rte_cryptodev_configure(slave_dev_id, config);
> + ret = rte_cryptodev_configure(worker_dev_id, config);
> if (ret < 0)
> break;
> }
> @@ -89,7 +89,7 @@ update_order_ring(struct rte_cryptodev *dev, uint16_t
> qp_id)
> if (sched_ctx->reordering_enabled) {
> char order_ring_name[RTE_CRYPTODEV_NAME_MAX_LEN];
> uint32_t buff_size = rte_align32pow2(
> - sched_ctx->nb_slaves * PER_SLAVE_BUFF_SIZE);
> + sched_ctx->nb_workers * PER_WORKER_BUFF_SIZE);
>
> if (qp_ctx->order_ring) {
> rte_ring_free(qp_ctx->order_ring);
> @@ -135,10 +135,10 @@ scheduler_pmd_start(struct rte_cryptodev *dev)
> if (dev->data->dev_started)
> return 0;
>
> - /* although scheduler_attach_init_slave presents multiple times,
> + /* although scheduler_attach_init_worker presents multiple times,
> * there will be only 1 meaningful execution.
> */
> - ret = scheduler_attach_init_slave(dev);
> + ret = scheduler_attach_init_worker(dev);
> if (ret < 0)
> return ret;
>
> @@ -155,18 +155,18 @@ scheduler_pmd_start(struct rte_cryptodev *dev)
> return -1;
> }
>
> - if (!sched_ctx->nb_slaves) {
> - CR_SCHED_LOG(ERR, "No slave in the scheduler");
> + if (!sched_ctx->nb_workers) {
> + CR_SCHED_LOG(ERR, "No worker in the scheduler");
> return -1;
> }
>
> - RTE_FUNC_PTR_OR_ERR_RET(*sched_ctx->ops.slave_attach, -
> ENOTSUP);
> + RTE_FUNC_PTR_OR_ERR_RET(*sched_ctx->ops.worker_attach, -
> ENOTSUP);
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
>
> - if ((*sched_ctx->ops.slave_attach)(dev, slave_dev_id) < 0) {
> - CR_SCHED_LOG(ERR, "Failed to attach slave");
> + if ((*sched_ctx->ops.worker_attach)(dev, worker_dev_id) <
> 0) {
> + CR_SCHED_LOG(ERR, "Failed to attach worker");
> return -ENOTSUP;
> }
> }
> @@ -178,16 +178,16 @@ scheduler_pmd_start(struct rte_cryptodev *dev)
> return -1;
> }
>
> - /* start all slaves */
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev *slave_dev =
> - rte_cryptodev_pmd_get_dev(slave_dev_id);
> + /* start all workers */
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev *worker_dev =
> +
> rte_cryptodev_pmd_get_dev(worker_dev_id);
>
> - ret = (*slave_dev->dev_ops->dev_start)(slave_dev);
> + ret = (*worker_dev->dev_ops->dev_start)(worker_dev);
> if (ret < 0) {
> - CR_SCHED_LOG(ERR, "Failed to start slave dev %u",
> - slave_dev_id);
> + CR_SCHED_LOG(ERR, "Failed to start worker dev %u",
> + worker_dev_id);
> return ret;
> }
> }
> @@ -205,23 +205,23 @@ scheduler_pmd_stop(struct rte_cryptodev *dev)
> if (!dev->data->dev_started)
> return;
>
> - /* stop all slaves first */
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev *slave_dev =
> - rte_cryptodev_pmd_get_dev(slave_dev_id);
> + /* stop all workers first */
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev *worker_dev =
> +
> rte_cryptodev_pmd_get_dev(worker_dev_id);
>
> - (*slave_dev->dev_ops->dev_stop)(slave_dev);
> + (*worker_dev->dev_ops->dev_stop)(worker_dev);
> }
>
> if (*sched_ctx->ops.scheduler_stop)
> (*sched_ctx->ops.scheduler_stop)(dev);
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
>
> - if (*sched_ctx->ops.slave_detach)
> - (*sched_ctx->ops.slave_detach)(dev, slave_dev_id);
> + if (*sched_ctx->ops.worker_detach)
> + (*sched_ctx->ops.worker_detach)(dev,
> worker_dev_id);
> }
> }
>
> @@ -237,13 +237,13 @@ scheduler_pmd_close(struct rte_cryptodev *dev)
> if (dev->data->dev_started)
> return -EBUSY;
>
> - /* close all slaves first */
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev *slave_dev =
> - rte_cryptodev_pmd_get_dev(slave_dev_id);
> + /* close all workers first */
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev *worker_dev =
> +
> rte_cryptodev_pmd_get_dev(worker_dev_id);
>
> - ret = (*slave_dev->dev_ops->dev_close)(slave_dev);
> + ret = (*worker_dev->dev_ops->dev_close)(worker_dev);
> if (ret < 0)
> return ret;
> }
> @@ -283,19 +283,19 @@ scheduler_pmd_stats_get(struct rte_cryptodev
> *dev,
> struct scheduler_ctx *sched_ctx = dev->data->dev_private;
> uint32_t i;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev *slave_dev =
> - rte_cryptodev_pmd_get_dev(slave_dev_id);
> - struct rte_cryptodev_stats slave_stats = {0};
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev *worker_dev =
> +
> rte_cryptodev_pmd_get_dev(worker_dev_id);
> + struct rte_cryptodev_stats worker_stats = {0};
>
> - (*slave_dev->dev_ops->stats_get)(slave_dev, &slave_stats);
> + (*worker_dev->dev_ops->stats_get)(worker_dev,
> &worker_stats);
>
> - stats->enqueued_count += slave_stats.enqueued_count;
> - stats->dequeued_count += slave_stats.dequeued_count;
> + stats->enqueued_count += worker_stats.enqueued_count;
> + stats->dequeued_count += worker_stats.dequeued_count;
>
> - stats->enqueue_err_count +=
> slave_stats.enqueue_err_count;
> - stats->dequeue_err_count +=
> slave_stats.dequeue_err_count;
> + stats->enqueue_err_count +=
> worker_stats.enqueue_err_count;
> + stats->dequeue_err_count +=
> worker_stats.dequeue_err_count;
> }
> }
>
> @@ -306,12 +306,12 @@ scheduler_pmd_stats_reset(struct rte_cryptodev
> *dev)
> struct scheduler_ctx *sched_ctx = dev->data->dev_private;
> uint32_t i;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev *slave_dev =
> - rte_cryptodev_pmd_get_dev(slave_dev_id);
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev *worker_dev =
> +
> rte_cryptodev_pmd_get_dev(worker_dev_id);
>
> - (*slave_dev->dev_ops->stats_reset)(slave_dev);
> + (*worker_dev->dev_ops->stats_reset)(worker_dev);
> }
> }
>
> @@ -329,32 +329,32 @@ scheduler_pmd_info_get(struct rte_cryptodev
> *dev,
> if (!dev_info)
> return;
>
> - /* although scheduler_attach_init_slave presents multiple times,
> + /* although scheduler_attach_init_worker presents multiple times,
> * there will be only 1 meaningful execution.
> */
> - scheduler_attach_init_slave(dev);
> + scheduler_attach_init_worker(dev);
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev_info slave_info;
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev_info worker_info;
>
> - rte_cryptodev_info_get(slave_dev_id, &slave_info);
> - uint32_t dev_max_sess = slave_info.sym.max_nb_sessions;
> + rte_cryptodev_info_get(worker_dev_id, &worker_info);
> + uint32_t dev_max_sess =
> worker_info.sym.max_nb_sessions;
> if (dev_max_sess != 0) {
> if (max_nb_sess == 0 || dev_max_sess <
> max_nb_sess)
> - max_nb_sess =
> slave_info.sym.max_nb_sessions;
> + max_nb_sess =
> worker_info.sym.max_nb_sessions;
> }
>
> - /* Get the max headroom requirement among slave PMDs
> */
> - headroom_sz = slave_info.min_mbuf_headroom_req >
> + /* Get the max headroom requirement among worker PMDs
> */
> + headroom_sz = worker_info.min_mbuf_headroom_req >
> headroom_sz ?
> - slave_info.min_mbuf_headroom_req :
> + worker_info.min_mbuf_headroom_req :
> headroom_sz;
>
> - /* Get the max tailroom requirement among slave PMDs */
> - tailroom_sz = slave_info.min_mbuf_tailroom_req >
> + /* Get the max tailroom requirement among worker PMDs
> */
> + tailroom_sz = worker_info.min_mbuf_tailroom_req >
> tailroom_sz ?
> - slave_info.min_mbuf_tailroom_req :
> + worker_info.min_mbuf_tailroom_req :
> tailroom_sz;
> }
>
> @@ -409,15 +409,15 @@ scheduler_pmd_qp_setup(struct rte_cryptodev
> *dev, uint16_t qp_id,
> if (dev->data->queue_pairs[qp_id] != NULL)
> scheduler_pmd_qp_release(dev, qp_id);
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_id = sched_ctx->slaves[i].dev_id;
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_id = sched_ctx->workers[i].dev_id;
>
> /*
> - * All slaves will share the same session mempool
> + * All workers will share the same session mempool
> * for session-less operations, so the objects
> * must be big enough for all the drivers used.
> */
> - ret = rte_cryptodev_queue_pair_setup(slave_id, qp_id,
> + ret = rte_cryptodev_queue_pair_setup(worker_id, qp_id,
> qp_conf, socket_id);
> if (ret < 0)
> return ret;
> @@ -434,12 +434,12 @@ scheduler_pmd_qp_setup(struct rte_cryptodev
> *dev, uint16_t qp_id,
>
> dev->data->queue_pairs[qp_id] = qp_ctx;
>
> - /* although scheduler_attach_init_slave presents multiple times,
> + /* although scheduler_attach_init_worker presents multiple times,
> * there will be only 1 meaningful execution.
> */
> - ret = scheduler_attach_init_slave(dev);
> + ret = scheduler_attach_init_worker(dev);
> if (ret < 0) {
> - CR_SCHED_LOG(ERR, "Failed to attach slave");
> + CR_SCHED_LOG(ERR, "Failed to attach worker");
> scheduler_pmd_qp_release(dev, qp_id);
> return ret;
> }
> @@ -461,10 +461,10 @@ scheduler_pmd_sym_session_get_size(struct
> rte_cryptodev *dev __rte_unused)
> uint8_t i = 0;
> uint32_t max_priv_sess_size = 0;
>
> - /* Check what is the maximum private session size for all slaves */
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - uint8_t slave_dev_id = sched_ctx->slaves[i].dev_id;
> - struct rte_cryptodev *dev = &rte_cryptodevs[slave_dev_id];
> + /* Check what is the maximum private session size for all workers */
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + uint8_t worker_dev_id = sched_ctx->workers[i].dev_id;
> + struct rte_cryptodev *dev =
> &rte_cryptodevs[worker_dev_id];
> uint32_t priv_sess_size = (*dev->dev_ops-
> >sym_session_get_size)(dev);
>
> if (max_priv_sess_size < priv_sess_size)
> @@ -484,10 +484,10 @@ scheduler_pmd_sym_session_configure(struct
> rte_cryptodev *dev,
> uint32_t i;
> int ret;
>
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - struct scheduler_slave *slave = &sched_ctx->slaves[i];
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + struct scheduler_worker *worker = &sched_ctx->workers[i];
>
> - ret = rte_cryptodev_sym_session_init(slave->dev_id, sess,
> + ret = rte_cryptodev_sym_session_init(worker->dev_id, sess,
> xform, mempool);
> if (ret < 0) {
> CR_SCHED_LOG(ERR, "unable to config sym session");
> @@ -506,11 +506,11 @@ scheduler_pmd_sym_session_clear(struct
> rte_cryptodev *dev,
> struct scheduler_ctx *sched_ctx = dev->data->dev_private;
> uint32_t i;
>
> - /* Clear private data of slaves */
> - for (i = 0; i < sched_ctx->nb_slaves; i++) {
> - struct scheduler_slave *slave = &sched_ctx->slaves[i];
> + /* Clear private data of workers */
> + for (i = 0; i < sched_ctx->nb_workers; i++) {
> + struct scheduler_worker *worker = &sched_ctx->workers[i];
>
> - rte_cryptodev_sym_session_clear(slave->dev_id, sess);
> + rte_cryptodev_sym_session_clear(worker->dev_id, sess);
> }
> }
>
> diff --git a/drivers/crypto/scheduler/scheduler_pmd_private.h
> b/drivers/crypto/scheduler/scheduler_pmd_private.h
> index e1531d1da..adb4eb063 100644
> --- a/drivers/crypto/scheduler/scheduler_pmd_private.h
> +++ b/drivers/crypto/scheduler/scheduler_pmd_private.h
> @@ -10,7 +10,7 @@
> #define CRYPTODEV_NAME_SCHEDULER_PMD crypto_scheduler
> /**< Scheduler Crypto PMD device name */
>
> -#define PER_SLAVE_BUFF_SIZE (256)
> +#define PER_WORKER_BUFF_SIZE (256)
>
> extern int scheduler_logtype_driver;
>
> @@ -18,7 +18,7 @@ extern int scheduler_logtype_driver;
> rte_log(RTE_LOG_ ## level, scheduler_logtype_driver, \
> "%s() line %u: "fmt "\n", __func__, __LINE__, ##args)
>
> -struct scheduler_slave {
> +struct scheduler_worker {
> uint8_t dev_id;
> uint16_t qp_id;
> uint32_t nb_inflight_cops;
> @@ -35,8 +35,8 @@ struct scheduler_ctx {
>
> uint32_t max_nb_queue_pairs;
>
> - struct scheduler_slave
> slaves[RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES];
> - uint32_t nb_slaves;
> + struct scheduler_worker
> workers[RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS];
> + uint32_t nb_workers;
>
> enum rte_cryptodev_scheduler_mode mode;
>
> @@ -49,8 +49,8 @@ struct scheduler_ctx {
> uint16_t wc_pool[RTE_MAX_LCORE];
> uint16_t nb_wc;
>
> - char
> *init_slave_names[RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES];
> - int nb_init_slaves;
> + char
> *init_worker_names[RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS];
> + int nb_init_workers;
> } __rte_cache_aligned;
>
> struct scheduler_qp_ctx {
> diff --git a/drivers/crypto/scheduler/scheduler_roundrobin.c
> b/drivers/crypto/scheduler/scheduler_roundrobin.c
> index 9b891d978..bc4a63210 100644
> --- a/drivers/crypto/scheduler/scheduler_roundrobin.c
> +++ b/drivers/crypto/scheduler/scheduler_roundrobin.c
> @@ -9,11 +9,11 @@
> #include "scheduler_pmd_private.h"
>
> struct rr_scheduler_qp_ctx {
> - struct scheduler_slave
> slaves[RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES];
> - uint32_t nb_slaves;
> + struct scheduler_worker
> workers[RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS];
> + uint32_t nb_workers;
>
> - uint32_t last_enq_slave_idx;
> - uint32_t last_deq_slave_idx;
> + uint32_t last_enq_worker_idx;
> + uint32_t last_deq_worker_idx;
> };
>
> static uint16_t
> @@ -21,8 +21,8 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> {
> struct rr_scheduler_qp_ctx *rr_qp_ctx =
> ((struct scheduler_qp_ctx *)qp)->private_qp_ctx;
> - uint32_t slave_idx = rr_qp_ctx->last_enq_slave_idx;
> - struct scheduler_slave *slave = &rr_qp_ctx->slaves[slave_idx];
> + uint32_t worker_idx = rr_qp_ctx->last_enq_worker_idx;
> + struct scheduler_worker *worker = &rr_qp_ctx-
> >workers[worker_idx];
> uint16_t i, processed_ops;
>
> if (unlikely(nb_ops == 0))
> @@ -31,13 +31,13 @@ schedule_enqueue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> for (i = 0; i < nb_ops && i < 4; i++)
> rte_prefetch0(ops[i]->sym->session);
>
> - processed_ops = rte_cryptodev_enqueue_burst(slave->dev_id,
> - slave->qp_id, ops, nb_ops);
> + processed_ops = rte_cryptodev_enqueue_burst(worker->dev_id,
> + worker->qp_id, ops, nb_ops);
>
> - slave->nb_inflight_cops += processed_ops;
> + worker->nb_inflight_cops += processed_ops;
>
> - rr_qp_ctx->last_enq_slave_idx += 1;
> - rr_qp_ctx->last_enq_slave_idx %= rr_qp_ctx->nb_slaves;
> + rr_qp_ctx->last_enq_worker_idx += 1;
> + rr_qp_ctx->last_enq_worker_idx %= rr_qp_ctx->nb_workers;
>
> return processed_ops;
> }
> @@ -64,34 +64,35 @@ schedule_dequeue(void *qp, struct rte_crypto_op
> **ops, uint16_t nb_ops)
> {
> struct rr_scheduler_qp_ctx *rr_qp_ctx =
> ((struct scheduler_qp_ctx *)qp)->private_qp_ctx;
> - struct scheduler_slave *slave;
> - uint32_t last_slave_idx = rr_qp_ctx->last_deq_slave_idx;
> + struct scheduler_worker *worker;
> + uint32_t last_worker_idx = rr_qp_ctx->last_deq_worker_idx;
> uint16_t nb_deq_ops;
>
> - if (unlikely(rr_qp_ctx->slaves[last_slave_idx].nb_inflight_cops == 0))
> {
> + if (unlikely(rr_qp_ctx->workers[last_worker_idx].nb_inflight_cops
> + == 0)) {
> do {
> - last_slave_idx += 1;
> + last_worker_idx += 1;
>
> - if (unlikely(last_slave_idx >= rr_qp_ctx->nb_slaves))
> - last_slave_idx = 0;
> + if (unlikely(last_worker_idx >= rr_qp_ctx-
> >nb_workers))
> + last_worker_idx = 0;
> /* looped back, means no inflight cops in the queue
> */
> - if (last_slave_idx == rr_qp_ctx->last_deq_slave_idx)
> + if (last_worker_idx == rr_qp_ctx-
> >last_deq_worker_idx)
> return 0;
> - } while (rr_qp_ctx->slaves[last_slave_idx].nb_inflight_cops
> + } while (rr_qp_ctx-
> >workers[last_worker_idx].nb_inflight_cops
> == 0);
> }
>
> - slave = &rr_qp_ctx->slaves[last_slave_idx];
> + worker = &rr_qp_ctx->workers[last_worker_idx];
>
> - nb_deq_ops = rte_cryptodev_dequeue_burst(slave->dev_id,
> - slave->qp_id, ops, nb_ops);
> + nb_deq_ops = rte_cryptodev_dequeue_burst(worker->dev_id,
> + worker->qp_id, ops, nb_ops);
>
> - last_slave_idx += 1;
> - last_slave_idx %= rr_qp_ctx->nb_slaves;
> + last_worker_idx += 1;
> + last_worker_idx %= rr_qp_ctx->nb_workers;
>
> - rr_qp_ctx->last_deq_slave_idx = last_slave_idx;
> + rr_qp_ctx->last_deq_worker_idx = last_worker_idx;
>
> - slave->nb_inflight_cops -= nb_deq_ops;
> + worker->nb_inflight_cops -= nb_deq_ops;
>
> return nb_deq_ops;
> }
> @@ -109,15 +110,15 @@ schedule_dequeue_ordering(void *qp, struct
> rte_crypto_op **ops,
> }
>
> static int
> -slave_attach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_attach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
>
> static int
> -slave_detach(__rte_unused struct rte_cryptodev *dev,
> - __rte_unused uint8_t slave_id)
> +worker_detach(__rte_unused struct rte_cryptodev *dev,
> + __rte_unused uint8_t worker_id)
> {
> return 0;
> }
> @@ -142,19 +143,19 @@ scheduler_start(struct rte_cryptodev *dev)
> qp_ctx->private_qp_ctx;
> uint32_t j;
>
> - memset(rr_qp_ctx->slaves, 0,
> -
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_SLAVES *
> - sizeof(struct scheduler_slave));
> - for (j = 0; j < sched_ctx->nb_slaves; j++) {
> - rr_qp_ctx->slaves[j].dev_id =
> - sched_ctx->slaves[j].dev_id;
> - rr_qp_ctx->slaves[j].qp_id = i;
> + memset(rr_qp_ctx->workers, 0,
> +
> RTE_CRYPTODEV_SCHEDULER_MAX_NB_WORKERS *
> + sizeof(struct scheduler_worker));
> + for (j = 0; j < sched_ctx->nb_workers; j++) {
> + rr_qp_ctx->workers[j].dev_id =
> + sched_ctx->workers[j].dev_id;
> + rr_qp_ctx->workers[j].qp_id = i;
> }
>
> - rr_qp_ctx->nb_slaves = sched_ctx->nb_slaves;
> + rr_qp_ctx->nb_workers = sched_ctx->nb_workers;
>
> - rr_qp_ctx->last_enq_slave_idx = 0;
> - rr_qp_ctx->last_deq_slave_idx = 0;
> + rr_qp_ctx->last_enq_worker_idx = 0;
> + rr_qp_ctx->last_deq_worker_idx = 0;
> }
>
> return 0;
> @@ -191,8 +192,8 @@ scheduler_create_private_ctx(__rte_unused struct
> rte_cryptodev *dev)
> }
>
> static struct rte_cryptodev_scheduler_ops scheduler_rr_ops = {
> - slave_attach,
> - slave_detach,
> + worker_attach,
> + worker_detach,
> scheduler_start,
> scheduler_stop,
> scheduler_config_qp,
> @@ -204,7 +205,7 @@ static struct rte_cryptodev_scheduler_ops
> scheduler_rr_ops = {
> static struct rte_cryptodev_scheduler scheduler = {
> .name = "roundrobin-scheduler",
> .description = "scheduler which will round robin burst across "
> - "slave crypto devices",
> + "worker crypto devices",
> .mode = CDEV_SCHED_MODE_ROUNDROBIN,
> .ops = &scheduler_rr_ops
> };
> diff --git a/examples/l2fwd-crypto/main.c b/examples/l2fwd-crypto/main.c
> index 827da9b3e..42e80bc3f 100644
> --- a/examples/l2fwd-crypto/main.c
> +++ b/examples/l2fwd-crypto/main.c
> @@ -2277,11 +2277,11 @@ initialize_cryptodevs(struct
> l2fwd_crypto_options *options, unsigned nb_ports,
> */
> if (!strcmp(dev_info.driver_name, "crypto_scheduler")) {
> #ifdef RTE_LIBRTE_PMD_CRYPTO_SCHEDULER
> - uint32_t nb_slaves =
> -
> rte_cryptodev_scheduler_slaves_get(cdev_id,
> + uint32_t nb_workers =
> +
> rte_cryptodev_scheduler_workers_get(cdev_id,
> NULL);
>
> - sessions_needed = enabled_cdev_count * nb_slaves;
> + sessions_needed = enabled_cdev_count *
> nb_workers;
> #endif
> } else
> sessions_needed = enabled_cdev_count;
> --
> 2.25.1
Looks good from ABI perspective.
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
next prev parent reply other threads:[~2020-09-28 15:12 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-08-26 15:34 [dpdk-dev] [PATCH] " Adam Dybkowski
2020-08-27 15:18 ` Zhang, Roy Fan
2020-09-28 2:49 ` Ruifeng Wang
2020-09-28 10:17 ` Dybkowski, AdamX
2020-09-28 14:16 ` [dpdk-dev] [PATCH v2 0/1] " Adam Dybkowski
2020-09-28 14:16 ` [dpdk-dev] [PATCH v2 1/1] " Adam Dybkowski
2020-09-28 15:12 ` Ruifeng Wang [this message]
2020-10-06 20:49 ` Akhil Goyal
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=VI1PR0802MB2351C78B98AF0B794538B27D9E350@VI1PR0802MB2351.eurprd08.prod.outlook.com \
--to=ruifeng.wang@arm.com \
--cc=adamx.dybkowski@intel.com \
--cc=akhil.goyal@nxp.com \
--cc=dev@dpdk.org \
--cc=fiona.trahe@intel.com \
--cc=nd@arm.com \
--cc=roy.fan.zhang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).