From mboxrd@z Thu Jan 1 00:00:00 1970 From: Fan Zhang Subject: [PATCH] crypto/scheduler: fix slave init attach Date: Wed, 19 Apr 2017 14:42:17 +0100 Message-ID: <1492609337-44733-1-git-send-email-roy.fan.zhang@intel.com> Cc: pablo.de.lara.guarch@intel.com, stable@dpdk.com To: dev@dpdk.org Return-path: Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by dpdk.org (Postfix) with ESMTP id C4440388F for ; Wed, 19 Apr 2017 15:40:29 +0200 (CEST) List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Fixes: 881851676940 ("crypto/scheduler: improve commandline parsing") This patch fix the slaves attaching initially. Originally, the scheduler will try to attach the slaves passed by the cmdline option in the initial stage, this may break the sequence of the slaves inside the scheduler. Signed-off-by: Fan Zhang --- drivers/crypto/scheduler/scheduler_pmd.c | 21 --------------------- 1 file changed, 21 deletions(-) diff --git a/drivers/crypto/scheduler/scheduler_pmd.c b/drivers/crypto/scheduler/scheduler_pmd.c index f2a1d2a..2a92d1e 100644 --- a/drivers/crypto/scheduler/scheduler_pmd.c +++ b/drivers/crypto/scheduler/scheduler_pmd.c @@ -160,27 +160,6 @@ cryptodev_scheduler_create(const char *name, } for (i = 0; i < init_params->nb_slaves; i++) { - struct rte_cryptodev *slave_dev = - rte_cryptodev_pmd_get_named_dev( - init_params->slave_names[i]); - - /* if the slave dev exists in the system, attach it */ - if (slave_dev) { - int ret = rte_cryptodev_scheduler_slave_attach( - dev->data->dev_id, - slave_dev->data->dev_id); - - if (ret < 0) - return ret; - - CS_LOG_INFO("Scheduler %s attached slave %s\n", - dev->data->name, - init_params->slave_names[i]); - - continue; - } - - /* if the slave doesn't exist, store and attach later*/ sched_ctx->init_slave_names[sched_ctx->nb_init_slaves] = rte_zmalloc_socket( NULL, -- 2.7.4