From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from EUR01-DB5-obe.outbound.protection.outlook.com (mail-db5eur01on0055.outbound.protection.outlook.com [104.47.2.55]) by dpdk.org (Postfix) with ESMTP id B55A92C52 for ; Tue, 24 Apr 2018 13:50:38 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=nxp.com; s=selector1; h=From:Date:Subject:Message-ID:Content-Type:MIME-Version; bh=lIK5CtC5VuN6RaSxiXPIIIVqudKRyXdfa4hS3aP3Rjk=; b=iCmCTjpnoxH9li2u+2RPzNATj+QBuhiCElVZwc6fXvQ3+yMsy6wooPeb10x5woJaa/aC8tX/hyaiEpWN4z8CJrzysM5QMy3E+zhQ62o1CldQvnmQYz660s4RJboV2b4cCoCiVJtxyCdlTo+cRvi95xcCsCM5Ts2ETgsnNK/KDZw= Authentication-Results: spf=none (sender IP is ) smtp.mailfrom=nipun.gupta@nxp.com; Received: from b27504-OptiPlex-790.ap.freescale.net (14.142.187.166) by DB6PR0401MB2424.eurprd04.prod.outlook.com (2603:10a6:4:4a::27) with Microsoft SMTP Server (version=TLS1_2, cipher=TLS_ECDHE_RSA_WITH_AES_256_CBC_SHA384_P256) id 15.20.696.15; Tue, 24 Apr 2018 11:50:36 +0000 From: Nipun Gupta To: shreyansh.jain@nxp.com, thomas@monjalon.net, hemant.agrawal@nxp.com Cc: dev@dpdk.org, Nipun Gupta Date: Tue, 24 Apr 2018 17:19:47 +0530 Message-Id: <1524570589-20148-7-git-send-email-nipun.gupta@nxp.com> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1524570589-20148-1-git-send-email-nipun.gupta@nxp.com> References: <1523114224-9852-1-git-send-email-nipun.gupta@nxp.com> <1524570589-20148-1-git-send-email-nipun.gupta@nxp.com> MIME-Version: 1.0 Content-Type: text/plain X-Originating-IP: [14.142.187.166] X-ClientProxiedBy: PN1PR01CA0085.INDPRD01.PROD.OUTLOOK.COM (2603:1096:c00:1::25) To DB6PR0401MB2424.eurprd04.prod.outlook.com (2603:10a6:4:4a::27) X-MS-PublicTrafficType: Email X-MS-Office365-Filtering-HT: Tenant X-Microsoft-Antispam: UriScan:; BCL:0; PCL:0; RULEID:(7020095)(4652020)(5600026)(48565401081)(2017052603328)(7153060)(7193020); SRVR:DB6PR0401MB2424; X-Microsoft-Exchange-Diagnostics: 1; DB6PR0401MB2424; 3:X5azDcyEnpDXnQCKnwcXpSgGXiAp9i/GlOqu5UiH4woH+0ehSr9//sTuYVUGan8j7dvdHrauRvgcDMLYzOihxrOCF9tXIQqyaOmn2ZZgamwH/uO9+DvzpGDrSSfCCqpOTDWgoMCIDhyVqHE3Iw3ZipUCdG9VKPOAjWM0f9SZLDemk+cvktd6ZChCh8pXOyxV5C694/ttR6RQ1bB+IkTSxDArlnaNkiFvhEbzwFWNcB5XN7zGcGeRemyZw2WOJ+LN; 25:zuIrsSsKlL+68UOQxS4ef0mHoZu8flgOyLpbaQ/eUwWN1NmfPAJz4vnp/4tacP8i03sAYDa0wSZo1tf5YpZPnZKhlkPMYJfsI7oDPHJfiGAjzMOVfsVwzRvAS+klSVYXy9dCK1P5RYz09+XsavuUVm0Btero6Ton2A0NZ4Ptvue6G317Dbp9KBJB7uJcJKYV3zXe6jdWnWfZqaRcU3Y6yqyMKopa+6buO7fNlikKyzMjQDiyMO2AUXGoRX6wrJKusA5M7GrcIOPzESXQ7dUqPRqDlR/NVfhyxYEnMcSD4uq1WfG9Z33PLvMbffSAimOLvWORnFKOYyouK0eRJHTH8g==; 31:utzyI5zLTnbyJXliiFc3/K0jpAu93ZPUamjKzFt6DPpfKHnN7CgdktJsyy0e5NMX06VG1KSCLouZVh7dzQyCE00oXSbV14cv56GiRZaa6SuayaxJYkIQUaUsjjzD5Ks6nS4I6Fo7ZX1WrbrsfctC1myFHiqYLDdKwvYU1xqZPsnfAZc9HP3Xd/BTb1Rgf7mT390X1bWo7nPqPV9lImZF4lHpowFdFaW5FlHP2o7sCZY= X-MS-TrafficTypeDiagnostic: DB6PR0401MB2424: X-Microsoft-Exchange-Diagnostics: 1; DB6PR0401MB2424; 20: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; 4:Z6EhlcBeuvF/Dkr0PvEJbe9Z8CEB2f+lSP1Jcqu9HNjfgibLa5mPQuBzRQLIgb54hqGL0W8m8PsXJFSSrGNDWg/0j6PfBcOu1Erf4EHbwQWEtPalNV7MUYrmJStmt5E00AylYKafvcI6M1jTB8g27ml/0Wuagc3ccvq+cVv0i0D1bRfXnhVfbDfdoJ3fcT9aWieasDIKl/Qmp7W1G1GN7ppWEfHxjgSUQv5dz9on1HLtaStujUDtKMKioyDOdrl66gb28MjwkfrH7YbeXtii9G3mBYCrtB/bAEzLJ4F8kEsKpw6/uAXWtXHYQhoHYAzM X-Microsoft-Antispam-PRVS: X-Exchange-Antispam-Report-Test: UriScan:(185117386973197); X-Exchange-Antispam-Report-CFA-Test: BCL:0; PCL:0; RULEID:(8211001083)(6040522)(2401047)(5005006)(8121501046)(10201501046)(93006095)(93001095)(3002001)(3231232)(944501410)(52105095)(6055026)(6041310)(20161123558120)(201703131423095)(201702281528075)(20161123555045)(201703061421075)(201703061406153)(20161123560045)(20161123562045)(20161123564045)(6072148)(201708071742011); SRVR:DB6PR0401MB2424; BCL:0; PCL:0; RULEID:; SRVR:DB6PR0401MB2424; X-Forefront-PRVS: 0652EA5565 X-Forefront-Antispam-Report: SFV:NSPM; SFS:(10009020)(376002)(346002)(39860400002)(396003)(39380400002)(366004)(199004)(189003)(3846002)(97736004)(316002)(486006)(305945005)(6486002)(6636002)(476003)(2616005)(956004)(4326008)(25786009)(5009440100003)(446003)(86362001)(6512007)(6666003)(11346002)(5660300001)(478600001)(44832011)(53936002)(52116002)(16526019)(76176011)(551934003)(68736007)(7736002)(47776003)(51416003)(6506007)(106356001)(66066001)(26005)(36756003)(386003)(59450400001)(50466002)(81166006)(8676002)(6116002)(105586002)(16586007)(48376002)(50226002)(81156014)(8936002)(2906002)(110426005); DIR:OUT; SFP:1101; SCL:1; SRVR:DB6PR0401MB2424; H:b27504-OptiPlex-790.ap.freescale.net; FPR:; SPF:None; LANG:en; PTR:InfoNoRecords; A:1; MX:1; Received-SPF: None (protection.outlook.com: nxp.com does not designate permitted sender hosts) X-Microsoft-Exchange-Diagnostics: =?us-ascii?Q?1; DB6PR0401MB2424; 23:BppBKpY5XVzk34398zwlMsLtzuigF5TcGqeGyVL?= =?us-ascii?Q?Ev+2wTmCRkYmQZwP/vUogmKR+G7xNrCtWdz6+xGdbCvqrF+O6X37ehDdMtu3?= =?us-ascii?Q?wbbF4N/t4RKCvLIi13M/CXwOmEL4LvLrxRyY2uVNEmpyt4LTOmbdsiqf9/O9?= =?us-ascii?Q?vfqhdY0zMH0xJ8HyWFQGzXbbTV6A6BKjXdU7vnY4eAqYDXiN1CVMTjA5l/8h?= =?us-ascii?Q?4HH4RuOi0dXbCb1L3Rhx/3HHlLNP5SkahwsfsCCWwT6QTDezweqKEX1l0Fqt?= =?us-ascii?Q?Wy0yJgEFrOYQizD0ayAhTf6BMmYe/nmI/3Myqyqec3VeQeh0hHzUP6BLKL04?= =?us-ascii?Q?VQ1R2g/+tY591BkqxFkNTLuWdm6q0m0qnmQh0H2l+M/Pcv0qjuXRGa+BDa11?= =?us-ascii?Q?3vgKM81J1hj32JuNKjAtNyfplNlzhaSr4z8iqDqzw80lo/lfShcmahZuQKhS?= =?us-ascii?Q?wI01oW9bAAiELQVdhFfknqW38ouY4g0i/8wtbqNL6EQ4bQ0VlAl5C0UoGuKs?= =?us-ascii?Q?d1dhFzOPzmFXWkTVxsYvFRp3BuLZhG35z0TVqG26rQBc1FP4b7AIkfwj1Xhq?= =?us-ascii?Q?68eVyNkbIHMV/1APANjGgjMPdTnDTrKysciDz79SNzkEbQRqzZpoYbiOzRkh?= =?us-ascii?Q?Zl9E/1EkSxxYgK8JOj40KrsyqvVxcgpNMTLFSi4S5feVEyNQOjmGHeHDNFc+?= =?us-ascii?Q?s8l+wCs26YRFklDgNoK3qycjng5NGAMPbPVukd2o9Q5077mxsRd1vK+XrsSc?= =?us-ascii?Q?Hz8yCC2n42uGs2FzhLGfLtg8Anbr8zYUu6ywyDPj9k2ZX+l8YXbJIrxzRThi?= =?us-ascii?Q?gm8RDLJy9MSNruNULWG62gaYFIgSgj+7NrR9PvmImnccvZqOVi+t4vHlXR3X?= =?us-ascii?Q?aMxKlrQSfVKZBfDuSS88FIfw6Nyb6QhIkO0KXQEF4t9+Gkg/gcYxm6SSqKgH?= =?us-ascii?Q?At0k7N+urCIYLMzWJP97K6rKdLbKYdzdm2RPsvRc7mEo2v/nx+ByFvy3F7Ky?= =?us-ascii?Q?agNgL3PfsrSBKO3LPzdkvYJK+oST+rRFauaYwUA6Q+lcTiHTkm3e09aSK+fY?= =?us-ascii?Q?TDFEIWXcI5CxxIf5H3PIXXW6Ci6tvqgkjOEYhFF+g2svLN8u+IUKsEUS00fz?= =?us-ascii?Q?xHCBBUPzGI7JmJwWCIuOVDy1Psd9am09sBf3SVIm/RE7ma59hNXNyHdBIjgx?= =?us-ascii?Q?DmYJH0hscpvz1abSyBJmu4IbVVCRT5yl38H6UV2XM9RLx7WDho0uCGuOCmW1?= =?us-ascii?Q?yr0HKQB0El1727uS4VN4aQP8+JYnO3AA8XbHlSu9w2Fy7TH2nVlfImE86EIc?= =?us-ascii?Q?3xQ=3D=3D?= X-Microsoft-Antispam-Message-Info: yoPHBiLYD2pRPZx0mRFb2og2KAPaBORwtkk1BVN0epA5jAyavEPKEMHpsc0dFX+MA5n8kag/rQ9FF62oqktnW9TsNhlg32OkwUDDnC7JfDj3oj4OFCRhznOEUXmtN8RicewL1J3IlVKyzVexnwbttJEKG54zMYYWH8itvcmPSa6WdHonwhfMbvLikie/I9yx X-Microsoft-Exchange-Diagnostics: 1; DB6PR0401MB2424; 6:sRVwU8fQpG2n0OOoaXdgcNe9ZOwPCrDUw8w+X55dmLndfsjlub7Qi33WwCfhFfJs3rDL/LVtykbezVuTN1ulltnOrVjGtUpeT5zff35etqZBeRHKJgLJlcoX3GyYa1DqbGeOi3RbDv2szqm/ZsE2UXNqMFtfLj5EOFKCGwhzI5VIBkX95Xzk9W4fZfEptHHCqfOj5rz3C91oLkz9bXQj2zkyr7nz7mUP1dedO7wPODCymOmjP7euDodkEkAT33E2z1P6ua95fuIywEFAKkd1jrKilVb/BtdA2YttDj8hdYnEgTvXDTVYvYrbfKgbYDuhVKIu4SeRUN9+pbcT/0GF3ASAgnVbwE5fGg40jeWJGa6TewVb3b9T8sAXwvqRuoKS3jGBy045bGbCLID56e6glEjsytDO//gmTxNxKo2LD9gJuWlFpNlMivyeo7wki2TX6SLhSmdu4M7yl03YgeePRQ==; 5:yL1C0fYaIIzPdMydz+pxppr3t7mgICmbx1Lg+QmvozAEnnePdJ+aqhjlwEUnYWW6vG4i2mTtwzLKuFCWnO6ZrFFSiZs6VAdIFXpS1i4EnkBLFF6bCSGnfGGyme0KpxaJuJZ7HO1qHZYTHSCA65YT2ZQ7FECgeJ9tRtu8Pq5wc8w=; 24:lnZe27jUmPthZWQQyGY5+foG5HxQw6JfMH0n4cQJixrr8gJfvwTDlApjSw4r/CZo45fFsLLH/rGBqEGAxxPByR0xzKmyGRVyePmlCX2vGiE= SpamDiagnosticOutput: 1:99 SpamDiagnosticMetadata: NSPM X-Microsoft-Exchange-Diagnostics: 1; DB6PR0401MB2424; 7:1vv6S2RTmtFh36iyINB0aruI36/F1wJVNi1qyrzn6yN3SgZzgT+GkE2xOs1xbgIKhzzQNs6sSRjAXkekwOpdt/U5rw8KCUv2mAyPW9hcK8rcwKV9H8dxaGKVm6C6lHhSkCFOTyaU6wzEGeHwnVG1RmTPtfV2kN8OFtx3Mrcl+1r6fjAawz5OX1FNdTLwMniENZAl8HoSUniH5tHcWsOaKr+BMBpl8CIIbv9srhMwim7MDqPpJmAd0rb6evcpsIY+ X-MS-Office365-Filtering-Correlation-Id: f0a91fe0-5b46-45a4-54d0-08d5a9d99870 X-OriginatorOrg: nxp.com X-MS-Exchange-CrossTenant-OriginalArrivalTime: 24 Apr 2018 11:50:36.2039 (UTC) X-MS-Exchange-CrossTenant-Network-Message-Id: f0a91fe0-5b46-45a4-54d0-08d5a9d99870 X-MS-Exchange-CrossTenant-FromEntityHeader: Hosted X-MS-Exchange-CrossTenant-Id: 686ea1d3-bc2b-4c6f-a92c-d99c5c301635 X-MS-Exchange-Transport-CrossTenantHeadersStamped: DB6PR0401MB2424 Subject: [dpdk-dev] [PATCH 6/8 v4] raw/dpaa2_qdma: support configuration APIs X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 24 Apr 2018 11:50:39 -0000 Signed-off-by: Nipun Gupta --- doc/api/doxy-api-index.md | 1 + doc/api/doxy-api.conf | 1 + drivers/raw/dpaa2_qdma/Makefile | 2 + drivers/raw/dpaa2_qdma/dpaa2_qdma.c | 375 +++++++++++++++++++++ drivers/raw/dpaa2_qdma/dpaa2_qdma.h | 63 ++++ drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma.h | 216 ++++++++++++ .../raw/dpaa2_qdma/rte_pmd_dpaa2_qdma_version.map | 12 + 7 files changed, 670 insertions(+) create mode 100644 drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma.h diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md index 26ce7b4..437d903 100644 --- a/doc/api/doxy-api-index.md +++ b/doc/api/doxy-api-index.md @@ -38,6 +38,7 @@ The public API headers are grouped by topics: [i40e] (@ref rte_pmd_i40e.h), [bnxt] (@ref rte_pmd_bnxt.h), [dpaa] (@ref rte_pmd_dpaa.h), + [dpaa2_qdma] (@ref rte_pmd_dpaa2_qdma.h), [crypto_scheduler] (@ref rte_cryptodev_scheduler.h) - **memory**: diff --git a/doc/api/doxy-api.conf b/doc/api/doxy-api.conf index ad8bdcf..18a7ed4 100644 --- a/doc/api/doxy-api.conf +++ b/doc/api/doxy-api.conf @@ -37,6 +37,7 @@ INPUT = doc/api/doxy-api-index.md \ drivers/net/i40e \ drivers/net/ixgbe \ drivers/net/softnic \ + drivers/raw/dpaa2_qdma \ lib/librte_eal/common/include \ lib/librte_eal/common/include/generic \ lib/librte_acl \ diff --git a/drivers/raw/dpaa2_qdma/Makefile b/drivers/raw/dpaa2_qdma/Makefile index 3b5a45a..264eca8 100644 --- a/drivers/raw/dpaa2_qdma/Makefile +++ b/drivers/raw/dpaa2_qdma/Makefile @@ -31,4 +31,6 @@ LIBABIVER := 1 # SRCS-$(CONFIG_RTE_LIBRTE_PMD_DPAA2_QDMA_RAWDEV) += dpaa2_qdma.c +SYMLINK-$(CONFIG_RTE_LIBRTE_PMD_DPAA2_QDMA_RAWDEV)-include += rte_pmd_dpaa2_qdma.h + include $(RTE_SDK)/mk/rte.lib.mk diff --git a/drivers/raw/dpaa2_qdma/dpaa2_qdma.c b/drivers/raw/dpaa2_qdma/dpaa2_qdma.c index 9288350..b38a282 100644 --- a/drivers/raw/dpaa2_qdma/dpaa2_qdma.c +++ b/drivers/raw/dpaa2_qdma/dpaa2_qdma.c @@ -20,6 +20,7 @@ #include "dpaa2_qdma.h" #include "dpaa2_qdma_logs.h" +#include "rte_pmd_dpaa2_qdma.h" /* Dynamic log type identifier */ int dpaa2_qdma_logtype; @@ -32,6 +33,380 @@ static struct qdma_hw_queue_list qdma_queue_list = TAILQ_HEAD_INITIALIZER(qdma_queue_list); +/* QDMA Virtual Queues */ +struct qdma_virt_queue *qdma_vqs; + +/* QDMA per core data */ +struct qdma_per_core_info qdma_core_info[RTE_MAX_LCORE]; + +static struct qdma_hw_queue * +alloc_hw_queue(uint32_t lcore_id) +{ + struct qdma_hw_queue *queue = NULL; + + DPAA2_QDMA_FUNC_TRACE(); + + /* Get a free queue from the list */ + TAILQ_FOREACH(queue, &qdma_queue_list, next) { + if (queue->num_users == 0) { + queue->lcore_id = lcore_id; + queue->num_users++; + break; + } + } + + return queue; +} + +static void +free_hw_queue(struct qdma_hw_queue *queue) +{ + DPAA2_QDMA_FUNC_TRACE(); + + queue->num_users--; +} + + +static struct qdma_hw_queue * +get_hw_queue(uint32_t lcore_id) +{ + struct qdma_per_core_info *core_info; + struct qdma_hw_queue *queue, *temp; + uint32_t least_num_users; + int num_hw_queues, i; + + DPAA2_QDMA_FUNC_TRACE(); + + core_info = &qdma_core_info[lcore_id]; + num_hw_queues = core_info->num_hw_queues; + + /* + * Allocate a HW queue if there are less queues + * than maximum per core queues configured + */ + if (num_hw_queues < qdma_dev.max_hw_queues_per_core) { + queue = alloc_hw_queue(lcore_id); + if (queue) { + core_info->hw_queues[num_hw_queues] = queue; + core_info->num_hw_queues++; + return queue; + } + } + + queue = core_info->hw_queues[0]; + /* In case there is no queue associated with the core return NULL */ + if (!queue) + return NULL; + + /* Fetch the least loaded H/W queue */ + least_num_users = core_info->hw_queues[0]->num_users; + for (i = 0; i < num_hw_queues; i++) { + temp = core_info->hw_queues[i]; + if (temp->num_users < least_num_users) + queue = temp; + } + + if (queue) + queue->num_users++; + + return queue; +} + +static void +put_hw_queue(struct qdma_hw_queue *queue) +{ + struct qdma_per_core_info *core_info; + int lcore_id, num_hw_queues, i; + + DPAA2_QDMA_FUNC_TRACE(); + + /* + * If this is the last user of the queue free it. + * Also remove it from QDMA core info. + */ + if (queue->num_users == 1) { + free_hw_queue(queue); + + /* Remove the physical queue from core info */ + lcore_id = queue->lcore_id; + core_info = &qdma_core_info[lcore_id]; + num_hw_queues = core_info->num_hw_queues; + for (i = 0; i < num_hw_queues; i++) { + if (queue == core_info->hw_queues[i]) + break; + } + for (; i < num_hw_queues - 1; i++) + core_info->hw_queues[i] = core_info->hw_queues[i + 1]; + core_info->hw_queues[i] = NULL; + } else { + queue->num_users--; + } +} + +int __rte_experimental +rte_qdma_init(void) +{ + DPAA2_QDMA_FUNC_TRACE(); + + rte_spinlock_init(&qdma_dev.lock); + + return 0; +} + +void __rte_experimental +rte_qdma_attr_get(struct rte_qdma_attr *qdma_attr) +{ + DPAA2_QDMA_FUNC_TRACE(); + + qdma_attr->num_hw_queues = qdma_dev.num_hw_queues; +} + +int __rte_experimental +rte_qdma_reset(void) +{ + struct qdma_hw_queue *queue; + int i; + + DPAA2_QDMA_FUNC_TRACE(); + + /* In case QDMA device is not in stopped state, return -EBUSY */ + if (qdma_dev.state == 1) { + DPAA2_QDMA_ERR( + "Device is in running state. Stop before reset."); + return -EBUSY; + } + + /* In case there are pending jobs on any VQ, return -EBUSY */ + for (i = 0; i < qdma_dev.max_vqs; i++) { + if (qdma_vqs[i].in_use && (qdma_vqs[i].num_enqueues != + qdma_vqs[i].num_dequeues)) + DPAA2_QDMA_ERR("Jobs are still pending on VQ: %d", i); + return -EBUSY; + } + + /* Reset HW queues */ + TAILQ_FOREACH(queue, &qdma_queue_list, next) + queue->num_users = 0; + + /* Reset and free virtual queues */ + for (i = 0; i < qdma_dev.max_vqs; i++) { + if (qdma_vqs[i].status_ring) + rte_ring_free(qdma_vqs[i].status_ring); + } + if (qdma_vqs) + rte_free(qdma_vqs); + qdma_vqs = NULL; + + /* Reset per core info */ + memset(&qdma_core_info, 0, + sizeof(struct qdma_per_core_info) * RTE_MAX_LCORE); + + /* Free the FLE pool */ + if (qdma_dev.fle_pool) + rte_mempool_free(qdma_dev.fle_pool); + + /* Reset QDMA device structure */ + qdma_dev.mode = RTE_QDMA_MODE_HW; + qdma_dev.max_hw_queues_per_core = 0; + qdma_dev.fle_pool = NULL; + qdma_dev.fle_pool_count = 0; + qdma_dev.max_vqs = 0; + + return 0; +} + +int __rte_experimental +rte_qdma_configure(struct rte_qdma_config *qdma_config) +{ + int ret; + + DPAA2_QDMA_FUNC_TRACE(); + + /* In case QDMA device is not in stopped state, return -EBUSY */ + if (qdma_dev.state == 1) { + DPAA2_QDMA_ERR( + "Device is in running state. Stop before config."); + return -1; + } + + /* Reset the QDMA device */ + ret = rte_qdma_reset(); + if (ret) { + DPAA2_QDMA_ERR("Resetting QDMA failed"); + return ret; + } + + /* Set mode */ + qdma_dev.mode = qdma_config->mode; + + /* Set max HW queue per core */ + if (qdma_config->max_hw_queues_per_core > MAX_HW_QUEUE_PER_CORE) { + DPAA2_QDMA_ERR("H/W queues per core is more than: %d", + MAX_HW_QUEUE_PER_CORE); + return -EINVAL; + } + qdma_dev.max_hw_queues_per_core = + qdma_config->max_hw_queues_per_core; + + /* Allocate Virtual Queues */ + qdma_vqs = rte_malloc("qdma_virtual_queues", + (sizeof(struct qdma_virt_queue) * qdma_config->max_vqs), + RTE_CACHE_LINE_SIZE); + if (!qdma_vqs) { + DPAA2_QDMA_ERR("qdma_virtual_queues allocation failed"); + return -ENOMEM; + } + qdma_dev.max_vqs = qdma_config->max_vqs; + + /* Allocate FLE pool */ + qdma_dev.fle_pool = rte_mempool_create("qdma_fle_pool", + qdma_config->fle_pool_count, QDMA_FLE_POOL_SIZE, + QDMA_FLE_CACHE_SIZE(qdma_config->fle_pool_count), 0, + NULL, NULL, NULL, NULL, SOCKET_ID_ANY, 0); + if (!qdma_dev.fle_pool) { + DPAA2_QDMA_ERR("qdma_fle_pool create failed"); + rte_free(qdma_vqs); + qdma_vqs = NULL; + return -ENOMEM; + } + qdma_dev.fle_pool_count = qdma_config->fle_pool_count; + + return 0; +} + +int __rte_experimental +rte_qdma_start(void) +{ + DPAA2_QDMA_FUNC_TRACE(); + + qdma_dev.state = 1; + + return 0; +} + +int __rte_experimental +rte_qdma_vq_create(uint32_t lcore_id, uint32_t flags) +{ + char ring_name[32]; + int i; + + DPAA2_QDMA_FUNC_TRACE(); + + rte_spinlock_lock(&qdma_dev.lock); + + /* Get a free Virtual Queue */ + for (i = 0; i < qdma_dev.max_vqs; i++) { + if (qdma_vqs[i].in_use == 0) + break; + } + + /* Return in case no VQ is free */ + if (i == qdma_dev.max_vqs) { + rte_spinlock_unlock(&qdma_dev.lock); + return -ENODEV; + } + + if (qdma_dev.mode == RTE_QDMA_MODE_HW || + (flags & RTE_QDMA_VQ_EXCLUSIVE_PQ)) { + /* Allocate HW queue for a VQ */ + qdma_vqs[i].hw_queue = alloc_hw_queue(lcore_id); + qdma_vqs[i].exclusive_hw_queue = 1; + } else { + /* Allocate a Ring for Virutal Queue in VQ mode */ + sprintf(ring_name, "status ring %d", i); + qdma_vqs[i].status_ring = rte_ring_create(ring_name, + qdma_dev.fle_pool_count, rte_socket_id(), 0); + if (!qdma_vqs[i].status_ring) { + DPAA2_QDMA_ERR("Status ring creation failed for vq"); + rte_spinlock_unlock(&qdma_dev.lock); + return rte_errno; + } + + /* Get a HW queue (shared) for a VQ */ + qdma_vqs[i].hw_queue = get_hw_queue(lcore_id); + qdma_vqs[i].exclusive_hw_queue = 0; + } + + if (qdma_vqs[i].hw_queue == NULL) { + DPAA2_QDMA_ERR("No H/W queue available for VQ"); + if (qdma_vqs[i].status_ring) + rte_ring_free(qdma_vqs[i].status_ring); + qdma_vqs[i].status_ring = NULL; + rte_spinlock_unlock(&qdma_dev.lock); + return -ENODEV; + } + + qdma_vqs[i].in_use = 1; + qdma_vqs[i].lcore_id = lcore_id; + + rte_spinlock_unlock(&qdma_dev.lock); + + return i; +} + +void __rte_experimental +rte_qdma_vq_stats(uint16_t vq_id, + struct rte_qdma_vq_stats *vq_status) +{ + struct qdma_virt_queue *qdma_vq = &qdma_vqs[vq_id]; + + DPAA2_QDMA_FUNC_TRACE(); + + if (qdma_vq->in_use) { + vq_status->exclusive_hw_queue = qdma_vq->exclusive_hw_queue; + vq_status->lcore_id = qdma_vq->lcore_id; + vq_status->num_enqueues = qdma_vq->num_enqueues; + vq_status->num_dequeues = qdma_vq->num_dequeues; + vq_status->num_pending_jobs = vq_status->num_enqueues - + vq_status->num_dequeues; + } +} + +int __rte_experimental +rte_qdma_vq_destroy(uint16_t vq_id) +{ + struct qdma_virt_queue *qdma_vq = &qdma_vqs[vq_id]; + + DPAA2_QDMA_FUNC_TRACE(); + + /* In case there are pending jobs on any VQ, return -EBUSY */ + if (qdma_vq->num_enqueues != qdma_vq->num_dequeues) + return -EBUSY; + + rte_spinlock_lock(&qdma_dev.lock); + + if (qdma_vq->exclusive_hw_queue) + free_hw_queue(qdma_vq->hw_queue); + else { + if (qdma_vqs->status_ring) + rte_ring_free(qdma_vqs->status_ring); + + put_hw_queue(qdma_vq->hw_queue); + } + + memset(qdma_vq, 0, sizeof(struct qdma_virt_queue)); + + rte_spinlock_lock(&qdma_dev.lock); + + return 0; +} + +void __rte_experimental +rte_qdma_stop(void) +{ + DPAA2_QDMA_FUNC_TRACE(); + + qdma_dev.state = 0; +} + +void __rte_experimental +rte_qdma_destroy(void) +{ + DPAA2_QDMA_FUNC_TRACE(); + + rte_qdma_reset(); +} + static const struct rte_rawdev_ops dpaa2_qdma_ops; static int diff --git a/drivers/raw/dpaa2_qdma/dpaa2_qdma.h b/drivers/raw/dpaa2_qdma/dpaa2_qdma.h index 8b3b1b9..fe1da41 100644 --- a/drivers/raw/dpaa2_qdma/dpaa2_qdma.h +++ b/drivers/raw/dpaa2_qdma/dpaa2_qdma.h @@ -5,6 +5,22 @@ #ifndef __DPAA2_QDMA_H__ #define __DPAA2_QDMA_H__ +struct qdma_sdd; +struct qdma_io_meta; + +#define DPAA2_QDMA_MAX_FLE 3 +#define DPAA2_QDMA_MAX_SDD 2 + +/** FLE pool size: 3 Frame list + 2 source/destination descriptor */ +#define QDMA_FLE_POOL_SIZE (sizeof(struct qdma_io_meta) + \ + sizeof(struct qbman_fle) * DPAA2_QDMA_MAX_FLE + \ + sizeof(struct qdma_sdd) * DPAA2_QDMA_MAX_SDD) +/** FLE pool cache size */ +#define QDMA_FLE_CACHE_SIZE(_num) (_num/(RTE_MAX_LCORE * 2)) + +/** Maximum possible H/W Queues on each core */ +#define MAX_HW_QUEUE_PER_CORE 64 + /** * Represents a QDMA device. * A single QDMA device exists which is combination of multiple DPDMAI rawdev's. @@ -45,6 +61,53 @@ struct qdma_hw_queue { uint32_t num_users; }; +/** Represents a QDMA virtual queue */ +struct qdma_virt_queue { + /** Status ring of the virtual queue */ + struct rte_ring *status_ring; + /** Associated hw queue */ + struct qdma_hw_queue *hw_queue; + /** Associated lcore id */ + uint32_t lcore_id; + /** States if this vq is in use or not */ + uint8_t in_use; + /** States if this vq has exclusively associated hw queue */ + uint8_t exclusive_hw_queue; + /* Total number of enqueues on this VQ */ + uint64_t num_enqueues; + /* Total number of dequeues from this VQ */ + uint64_t num_dequeues; +}; + +/** Represents a QDMA per core hw queues allocation in virtual mode */ +struct qdma_per_core_info { + /** list for allocated hw queues */ + struct qdma_hw_queue *hw_queues[MAX_HW_QUEUE_PER_CORE]; + /* Number of hw queues allocated for this core */ + uint16_t num_hw_queues; +}; + +/** Metadata which is stored with each operation */ +struct qdma_io_meta { + /** + * Context which is stored in the FLE pool (just before the FLE). + * QDMA job is stored as a this context as a part of metadata. + */ + uint64_t cnxt; + /** VQ ID is stored as a part of metadata of the enqueue command */ + uint64_t id; +}; + +/** Source/Destination Descriptor */ +struct qdma_sdd { + uint32_t rsv; + /** Stride configuration */ + uint32_t stride; + /** Route-by-port command */ + uint32_t rbpcmd; + uint32_t cmd; +} __attribute__((__packed__)); + /** Represents a DPDMAI raw device */ struct dpaa2_dpdmai_dev { /** Pointer to Next device instance */ diff --git a/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma.h b/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma.h new file mode 100644 index 0000000..29a1e4b --- /dev/null +++ b/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma.h @@ -0,0 +1,216 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright 2018 NXP + */ + +#ifndef __RTE_PMD_DPAA2_QDMA_H__ +#define __RTE_PMD_DPAA2_QDMA_H__ + +/** + * @file + * + * NXP dpaa2 QDMA specific structures. + * + */ + +/** Determines the mode of operation */ +enum { + /** + * Allocate a H/W queue per VQ i.e. Exclusive hardware queue for a VQ. + * This mode will have best performance. + */ + RTE_QDMA_MODE_HW, + /** + * A VQ shall not have an exclusive associated H/W queue. + * Rather a H/W Queue will be shared by multiple Virtual Queues. + * This mode will have intermediate data structures to support + * multi VQ to PQ mappings thus having some performance implications. + * Note: Even in this mode there is an option to allocate a H/W + * queue for a VQ. Please see 'RTE_QDMA_VQ_EXCLUSIVE_PQ' flag. + */ + RTE_QDMA_MODE_VIRTUAL +}; + +/** + * If user has configued a Virtual Queue mode, but for some particular VQ + * user needs an exclusive H/W queue associated (for better performance + * on that particular VQ), then user can pass this flag while creating the + * Virtual Queue. A H/W queue will be allocated corresponding to + * VQ which uses this flag. + */ +#define RTE_QDMA_VQ_EXCLUSIVE_PQ (1ULL) + +/** States if the source addresses is physical. */ +#define RTE_QDMA_JOB_SRC_PHY (1ULL) + +/** States if the destination addresses is physical. */ +#define RTE_QDMA_JOB_DEST_PHY (1ULL << 1) + +/** Provides QDMA device attributes */ +struct rte_qdma_attr { + /** total number of hw QDMA queues present */ + uint16_t num_hw_queues; +}; + +/** QDMA device configuration structure */ +struct rte_qdma_config { + /** Number of maximum hw queues to allocate per core. */ + uint16_t max_hw_queues_per_core; + /** Maximum number of VQ's to be used. */ + uint16_t max_vqs; + /** mode of operation - physical(h/w) or virtual */ + uint8_t mode; + /** + * User provides this as input to the driver as a size of the FLE pool. + * FLE's (and corresponding source/destination descriptors) are + * allocated by the driver at enqueue time to store src/dest and + * other data and are freed at the dequeue time. This determines the + * maximum number of inflight jobs on the QDMA device. This should + * be power of 2. + */ + int fle_pool_count; +}; + +/** Provides QDMA device statistics */ +struct rte_qdma_vq_stats { + /** States if this vq has exclusively associated hw queue */ + uint8_t exclusive_hw_queue; + /** Associated lcore id */ + uint32_t lcore_id; + /* Total number of enqueues on this VQ */ + uint64_t num_enqueues; + /* Total number of dequeues from this VQ */ + uint64_t num_dequeues; + /* total number of pending jobs in this VQ */ + uint64_t num_pending_jobs; +}; + +/** Determines a QDMA job */ +struct rte_qdma_job { + /** Source Address from where DMA is (to be) performed */ + uint64_t src; + /** Destination Address where DMA is (to be) done */ + uint64_t dest; + /** Length of the DMA operation in bytes. */ + uint32_t len; + /** See RTE_QDMA_JOB_ flags */ + uint32_t flags; + /** + * User can specify a context which will be maintained + * on the dequeue operation. + */ + uint64_t cnxt; + /** + * Status of the transaction. + * This is filled in the dequeue operation by the driver. + */ + uint8_t status; +}; + +/** + * Initialize the QDMA device. + * + * @returns + * - 0: Success. + * - <0: Error code. + */ +int __rte_experimental +rte_qdma_init(void); + +/** + * Get the QDMA attributes. + * + * @param qdma_attr + * QDMA attributes providing total number of hw queues etc. + */ +void __rte_experimental +rte_qdma_attr_get(struct rte_qdma_attr *qdma_attr); + +/** + * Reset the QDMA device. This API will completely reset the QDMA + * device, bringing it to original state as if only rte_qdma_init() API + * has been called. + * + * @returns + * - 0: Success. + * - <0: Error code. + */ +int __rte_experimental +rte_qdma_reset(void); + +/** + * Configure the QDMA device. + * + * @returns + * - 0: Success. + * - <0: Error code. + */ +int __rte_experimental +rte_qdma_configure(struct rte_qdma_config *qdma_config); + +/** + * Start the QDMA device. + * + * @returns + * - 0: Success. + * - <0: Error code. + */ +int __rte_experimental +rte_qdma_start(void); + +/** + * Create a Virtual Queue on a particular lcore id. + * This API can be called from any thread/core. User can create/destroy + * VQ's at runtime. + * + * @param lcore_id + * LCORE ID on which this particular queue would be associated with. + * @param flags + * RTE_QDMA_VQ_ flags. See macro definitions. + * + * @returns + * - >= 0: Virtual queue ID. + * - <0: Error code. + */ +int __rte_experimental +rte_qdma_vq_create(uint32_t lcore_id, uint32_t flags); + +/** + * Get a Virtual Queue statistics. + * + * @param vq_id + * Virtual Queue ID. + * @param vq_stats + * VQ statistics structure which will be filled in by the driver. + */ +void __rte_experimental +rte_qdma_vq_stats(uint16_t vq_id, + struct rte_qdma_vq_stats *vq_stats); + +/** + * Destroy the Virtual Queue specified by vq_id. + * This API can be called from any thread/core. User can create/destroy + * VQ's at runtime. + * + * @param vq_id + * Virtual Queue ID which needs to be deinialized. + * + * @returns + * - 0: Success. + * - <0: Error code. + */ +int __rte_experimental +rte_qdma_vq_destroy(uint16_t vq_id); + +/** + * Stop QDMA device. + */ +void __rte_experimental +rte_qdma_stop(void); + +/** + * Destroy the QDMA device. + */ +void __rte_experimental +rte_qdma_destroy(void); + +#endif /* __RTE_PMD_DPAA2_QDMA_H__*/ diff --git a/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma_version.map b/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma_version.map index 33d2379..7335c35 100644 --- a/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma_version.map +++ b/drivers/raw/dpaa2_qdma/rte_pmd_dpaa2_qdma_version.map @@ -1,4 +1,16 @@ EXPERIMENTAL { + global: + + rte_qdma_attr_get; + rte_qdma_configure; + rte_qdma_destroy; + rte_qdma_init; + rte_qdma_reset; + rte_qdma_start; + rte_qdma_stop; + rte_qdma_vq_create; + rte_qdma_vq_destroy; + rte_qdma_vq_stats; local: *; }; -- 1.9.1