From patchwork Mon Jan 8 08:27:49 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Amit Prakash Shukla X-Patchwork-Id: 135797 X-Patchwork-Delegate: thomas@monjalon.net Return-Path: X-Original-To: patchwork@inbox.dpdk.org Delivered-To: patchwork@inbox.dpdk.org Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 14E3443861; Mon, 8 Jan 2024 09:28:11 +0100 (CET) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id E001D402B0; Mon, 8 Jan 2024 09:28:10 +0100 (CET) Received: from mx0b-0016f401.pphosted.com (mx0b-0016f401.pphosted.com [67.231.156.173]) by mails.dpdk.org (Postfix) with ESMTP id D8BB540263 for ; Mon, 8 Jan 2024 09:28:09 +0100 (CET) Received: from pps.filterd (m0045851.ppops.net [127.0.0.1]) by mx0b-0016f401.pphosted.com (8.17.1.24/8.17.1.24) with ESMTP id 407NP5H8009772; Mon, 8 Jan 2024 00:28:09 -0800 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=marvell.com; h= from:to:cc:subject:date:message-id:mime-version :content-transfer-encoding:content-type; s=pfpt0220; bh=2c1Kl4RB jd2SPEwjyzhQ0QbvbnwwUM8h1YcYYQ4Npuk=; b=ZerVDJ9cPBB2fnoVP8rrx80b 0jrKmIF+kn3uddX2UzPLyTvcwvUSDvqqeWEp7kbgCb5WNrUElddM4S61WgxQiEE7 FZvA0xOlxSqp5X1OabwktmufvixQEsjxUmHkSsp8wdZk0f+8DJlsQuBC9yCaj2Md A5W+ZZ4k/Hyn4KV9o8b1KZWMoashClPmjJ5cOJcvCrMC7QrKBP42C/ojC3SzeUbE pm15oMmGQUxzVMFzCKJ7D7TN5Mb7i+puvdbjmIsLBPDXponXvsi6TOGsm9nEwZRj 5H9KCpPgYzxscZHpFRxW52tCx7JVaVKi/hVHJaBQ55qLMea3q0US/buTBWN7Qg== Received: from dc5-exch02.marvell.com ([199.233.59.182]) by mx0b-0016f401.pphosted.com (PPS) with ESMTPS id 3vf78n4412-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-SHA384 bits=256 verify=NOT); Mon, 08 Jan 2024 00:28:09 -0800 (PST) Received: from DC5-EXCH01.marvell.com (10.69.176.38) by DC5-EXCH02.marvell.com (10.69.176.39) with Microsoft SMTP Server (TLS) id 15.0.1497.48; Mon, 8 Jan 2024 00:28:06 -0800 Received: from maili.marvell.com (10.69.176.80) by DC5-EXCH01.marvell.com (10.69.176.38) with Microsoft SMTP Server id 15.0.1497.48 via Frontend Transport; Mon, 8 Jan 2024 00:28:06 -0800 Received: from cavium-OptiPlex-5090-BM14.. (unknown [10.28.36.157]) by maili.marvell.com (Postfix) with ESMTP id 057603F704B; Mon, 8 Jan 2024 00:28:03 -0800 (PST) From: Amit Prakash Shukla To: Cheng Jiang CC: , , , , , , , Amit Prakash Shukla Subject: [PATCH] app/dma-perf: support bi-directional transfer Date: Mon, 8 Jan 2024 13:57:49 +0530 Message-ID: <20240108082749.1016345-1-amitprakashs@marvell.com> X-Mailer: git-send-email 2.34.1 MIME-Version: 1.0 X-Proofpoint-ORIG-GUID: q94BP3opKxn8Uq0eQnXWV9PxqfRqhf0V X-Proofpoint-GUID: q94BP3opKxn8Uq0eQnXWV9PxqfRqhf0V X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.272,Aquarius:18.0.997,Hydra:6.0.619,FMLib:17.11.176.26 definitions=2023-12-09_02,2023-12-07_01,2023-05-22_02 X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Adds bi-directional DMA transfer support to test performance. Signed-off-by: Amit Prakash Shukla --- Depends-on: series-30357 ("PCI Dev and SG copy support") app/test-dma-perf/benchmark.c | 89 +++++++++++++++++++++++++---------- app/test-dma-perf/config.ini | 5 ++ app/test-dma-perf/main.c | 21 +++++++-- app/test-dma-perf/main.h | 1 + 4 files changed, 88 insertions(+), 28 deletions(-) diff --git a/app/test-dma-perf/benchmark.c b/app/test-dma-perf/benchmark.c index 4530bd98ce..91ba0f4718 100644 --- a/app/test-dma-perf/benchmark.c +++ b/app/test-dma-perf/benchmark.c @@ -144,12 +144,19 @@ cache_flush_buf(__rte_unused struct rte_mbuf **array, static int vchan_data_populate(uint32_t dev_id, struct rte_dma_vchan_conf *qconf, - struct test_configure *cfg) + struct test_configure *cfg, uint16_t dev_num) { struct rte_dma_info info; qconf->direction = cfg->transfer_dir; + /* If its a bi-directional test, configure odd device for inbound dma + * transfer and even device for outbound dma transfer. + */ + if (cfg->is_bidir) + qconf->direction = (dev_num % 2) ? RTE_DMA_DIR_MEM_TO_DEV : + RTE_DMA_DIR_DEV_TO_MEM; + rte_dma_info_get(dev_id, &info); if (!(RTE_BIT64(qconf->direction) & info.dev_capa)) return -1; @@ -181,14 +188,15 @@ vchan_data_populate(uint32_t dev_id, struct rte_dma_vchan_conf *qconf, /* Configuration of device. */ static void -configure_dmadev_queue(uint32_t dev_id, struct test_configure *cfg, uint8_t ptrs_max) +configure_dmadev_queue(uint32_t dev_id, struct test_configure *cfg, uint8_t ptrs_max, + uint16_t dev_num) { uint16_t vchan = 0; struct rte_dma_info info; struct rte_dma_conf dev_config = { .nb_vchans = 1 }; struct rte_dma_vchan_conf qconf = { 0 }; - if (vchan_data_populate(dev_id, &qconf, cfg) != 0) + if (vchan_data_populate(dev_id, &qconf, cfg, dev_num) != 0) rte_exit(EXIT_FAILURE, "Error with vchan data populate.\n"); if (rte_dma_configure(dev_id, &dev_config) != 0) @@ -235,7 +243,7 @@ config_dmadevs(struct test_configure *cfg) } ldm->dma_ids[i] = dev_id; - configure_dmadev_queue(dev_id, cfg, ptrs_max); + configure_dmadev_queue(dev_id, cfg, ptrs_max, nb_dmadevs); ++nb_dmadevs; } @@ -433,7 +441,8 @@ setup_memory_env(struct test_configure *cfg, struct rte_dma_sge **src_sges, struct rte_dma_sge **dst_sges) { static struct rte_mbuf_ext_shared_info *ext_buf_info; - unsigned int buf_size = cfg->buf_size.cur; + unsigned int cur_buf_size = cfg->buf_size.cur; + unsigned int buf_size = cur_buf_size + RTE_PKTMBUF_HEADROOM; unsigned int nr_sockets; uint32_t nr_buf = cfg->nr_buf; uint32_t i; @@ -449,7 +458,7 @@ setup_memory_env(struct test_configure *cfg, nr_buf, 0, 0, - buf_size + RTE_PKTMBUF_HEADROOM, + buf_size, cfg->src_numa_node); if (src_pool == NULL) { PRINT_ERR("Error with source mempool creation.\n"); @@ -460,7 +469,7 @@ setup_memory_env(struct test_configure *cfg, nr_buf, 0, 0, - buf_size + RTE_PKTMBUF_HEADROOM, + buf_size, cfg->dst_numa_node); if (dst_pool == NULL) { PRINT_ERR("Error with destination mempool creation.\n"); @@ -490,8 +499,8 @@ setup_memory_env(struct test_configure *cfg, } for (i = 0; i < nr_buf; i++) { - memset(rte_pktmbuf_mtod((*srcs)[i], void *), rte_rand(), buf_size); - memset(rte_pktmbuf_mtod((*dsts)[i], void *), 0, buf_size); + memset(rte_pktmbuf_mtod((*srcs)[i], void *), rte_rand(), cur_buf_size); + memset(rte_pktmbuf_mtod((*dsts)[i], void *), 0, cur_buf_size); } if (cfg->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM || @@ -503,24 +512,38 @@ setup_memory_env(struct test_configure *cfg, } } - if (cfg->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM) { + if (cfg->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM && !cfg->is_bidir) { + ext_buf_info->free_cb = dummy_free_ext_buf; + ext_buf_info->fcb_opaque = NULL; + for (i = 0; i < nr_buf; i++) { + /* Using mbuf structure to hold remote iova address. */ + rte_pktmbuf_attach_extbuf((*srcs)[i], (void *)(cfg->raddr + (i * buf_size)), + (rte_iova_t)(cfg->raddr + (i * buf_size)), 0, + ext_buf_info); + rte_mbuf_ext_refcnt_update(ext_buf_info, 1); + } + } + + if (cfg->transfer_dir == RTE_DMA_DIR_MEM_TO_DEV && !cfg->is_bidir) { ext_buf_info->free_cb = dummy_free_ext_buf; ext_buf_info->fcb_opaque = NULL; for (i = 0; i < nr_buf; i++) { /* Using mbuf structure to hold remote iova address. */ - rte_pktmbuf_attach_extbuf((*srcs)[i], (void *)cfg->raddr, - (rte_iova_t)cfg->raddr, 0, ext_buf_info); + rte_pktmbuf_attach_extbuf((*dsts)[i], (void *)(cfg->raddr + (i * buf_size)), + (rte_iova_t)(cfg->raddr + (i * buf_size)), 0, + ext_buf_info); rte_mbuf_ext_refcnt_update(ext_buf_info, 1); } } - if (cfg->transfer_dir == RTE_DMA_DIR_MEM_TO_DEV) { + if (cfg->is_bidir) { ext_buf_info->free_cb = dummy_free_ext_buf; ext_buf_info->fcb_opaque = NULL; for (i = 0; i < nr_buf; i++) { /* Using mbuf structure to hold remote iova address. */ - rte_pktmbuf_attach_extbuf((*dsts)[i], (void *)cfg->raddr, - (rte_iova_t)cfg->raddr, 0, ext_buf_info); + rte_pktmbuf_attach_extbuf((*srcs)[i], (void *)(cfg->raddr + (i * buf_size)), + (rte_iova_t)(cfg->raddr + (i * buf_size)), 0, + ext_buf_info); rte_mbuf_ext_refcnt_update(ext_buf_info, 1); } } @@ -646,16 +669,30 @@ mem_copy_benchmark(struct test_configure *cfg) lcores[i]->nr_buf = (uint32_t)(nr_buf / nb_workers); lcores[i]->buf_size = buf_size; lcores[i]->test_secs = test_secs; - lcores[i]->srcs = srcs + offset; - lcores[i]->dsts = dsts + offset; lcores[i]->scenario_id = cfg->scenario_id; lcores[i]->lcore_id = lcore_id; - if (cfg->is_sg) { - lcores[i]->src_ptrs = cfg->src_ptrs; - lcores[i]->dst_ptrs = cfg->dst_ptrs; - lcores[i]->src_sges = src_sges + (nr_sgsrc / nb_workers * i); - lcores[i]->dst_sges = dst_sges + (nr_sgdst / nb_workers * i); + /* Number of workers is equal to number of devices. In case of bi-directional + * dma, use 1 device for mem-to-dev and 1 device for dev-to-mem. + */ + if (cfg->is_dma && cfg->is_bidir && (i % 2 != 0)) { + lcores[i]->dsts = srcs + offset; + lcores[i]->srcs = dsts + offset; + if (cfg->is_sg) { + lcores[i]->dst_ptrs = cfg->src_ptrs; + lcores[i]->src_ptrs = cfg->dst_ptrs; + lcores[i]->dst_sges = src_sges + (nr_sgsrc / nb_workers * i); + lcores[i]->src_sges = dst_sges + (nr_sgdst / nb_workers * i); + } + } else { + lcores[i]->srcs = srcs + offset; + lcores[i]->dsts = dsts + offset; + if (cfg->is_sg) { + lcores[i]->src_ptrs = cfg->src_ptrs; + lcores[i]->dst_ptrs = cfg->dst_ptrs; + lcores[i]->src_sges = src_sges + (nr_sgsrc / nb_workers * i); + lcores[i]->dst_sges = dst_sges + (nr_sgdst / nb_workers * i); + } } if (cfg->is_dma) { @@ -699,7 +736,7 @@ mem_copy_benchmark(struct test_configure *cfg) rte_eal_mp_wait_lcore(); - if (!cfg->is_sg) { + if (!cfg->is_sg && cfg->transfer_dir == RTE_DMA_DIR_MEM_TO_MEM) { for (i = 0; i < (nr_buf / nb_workers) * nb_workers; i++) { if (memcmp(rte_pktmbuf_mtod(srcs[i], void *), rte_pktmbuf_mtod(dsts[i], void *), @@ -709,7 +746,7 @@ mem_copy_benchmark(struct test_configure *cfg) goto out; } } - } else { + } else if (cfg->is_sg && cfg->transfer_dir == RTE_DMA_DIR_MEM_TO_MEM) { size_t src_remsz = buf_size % cfg->src_ptrs; size_t dst_remsz = buf_size % cfg->dst_ptrs; size_t src_sz = buf_size / cfg->src_ptrs; @@ -756,6 +793,8 @@ mem_copy_benchmark(struct test_configure *cfg) calc_result(buf_size, nr_buf, nb_workers, test_secs, lcores[i]->worker_info.test_cpl, &memory, &avg_cycles, &bandwidth, &mops); + if (cfg->is_bidir) + printf("%s direction\n", i % 2 ? "MEM-to-DEV" : "DEV-to-MEM"); output_result(cfg, lcores[i], kick_batch, avg_cycles, buf_size, nr_buf / nb_workers, memory, bandwidth, mops); mops_total += mops; @@ -769,7 +808,7 @@ mem_copy_benchmark(struct test_configure *cfg) out: - if (cfg->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM) + if (cfg->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM || cfg->is_bidir) m = srcs; else if (cfg->transfer_dir == RTE_DMA_DIR_MEM_TO_DEV) m = dsts; diff --git a/app/test-dma-perf/config.ini b/app/test-dma-perf/config.ini index f460b93414..a764ef7e7f 100644 --- a/app/test-dma-perf/config.ini +++ b/app/test-dma-perf/config.ini @@ -55,6 +55,10 @@ ; "pfid" denotes PF-id to be used for data transfer ; "vfid" denotes VF-id of PF-id to be used for data transfer. +; "xfer_mode" denotes mode of data transfer. It can take 2 values: +; 0 - unidirection transfer based on direction configured (default). +; 1 - Bi-directional transfer based on direction configured (mem-to-dev and dev-to-mem). + ; =========== End of "mem to dev" and "dev to mem" config parameters. ============== [case1] @@ -89,6 +93,7 @@ eal_args=--in-memory --file-prefix=test skip=1 type=DMA_MEM_COPY direction=2 +xfer_mode=0 raddr=0x200000000 scoreid=0 dcoreid=0 diff --git a/app/test-dma-perf/main.c b/app/test-dma-perf/main.c index e81eca14e1..be91405305 100644 --- a/app/test-dma-perf/main.c +++ b/app/test-dma-perf/main.c @@ -342,6 +342,7 @@ load_configs(const char *path) *src_ptrs_str, *dst_ptrs_str; const char *skip; const char *raddr, *scoreid, *dcoreid, *vfid, *pfid; + const char *xfer_mode; int args_nr, nb_vp; bool is_dma; @@ -393,6 +394,20 @@ load_configs(const char *path) test_case->is_valid = false; continue; } + xfer_mode = rte_cfgfile_get_entry(cfgfile, section_name, "xfer_mode"); + if (xfer_mode) { + int xmode = atoi(xfer_mode); + if (xmode == 1) { + if (test_case->transfer_dir == RTE_DMA_DIR_MEM_TO_MEM) { + printf("Error: Invalid configuration. For mem to" + " mem dma transfer bi-directional cannot be" + " configured.\n"); + test_case->is_valid = false; + continue; + } + test_case->is_bidir = true; + } + } is_dma = true; } else if (strcmp(case_type, CPU_MEM_COPY) == 0) { test_case->test_type = TEST_TYPE_CPU_MEM_COPY; @@ -405,7 +420,7 @@ load_configs(const char *path) } if (test_case->transfer_dir == RTE_DMA_DIR_MEM_TO_DEV || - test_case->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM) { + test_case->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM) { char *endptr; raddr = rte_cfgfile_get_entry(cfgfile, section_name, "raddr"); @@ -434,7 +449,7 @@ load_configs(const char *path) } - if (test_case->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM) { + if (test_case->transfer_dir == RTE_DMA_DIR_DEV_TO_MEM || test_case->is_bidir) { scoreid = rte_cfgfile_get_entry(cfgfile, section_name, "scoreid"); if (scoreid == NULL) { printf("Error: No scoreid configured for case%d.\n", i + 1); @@ -444,7 +459,7 @@ load_configs(const char *path) test_case->scoreid = (uint8_t)atoi(scoreid); } - if (test_case->transfer_dir == RTE_DMA_DIR_MEM_TO_DEV) { + if (test_case->transfer_dir == RTE_DMA_DIR_MEM_TO_DEV || test_case->is_bidir) { dcoreid = rte_cfgfile_get_entry(cfgfile, section_name, "dcoreid"); if (dcoreid == NULL) { printf("Error: No dcoreid configured for case%d.\n", i + 1); diff --git a/app/test-dma-perf/main.h b/app/test-dma-perf/main.h index 31e0bf71c9..478c2a1c6d 100644 --- a/app/test-dma-perf/main.h +++ b/app/test-dma-perf/main.h @@ -66,6 +66,7 @@ struct test_configure { uint8_t pfid; uint16_t vfid; uintptr_t raddr; + bool is_bidir; }; int mem_copy_benchmark(struct test_configure *cfg);