aclmdlExecuteAsync后调用aclrtSynchronizeStream返回507011。日志见附件

请问各位大佬，用Atlas200可以通过PCIe的方式给飞腾处理器D2000进行AI业务的加速吗？

如何设置开机自启动我的程序放在了docker里，也就是华为官方提供的infer-modelzoo:21.0.4里面，现在我希望开机时能够自启动docker里的脚本程序目前：我设置了开机自启动docker，即docker containerID update --restart=always然后我在/etc/rc.local文件里加上了 docker exec containerid -d bash test.sh然而重启后并没有运行，所以Atlas该如何设置开机自启脚本

问题现象：SD卡模块下，系统可正常启动，USB虚拟网卡及千兆以太网 均可正常通讯。切换为EMMC模式后，电脑测显示千兆以太网可连接，底板千兆网网络指示灯在日志输出加载内核5S左右后闪烁，但是电脑无法Ping通192.168.0.2？？？电脑端显示设备以太网连接如下：EMMC 模式底板配置电阻件附件“Atlas200模块_EMMC模式配置电阻图纸”整个Atlas200模块设计图纸见附件。EMMC模式串口日志如下：52707178b806f00080current[1], last[0].00000105Get board ID:4SRAM INIT timeout:441msSNPS PHY PLL state Error [0x868]:0x00000000 [0x86C]:0x00000000 654ms0000010c0000010c - 0000010c0000010f00000000000000000000029200000292 - 0000029500000000 0000031e 00000308pmu second pmu initial:0x00000003.current[2], last[0].current[3], last[0].0.5c-init okPartition 0 is not a vaild NVE partition[NVE]current_id = 1,nve_age = 1,age = 0[NVE]current_id = 2,nve_age = 2,age = 1Partition 3 is not a vaild NVE partition[NVE]current_id = 2 number of nvbin = 24, version = -1828191725[NVE]nve_offser = 0x000038a0 nve_valid_items = 22current[4], las0].cpu mode 3, efuse:268!cpu pll recfg function,success---200764!current[5], last[0].current[6], last[0].current[7], last[0].Ddr_freq->ddr_freq 1600 !uc sector[0] valid flag not enabled!uc sector[1] valid flag not enabled![UC]ddr_iecc_en:0freq 60 vol 2:ch_mask: 0x000000ffch_lr: 0x00000003ch_num: 0x00000008rank_num: 0x00000001Manufacturer ID[0]:0x00000013Manufacturer ID[1]:0x00000013Manufacturer ID[2]:0x00000013Manufacturer ID[3]:0x00000013Manufacturer ID[4]:0x00000013Manufacturer ID[5]:0x00000013Manufacturer ID[6]:0x00000013Manufacturer ID[7]:0x00000013Total size:0x00000008DDR info:0x00000813DDR_TYPE_8CH_ALL 1_rank 8G !DDR HHA interleave enabled!idx: 0 start_addr_l: 0x00000000 start_addr_h: 0x00000000 length: 0x00000800idx: 1 start_addr_l: 0x80000000 start_addr_h: 0x00000008 length: 0x00001800uce init entryuce profile total is 0x0000032cuce init endfreq 400 vol 2:freq 1600 vol 2:HVrefThres:8uce init entryuce profile total is 0x00000298uce init endtransfer ddr data to l2_buf 0x40000000.DDR Total Capacity: 8192 MB.DDR Rank Type: single.DDR Channel with DRAM: A B C D E F G HDDR Init 847 mscurrent[8], last[0].maintain:00000000, static:00000000current[9], last[0].[STID]adcin7:0x0000009c.[STID]adcin7_remap:0x00000001.[STID]adcin8:0x0000009d.[STID]adcin8_remap:0x00000001.current[10], last[0].SPI FLASH download uefi.current[25], last[0].g_imgNum = 1current[26], last[0].SPI FLASH download uefi finish.sicurrent[34], last[0].ns, succurrent[35], last[0].si succurrent[36], last[0].s vr bgns vr, sccurrent[37], last[0].sv sucsv suchigh mntn_config=0xa0000000, low mntn_config=0x00000000lpm3 will WFEdisreset A55.current[0], last[0].aicore_power_on:312----core1 step1aicore_pmu_up:233----repair step 5_aicore0_buck_on:164---state -0x00000007--cfg:0x00000064--buck voltage:0x0000006 4_aicore1_buck_on:222---state -0x0000000f--cfg:0x00000064--buck voltage:0x0000006 4_aimemory0_1_buck_on:193---state -0x0000000f--cfg:0x00000064--buck voltage:0x000 00064aicore_pmu_up:257----pmu up step1aicore0_power_reset:74aicore0_power_reset:83aicore1_power_reset:88aicore1_power_reset:97aicore1_isolation_enable:112----repair step 4aicore_power_on:331----power on success.DDR HHA interleave endump static ddr 0x0@0x0rc_ep_mode = 1dump static ddr 0x0@0x0dump static ddr 0x0@0x0UEFI Platform GICD init...OKUEFI ltf startUEFI ctf okNOTICE: B T FNOTICE: 2NOTICE: 31UEFI ltf finishInstall gSpiFlashProtocol Protocol Interface ok!AgingResultSaveDxeEntry ............. finishPcie Port Init Start.Finish pcie ep mode register config.Start nve init(4626).UEFI get board id from DDR:4[nve_find_valid_partition]current_id = 2 valid_items = 22, version = -1, sha_sup port = -1Finish nve init(5159).not mini rc environment: not install SDDriverUEFI get slot id from DDR:11Now start to auto boot linux.(5166)This is Flash Boot EntryStart dump(5171)mntn config is 0xA000000000000000pcie link status=0x0Finish dump(5178)Enter4BCmdReadEnter4BCmd(2490):ok!Enter4BCmd..............................OKcheck img(0x13C0000) precode OKcheck img(0x8C0000) precode OKcheck img(0x14C0000) precode OKcheck img(0x400000) precode OKcheck img(0x840000) precode OKcopy img(0x13C0000 | 0xDD7C000 | 0xFC000)copy img(0x8C0000 | 0x7E7C000 | 0xAFC000)copy img(0x14C0000 | 0x880000000 | 0x27FC000)copy img(0x400000 | 0xA40000 | 0x3C000)copy img(0x840000 | 0x12E80000 | 0x7C000)Exit4BCmdReadExit4BCmd(2449):ok!Exit4BCmdRead..............................OK[get_reboot_reason_from_pmu]:[708L] This is cold bootV-TEENOTICE: t-p12.C-dtbu ns.C-Imageu ns.C-filesystemu ns.boardid = 4Update FDT[0] base = 0x0, length = 0x80000000[1] base = 0x880000000, length = 0x180000000soc node = 38328Lpapnv->cpu_idle_en = 0Lpapnv->acpu_num = 8cpus node = 184cpu-map node = 1440cluster0 node = 1452cpus node = 184[cpu@0] subnode = 228del cpu-idle-states[cpu@1] subnode = 356del cpu-idle-states[cpu@2] subnode = 484del cpu-idle-states[cpu@3] subnode = 612del cpu-idle-states[cpu@4] subnode = 740del cpu-idle-states[cpu@5] subnode = 868del cpu-idle-states[cpu@6] subnode = 996del cpu-idle-states[cpu@7] subnode = 1124del cpu-idle-states[GetKernelLogBuf]:[468L] tmp=0x1fc000[GetKernelLogBuf]:[481L] end=10 argsBuf[0x6fe04000 default_hugepagesz=2M loglevel=1 reboot_reason=AP_S_COLDBOOT himntn=1010000000000000000000000000000000000000000000000000000000000000 slotid=11 boardid=004][GetKernelLogBuf]:[489L] kernel log buffer[0x6FE04000][0x1FC000]Start to boot LinuxV-LPM3LPMCU has covered bootrom keypoint.NOTICE: l-p12.W-LPM3Start to jump Linux kernel(26476)

MindX SDK -- 医学人体关键点检测参考设计案例1 案例概述1.1 概要描述在本系统中，目的是基于MindX SDK，在华为云昇腾平台上，开发端到端医学人体关键点检测的参考设计，实现对应体态的关键点检测的功能，达到精度要求。1.2 模型介绍本项目用到了两个模型：1)用于人体目标检测的FasterRcnn模型FasterRcnn模型相关文件可以在此处Faster R-CNN-昇腾社区 (hiascend.com)。2)用于人体关键点检测的HRNet-w48模型模型相关文件可以在此处获取：1.3 实现流程1、基础环境： MindX SDK ： 版本2.0.4 获取方式：cid:link_6 ubuntu ： 版本:18.04 获取方式：请上ubuntu官网获取 Ascend-CANN-toolkit: 版本：5.0.4 获取方式：cid:link_5 2、模型转换：利用atc工具完成模型转化：pose_model_384x288.onnx.onxx --> pose_model_384_288_noAipp_fp16.om3、python主程序代码开发技术流程图如下：1.4 代码地址cid:link_32 软件方案介绍2.1 项目方案架构介绍本系统设计了不同的功能模块。主要流程为：1） 待检测试图片由插件appsrc0输入并解码，同时通过opencv获取图像的信息由appsrc1送入推理，用mxpi_imageresize0插件将图像放缩至满足检测模型要求的输入图像大小要求。随后，图片输入模型推理插件mxpi_tensorinfer0进行处理。获取得到人像识别框，并根据该框对原图像进行仿射变换，进行数据增强。2） 对仿射变换后的图片进行转向量与归一化的操作，通过sendprotobuf接口将数据送入HRNET-w48模型进行推理，可以得到80个点位的像素值及原图的体位分类值，通过后处理可对图像进行可视化。表2.1 系统方案（1）中各模块功能：序号子系统功能描述1图片输入接收外部调用接口的输入图片。2图片数据输入接受外部调用接口的输入图片尺寸信息。3图片解码用于解码，将jpg格式图片解码为YUV4图像处理对解码后的YUV格式的图像进行放缩。5模型推理插件fasterrcnn人体目标检测。6输出输出检测信息表2.2 系统方案（2）中各模块功能：序号子系统功能描述1处理后的数据输入接收外部调用接口的输入数据。2模型推理插件hrnet-w48人体目标检测。3输出输出检测信息(体位分类 与 体表检测)2.2 代码目录结构与说明本项目名为hrnet-w48医学人体关键点检测，项目目录如下所示：. |-------- data // 存放测试图片 |-------- model // 存放模型 |-------- main.py // 主程序 |-------- pipeline | |---- model1.pipeline // 模型1pipeline配置文件 | |---- model2.pipeline // 模型2pipeline配置文件 |-------- evaluate.py // 精度测试程序 |-------- README.md |-------- run.sh3 开发介绍3.1 环境依赖说明推荐系统为ubuntu 18.04,环境软件和版本如下：软件名称版本说明获取方式MindX SDK2.0.4mxVision软件包链接ubuntu18.04操作系统请上ubuntu官网获取Ascend-CANN-toolkit5.0.4Ascend-cann-toolkit开发套件包链接3.2 环境搭建在编译运行项目前，需要设置环境变量：MindSDK 环境变量：. ${SDK-path}/set_env.shCANN 环境变量：. ${ascend-toolkit-path}/set_env.sh环境变量介绍SDK-path: SDK mxVision 安装路径 ascend-toolkit-path: CANN 安装路径。3.3 hrnet-w48模型转换医学人体关键点检测采用自行训练的hrnet-w48模型。我们需要借助于ATC工具将onnx模型转化为对应的om模型。具体步骤如下：步骤1 获取模型onnx文件 ，下载链接为下载步骤2 将获取到的HRnet-w48模型onnx文件存放至：“项目所在目录/model”步骤3 模型转换在确保环境变量设置正确后，在onnx文件所在目录下执行以下命令：atc --model=pose_model_384x288.onnx --framework=5 --output=pose_model_384_288_noAipp_fp16 --soc_version=Ascend310 --output_type=FP16执行完模型转换脚本后，若提示如下信息说明模型转换成功，会在output参数指定的路径下生成pose_model_384_288_noAipp_fp16.om模型文件。ATC run success 4 编译与运行接下来进行模型的安装运行，具体步骤如下：步骤1 获取om模型（见前一章节）步骤2 配置pipeline根据所需场景，配置pipeline1与pipeline2文件，调整路径参数等。fasterrcnn.pipeline:"mxpi_tensorinfer0": { "props": { "modelPath": "./model/fasterrcnn_mindspore_dvpp.om" }, "factory": "mxpi_tensorinfer", "next":"appsink0" } #修改om文件存放的路径hrnet.pipeline:"mxpi_tensorinfer0": { "props": { "modelPath": "./model/pose_model_384_288_noAipp_fp16.om" }, "factory": "mxpi_tensorinfer", "next":"appsink0" } #修改om文件存放的路径步骤3 存放图片，执行模型进行测试将测试图片存放至主目录下，修改main.py中的图片存放路径以及人像分割后的存储路径的相关代码： 测试图片 【注】本模型定位场景为医学应用下的人体关键点检测，故只针对特定体态进行识别，建议从提供的测试图片或根据对应体态自行拍摄图片进行测试。FILE_PATH = '' #测试图片路径 SAVEPATH = '' #测试结果保存路径,图片保存路径需提前手动创建然后执行run.sh文件(执行前需要通过命令提供执行权限）：./run.sh输出的图片即为样例的人像分割后的图片。5 精度测试对测试集中的3651张图片进行精度测试，具体步骤如下：步骤1 获取测试集的图片,确保测试集的输入图片为jpg格式。获取测试集的标注文件（person_keypoints_val2017.json）。存放于data路径下。 （数据集暂未开源）步骤2 修改evaluate.py中的测试集图片存放路径：coco_anno = ' ' #485行，修改为测试集的标注文件路径 coco_anno = ' ' #557行，修改为测试集的标注文件路径 IMAGEFOLDER = ' ' #677行，修改为测试集的路径 FILE_ANNO = ' ' #678行，修改为测试集的标注文件路径 TXT = '' #689行，体位分类情况,请提前在文件夹对应位置创建同名文本文件步骤3 找到pycocotools源代码（cocoeval.py文件），修改kpt_oks_sigmas的值如下：（此为自制数据集的标注误差）self.kpt_oks_sigmas = np.array( [0.008784891334757884, 0.007934237422930233, 0.008130125823095129, 0.010598041983050829, 0.010611059330895747, 0.013854384373318705, 0.012312922863012592, 0.011434032335303733, 0.011415450225720266, 0.013836090153967652, 0.010062551495331999, 0.012966371944947402, 0.014659796595795808, 0.017770972497132317, 0.030104085032138312, 0.014557348480916475, 0.01536017320369201, 0.012869410502722093, 0.015257251871497391, 0.010375051254335216, 0.011380203661720942, 0.01414838952723429, 0.01382237259730266, 0.013735144644847687, 0.01472162976531488, 0.009619962495050469, 0.008610254442541486, 0.029481674119348003, 0.024059201435679762, 0.042427130812542665, 0.040127749201969376, 0.0525428294612648, 0.026481463695494662, 0.027975991688555843, 0.025593756110373064, 0.027253280146480565, 0.027454019180679124, 0.025064436713274765, 0.02099536659024268, 0.020033183495373304, 0.015050220402346052, 0.015267670545143495, 0.02596082400646692, 0.02923469684259541, 0.027801624530848095, 0.021826487778693508, 0.015060326275153427, 0.014315148040325337, 0.02596082400646692, 0.02923469684259541, 0.027801624530848095, 0.021826487778693508, 0.015060326275153427, 0.014315148040325337, 0.007395141152098735, 0.00762250652117458, 0.010602288831543757, 0.0090104333962127, 0.010453948805939391, 0.010370213072066961, 0.022732842847710055, 0.06232777881671897, 0.045665339866036256, 0.022732842847710055, 0.06232777881671897, 0.045665339866036256, 0.02359125146550032, 0.02359125146550032, 0.031776475097392176, 0.0270604129014659, 0.03426748613886749, 0.03807683079471247, 0.01993400901401409, 0.01993400901401409, 0.032237205934095854, 0.032237205934095854, 0.017419685722172645, 0.017419685722172645, 0.03147083587813606, 0.03147083587813606], dtype=np.float32 )步骤4 修改run.sh最后的执行文件名称：python3.9 evaluate.py并执行：./run.sh最终可以得到测试的精度结果及性能结果。精度测试指标如下：分类指标-体位识别准确率姿态估计-关键点检测精度APPytorch框架96.47%78.3Mindx-sdk架构95.2%77.26注：由于模型一（FasterRcnn）未能正常完成模型转换（算子未适配），是在华为仓内找到的替换模型，精度误差来自相似模型之间的精度差距，与模型二（HRNet-w48）无关。性能测试结果如下：在Atlas-200dk上进行对整个测试集上进行性能测试，未进行模型1的推理时间统计（因为该模型来自华为仓内，未进行任何调整，无需加入性能测试），模型2(hrnet-w48)的推理总花费时间为161.89s，fps=3651/161=22.67.原modelzoo下的hrnet-w48在五百张图片上推理耗时551s，fps=1.1.可见本模型在模型上的优化可以很好的提升性能。6 参考链接模型类型模型名称参考链接人体检测模型FasterrcnnFaster R-CNN-昇腾社区 (hiascend.com)关键点检测模型-自训练HRNet-w48models: Models of MindSpore - Gitee.com

但是你好，我想问一下我运行图片分类样例就是resnet50那个的时候，下载图片之后，data里的两个.jpg的图片是红色的然后运行Python文件trandferPic.py的时候虽然运行成功，但是并没有把我的图片转换成.bin格式的，不知道为啥。

请问 Atlas200DK 可以作为第三方机载平台搭载到 大疆M300RTK吗？

安装过程 Verifying archive integrity...  100%   SHA256 checksums are OK. All good. Uncompressing ASCEND_RUN_PACKAGE  100% [Toolkit] [20221103-05:16:30] [INFO] LogFile:/var/log/ascend_seclog/ascend_toolkit_install.log [Toolkit] [20221103-05:16:30] [INFO] install start [Toolkit] [20221103-05:16:30] [INFO] The install path is /usr/local/Ascend ! [Toolkit] [20221103-05:16:30] [INFO] install package Ascend-acllib-1.77.t21.0.b210-linux.aarch64.run start [Toolkit] [20221103-05:16:36] [INFO] Ascend-acllib-1.77.t21.0.b210-linux.aarch64.run --full --quiet --nox11 install success [Toolkit] [20221103-05:16:36] [INFO] install package Ascend-pyACL-20.2.rc1-linux.aarch64.run start [Toolkit] [20221103-05:16:37] [INFO] Ascend-pyACL-20.2.rc1-linux.aarch64.run --full --quiet --nox11 install success [Toolkit] [20221103-05:16:37] [INFO] install package Ascend-atc-1.77.t21.0.b210-linux.aarch64.run start WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv [Toolkit] [20221103-05:19:52] [INFO] Ascend-atc-1.77.t21.0.b210-linux.aarch64.run --full --pylocal --quiet --nox11 install success [Toolkit] [20221103-05:19:52] [INFO] install package Ascend-opp-1.77.t21.0.b210-linux.aarch64.run start [Toolkit] [20221103-05:20:37] [INFO] Ascend-opp-1.77.t21.0.b210-linux.aarch64.run --full --quiet --nox11 install success [Toolkit] [20221103-05:20:37] [INFO] install package Ascend-toolkit-1.77.t21.0.b210-linux.aarch64.run start WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv [Toolkit] [20221103-05:22:24] [INFO] Ascend-toolkit-1.77.t21.0.b210-linux.aarch64.run --full --pylocal --quiet --nox11 install success [Toolkit] [20221103-05:22:24] [INFO] install package Ascend-fwkacllib-1.77.t21.0.b210-linux.aarch64.run start ERROR: Exception: Traceback (most recent call last):   File "/usr/local/python3.7.5/lib/python3.7/shutil.py", line 566, in move     os.rename(src, real_dst) OSError: [Errno 18] Invalid cross-device link: '/tmp/pip-target-0nxv_a2l/lib/python/te' -> '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te'  During handling of the above exception, another exception occurred:  Traceback (most recent call last):   File "/usr/local/python3.7.5/lib/python3.7/site-packages/pip/_internal/cli/base_command.py", line 167, in exc_logging_wrapper     status = run_func(*args)   File "/usr/local/python3.7.5/lib/python3.7/site-packages/pip/_internal/cli/req_command.py", line 247, in wrapper     return func(self, options, args)   File "/usr/local/python3.7.5/lib/python3.7/site-packages/pip/_internal/commands/install.py", line 521, in run     options.target_dir, target_temp_dir, options.upgrade   File "/usr/local/python3.7.5/lib/python3.7/site-packages/pip/_internal/commands/install.py", line 579, in _handle_target_dir     shutil.move(os.path.join(lib_dir, item), target_item_dir)   File "/usr/local/python3.7.5/lib/python3.7/shutil.py", line 577, in move     symlinks=True)   File "/usr/local/python3.7.5/lib/python3.7/shutil.py", line 368, in copytree     raise Error(errors) shutil.Error: [('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/cube/cube_common.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/cube/cube_common.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/cube/conv2d.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/cube/conv2d.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/cube/__init__.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/cube/__init__.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_vector_api.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_vector_api.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_vector.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_vector.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_tensor.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_tensor.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_scalar_api.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_scalar_api.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_scalar.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_scalar.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_reduce_api.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_reduce_api.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_ir_builder.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_ir_builder.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_dprofile.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_dprofile.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_data_operation_api.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_data_operation_api.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_conf.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_conf.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_cmp_api.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_cmp_api.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/tik_build.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/tik_build.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/api/__init__.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/api/__init__.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/debugger.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/debugger.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/tik/__init__.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/tik/__init__.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/platform', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/platform', "[Errno 28] No space left on device: '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/platform'"), ('/tmp/pip-target-0nxv_a2l/lib/python/te/lang', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/lang', "[Errno 28] No space left on device: '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/lang'"), ('/tmp/pip-target-0nxv_a2l/lib/python/te/domain', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/domain', "[Errno 28] No space left on device: '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/domain'"), ('/tmp/pip-target-0nxv_a2l/lib/python/te/version.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/version.py', '[Errno 28] No space left on device'), ('/tmp/pip-target-0nxv_a2l/lib/python/te/__init__.py', '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/te/__init__.py', '[Errno 28] No space left on device')] chmod: cannot access '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/auto_tune.egg': No such file or directory chmod: cannot access '/usr/local/Ascend/ascend-toolkit/3.3.0.alpha001/arm64-linux/fwkacllib/python/site-packages/schedule_search.egg': No such file or directory [Toolkit] [20221103-05:24:49] [INFO] start uninstall Ascend-acllib-1.77.t21.0.b210-linux.aarch64.run acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device acllib/scripts/install.sh: line 179: echo: write error: No space left on device /tmp/selfgz1687424871/acllib/scripts/run_acllib_uninstall.sh: line 38: echo: write error: No space left on device /tmp/selfgz1687424871/acllib/scripts/run_acllib_uninstall.sh: line 38: echo: write error: No space left on device cp: error writing '/root/filelist_NjaG.csv': No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 923: echo: write error: No space left on device acllib/scripts/install.sh: line 187: echo: write error: No space left on device [Toolkit] [20221103-05:24:53] [INFO] start uninstall Ascend-pyACL-20.2.rc1-linux.aarch64.run ./script/install.sh: line 72: echo: write error: No space left on device [Toolkit] [20221103-05:24:53] [INFO] start uninstall Ascend-atc-1.77.t21.0.b210-linux.aarch64.run [Toolkit] [20221103-05:25:37] [INFO] start uninstall Ascend-opp-1.77.t21.0.b210-linux.aarch64.run [Toolkit] [20221103-05:25:51] [INFO] start uninstall Ascend-toolkit-1.77.t21.0.b210-linux.aarch64.run [Toolkit] [20221103-05:26:29] [INFO] start uninstall Ascend-fwkacllib-1.77.t21.0.b210-linux.aarch64.run [Toolkit] [20221103-05:27:11] [ERROR] packe_name=Ascend-fwkacllib-1.77.t21.0.b210-linux.aarch64.run install failed [Toolkit] [20221103-05:27:11] [ERROR] package install failed,Please refer to the log for more details: /var/log/ascend_seclog/ascend_toolkit_install.log 

想要在开发板上部署基于yolov5+deepsort的目标跟踪项目，苦于没有相关教程，想请教一下有没有相关的部署教程，能够实现项目的部署。

-+在root用户下安装环境依赖,并在root下安装cann,执行步骤./Ascend-cann-toolkit_20.1.rc1_linux-x86_64.run --install时报错root@davinci-mini:~/CANN# ./Ascend-cann-toolkit_3.3.0.alpha001_linux-aarch64.run --install --chip=Ascend310-minircVerifying archive integrity... 100% SHA256 checksums are OK. All good.Uncompressing ASCEND_RUN_PACKAGE 100%[Toolkit] [20221102-16:39:35] [INFO] LogFile:/var/log/ascend_seclog/ascend_toolkit_install.log[Toolkit] [20221102-16:39:35] [INFO] install start[Toolkit] [20221102-16:39:35] [INFO] The install path is /usr/local/Ascend ![Toolkit] [20221102-16:39:35] [INFO] install package Ascend-acllib-1.77.t21.0.b210-linux.aarch64.run start[Toolkit] [20221102-16:39:44] [INFO] Ascend-acllib-1.77.t21.0.b210-linux.aarch64.run --full --quiet --nox11 install success[Toolkit] [20221102-16:39:44] [INFO] install package Ascend-pyACL-20.2.rc1-linux.aarch64.run start[Toolkit] [20221102-16:39:45] [INFO] Ascend-pyACL-20.2.rc1-linux.aarch64.run --full --quiet --nox11 install success[Toolkit] [20221102-16:39:45] [INFO] install package Ascend-atc-1.77.t21.0.b210-linux.aarch64.run startWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venvWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venvWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venvWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv[Toolkit] [20221102-16:42:51] [INFO] Ascend-atc-1.77.t21.0.b210-linux.aarch64.run --full --pylocal --quiet --nox11 install success[Toolkit] [20221102-16:42:51] [INFO] install package Ascend-opp-1.77.t21.0.b210-linux.aarch64.run start[Toolkit] [20221102-16:43:47] [INFO] Ascend-opp-1.77.t21.0.b210-linux.aarch64.run --full --quiet --nox11 install success[Toolkit] [20221102-16:43:47] [INFO] install package Ascend-toolkit-1.77.t21.0.b210-linux.aarch64.run startWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venvWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv[Toolkit] [20221102-16:45:55] [INFO] start uninstall Ascend-acllib-1.77.t21.0.b210-linux.aarch64.run[Toolkit] [20221102-16:45:59] [INFO] start uninstall Ascend-pyACL-20.2.rc1-linux.aarch64.run[Toolkit] [20221102-16:45:59] [INFO] start uninstall Ascend-atc-1.77.t21.0.b210-linux.aarch64.run[Toolkit] [20221102-16:46:42] [INFO] start uninstall Ascend-opp-1.77.t21.0.b210-linux.aarch64.run[Toolkit] [20221102-16:46:56] [INFO] start uninstall Ascend-toolkit-1.77.t21.0.b210-linux.aarch64.run[Toolkit] [20221102-16:47:17] [ERROR] packe_name=Ascend-toolkit-1.77.t21.0.b210-linux.aarch64.run install failed[Toolkit] [20221102-16:47:17] [ERROR] package install failed,Please refer to the log for more details: /var/log/ascend_seclog/ascend_toolkit_install.log部分错误日志

以下两个驱动包的mntn_bbox.ko，一个有netlink初始化的打印，另一个没有。两个版本对netlink的处理是否有变化？Ascend310-driver-20.2.0-ubuntu18.04.aarch64-minirc.tar.gz[    8.584664] [bbox] ============ wait for link ready start ============== [    8.584681] [bbox] Success register netlink interface. [    8.584693] [bbox] Netlink channel init, num: 1. [    8.584715] [bbox] Succ to create netlink socket. [    8.585118] [bbox] Need create 1 channel.Ascend310-driver-21.0.4-ubuntu18.04.aarch64-minirc.tar.gz

没有这个gg我在运行 YOLOV4_coco_detection_picture时首先按照python依赖安装配置了环境，我的环境变量设置如下cann版本如下我在运行atc -help显示没有这个命令，请问是环境配置有问题嘛？我是在docker容器中运行的。

按照此处配置环境变量：https://gitee.com/ascend/samples/tree/master/python/environment然后根据https://gitee.com/ascend/samples/tree/v0.3.0/python/level2_simple_inference/2_object_detection/YOLOV3_coco_detection_picture执行预测src下的py文件，报错：Traceback (most recent call last):   File "object_detect.py", line 9, in <module>     from acllite_imageproc import AclLiteImageProc   File "/home/scxys/Ascend/thirdpart/aarch64/acllite/acllite_imageproc.py", line 10, in <module>     from acllite_image import AclLiteImage   File "/home/scxys/Ascend/thirdpart/aarch64/acllite/acllite_image.py", line 10, in <module>     class AclLiteImage(object):   File "/home/scxys/Ascend/thirdpart/aarch64/acllite/acllite_image.py", line 28, in AclLiteImage     _run_mode, _ = acl.rt.get_run_mode() AttributeError: module 'acl' has no attribute 'rt' 

我的电脑在虚拟机上运行Ubuntu十分卡顿,所以我采取了双系统的方式运行Ubuntu,但官方文档中"配置网络连接"这一步是在虚拟机上完成的,我在在双系统上操作时发现操作逻辑非常不同,请问怎么单纯在Ubuntu系统下完成这一步?

加载内核时，无法注册注册netlink，见dmesg打印。1.76.22.3.220(linux 4.19.95)可以注册。