• Versal AIE 上手尝鲜 2 -- Linux例程


    1. 介绍

    最近陆陆续续有工程师拿到了VCK190单板。 VCK190集成了Xilinx的7nm AIE,有很强的处理能力。 本文介绍怎么运行Xilinx AIE的例程,熟悉AIE开发流程。

    前一篇文章,Versal AIE 上手尝鲜 -- Standalone例程介绍了进行Standalone(BareMetal)程序开发的例子。

    这一篇文章,在Xilinx提供的Linux平台基础上,介绍怎么进行Linux程序开发,使用了Vitis_Accel_Examples中的aie_adder作为例子。

    2. 准备工作

    2.1. License

    在上手之前,需要注意是VCK190 Production单板,还是VCK190 ES单板。如果是VCK190 Production单板,使用VCK190 Voucher,在Xilinx网站,可以申请到License。安装License后,License的状态窗口下,能看到下列项目。

    AIEBuild
    AIESim
    MEBuild
    MESim
    

    如果是VCK190 ES单板,需要在Lounge里申请"Versal Tools Early Eacess"; "Versal Tools PDI Early Eacess"的License,并在Vivado里使能ES器件。在Vivado/2020.2/scripts/init.tcl的文件里,添加“enable_beta_device xcvc*”,可以自动使能ES器件。

    2.2. Platform

    在进行开发之前,需要准备Platform。 VCK190 Production单板和VCK190 ES单板使用的Platform不一样,可以从下面链接下载各自的Platform,再复制到目录“Xilinx/Vitis/2020.2/platforms/”下。
    VCK190 Production Platform
    VCK190 ES Platform

    准备好后,目录结构与下面类似。

    Versal Platform Direcoty Stucture

    Versal Platform Direcoty Stucture

    2.3. Common Images

    Xilinx现在还提供了Common Images,包含对应单板的Linux启动文件,和编译器、sysroots(头文件、应用程序库)等。可以在Xilinx Download下载Versal common image

    2.4. 测试环境

    Host OS: Ubuntu 18.04
    Vitis 2020.2
    PetaLinux 2020.2
    VCK190 Production

    3. aie_adder介绍

    AIE的aie_adder,相当于C语言的helloword例子,它创建了AIE Kernel、用于为AIE Kernel搬移数据的PL Kernel,并以仿真方式、或者硬件方式运行。

    3.1. 文件列表

    aie_adder有下列文件。 稍后也对主要文件,进行简要介绍。

    aie_adder:
    │  description.json
    │  details.rst
    │  Makefile
    │  qor.json
    │  README.rst
    │  system.cfg
    │  utils.mk
    │  xrt.ini
    │  
    ├─data
    │      golden.txt
    │      input0.txt
    │      input1.txt
    │      
    └─src
            aie_adder.cc
            aie_graph.cpp
            aie_graph.h
            aie_kernel.h
            host.cpp
            pl_mm2s.cpp
            pl_s2mm.cpp
    

    3.2. aie_adder.cc

    aie_adder.cc是定义AIE Kernel的文件,也是最重要的文件,仿真和实际运行都需要。

    AIE Kernel也很简单,相当于是C语言编程的HelloWorld, 只是读取2个向量,做加法运算后,再写出去。

    void aie_adder(input_stream_int32* in0, input_stream_int32* in1, output_stream_int32* out) {
        v4int32 a = readincr_v4(in0);
        v4int32 b = readincr_v4(in1);
        v4int32 c = operator+(a, b);
        writeincr_v4(out, c);
    }
    

    3.3. aie_graph.cpp

    aie_graph.cpp定义和控制运算的graph,这个例子中,只用于仿真。

    PLIO* in0 = new PLIO("DataIn0", adf::plio_32_bits, "data/input0.txt");
    PLIO* in1 = new PLIO("DataIn1", adf::plio_32_bits, "data/input1.txt");
    PLIO* out = new PLIO("DataOut", adf::plio_32_bits, "data/output.txt");
    
    // Hank: only for simulation??
    simulation::platform<2, 1> platform(in0, in1, out);
    
    simpleGraph addergraph;
    
    connect<> net0(platform.src[0], addergraph.in0);
    connect<> net1(platform.src[1], addergraph.in1);
    
    connect<> net2(addergraph.out, platform.sink[0]);
    
    # 2. ifdef __AIESIM__
    int main(int argc, char** argv) {
        addergraph.init();
        addergraph.run(4);
        addergraph.end();
        return 0;
    }
    
    # 3. endif
    
    

    3.4. aie_graph.h

    aie_graph.h定义了运算的graph,连接了stream数据流和AIE kernel,仿真和实际运行都需要。

    using namespace adf;
    
    class simpleGraph : public graph {
       private:
        kernel adder;
    
       public:
        port<input> in0, in1;
        port<output> out;
    
        simpleGraph() {
            adder = kernel::create(aie_adder);
    
            connect<stream>(in0, adder.in[0]);
            connect<stream>(in1, adder.in[1]);
            connect<stream>(adder.out[0], out);
    
            source(adder) = "aie_adder.cc";
    
            runtime<ratio>(adder) = 0.1;
        };
    };
    

    3.5. aie_kernel.h

    aie_kernel.h最简单,只声明了aie_adder的原型,仿真和实际运行都需要。

    void aie_adder(input_stream_int32* in0, input_stream_int32* in1, output_stream_int32* out);
    

    3.6. host.cpp

    host.cpp会申请内存,加载数据, 加载xclbin, 运行AIE Kernel。

    simpleGraph addergraph;
    
    static std::vector<char> load_xclbin(xrtDeviceHandle device, const std::string& fnm) {
    
        // load bit stream
        std::ifstream stream(fnm);
        stream.seekg(0, stream.end);
        size_t size = stream.tellg();
        stream.seekg(0, stream.beg);
    
        std::vector<char> header(size);
        stream.read(header.data(), size);
    
        auto top = reinterpret_cast<const axlf*>(header.data());
        xrtDeviceLoadXclbin(device, top);
    
        return header;
    }
    
    int main(int argc, char** argv) {
    
        // Open xclbin
        auto dhdl = xrtDeviceOpen(0); // Open Device the local device
        auto xclbin = load_xclbin(dhdl, "krnl_adder.xclbin");
        auto top = reinterpret_cast<const axlf*>(xclbin.data());
        adf::registerXRT(dhdl, top->m_header.uuid);
    
        int DataInput0[sizeIn], DataInput1[sizeIn];
        for (int i = 0; i < sizeIn; i++) {
            DataInput0[i] = rand() % 100;
            DataInput1[i] = rand() % 100;
        }
    
        // input memory
        // Allocating the input size of sizeIn to MM2S
        // This is using low-level XRT call xclAllocBO to allocate the memory
    
        xrtBufferHandle in_bohdl0 = xrtBOAlloc(dhdl, sizeIn * sizeof(int), 0, 0);
        auto in_bomapped0 = reinterpret_cast<uint32_t*>(xrtBOMap(in_bohdl0));
        memcpy(in_bomapped0, DataInput0, sizeIn * sizeof(int));
        printf("Input memory virtual addr 0x%px
    ", in_bomapped0);
    
        xrtBufferHandle in_bohdl1 = xrtBOAlloc(dhdl, sizeIn * sizeof(int), 0, 0);
        auto in_bomapped1 = reinterpret_cast<uint32_t*>(xrtBOMap(in_bohdl1));
        memcpy(in_bomapped1, DataInput1, sizeIn * sizeof(int));
        printf("Input memory virtual addr 0x%px
    ", in_bomapped1);
    
        // output memory
        // Allocating the output size of sizeOut to S2MM
        // This is using low-level XRT call xclAllocBO to allocate the memory
    
        xrtBufferHandle out_bohdl = xrtBOAlloc(dhdl, sizeOut * sizeof(int), 0, 0);
        auto out_bomapped = reinterpret_cast<uint32_t*>(xrtBOMap(out_bohdl));
        memset(out_bomapped, 0xABCDEF00, sizeOut * sizeof(int));
        printf("Output memory virtual addr 0x%px
    ", out_bomapped);
    
        // mm2s ip
        // Using the xrtPLKernelOpen function to manually control the PL Kernel
        // that is outside of the AI Engine graph
    
        xrtKernelHandle mm2s_khdl1 = xrtPLKernelOpen(dhdl, top->m_header.uuid, "pl_mm2s:{pl_mm2s_1}");
        // Need to provide the kernel handle, and the argument order of the kernel arguments
        // Here the in_bohdl is the input buffer, the nullptr is the streaming interface and must be null,
        // lastly, the size of the data. This info can be found in the kernel definition.
        xrtRunHandle mm2s_rhdl1 = xrtKernelRun(mm2s_khdl1, in_bohdl0, nullptr, sizeIn);
        printf("run pl_mm2s_1
    ");
    
        xrtKernelHandle mm2s_khdl2 = xrtPLKernelOpen(dhdl, top->m_header.uuid, "pl_mm2s:{pl_mm2s_2}");
        xrtRunHandle mm2s_rhdl2 = xrtKernelRun(mm2s_khdl2, in_bohdl1, nullptr, sizeIn);
        printf("run pl_mm2s_2
    ");
    
        // s2mm ip
        // Using the xrtPLKernelOpen function to manually control the PL Kernel
        // that is outside of the AI Engine graph
    
        xrtKernelHandle s2mm_khdl = xrtPLKernelOpen(dhdl, top->m_header.uuid, "pl_s2mm");
        // Need to provide the kernel handle, and the argument order of the kernel arguments
        // Here the out_bohdl is the output buffer, the nullptr is the streaming interface and must be null,
        // lastly, the size of the data. This info can be found in the kernel definition.
        xrtRunHandle s2mm_rhdl = xrtKernelRun(s2mm_khdl, out_bohdl, nullptr, sizeOut);
        printf("run pl_s2mm
    ");
    
        // graph execution for AIE
        printf("graph init. This does nothing because CDO in boot PDI already configures AIE.
    ");
        addergraph.init();
    
        printf("graph run
    ");
        addergraph.run(N_ITER);
    
        addergraph.end();
        printf("graph end
    ");
    
        // wait for mm2s done
        auto state = xrtRunWait(mm2s_rhdl1);
        std::cout << "mm2s_1 completed with status(" << state << ")
    ";
        xrtRunClose(mm2s_rhdl1);
        xrtKernelClose(mm2s_khdl1);
    
        state = xrtRunWait(mm2s_rhdl2);
        std::cout << "mm2s_2 completed with status(" << state << ")
    ";
        xrtRunClose(mm2s_rhdl2);
        xrtKernelClose(mm2s_khdl2);
    
        // wait for s2mm done
        state = xrtRunWait(s2mm_rhdl);
        std::cout << "s2mm completed with status(" << state << ")
    ";
        xrtRunClose(s2mm_rhdl);
        xrtKernelClose(s2mm_khdl);
    
        // Comparing the execution data to the golden data
    
        // clean up XRT
        std::cout << "Releasing remaining XRT objects...
    ";
        xrtBOFree(in_bohdl0);
        xrtBOFree(in_bohdl1);
        xrtBOFree(out_bohdl);
        xrtDeviceClose(dhdl);
    
        return errorCount;
    }
    
    

    3.7. pl_mm2s.cpp

    pl_mm2s.cpp是利用HLS做的PL设计,用于从内存搬移数据到AIE Kernel。

    void pl_mm2s(ap_int<32>* mem, hls::stream<qdma_axis<32, 0, 0, 0> >& s, int size) {
    data_mover:
        for (int i = 0; i < size; i++) {
            qdma_axis<32, 0, 0, 0> x;
            x.data = mem[i];
            x.keep_all();
            s.write(x);
        }
    }
    

    3.8. pl_s2mm.cpp

    pl_mm2s.cpp也是利用HLS做的PL设计,用于从AIE Kernel搬移数据到内存。

    void pl_s2mm(ap_int<32>* mem, hls::stream<qdma_axis<32, 0, 0, 0> >& s, int size) {
    data_mover:
        for (int i = 0; i < size; i++) {
            qdma_axis<32, 0, 0, 0> x = s.read();
            mem[i] = x.data;
        }
    }
    

    4. 经验

    aie_adder 基本可以顺利完成。 在实验过程中,可能遇到下列问题。

    4.1. DEVICE和EDGE_COMMON_SW

    aie_adder 的说明中,没有提到编译命令。 Makefile中提供了多个命令,基本模式如下:

    make all TARGET=<sw_emu/hw_emu/hw> DEVICE=<FPGA platform> HOST_ARCH=<aarch32/aarch64/x86> EDGE_COMMON_SW=<rootfs and kernel image path
    

    检查Makefile,发现下列语句。

    ifneq ($(findstring vck190, $(DEVICE)), vck190)
    $(warning [WARNING]: This example has not been tested for $(DEVICE). It may or may not work.)
    endif
    

    于是指定DEVICE为vck190。

    对于EDGE_COMMON_SW,在Xilinx下载网站找到了common image,包含rootfs 和 Linux kernel image。于是下载Versal common image,再在编译命令里指定解压后的目录“/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2”。

    总的编译命令如下。

    make sd_card TARGET=hw DEVICE=vck190 HOST_ARCH=aarch64 EDGE_COMMON_SW=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2
    

    编译后,报告找不到对应的platform(平台)。

    Running Dispatch Server on port:41287
    INFO: [v++ 60-1548] Creating build summary session with primary output /proj/hankf/vck190/Vitis_Accel_Examples-master-2021-0625/aie_kernels/aie_adder/pl_s2mm.xo.compile_summary, at Tue Jun 29 16:36:21 2021
    INFO: [v++ 60-1316] Initiating connection to rulecheck server, at Tue Jun 29 16:36:21 2021
    Running Rule Check Server on port:39067
    INFO: [v++ 60-1315] Creating rulecheck session with output '/proj/hankf/vck190/Vitis_Accel_Examples-master-2021-0625/aie_kernels/aie_adder/_x/reports/pl_s2mm/v++_compile_pl_s2mm_guidance.html', at Tue Jun 29 16:36:22 2021
    ERROR: [v++ 60-1258] No valid platform was found that matches 'vck190'. Please make sure that the platform is specified correctly, and the platform has the right version number. The platform repo paths are:
    	/opt/Xilinx/Vitis/2020.2/platforms
    The valid platforms found from the above repo paths are:
    	/opt/Xilinx/Vitis/2020.2/platforms/xilinx_vck190_base_202020_1/xilinx_vck190_base_202020_1.xpfm
    	/opt/Xilinx/Vitis/2020.2/platforms/xilinx_vck190_es1_base_202020_1/xilinx_vck190_es1_base_202020_1.xpfm
    

    根据提示,把device换成xilinx_vck190_es1_base_202110_1,使用下列命令编译,同样的问题消失。新的编译命令如下。

    make sd_card TARGET=hw DEVICE=xilinx_vck190_es1_base_202110_1 HOST_ARCH=aarch64 EDGE_COMMON_SW=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2
    

    4.2. ES Device

    如果没有使能ES Device,会得到错误“ERROR: [HLS 200-1023] Part 'xcvc1902-vsva2197-2MP-e-S-es1' is not installed.”。 需要在Vivado里使能ES器件。

    ===>The following messages were generated while  performing high-level synthesis for kernel: pl_s2mm Log file: /proj/hankf/vck190/Vitis_Accel_Examples-master-2021-0625/aie_kernels/aie_adder/_x/pl_s2mm/pl_s2mm/vitis_hls.log :
    ERROR: [v++ 200-1023] Part 'xcvc1902-vsva2197-2MP-e-S-es1' is not installed.
    ERROR: [v++ 60-300] Failed to build kernel(ip) pl_s2mm, see log for details: /proj/hankf/vck190/Vitis_Accel_Examples-master-2021-0625/aie_kernels/aie_adder/_x/pl_s2mm/pl_s2mm/vitis_hls.log
    ERROR: [v++ 60-773] In '/proj/hankf/vck190/Vitis_Accel_Examples-master-2021-0625/aie_kernels/aie_adder/_x/pl_s2mm/pl_s2mm/vitis_hls.log', caught Tcl error: ERROR: [HLS 200-1023] Part 'xcvc1902-vsva2197-2MP-e-S-es1' is not installed.
    ERROR: [v++ 60-599] Kernel compilation failed to complete
    ERROR: [v++ 60-592] Failed to finish compilation
    INFO: [v++ 60-1653] Closing dispatch client.
    Makefile:144: recipe for target 'pl_s2mm.xo' failed
    make: *** [pl_s2mm.xo] Error 1
    

    4.3. iostream

    编译时,得到错误“fatal error: iostream: No such file or directory”。

    INFO: [v++ 60-791] Total elapsed time: 0h 0m 57s
    INFO: [v++ 60-1653] Closing dispatch client.
    /opt/Xilinx/Vitis/2020.2/gnu/aarch64/lin/aarch64-linux/bin/aarch64-linux-gnu-g++ -Wall -c -std=c++14 -Wno-int-to-pointer-cast --sysroot=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/sysroots/aarch64-xilinx-linux  -I/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/sysroots/aarch64-xilinx-linux/usr/include/xrt -I/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/sysroots/aarch64-xilinx-linux/usr/include -I./ -I/opt/Xilinx/Vitis/2020.2/aietools/include -I/opt/Xilinx/Vitis/2020.2/include -o aie_control_xrt.o ./Work/ps/c_rts/aie_control_xrt.cpp
    ./Work/ps/c_rts/aie_control_xrt.cpp:1:10: fatal error: iostream: No such file or directory
        1 | #include <iostream>
          |          ^~~~~~~~~~
    compilation terminated.
    Makefile:170: recipe for target 'host' failed
    make: *** [host] Error 1
    

    交叉编译时,引用的头文件一般在sysroots里。
    Versal common image解压后,有文件sdk.sh。执行sdk.sh,能得到sysroots。
    于是在Versal的sysroots里查找iostream,果然有文件iostream。

    hankf@XSZGS4:/opt/Xilinx/peta/2020.2.sdk/sysroots$ ls -l -h
    total 8.0K
    drwxr-xr-x 17 hankf hankf 4.0K Jun 30 14:32 aarch64-xilinx-linux
    drwxr-xr-x  8 hankf hankf 4.0K Jun 30 14:33 x86_64-petalinux-linux
    
    hankf@XSZGS4:/opt/Xilinx/peta/2020.2.sdk/sysroots$ find -name iostream
    ./aarch64-xilinx-linux/usr/include/c++/9.2.0/iostream
    ./x86_64-petalinux-linux/usr/include/c++/9.2.0/iostream
    hankf@XSZGS4:/opt/Xilinx/peta/2020.2.sdk/sysroots$
    

    根据编译命令中的选项,“--sysroot=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/sysroots/aarch64-xilinx-linux”,想到要把sysroots放在目录/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2里。于是在目录/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2里为sysroots创建链接,从而使目录/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2里有了sysroots。

    hankf@XSZGS4:/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2$ ln -s /opt/Xilinx/peta/2020.2.sdk/sysroots/ ./sysroots
    hankf@XSZGS4:/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2$ ls -l -h
    total 3.1G
    -rw-r--r-- 1 hankf hankf 657K Nov 19  2020 bl31.elf
    -rw-r--r-- 1 hankf hankf 2.0K Nov 19  2020 boot.scr
    -rw-r--r-- 1 hankf hankf  17M Nov 19  2020 Image
    -rw-r--r-- 1 hankf hankf 1.6K Nov 19  2020 README.txt
    -rw-r--r-- 1 hankf hankf 2.3G Nov 19  2020 rootfs.ext4
    -rw-r--r-- 1 hankf hankf  44K Nov 19  2020 rootfs.manifest
    -rw-r--r-- 1 hankf hankf 221M Nov 19  2020 rootfs.tar.gz
    -rwxr-xr-x 1 hankf hankf 666M Nov 19  2020 sdk.sh
    lrwxrwxrwx 1 hankf hankf   37 Jun 30 14:45 sysroots -> /opt/Xilinx/peta/2020.2.sdk/sysroots/
    -rw-r--r-- 1 hankf hankf 946K Nov 19  2020 u-boot.elf
    
    hankf@XSZGS4:/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2$ ls -l ./sysroots/aarch64-xilinx-linux/
    total 60
    drwxr-xr-x  3 hankf hankf 4096 Jun 30 14:32 bin
    drwxr-xr-x  3 hankf hankf 4096 Jun 30 14:32 boot
    drwxr-xr-x  2 hankf hankf 4096 Jun 30 14:32 dev
    drwxr-xr-x 41 hankf hankf 4096 Jun 30 14:32 etc
    drwxr-xr-x  3 hankf hankf 4096 Jun 30 14:32 home
    drwxr-xr-x  8 hankf hankf 4096 Jun 30 14:32 lib
    drwxr-xr-x  2 hankf hankf 4096 Jun 30 14:32 media
    drwxr-xr-x  2 hankf hankf 4096 Jun 30 14:32 mnt
    dr-xr-xr-x  2 hankf hankf 4096 Jun 30 14:32 proc
    drwxr-xr-x  2 hankf hankf 4096 Jun 30 14:32 run
    drwxr-xr-x  3 hankf hankf 4096 Jun 30 14:32 sbin
    dr-xr-xr-x  2 hankf hankf 4096 Jun 30 14:32 sys
    drwxrwxr-x  2 hankf hankf 4096 Jun 30 14:32 tmp
    drwxr-xr-x 10 hankf hankf 4096 Jun 30 14:32 usr
    drwxr-xr-x  9 hankf hankf 4096 Jun 30 14:32 var
    

    4.4. Source file does not exist: adder.xclbin

    aie_adder 的Makefile中提供了多个命令。考虑到VCK190使用SD(TF)卡启动,于是使用了目标为sd_card的下列命令编译。但是编译后,遇到了错误“Source file does not exist: adder.xclbin”。

    /opt/Xilinx/Vitis/2020.2/gnu/aarch64/lin/aarch64-linux/bin/aarch64-linux-gnu-g++ *.o -lxaiengine -ladf_api_xrt -lxrt_core -lxrt_coreutil -L/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/sysroots/aarch64-xilinx-linux/usr/lib --sysroot=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/sysroots/aarch64-xilinx-linux -L/opt/Xilinx/Vitis/2020.2/aietools/lib/aarch64.o -o ./aie_adder
    COMPLETE: Host application created.
    rm -rf run_app.sh
    v++ -p -t hw 
    	--platform xilinx_vck190_es1_base_202020_1 
    	--package.out_dir ./package.hw 
    	--package.rootfs /opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/rootfs.ext4 
    	--package.image_format=ext4 
    	--package.boot_mode=sd 
    	--package.kernel_image=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2/Image 
    	--package.defer_aie_run 
    	--package.sd_file ./run_app.sh 
    	--package.sd_file aie_adder adder.xclbin libadf.a -o krnl_adder.xclbin
    Option Map File Used: '/opt/Xilinx/Vitis/2020.2/data/vitis/vpp/optMap.xml'
    
    ****** v++ v2020.2 (64-bit)
      **** SW Build (by xbuild) on 2020-11-18-05:13:29
        ** Copyright 1986-2020 Xilinx, Inc. All Rights Reserved.
    
    ERROR: [v++ 60-602] Source file does not exist: /proj/hankf/vck190/Vitis_Accel_Examples-master-2021-0625/aie_kernels/aie_adder/adder.xclbin
    INFO: [v++ 60-1662] Stopping dispatch session having empty uuid.
    INFO: [v++ 60-1653] Closing dispatch client.
    Makefile:192: recipe for target 'sd_card' failed
    make: *** [sd_card] Error 1
    

    后来尝试命令“all”,能编译成功。

    make all TARGET=hw DEVICE=vck190 HOST_ARCH=aarch64 EDGE_COMMON_SW=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2
    

    4.5. SD Card cannot boot

    使用编译好的文件,复制到TF卡,启动vck190单板,发现单板不能启动。

    检查发现,手上的单板是production芯片,换用xilinx_vck190_base_202020_1,编译出来的文件能够启动。

    最后成功编译,而且产生的映像能在vck190 production单板正常启动的编译命令如下:

    make all TARGET=hw DEVICE=xilinx_vck190_base_202020_1 HOST_ARCH=aarch64 EDGE_COMMON_SW=/opt/Xilinx/download/2020/xilinx-versal-common-v2020.2
    
  • 相关阅读:
    (转)Mat, vector<point2f>,Iplimage等等常见类型转换
    人脸检测与识别的趋势和分析
    经典网络LeNet5看卷积神经网络各层的维度变化
    C++: int int& int * int**的区别、联系和用途
    c++从txt中读取数据,数据并不是一行路径(实用)
    Spyder常用快捷键
    批量读写变换图片(转)
    OpenCV属性页配置问题~
    视觉目标检测和识别之过去,现在及可能(2017.06.28)
    数据库---SQL Server
  • 原文地址:https://www.cnblogs.com/hankfu/p/15108882.html
Copyright © 2020-2023  润新知