Intel核显--OpenCL环境--Linux
Machine :Intel Xeon i7 6700(core) + Intel HD Graphics 530(skyLake) +Linux (CentOS 7.2或7.3，我的7.3)+eclipse(Neon)安装OpenCL环境，希望安装成功后将Intel i7 6700作为host，HD Graphics 530作为device。
搞了半天，才知道Linux下 HD Graphics 530并没有driver，https://downloadcenter.intel.com/product/88345/Intel-HD-Graphics-530 官网给出的都是windows下的。所以driver这步就省了。
按照：https://software.intel.com/en-us/articles/sdk-for-opencl-gsg 来安装。
一、安装Intel OpenCL driver
官网给出了脚本直接安装：https://software.intel.com/sites/default/files/managed/f6/77/install_OCL_driver.sh_.txt 我之前都是自己新建一个txt将这段脚本复制过去，结果安装时总是出错，不能复制！要直接下载。我放在 http://download.csdn.net/download/wd1603926823/10219725 可以直接下载。下载完毕以后，直接按照官网的步骤开始安装：
$ mv install_OCL_driver.sh_.txt install_OCL_driver.sh
$ chmod 755 install_OCL_driver.sh
$ sudo su
$ ./install_OCL_driver.sh install一直选择y 即yes，然后出现：
然后出现一个error：因为我的linux kernel是3.10.0-514.el7.X86_64 所以报错。不满足install的条件。
这个提示出来，我确认我有足够的磁盘空间，于是我又运行install一次，这次就不会报错了。漫长的等待----安装完毕。
然后我按照建议1 将用户root加入到video group了。 第2点不满足就不管它了，然后重启。
也可以自己打开提示的这个网页去手动下载linux-4.7.tar.xz 然后看完 https://linoxide.com/linux-how-to/upgrade-linux-kernel-stable-4-7-centos-7-x/ 的后部分 就会安装了。
二、Intel_OpenCL_SDK
https://software.intel.com/en-us/articles/sdk-for-opencl-gsg 这里给出的是Ubuntu的，没有给CentOS的。但我还是想试试，于是下载这个(不要复制，而是直接下载)，然后按照运行：
$ mv install_SDK_prereq_ubuntu.sh_.txt install_SDK_prereq_ubuntu.sh
$ sudo su
$ ./install_SDK_prereq_ubuntu.sh
结果显示：bash: ./install_SDK_prereq_ubuntu.sh: 权限不够 ！！！但我是root登录的！我输入clinfo也说我未找到命令！？
后来我想了想，还是自己手动安装SDK，去官网 https://registrationcenter.intel.com/en/products/postregistration/?sn=c69g-kd7n5h3j&EmailID=1603926823@qq.com&Sequence=2151318&dnld=t 下的合适我的SDK，当然怎么找合适的，要先看 release note ：https://software.intel.com/en-us/articles/opencl-code-builder-release-notes 看哪个版本适合我。 可以通过install.sh或者install_GUI.sh安装，我是通过sh install_GUI.sh安装的，在安装期间，说我缺少dkms这个包，于是我yum install epel-release dkms 安装，但装了两遍才装成功dkms，然后继续中断的安装步骤。安装完毕会显示成功，出来这个界面。
所以SDK就装完了。
装完后可以看到/opt/instel/下有4个文件(夹)：ism opencl opencl-sdk和另一个文件。打开文件夹opencl-sdk可以看到有include和lib64文件夹。。。应该是真的成功了。
三、检查对OpenCL的设备支持
但输入clinfo说权限不够，于是自己去手动下载了clinfo的rpm包，然后：rpm -ivh clinfo-2.1.17.02.09-1.el7.x86_64.rpm 安装成功。然后输入clinfo 即可：哈哈哈终于出现2个平台和2个devices了哈哈哈！！！！！
平台1：Intel(R) OpenCL （设备是：Intel(R) HD Graphics 与 Intel(R) Core(TM) i7-6700 CPU）；
平台2：Experimental OpenCL 2.1 CPU Only Platform （设备是：Intel(R) Core(TM) i7-6700 CPU）;
Number of platforms 2Platform Name Intel(R) OpenCLPlatform Vendor Intel(R) CorporationPlatform Version OpenCL 2.0 Platform Profile FULL_PROFILEPlatform Extensions cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_image2d_from_buffer cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_spirPlatform Extensions function suffix INTELPlatform Name Experimental OpenCL 2.1 CPU Only PlatformPlatform Vendor Intel(R) CorporationPlatform Version OpenCL 2.1 LINUXPlatform Profile FULL_PROFILEPlatform Extensions cl_khr_icd cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_intel_exec_by_local_thread cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer Platform Host timer resolution 1nsPlatform Extensions function suffix INTELPlatform Name Intel(R) OpenCL
Number of devices 2Device Name Intel(R) HD GraphicsDevice Vendor Intel(R) CorporationDevice Vendor ID 0x8086Device Version OpenCL 2.0 Driver Version r5.0.63503Device OpenCL C Version OpenCL C 2.0 Device Type GPUDevice Available YesDevice Profile FULL_PROFILEMax compute units 24Max clock frequency 1150MHzDevice Partition (core)Max number of sub-devices 0Supported partition types NoneMax work item dimensions 3Max work item sizes 256x256x256Max work group size 256Compiler Available YesLinker Available YesPreferred work group size multiple 32Sub-group sizes (Intel) 8x16x32Preferred / native vector sizes char 16 / 16 short 8 / 8 int 4 / 4 long 1 / 1 half 8 / 8 (cl_khr_fp16)float 1 / 1 double 1 / 1 (cl_khr_fp64)Half-precision Floating-point support (cl_khr_fp16)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero YesRound to infinity YesIEEE754-2008 fused multiply-add YesSupport is emulated in software NoCorrectly-rounded divide and sqrt operations NoSingle-precision Floating-point support (core)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero YesRound to infinity YesIEEE754-2008 fused multiply-add YesSupport is emulated in software NoCorrectly-rounded divide and sqrt operations YesDouble-precision Floating-point support (cl_khr_fp64)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero YesRound to infinity YesIEEE754-2008 fused multiply-add YesSupport is emulated in software NoCorrectly-rounded divide and sqrt operations NoAddress bits 64, Little-EndianGlobal memory size 13238812672 (12.33GiB)Error Correction support NoMax memory allocation 4294959103 (4GiB)Unified memory for Host and Device YesShared Virtual Memory (SVM) capabilities (core)Coarse-grained buffer sharing YesFine-grained buffer sharing NoFine-grained system sharing NoAtomics NoMinimum alignment for any data type 128 bytesAlignment of base address 1024 bits (128 bytes)Preferred alignment for atomics SVM 64 bytesGlobal 64 bytesLocal 64 bytesMax size for global variable 65536 (64KiB)Preferred total size of global vars 4294959103 (4GiB)Global Memory cache type Read/WriteGlobal Memory cache size 524288 (512KiB)Global Memory cache line 64 bytesImage support YesMax number of samplers per kernel 16Max size for 1D images from buffer 268434943 pixelsMax 1D or 2D image array size 2048 imagesBase address alignment for 2D image buffers 4 bytesPitch alignment for 2D image buffers 4 bytesMax 2D image size 16384x16384 pixelsMax planar YUV image size 16384x16380 pixelsMax 3D image size 16384x16384x2048 pixelsMax number of read image args 128Max number of write image args 128Max number of read/write image args 128Max number of pipe args 16Max active pipe reservations 1Max pipe packet size 1024Local memory type LocalLocal memory size 65536 (64KiB)Max constant buffer size 4294959103 (4GiB)Max number of constant args 8Max size of kernel argument 1024Queue properties (on host) Out-of-order execution YesProfiling YesQueue properties (on device) Out-of-order execution YesProfiling YesPreferred size 131072 (128KiB)Max size 67108864 (64MiB)Max queues on device 1Max events on device 1024Prefer user sync for interop YesProfiling timer resolution 83nsExecution capabilities Run OpenCL kernels YesRun native kernels NoSPIR versions 1.2 printf() buffer size 4194304 (4MiB)Built-in kernels block_motion_estimate_intel;block_advanced_motion_estimate_check_intel;block_advanced_motion_estimate_bidirectional_check_intelMotion Estimation accelerator version (Intel) 2Device-side AVC Motion Estimation version 1Supports texture sampler use YesSupports preemption NoDevice Extensions cl_intel_accelerator cl_intel_advanced_motion_estimation cl_intel_device_side_avc_motion_estimation cl_intel_driver_diagnostics cl_intel_media_block_io cl_intel_motion_estimation cl_intel_planar_yuv cl_intel_packed_yuv cl_intel_required_subgroup_size cl_intel_subgroups cl_intel_subgroups_short cl_intel_va_api_media_sharing cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_fp16 cl_khr_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_icd cl_khr_image2d_from_buffer cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_khr_spir cl_khr_subgroups Device Name Intel(R) Core(TM) i7-6700 CPU @ 3.40GHzDevice Vendor Intel(R) CorporationDevice Vendor ID 0x8086Device Version OpenCL 2.0 (Build 475)Driver Version 1.2.0.475Device OpenCL C Version OpenCL C 2.0 Device Type CPUDevice Available YesDevice Profile FULL_PROFILEMax compute units 8Max clock frequency 3400MHzDevice Partition (core)Max number of sub-devices 8Supported partition types by counts, equally, by names (Intel)Max work item dimensions 3Max work item sizes 8192x8192x8192Max work group size 8192Compiler Available YesLinker Available YesPreferred work group size multiple 128Preferred / native vector sizes char 1 / 32 short 1 / 16 int 1 / 8 long 1 / 4 half 0 / 0 (n/a)float 1 / 8 double 1 / 4 (cl_khr_fp64)Half-precision Floating-point support (n/a)Single-precision Floating-point support (core)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero NoRound to infinity NoIEEE754-2008 fused multiply-add NoSupport is emulated in software NoCorrectly-rounded divide and sqrt operations NoDouble-precision Floating-point support (cl_khr_fp64)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero YesRound to infinity YesIEEE754-2008 fused multiply-add YesSupport is emulated in software NoCorrectly-rounded divide and sqrt operations NoAddress bits 64, Little-EndianGlobal memory size 16559001600 (15.42GiB)Error Correction support NoMax memory allocation 4139750400 (3.855GiB)Unified memory for Host and Device YesShared Virtual Memory (SVM) capabilities (core)Coarse-grained buffer sharing YesFine-grained buffer sharing NoFine-grained system sharing NoAtomics NoMinimum alignment for any data type 128 bytesAlignment of base address 1024 bits (128 bytes)Preferred alignment for atomics SVM 64 bytesGlobal 64 bytesLocal 0 bytesMax size for global variable 65536 (64KiB)Preferred total size of global vars 65536 (64KiB)Global Memory cache type Read/WriteGlobal Memory cache size 262144 (256KiB)Global Memory cache line 64 bytesImage support YesMax number of samplers per kernel 480Max size for 1D images from buffer 258734400 pixelsMax 1D or 2D image array size 2048 imagesBase address alignment for 2D image buffers 64 bytesPitch alignment for 2D image buffers 64 bytesMax 2D image size 16384x16384 pixelsMax 3D image size 2048x2048x2048 pixelsMax number of read image args 480Max number of write image args 480Max number of read/write image args 480Max number of pipe args 16Max active pipe reservations 32767Max pipe packet size 1024Local memory type GlobalLocal memory size 32768 (32KiB)Max constant buffer size 131072 (128KiB)Max number of constant args 480Max size of kernel argument 3840 (3.75KiB)Queue properties (on host) Out-of-order execution YesProfiling YesLocal thread execution (Intel) YesQueue properties (on device) Out-of-order execution YesProfiling YesPreferred size 4294967295 (4GiB)Max size 4294967295 (4GiB)Max queues on device 4294967295Max events on device 4294967295Prefer user sync for interop NoProfiling timer resolution 1nsExecution capabilities Run OpenCL kernels YesRun native kernels YesSPIR versions 1.2printf() buffer size 1048576 (1024KiB)Built-in kernels Device Extensions cl_khr_icd cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_intel_exec_by_local_thread cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer Platform Name Experimental OpenCL 2.1 CPU Only Platform
Number of devices 1Device Name Intel(R) Core(TM) i7-6700 CPU @ 3.40GHzDevice Vendor Intel(R) CorporationDevice Vendor ID 0x8086Device Version OpenCL 2.1 (Build 10)Driver Version 1.2.0.10Device OpenCL C Version OpenCL C 2.0 Device Type CPUDevice Available YesDevice Profile FULL_PROFILEMax compute units 8Max clock frequency 3400MHzDevice Partition (core)Max number of sub-devices 8Supported partition types by counts, equally, by names (Intel)Max work item dimensions 3Max work item sizes 8192x8192x8192Max work group size 8192Compiler Available YesLinker Available YesPreferred work group size multiple 128Max sub-groups per work group 1Preferred / native vector sizes char 1 / 32 short 1 / 16 int 1 / 8 long 1 / 4 half 0 / 0 (n/a)float 1 / 8 double 1 / 4 (cl_khr_fp64)Half-precision Floating-point support (n/a)Single-precision Floating-point support (core)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero NoRound to infinity NoIEEE754-2008 fused multiply-add NoSupport is emulated in software NoCorrectly-rounded divide and sqrt operations NoDouble-precision Floating-point support (cl_khr_fp64)Denormals YesInfinity and NANs YesRound to nearest YesRound to zero YesRound to infinity YesIEEE754-2008 fused multiply-add YesSupport is emulated in software NoCorrectly-rounded divide and sqrt operations NoAddress bits 64, Little-EndianGlobal memory size 16559001600 (15.42GiB)Error Correction support NoMax memory allocation 4139750400 (3.855GiB)Unified memory for Host and Device YesShared Virtual Memory (SVM) capabilities (core)Coarse-grained buffer sharing YesFine-grained buffer sharing YesFine-grained system sharing YesAtomics YesMinimum alignment for any data type 128 bytesAlignment of base address 1024 bits (128 bytes)Preferred alignment for atomics SVM 64 bytesGlobal 64 bytesLocal 0 bytesMax size for global variable 65536 (64KiB)Preferred total size of global vars 65536 (64KiB)Global Memory cache type Read/WriteGlobal Memory cache size 262144 (256KiB)Global Memory cache line 64 bytesImage support YesMax number of samplers per kernel 480Max size for 1D images from buffer 258734400 pixelsMax 1D or 2D image array size 2048 imagesBase address alignment for 2D image buffers 64 bytesPitch alignment for 2D image buffers 64 bytesMax 2D image size 16384x16384 pixelsMax 3D image size 2048x2048x2048 pixelsMax number of read image args 480Max number of write image args 480Max number of read/write image args 480Max number of pipe args 16Max active pipe reservations 32767Max pipe packet size 1024Local memory type GlobalLocal memory size 32768 (32KiB)Max constant buffer size 131072 (128KiB)Max number of constant args 480Max size of kernel argument 3840 (3.75KiB)Queue properties (on host) Out-of-order execution YesProfiling YesLocal thread execution (Intel) YesQueue properties (on device) Out-of-order execution YesProfiling YesPreferred size 4294967295 (4GiB)Max size 4294967295 (4GiB)Max queues on device 4294967295Max events on device 4294967295Prefer user sync for interop NoProfiling timer resolution 1nsExecution capabilities Run OpenCL kernels YesRun native kernels YesSub-group independent forward progress NoIL version SPIR-V_1.0SPIR versions 1.2printf() buffer size 1048576 (1024KiB)Built-in kernels Device Extensions cl_khr_icd cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_depth_images cl_khr_3d_image_writes cl_intel_exec_by_local_thread cl_khr_spir cl_khr_fp64 cl_khr_image2d_from_buffer NULL platform behaviorclGetPlatformInfo(NULL, CL_PLATFORM_NAME, ...) No platformclGetDeviceIDs(NULL, CL_DEVICE_TYPE_ALL, ...) No platformclCreateContext(NULL, ...) [default] No platformclCreateContext(NULL, ...) [other] Success [INTEL]clCreateContextFromType(NULL, CL_DEVICE_TYPE_CPU) No platformclCreateContextFromType(NULL, CL_DEVICE_TYPE_GPU) No platformclCreateContextFromType(NULL, CL_DEVICE_TYPE_ACCELERATOR) No platformclCreateContextFromType(NULL, CL_DEVICE_TYPE_CUSTOM) No platformclCreateContextFromType(NULL, CL_DEVICE_TYPE_ALL) No platform太开心了！！！！Linux下Intel核显for OpenCL环境搭建完毕！！！
四、用eclipse创建一个OpenCL项目测试
在相应地方加上/opt/intel/opencl-sdk/include和.../lib64即可！！！
五、Intel opencl samples
可以看看 https://software.intel.com/en-us/intel-opencl-support/code-samples 以及https://software.intel.com/en-us/articles/getting-the-most-from-opencl-12-how-to-increase-performance-by-minimizing-buffer-copies-on-intel-processor-graphics 因为很怕本来0传输的被自己写成内存到内存的传输 那么用核显就没有意义了。。。
https://software.intel.com/en-us/articles/getting-the-most-from-opencl-12-how-to-increase-performance-by-minimizing-buffer-copies-on-intel-processor-graphics
https://arrayfire.com/zero-copy-on-integrated-gpus/
https://software.intel.com/en-us/forums/opencl/topic/708049
http://blog.csdn.net/wcj0626/article/details/46360605
http://www.cnblogs.com/lifan3a/articles/4613858.html
六、Intel的调试及性能查看tool for linux
我在官网看了那个Amplifier XE要收费，后来查到有Intel VTune Performance Analyzer，我在 http://download.csdn.net/download/morre/8557459 下载的这个工具，这里也有下载链接：http://intel.brothersoft.com/intel-vtune-performance-analyzer-for-linux.html?fromsearch 。另外：http://qcd.phys.cmu.edu/QCDcluster/intel/VTLRELEASENOTES.htm 这里还有各个版本的不同比较。http://blog.csdn.net/yangsc1984/article/details/3908823 这里是别人的中文介绍。https://www.viva64.com/en/t/0091/ 这里是别人的英文介绍。 http://blog.csdn.net/honey_yyang/article/details/7848923 这个是与eclipse结合后的强大功能介绍。 https://software.intel.com/en-us/code-builder-user-manual-installing-opencl-kernel-debugger-for-linux 官网说的调试工具就在sdk里？ http://opencl38.rssing.com/chan-13188063/all_p8.html 这个是论坛上别人调试时遇到的问题。 解压准备安装，但提示说需要序列号？！！！！我没有序列号 ，然后Intel官网现在已经没有这个Intel VTune Performance Analyzer 了？！！！！！现在官网上linux的工具只有：https://software.intel.com/en-us/intel-devtools-by-os/linux 只有这些。没有序列号，无法安装免费的opencl工具，所以只能装这个 https://software.intel.com/en-us/intel-vtune-amplifier-xe 使用30的，按照https://software.intel.com/en-us/vtune-amplifier-install-guide-linux-user-interface-install 来安装。
七、一个例子与以前AMD时的对比
以前用AMD 560显卡时：CPU版本耗时约19s，OpenCL版本耗时约4.5s (单线程多线程均如此)；
现在用Intel skylake核显：CPU版本耗时约17.3s，单线程的OpenCL版本耗时约5s，但多线程的OpenCL版本竟然要38s？！（将OpenCL版本由单线程改成多线程时应该有问题）
Intel核显--OpenCL环境--Linux相关推荐
- VPU?APU?GPU?折腾从AMD-GPU回到Intel核显+OpenCL+Linux环境搭建
大神说: 也就是说APU比较适合计算部分少,但传输耗时多的情况:而GPU相反. 请注意APU的一些细节, 如何才能做到无传输. 因为APU上稍微不注意, 按照传统写法很容易造成一次内存内部的传输的. ...
- 文明6游戏 linux,《文明6》正式支持Linux/SteamOS 但不支持Intel核显和A卡
2K旗下策略游戏"文明"系列最新作<文明6>正式登陆Linux/SteamOS平台,在<文明6>的Steam页面上显示其已经支持Linux和SteamOS版 ...
- Ubuntu20.04 ffmpeg添加 Intel核显QSV加速支持
背景介绍 ffmpeg 是一款非常常用的视频编码解码工具,而且 ffmpeg是支持 NV 显卡硬件加速以及 Intel 核显的 QSV 加速.在 Window 下,使用 ffmpeg 的 qsv 加速 ...
- OpenCL入门一：Intel核心显卡OpenCL环境搭建
本文在Windows 10 64位系统上搭建OpenCL开发环境,笔记本显卡的类型(使用英特尔® 驱动程序和支持助理查询): NVIDIA GeForce 820M Intel® HD Graphic ...
- 桌面窗口管理器dwm.exe内存过高，intel核显内存泄漏问题（附核显升级链接）
系统 win10笔记本 intel核显630 问题概述 电脑长时间运行后突然变得很卡,打开任务管理器后发现[桌面窗口管理器]占内存过高,甚至能吃10个G,如图: 分析 查询各种论坛后基本确定问题为内存 ...
- UnRaid利用iGVT-g插件实现Nas宿主机、虚拟机同时使用intel核显输出或硬解（硬件加速）功能
文章目录 0.前言 1.准备工作 1.1.1台用intel处理器的组建的Nas主机(UnRaid系统必须是6.9.2及以上) 1.2.1台或多台显示器 1.3.1个DisplayLink转接器(USB ...
- 第一代intel核显id：0046的10.9驱动安装详解（转）
一代0046 intel核显hd1000m 10.8的驱动已经失效了,开不了QE/CI的 从tonymac找来的驱动,并完善一下 直接上驱动啦 安装步骤务必按照顺序进行,不然是驱动不起来的 第一步:首 ...
- 台式机ubuntu，使用intel核显作显示输出，nvidia独显做cuda运算
问题来源: 将显示线插到独显上安装完ubuntu之后装完nvidia显卡驱动,使用nvidia-smi命令会发现Xorg和compiz两个服务占用了一部分显存,随着ubuntu不关机使用越来越久,甚至 ...
- intel 核显linux驱动程序,10nm来也！Intel 11代核显Linux驱动功能冻结
Intel今年初宣布了一系列10nm工艺产品,其中面向轻薄笔记本的Ice Lake U系列将在年底购物季期间上市,而在日前发布一季度财报后,Intel又透露今年第二季度会完成对10nm Ice Lak ...
最新文章
- cpu多核 node 单线程_详解node单线程实现高并发原理与node异步I/O
- WebService在开发中的实际问题
- Win32编程day04 学习笔记
- chrome自动调节窗口大小插件_高效使用Chrome浏览器的10个技巧
- 指令和数据都用二进制代码存放在内存中,从时空观角度回答CPU如何区分读出的代码是指令还是数据
- QT 开发openSSL CSR证书请求工具
- Inter Thread Latency
- ESXi 中重新启动管理代理
- python 打印皮卡丘_Python干掉了97%的办公软件？
- virtualbox centos并配置网络使宿主机和虚拟机能够互相访问 host-only
- 【UI】NGUI和UGUI
- QT 字符乱码的原因
- gmail邮箱服务器被禁,Gmail邮箱失联：谷歌回应服务器没问题
- python办公室自动化之office颜值担当-PPT
- vi设计管理手册的体系
- 华为手机各代系拆机图
- 爬虫路线Requests-Re-BeautifulSoup技术路线总结
- 提升大数据数据分析性能的方法及技术（二）
- 婚宴座位图html5,婚礼酒席座位该怎么安排好？婚宴座位安排图解
- R语言绘制gif动图