INTRODUCTION REQUIREMENTS CREATE A NEW PROJECT GPU EXAMPLE GSTREAMER EXAMPLE 1. INTRODUCTION:      The below steps show how to create different application examples using Elipse IDE. 2. REQUIREMENTS:      A fully working image and meta-toolchain generated in Yocto . You can follow the  next training: Yocto Training - HOME      Install and configure the Yocto Eclipse Plug-in. For more details about this requirement please refer to Setting up the Eclipse IDE for Yocto Application Development         To demonstrate the steps, L3.14.28  BSP, fsl-image-qt5 image and i.MX6Q SABRE-SDP board were used. 3. CREATE A NEW PROJECT      Follow the section Creating a Hello World Project of this document Setting up the Eclipse IDE for Yocto Application Development 4. GPU EXAMPLE           For this project we use the source code found in the fsl-gpu-sdk that can be downloaded from:      https://www.freescale.com/webapp/Download?colCode=IMX6_GPU_SDK&location=null&Parent_nodeId=1337637154535695831062&Parent…      Follow section 3 and create a new project named gputest.      From the IMX6_GPU_SDK choose one of the examples of GLES2.0 folder. In this case the 01_SimpleTriangle is chosen.      Copy the .c and .h files to the src directory of the gputest project. The Project Explorer window should look like this:              Add the needed files and libraries to compile and link in the Makefile.am file found in the ´src´ folder. The Makefile.am file should have the below content:          bin_PROGRAMS = gputest          gputest_SOURCES = gputest.c fsl_egl.c fslutil.c          AM_CFLAGS = @gputest_CFLAGS@          AM_LDFLAGS = @gputest_LIBS@ -lstdc++ -lm -lGLESv2 -lEGL -lX11 -ldl          CLEANFILES = *~ ​    Add the PATH to CFLAGS where the compiler will look for the headers at Project->Properties->Autotools->configure:           In this project there is no need to add extra PATHs for the headers. Apply the changes by clicking on Reconfigure Project. Build the project To test the file you can send the executable to the board with:           $ scp gputest root@<board_ip>:/home/root      $./gputest      You should get the next output in the display: 5. GSTREAMER EXAMPLE      For this project we use the source code found at Basic tutorial 1: Hello world! - GStreamer SDK documentation - GStreamer SDK documentation    Follow section 3 and create a new project named Gstreamer.    Copy the code of the basic tutorial to your Gstreamer.c file.    Add the needed files and libraries to compile and link in the Makefile.am file found in the ´src´ folder. The Makefile.am file should have the below content:                           bin_PROGRAMS = Gstreamer      Gstreamer_SOURCES = Gstreamer.c      AM_CFLAGS = @Gstreamer_CFLAGS@      AM_LDFLAGS = @Gstreamer_LIBS@ -lstdc++  -lVDK -lm -lGLESv2 -lGAL -lEGL  -ldl -lgstreamer-0.10 -lgobject-2.0 -lgmodule-2.0 -lgthread-2.0 -lrt -lxml2 -lglib-2.0      CLEANFILES = *~         ​    Add the PATH to CFLAGS where the compiler will look for the headers at Project->Properties->Autotools->configure:           For this example the next lines are added             -I${Sysroot}/usr/include/gstreamer-1.0        -I${Sysroot}/usr/include/glib-2.0        -I${Sysroot}/usr/include/libxml2        -I${Sysroot}/usr/lib/glib-2.0/include      Apply the changes by clicking on Reconfigure Project. Build the project To test the file you can send the executable to the board with:           $ scp Gstreamer root@<board_ip>:/home/root To execute the application on the board:      $./Gstreamer The board should have internet access and the application should play the video found at http://docs.gstreamer.com/media/sintel_trailer-480p.webm

i.MX6UL CSI (CMOS Sensor Interface) host port support BT.656(CCIR656) Interlace mode by hardware.  It can received  PAL(720pix X 576pix) and NTSC(720pix X 480 pix) format data from camera. This document introduce how to add this feature to Linux L3.14.38-ga and receive ADV7180 output. Software: yocto L3.14.38-ga. Hareware:  i.mx6ul-evk + ADV7180. ADV7180: PAL 720pix X576pix  ,  YUV4:2:2(UYVY) interlace output. LCD Display screen resolution:  800 X 480,  RGB565. 1) Note : For BT.656 mode,   parallel data port is  CSI_DATA[9:2], not CSI_DATA[13:6].    So,  "pinctrl_csi1" is the same as "ov5640" in "imx6ul-14x14-evk.dts": pinctrl_csi1: csi1grp {    fsl,pins = <     MX6UL_PAD_CSI_MCLK__CSI_MCLK  0x1b088     MX6UL_PAD_CSI_PIXCLK__CSI_PIXCLK 0x1b088     MX6UL_PAD_CSI_VSYNC__CSI_VSYNC  0x1b088     MX6UL_PAD_CSI_HSYNC__CSI_HSYNC  0x1b088     MX6UL_PAD_CSI_DATA00__CSI_DATA02 0x1b088     MX6UL_PAD_CSI_DATA01__CSI_DATA03 0x1b088     MX6UL_PAD_CSI_DATA02__CSI_DATA04 0x1b088     MX6UL_PAD_CSI_DATA03__CSI_DATA05 0x1b088     MX6UL_PAD_CSI_DATA04__CSI_DATA06 0x1b088     MX6UL_PAD_CSI_DATA05__CSI_DATA07 0x1b088     MX6UL_PAD_CSI_DATA06__CSI_DATA08 0x1b088     MX6UL_PAD_CSI_DATA07__CSI_DATA09 0x1b088     MX6UL_PAD_SNVS_TAMPER5__GPIO5_IO05 0x17059            /* configue csi_reset in this case */     MX6UL_PAD_SNVS_TAMPER6__GPIO5_IO06 0x17059            /* configure csi_en in this case*/    > 2) Applay the video driver patches as attatched.   $git am 0001-Enable-CSI-support-BT656-interlace-and-add-adv7180.patch 3) Build the kernel.    make imx_v7_defconfig    make -j4 4) Use  unit_tests "mx6s_v4l2_capture_uyvy.out" to test it,  source code is attached "mx6s_v4l2_capture.c": Copy "mx6s_v4l2_capture_uyvy.out" to target device FS  "/unit_tests" folder, and run it like this: ./mx6s_v4l2_capture_uyvy.out -m 0 -t 50 -d /dev/video0 5) In this "mx6s_v4l2_capture.c" demo test code,   it can utilize software algorithm to implement CSC(Color Space Conversion) from YUV4:2:2 to RGB. PXP module can realize hardware CSC and image resize. 6) How to use CSI + PXP to preview camera, refer to  unit_tests "imx-test" package "pxp_v4l2_test" - "pxp_v4l2_test.c"  . Run following command in Target device FS:   /unit_tests /*Record raw camera UYVY data(720x576) to save in test1.yuv */ ./mx6s_v4l2_capture_uyvy.out -m 0 -t 10 -of test1.yuv -d /dev/video0 /*Play this test1.yuv file(UYVY) by PXP engine and resize to full screan(800x480) */ ./pxp_v4l2_test.out -sx 800 -sy 480 -res 720:576 -dst 0:0:800:480 -a 100 -f 5 test1.yuv BLANK 7) The "pxp_v4l2_preview_test.c" demo attached  is  a TV-in demo for i.MX6ul which support ADV7180 camera preview by PXP in time. This demo implements scaling image frame to full screen, Alpha blending and Composite two image together function by PXP hardware.

Overview As more and more communication required between online and offline, the QR code is widely used in the mobile payment, mobile small apps, industry things identification and etc. The i.MX6UL/ULL has the IP of CSI and PXP for camera connection and image CSC/FLIP/ROTATION acceleration. A LCDIF IP is supporting the display, but no 3D IP support. This means this low power and low end AP is very suitable for the industry HMI segment, which does not require a cool 3D graphic display, but a simple and straightforward GUI for interaction. QR code scanner is one of the use cases in the industry segment, which more and more customer are focusing on. The i.MX6UL CPU freq of i.MX6UL is about 500Mhz, and it does not have GPU IP, so a lightweight GUI and window system is required. Here we recommend the QT with wayland backend (without X11), which would make the window system small and faster than traditional X11 UI. Why chose QT is because of it has open source version, rich components, platform independent, good performance for embedded system and strong development staffs like QtCreator for creating application. How to enable the QT development environment, check this: Enable QT developement for i.MX6UL (v2)  Here I made a QR code scanner demo based on QT5.6 + QZXing (QR/Bar code scan engine) running on the i.MX6UL EVK board with a UVC camera (at least 640x480 resolution is required) and 480x272px LCD. Source code is open here (License Apache2.0): https://github.com/muddog/QRScanner  Implementation To do camera preview and capture, you must think on the gstreamer first, which is easy use and has the acceleration pads which implemented by NXP for i.MX6UL. Yes, it's very easy for you to enable the preview in console like: $ gst-launch-1.0 v4l2src device=/dev/video1 ! video/x-raw,format=YUY2,width=640,height=320 ! imxvideoconvert_pxp ! video/x-raw,format=RGB16 ! waylandsink It works under the i.MX6UL EVK, with PXP IP to do color space convert from YUY2 -> RGB16 acceleration, also the potential scaling of the image. The CPU loading of this is about 20-30%, but if you use the component of "videoconvert" to replace the "imxvideoconvert_pxp", we do CSC and scale by CPU, then the loading would increase to 50-60%. The "/dev/video1" is the device node for UVC camera, it may different in your environment. So our target is clear, create such pipeline (with PXP acceleration) in the QT application, and use a appsink to get preview images, do simple "sink" to one QWidget by drawing this image on the widget surface for preview (say every 50ms for 20fps). Then in other thread, we fetch the preview buffer in a fixed frequency (like every 0.5s), then feed it into the ZXing engine to decode the strings inside this image. Here are the class created inside the source code: ScannerQWidgetSink It act as a gstreamer sink for preview rendering. Init the pipeline, create a timer with timeout every 50ms. In the timer handler, we use appsink to copy the camera buffer from gstreamer, and tell the ViewfinderWidget to do update (re-draw event). ViewfinderWidget This class inherit from the QWidget, which draw the preview buffer as a QImage onto it's own surface by using QPainter. The QImage is created at the very begining with the image buffer created by the ScannerQWidgetSink. Because QImage itself does not maintain the image buffer, so the buffer must be alive during it's usage. So we keep this buffer during the ScannerQWidgetSink life cycle, copy the appsink buffer from pipeline to it for preview. MainWindow Create main window, which does not have title bar and border. Start any animation for the red line scan bar. Create instance of DecoderThread and ScannerQWidgetSink. Setup and start them. DecoderThread A infinite loop, to wait for a available buffer released by the ScannerQWidgetSink every 0.5s. Copy the buffer data to it's own buffer (imgData) to avoid any change to the buffer by sink when doing decoding. Then feed this copy of buffer into ZXing engine to get decoder result. Then show on the QLabel. Screenshot under wayland (weston) desktop: Customize Camera instance Now I use the UVC camera which pluged in the USB host, which device node is /dev/video1. If you want to use CSI or other device, please change the construction parameters for ScannerQWidgetSink(): sink = new ScannerQWidgetSink(ui->widget, QString("v4l2src device=/dev/video1")); Image resolution captured and review Change the static member value of ScannerQWidgetSink class: uint ScannerQWidgetSink::CAPTURE_HEIGHT = 480; uint ScannerQWidgetSink::CAPTURE_WIDTH = 640; Preview fps and decoding frequency Find the "framerate=20/1" strings in the ScannerQWidgetSink::GstPipelineInit(), change to your fps. You also have to change the renderTimer start timeout value in the ::StartRender(). The decoding frequency is determined by renderCnt, which determine after how many preview frames showed to feed the decoder. Main window size It's fixed size of main window, you have to change the mainwindow.ui. It's easy to do in the QtCreate Designer. FAQ Why not use CSI camera in demo? Honestly, I do not have CSI camera module, it's also DNP when you buying the board on NXP.com. So a widely used UVC camera is preferred, it's also easy for you to scan QR code on your phone, your display panel etc. Why not use QCamera to do preview and capture? The QCamera class in the Qtmultimedia component uses the camerabin2 gstreamer plugin, which create a very long pipeline for different usage of viewfinder, image capture and video encoder. Camerabin2 would eat too much CPU and memory resource, take picture and recording are very very slow. The preview of 30fps would eat about 70-80% CPU loading even I hacked it using imxvideoconvert_pxp instread of software videoconvert. Finally I give up to implement the QRScanner based on QCamera. How to make sure only one instance of QT app is running? We can use QSharedMemory to create a share memory with a unique KEY. When second instance of app is started, it would check if the share memory with this KEY is created or not. If the shm is there, it means there's already one instance running, it has to exit(). But as the QT mentioned, the QSharedMemory can not be destroyed correctly when app crashed, this means we have to handle each terminate signal, and do delete by ourselves: static QSharedMemory *gShm = NULL; static void terminate(int signum) {    if (gShm) {       delete gShm;       gShm = NULL;    }    qDebug() << "Terminate with signal:" << signum;    exit(128 + signum); } int main(int argc, char *argv[]) {    QApplication a(argc, argv);    // Handle any further termination signals to ensure the    // QSharedMemory block is deleted even if the process crashes    signal(SIGHUP, terminate ); // 1    signal(SIGINT, terminate ); // 2    signal(SIGQUIT, terminate ); // 3    signal(SIGILL, terminate ); // 4    signal(SIGABRT, terminate ); // 6    signal(SIGFPE, terminate ); // 8    signal(SIGBUS, terminate ); // 10    signal(SIGSEGV, terminate ); // 11    signal(SIGSYS, terminate ); // 12    signal(SIGPIPE, terminate ); // 13    signal(SIGALRM, terminate ); // 14    signal(SIGTERM, terminate ); // 15    signal(SIGXCPU, terminate ); // 24    signal(SIGXFSZ, terminate ); // 25    gShm = new QSharedMemory("QRScannerNXP");    if (!gShm->create(4, QSharedMemory::ReadWrite)) {       delete gShm;       qDebug() << "Only allow one instance of QRScanner";       exit(0);    } .....

Many times I came across one issue while using Redlib in MXUXpresso IDE project. I like to provide some guidance here to match the proper solution that can help others. Problem Statement : printf or sprintf doesn't print anything or printing random characters while using Redlib library. Reason : When you are creating your project you may ask to choose the c/c++ library setting to select either of the c libray provided by IDE in Advanced project setting wizard. If you have not checked the option "Redlib: Use floating point version of printf" (which will use the floating-point variant of printf) have tried to print the floating point value, You will end up with the problem mentioned above. Solution : You need to enable the floating support by modifying some preprocessor directives in "Defined symbols (-D)" wizard of your project. Path :  Your Project > properties > C/C++ Build > Setiings > Tool Settings > MCU C Compiler > Preprocessor. These are: PRINTF_FLOAT_ENABLE - keep the directive value to "1" SCANF_FLOAT_ENABLE - keep the directive value to "1" CR_INTEGER_PRINTF - Undefine/Remove this directive Click on Apply and close. That is it. Now you will have your expected prints for floating point values in your debugger console.

Wayland:   Wayland is a display SERVER and COMPOSITION protocol. It is relatively new, as its first release was in 2012. The protocol enables applications to allocate their own off-screen buffers and render their window contents directly, using hardware accelerated libraries like OpenGL ES, or high quality software implementations like Cairo. Wayland is ONLY a display server protocol, not a display server itself. Weston is the reference Wayland protocol implementation.   YOCTO Setup . $ mkdir ~/bin $ curl http://commondatastorage.googleapis.com/git-repo-downloads/repo > ~/bin/repo $ chmod a+x ~/bin/repo $ export PATH=~/bin:$PATH $ git config --global user.name "Your Name" $ git config --global user.email "Your Email" $ git config –list $ mkdir fsl-release-bsp $ cd fsl-release-bsp $ repo init -u git://git.freescale.com/imx/fsl-arm-yocto-bsp.git -b imx-3.14.52-1.1.0_ga $ repo sync     you will be able to build Yocto and also have all the recipes to do so, we need to add WAYLAND, then execute the following steps: $ DISTRO=fsl-imx-wayland MACHINE=imx6qsabresd source fsl-setup-release.sh -b build-wayland $ bitbake fsl-image-gui After these steps, you will have a wayland based i.MX6Q image where you will be able to play with all the knowledge we provided here.   Once your image has been properly generated, you will find the Weston source codes in: <YOUR YOCTODIR>/build-wayland/tmp/work/cortexa9hf-vfp-neon-mx6qdl-poky-linux-gnueabi/weston/1.9.0-r0/weston-1.9.0     Wayland application for extended desktop: This functionality is only supported using the GAL2D blitter, in order to enable a multiple desktop approach, you need to pass the following parameters to your weston command: /etc/init.d/weston stop echo 0 > /sys/class/graphics/fb4/blank weston --tty=1 --use-gal2d=1 --use-gl=0 --device=/dev/fb0,/dev/fb4 &     Xwayland: Wayland is a complete window system in itself, but even so, if we're migrating away from X, it makes sense to have a good backwards compatibility story. With a few changes, the Xorg server can be modified to use wayland input devices for input and forward either the root window or individual top-level windows as wayland surfaces.   DISTRO=fsl-imx-xwayland MACHINE=imx6qsabresd source ./fsl-setup-release.sh -b build-xwayland bitbake fsl-image-gui Once you have the image your Wayland/Weston image will be able to run X11 applications   Excepting X11 applications that use EGL, we don’t support that, if you plan to use EGL apps, please use the Wayland provided functions to create the buffer.   Application for rotation: Weston allows rotating windows with super-key + middle mouse button. As this works for Wayland clients only, you can run Xwayland in weston, run your X application on Xwayland, and rotate the Xwayland display. For another option: Create a file ~/.config/weston.ini with this content: [core] modules=xwayland.so shell=desktop-shell.so idle-time=0 [shell] background-color=0xff002244 locking=false # panel-location=none    [launcher] icon=/usr/share/icons/gnome/24x24/apps/utilities-terminal.png path=/usr/bin/weston-terminal [launcher] icon=/usr/share/icons/hicolor/48x48/apps/firefox.png path=/usr/bin/firefox [output] name=X1 mode=640x800 transform=90 # wanna get mad? use: transform=flipped-270 scale=1 This weston.ini enables a rootless xwayland.so in weston. The [output] section with name=X1 defines weston's appearance as X client. transform=90 rotates the weston display.   the [launcher] sections can be used to create custom panel starters for your X applications. See  /usr/share/doc/weston/examples/weston.ini for more detailed information for further cases, I will attach in the future.

Requirements: Host machine with Ubuntu 14.04 UDOO Quad/Dual Board uSD card with at least 8 GB Download documentation and install latest Official Udoobuntu OS (at the moment of writing: UDOObuntu 2.1.2), https://www.udoo.org/downloads/   Overview: This document describes how to install and test Keras (Open source neural network library) and Theano (numerical computation library for python ) for deep learning library usage on i.MX6QD UDOO board.  Installation: $ sudo apt-get update && sudo apt-get upgrade update your date system: e.g. $ sudo date -s “07/08/2017 12:00” First satisfy the run-time and build time dependencies: $ sudo apt-get install python-software-properties software-properties-common make unzip zlib1g-dev git pkg-config autoconf automake libtool curl  python-pip python-numpy libblas-dev liblapack-dev python-dev libatlas-base-dev gfortran libhdf5-serial-dev libhdf5-dev python-setuptools libyaml-dev libpython2.7-dev $ sudo easy_install scipy The last step is installing scipy through pip, and can take several hours. Theano First, we have a few more dependencies to get: $sudo pip install scikit-learn $sudo pip install pillow $sudo pip install h5py With these dependencies met, we can install a stable Theano release from the git source: $ git clone https://github.com/Theano/Theano $ cd Theano Numpy 1.9 cause conflicts with armv7, so we need to change the setup.py configuration: $ sudo nano setup.py Remove line    #       install_requires=['numpy>=1.9.1', 'scipy>=0.14', 'six>=1.9.0'], And add setup_requires=["numpy"], install_requires=["numpy"], Then install it: $ sudo python setup.py install Keras The installation can occur with the command: (this could take a lot of time!!!) $ cd .. $ git clone https://github.com/fchollet/keras.git $ cd keras $ sudo python setup.py install $ LC_ALL=C $sudo pip install --upgrade keras After Keras is installed, you will want to edit the Keras configuration file ~/.keras/keras.json to use Theano instead of the default TensorFlow backend. If it isn't there, you can create it. This requires changing two lines. The first change is: "image_dim_ordering": "tf"  --> "image_dim_ordering": "th" and the second: "backend": "tensorflow" --> "backend": "theano" (The final file should look like the example below) sudo nano ~/.keras/keras.json {     "image_dim_ordering": "th",     "epsilon": 1e-07,     "floatx": "float32",     "image_data_format": "channels_last",     "backend": "theano" } You can also define the environment variable KERAS_BACKEND and this will override what is defined in your config file : $ KERAS_BACKEND=theano python -c "from keras import backend" Testing Quick test: udooer@udoo:~$ python Python 2.7.6 (default, Oct 26 2016, 20:46:32) [GCC 4.8.4] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import keras Using Theano backend. >>>  Test 2: Be aware this test take some time (~1hr on udoo dual): $ curl -sSL -k https://github.com/fchollet/keras/raw/master/examples/mnist_mlp.py | python Output: For demonstration, deep-learning-models repository provided by pyimagesearch and from fchollet git, and also have three Keras models (VGG16, VGG19, and ResNet50) online — these networks are pre-trained on the ImageNet dataset, meaning that they can recognize 1,000 common object classes out-of-the-box. $ cd keras $ git clone https://github.com/fchollet/deep-learning-models $ Cd deep-learning-models $ ls -l Notice how we have four Python files. The resnet50.py , vgg16.py , and vgg19.py  files correspond to their respective network architecture definitions. The imagenet_utils  file, as the name suggests, contains a couple helper functions that allow us to prepare images for classification as well as obtain the final class label predictions from the network Classify ImageNet classes with ResNet50 ResNet50 model, with weights pre-trained on ImageNet. This model is available for both the Theano and TensorFlow backend, and can be built both with "channels_first" data format (channels, height, width) or "channels_last" data format (height, width, channels). The default input size for this model is 224x224. We are now ready to write some Python code to classify image contents utilizing  convolutional Neural Networks (CNNs) pre-trained on the ImageNet dataset. For udoo Quad/Dual use ResNet50 due to avoid space conflict. Also we are going to use ImageNet (http://image-net.org/) that is an image database organized according to the WordNet hierarchy, in which each node of the hierarchy is depicted by hundreds and thousands of images. from keras.applications.resnet50 import ResNet50 from keras.preprocessing import image from keras.applications.resnet50 import preprocess_input, decode_predictions import numpy as np   model = ResNet50(weights='imagenet')   #for this sample I download the image from: http://i.imgur.com/wpxMwsR.jpg  img_path = 'elephant.jpg' img = image.load_img(img_path, target_size=(224, 224)) x = image.img_to_array(img) x = np.expand_dims(x, axis=0) x = preprocess_input(x)   preds = model.predict(x) # decode the results into a list of tuples (class, description, probability) # (one such list for each sample in the batch) print('Predicted:', decode_predictions(preds, top=3)[0]) Save the file an run it. Results for elephant image: Top prediction was 0.8890 for African Elephant Testing with this image: http://i.imgur.com/4FIOwAN.jpg Results: Top prediction was: 0.7799 for golden_retriever. Now your Udoo is ready to use Keras and Theano as Deep Learning libraries, next time we are going to show some usage example for image classification models with OpenCV. References: GitHub - fchollet/keras: Deep Learning library for Python. Runs on TensorFlow, Theano, or CNTK.  GitHub - Theano/Theano: Theano is a Python library that allows you to define, optimize, and evaluate mathematical expres…  GitHub - fchollet/deep-learning-models: Keras code and weights files for popular deep learning models.  Installing Keras for deep learning - PyImageSearch 

Building Linux Kernel Building Linux Kernel Building Using LTIB Building Outside LTIB Downloading and installing GNU Toolchain and git Building Kernel from Freescale git repository Building Kernel Mainline About Linux Building Using LTIB Linux kernel can be easily built using Ltib. On Ltib menu, just select: [*] Configure the Kernel When you exit this menu, Ltib will show the Kernel Menuconfig as below: This is the Kernel Menuconfig, where it's possible to configure kernel options and drivers. After exit this menu, kernel will be built and stored at: <Ltib directory>/rootfs/boot Building Outside LTIB Downloading and installing GNU Toolchain and git When you install LTIB, a GNU toolchain is automatically installed on /opt/freescale/usr/local/ Kernel releases newer than 2.6.34 doesn't build on Toolchain 4.1.2, only on 4.4.1 or later Check on your host at /opt/freescale/usr/local/ the current installed Toolchain. Next step is to install GIT on host. For Ubuntu machines, use: sudo apt-get install git-core Building Kernel from Freescale git repository Freescale provides access to their own git kernel repository and can be viewed at: Freescale Public GIT To download the kernel source code, create a new folder and use the command: git clone git://git.freescale.com/imx/linux-2.6-imx.git OR git clone http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git After some minutes, a folder called linux-2.6-imx will be created containing the Linux kernel Create a local git branch from a remote branch you want to use. Let's use branch origin/imx_3.0.15 as example: cd linux-2.6-imx git checkout -b localbranch origin/imx_3.0.15 To check all available remote branches, use: git branch -r Export the cross compiler, architecture and the toolchain path: export ARCH=arm export CROSS_COMPILE=arm-none-linux-gnueabi- If using Toolchain 4.1.2: export PATH="$PATH:/opt/freescale/usr/local/gcc-4.1.2-glibc-2.5-nptl-3/arm-none-linux-gnueabi/bin/" OR If using Toolchain 4.4.4: export PATH="$PATH:/opt/freescale/usr/local/gcc-4.4.4-glibc-2.11.1-multilib-1.0/arm-fsl-linux-gnueabi/bin/" Copy the config file for the wanted platform on linux folder as example: cp arch/arm/configs/imx6_defconfig .config All platform config files are located at <linux directory>/arch/arm/configs/ Call menuconfig and change configuration (if needed) make menuconfig Now it's ready to be built: make uImage The zImage and uImage will be located at /arch/arm/boot/ folder. Building Kernel Mainline Mainline Kernel can be viewed on this link: https://git.kernel.org/cgit/linux/kernel/git/stable/linux-stable.git To download the kernel source code, create a new folder and use the command: git clone git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git OR git clone http://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git OR git clone https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git After some minutes, a folder called linux will be created containing the Linux kernel Create a local git branch from a remote branch you want to use. Let's use branch origin/linux-3.8.y as example: cd linux git checkout -b localbranch origin/linux-3.8.y To check all available remote branches, use: git branch -r Export the cross compiler, architecture and the toolchain path: export ARCH=arm export CROSS_COMPILE=arm-none-linux-gnueabi- If using Toolchain 4.4.4: export PATH="$PATH:/opt/freescale/usr/local/gcc-4.4.4-glibc-2.11.1-multilib-1.0/arm-fsl-linux-gnueabi/bin/" Configure to the platform you want to build kernel. For i.MX family, use imx_v6_v7_defconfig: make imx_v6_v7_defconfig All platform config files are located at <linux directory>/arch/arm/configs/ Call menuconfig and change configuration (only if needed, this is an optional step!) make menuconfig Now it's ready to be built: make -j4 uImage LOADADDR=0x70008000 - Use -j4 option to speed up your build in case or PC has 4 cores. It's optional. - IMPORTANT: Use the correct address for each processor. You can check the correct address value at linux/arch/arm/mach-imx/Makefile.boot. After build the uImage, build the dtb file (device tree binary). For i.MX53 QSB use: make imx53-qsb.dtb The uImage will be located at: linux/arch/arm/boot/ folder and dtb binary will be located at: linux/arch/arm/boot/dts About Linux For general Linux information, see About Linux

This is a HW design checklist for customer's reference. Please read and fill it in carefully before requesting a schematic review.

Recipes to include Amazon's Alexa Voice Services in your applications. Step 1 : Get iMX Yocto AVS setup environment Review the steps under Chapter 3 of the i.MX_Yocto_Project_User'sGuide.pdf on the L4.X LINUX_DOCS to prepare your host machine. Including at least the following essential Yocto packages $ sudo apt-get install gawk wget git-core diffstat unzip texinfo \ gcc-multilib build-essential chrpath socat libsdl1.2-dev u-boot-tools Install the i.MX NXP AVS repo Create/Move to a directory where you want to install the AVS yocto build enviroment. Let's call this as <yocto_dir> $ cd <yocto_dir> $ repo init -u https://source.codeaurora.org/external/imxsupport/meta-avs-demos -b master -m imx-alexa-sdk-4.9.51-8mq_ga.xml Download the AVS BSP build environment: $ repo sync Step 2: Setup yocto for Alexa_SDK image with AVS-SETUP-DEMO script: Run the avs-setup-demo script as follows to setup your environment for the imx8mqevk board: $ MACHINE=imx8mqevk DISTRO=fsl-imx-xwayland source avs-setup-demo.sh -b <build_sdk_8M> Where <build_sdk> is the name you will give to your build folder. After acepting the EULA the script will prompt if you want to enable: Sound Card selection The following Sound Cards are supported on the build: 2-Mic Synaptics/Conexant 2-Mic TechNexion Voice Hat (with DSPConcepts SW) The script will prompt to select the soundcard you will be using: Which Sound Card are you going to use? Synaptics/Conexant .................... 1 VoiceHat (for DSPConcepts SW) ......... 2 Type the number of your selection and press Enter... Install Alexa SDK Next option is to select if you want to pre-install the AVS SDK software on the image. Do you want to build/include the AVS_SDK package on this image(Y/N)? If you select YES, then your image will contain the AVS SDK ready to use (after authentication). Note this AVS_SDK will not have WakeWord detection support, but it can be added on runtime. If your selection was NO, then you can always manually fetch and build the AVS_SDK on runtime. All the packages dependencies will be already there, so only fetching the AVS_SDK source code and building it is required. Finish avs-image configuration At the end you will see a text according with the configuration you select for your image build. Next is an example for a Preinstalled AVS_SDK with Synaptics Sound Card support ============================================================ AVS configuration is now ready at conf/local.conf - Sound Card = Synaptics - Alexa SDK 1.7 pre-installed - Wifi supported You are ready to bitbake your AVS demo image now:   bitbake avs-image If you want to use QT5DisplayCards, use then:   bitbake avs-image-qt5 ============================================================ Step 3: Build the AVS image Go to your <build_sdk> directory and start the build of the avs-image There are 2 options Regular Build: $ cd  <yocto_dir>/<build_sdk>   $ bitbake avs-image With QT5 support included: $ cd  <yocto_dir>/<build_sdk>   $ bitbake avs-image-qt5 The image with QT5 is useful if you want to add some GUI for example to render DisplayCards. Step 4 : Deploying the built images to SD/MMC card to boot on target board. After a build has succesfully completed, the created image resides at <build_sdk>/tmp/deploy/images/imx8mqevk/ In this directory, you will find imx8mqevk-avs--.sdcard image or imx8mqevk-avs-qt5--.sdcard, depending on the build you chose on Step3. To Flash the .sdcard image into an SD Card follow the next steps: Extract and copy the .sdcard file to your SD Card $ cd <build_sdk>/tmp/deploy/images/imx8mqevk/ $ cp -v imx8mqevk-avs-synaptics-1.7.sdcard.bz2 <workdir> $ cd <workdir> $ sudo bzip2 -d imx8mqevk-avs-synaptics-1.7.sdcard.bz2 $ sudo dd if=imx8mqevk-avs-synaptics-1.7.sdcard.bz2 of=/dev/sd<part> bs=1M && sync $ sync Properly eject the SD Card: $ sudo eject /dev/sd<part> Insert the flashed SD Card on the 8M EVK and boot. Follow the instructions at startup to setup your AVS and run the SampleApp. NXP Documentation For a more comprehensive understanding of Yocto, its features and setup; more image build and deployment options and customization, please take a look at the i.MX_Yocto_Project_User's_Guide.pdf document from the Linux documents bundle mentioned at the beginning of this document. For a more detailed description of the Linux BSP, u-boot use and configuration, please take a look at the i.MX_Linux_User's_Guide.pdf document from the Linux documents bundle mentioned at the beginning of this document.

Script which patches the ltib folder on Ubuntu 12.04. Steps: $ cp patch-ltib-ubuntu12.04.sh <your ltib folder> $ cd <your ltib folder> $ chmod +x patch-ltib-ubuntu12.04.sh $ ./patch-ltib-ubuntu12.04.sh

Here are my experiences for compiling Qt5.3.0-beta1 on Yocto. Special thanks to Martin Jansa, the maintainer of the meta-qt5 layer and his help on this. My original procedure was based on this tutorial: Building Qt5 using yocto on Wandboard - Wandboard Wiki Reason: Qt5.3 contains a nice new plugin that allows the use of gstreamer output for textures without the CPU intensive step of copying them (Gerrit Code Review). This allows to play even full HD videos and apply all the power of Qt5 (e.g. shaders) to them. Steps: Setup your repo: repo init -u https://github.com/Freescale/fsl-community-bsp-platform -b master-next; repo sync Download meta-qt5 branch: cd sources; git clone -b jansa/qt5-5.3.0-beta1 https://github.com/meta-qt5/meta-qt5.git Checkout a specific revision: cd meta-qt5; git checkout 92be18a3a14deed9d38b8fc6e89f09ba4d730597 Apply the following patch (maybe later no longer needed): diff --git a/recipes-qt/qt5/qt5.inc b/recipes-qt/qt5/qt5.inc index dfc1c76..a2f9a73 100644 --- a/recipes-qt/qt5/qt5.inc +++ b/recipes-qt/qt5/qt5.inc @@ -54,6 +54,7 @@ FILES_${PN}-tools-dbg = " \ " FILES_${PN}-plugins-dbg = " \      ${OE_QMAKE_PATH_PLUGINS}/*/.debug/* \ +    ${OE_QMAKE_PATH_PLUGINS}/*/*/.debug/* \ " # extra packages @@ -98,6 +99,7 @@ FILES_${PN}-tools = " \ " FILES_${PN}-plugins = " \      ${OE_QMAKE_PATH_PLUGINS}/*/*${SOLIBSDEV} \ +    ${OE_QMAKE_PATH_PLUGINS}/*/*/*${SOLIBSDEV} \ " FILES_${PN}-mkspecs = "\      ${OE_QMAKE_PATH_ARCHDATA}/mkspecs \ Define your machine: export MACHINE=xxx (replace with your board) Setup build environment: cd .. ; . setup-environment build edit your local layer conf ("conf/bblayers.conf") and add the following two lines:   ${BSPDIR}/sources/meta-openembedded/meta-ruby \   ${BSPDIR}/sources/meta-qt5 \ edit your local.conf and add the following lines: DISTRO_FEATURES_remove = "x11 wayland" IMAGE_INSTALL_append = " \     firmware-imx-vpu-imx6q \     firmware-imx-vpu-imx6d \ " IMAGE_INSTALL_append = " \     cpufrequtils \     nano \     packagegroup-fsl-gstreamer \     packagegroup-fsl-tools-testapps \     packagegroup-fsl-tools-benchmark \     gstreamer \     gst-plugins-base-app \     gst-plugins-base \     gst-plugins-good \     gst-plugins-good-rtsp \     gst-plugins-good-udp \     gst-plugins-good-rtpmanager \     gst-plugins-good-rtp \     gst-plugins-good-video4linux2 \     qtbase-fonts \     qtbase-plugins \     qtbase-tools \     qtbase-examples \     qtdeclarative \     qtdeclarative-plugins \     qtdeclarative-tools \     qtdeclarative-examples \     qtdeclarative-qmlplugins \     qtmultimedia \     qtmultimedia-plugins \     qtmultimedia-examples \     qtmultimedia-qmlplugins \     qtsvg \     qtsvg-plugins \     qtsensors \     qtimageformats-plugins \     qtsystems \     qtsystems-tools \     qtsystems-examples \     qtsystems-qmlplugins \     qtscript \     qt3d \     qt3d-examples \     qt3d-qmlplugins \     qt3d-tools \     qtwebkit \     qtwebkit-examples-examples \     qtwebkit-qmlplugins \     cinematicexperience \     " PACKAGECONFIG_append_pn-qtmultimedia = " gstreamer010" QT5_VERSION = "5.2.1+5.3.0-beta1+git%" PREFERRED_VERSION_qtbase-native = "${QT5_VERSION}" PREFERRED_VERSION_qtbase = "${QT5_VERSION}" PREFERRED_VERSION_qtdeclarative = "${QT5_VERSION}" PREFERRED_VERSION_qtjsbackend = "${QT5_VERSION}" PREFERRED_VERSION_qtjsbackend-native = "${QT5_VERSION}" PREFERRED_VERSION_qtgraphicaleffects = "${QT5_VERSION}" PREFERRED_VERSION_qtimageformats = "${QT5_VERSION}" PREFERRED_VERSION_qtmultimedia = "${QT5_VERSION}" PREFERRED_VERSION_qtquick1 = "${QT5_VERSION}" PREFERRED_VERSION_qtquickcontrols = "${QT5_VERSION}" PREFERRED_VERSION_qtsensors = "${QT5_VERSION}" PREFERRED_VERSION_qtserialport = "${QT5_VERSION}" PREFERRED_VERSION_qtscript = "${QT5_VERSION}" PREFERRED_VERSION_qtsvg = "${QT5_VERSION}" PREFERRED_VERSION_qttools-native = "${QT5_VERSION}" PREFERRED_VERSION_qtwebkit = "${QT5_VERSION}" PREFERRED_VERSION_qtwebkit-examples = "${QT5_VERSION}" PREFERRED_VERSION_qtxmlpatterns = "${QT5_VERSION}" build an image: bitbake core-image-minimal This image will build QT5.3 for framebuffer. If you want to use it with X11, then adapt according to this tutorial: Integrate Qt5 into yocto sato image on Wandboard - Wandboard Wiki Please tell me, if I missed something. I wrote this as I remembered the steps.

Question: How is mx6 PMIC_ON_REQ under SW control? mx6 PMIC_ON_REQ is hooked up to the PFUZE100's PWRON and Linux and our 3.0.35bsp is used. Mx6 SW control is to drive the PMIC_ON_REQ pin low.  It appears from the documentation that this pin can be controlled by either another imx6 pin OR through SW control. The issue is that the reference manual is not clear on how to do this. While doing an SR search (SR 1-877711457), it does appear the PMIC_ON_REQ is controlled by SW. Answer: In latest RM version, Figure 60-3. Chip on/off state flow diagram and Table 60-3. Power mode transitions in IMX6DQRM.pdf show two ways to make PMIC_ON_REQ go low. I'm sure in latest BSP SW method had been included. It turns out the SNVS module on the mx6s/dl is different from the mx6q/d which is again different from the mx6slx. The bottom line is that the requirements for the SNVS functionality came primarily from the Android market so many of the Linux use cases are not supported. SW control of the PMIC_ON_REQ pin is an example of this. This means that you are correct, there only 2 ways to get PMIC_ON_REQ to power up for the mx6q/d 1 -  a low on the ON/OFF pin greater than the debounce time (750ms) 2 - a wake-up/tamper event. For the mx6s/dl, there are 3 ways to get PMIC_ON_REQ to power up 1 - power-on-reset on the VSNVS  (i.e first applying VSNVS) 2 -  a low on the ON/OFF pin greater than the debounce time (750ms) 3 - a wake-up/tamper event. Note, in my case, where there is an external input that actually wakes up the system, turns on the PMIC and brings up the mx6 there is only 1 way to get PMIC_ON_REQ to go back high 1 - a low on the ON/OFF pin greater than the debounce time (750ms) As it turns out, when the VSNVS_HP section is powered (i.e VDDHIGH is applied), it gates off the wake-up timer.

Contents 1 创建 i.MX8QXP Linux 4.14.98_ga 板级开发包编译环境 2 1.1 下载板级开发包 ...................................................... 2 1.2 创建yocto编译环境: ................................................ 3 2 Device Tree ............................................................. 15 2.1 恩智浦的device Tree结构 ..................................... 15 2.2 device Tree的由来(no updates) ............................ 18 2.3 device Tree的基础与语法(no updates) ................. 20 2.4 device Tree的代码分析(no updates) .................... 42 3 恩智浦i.MX8XBSP 包文件目录结构 ......................... 75 4 恩智浦i.MX8XBSP的编译(no updates) .................... 77 4.1 需要编译哪些文件 ................................................ 77 4.2 如何编译这些文件 ................................................ 78 4.3 如何链接为目标文件及链接顺序 ........................... 79 4.4 kernel Kconfig ...................................................... 81 5 恩智浦BSP的内核初始化过程(no updates) .............. 81 5.1 初始化的汇编代码 ................................................ 83 5.2 初始化的C代码 ..................................................... 87 5.3 init_machine ....................................................... 100 6 恩智浦BSP的内核定制 ........................................... 103 6.1 DDR修改 ............................................................ 103 6.2 IO管脚配置与Pinctrl驱动 .................................... 105 6.3 新板bringup ........................................................ 120 6.4 更改调试串口 ...................................................... 128 6.5 uSDHC设备定制(eMMC flash,SDcard, SDIOcard) 135 6.6 LVDS LCD 驱动定制 .......................................... 144 6.7 GPIO_Key 驱动定制 .......................................... 147 6.8 GPIO_LED 驱动定制 ......................................... 151 6.9 Fuse nvram驱动 ................................................. 154 6.10 SPI与SPI Slave驱动 ........................................... 155 6.11 USB 3.0 TypeC 改成 USB 3.0 TypeA(未验证) ... 162 6.12 汽车级以太网驱动定制 ....................................... 162 6.13 i.MX8DX MEK支持 ............................................. 180 6.14 NAND Flash支持与烧录 ..................................... 181

In i.MX8MQ and i.MX8M Mini, the codec used is WM8524, which only supports audio playback. Although 8M Mini does have PDM microphone interface (MICFIL), there is no support for audio record via I2S. This guide will show you how to add audio recording driver in i.MX8MQ/8MM step by step.   Hardware: i.MX8MQ/8MM Evk, I2S output digital microphone OS: Android/Linux Kernel version: 4.14.78 For detailed steps, please see attachment.

目录 1    硬件资源，文档及工具下载... 2 1.1    硬件资源... 2 1.2    内存配置测试相关的文档... 3 1.3    内存压力测试工具. 3 1.4    内存配置工具. 4 2    内存设计要求... 4 3    LPDDR4基础... 4 4    硬件连接... 6 5    i.MX8QXP/DXP+LPDDR4内存配置与测试步骤... 8 5.1    生成LPDDR4初始化脚本... 8 5.2    使用内存测试工具测试内存... 13 5.3    编译内存测试工具所用的SCFW镜像... 17 5.4    其它尺寸的LPDDR4配置... 18 6    i.MX8DX+DDR3L内存配置... 23 7    测试失败的DEBUG.. 26 8    内存参数应用到SCFW中... 30

Platform: i.mx8qm/qxp OS: imx-yocto-L4.14.98_2.0.0_ga Camera: max9286 deserializer <=> max96705 serializer  + ar0144 or: max9286 deserializer <=> max96705 serializer + ov9284 Note that currently only one camera is support and the serializer should be connected to the IN0 of max9286. Data format: ar0144: mono raw 12bit. ov9284: mono raw 10bit. On imx8qm/qxp the data will be recieved as raw 16bit and the valid data bit start from bit[13] to LSB. for mono raw 12bit the valid data bit is 0bxxdd_dddd_dddd_ddxx for mono raw 10bit the valid data bit is 0bxxdd_dddd_dddd_xxxx max9286 and max96705 configuration: dbl bws PXL_CRC/edc hven hibw lccen him should be the same on both sides, this can be achieved by pin or register configurations. The crossbar function of max96705 can be used to fix the reversed data bit. for example, reversed 12bit with dbl to 1. 0x20 0xb 0x21 0xa 0x22 0x9 ....... 0x2b 0x0 0x30 0xb 0x31 0xa .... 0x3b 0x0 0x20 to 0x2b and 0x30 to 0x3b are the registers of max96705. Patch apply: 1. push the kernel-patch to the kernel source and apply it. 2. reconfig the kernel setting, make sure there is only CONFIG_MAX9286_AR0144 or        CONFIG_MAX9286_WISSEN(ov9284) enabled, all other max9286 related are disabled. You can run menuconfig to achieve this. 3. For testing copy the vulkan-v4l2.tar to the board, and run vulkan-v4l2.     the source code is at https://github.com/sheeaza/vulkan-v4l2 branch ar0144 for ar0144, branch ov9284 for ov9284. =========== updated patch for data format.

Introduction The Intel® Neural Compute Stick 2 (Intel® NCS 2) is Intel’s newest deep learning inference development kit. Packed in an affordable USB-stick form factor, the Intel® NCS 2 is powered by latest VPU (vision processing unit) – the Intel® Movidius™ Myriad X, which includes an on-chip neural network accelerator called the Neural Compute Engine. With 16 SHAVE cores and a dedicated hardware neural network accelerator, the NCS 2 offers up to 8x performance improvement+ over the previous generation. Ref: https://software.intel.com/en-us/articles/run-intel-openvino-models-on-intel-neural-compute-stick-2   The NCS 2 officially supported hardware platform is x86 PC and Raspberry Pi. In this guide, we will introduce how to implement in i.MX8MQ. Please see attached guide for more details.

Check new updated version for with Morty here Step 1 : Get iMX Yocto AVS setup environment Review the steps under Chapter 3 of the i.MX_Yocto_Project_User'sGuide.pdf on the L4.X LINUX_DOCS to prepare your host machine. Including at least the following essential Yocto packages $ sudo apt-get install gawk wget git-core diffstat unzip texinfo \   gcc-multilib build-essential chrpath socat libsdl1.2-dev u-boot-tools Install the i.MX NXP AVS repo Create/Move to a directory where you want to install the AVS yocto build enviroment. Let's call this as <yocto_dir> $ cd <yocto_dir> $ repo init -u https://source.codeaurora.org/external/imxsupport/meta-avs-demos -b master -m imx7d-pico-avs-sdk_4.1.15-1.0.0.xml Download the AVS BSP build environment: $ repo sync Step 2: Setup yocto for Alexa_SDK image with AVS-SETUP-DEMO script: Run the avs-setup-demo script as follows to setup your environment for the imx7d-pico board: $ MACHINE=imx7d-pico DISTRO=fsl-imx-x11 source avs-setup-demo.sh -b <build_sdk> Where <build_sdk> is the name you will give to your build folder. After acepting the EULA the script will prompt if you want to enable: Sound Card selection The following Sound Cards are supported on the build: SGTL (In-board Audio Codec for PicoPi) 2-Mic Conexant The script will prompt if you are going to use the Conexant Card. If not then SGTL will be assumed as your selection Are you going to use Conexant Sound Card [Y/N]? Install Alexa SDK Next option is to select if you want to pre-install the AVS SDK software on the image. Do you want to build/include the AVS_SDK package on this image(Y/N)? If you select YES, then your image will contain the AVS SDK ready to use (after authentication). Note this AVS_SDK will not have WakeWord detection support, but it can be added on runtime. If your selection was NO, then you can always manually fetch and build the AVS_SDK on runtime. All the packages dependencies will be already there, so only fetching the AVS_SDK source code and building it is required. Finish avs-image configuration At the end you will see a text according with the configuration you select for your image build. Next is an example for a Preinstalled AVS_SDK with Conxant Sound Card support and WiFi/BT not enabled. ==========================================================   AVS configuration is now ready at conf/local.conf             - Sound Card = Conexant                                     - AVS_SDK pre-installed                                       You are ready to bitbake your AVS demo image now:               bitbake avs-image                                        ========================================================== Step 3: Build the AVS image Go to your <build_sdk> directory and start the build of the avs-image There are 2 options Regular Build: $ cd <yocto_dir>/<build_sdk> $ bitbake avs-image With QT5 support included: $ cd <yocto_dir>/<build_sdk> $ bitbake avs-image-qt5 The image with QT5 is useful if you want to add some GUI for example to render DisplayCards. Step 4 : Deploying the built images to SD/MMC card to boot on target board. After a build has succesfully completed, the created image resides at <build_sdk>/tmp/deploy/images/imx7d-pico/ In this directory, you will find the imx7d-pico-avs.sdcard image or imx7d-pico-avs-qt5.sdcard, depending on the build you chose on Step3. To Flash the .sdcard image into the eMMC device of your PicoPi board follow the next steps: Download the bootbomb flasher Follow the instruction on Section 4. Board Reflashing of the Quick Start Guide for AVS kit to setup your board on flashing mode. Copy the built SDCARD file $ sudo dd if=imx7d-pico-avs.sdcard of=/dev/sd bs=1M && sync $ sync Properly eject the pico-imx7d board: $ sudo eject /dev/sd NXP Documentation Refer to the Quick Start Quide for AVS SDK to fully setup your PicoPi board with Synaptics 2Mic and PicoPi i.mx7D For a more comprehensive understanding of Yocto, its features and setup; more image build and deployment options and customization, please take a look at the i.MX_Yocto_Project_User's_Guide.pdf document from the Linux documents bundle mentioned at the beginning of this document. For a more detailed description of the Linux BSP, u-boot use and configuration, please take a look at the i.MX_Linux_User's_Guide.pdf document from the Linux documents bundle mentioned at the beginning of this document.

Basic Linear Algebra Subprograms (BLAS) is a specification that prescribes a set of low-level routines for performing common linear algebra operations such as vector addition, scalar multiplication, dot products, linear combinations, and matrix multiplication. OpenBLAS is an optimized BLAS library which is uesd for deep learning accelerator in Caffe/Caffe2. I enable it in Yocto (Rocko) by adding bb file. And I build on i.MX6QP, i.MX7ULP and i.MX8MQ and also run its test example successfully. You can find test example(openblas_utest) under folder image/opt/openblas/bin of OpenBLAS work directory. Currently, version 0.3.0 is supported in the bb file. +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ update to v 0.3.6 and enable mutli-thread by set USE_OPENMP=1 and USE_THREAD=4 when compiling this library.

In defaut Linux BSP, NXP implemented LVDS to HDMI(it6263) and MIPI-DSI to HDMI(adv7535) bridge chip drivers. And these drivers need read the EDID from display, then apply the timing parameters to DRM driver. But for the use case that bridge chip -> Serializer -> Deserializer -> LCD Panel use case, there is no EDID. The attached are reference patches for such use case, it combined the bridge chip to panel directly, and no EDID is needed. The patches are tested on iMX8QXP MEK with bridge chip + panel mode, both of them can see the fb0 device under /sys/class/graphics/ folder, also can see card under  /sys/class/drm/. Display works fine with DTS selected 720P panel mode. [2020-06-24]: Add patches for L4.14.98 kernel: Android_Auto_P9.0.0_GA2.1.0_Kernel_No_EDID_IT6263.patch L4.14.98-iMX8QXP-MEK-ADV7535-MIPI-DSI-to-HDMI-bridge-chip-com.patch