<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <meta name="generator" content= "HTML Tidy for Linux/x86 (vers 1 September 2005), see www.w3.org"> <meta http-equiv="Content-Type" content= "text/html; charset=us-ascii"> <title>Chapter 5. Listing of Installed Components</title> <meta name="generator" content="DocBook XSL Stylesheets V1.68.1"> <link rel="start" href="index.html" title= "NVIDIA Accelerated Linux Graphics Driver README and Installation Guide"> <link rel="up" href="installationandconfiguration.html" title= "Part I. Installation and Configuration Instructions"> <link rel="prev" href="installdriver.html" title= "Chapter 4. Installing the NVIDIA Driver"> <link rel="next" href="editxconfig.html" title= "Chapter 6. Configuring X for the NVIDIA Driver"> </head> <body> <div class="navheader"> <table width="100%" summary="Navigation header"> <tr> <th colspan="3" align="center">Chapter 5. Listing of Installed Components</th> </tr> <tr> <td width="20%" align="left"><a accesskey="p" href= "installdriver.html">Prev</a> </td> <th width="60%" align="center">Part I. Installation and Configuration Instructions</th> <td width="20%" align="right"> <a accesskey="n" href= "editxconfig.html">Next</a></td> </tr> </table> <hr></div> <div class="chapter" lang="en"> <div class="titlepage"> <div> <div> <h2 class="title"><a name="installedcomponents" id= "installedcomponents"></a>Chapter 5. Listing of Installed Components</h2> </div> </div> </div> <p>The NVIDIA Accelerated Linux Graphics Driver consists of the following components (filenames in parentheses are the full names of the components after installation). Some paths may be different on different systems (e.g., X modules may be installed in /usr/X11R6/ rather than /usr/lib/xorg/).</p> <div class="itemizedlist"> <ul type="disc"> <li> <p>An X driver (<code class= "filename">/usr/lib/xorg/modules/drivers/nvidia_drv.so</code>); this driver is needed by the X server to use your NVIDIA hardware.</p> </li> <li> <p>A GLX extension module for X (<code class= "filename">/usr/lib/xorg/modules/extensions/libglxserver_nvidia.so.430.64</code>); this module is used by the X server to provide server-side GLX support.</p> </li> <li> <p>EGL and OpenGL ES libraries ( <code class= "filename">/usr/lib/libEGL.so.1</code>, <code class= "filename">/usr/lib/libGLESv1_CM.so.430.64</code>, and <code class="filename">/usr/lib/libGLESv2.so.430.64</code> ); these libraries provide the API entry points for all OpenGL ES and EGL function calls. They are loaded at run-time by applications.</p> </li> <li> <p>A Wayland EGL external platform library (<code class= "filename">/usr/lib/libnvidia-egl-wayland.so.1</code>) and its corresponding configuration file ( <code class= "filename">/usr/share/egl/egl_external_platform.d/10_nvidia_wayland.json</code> ); this library provides client-side Wayland support on top of the EGLDevice and EGLStream families of extensions, for use in combination with an EGLStream-enabled Wayland compositor: <a href= "https://cgit.freedesktop.org/~jjones/weston/" target= "_top">https://cgit.freedesktop.org/~jjones/weston/</a></p> <p>More information can be found along with the EGL external interface and Wayland library source code at <a href= "https://github.com/NVIDIA/eglexternalplatform" target= "_top">https://github.com/NVIDIA/eglexternalplatform</a> and <a href="https://github.com/NVIDIA/egl-wayland" target= "_top">https://github.com/NVIDIA/egl-wayland</a>.</p> </li> <li> <p>Vendor neutral graphics libraries provided by libglvnd (<code class="filename">/usr/lib/libOpenGL.so.0</code>, <code class="filename">/usr/lib/libGLX.so.0</code>, and <code class="filename">/usr/lib/libGLdispatch.so.0</code>); these libraries are currently used to provide full OpenGL dispatching support to NVIDIA's implementation of EGL.</p> <p>Source code for libglvnd is available at <a href= "https://github.com/NVIDIA/libglvnd" target= "_top">https://github.com/NVIDIA/libglvnd</a></p> </li> <li> <p>GLVND vendor implementation libraries for GLX (<code class= "filename">/usr/lib/libGLX_nvidia.so.0</code>) and EGL (<code class="filename">/usr/lib/libEGL_nvidia.so.0</code>); these libraries provide NVIDIA implementations of OpenGL functionality which may be accessed using the GLVND client-facing libraries.</p> </li> <li> <p>A GLX client library and Vulkan ICD (<code class= "filename">/usr/lib/libGL.so.1</code>), either as part of the GLVND infrastructure, or a legacy, non-GLVND GLX client library. This library provides API entry points for all GLX function calls, and is loaded at run-time by applications. Users may choose one or the other at installation time by using either the <code class= "option">--glvnd-glx-client</code> or the <code class= "option">--no-glvnd-glx-client</code> command line option to <span><strong class="command">nvidia-installer</strong></span>.</p> <p>Note that although both the GLVND and non-GLVND GLX client libraries share the same SONAME of libGL.so.1, only one of them at a time may be installed at a time. <code class= "filename">/usr/lib/libGL.so.430.64</code> is the non-GLVND GLX client library, and <code class= "filename">/usr/lib/libGL.so.1.0.0</code> is the GLVND GLX client library.</p> <p>This library is also used as the Vulkan ICD. Its configuration file is installed as <code class= "filename">/etc/vulkan/icd.d/nvidia_icd.json</code>.</p> <p>Repackagers of the driver are encouraged to provide the GLVND-based driver stack to promote adoption of the new infrastructure, but those who choose to package the legacy GLX client library instead of, or as an alternative to, the GLVND GLX client library should be aware that the NVIDIA EGL driver depends upon GLVND for proper functionality. The legacy GLX client library may coexist with most GLVND libraries, with the exception of <code class="filename">libGL.so.1</code> and <code class= "filename">libGLX.so.0</code>, so it is possible to support both NVIDIA EGL and legacy, non-GLVND NVIDIA GLX by installing all of the GLVND libraries except for libGL and libGLX alongside the legacy libGL.</p> </li> <li> <p>Various libraries that are used internally by other driver components. These include <code class= "filename">/usr/lib/libnvidia-cfg.so.430.64</code>, <code class= "filename">/usr/lib/libnvidia-compiler.so.430.64</code>, <code class= "filename">/usr/lib/libnvidia-eglcore.so.430.64</code>, <code class= "filename">/usr/lib/libnvidia-glcore.so.430.64</code>, <code class= "filename">/usr/lib/libnvidia-glsi.so.430.64</code>, <code class= "filename">/usr/lib/libnvidia-glvkspirv.so.430.64</code>, <code class= "filename">/usr/lib/libnvidia-rtcore.so.430.64</code>, and <code class= "filename">/usr/lib/libnvidia-cbl.so.430.64</code>.</p> </li> <li> <p>A VDPAU (Video Decode and Presentation API for Unix-like systems) library for the NVIDIA vendor implementation, (<code class="filename">/usr/lib/vdpau/libvdpau_nvidia.so.430.64</code>); see <a href="vdpausupport.html" title= "Appendix G. VDPAU Support">Appendix G, <i>VDPAU Support</i></a> for details.</p> </li> <li> <p>The CUDA library (<code class= "filename">/usr/lib/libcuda.so.430.64</code>) which provides runtime support for CUDA (high-performance computing on the GPU) applications.</p> </li> <li> <p>The Fatbinary Loader library (<code class= "filename">/usr/lib/libnvidia-fatbinaryloader.so.430.64</code>) provides support for the CUDA driver to work with CUDA fatbinaries. Fatbinary is a container format which can package multiple PTX and Cubin files compiled for different SM architectures.</p> </li> <li> <p>The PTX JIT Compiler library (<code class= "filename">/usr/lib/libnvidia-ptxjitcompiler.so.430.64</code>) is a JIT compiler which compiles PTX into GPU machine code and is used by the CUDA driver.</p> </li> <li> <p>Two OpenCL libraries (<code class= "filename">/usr/lib/libOpenCL.so.1.0.0</code>, <code class= "filename">/usr/lib/libnvidia-opencl.so.430.64</code>); the former is a vendor-independent Installable Client Driver (ICD) loader, and the latter is the NVIDIA Vendor ICD. A config file <code class="filename">/etc/OpenCL/vendors/nvidia.icd</code> is also installed, to advertise the NVIDIA Vendor ICD to the ICD Loader.</p> </li> <li> <p>The <code class="filename">nvidia-cuda-mps-control</code> and <code class="filename">nvidia-cuda-mps-server</code> applications, which allow MPI processes to run concurrently on a single GPU.</p> </li> <li> <p>A kernel module (<code class="filename">/lib/modules/`uname -r`/kernel/drivers/video/nvidia-modeset.ko</code>); this kernel module is responsible for programming the display engine of the GPU. User-mode NVIDIA driver components such as the NVIDIA X driver, OpenGL driver, and VDPAU driver communicate with nvidia-modeset.ko through the /dev/nvidia-modeset device file.</p> </li> <li> <p>A kernel module (<code class="filename">/lib/modules/`uname -r`/kernel/drivers/video/nvidia.ko</code>); this kernel module provides low-level access to your NVIDIA hardware for all of the above components. It is generally loaded into the kernel when the X server is started, and is used by the X driver and OpenGL. nvidia.ko consists of two pieces: the binary-only core, and a kernel interface that must be compiled specifically for your kernel version. Note that the Linux kernel does not have a consistent binary interface like the X server, so it is important that this kernel interface be matched with the version of the kernel that you are using. This can either be accomplished by compiling yourself, or using precompiled binaries provided for the kernels shipped with some of the more common Linux distributions.</p> </li> <li> <p>NVIDIA Unified Memory kernel module (<code class= "filename">/lib/modules/`uname -r`/kernel/drivers/video/nvidia-uvm.ko</code>); this kernel module provides functionality for sharing memory between the CPU and GPU in CUDA programs. It is generally loaded into the kernel when a CUDA program is started, and is used by the CUDA driver on supported platforms.</p> </li> <li> <p>The nvidia-tls library (<code class= "filename">/usr/lib/libnvidia-tls.so.430.64</code>); this file provides thread local storage support for the NVIDIA OpenGL libraries (libGLX_nvidia, libnvidia-glcore, and libglxserver_nvidia).</p> </li> <li> <p>The nvidia-ml library (<code class= "filename">/usr/lib/libnvidia-ml.so.430.64</code>); The NVIDIA Management Library provides a monitoring and management API. See <a href="nvidia-ml.html" title= "Chapter 26. The NVIDIA Management Library">Chapter 26, <i>The NVIDIA Management Library</i></a> for more information.</p> </li> <li> <p>The application nvidia-installer (<code class= "filename">/usr/bin/nvidia-installer</code>) is NVIDIA's tool for installing and updating NVIDIA drivers. See <a href= "installdriver.html" title= "Chapter 4. Installing the NVIDIA Driver">Chapter 4, <i>Installing the NVIDIA Driver</i></a> for a more thorough description.</p> <p>Source code is available at <a href= "https://download.nvidia.com/XFree86/nvidia-installer/" target= "_top">https://download.nvidia.com/XFree86/nvidia-installer/</a>.</p> </li> <li> <p>The application nvidia-modprobe (<code class= "filename">/usr/bin/nvidia-modprobe</code>) is installed as setuid root and is used to load the NVIDIA kernel module and create the <code class="filename">/dev/nvidia*</code> device nodes by processes (such as CUDA applications) that don't run with sufficient privileges to do those things themselves.</p> <p>Source code is available at <a href= "https://download.nvidia.com/XFree86/nvidia-modprobe/" target= "_top">https://download.nvidia.com/XFree86/nvidia-modprobe/</a>.</p> </li> <li> <p>The application nvidia-xconfig (<code class= "filename">/usr/bin/nvidia-xconfig</code>) is NVIDIA's tool for manipulating X server configuration files. See <a href= "editxconfig.html" title= "Chapter 6. Configuring X for the NVIDIA Driver">Chapter 6, <i>Configuring X for the NVIDIA Driver</i></a> for more information.</p> <p>Source code is available at <a href= "https://download.nvidia.com/XFree86/nvidia-xconfig/" target= "_top">https://download.nvidia.com/XFree86/nvidia-xconfig/</a>.</p> </li> <li> <p>The application nvidia-settings (<code class= "filename">/usr/bin/nvidia-settings</code>) is NVIDIA's tool for dynamic configuration while the X server is running. See <a href= "nvidiasettings.html" title= "Chapter 23. Using the nvidia-settings Utility">Chapter 23, <i>Using the nvidia-settings Utility</i></a> for more information.</p> </li> <li> <p>The libnvidia-gtk libraries (<code class= "filename">/usr/lib/libnvidia-gtk2.so.430.64</code> and on some platforms <code class= "filename">/usr/lib/libnvidia-gtk3.so.430.64</code>); these libraries are required to provide the nvidia-settings user interface.</p> <p>Source code is available at <a href= "https://download.nvidia.com/XFree86/nvidia-settings/" target= "_top">https://download.nvidia.com/XFree86/nvidia-settings/</a>.</p> </li> <li> <p>The application nvidia-smi (<code class= "filename">/usr/bin/nvidia-smi</code>) is the NVIDIA System Management Interface for management and monitoring functionality. See <a href="nvidia-smi.html" title= "Chapter 25. Using the nvidia-smi Utility">Chapter 25, <i>Using the nvidia-smi Utility</i></a> for more information.</p> </li> <li> <p>The application nvidia-debugdump (<code class= "filename">/usr/bin/nvidia-debugdump</code>) is NVIDIA's tool for collecting internal GPU state. It is normally invoked by the nvidia-bug-report.sh (<code class= "filename">/usr/bin/nvidia-bug-report.sh</code>) script. See <a href="nvidia-debugdump.html" title= "Chapter 27. Using the nvidia-debugdump Utility">Chapter 27, <i>Using the nvidia-debugdump Utility</i></a> for more information.</p> </li> <li> <p>The daemon nvidia-persistenced (<code class= "filename">/usr/bin/nvidia-persistenced</code>) is the NVIDIA Persistence Daemon for allowing the NVIDIA kernel module to maintain persistent state when no other NVIDIA driver components are running. See <a href="nvidia-persistenced.html" title= "Chapter 28. Using the nvidia-persistenced Utility">Chapter 28, <i>Using the nvidia-persistenced Utility</i></a> for more information.</p> <p>Source code is available at <a href= "https://download.nvidia.com/XFree86/nvidia-persistenced/" target= "_top">https://download.nvidia.com/XFree86/nvidia-persistenced/</a>.</p> </li> <li> <p>The NVCUVID library (<code class= "filename">/usr/lib/libnvcuvid.so.430.64</code>); The NVIDIA CUDA Video Decoder (NVCUVID) library provides an interface to hardware video decoding capabilities on NVIDIA GPUs with CUDA.</p> </li> <li> <p>The NvEncodeAPI library (<code class= "filename">/usr/lib/libnvidia-encode.so.430.64</code>); The NVENC Video Encoding library provides an interface to video encoder hardware on supported NVIDIA GPUs.</p> </li> <li> <p>The NvIFROpenGL library (<code class= "filename">/usr/lib/libnvidia-ifr.so.430.64</code>); The NVIDIA OpenGL-based Inband Frame Readback library provides an interface to capture and optionally encode an OpenGL framebuffer.</p> </li> <li> <p>The NvFBC library (<code class= "filename">/usr/lib/libnvidia-fbc.so.430.64</code>); The NVIDIA Framebuffer Capture library provides an interface to capture and optionally encode the framebuffer of an X server screen.</p> </li> <li> <p>An X driver configuration file (<code class= "filename">/usr/share/X11/xorg.conf.d/nvidia-drm-outputclass.conf</code>); If the X server is sufficiently new, this file will be installed to configure the X server to load the <code class= "filename">nvidia_drv.so</code> driver automatically if it is started after the NVIDIA DRM kernel module (nvidia-drm.ko) is loaded. This feature is supported in X.Org xserver 1.16 and higher when running on Linux kernel 3.13 or higher with CONFIG_DRM enabled.</p> </li> <li> <p>Predefined application profile keys and documentation for those keys can be found in the following files in the directory <code class="filename">/usr/share/nvidia/</code>: <code class= "filename">nvidia-application-profiles-430.64-rc</code>, <code class= "filename">nvidia-application-profiles-430.64-key-documentation</code>.</p> <p>See <a href="profiles.html" title= "Appendix J. Application Profiles">Appendix J, <i>Application Profiles</i></a> for more information.</p> </li> <li> <p>The OptiX library (<code class= "filename">/usr/lib/libnvoptix.so.1</code>); This library implements the OptiX ray tracing engine. It is loaded by the <code class="filename">liboptix.so.*</code> library bundled with applications that use the OptiX API.</p> </li> <li> <p>The NVIDIA Optical Flow library (<code class= "filename">/usr/lib/libnvidia-opticalflow.so.430.64</code>); The NVIDIA Optical Flow library can be used for hardware-accelerated computation of optical flow vectors and stereo disparity values on Turing and later NVIDIA GPUs. This is useful for some forms of computer vision and image analysis. The Optical Flow library depends on the NVCUVID library, which in turn depends on the CUDA library.</p> </li> </ul> </div> <p></p> <p>Problems will arise if applications use the wrong version of a library. This can be the case if there are either old libGL libraries or stale symlinks left lying around. If you think there may be something awry in your installation, check that the following files are in place (these are all the files of the NVIDIA Accelerated Linux Graphics Driver, as well as their symlinks):</p> <pre class="screen"> /usr/lib/xorg/modules/drivers/nvidia_drv.so /usr/lib/xorg/modules/extensions/libglxserver_nvidia.so.430.64 /usr/lib/xorg/modules/extensions/libglxserver_nvidia.so -> libglxserver_nvidia.so.430.64 /usr/lib/libGL.so.430.64 /usr/lib/libGL.so.1 -> libGL.so.430.64 /usr/lib/libGL.so -> libGL.so.1 (on GLVND-based installations, libGL.so.1 from GLVND will be used instead of libGL.so.430.64 as shown above.) /usr/lib/libnvidia-glcore.so.430.64 /usr/lib/libcuda.so.430.64 /usr/lib/libcuda.so -> libcuda.so.430.64 /lib/modules/`uname -r`/video/nvidia.{o,ko}, or /lib/modules/`uname -r`/kernel/drivers/video/nvidia.{o,ko} </pre> <p>If there are other libraries whose "soname" conflicts with that of the NVIDIA libraries, ldconfig may create the wrong symlinks. It is recommended that you manually remove or rename conflicting libraries (be sure to rename clashing libraries to something that ldconfig will not look at -- we have found that prepending "XXX" to a library name generally does the trick), rerun 'ldconfig', and check that the correct symlinks were made. An example of a library that often creates conflicts is "/usr/lib/mesa/libGL.so*".</p> <p>If the libraries appear to be correct, then verify that the application is using the correct libraries. For example, to check that the application /usr/bin/glxgears is using the NVIDIA libraries, run:</p> <pre class="screen"> % ldd /usr/bin/glxgears linux-gate.so.1 => (0xffffe000) libGL.so.1 => /usr/lib/libGL.so.1 (0xb7ed1000) libXext.so.6 => /usr/lib/libXext.so.6 (0xb7ec0000) libX11.so.6 => /usr/lib/libX11.so.6 (0xb7de0000) libpthread.so.0 => /lib/tls/libpthread.so.0 (0x00946000) libm.so.6 => /lib/tls/libm.so.6 (0x0075d000) libc.so.6 => /lib/tls/libc.so.6 (0x00631000) libnvidia-tls.so.430.64 => /usr/lib/tls/libnvidia-tls.so.430.64 (0xb7ddd000) libnvidia-glcore.so.430.64 => /usr/lib/libnvidia-glcore.so.430.64 (0xb5d1f000) libdl.so.2 => /lib/libdl.so.2 (0x00782000) /lib/ld-linux.so.2 (0x00614000) </pre> <p>In the example above, the list of libraries reported by <span><strong class="command">ldd</strong></span> includes <code class="filename">libnvidia-tls.so.430.64</code> and <code class="filename">libnvidia-glcore.so.430.64</code>: this is because <span><strong class="command">glxgears</strong></span> links <code class="filename">libGL.so.1</code>, which in this case is the legacy, non-GLVND NVIDIA GLX client library. When <code class="filename">libGL.so.1</code> is provided by GLVND instead, <code class="filename">libGLX.so.0</code> and <code class= "filename">libGLdispatch.so.0</code> should appear in the output of <span><strong class="command">ldd</strong></span>. If the GLX client library is something other than the NVIDIA or GLVND <code class="filename">libGL.so.1</code>, then you will need to either remove the library that is getting in the way or adjust your dynamic loader search path using the <code class= "envar">LD_LIBRARY_PATH</code> environment variable. You may want to consult the man pages for <span><strong class= "command">ldconfig</strong></span> and <span><strong class= "command">ldd</strong></span>.</p> </div> <div class="navfooter"> <hr> <table width="100%" summary="Navigation footer"> <tr> <td width="40%" align="left"><a accesskey="p" href= "installdriver.html">Prev</a> </td> <td width="20%" align="center"><a accesskey="u" href= "installationandconfiguration.html">Up</a></td> <td width="40%" align="right"> <a accesskey="n" href= "editxconfig.html">Next</a></td> </tr> <tr> <td width="40%" align="left" valign="top"> Chapter 4. Installing the NVIDIA Driver </td> <td width="20%" align="center"><a accesskey="h" href= "index.html">Home</a></td> <td width="40%" align="right" valign="top"> Chapter 6. Configuring X for the NVIDIA Driver</td> </tr> </table> </div> </body> </html>