xpra icon
Bug tracker and wiki

This bug tracker and wiki are being discontinued
please use https://github.com/Xpra-org/xpra instead.


Changes between Version 37 and Version 38 of Encodings/nvenc


Ignore:
Timestamp:
03/30/15 07:50:20 (6 years ago)
Author:
Antoine Martin
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • Encodings/nvenc

    v37 v38  
    1111This encoder requires a [https://developer.nvidia.com/nvidia-video-codec-sdk#gpulist supported NVIDIA graphics card]:
    1212* Tesla onwards cards: Quadro K4000 and up, .. (no license key required)
    13 * Some consumer cards (ie: GTX 680, GTX 750 and up for sure) and a license key
     13* Some consumer cards (ie: GTX 680, GTX 750 and up for sure) and a license key (best with SDK version 4 or older)
    1414}}}
    1515
     
    1818== Software Requirements ==
    1919You must have [http://mathema.tician.de/software/pycuda/ PyCUDA] installed, and a recent enough version of the nvidia drivers.
    20 How you install those drivers is entirely up to you: [http://www.nvidia.com/object/unix.html nvidia installers], [http://negativo17.org/nvidia-driver/ negativo17 repository], [http://rpmfusion.org/ rpmfusion], [http://elrepo.org/tiki/tiki-index.php elrepo], etc..
     20How you install those drivers is entirely up to you:
     21* [http://www.nvidia.com/object/unix.html nvidia installers]
     22* [http://negativo17.org/nvidia-driver/ negativo17 repository]
     23* [http://rpmfusion.org/ rpmfusion]
     24* [http://elrepo.org/tiki/tiki-index.php elrepo]
     25etc..
    2126
    2227If your CUDA ({{{libcuda.so}}}) or NVENC ({{{libnvidia-encode.so}}}) libraries are installed in an unusual location, it is your responsibility to ensure they can be loaded at runtime, usually by adding the directory to the {{{LD_LIBRARY_PATH}}}.
    2328
    24 Note: xpra version 0.15 or newer is required. (older versions require the full CUDA SDK at runtime)
     29Note: xpra version 0.15 or newer is required. (older versions require the full CUDA SDK at runtime and contain unfixable bugs)
    2530}}}
    2631
     
    6772{{{#!div class="box"
    6873== Building ==
    69 * Download the [https://developer.nvidia.com/cuda-downloads CUDA SDK] and install it. If present, you should remove any previously installed nvidia drivers: both "nouveau" and nvidia's proprietary drivers - either install the drivers bundled with CUDA or a sufficiently recent version, preferably directly from nvidia (ie: {{{331.20}}}, {{{331.49}}} and {{{334.21}}} are known to work with the version 3 SDK - {{{337.12}}} probably requires SDK version 4 or different keys, {{{331.79}}} also works but may require newer license keys, more information here: #595)
    70 * Download the [/attachment/wiki/Encodings/nvenc/cuda.pc cuda.pc] pkgconfig file (missing from the SDK) and install it (usually in {{{/usr/lib64/pkgconfig}}} for 64-bit systems)
     74* Download the [https://developer.nvidia.com/cuda-downloads CUDA SDK] and install it. If present, you should remove any previously installed nvidia drivers: both "nouveau" and nvidia's proprietary drivers - either install the drivers bundled with CUDA or a sufficiently recent version, preferably directly from nvidia (ie: {{{331.20}}}, {{{331.49}}} and {{{334.21}}} are known to work with the version 3 SDK - {{{331.79}}} and {{{337.12}}} require a different set of license keys, more information here: #595)
     75* Download the [/attachment/wiki/Encodings/nvenc/cuda.pc cuda.pc] pkgconfig file (missing from the SDK) and install it (usually {{{/usr/lib/pkgconfig}}} or {{{/usr/lib64/pkgconfig}}} on some 64-bit systems)
    7176* [http://wiki.tiker.net/PyCuda/Installation/Linux Install PyCuda]
    72 * Download the latest [https://developer.nvidia.com/nvidia-video-codec-sdk NVENC SDK] (aka "NVIDIA VIDEO CODEC SDK") and install it (just unzip into {{{/usr/local/}}} - then create a symlink named just {{{nvenc}}})
    73 * Download the [/attachment/wiki/Encodings/nvenc/nvenc.pc nvenc.pc] pkgconfig file (also missing from the SDK) and install it
    74 * Build xpra version with nvenc support:
     77* Download the [https://developer.nvidia.com/nvidia-video-codec-sdk NVENC SDK], aka "NVIDIA VIDEO CODEC SDK" and install it (just unzip into {{{/opt/}}}), then create a symlink named just {{{/opt/nvencV}}} (where V is the SDK version, ie: {{{nvenc4}}}). You may want to stick with SDK versions 3 or 4 if you plan on using consumer cards with a license key, as the SDK version 5 removes the ability to use keys (see #825).
     78* Download the pkgconfig file matching your SDK version (ie: [/attachment/wiki/Encodings/nvenc/nvenc4.pc nvenc4.pc]) and install it (same location as the {{{cuda.pc}}})
     79* When building xpra, nvenc support should be auto-detected, but you can try forcing it to verify, ie: for version 4:
    7580{{{
    76 ./setup.py install --with-nvenc
     81./setup.py install --with-nvenc4
    7782}}}
    7883
    7984Notes:
    80 * the files given here are for the current versions of the {{{SDK}}}s and for 64 bit systems only, adjust the files and locations accordingly
     85* you may need to adjust some paths
    8186* If CUDA refuses to build and complains about:
    8287{{{