README.md

% Embree: High Performance Ray Tracing Kernels 4.3.2
% Intel Corporation

Intel® Embree Overview
======================

Intel® Embree is a high-performance ray tracing library developed at
Intel, which is released as open source under the [Apache 2.0
license](http://www.apache.org/licenses/LICENSE-2.0). Intel® Embree
supports x86 CPUs under Linux, macOS, and Windows; ARM CPUs on Linux
and macOS; as well as Intel® GPUs under Linux and Windows.

Intel® Embree targets graphics application developers to improve the
performance of photo-realistic rendering applications. Embree is
optimized towards production rendering, by putting focus on incoherent
ray performance, high quality acceleration structure construction, a
rich feature set, accurate primitive intersection, and low memory
consumption.

Embree's feature set includes various primitive types such as
triangles (as well quad and grids for lower memory consumption);
Catmull-Clark subdivision surfaces; various types of curve primitives,
such as flat curves (for distant views), round curves (for closeup
views), and normal oriented curves, all supported with different basis
functions (linear, Bézier, B-spline, Hermite, and Catmull Rom);
point-like primitives, such as ray oriented discs, normal oriented
discs, and spheres; user defined geometries with a procedural
intersection function; multi-level instancing; filter callbacks
invoked for any hit encountered; motion blur including multi-segment
motion blur, deformation blur, and quaternion motion blur; and ray
masking.

Intel® Embree contains ray tracing kernels optimized for the latest
x86 processors with support for SSE, AVX, AVX2, and AVX-512
instructions, and uses runtime code selection to choose between these
kernels. Intel® Embree contains algorithms optimized for incoherent
workloads (e.g.  Monte Carlo ray tracing algorithms) and coherent
workloads (e.g. primary visibility and hard shadow rays) as well as
supports for dynamic scenes by implementing high-performance two-level
spatial index structure construction algorithms.

Intel® Embree supports applications written with the Intel® Implicit
SPMD Program Compiler (Intel® ISPC, <https://ispc.github.io/>) by
providing an ISPC interface to the core ray tracing
algorithms. This makes it possible to write a renderer that
automatically vectorizes and leverages SSE, AVX, AVX2, and AVX-512
instructions.

Intel® Embree supports Intel GPUs through the
[SYCL](https://www.khronos.org/sycl/) open standard programming
language. SYCL allows to write C++ code that can be run on various
devices, such as CPUs and GPUs. Using Intel® Embree application
developers can write a single source renderer that executes
efficiently on CPUs and GPUs. Maintaining just one code base
this way can significantly improve productivity and eliminate
inconsistencies between a CPU and GPU version of the renderer. Embree
supports GPUs based on the Xe HPG and Xe HPC microarchitecture,
which support hardware accelerated ray tracing do deliver excellent
levels of ray tracing performance.

Supported Platforms
-------------------

Embree supports Windows (32-bit and 64-bit), Linux (64-bit), and macOS
(64-bit). Under Windows, Linux and macOS x86 based CPUs are supported,
while ARM CPUs are currently only supported under Linux and macOS (e.g. 
Apple M1). ARM support for Windows experimental.

Embree supports Intel GPUs based on the Xe HPG microarchitecture
(Intel® Arc™ GPU) under Linux and Windows and Xe HPC microarchitecture
(Intel® Data Center GPU Flex Series and Intel® Data Center GPU Max
Series) under Linux.

The code compiles with the Intel® Compiler, Intel® oneAPI DPC++
Compiler, GCC, Clang, and the Microsoft Compiler. To use Embree on the
GPU the Intel® oneAPI DPC++ Compiler must be used. Please see section
[Compiling Embree] for details on tested compiler versions.

Embree requires at least an x86 CPU with support for
SSE2 or an Apple M1 CPU.

Embree Support and Contact
--------------------------

If you encounter bugs please report them via [Embree's GitHub Issue
Tracker](https://github.com/embree/embree/issues).

For questions and feature requests please write us at
<embree_support@intel.com>.

To receive notifications of updates and new features of Embree please
subscribe to the [Embree mailing
list](https://groups.google.com/d/forum/embree/).

Installation of Embree
======================


Windows Installation
--------------------

A pre-built version of Embree for Windows is provided as a ZIP archive
[embree-4.3.2.x64.windows.zip](https://github.com/embree/embree/releases/download/v4.3.2/embree-4.3.2.x64.windows.zip). After
unpacking this ZIP file, you should set the path to the `lib` folder
manually to your `PATH` environment variable for applications to find
Embree.


Linux Installation
------------------

A pre-built version of Embree for Linux is provided as a `tar.gz` archive:
[embree-4.3.2.x86_64.linux.tar.gz](https://github.com/embree/embree/releases/download/v4.3.2/embree-4.3.2.x86_64.linux.tar.gz). Unpack
this file using `tar` and source the provided `embree-vars.sh` (if you
are using the bash shell) or `embree-vars.csh` (if you are using the C
shell) to set up the environment properly:

    tar xzf embree-4.3.2.x86_64.linux.tar.gz
    source embree-4.3.2.x86_64.linux/embree-vars.sh

We recommend adding a relative `RPATH` to your application that points
to the location where Embree (and TBB) can be found, e.g. `$ORIGIN/../lib`.


macOS Installation
------------------

The macOS version of Embree is also delivered as a ZIP file:
[embree-4.3.2.x86_64.macosx.zip](https://github.com/embree/embree/releases/download/v4.3.2/embree-4.3.2.x86_64.macosx.zip). Unpack
this file using `tar` and source the provided `embree-vars.sh` (if you
are using the bash shell) or `embree-vars.csh` (if you are using the C
shell) to set up the environment properly:

    unzip embree-4.3.2.x64.macosx.zip    source embree-4.3.2.x64.macosx/embree-vars.sh

If you want to ship Embree with your application, please use the Embree
library of the provided ZIP file. The library name of that Embree
library is of the form `@rpath/libembree.4.dylib`
(and similar also for the included TBB library). This ensures that you
can add a relative `RPATH` to your application that points to the location
where Embree (and TBB) can be found, e.g. `@loader_path/../lib`.


Building Embree Applications
----------------------------

The most convenient way to build an Embree application is through
CMake. Just let CMake find your unpacked Embree package using the
`FIND_PACKAGE` function inside your `CMakeLists.txt` file:

     FIND_PACKAGE(embree 4 REQUIRED)

For CMake to properly find Embree you need to set the `embree_DIR` variable to
the folder containing the `embree_config.cmake` file. You might also have to
set the `TBB_DIR` variable to the path containing `TBB-config.cmake` of a local
TBB install, in case you do not have TBB installed globally on your system,
e.g:

    cmake -D embree_DIR=path_to_embree_package/lib/cmake/embree-4.3.2/ \
          -D TBB_DIR=path_to_tbb_package/lib/cmake/tbb/ \
          ..

The `FIND_PACKAGE` function will create an `embree` target that
you can add to your target link libraries:

    TARGET_LINK_LIBRARIES(application embree)

For a full example on how to build an Embree application please have a
look at the `minimal` tutorial provided in the `src` folder of the
Embree package and also the contained `README.txt` file.


Building Embree SYCL Applications
----------------------------------

Building Embree SYCL applications is also best done using
CMake. Please first get some compatible SYCL compiler and setup the
environment as decribed in sections [Linux SYCL Compilation] and
[Windows SYCL Compilation].

Also perform the setup steps from the previous [Building Embree
Applications] section.

Please also have a look at the [Minimal] tutorial that is provided
with the Embree release, for an example how to build a simple SYCL
application using CMake and Embree.

To properly compile your SYCL application you have to add additional
SYCL compile flags for each C++ file that contains SYCL device side
code or kernels as described next.


### JIT Compilation

We recommend using just in time compilation (JIT compilation) together
with [SYCL JIT caching] to compile Embree SYCL applications. For JIT
compilation add these options to the compilation phase of all C++
files that contain SYCL code:

    -fsycl -Xclang -fsycl-allow-func-ptr -fsycl-targets=spir64

These options enable SYCL two phase compilation (`-fsycl` option),
enable function pointer support (`-Xclang -fsycl-allow-func-ptr`
option), and just in time (JIT) compilation only
(`-fsycl-targets=spir64` option).

The following link options have to get added to the linking stage of
your application when using just in time compilation:

    -fsycl -fsycl-targets=spir64

For a full example on how to build an Embree SYCL application please
have a look at the SYCL version of the `minimal` tutorial provided in
the `src` folder of the Embree package and also the contained
`README.txt` file.

Please have a look at the [Compiling Embree] section on how to create
an Embree package from sources if required.


### AOT Compilation

Ahead of time compilation (AOT compilation) allows to speed up first
application start up time as device binaries are precompiled. We do
not recommend using AOT compilation as it does not allow the usage of
specialization constants to reduce code complexity.

For ahead of time compilation add these compile options to the
compilation phase of all C++ files that contain SYCL code:

    -fsycl -Xclang -fsycl-allow-func-ptr -fsycl-targets=spir64_gen

These options enable SYCL two phase compilation (`-fsycl` option),
enable function pointer support (`-Xclang -fsycl-allow-func-ptr`
option), and ahead of time (AOT) compilation
(`-fsycl-targets=spir64_gen` option).

The following link options have to get added to the linking stage of
your application when compiling ahead of time for Xe HPG devices:

    -fsycl -fsycl-targets=spir64_gen
    -Xsycl-target-backend=spir64_gen "-device XE_HPG_CORE"

This in particular configures the devices for AOT compilation to
`XE_HPG_CORE`.

To get a list of all device supported by AOT compilation look at the
help of the device option in ocloc tool:

    ocloc compile --help


Building Embree Tests
---------------------

Embree is released with a bundle of tests in an optional testing package.
To run these tests extract the testing package in the same folder as your embree installation.
e.g.:
    
    tar -xzf embree-4.3.2-testing.zip -C /path/to/installed/embree

The tests are extracted into a new folder inside you embree installation and can be run with:

    cd /path/to/installed/embree/testing
    cmake -B build
    cmake --build build target=tests


Compiling Embree
================

We recommend using the prebuild Embree packages from
[https://github.com/embree/embree/releases](https://github.com/embree/embree/releases). If
you need to compile Embree yourself you need to use CMake as described
in the following.

Do not enable fast-math optimizations in your compiler as this mode is
not supported by Embree.

Linux and macOS
---------------

To compile Embree you need a modern C++ compiler that supports
C++11. Embree is tested with the following compilers:

Linux

  - Intel® oneAPI DPC++/C++ Compiler 2024.0.2
  - oneAPI DPC++/C++ Compiler 2023-10-26
  - Clang 5.0.0
  - Clang 4.0.0
  - GCC 10.0.1 (Fedora 32) AVX512 support
  - GCC  8.3.1 (Fedora 29) AVX512 support
  - Intel® Implicit SPMD Program Compiler 1.22.0

macOS x86_64

  - Apple Clang 15

macOS Arm64

  - Apple Clang 14

Embree supports using the Intel® Threading Building Blocks (TBB) as the
tasking system. For performance and flexibility reasons we recommend
using Embree with the Intel® Threading Building Blocks (TBB) and best
also use TBB inside your application. Optionally you can disable TBB
in Embree through the `EMBREE_TASKING_SYSTEM` CMake variable.

Embree supports the Intel® Implicit SPMD Program Compiler (Intel® ISPC), which allows
straightforward parallelization of an entire renderer. If you
want to use Intel® ISPC then you can enable `EMBREE_ISPC_SUPPORT` in
CMake. Download and install the Intel® ISPC binaries from
[ispc.github.io](https://ispc.github.io/downloads.html). After
installation, put the path to `ispc` permanently into your `PATH` environment
variable or you set the `EMBREE_ISPC_EXECUTABLE` variable to point at the ISPC
executable during CMake configuration.

You additionally have to install CMake 3.1.0 or higher and the developer
version of [GLFW](https://www.glfw.org/) version 3.

Under macOS, all these dependencies can be installed
using [MacPorts](http://www.macports.org/):

    sudo port install cmake tbb glfw-devel

Depending on your Linux distribution you can install these dependencies
using `yum` or `apt-get`.  Some of these packages might already be
installed or might have slightly different names.

Type the following to install the dependencies using `yum`:

    sudo yum install cmake
    sudo yum install tbb-devel
    sudo yum install glfw-devel

Type the following to install the dependencies using `apt-get`:

    sudo apt-get install cmake-curses-gui
    sudo apt-get install libtbb-dev
    sudo apt-get install libglfw3-dev

Finally, you can compile Embree using CMake. Create a build directory
inside the Embree root directory and execute `ccmake ..` inside this
build directory.

    mkdir build
    cd build
    ccmake ..

Per default, CMake will use the compilers specified with the `CC` and
`CXX` environment variables. Should you want to use a different
compiler, run `cmake` first and set the `CMAKE_CXX_COMPILER` and
`CMAKE_C_COMPILER` variables to the desired compiler. For example, to
use the Clang compiler instead of the default GCC on most Linux machines
(`g++` and `gcc`), execute

    cmake -DCMAKE_CXX_COMPILER=clang++ -DCMAKE_C_COMPILER=clang ..

Running `ccmake` will open a dialog where you can perform various
configurations as described below in [CMake Configuration]. After having
configured Embree, press `c` (for configure) and `g` (for generate) to
generate a Makefile and leave the configuration. The code can be
compiled by executing make.

    make -j 8

The executables will be generated inside the build folder. We recommend
installing the Embree library and header files on your
system. Therefore set the `CMAKE_INSTALL_PREFIX` to `/usr` in cmake
and type:

    sudo make install

If you keep the default `CMAKE_INSTALL_PREFIX` of `/usr/local` then
you have to make sure the path `/usr/local/lib` is in your
`LD_LIBRARY_PATH`.

You can also uninstall Embree again by executing:

    sudo make uninstall

You can also create an Embree package using the following command:

    make package

Please see the [Building Embree Applications] section on how to build
your application with such an Embree package.

Linux SYCL Compilation
-----------------------

There are two options to compile Embree with SYCL support:
The open source ["oneAPI DPC++ Compiler"](https://github.com/intel/llvm/) or
the ["Intel(R) oneAPI DPC++/C++ Compiler"](https://www.intel.com/content/www/us/en/developer/articles/tool/oneapi-standalone-components.html#dpcpp-cpp).
Other SYCL compilers are not supported.

The "oneAPI DPC++ Compiler" is more up-to-date than the "Intel(R) oneAPI
DPC++/C++ Compiler" but less stable. The current tested version of the "oneAPI
DPC++ compiler is

  - [oneAPI DPC++ Compiler 2023-10-26](https://github.com/intel/llvm/releases/tag/nightly-2023-10-26)
  
The compiler can be downloaded and simply extracted. The oneAPI DPC++ compiler
can be set up executing the following commands in a Linux (bash) shell:

    export SYCL_BUNDLE_ROOT=path_to_dpcpp_compiler
    export PATH=$SYCL_BUNDLE_ROOT/bin:$PATH
    export CPATH=$SYCL_BUNDLE_ROOT/include:$CPATH
    export LIBRARY_PATH=$SYCL_BUNDLE_ROOT/lib:$LIBRARY_PATH
    export LD_LIBRARY_PATH=$SYCL_BUNDLE_ROOT/lib:$LD_LIBRARY_PATH
    export LD_LIBRARY_PATH=$SYCL_BUNDLE_ROOT/linux/lib/x64:$LD_LIBRARY_PATH

where the `path_to_dpcpp_compiler` should point to the unpacked oneAPI DPC++
compiler. This will put `clang++` and `clang` from the oneAPI DPC++ Compiler
into your path.

Please also install all Linux packages described in the previous
section.

Now, you can configure Embree using CMake by executing the following command
in the Embree root directory:

    cmake -B build \
          -DCMAKE_CXX_COMPILER=clang++ \
          -DCMAKE_C_COMPILER=clang \
          -DEMBREE_SYCL_SUPPORT=ON

This will create a directory `build` to use as the CMake build directory,
configure the usage of the oneAPI DPC++ Compiler, and turn on SYCL support
through `EMBREE_SYCL_SUPPORT=ON`.

Alternatively, you can download and run the installer of the

 - [Intel(R) oneAPI DPC++/C++ Compiler](https://www.intel.com/content/www/us/en/developer/articles/tool/oneapi-standalone-components.html#dpcpp-cpp).

After installation, you can set up the compiler by sourcing the
`vars.sh` script in the `env` directory of the compiler install directory, for example,

    source /opt/intel/oneAPI/compiler/latest/env/vars.sh

This script will put the `icpx` and `icx` compiler executables from the
Intel(R) oneAPI DPC++/C++ Compiler in your path.

Now, you can configure Embree using CMake by executing the following command
in the Embree root directory:

    cmake -B build \
          -DCMAKE_CXX_COMPILER=icpx \
          -DCMAKE_C_COMPILER=icx \
          -DEMBREE_SYCL_SUPPORT=ON

More information about setting up the Intel(R) oneAPI DPC++/C++ compiler can be
found in the [Development Reference Guide](https://www.intel.com/content/www/us/en/develop/documentation/oneapi-dpcpp-cpp-compiler-dev-guide-and-reference/top/compiler-setup.html). Please note, that the Intel(R) oneAPI DPC++/C++ compiler
requires [at least CMake version 3.20.5 on Linux](https://www.intel.com/content/www/us/en/develop/documentation/oneapi-dpcpp-cpp-compiler-dev-guide-and-reference/top/compiler-setup/use-the-command-line/use-cmake-with-the-compiler.html).

Independent of the DPC++ compiler choice, you can now build Embree using

    cmake --build build -j 8

The executables will be generated inside the build folder. The
executable names of the SYCL versions of the tutorials end with
`_sycl`.


### Linux Graphics Driver Installation

To run the SYCL code you need to install the latest GPGPU drivers for
your Intel Xe HPG/HPC GPUs from here
[https://dgpu-docs.intel.com/](https://dgpu-docs.intel.com/). Follow
the driver installation instructions for your graphics card and
operating system.


Windows
-------

Embree is tested using the following compilers under Windows:

  - Intel® oneAPI DPC++/C++ Compiler 2024.0.2
  - oneAPI DPC++/C++ Compiler 2023-10-26
  - Visual Studio 2022
  - Visual Studio 2019
  - Visual Studio 2017
  - Intel® Implicit SPMD Program Compiler 1.22.0

To compile Embree for AVX-512 you have to use the Intel® Compiler.

Embree supports using the Intel® Threading Building Blocks (TBB) as the
tasking system. For performance and flexibility reasons we recommend
using use Embree with the Intel® Threading Building Blocks (TBB) and best
also use TBB inside your application. Optionally you can disable TBB
in Embree through the `EMBREE_TASKING_SYSTEM` CMake variable.

Embree will either find the Intel® Threading Building Blocks (TBB)
installation that comes with the Intel® Compiler, or you can install the
binary distribution of TBB directly from
[https://github.com/oneapi-src/oneTBB/releases](https://github.com/oneapi-src/oneTBB/releases)
into a folder named `tbb` into your Embree root directory. You also have
to make sure that the libraries `tbb.dll` and `tbb_malloc.dll` can be
found when executing your Embree applications, e.g. by putting the path
to these libraries into your `PATH` environment variable.

Embree supports the Intel® Implicit SPMD Program Compiler (Intel® ISPC), which
allows straightforward parallelization of an entire renderer. When installing
Intel® ISPC, make sure to download an Intel® ISPC version from
[ispc.github.io](https://ispc.github.io/downloads.html) that is compatible with
your Visual Studio version. After installation, put the path to `ispc.exe`
permanently into your `PATH` environment variable or you need to correctly set
the `EMBREE_ISPC_EXECUTABLE` variable during CMake configuration to point to
the ISPC executable. If you want to use Intel® ISPC, you have to enable
`EMBREE_ISPC_SUPPORT` in CMake.

You additionally have to install [CMake](http://www.cmake.org/download/)
(version 3.1 or higher). Note that you need a native Windows CMake
installation because CMake under Cygwin cannot generate solution files
for Visual Studio.

### Using the IDE

Run `cmake-gui`, browse to the Embree sources, set the build directory
and click Configure. Now you can select the Generator, e.g. "Visual
Studio 12 2013" for a 32-bit build or "Visual Studio 12 2013 Win64"
for a 64-bit build.

To use a different compiler than the Microsoft Visual C++ compiler, you
additionally need to specify the proper compiler toolset through the
option "Optional toolset to use (-T parameter)". E.g. to use Clang for
compilation set the toolset to "LLVM_v142".

Do not change the toolset manually in a solution file (neither through
the project properties dialog nor through the "Use Intel Compiler"
project context menu), because then some compiler-specific command line
options cannot be set by CMake.

Most configuration parameters described in the [CMake Configuration]
can be set under Windows as well. Finally, click "Generate" to create
the Visual Studio solution files. 

The following CMake options are only available under Windows:

+ `CMAKE_CONFIGURATION_TYPE`:  List of generated
  configurations. The default value is Debug;Release;RelWithDebInfo.

+  `USE_STATIC_RUNTIME`: Use the static version of the C/C++ runtime
  library. This option is turned OFF by default.

Use the generated Visual Studio solution file `embree4.sln` to compile
the project.

We recommend enabling syntax highlighting for the `.ispc` source and
`.isph` header files. To do so open Visual Studio, go to Tools ⇒
Options ⇒ Text Editor ⇒ File Extension and add the `isph` and `ispc`
extensions for the "Microsoft Visual C++" editor.

### Using the Command Line

Embree can also be configured and built without the IDE using the Visual
Studio command prompt:

    cd path\to\embree
    mkdir build
    cd build
    cmake -G "Visual Studio 16 2019" ..
    cmake --build . --config Release

You can also build only some projects with the `--target` switch.
Additional parameters after "`--`" will be passed to `msbuild`. For
example, to build the Embree library in parallel use

    cmake --build . --config Release --target embree -- /m

### Building Embree - Using vcpkg

You can download and install Embree using the [vcpkg](https://github.com/Microsoft/vcpkg) dependency manager:

    git clone https://github.com/Microsoft/vcpkg.git
    cd vcpkg
    ./bootstrap-vcpkg.sh
    ./vcpkg integrate install
    ./vcpkg install embree3

The Embree port in vcpkg is kept up to date by Microsoft team members
and community contributors. If the version is out of date, please
[create an issue or pull request](https://github.com/Microsoft/vcpkg)
on the vcpkg repository.


Windows SYCL Compilation
-------------------------

There are two options to compile Embree with SYCL support:
The open source ["oneAPI DPC++ Compiler"](https://github.com/intel/llvm/) or
the ["Intel(R) oneAPI DPC++/C++ Compiler"](https://www.intel.com/content/www/us/en/developer/articles/tool/oneapi-standalone-components.html#dpcpp-cpp).
Other SYCL compilers are not supported. You will also need an installed version
of Visual Studio that supports the C++17 standard, e.g. Visual Studio 2019.

The "oneAPI DPC++ Compiler" is more up-to-date than the "Intel(R) oneAPI
DPC++/C++ Compiler" but less stable. The current tested version of the oneAPI
DPC++ compiler is

  - [oneAPI DPC++ Compiler 2023-10-26](https://github.com/intel/llvm/releases/tag/nightly-2023-10-26)

Download and unpack the archive and open the "x64 Native Tools Command Prompt"
of Visual Studio and execute the following lines to properly configure the
environment to use the oneAPI DPC++ compiler:

    set "DPCPP_DIR=path_to_dpcpp_compiler"
    set "PATH=%DPCPP_DIR%\bin;%PATH%"
    set "PATH=%DPCPP_DIR%\lib;%PATH%"
    set "CPATH=%DPCPP_DIR%\include;%CPATH%"
    set "INCLUDE=%DPCPP_DIR%\include;%INCLUDE%"
    set "LIB=%DPCPP_DIR%\lib;%LIB%"

The `path_to_dpcpp_compiler` should point to the unpacked oneAPI DPC++
compiler.

Now, you can configure Embree using CMake by executing the following command
in the Embree root directory:

    cmake -B build
          -G Ninja
          -D CMAKE_BUILD_TYPE=Release
          -D CMAKE_CXX_COMPILER=clang++
          -D CMAKE_C_COMPILER=clang
          -D EMBREE_SYCL_SUPPORT=ON
          -D TBB_ROOT=path_to_tbb\lib\cmake\tbb

This will create a directory `build` to use as the CMake build directory, and
configure a release build that uses `clang++` and `clang` from the oneAPI DPC++
compiler.

The [Ninja](https://ninja-build.org/) generator is currently the easiest way to
use the oneAPI DPC++ compiler.

We also enable SYCL support in Embree using the `EMBREE_SYCL_SUPPORT` CMake
option.

Alternatively, you can download and run the installer of the

 - [Intel(R) oneAPI DPC++/C++ Compiler](https://www.intel.com/content/www/us/en/developer/articles/tool/oneapi-standalone-components.html#dpcpp-cpp).

After installation, you can either open a regular `Command Prompt` and execute
the `vars.bat` script in the `env` directory of the compiler install directory,
for example

    C:\Program Files (x86)\Intel\oneAPI\compiler\latest\env\vars.bat

or simply open the installed "Intel oneAPI command prompt for Intel 64 for Visual Studio".

Both ways will put the `icx` compiler executable from the
Intel(R) oneAPI DPC++/C++ compiler in your path.

Now, you can configure Embree using CMake by executing the following command
in the Embree root directory:

    cmake -B build
          -G Ninja
          -D CMAKE_BUILD_TYPE=Release
          -D CMAKE_CXX_COMPILER=icx
          -D CMAKE_C_COMPILER=icx
          -D EMBREE_SYCL_SUPPORT=ON
          -D TBB_ROOT=path_to_tbb\lib\cmake\tbb

More information about setting up the Intel(R) oneAPI DPC++/C++ compiler can be
found in the [Development Reference Guide](https://www.intel.com/content/www/us/en/develop/documentation/oneapi-dpcpp-cpp-compiler-dev-guide-and-reference/top/compiler-setup.html). Please note, that the Intel(R) oneAPI DPC++/C++ compiler
requires [at least CMake version 3.23 on Windows](https://www.intel.com/content/www/us/en/develop/documentation/oneapi-dpcpp-cpp-compiler-dev-guide-and-reference/top/compiler-setup/use-the-command-line/use-cmake-with-the-compiler.html).

Independent of the DPC++ compiler choice, you can now build Embree using

    cmake --build build

If you have problems with Ninja re-running CMake in an infinite loop,
then first remove the "Re-run CMake if any of its inputs changed."
section from the `build.ninja` file and run the above command again.

You can also create an Embree package using the following command:

    cmake --build build --target package

Please see the [Building Embree SYCL Applications] section on how to build
your application with such an Embree package.


### Windows Graphics Driver Installation

In order to run the SYCL tutorials on HPG hardware, you first need to
install the graphics drivers for your graphics card from
[https://www.intel.com](https://www.intel.com). Please make sure to
have installed version 31.0.101.4644 or newer.


CMake Configuration
-------------------

The default CMake configuration in the configuration dialog should be
appropriate for most usages. The following list describes all
parameters that can be configured in CMake:

+ `CMAKE_BUILD_TYPE`: Can be used to switch between Debug mode
  (Debug), Release mode (Release) (default), and Release mode with
  enabled assertions and debug symbols (RelWithDebInfo).

+ `EMBREE_STACK_PROTECTOR`: Enables protection of return address
  from buffer overwrites. This option is OFF by default.

+ `EMBREE_ISPC_SUPPORT`: Enables Intel® ISPC support of Embree. This option
  is OFF by default.

+ `EMBREE_SYCL_SUPPORT`: Enables GPU support using SYCL. When this
  option is enabled you have to use some DPC++ compiler. Please see
  the sections [Linux SYCL Compilation] and [Windows SYCL Compilation]
  on supported DPC++ compilers. This option is OFF by default.

+ `EMBREE_SYCL_AOT_DEVICES`: Selects a list of GPU devices for
  ahead-of-time (AOT) compilation of device code. Possible values are
  either, "none" which enables only just in time (JIT) compilation, or
  a list of the Embree-supported Xe GPUs for AOT compilation:

  * XE_HPG_CORE : Xe HPG devices
  * XE_HPC_CORE : Xe HPC devices

  One can also specify multiple devices separated by comma to
  compile ahead of time for multiple devices,
  e.g. "XE_HPG_CORE,XE_HP_CORE". When enabling AOT compilation for one
  or multiple devices, JIT compilation will always additionally be
  enabled in case the code is executed on a device no code is
  precompiled for.

  Execute "ocloc compile --help" for more details of possible devices
  to pass. Embree is only supported on Xe HPG/HPC and newer devices.

  Per default, this option is set to "none" to enable JIT
  compilation. We recommend using JIT compilation as this enables the
  use of specialization constants to reduce code complexity.

+ `EMBREE_STATIC_LIB`: Builds Embree as a static library (OFF by
  default). Further multiple static libraries are generated for the
  different ISAs selected (e.g. `embree4.a`, `embree4_sse42.a`,
  `embree4_avx.a`, `embree4_avx2.a`, `embree4_avx512.a`). You have
  to link these libraries in exactly this order of increasing ISA.

+ `EMBREE_API_NAMESPACE`: Specifies a namespace name to put all Embree
  API symbols inside. By default, no namespace is used and plain C symbols
  are exported.

+ `EMBREE_LIBRARY_NAME`: Specifies the name of the Embree library file
  created. By default, the name embree4 is used.

+ `EMBREE_IGNORE_CMAKE_CXX_FLAGS`: When enabled, Embree ignores
  default CMAKE_CXX_FLAGS. This option is turned ON by default.

+ `EMBREE_TUTORIALS`: Enables build of Embree tutorials (default ON).

+ `EMBREE_BACKFACE_CULLING`: Enables backface culling, i.e. only
  surfaces facing a ray can be hit. This option is turned OFF by
  default.

+ `EMBREE_BACKFACE_CULLING_CURVES`: Enables backface culling for curves,
  i.e. only surfaces facing a ray can be hit. This option is turned OFF
  by default.

+ `EMBREE_BACKFACE_CULLING_SPHERES`: Enables backface culling for spheres,
  i.e. only surfaces facing a ray can be hit. This option is turned OFF
  by default.

+ `EMBREE_COMPACT_POLYS`: Enables compact tris/quads, i.e. only
  geomIDs and primIDs are stored inside the leaf nodes.  

+ `EMBREE_FILTER_FUNCTION`: Enables the intersection filter function
  feature (ON by default).

+ `EMBREE_RAY_MASK`: Enables the ray masking feature (OFF by default).

+ `EMBREE_RAY_PACKETS`: Enables ray packet traversal kernels. This
  feature is turned ON by default. When turned on packet traversal is
  used internally and packets passed to rtcIntersect4/8/16 are kept
  intact in callbacks (when the ISA of appropriate width is enabled).

+ `EMBREE_IGNORE_INVALID_RAYS`: Makes code robust against the risk of
  full-tree traversals caused by invalid rays (e.g. rays containing
  INF/NaN as origins). This option is turned OFF by default.

+ `EMBREE_TASKING_SYSTEM`: Chooses between Intel® Threading TBB
  Building Blocks (TBB), Parallel Patterns Library (PPL) (Windows
  only), or an internal tasking system (INTERNAL). By default, TBB is
  used.

+ `EMBREE_TBB_ROOT`: If Intel® Threading Building Blocks (TBB)
  is used as a tasking system, search the library in this directory
  tree.

+ `EMBREE_TBB_COMPONENT`: The component/library name of Intel® Threading 
  Building Blocks (TBB). Embree searches for this library name (default: tbb)
  when TBB is used as the tasking system.

+ `EMBREE_TBB_POSTFIX`: If Intel® Threading Building Blocks (TBB)
  is used as a tasking system, link to tbb<EMBREE_TBB_POSTFIX>.(so,dll,lib).
  Defaults to the empty string.

+ `EMBREE_TBB_DEBUG_ROOT`: If Intel® Threading Building Blocks (TBB)
  is used as a tasking system, search the library in this directory
  tree in Debug mode. Defaults to `EMBREE_TBB_ROOT`.

+ `EMBREE_TBB_DEBUG_POSTFIX`: If Intel® Threading Building Blocks (TBB)
  is used as a tasking system, link to tbb<EMBREE_TBB_DEBUG_POSTFIX>.(so,dll,lib)
  in Debug mode. Defaults to "_debug".

+ `EMBREE_MAX_ISA`: Select highest supported ISA (SSE2, SSE4.2, AVX,
  AVX2, AVX512, or NONE). When set to NONE the
  EMBREE_ISA_* variables can be used to enable ISAs individually. By
  default, the option is set to AVX2.

+ `EMBREE_ISA_SSE2`: Enables SSE2 when EMBREE_MAX_ISA is set to
  NONE. By default, this option is turned OFF.

+ `EMBREE_ISA_SSE42`: Enables SSE4.2 when EMBREE_MAX_ISA is set to
  NONE. By default, this option is turned OFF.

+ `EMBREE_ISA_AVX`: Enables AVX when EMBREE_MAX_ISA is set to NONE. By
  default, this option is turned OFF.

+ `EMBREE_ISA_AVX2`: Enables AVX2 when EMBREE_MAX_ISA is set to
  NONE. By default, this option is turned OFF.

+ `EMBREE_ISA_AVX512`: Enables AVX-512 for Skylake when
  EMBREE_MAX_ISA is set to NONE. By default, this option is turned OFF.

+ `EMBREE_GEOMETRY_TRIANGLE`: Enables support for triangle geometries
  (ON by default).

+ `EMBREE_GEOMETRY_QUAD`: Enables support for quad geometries (ON by
  default).

+ `EMBREE_GEOMETRY_CURVE`: Enables support for curve geometries (ON by
  default).

+ `EMBREE_GEOMETRY_SUBDIVISION`: Enables support for subdivision
  geometries (ON by default).

+ `EMBREE_GEOMETRY_INSTANCE`: Enables support for instances (ON by
  default).

+ `EMBREE_GEOMETRY_INSTANCE_ARRAY`: Enables support for instance arrays (ON by
  default).

+ `EMBREE_GEOMETRY_USER`: Enables support for user-defined geometries
  (ON by default).

+ `EMBREE_GEOMETRY_POINT`: Enables support for point geometries
  (ON by default).

+ `EMBREE_CURVE_SELF_INTERSECTION_AVOIDANCE_FACTOR`: Specifies a
  factor that controls the self-intersection avoidance feature for flat
  curves. Flat curve intersections which are closer than
  curve_radius*`EMBREE_CURVE_SELF_INTERSECTION_AVOIDANCE_FACTOR` to
  the ray origin are ignored. A value of 0.0f disables self-intersection
  avoidance while 2.0f is the default value.

+ `EMBREE_DISC_POINT_SELF_INTERSECTION_AVOIDANCE`: Enables self-intersection
  avoidance for RTC_GEOMETRY_TYPE_DISC_POINT geometry type (ON by default).
  When enabled intersections are skipped if the ray origin lies inside the
  sphere defined by the point primitive.

+ `EMBREE_MIN_WIDTH`: Enabled the min-width feature, which allows
  increasing the radius of curves and points to match some amount of
  pixels. See [rtcSetGeometryMaxRadiusScale] for more details.

+ `EMBREE_MAX_INSTANCE_LEVEL_COUNT`: Specifies the maximum number of nested
  instance levels. Should be greater than 0; the default value is 1.
  Instances nested any deeper than this value will silently disappear in
  release mode, and cause assertions in debug mode.




Embree API
==========

The Embree API is a low-level C99 ray tracing API which can be used to
build spatial index structures for 3D scenes and perform ray queries of
different types.

The API can get used on the CPU using standard C, C++, and ISPC code
and Intel GPUs by using SYCL code.

The Intel® Implicit SPMD Program Compiler (Intel® ISPC) version of the
API, is almost identical to the standard C99 version, but contains
additional functions that operate on ray packets with a size of the
native SIMD width used by Intel® ISPC.

The SYCL version of the API is also mostly identical to the C99 version
of the API, with some exceptions listed in section [Embree SYCL API].

For simplicity this document refers to the C99 version of the API
functions. For changes when upgrading from the Embree 3 to the current
Embree 4 API see Section [Upgrading from Embree 3 to Embree 4].

All API calls carry the prefix `rtc` (or `RTC` for types) which stands
for **r**ay **t**racing **c**ore. The API supports scenes consisting of
different geometry types such as triangle meshes, quad meshes (triangle
pairs), grid meshes, flat curves, round curves, oriented curves,
subdivision meshes, instances, and user-defined geometries. See Section
[Scene Object](#scene-object) for more information.

Finding the closest hit of a ray segment with the scene
(`rtcIntersect`-type functions), and determining whether any hit
between a ray segment and the scene exists (`rtcOccluded`-type
functions) are both supported. The API supports queries for single rays
and ray packets. See Section [Ray Queries](#ray-queries) for more
information.

The API is designed in an object-oriented manner, e.g. it contains
device objects (`RTCDevice` type), scene objects (`RTCScene` type),
geometry objects (`RTCGeometry` type), buffer objects (`RTCBuffer`
type), and BVH objects (`RTCBVH` type). All objects are reference
counted, and handles can be released by calling the appropriate release
function (e.g. `rtcReleaseDevice`) or retained by incrementing the
reference count (e.g. `rtcRetainDevice`). In general, API calls that
access the same object are not thread-safe, unless specified otherwise.
However, attaching geometries to the same scene and performing ray
queries in a scene is thread-safe.

Device Object
-------------

Embree supports a device concept, which allows different components of
the application to use the Embree API without interfering with each
other. An application typically first creates a device using the
[rtcNewDevice] function (or [rtcNewSYCLDevice] when using SYCL for
the GPU). This device can then be used to construct further objects,
such as scenes and geometries. Before the application exits, it should
release all devices by invoking [rtcReleaseDevice]. An application
typically creates only a single device. If required differently, it
should only use a small number of devices at any given time.

Each user thread has its own error flag per device. If an error occurs
when invoking an API function, this flag is set to an error code (if it
isn't already set by a previous error). See Section
[rtcGetDeviceError] for information on how to read the error code and
Section [rtcSetDeviceErrorFunction] on how to register a callback
that is invoked for each error encountered. It is recommended to always
set a error callback function, to detect all errors.

Scene Object
------------

A scene is a container for a set of geometries, and contains a spatial
acceleration structure which can be used to perform different types of
ray queries.

A scene is created using the `rtcNewScene` function call, and released
using the `rtcReleaseScene` function call. To populate a scene with
geometries use the `rtcAttachGeometry` call, and to detach them use the
`rtcDetachGeometry` call. Once all scene geometries are attached, an
`rtcCommitScene` call (or `rtcJoinCommitScene` call) will finish the
scene description and trigger building of internal data structures.
After the scene got committed, it is safe to perform ray queries (see
Section [Ray Queries](#ray-queries)) or to query the scene bounding box
(see [rtcGetSceneBounds] and [rtcGetSceneLinearBounds]).

If scene geometries get modified or attached or detached, the
`rtcCommitScene` call must be invoked before performing any further ray
queries for the scene; otherwise the effect of the ray query is
undefined. The modification of a geometry, committing the scene, and
tracing of rays must always happen sequentially, and never at the same
time. Any API call that sets a property of the scene or geometries
contained in the scene count as scene modification, e.g. including
setting of intersection filter functions.

Scene flags can be used to configure a scene to use less memory
(`RTC_SCENE_FLAG_COMPACT`), use more robust traversal algorithms
(`RTC_SCENE_FLAG_ROBUST`), and to optimize for dynamic content. See
Section [rtcSetSceneFlags] for more details.

A build quality can be specified for a scene to balance between
acceleration structure build performance and ray query performance. See
Section [rtcSetSceneBuildQuality] for more details on build quality.

Geometry Object
---------------

A new geometry is created using the `rtcNewGeometry` function.
Depending on the geometry type, different buffers must be bound (e.g.
using `rtcSetSharedGeometryBuffer`) to set up the geometry data. In
most cases, binding of a vertex and index buffer is required. The
number of primitives and vertices of that geometry is typically
inferred from the size of these bound buffers.

Changes to the geometry always must be committed using the
`rtcCommitGeometry` call before using the geometry. After committing, a
geometry is not included in any scene. A geometry can be added to a
scene by using the `rtcAttachGeometry` function (to automatically
assign a geometry ID) or using the `rtcAttachGeometryById` function (to
specify the geometry ID manually). A geometry can get attached to
multiple scenes.

All geometry types support multi-segment motion blur with an arbitrary
number of equidistant time steps (in the range of 2 to 129) inside a
user specified time range. Each geometry can have a different number of
time steps and a different time range. The motion blur geometry is
defined by linearly interpolating the geometries of neighboring time
steps. To construct a motion blur geometry, first the number of time
steps of the geometry must be specified using the
`rtcSetGeometryTimeStepCount` function, and then a vertex buffer for
each time step must be bound, e.g. using the
`rtcSetSharedGeometryBuffer` function. Optionally, a time range
defining the start (and end time) of the first (and last) time step can
be set using the `rtcSetGeometryTimeRange` function. This feature will
also allow geometries to appear and disappear during the camera shutter
time if the time range is a sub range of [0,1].

Ray Queries
-----------

The API supports finding the closest hit of a ray segment with the
scene (`rtcIntersect`-type functions), and determining whether any hit
between a ray segment and the scene exists (`rtcOccluded`-type
functions).

Supported are single ray queries (`rtcIntersect1` and `rtcOccluded1`)
as well as ray packet queries for ray packets of size 4
(`rtcIntersect4` and `rtcOccluded4`), ray packets of size 8
(`rtcIntersect8` and `rtcOccluded8`), and ray packets of size 16
(`rtcIntersect16` and `rtcOccluded16`).

See Sections [rtcIntersect1] and [rtcOccluded1] for a detailed
description of how to set up and trace a ray.

See tutorial [Triangle Geometry] for a complete example of how to
trace single rays and ray packets.

Point Queries
-------------

The API supports traversal of the BVH using a point query object that
specifies a location and a query radius. For all primitives
intersecting the according domain, a user defined callback function is
called which allows queries such as finding the closest point on the
surface geometries of the scene (see Tutorial [Closest Point]) or
nearest neighbour queries (see Tutorial [Voronoi]).

See Section [rtcPointQuery] for a detailed description of how to set
up point queries.

Collision Detection
-------------------

The Embree API also supports collision detection queries between two
scenes consisting only of user geometries. Embree only performs
broadphase collision detection, the narrow phase detection can be
performed through a callback function.

See Section [rtcCollide] for a detailed description of how to set up
collision detection.

Seen tutorial [Collision Detection](#collision-detection) for a
complete example of collision detection being used on a simple cloth
solver.

Filter Functions
----------------

The API supports filter functions that are invoked for each
intersection found during the `rtcIntersect`-type or `rtcOccluded`-type
calls.

The filter functions can be set per-geometry using the
`rtcSetGeometryIntersectFilterFunction` and
`rtcSetGeometryOccludedFilterFunction` calls. The former ones are
called geometry intersection filter functions, the latter ones geometry
occlusion filter functions. These filter functions are designed to be
used to ignore intersections outside of a user-defined silhouette of a
primitive, e.g. to model tree leaves using transparency textures.

The filter function can also get passed as arguments directly to the
traversal functions, see section [rtcInitIntersectArguments] and
[rtcInitOccludedArguments] for more details. These argument filter
functions are designed to change the semantics of the ray query,
e.g. to accumulate opacity for transparent shadows, count the number of
surfaces along a ray, collect all hits along a ray, etc. The argument
filter function must be enabled to be used for a scene using the
`RTC_SCENE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS` scene flag. The callback
is only invoked for geometries that enable the callback using the
`rtcSetGeometryEnableFilterFunctionFromArguments` call, or enabled for
all geometries when the `RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER` ray
query flag is set.

BVH Build API
-------------

The internal algorithms to build a BVH are exposed through the `RTCBVH`
object and `rtcBuildBVH` call. This call makes it possible to build a
BVH in a user-specified format over user-specified primitives. See the
documentation of the `rtcBuildBVH` call for more details.

Embree SYCL API
===============

Embree supports ray tracing on Intel GPUs by using the SYCL programming
language. SYCL is a Khronos standardized C++ based language for single
source heterogenous programming for acceleration offload, see the [SYCL
webpage](https://www.khronos.org/sycl/) for details.

The Embree SYCL API is designed for photorealistic rendering use cases,
where scene setup is performed on the host, and rendering on the
device. The Embree SYCL API is very similar to the standard Embree C99
API, and supports most of its features, such as all triangle-type
geometries, all curve types and basis functions, point geometry types,
user geometries, filter callbacks, multi-level instancing, and motion
blur.

To enable SYCL support you have to include the `sycl.hpp` file before
the Embree API headers:

    #include <sycl/sycl.hpp>
    #include <embree4/rtcore.h>

Next you need to initializes an Embree SYCL device using the
`rtcNewSYCLDevice` API function by providing a SYCL context.

Embree provides the `rtcIsSYCLDeviceSupported` API function to check if
some SYCL device is supported by Embree. You can also use the
`rtcSYCLDeviceSelector` to conveniently select the first SYCL device
that is supported by Embree, e.g.:

    sycl::device device(rtcSYCLDeviceSelector);
    sycl::queue queue(device, exception_handler);
    sycl::context context(device);
    RTCDevice device = rtcNewSYCLDevice(context,"");

Scenes created with an Embree SYCL device can only get used to trace
rays using SYCL on the GPU, it is not possible to trace rays on the CPU
with such a device. To render on the CPU and GPU in parallel, the user
has to create a second Embree device and create a second scene to be
used on the CPU.

Files containing SYCL code, have to get compiled with the Intel® oneAPI
DPC++ compiler. Please see section [Linux SYCL Compilation] and
[Windows SYCL Compilation] for supported compilers. The DPC++
compiler performs a two-phase compilation, where host code is compiled
in a first phase, and device code compiled in a second compilation
phase.

Standard Embree API functions for scene construction can get used on
the host but not the device. Data buffers that are shared with Embree
(e.g. for vertex of index buffers) have to get allocated as SYCL
unified shared memory (USM memory), using the `sycl::malloc` or
`sycl::aligned_alloc` calls with `sycl::usm::alloc::shared` property,
or the sycl::aligned\_alloc\_shared call, e.g:

    void* ptr = sycl::aligned_alloc(16, bytes, queue, sycl::usm::alloc::shared);

These shared allocations have to be valid during rendering, as Embree
may access contained data when tracing rays. Embree does not support
device-only memory allocations, as the BVH builder implemented on the
CPU relies on reading the data buffers.

Device side rendering can get invoked by submitting a SYCL
`parallel_for` to the SYCL queue:

    const sycl::specialization_id<RTCFeatureFlags> feature_mask;

    RTCFeatureFlags required_features = RTC_FEATURE_FLAG_TRIANGLE;

    queue.submit([=](sycl::handler& cgh)
    {
      cgh.set_specialization_constant<feature_mask>(required_features);
      
      cgh.parallel_for(sycl::range<1>(1),[=](sycl::id<1> item, sycl::kernel_handler kh)
      {
        RTCIntersectArguments args;
        rtcInitIntersectArguments(&args);

        const RTCFeatureFlags features = kh.get_specialization_constant<feature_mask>();
        args.feature_mask = features;

        struct RTCRayHit rayhit;
        rayhit.ray.org_x = ox;
        rayhit.ray.org_y = oy;
        rayhit.ray.org_z = oz;
        rayhit.ray.dir_x = dx;
        rayhit.ray.dir_y = dy;
        rayhit.ray.dir_z = dz;
        rayhit.ray.tnear = 0;
        rayhit.ray.tfar = std::numeric_limits<float>::infinity();
        rayhit.ray.mask = -1;
        rayhit.ray.flags = 0;
        rayhit.hit.geomID = RTC_INVALID_GEOMETRY_ID;
        rayhit.hit.instID[0] = RTC_INVALID_GEOMETRY_ID;

        rtcIntersect1(scene, &rayhit, &args);

        result->geomID = rayhit.hit.geomID;
        result->primID = rayhit.hit.primID;
        result->tfar = rayhit.ray.tfar;
      });
    });
    queue.wait_and_throw();

This example passes a feature mask using a specialization contant to
the `rtcIntersect1` function, which is recommended for GPU rendering.
For best performance, this feature mask should get used to enable only
features required by the application to render the scene, e.g. just
triangles in this example.

Inside the SYCL `parallel_for` loop you can use rendering related
functions, such as the `rtcIntersect1` and `rtcOccluded1` functions to
trace rays, `rtcForwardIntersect1/Ex` and `rtcForwardOccluded1/Ex` to
continue object traversal from inside a user geometry callback, and
`rtcGetGeometryUserDataFromScene` to get the user data pointer of some
geometry.

Have a look at the [Minimal] tutorial for a minimal SYCL example.

SYCL JIT caching
----------------

Compile times for just in time compilation (JIT compilation) can be
large. To resolve this issue we recommend enabling persistent JIT
compilation caching inside your application, by setting the
`SYCL_CACHE_PERSISTENT` environment variable to `1`, and the
`SYCL_CACHE_DIR` environment variable to some proper directory where
the JIT cache should get stored. These environment variables have to
get set before the SYCL device is created, e.g:

    setenv("SYCL_CACHE_PERSISTENT","1",1);
    setenv("SYCL_CACHE_DIR","cache_dir",1);

    sycl::device device(rtcSYCLDeviceSelector);
    ...

SYCL Memory Pooling
-------------------

Memory Pooling is a mechanism where small USM memory allocations are
packed into larger allocation blocks. This mode is required when your
application performs many small USM allocations, as otherwise only a
small fraction of GPU memory is usable and data transfer performance
will be low.

Memory pooling is supported for USM allocations that are read-only by
the device. The following example allocated device read-only memory
with memory pooling support:

    sycl::aligned_alloc_shared(align, bytes, queue,
      sycl::ext::oneapi::property::usm::device_read_only());

Embree SYCL Limitations
-----------------------

Embree only supports Xe HPC and HPG GPUs as SYCL devices, thus in
particular the CPU and other GPUs cannot get used as a SYCL device. To
render on the CPU just use the standard C99 API without relying on
SYCL.

The SYCL language spec puts some restrictions to device functions, such
as disallowing: global variable access, malloc, invokation of virtual
functions, function pointers, runtime type information, exceptions,
recursion, etc. See Section
`5.4. Language Restrictions for device functions` of the [SYCL
specification](https://www.khronos.org/registry/SYCL/specs/sycl-2020/html/sycl-2020.html#sec:language.restrictions.kernels)
for more details.

Using Intel's oneAPI DPC++ compiler invoking an indirectly called
function is allowed, but we do not recommend this for performance
reasons.

Some features are not supported by the Embree SYCL API thus cannot get
used on the GPU:

-   The packet tracing functions `rtcIntersect4/8/16` and
    `rtcOccluded4/8/16` are not supported in SYCL device side code.
    Using these functions makes no sense for SYCL, as the programming
    model is implicitely executed in SIMT mode on the GPU anyway.

-   Filter and user geometry callbacks stored inside the geometry
    objects are not supported on SYCL. Please use the alternative
    approach of passing the function pointer through the
    `RTCIntersectArguments` (or `RTCOccludedArguments`) structures to
    the tracing function, which enables inlining on the GPU.

-   The `rtcInterpolate` function cannot get used on the the device.
    For most primitive types the vertex data interpolation is anyway a
    trivial operation, and an API call just introduces overheads. On
    the CPU that overhead is acceptable, but on the GPU it is not. The
    `rtcInterpolate` function does not know the geometry type it is
    interpolating over, thus its implementation on the GPU would
    contain a large switch statement for all potential geometry types.

-   Tracing rays using `rtcIntersect1` and `rtcOccluded1` functions
    from user geometry callbacks is not supported in SYCL. Please use
    the tail recursive `rtcForwardIntersect1` and `rtcForwardOccluded1`
    calls instead.

-   Subdivision surfaces are not supported for Embree SYCL devices.

-   Collision detection (`rtcCollide` API call) is not supported in
    SYCL device side code.

-   Point queries (`rtcPointQuery` API call) are not supported in SYCL
    device side code.

Embree SYCL Known Issues
------------------------

-   The SYCL support of Embree is in beta phase. Current functionality,
    quality, and GPU performance may not reflect that of the final
    product.

-   Compilation with build configuration "debug" is currently not
    working on Windows.

Upgrading from Embree 3 to Embree 4
===================================

This section summarizes API changes between Embree 3 and Embree4. Most
of these changes are motivated by GPU performance and having a
consistent API that works properly for the CPU and GPU.

-   The API include folder got renamed from embree3 to embree4, to be
    able to install Embree 3 and Embree 4 side by side, without having
    conflicts in API folder.

-   The `RTCIntersectContext` is renamed to `RTCRayQueryContext` and
    the `RTCIntersectContextFlags` got renamed to `RTCRayQueryFlags`.

-   There are some changes to the `rtcIntersect` and `rtcOccluded`
    functions. Most members of the old intersect context have been
    moved to some optional `RTCIntersectArguments` (and
    `RTCOccludedArguments`) structures, which also contains a pointer
    to the new ray query context. The argument structs fulfill the task
    of providing additional advanced arguments to the traversal
    functions. The ray query context can get used to pass additional
    data to callbacks, and to maintain an instID stack in case
    instancing is done manually inside user geometry callbacks. The
    arguments struct is not available inside callbacks. This change was
    in particular necessary for SYCL to allow inlining of function
    pointers provided to the traversal functions, and to reduce the
    amount of state passed to callbacks, which both improves GPU
    performance. Most applications can just drop passing the ray query
    context to port to Embree 4.

-   The `rtcFilterIntersection` and `rtcFilterOcclusion` API calls that
    invoke both, the geometry and argument version of the filter
    callback, from a user geometry callback are no longer supported.
    Instead applications should use the
    `rtcInvokeIntersectFilterFromGeometry` and
    `rtcInvokeOccludedFilterFromGeometry` API calls that invoke just
    the geometry version of the filter function, and invoke the
    argument filter function manually if required.

-   The filter function passed as arguments to `rtcIntersect` and
    `rtcOccluded` functions is only invoked for some geometry if
    enabled through `rtcSetGeometryEnableFilterFunctionFromArguments`
    for that geometry. Alternatively, argument filter functions can get
    enabled for all geometries using the
    `RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER` ray query flag.

-   User geometry callbacks get a valid vector as input to identify
    valid and invalid rays. In Embree 3 the user geometry callback just
    had to update the ray hit members when an intersection was found
    and perform no operation otherwise. In Embree 4 the callback
    additionally has to return valid=-1 when a hit was found, and
    valid=0 when no hit was found. This allows Embree to properly pass
    the new hit distance to the ray tracing hardware only in the case a
    hit was found.

-   Further ray masking is enabled by default now as required by most
    applications and the default ray mask for geometries got changed
    from 0xFFFFFFFF to 0x1.

-   The stream tracing functions `rtcIntersect1M`, `rtcIntersect1Mp`,
    `rtcIntersectNM`, `rtcIntersectNp`, `rtcOccluded1M`,
    `rtcOccluded1Mp`, `rtcOccludedNM`, and `rtcOccludedNp` got removed
    as they were rarely used and did not provide relevant performance
    benefits. As alternative the application can just iterate over
    `rtcIntersect1` and potentially `rtcIntersect4/8/16` to get similar
    performance.

To use Embree through SYCL on the CPU and GPU additional changes are
required:

-   Embree 3 allows to use `rtcIntersect` recursively from a user
    geometry or intersection filter callback to continue a ray inside
    an instantiated object. In Embree 4 using `rtcIntersect`
    recursively is disallowed on the GPU but still supported on the
    CPU. To properly continue a ray inside an instantiated object use
    the new `rtcForwardIntersect1` and `rtcForwardOccluded1` functions.

-   The geometry object of Embree 4 is a host side only object, thus
    accessing it during rendering from the GPU is not allowed. Thus all
    API functions that take an RTCGeometry object as argument cannot
    get used during rendering. Thus in particular the
    `rtcGetGeometryUserData(RTCGeometry)` call cannot get used, but
    there is an alternative function
    `rtcGetGeometryUserDataFromScene(RTCScene scene,uint geomID)` that
    should get used instead.

-   The user geometry callback and filter callback functions should get
    passed through the intersection and occlusion argument structures
    to the `rtcIntersect1` and `rtcOccluded1` functions directly to
    allow inlining. The experimental geometry version of the callbacks
    is disabled in SYCL and should not get used.

-   The feature flags should get used in SYCL to minimal GPU code for
    optimal performance.

-   The `rtcInterpolate` function cannot get used on the device, and
    vertex data interpolation should get implemented by the
    application.

-   Indirectly called functions must be declared with
    `RTC_SYCL_INDIRECTLY_CALLABLE` when used as filter or user geometry
    callbacks.

```{=tex}

```
Embree API Reference
====================

rtcNewDevice
------------

#### NAME

    rtcNewDevice - creates a new device

#### SYNOPSIS

    #include <embree4/rtcore.h>

    RTCDevice rtcNewDevice(const char* config);

#### DESCRIPTION

This function creates a new device to be used for CPU ray tracing and
returns a handle to this device. The device object is reference counted
with an initial reference count of 1. The handle can be released using
the `rtcReleaseDevice` API call.

The device object acts as a class factory for all other object types.
All objects created from the device (like scenes, geometries, etc.)
hold a reference to the device, thus the device will not be destroyed
unless these objects are destroyed first.

Objects are only compatible if they belong to the same device, e.g it
is not allowed to create a geometry in one device and attach it to a
scene created with a different device.

A configuration string (`config` argument) can be passed to the device
construction. This configuration string can be `NULL` to use the
default configuration.

The following configuration is supported:

-   `threads=[int]`: Specifies a number of build threads to use. A
    value of 0 enables all detected hardware threads. By default all
    hardware threads are used.

-   `user_threads=[int]`: Sets the number of user threads that can be
    used to join and participate in a scene commit using
    `rtcJoinCommitScene`. The tasking system will only use
    threads-user\_threads many worker threads, thus if the app wants to
    solely use its threads to commit scenes, just set threads equal to
    user\_threads. This option only has effect with the Intel(R)
    Threading Building Blocks (TBB) tasking system.

-   `set_affinity=[0/1]`: When enabled, build threads are affinitized
    to hardware threads. This option is disabled by default on standard
    CPUs, and enabled by default on Xeon Phi Processors.

-   `start_threads=[0/1]`: When enabled, the build threads are started
    upfront. This can be useful for benchmarking to exclude thread
    creation time. This option is disabled by default.

-   `isa=[sse2,sse4.2,avx,avx2,avx512]`: Use specified ISA. By default
    the ISA is selected automatically.

-   `max_isa=[sse2,sse4.2,avx,avx2,avx512]`: Configures the automated
    ISA selection to use maximally the specified ISA.

-   `hugepages=[0/1]`: Enables or disables usage of huge pages. Under
    Linux huge pages are used by default but under Windows and macOS
    they are disabled by default.

-   `enable_selockmemoryprivilege=[0/1]`: When set to 1, this enables
    the `SeLockMemoryPrivilege` privilege with is required to use huge
    pages on Windows. This option has an effect only under Windows and
    is ignored on other platforms. See Section [Huge Page Support]
    for more details.

-   `verbose=[0,1,2,3]`: Sets the verbosity of the output. When set to
    0, no output is printed by Embree, when set to a higher level more
    output is printed. By default Embree does not print anything on the
    console.

-   `frequency_level=[simd128,simd256,simd512]`: Specifies the
    frequency level the application want to run on, which can be
    either:

    a)  simd128 to run at highest frequency
    b)  simd256 to run at AVX2-heavy frequency level
    c)  simd512 to run at heavy AVX512 frequency level. When some
        frequency level is specified, Embree will avoid doing
        optimizations that may reduce the frequency level below the
        level specified. E.g. if your app does not use AVX instructions
        setting "frequency\_level=simd128" will cause some CPUs to run
        at highest frequency, which may result in higher application
        performance if you do much shading. If you application heavily
        uses AVX code, you should best set the frequency level to
        simd256. Per default Embree tries to avoid reducing the
        frequency of the CPU by setting the simd256 level only when the
        CPU has no significant down clocking.

Different configuration options should be separated by commas, e.g.:

    rtcNewDevice("threads=1,isa=avx");

#### EXIT STATUS

On success returns a handle of the created device. On failure returns
`NULL` as device and sets a per-thread error code that can be queried
using `rtcGetDeviceError(NULL)`.

#### SEE ALSO

[rtcRetainDevice], [rtcReleaseDevice], [rtcNewSYCLDevice]

```{=tex}

```
rtcNewSYCLDevice
----------------

#### NAME {#name}

    rtcNewSYCLDevice - creates a new device to be used with SYCL

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCDevice rtcNewSYCLDevice(sycl::context context, const char* config);

#### DESCRIPTION {#description}

This function creates a new device to be used with SYCL for GPU
rendering and returns a handle to this device. The device object is
reference counted with an initial reference count of 1. The handle can
get released using the `rtcReleaseDevice` API call.

The passed SYCL context (`context` argument) is used to allocate GPU
data, thus only devices contained inside this context can be used for
rendering. By default the GPU data is allocated on the first GPU device
of the context, but this behavior can get changed with the
[rtcSetDeviceSYCLDevice] function.

The device object acts as a class factory for all other object types.
All objects created from the device (like scenes, geometries, etc.)
hold a reference to the device, thus the device will not be destroyed
unless these objects are destroyed first.

Objects are only compatible if they belong to the same device, e.g it
is not allowed to create a geometry in one device and attach it to a
scene created with a different device.

For an overview of configurations that can get passed (`config`
argument) please see the [rtcNewDevice] function description.

#### EXIT STATUS {#exit-status}

On success returns a handle of the created device. On failure returns
`NULL` as device and sets a per-thread error code that can be queried
using `rtcGetDeviceError(NULL)`.

#### SEE ALSO {#see-also}

[rtcRetainDevice], [rtcReleaseDevice], [rtcNewDevice]

```{=tex}

```
rtcIsSYCLDeviceSupported
------------------------

#### NAME {#name}

    rtcIsSYCLDeviceSupported - checks if some SYCL device is supported by Embree

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    bool rtcIsSYCLDeviceSupported(const sycl::device sycl_device);

#### DESCRIPTION {#description}

This function can be used to check if some SYCL device (`sycl_device`
argument) is supported by Embree.

#### EXIT STATUS {#exit-status}

The function returns true if the SYCL device is supported by Embree and
false otherwise. On failure an error code is set that can get queried
using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSYCLDeviceSelector]

```{=tex}

```
rtcSYCLDeviceSelector
---------------------

#### NAME {#name}

    rtcSYCLDeviceSelector - SYCL device selector function to select
      devices supported by Embree

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    int rtcSYCLDeviceSelector(const sycl::device sycl_device);

#### DESCRIPTION {#description}

This function checks if the passed SYCL device (`sycl_device`
arguments) is supported by Embree or not. This function can be used
directly to select some supported SYCL device by using it as SYCL
device selector function. For instance, the following code sequence
selects an Embree supported SYCL device and creates an Embree device
from it:

    sycl::device sycl_device(rtcSYCLDeviceSelector);
    sycl::queue sycl_queue(sycl_device);
    sycl::context(sycl_device);
    RTCDevice device = rtcNewSYCLDevice(sycl_context,nullptr);

#### EXIT STATUS {#exit-status}

The function returns -1 if the SYCL device is supported by Embree and 1
otherwise. On failure an error code is set that can get queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcIsSYCLDeviceSupported]

```{=tex}

```
rtcSetDeviceSYCLDevice
----------------------

#### NAME {#name}

    rtcSetDeviceSYCLDevice - sets the SYCL device to be used for memory allocations

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetDeviceSYCLDevice(RTCDevice device, const sycl::device sycl_device);

#### DESCRIPTION {#description}

This function sets the SYCL device (`sycl_device` argument) to be used
to allocate GPU memory when using the specified Embree device (`device`
argument). This SYCL device must be one of the SYCL devices contained
inside the SYCL context used to create the Embree device.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can get queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewSYCLDevice]

```{=tex}

```
rtcRetainDevice
---------------

#### NAME {#name}

    rtcRetainDevice - increments the device reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcRetainDevice(RTCDevice device);

#### DESCRIPTION {#description}

Device objects are reference counted. The `rtcRetainDevice` function
increments the reference count of the passed device object (`device`
argument). This function together with `rtcReleaseDevice` allows to use
the internal reference counting in a C++ wrapper class to manage the
ownership of the object.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewDevice], [rtcReleaseDevice]

```{=tex}

```
rtcReleaseDevice
----------------

#### NAME {#name}

    rtcReleaseDevice - decrements the device reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcReleaseDevice(RTCDevice device);

#### DESCRIPTION {#description}

Device objects are reference counted. The `rtcReleaseDevice` function
decrements the reference count of the passed device object (`device`
argument). When the reference count falls to 0, the device gets
destroyed.

All objects created from the device (like scenes, geometries, etc.)
hold a reference to the device, thus the device will not get destroyed
unless these objects are destroyed first.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewDevice], [rtcRetainDevice]

```{=tex}

```
rtcGetDeviceProperty
--------------------

#### NAME {#name}

    rtcGetDeviceProperty - queries properties of the device

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    ssize_t rtcGetDeviceProperty(
      RTCDevice device,
      enum RTCDeviceProperty prop
    );

#### DESCRIPTION {#description}

The `rtcGetDeviceProperty` function can be used to query properties
(`prop` argument) of a device object (`device` argument). The returned
property is an integer of type `ssize_t`.

Possible properties to query are:

-   `RTC_DEVICE_PROPERTY_VERSION`: Queries the combined version number
    (MAJOR.MINOR.PATCH) with two decimal digits per component. E.g. for
    Embree 2.8.3 the integer 208003 is returned.

-   `RTC_DEVICE_PROPERTY_VERSION_MAJOR`: Queries the major version
    number of Embree.

-   `RTC_DEVICE_PROPERTY_VERSION_MINOR`: Queries the minor version
    number of Embree.

-   `RTC_DEVICE_PROPERTY_VERSION_PATCH`: Queries the patch version
    number of Embree.

-   `RTC_DEVICE_PROPERTY_NATIVE_RAY4_SUPPORTED`: Queries whether the
    `rtcIntersect4` and `rtcOccluded4` functions preserve packet size
    and ray order when invoking callback functions. This is only the
    case if Embree is compiled with `EMBREE_RAY_PACKETS` and `SSE2` (or
    `SSE4.2`) enabled, and if the machine it is running on supports
    `SSE2` (or `SSE4.2`).

-   `RTC_DEVICE_PROPERTY_NATIVE_RAY8_SUPPORTED`: Queries whether the
    `rtcIntersect8` and `rtcOccluded8` functions preserve packet size
    and ray order when invoking callback functions. This is only the
    case if Embree is compiled with `EMBREE_RAY_PACKETS` and `AVX` (or
    `AVX2`) enabled, and if the machine it is running on supports `AVX`
    (or `AVX2`).

-   `RTC_DEVICE_PROPERTY_NATIVE_RAY16_SUPPORTED`: Queries whether the
    `rtcIntersect16` and `rtcOccluded16` functions preserve packet size
    and ray order when invoking callback functions. This is only the
    case if Embree is compiled with `EMBREE_RAY_PACKETS` and `AVX512`
    enabled, and if the machine it is running on supports `AVX512`.

-   `RTC_DEVICE_PROPERTY_RAY_MASK_SUPPORTED`: Queries whether ray masks
    are supported. This is only the case if Embree is compiled with
    `EMBREE_RAY_MASK` enabled.

-   `RTC_DEVICE_PROPERTY_BACKFACE_CULLING_ENABLED`: Queries whether
    back face culling is enabled. This is only the case if Embree is
    compiled with `EMBREE_BACKFACE_CULLING` enabled.

-   `RTC_DEVICE_PROPERTY_BACKFACE_CULLING_CURVES_ENABLED`: Queries
    whether back face culling for curves is enabled. This is only the
    case if Embree is compiled with `EMBREE_BACKFACE_CULLING_CURVES`
    enabled.

-   `RTC_DEVICE_PROPERTY_BACKFACE_CULLING_SPHERES_ENABLED`: Queries
    whether back face culling for spheres is enabled. This is only the
    case if Embree is compiled with `EMBREE_BACKFACE_CULLING_SPHERES`
    enabled.

-   `RTC_DEVICE_PROPERTY_COMPACT_POLYS_ENABLED`: Queries whether
    compact polys is enabled. This is only the case if Embree is
    compiled with `EMBREE_COMPACT_POLYS` enabled.

-   `RTC_DEVICE_PROPERTY_FILTER_FUNCTION_SUPPORTED`: Queries whether
    filter functions are supported, which is the case if Embree is
    compiled with `EMBREE_FILTER_FUNCTION` enabled.

-   `RTC_DEVICE_PROPERTY_IGNORE_INVALID_RAYS_ENABLED`: Queries whether
    invalid rays are ignored, which is the case if Embree is compiled
    with `EMBREE_IGNORE_INVALID_RAYS` enabled.

-   `RTC_DEVICE_PROPERTY_TRIANGLE_GEOMETRY_SUPPORTED`: Queries whether
    triangles are supported, which is the case if Embree is compiled
    with `EMBREE_GEOMETRY_TRIANGLE` enabled.

-   `RTC_DEVICE_PROPERTY_QUAD_GEOMETRY_SUPPORTED`: Queries whether
    quads are supported, which is the case if Embree is compiled with
    `EMBREE_GEOMETRY_QUAD` enabled.

-   `RTC_DEVICE_PROPERTY_SUBDIVISION_GEOMETRY_SUPPORTED`: Queries
    whether subdivision meshes are supported, which is the case if
    Embree is compiled with `EMBREE_GEOMETRY_SUBDIVISION` enabled.

-   `RTC_DEVICE_PROPERTY_CURVE_GEOMETRY_SUPPORTED`: Queries whether
    curves are supported, which is the case if Embree is compiled with
    `EMBREE_GEOMETRY_CURVE` enabled.

-   `RTC_DEVICE_PROPERTY_POINT_GEOMETRY_SUPPORTED`: Queries whether
    points are supported, which is the case if Embree is compiled with
    `EMBREE_GEOMETRY_POINT` enabled.

-   `RTC_DEVICE_PROPERTY_USER_GEOMETRY_SUPPORTED`: Queries whether user
    geometries are supported, which is the case if Embree is compiled
    with `EMBREE_GEOMETRY_USER` enabled.

-   `RTC_DEVICE_PROPERTY_TASKING_SYSTEM`: Queries the tasking system
    Embree is compiled with. Possible return values are:

    0.  internal tasking system
    1.  Intel Threading Building Blocks (TBB)
    2.  Parallel Patterns Library (PPL)

-   `RTC_DEVICE_PROPERTY_JOIN_COMMIT_SUPPORTED`: Queries whether
    `rtcJoinCommitScene` is supported. This is not the case when Embree
    is compiled with PPL or older versions of TBB.

-   `RTC_DEVICE_PROPERTY_PARALLEL_COMMIT_SUPPORTED`: Queries whether
    `rtcCommitScene` can get invoked from multiple TBB worker threads
    concurrently. This feature is only supported starting with TBB 2019
    Update 9.

#### EXIT STATUS {#exit-status}

On success returns the value of the queried property. For properties
returning a boolean value, the return value 0 denotes `false` and 1
denotes `true`.

On failure zero is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

```{=tex}

```
rtcGetDeviceError
-----------------

#### NAME {#name}

    rtcGetDeviceError - returns the error code of the device

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCError rtcGetDeviceError(RTCDevice device);

#### DESCRIPTION {#description}

Each thread has its own error code per device. If an error occurs when
calling an API function, this error code is set to the occurred error
if it stores no previous error. The `rtcGetDeviceError` function reads
and returns the currently stored error and clears the error code. This
assures that the returned error code is always the first error occurred
since the last invocation of `rtcGetDeviceError`.

Possible error codes returned by `rtcGetDeviceError` are:

-   `RTC_ERROR_NONE`: No error occurred.

-   `RTC_ERROR_UNKNOWN`: An unknown error has occurred.

-   `RTC_ERROR_INVALID_ARGUMENT`: An invalid argument was specified.

-   `RTC_ERROR_INVALID_OPERATION`: The operation is not allowed for the
    specified object.

-   `RTC_ERROR_OUT_OF_MEMORY`: There is not enough memory left to
    complete the operation.

-   `RTC_ERROR_UNSUPPORTED_CPU`: The CPU is not supported as it does
    not support the lowest ISA Embree is compiled for.

-   `RTC_ERROR_CANCELLED`: The operation got canceled by a memory
    monitor callback or progress monitor callback function.

When the device construction fails, `rtcNewDevice` returns `NULL` as
device. To detect the error code of a such a failed device
construction, pass `NULL` as device to the `rtcGetDeviceError`
function. For all other invocations of `rtcGetDeviceError`, a proper
device pointer must be specified.

#### EXIT STATUS {#exit-status}

Returns the error code for the device.

#### SEE ALSO {#see-also}

[rtcSetDeviceErrorFunction]

```{=tex}

```
rtcSetDeviceErrorFunction
-------------------------

#### NAME {#name}

    rtcSetDeviceErrorFunction - sets an error callback function for the device

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    typedef void (*RTCErrorFunction)(
      void* userPtr,
      RTCError code,
      const char* str
    );

    void rtcSetDeviceErrorFunction(
      RTCDevice device,
      RTCErrorFunction error,
      void* userPtr
    );

#### DESCRIPTION {#description}

Using the `rtcSetDeviceErrorFunction` call, it is possible to set a
callback function (`error` argument) with payload (`userPtr` argument),
which is called whenever an error occurs for the specified device
(`device` argument).

Only a single callback function can be registered per device, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

When the registered callback function is invoked, it gets passed the
user-defined payload (`userPtr` argument as specified at registration
time), the error code (`code` argument) of the occurred error, as well
as a string (`str` argument) that further describes the error.

The error code is also set if an error callback function is registered.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetDeviceError]

```{=tex}

```
rtcSetDeviceMemoryMonitorFunction
---------------------------------

#### NAME {#name}

    rtcSetDeviceMemoryMonitorFunction - registers a callback function
      to track memory consumption

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    typedef bool (*RTCMemoryMonitorFunction)(
      void* userPtr,
      ssize_t bytes,
      bool post
    );

    void rtcSetDeviceMemoryMonitorFunction(
      RTCDevice device,
      RTCMemoryMonitorFunction memoryMonitor,
      void* userPtr
    );

#### DESCRIPTION {#description}

Using the `rtcSetDeviceMemoryMonitorFunction` call, it is possible to
register a callback function (`memoryMonitor` argument) with payload
(`userPtr` argument) for a device (`device` argument), which is called
whenever internal memory is allocated or deallocated by objects of that
device. Using this memory monitor callback mechanism, the application
can track the memory consumption of an Embree device, and optionally
terminate API calls that consume too much memory.

Only a single callback function can be registered per device, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

Once registered, the Embree device will invoke the memory monitor
callback function before or after it allocates or frees important
memory blocks. The callback function gets passed the payload as
specified at registration time (`userPtr` argument), the number of
bytes allocated or deallocated (`bytes` argument), and whether the
callback is invoked after the allocation or deallocation took place
(`post` argument). The callback function might get called from multiple
threads concurrently.

The application can track the current memory usage of the Embree device
by atomically accumulating the `bytes` input parameter provided to the
callback function. This parameter will be \>0 for allocations and \<0
for deallocations.

Embree will continue its operation normally when returning `true` from
the callback function. If `false` is returned, Embree will cancel the
current operation with the `RTC_ERROR_OUT_OF_MEMORY` error code.
Issuing multiple cancel requests from different threads is allowed.
Canceling will only happen when the callback was called for allocations
(bytes \> 0), otherwise the cancel request will be ignored.

If a callback to cancel was invoked before the allocation happens
(`post == false`), then the `bytes` parameter should not be
accumulated, as the allocation will never happen. If the callback to
cancel was invoked after the allocation happened (`post == true`), then
the `bytes` parameter should be accumulated, as the allocation properly
happened and a deallocation will later free that data block.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewDevice]

```{=tex}

```
rtcNewScene
-----------

#### NAME {#name}

    rtcNewScene - creates a new scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCScene rtcNewScene(RTCDevice device);

#### DESCRIPTION {#description}

This function creates a new scene bound to the specified device
(`device` argument), and returns a handle to this scene. The scene
object is reference counted with an initial reference count of 1. The
scene handle can be released using the `rtcReleaseScene` API call.

#### EXIT STATUS {#exit-status}

On success a scene handle is returned. On failure `NULL` is returned
and an error code is set that can be queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcRetainScene], [rtcReleaseScene]

```{=tex}

```
rtcGetSceneDevice
-----------------

#### NAME {#name}

    rtcGetSceneDevice - returns the device the scene got created in

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCDevice rtcGetSceneDevice(RTCScene scene);

#### DESCRIPTION {#description}

This function returns the device object the scene got created in. The
returned handle own one additional reference to the device object, thus
you should need to call `rtcReleaseDevice` when the returned handle is
no longer required.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcReleaseDevice]

```{=tex}

```
rtcRetainScene
--------------

#### NAME {#name}

    rtcRetainScene - increments the scene reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcRetainScene(RTCScene scene);

#### DESCRIPTION {#description}

Scene objects are reference counted. The `rtcRetainScene` function
increments the reference count of the passed scene object (`scene`
argument). This function together with `rtcReleaseScene` allows to use
the internal reference counting in a C++ wrapper class to handle the
ownership of the object.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewScene], [rtcReleaseScene]

```{=tex}

```
rtcReleaseScene
---------------

#### NAME {#name}

    rtcReleaseScene - decrements the scene reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcReleaseScene(RTCScene scene);

#### DESCRIPTION {#description}

Scene objects are reference counted. The `rtcReleaseScene` function
decrements the reference count of the passed scene object (`scene`
argument). When the reference count falls to 0, the scene gets
destroyed.

The scene holds a reference to all attached geometries, thus if the
scene gets destroyed, all geometries get detached and their reference
count decremented.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewScene], [rtcRetainScene]

```{=tex}

```
rtcAttachGeometry
-----------------

#### NAME {#name}

    rtcAttachGeometry - attaches a geometry to the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    unsigned int rtcAttachGeometry(
      RTCScene scene,
      RTCGeometry geometry
    );

#### DESCRIPTION {#description}

The `rtcAttachGeometry` function attaches a geometry (`geometry`
argument) to a scene (`scene` argument) and assigns a geometry ID to
that geometry. All geometries attached to a scene are defined to be
included inside the scene. A geometry can get attached to multiple
scenes. The geometry ID is unique for the scene, and is used to
identify the geometry when hit by a ray during ray queries.

This function is thread-safe, thus multiple threads can attach
geometries to a scene in parallel.

The geometry IDs are assigned sequentially, starting from 0, as long as
no geometry got detached. If geometries got detached, the
implementation will reuse IDs in an implementation dependent way.
Consequently sequential assignment is no longer guaranteed, but a
compact range of IDs.

These rules allow the application to manage a dynamic array to
efficiently map from geometry IDs to its own geometry representation.
Alternatively, the application can also use per-geometry user data to
map to its geometry representation. See `rtcSetGeometryUserData` and
`rtcGetGeometryUserData` for more information.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryUserData], [rtcGetGeometryUserData]

```{=tex}

```
rtcAttachGeometryByID
---------------------

#### NAME {#name}

    rtcAttachGeometryByID - attaches a geometry to the scene
      using a specified geometry ID

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcAttachGeometryByID(
      RTCScene scene,
      RTCGeometry geometry,
      unsigned int geomID
    );

#### DESCRIPTION {#description}

The `rtcAttachGeometryByID` function attaches a geometry (`geometry`
argument) to a scene (`scene` argument) and assigns a user provided
geometry ID (`geomID` argument) to that geometry. All geometries
attached to a scene are defined to be included inside the scene. A
geometry can get attached to multiple scenes. The passed user-defined
geometry ID is used to identify the geometry when hit by a ray during
ray queries. Using this function, it is possible to share the same IDs
to refer to geometries inside the application and Embree.

This function is thread-safe, thus multiple threads can attach
geometries to a scene in parallel.

The user-provided geometry ID must be unused in the scene, otherwise
the creation of the geometry will fail. Further, the user-provided
geometry IDs should be compact, as Embree internally creates a vector
which size is equal to the largest geometry ID used. Creating very
large geometry IDs for small scenes would thus cause a memory
consumption and performance overhead.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcAttachGeometry]

```{=tex}

```
rtcDetachGeometry
-----------------

#### NAME {#name}

    rtcDetachGeometry - detaches a geometry from the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcDetachGeometry(RTCScene scene, unsigned int geomID);

#### DESCRIPTION {#description}

This function detaches a geometry identified by its geometry ID
(`geomID` argument) from a scene (`scene` argument). When detached, the
geometry is no longer contained in the scene.

This function is thread-safe, thus multiple threads can detach
geometries from a scene at the same time.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcAttachGeometry], [rtcAttachGeometryByID]

```{=tex}

```
rtcGetGeometry
--------------

#### NAME {#name}

    rtcGetGeometry - returns the geometry bound to
      the specified geometry ID

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry rtcGetGeometry(RTCScene scene, unsigned int geomID);

#### DESCRIPTION {#description}

The `rtcGetGeometry` function returns the geometry that is bound to the
specified geometry ID (`geomID` argument) for the specified scene
(`scene` argument). This function just looks up the handle and does
*not* increment the reference count. If you want to get ownership of
the handle, you need to additionally call `rtcRetainGeometry`.

This function is not thread safe and thus can be used during rendering.
However, it is generally recommended to store the geometry handle
inside the application's geometry representation and look up the
geometry handle from that representation directly.

If you need a thread safe version of this function please use
[rtcGetGeometryThreadSafe].

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcAttachGeometry], [rtcAttachGeometryByID],
[rtcGetGeometryThreadSafe]

```{=tex}

```
rtcGetGeometryThreadSafe
------------------------

#### NAME {#name}

    rtcGetGeometryThreadSafe - returns the geometry bound to
      the specified geometry ID

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry rtcGetGeometryThreadSafe(RTCScene scene, unsigned int geomID);

#### DESCRIPTION {#description}

The `rtcGetGeometryThreadSafe` function returns the geometry that is
bound to the specified geometry ID (`geomID` argument) for the
specified scene (`scene` argument). This function just looks up the
handle and does *not* increment the reference count. If you want to get
ownership of the handle, you need to additionally call
`rtcRetainGeometry`.

This function is thread safe and should NOT get used during rendering.
If you need a fast non-thread safe version during rendering please use
the [rtcGetGeometry] function.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcAttachGeometry], [rtcAttachGeometryByID], [rtcGetGeometry]

```{=tex}

```
rtcCommitScene
--------------

#### NAME {#name}

    rtcCommitScene - commits scene changes

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcCommitScene(RTCScene scene);

#### DESCRIPTION {#description}

The `rtcCommitScene` function commits all changes for the specified
scene (`scene` argument). This internally triggers building of a
spatial acceleration structure for the scene using all available worker
threads. Ray queries can be performed only after committing all scene
changes.

If the application uses TBB 2019 Update 9 or later for parallelization
of rendering, lazy scene construction during rendering is supported by
`rtcCommitScene`. Therefore `rtcCommitScene` can get called from
multiple TBB worker threads concurrently for the same scene. The
`rtcCommitScene` function will then internally isolate the scene
construction using a tbb::isolated\_task\_group. The alternative
approach of using `rtcJoinCommitScene` which uses an tbb:task\_arena
internally, is not recommended due to it's high runtime overhead.

If scene geometries get modified or attached or detached, the
`rtcCommitScene` call must be invoked before performing any further ray
queries for the scene; otherwise the effect of the ray query is
undefined. The modification of a geometry, committing the scene, and
tracing of rays must always happen sequentially, and never at the same
time. Any API call that sets a property of the scene or geometries
contained in the scene count as scene modification, e.g. including
setting of intersection filter functions.

The kind of acceleration structure built can be influenced using scene
flags (see `rtcSetSceneFlags`), and the quality can be specified using
the `rtcSetSceneBuildQuality` function.

Embree silently ignores primitives during spatial acceleration
structure construction that would cause numerical issues,
e.g. primitives containing NaNs, INFs, or values greater than 1.844E18f
(as no reasonable calculations can be performed with such values
without causing overflows).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcJoinCommitScene]

```{=tex}

```
rtcJoinCommitScene
------------------

#### NAME {#name}

    rtcJoinCommitScene - commits the scene from multiple threads

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcJoinCommitScene(RTCScene scene);

#### DESCRIPTION {#description}

The `rtcJoinCommitScene` function commits all changes for the specified
scene (`scene` argument). The scene commit internally triggers building
of a spatial acceleration structure for the scene. Ray queries can be
performed after scene changes got properly committed.

The `rtcJoinCommitScene` function can get called from multiple user
threads which will all cooperate in the build operation. All threads
calling into this function will return from `rtcJoinCommitScene` after
the scene commit is finished. All threads must consistently call
`rtcJoinCommitScene` and not `rtcCommitScene`.

In contrast to the `rtcCommitScene` function, the `rtcJoinCommitScene`
function can be called from multiple user threads, while the
`rtcCommitScene` can only get called from multiple TBB worker threads
when used concurrently. For optimal performance we strongly recommend
using TBB inside the application together with the `rtcCommitScene`
function and to avoid using the `rtcJoinCommitScene` function.

The `rtcJoinCommitScene` feature allows a flexible way to lazily create
hierarchies during rendering. A thread reaching a not-yet-constructed
sub-scene of a two-level scene can generate the sub-scene geometry and
call `rtcJoinCommitScene` on that just generated scene. During
construction, further threads reaching the not-yet-built scene can join
the build operation by also invoking `rtcJoinCommitScene`. A thread
that calls `rtcJoinCommitScene` after the build finishes will directly
return from the `rtcJoinCommitScene` call.

Multiple scene commit operations on different scenes can be running at
the same time, hence it is possible to commit many small scenes in
parallel, distributing the commits to many threads.

When using Embree with the Intel® Threading Building Blocks (which is
the default), threads that call `rtcJoinCommitScene` will join the
build operation, but other TBB worker threads might also participate in
the build. To avoid thread oversubscription, we recommend using TBB
also inside the application. Further, the join mode only works properly
starting with TBB v4.4 Update 1. For earlier TBB versions, threads that
call `rtcJoinCommitScene` to join a running build will just trigger the
build and wait for the build to finish. Further, old TBB versions with
`TBB_INTERFACE_VERSION_MAJOR < 8` do not support `rtcJoinCommitScene`,
and invoking this function will result in an error.

When using Embree with the internal tasking system, only threads that
call `rtcJoinCommitScene` will perform the build operation, and no
additional worker threads will be scheduled.

When using Embree with the Parallel Patterns Library (PPL),
`rtcJoinCommitScene` is not supported and calling that function will
result in an error.

To detect whether `rtcJoinCommitScene` is supported, use the
`rtcGetDeviceProperty` function.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcCommitScene], [rtcGetDeviceProperty]

```{=tex}

```
rtcSetSceneProgressMonitorFunction
----------------------------------

#### NAME {#name}

    rtcSetSceneProgressMonitorFunction - registers a callback
      to track build progress

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    typedef bool (*RTCProgressMonitorFunction)(
      void* ptr,
      double n
    );

    void rtcSetSceneProgressMonitorFunction(
      RTCScene scene,
      RTCProgressMonitorFunction progress,
      void* userPtr
    );

#### DESCRIPTION {#description}

Embree supports a progress monitor callback mechanism that can be used
to report progress of hierarchy build operations and to cancel build
operations.

The `rtcSetSceneProgressMonitorFunction` registers a progress monitor
callback function (`progress` argument) with payload (`userPtr`
argument) for the specified scene (`scene` argument).

Only a single callback function can be registered per scene, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

Once registered, Embree will invoke the callback function multiple
times during hierarchy build operations of the scene, by passing the
payload as set at registration time (`userPtr` argument), and a double
in the range $[0, 1]$ which estimates the progress of the operation
(`n` argument). The callback function might be called from multiple
threads concurrently.

When returning `true` from the callback function, Embree will continue
the build operation normally. When returning `false`, Embree will
cancel the build operation with the `RTC_ERROR_CANCELLED` error code.
Issuing multiple cancel requests for the same build operation is
allowed.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewScene]

```{=tex}

```
rtcSetSceneBuildQuality
-----------------------

#### NAME {#name}

    rtcSetSceneBuildQuality - sets the build quality for
      the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetSceneBuildQuality(
      RTCScene scene,
      enum RTCBuildQuality quality
    );

#### DESCRIPTION {#description}

The `rtcSetSceneBuildQuality` function sets the build quality
(`quality` argument) for the specified scene (`scene` argument).
Possible values for the build quality are:

-   `RTC_BUILD_QUALITY_LOW`: Create lower quality data structures,
    e.g. for dynamic scenes. A two-level spatial index structure is
    built when enabling this mode, which supports fast partial scene
    updates, and allows for setting a per-geometry build quality
    through the `rtcSetGeometryBuildQuality` function.

-   `RTC_BUILD_QUALITY_MEDIUM`: Default build quality for most usages.
    Gives a good compromise between build and render performance.

-   `RTC_BUILD_QUALITY_HIGH`: Create higher quality data structures for
    final-frame rendering. For certain geometry types this enables a
    spatial split BVH. When high quality mode is enabled, filter
    callbacks may be invoked multiple times for the same geometry.

Selecting a higher build quality results in better rendering
performance but slower scene commit times. The default build quality
for a scene is `RTC_BUILD_QUALITY_MEDIUM`.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryBuildQuality]

```{=tex}

```
rtcSetSceneFlags
----------------

#### NAME {#name}

    rtcSetSceneFlags - sets the flags for the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    enum RTCSceneFlags
    {
      RTC_SCENE_FLAG_NONE                    = 0,
      RTC_SCENE_FLAG_DYNAMIC                 = (1 << 0),
      RTC_SCENE_FLAG_COMPACT                 = (1 << 1),
      RTC_SCENE_FLAG_ROBUST                  = (1 << 2),
      RTC_SCENE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS = (1 << 3)
    };

    void rtcSetSceneFlags(RTCScene scene, enum RTCSceneFlags flags);

#### DESCRIPTION {#description}

The `rtcSetSceneFlags` function sets the scene flags (`flags` argument)
for the specified scene (`scene` argument). Possible scene flags are:

-   `RTC_SCENE_FLAG_NONE`: No flags set.

-   `RTC_SCENE_FLAG_DYNAMIC`: Provides better build performance for
    dynamic scenes (but also higher memory consumption).

-   `RTC_SCENE_FLAG_COMPACT`: Uses compact acceleration structures and
    avoids algorithms that consume much memory.

-   `RTC_SCENE_FLAG_ROBUST`: Uses acceleration structures that allow
    for robust traversal, and avoids optimizations that reduce
    arithmetic accuracy. This mode is typically used for avoiding
    artifacts caused by rays shooting through edges of neighboring
    primitives.

-   `RTC_SCENE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS`: Enables scene
    support for filter functions passed as argument to the traversal
    functions. See Section [rtcInitIntersectArguments] and
    [rtcInitOccludedArguments] for more details.

Multiple flags can be enabled using an `or` operation,
e.g. `RTC_SCENE_FLAG_COMPACT | RTC_SCENE_FLAG_ROBUST`.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetSceneFlags]

```{=tex}

```
rtcGetSceneFlags
----------------

#### NAME {#name}

    rtcGetSceneFlags - returns the flags of the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    enum RTCSceneFlags rtcGetSceneFlags(RTCScene scene);

#### DESCRIPTION {#description}

Queries the flags of a scene. This function can be useful when setting
individual flags, e.g. to just set the robust mode without changing
other flags the following way:

    RTCSceneFlags flags = rtcGetSceneFlags(scene);
    rtcSetSceneFlags(scene, RTC_SCENE_FLAG_ROBUST | flags);

#### EXIT STATUS {#exit-status}

On failure `RTC_SCENE_FLAG_NONE` is returned and an error code is set
that can be queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetSceneFlags]

```{=tex}

```
rtcGetSceneBounds
-----------------

#### NAME {#name}

    rtcGetSceneBounds - returns the axis-aligned bounding box of the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCORE_ALIGN(16) RTCBounds
    {
      float lower_x, lower_y, lower_z, align0;
      float upper_x, upper_y, upper_z, align1;
    };

    void rtcGetSceneBounds(
      RTCScene scene,
      struct RTCBounds* bounds_o
    );

#### DESCRIPTION {#description}

The `rtcGetSceneBounds` function queries the axis-aligned bounding box
of the specified scene (`scene` argument) and stores that bounding box
to the provided destination pointer (`bounds_o` argument). The stored
bounding box consists of lower and upper bounds for the x, y, and z
dimensions as specified by the `RTCBounds` structure.

The provided destination pointer must be aligned to 16 bytes. The
function may be invoked only after committing the scene; otherwise the
result is undefined.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetSceneLinearBounds], [rtcCommitScene], [rtcJoinCommitScene]

```{=tex}

```
rtcGetSceneLinearBounds
-----------------------

#### NAME {#name}

    rtcGetSceneLinearBounds - returns the linear bounds of the scene

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCORE_ALIGN(16) RTCLinearBounds
    {
      RTCBounds bounds0;
      RTCBounds bounds1;
    };

    void rtcGetSceneLinearBounds(
      RTCScene scene,
      struct RTCLinearBounds* bounds_o
    );

#### DESCRIPTION {#description}

The `rtcGetSceneLinearBounds` function queries the linear bounds of the
specified scene (`scene` argument) and stores them to the provided
destination pointer (`bounds_o` argument). The stored linear bounds
consist of bounding boxes for time 0 (`bounds0` member) and time 1
(`bounds1` member) as specified by the `RTCLinearBounds` structure.
Linearly interpolating these bounds to a specific time `t` yields
bounds for the geometry at that time.

The provided destination pointer must be aligned to 16 bytes. The
function may be called only after committing the scene, otherwise the
result is undefined.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetSceneBounds], [rtcCommitScene], [rtcJoinCommitScene]

```{=tex}

```
rtcNewGeometry
--------------

#### NAME {#name}

    rtcNewGeometry - creates a new geometry object

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    enum RTCGeometryType
    {
     RTC_GEOMETRY_TYPE_TRIANGLE,
     RTC_GEOMETRY_TYPE_QUAD,
     RTC_GEOMETRY_TYPE_SUBDIVISION,
     RTC_GEOMETRY_TYPE_FLAT_LINEAR_CURVE,
     RTC_GEOMETRY_TYPE_FLAT_BEZIER_CURVE,
     RTC_GEOMETRY_TYPE_FLAT_BSPLINE_CURVE,
     RTC_GEOMETRY_TYPE_FLAT_HERMITE_CURVE,
     RTC_GEOMETRY_TYPE_FLAT_CATMULL_ROM_CURVE,
     RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BEZIER_CURVE,
     RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BSPLINE_CURVE,
     RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_HERMITE_CURVE,
     RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_CATMULL_ROM_CURVE,
     RTC_GEOMETRY_TYPE_CONE_LINEAR_CURVE,
     RTC_GEOMETRY_TYPE_ROUND_LINEAR_CURVE,
     RTC_GEOMETRY_TYPE_ROUND_BEZIER_CURVE,
     RTC_GEOMETRY_TYPE_ROUND_BSPLINE_CURVE,
     RTC_GEOMETRY_TYPE_ROUND_HERMITE_CURVE,
     RTC_GEOMETRY_TYPE_ROUND_CATMULL_ROM_CURVE,
     RTC_GEOMETRY_TYPE_GRID,
     RTC_GEOMETRY_TYPE_SPHERE_POINT,
     RTC_GEOMETRY_TYPE_DISC_POINT,
     RTC_GEOMETRY_TYPE_ORIENTED_DISC_POINT,
     RTC_GEOMETRY_TYPE_USER,
     RTC_GEOMETRY_TYPE_INSTANCE,
     RTC_GEOMETRY_TYPE_INSTANCE_ARRAY,
    };

    RTCGeometry rtcNewGeometry(
      RTCDevice device,
      enum RTCGeometryType type
    );

#### DESCRIPTION {#description}

Geometries are objects that represent an array of primitives of the
same type. The `rtcNewGeometry` function creates a new geometry of
specified type (`type` argument) bound to the specified device
(`device` argument) and returns a handle to this geometry. The geometry
object is reference counted with an initial reference count of 1. The
geometry handle can be released using the `rtcReleaseGeometry` API
call.

Supported geometry types are triangle meshes
(`RTC_GEOMETRY_TYPE_TRIANGLE` type), quad meshes (triangle pairs)
(`RTC_GEOMETRY_TYPE_QUAD` type), Catmull-Clark subdivision surfaces
(`RTC_GEOMETRY_TYPE_SUBDIVISION` type), curve geometries with different
bases (`RTC_GEOMETRY_TYPE_FLAT_LINEAR_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_BEZIER_CURVE`,\
`RTC_GEOMETRY_TYPE_FLAT_BSPLINE_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_HERMITE_CURVE`,\
`RTC_GEOMETRY_TYPE_FLAT_CATMULL_ROM_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BSPLINE_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_HERMITE_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_CATMULL_ROM_CURVE`,
`RTC_GEOMETRY_TYPE_CONE_LINEAR_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_LINEAR_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_BSPLINE_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_HERMITE_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_CATMULL_ROM_CURVE` types) grid meshes
(`RTC_GEOMETRY_TYPE_GRID`), point geometries
(`RTC_GEOMETRY_TYPE_SPHERE_POINT`, `RTC_GEOMETRY_TYPE_DISC_POINT`,
`RTC_TYPE_ORIENTED_DISC_POINT`), user-defined geometries
(`RTC_GEOMETRY_TYPE_USER`), instances (`RTC_GEOMETRY_TYPE_INSTANCE`),
and instance arrays (`RTC_GEOMETRY_TYPE_INSTANCE_ARRAY`).

The types `RTC_GEOMETRY_TYPE_ROUND_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_BSPLINE_CURVE`, and
`RTC_GEOMETRY_TYPE_ROUND_CATMULL_ROM_CURVE` will treat the curve as a
sweep surface of a varying-radius circle swept tangentially along the
curve. The types `RTC_GEOMETRY_TYPE_FLAT_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_BSPLINE_CURVE`, and
`RTC_GEOMETRY_TYPE_FLAT_CATMULL_ROM_CURVE` use ray-facing ribbons as a
faster-to-intersect approximation.

After construction, geometries are enabled by default and not attached
to any scene. Geometries can be disabled (`rtcDisableGeometry` call),
and enabled again (`rtcEnableGeometry` call). A geometry can be
attached to multiple scenes using the `rtcAttachGeometry` call (or
`rtcAttachGeometryByID` call), and detached using the
`rtcDetachGeometry` call. During attachment, a geometry ID is assigned
to the geometry (or assigned by the user when using the
`rtcAttachGeometryByID` call), which uniquely identifies the geometry
inside that scene. This identifier is returned when primitives of the
geometry are hit in later ray queries for the scene.

Geometries can also be modified, including their vertex and index
buffers. After modifying a buffer, `rtcUpdateGeometryBuffer` must be
called to notify that the buffer got modified.

The application can use the `rtcSetGeometryUserData` function to set a
user data pointer to its own geometry representation, and later read
out this pointer using the `rtcGetGeometryUserData` function.

After setting up the geometry or modifying it, `rtcCommitGeometry` must
be called to finish the geometry setup. After committing the geometry,
vertex data interpolation can be performed using the `rtcInterpolate`
and `rtcInterpolateN` functions.

A build quality can be specified for a geometry using the
`rtcSetGeometryBuildQuality` function, to balance between acceleration
structure build performance and ray query performance. The build
quality per geometry will be used if a two-level acceleration structure
is built internally, which is the case if the `RTC_BUILD_QUALITY_LOW`
is set as the scene build quality. See Section
[rtcSetSceneBuildQuality] for more details.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcEnableGeometry], [rtcDisableGeometry], [rtcAttachGeometry],
[rtcAttachGeometryByID], [rtcUpdateGeometryBuffer],
[rtcSetGeometryUserData], [rtcGetGeometryUserData],
[rtcCommitGeometry], [rtcInterpolate], [rtcInterpolateN],
[rtcSetGeometryBuildQuality], [rtcSetSceneBuildQuality],
[RTC\_GEOMETRY\_TYPE\_TRIANGLE], [RTC\_GEOMETRY\_TYPE\_QUAD],
[RTC\_GEOMETRY\_TYPE\_SUBDIVISION], [RTC\_GEOMETRY\_TYPE\_CURVE],
[RTC\_GEOMETRY\_TYPE\_GRID], [RTC\_GEOMETRY\_TYPE\_POINT],
[RTC\_GEOMETRY\_TYPE\_USER], [RTC\_GEOMETRY\_TYPE\_INSTANCE],
[RTC\_GEOMETRY\_TYPE\_INSTANCE\_ARRAY]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_TRIANGLE
-----------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_TRIANGLE - triangle geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
      rtcNewGeometry(device, RTC_GEOMETRY_TYPE_TRIANGLE);

#### DESCRIPTION {#description}

Triangle meshes are created by passing `RTC_GEOMETRY_TYPE_TRIANGLE` to
the `rtcNewGeometry` function call. The triangle indices can be
specified by setting an index buffer (`RTC_BUFFER_TYPE_INDEX` type) and
the triangle vertices by setting a vertex buffer
(`RTC_BUFFER_TYPE_VERTEX` type). See `rtcSetGeometryBuffer` and
`rtcSetSharedGeometryBuffer` for more details on how to set buffers.
The index buffer must contain an array of three 32-bit indices per
triangle (`RTC_FORMAT_UINT3` format) and the number of primitives is
inferred from the size of that buffer. The vertex buffer must contain
an array of single precision `x`, `y`, `z` floating point coordinates
(`RTC_FORMAT_FLOAT3` format), and the number of vertices are inferred
from the size of that buffer. The vertex buffer can be at most 16 GB
large.

The parametrization of a triangle uses the first vertex `p0` as base
point, the vector `p1 - p0` as u-direction and the vector `p2 - p0` as
v-direction. Thus vertex attributes `t0,t1,t2` can be linearly
interpolated over the triangle the following way:

    t_uv = (1-u-v)*t0 + u*t1 + v*t2
         = t0 + u*(t1-t0) + v*(t2-t0)

A triangle whose vertices are laid out counter-clockwise has its
geometry normal pointing upwards outside the front face, like
illustrated in the following picture:

![][imgTriangleUV]

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` call. Then a vertex
buffer for each time step can be set using different buffer slots, and
all these buffers have to have the same stride and size.

Also see tutorial [Triangle Geometry] for an example of how to create
triangle meshes.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that be get
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_QUAD
-------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_QUAD - quad geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
      rtcNewGeometry(device, RTC_GEOMETRY_TYPE_QUAD);

#### DESCRIPTION {#description}

Quad meshes are created by passing `RTC_GEOMETRY_TYPE_QUAD` to the
`rtcNewGeometry` function call. The quad indices can be specified by
setting an index buffer (`RTC_BUFFER_TYPE_INDEX` type) and the quad
vertices by setting a vertex buffer (`RTC_BUFFER_TYPE_VERTEX` type).
See `rtcSetGeometryBuffer` and `rtcSetSharedGeometryBuffer` for more
details on how to set buffers. The index buffer contains an array of
four 32-bit indices per quad (`RTC_FORMAT_UINT4` format), and the
number of primitives is inferred from the size of that buffer. The
vertex buffer contains an array of single precision `x`, `y`, `z`
floating point coordinates (`RTC_FORMAT_FLOAT3` format), and the number
of vertices is inferred from the size of that buffer. The vertex buffer
can be at most 16 GB large.

A quad is internally handled as a pair of two triangles `v0,v1,v3` and
`v2,v3,v1`, with the `u'`/`v'` coordinates of the second triangle
corrected by `u = 1-u'` and `v = 1-v'` to produce a quad
parametrization where `u` and `v` are in the range 0 to 1. Thus the
parametrization of a quad uses the first vertex `p0` as base point, and
the vector `p1 - p0` as `u`-direction, and `p3 - p0` as v-direction.
Thus vertex attributes `t0,t1,t2,t3` can be bilinearly interpolated
over the quadrilateral the following way:

    t_uv = (1-v)((1-u)*t0 + u*t1) + v*((1-u)*t3 + u*t2)

Mixed triangle/quad meshes are supported by encoding a triangle as a
quad, which can be achieved by replicating the last triangle vertex
(`v0,v1,v2` -\> `v0,v1,v2,v2`). This way the second triangle is a line
(which can never get hit), and the parametrization of the first
triangle is compatible with the standard triangle parametrization.

A quad whose vertices are laid out counter-clockwise has its geometry
normal pointing upwards outside the front face, like illustrated in the
following picture.

![][imgQuadUV]

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` call. Then a vertex
buffer for each time step can be set using different buffer slots, and
all these buffers must have the same stride and size.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_GRID
-------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_GRID - grid geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
      rtcNewGeometry(device, RTC_GEOMETRY_TYPE_GRID);

#### DESCRIPTION {#description}

Grid meshes are created by passing `RTC_GEOMETRY_TYPE_GRID` to the
`rtcNewGeometry` function call, and contain an array of grid
primitives. This array of grids can be specified by setting up a grid
buffer (with `RTC_BUFFER_TYPE_GRID` type and `RTC_FORMAT_GRID` format)
and the grid mesh vertices by setting a vertex buffer
(`RTC_BUFFER_TYPE_VERTEX` type). See `rtcSetGeometryBuffer` and
`rtcSetSharedGeometryBuffer` for more details on how to set buffers.
The number of grid primitives in the grid mesh is inferred from the
size of the grid buffer.

The vertex buffer contains an array of single precision `x`, `y`, `z`
floating point coordinates (`RTC_FORMAT_FLOAT3` format), and the number
of vertices is inferred from the size of that buffer.

Each grid in the grid buffer is of the type `RTCGrid`:

    struct RTCGrid
    {
      unsigned int startVertexID;
      unsigned int stride;
      unsigned short width,height; 
    };

The `RTCGrid` structure describes a 2D grid of vertices (with respect
to the vertex buffer of the grid mesh). The `width` and `height`
members specify the number of vertices in u and v direction,
e.g. setting both `width` and `height` to 3 sets up a 3×3 vertex grid.
The maximum allowed `width` and `height` is 32767. The `startVertexID`
specifies the ID of the top-left vertex in the vertex grid, while the
`stride` parameter specifies a stride (in number of vertices) used to
step to the next row.

A vertex grid of dimensions `width` and `height` is treated as a
`(width-1)` x `(height-1)` grid of `quads` (triangle-pairs), with the
same shared edge handling as for regular quad meshes. However, the
`u`/`v` coordinates have the uniform range `[0..1]` for an entire
vertex grid. The `u` direction follows the `width` of the grid while
the `v` direction the `height`.

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` call. Then a vertex
buffer for each time step can be set using different buffer slots, and
all these buffers must have the same stride and size.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_SUBDIVISION
--------------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_SUBDIVISION - subdivision geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
      rtcNewGeometry(device, RTC_GEOMETRY_TYPE_SUBDIVISION);

#### DESCRIPTION {#description}

Catmull-Clark subdivision meshes are supported, including support for
edge creases, vertex creases, holes, non-manifold geometry, and
face-varying interpolation. The number of vertices per face can be in
the range of 3 to 15 vertices (triangles, quadrilateral, pentagons,
etc).

Subdivision meshes are created by passing
`RTC_GEOMETRY_TYPE_SUBDIVISION` to the `rtcNewGeometry` function.
Various buffers need to be set by the application to set up the
subdivision mesh. See `rtcSetGeometryBuffer` and
`rtcSetSharedGeometryBuffer` for more details on how to set buffers.
The face buffer (`RTC_BUFFER_TYPE_FACE` type and `RTC_FORMAT_UINT`
format) contains the number of edges/indices of each face (3 to 15),
and the number of faces is inferred from the size of this buffer. The
index buffer (`RTC_BUFFER_TYPE_INDEX` type) contains multiple (3 to 15)
32-bit vertex indices (`RTC_FORMAT_UINT` format) for each face, and the
number of edges is inferred from the size of this buffer. The vertex
buffer (`RTC_BUFFER_TYPE_VERTEX` type) stores an array of single
precision `x`, `y`, `z` floating point coordinates (`RTC_FORMAT_FLOAT3`
format), and the number of vertices is inferred from the size of this
buffer.

Optionally, the application may set additional index buffers using
different buffer slots if multiple topologies are required for
face-varying interpolation. The standard vertex buffers
(`RTC_BUFFER_TYPE_VERTEX`) are always bound to the geometry topology
(topology 0) thus use `RTC_BUFFER_TYPE_INDEX` with buffer slot 0. User
vertex data interpolation may use different topologies as described
later.

Optionally, the application can set up the hole buffer
(`RTC_BUFFER_TYPE_HOLE`) which contains an array of 32-bit indices
(`RTC_FORMAT_UINT` format) of faces that should be considered
non-existing in all topologies. The number of holes is inferred from
the size of this buffer.

Optionally, the application can fill the level buffer
(`RTC_BUFFER_TYPE_LEVEL`) with a tessellation rate for each of the
edges of each face. This buffer must have the same size as the index
buffer. The tessellation level is a positive floating point value
(`RTC_FORMAT_FLOAT` format) that specifies how many quads along the
edge should be generated during tessellation. If no level buffer is
specified, a level of 1 is used. The maximally supported edge level is
4096, and larger levels are clamped to that value. Note that edges may
be shared between (typically 2) faces. To guarantee a watertight
tessellation, the level of these shared edges should be identical. A
uniform tessellation rate for an entire subdivision mesh can be set by
using the `rtcSetGeometryTessellationRate` function. The existence of a
level buffer has precedence over the uniform tessellation rate.

Optionally, the application can fill the sparse edge crease buffers to
make edges appear sharper. The edge crease index buffer
(`RTC_BUFFER_TYPE_EDGE_CREASE_INDEX`) contains an array of pairs of
32-bit vertex indices (`RTC_FORMAT_UINT2` format) that specify
unoriented edges in the geometry topology. The edge crease weight
buffer (`RTC_BUFFER_TYPE_EDGE_CREASE_WEIGHT`) stores for each of these
crease edges a positive floating point weight (`RTC_FORMAT_FLOAT`
format). The number of edge creases is inferred from the size of these
buffers, which has to be identical. The larger a weight, the sharper
the edge. Specifying a weight of infinity is supported and marks an
edge as infinitely sharp. Storing an edge multiple times with the same
crease weight is allowed, but has lower performance. Storing an edge
multiple times with different crease weights results in undefined
behavior. For a stored edge (i,j), the reverse direction edges (j,i) do
not have to be stored, as both are considered the same unoriented edge.
Edge crease features are shared between all topologies.

Optionally, the application can fill the sparse vertex crease buffers
to make vertices appear sharper. The vertex crease index buffer
(`RTC_BUFFER_TYPE_VERTEX_CREASE_INDEX`), contains an array of 32-bit
vertex indices (`RTC_FORMAT_UINT` format) to specify a set of vertices
from the geometry topology. The vertex crease weight buffer
(`RTC_BUFFER_TYPE_VERTEX_CREASE_WEIGHT`) specifies for each of these
vertices a positive floating point weight (`RTC_FORMAT_FLOAT` format).
The number of vertex creases is inferred from the size of these
buffers, and has to be identical. The larger a weight, the sharper the
vertex. Specifying a weight of infinity is supported and makes the
vertex infinitely sharp. Storing a vertex multiple times with the same
crease weight is allowed, but has lower performance. Storing a vertex
multiple times with different crease weights results in undefined
behavior. Vertex crease features are shared between all topologies.

Subdivision modes can be used to force linear interpolation for parts
of the subdivision mesh; see `rtcSetGeometrySubdivisionMode` for more
details.

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` call. Then a vertex
buffer for each time step can be set using different buffer slots, and
all these buffers have to have the same stride and size.

Also see tutorial [Subdivision Geometry] for an example of how to
create subdivision surfaces.

#### Parametrization

The parametrization for subdivision faces is different for
quadrilaterals and non-quadrilateral faces.

The parametrization of a quadrilateral face uses the first vertex `p0`
as base point, and the vector `p1 - p0` as u-direction and `p3 - p0` as
v-direction.

The parametrization for all other face types (with number of vertices
not equal 4), have a special parametrization where the subpatch ID `n`
(of the `n`-th quadrilateral that would be obtained by a single
subdivision step) and the local hit location inside this quadrilateral
are encoded in the UV coordinates. The following code extracts the
sub-patch ID `i` and local UVs of this subpatch:

    unsigned int l = floorf(0.5f*U);
    unsigned int h = floorf(0.5f*V);
    unsigned int i = 4*h+l;
    float u = 2.0f*fracf(0.5f*U)-0.5f;
    float v = 2.0f*fracf(0.5f*V)-0.5f;

This encoding allows local subpatch UVs to be in the range `[-0.5,1.5[`
thus negative subpatch UVs can be passed to `rtcInterpolate` to sample
subpatches slightly out of bounds. This can be useful to calculate
derivatives using finite differences if required. The encoding further
has the property that one can just move the value `u` (or `v`) on a
subpatch by adding `du` (or `dv`) to the special UV encoding as long as
it does not fall out of the `[-0.5,1.5[` range.

To smoothly interpolate vertex attributes over the subdivision surface
we recommend using the `rtcInterpolate` function, which will apply the
standard subdivision rules for interpolation and automatically takes
care of the special UV encoding for non-quadrilaterals.

#### Face-Varying Data

Face-varying interpolation is supported through multiple topologies per
subdivision mesh and binding such topologies to vertex attribute
buffers to interpolate. This way, texture coordinates may use a
different topology with additional boundaries to construct separate UV
regions inside one subdivision mesh.

Each such topology `i` has a separate index buffer (specified using
`RTC_BUFFER_TYPE_INDEX` with buffer slot `i`) and separate subdivision
mode that can be set using `rtcSetGeometrySubdivisionMode`. A vertex
attribute buffer `RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE` bound to a buffer
slot `j` can be assigned to use a topology for interpolation using the
`rtcSetGeometryVertexAttributeTopology` call.

The face buffer (`RTC_BUFFER_TYPE_FACE` type) is shared between all
topologies, which means that the `n`-th primitive always has the same
number of vertices (e.g. being a triangle or a quad) for each topology.
However, the indices of the topologies themselves may be different.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_CURVE
--------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_FLAT_LINEAR_CURVE -
      flat curve geometry with linear basis

    RTC_GEOMETRY_TYPE_FLAT_BEZIER_CURVE -
      flat curve geometry with cubic Bézier basis

    RTC_GEOMETRY_TYPE_FLAT_BSPLINE_CURVE - 
      flat curve geometry with cubic B-spline basis

    RTC_GEOMETRY_TYPE_FLAT_HERMITE_CURVE - 
      flat curve geometry with cubic Hermite basis

    RTC_GEOMETRY_TYPE_FLAT_CATMULL_ROM_CURVE - 
      flat curve geometry with Catmull-Rom basis

    RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BEZIER_CURVE -
      flat normal oriented curve geometry with cubic Bézier basis

    RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BSPLINE_CURVE - 
      flat normal oriented curve geometry with cubic B-spline basis

    RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_HERMITE_CURVE - 
      flat normal oriented curve geometry with cubic Hermite basis

    RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_CATMULL_ROM_CURVE - 
      flat normal oriented curve geometry with Catmull-Rom basis

    RTC_GEOMETRY_TYPE_CONE_LINEAR_CURVE -
      capped cone curve geometry with linear basis - discontinuous at edge boundaries

    RTC_GEOMETRY_TYPE_ROUND_LINEAR_CURVE -
      capped cone curve geometry with linear basis and spherical ending

    RTC_GEOMETRY_TYPE_ROUND_BEZIER_CURVE -
      swept surface curve geometry with cubic Bézier basis

    RTC_GEOMETRY_TYPE_ROUND_BSPLINE_CURVE -
      swept surface curve geometry with cubic B-spline basis

    RTC_GEOMETRY_TYPE_ROUND_HERMITE_CURVE -
      swept surface curve geometry with cubic Hermite basis

    RTC_GEOMETRY_TYPE_ROUND_CATMULL_ROM_CURVE -
      swept surface curve geometry with Catmull-Rom basis

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_FLAT_LINEAR_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_FLAT_BEZIER_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_FLAT_BSPLINE_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_FLAT_HERMITE_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_FLAT_CATMULL_ROM_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BEZIER_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_BSPLINE_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_HERMITE_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_CATMULL_ROM_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_CONE_LINEAR_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_ROUND_LINEAR_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_ROUND_BEZIER_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_ROUND_BSPLINE_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_ROUND_HERMITE_CURVE);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_ROUND_CATMULL_ROM_CURVE);

#### DESCRIPTION {#description}

Curves with per vertex radii are supported with linear, cubic Bézier,
cubic B-spline, and cubic Hermite bases. Such curve geometries are
created by passing `RTC_GEOMETRY_TYPE_FLAT_LINEAR_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_BSPLINE_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_HERMITE_CURVE`,
`RTC_GEOMETRY_TYPE_FLAT_CATMULL_ROM_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_FLAT_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_FLAT_BSPLINE_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_FLAT_HERMITE_CURVE`,
`RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_FLAT_CATMULL_ROM_CURVE`,
`RTC_GEOMETRY_TYPE_CONE_LINEAR_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_LINEAR_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_BEZIER_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_BSPLINE_CURVE`,
`RTC_GEOMETRY_TYPE_ROUND_HERMITE_CURVE`, or
`RTC_GEOMETRY_TYPE_ROUND_CATMULL_ROM_CURVE` to the `rtcNewGeometry`
function. The curve indices can be specified through an index buffer
(`RTC_BUFFER_TYPE_INDEX`) and the curve vertices through a vertex
buffer (`RTC_BUFFER_TYPE_VERTEX`). For the Hermite basis a tangent
buffer (`RTC_BUFFER_TYPE_TANGENT`), normal oriented curves a normal
buffer (`RTC_BUFFER_TYPE_NORMAL`), and for normal oriented Hermite
curves a normal derivative buffer (`RTC_BUFFER_TYPE_NORMAL_DERIVATIVE`)
has to get specified additionally. See `rtcSetGeometryBuffer` and
`rtcSetSharedGeometryBuffer` for more details on how to set buffers.

The index buffer contains an array of 32-bit indices (`RTC_FORMAT_UINT`
format), each pointing to the first control vertex in the vertex
buffer, but also to the first tangent in the tangent buffer, and first
normal in the normal buffer if these buffers are present.

The vertex buffer stores each control vertex in the form of a single
precision position and radius stored in (`x`, `y`, `z`, `r`) order in
memory (`RTC_FORMAT_FLOAT4` format). The number of vertices is inferred
from the size of this buffer. The radii may be smaller than zero but
the interpolated radii should always be greater or equal to zero.
Similarly, the tangent buffer stores the derivative of each control
vertex (`x`, `y`, `z`, `r` order and `RTC_FORMAT_FLOAT4` format) and
the normal buffer stores a single precision normal per control vertex
(`x`, `y`, `z` order and `RTC_FORMAT_FLOAT3` format).

##### Linear Basis

For the linear basis the indices point to the first of 2 consecutive
control points in the vertex buffer. The first control point is the
start and the second control point the end of the line segment. When
constructing hair strands in this basis, the end-point can be shared
with the start of the next line segment.

For the linear basis the user optionally can provide a flags buffer of
type `RTC_BUFFER_TYPE_FLAGS` which contains bytes that encode if the
left neighbor segment (`RTC_CURVE_FLAG_NEIGHBOR_LEFT` flag) and/or
right neighbor segment (`RTC_CURVE_FLAG_NEIGHBOR_RIGHT` flags) exist
(see [RTCCurveFlags]). If this buffer is not set, than the left/right
neighbor bits are automatically calculated base on the index buffer
(left segment exists if segment(id-1)+1 == segment(id) and right
segment exists if segment(id+1)-1 == segment(id)).

A left neighbor segment is assumed to end at the start vertex of the
current segment, and to start at the previous vertex in the vertex
buffer. Similarly, the right neighbor segment is assumed to start at
the end vertex of the current segment, and to end at the next vertex in
the vertex buffer.

Only when the left and right bits are properly specified the current
segment can properly attach to the left and/or right neighbor,
otherwise the touching area may not get rendered properly.

##### Bézier Basis

For the cubic Bézier basis the indices point to the first of 4
consecutive control points in the vertex buffer. These control points
use the cubic Bézier basis, where the first control point represents
the start point of the curve, and the 4th control point the end point
of the curve. The Bézier basis is interpolating, thus the curve does go
exactly through the first and fourth control vertex.

##### B-spline Basis

For the cubic B-spline basis the indices point to the first of 4
consecutive control points in the vertex buffer. These control points
make up a cardinal cubic B-spline (implicit equidistant knot vector).
This basis is not interpolating, thus the curve does in general not go
through any of the control points directly. A big advantage of this
basis is that 3 control points can be shared for two continuous
neighboring curve segments, e.g. the curves (p0,p1,p2,p3) and
(p1,p2,p3,p4) are C1 continuous. This feature makes this basis a good
choice to construct continuous multi-segment curves, as memory
consumption can be kept minimal.

##### Hermite Basis

For the cubic Hermite basis the indices point to the first of 2
consecutive points in the vertex buffer, and the first of 2 consecutive
tangents in the tangent buffer. These two points and two tangents make
up a cubic Hermite curve. This basis is interpolating, thus does
exactly go through the first and second control point, and the first
order derivative at the begin and end matches exactly the value
specified in the tangent buffer. When connecting two segments
continuously, the end point and tangent of the previous segment can be
shared. Different versions of Catmull-Rom splines can be easily
constructed using the Hermite basis, by calculating a proper tangent
buffer from the control points.

##### Catmull-Rom Basis

For the Catmull-Rom basis the indices point to the first of 4
consecutive control points in the vertex buffer. This basis goes
through p1 and p2, with tangents (p2-p0)/2 and (p3-p1)/2.

##### Flat Curves

The `RTC_GEOMETRY_TYPE_FLAT_*` flat mode is a fast mode designed to
render distant hair. In this mode the curve is rendered as a connected
sequence of ray facing quads. Individual quads are considered to have
subpixel size, and zooming onto the curve might show geometric
artifacts. The number of quads to subdivide into can be specified
through the `rtcSetGeometryTessellationRate` function. By default the
tessellation rate is 4.

##### Normal Oriented Curves

The `RTC_GEOMETRY_TYPE_NORMAL_ORIENTED_*` mode is a mode designed to
render blades of grass. In this mode a vertex spline has to get
specified as for the previous modes, but additionally a normal spline
is required. If the Hermite basis is used, the `RTC_BUFFER_TYPE_NORMAL`
and `RTC_BUFFER_TYPE_NORMAL_DERIVATIVE` buffers have both to be set.

The curve is rendered as a flat band whose center approximately follows
the provided vertex spline, whose half width approximately follows the
provided radius spline, and whose normal orientation approximately
follows the provided normal spline.

To intersect the normal oriented curve, we perform a newton-raphson
style intersection of a ray with a tensor product surface of a linear
basis (perpendicular to the curve) and cubic Bézier basis (along the
curve). We use a guide curve and its derivatives to construct the
control points of that surface. The guide curve is defined by a sweep
surface defined by sweeping a line centered at the vertex spline
location along the curve. At each parameter value the half width of the
line matches the radius spline, and the direction matches the cross
product of the normal from the normal spline and tangent of the vertex
spline. Note that this construction does not work when the provided
normals are parallel to the curve direction. For this reason the
provided normals should best be kept as perpendicular to the curve
direction as possible. We further assume second order derivatives of
the center curve to be zero for this construction, as otherwise very
large curvatures occurring in corner cases, can thicken the constructed
curve significantly.

##### Round Curves

In the `RTC_GEOMETRY_TYPE_ROUND_*` round mode, a real geometric surface
is rendered for the curve, which is more expensive but allows closeup
views.

For the linear basis the round mode renders a cone that tangentially
touches a start-sphere and end-sphere. The start sphere is rendered
when no previous segments is indicated by the neighbor bits. The end
sphere is always rendered but parts that lie inside the next segment
are clipped away (if that next segment exists). This way a curve is
closed on both ends and the interior will render properly as long as
only neighboring segments penetrate into a segment. For this to work
properly it is important that the flags buffer is properly populated
with neighbor information.

For the cubic polynomial bases, the round mode renders a sweep surface
by sweeping a varying radius circle tangential along the curve. As a
limitation, the radius of the curve has to be smaller than the
curvature radius of the curve at each location on the curve.

The intersection with the curve segment stores the parametric hit
location along the curve segment as u-coordinate (range 0 to +1).

For flat curves, the v-coordinate is set to the normalized distance in
the range -1 to +1. For normal oriented curves the v-coordinate is in
the range 0 to 1. For the linear basis and in round mode the
v-coordinate is set to zero.

In flat mode, the geometry normal `Ng` is set to the tangent of the
curve at the hit location. In round mode and for normal oriented
curves, the geometry normal `Ng` is set to the non-normalized geometric
normal of the surface.

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` call. Then a vertex
buffer for each time step can be set using different buffer slots, and
all these buffers must have the same stride and size. For the Hermite
basis also a tangent buffer has to be set for each time step and for
normal oriented curves a normal buffer has to get specified for each
time step.

Also see tutorials [Hair] and [Curves] for examples of how to
create and use curve geometries.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [RTCCurveFlags]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_POINT
--------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_SPHERE_POINT -
      point geometry spheres

    RTC_GEOMETRY_TYPE_DISC_POINT -
      point geometry with ray-oriented discs

    RTC_GEOMETRY_TYPE_ORIENTED_DISC_POINT -
      point geometry with normal-oriented discs

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_SPHERE_POINT);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_DISC_POINT);
    rtcNewGeometry(device, RTC_GEOMETRY_TYPE_ORIENTED_DISC_POINT);

#### DESCRIPTION {#description}

Points with per vertex radii are supported with sphere, ray-oriented
discs, and normal-oriented discs geometric representations. Such point
geometries are created by passing `RTC_GEOMETRY_TYPE_SPHERE_POINT`,
`RTC_GEOMETRY_TYPE_DISC_POINT`, or
`RTC_GEOMETRY_TYPE_ORIENTED_DISC_POINT` to the `rtcNewGeometry`
function. The point vertices can be specified t through a vertex buffer
(`RTC_BUFFER_TYPE_VERTEX`). For the normal oriented discs a normal
buffer (`RTC_BUFFER_TYPE_NORMAL`) has to get specified additionally.
See `rtcSetGeometryBuffer` and `rtcSetSharedGeometryBuffer` for more
details on how to set buffers.

The vertex buffer stores each control vertex in the form of a single
precision position and radius stored in (`x`, `y`, `z`, `r`) order in
memory (`RTC_FORMAT_FLOAT4` format). The number of vertices is inferred
from the size of this buffer. Similarly, the normal buffer stores a
single precision normal per control vertex (`x`, `y`, `z` order and
`RTC_FORMAT_FLOAT3` format).

In the `RTC_GEOMETRY_TYPE_SPHERE_POINT` mode, a real geometric surface
is rendered for the curve, which is more expensive but allows closeup
views.

The `RTC_GEOMETRY_TYPE_DISC_POINT` flat mode is a fast mode designed to
render distant points. In this mode the point is rendered as a ray
facing disc.

The `RTC_GEOMETRY_TYPE_ORIENTED_DISC_POINT` mode is a mode designed as
a midpoint geometrically between ray facing discs and spheres. In this
mode the point is rendered as a normal oriented disc.

For all point types, only the hit distance and geometry normal is
returned as hit information, u and v are set to zero.

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` call. Then a vertex
buffer for each time step can be set using different buffer slots, and
all these buffers must have the same stride and size.

Also see tutorial [Points] for an example of how to create and use
point geometries.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_USER
-------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_USER - user geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
      rtcNewGeometry(device, RTC_GEOMETRY_TYPE_USER);

#### DESCRIPTION {#description}

User-defined geometries contain a number of user-defined primitives,
just like triangle meshes contain multiple triangles. The shape of the
user-defined primitives is specified through registered callback
functions, which enable extending Embree with arbitrary types of
primitives.

User-defined geometries are created by passing `RTC_GEOMETRY_TYPE_USER`
to the `rtcNewGeometry` function call. One has to set the number of
primitives (see `rtcSetGeometryUserPrimitiveCount`), a user data
pointer (see `rtcSetGeometryUserData`), a bounding function closure
(see `rtcSetGeometryBoundsFunction`), as well as user-defined intersect
(see `rtcSetGeometryIntersectFunction`) and occluded (see
`rtcSetGeometryOccludedFunction`) callback functions. The bounding
function is used to query the bounds of all time steps of a user
primitive, while the intersect and occluded callback functions are
called to intersect the primitive with a ray. The user data pointer is
passed to each callback invocation and can be used to point to the
application's representation of the user geometry.

The creation of a user geometry typically looks the following:

    RTCGeometry geometry = rtcNewGeometry(device, RTC_GEOMETRY_TYPE_USER);
    rtcSetGeometryUserPrimitiveCount(geometry, numPrimitives);
    rtcSetGeometryUserData(geometry, userGeometryRepresentation);
    rtcSetGeometryBoundsFunction(geometry, boundsFunction);
    rtcSetGeometryIntersectFunction(geometry, intersectFunction);
    rtcSetGeometryOccludedFunction(geometry, occludedFunction);

Please have a look at the `rtcSetGeometryBoundsFunction`,
`rtcSetGeometryIntersectFunction`, and `rtcSetGeometryOccludedFunction`
functions on the implementation of the callback functions.

Primitives of a user geometry are ignored during rendering when their
bounds are empty, thus bounds have lower\>upper in at least one
dimension.

See tutorial [User Geometry] for an example of how to use the
user-defined geometries.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcSetGeometryUserPrimitiveCount],
[rtcSetGeometryUserData], [rtcSetGeometryBoundsFunction],
[rtcSetGeometryIntersectFunction], [rtcSetGeometryOccludedFunction]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_INSTANCE
-----------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_INSTANCE - instance geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
       rtcNewGeometry(device, RTC_GEOMETRY_TYPE_INSTANCE);

#### DESCRIPTION {#description}

Embree supports instancing of scenes using affine transformations (3×3
matrix plus translation). As the instanced scene is stored only a
single time, even if instanced to multiple locations, this feature can
be used to create very complex scenes with small memory footprint.

Embree supports both single-level instancing and multi-level
instancing. The maximum instance nesting depth is
`RTC_MAX_INSTANCE_LEVEL_COUNT`; it can be configured at compile-time
using the constant `EMBREE_MAX_INSTANCE_LEVEL_COUNT`. Users should
adapt this constant to their needs: instances nested any deeper are
silently ignored in release mode, and cause assertions in debug mode.

Instances are created by passing `RTC_GEOMETRY_TYPE_INSTANCE` to the
`rtcNewGeometry` function call. The instanced scene can be set using
the `rtcSetGeometryInstancedScene` call, and the affine transformation
can be set using the `rtcSetGeometryTransform` function.

Please note that `rtcCommitScene` on the instanced scene should be
called first, followed by `rtcCommitGeometry` on the instance, followed
by `rtcCommitScene` for the top-level scene containing the instance.

If a ray hits the instance, the `geomID` and `primID` members of the
hit are set to the geometry ID and primitive ID of the hit primitive in
the instanced scene, and the `instID` member of the hit is set to the
geometry ID of the instance in the top-level scene.

The instancing scheme can also be implemented using user geometries. To
achieve this, the user geometry code should set the `instID` member of
the ray query context to the geometry ID of the instance, then trace
the transformed ray, and finally set the `instID` field of the ray
query context again to -1. The `instID` field is copied automatically
by each primitive intersector into the `instID` field of the hit
structure when the primitive is hit. See the [User Geometry] tutorial
for an example.

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` function. Then a
transformation for each time step can be specified using the
`rtcSetGeometryTransform` function.

See tutorials [Instanced Geometry] and [Multi Level Instancing] for
examples of how to use instances.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcSetGeometryInstancedScene],
[rtcSetGeometryTransform]

```{=tex}

```
RTC\_GEOMETRY\_TYPE\_INSTANCE\_ARRAY
------------------------------------

#### NAME {#name}

    RTC_GEOMETRY_TYPE_INSTANCE_ARRAY - instance array geometry type

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCGeometry geometry =
       rtcNewGeometry(device, RTC_GEOMETRY_TYPE_INSTANCE_ARRAY);

#### DESCRIPTION {#description}

Embree supports instance arrays, which is a more memory efficient way
to represent large amounts of instances of the same or a small set of
(sub)scenes. The main difference to regular Embree instances is that
Embree instance arrays have a buffer of transformations (either affine
transformations or quaternion decompositions
[RTCQuaternionDecomposition]) that can be allocated by Embree or a
shared buffer, similar to vertex buffers for triangle meshes.
Optionally, instead of instancing only one scene, an instance array can
instance multiple scenes by passing an array of scenes and a
corresponding index buffer that specifies which instance of the
instance array instances which of the scenes in the scenes array.

Instance arrays are created by passing
`RTC_GEOMETRY_TYPE_INSTANCE_ARRAY` to the `rtcNewGeometry` function
call. The instanced scene can be either be set using the
`rtcSetGeometryInstancedScene` call, or if multiple scenes should be
instanced by passing an array of scenes using
`rtcSetGeometryInstancedScenes`. The latter also requires to specify an
index buffer using `rtcSetNewGeometryBuffer` or
`rtcSetSharedGeometryBuffer` with `RTC_BUFFER_TYPE_INDEX` as the buffer
type.

Because the transformation information can become large for a large
amount of instances, the instance array allows to share the
transformation buffer between the user application and Embree. It can
be either stored in a buffer created by Embree with
`rtcSetNewGeometryBuffer` or an already existing buffer can be shared
using `rtcSetSharedGeometryBuffer`. In either case, the buffer type has
to be `RTC_BUFFER_TYPE_TRANSFORM` and the allowed formats are
`RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR`, `RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR`,
`RTC_FORMAT_FLOAT3X4_ROW_MAJOR`, and
`RTC_FORMAT_QUATERNION_DECOMPOSITION`. Embree will not modify the data
in the transformation buffer.

Embree instance arrays support both single-level instancing and
multi-level instancing. The maximum instance nesting depth is
`RTC_MAX_INSTANCE_LEVEL_COUNT`; it can be configured at compile-time
using the constant `EMBREE_MAX_INSTANCE_LEVEL_COUNT`. Users should
adapt this constant to their needs: instances nested any deeper are
silently ignored in release mode, and cause assertions in debug mode.

Please note that `rtcCommitScene` on the instanced scene(s) should be
called first, followed by `rtcCommitGeometry` on the instance array,
followed by `rtcCommitScene` for the top-level scene containing the
instance array.

If a ray hits the instance, the `geomID` and `primID` members of the
hit are set to the geometry ID and primitive ID of the hit primitive in
the instanced scene. The `instID` member of the hit is set to the
geometry ID of the instance array in the top-level scene and the
`instPrimID` member is set to the index of the hit instance of the
instance array.

For multi-segment motion blur, the number of time steps must be first
specified using the `rtcSetGeometryTimeStepCount` function. Then a
transformation for each time step can be specified using the
`rtcSetNewGeometryBuffer` or `rtcSetSharedGeometryBuffer` function and
passing the time step as the `slot` parameter of these calls.

See the [Instance Array Geometry] tutorial for an example of how to
use instance arrays.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcSetGeometryInstancedScene],
[rtcSetGeometryInstancedScenes], [rtcSetNewGeometryBuffer],
[rtcSetSharedGeometryBuffer], [rtcGetGeometryTransformEx]

```{=tex}

```
RTCCurveFlags
-------------

#### NAME {#name}

    RTCCurveFlags - per segment flags for curve geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    enum RTCCurveFlags
    {
      RTC_CURVE_FLAG_NEIGHBOR_LEFT  = (1 << 0), 
      RTC_CURVE_FLAG_NEIGHBOR_RIGHT = (1 << 1) 
    };

#### DESCRIPTION {#description}

The RTCCurveFlags type is used for linear curves to determine if the
left and/or right neighbor segment exist. Therefore one attaches a
buffer of type RTC\_BUFFER\_TYPE\_FLAGS to the curve geometry which
stores an individual byte per curve segment.

If the RTC\_CURVE\_FLAG\_NEIGHBOR\_LEFT flag in that byte is enabled
for a curve segment, then the left segment exists (which starts one
vertex before the start vertex of the current curve) and the current
segment is rendered to properly attach to that segment.

If the RTC\_CURVE\_FLAG\_NEIGHBOR\_RIGHT flag in that byte is enabled
for a curve segment, then the right segment exists (which ends one
vertex after the end vertex of the current curve) and the current
segment is rendered to properly attach to that segment.

When not properly specifying left and right flags for linear curves,
the rendering at the ending of these curves may not look correct, in
particular when round linear curves are viewed from the inside.

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_CURVE]

```{=tex}

```
rtcRetainGeometry
-----------------

#### NAME {#name}

    rtcRetainGeometry - increments the geometry reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcRetainGeometry(RTCGeometry geometry);

#### DESCRIPTION {#description}

Geometry objects are reference counted. The `rtcRetainGeometry`
function increments the reference count of the passed geometry object
(`geometry` argument). This function together with `rtcReleaseGeometry`
allows to use the internal reference counting in a C++ wrapper class to
handle the ownership of the object.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcReleaseGeometry]

```{=tex}

```
rtcReleaseGeometry
------------------

#### NAME {#name}

    rtcReleaseGeometry - decrements the geometry reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcReleaseGeometry(RTCGeometry geometry);

#### DESCRIPTION {#description}

Geometry objects are reference counted. The `rtcReleaseGeometry`
function decrements the reference count of the passed geometry object
(`geometry` argument). When the reference count falls to 0, the
geometry gets destroyed.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcRetainGeometry]

```{=tex}

```
rtcCommitGeometry
-----------------

#### NAME {#name}

    rtcCommitGeometry - commits geometry changes

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcCommitGeometry(RTCGeometry geometry);

#### DESCRIPTION {#description}

The `rtcCommitGeometry` function is used to commit all geometry changes
performed to a geometry (`geometry` parameter). After a geometry gets
modified, this function must be called to properly update the internal
state of the geometry to perform interpolations using `rtcInterpolate`
or to commit a scene containing the geometry using `rtcCommitScene`.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcInterpolate], [rtcCommitScene]

```{=tex}

```
rtcEnableGeometry
-----------------

#### NAME {#name}

    rtcEnableGeometry - enables the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcEnableGeometry(RTCGeometry geometry);

#### DESCRIPTION {#description}

The `rtcEnableGeometry` function enables the specified geometry
(`geometry` argument). Only enabled geometries are rendered. Each
geometry is enabled by default at construction time.

After enabling a geometry, the scene containing that geometry must be
committed using `rtcCommitScene` for the change to have effect.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcDisableGeometry], [rtcCommitScene]

```{=tex}

```
rtcDisableGeometry
------------------

#### NAME {#name}

    rtcDisableGeometry - disables the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcDisableGeometry(RTCGeometry geometry);

#### DESCRIPTION {#description}

The `rtcDisableGeometry` function disables the specified geometry
(`geometry` argument). A disabled geometry is not rendered. Each
geometry is enabled by default at construction time.

After disabling a geometry, the scene containing that geometry must be
committed using `rtcCommitScene` for the change to have effect.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcEnableGeometry], [rtcCommitScene]

```{=tex}

```
rtcSetGeometryTimeStepCount
---------------------------

#### NAME {#name}

    rtcSetGeometryTimeStepCount - sets the number of time steps of the
      geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryTimeStepCount(
      RTCGeometry geometry,
      unsigned int timeStepCount
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryTimeStepCount` function sets the number of time
steps for multi-segment motion blur (`timeStepCount` parameter) of the
specified geometry (`geometry` parameter).

For triangle meshes (`RTC_GEOMETRY_TYPE_TRIANGLE`), quad meshes
(`RTC_GEOMETRY_TYPE_QUAD`), curves (`RTC_GEOMETRY_TYPE_CURVE`), points
(`RTC_GEOMETRY_TYPE_POINT`), and subdivision geometries
(`RTC_GEOMETRY_TYPE_SUBDIVISION`), the number of time steps directly
corresponds to the number of vertex buffer slots available
(`RTC_BUFFER_TYPE_VERTEX` buffer type). For these geometries, one
vertex buffer per time step must be specified when creating
multi-segment motion blur geometries.

For instance geometries (`RTC_GEOMETRY_TYPE_INSTANCE`), a
transformation must be specified for each time step (see
`rtcSetGeometryTransform`).

For user geometries, the registered bounding callback function must
provide a bounding box per primitive and time step, and the
intersection and occlusion callback functions should properly intersect
the motion-blurred geometry at the ray time.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcSetGeometryTimeRange]

```{=tex}

```
rtcSetGeometryTimeRange
-----------------------

#### NAME {#name}

    rtcSetGeometryTimeRange - sets the time range for a motion blur geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryTimeRange(
      RTCGeometry geometry,
      float startTime,
      float endTime
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryTimeRange` function sets a time range which defines
the start (and end time) of the first (and last) time step of a motion
blur geometry. The time range is defined relative to the camera shutter
interval [0,1] but it can be arbitrary. Thus the startTime can be
smaller, equal, or larger 0, indicating a geometry whose animation
definition start before, at, or after the camera shutter opens. Similar
the endTime can be smaller, equal, or larger than 1, indicating a
geometry whose animation definition ends after, at, or before the
camera shutter closes. The startTime has to be smaller or equal to the
endTime.

The default time range when this function is not called is the entire
camera shutter [0,1]. For best performance at most one time segment
of the piece wise linear definition of the motion should fall outside
the shutter window to the left and to the right. Thus do not set the
startTime or endTime too far outside the [0,1] interval for best
performance.

This time range feature will also allow geometries to appear and
disappear during the camera shutter time if the specified time range is
a sub range of [0,1].

Please also have a look at the `rtcSetGeometryTimeStepCount` function
to see how to define the time steps for the specified time range.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryTimeStepCount]

```{=tex}

```
rtcSetGeometryVertexAttributeCount
----------------------------------

#### NAME {#name}

    rtcSetGeometryVertexAttributeCount - sets the number of vertex
      attributes of the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryVertexAttributeCount(
      RTCGeometry geometry,
      unsigned int vertexAttributeCount
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryVertexAttributeCount` function sets the number of
slots (`vertexAttributeCount` parameter) for vertex attribute buffers
(`RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE`) that can be used for the specified
geometry (`geometry` parameter).

This function is supported only for triangle meshes
(`RTC_GEOMETRY_TYPE_TRIANGLE`), quad meshes (`RTC_GEOMETRY_TYPE_QUAD`),
curves (`RTC_GEOMETRY_TYPE_CURVE`), points (`RTC_GEOMETRY_TYPE_POINT`),
and subdivision geometries (`RTC_GEOMETRY_TYPE_SUBDIVISION`).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [RTCBufferType]

```{=tex}

```
rtcSetGeometryMask
------------------

#### NAME {#name}

    rtcSetGeometryMask - sets the geometry mask

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryMask(
      RTCGeometry geometry,
      unsigned int mask
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryMask` function sets a 32-bit geometry mask (`mask`
argument) for the specified geometry (`geometry` argument).

This geometry mask is used together with the ray mask stored inside the
`mask` field of the ray. The primitives of the geometry are hit by the
ray only if the bitwise `and` operation of the geometry mask with the
ray mask is not 0. This feature can be used to disable selected
geometries for specifically tagged rays, e.g. to disable shadow casting
for certain geometries.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTCRay], [rtcGetDeviceProperty]

```{=tex}

```
rtcSetGeometryBuildQuality
--------------------------

#### NAME {#name}

    rtcSetGeometryBuildQuality - sets the build quality for the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryBuildQuality(
      RTCGeometry geometry,
      enum RTCBuildQuality quality
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryBuildQuality` function sets the build quality
(`quality` argument) for the specified geometry (`geometry` argument).
The per-geometry build quality is only a hint and may be ignored.
Embree currently uses the per-geometry build quality when the scene
build quality is set to `RTC_BUILD_QUALITY_LOW`. In this mode a
two-level acceleration structure is build, and geometries build a
separate acceleration structure using the geometry build quality. The
per-geometry build quality can be one of:

-   `RTC_BUILD_QUALITY_LOW`: Creates lower quality data structures,
    e.g. for dynamic scenes.

-   `RTC_BUILD_QUALITY_MEDIUM`: Default build quality for most usages.
    Gives a good compromise between build and render performance.

-   `RTC_BUILD_QUALITY_HIGH`: Creates higher quality data structures
    for final-frame rendering. Enables a spatial split builder for
    certain primitive types.

-   `RTC_BUILD_QUALITY_REFIT`: Uses a BVH refitting approach when
    changing only the vertex buffer.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetSceneBuildQuality]

```{=tex}

```
rtcSetGeometryMaxRadiusScale
----------------------------

#### NAME {#name}

    rtcSetGeometryMaxRadiusScale - assigns a maximal curve radius scale factor for min-width feature

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryMaxRadiusScale(RTCGeometry geometry, float maxRadiusScale);

#### DESCRIPTION {#description}

The `rtcSetMaxGeometryScale` function specifies a maximal scaling
factor for curve radii used by the min-width feature.

The min-width feature can increase the radius of curves and points, in
order to reduce aliasing and improve render times. The feature is
disabled by default and has to get enabled using the EMBREE\_MIN\_WIDTH
cmake option.

To use the feature, one has to specify a maximal curve radius scaling
factor using the [rtcSetGeometryMaxRadiusScale] function. This factor
should be a small number (e.g. 4) as the constructed BVH bounds get
increased in order to bound the curve in the worst case of maximal
radii.

One also has to set the minWidthDistanceFactor in the
RTCRayQueryContext when tracing a ray. This factor controls the target
radius size of a curve or point at some distance away of the ray
origin.

For each control point p with radius r of a curve or point primitive,
the primitive intersectors first calculate a target radius r' as:

    r' = length(p-ray_org) * minWidthDistanceFactor

Typically the minWidthDistanceFactor is set by the application such
that the target radius projects to the width of half a pixel (thus
primitive diameter is pixel sized).

The target radius r' is then clamped against the minimal bound r and
maximal bound maxRadiusScale\*r to obtain the final radius r'':

    r'' = max(r, min(r', maxRadiusScale*r))

Thus curves or points close to the camera are rendered with a normal
radii r, and curves or points far from the camera are not enlarged too
much, as this would be very expensive to render.

When `rtcSetGeometryMaxRadiusScale` function is not invoked for a curve
or point geometry (or if the maximal scaling factor is set to 1.0),
then the curve or point geometry renders normally, with radii not
modified by the min-width feature.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcInitRayQueryContext]

```{=tex}

```
rtcSetGeometryBuffer
--------------------

#### NAME {#name}

    rtcSetGeometryBuffer - assigns a view of a buffer to the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryBuffer(
      RTCGeometry geometry,
      enum RTCBufferType type,
      unsigned int slot,
      enum RTCFormat format,
      RTCBuffer buffer,
      size_t byteOffset,
      size_t byteStride,
      size_t itemCount
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryBuffer` function binds a view of a buffer object
(`buffer` argument) to a geometry buffer type and slot (`type` and
`slot` argument) of the specified geometry (`geometry` argument).

One can specify the start of the first buffer element in bytes
(`byteOffset` argument), the byte stride between individual buffer
elements (`byteStride` argument), the format of the buffer elements
(`format` argument), and the number of elements to bind (`itemCount`).

The start address (`byteOffset` argument) and stride (`byteStride`
argument) must be both aligned to 4 bytes, otherwise the
`rtcSetGeometryBuffer` function will fail.

After successful completion of this function, the geometry will hold a
reference to the buffer object.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetSharedGeometryBuffer], [rtcSetNewGeometryBuffer]

```{=tex}

```
rtcSetSharedGeometryBuffer
--------------------------

#### NAME {#name}

    rtcSetSharedGeometryBuffer - assigns a view of a shared data buffer
      to a geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetSharedGeometryBuffer(
      RTCGeometry geometry,
      enum RTCBufferType type,
      unsigned int slot,
      enum RTCFormat format,
      const void* ptr,
      size_t byteOffset,
      size_t byteStride,
      size_t itemCount
    );

#### DESCRIPTION {#description}

The `rtcSetSharedGeometryBuffer` function binds a view of a shared
user-managed data buffer (`ptr` argument) to a geometry buffer type and
slot (`type` and `slot` argument) of the specified geometry (`geometry`
argument).

One can specify the start of the first buffer element in bytes
(`byteOffset` argument), the byte stride between individual buffer
elements (`byteStride` argument), the format of the buffer elements
(`format` argument), and the number of elements to bind (`itemCount`).

The start address (`byteOffset` argument) and stride (`byteStride`
argument) must be both aligned to 4 bytes; otherwise the
`rtcSetSharedGeometryBuffer` function will fail.

When the buffer will be used as a vertex buffer
(`RTC_BUFFER_TYPE_VERTEX` and `RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE`), the
last buffer element must be readable using 16-byte SSE load
instructions, thus padding the last element is required for certain
layouts. E.g. a standard `float3` vertex buffer layout should add
storage for at least one more float to the end of the buffer.

The buffer data must remain valid for as long as the buffer may be
used, and the user is responsible for freeing the buffer data when no
longer required.

Sharing buffers can significantly reduce the memory required by the
application, thus we recommend using this feature. When enabling the
`RTC_SCENE_FLAG_COMPACT` scene flag, the spatial index structures index
into the vertex buffer, resulting in even higher memory savings.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryBuffer], [rtcSetNewGeometryBuffer]

```{=tex}

```
rtcSetNewGeometryBuffer
-----------------------

#### NAME {#name}

    rtcSetNewGeometryBuffer - creates and assigns a new data buffer to
      the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void* rtcSetNewGeometryBuffer(
      RTCGeometry geometry,
      enum RTCBufferType type,
      unsigned int slot,
      enum RTCFormat format,
      size_t byteStride,
      size_t itemCount
    );

#### DESCRIPTION {#description}

The `rtcSetNewGeometryBuffer` function creates a new data buffer of
specified format (`format` argument), byte stride (`byteStride`
argument), and number of items (`itemCount` argument), and assigns it
to a geometry buffer slot (`type` and `slot` argument) of the specified
geometry (`geometry` argument). The buffer data is managed internally
and automatically freed when the geometry is destroyed.

The byte stride (`byteStride` argument) must be aligned to 4 bytes;
otherwise the `rtcSetNewGeometryBuffer` function will fail.

The allocated buffer will be automatically over-allocated slightly when
used as a vertex buffer, where a requirement is that each buffer
element should be readable using 16-byte SSE load instructions.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryBuffer], [rtcSetSharedGeometryBuffer]

```{=tex}

```
RTCFormat
---------

#### NAME {#name}

    RTCFormat - specifies format of data in buffers

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    enum RTCFormat
    {
      RTC_FORMAT_UINT,
      RTC_FORMAT_UINT2,
      RTC_FORMAT_UINT3,
      RTC_FORMAT_UINT4,

      RTC_FORMAT_FLOAT,
      RTC_FORMAT_FLOAT2,
      RTC_FORMAT_FLOAT3,
      RTC_FORMAT_FLOAT4,
      RTC_FORMAT_FLOAT5,
      RTC_FORMAT_FLOAT6,
      RTC_FORMAT_FLOAT7,
      RTC_FORMAT_FLOAT8,
      RTC_FORMAT_FLOAT9,
      RTC_FORMAT_FLOAT10,
      RTC_FORMAT_FLOAT11,
      RTC_FORMAT_FLOAT12,
      RTC_FORMAT_FLOAT13,
      RTC_FORMAT_FLOAT14,
      RTC_FORMAT_FLOAT15,
      RTC_FORMAT_FLOAT16,

      RTC_FORMAT_FLOAT3X4_ROW_MAJOR,
      RTC_FORMAT_FLOAT4X4_ROW_MAJOR,

      RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR,
      RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR,

      RTC_FORMAT_GRID,

      RTC_FORMAT_QUATERNION_DECOMPOSITION
    };

#### DESCRIPTION {#description}

The `RTFormat` structure defines the data format stored in data buffers
provided to Embree using the [rtcSetGeometryBuffer],
[rtcSetSharedGeometryBuffer], and [rtcSetNewGeometryBuffer] API
calls.

The `RTC_FORMAT_UINT/2/3/4` format are used to specify that data
buffers store unsigned integers, or unsigned integer vectors of size
2,3 or 4. This format has typically to get used when specifying index
buffers, e.g. `RTC_FORMAT_UINT3` for triangle meshes.

The `RTC_FORMAT_FLOAT/2/3/4...` format are used to specify that data
buffers store single precision floating point values, or vectors there
of (size 2,3,4, etc.). This format is typcally used to specify to
format of vertex buffers, e.g. the `RTC_FORMAT_FLOAT3` type for vertex
buffers of triangle meshes.

The `RTC_FORMAT_FLOAT3X4_ROW_MAJOR` and
`RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR` formats, specify a 3x4 floating
point matrix layed out either row major or column major. The
`RTC_FORMAT_FLOAT4X4_ROW_MAJOR` and `RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR`
formats, specify a 4x4 floating point matrix layed out either row major
or column major. The `RTC_FORMAT_QUATERNION_DECOMPOSITION` format
specifies a structure that represents a quaternion decomposition (see
[RTCQuaternionDecomposition]) of an affine transformation. These
formats are used in the [rtcSetGeometryTransform] function or in
geometry buffers with type `RTC_BUFFER_TYPE_TRANSFORM` in order to set
a transformation matrix for instance and instance array geometries.

The `RTC_FORMAT_GRID` is a special data format used to specify grid
primitives of layout RTCGrid when creating grid geometries (see
[RTC\_GEOMETRY\_TYPE\_GRID]).

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[rtcSetGeometryBuffer], [rtcSetSharedGeometryBuffer],
[rtcSetNewGeometryBuffer], [rtcSetGeometryTransform]
[RTCQuaternionDecomposition]

```{=tex}

```
RTCBufferType
-------------

#### NAME {#name}

    RTCFormat - specifies format of data in buffers

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    enum RTCBufferType
    {
      RTC_BUFFER_TYPE_INDEX            = 0,
      RTC_BUFFER_TYPE_VERTEX           = 1,
      RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE = 2,
      RTC_BUFFER_TYPE_NORMAL           = 3,
      RTC_BUFFER_TYPE_TANGENT          = 4,
      RTC_BUFFER_TYPE_NORMAL_DERIVATIVE = 5,

      RTC_BUFFER_TYPE_GRID                 = 8,

      RTC_BUFFER_TYPE_FACE                 = 16,
      RTC_BUFFER_TYPE_LEVEL                = 17,
      RTC_BUFFER_TYPE_EDGE_CREASE_INDEX    = 18,
      RTC_BUFFER_TYPE_EDGE_CREASE_WEIGHT   = 19,
      RTC_BUFFER_TYPE_VERTEX_CREASE_INDEX  = 20,
      RTC_BUFFER_TYPE_VERTEX_CREASE_WEIGHT = 21,
      RTC_BUFFER_TYPE_HOLE                 = 22,
      
      RTC_BUFFER_TYPE_TRANSFORM            = 23,

      RTC_BUFFER_TYPE_FLAGS = 32
    };

#### DESCRIPTION {#description}

The `RTBufferType` structure defines slots to assign data buffers to
using the [rtcSetGeometryBuffer], [rtcSetSharedGeometryBuffer], and
[rtcSetNewGeometryBuffer] API calls.

For most geometry types the `RTC_BUFFER_TYPE_INDEX` slot is used to
assign an index buffer, while the `RTC_BUFFER_TYPE_VERTEX` is used to
assign the corresponding vertex buffer.

The `RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE` slot can get used to assign
arbitrary additional vertex data which can get interpolated using the
[rtcInterpolate] API call.

The `RTC_BUFFER_TYPE_NORMAL`, `RTC_BUFFER_TYPE_TANGENT`, and
`RTC_BUFFER_TYPE_NORMAL_DERIVATIVE` are special buffers required to
assign per vertex normals, tangents, and normal derivatives for some
curve types.

The `RTC_BUFFER_TYPE_GRID` buffer is used to assign the grid primitive
buffer for grid geometries (see [RTC\_GEOMETRY\_TYPE\_GRID]).

The `RTC_BUFFER_TYPE_FACE`, `RTC_BUFFER_TYPE_LEVEL`,
`RTC_BUFFER_TYPE_EDGE_CREASE_INDEX`,
`RTC_BUFFER_TYPE_EDGE_CREASE_WEIGHT`,
`RTC_BUFFER_TYPE_VERTEX_CREASE_INDEX`,
`RTC_BUFFER_TYPE_VERTEX_CREASE_WEIGHT`, and `RTC_BUFFER_TYPE_HOLE` are
special buffers required to create subdivision meshes (see
[RTC\_GEOMETRY\_TYPE\_SUBDIVISION]).

The `RTC_BUFFER_TYPE_TRANSFORM` buffer is used to provide instance
transformation information for instance array geometries (see
[RTC\_GEOMETRY\_TYPE\_INSTANCE\_ARRAY]).

The `RTC_BUFFER_TYPE_FLAGS` can get used to add additional flag per
primitive of a geometry, and is currently only used for linear curves.

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[rtcSetGeometryBuffer], [rtcSetSharedGeometryBuffer],
[rtcSetNewGeometryBuffer]

```{=tex}

```
rtcGetGeometryBufferData
------------------------

#### NAME {#name}

    rtcGetGeometryBufferData - gets pointer to
      the first buffer view element

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void* rtcGetGeometryBufferData(
      RTCGeometry geometry,
      enum RTCBufferType type,
      unsigned int slot
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryBufferData` function returns a pointer to the first
element of the buffer view attached to the specified buffer type and
slot (`type` and `slot` argument) of the geometry (`geometry`
argument).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryBuffer], [rtcSetSharedGeometryBuffer],
[rtcSetNewGeometryBuffer]

```{=tex}

```
rtcUpdateGeometryBuffer
-----------------------

#### NAME {#name}

    rtcUpdateGeometryBuffer - marks a buffer view bound to the geometry
      as modified

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcUpdateGeometryBuffer(
      RTCGeometry geometry,
      enum RTCBufferType type,
      unsigned int slot
    );

#### DESCRIPTION {#description}

The `rtcUpdateGeometryBuffer` function marks the buffer view bound to
the specified buffer type and slot (`type` and `slot` argument) of a
geometry (`geometry` argument) as modified.

If a data buffer is changed by the application, the
`rtcUpdateGeometryBuffer` call must be invoked for that buffer. Each
buffer view assigned to a buffer slot is initially marked as modified,
thus this function needs to be called only when doing buffer
modifications after the first `rtcCommitScene`.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewGeometry], [rtcCommitScene]

```{=tex}

```
rtcSetGeometryIntersectFilterFunction
-------------------------------------

#### NAME {#name}

    rtcSetGeometryIntersectFilterFunction - sets the intersection filter
      for the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCFilterFunctionNArguments
    {
      int* valid;
      void* geometryUserPtr;
      const struct RTCRayQueryContext* context;
      struct RTCRayN* ray;
      struct RTCHitN* hit;
      unsigned int N;
    };

    typedef void (*RTCFilterFunctionN)(
      const struct RTCFilterFunctionNArguments* args
    );

    void rtcSetGeometryIntersectFilterFunction(
      RTCGeometry geometry,
      RTCFilterFunctionN filter
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryIntersectFilterFunction` function registers an
intersection filter callback function (`filter` argument) for the
specified geometry (`geometry` argument).

Only a single callback function can be registered per geometry, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

The registered intersection filter function is invoked for every hit
encountered during the `rtcIntersect`-type ray queries and can accept
or reject that hit. The feature can be used to define a silhouette for
a primitive and reject hits that are outside the silhouette. E.g. a
tree leaf could be modeled with an alpha texture that decides whether
hit points lie inside or outside the leaf.

If the `RTC_BUILD_QUALITY_HIGH` mode is set, the filter functions may
be called multiple times for the same primitive hit. Further, rays
hitting exactly the edge might also report two hits for the same
surface. For certain use cases, the application may have to work around
this limitation by collecting already reported hits (`geomID`/`primID`
pairs) and ignoring duplicates.

The filter function callback of type `RTCFilterFunctionN` gets passed a
number of arguments through the `RTCFilterFunctionNArguments`
structure. The `valid` parameter of that structure points to an integer
valid mask (0 means invalid and -1 means valid). The `geometryUserPtr`
member is a user pointer optionally set per geometry through the
`rtcSetGeometryUserData` function. The `context` member points to the
ray query context passed to the ray query function. The `ray` parameter
points to `N` rays in SOA layout. The `hit` parameter points to `N`
hits in SOA layout to test. The `N` parameter is the number of rays and
hits in `ray` and `hit`. The hit distance is provided as the `tfar`
value of the ray. If the hit geometry is instanced, the `instID` member
of the ray is valid, and the ray and the potential hit are in object
space.

The filter callback function has the task to check for each valid ray
whether it wants to accept or reject the corresponding hit. To reject a
hit, the filter callback function just has to write `0` to the integer
valid mask of the corresponding ray. To accept the hit, it just has to
leave the valid mask set to `-1`. When accepting a hit, the filter
function is further allowed to change the hit and decrease the `tfar`
value of the ray but it should not modify other ray data nor any
inactive components of the ray or hit.

When performing ray queries using `rtcIntersect1`, it is guaranteed
that the packet size is 1 when the callback is invoked. When performing
ray queries using the `rtcIntersect4/8/16` functions, it is not
generally guaranteed that the ray packet size (and order of rays inside
the packet) passed to the callback matches the initial ray packet.
However, under some circumstances these properties are guaranteed, and
whether this is the case can be queried using `rtcGetDeviceProperty`.

For many usage scenarios, repacking and re-ordering of rays does not
cause difficulties in implementing the callback function. However,
algorithms that need to extend the ray with additional data must use
the `rayID` component of the ray to identify the original ray to access
the per-ray data.

The implementation of the filter function can choose to implement a
single code path that uses the ray access helper functions `RTCRay_XXX`
and hit access helper functions `RTCHit_XXX` to access ray and hit
data. Alternatively the code can branch to optimized implementations
for specific sizes of `N` and cast the `ray` and `hit` inputs to the
proper packet types.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryOccludedFilterFunction]

```{=tex}

```
rtcSetGeometryOccludedFilterFunction
------------------------------------

#### NAME {#name}

    rtcSetGeometryOccludedFilterFunction - sets the occlusion filter
      for the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryOccludedFilterFunction(
      RTCGeometry geometry,
      RTCFilterFunctionN filter
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryOccludedFilterFunction` function registers an
occlusion filter callback function (`filter` argument) for the
specified geometry (`geometry` argument).

Only a single callback function can be registered per geometry, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

The registered occlusion filter function is invoked for every hit
encountered during the `rtcOccluded`-type ray queries and can accept or
reject that hit. The feature can be used to define a silhouette for a
primitive and reject hits that are outside the silhouette. E.g. a tree
leaf could be modeled with an alpha texture that decides whether hit
points lie inside or outside the leaf.

Please see the description of the
`rtcSetGeometryIntersectFilterFunction` for a description of the filter
callback function.

The `rtcOccluded`-type functions terminate traversal when a hit got
committed. As filter functions can only set the `tfar` distance of the
ray for a committed hit, the occlusion filter cannot influence the
`tfar` value of subsequent traversal, as that directly ends. For that
reason `rtcOccluded` and occlusion filters cannot get used to gather
the next n-hits, and `rtcIntersect` and intersection filters should get
used instead.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryIntersectFilterFunction]

```{=tex}

```
rtcSetGeometryEnableFilterFunctionFromArguments
-----------------------------------------------

#### NAME {#name}

    rtcSetGeometryEnableFilterFunctionFromArguments - enables
      argument filter functions for the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryEnableFilterFunctionFromArguments(
       RTCGeometry geometry, bool enable);

#### DESCRIPTION {#description}

This function enables invokation the filter function passed through
`RTCIntersectArguments` or `RTCOccludedArguments` to the intersect and
occluded queries. If enable is true the argument filter function
invokation is enabled for the geometry or disabled otherwise. By
default the invokation of the argument filter function is disabled for
some geometry.

The argument filter function invokation can also get enforced for each
geometry by using the `RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER` ray
query flag that can get passed to `rtcIntersect` and `rtcOccluded`
functions. See Section [rtcInitIntersectArguments] and
[rtcInitOccludedArguments] for more details.

In order to use the argument filter function for some scene, that
feature additionally has to get enabled using the
`RTC_SCENE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS` scene flag. See Section
[rtcSetSceneFlags] for more details.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can get queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcInitIntersectArguments], [rtcInitOccludedArguments],
[rtcSetSceneFlags]

```{=tex}

```
rtcInvokeIntersectFilterFromGeometry
------------------------------------

#### NAME {#name}

    rtcInvokeIntersectFilterFromGeometry - invokes the
      intersection filter function from the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcInvokeIntersectFilterFromGeometry(
      const struct RTCIntersectFunctionNArguments* args,
      const struct RTCFilterFunctionNArguments* filterArgs
    );

#### DESCRIPTION {#description}

The `rtcInvokeIntersectFilterFromGeometry` function can be called
inside an `RTCIntersectFunctionN` user geometry callback function to
invoke the intersection filter registered to the geometry. For this an
`RTCFilterFunctionNArguments` structure must be created (see
`rtcSetGeometryIntersectFilterFunction`) which basically consists of a
valid mask, a hit packet to filter, the corresponding ray packet, and
the packet size. After the invocation of
`rtcInvokeIntersectFilterFromGeometry`, only rays that are still valid
(valid mask set to -1) should update a hit.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcInvokeOccludedFilterFromGeometry],
[rtcSetGeometryIntersectFunction]

```{=tex}

```
rtcInvokeOccludedFilterFromGeometry
-----------------------------------

#### NAME {#name}

    rtcInvokeOccludedFilterFromGeometry - invokes the occlusion
      filter function from the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcInvokeOccludedFilterFromGeometry(
      const struct RTCOccludedFunctionNArguments* args,
      const struct RTCFilterFunctionNArguments* filterArgs
    );

#### DESCRIPTION {#description}

The `rtcInvokeOccludedFilterFromGeometry` function can be called inside
an `RTCOccludedFunctionN` user geometry callback function to invoke the
occlusion filter registered to the geometry. For this an
`RTCFilterFunctionNArguments` structure must be created (see
`rtcSetGeometryIntersectFilterFunction`) which basically consists of a
valid mask, a hit packet to filter, the corresponding ray packet, and
the packet size. After the invocation of
`rtcInvokeOccludedFilterFromGeometry` only rays that are still valid
(valid mask set to -1) should signal an occlusion.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcInvokeIntersectFilterFromGeometry],
[rtcSetGeometryOccludedFunction]

```{=tex}

```
rtcSetGeometryUserData
----------------------

#### NAME {#name}

    rtcSetGeometryUserData - sets the user-defined data pointer of the
      geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryUserData(RTCGeometry geometry, void* userPtr);

#### DESCRIPTION {#description}

The `rtcSetGeometryUserData` function sets the user-defined data
pointer (`userPtr` argument) for a geometry (`geometry` argument). This
user data pointer is intended to be pointing to the application's
representation of the geometry, and is passed to various callback
functions. The application can use this pointer inside the callback
functions to access its geometry representation.

The `rtcGetGeometryUserData` function can be used to query an already
set user data pointer of a geometry.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetGeometryUserData]

```{=tex}

```
rtcGetGeometryUserData
----------------------

#### NAME {#name}

    rtcGetGeometryUserData - returns the user data pointer
      of the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void* rtcGetGeometryUserData(RTCGeometry geometry);

#### DESCRIPTION {#description}

The `rtcGetGeometryUserData` function queries the user data pointer
previously set with `rtcSetGeometryUserData`. When
`rtcSetGeometryUserData` was not called yet, `NULL` is returned.

This function is supposed to be used during rendering, but only
supported on the CPU and not inside SYCL kernels on the GPU. Inside a
SYCL kernel the `rtcGetGeometryUserDataFromScene` function has to get
used.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryUserData], [rtcGetGeometryUserDataFromScene]

```{=tex}

```
rtcGetGeometryUserDataFromScene
-------------------------------

#### NAME {#name}

    rtcGetGeometryUserDataFromScene - returns the user data pointer
      of the geometry through the scene object

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void* rtcGetGeometryUserDataFromScene(RTCScene scene, unsigned int geomID);

#### DESCRIPTION {#description}

The `rtcGetGeometryUserDataFromScene` function queries the user data
pointer previously set with `rtcSetGeometryUserData` from the geometry
with index `geomID` from the specified scene `scene`. When
`rtcSetGeometryUserData` was not called yet, `NULL` is returned.

In contrast to the `rtcGetGeometryUserData` function, the
`rtcGetGeometryUserDataFromScene` function an get used during rendering
inside a SYCL kernel.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryUserData], [rtcGetGeometryUserData]

```{=tex}

```
rtcSetGeometryUserPrimitiveCount
--------------------------------

#### NAME {#name}

    rtcSetGeometryUserPrimitiveCount - sets the number of primitives
      of a user-defined geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryUserPrimitiveCount(
      RTCGeometry geometry,
      unsigned int userPrimitiveCount
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryUserPrimitiveCount` function sets the number of
user-defined primitives (`userPrimitiveCount` parameter) of the
specified user-defined geometry (`geometry` parameter).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_USER]

```{=tex}

```
rtcSetGeometryBoundsFunction
----------------------------

#### NAME {#name}

    rtcSetGeometryBoundsFunction - sets a callback to query the
      bounding box of user-defined primitives

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCBoundsFunctionArguments
    {
      void* geometryUserPtr;
      unsigned int primID;
      unsigned int timeStep;
      struct RTCBounds* bounds_o;
    };

    typedef void (*RTCBoundsFunction)(
      const struct RTCBoundsFunctionArguments* args
    );

    void rtcSetGeometryBoundsFunction(
      RTCGeometry geometry,
      RTCBoundsFunction bounds,
      void* userPtr
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryBoundsFunction` function registers a bounding box
callback function (`bounds` argument) with payload (`userPtr` argument)
for the specified user geometry (`geometry` argument).

Only a single callback function can be registered per geometry, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

In SYCL mode the BVH construction is done on the host and the passed
function pointer must be a host-side function pointer.

The registered bounding box callback function is invoked to calculate
axis-aligned bounding boxes of the primitives of the user-defined
geometry during spatial acceleration structure construction. The
bounding box callback of `RTCBoundsFunction` type is invoked with a
pointer to a structure of type `RTCBoundsFunctionArguments` which
contains various arguments, such as: the user data of the geometry
(`geometryUserPtr` member), the ID of the primitive to calculate the
bounds for (`primID` member), the time step at which to calculate the
bounds (`timeStep` member), and a memory location to write the
calculated bound to (`bounds_o` member).

In a typical usage scenario one would store a pointer to the internal
representation of the user geometry object using
`rtcSetGeometryUserData`. The callback function can then read that
pointer from the `geometryUserPtr` field and calculate the proper
bounding box for the requested primitive and time, and store that
bounding box to the destination structure (`bounds_o` member).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_USER]

```{=tex}

```
rtcSetGeometryIntersectFunction
-------------------------------

#### NAME {#name}

    rtcSetGeometryIntersectFunction - sets the callback function to
      intersect a user geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCIntersectFunctionNArguments
    {
      int* valid;
      void* geometryUserPtr;
      unsigned int primID;
      struct RTCRayQueryContext* context;
      struct RTCRayHitN* rayhit;
      unsigned int N;
      unsigned int geomID;
    };

    typedef void (*RTCIntersectFunctionN)(
      const struct RTCIntersectFunctionNArguments* args
    );

    void rtcSetGeometryIntersectFunction(
      RTCGeometry geometry,
      RTCIntersectFunctionN intersect
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryIntersectFunction` function registers a
ray/primitive intersection callback function (`intersect` argument) for
the specified user geometry (`geometry` argument).

Only a single callback function can be registered per geometry and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

The registered callback function is invoked by `rtcIntersect`-type ray
queries to calculate the intersection of a ray packet of variable size
with one user-defined primitive. The callback function of type
`RTCIntersectFunctionN` gets passed a number of arguments through the
`RTCIntersectFunctionNArguments` structure. The value `N` specifies the
ray packet size, `valid` points to an array of integers that specify
whether the corresponding ray is valid (-1) or invalid (0), the
`geometryUserPtr` member points to the geometry user data previously
set through `rtcSetGeometryUserData`, the `context` member points to
the ray query context passed to the ray query, the `rayhit` member
points to a ray and hit packet of variable size `N`, and the `geomID`
and `primID` member identifies the geometry ID and primitive ID of the
primitive to intersect.

The `ray` component of the `rayhit` structure contains valid data, in
particular the `tfar` value is the current closest hit distance found.
All data inside the `hit` component of the `rayhit` structure are
undefined and should not be read by the function.

The task of the callback function is to intersect each active ray from
the ray packet with the specified user primitive. If the user-defined
primitive is missed by a ray of the ray packet, the function should
return without modifying the ray or hit. If an intersection of the
user-defined primitive with the ray was found in the valid range (from
`tnear` to `tfar`), it should update the hit distance of the ray
(`tfar` member) and the hit (`u`, `v`, `Ng`, `instID`, `geomID`,
`primID` members). In particular, the currently intersected instance is
stored in the `instID` field of the ray query context, which must be
deep copied into the `instID` member of the hit.

As a primitive might have multiple intersections with a ray, the
intersection filter function needs to be invoked by the user geometry
intersection callback for each encountered intersection, if filtering
of intersections is desired. This can be achieved through the
`rtcInvokeIntersectFilterFromGeometry` call.

Within the user geometry intersect function, it is safe to trace new
rays and create new scenes and geometries.

When performing ray queries using `rtcIntersect1`, it is guaranteed
that the packet size is 1 when the callback is invoked. When performing
ray queries using the `rtcIntersect4/8/16` functions, it is not
generally guaranteed that the ray packet size (and order of rays inside
the packet) passed to the callback matches the initial ray packet.
However, under some circumstances these properties are guaranteed, and
whether this is the case can be queried using `rtcGetDeviceProperty`.

For many usage scenarios, repacking and re-ordering of rays does not
cause difficulties in implementing the callback function. However,
algorithms that need to extend the ray with additional data must use
the `rayID` component of the ray to identify the original ray to access
the per-ray data.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryOccludedFunction], [rtcSetGeometryUserData],
[rtcInvokeIntersectFilterFromGeometry]

```{=tex}

```
rtcSetGeometryOccludedFunction
------------------------------

#### NAME {#name}

    rtcSetGeometryOccludedFunction - sets the callback function to
      test a user geometry for occlusion

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCOccludedFunctionNArguments
    {
      int* valid;
      void* geometryUserPtr;
      unsigned int primID;
      struct RTCRayQueryContext* context;
      struct RTCRayN* ray;
      unsigned int N;
      unsigned int geomID;
    };

    typedef void (*RTCOccludedFunctionN)(
      const struct RTCOccludedFunctionNArguments* args
    );

    void rtcSetGeometryOccludedFunction(
      RTCGeometry geometry,
      RTCOccludedFunctionN filter
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryOccludedFunction` function registers a ray/primitive
occlusion callback function (`filter` argument) for the specified user
geometry (`geometry` argument).

Only a single callback function can be registered per geometry, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

The registered callback function is invoked by `rtcOccluded`-type ray
queries to test whether the rays of a packet of variable size are
occluded by a user-defined primitive. The callback function of type
`RTCOccludedFunctionN` gets passed a number of arguments through the
`RTCOccludedFunctionNArguments` structure. The value `N` specifies the
ray packet size, `valid` points to an array of integers which specify
whether the corresponding ray is valid (-1) or invalid (0), the
`geometryUserPtr` member points to the geometry user data previously
set through `rtcSetGeometryUserData`, the `context` member points to
the ray query context passed to the ray query, the `ray` member points
to a ray packet of variable size `N`, and the `geomID` and `primID`
member identifies the geometry ID and primitive ID of the primitive to
intersect.

The task of the callback function is to intersect each active ray from
the ray packet with the specified user primitive. If the user-defined
primitive is missed by a ray of the ray packet, the function should
return without modifying the ray. If an intersection of the
user-defined primitive with the ray was found in the valid range (from
`tnear` to `tfar`), it should set the `tfar` member of the ray to
`-inf`.

As a primitive might have multiple intersections with a ray, the
occlusion filter function needs to be invoked by the user geometry
occlusion callback for each encountered intersection, if filtering of
intersections is desired. This can be achieved through the
`rtcInvokeOccludedFilterFromGeometry` call.

Within the user geometry occlusion function, it is safe to trace new
rays and create new scenes and geometries.

When performing ray queries using `rtcOccluded1`, it is guaranteed that
the packet size is 1 when the callback is invoked. When performing ray
queries using the `rtcOccluded4/8/16` functions, it is not generally
guaranteed that the ray packet size (and order of rays inside the
packet) passed to the callback matches the initial ray packet. However,
under some circumstances these properties are guaranteed, and whether
this is the case can be queried using `rtcGetDeviceProperty`.

For many usage scenarios, repacking and re-ordering of rays does not
cause difficulties in implementing the callback function. However,
algorithms that need to extend the ray with additional data must use
the `rayID` component of the ray to identify the original ray to access
the per-ray data.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryIntersectFunction], [rtcSetGeometryUserData],
[rtcInvokeOccludedFilterFromGeometry]

```{=tex}

```
rtcSetGeometryPointQueryFunction
--------------------------------

#### NAME {#name}

    rtcSetGeometryPointQueryFunction - sets the point query callback function
      for a geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCPointQueryFunctionArguments
    {
      // the (world space) query object that was passed as an argument of rtcPointQuery.
      struct RTCPointQuery* query;

      // used for user input/output data. Will not be read or modified internally.
      void* userPtr;

      // primitive and geometry ID of primitive
      unsigned int  primID;        
      unsigned int  geomID;    

      // the context with transformation and instance ID stack
      struct RTCPointQueryContext* context;

      // scaling factor indicating whether the current instance transformation
      // is a similarity transformation.
      float similarityScale;
    };

    typedef bool (*RTCPointQueryFunction)(
      struct RTCPointQueryFunctionArguments* args
    );

    void rtcSetGeometryPointQueryFunction(
      RTCGeometry geometry,
      RTCPointQueryFunction queryFunc
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryPointQueryFunction` function registers a point query
callback function (`queryFunc` argument) for the specified geometry
(`geometry` argument).

Only a single callback function can be registered per geometry and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

The registered callback function is invoked by [rtcPointQuery] for
every primitive of the geometry that intersects the corresponding point
query domain. The callback function of type `RTCPointQueryFunction`
gets passed a number of arguments through the
`RTCPointQueryFunctionArguments` structure. The `query` object is the
original point query object passed into [rtcPointQuery], `usrPtr` is
an arbitrary pointer to pass input into and store results of the
callback function. The `primID`, `geomID` and `context` (see
[rtcInitPointQueryContext] for details) can be used to identify the
geometry data of the primitive.

A `RTCPointQueryFunction` can also be passed directly as an argument to
[rtcPointQuery]. In this case the callback is invoked for all
primitives in the scene that intersect the query domain. If a callback
function is passed as an argument to [rtcPointQuery] and (a
potentially different) callback function is set for a geometry with
[rtcSetGeometryPointQueryFunction] both callback functions are
invoked and the callback function passed to [rtcPointQuery] will be
called before the geometry specific callback function.

If instancing is used, the parameter `simliarityScale` indicates
whether the current instance transform (top element of the stack in
`context`) is a similarity transformation or not. Similarity
transformations are composed of translation, rotation and uniform
scaling and if a matrix M defines a similarity transformation, there is
a scaling factor D such that for all x,y: dist(Mx, My) = D \* dist(x,
y). In this case the parameter `scalingFactor` is this scaling factor D
and otherwise it is 0. A valid similarity scale (`similarityScale` \>
0) allows to compute distance information in instance space and scale
the distances into world space (for example, to update the query
radius, see below) by dividing the instance space distance with the
similarity scale. If the current instance transform is not a similarity
transform (`similarityScale` is 0), the distance computation has to be
performed in world space to ensure correctness. In this case the
instance to world transformations given with the `context` should be
used to transform the primitive data into world space. Otherwise, the
query location can be transformed into instance space which can be more
efficient. If there is no instance transform, the similarity scale is
1.

The callback function will potentially be called for primitives outside
the query domain for two reasons: First, the callback is invoked for
all primitives inside a BVH leaf node since no geometry data of
primitives is determined internally and therefore individual primitives
are not culled (only their (aggregated) bounding boxes). Second, in
case non similarity transformations are used, the resulting ellipsoidal
query domain (in instance space) is approximated by its axis aligned
bounding box internally and therefore inner nodes that do not intersect
the original domain might intersect the approximative bounding box
which results in unnecessary callbacks. In any case, the callbacks are
conservative, i.e. if a primitive is inside the query domain a callback
will be invoked but the reverse is not necessarily true.

For efficiency, the radius of the `query` object can be decreased (in
world space) inside the callback function to improve culling of
geometry during BVH traversal. If the query radius was updated, the
callback function should return `true` to issue an update of internal
traversal information. Increasing the radius or modifying the time or
position of the query results in undefined behaviour.

Within the callback function, it is safe to call [rtcPointQuery]
again, for example when implementing instancing manually. In this case
the instance transformation should be pushed onto the stack in
`context`. Embree will internally compute the point query information
in instance space using the top element of the stack in `context` when
[rtcPointQuery] is called.

For a reference implementation of a closest point traversal of triangle
meshes using instancing and user defined instancing see the tutorial
[ClosestPoint].

#### SEE ALSO {#see-also}

[rtcPointQuery], [rtcInitPointQueryContext]

```{=tex}

```
rtcGetSYCLDeviceFunctionPointer
-------------------------------

#### NAME {#name}

    rtcGetSYCLDeviceFunctionPointer - obtains a device side
      function pointer for some SYCL function

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    template<auto F>
    inline decltype(F) rtcGetSYCLDeviceFunctionPointer(sycl::queue& queue);

#### DESCRIPTION {#description}

This function returns a device side function pointer for some function
F. This function F must be defined using the
`RTC_SYCL_INDIRECTLY_CALLABLE` attribute, e.g.:

    RTC_SYCL_INDIRECTLY_CALLABLE void filter(
      const RTCFilterFunctionNArguments* args) { ... }

    RTCFilterFunctionN fptr = rtcGetSYCLDeviceFunctionPointer<filter>(queue);

Such a device side function pointers of some filter callbacks can get
assigned to a geometry using the
`rtcSetGeometryIntersectFilterFunction` and
`rtcSetGeometryOccludedFilterFunction` API functions.

Further, device side function pointers for user geometry callbacks can
be assigned to geometries using the `rtcSetGeometryIntersectFunction`
and `rtcSetGeometryOccludedFunction` API calls.

These geometry versions of the callback functions are disabled in SYCL
by default, and we recommend not using them for performance reasons.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometryIntersectFunction],
[rtcSetGeometryOccludedFunction],
[rtcSetGeometryIntersectFilterFunction],
[rtcSetGeometryOccludedFilterFunction]

```{=tex}

```
rtcSetGeometryInstancedScene
----------------------------

#### NAME {#name}

    rtcSetGeometryInstancedScene - sets the instanced scene of
      an instance geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryInstancedScene(
      RTCGeometry geometry,
      RTCScene scene
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryInstancedScene` function sets the instanced scene
(`scene` argument) of the specified instance geometry (`geometry`
argument).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_INSTANCE], [rtcSetGeometryTransform]

```{=tex}

```
rtcSetGeometryInstancedScenes
-----------------------------

#### NAME {#name}

    rtcSetGeometryInstancedScenes - sets an array of scenes that can be
    instanced by an instance array geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryInstancedScenes(
      RTCGeometry geometry,
      RTCScene* scene,
      size_t numScenes
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryInstancedScenes` function sets an array of type
`RTCScene` with `numScenes` elements that the specified instance
geometry (`geometry` argument) can instance. This call also requires
setting an index buffer using either `rtcSetSharedGeometryBuffer` or
`rtcSetNewGeometryBuffer` (similar to index buffers for triangle
meshes), that specifies which instance of the instance array instances
which scene in the scene array. If only one scene should be instanced
the call `rtcSetGeometryInstancedScene` should be preferred.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_INSTANCE\_ARRAY], [rtcSetNewGeometryBuffer],
[rtcSetSharedGeometryBuffer], [rtcSetGeometryInstancedScene]

```{=tex}

```
rtcSetGeometryTransform
-----------------------

#### NAME {#name}

    rtcSetGeometryTransform - sets the transformation for a particular
      time step of an instance geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryTransform(
      RTCGeometry geometry,
      unsigned int timeStep,
      enum RTCFormat format,
      const float* xfm
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryTransform` function sets the local-to-world affine
transformation (`xfm` parameter) of an instance geometry (`geometry`
parameter) for a particular time step (`timeStep` parameter). The
transformation is specified as a 3×4 matrix (3×3 linear transformation
plus translation), for which the following formats (`format` parameter)
are supported:

-   `RTC_FORMAT_FLOAT3X4_ROW_MAJOR`: The 3×4 float matrix is laid out
    in row-major form.

-   `RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form.

-   `RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form as a 4×4 homogeneous matrix with the last
    row being equal to (0, 0, 0, 1).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_INSTANCE]

```{=tex}

```
rtcSetGeometryTransformQuaternion
---------------------------------

#### NAME {#name}

    rtcSetGeometryTransformQuaternion - sets the transformation for a particular
      time step of an instance geometry as a decomposition of the
      transformation matrix using quaternions to represent the rotation.

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryTransformQuaternion(
      RTCGeometry geometry,
      unsigned int timeStep,
      const struct RTCQuaternionDecomposition* qd
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryTransformQuaternion` function sets the
local-to-world affine transformation (`qd` parameter) of an instance
geometry (`geometry` parameter) for a particular time step (`timeStep`
parameter). The transformation is specified as a
[RTCQuaternionDecomposition], which is a decomposition of an affine
transformation that represents the rotational component of an affine
transformation as a quaternion. This allows interpolating rotational
transformations exactly using spherical linear interpolation (such as a
turning wheel).

For more information about the decomposition see
[RTCQuaternionDecomposition]. The quaternion given in the
`RTCQuaternionDecomposition` struct will be normalized internally.

For correct results, the transformation matrices for all time steps
must be set either using `rtcSetGeometryTransform` or
`rtcSetGeometryTransformQuaternion`. Mixing both representations is not
allowed. Spherical linear interpolation will be used, iff the
transformation matizes are set with
`rtcSetGeometryTransformQuaternion`.

For an example of this feature see the tutorial [Quaternion Motion
Blur].

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcInitQuaternionDecomposition], [rtcSetGeometryTransform]

```{=tex}

```
rtcGetGeometryTransform
-----------------------

#### NAME {#name}

    rtcGetGeometryTransform - returns the interpolated instance
      transformation for the specified time

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcGetGeometryTransform(
      RTCGeometry geometry,
      float time,
      enum RTCFormat format,
      void* xfm
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryTransform` function returns the interpolated local
to world transformation (`xfm` parameter) of an instance geometry
(`geometry` parameter) for a particular time (`time` parameter in range
$[0,1]$) in the specified format (`format` parameter).

Possible formats for the returned matrix are:

-   `RTC_FORMAT_FLOAT3X4_ROW_MAJOR`: The 3×4 float matrix is laid out
    in row-major form.

-   `RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form.

-   `RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form as a 4×4 homogeneous matrix with last row
    equal to (0, 0, 0, 1).

This function is supposed to be used during rendering, but only
supported on the CPU and not inside SYCL kernels on the GPU. Inside a
SYCL kernel the `rtcGetGeometryTransformFromScene` function has to get
used.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_INSTANCE], [rtcSetGeometryTransform],
[rtcGetGeometryTransformFromScene]

```{=tex}

```
rtcGetGeometryTransformEx
-------------------------

#### NAME {#name}

    rtcGetGeometryTransformEx - returns the interpolated instance
    transformation for an instance of an instance array geometry for the
    specified time.

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcGetGeometryTransformEx(
      RTCGeometry geometry,
      unsigned int instPrimID,
      float time,
      enum RTCFormat format,
      void* xfm
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryTransformEx` function returns the interpolated local
to world transformation (`xfm` parameter) of the `instPrimID`-th
instance of an instance array geometry (`geometry` parameter) for a
particular time (`time` parameter in range $[0,1]$) in the specified
format (`format` parameter). The function can also be used when
`geometry` refers to a regular instance, but then the `instPrimID` has
to be $0$.

Possible formats for the returned matrix are:

-   `RTC_FORMAT_FLOAT3X4_ROW_MAJOR`: The 3×4 float matrix is laid out
    in row-major form.

-   `RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form.

-   `RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form as a 4×4 homogeneous matrix with last row
    equal to (0, 0, 0, 1).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_INSTANCE\_ARRAY]

```{=tex}

```
rtcGetGeometryTransformFromScene
--------------------------------

#### NAME {#name}

    rtcGetGeometryTransformFromScene - returns the interpolated instance
      transformation for the specified time

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcGetGeometryTransformFromScene(
      RTCScene scene,
      unsigned int geomID,
      float time,
      enum RTCFormat format,
      void* xfm
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryTransformFromScene` function returns the
interpolated local to world transformation (`xfm` output parameter) of
an instance geometry specified by its geometry ID (`geomID` parameter)
of a scene (`scene` parameter) for a particular time (`time` parameter
in range $[0,1]$) in the specified format (`format` parameter).

Possible formats for the returned matrix are:

-   `RTC_FORMAT_FLOAT3X4_ROW_MAJOR`: The 3×4 float matrix is laid out
    in row-major form.

-   `RTC_FORMAT_FLOAT3X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form.

-   `RTC_FORMAT_FLOAT4X4_COLUMN_MAJOR`: The 3×4 float matrix is laid
    out in column-major form as a 4×4 homogeneous matrix with last row
    equal to (0, 0, 0, 1).

In contrast to the `rtcGetGeometryTransform` function, the
`rtcGetGeometryTransformFromScene` function can get used during
rendering inside a SYCL kernel.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_INSTANCE], [rtcSetGeometryTransform],
[rtcGetGeometryTransform]

```{=tex}

```
rtcSetGeometryTessellationRate
------------------------------

#### NAME {#name}

    rtcSetGeometryTessellationRate - sets the tessellation rate of the
      geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryTessellationRate(
      RTCGeometry geometry,
      float tessellationRate
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryTessellationRate` function sets the tessellation
rate (`tessellationRate` argument) for the specified geometry
(`geometry` argument). The tessellation rate can only be set for flat
curves and subdivision geometries. For curves, the tessellation rate
specifies the number of ray-facing quads per curve segment. For
subdivision surfaces, the tessellation rate specifies the number of
quads along each edge.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_CURVE], [RTC\_GEOMETRY\_TYPE\_SUBDIVISION]

```{=tex}

```
rtcSetGeometryTopologyCount
---------------------------

#### NAME {#name}

    rtcSetGeometryTopologyCount - sets the number of topologies of
      a subdivision geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryTopologyCount(
      RTCGeometry geometry,
      unsigned int topologyCount
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryTopologyCount` function sets the number of
topologies (`topologyCount` parameter) for the specified subdivision
geometry (`geometry` parameter). The number of topologies of a
subdivision geometry must be greater or equal to 1.

To use multiple topologies, first the number of topologies must be
specified, then the individual topologies can be configured using
`rtcSetGeometrySubdivisionMode` and by setting an index buffer
(`RTC_BUFFER_TYPE_INDEX`) using the topology ID as the buffer slot.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_SUBDIVISION], [rtcSetGeometrySubdivisionMode]

```{=tex}

```
rtcSetGeometrySubdivisionMode
-----------------------------

#### NAME {#name}

    rtcSetGeometrySubdivisionMode - sets the subdivision mode
      of a subdivision geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometrySubdivisionMode(
      RTCGeometry geometry,
      unsigned int topologyID,
      enum RTCSubdivisionMode mode
    );

#### DESCRIPTION {#description}

The `rtcSetGeometrySubdivisionMode` function sets the subdivision mode
(`mode` parameter) for the topology (`topologyID` parameter) of the
specified subdivision geometry (`geometry` parameter).

The subdivision modes can be used to force linear interpolation for
certain parts of the subdivision mesh:

-   `RTC_SUBDIVISION_MODE_NO_BOUNDARY`: Boundary patches are ignored.
    This way each rendered patch has a full set of control vertices.

-   `RTC_SUBDIVISION_MODE_SMOOTH_BOUNDARY`: The sequence of boundary
    control points are used to generate a smooth B-spline boundary
    curve (default mode).

-   `RTC_SUBDIVISION_MODE_PIN_CORNERS`: Corner vertices are pinned to
    their location during subdivision.

-   `RTC_SUBDIVISION_MODE_PIN_BOUNDARY`: All vertices at the border are
    pinned to their location during subdivision. This way the boundary
    is interpolated linearly. This mode is typically used for texturing
    to also map texels at the border of the texture to the mesh.

-   `RTC_SUBDIVISION_MODE_PIN_ALL`: All vertices at the border are
    pinned to their location during subdivision. This way all patches
    are linearly interpolated.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_SUBDIVISION]

```{=tex}

```
rtcSetGeometryVertexAttributeTopology
-------------------------------------

#### NAME {#name}

    rtcSetGeometryVertexAttributeTopology - binds a vertex
      attribute to a topology of the geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcSetGeometryVertexAttributeTopology(
      RTCGeometry geometry,
      unsigned int vertexAttributeID,
      unsigned int topologyID
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryVertexAttributeTopology` function binds a vertex
attribute buffer slot (`vertexAttributeID` argument) to a topology
(`topologyID` argument) for the specified subdivision geometry
(`geometry` argument). Standard vertex buffers are always bound to the
default topology (topology 0) and cannot be bound differently. A vertex
attribute buffer always uses the topology it is bound to when used in
the `rtcInterpolate` and `rtcInterpolateN` calls.

A topology with ID `i` consists of a subdivision mode set through
`rtcSetGeometrySubdivisionMode` and the index buffer bound to the index
buffer slot `i`. This index buffer can assign indices for each face of
the subdivision geometry that are different to the indices of the
default topology. These new indices can for example be used to
introduce additional borders into the subdivision mesh to map multiple
textures onto one subdivision geometry.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcSetGeometrySubdivisionMode], [rtcInterpolate],
[rtcInterpolateN]

```{=tex}

```
rtcSetGeometryDisplacementFunction
----------------------------------

#### NAME {#name}

    rtcSetGeometryDisplacementFunction - sets the displacement function
      for a subdivision geometry

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCDisplacementFunctionNArguments
    {
      void* geometryUserPtr;
      RTCGeometry geometry;
      unsigned int primID;
      unsigned int timeStep;
      const float* u;
      const float* v;
      const float* Ng_x;
      const float* Ng_y;
      const float* Ng_z;
      float* P_x;
      float* P_y;
      float* P_z;
      unsigned int N;
    };

    typedef void (*RTCDisplacementFunctionN)(
       const struct RTCDisplacementFunctionNArguments* args
    );

    void rtcSetGeometryDisplacementFunction(
      RTCGeometry geometry,
      RTCDisplacementFunctionN displacement
    );

#### DESCRIPTION {#description}

The `rtcSetGeometryDisplacementFunction` function registers a
displacement callback function (`displacement` argument) for the
specified subdivision geometry (`geometry` argument).

Only a single callback function can be registered per geometry, and
further invocations overwrite the previously set callback function.
Passing `NULL` as function pointer disables the registered callback
function.

The registered displacement callback function is invoked to displace
points on the subdivision geometry during spatial acceleration
structure construction, during the `rtcCommitScene` call.

The callback function of type `RTCDisplacementFunctionN` is invoked
with a number of arguments stored inside the
`RTCDisplacementFunctionNArguments` structure. The provided user data
pointer of the geometry (`geometryUserPtr` member) can be used to point
to the application's representation of the subdivision mesh. A number
`N` of points to displace are specified in a structure of array layout.
For each point to displace, the local patch UV coordinates (`u` and `v`
arrays), the normalized geometry normal (`Ng_x`, `Ng_y`, and `Ng_z`
arrays), and the position (`P_x`, `P_y`, and `P_z` arrays) are
provided. The task of the displacement function is to use this
information and change the position data.

The geometry handle (`geometry` member) and primitive ID (`primID`
member) of the patch to displace are additionally provided as well as
the time step `timeStep`, which can be important if the displacement is
time-dependent and motion blur is used.

All passed arrays must be aligned to 64 bytes and properly padded to
make wide vector processing inside the displacement function easily
possible.

Also see tutorial [Displacement Geometry] for an example of how to
use the displacement mapping functions.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[RTC\_GEOMETRY\_TYPE\_SUBDIVISION]

```{=tex}

```
rtcGetGeometryFirstHalfEdge
---------------------------

#### NAME {#name}

    rtcGetGeometryFirstHalfEdge - returns the first half edge of a face

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    unsigned int rtcGetGeometryFirstHalfEdge(
      RTCGeometry geometry,
      unsigned int faceID
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryFirstHalfEdge` function returns the ID of the first
half edge belonging to the specified face (`faceID` argument). For
instance in the following example the first half edge of face `f1` is
`e4`.

![][imgHalfEdges]

This function can only be used for subdivision geometries. As all
topologies of a subdivision geometry share the same face buffer the
function does not depend on the topology ID.

Here f0 to f7 are 8 quadrilateral faces with 4 vertices each. The edges
e0 to e23 of these faces are shown with their orientation. For each
face the ID of the edges corresponds to the slots the face occupies in
the index array of the geometry. E.g. as the indices of face f1 start
at location 4 of the index array, the first edge is edge e4, the next
edge e5, etc.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetGeometryFirstHalfEdge], [rtcGetGeometryFace],
[rtcGetGeometryOppositeHalfEdge], [rtcGetGeometryNextHalfEdge],
[rtcGetGeometryPreviousHalfEdge]

```{=tex}

```
rtcGetGeometryFace
------------------

#### NAME {#name}

    rtcGetGeometryFace - returns the face of some half edge

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    unsigned int rtcGetGeometryFace(
      RTCGeometry geometry,
      unsigned int edgeID
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryFace` function returns the ID of the face the
specified half edge (`edgeID` argument) belongs to. For instance in the
following example the face `f1` is returned for edges `e4`, `e5`, `e6`,
and `e7`.

![][imgHalfEdges]

This function can only be used for subdivision geometries. As all
topologies of a subdivision geometry share the same face buffer the
function does not depend on the topology ID.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetGeometryFirstHalfEdge], [rtcGetGeometryFace],
[rtcGetGeometryOppositeHalfEdge], [rtcGetGeometryNextHalfEdge],
[rtcGetGeometryPreviousHalfEdge]

```{=tex}

```
rtcGetGeometryNextHalfEdge
--------------------------

#### NAME {#name}

    rtcGetGeometryNextHalfEdge - returns the next half edge

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    unsigned int rtcGetGeometryNextHalfEdge(
      RTCGeometry geometry,
      unsigned int edgeID
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryNextHalfEdge` function returns the ID of the next
half edge of the specified half edge (`edgeID` argument). For instance
in the following example the next half edge of `e10` is `e11`.

![][imgHalfEdges]

This function can only be used for subdivision geometries. As all
topologies of a subdivision geometry share the same face buffer the
function does not depend on the topology ID.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetGeometryFirstHalfEdge], [rtcGetGeometryFace],
[rtcGetGeometryOppositeHalfEdge], [rtcGetGeometryNextHalfEdge],
[rtcGetGeometryPreviousHalfEdge]

```{=tex}

```
rtcGetGeometryPreviousHalfEdge
------------------------------

#### NAME {#name}

    rtcGetGeometryPreviousHalfEdge - returns the previous half edge

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    unsigned int rtcGetGeometryPreviousHalfEdge(
      RTCGeometry geometry,
      unsigned int edgeID
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryPreviousHalfEdge` function returns the ID of the
previous half edge of the specified half edge (`edgeID` argument). For
instance in the following example the previous half edge of `e6` is
`e5`.

![][imgHalfEdges]

This function can only be used for subdivision geometries. As all
topologies of a subdivision geometry share the same face buffer the
function does not depend on the topology ID.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetGeometryFirstHalfEdge], [rtcGetGeometryFace],
[rtcGetGeometryOppositeHalfEdge], [rtcGetGeometryNextHalfEdge],
[rtcGetGeometryPreviousHalfEdge]

```{=tex}

```
rtcGetGeometryOppositeHalfEdge
------------------------------

#### NAME {#name}

    rtcGetGeometryOppositeHalfEdge - returns the opposite half edge

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    unsigned int rtcGetGeometryOppositeHalfEdge(
      RTCGeometry geometry,
      unsigned int topologyID,
      unsigned int edgeID
    );

#### DESCRIPTION {#description}

The `rtcGetGeometryOppositeHalfEdge` function returns the ID of the
opposite half edge of the specified half edge (`edgeID` argument) in
the specified topology (`topologyID` argument). For instance in the
following example the opposite half edge of `e6` is `e16`.

![][imgHalfEdges]

An opposite half edge does not exist if the specified half edge has
either no neighboring face, or more than 2 neighboring faces. In these
cases the function just returns the same edge `edgeID` again.

This function can only be used for subdivision geometries. The function
depends on the topology as the topologies of a subdivision geometry
have different index buffers assigned.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcGetGeometryFirstHalfEdge], [rtcGetGeometryFace],
[rtcGetGeometryOppositeHalfEdge], [rtcGetGeometryNextHalfEdge],
[rtcGetGeometryPreviousHalfEdge]

```{=tex}

```
rtcInterpolate
--------------

#### NAME {#name}

    rtcInterpolate - interpolates vertex attributes

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCInterpolateArguments
    {
      RTCGeometry geometry;
      unsigned int primID;
      float u;
      float v;
      enum RTCBufferType bufferType;
      unsigned int bufferSlot;
      float* P;
      float* dPdu;
      float* dPdv;
      float* ddPdudu;
      float* ddPdvdv;
      float* ddPdudv;
      unsigned int valueCount;
    };

    void rtcInterpolate(
      const struct RTCInterpolateArguments* args
    );

#### DESCRIPTION {#description}

The `rtcInterpolate` function smoothly interpolates per-vertex data
over the geometry. This interpolation is supported for triangle meshes,
quad meshes, curve geometries, and subdivision geometries. Apart from
interpolating the vertex attribute itself, it is also possible to get
the first and second order derivatives of that value. This
interpolation ignores displacements of subdivision surfaces and always
interpolates the underlying base surface.

The `rtcInterpolate` call gets passed a number of arguments inside a
structure of type `RTCInterpolateArguments`. For some geometry
(`geometry` parameter) this function smoothly interpolates the
per-vertex data stored inside the specified geometry buffer
(`bufferType` and `bufferSlot` parameters) to the u/v location (`u` and
`v` parameters) of the primitive (`primID` parameter). The number of
floating point values to interpolate and store to the destination
arrays can be specified using the `valueCount` parameter. As
interpolation buffer, one can specify vertex buffers
(`RTC_BUFFER_TYPE_VERTEX`) and vertex attribute buffers
(`RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE`) as well.

The `rtcInterpolate` call stores `valueCount` number of interpolated
floating point values to the memory location pointed to by `P`. One can
avoid storing the interpolated value by setting `P` to `NULL`.

The first order derivative of the interpolation by u and v are stored
at the `dPdu` and `dPdv` memory locations. One can avoid storing first
order derivatives by setting both `dPdu` and `dPdv` to `NULL`.

The second order derivatives are stored at the `ddPdudu`, `ddPdvdv`,
and `ddPdudv` memory locations. One can avoid storing second order
derivatives by setting these three pointers to `NULL`.

To use `rtcInterpolate` for a geometry, all changes to that geometry
must be properly committed using `rtcCommitGeometry`.

All input buffers and output arrays must be padded to 16 bytes, as the
implementation uses 16-byte SSE instructions to read and write into
these buffers.

See tutorial [Interpolation] for an example of using the
`rtcInterpolate` function.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcInterpolateN]

```{=tex}

```
rtcInterpolateN
---------------

#### NAME {#name}

    rtcInterpolateN - performs N interpolations of vertex attribute data

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCInterpolateNArguments
    {
      RTCGeometry geometry;
      const void* valid;
      const unsigned int* primIDs;
      const float* u;
      const float* v;
      unsigned int N;
      enum RTCBufferType bufferType;
      unsigned int bufferSlot;
      float* P;
      float* dPdu;
      float* dPdv;
      float* ddPdudu;
      float* ddPdvdv;
      float* ddPdudv;
      unsigned int valueCount;
    };

    void rtcInterpolateN(
      const struct RTCInterpolateNArguments* args
    );

#### DESCRIPTION {#description}

The `rtcInterpolateN` is similar to `rtcInterpolate`, but performs `N`
many interpolations at once. It additionally gets an array of u/v
coordinates and a valid mask (`valid` parameter) that specifies which
of these coordinates are valid. The valid mask points to `N` integers,
and a value of -1 denotes valid and 0 invalid. If the valid pointer is
`NULL` all elements are considers valid. The destination arrays are
filled in structure of array (SOA) layout. The value `N` must be
divisible by 4.

To use `rtcInterpolateN` for a geometry, all changes to that geometry
must be properly committed using `rtcCommitGeometry`.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcInterpolate]

```{=tex}

```
rtcNewBuffer
------------

#### NAME {#name}

    rtcNewBuffer - creates a new data buffer

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCBuffer rtcNewBuffer(
      RTCDevice device,
      size_t byteSize
    );

#### DESCRIPTION {#description}

The `rtcNewBuffer` function creates a new data buffer object of
specified size in bytes (`byteSize` argument) that is bound to the
specified device (`device` argument). The buffer object is reference
counted with an initial reference count of 1. The returned buffer
object can be released using the `rtcReleaseBuffer` API call. The
specified number of bytes are allocated at buffer construction time and
deallocated when the buffer is destroyed.

When the buffer will be used as a vertex buffer
(`RTC_BUFFER_TYPE_VERTEX` and `RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE`), the
last buffer element must be readable using 16-byte SSE load
instructions, thus padding the last element is required for certain
layouts. E.g. a standard `float3` vertex buffer layout should add
storage for at least one more float to the end of the buffer.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcRetainBuffer], [rtcReleaseBuffer]

```{=tex}

```
rtcNewSharedBuffer
------------------

#### NAME {#name}

    rtcNewSharedBuffer - creates a new shared data buffer

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCBuffer rtcNewSharedBuffer(
      RTCDevice device,
      void* ptr,
      size_t byteSize
    );

#### DESCRIPTION {#description}

The `rtcNewSharedBuffer` function creates a new shared data buffer
object bound to the specified device (`device` argument). The buffer
object is reference counted with an initial reference count of 1. The
buffer can be released using the `rtcReleaseBuffer` function.

At construction time, the pointer to the user-managed buffer data
(`ptr` argument) including its size in bytes (`byteSize` argument) is
provided to create the buffer. At buffer construction time no buffer
data is allocated, but the buffer data provided by the application is
used. The buffer data must remain valid for as long as the buffer may
be used, and the user is responsible to free the buffer data when no
longer required.

When the buffer will be used as a vertex buffer
(`RTC_BUFFER_TYPE_VERTEX` and `RTC_BUFFER_TYPE_VERTEX_ATTRIBUTE`), the
last buffer element must be readable using 16-byte SSE load
instructions, thus padding the last element is required for certain
layouts. E.g. a standard `float3` vertex buffer layout should add
storage for at least one more float to the end of the buffer.

The data pointer (`ptr` argument) must be aligned to 4 bytes; otherwise
the `rtcNewSharedBuffer` function will fail.

#### EXIT STATUS {#exit-status}

On failure `NULL` is returned and an error code is set that can be
queried using `rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcRetainBuffer], [rtcReleaseBuffer]

```{=tex}

```
rtcRetainBuffer
---------------

#### NAME {#name}

    rtcRetainBuffer - increments the buffer reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcRetainBuffer(RTCBuffer buffer);

#### DESCRIPTION {#description}

Buffer objects are reference counted. The `rtcRetainBuffer` function
increments the reference count of the passed buffer object (`buffer`
argument). This function together with `rtcReleaseBuffer` allows to use
the internal reference counting in a C++ wrapper class to handle the
ownership of the object.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewBuffer], [rtcReleaseBuffer]

```{=tex}

```
rtcReleaseBuffer
----------------

#### NAME {#name}

    rtcReleaseBuffer - decrements the buffer reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcReleaseBuffer(RTCBuffer buffer);

#### DESCRIPTION {#description}

Buffer objects are reference counted. The `rtcReleaseBuffer` function
decrements the reference count of the passed buffer object (`buffer`
argument). When the reference count falls to 0, the buffer gets
destroyed.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewBuffer], [rtcRetainBuffer]

```{=tex}

```
rtcGetBufferData
----------------

#### NAME {#name}

    rtcGetBufferData - gets a pointer to the buffer data

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void* rtcGetBufferData(RTCBuffer buffer);

#### DESCRIPTION {#description}

The `rtcGetBufferData` function returns a pointer to the buffer data of
the specified buffer object (`buffer` argument).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewBuffer]

```{=tex}

```
RTCRay
------

#### NAME {#name}

    RTCRay - single ray structure

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    struct RTC_ALIGN(16) RTCRay
    {
      float org_x;        // x coordinate of ray origin
      float org_y;        // y coordinate of ray origin
      float org_z;        // z coordinate of ray origin
      float tnear;        // start of ray segment

      float dir_x;        // x coordinate of ray direction
      float dir_y;        // y coordinate of ray direction
      float dir_z;        // z coordinate of ray direction
      float time;         // time of this ray for motion blur

      float tfar;         // end of ray segment (set to hit distance)
      unsigned int mask;  // ray mask
      unsigned int id;    // ray ID
      unsigned int flags; // ray flags
    };

#### DESCRIPTION {#description}

The `RTCRay` structure defines the ray layout for a single ray. The ray
contains the origin (`org_x`, `org_y`, `org_z` members), direction
vector (`dir_x`, `dir_y`, `dir_z` members), and ray segment (`tnear`
and `tfar` members). The ray direction does not have to be normalized,
and only the parameter range specified by the `tnear`/`tfar` interval
is considered valid.

The ray segment must be in the range $[0, \infty]$, thus ranges that
start behind the ray origin are not allowed, but ranges can reach to
infinity.

The ray further contains a motion blur time in the range $[0, 1]$
(`time` member), a ray mask (`mask` member), a ray ID (`id` member),
and ray flags (`flags` member). The ray mask can be used to mask out
some geometries for some rays (see `rtcSetGeometryMask` for more
details). The ray ID can be used to identify a ray inside a callback
function, even if the order of rays inside a ray packet has changed.

The `embree4/rtcore_ray.h` header additionally defines the same ray
structure in structure of array (SOA) layout for API functions
accepting ray packets of size 4 (`RTCRay4` type), size 8 (`RTCRay8`
type), and size 16 (`RTCRay16` type). The header additionally defines
an `RTCRayNt` template for ray packets of an arbitrary compile-time
size.

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTCHit]

```{=tex}

```
RTCHit
------

#### NAME {#name}

    RTCHit - single hit structure

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCHit
    {
      float Ng_x;                                        // x coordinate of geometry normal
      float Ng_y;                                        // y coordinate of geometry normal
      float Ng_z;                                        // z coordinate of geometry normal

      float u;                                           // barycentric u coordinate of hit
      float v;                                           // barycentric v coordinate of hit

      unsigned int primID;                               // geometry ID
      unsigned int geomID;                               // primitive ID
      unsigned int instID[RTC_MAX_INSTANCE_LEVEL_COUNT]; // instance ID
    };

#### DESCRIPTION {#description}

The `RTCHit` type defines the type of a ray/primitive intersection
result. The hit contains the unnormalized geometric normal in object
space at the hit location (`Ng_x`, `Ng_y`, `Ng_z` members), the
barycentric u/v coordinates of the hit (`u` and `v` members), as well
as the primitive ID (`primID` member), geometry ID (`geomID` member),
and instance ID stack (`instID` member) of the hit. The parametric
intersection distance is not stored inside the hit, but stored inside
the `tfar` member of the ray.

The `embree4/rtcore_ray.h` header additionally defines the same hit
structure in structure of array (SOA) layout for hit packets of size 4
(`RTCHit4` type), size 8 (`RTCHit8` type), and size 16 (`RTCHit16`
type). The header additionally defines an `RTCHitNt` template for hit
packets of an arbitrary compile-time size.

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTCRay], [Multi-Level Instancing]

```{=tex}

```
RTCRayHit
---------

#### NAME {#name}

    RTCRayHit - combined single ray/hit structure

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    struct RTCORE_ALIGN(16) RTCRayHit
    {
      struct RTCRay ray;
      struct RTCHit hit;
    };

#### DESCRIPTION {#description}

The `RTCRayHit` structure is used as input for the `rtcIntersect`-type
functions and stores the ray to intersect and some hit fields that hold
the intersection result afterwards.

The `embree4/rtcore_ray.h` header additionally defines the same ray/hit
structure in structure of array (SOA) layout for API functions
accepting ray packets of size 4 (`RTCRayHit4` type), size 8
(`RTCRayHit8` type), and size 16 (`RTCRayHit16` type). The header
additionally defines an `RTCRayHitNt` template to generate ray/hit
packets of an arbitrary compile-time size.

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTCRay], [RTCHit]

```{=tex}

```
RTCRayN
-------

#### NAME {#name}

    RTCRayN - ray packet of runtime size

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    struct RTCRayN;

    float& RTCRayN_org_x(RTCRayN* ray, unsigned int N, unsigned int i);
    float& RTCRayN_org_y(RTCRayN* ray, unsigned int N, unsigned int i);
    float& RTCRayN_org_z(RTCRayN* ray, unsigned int N, unsigned int i);
    float& RTCRayN_tnear(RTCRayN* ray, unsigned int N, unsigned int i);

    float& RTCRayN_dir_x(RTCRayN* ray, unsigned int N, unsigned int i);
    float& RTCRayN_dir_y(RTCRayN* ray, unsigned int N, unsigned int i);
    float& RTCRayN_dir_z(RTCRayN* ray, unsigned int N, unsigned int i);
    float& RTCRayN_time (RTCRayN* ray, unsigned int N, unsigned int i);

    float&        RTCRayN_tfar (RTCRayN* ray, unsigned int N, unsigned int i);
    unsigned int& RTCRayN_mask (RTCRayN* ray, unsigned int N, unsigned int i);
    unsigned int& RTCRayN_id   (RTCRayN* ray, unsigned int N, unsigned int i);
    unsigned int& RTCRayN_flags(RTCRayN* ray, unsigned int N, unsigned int i);

#### DESCRIPTION {#description}

When the ray packet size is not known at compile time (e.g. when Embree
returns a ray packet in the `RTCFilterFuncN` callback function), Embree
uses the `RTCRayN` type for ray packets. These ray packets can only
have sizes of 1, 4, 8, or 16. No other packet size will be used.

You can either implement different special code paths for each of these
possible packet sizes and cast the ray to the appropriate ray packet
type, or implement one general code path that uses the `RTCRayN_XXX`
helper functions to access the ray packet components.

These helper functions get a pointer to the ray packet (`ray`
argument), the packet size (`N` argument), and returns a reference to a
component (e.g. x-component of origin) of the the i-th ray of the
packet (`i` argument).

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTCHitN]

```{=tex}

```
RTCHitN
-------

#### NAME {#name}

    RTCHitN - hit packet of runtime size

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct HitN;

    float& RTCHitN_Ng_x(RTCHitN* hit, unsigned int N, unsigned int i);
    float& RTCHitN_Ng_y(RTCHitN* hit, unsigned int N, unsigned int i);
    float& RTCHitN_Ng_z(RTCHitN* hit, unsigned int N, unsigned int i);

    float& RTCHitN_u(RTCHitN* hit, unsigned int N, unsigned int i);
    float& RTCHitN_v(RTCHitN* hit, unsigned int N, unsigned int i);

    unsigned& RTCHitN_primID(RTCHitN* hit, unsigned int N, unsigned int i);
    unsigned& RTCHitN_geomID(RTCHitN* hit, unsigned int N, unsigned int i);
    unsigned& RTCHitN_instID(RTCHitN* hit, unsigned int N, unsigned int i, unsigned int level);

#### DESCRIPTION {#description}

When the hit packet size is not known at compile time (e.g. when Embree
returns a hit packet in the `RTCFilterFuncN` callback function), Embree
uses the `RTCHitN` type for hit packets. These hit packets can only
have sizes of 1, 4, 8, or 16. No other packet size will be used.

You can either implement different special code paths for each of these
possible packet sizes and cast the hit to the appropriate hit packet
type, or implement one general code path that uses the `RTCHitN_XXX`
helper functions to access hit packet components.

These helper functions get a pointer to the hit packet (`hit`
argument), the packet size (`N` argument), and returns a reference to a
component (e.g. x component of `Ng`) of the the i-th hit of the packet
(`i` argument).

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTCRayN]

```{=tex}

```
RTCRayHitN
----------

#### NAME {#name}

    RTCRayHitN - combined ray/hit packet of runtime size

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    struct RTCRayHitN;

    struct RTCRayN* RTCRayHitN_RayN(struct RTCRayHitN* rayhit, unsigned int N);
    struct RTCHitN* RTCRayHitN_HitN(struct RTCRayHitN* rayhit, unsigned int N);

#### DESCRIPTION {#description}

When the packet size of a ray/hit structure is not known at compile
time (e.g. when Embree returns a ray/hit packet in the
`RTCIntersectFunctionN` callback function), Embree uses the
`RTCRayHitN` type for ray packets. These ray/hit packets can only have
sizes of 1, 4, 8, or 16. No other packet size will be used.

You can either implement different special code paths for each of these
possible packet sizes and cast the ray/hit to the appropriate ray/hit
packet type, or extract the `RTCRayN` and `RTCHitN` components using
the `rtcGetRayN` and `rtcGetHitN` helper functions and use the
`RTCRayN_XXX` and `RTCHitN_XXX` functions to access the ray and hit
parts of the structure.

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[RTCHitN]

```{=tex}

```
RTCFeatureFlags
---------------

#### NAME {#name}

    RTCFeatureFlags - specifies features to enable
      for ray queries

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore_ray.h>

    enum RTCFeatureFlags
    {
      RTC_FEATURE_FLAG_NONE = 0,
      
      RTC_FEATURE_FLAG_MOTION_BLUR = 1 << 0,

      RTC_FEATURE_FLAG_TRIANGLE = 1 << 1,
      RTC_FEATURE_FLAG_QUAD = 1 << 2,
      RTC_FEATURE_FLAG_GRID = 1 << 3,
      RTC_FEATURE_FLAG_SUBDIVISION = 1 << 4,
      RTC_FEATURE_FLAG_POINT = ... ,
      RTC_FEATURE_FLAG_CURVES = ... ,
     
      RTC_FEATURE_FLAG_CONE_LINEAR_CURVE = 1 << 5,
      RTC_FEATURE_FLAG_ROUND_LINEAR_CURVE  = 1 << 6,
      RTC_FEATURE_FLAG_FLAT_LINEAR_CURVE = 1 << 7,

      RTC_FEATURE_FLAG_ROUND_BEZIER_CURVE = 1 << 8,
      RTC_FEATURE_FLAG_FLAT_BEZIER_CURVE = 1 << 9,
      RTC_FEATURE_FLAG_NORMAL_ORIENTED_BEZIER_CURVE = 1 << 10,

      RTC_FEATURE_FLAG_ROUND_BSPLINE_CURVE = 1 << 11,
      RTC_FEATURE_FLAG_FLAT_BSPLINE_CURVE = 1 << 12,
      RTC_FEATURE_FLAG_NORMAL_ORIENTED_BSPLINE_CURVE = 1 << 13,

      RTC_FEATURE_FLAG_ROUND_HERMITE_CURVE = 1 << 14,
      RTC_FEATURE_FLAG_FLAT_HERMITE_CURVE = 1 << 15,
      RTC_FEATURE_FLAG_NORMAL_ORIENTED_HERMITE_CURVE = 1 << 16,

      RTC_FEATURE_FLAG_ROUND_CATMULL_ROM_CURVE = 1 << 17,
      RTC_FEATURE_FLAG_FLAT_CATMULL_ROM_CURVE = 1 << 18,
      RTC_FEATURE_FLAG_NORMAL_ORIENTED_CATMULL_ROM_CURVE = 1 << 19,

      RTC_FEATURE_FLAG_SPHERE_POINT = 1 << 20,
      RTC_FEATURE_FLAG_DISC_POINT = 1 << 21,
      RTC_FEATURE_FLAG_ORIENTED_DISC_POINT = 1 << 22,

      RTC_FEATURE_FLAG_ROUND_CURVES = ... ,
      RTC_FEATURE_FLAG_FLAT_CURVES = ... ,
      RTC_FEATURE_FLAG_NORMAL_ORIENTED_CURVES = ... ,
      
      RTC_FEATURE_FLAG_LINEAR_CURVES = ... ,
      RTC_FEATURE_FLAG_BEZIER_CURVES = ... ,
      RTC_FEATURE_FLAG_BSPLINE_CURVES = ... ,
      RTC_FEATURE_FLAG_HERMITE_CURVES = ... ,
      
      RTC_FEATURE_FLAG_INSTANCE = 1 << 23,

      RTC_FEATURE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS = 1 << 24,
      RTC_FEATURE_FLAG_FILTER_FUNCTION_IN_GEOMETRY = 1 << 25,
      RTC_FEATURE_FLAG_FILTER_FUNCTION = ... ,

      RTC_FEATURE_FLAG_USER_GEOMETRY_CALLBACK_IN_ARGUMENTS = 1 << 26,
      RTC_FEATURE_FLAG_USER_GEOMETRY_CALLBACK_IN_GEOMETRY = 1 << 27,
      RTC_FEATURE_FLAG_USER_GEOMETRY = ... ,

      RTC_FEATURE_FLAG_32_BIT_RAY_MASK = 1 << 28,

      RTC_FEATURE_FLAG_ALL = 0xffffffff
    };

#### DESCRIPTION {#description}

The `RTCFeatureFlags` enum specify a bit mask to enable specific ray
tracing features for ray query operations. The feature flags are passed
to the `rtcIntersect1/4/8/16` and `rtcOccluded1/4/8/16` functions
through the `RTCIntersectArguments` and `RTCOccludedArguments`
structures. Only a ray tracing feature whose bit is enabled in the
feature mask can get used. If a feature bit is not set, the behaviour
is undefined, thus the feature may work or not. To enable multiple
features the respective features have to get combined using a bitwise
`OR` operation.

The purpose of feature flags is to reduce code size on the GPU by
enabling just the features required to render the scene. On the CPU
there is no need to use feature flags, and the default of all features
enabled (`RTC_FEATURE_FLAG_ALL`) can just be kept.

The following features can get enabled using feature flags:

-   RTC\_FEATURE\_FLAG\_MOTION\_BLUR: Enables motion blur for all
    geometry types.

-   RTC\_FEATURE\_FLAG\_TRIANGLE: Enables triangle geometries
    (RTC\_GEOMETRY\_TYPE\_TRIANGLE).

-   RTC\_FEATURE\_FLAG\_QUAD: Enables quad geometries
    (RTC\_GEOMETRY\_TYPE\_QUAD).

-   RTC\_FEATURE\_FLAG\_GRID: Enables grid geometries
    (RTC\_GEOMETRY\_TYPE\_GRID).

-   RTC\_FEATURE\_FLAG\_SUBDIVISION: Enables subdivision geometries
    (RTC\_GEOMETRY\_TYPE\_SUBDIVISION).

-   RTC\_FEATURE\_FLAG\_POINT: Enables all point geometry types
    (RTC\_GEOMETRY\_TYPE\_XXX\_POINT)

-   RTC\_FEATURE\_FLAG\_CURVES: Enables all curve geometry types
    (RTC\_GEOMETRY\_TYPE\_XXX\_YYY\_CURVE)

-   RTC\_FEATURE\_FLAG\_ROUND\_CURVES: Enables all round curves
    (RTC\_GEOMETRY\_TYPE\_ROUND\_XXX\_CURVE).

-   RTC\_FEATURE\_FLAG\_FLAT\_CURVES: Enables all flat curves
    (RTC\_GEOMETRY\_TYPE\_FLAT\_XXX\_CURVE).

-   RTC\_FEATURE\_FLAG\_NORMAL\_ORIENTED\_CURVES: Enables all normal
    oriented curves
    (RTC\_GEOMETRY\_TYPE\_NORMAL\_ORIENTED\_XXX\_CURVE).

-   RTC\_FEATURE\_FLAG\_LINEAR\_CURVES: Enables all linear curves
    (RTC\_GEOMETRY\_TYPE\_XXX\_LINEAR\_CURVE).

-   RTC\_FEATURE\_FLAG\_BEZIER\_CURVES: Enables all Bézier curves
    (RTC\_GEOMETRY\_TYPE\_XXX\_BEZIER\_CURVE).

-   RTC\_FEATURE\_FLAG\_BSPLINE\_CURVES: Enables all B-spline curves
    (RTC\_GEOMETRY\_TYPE\_XXX\_BSPLINE\_CURVE).

-   RTC\_FEATURE\_FLAG\_HERMITE\_CURVES: Enables all Hermite curves
    (RTC\_GEOMETRY\_TYPE\_XXX\_HERMITE\_CURVE).

-   RTC\_FEATURE\_FLAG\_CONE\_LINEAR\_CURVE: Enables cone geometry type
    (RTC\_GEOMETRY\_TYPE\_CONE\_LINEAR\_CURVE).

-   RTC\_FEATURE\_FLAG\_ROUND\_LINEAR\_CURVE: Enables round linear
    curves (RTC\_GEOMETRY\_TYPE\_ROUND\_LINEAR\_CURVE).

-   RTC\_FEATURE\_FLAG\_FLAT\_LINEAR\_CURVE: Enables flat linear curves
    (RTC\_GEOMETRY\_TYPE\_FLAT\_LINEAR\_CURVE).

-   RTC\_FEATURE\_FLAG\_ROUND\_BEZIER\_CURVE: Enables round Bézier
    curves (RTC\_GEOMETRY\_TYPE\_ROUND\_BEZIER\_CURVE).

-   RTC\_FEATURE\_FLAG\_FLAT\_BEZIER\_CURVE: Enables flat Bézier curves
    (RTC\_GEOMETRY\_TYPE\_FLAT\_BEZIER\_CURVE).

-   RTC\_FEATURE\_FLAG\_NORMAL\_ORIENTED\_BEZIER\_CURVE: Enables normal
    oriented Bézier curves
    (RTC\_GEOMETRY\_TYPE\_NORMAL\_ORIENTED\_BEZIER\_CURVE).

-   RTC\_FEATURE\_FLAG\_ROUND\_BSPLINE\_CURVE: Enables round B-spline
    curves (RTC\_GEOMETRY\_TYPE\_ROUND\_BSPLINE\_CURVE).

-   RTC\_FEATURE\_FLAG\_FLAT\_BSPLINE\_CURVE: Enables flat B-spline
    curves (RTC\_GEOMETRY\_TYPE\_FLAT\_BSPLINE\_CURVE).

-   RTC\_FEATURE\_FLAG\_NORMAL\_ORIENTED\_BSPLINE\_CURVE: Enables
    normal oriented B-spline curves
    (RTC\_GEOMETRY\_TYPE\_NORMAL\_ORIENTED\_BSPLINE\_CURVE).

-   RTC\_FEATURE\_FLAG\_ROUND\_HERMITE\_CURVE: Enables round Hermite
    curves (RTC\_GEOMETRY\_TYPE\_ROUND\_HERMITE\_CURVE).

-   RTC\_FEATURE\_FLAG\_FLAT\_HERMITE\_CURVE: Enables flat Hermite
    curves (RTC\_GEOMETRY\_TYPE\_FLAT\_HERMITE\_CURVE).

-   RTC\_FEATURE\_FLAG\_NORMAL\_ORIENTED\_HERMITE\_CURVE: Enables
    normal oriented Hermite curves
    (RTC\_GEOMETRY\_TYPE\_NORMAL\_ORIENTED\_HERMITE\_CURVE).

-   RTC\_FEATURE\_FLAG\_ROUND\_CATMULL\_ROM\_CURVE: Enables round
    Catmull Rom curves
    (RTC\_GEOMETRY\_TYPE\_ROUND\_CATMULL\_ROM\_CURVE).

-   RTC\_FEATURE\_FLAG\_FLAT\_CATMULL\_ROM\_CURVE: Enables flat Catmull
    Rom curves (RTC\_GEOMETRY\_TYPE\_FLAT\_CATMULL\_ROM\_CURVE).

-   RTC\_FEATURE\_FLAG\_NORMAL\_ORIENTED\_CATMULL\_ROM\_CURVE: Enables
    normal oriented Catmull Rom curves
    (RTC\_GEOMETRY\_TYPE\_NORMAL\_ORIENTED\_CATMULL\_ROM\_CURVE).

-   RTC\_FEATURE\_FLAG\_SPHERE\_POINT: Enables sphere geometry type
    (RTC\_GEOMETRY\_TYPE\_SPHERE\_POINT).

-   RTC\_FEATURE\_FLAG\_DISC\_POINT: Enables disc geometry type
    (RTC\_GEOMETRY\_TYPE\_DISC\_POINT).

-   RTC\_FEATURE\_FLAG\_ORIENTED\_DISC\_POINT: Enables oriented disc
    geometry types (RTC\_GEOMETRY\_TYPE\_ORIENTED\_DISC\_POINT).

-   RTC\_FEATURE\_FLAG\_INSTANCE: Enables instance geometries
    (RTC\_GEOMETRY\_TYPE\_INSTANCE).

-   RTC\_FEATURE\_FLAG\_FILTER\_FUNCTION\_IN\_ARGUMENTS: Enables filter
    functions passed through intersect arguments.

-   RTC\_FEATURE\_FLAG\_FILTER\_FUNCTION\_IN\_GEOMETRY: Enable filter
    functions passed through geometry.

-   RTC\_FEATURE\_FLAG\_FILTER\_FUNCTION: Enables filter functions
    (argument and geometry version).

-   RTC\_FEATURE\_FLAG\_USER\_GEOMETRY\_CALLBACK\_IN\_ARGUMENTS:
    Enables RTC\_GEOMETRY\_TYPE\_USER with function pointer passed
    through intersect arguments.

-   RTC\_FEATURE\_FLAG\_USER\_GEOMETRY\_CALLBACK\_IN\_GEOMETRY: Enables
    RTC\_GEOMETRY\_TYPE\_USER with function pointer passed through
    geometry object.

-   RTC\_FEATURE\_FLAG\_USER\_GEOMETRY: Enables
    RTC\_GEOMETRY\_TYPE\_USER geometries (both argument and geometry
    callback versions).

-   RTC\_FEATURE\_FLAG\_32\_BIT\_RAY\_MASK: Enables full 32 bit ray
    masks. If not used, only the lower 7 bits in the ray mask are
    handled correctly.

-   RTC\_FEATURE\_FLAG\_ALL: Enables all features (default).

#### EXIT STATUS {#exit-status}

#### SEE ALSO {#see-also}

[rtcIntersect1], [rtcIntersect4/8/16], [rtcOccluded1],
[rtcOccluded4/8/16],

```{=tex}

```
rtcInitIntersectArguments
-------------------------

#### NAME {#name}

    rtcInitIntersectArguments - initializes the intersect arguments struct

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    enum RTCRayQueryFlags
    {
      RTC_RAY_QUERY_FLAG_NONE,
      RTC_RAY_QUERY_FLAG_INCOHERENT,
      RTC_RAY_QUERY_FLAG_COHERENT,
      RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER
    };

    struct RTCIntersectArguments
    {
      enum RTCRayQueryFlags flags;
      enum RTCFeatureFlags feature_mask;
      struct RTCRayQueryContext* context;
      RTCFilterFunctionN filter;
      RTCIntersectFunctionN intersect;
    #if RTC_MIN_WIDTH
      float minWidthDistanceFactor;
    #endif
    };

    void rtcInitIntersectArguments(
      struct RTCIntersectArguments* args
    );

#### DESCRIPTION {#description}

The `rtcInitIntersectArguments` function initializes the optional
argument struct that can get passed to the `rtcIntersect1/4/8/16`
functions to default values. The arguments struct needs to get used for
more advanced Embree features as described here.

The `flags` member can get used to enable special traversal mode. Using
the `RTC_RAY_QUERY_FLAG_INCOHERENT` flag uses an optimized traversal
algorithm for incoherent rays (default), while
`RTC_RAY_QUERY_FLAG_COHERENT` uses an optimized traversal algorithm for
coherent rays (e.g. primary camera rays).

The `feature_mask` member should get used in SYCL to just enable ray
tracing features required to render a given scene. Please see section
[RTCFeatureFlags] for a more detailed description.

The `context` member can get used to pass an optional intersection
context. It is guaranteed that the pointer to the context passed to a
ray query is directly passed to all callback functions. This way it is
possible to attach arbitrary data to the end of the context, such as a
per-ray payload. Please note that the ray pointer is not guaranteed to
be passed to the callback functions, thus reading additional data from
the ray pointer passed to callbacks is not possible. See section
[rtcInitRayQueryContext] for more details.

The `filter` member specifies a filter function to invoke for each
encountered hit. The support for the argument filter function must be
enabled for a scene by using the
`RTC_SCENE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS` scene flag. In case of
instancing this feature has to get enabled also for each instantiated
scene.

The argument filter function is invoked for each geometry for which it
got explicitely enabled using the
`rtcSetGeometryEnableFilterFunctionFromArguments` function. The
invokation of the argument filter function can also get enfored for
each geometry using the `RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER` ray
query flag. This argument filter function is invoked as a second filter
stage after the per-geometry filter function is invoked. Only rays that
passed the first filter stage are valid in this second filter stage.
Having such a per ray-query filter function can be useful to implement
modifications of the behavior of the query, such as collecting all hits
or accumulating transparencies.

The `intersect` member specifies the user geometry callback to get
invoked for each user geometry encountered during traversal. The user
geometry callback specified this way has preference over the one
specified inside the geometry.

The `minWidthDistanceFactor` value controls the target size of the
curve radii when the min-width feature is enabled. Please see the
[rtcSetGeometryMaxRadiusScale] function for more details on the
min-width feature.

#### EXIT STATUS {#exit-status}

No error code is set by this function.

#### SEE ALSO {#see-also}

[rtcIntersect1], [rtcIntersect4/8/16], [RTCFeatureFlags],
[rtcInitRayQueryContext], [RTC\_GEOMETRY\_TYPE\_USER],
[rtcSetGeometryMaxRadiusScale]

```{=tex}

```
rtcInitOccludedArguments
------------------------

#### NAME {#name}

    rtcInitOccludedArguments - initializes the occluded arguments struct

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    enum RTCRayQueryFlags
    {
      RTC_RAY_QUERY_FLAG_NONE,
      RTC_RAY_QUERY_FLAG_INCOHERENT,
      RTC_RAY_QUERY_FLAG_COHERENT,
      RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER
    };

    struct RTCOccludedArguments
    {
      enum RTCRayQueryFlags flags;
      enum RTCFeatureFlags feature_mask;
      struct RTCRayQueryContext* context;
      RTCFilterFunctionN filter;
      RTCOccludedFunctionN intersect;
    #if RTC_MIN_WIDTH
      float minWidthDistanceFactor;
    #endif
    };

    void rtcInitOccludedArguments(
      struct RTCOccludedArguments* args
    );

#### DESCRIPTION {#description}

The `rtcInitOccludedArguments` function initializes the optional
argument struct that can get passed to the `rtcOccluded1/4/8/16`
functions to default values. The arguments struct needs to get used for
more advanced Embree features as described here.

The `flags` member can get used to enable special traversal mode. Using
the `RTC_RAY_QUERY_FLAG_INCOHERENT` flag uses an optimized traversal
algorithm for incoherent rays (default), while
`RTC_RAY_QUERY_FLAG_COHERENT` uses an optimized traversal algorithm for
coherent rays (e.g. primary camera rays).

The `feature_mask` member should get used in SYCL to just enable ray
tracing features required to render a given scene. Please see section
[RTCFeatureFlags] for a more detailed description.

The `context` member can get used to pass an optional intersection
context. It is guaranteed that the pointer to the context passed to a
ray query is directly passed to all callback functions. This way it is
possible to attach arbitrary data to the end of the context, such as a
per-ray payload. Please note that the ray pointer is not guaranteed to
be passed to the callback functions, thus reading additional data from
the ray pointer passed to callbacks is not possible. See section
[rtcInitRayQueryContext] for more details.

The `filter` member specifies a filter function to invoked for each
encountered hit. The support for the argument filter function must be
enabled for a scene by using the
`RTC_SCENE_FLAG_FILTER_FUNCTION_IN_ARGUMENTS` scene flag. In case of
instancing this feature has to get enabled also for each instantiated
scene.

The argument filter function is invoked for each geometry for which it
got explicitely enabled using the
`rtcSetGeometryEnableFilterFunctionFromArguments` function. The
invokation of the argument filter function can also get enfored for
each geometry using the `RTC_RAY_QUERY_FLAG_INVOKE_ARGUMENT_FILTER` ray
query flag. This argument filter function is invoked as a second filter
stage after the per-geometry filter function is invoked. Only rays that
passed the first filter stage are valid in this second filter stage.
Having such a per ray-query filter function can be useful to implement
modifications of the behavior of the query, such as collecting all hits
or accumulating transparencies.

The `intersect` member specifies the user geometry callback to get
invoked for each user geometry encountered during traversal. The user
geometry callback specified this way has preference over the one
specified inside the geometry.

The `minWidthDistanceFactor` value controls the target size of the
curve radii when the min-width feature is enabled. Please see the
[rtcSetGeometryMaxRadiusScale] function for more details on the
min-width feature.

#### EXIT STATUS {#exit-status}

No error code is set by this function.

#### SEE ALSO {#see-also}

[rtcOccluded1], [rtcOccluded4/8/16], [RTCFeatureFlags],
[rtcInitRayQueryContext], [RTC\_GEOMETRY\_TYPE\_USER],
[rtcSetGeometryMaxRadiusScale]

```{=tex}

```
rtcInitRayQueryContext
----------------------

#### NAME {#name}

    rtcInitRayQueryContext - initializes the ray query context

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCRayQueryContext
    {
      #if RTC_MAX_INSTANCE_LEVEL_COUNT > 1
        unsigned int instStackSize;
      #endif
      
      unsigned int instID[RTC_MAX_INSTANCE_LEVEL_COUNT];
    };

    void rtcInitRayQueryContext(
      struct RTCRayQueryContext* context
    );

#### DESCRIPTION {#description}

The `rtcInitRayQueryContext` function initializes the intersection
context to default values and should be called to initialize every ray
query context.

It is guaranteed that the pointer to the ray query context
(`RTCRayQueryContext` type) is passed to the registered callback
functions. This way it is possible to attach arbitrary data to the end
of the ray query context, such as a per-ray payload.

Inside the user geometry callback the ray query context can get used to
access the `instID` stack to know which instance the user geometry
object resides.

If not ray query context is specified when tracing a ray, a default
context is used.

#### EXIT STATUS {#exit-status}

No error code is set by this function.

#### SEE ALSO {#see-also}

[rtcIntersect1], [rtcIntersect4/8/16], [rtcOccluded1],
[rtcOccluded4/8/16]

```{=tex}

```
rtcIntersect1
-------------

#### NAME {#name}

    rtcIntersect1 - finds the closest hit for a single ray

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcIntersect1(
      RTCScene scene,
      struct RTCRayHit* rayhit
      struct RTCIntersectArguments* args = NULL
    );

#### DESCRIPTION {#description}

The `rtcIntersect1` function finds the closest hit of a single ray
(`rayhit` argument) with the scene (`scene` argument). The provided
ray/hit structure contains the ray to intersect and some hit output
fields that are filled when a hit is found. The passed optional
arguments struct (`args` argument) can get used for advanced use cases,
see section [rtcInitIntersectArguments] for more details.

To trace a ray, the user has to initialize the ray origin (`org` ray
member), ray direction (`dir` ray member), ray segment (`tnear`, `tfar`
ray members), ray mask (`mask` ray member), and set the ray flags to
`0` (`flags` ray member). The ray time (`time` ray member) must be
initialized to a value in the range \$[0, 1]. The ray segment has to
be in the range $[0, \infty]$, thus ranges that start behind the ray
origin are not valid, but ranges can reach to infinity. See Section
[RTCRay] for the ray layout description.

The geometry ID (`geomID` hit member) of the hit data must be
initialized to `RTC_INVALID_GEOMETRY_ID` (-1).

When no intersection is found, the ray/hit data is not updated. When an
intersection is found, the hit distance is written into the `tfar`
member of the ray and all hit data is set, such as unnormalized
geometry normal in object space (`Ng` hit member), local hit
coordinates (`u`, `v` hit member), instance ID stack (`instID` hit
member), geometry ID (`geomID` hit member), and primitive ID (`primID`
hit member). See Section [RTCHit] for the hit layout description.

If the instance ID stack has a prefix of values not equal to
`RTC_INVALID_GEOMETRY_ID`, the instance ID on each level corresponds to
the geometry ID of the hit instance of the higher-level scene, the
geometry ID corresponds to the hit geometry inside the hit instanced
scene, and the primitive ID corresponds to the n-th primitive of that
geometry.

If level 0 of the instance ID stack is equal to
`RTC_INVALID_GEOMETRY_ID`, the geometry ID corresponds to the hit
geometry inside the top-level scene, and the primitive ID corresponds
to the n-th primitive of that geometry.

The implementation makes no guarantees that primitives whose hit
distance is exactly at (or very close to) `tnear` or `tfar` are hit or
missed. If you want to exclude intersections at `tnear` just pass a
slightly enlarged `tnear`, and if you want to include intersections at
`tfar` pass a slightly enlarged `tfar`.

The ray pointer passed to callback functions is not guaranteed to be
identical to the original ray provided. To extend the ray with
additional data to be accessed in callback functions, use the ray query
context. See section [rtcInitRayQueryContext] for more details.

The ray/hit structure must be aligned to 16 bytes.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcOccluded1], [rtcIntersect4/8/16], [RTCRayHit],
[rtcInitIntersectArguments]

```{=tex}

```
rtcOccluded1
------------

#### NAME {#name}

    rtcOccluded1 - finds any hit for a single ray

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcOccluded1(
      RTCScene scene,
      struct RTCRay* ray,
      struct RTCOccludedArguments* args = NULL
    );

#### DESCRIPTION {#description}

The `rtcOccluded1` function checks for a single ray (`ray` argument)
whether there is any hit with the scene (`scene` argument). The passed
optional arguments struct (`args` argument) can get used for advanced
use cases, see section [rtcInitOccludedArguments] for more details.

To trace a ray, the user must initialize the ray origin (`org` ray
member), ray direction (`dir` ray member), ray segment (`tnear`, `tfar`
ray members), ray mask (`mask` ray member), and must set the ray flags
to `0` (`flags` ray member). The ray time (`time` ray member) must be
initialized to a value in the range $[0, 1]$. The ray segment must be
in the range $[0, \infty]$, thus ranges that start behind the ray
origin are not valid, but ranges can reach to infinity. See Section
[RTCRay] for the ray layout description.

When no intersection is found, the ray data is not updated. In case a
hit was found, the `tfar` component of the ray is set to `-inf`.

The implementation makes no guarantees that primitives whose hit
distance is exactly at (or very close to) `tnear` or `tfar` are hit or
missed. If you want to exclude intersections at `tnear` just pass a
slightly enlarged `tnear`, and if you want to include intersections at
`tfar` pass a slightly enlarged `tfar`.

The ray pointer passed to callback functions is not guaranteed to be
identical to the original ray provided. To extend the ray with
additional data to be accessed in callback functions, use the ray query
context. See section [rtcInitRayQueryContext] for more details.

The ray must be aligned to 16 bytes.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcIntersect1], [rtcOccluded4/8/16], [RTCRay],
[rtcInitOccludedArguments]

```{=tex}

```
rtcIntersect4/8/16
------------------

#### NAME {#name}

    rtcIntersect4/8/16 - finds the closest hits for a ray packet

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcIntersect4(
      const int* valid,
      RTCScene scene,
      struct RTCRayHit4* rayhit,
      struct RTCIntersectArguments* args = NULL
    );

    void rtcIntersect8(
      const int* valid,
      RTCScene scene,
      struct RTCRayHit8* rayhit,
      struct RTCIntersectArguments* args = NULL
    );

    void rtcIntersect16(
      const int* valid,
      RTCScene scene,
      struct RTCRayHit16* rayhit,
      struct RTCIntersectArguments* args = NULL
    );

#### DESCRIPTION {#description}

The `rtcIntersect4/8/16` functions finds the closest hits for a ray
packet of size 4, 8, or 16 (`rayhit` argument) with the scene (`scene`
argument). The ray/hit input contains a ray packet and hit packet. The
passed optional arguments struct (`args` argument) are used to pass
additional arguments for advanced features. See Section
[rtcIntersect1] for more details and a description of how to set up
and trace rays.

A ray valid mask must be provided (`valid` argument) which stores one
32-bit integer (`-1` means valid and `0` invalid) per ray in the
packet. Only active rays are processed, and hit data of inactive rays
is not changed.

The ray pointer passed to callback functions is not guaranteed to be
identical to the original ray provided. To extend the ray with
additional data to be accessed in callback functions, use the ray query
context. See section [rtcInitRayQueryContext] for more details.

For `rtcIntersect4` the ray packet must be aligned to 16 bytes, for
`rtcIntersect8` the alignment must be 32 bytes, and for
`rtcIntersect16` the alignment must be 64 bytes.

The `rtcIntersect4`, `rtcIntersect8` and `rtcIntersect16` functions may
change the ray packet size and ray order when calling back into filter
functions or user geometry callbacks. Under some conditions the
application can assume packets to stay intakt, which can determined by
querying the `RTC_DEVICE_PROPERTY_NATIVE_RAY4_SUPPORTED`,
`RTC_DEVICE_PROPERTY_NATIVE_RAY8_SUPPORTED`,
`RTC_DEVICE_PROPERTY_NATIVE_RAY16_SUPPORTED` properties through the
`rtcGetDeviceProperty` function. See [rtcGetDeviceProperty] for more
information.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcIntersect1], [rtcOccluded4/8/16], [rtcInitIntersectArguments]

```{=tex}

```
rtcOccluded4/8/16
-----------------

#### NAME {#name}

    rtcOccluded4/8/16 - finds any hits for a ray packet

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcOccluded4(
      const int* valid,
      RTCScene scene,
      struct RTCRay4* ray,
      struct RTCOccludedArguments* args = NULL
    );

    void rtcOccluded8(
      const int* valid,
      RTCScene scene,
      struct RTCRay8* ray,
      struct RTCOccludedArguments* args = NULL
    );

    void rtcOccluded16(
      const int* valid,
      RTCScene scene,
      struct RTCRay16* ray,
      struct RTCOccludedArguments* args = NULL
    );

#### DESCRIPTION {#description}

The `rtcOccluded4/8/16` functions checks for each active ray of the ray
packet of size 4, 8, or 16 (`ray` argument) whether there is any hit
with the scene (`scene` argument). The passed optional arguments struct
(`args` argument) can get used for advanced use cases, see section
[rtcInitOccludedArguments] for more details. See Section
[rtcOccluded1] for more details and a description of how to set up
and trace occlusion rays.

A ray valid mask must be provided (`valid` argument) which stores one
32-bit integer (`-1` means valid and `0` invalid) per ray in the
packet. Only active rays are processed, and hit data of inactive rays
is not changed.

The ray pointer passed to callback functions is not guaranteed to be
identical to the original ray provided. To extend the ray with
additional data to be accessed in callback functions, use the ray query
context. See section [rtcInitRayQueryContext] for more details.

For `rtcOccluded4` the ray packet must be aligned to 16 bytes, for
`rtcOccluded8` the alignment must be 32 bytes, and for `rtcOccluded16`
the alignment must be 64 bytes.

The `rtcOccluded4`, `rtcOccluded8` and `rtcOccluded16` functions may
change the ray packet size and ray order when calling back into
intersect filter functions or user geometry callbacks. Under some
conditions the application can assume packets to stay intakt, which can
determined by querying the `RTC_DEVICE_PROPERTY_NATIVE_RAY4_SUPPORTED`,
`RTC_DEVICE_PROPERTY_NATIVE_RAY8_SUPPORTED`,
`RTC_DEVICE_PROPERTY_NATIVE_RAY16_SUPPORTED` properties through the
`rtcGetDeviceProperty` function. See [rtcGetDeviceProperty] for more
information.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcOccluded1], [rtcIntersect4/8/16], [rtcInitOccludedArguments]

```{=tex}

```
rtcForwardIntersect1
--------------------

#### NAME {#name}

    rtcForwardIntersect1/Ex - forwards a single ray to new scene
      from user geometry callback

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcForwardIntersect1(
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay* ray,
      unsigned int instID
    );

    void rtcForwardIntersect1Ex(
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay* ray,
      unsigned int instID,
      unsigned int instPrimID,
    );

#### DESCRIPTION {#description}

The `rtcForwardIntersect1` and `rtcForwardIntersect1Ex` functions
forward the traversal of a transformed ray (`ray` argument) into a
scene (`scene` argument) from a user geometry callback. The function
can only get invoked from a user geometry callback for a ray traversal
initiated with the `rtcIntersect1` function. The callback arguments
structure of the callback invokation has to get passed to the ray
forwarding (`args` argument). The user geometry callback should
instantly terminate after invoking the `rtcForwardIntersect1/Ex`
function.

Only the ray origin and ray direction members of the ray argument are
used for forwarding, all additional ray properties are inherited from
the initial ray traversal invokation of `rtcIntersect1`.

The implementation of the `rtcForwardIntersect1` function recursively
continues the ray traversal into the specified scene and pushes the
provided instance ID (`instID` argument) to the instance ID stack. Hit
information is updated into the ray hit structure passed to the
original `rtcIntersect1` invokation.

This function can get used to implement user defined instancing using
user geometries, e.g. by transforming the ray in a special way, and/or
selecting between different scenes to instantiate.

For user defined instance arrays, the `rtcForwardIntersect1Ex` variant
has an additional `instPrimID` argument which is pushed to the instance
primitive ID stack. Instance primitive IDs identify which instance of
an instance array was hit.

When using Embree on the CPU it is possible to recursively invoke
`rtcIntersect1` directly from a user geometry callback. However, when
SYCL is used, recursively tracing rays is not directly supported, and
the `rtcForwardIntersect1/Ex` functions must be used.

The ray structure must be aligned to 16 bytes.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcIntersect1], [RTCRay]

```{=tex}

```
rtcForwardOccluded1
-------------------

#### NAME {#name}

    rtcForwardOccluded1/Ex - forwards a single ray to new scene
      from user geometry callback

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcForwardOccluded1(
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay* ray,
      unsigned int instID
    );

    void rtcForwardOccluded1(
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

#### DESCRIPTION {#description}

The `rtcForwardOccluded1` and `rtcForwardOccluded1Ex` functions forward
the traversal of a transformed ray (`ray` argument) into a scene
(`scene` argument) from a user geometry callback. The function can only
get invoked from a user geometry callback for a ray traversal initiated
with the `rtcOccluded1` function. The callback arguments structure of
the callback invokation has to get passed to the ray forwarding (`args`
argument). The user geometry callback should instantly terminate after
invoking the `rtcForwardOccluded1/Ex` function.

Only the ray origin and ray direction members of the ray argument are
used for forwarding, all additional ray properties are inherited from
the initial ray traversal invokation of `rtcOccluded1`.

The implementation of the `rtcForwardOccluded1` function recursively
continues the ray traversal into the specified scene and pushes the
provided instance ID (`instID` argument) to the instance ID stack. Hit
information is updated into the ray structure passed to the original
`rtcOccluded1` invokation.

This function can get used to implement user defined instancing using
user geometries, e.g. by transforming the ray in a special way, and/or
selecting between different scenes to instantiate.

For user defined instance arrays, the `rtcForwardIntersect1Ex` variant
has an additional `instPrimID` argument which is pushed to the instance
primitive ID stack. Instance primitive IDs identify which instance of
an instance array was hit.

When using Embree on the CPU it is possible to recursively invoke
`rtcOccluded1` directly from a user geometry callback. However, when
SYCL is used, recursively tracing rays is not directly supported, and
the `rtcForwardOccluded1/Ex` function must be used.

The ray structure must be aligned to 16 bytes.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcOccluded1], [RTCRay]

```{=tex}

```
rtcForwardIntersect4/8/16
-------------------------

#### NAME {#name}

    rtcForwardIntersect4/8/16/Ex - forwards a ray packet to new scene
      from user geometry callback

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcForwardIntersect4(
      void int* valid,
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay4* ray,
      unsigned int instID
    );

    void rtcForwardIntersect8(
      void int* valid,
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay8* ray,
      unsigned int instID
    );

    void rtcForwardIntersect16(
      void int* valid,
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay16* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

    void rtcForwardIntersect4Ex(
      void int* valid,
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay4* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

    void rtcForwardIntersect8Ex(
      void int* valid,
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay8* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

    void rtcForwardIntersect16Ex(
      void int* valid,
      const struct RTCIntersectFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay16* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

#### DESCRIPTION {#description}

The `rtcForwardIntersect4/8/16` and `rtcForwardIntersect4/8/16Ex`
functions forward the traversal of a transformed ray packet (`ray`
argument) into a scene (`scene` argument) from a user geometry
callback. The function can only get invoked from a user geometry
callback for a ray traversal initiated with the `rtcIntersect4/8/16`
function. The callback arguments structure of the callback invokation
has to get passed to the ray forwarding (`args` argument). The user
geometry callback should instantly terminate after invoking the
`rtcForwardIntersect4/8/16/Ex` function.

Only the ray origin and ray direction members of the ray argument are
used for forwarding, all additional ray properties are inherited from
the initial ray traversal invokation of `rtcIntersect4/8/16`.

The implementation of the `rtcForwardIntersect4/8/16` function
recursively continues the ray traversal into the specified scene and
pushes the provided instance ID (`instID` argument) to the instance ID
stack. Hit information is updated into the ray hit structure passed to
the original `rtcIntersect4/8/16` invokation.

This function can get used to implement user defined instancing using
user geometries, e.g. by transforming the ray in a special way, and/or
selecting between different scenes to instantiate.

For user defined instance arrays, the `rtcForwardIntersect4/8/16Ex`
variant has an additional `instPrimID` argument which is pushed to the
instance primitive ID stack. Instance primitive IDs identify which
instance of an instance array was hit.

When using Embree on the CPU it is possible to recursively invoke
`rtcIntersect4/8/16` directly from a user geometry callback. However,
when SYCL is used, recursively tracing rays is not directly supported,
and the `rtcForwardIntersect4/8/16` function must be used.

For `rtcForwardIntersect4` the ray packet must be aligned to 16 bytes,
for `rtcForwardIntersect8` the alignment must be 32 bytes, and for
`rtcForwardIntersect16` the alignment must be 64 bytes.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcIntersect4/8/16]

```{=tex}

```
rtcForwardOccluded4/8/16
------------------------

#### NAME {#name}

    rtcForwardOccluded4/8/16/Ex - forwards a ray packet to new scene
      from user geometry callback

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcForwardOccluded4(
      void int* valid,
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay4* ray,
      unsigned int instID
    );

    void rtcForwardOccluded8(
      void int* valid,
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay8* ray,
      unsigned int instID
    );

    void rtcForwardOccluded16(
      void int* valid,
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay16* ray,
      unsigned int instID
    );

    void rtcForwardOccluded4Ex(
      void int* valid,
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay4* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

    void rtcForwardOccluded8Ex(
      void int* valid,
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay8* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

    void rtcForwardOccluded16Ex(
      void int* valid,
      const struct RTCOccludedFunctionNArguments* args,
      RTCScene scene,
      struct RTCRay16* ray,
      unsigned int instID,
      unsigned int instPrimID
    );

#### DESCRIPTION {#description}

The `rtcForwardOccluded4/8/16` and `rtcForwardOccluded4/8/16Ex`
functions forward the traversal of a transformed ray packet (`ray`
argument) into a scene (`scene` argument) from a user geometry
callback. The function can only get invoked from a user geometry
callback for a ray traversal initiated with the `rtcOccluded4/8/16`
function. The callback arguments structure of the callback invokation
has to get passed to the ray forwarding (`args` argument). The user
geometry callback should instantly terminate after invoking the
`rtcForwardOccluded4/8/16/Ex` function.

Only the ray origin and ray direction members of the ray argument are
used for forwarding, all additional ray properties are inherited from
the initial ray traversal invokation of `rtcOccluded4/8/16`.

The implementation of the `rtcForwardOccluded4/8/16` function
recursively continues the ray traversal into the specified scene and
pushes the provided instance ID (`instID` argument) to the instance ID
stack. Hit information is updated into the ray structure passed to the
original `rtcOccluded4/8/16` invokation.

This function can get used to implement user defined instancing using
user geometries, e.g. by transforming the ray in a special way, and/or
selecting between different scenes to instantiate.

For user defined instance arrays, the `rtcForwardIntersect4/8/16Ex`
variant has an additional `instPrimID` argument which is pushed to the
instance primitive ID stack. Instance primitive IDs identify which
instance of an instance array was hit.

When using Embree on the CPU it is possible to recursively invoke
`rtcOccluded4/8/16` directly from a user geometry callback. However,
when SYCL is used, recursively tracing rays is not directly supported,
and the `rtcForwardOccluded4/8/16` function must be used.

For `rtcForwardOccluded4` the ray packet must be aligned to 16 bytes,
for `rtcForwardOccluded8` the alignment must be 32 bytes, and for
`rtcForwardOccluded16` the alignment must be 64 bytes.

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcOccluded4/8/16]

```{=tex}

```
rtcInitPointQueryContext
------------------------

#### NAME {#name}

    rtcInitPointQueryContext - initializes the context information (e.g.
      stack of (multilevel-)instance transformations) for point queries

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTC_ALIGN(16) RTCPointQueryContext
    {
      // accumulated 4x4 column major matrices from world to instance space.
      float world2inst[RTC_MAX_INSTANCE_LEVEL_COUNT][16];
      
      // accumulated 4x4 column major matrices from instance to world space.
      float inst2world[RTC_MAX_INSTANCE_LEVEL_COUNT][16];

      // instance ids.
      unsigned int instID[RTC_MAX_INSTANCE_LEVEL_COUNT];
      
      // number of instances currently on the stack.
      unsigned int instStackSize;
    };

    void rtcInitPointQueryContext(
      struct RTCPointQueryContext* context
    );

#### DESCRIPTION {#description}

A stack (`RTCPointQueryContext` type) which stores the IDs and instance
transformations during a BVH traversal for a point query. The
transformations are assumed to be affine transformations (3×3 matrix
plus translation) and therefore the last column is ignored (see
[RTC\_GEOMETRY\_TYPE\_INSTANCE] for details).

The `rtcInitPointContext` function initializes the context to default
values and should be called for initialization.

The context will be passed as an argument to the point query callback
function (see [rtcSetGeometryPointQueryFunction]) and should be used
to pass instance information down the instancing chain for user defined
instancing (see tutorial [ClosestPoint] for a reference
implementation of point queries with user defined instancing).

The context is an necessary argument to [rtcPointQuery] and Embree
internally uses the topmost instance transformation of the stack to
transform the point query into instance space.

#### EXIT STATUS {#exit-status}

No error code is set by this function.

#### SEE ALSO {#see-also}

[rtcPointQuery], [rtcSetGeometryPointQueryFunction]

```{=tex}

```
rtcPointQuery
-------------

#### NAME {#name}

    rtcPointQuery - traverses the BVH with a point query object

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTC_ALIGN(16) RTCPointQuery
    {
      // location of the query
      float x;
      float y;
      float z;

      // radius and time of the query
      float radius;
      float time;
    };

    void rtcPointQuery(
      RTCScene scene,
      struct RTCPointQuery* query,
      struct RTCPointQueryContext* context,
      struct RTCPointQueryFunction* queryFunc,
      void* userPtr
    );

#### DESCRIPTION {#description}

The `rtcPointQuery` function traverses the BVH using a `RTCPointQuery`
object (`query` argument) and calls a user defined callback function
(e.g `queryFunc` argument) for each primitive of the scene (`scene`
argument) that intersects the query domain.

The user has to initialize the query location (`x`, `y` and `z` member)
and query radius in the range $[0, \infty]$. If the scene contains
motion blur geometries, also the query time (`time` member) must be
initialized to a value in the range $[0, 1]$.

Further, a `RTCPointQueryContext` (`context` argument) must be created
and initialized. It contains ID and transformation information of the
instancing hierarchy if (multilevel-)instancing is used. See
[rtcInitPointQueryContext] for further information.

For every primitive that intersects the query domain, the callback
function (`queryFunc` argument) is called, in which distance
computations to the primitive can be implemented. The user will be
provided with the primID and geomID of the according primitive,
however, the geometry information (e.g. triangle index and vertex data)
has to be determined manually. The `userPtr` argument can be used to
input geometry data of the scene or output results of the point query
(e.g. closest point currently found on surface geometry (see tutorial
[ClosestPoint])).

The parameter `queryFunc` is optional and can be NULL, in which case
the callback function is not invoked. However, a callback function can
still get attached to a specific `RTCGeometry` object using
[rtcSetGeometryPointQueryFunction]. If a callback function is
attached to a geometry and (a potentially different) callback function
is passed as an argument to `rtcPointQuery`, both functions are called
for the primitives of the according geometries.

The query radius can be decreased inside the callback function, which
allows to efficiently cull parts of the scene during BVH traversal.
Increasing the query radius and modifying time or location of the query
will result in undefined behaviour.

The callback function will be called for all primitives in a leaf node
of the BVH even if the primitive is outside the query domain, since
Embree does not gather geometry information of primitives internally.

Point queries can be used with (multilevel)-instancing. However, care
has to be taken when the instance transformation contains anisotropic
scaling or sheering. In these cases distance computations have to be
performed in world space to ensure correctness and the ellipsoidal
query domain (in instance space) will be approximated with its axis
aligned bounding box internally. Therefore, the callback function might
be invoked even for primitives in inner BVH nodes that do not intersect
the query domain. See [rtcSetGeometryPointQueryFunction] for details.

The point query structure must be aligned to 16 bytes.

#### SUPPORTED PRIMITIVES

Currently, all primitive types are supported by the point query API
except of points (see [RTC\_GEOMETRY\_TYPE\_POINT]), curves (see
[RTC\_GEOMETRY\_TYPE\_CURVE]) and sudivision surfaces (see
[RTC\_GEOMETRY\_SUBDIVISION]).

#### EXIT STATUS {#exit-status}

For performance reasons this function does not do any error checks,
thus will not set any error flags on failure.

#### SEE ALSO {#see-also}

[rtcSetGeometryPointQueryFunction], [rtcInitPointQueryContext]

```{=tex}

```
rtcCollide
----------

#### NAME {#name}

    rtcCollide - intersects one BVH with another

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTCCollision {
      unsigned int geomID0, primID0;
      unsigned int geomID1, primID1;
    };

    typedef void (*RTCCollideFunc) (
      void* userPtr,
      RTCCollision* collisions,
      size_t num_collisions);

    void rtcCollide (
        RTCScene hscene0, 
        RTCScene hscene1, 
        RTCCollideFunc callback, 
        void* userPtr
    );

#### DESCRIPTION {#description}

The `rtcCollide` function intersects the BVH of `hscene0` with the BVH
of scene `hscene1` and calls a user defined callback function (e.g
`callback` argument) for each pair of intersecting primitives between
the two scenes. A user defined data pointer (`userPtr` argument) can
also be passed in.

For every pair of primitives that may intersect each other, the
callback function (`callback` argument) is called. The user will be
provided with the primID's and geomID's of multiple potentially
intersecting primitive pairs. Currently, only scene entirely composed
of user geometries are supported, thus the user is expected to
implement a primitive/primitive intersection to filter out false
positives in the callback function. The `userPtr` argument can be used
to input geometry data of the scene or output results of the
intersection query.

#### SUPPORTED PRIMITIVES {#supported-primitives}

Currently, the only supported type is the user geometry type (see
[RTC\_GEOMETRY\_TYPE\_USER]).

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

```{=tex}

```
rtcNewBVH
---------

#### NAME {#name}

    rtcNewBVH - creates a new BVH object

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    RTCBVH rtcNewBVH(RTCDevice device);

#### DESCRIPTION {#description}

This function creates a new BVH object and returns a handle to this
BVH. The BVH object is reference counted with an initial reference
count of 1. The handle can be released using the `rtcReleaseBVH` API
call.

The BVH object can be used to build a BVH in a user-specified format
over user-specified primitives. See the documentation of the
`rtcBuildBVH` call for more details.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcRetainBVH], [rtcReleaseBVH], [rtcBuildBVH]

```{=tex}

```
rtcRetainBVH
------------

#### NAME {#name}

    rtcRetainBVH - increments the BVH reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcRetainBVH(RTCBVH bvh);

#### DESCRIPTION {#description}

BVH objects are reference counted. The `rtcRetainBVH` function
increments the reference count of the passed BVH object (`bvh`
argument). This function together with `rtcReleaseBVH` allows to use
the internal reference counting in a C++ wrapper class to handle the
ownership of the object.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewBVH], [rtcReleaseBVH]

```{=tex}

```
rtcReleaseBVH
-------------

#### NAME {#name}

    rtcReleaseBVH - decrements the BVH reference count

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    void rtcReleaseBVH(RTCBVH bvh);

#### DESCRIPTION {#description}

BVH objects are reference counted. The `rtcReleaseBVH` function
decrements the reference count of the passed BVH object (`bvh`
argument). When the reference count falls to 0, the BVH gets destroyed.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewBVH], [rtcRetainBVH]

```{=tex}

```
rtcBuildBVH
-----------

#### NAME {#name}

    rtcBuildBVH - builds a BVH

#### SYNOPSIS {#synopsis}

    #include <embree4/rtcore.h>

    struct RTC_ALIGN(32) RTCBuildPrimitive
    {
      float lower_x, lower_y, lower_z; 
      unsigned int geomID;
      float upper_x, upper_y, upper_z;
      unsigned int primID;
    };

    typedef void* (*RTCCreateNodeFunction) (
      RTCThreadLocalAllocator allocator,
      unsigned int childCount,
      void* userPtr
    );

    typedef void (*RTCSetNodeChildrenFunction) (
      void* nodePtr,
      void** children,
      unsigned int childCount,
      void* userPtr
    );

    typedef void (*RTCSetNodeBoundsFunction) (
      void* nodePtr,
      const struct RTCBounds** bounds,
      unsigned int childCount,
      void* userPtr
    );

    typedef void* (*RTCCreateLeafFunction) (
      RTCThreadLocalAllocator allocator,
      const struct RTCBuildPrimitive* primitives,
      size_t primitiveCount,
      void* userPtr
    );

    typedef void (*RTCSplitPrimitiveFunction) (
      const struct RTCBuildPrimitive* primitive,
      unsigned int dimension,
      float position,
      struct RTCBounds* leftBounds,
      struct RTCBounds* rightBounds,
      void* userPtr
    );

    typedef bool (*RTCProgressMonitorFunction)(
      void* userPtr, double n
    );

    enum RTCBuildFlags
    {
      RTC_BUILD_FLAG_NONE,
      RTC_BUILD_FLAG_DYNAMIC
    };

    struct RTCBuildArguments
    {
      size_t byteSize;

      enum RTCBuildQuality buildQuality;
      enum RTCBuildFlags buildFlags;
      unsigned int maxBranchingFactor;
      unsigned int maxDepth;
      unsigned int sahBlockSize;
      unsigned int minLeafSize;
      unsigned int maxLeafSize;
      float traversalCost;
      float intersectionCost;

      RTCBVH bvh;
      struct RTCBuildPrimitive* primitives;
      size_t primitiveCount;
      size_t primitiveArrayCapacity;
      
      RTCCreateNodeFunction createNode;
      RTCSetNodeChildrenFunction setNodeChildren;
      RTCSetNodeBoundsFunction setNodeBounds;
      RTCCreateLeafFunction createLeaf;
      RTCSplitPrimitiveFunction splitPrimitive;
      RTCProgressMonitorFunction buildProgress;
      void* userPtr;
    };

    struct RTCBuildArguments rtcDefaultBuildArguments();

    void* rtcBuildBVH(
      const struct RTCBuildArguments* args
    );

#### DESCRIPTION {#description}

The `rtcBuildBVH` function can be used to build a BVH in a user-defined
format over arbitrary primitives. All arguments to the function are
provided through the `RTCBuildArguments` structure. The first member of
that structure must be set to the size of the structure in bytes
(`bytesSize` member) which allows future extensions of the structure.
It is recommended to initialize the build arguments structure using the
`rtcDefaultBuildArguments` function.

The `rtcBuildBVH` function gets passed the BVH to build (`bvh` member),
the array of primitives (`primitives` member), the capacity of that
array (`primitiveArrayCapacity` member), the number of primitives
stored inside the array (`primitiveCount` member), callback function
pointers, and a user-defined pointer (`userPtr` member) that is passed
to all callback functions when invoked. The `primitives` array can be
freed by the application after the BVH is built. All callback functions
are typically called from multiple threads, thus their implementation
must be thread-safe.

Four callback functions must be registered, which are invoked during
build to create BVH nodes (`createNode` member), to set the pointers to
all children (`setNodeChildren` member), to set the bounding boxes of
all children (`setNodeBounds` member), and to create a leaf node
(`createLeaf` member).

The function pointer to the primitive split function (`splitPrimitive`
member) may be `NULL`, however, then no spatial splitting in high
quality mode is possible. The function pointer used to report the build
progress (`buildProgress` member) is optional and may also be `NULL`.

Further, some build settings are passed to configure the BVH build.
Using the build quality settings (`buildQuality` member), one can
select between a faster, low quality build which is good for dynamic
scenes, and a standard quality build for static scenes. One can also
specify the desired maximum branching factor of the BVH
(`maxBranchingFactor` member), the maximum depth the BVH should have
(`maxDepth` member), the block size for the SAH heuristic
(`sahBlockSize` member), the minimum and maximum leaf size
(`minLeafSize` and `maxLeafSize` member), and the estimated costs of
one traversal step and one primitive intersection (`traversalCost` and
`intersectionCost` members). When enabling the `RTC_BUILD_FLAG_DYNAMIC`
build flags (`buildFlags` member), re-build performance for dynamic
scenes is improved at the cost of higher memory requirements.

To spatially split primitives in high quality mode, the builder needs
extra space at the end of the build primitive array to store split
primitives. The total capacity of the build primitive array is passed
using the `primitiveArrayCapacity` member, and should be about twice
the number of primitives when using spatial splits.

The `RTCCreateNodeFunc` and `RTCCreateLeafFunc` callbacks are passed a
thread local allocator object that should be used for fast allocation
of nodes using the `rtcThreadLocalAlloc` function. We strongly
recommend using this allocation mechanism, as alternative approaches
like standard `malloc` can be over 10× slower. The allocator object
passed to the create callbacks may be used only inside the current
thread. Memory allocated using `rtcThreadLocalAlloc` is automatically
freed when the `RTCBVH` object is deleted. If you use your own memory
allocation scheme you have to free the memory yourself when the
`RTCBVH` object is no longer used.

The `RTCCreateNodeFunc` callback additionally gets the number of
children for this node in the range from 2 to `maxBranchingFactor`
(`childCount` argument).

The `RTCSetNodeChildFunc` callback function gets a pointer to the node
as input (`nodePtr` argument), an array of pointers to the children
(`childPtrs` argument), and the size of this array (`childCount`
argument).

The `RTCSetNodeBoundsFunc` callback function gets a pointer to the node
as input (`nodePtr` argument), an array of pointers to the bounding
boxes of the children (`bounds` argument), and the size of this array
(`childCount` argument).

The `RTCCreateLeafFunc` callback additionally gets an array of
primitives as input (`primitives` argument), and the size of this array
(`primitiveCount` argument). The callback should read the `geomID` and
`primID` members from the passed primitives to construct the leaf.

The `RTCSplitPrimitiveFunc` callback is invoked in high quality mode to
split a primitive (`primitive` argument) at the specified position
(`position` argument) and dimension (`dimension` argument). The
callback should return bounds of the clipped left and right parts of
the primitive (`leftBounds` and `rightBounds` arguments).

The `RTCProgressMonitorFunction` callback function is called with the
estimated completion rate `n` in the range $[0,1]$. Returning `true`
from the callback lets the build continue; returning `false` cancels
the build.

#### EXIT STATUS {#exit-status}

On failure an error code is set that can be queried using
`rtcGetDeviceError`.

#### SEE ALSO {#see-also}

[rtcNewBVH]

```{=tex}

```
RTCQuaternionDecomposition
--------------------------

#### NAME {#name}

    RTCQuaternionDecomposition - structure that represents a quaternion
      decomposition of an affine transformation

#### SYNOPSIS {#synopsis}

    struct RTCQuaternionDecomposition
    {
      float scale_x, scale_y, scale_z;
      float skew_xy, skew_xz, skew_yz;
      float shift_x, shift_y, shift_z;
      float quaternion_r, quaternion_i, quaternion_j, quaternion_k;
      float translation_x, translation_y, translation_z;
    };

#### DESCRIPTION {#description}

The struct `RTCQuaternionDecomposition` represents an affine
transformation decomposed into three parts. An upper triangular
scaling/skew/shift matrix

$$
S = \left( \begin{array}{cccc}
scale_x & skew_{xy} & skew_{xz} & shift_x \ 
0 & scale_y & skew_{yz} & shift_y \ 
0 & 0 & scale_z & shift_z \ 
0 & 0 & 0 & 1 \ 
\end{array} \right),
$$

a translation matrix

$$
T = \left( \begin{array}{cccc}
1 & 0 & 0 & translation_x \ 
0 & 1 & 0 & translation_y \ 
0 & 0 & 1 & translation_z \ 
0 & 0 & 0 & 1 \ 
\end{array} \right),
$$

and a rotation matrix $R$, represented as a quaternion

$quaternion_r + quaternion_i  \mathbf{i} + quaternion_j  \mathbf{i} + quaternion_k  \mathbf{k}$

where $\mathbf{i}$, $\mathbf{j}$ $\mathbf{k}$ are the imaginary
quaternion units. The passed quaternion will be normalized internally.

The affine transformation matrix corresponding to a
`RTCQuaternionDecomposition` is $TRS$ and a point
$p = (p_x, p_y, p_z, 1)^T$ will be transformed as
$$p' = T  R  S  p.$$

The functions `rtcInitQuaternionDecomposition`,
`rtcQuaternionDecompositionSetQuaternion`,
`rtcQuaternionDecompositionSetScale`,
`rtcQuaternionDecompositionSetSkew`,
`rtcQuaternionDecompositionSetShift`, and
`rtcQuaternionDecompositionSetTranslation` allow to set the fields of
the structure more conveniently.

#### EXIT STATUS {#exit-status}

No error code is set by this function.

#### SEE ALSO {#see-also}

[rtcSetGeometryTransformQuaternion],
[rtcInitQuaternionDecomposition]

```{=tex}

```
rtcInitQuaternionDecomposition
------------------------------

#### NAME {#name}

    rtcInitQuaternionDecomposition - initializes quaternion decomposition

#### SYNOPSIS {#synopsis}

    void rtcInitQuaternionDecomposition(
      struct RTCQuaternionDecomposition* qd
    );

#### DESCRIPTION {#description}

The `rtcInitQuaternionDecomposition` function initializes a
`RTCQuaternionDecomposition` structure to represent an identity
transformation.

#### EXIT STATUS {#exit-status}

No error code is set by this function.

#### SEE ALSO {#see-also}

[rtcSetGeometryTransformQuaternion], [RTCQuaternionDecomposition]

```{=tex}

```
CPU Performance Recommendations
===============================

MXCSR control and status register
---------------------------------

It is strongly recommended to have the `Flush to Zero` and
`Denormals are Zero` mode of the MXCSR control and status register
enabled for each thread before calling the `rtcIntersect`-type and
`rtcOccluded`-type functions. Otherwise, under some circumstances
special handling of denormalized floating point numbers can
significantly reduce application and Embree performance. When using
Embree together with the Intel® Threading Building Blocks, it is
sufficient to execute the following code at the beginning of the
application main thread (before the creation of the
`tbb::task_scheduler_init` object):

    #include <xmmintrin.h>
    #include <pmmintrin.h>
    ...
    _MM_SET_FLUSH_ZERO_MODE(_MM_FLUSH_ZERO_ON);
    _MM_SET_DENORMALS_ZERO_MODE(_MM_DENORMALS_ZERO_ON);

If using a different tasking system, make sure each rendering thread
has the proper mode set.

Thread Creation and Affinity Settings
-------------------------------------

Tasking systems like TBB create worker threads on demand, which will
add a runtime overhead for the very first `rtcCommitScene` call. In
case you want to benchmark the scene build time, you should start the
threads at application startup. You can let Embree start TBB threads by
passing `start_threads=1` to the `cfg` parameter of `rtcNewDevice`.

On machines with a high thread count (e.g. dual-socket Xeon or Xeon Phi
machines), affinitizing TBB worker threads increases build and
rendering performance. You can let Embree affinitize TBB worker threads
by passing `set_affinity=1` to the `cfg` parameter of `rtcNewDevice`.
By default, threads are not affinitized by Embree with the exception of
Xeon Phi Processors where they are affinitized by default.

All Embree tutorials automatically start and affinitize TBB worker
threads by passing `start_threads=1,set_affinity=1` to `rtcNewDevice`.

Fast Coherent Rays
------------------

For getting the highest performance for highly coherent rays, e.g.
primary or hard shadow rays, it is recommended to use packets with
setting the `RTC_RAY_QUERY_FLAG_COHERENT` flag in the
`RTCIntersectArguments` struct passed to the
`rtcIntersect`/`rtcOccluded` calls. The rays inside each packet should
be grouped as coherent as possible.

Huge Page Support
-----------------

It is recommended to use huge pages under Linux to increase rendering
performance. Embree supports 2MB huge pages under Windows, Linux, and
macOS. Under Linux huge page support is enabled by default, and under
Windows and macOS disabled by default. Huge page support can be enabled
in Embree by passing `hugepages=1` to `rtcNewDevice` or disabled by
passing `hugepages=0` to `rtcNewDevice`.

We recommend using 2MB huge pages with Embree under Linux as this
improves ray tracing performance by about 5-10%. Under Windows using
huge pages requires the application to run in elevated mode which is a
security issue, thus likely not an option for most use cases. Under
macOS huge pages are rarely available as memory tends to get quickly
fragmented, thus we do not recommend using huge pages on macOS.

### Huge Pages under Linux

Linux supports transparent huge pages and explicit huge pages. To
enable transparent huge page support under Linux, execute the following
as root:

    echo always > /sys/kernel/mm/transparent_hugepage/enabled

When transparent huge pages are enabled, the kernel tries to merge 4KB
pages to 2MB pages when possible as a background job. Many Linux
distributions have transparent huge pages enabled by default. See the
following webpage for more information on [transparent huge pages under
Linux](https://www.kernel.org/doc/Documentation/vm/transhuge.txt). In
this mode each application, including your rendering application based
on Embree, will automatically tend to use huge pages.

Using transparent huge pages, the transitioning from 4KB to 2MB pages
might take some time. For that reason Embree also supports allocating
2MB pages directly when a huge page pool is configured. Such a pool can
be configured by writing some number of huge pages to allocate to
`/proc/sys/vm/nr_overcommit_hugepages` as root user. E.g. to configure
2GB of address space for huge page allocation, execute the following as
root:

    echo 1000 > /proc/sys/vm/nr_overcommit_hugepages

See the following webpage for more information on [huge pages under
Linux](https://www.kernel.org/doc/Documentation/vm/hugetlbpage.txt).

### Huge Pages under Windows

To use huge pages under Windows, the current user must have the "Lock
pages in memory" (SeLockMemoryPrivilege) assigned. This can be
configured through the "Local Security Policy" application, by adding a
user to "Local Policies" -\> "User Rights Assignment" -\> "Lock pages
in memory". You have to log out and in again for this change to take
effect.

Further, your application must be executed as an elevated process ("Run
as administrator") and the "SeLockMemoryPrivilege" must be explicitly
enabled by your application. Example code on how to enable this
privilege can be found in the "common/sys/alloc.cpp" file of Embree.
Alternatively, Embree will try to enable this privilege when passing
`enable_selockmemoryprivilege=1` to `rtcNewDevice`. Further, huge pages
should be enabled in Embree by passing `hugepages=1` to `rtcNewDevice`.

When the system has been running for a while, physical memory gets
fragmented, which can slow down the allocation of huge pages
significantly under Windows.

### Huge Pages under macOS

To use huge pages under macOS you have to pass `hugepages=1` to
`rtcNewDevice` to enable that feature in Embree.

When the system has been running for a while, physical memory gets
quickly fragmented, and causes huge page allocations to fail. For this
reason, huge pages are not very useful under macOS in practice.

Avoid store-to-load forwarding issues with single rays
------------------------------------------------------

We recommend to use a single SSE store to set up the `org` and `tnear`
components, and a single SSE store to set up the `dir` and `time`
components of a single ray (`RTCRay` type). Storing these values using
scalar stores causes a store-to-load forwarding penalty because Embree
is reading these components using SSE loads later on.

```{=tex}

```
GPU Performance Recommendations
===============================

Low Code Complexity
-------------------

As a general rule try to keep code complexity low, to avoid spill code
generation. To achieve this we recommend splitting your renderer into
separate kernels instead of using a single Uber kernel invokation.

Code can further get reduced by using SYCL specialization constants to
just enable rendering features required to render a given scene.

Feature Flags
-------------

Use SYCL specialization constants and the feature flags (see section
[RTCFeatureFlags]) of the `rtcIntersect1` and `rtcOccluded1` calls to
JIT compile minimal code. The passed feature flags should just contain
features required to render the current scene. If JIT compile times are
an issue, reduce the number of feature masks used and use JIT caching
(see section [SYCL JIT caching](#sycl-jit-caching)).

Inline Indirect Calls
---------------------

Attaching user geometry and intersection filter callbacks to the
geometries of the scene is not supported in SYCL for performance
reasons.

Instead directly pass the user geometry and intersection filter
callback functions through the `RTCIntersectArguments` (and
`RTCOccludedArguments`) struct to `rtcIntersect1` (and `rtcOccluded1`)
API functions as in the following example:

    RTC_SYCL_INDIRECTLY_CALLABLE void intersectionFilter(
      const RTCFilterFunctionNArguments* args
    ) { ... }

    RTCIntersectArguments args;
    rtcInitIntersectArguments(&args);
    args.filter = intersectionFilter;

    rtcIntersect1(scene,&ray,&args);

If the callback function is directly passed that way, the SYCL compiler
can inline the indirect call, which gives a huge performance benefit.
Do *not* read a function pointer form some memory location and pass it
to `rtcIntersect1` (and `rtcOccluded1`) as this will also prevent
inlining.

7 Bit Ray Mask
--------------

Use just the lower 7 bits of the ray and geometry mask if possible,
even though Embree supports 32 bit ray masks for geometry masking. On
the CPU using any of the 32 bits yields the same performance, but the
ray tracing hardware only supports an 8 bit mask, thus Embree has to
emulate 32 bit masking if used. For that reason the lower 7 mask bits
are hardware accelerated and fast, while the mask bits 7-31 require
some software intervention and using them reduces performance. To turn
on 32 bit ray masks use the RTC\_FEATURE\_FLAG\_32\_BIT\_RAY\_MASK (see
section [RTCFeatureFlags]).

Limit Motion Blur Motions
-------------------------

The motion blur implementation on SYCL has some limitations regarding
supported motion. Primitive motion should be maximally as large as a
small multiple of the primitive size, otherwise performance can degrade
a lot. If detailed geometry moves fast, best put the geometry into an
instance, and apply motion blur to the instance itself, which
efficiently allows larger motions. As a fallback, problematic scenes
can always still get rendered robustly on the CPU.

Generic Pointers
----------------

Embree uses standard C++ pointers in its implementation. SYCL might not
be able to detect the memory space these pointers refer to and has to
treat them as generic pointers which are not performing optimal. The
DPC++ compiler has advanced optimizations to infer the proper address
space to avoid usage of generic pointers.

However, if you still encounter the following warning during ahead of
time compilation of SYCL kernels, then loads from generic pointer are
present:

    warning: Adding XX occurrences of additional control flow due to presence
             of generic address space operations in function YYY.

To work around this issue we recommend:

-   Do not use local memory inside kernels that trace rays. In this
    case the DPC++ compiler knows that no local memory pointer can
    exist and will optimize generic loads. As this is typically the
    case for renderers, generic pointer will typically not cause
    issues.

-   Indirectly callable functions may still cause problems, even if
    your kernel does not use local memory. Thus best use SYCL pointers
    like sycl::global\_ptr`<T>`{=html} and
    sycl::private\_ptr`<T>`{=html} in indirectly callable functions to
    avoid generic address space usage.

-   You can also enforce usage of global pointers using the following
    DPC++ compile flags:
    `-cl-intel-force-global-mem-allocation   -cl-intel-no-local-to-generic`.

Embree Tutorials
================

Embree comes with a set of tutorials aimed at helping users understand
how Embree can be used and extended. There is a very basic minimal
that can be compiled as both C and C++, which should get new users started quickly. 
All other tutorials exist in an Intel® ISPC and C++ version to demonstrate 
the two versions of the API. Look for files
named `tutorialname_device.ispc` for the Intel® ISPC implementation of the
tutorial, and files named `tutorialname_device.cpp` for the single ray C++
version of the tutorial. To start the C++ version use the `tutorialname`
executables, to start the Intel® ISPC version use the `tutorialname_ispc`
executables. All tutorials can print available command line options
using the `--help` command line parameter.

For all tutorials except minimal, you can select an initial camera using 
the `--vp` (camera position), `--vi` (camera look-at point), `--vu` 
(camera up vector), and `--fov` (vertical field of view) command line 
parameters:

    ./triangle_geometry --vp 10 10 10 --vi 0 0 0

You can select the initial window size using the `--size` command line
parameter, or start the tutorials in full screen using the `--fullscreen`
parameter:

    ./triangle_geometry --size 1024 1024
    ./triangle_geometry --fullscreen

The initialization string for the Embree device (`rtcNewDevice` call)
can be passed to the ray tracing core through the `--rtcore` command
line parameter, e.g.:

    ./triangle_geometry --rtcore verbose=2,threads=1

The navigation in the interactive display mode follows the camera orbit
model, where the camera revolves around the current center of interest.
With the left mouse button you can rotate around the center of interest
(the point initially set with `--vi`). Holding Control pressed while
clicking the left mouse button rotates the camera around its location.
You can also use the arrow keys for navigation.

You can use the following keys:

F1
:   Default shading

F2
:   Gray EyeLight shading

F3
:   Traces occlusion rays only.

F4
:   UV Coordinate visualization

F5
:   Geometry normal visualization

F6
:   Geometry ID visualization

F7
:   Geometry ID and Primitive ID visualization

F8
:   Simple shading with 16 rays per pixel for benchmarking.

F9
:   Switches to render cost visualization. Pressing again reduces
    brightness.

F10
:   Switches to render cost visualization. Pressing again increases
    brightness.

f
:   Enters or leaves full screen mode.

c
:   Prints camera parameters.

ESC
:   Exits the tutorial.

q
:   Exits the tutorial.

Minimal
-------

This tutorial is designed to get new users started with Embree.
It can be compiled as both C and C++. It demonstrates how to initialize
a device and scene, and how to intersect rays with the scene.
There is no image output to keep the tutorial as simple as possible.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/minimal/minimal.cpp)

Triangle Geometry
-----------------

[![][imgTriangleGeometry]](https://github.com/embree/embree/blob/master/tutorials/triangle_geometry/triangle_geometry_device.cpp)

This tutorial demonstrates the creation of a static cube and ground
plane using triangle meshes. It also demonstrates the use of the
`rtcIntersect1` and `rtcOccluded1` functions to render primary visibility
and hard shadows. The cube sides are colored based on the ID of the hit
primitive.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/triangle_geometry/triangle_geometry_device.cpp)

Dynamic Scene
-------------

[![][imgDynamicScene]](https://github.com/embree/embree/blob/master/tutorials/dynamic_scene/dynamic_scene_device.cpp)

This tutorial demonstrates the creation of a dynamic scene, consisting
of several deforming spheres. Half of the spheres use the
`RTC_BUILD_QUALITY_REFIT` geometry build quality, which allows Embree
to use a refitting strategy for these spheres, the other half uses the
`RTC_BUILD_QUALITY_LOW` geometry build quality, causing a high
performance rebuild of their spatial data structure each frame. The
spheres are colored based on the ID of the hit sphere geometry.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/dynamic_scene/dynamic_scene_device.cpp)

Multi Scene Geometry
-------------

[![][imgDynamicScene]](https://github.com/embree/embree/blob/master/tutorials/multiscene_geometry/multiscene_geometry_device.cpp)

This tutorial demonstrates the creation of multiple scenes sharing the
same geometry objects.  Here, three scenes are built.  One with all
the dynamic spheres of the Dynamic Scene test and two others each with
half.  The ground plane is shared by all three scenes.  The space bar
is used to cycle the scene chosen for rendering.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/multiscene_geometry/multiscene_geometry_device.cpp)

User Geometry
-------------

[![][imgUserGeometry]](https://github.com/embree/embree/blob/master/tutorials/user_geometry/user_geometry_device.cpp)

This tutorial shows the use of user-defined geometry, to re-implement
instancing, and to add analytic spheres. A two-level scene is created,
with a triangle mesh as ground plane, and several user geometries that
instance other scenes with a small number of spheres of different kinds.
The spheres are colored using the instance ID and geometry ID of the hit
sphere, to demonstrate how the same geometry instanced in different
ways can be distinguished.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/user_geometry/user_geometry_device.cpp)

Viewer
------

[![][imgViewer]](https://github.com/embree/embree/blob/master/tutorials/viewer/viewer_device.cpp)

This tutorial demonstrates a simple OBJ viewer that traces primary
visibility rays only. A scene consisting of multiple meshes is created,
each mesh sharing the index and vertex buffer with the application.
It also demonstrates how to support additional per-vertex data, such as
shading normals.

You need to specify an OBJ file at the command line for this tutorial to
work:

    ./viewer -i model.obj

[Source Code](https://github.com/embree/embree/blob/master/tutorials/viewer/viewer_device.cpp)

Intersection Filter
-------------------

[![][imgIntersectionFilter]](https://github.com/embree/embree/blob/master/tutorials/intersection_filter/intersection_filter_device.cpp)

This tutorial demonstrates the use of filter callback functions to
efficiently implement transparent objects. The filter function used for
primary rays lets the ray pass through the geometry if it is entirely
transparent. Otherwise, the shading loop handles the transparency
properly, by potentially shooting secondary rays. The filter function
used for shadow rays accumulates the transparency of all surfaces along
the ray, and terminates traversal if an opaque occluder is hit.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/intersection_filter/intersection_filter_device.cpp)

Instanced Geometry
------------------

[![][imgInstancedGeometry]](https://github.com/embree/embree/blob/master/tutorials/instanced_geometry/instanced_geometry_device.cpp)

This tutorial demonstrates the in-build instancing feature of Embree, by
instancing a number of other scenes built from triangulated spheres. The
spheres are again colored using the instance ID and geometry ID of the
hit sphere, to demonstrate how the same geometry instanced in different
ways can be distinguished.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/instanced_geometry/instanced_geometry_device.cpp)

Instance Array Geometry
----------------------

[![][imgForest]](https://github.com/embree/embree/blob/master/tutorials/forest/forest_device.cpp)

This tutorial demonstrates the usage of instance arrays in Embree. Instance arrays
are large collections of similar objects. Examples are sand dunes that consist
of millions of instances of a few grain models or, like here, a forest consisting of
many instances of a few tree models.

In this application can switch between representing the scene with regular
instances or (one!) instance array. It also prints several stats, that
demonstrate the memory savings and faster BVH build times when using instance
arrays for such scenes. Instance arrays come with a small overhead on CPU and
should be preferred if memory consumption is more important than raytracing
performance.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/forest/forest_device.cpp)

Multi Level Instancing
----------------------

[![][imgMultiLevelInstancing]](https://github.com/embree/embree/blob/master/tutorials/multi_instanced_geometry/multi_instanced_geometry_device.cpp)

This tutorial demonstrates multi-level instancing, i.e., nesting instances
into instances. To enable the tutorial, set the compile-time variable
`EMBREE_MAX_INSTANCE_LEVEL_COUNT` to a value other than the default 1.
This variable is available in the code as `RTC_MAX_INSTANCE_LEVEL_COUNT`.

The renderer uses a basic path tracing approach, and the
image will progressively refine over time.
There are two levels of instances in this scene: multiple instances of
the same tree nest instances of a twig.
Intersections on up to `RTC_MAX_INSTANCE_LEVEL_COUNT` nested levels of
instances work out of the box. Users may obtain the *instance ID stack* for
a given hitpoint from the `instID` member.
During shading, the instance ID stack is used to accumulate
normal transformation matrices for each hit. The tutorial visualizes
transformed normals as colors.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/multi_instanced_geometry/multi_instanced_geometry_device.cpp)

Path Tracer
-----------

[![][imgPathtracer]](https://github.com/embree/embree/blob/master/tutorials/pathtracer/pathtracer_device.cpp)

This tutorial is a simple path tracer, based on the viewer tutorial.

You need to specify an OBJ file and light source at the command line for
this tutorial to work:

    ./pathtracer -i model.obj --ambientlight 1 1 1

As example models we provide the "Austrian Imperial Crown" model by
[Martin Lubich](http://www.loramel.net) and the "Asian Dragon" model from the
[Stanford 3D Scanning Repository](http://graphics.stanford.edu/data/3Dscanrep/).

[crown.zip](https://github.com/embree/models/releases/download/release/crown.zip)

[asian_dragon.zip](https://github.com/embree/models/releases/download/release/asian_dragon.zip)

To render these models execute the following:

    ./pathtracer -c crown/crown.ecs
    ./pathtracer -c asian_dragon/asian_dragon.ecs

[Source Code](https://github.com/embree/embree/blob/master/tutorials/pathtracer/pathtracer_device.cpp)

Hair
----

[![][imgHairGeometry]](https://github.com/embree/embree/blob/master/tutorials/hair_geometry/hair_geometry_device.cpp)

This tutorial demonstrates the use of the hair geometry to render a
hairball.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/hair_geometry/hair_geometry_device.cpp)

Curve Geometry
--------------

[![][imgCurveGeometry]](https://github.com/embree/embree/blob/master/tutorials/curve_geometry/curve_geometry_device.cpp)

This tutorial demonstrates the use of the Linear Basis, B-Spline, and Catmull-Rom curve geometries.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/curve_geometry/curve_geometry_device.cpp)

Subdivision Geometry
--------------------

[![][imgSubdivisionGeometry]](https://github.com/embree/embree/blob/master/tutorials/subdivision_geometry/subdivision_geometry_device.cpp)

This tutorial demonstrates the use of Catmull-Clark subdivision
surfaces.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/subdivision_geometry/subdivision_geometry_device.cpp)

Displacement Geometry
---------------------

[![][imgDisplacementGeometry]](https://github.com/embree/embree/blob/master/tutorials/displacement_geometry/displacement_geometry_device.cpp)

This tutorial demonstrates the use of Catmull-Clark subdivision
surfaces with procedural displacement mapping using a constant edge
tessellation level.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/displacement_geometry/displacement_geometry_device.cpp)

Grid Geometry
---------------------

[![][imgGridGeometry]](https://github.com/embree/embree/tree/master/tutorials/grid_geometry)

This tutorial demonstrates the use of the memory efficient grid
primitive to handle highly tessellated and displaced geometry.

[Source Code](https://github.com/embree/embree/tree/master/tutorials/grid_geometry)

Point Geometry
---------------------

[![][imgPointGeometry]](https://github.com/embree/embree/blob/master/tutorials/point_geometry/point_geometry_device.cpp)

This tutorial demonstrates the use of the three representations
of point geometry.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/point_geometry/point_geometry_device.cpp)

Motion Blur Geometry
--------------------

[![][imgMotionBlurGeometry]](https://github.com/embree/embree/blob/master/tutorials/motion_blur_geometry/motion_blur_geometry_device.cpp)

This tutorial demonstrates rendering of motion blur using the
multi-segment motion blur feature. Shown is motion blur of a triangle mesh,
quad mesh, subdivision surface, line segments, hair geometry, Bézier
curves, instantiated triangle mesh where the instance moves,
instantiated quad mesh where the instance and the quads move, and user
geometry.

The number of time steps used can be configured using the `--time-steps
<int>` and `--time-steps2 <int>` command line parameters, and the
geometry can be rendered at a specific time using the the `--time
<float>` command line parameter.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/motion_blur_geometry/motion_blur_geometry_device.cpp)

Quaternion Motion Blur
----------------------

[![][imgQuaternionMotionBlur]](https://github.com/embree/embree/blob/master/tutorials/quaternion_motion_blur/quaternion_motion_blur_device.cpp)

This tutorial demonstrates rendering of motion blur using quaternion
interpolation. Shown is motion blur using spherical linear interpolation of
the rotational component of the instance transformation on the left and
simple linear interpolation of the instance transformation on the right. The
number of time steps can be modified as well.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/quaternion_motion_blur/quaternion_motion_blur_device.cpp)

Interpolation
-------------

[![][imgInterpolation]](https://github.com/embree/embree/blob/master/tutorials/interpolation/interpolation_device.cpp)

This tutorial demonstrates interpolation of user-defined per-vertex data.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/interpolation/interpolation_device.cpp)

Closest Point
----------------------

[![][imgClosestPoint]](https://github.com/embree/embree/blob/master/tutorials/closest_point/closest_point_device.cpp)

This tutorial demonstrates a use-case of the point query API. The scene
consists of a simple collection of objects that are instanced and for several
point in the scene (red points) the closest point on the surfaces of the
scene are computed (white points). The closest point functionality is
implemented for Embree internal and for user-defined instancing. The tutorial
also illustrates how to handle instance transformations that are not
similarity transforms.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/closest_point/closest_point_device.cpp)

Voronoi
----------------------

[![][imgVoronoi]](https://github.com/embree/embree/blob/master/tutorials/voronoi/voronoi_device.cpp)

This tutorial demonstrates how to implement nearest neighbour lookups using
the point query API. Several colored points are located on a plane and the
corresponding voroni regions are illustrated.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/voronoi/voronoi_device.cpp)

Collision Detection
----------------------

[![][imgCollision]](https://github.com/embree/embree/blob/master/tutorials/collide/collide_device.cpp)

This tutorial demonstrates how to implement collision detection using
the collide API. A simple cloth solver is setup to collide with a sphere.

The cloth can be reset with the `space` bar.  The sim stepped once with `n` 
and continuous simulation started and paused with `p`.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/collide/collide_device.cpp)

BVH Builder
-----------

This tutorial demonstrates how to use the templated hierarchy builders
of Embree to build a bounding volume hierarchy with a user-defined
memory layout using a high-quality SAH builder using spatial splits, a
standard SAH builder, and a very fast Morton builder.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/bvh_builder/bvh_builder_device.cpp)

BVH Access
-----------

This tutorial demonstrates how to access the internal triangle
acceleration structure build by Embree. Please be aware that the
internal Embree data structures might change between Embree updates.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/bvh_access/bvh_access.cpp)

Find Embree
-----------

This tutorial demonstrates how to use the `FIND_PACKAGE` CMake feature
to use an installed Embree. Under Linux and macOS the tutorial finds
the Embree installation automatically, under Windows the `embree_DIR`
CMake variable must be set to the following folder of the Embree
installation: `C:\Program Files\Intel\Embree3`.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/find_embree/CMakeLists.txt)

Next Hit
-----------

This tutorial demonstrates how to robustly enumerate all hits along
the ray using multiple ray queries and an intersection filter
function. To improve performance, the tutorial also supports
collecting the next N hits in a single ray query.

[Source Code](https://github.com/embree/embree/blob/master/tutorials/next_hit/next_hit_device.cpp)



[Embree API]: #embree-api
[Embree Tutorials]: #embree-tutorials
[Ray Layout]: #ray-layout
[Extending the Ray Structure]: #extending-the-ray-structure
[Embree Example Renderer]: https://embree.github.io/renderer.html
[Triangle Geometry]: #triangle-geometry
[Stream Viewer]: #stream-viewer
[User Geometry]: #user-geometry
[Instanced Geometry]: #instanced-geometry
[Instance Array Geometry]: #instance-array-geometry
[Multi Level Instancing]: #multi-level-instancing
[Intersection Filter]: #intersection-filter
[Hair]: #hair
[Curves]: #bézier-curves
[Subdivision Geometry]: #subdivision-geometry
[Displacement Geometry]: #displacement-geometry
[Quaternion Motion Blur]: #quaternion-motion-blur
[BVH Builder]: #bvh-builder
[Interpolation]: #interpolation
[Closest Point]: #closest-point
[Voronoi]: #voronoi
[imgHalfEdges]: https://embree.github.io/images/half_edges.png
[imgTriangleUV]: https://embree.github.io/images/triangle_uv.png
[imgQuadUV]: https://embree.github.io/images/quad_uv.png
[imgTriangleGeometry]: https://embree.github.io/images/triangle_geometry.jpg
[imgDynamicScene]: https://embree.github.io/images/dynamic_scene.jpg
[imgUserGeometry]: https://embree.github.io/images/user_geometry.jpg
[imgViewer]: https://embree.github.io/images/viewer.jpg
[imgInstancedGeometry]: https://embree.github.io/images/instanced_geometry.jpg
[imgForest]: https://embree.github.io/images/forest.jpg
[imgMultiLevelInstancing]: https://embree.github.io/images/multi_level_instancing.jpg
[imgIntersectionFilter]: https://embree.github.io/images/intersection_filter.jpg
[imgPathtracer]: https://embree.github.io/images/pathtracer.jpg
[imgHairGeometry]: https://embree.github.io/images/hair_geometry.jpg
[imgCurveGeometry]: https://embree.github.io/images/curve_geometry.jpg
[imgSubdivisionGeometry]: https://embree.github.io/images/subdivision_geometry.jpg
[imgDisplacementGeometry]: https://embree.github.io/images/displacement_geometry.jpg
[imgGridGeometry]: https://embree.github.io/images/grid_geometry.jpg
[imgPointGeometry]: https://embree.github.io/images/point_geometry.jpg
[imgMotionBlurGeometry]: https://embree.github.io/images/motion_blur_geometry.jpg
[imgQuaternionMotionBlur]: https://embree.github.io/images/quaternion_motion_blur.jpg
[imgInterpolation]: https://embree.github.io/images/interpolation.jpg
[imgClosestPoint]: https://embree.github.io/images/closest_point.jpg
[imgVoronoi]: https://embree.github.io/images/voronoi.jpg
[imgCollision]: https://embree.github.io/images/collide.jpg