• Stars
    star
    356
  • Rank 119,446 (Top 3 %)
  • Language
    C
  • License
    BSD 3-Clause "New...
  • Created almost 9 years ago
  • Updated about 1 year ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Intel QuickAssist Technology( QAT) OpenSSL Engine (an OpenSSL Plug-In Engine) which provides cryptographic acceleration for both hardware and optimized software using Intel QuickAssist Technology enabled Intel platforms. https://developer.intel.com/quickassist

Intel® QuickAssist Technology(QAT) OpenSSL* Engine

Intel® QuickAssist Technology OpenSSL* Engine (QAT_Engine) supports acceleration for both hardware as well as optimized software based on vectorized instructions. This change starting with the 3rd Generation Intel® Xeon® Scalable Processors family provides users more options to accelerate their workloads. The QAT OpenSSL* engine now supports the ability to accelerate from the stand OpenSSL* to basic Intel instruction set, to either Hardware acceleration path (via the qat_hw path) or via the optimized SW path (qat_sw lib). This document details the capabilities and interfaces for both internal libraries which maintaining the cohesiveness of being all packaged up with the QAT_Engine.

The image below illustrates the high-level software architecture of the QAT_Engine. Applications such as NGINX and HAProxy are common applications which interfaces to OpenSSL*. OpenSSL* is a toolkit for TLS/SSL protocols and has developed a modular system to plugin device-specific engines starting with version 1.1.0. As mentioned above, within the QAT_Engine are two separate internal entities by which acceleration can be performed. Depending on your particular use case, the QAT_Engine can be configured to meet your specific acceleration needs. This document details the capabilities of the QAT engine as well as the limitations. Both the hardware and software requirements are explained followed by detailed instructions how to install QAT_Engine. If you have any issues integrating the QAT_Engine, we’ve added a troubleshooting section.

drawing

Features

Features of the QAT_Engine are described here.

Limitations and Known Issues

Limitations and known issues for the QAT_Engine are described here.

Requirements

Installation Instructions

Installation consists of the following:

Install Prerequisites
qat_hw Prerequisites


Install the Intel® QuickAssist Technology Driver using instructions from the Getting Started Guide

Platform Getting Started Guide
Intel® Xeon® with Intel® C62X Series Chipset
Intel® Atom™ Processor with Intel&reg C3xx series
Intel® Communications Chipset 8925 to 8955 Series
Intel® QAT Software for Linux* - Getting Started Guide Hardware v1.x CE Release
Intel® Xeon® Scalable Processor family with Intel® QAT Gen4/Gen4m Intel® QAT Software for Linux* - Getting Started Guide Hardware v2.0

Other technical collaterals of the Intel® QuickAssist Technology driver can be found in the below page.

Contiguous memory driver

The Intel® QAT API requires many of the data structures (those that will be passed to the hardware) to be allocated in contiguous pinned memory in order to support DMA operations. You must either supply your own contiguous memory driver and make changes to the engine to make use of it or use one of the following drivers:

User Space DMA-able Memory (USDM) Component

The Intel® QAT Driver for QAT_HW comes with its own contiguous pinned memory driver that is compatible with the Intel® QAT OpenSSL* Engine. The USDM component is of a higher quality than the qat_contig_mem driver provided within the Intel® QAT OpenSSL* Engine, and is the preferred option. The USDM component is used by the Intel® QAT Driver itself, and also has the following additional features:

  • Support for virtualization
  • Support for configurable slab sizes
  • Support for configurable secure freeing of memory (overwrite with zeros)
  • Support for configurable slab caching
  • Support for newer kernels
  • Support for thread specific memory to avoid locks (QAT_HW Version 1.7 & 1.8 only)

The USDM component is located within the Intel® QAT Driver for QAT_HW source code in the following subdirectory: quickassist/utilities/libusdm_drv. As the USDM component is also used by the QAT_HW driver itself it will have already been built when the driver was built. It may also already be loaded as well, and you can check by running lsmod and looking for usdm_drv in the list. If not present it can be loaded as follows:

modprobe usdm_drv.ko

The USDM thread specific memory can be enabled in QAT_HW driver using the below configure flags in driver build which is only needed for multithreaded application for performance improvement. This is supported from version 4.20 of QAT_HW Version 1.7 driver only.

./configure --enable-icp-thread-specific-usdm --enable-128k-slab

Example contiguous memory driver - qat_contig_mem

This step is not required if using the default USDM driver above. The Intel® QAT OpenSSL* Engine comes with an example kernel space contiguous memory driver that can be used to try out operation of the engine. It is considered to be an example only and is not written to be a production quality driver. The use of the qat_contig_mem driver can be enabled using the configure option --enable-qat_hw_contig_mem that tells the build that the Intel® QAT OpenSSL* Engine should be compiled to use the qat_contig_mem component instead of the USDM memory driver above.

Building and loading the qat_contig_mem driver assumming:

  • The Intel® QAT OpenSSL* Engine was cloned to its own location at the root of the drive: /.

To build/install the qat_contig_mem driver follow these steps:

cd /QAT_Engine/qat_contig_mem
make
make load
make test

The expected output from make test should be something similar to the following:

seg mapped to 0x7f9eedd6e000, virtualAddress in seg 0xffff880ac9c0c000,
length 64
Hello world!
# PASS Verify for QAT Contig Mem Test
qat_sw Prerequisites
Install OpenSSL* (Note this step is not required if OpenSSL* 1.1.1 or 3.0 is already installed)

Build OpenSSL*

This step is not required if building the Intel® QAT OpenSSL* Engine against system prebuilt OpenSSL* 1.1.1 or 3.0. When using the prebuild system OpenSSL library the engine library is installed in the system OpenSSL engines directory.

Clone OpenSSL* from Github* at the following location:

git clone https://github.com/openssl/openssl.git

It is recommended to checkout and build against the OpenSSL* git tag specified in the Software Requirements section. OpenSSL* Version 1.1.1 and 3.0 are only supported.

Due to the nature of the Intel® QAT OpenSSL* Engine being a dynamic engine it can only be used with shared library builds of OpenSSL*.

Note: It is not recommended to install the accelerated version of OpenSSL* as your default system library. If you do, you may find that acceleration is used unexpectedly by other applications on the system resulting in undesired/unsupported behaviour. The --prefix can be used with the ./config command to specify the location that make install will copy files to. Please see the OpenSSL* INSTALL file for full details on usage of the --prefix option.

With OpenSSL* version 1.1.0 and on, binaries are installed in standard directories by default, and the addition of runpath directories is not done automatically. If you wish to install OpenSSL* in a non-standard location (recommended), the runpath directories can be specified via the OpenSSL* Configure command, which recognises the arguments -rpath and -R to support user-added rpaths. For convenience, a Makefile variable LIBRPATH has also been added which is defined as the full path to a subdirectory of the installation directory. The subdirectory is named lib by default. If you do not wish to use LIBRPATH, the rpath can be specified directly. The syntax for specifying a rpath is as follows:

./config [options] -Wl,-rpath,\${LIBRPATH}

The -rpath can be replaced with -R for brevity. If you do not wish to use the built-in variable LIBRPATH, the syntax for specifying a rpath of /usr/local/ssl/lib for example would be:

./config [options] -Wl,-rpath,/usr/local/ssl/lib

Alternatively, you can specify the rpath by adding it to the environment variable LD_LIBRARY_PATH via the command:

export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:`RPATH`

where RPATH is the full path(s) to the shared libraries. This is not the preferred method though. ` The following example is assuming:

  • The OpenSSL* source was cloned from Github* to its own location at the root of the drive: /.
  • You want OpenSSL* to be installed to /usr/local/ssl.

An example build would be:

cd /openssl
./config --prefix=/usr/local/ssl -Wl,-rpath,\${LIBRPATH}
make depend (if recommended by the OpenSSL* build system)
make
make install

As the Intel® QAT OpenSSL* Engine will be built as a dynamic engine it is important to tell OpenSSL* where to find the dynamic engines at runtime. This is achieved by exporting the following environment variable (assuming the example paths above):

OpenSSL 1.1.1: export OPENSSL_ENGINES=/usr/local/ssl/lib/engines-1.1

OpenSSL 3.0: export OPENSSL_ENGINES=/usr/local/ssl/lib64/engines-3

Note: This variable will need to be present in the environment whenever the engine is used and not needed when provider is used.

Load/Initialize Engine using the OpenSSL* config file is located here

Further information on building OpenSSL* can be found in the INSTALL file distributed with the OpenSSL* source code or on the official OpenSSL* Wiki in the Compilation and Installation section: https://wiki.openssl.org/index.php/Compilation_and_Installation

Build the Intel® QuickAssist Technology OpenSSL* Engine

Build the Intel® QuickAssist Technology OpenSSL* Engine

Clone the Github* repository containing the Intel® QAT OpenSSL* Engine:

git clone https://github.com/intel/QAT_Engine.git

When building it is possible to specify command line options that can be used to turn engine functionality on and off. The complete list of these options is available here.

The prerequisite to run autogen.sh is to have autotools (autoconf, automake, libtool and pkg-config) installed in the system.

cd /QAT_Engine
./autogen.sh

./autogen.sh will regenerate autoconf tools files.

Example Builds

Here are a few example builds that demonstrate how the Intel® QAT OpenSSL* Engine can be configured to use qat_hw and/or qat_sw.

Example 1: qat_hw target with OpenSSL\* 1.1.1 or 3.0 built from source

The following example is assuming:

  • The Intel® QAT OpenSSL* Engine was cloned to its own location at the root of the drive: /.
  • The Intel® QAT Driver version 1.7 or 2.0 was unpacked within /QAT and using the USDM component.
  • OpenSSL* 1.1.1 or 3.0 built from source is being used and installed to /usr/local/ssl.

To build and install the Intel® QAT OpenSSL* Engine:

cd /QAT_Engine
./configure \
--with-qat_hw_dir=/QAT \
--with-openssl_install_dir=/usr/local/ssl
make
make install

In the above example this will create the file qatengine.so and copy it to /usr/local/ssl/lib/engines-1.1.

For building QAT Engine against qatlib(intree driver) from source which is installed to default location "/usr/local" use --with-qat_hw_dir=/usr/local or provide the path that is used in the prefix to build qatlib.

If qatlib is installed via RPM then -with-qat_hw_dir is not needed as qatengine automatically picks qatlib libraries and header from default location /usr/lib64.


Example 2: qat_hw target with Prebuilt OpenSSL\* 1.1.1 or 3.0

The following example is assuming:

  • The Intel® QAT OpenSSL* Engine was cloned to its own location at the root of the drive: /.
  • The Intel® QAT Driver was unpacked within /QAT and using the USDM component.
  • Prebuilt OpenSSL* (both library and devel RPM packages) are installed in the system and the OpenSSL* version is in the 1.1.1 or 3.0 series.

To build and install the Intel® QAT OpenSSL* Engine:

cd /QAT_Engine
./configure  --with-qat_hw_dir=/QAT
make
make install

In the above example this will create the file qatengine.so and copy it to the engines dir of the system which can be checked using pkg-config --variable=enginesdir libcrypto.


Example 3: qat_hw + qat_sw target with Prebuilt OpenSSL\* 1.1.1 or 3.0

The following example is assuming:

  • The Intel® QAT OpenSSL* Engine was cloned to its own location at the root of the drive: /.
  • The Intel® QAT Driver was unpacked within /QAT and using the USDM component.
  • Intel® Multi-Buffer Crypto for IPsec Library was installed to the default path
  • OpenSSL* 1.1.1 or 3.0 built from source is being used and installed to /usr/local/ssl.

To build and install the Intel® QAT OpenSSL* Engine:

cd /QAT_Engine
./configure \
--with-qat_hw_dir=/QAT \
--enable-qat_sw \
--with-openssl_install_dir=/usr/local/ssl
make
make install
  • In the above example this will create the file qatengine.so and copy it to /usr/local/ssl/lib/engines-1.1.
  • AES-GCM operations are handled by qat_sw
  • Other cryptographic operations are handled by qat_hw

Example 4: qat_sw target with Prebuilt OpenSSL\* 1.1.1 or 3.0

The following example is assuming:

  • The Intel® QAT OpenSSL* Engine was cloned to its own location at the root of the drive: /.
  • The Intel® Crypto Multi-buffer library was installed to the default path (/usr/local).
  • The Intel® Multi-Buffer crypto for IPsec Library was installed to its default path (/usr/). (Optional if QAT SW AES-GCM support is not needed).
  • Prebuilt OpenSSL* 1.1.1 or 3.0 from the system is used.

To build and install the Intel® QAT OpenSSL* Engine with QAT Software support:

cd /QAT_Engine
./configure --enable-qat_sw
make
make install

In the above example, --disable-qat_hw needs to be provided if the system has qatlib installed. Note : --enable-qat_sw checks crypto_mb and IPSec_MB libraries in its respective default path (/usr/local/lib and /usr/lib) or in the path provided in the config flag --with-qat_sw_crypto_mb_install_dir (for crypto_mb) and --with-qat_sw_ipsec_mb_install_dir (for ipsec_mb). If any of the libraries is not installed then their corresponding algorithm support is disabled (crypto_mb library for PKE algorithms and IPSec_mb library for AES-GCM).

Enable plock optimization in QAT Engine for Multithread applications

Progressive Lock (plock) optimization in QAT Engine provides an alternative to pthread's rwlock as in Haproxy. This plock is an alternative to read/write locks in the pthread library and optimizes read/write locking with a series of atomic operations and there will be no more syscalls in case of contention. The benefit of plock is that it saves lot of CPU and improves performance with higher number of threads for multithread applications. In case of Haproxy, this can directly enabled in the Haproxy build with addition build flag "USE_PTHREAD_EMULATION=1".

Build Flag to enable plock optimization

QAT Engine generates dynamic libraries and currently enables Plock by preloading.

  • When compiling the qat engine, add the compilation option --enable-qat_plock, and the libplock.so will be generated. eg: ./configure --enable-qat_sw --enable-qat_plock --with-openssl_install_dir=${OPENSSL_INSTALL_DIR}
  • Set the environment variable ‘LD_PRELOAD’ to use the plock optimization. eg: export LD_PRELOAD = ${OPENSSL_INSTALL_DIR}/lib/engine-1.1/libplock.so Note: The LD_PRELOAD must be set before launching the multithread Appliaction.

[For BoringSSL only] Build the Intel® QuickAssist Technology BoringSSL* Library

Please refer BoringSSL section for steps to build the Intel® QuickAssist Technology BoringSSL* library which supports RSA and ECDSA QAT Hardware and QAT Software Acceleration using BoringSSL.

Copy the Intel® QuickAssist Technology Driver config file(s)

Copy the Intel® QuickAssist Technology Driver config files for qat_hw

This step is not required for qat_sw target. The Intel® QAT OpenSSL* Engine includes example conf files to use with the Intel® QAT Driver. The Intel® QAT OpenSSL* Engine will not function with the default Intel® QAT Driver conf file because the default conf does not contain a [SHIM] section which the Intel® QAT OpenSSL* Engine requires by default. The default section name in the QAT OpenSSL* Engine can be modified if required by either using the engine ctrl command SET_CONFIGURATION_SECTION_NAME or by setting the environment variable "QAT_SECTION_NAME". The conf files are located at:

/path/to/qat_engine/qat_hw_config

The files are grouped by acceleration device(dh895xcc or c6xx or c3xxx or 200xx or c4xxx or 4xxx), please choose the files appropriate to your acceleration device only.

The files are also split into multi_process and multi_thread.

If your application runs one (or very few) processes, but has multiple threads in each process, each accessing the acceleration device, then you should pick the multi_thread config files. An example of this is a webserver that creates a new thread for each incoming connection.

If your application scales by creating new processes, then you should pick the multi_process config files. An example of this is an event driven application that runs as a single thread in an event loop. In this type of application it is usual for the application to create at least one new process for each cpu core you want to utilize.

For event driven polling based application, change the parameter Cy$nIsPolled=1 to Cy$nIsPolled=2 for each instances($n) in the respective config file to use event driven polling support. Event driven config files are only supported in Linux. Once you have decided which config file you should use, or created your own you should follow the procedure below to install it:

  1. Stop the acceleration driver as decribed in the Section 3.4 Starting/Stopping the Acceleration software from the Getting Started Guide available in Intel® QuickAssist Technology Driver

  2. Copy the appropriate .conf file to /etc for n number of QAT devices

  3. Start the acceleration driver as decribed in the Section 3.4 Starting/Stopping the Acceleration software from the Getting Started Guide available in Intel® QuickAssist Technology Driver

Test the Intel® QuickAssist Technology OpenSSL* Engine

Test the Intel® QuickAssist Technology OpenSSL* Engine

Run this command to verify the Intel® QAT OpenSSL* Engine is loaded correctly:

cd /path/to/openssl_install/bin
./openssl engine -t -c -v qatengine

qat_hw target output will be:

(qatengine) Reference implementation of QAT crypto engine(qat_hw) <qatengine version>
 [RSA, DSA, DH, AES-128-CBC-HMAC-SHA1, AES-128-CBC-HMAC-SHA256,
 AES-256-CBC-HMAC-SHA1, AES-256-CBC-HMAC-SHA256, TLS1-PRF, HKDF, X25519, X448]
    [ available ]
    ENABLE_EXTERNAL_POLLING, POLL, SET_INSTANCE_FOR_THREAD,
    GET_NUM_OP_RETRIES, SET_MAX_RETRY_COUNT, SET_INTERNAL_POLL_INTERVAL,
    GET_EXTERNAL_POLLING_FD, ENABLE_EVENT_DRIVEN_POLLING_MODE,
    GET_NUM_CRYPTO_INSTANCES, DISABLE_EVENT_DRIVEN_POLLING_MODE,
    SET_EPOLL_TIMEOUT, SET_CRYPTO_SMALL_PACKET_OFFLOAD_THRESHOLD,
    ENABLE_INLINE_POLLING, ENABLE_HEURISTIC_POLLING,
    GET_NUM_REQUESTS_IN_FLIGHT, INIT_ENGINE, SET_CONFIGURATION_SECTION_NAME,
    ENABLE_SW_FALLBACK, HEARTBEAT_POLL, DISABLE_QAT_OFFLOAD

qat_sw target output will be:

(qatengine) Reference implementation of QAT crypto engine(qat_sw) <qatengine version>
 [RSA, id-aes128-GCM, id-aes192-GCM, id-aes256-GCM, X25519]
     [ available ]
     ENABLE_EXTERNAL_POLLING, POLL, ENABLE_HEURISTIC_POLLING,
     GET_NUM_REQUESTS_IN_FLIGHT, INIT_ENGINE

Detailed information about the engine specific messages is available here. Also ./openssl engine -t -c -vvvv qatengine gives brief decription about each ctrl command.

Testing the Intel® QuickAssist Technology OpenSSL* Engine using OpenSSL* speed utility

Testing the Intel® QuickAssist Technology OpenSSL* Engine using OpenSSL* speed utility

cd /path/to/openssl_install/bin

qat_hw

* RSA 2K
  * Asynchronous
  ./openssl speed -engine qatengine -elapsed -async_jobs 72 rsa2048
  * Synchronous
  ./openssl speed -engine qatengine -elapsed rsa2048
  * OpenSSL Software
  ./openssl speed -elapsed rsa2048
* ECDH Compute Key
  * Asynchronous
  ./openssl speed -engine qatengine -elapsed -async_jobs 36 ecdh
  * Synchronous
  ./openssl speed -engine qatengine -elapsed ecdh
  * OpenSSL Software
  ./openssl speed -elapsed ecdh
* Chained Cipher: aes-128-cbc-hmac-sha1
  * Asynchronous
  ./openssl speed -engine qatengine -elapsed -async_jobs 128 -multi 2 -evp aes-128-cbc-hmac-sha1
  * Synchronous
  ./openssl speed -engine qatengine -elapsed -multi 2 -evp aes-128-cbc-hmac-sha1
  * OpenSSL Software
  ./openssl speed -elapsed -multi 2 -evp aes-128-cbc-hmac-sha1

qat_sw (Intel(R) Crypto Multi-buffer library)

* RSA 2K
  ./openssl speed -engine qatengine -elapsed -async_jobs 8 rsa2048
* ECDH X25519
  ./openssl speed -engine qatengine -elapsed -async_jobs 8 ecdhx25519
* ECDH P-256
  ./openssl speed -engine qatengine -elapsed -async_jobs 8 ecdhp256
* ECDSA P-256
  ./openssl speed -engine qatengine -elapsed -async_jobs 8 ecdsap256
* ECDH P-384
  ./openssl speed -engine qatengine -elapsed -async_jobs 8 ecdhp384
* ECDSA P-384
  ./openssl speed -engine qatengine -elapsed -async_jobs 8 ecdsap384

qat_sw (Intel(R) Multi-Buffer Crypto for IPsec)

* AES-128-GCM
  ./openssl speed -engine qatengine -elapsed -evp aes-128-gcm
* AES-192-GCM
  ./openssl speed -engine qatengine -elapsed -evp aes-192-gcm
* AES-256-GCM
  ./openssl speed -engine qatengine -elapsed -evp aes-256-gcm
Testing the Intel® QuickAssist Technology OpenSSL* Engine using testapp utility

Testing the Intel® QuickAssist Technology OpenSSL* Engine using testapp utility

cd /path/to/qat_engine
make test
./testapp.sh QAT_HW (For testing algorithms supported by QAT_HW)
./testapp.sh QAT_SW (For testing algorithms supported by QAT_SW)

The testapp.sh script will run the corresponding functional tests supported by QAT_HW and QAT_SW. Please note that the QAT Engine should be built with that support for the tests.

Additional information for testapp tests available with the help option ./testapp -help

Troubleshooting

Troubleshooting information is available here.

Further Reading

Links to additional content is available here.

Licensing Information

Licensing information is available here.

Legal

Legal information is available here.

More Repositories

1

hyperscan

High-performance regular expression matching library
C++
4,478
star
2

acat

Assistive Context-Aware Toolkit (ACAT)
C#
3,191
star
3

haxm

Intel® Hardware Accelerated Execution Manager (Intel® HAXM)
C
3,029
star
4

appframework

The definitive HTML5 mobile javascript framework
CSS
2,435
star
5

neural-compressor

SOTA low-bit LLM quantization (INT8/FP8/INT4/FP4/NF4) & sparsity; leading model compression techniques on TensorFlow, PyTorch, and ONNX Runtime
Python
2,182
star
6

intel-extension-for-transformers

⚡ Build your chatbot within minutes on your favorite device; offer SOTA compression techniques for LLMs; run LLMs efficiently on Intel Platforms⚡
Python
2,122
star
7

pcm

Intel® Performance Counter Monitor (Intel® PCM)
C++
2,083
star
8

intel-extension-for-pytorch

A Python package for extending the official PyTorch that can easily obtain performance on Intel platform
Python
1,203
star
9

linux-sgx

Intel SGX for Linux*
C++
1,180
star
10

scikit-learn-intelex

Intel(R) Extension for Scikit-learn is a seamless way to speed up your Scikit-learn application
Python
954
star
11

llvm

Intel staging area for llvm.org contribution. Home for Intel LLVM-based projects.
918
star
12

nemu

ARCHIVED: Modern Hypervisor for the Cloud. See https://github.com/cloud-hypervisor/cloud-hypervisor instead
C
915
star
13

compute-runtime

Intel® Graphics Compute Runtime for oneAPI Level Zero and OpenCL™ Driver
C++
912
star
14

caffe

This fork of BVLC/Caffe is dedicated to improving performance of this deep learning framework when running on CPU, in particular Intel® Xeon processors.
C++
850
star
15

isa-l

Intelligent Storage Acceleration Library
C
816
star
16

media-driver

C
783
star
17

cve-bin-tool

The CVE Binary Tool helps you determine if your system includes known vulnerabilities. You can scan binaries for over 200 common, vulnerable components (openssl, libpng, libxml2, expat and others), or if you know the components used, you can get a list of known vulnerabilities associated with an SBOM or a list of components and versions.
Python
721
star
18

intel-cmt-cat

User space software for Intel(R) Resource Director Technology
C
630
star
19

fastuidraw

C++
603
star
20

optimization-manual

Contains the source code examples described in the "Intel® 64 and IA-32 Architectures Optimization Reference Manual"
Assembly
602
star
21

libipt

libipt - an Intel(R) Processor Trace decoder library
C
594
star
22

libxcam

libXCam is a project for extended camera(not limited in camera) features and focus on image quality improvement and video analysis. There are lots features supported in image pre-processing, image post-processing and smart analysis. This library makes GPU/CPU/ISP working together to improve image quality. OpenCL is used to improve performance in different platforms.
C++
590
star
23

clDNN

Compute Library for Deep Neural Networks (clDNN)
C++
573
star
24

libva

Libva is an implementation for VA-API (Video Acceleration API)
C
558
star
25

intel-graphics-compiler

C++
503
star
26

wds

Wireless Display Software For Linux OS (WDS)
C++
496
star
27

thermal_daemon

Thermal daemon for IA
C++
485
star
28

x86-simd-sort

C++ header file library for high performance SIMD based sorting algorithms for primitive datatypes
C++
485
star
29

Intel-Linux-Processor-Microcode-Data-Files

466
star
30

gvt-linux

C
463
star
31

kernel-fuzzer-for-xen-project

Kernel Fuzzer for Xen Project (KF/x) - Hypervisor-based fuzzing using Xen VM forking, VMI & AFL
C
441
star
32

tinycbor

Concise Binary Object Representation (CBOR) Library
C
432
star
33

openfl

An open framework for Federated Learning.
Python
427
star
34

cc-oci-runtime

OCI (Open Containers Initiative) compatible runtime for Intel® Architecture
C
415
star
35

tinycrypt

tinycrypt is a library of cryptographic algorithms with a focus on small, simple implementation.
C
373
star
36

compile-time-init-build

C++ library for composing modular firmware at compile-time.
C++
372
star
37

ARM_NEON_2_x86_SSE

The platform independent header allowing to compile any C/C++ code containing ARM NEON intrinsic functions for x86 target systems using SIMD up to SSE4 intrinsic functions
C
369
star
38

yarpgen

Yet Another Random Program Generator
C++
357
star
39

intel-device-plugins-for-kubernetes

Collection of Intel device plugins for Kubernetes
Go
356
star
40

linux-sgx-driver

Intel SGX Linux* Driver
C
334
star
41

safestringlib

C
328
star
42

xess

C
313
star
43

idlf

Intel® Deep Learning Framework
C++
311
star
44

ad-rss-lib

Library implementing the Responsibility Sensitive Safety model (RSS) for Autonomous Vehicles
C++
298
star
45

intel-vaapi-driver

VA-API user mode driver for Intel GEN Graphics family
C
289
star
46

ipp-crypto

C
269
star
47

rohd

The Rapid Open Hardware Development (ROHD) framework is a framework for describing and verifying hardware in the Dart programming language. ROHD enables you to build and traverse a graph of connectivity between module objects using unrestricted software.
Dart
256
star
48

opencl-intercept-layer

Intercept Layer for Debugging and Analyzing OpenCL Applications
C++
255
star
49

FSP

Intel(R) Firmware Support Package (FSP)
C
244
star
50

dffml

The easiest way to use Machine Learning. Mix and match underlying ML libraries and data set sources. Generate new datasets or modify existing ones with ease.
Python
244
star
51

userspace-cni-network-plugin

Go
242
star
52

intel-ipsec-mb

Intel(R) Multi-Buffer Crypto for IPSec
C
238
star
53

isa-l_crypto

Assembly
232
star
54

confidential-computing-zoo

Confidential Computing Zoo provides confidential computing solutions based on Intel SGX, TDX, HEXL, etc. technologies.
CMake
229
star
55

bmap-tools

BMAP Tools
Python
227
star
56

intel-extension-for-tensorflow

Intel® Extension for TensorFlow*
C++
226
star
57

ozone-wayland

Wayland implementation for Chromium Ozone classes
C++
214
star
58

intel-qs

High-performance simulator of quantum circuits
C++
202
star
59

SGXDataCenterAttestationPrimitives

C++
202
star
60

intel-sgx-ssl

Intel® Software Guard Extensions SSL
C
197
star
61

msr-tools

C
195
star
62

depth-camera-web-demo

JavaScript
194
star
63

rmd

Go
189
star
64

CPU-Manager-for-Kubernetes

Kubernetes Core Manager for NFV workloads
Python
187
star
65

asynch_mode_nginx

C
186
star
66

hexl

Intel®️ Homomorphic Encryption Acceleration Library accelerates modular arithmetic operations used in homomorphic encryption
C++
181
star
67

ros_object_analytics

C++
177
star
68

zephyr.js

JavaScript* Runtime for Zephyr* OS
C
176
star
69

generic-sensor-demos

HTML
175
star
70

ipmctl

C
172
star
71

sgx-ra-sample

C++
171
star
72

lmbench

C
171
star
73

cri-resource-manager

Kubernetes Container Runtime Interface proxy service with hardware resource aware workload placement policies
Go
170
star
74

platform-aware-scheduling

Enabling Kubernetes to make pod placement decisions with platform intelligence.
Go
165
star
75

virtual-storage-manager

Python
165
star
76

PerfSpect

System performance characterization tool based on linux perf
Python
164
star
77

he-transformer

nGraph-HE: Deep learning with Homomorphic Encryption (HE) through Intel nGraph
C++
163
star
78

systemc-compiler

This tool translates synthesizable SystemC code to synthesizable SystemVerilog.
C++
155
star
79

webml-polyfill

Deprecated, the Web Neural Network Polyfill project has been moved to https://github.com/webmachinelearning/webnn-polyfill
Python
153
star
80

pmem-csi

Persistent Memory Container Storage Interface Driver
Go
151
star
81

libyami

Yet Another Media Infrastructure. it is core part of media codec with hardware acceleration, it is yummy to your video experience on Linux like platform.
C++
148
star
82

ros_openvino_toolkit

C++
147
star
83

rib

Rapid Interface Builder (RIB) is a browser-based design tool for quickly prototyping and creating the user interface for web applications. Layout your UI by dropping widgets onto a canvas. Run the UI in an interactive "Preview mode". Export the generated HTML and Javascript. It's that simple!
JavaScript
147
star
84

ideep

Intel® Optimization for Chainer*, a Chainer module providing numpy like API and DNN acceleration using MKL-DNN.
C++
145
star
85

libva-utils

Libva-utils is a collection of tests for VA-API (VIdeo Acceleration API)
C
144
star
86

gmmlib

C++
141
star
87

numatop

NumaTOP is an observation tool for runtime memory locality characterization and analysis of processes and threads running on a NUMA system.
C
139
star
88

ros2_grasp_library

C++
138
star
89

XBB

C++
133
star
90

tdx-tools

Cloud Stack and Tools for Intel TDX (Trust Domain Extension)
C
131
star
91

ros2_intel_realsense

This project is deprecated and no more maintained. Please visit https://github.com/IntelRealSense/realsense-ros for ROS2 wrapper.
C++
131
star
92

linux-intel-lts

C
131
star
93

CeTune

Python
130
star
94

cm-compiler

C++
130
star
95

pti-gpu

Profiling Tools Interfaces for GPU (PTI for GPU) is a set of Getting Started Documentation and Tools Library to start performance analysis on Intel(R) Processor Graphics easily
C++
129
star
96

fMBT

Free Model Based tool
Python
129
star
97

zlib

C
128
star
98

ros_intel_movidius_ncs

C++
126
star
99

mpi-benchmarks

C
125
star
100

mOS

C
124
star