* Remove all openvino related recipes, tests and other data from meta-intel
layer as a new layer (meta-oepnvino) specific to openvino has been created.
* Update openvino documentation
meta-openvino layer URL:
https://github.com/intel/meta-openvino
Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Language changes:
- Added support for non-type template parameters. Uniform integers
and enums can be used now as template parameters.
- Added dot product functions for unsigned and signed int8 and int16
types. They leverage AVX-VNNI and AVX512-VNNI instructions if
supported by targets (docs).
- Added macro definitions for numeric limits.
New targets:
- avx2vnni-i32x4, avx2vnni-i32x8, avx2vnni-i32x16 with AVX-VNNI
instruction support,
- avx512icl-x4, avx512icl-x8, avx512icl-x16, avx512icl-x32 and
avx512icl-x64 with AVX512-VNNI instruction support.
Code generation:
- Fixed generation of code for GPU when unnecessary vectorized
instruction are used during address arithmetic, e.g., for
accessing fields of varying structures (#2846).
- Improved generated code for cases when foreach loop iteration
domain is less than the target width (#2836 ).
Compiler switches behavior:
- --pic command line flag now corresponds to the -fpic flag of Clang
and GCC, whereas the newly introduced --PIC corresponds to -fPIC.
Bug fixes:
- The implementation of round standard library function was aligned
across all targets. It may potentially affect the results of the
code that uses this function for the following targets: avx2-i16x16,
avx2-i8x32 and all avx512 targets (#2793).
- Fixed cases when unwind info were not generated for functions.
This impacted debugging and profiling on Windows (#2842).
- Fixed broken targets sse4-i8xN and avx2-i8xN (#2800).
More details:
https://github.com/ispc/ispc/releases/tag/v1.24.0
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
'syrah' is not a license name. It should be BSD-3-Clause as per the
license listed in third-party-programs.txt.
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
This recipe enables User Mode Driver for Intel® NPU device.
Intel® NPU device is an AI inference accelerator integrated
with Intel client CPUs, starting from Intel® Core™ Ultra generation
of CPUs (formerly known as Meteor Lake).
It enables energy-efficient execution of artificial neural network tasks.
https://github.com/intel/linux-npu-driver
Signed-off-by: Naveen Saini <naveen.kumar.saini@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
We no longer allow building with 14.0 version of LLVM so remove the
recipe and remove the logic to select that recipe version based on
LLVMVERSION value.
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
- Disable NPU plugin for now and enable it in a later change
- Drop backported patch which is now avialable in this version
- Drop patch which is already merged upstream
- Refresh patches
Release Notes:
https://github.com/openvinotoolkit/openvino/releases/tag/2024.1.0
Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Replace usage of WORKDIR in do_compile/do_install by ${S} or ${B}. It
also helps with cases when externalsrc is used like with devtool.
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
- Removed the Cython patch, which is no longer relevant
as Legacy IE python API has been removed:
a561a2a484
- Remove cython dependency as well from recipe
- Remove the switch disabling MLAS support. It's enabled by default.
Fetch the submodule as well and include the file pointing to its
Apache-2.0 license in LIC_FILES_CHKSUM.
- Update conditional check to exclude directories which have "/usr/include"
in them to not add <sysroot>/usr/include as well.
- Backport a patch which solves the build failure caused by setting the
"ENABLE_OV_ONNX_FRONTEND" option to "OFF"
Release Notes:
https://github.com/openvinotoolkit/openvino/releases/tag/2024.0.0
Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Create a separate variable for branch value and derive the value from PV
so AUH is able to upgrade the recipe.
Release Notes:
https://github.com/intel/thermal_daemon/releases/tag/v2.5.6
Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Create a separate variable for branch value and derive the value from PV
so AUH is able to upgrade the recipe.
Release Notes:
https://github.com/oneapi-src/oneDNN/releases/tag/v3.4
Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Drop the patch that has already been merged.
License-Update: OpenCL-Headers are now Apache-2.0 licensed.
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
The instrinsics Python scripts now need mako module as well. Use the
native Python to ensure builds are not dependent on what is installed on
the host.
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
