summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* libva-intel-utils: fix mpeg2vldemo build with GCC 16HEADmasterYogesh Tyagi2026-06-102-1/+49
| | | | | | | | | GCC 16 defaults to C++20 and rejects the mixed designated/non-designated clauses in the VAPictureParameterBufferMPEG2 initializer. Backport the upstream fix (commit 5fcab2d02c5a), adapted to the v2.22 release branch which uses GNU colon-style designators. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-compute-runtime: fix build with GCC 16Yogesh Tyagi2026-06-104-0/+176
| | | | | | | | | | Newer GCC (16.x) no longer pulls in some standard headers transitively, breaking the build. For ULONG_MAX, backport the upstream commit that switches to std::numeric_limits (0004). For the remaining fixed-width integer uses, add the missing <cstdint> includes (0005, 0006). Also relax the new -Wsfinae-incomplete diagnostic so -Werror does not fail. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* recipes-support: intel-crypto-mb: avoid Werror failure on SM4Yogesh Tyagi2026-06-011-0/+4
| | | | Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* linux-intel/linux-intel-rt: bump SRCREV_meta to fbc4b73674baYogesh Tyagi2026-05-202-2/+2
| | | | | | Update kernel meta SRCREV for 6.18.15 to pick up latest kernel-cache updates. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-oneapi-toolkit: split mkl and ipp into runtime/sycl/staticdev/dev ↵Yogesh Tyagi2026-05-171-13/+100
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | sub-packages The 2026.0 oneAPI MKL component grew to 1.9 GB on disk, dominated by ~900 MB of static archives, ~330 MB of SYCL/GPU offload libraries and ~80 MB of headers / cmake metadata / sample sources. The IPP component similarly carries ~280 MB of static archives on top of a 537 MB runtime payload. Consumers that only need a CPU BLAS/FFT backend (PyTorch CPU, NumPy, image-processing pipelines) were paying for all of it whenever they installed intel-oneapi-toolkit-mkl or intel-oneapi-toolkit-ipp. Split the two heaviest components along the usual runtime/dev/static seams: intel-oneapi-toolkit-mkl CPU shared libs + dispatch libraries + env/etc/bin glue (~640 MB) intel-oneapi-toolkit-mkl-sycl SYCL / GPU offload shared libs (~330 MB, RDEPENDS -runtime) intel-oneapi-toolkit-mkl-staticdev Static archives, lib/*.a (~900 MB) intel-oneapi-toolkit-mkl-dev Headers, cmake modules, pkgconfig files, share/ examples (~85 MB) intel-oneapi-toolkit-ipp ipp + ippcp runtime shared libs (~540 MB) intel-oneapi-toolkit-ipp-staticdev Static archives (~280 MB) intel-oneapi-toolkit-ipp-dev Headers, cmake, pkgconfig, share PACKAGES ordering places the specific sub-sub-packages before the catch-all component package so first-match-wins puts the heavyweight bits in their dedicated package. FILES patterns use [0-9]* to match the versioned directory only (the sibling 'latest' compatibility symlink is captured by the parent component package). RDEPENDS form a closed chain: -dev depends on the runtime package, -staticdev on -dev, -sycl on -runtime + the CPU runtime, so users that explicitly install a sub-sub-package still get a working environment. The top-level meta-package RDEPENDS:${PN} continues to pull all of the new sub-packages, so an image that asks for the unqualified intel-oneapi-toolkit recipe is unchanged. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com> Tested-by: Zoltán Böszörményi <zboszor@gmail.com>
* libipt: drop pttc nasm patch and runtime QA caseYogesh Tyagi2026-05-123-203/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The 0001-pttc-use-nasm-instead-of-yasm.patch local patch reworked upstream pttc to invoke nasm in place of the unmaintained yasm assembler. The patch was submitted upstream (https://github.com/intel/libipt/pull/120) but has not been merged, and carrying it locally is the only reason meta-intel needs to diverge from a pristine stable/v2.2 checkout. Drop the patch and let pttc fall back to upstream's stock yasm-based behaviour. This matches what Fedora/RHEL already does for the libipt package: ship libipt without yasm support and let libipt's own CMake skip the yasm-driven ptt tests automatically when yasm isn't available. Native nasm support in pttc remains a future upstream enhancement; when it lands we can revisit and re-add a runtime case that exercises pttc end-to-end against nasm. The libipt runtime test (lib/oeqa/runtime/cases/libipt.py) gated on the 'nasm' package and ran pttc on the shipped sample .ptt files. With the patch removed pttc invokes yasm at runtime, so the test would fail on every DUT (yasm is not pulled in as an RDEPENDS anywhere in meta-intel). The test exercised pttc/ptdump end-to-end, not the libipt library itself, and library coverage is provided indirectly by anything that links against libipt2; the runtime case has no remaining utility once the assembler-substitution patch is gone. No change to the recipe's PACKAGES, FILES, or EXTRA_OECMAKE: the libipt-test sub-package still ships pttc, ptdump, ptseg and the .ptt sample sources under ${bindir}/libipt for any user that installs yasm out-of-band and wants to drive pttc manually. This matches the Fedora libipt RPM behaviour. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-vaapi-driver: drop archived i965 VA-API backendYogesh Tyagi2026-05-124-63/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Upstream https://github.com/intel/intel-vaapi-driver was archived in May 2024 and no longer receives fixes. The i965 driver targeted Gen4-Gen9 Intel GPUs and has been functionally superseded on Gen8+ hardware by intel-media-driver (iHD), which is already the VA-API backend used by every meta-intel BSP machine that ships modern graphics. Remove the recipe, its runtime test case, the maintainer entry, and the only in-tree consumer (intel-core2-32's MACHINE_HWCODECS, which now keeps just the gstreamer1.0-plugins-bad codec set since no maintained VA-API backend supports that legacy 32-bit Core 2 hardware anyway). Recipes / files removed: * recipes-multimedia/libva/intel-vaapi-driver_2.4.1.bb * lib/oeqa/runtime/cases/intel_vaapi_driver.py Other changes: * conf/machine/intel-core2-32.conf: drop intel-vaapi-driver from MACHINE_HWCODECS * conf/include/maintainers.inc: drop pn-intel-vaapi-driver entry Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-oneapi: drop obsolete standalone component recipesYogesh Tyagi2026-04-296-234/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove the per-component apt-package recipes superseded by the unified intel-oneapi-toolkit_2026.0.0.198.bb recipe, which packages the same components as sub-packages of a single offline installer (intel-oneapi-toolkit-{compiler,runtime,mkl,tbb,dpl,ipp,...}). Consumers should reference the canonical toolkit sub-package names directly. Recipes removed: * intel-oneapi-dpcpp-cpp_2024.0.0-49819.bb * intel-oneapi-dpcpp-cpp-runtime_2024.0.0-49819.bb * intel-oneapi-mkl_2024.0.0-49656.bb * intel-oneapi-ipp_2021.10.0-653.bb * setup-intel-oneapi-env_2023.0.0-25370.bb (and its intel-oneapi-runtime.conf helper) The setup-intel-oneapi-env helper installed a single ld.so.conf.d snippet pointing the dynamic loader at /opt/intel/oneapi/lib (and lib/intel64, lib/ia32, lib/emu). It dates back to the 2023.0 oneAPI layout. The unified toolkit installs libs under per-component directories such as /opt/intel/oneapi/{compiler,mkl, tbb,...}/2026.0/lib, not the flat /opt/intel/oneapi/lib that the conf file targets, so the ld.so.conf entries no longer point at real directories. Loadability of oneAPI binaries on the target is provided by: * setvars.sh / per-component env/vars.sh sourcing (the canonical Intel-supported entry point that every test and example uses) * $ORIGIN-relative RUNPATH baked into the shipped shared objects by the Intel installer * /etc/OpenCL/vendors/intel-cpu.icd written by the unified toolkit recipe's do_install for ICD loader discovery * virtual-opencl-icd / level-zero-loader RDEPENDS on intel-oneapi-toolkit-runtime for the OpenCL / Level Zero loaders The setup-intel-oneapi-env filename version (2023.0.0-25370) was a stale carry-over from when the recipe was first added; nothing in the recipe ever referenced ${PV}. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* packagegroup-sdk: switch ICXSDK to intel-oneapi-toolkit sub-packagesYogesh Tyagi2026-04-292-2/+2
| | | | | | | | | | | | | | | | | The 2024.0 standalone recipes provided per-component package names (intel-oneapi-dpcpp-cpp{,-dev}, intel-oneapi-dpcpp-cpp-runtime{,-dev}) that have been replaced in the unified intel-oneapi-toolkit_2026 recipe by the canonical sub-package layout. Update the SDK host and target packagegroup bbappends to reference the new names directly: intel-oneapi-toolkit-compiler - C/C++/SYCL compiler + headers intel-oneapi-toolkit-runtime - shared runtime libraries The toolkit-compiler sub-package already contains the headers and CMake/pkg-config files the prior *-dev sub-packages exposed, so a separate -dev append is no longer needed. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-oneapi-toolkit: add unified 2026.0.0.198 recipeYogesh Tyagi2026-04-292-0/+353
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Intel oneAPI 2026.0 ships the former Base and HPC toolkits as a single offline self-extracting installer instead of as the per- component apt packages used in the 2021/2024 layout. Add a recipe that consumes intel-oneapi-toolkit-2026.0.0.198_offline.sh via a pseudo-free do_unpack_vendor task and stages /opt/intel/oneapi into the rootfs. The recipe partitions the installed tree into sub-packages along component boundaries: intel-oneapi-toolkit-{compiler,runtime,mkl, tbb,dpl,debugger,ipp,ccl,vtune,common,licensing}. Consumers should reference these canonical sub-package names directly rather than the per-component apt names that the 2021/2024 standalone recipes exposed. The runtime sub-package additionally: * registers /opt/intel/oneapi/compiler/2026.0/lib/libintelocl.so as /etc/OpenCL/vendors/intel-cpu.icd so systems without an Intel GPU expose an Intel OpenCL platform via the standard ICD mechanism; * installs a libxml2.so.2 -> libxml2.so.16 SONAME compat symlink, because the prebuilt libcommon_clang.so / libintelocl.so are linked against the libxml2 2.x SONAME while current oe-core ships libxml2 3.x as libxml2.so.16. libxml2 3.x is API-compatible with the option-parsing surface the binaries use. The bundle is x86_64-only so COMPATIBLE_HOST is restricted. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* classes: add oneapi-vendor-compat bbclassYogesh Tyagi2026-04-291-0/+76
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Intel oneAPI DPC++/C++ compiler binaries (icx, icpx) shipped by meta-intel as prebuilt installers have a fixed default -target triple baked into the LLVM driver. As of the 2025.3.x bundle the triple is 'x86_64-unknown-linux-gnu' (LLVM's portable default); prior 2024.x bundles used 'x86_64-oe-linux' (matching the OpenEmbedded nodistro default of TARGET_VENDOR=oe). icx also emits 'x86_64-linux-gnu' (the Debian/Ubuntu host triple) in some search paths. Yocto distros set TARGET_VENDOR differently: poky uses 'poky' (TARGET_SYS=x86_64-poky-linux), nodistro uses 'oe', etc. The result is a vendor-triple mismatch: icx looks for crtbeginS.o / bits/c++config.h and similar files under one triple while they are installed at another. Until Intel either ships per-vendor variants or the meta-intel oneAPI recipes patch the prebuilt binaries, install a set of compatibility symlinks at rootfs time so every spelling of the multilib vendor triple resolves to the same files on the image. The alias set is configurable via ONEAPI_VENDOR_TRIPLE_ALIASES; entries that already match the native target triple are skipped, so the class is a no-op on images whose native vendor matches every alias. Usage: add to IMAGE_CLASSES in a distro / local config that consumes the Intel oneAPI compiler: IMAGE_CLASSES += "oneapi-vendor-compat" Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* ispc: isolate LLVM superbuild from recipe-sysroot-nativeYogesh Tyagi2026-04-081-37/+46
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The ISPC superbuild internally builds LLVM (stage1/stage2). The generated toolchain files inherit Yocto flags and sysroot paths, causing CMake to locate libraries from recipe-sysroot-native. This leads to linking failures when host compiler toolchains are used together with libraries built against a newer glibc, resulting in errors such as: undefined reference to `__isoc23_wcstol@GLIBC_2.38' Fix this by isolating the LLVM superbuild from Yocto sysroot: * Patch generated stage1/stage2 toolchain files: - Replace symbolic compiler names (cc/c++) with ${BUILD_CC}/${BUILD_CXX} - Clear injected CMake compiler and linker flags * Prevent CMake from searching recipe-sysroot-native: - Set CMAKE_FIND_ROOT_PATH_MODE_* to NEVER * Disable optional LLVM features requiring external libraries (libedit, terminfo, zstd, zlib) * Scope environment cleanup to the cmake build invocation to avoid leaking Yocto flags into the superbuild This ensures the internal LLVM build uses a consistent host environment, avoiding GLIBC ABI mismatches. Remove ncurses dependency as terminfo/libedit are disabled. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* meta-intel/wic: Move to files/wicYogesh Tyagi2026-04-085-0/+0
| | | | | | To align with the new search paths, move the files to the new location. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* perf: drop libbpf const qualifier workaroundYogesh Tyagi2026-03-201-17/+0
| | | | | | | | | | | | | Remove the bbappend that applied const qualifier fixes to libbpf.c. The workaround was required for older kernel versions where building perf with -Werror triggered warnings due to discarded const qualifiers from strstr() and strchr(). The issue is resolved in the currently used linux-intel 6.18 kernel, making this workaround no longer necessary. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* layer.conf: Add the new release layer seriesYogesh Tyagi2026-03-191-1/+1
| | | | | | Add the wrynose release as being compatible with the layer. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* meta-intel.inc: set PREFERRED_VERSION for linux-intel to 6.18Yogesh Tyagi2026-03-181-4/+4
| | | | | | | | | Update default kernel version preference to 6.18 for both RT and Non RT variants. 6.18 continues to use a single branch for both RT and non-RT kernels. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* linux-intel-rt: upgrade 6.12.61 -> 6.18.15Yogesh Tyagi2026-03-181-6/+6
| | | | | | | | | | | Upgrade to Intel LTS RT kernel 6.18.15. Uses same refreshed patches as linux-intel. Release Notes: https://github.com/intel/linux-intel-lts/releases/tag/lts-v6.18.15-linux-260310T050801Z Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* linux-intel: upgrade 6.12.61 -> 6.18.15Yogesh Tyagi2026-03-181-6/+6
| | | | | | | | | Upgrade to Intel LTS kernel 6.18.15. Release Notes: https://github.com/intel/linux-intel-lts/releases/tag/lts-v6.18.15-linux-260310T050801Z Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* conf/machine, oeqa: replace gstreamer1.0-vaapi with gst-plugins-badYogesh Tyagi2026-03-184-4/+4
| | | | | | | | | | | | | | | | | | | | | | | gstreamer1.0-vaapi has been removed from openembedded-core in favor of the va plugin in gst-plugins-bad and is no longer available. This updates all Intel machine configurations to use gstreamer1.0-plugins-bad instead of the obsolete gstreamer1.0-vaapi package in MACHINE_HWCODECS. The VA-API functionality is now provided by the va plugin within gst-plugins-bad, which contains the same VA-API GStreamer elements (vah264dec, vah265dec, vapostproc, etc.). Also updates the runtime test case to check for gstreamer1.0-plugins-bad instead of gstreamer1.0-vaapi. Affected machines: - intel-core2-32 - intel-corei7-64 - intel-skylake-64 Reference: openembedded-core commit 9e2d2a5b0c9e062f13651093bb1e459f210618e6 (gstreamer1.0-vaapi: remove recipe) Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* perf: Fix const qualifier warnings in libbpf.cYogesh Tyagi2026-03-131-0/+17
| | | | | | | | | | | | | | | | | | | | | | Fix compilation warnings when building perf with -Werror where const qualifiers are discarded from pointers returned by strstr() and strchr() functions in libbpf.c. Both strstr() and strchr() return 'const char *' when given 'const char *' input parameters. The receiving variables must preserve this const qualification to maintain const-correctness and avoid compiler warnings. Errors fixed: libbpf.c:8179:13: error: assignment discards 'const' qualifier from pointer target type [-Werror=discarded-qualifiers] libbpf.c:11967:35: error: assignment discards 'const' qualifier from pointer target type [-Werror=discarded-qualifiers] This is a workaround for kernel 6.12.x. The issue is already fixed upstream in kernel 6.13+ and can be removed when the kernel is updated. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* oeqa : drop libxcam test scriptYogesh Tyagi2026-03-031-37/+0
| | | | | | libxcam recipe was already dropped. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* oidn : drop old recipeYogesh Tyagi2026-03-031-27/+0
| | | | Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* thermald: remove and migrate to meta-oeAnuj Mittal2026-02-252-44/+0
| | | | | | | | | | thermald now supports ARM platforms [1], so move it out of meta-intel to allow broader reuse. [1] thermal_daemon 1931a12e7e ("Add ARM backend and enable ARM platform detection") Signed-off-by: Anuj Mittal <anuj.mittal@oss.qualcomm.com> Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-microcode: upgrade 20251111 -> 20260210Adam Duskett2026-02-251-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Update for functional issues. Refer to processor specification updates for details. Updated Platforms: - Core Gen12 - Core i3-N305/N300, N50/N97/N100/N200, Atom x7211E/x7213E/x7425E - Core Ultra Processor (Series 2) (ARL-H, ARL-S/HX (8P), ARL-U) - Atom C1100 - Xeon Scalable Gen5 - Xeon 6900/6700/6500 Series Processors with P-Cores - Xeon 6700P-B/6500P-B Series SoC with P-Cores - Xeon 6700/6500-Series Processors with P-Cores - Xeon D-17xx, D-27xx - Core Gen10 Mobile - Xeon Scalable Gen3 - Core Ultra Processor - Core Gen11 - Core Gen13/Gen14 (RPL-E/HX/S, RPL-S, RPL-HX/S) - Core Gen13 (RPL-H/P/PX 6+8, RPL-U 2+8) - Xeon Max - Xeon Scalable Gen4 - Core Gen11 Mobile (TGL, TGL-H, TGL-R) - Core i3-N305/N300, N50/N97/N100/N200, Atom x7211E/x7213E/x7425E Release notes: https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files/releases/tag/microcode-20260210-rev1 Signed-off-by: Adam Duskett <adam.duskett@amarulasolutions.com> Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* maintainers: Update recipe maintainersYogesh Tyagi2026-02-241-43/+43
| | | | Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* Revert "intel-media-driver : upgrade 25.1.4 -> 26.1.0"Yogesh Tyagi2026-02-181-1/+1
| | | | | | This reverts commit deccd450234efe2fcf3565d0cbd976d9f08c1816. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* metee: Use targeted pragma to suppress sign-conversion warningsYogesh Tyagi2026-02-182-1/+63
| | | | Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* embree, openvkl, ospray: Skip buildpaths QA checkYogesh Tyagi2026-02-173-0/+8
| | | | | | | | | | | | | | | | | | | | | | | | | Add INSANE_SKIP for buildpaths QA check in RenderKit recipes that use ISPC for code generation. ISPC (Intel SPMD Program Compiler) embeds absolute build paths in generated binaries and debug information by design. This is necessary for: 1. Debug information accuracy - ISPC generates C++ wrapper code with #line directives pointing to original .ispc source files 2. Error reporting during kernel execution - backtraces reference original source locations 3. Profiling support - performance tools need source file locations Affected packages: - embree: ISPC kernels for ray tracing primitives - openvkl: ISPC kernels for volume rendering - ospray: ISPC kernels for rendering algorithms This resolves QA errors that prevent SDK image builds when RenderKit stack is included. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* oidn: Add recipe for version 2.4.1Yogesh Tyagi2026-02-171-0/+32
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add Intel Open Image Denoise (OIDN) 2.4.1, an open-source library for high-performance, high-quality denoising of images rendered with ray tracing. Key features: - Machine learning-based denoising filters optimized for Intel CPUs - ISPC-accelerated kernels for vectorization - Supports HDR and LDR images - Multiple filter types: RT, RTLightmap Dependencies: - ISPC 1.25.0+ for kernel compilation - TBB for task parallelism - CMake 3.1+ Build configuration: - GPU support disabled (OIDN_DEVICE_GPU=OFF) - Apps disabled for minimal footprint - Static TBB linkage Part of Intel RenderKit stack upgrade coordinated with: - ISPC 1.29.1 - embree 4.4.0 - openvkl 2.0.1 - ospray 3.2.0 - rkcommon 1.15.2 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* ispc: Upgrade to version 1.29.1 with build fixes and ptest supportYogesh Tyagi2026-02-1711-205/+384
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Upgrade from 1.24.0 to 1.29.1 with bundled LLVM 20.1 using superbuild. Recipe changes: - Upgrade to ISPC 1.29.1 with bundled LLVM 20.1 using superbuild - Add patches for multiarch header detection and CMake policy updates - Split do_install() into base (native+target) and target-only sections * Base: Install compiler binaries (ispc, check_isa) to ${bindir} * Target-only: Install test suite to ${libdir}/ispc/gio/ - Make do_install_ptest() target-only to avoid native build issues - Create separate test packages with class-specific definitions: * ispc-ptest: Yocto ptest integration at /usr/lib/ispc/ptest/ - Add run-ptest script for ptest-runner integration - Add pkg_postinst to create /lib64 symlink for binary compatibility Dependencies and packaging: - Add ncurses to DEPENDS for libedit linking (terminfo functions) - Skip file-rdeps and buildpaths QA checks for bundled LLVM - Set RDEPENDS for ptest: bash, python3-multiprocessing - Make PACKAGES, FILES, and RDEPENDS class-specific for proper separation Build configuration: - Use superbuild with bundled LLVM 20.1, vc-intrinsics, SPIRV-Translator - Forward ISPCRT_BUILD_TASK_MODEL for TBB support - Set custom build/install directories for superbuild stage2 output - Export empty CFLAGS/CXXFLAGS/LDFLAGS_FOR_BUILD for clean builds Fixes: - ispc-native build failures when test packages were defined - Link errors with libedit.so requiring ncurses terminfo symbols - QA errors from duplicate PACKAGES definitions - Proper class separation between native and target builds - GLIBC_2.42 symbol resolution issues Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* ixgbevf: upgrade 4.19.10 -> 5.2.2Yogesh Tyagi2026-02-101-5/+5
| | | | | | | Updated source URL from Sourceforge to GitHub as Intel ethernet driver project has migrated to github.com/intel/ethernet-linux-ixgbevf. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* ixgbe: upgrade 5.20.10 -> 6.3.4Yogesh Tyagi2026-02-101-5/+5
| | | | | | | Updated source URL from Sourceforge to GitHub as Intel ethernet driver project has migrated to github.com/intel/ethernet-linux-ixgbe. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* recipes: inherit pkgconfigRoss Burton2026-02-106-12/+14
| | | | | | | | | | Instead of depending directly on pkgconfig-native, inherit the pkgconfig class. At the moment this just adds a layer of indirection to the right provider of pkgconfig, but there are patches in flight to move exports into the class to isolate the variables. Signed-off-by: Ross Burton <ross.burton@arm.com> Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* metee: Fix build failure with musl libcYogesh Tyagi2026-02-062-0/+49
| | | | | | | | | | | | | Add patch to fix sign-conversion warnings with musl ioctl macros. musl's ioctl request macros expand to unsigned long values, causing sign-conversion warnings when passed to ioctl() which expects an int. Cast the macros explicitly to int in three ioctl calls: - IOCTL_MEI_CONNECT_CLIENT - IOCTL_MEI_CONNECT_CLIENT_VTAG - IOCTL_MEI_NOTIFY_GET Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* linux-npu-driver: Set COMPATIBLE_HOST to restrict compatibility to x86_64 hostsYongxin Liu2026-02-051-0/+2
| | | | | | | | | | | | | - Added "COMPATIBLE_HOST" to restrict the "linux-npu-driver" recipe to x86_64 Linux-based build systems. - Ensured that the recipe is not compatible with systems using musl libc. This change ensures the recipe builds only on x86_64 hosts with glibc and avoids issues with musl-based systems. Signed-off-by: Yongxin Liu <yongxin.liu@windriver.com> Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* libipt: Switch from yasm to nasm assemblerYogesh Tyagi2026-02-022-1/+180
| | | | | | | | | | | | | | | | | | | | Replace yasm with nasm as yasm is no longer maintained and has unpatched CVEs (CVE-2024-45534, CVE-2024-45535). Nasm is actively maintained and generates compatible listing files. The patch includes: - Changed assembler invocation from yasm to nasm - Removed unsupported -L nasm parameter - Added dual-format parser for org directives (yasm and nasm syntax) - Implemented 1:1 line mapping fallback for nasm listing format - Updated documentation Both test cases (loop-tnt.ptt and dump-all-packets.ptt) now pass on target system. Upstream-Status: Submitted [https://github.com/intel/libipt/pull/120] Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-compute-runtime: add conflict with libopencl-mesaYogesh Tyagi2026-02-011-0/+4
| | | | | | | | | | | | | | | | | | | | | | Add RCONFLICTS to prevent libopencl-mesa (mesa's rusticl) and intel-compute-runtime from being installed together. Both packages provide OpenCL ICD implementations that conflict at runtime. The conflict manifests when both ICDs are present in /etc/OpenCL/vendors/, leading to: - Multiple OpenCL platforms being detected - Wrong platform being selected by applications - Segmentation faults due to LLVM version conflicts between rusticl and intel-compute-runtime - intel-compute-runtime tests failing Add RPROVIDES for virtual-opencl-icd so intel-compute-runtime can satisfy OpenCL ICD dependencies from other packages (e.g., intel-oneapi-dpcpp-cpp-runtime). This ensures only one OpenCL implementation is active, preventing runtime conflicts for all users. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* ospray : add patch for upgraded dependenciesYogesh Tyagi2026-01-272-0/+44
| | | | | | | | | | | | | | | | | | | | Add patch to update ospray's dependency version requirements: - rkcommon: 1.14.0 -> 1.15.2 (remove EXACT matching) - embree: 4.3.3 -> 4.4.0 Removing EXACT version matching for rkcommon is safe because: 1. rkcommon declares COMPATIBILITY SameMajorVersion in CMake 2. Changes from 1.14.0 to 1.15.2 are backward compatible: - Added shutdownTaskingSystem() function (new, non-breaking) - Optimized frame() math algorithm (same result, better performance) - Bug fixes in tasking system (improved reliability) - Compiler compatibility improvements (icx, ARM) 3. No API breakage: zero functions removed or signatures changed The 1.15.2 version provides performance improvements and bug fixes while maintaining full API compatibility with 1.14.0. Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* embree : upgrade 4.3.3 -> 4.4.0Yogesh Tyagi2026-01-271-1/+1
| | | | | | | Release Notes: https://github.com/embree/embree/releases/tag/v4.4.0 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* rkcommon : upgrade 1.14.0 -> 1.15.2Yogesh Tyagi2026-01-271-2/+1
| | | | | | | | | | Dropped 0001-use-fully-qualified-rkcommon-math-rsqrt-to-avoid-ove.patch as it has been merged upstream. Release Notes: https://github.com/ospray/rkcommon/releases/tag/v1.15.2 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* onedpl : upgrade 2022.8.0 -> 2022.11.0Yogesh Tyagi2026-01-271-2/+2
| | | | | | | Release Notes: https://github.com/uxlfoundation/oneDPL/releases/tag/oneDPL-2022.11.0-release Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* libipt : upgrade 2.1.2 -> 2.2Yogesh Tyagi2026-01-271-3/+4
| | | | | | | | | | License update : Update copyright format from 'Copyright (c)' to 'Copyright (C)' in BSD-3-Clause LICENSE file. Release Notes: https://github.com/intel/libipt/releases/tag/v2.2 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* vpl-gpu-rt : upgrade 25.1.4 -> 26.1.0Yogesh Tyagi2026-01-271-1/+1
| | | | | | | Release Notes: https://github.com/intel/vpl-gpu-rt/releases/tag/intel-onevpl-26.1.0 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* libvpl-tools : upgrade 1.3.0 -> 1.5.0Yogesh Tyagi2026-01-271-3/+1
| | | | | | | | | | | Drop 2 patches as they are merged upstream: - 0001-Update-vpl-inspect-to-remove-ReportedStats.patch - 0001-Add-cstdint-header-to-resolve-build-issue-with-gcc-1.patch Release Notes: https://github.com/intel/libvpl-tools/releases/tag/v1.5.0 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* libvpl : upgrade 2.14.0 -> 2.16.0Yogesh Tyagi2026-01-271-8/+3
| | | | | | | | | | | | | License update : Remove Microsoft Visual C++ Redistributable, Intel Media SDK, Intel GMMLib and Intel Media Driver entries from third-party-programs.txt. Remove do_install:append as examples are now installed as source. Release Notes: https://github.com/intel/libvpl/releases/tag/v2.16.0 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* intel-media-driver : upgrade 25.1.4 -> 26.1.0Yogesh Tyagi2026-01-271-1/+1
| | | | | | | | Release Notes: https://github.com/intel/media-driver/releases/tag/ intel-media-26.1.0 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* itt : upgrade 3.26.1 -> 3.26.5Yogesh Tyagi2026-01-271-1/+1
| | | | | | | Release Notes: https://github.com/intel/ittapi/releases/tag/v3.26.5 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* linux-intel-rt : upgrade 6.12.55 -> 6.12.61Yogesh Tyagi2026-01-271-3/+3
| | | | | | | | | | Uses same refreshed patches as linux-intel. Release Notes: https://github.com/intel/linux-intel-lts/releases/tag/ lts-v6.12.61-linux-251229T065211Z Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* linux-intel : upgrade 6.12.55 -> 6.12.61Yogesh Tyagi2026-01-272-23/+24
| | | | | | | | | | Refresh 0002-mconf patch for new kconfig infrastructure. Release Notes: https://github.com/intel/linux-intel-lts/releases/tag/ lts-v6.12.61-linux-251229T065211Z Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
* backport-iwlwifi : upgrade core96 -> core102Yogesh Tyagi2026-01-271-3/+3
| | | | | | | | | | | | | Update to latest upstream release/core102 branch. This release includes various bug fixes and improvements to the Intel wireless LinuxCore kernel driver. Release Notes: https://git.kernel.org/pub/scm/linux/kernel/git/iwlwifi/ backport-iwlwifi.git/log/?h=release/core102 Signed-off-by: Yogesh Tyagi <yogesh.tyagi@intel.com>
generated by cgit v1.2.3-54-g00ecf (git 2.39.0) at 2026-10-14 01:38:23 +0000