Searched +full:build +full:- +full:docker +full:- +full:cpu +full:- +full:aarch64 (Results 1 – 25 of 30) sorted by relevance
12
1 name: Build manywheel docker images7 - main8 - release/*10 … # NOTE: Binary build pipelines should only get triggered on release candidate or nightly builds11 # Release candidate tags look like: v1.11.0-rc112 - v[0-9]+.[0-9]+.[0-9]+-rc[0-9]+14 - '.ci/docker/manywheel/*'15 - '.ci/docker/manywheel/build_scripts/*'16 - '.ci/docker/common/*'17 - .github/workflows/build-manywheel-images.yml[all …]
1 name: inductor-perf-nightly-aarch645 # - cron: 0 7 * * 1-66 # - cron: 0 7 * * 07 # Does not perform max_autotune on CPU, so skip the weekly run setup10 - cron: 0 */4 * * *15 # CPU for training is not typical, but leave the option open here47 …-${{ github.event.pull_request.number || github.ref_name }}-${{ github.ref_type == 'branch' && git…48 cancel-in-progress: true50 permissions: read-all53 linux-jammy-aarch64-py3_10-inductor-build:[all …]
4 set -eou pipefail6 TOPDIR=$(git rev-parse --show-toplevel)11 if [ -z "${image}" ]; then18 DOCKER_REGISTRY="${DOCKER_REGISTRY:-docker.io}"20 GPU_ARCH_TYPE=${GPU_ARCH_TYPE:-cpu}21 GPU_ARCH_VERSION=${GPU_ARCH_VERSION:-}22 MANY_LINUX_VERSION=${MANY_LINUX_VERSION:-}23 DOCKERFILE_SUFFIX=${DOCKERFILE_SUFFIX:-}24 WITH_PUSH=${WITH_PUSH:-}27 cpu)[all …]
2 - style3 - build4 - test6 .debian-amd64-common:7 image: registry.videolan.org/dav1d-debian-unstable:202404061425518 stage: build10 - docker11 - amd6413 .debian-amd64-minimum:14 image: registry.videolan.org/dav1d-debian-minimum:20240406142551[all …]
3 This page describes how to build the TensorFlow Lite library for various ARM6 The following instructions have been tested on Ubuntu 16.04.3 64-bit PC (AMD64)7 , TensorFlow devel docker image8 [tensorflow/tensorflow:devel](https://hub.docker.com/r/tensorflow/tensorflow/tags/).15 [Build TensorFlow Lite with CMake](https://www.tensorflow.org/lite/guide/build_cmake)21 and Mendel Linux 4.0. Depending on your target glibc version and CPU22 capabilities, you may need to use different version of toolchain and build28 ldd --version31 <pre class="tfo-notebook-code-cell-output">32 ldd (Debian GLIBC 2.28-10) 2.28[all …]
1 """Repository rule to create a platform for a docker image to be used with RBE."""14 cpu = "x86_64"15 machine_type = repository_ctx.execute(["bash", "-c", "echo $MACHTYPE"]).stdout18 cpu = "ppc"20 cpu = "s390x"21 elif machine_type.startswith("aarch64"):22 cpu = "aarch64"24 cpu = "aarch64"26 cpu = "arm"28 cpu = "mips64"[all …]
1 # Arm NN Build Tool2 …Build Tool_** provides a collection of **parameterized bash scripts** accompanied by a **Dockerfil…3 …places/supersedes the majority of the existing Arm NN build guides as a user-friendly way to build…4 …rm NN from scratch is the ability to **exactly choose which components to build, targeted for your…5 The Arm NN Build Tool is tested on **x86_64 (Intel) and aarch64 (Arm) build hosts** for the **Ubunt…6 Other host platforms such as Windows and Mac **should** work (with Docker installed), but have not …7 …orts **targeting Linux devices (from Ubuntu 18.04 onwards) on x86_64 and aarch64** architectures.<…8 …the Arm NN Build Tool through the use of Docker. However, the scripts may be [executed directly on…10 …Docker installed** and want to quickly build the Arm NN Dockerfile with some default build argumen…13 * [System Requirements](#system-requirements)[all …]
2 - torch/_dynamo/**3 - torch/csrc/dynamo/**4 - benchmarks/dynamo/**5 - test/dynamo/**8 - torch/_inductor/**9 - test/inductor/**12 - torch/_decomp/**13 - torch/_dynamo/**14 - torch/_export/**15 - torch/_inductor/**[all …]
1 # syntax=docker/dockerfile:13 # NOTE: Building this image require's docker version >= 23.0.6 # - https://docs.docker.com/build/dockerfile/frontend/#stable-channel11 FROM ${BASE_IMAGE} as dev-base12 RUN apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends \13 build-essential \14 ca-certificates \19 libjpeg-dev \20 libpng-dev && \21 rm -rf /var/lib/apt/lists/*[all …]
3 <!-- toc -->5 - [Release Compatibility Matrix](#release-compatibility-matrix)6 - [Release Cadence](#release-cadence)7 - [General Overview](#general-overview)8 - [Frequently Asked Questions](#frequently-asked-questions)9 - [Cutting a release branch preparations](#cutting-a-release-branch-preparations)10 - [Cutting release branches](#cutting-release-branches)11 - [`pytorch/pytorch`](#pytorchpytorch)12 - [`pytorch/builder` / PyTorch domain libraries](#pytorchbuilder--pytorch-domain-libraries)13 …- [Making release branch specific changes for PyTorch](#making-release-branch-specific-changes-for…[all …]
3 set -ex6 # (This is set by default in the Docker images we build, so you don't11 # shellcheck source=./common-build.sh12 source "$(dirname "${BASH_SOURCE[0]}")/common-build.sh"14 if [[ "$BUILD_ENVIRONMENT" == *-mobile-*build* ]]; then15 exec "$(dirname "${BASH_SOURCE[0]}")/build-mobile.sh" "$@"19 python --version22 gcc --version25 cmake --version32 export LD_PRELOAD=/usr/lib/x86_64-linux-gnu/libjemalloc.so.2[all …]
4 # (This is set by default in the Docker images we build, so you don't7 set -ex18 # Workaround for dind-rootless userid mapping (https://github.com/pytorch/ci-infra/issues/96)19 WORKSPACE_ORIGINAL_OWNER_ID=$(stat -c '%u' "/var/lib/jenkins/workspace")23 echo "For more details refer to https://github.com/sudo-project/sudo/issues/42"24 sudo chown -R "$WORKSPACE_ORIGINAL_OWNER_ID" /var/lib/jenkins/workspace29 sudo chown -R jenkins /var/lib/jenkins/workspace30 git config --global --add safe.directory /var/lib/jenkins/workspace36 TORCH_INSTALL_DIR=$(python -c "import site; print(site.getsitepackages()[0])")/torch41 BUILD_DIR="build"[all …]
8 # http://www.apache.org/licenses/LICENSE-2.031 if [[ $? -ne 0 ]]; then49 set -x55 mkdir -p /tmpfs/bazel_output62 case "$(uname -s)" in63 Darwin) local name=bazelisk-darwin-amd64 ;;65 case "$(uname -m)" in66 x86_64) local name=bazelisk-linux-amd64 ;;67 aarch64) local name=bazelisk-linux-arm64 ;;68 *) die "Unknown machine type: $(uname -m)" ;;[all …]
8 # http://www.apache.org/licenses/LICENSE-2.0183 """Returns True if running running as a --use_docker child."""198 build = (config_vars.shell + config_vars.builder +205 return PythonConfig(name, build, run)219 # Bit-ness is handled by the test machine's environment262 …# cmake + ninja build is the default because it is faster and supports boringssl assembly optimiza…274 # required to pass as cmake "-A" configuration for VS builds (but not for Ninja)278 … # "x64_x86" means create 32bit binaries, but use 64bit toolkit to secure more memory for the build284 # something illegal when not running under docker.295 self._cmake_configure_extra_args.append('-DOPENSSL_NO_ASM=ON')[all …]
8 # http://www.apache.org/licenses/LICENSE-2.0197 """Returns True if running running as a --use_docker child."""215 build = (227 return PythonConfig(name, build, run, python_path)248 # Bit-ness is handled by the test machine's environment300 …# cmake + ninja build is the default because it is faster and supports boringssl assembly optimiza…319 # required to pass as cmake "-A" configuration for VS builds (but not for Ninja)325 … # "x64_x86" means create 32bit binaries, but use 64bit toolkit to secure more memory for the build333 # something illegal when not running under docker.352 # cmake doesn't build boringssl tests[all …]
... xla hash.", 13 "headRefName": "update-xla-commit-hash/5573005593-54-1", ...
2 fix: compression bug in 32-bit mode associated with long-lasting sessions7 perf: much faster speed for `--patch-from` at high compression levels (#4276)8 perf: higher `--patch-from` compression ratios, notably at high levels (#4288)12 perf: runtime bmi2 detection enabled on x86 32-bit mode (#4251)13 cli: multi-threading as default CLI setting, by @daniellerozenblit14 cli: new `--max` command (#4290)15 build: improve `msbuild` version autodetection, support VS2022, by @ManuelBlanc16 build: fix `meson` build by @artem and @Victor-C-Zhang, and on Windows by @bgilbert17 build: compatibility with Apple Framework, by @Treata1118 build: improve icc/icx compatibility, by @josepho0918 and @luau-project[all …]
1 Mesa 19.3.0 Release Notes / 2019-12-1221 ---------------25 5fa0e4e9dca79560f6882e362f9db36d81cf96da16cf6a84e0ada7466a99a5d7 mesa-19.3.0.tar.xz28 ------------30 - GL_ARB_gl_spirv on i965, iris.31 - GL_ARB_spirv_extensions on i965, iris.32 - GL_EXT_demote_to_helper_invocation on iris, i965.33 - OpenGL 4.6 on i965, iris.34 - EGL_EXT_image_flush_external35 - VK_ANDROID_external_memory_android_hardware_buffer on RADV.[all …]
1 Mesa 23.3.0 Release Notes / 2023-11-2920 ---------------24 50f729dd60ed6335b989095baad81ef5edf7cfdd4b4b48b9b955917cb07d69c5 mesa-23.3.0.tar.xz28 -----------29 - NVK: A Vulkan driver for Nvidia hardware32 ------------33 - VK_EXT_pipeline_robustness on ANV34 - VK_KHR_maintenance5 on RADV35 - OpenGL ES 3.1 on Asahi36 - GL_ARB_compute_shader on Asahi[all …]
1 Mesa 23.2.1 Release Notes / 2023-09-2823 ---------------27 64de0616fc2d801f929ab1ac2a4f16b3e2783c4309a724c8a259b20df8bbc1cc mesa-23.2.1.tar.xz32 ------------34 - VK_EXT_attachment_feedback_loop_dynamic_state on RADV36 - extendedDynamicState3SampleLocationsEnable on RADV38 - VK_EXT_dynamic_rendering_unused_attachments on RADV40 - VK_EXT_mesh_shader on lavapipe42 - OpenGL 3.1 on Asahi44 - OpenGL ES 3.0 on Asahi[all …]
3 # Use of this source code is governed by a BSD-style license that can be13 # pylint: disable=missing-docstring69 # pylint: disable=wildcard-import110 '_-+=>|')242 # large performance hit due to a docker + python2 bug. Eventually297 if len(data[-1]) == 0:368 return (1 << 32) - (1 << 32-bits)414 cpu 324 345 34 5 345470 Read a key-value pair format file into a dictionary, and return it.484 pattern = r'^([-\.\w]+)\{%s\}=(.*)$' % type_tag[all …]
1 # mypy: allow-untyped-defs86 from triton.fb.build import _run_build_command105 def use_global_cache() -> bool:119 def cpp_wrapper_cache_dir(name: str) -> str:121 "cpu"141 def get_system() -> Dict[str, Any]:168 json.dumps(system, sort_keys=True).encode("utf-8")176 def get_local_cache_path() -> Path:181 def get_global_cache_path() -> Optional[Path]:188 def __init__(self) -> None:[all …]
3 = Mbed TLS 3.5.2 branch released 2024-01-2614 could result in an integer overflow, causing a zero-length buffer to be18 = Mbed TLS 3.5.1 branch released 2023-11-0621 * Mbed TLS is now released under a dual Apache-2.0 OR GPL-2.0-or-later28 = Mbed TLS 3.5.0 branch released 2023-10-0531 * Mbed TLS 3.4 introduced support for omitting the built-in implementation32 of ECDSA and/or EC J-PAKE when those are provided by a driver. However,33 there was a flaw in the logic checking if the built-in implementation, in36 accelerated and still have the built-in implementation compiled out.39 considered not accelerated, and the built-in implementation of the curves[all …]
3 = Mbed TLS 3.6.0 branch released 2024-03-2842 * Support Armv8-A Crypto Extension acceleration for SHA-25643 when compiling for Thumb (T32) or 32-bit Arm (A32).44 * AES-NI is now supported in Windows builds with clang and clang-cl.50 This affects both the low-level modules and the high-level APIs53 * Support use of Armv8-A Cryptographic Extensions for hardware acclerated54 AES when compiling for Thumb (T32) or 32-bit Arm (A32).55 * If a cipher or AEAD mechanism has a PSA driver, you can now build the56 library without the corresponding built-in implementation. Generally58 or they'll both be built in. However, for CCM and GCM the built-in[all …]
78 …xperimental/SharpnessAwareMinimization). This class implements the sharpness-aware minimization te…82 …* Added support for cross-trainer data caching in tf.data service. This saves computation resour…86 …_BASED, then it runs a new algorithm that can get the same performance with lower CPU/memory usage.91 …mptionCheckpointHandler) to handle worker preemption/maintenance and cluster-wise consistent error…95 …ge from 8 to 100 times depending on the size of k. When running on CPU and GPU, a non-optimized XL…117 …inistically (and `tf.errors.UnimplementedError` is no longer thrown) when op-determinism has been …141 …-Review-Doctor, DEKHTIARJonathan, Deven Desai, Djacon, Duncan Riach, fedotoff, fo40225, Frederic B…163 …ers should switch to [TensorFlow Decision Forests](https://github.com/tensorflow/decision-forests).164 * Build, Compilation and Packaging166 …eels now specifically conform to [manylinux2014](https://peps.python.org/pep-0599/), an upgrade fr…[all …]